From 295c01f90db2b8b8394e1f7cde59b96d71f98e07 Mon Sep 17 00:00:00 2001 From: Christian Hohnstaedt Date: Sun, 12 Apr 2009 13:01:44 +0800 Subject: [PATCH 001/267] crypto: ixp4xx - check firmware for crypto support - the loaded firmware may not support crypto at all or only support DES and 3DES but not AES or support DES, 3DES and AES. - in case of no crypto support of the firmware, the module load will fail. - in case of missing AES support, the AES algorithms are not registered and a warning is printed during module load. Signed-off-by: Christian Hohnstaedt Signed-off-by: Herbert Xu --- drivers/crypto/ixp4xx_crypto.c | 33 ++++++++++++++++++++++++++++++++- 1 file changed, 32 insertions(+), 1 deletion(-) diff --git a/drivers/crypto/ixp4xx_crypto.c b/drivers/crypto/ixp4xx_crypto.c index af9761ccf9f1..9224c1f49311 100644 --- a/drivers/crypto/ixp4xx_crypto.c +++ b/drivers/crypto/ixp4xx_crypto.c @@ -415,6 +415,7 @@ static void crypto_done_action(unsigned long arg) static int init_ixp_crypto(void) { int ret = -ENODEV; + u32 msg[2] = { 0, 0 }; if (! ( ~(*IXP4XX_EXP_CFG2) & (IXP4XX_FEATURE_HASH | IXP4XX_FEATURE_AES | IXP4XX_FEATURE_DES))) { @@ -426,9 +427,35 @@ static int init_ixp_crypto(void) return ret; if (!npe_running(npe_c)) { - npe_load_firmware(npe_c, npe_name(npe_c), dev); + ret = npe_load_firmware(npe_c, npe_name(npe_c), dev); + if (ret) { + return ret; + } + if (npe_recv_message(npe_c, msg, "STATUS_MSG")) + goto npe_error; + } else { + if (npe_send_message(npe_c, msg, "STATUS_MSG")) + goto npe_error; + + if (npe_recv_message(npe_c, msg, "STATUS_MSG")) + goto npe_error; } + switch ((msg[1]>>16) & 0xff) { + case 3: + printk(KERN_WARNING "Firmware of %s lacks AES support\n", + npe_name(npe_c)); + support_aes = 0; + break; + case 4: + case 5: + support_aes = 1; + break; + default: + printk(KERN_ERR "Firmware of %s lacks crypto support\n", + npe_name(npe_c)); + return -ENODEV; + } /* buffer_pool will also be used to sometimes store the hmac, * so assure it is large enough */ @@ -459,6 +486,10 @@ static int init_ixp_crypto(void) qmgr_enable_irq(RECV_QID); return 0; + +npe_error: + printk(KERN_ERR "%s not responding\n", npe_name(npe_c)); + ret = -EIO; err: if (ctx_pool) dma_pool_destroy(ctx_pool); From abe5fa7899fb5809ddc7336d8dd0edd5b2b96665 Mon Sep 17 00:00:00 2001 From: Steffen Klassert Date: Wed, 15 Apr 2009 20:45:03 +0800 Subject: [PATCH 002/267] crypto: eseqiv - Fix IV generation for sync algorithms If crypto_ablkcipher_encrypt() returns synchronous, eseqiv_complete2() is called even if req->giv is already the pointer to the generated IV. The generated IV is overwritten with some random data in this case. This patch fixes this by calling eseqiv_complete2() just if the generated IV has to be copied to req->giv. Signed-off-by: Steffen Klassert Signed-off-by: Herbert Xu --- crypto/eseqiv.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/crypto/eseqiv.c b/crypto/eseqiv.c index 2a342c8e52b3..3ca3b669d5d5 100644 --- a/crypto/eseqiv.c +++ b/crypto/eseqiv.c @@ -153,7 +153,8 @@ static int eseqiv_givencrypt(struct skcipher_givcrypt_request *req) if (err) goto out; - eseqiv_complete2(req); + if (giv != req->giv) + eseqiv_complete2(req); out: return err; From 37fc334cc8eb84f5fe0a5a1cbe6a6a68049e142a Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 21 Apr 2009 13:27:16 +0800 Subject: [PATCH 003/267] crypto: api - Fix algorithm module auto-loading The commit a760a6656e6f00bb0144a42a048cf0266646e22c (crypto: api - Fix module load deadlock with fallback algorithms) broke the auto-loading of algorithms that require fallbacks. The problem is that the fallback mask check is missing an and which cauess bits that should be considered to interfere with the result. Reported-by: Chuck Ebbert Signed-off-by: Herbert Xu --- crypto/api.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/crypto/api.c b/crypto/api.c index 314dab96840e..fd2545decb28 100644 --- a/crypto/api.c +++ b/crypto/api.c @@ -221,7 +221,8 @@ struct crypto_alg *crypto_larval_lookup(const char *name, u32 type, u32 mask) request_module(name); - if (!((type ^ CRYPTO_ALG_NEED_FALLBACK) & mask) && + if (!((type ^ CRYPTO_ALG_NEED_FALLBACK) & mask & + CRYPTO_ALG_NEED_FALLBACK) && snprintf(tmp, sizeof(tmp), "%s-all", name) < sizeof(tmp)) request_module(tmp); From acd246b7494c629aa617da49716409566cf52149 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 21 Apr 2009 13:55:20 +0800 Subject: [PATCH 004/267] crypto: padlock - Revert aes-all alias to aes Since the padlock-aes driver doesn't require a fallback (it's only padlock-sha that does), it should use the aes alias rather than aes-all so that ones that do need a fallback can use it. Signed-off-by: Herbert Xu --- drivers/crypto/padlock-aes.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/crypto/padlock-aes.c b/drivers/crypto/padlock-aes.c index 3f0fdd18255d..856b3cc25583 100644 --- a/drivers/crypto/padlock-aes.c +++ b/drivers/crypto/padlock-aes.c @@ -489,4 +489,4 @@ MODULE_DESCRIPTION("VIA PadLock AES algorithm support"); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Michal Ludvig"); -MODULE_ALIAS("aes-all"); +MODULE_ALIAS("aes"); From 818cf5909701806285d977f7a9365c5cadb062a7 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Thu, 23 Apr 2009 09:58:22 +0300 Subject: [PATCH 005/267] slub: enforce MAX_ORDER slub_max_order may not be equal to or greater than MAX_ORDER. Additionally, if a single object cannot be placed in a slab of slub_max_order, it still must allocate slabs below MAX_ORDER. Acked-by: Christoph Lameter Signed-off-by: David Rientjes Signed-off-by: Pekka Enberg --- mm/slub.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/mm/slub.c b/mm/slub.c index 7ab54ecbd3f3..0e1247ed2a06 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -1909,7 +1909,7 @@ static inline int calculate_order(int size) * Doh this slab cannot be placed using slub_max_order. */ order = slab_order(size, 1, MAX_ORDER, 1); - if (order <= MAX_ORDER) + if (order < MAX_ORDER) return order; return -ENOSYS; } @@ -2522,6 +2522,7 @@ __setup("slub_min_order=", setup_slub_min_order); static int __init setup_slub_max_order(char *str) { get_option(&str, &slub_max_order); + slub_max_order = min(slub_max_order, MAX_ORDER - 1); return 1; } From 45cb50e646d1560eff01c5ad0f0df3c7fd6148dd Mon Sep 17 00:00:00 2001 From: Zhang Rui Date: Fri, 24 Apr 2009 12:13:18 -0400 Subject: [PATCH 006/267] ACPI video: dmi check for broken _BQC on Acer Aspire 5720 On Acer Aspire 5720, _BQC always returns a value 9 smaller than the actual brightness level. Add dmi quirk for this laptop. http://bugzilla.kernel.org/show_bug.cgi?id=13121 Tested-by: Maxim Levitsky Signed-off-by: Zhang Rui Signed-off-by: Len Brown --- drivers/acpi/video.c | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c index cd4fb7543a90..7e8ee49ebd40 100644 --- a/drivers/acpi/video.c +++ b/drivers/acpi/video.c @@ -538,6 +538,33 @@ acpi_video_device_lcd_set_level(struct acpi_video_device *device, int level) return -EINVAL; } +/* + * For some buggy _BQC methods, we need to add a constant value to + * the _BQC return value to get the actual current brightness level + */ + +static int bqc_offset_aml_bug_workaround; +static int __init video_set_bqc_offset(const struct dmi_system_id *d) +{ + bqc_offset_aml_bug_workaround = 9; + return 0; +} + +static struct dmi_system_id video_dmi_table[] __initdata = { + /* + * Broken _BQC workaround http://bugzilla.kernel.org/show_bug.cgi?id=13121 + */ + { + .callback = video_set_bqc_offset, + .ident = "Acer Aspire 5720", + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "Acer"), + DMI_MATCH(DMI_PRODUCT_NAME, "Aspire 5720"), + }, + }, + {} +}; + static int acpi_video_device_lcd_get_level_current(struct acpi_video_device *device, unsigned long long *level) @@ -557,6 +584,7 @@ acpi_video_device_lcd_get_level_current(struct acpi_video_device *device, *level = device->brightness->levels[*level + 2]; } + *level += bqc_offset_aml_bug_workaround; device->brightness->curr = *level; return 0; } else { @@ -2287,6 +2315,8 @@ EXPORT_SYMBOL(acpi_video_register); static int __init acpi_video_init(void) { + dmi_check_system(video_dmi_table); + if (intel_opregion_present()) return 0; From 51e95bed6263f6a2e558ca04fc4ff877f3b0cca3 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Fri, 24 Apr 2009 16:26:25 -0700 Subject: [PATCH 007/267] slub: add Documentation/ABI/testing/sysfs-kernel-slab Adds documentation for the slub ABI. This is placed in the `testing' directory since the meanings of these files are still subject to change as slub is developed. Acked-by: Christoph Lameter Cc: Randy Dunlap Signed-off-by: David Rientjes Signed-off-by: Pekka Enberg --- Documentation/ABI/testing/sysfs-kernel-slab | 479 ++++++++++++++++++++ 1 file changed, 479 insertions(+) create mode 100644 Documentation/ABI/testing/sysfs-kernel-slab diff --git a/Documentation/ABI/testing/sysfs-kernel-slab b/Documentation/ABI/testing/sysfs-kernel-slab new file mode 100644 index 000000000000..6dcf75e594fb --- /dev/null +++ b/Documentation/ABI/testing/sysfs-kernel-slab @@ -0,0 +1,479 @@ +What: /sys/kernel/slab +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The /sys/kernel/slab directory contains a snapshot of the + internal state of the SLUB allocator for each cache. Certain + files may be modified to change the behavior of the cache (and + any cache it aliases, if any). +Users: kernel memory tuning tools + +What: /sys/kernel/slab/cache/aliases +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The aliases file is read-only and specifies how many caches + have merged into this cache. + +What: /sys/kernel/slab/cache/align +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The align file is read-only and specifies the cache's object + alignment in bytes. + +What: /sys/kernel/slab/cache/alloc_calls +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The alloc_calls file is read-only and lists the kernel code + locations from which allocations for this cache were performed. + The alloc_calls file only contains information if debugging is + enabled for that cache (see Documentation/vm/slub.txt). + +What: /sys/kernel/slab/cache/alloc_fastpath +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The alloc_fastpath file is read-only and specifies how many + objects have been allocated using the fast path. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/alloc_from_partial +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The alloc_from_partial file is read-only and specifies how + many times a cpu slab has been full and it has been refilled + by using a slab from the list of partially used slabs. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/alloc_refill +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The alloc_refill file is read-only and specifies how many + times the per-cpu freelist was empty but there were objects + available as the result of remote cpu frees. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/alloc_slab +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The alloc_slab file is read-only and specifies how many times + a new slab had to be allocated from the page allocator. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/alloc_slowpath +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The alloc_slowpath file is read-only and specifies how many + objects have been allocated using the slow path because of a + refill or allocation from a partial or new slab. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/cache_dma +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The cache_dma file is read-only and specifies whether objects + are from ZONE_DMA. + Available when CONFIG_ZONE_DMA is enabled. + +What: /sys/kernel/slab/cache/cpu_slabs +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The cpu_slabs file is read-only and displays how many cpu slabs + are active and their NUMA locality. + +What: /sys/kernel/slab/cache/cpuslab_flush +Date: April 2009 +KernelVersion: 2.6.31 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The file cpuslab_flush is read-only and specifies how many + times a cache's cpu slabs have been flushed as the result of + destroying or shrinking a cache, a cpu going offline, or as + the result of forcing an allocation from a certain node. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/ctor +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The ctor file is read-only and specifies the cache's object + constructor function, which is invoked for each object when a + new slab is allocated. + +What: /sys/kernel/slab/cache/deactivate_empty +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The file deactivate_empty is read-only and specifies how many + times an empty cpu slab was deactivated. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/deactivate_full +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The file deactivate_full is read-only and specifies how many + times a full cpu slab was deactivated. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/deactivate_remote_frees +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The file deactivate_remote_frees is read-only and specifies how + many times a cpu slab has been deactivated and contained free + objects that were freed remotely. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/deactivate_to_head +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The file deactivate_to_head is read-only and specifies how + many times a partial cpu slab was deactivated and added to the + head of its node's partial list. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/deactivate_to_tail +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The file deactivate_to_tail is read-only and specifies how + many times a partial cpu slab was deactivated and added to the + tail of its node's partial list. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/destroy_by_rcu +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The destroy_by_rcu file is read-only and specifies whether + slabs (not objects) are freed by rcu. + +What: /sys/kernel/slab/cache/free_add_partial +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The file free_add_partial is read-only and specifies how many + times an object has been freed in a full slab so that it had to + added to its node's partial list. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/free_calls +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The free_calls file is read-only and lists the locations of + object frees if slab debugging is enabled (see + Documentation/vm/slub.txt). + +What: /sys/kernel/slab/cache/free_fastpath +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The free_fastpath file is read-only and specifies how many + objects have been freed using the fast path because it was an + object from the cpu slab. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/free_frozen +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The free_frozen file is read-only and specifies how many + objects have been freed to a frozen slab (i.e. a remote cpu + slab). + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/free_remove_partial +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The file free_remove_partial is read-only and specifies how + many times an object has been freed to a now-empty slab so + that it had to be removed from its node's partial list. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/free_slab +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The free_slab file is read-only and specifies how many times an + empty slab has been freed back to the page allocator. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/free_slowpath +Date: February 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The free_slowpath file is read-only and specifies how many + objects have been freed using the slow path (i.e. to a full or + partial slab). + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/hwcache_align +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The hwcache_align file is read-only and specifies whether + objects are aligned on cachelines. + +What: /sys/kernel/slab/cache/min_partial +Date: February 2009 +KernelVersion: 2.6.30 +Contact: Pekka Enberg , + David Rientjes +Description: + The min_partial file specifies how many empty slabs shall + remain on a node's partial list to avoid the overhead of + allocating new slabs. Such slabs may be reclaimed by utilizing + the shrink file. + +What: /sys/kernel/slab/cache/object_size +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The object_size file is read-only and specifies the cache's + object size. + +What: /sys/kernel/slab/cache/objects +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The objects file is read-only and displays how many objects are + active and from which nodes they are from. + +What: /sys/kernel/slab/cache/objects_partial +Date: April 2008 +KernelVersion: 2.6.26 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The objects_partial file is read-only and displays how many + objects are on partial slabs and from which nodes they are + from. + +What: /sys/kernel/slab/cache/objs_per_slab +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The file objs_per_slab is read-only and specifies how many + objects may be allocated from a single slab of the order + specified in /sys/kernel/slab/cache/order. + +What: /sys/kernel/slab/cache/order +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The order file specifies the page order at which new slabs are + allocated. It is writable and can be changed to increase the + number of objects per slab. If a slab cannot be allocated + because of fragmentation, SLUB will retry with the minimum order + possible depending on its characteristics. + +What: /sys/kernel/slab/cache/order_fallback +Date: April 2008 +KernelVersion: 2.6.26 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The file order_fallback is read-only and specifies how many + times an allocation of a new slab has not been possible at the + cache's order and instead fallen back to its minimum possible + order. + Available when CONFIG_SLUB_STATS is enabled. + +What: /sys/kernel/slab/cache/partial +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The partial file is read-only and displays how long many + partial slabs there are and how long each node's list is. + +What: /sys/kernel/slab/cache/poison +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The poison file specifies whether objects should be poisoned + when a new slab is allocated. + +What: /sys/kernel/slab/cache/reclaim_account +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The reclaim_account file specifies whether the cache's objects + are reclaimable (and grouped by their mobility). + +What: /sys/kernel/slab/cache/red_zone +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The red_zone file specifies whether the cache's objects are red + zoned. + +What: /sys/kernel/slab/cache/remote_node_defrag_ratio +Date: January 2008 +KernelVersion: 2.6.25 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The file remote_node_defrag_ratio specifies the percentage of + times SLUB will attempt to refill the cpu slab with a partial + slab from a remote node as opposed to allocating a new slab on + the local node. This reduces the amount of wasted memory over + the entire system but can be expensive. + Available when CONFIG_NUMA is enabled. + +What: /sys/kernel/slab/cache/sanity_checks +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The sanity_checks file specifies whether expensive checks + should be performed on free and, at minimum, enables double free + checks. Caches that enable sanity_checks cannot be merged with + caches that do not. + +What: /sys/kernel/slab/cache/shrink +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The shrink file is written when memory should be reclaimed from + a cache. Empty partial slabs are freed and the partial list is + sorted so the slabs with the fewest available objects are used + first. + +What: /sys/kernel/slab/cache/slab_size +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The slab_size file is read-only and specifies the object size + with metadata (debugging information and alignment) in bytes. + +What: /sys/kernel/slab/cache/slabs +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The slabs file is read-only and displays how long many slabs + there are (both cpu and partial) and from which nodes they are + from. + +What: /sys/kernel/slab/cache/store_user +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The store_user file specifies whether the location of + allocation or free should be tracked for a cache. + +What: /sys/kernel/slab/cache/total_objects +Date: April 2008 +KernelVersion: 2.6.26 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The total_objects file is read-only and displays how many total + objects a cache has and from which nodes they are from. + +What: /sys/kernel/slab/cache/trace +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + The trace file specifies whether object allocations and frees + should be traced. + +What: /sys/kernel/slab/cache/validate +Date: May 2007 +KernelVersion: 2.6.22 +Contact: Pekka Enberg , + Christoph Lameter +Description: + Writing to the validate file causes SLUB to traverse all of its + cache's objects and check the validity of metadata. From 1eb5ac6466d4be7b15b38ce3ab709600f1bc891f Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Tue, 5 May 2009 19:13:44 +1000 Subject: [PATCH 008/267] mm: SLUB fix reclaim_state SLUB does not correctly account reclaim_state.reclaimed_slab, so it will break memory reclaim. Account it like SLAB does. Cc: stable@kernel.org Cc: linux-mm@kvack.org Cc: Matt Mackall Acked-by: Christoph Lameter Signed-off-by: Nick Piggin Signed-off-by: Pekka Enberg --- mm/slub.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/mm/slub.c b/mm/slub.c index 7ab54ecbd3f3..aa34913a41ea 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -9,6 +9,7 @@ */ #include +#include /* struct reclaim_state */ #include #include #include @@ -1170,6 +1171,8 @@ static void __free_slab(struct kmem_cache *s, struct page *page) __ClearPageSlab(page); reset_page_mapcount(page); + if (current->reclaim_state) + current->reclaim_state->reclaimed_slab += pages; __free_pages(page, order); } From 1f0532eb617d28f65c93593a1491f662f14f7eac Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Tue, 5 May 2009 19:13:45 +1000 Subject: [PATCH 009/267] mm: SLOB fix reclaim_state SLOB does not correctly account reclaim_state.reclaimed_slab, so it will break memory reclaim. Account it like SLAB does. Cc: stable@kernel.org Cc: linux-mm@kvack.org Acked-by: Matt Mackall Acked-by: Christoph Lameter Signed-off-by: Nick Piggin Signed-off-by: Pekka Enberg --- mm/slob.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/mm/slob.c b/mm/slob.c index a2d4ab32198d..f92e66d558bd 100644 --- a/mm/slob.c +++ b/mm/slob.c @@ -60,6 +60,7 @@ #include #include #include +#include /* struct reclaim_state */ #include #include #include @@ -255,6 +256,8 @@ static void *slob_new_pages(gfp_t gfp, int order, int node) static void slob_free_pages(void *b, int order) { + if (current->reclaim_state) + current->reclaim_state->reclaimed_slab += 1 << order; free_pages((unsigned long)b, order); } @@ -407,7 +410,7 @@ static void slob_free(void *block, int size) spin_unlock_irqrestore(&slob_lock, flags); clear_slob_page(sp); free_slob_page(sp); - free_page((unsigned long)b); + slob_free_pages(b, 0); return; } From 54f2c841fa0007e5fee3b7d01a911c774f0a6cda Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Thu, 7 May 2009 17:28:59 +0200 Subject: [PATCH 010/267] oprofile: fix cpu buffer size The unit of oprofile_cpu_buffer_size is in samples, but was allocated in bytes. This led to the allocation of too small cpu buffers. This patch recalculates the buffer size in bytes taking also the ring_buffer_event header size into account. Reported-by: Suravee Suthikulpanit Signed-off-by: Robert Richter --- drivers/oprofile/cpu_buffer.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/oprofile/cpu_buffer.c b/drivers/oprofile/cpu_buffer.c index f0e99d4c066b..242257b19441 100644 --- a/drivers/oprofile/cpu_buffer.c +++ b/drivers/oprofile/cpu_buffer.c @@ -78,16 +78,20 @@ void free_cpu_buffers(void) op_ring_buffer_write = NULL; } +#define RB_EVENT_HDR_SIZE 4 + int alloc_cpu_buffers(void) { int i; unsigned long buffer_size = oprofile_cpu_buffer_size; + unsigned long byte_size = buffer_size * (sizeof(struct op_sample) + + RB_EVENT_HDR_SIZE); - op_ring_buffer_read = ring_buffer_alloc(buffer_size, OP_BUFFER_FLAGS); + op_ring_buffer_read = ring_buffer_alloc(byte_size, OP_BUFFER_FLAGS); if (!op_ring_buffer_read) goto fail; - op_ring_buffer_write = ring_buffer_alloc(buffer_size, OP_BUFFER_FLAGS); + op_ring_buffer_write = ring_buffer_alloc(byte_size, OP_BUFFER_FLAGS); if (!op_ring_buffer_write) goto fail; From 51991763416251f259efbfe459c16f331f31d01f Mon Sep 17 00:00:00 2001 From: Len Brown Date: Thu, 7 May 2009 15:24:31 -0400 Subject: [PATCH 011/267] ACPICA: use acpi.* modparam namespace Add acpi/acpica/*.c to the acpi.* modparam namespace so that any modparams we stick into ACPICA do not expose ACPICA filenames to users. There are currently only two modparams in ACPICA, just recently added for http://bugzilla.kernel.org/show_bug.cgi?id=13041 With this change, they become acpi.gts=1 acpi.bfs=1 rather than hwsleep.gts=1 hwsleep.bfs=1 Signed-off-by: Len Brown --- drivers/acpi/acpica/Makefile | 27 +++++++++++++++------------ 1 file changed, 15 insertions(+), 12 deletions(-) diff --git a/drivers/acpi/acpica/Makefile b/drivers/acpi/acpica/Makefile index 17e50824a6f1..72ac28da14e3 100644 --- a/drivers/acpi/acpica/Makefile +++ b/drivers/acpi/acpica/Makefile @@ -5,40 +5,43 @@ ccflags-y := -Os ccflags-$(CONFIG_ACPI_DEBUG) += -DACPI_DEBUG_OUTPUT -obj-y := dsfield.o dsmthdat.o dsopcode.o dswexec.o dswscope.o \ +# use acpi.o to put all files here into acpi.o modparam namespace +obj-y += acpi.o + +acpi-y := dsfield.o dsmthdat.o dsopcode.o dswexec.o dswscope.o \ dsmethod.o dsobject.o dsutils.o dswload.o dswstate.o \ dsinit.o -obj-y += evevent.o evregion.o evsci.o evxfevnt.o \ +acpi-y += evevent.o evregion.o evsci.o evxfevnt.o \ evmisc.o evrgnini.o evxface.o evxfregn.o \ evgpe.o evgpeblk.o -obj-y += exconfig.o exfield.o exnames.o exoparg6.o exresolv.o exstorob.o\ +acpi-y += exconfig.o exfield.o exnames.o exoparg6.o exresolv.o exstorob.o\ exconvrt.o exfldio.o exoparg1.o exprep.o exresop.o exsystem.o\ excreate.o exmisc.o exoparg2.o exregion.o exstore.o exutils.o \ exdump.o exmutex.o exoparg3.o exresnte.o exstoren.o -obj-y += hwacpi.o hwgpe.o hwregs.o hwsleep.o hwxface.o hwvalid.o +acpi-y += hwacpi.o hwgpe.o hwregs.o hwsleep.o hwxface.o hwvalid.o -obj-$(ACPI_FUTURE_USAGE) += hwtimer.o +acpi-$(ACPI_FUTURE_USAGE) += hwtimer.o -obj-y += nsaccess.o nsload.o nssearch.o nsxfeval.o \ +acpi-y += nsaccess.o nsload.o nssearch.o nsxfeval.o \ nsalloc.o nseval.o nsnames.o nsutils.o nsxfname.o \ nsdump.o nsinit.o nsobject.o nswalk.o nsxfobj.o \ nsparse.o nspredef.o -obj-$(ACPI_FUTURE_USAGE) += nsdumpdv.o +acpi-$(ACPI_FUTURE_USAGE) += nsdumpdv.o -obj-y += psargs.o psparse.o psloop.o pstree.o pswalk.o \ +acpi-y += psargs.o psparse.o psloop.o pstree.o pswalk.o \ psopcode.o psscope.o psutils.o psxface.o -obj-y += rsaddr.o rscreate.o rsinfo.o rsio.o rslist.o rsmisc.o rsxface.o \ +acpi-y += rsaddr.o rscreate.o rsinfo.o rsio.o rslist.o rsmisc.o rsxface.o \ rscalc.o rsirq.o rsmemory.o rsutils.o -obj-$(ACPI_FUTURE_USAGE) += rsdump.o +acpi-$(ACPI_FUTURE_USAGE) += rsdump.o -obj-y += tbxface.o tbinstal.o tbutils.o tbfind.o tbfadt.o tbxfroot.o +acpi-y += tbxface.o tbinstal.o tbutils.o tbfind.o tbfadt.o tbxfroot.o -obj-y += utalloc.o utdebug.o uteval.o utinit.o utmisc.o utxface.o \ +acpi-y += utalloc.o utdebug.o uteval.o utinit.o utmisc.o utxface.o \ utcopy.o utdelete.o utglobal.o utmath.o utobject.o \ utstate.o utmutex.o utobject.o utresrc.o utlock.o From 5afc4abe7902b8453c248321daa9b13b12d9c838 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Thu, 7 May 2009 21:11:56 -0400 Subject: [PATCH 012/267] ACPI: video: DMI workaround another broken Acer BIOS enabling display brightness http://bugzilla.kernel.org/show_bug.cgi?id=13121 Signed-off-by: Len Brown --- drivers/acpi/video.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c index 7e8ee49ebd40..3a3d1565a7dd 100644 --- a/drivers/acpi/video.c +++ b/drivers/acpi/video.c @@ -562,6 +562,14 @@ static struct dmi_system_id video_dmi_table[] __initdata = { DMI_MATCH(DMI_PRODUCT_NAME, "Aspire 5720"), }, }, + { + .callback = video_set_bqc_offset, + .ident = "Acer Aspire 5710Z", + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "Acer"), + DMI_MATCH(DMI_PRODUCT_NAME, "Aspire 5710Z"), + }, + }, {} }; From ddc50b6ad634d9ce2526a777d4b7da80effdfb60 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Fri, 8 May 2009 00:07:30 -0400 Subject: [PATCH 013/267] ACPI: power: update error message "Transitioning device [%s] to D%d" is not correct. We print this line when we attempted to transition the device, and it failed. So instead, print "Device [%s] failed to transition to D%d\n" This can happen under two conditions: 1. acpi_power_transition() fails when trying to handle the _ON/_OFF for associated power resource. 2. acpi_evaluate_object() on the explicit _PS0/_PS3 for that actual device could fail. this change clarifies, but doesn't fix http://bugzilla.kernel.org/show_bug.cgi?id=13243 Signed-off-by: Len Brown --- drivers/acpi/bus.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c index e8f7b64e92da..ae862f1798dc 100644 --- a/drivers/acpi/bus.c +++ b/drivers/acpi/bus.c @@ -312,7 +312,7 @@ int acpi_bus_set_power(acpi_handle handle, int state) end: if (result) printk(KERN_WARNING PREFIX - "Transitioning device [%s] to D%d\n", + "Device [%s] failed to transition to D%d\n", device->pnp.bus_id, state); else { device->power.state = state; From 19bde778c1fd2574cc020a618d7d576f260271ca Mon Sep 17 00:00:00 2001 From: Len Brown Date: Fri, 8 May 2009 00:22:29 -0400 Subject: [PATCH 014/267] ACPI: suspend: don't let device _PS3 failure prevent suspend 6328a57401dc5f5cf9931738eb7268fcd8058c49 "Enable PNPACPI _PSx Support, v3" added a call to acpi_bus_set_power(handle, ACPI_STATE_D3) to pnpacpi_disable_resource() before the existing call to evaluate _DIS on the device. This caused suspend to fail on the system in http://bugzilla.kernel.org/show_bug.cgi?id=13243 because the sanity check to verify we entered _PS3 failed on the serial port. As a work-around, that sanity check can be disabled system-wide with "acpi.power_nocheck=1" Or perhaps we should just shrug off the _PS3 failure and carry on with _DIS like we used to -- which is what this patch does. Signed-off-by: Len Brown --- drivers/pnp/pnpacpi/core.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/drivers/pnp/pnpacpi/core.c b/drivers/pnp/pnpacpi/core.c index 9a3a682c6981..9496494f340e 100644 --- a/drivers/pnp/pnpacpi/core.c +++ b/drivers/pnp/pnpacpi/core.c @@ -110,11 +110,9 @@ static int pnpacpi_disable_resources(struct pnp_dev *dev) /* acpi_unregister_gsi(pnp_irq(dev, 0)); */ ret = 0; - if (acpi_bus_power_manageable(handle)) { - ret = acpi_bus_set_power(handle, ACPI_STATE_D3); - if (ret) - return ret; - } + if (acpi_bus_power_manageable(handle)) + acpi_bus_set_power(handle, ACPI_STATE_D3); + /* continue even if acpi_bus_set_power() fails */ if (ACPI_FAILURE(acpi_evaluate_object(handle, "_DIS", NULL, NULL))) ret = -ENODEV; return ret; From 92d23f703c608fcb2c8edd74a3fd0f4031e18606 Mon Sep 17 00:00:00 2001 From: Ron Date: Fri, 8 May 2009 22:54:49 +0930 Subject: [PATCH 015/267] sched: Fix fallback sched_clock()'s offset when using jiffies Account for the initial offset to the jiffy count. [ Impact: fix printk timestamps on architectures using fallback sched_clock() ] Signed-off-by: Ron Lee Cc: Andrew Morton Signed-off-by: Ingo Molnar --- kernel/sched_clock.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/sched_clock.c b/kernel/sched_clock.c index 819f17ac796e..e1d16c9a7680 100644 --- a/kernel/sched_clock.c +++ b/kernel/sched_clock.c @@ -38,7 +38,8 @@ */ unsigned long long __attribute__((weak)) sched_clock(void) { - return (unsigned long long)jiffies * (NSEC_PER_SEC / HZ); + return (unsigned long long)(jiffies - INITIAL_JIFFIES) + * (NSEC_PER_SEC / HZ); } static __read_mostly int sched_clock_running; From b74d446f1f337e3fe906169a3266cb65ffa4179e Mon Sep 17 00:00:00 2001 From: Sam Ravnborg Date: Sat, 9 May 2009 15:35:10 +0600 Subject: [PATCH 016/267] x86: Fix false positive section mismatch warnings in the apic code [ Impact: reduce kernel image size a bit, annotate away warnings ] Signed-off-by: Sam Ravnborg [ modified and tested it ] Signed-off-by: Rakib Mullick Cc: Marcin Slusarz LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic/es7000_32.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/x86/kernel/apic/es7000_32.c b/arch/x86/kernel/apic/es7000_32.c index 1c11b819f245..302947775575 100644 --- a/arch/x86/kernel/apic/es7000_32.c +++ b/arch/x86/kernel/apic/es7000_32.c @@ -254,7 +254,7 @@ static int parse_unisys_oem(char *oemptr) } #ifdef CONFIG_ACPI -static int find_unisys_acpi_oem_table(unsigned long *oem_addr) +static int __init find_unisys_acpi_oem_table(unsigned long *oem_addr) { struct acpi_table_header *header = NULL; struct es7000_oem_table *table; @@ -285,7 +285,7 @@ static int find_unisys_acpi_oem_table(unsigned long *oem_addr) return 0; } -static void unmap_unisys_acpi_oem_table(unsigned long oem_addr) +static void __init unmap_unisys_acpi_oem_table(unsigned long oem_addr) { if (!oem_addr) return; @@ -306,7 +306,7 @@ static int es7000_check_dsdt(void) static int es7000_acpi_ret; /* Hook from generic ACPI tables.c */ -static int es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +static int __init es7000_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { unsigned long oem_addr = 0; int check_dsdt; @@ -717,7 +717,7 @@ struct apic apic_es7000_cluster = { .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, }; -struct apic apic_es7000 = { +struct apic __refdata apic_es7000 = { .name = "es7000", .probe = probe_es7000, From 917a0153621572e88aeb2d5df025ad2e81027287 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 6 May 2009 21:36:16 -0700 Subject: [PATCH 017/267] x86: mtrr: Fix high_width computation when phys-addr is >= 44bit found one system where cpu address line is 44bits, mtrr printout is not right: [ 0.000000] MTRR variable ranges enabled: [ 0.000000] 0 base 0 00000000 mask FF0 00000000 write-back [ 0.000000] 1 base 10 00000000 mask FFF 80000000 write-back [ 0.000000] 2 base 0 80000000 mask FFF 80000000 uncachable [ 0.000000] 3 base 0 7F800000 mask FFF FF800000 uncachable Li Zefan and Frederic pointed out the high_width could be -4 some how. It turns out when phys_addr is 44bit, size_or_mask will be ffffffff,00000000 so ffs(size_or_mask) will be 0. Try to check low 32 bit, to get correct high_width. Signed-off-by: Yinghai Lu Also-analyzed-by: Frederic Weisbecker Also-analyzed-by: Li Zefan Cc: Jeremy Fitzhardinge Cc: Zhaolei Cc: Steven Rostedt Cc: Vegard Nossum Cc: Andrew Morton LKML-Reference: <4A026540.8060504@kernel.org> Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/mtrr/generic.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index 0b776c09aff3..d21d4fb161f7 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -275,7 +275,11 @@ static void __init print_mtrr_state(void) } printk(KERN_DEBUG "MTRR variable ranges %sabled:\n", mtrr_state.enabled & 2 ? "en" : "dis"); - high_width = ((size_or_mask ? ffs(size_or_mask) - 1 : 32) - (32 - PAGE_SHIFT) + 3) / 4; + if (size_or_mask & 0xffffffffUL) + high_width = ffs(size_or_mask & 0xffffffffUL) - 1; + else + high_width = ffs(size_or_mask>>32) + 32 - 1; + high_width = (high_width - (32 - PAGE_SHIFT) + 3) / 4; for (i = 0; i < num_var_ranges; ++i) { if (mtrr_state.var_ranges[i].mask_lo & (1 << 11)) printk(KERN_DEBUG " %u base %0*X%05X000 mask %0*X%05X000 %s\n", From 3c598766a2bae1b208470e7cc934ac462561e3cb Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Mon, 11 May 2009 16:49:28 +0100 Subject: [PATCH 018/267] x86: fix percpu_{to,from}_op() - the byte operand constraints were wrong for 32-bit - the to-op's input operands weren't properly parenthesized [ Impact: fix possible miscompilation or build failure ] Signed-off-by: Jan Beulich Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/percpu.h | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/x86/include/asm/percpu.h b/arch/x86/include/asm/percpu.h index aee103b26d01..02ecb30982a3 100644 --- a/arch/x86/include/asm/percpu.h +++ b/arch/x86/include/asm/percpu.h @@ -82,22 +82,22 @@ do { \ case 1: \ asm(op "b %1,"__percpu_arg(0) \ : "+m" (var) \ - : "ri" ((T__)val)); \ + : "qi" ((T__)(val))); \ break; \ case 2: \ asm(op "w %1,"__percpu_arg(0) \ : "+m" (var) \ - : "ri" ((T__)val)); \ + : "ri" ((T__)(val))); \ break; \ case 4: \ asm(op "l %1,"__percpu_arg(0) \ : "+m" (var) \ - : "ri" ((T__)val)); \ + : "ri" ((T__)(val))); \ break; \ case 8: \ asm(op "q %1,"__percpu_arg(0) \ : "+m" (var) \ - : "re" ((T__)val)); \ + : "re" ((T__)(val))); \ break; \ default: __bad_percpu_size(); \ } \ @@ -109,7 +109,7 @@ do { \ switch (sizeof(var)) { \ case 1: \ asm(op "b "__percpu_arg(1)",%0" \ - : "=r" (ret__) \ + : "=q" (ret__) \ : "m" (var)); \ break; \ case 2: \ From 72af2b363107df9cd67985ecb8495bbdff3c9857 Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Mon, 11 May 2009 09:58:19 -0700 Subject: [PATCH 019/267] ARM: OMAP: Fix printing of reserved memory for frambuffer Print reserved memory only if it was actually reserved. Signed-off-by: Tomi Valkeinen Signed-off-by: Tony Lindgren --- arch/arm/plat-omap/fb.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/arm/plat-omap/fb.c b/arch/arm/plat-omap/fb.c index ce6b4baeedec..3746222bed10 100644 --- a/arch/arm/plat-omap/fb.c +++ b/arch/arm/plat-omap/fb.c @@ -206,9 +206,10 @@ void __init omapfb_reserve_sdram(void) config_invalid = 1; return; } - if (rg.paddr) + if (rg.paddr) { reserve_bootmem(rg.paddr, rg.size, BOOTMEM_DEFAULT); - reserved += rg.size; + reserved += rg.size; + } omapfb_config.mem_desc.region[i] = rg; configured_regions++; } From 7b6c6c77732ca1d2498eda7eabb64f9648896e96 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Mon, 11 May 2009 17:03:00 -0400 Subject: [PATCH 020/267] x86, 32-bit: fix kernel_trap_sp() Use ®s->sp instead of regs for getting the top of stack in kernel mode. (on x86-64, regs->sp always points the top of stack) [ Impact: Oprofile decodes only stack for backtracing on i386 ] Signed-off-by: Masami Hiramatsu [ v2: rename the API to kernel_stack_pointer(), move variable inside ] Acked-by: Linus Torvalds Cc: systemtap@sources.redhat.com Cc: Harvey Harrison Cc: Jan Blunck Cc: Christoph Hellwig LKML-Reference: <20090511210300.17332.67549.stgit@localhost.localdomain> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/ptrace.h | 7 ++++--- arch/x86/oprofile/backtrace.c | 2 +- 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/arch/x86/include/asm/ptrace.h b/arch/x86/include/asm/ptrace.h index e304b66abeea..624f133943ed 100644 --- a/arch/x86/include/asm/ptrace.h +++ b/arch/x86/include/asm/ptrace.h @@ -187,14 +187,15 @@ static inline int v8086_mode(struct pt_regs *regs) /* * X86_32 CPUs don't save ss and esp if the CPU is already in kernel mode - * when it traps. So regs will be the current sp. + * when it traps. The previous stack will be directly underneath the saved + * registers, and 'sp/ss' won't even have been saved. Thus the '®s->sp'. * * This is valid only for kernel mode traps. */ -static inline unsigned long kernel_trap_sp(struct pt_regs *regs) +static inline unsigned long kernel_stack_pointer(struct pt_regs *regs) { #ifdef CONFIG_X86_32 - return (unsigned long)regs; + return (unsigned long)(®s->sp); #else return regs->sp; #endif diff --git a/arch/x86/oprofile/backtrace.c b/arch/x86/oprofile/backtrace.c index 04df67f8a7ba..044897be021f 100644 --- a/arch/x86/oprofile/backtrace.c +++ b/arch/x86/oprofile/backtrace.c @@ -76,9 +76,9 @@ void x86_backtrace(struct pt_regs * const regs, unsigned int depth) { struct frame_head *head = (struct frame_head *)frame_pointer(regs); - unsigned long stack = kernel_trap_sp(regs); if (!user_mode_vm(regs)) { + unsigned long stack = kernel_stack_pointer(regs); if (depth) dump_trace(NULL, regs, (unsigned long *)stack, 0, &backtrace_ops, &depth); From d80c19df5fcceb8c741e96f09f275c2da719efef Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 12 May 2009 16:29:13 +0200 Subject: [PATCH 021/267] lockdep: increase MAX_LOCKDEP_ENTRIES and MAX_LOCKDEP_CHAINS Now that lockdep coverage has increased it has become easier to run out of entries: [ 21.401387] BUG: MAX_LOCKDEP_ENTRIES too low! [ 21.402007] turning off the locking correctness validator. [ 21.402007] Pid: 1555, comm: S99local Not tainted 2.6.30-rc5-tip #2 [ 21.402007] Call Trace: [ 21.402007] [] add_lock_to_list+0x53/0xba [ 21.402007] [] ? lookup_mnt+0x19/0x53 [ 21.402007] [] check_prev_add+0x14b/0x1c7 [ 21.402007] [] validate_chain+0x474/0x52a [ 21.402007] [] __lock_acquire+0x342/0x3c7 [ 21.402007] [] lock_acquire+0xc1/0xe5 [ 21.402007] [] ? lookup_mnt+0x19/0x53 [ 21.402007] [] _spin_lock+0x31/0x66 Double the size - as we've done in the past. [ Impact: allow lockdep to cover more locks ] Acked-by: Peter Zijlstra LKML-Reference: Signed-off-by: Ingo Molnar --- kernel/lockdep_internals.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/lockdep_internals.h b/kernel/lockdep_internals.h index a2cc7e9a6e84..699a2ac3a0d7 100644 --- a/kernel/lockdep_internals.h +++ b/kernel/lockdep_internals.h @@ -54,9 +54,9 @@ enum { * table (if it's not there yet), and we check it for lock order * conflicts and deadlocks. */ -#define MAX_LOCKDEP_ENTRIES 8192UL +#define MAX_LOCKDEP_ENTRIES 16384UL -#define MAX_LOCKDEP_CHAINS_BITS 14 +#define MAX_LOCKDEP_CHAINS_BITS 15 #define MAX_LOCKDEP_CHAINS (1UL << MAX_LOCKDEP_CHAINS_BITS) #define MAX_LOCKDEP_CHAIN_HLOCKS (MAX_LOCKDEP_CHAINS*5) From 6f733a349c2b70a83b5451fbe61ef2cf67d41cdf Mon Sep 17 00:00:00 2001 From: Paul Walmsley Date: Mon, 11 May 2009 09:58:19 -0700 Subject: [PATCH 022/267] ARM: OMAP: GPIO de-bounce clocks don't affect module idle state GPIO de-bounce clocks don't have any impact on the module idle state, so the clock code should not wait for the module to enable after the de-bounce clocks are enabled. Problem found by Kevin Hilman . Signed-off-by: Paul Walmsley Signed-off-by: Kevin Hilman Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/clock34xx.h | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/arm/mach-omap2/clock34xx.h b/arch/arm/mach-omap2/clock34xx.h index 6763b8f73028..017a30e9aa1d 100644 --- a/arch/arm/mach-omap2/clock34xx.h +++ b/arch/arm/mach-omap2/clock34xx.h @@ -2182,7 +2182,7 @@ static struct clk wkup_32k_fck = { static struct clk gpio1_dbck = { .name = "gpio1_dbck", - .ops = &clkops_omap2_dflt_wait, + .ops = &clkops_omap2_dflt, .parent = &wkup_32k_fck, .enable_reg = OMAP_CM_REGADDR(WKUP_MOD, CM_FCLKEN), .enable_bit = OMAP3430_EN_GPIO1_SHIFT, @@ -2427,7 +2427,7 @@ static struct clk per_32k_alwon_fck = { static struct clk gpio6_dbck = { .name = "gpio6_dbck", - .ops = &clkops_omap2_dflt_wait, + .ops = &clkops_omap2_dflt, .parent = &per_32k_alwon_fck, .enable_reg = OMAP_CM_REGADDR(OMAP3430_PER_MOD, CM_FCLKEN), .enable_bit = OMAP3430_EN_GPIO6_SHIFT, @@ -2437,7 +2437,7 @@ static struct clk gpio6_dbck = { static struct clk gpio5_dbck = { .name = "gpio5_dbck", - .ops = &clkops_omap2_dflt_wait, + .ops = &clkops_omap2_dflt, .parent = &per_32k_alwon_fck, .enable_reg = OMAP_CM_REGADDR(OMAP3430_PER_MOD, CM_FCLKEN), .enable_bit = OMAP3430_EN_GPIO5_SHIFT, @@ -2447,7 +2447,7 @@ static struct clk gpio5_dbck = { static struct clk gpio4_dbck = { .name = "gpio4_dbck", - .ops = &clkops_omap2_dflt_wait, + .ops = &clkops_omap2_dflt, .parent = &per_32k_alwon_fck, .enable_reg = OMAP_CM_REGADDR(OMAP3430_PER_MOD, CM_FCLKEN), .enable_bit = OMAP3430_EN_GPIO4_SHIFT, @@ -2457,7 +2457,7 @@ static struct clk gpio4_dbck = { static struct clk gpio3_dbck = { .name = "gpio3_dbck", - .ops = &clkops_omap2_dflt_wait, + .ops = &clkops_omap2_dflt, .parent = &per_32k_alwon_fck, .enable_reg = OMAP_CM_REGADDR(OMAP3430_PER_MOD, CM_FCLKEN), .enable_bit = OMAP3430_EN_GPIO3_SHIFT, @@ -2467,7 +2467,7 @@ static struct clk gpio3_dbck = { static struct clk gpio2_dbck = { .name = "gpio2_dbck", - .ops = &clkops_omap2_dflt_wait, + .ops = &clkops_omap2_dflt, .parent = &per_32k_alwon_fck, .enable_reg = OMAP_CM_REGADDR(OMAP3430_PER_MOD, CM_FCLKEN), .enable_bit = OMAP3430_EN_GPIO2_SHIFT, From eaf9393bb79a9c7c8e708669f9581ca466122430 Mon Sep 17 00:00:00 2001 From: Jarkko Nikula Date: Tue, 12 May 2009 11:20:02 -0700 Subject: [PATCH 023/267] ARM: OMAP2: Fix tusb6010 init error and compilation warning Fix "tusb6010 init error 5, -19" and compilation warning from function tusb6010_platform_retime "warning: 'sysclk_ps' is used uninitialized in this function". I suppose commit c094ba34b8f780885d029ce3c2715a194b780e5d was meant to test for zero fclk_ps instead of sysclk_ps. Signed-off-by: Jarkko Nikula Cc: Roel Kluin Tested-by: Kalle Valo Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/usb-tusb6010.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-omap2/usb-tusb6010.c b/arch/arm/mach-omap2/usb-tusb6010.c index 8df55f40f4c0..8622c24cd270 100644 --- a/arch/arm/mach-omap2/usb-tusb6010.c +++ b/arch/arm/mach-omap2/usb-tusb6010.c @@ -187,7 +187,7 @@ int tusb6010_platform_retime(unsigned is_refclk) unsigned sysclk_ps; int status; - if (!refclk_psec || sysclk_ps == 0) + if (!refclk_psec || fclk_ps == 0) return -ENODEV; sysclk_ps = is_refclk ? refclk_psec : TUSB6010_OSCCLK_60; From c8a799b03a981b4b27d8fd7320c96cfbe781b2fb Mon Sep 17 00:00:00 2001 From: Jarkko Nikula Date: Tue, 12 May 2009 11:20:02 -0700 Subject: [PATCH 024/267] ARM: OMAP2: Fix SPI driver failure on 2420 when running multi-omap config SPI driver will do unhandled fault on OMAP2420 if trying to probe non-existing SPI busses. Register those additional busses runtime only for cpus having them. Signed-off-by: Jarkko Nikula Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/devices.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-omap2/devices.c b/arch/arm/mach-omap2/devices.c index 496983ade97e..894cc355818a 100644 --- a/arch/arm/mach-omap2/devices.c +++ b/arch/arm/mach-omap2/devices.c @@ -354,10 +354,12 @@ static void omap_init_mcspi(void) platform_device_register(&omap2_mcspi1); platform_device_register(&omap2_mcspi2); #if defined(CONFIG_ARCH_OMAP2430) || defined(CONFIG_ARCH_OMAP3) - platform_device_register(&omap2_mcspi3); + if (cpu_is_omap2430() || cpu_is_omap343x()) + platform_device_register(&omap2_mcspi3); #endif #ifdef CONFIG_ARCH_OMAP3 - platform_device_register(&omap2_mcspi4); + if (cpu_is_omap343x()) + platform_device_register(&omap2_mcspi4); #endif } From 4ea60b0c7a8487af5de736d394b147baf7691f3c Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Tue, 12 May 2009 11:20:03 -0700 Subject: [PATCH 025/267] ARM: OMAP2/3: Add name for musb clocks With the clkdev, musb_core.c needs to register clock with name "ick". Once all the platforms using the musb driver have been converted to use clockdev, the clock name does not need to be passed from the low-level init code. Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/clock24xx.c | 2 +- arch/arm/mach-omap2/clock34xx.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-omap2/clock24xx.c b/arch/arm/mach-omap2/clock24xx.c index efc59c49341b..ff9b4c0b00ae 100644 --- a/arch/arm/mach-omap2/clock24xx.c +++ b/arch/arm/mach-omap2/clock24xx.c @@ -206,7 +206,7 @@ static struct omap_clk omap24xx_clks[] = { CLK(NULL, "aes_ick", &aes_ick, CK_243X | CK_242X), CLK(NULL, "pka_ick", &pka_ick, CK_243X | CK_242X), CLK(NULL, "usb_fck", &usb_fck, CK_243X | CK_242X), - CLK(NULL, "usbhs_ick", &usbhs_ick, CK_243X), + CLK("musb_hdrc", "ick", &usbhs_ick, CK_243X), CLK("mmci-omap-hs.0", "ick", &mmchs1_ick, CK_243X), CLK("mmci-omap-hs.0", "fck", &mmchs1_fck, CK_243X), CLK("mmci-omap-hs.1", "ick", &mmchs2_ick, CK_243X), diff --git a/arch/arm/mach-omap2/clock34xx.c b/arch/arm/mach-omap2/clock34xx.c index 0a14dca31e30..a0575393d0ea 100644 --- a/arch/arm/mach-omap2/clock34xx.c +++ b/arch/arm/mach-omap2/clock34xx.c @@ -157,7 +157,7 @@ static struct omap_clk omap34xx_clks[] = { CLK(NULL, "ssi_ssr_fck", &ssi_ssr_fck, CK_343X), CLK(NULL, "ssi_sst_fck", &ssi_sst_fck, CK_343X), CLK(NULL, "core_l3_ick", &core_l3_ick, CK_343X), - CLK(NULL, "hsotgusb_ick", &hsotgusb_ick, CK_343X), + CLK("musb_hdrc", "ick", &hsotgusb_ick, CK_343X), CLK(NULL, "sdrc_ick", &sdrc_ick, CK_343X), CLK(NULL, "gpmc_fck", &gpmc_fck, CK_343X), CLK(NULL, "security_l3_ick", &security_l3_ick, CK_343X), From 44408ad7368906c84000e87a99c14a16dbb867fd Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 12 May 2009 13:31:40 -0700 Subject: [PATCH 026/267] xen: use header for EXPORT_SYMBOL_GPL mmu.c needs to #include module.h to prevent these warnings: arch/x86/xen/mmu.c:239: warning: data definition has no type or storage class arch/x86/xen/mmu.c:239: warning: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL' arch/x86/xen/mmu.c:239: warning: parameter names (without types) in function declaration [ Impact: cleanup ] Signed-off-by: Randy Dunlap Acked-by: Jeremy Fitzhardinge Signed-off-by: Andrew Morton LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/xen/mmu.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index e25a78e1113a..fba55b1a4021 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -42,6 +42,7 @@ #include #include #include +#include #include #include From aa512a27e9e8ed32f31b15eec67ab1ceca33839b Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 13 May 2009 13:52:19 -0400 Subject: [PATCH 027/267] x86/function-graph: fix constraint for recording old return value After upgrading from gcc 4.2.2 to 4.4.0, the function graph tracer broke. Investigating, I found that in the asm that replaces the return value, gcc was using the same register for the old value as it was for the new value. mov (addr), old mov new, (addr) But if old and new are the same register, we clobber new with old! I first thought this was a bug in gcc 4.4.0 and reported it: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40132 Andrew Pinski responded (quickly), saying that it was correct gcc behavior and the code needed to denote old as an "early clobber". Instead of "=r"(old), we need "=&r"(old). [Impact: keep function graph tracer from breaking with gcc 4.4.0 ] Signed-off-by: Steven Rostedt --- arch/x86/kernel/ftrace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index 18dfa30795c9..b79c5533c421 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -442,7 +442,7 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr) _ASM_EXTABLE(1b, 4b) _ASM_EXTABLE(2b, 4b) - : [old] "=r" (old), [faulted] "=r" (faulted) + : [old] "=&r" (old), [faulted] "=r" (faulted) : [parent] "r" (parent), [return_hooker] "r" (return_hooker) : "memory" ); From 5df6d737dd4b0fe9eccf943abb3677cfea05a6c4 Mon Sep 17 00:00:00 2001 From: Abhijeet Joglekar Date: Fri, 17 Apr 2009 18:33:26 -0700 Subject: [PATCH 028/267] [SCSI] fnic: Add new Cisco PCI-Express FCoE HBA fnic is a driver for the Cisco PCI-Express FCoE HBA Signed-off-by: Abhijeet Joglekar Signed-off-by: Joe Eykholt Signed-off-by: Mike Christie Signed-off-by: James Bottomley --- MAINTAINERS | 8 + drivers/scsi/Kconfig | 11 + drivers/scsi/Makefile | 1 + drivers/scsi/fnic/Makefile | 15 + drivers/scsi/fnic/cq_desc.h | 78 ++ drivers/scsi/fnic/cq_enet_desc.h | 167 +++ drivers/scsi/fnic/cq_exch_desc.h | 182 +++ drivers/scsi/fnic/fcpio.h | 780 ++++++++++++ drivers/scsi/fnic/fnic.h | 265 +++++ drivers/scsi/fnic/fnic_attrs.c | 56 + drivers/scsi/fnic/fnic_fcs.c | 742 ++++++++++++ drivers/scsi/fnic/fnic_io.h | 67 ++ drivers/scsi/fnic/fnic_isr.c | 332 ++++++ drivers/scsi/fnic/fnic_main.c | 942 +++++++++++++++ drivers/scsi/fnic/fnic_res.c | 444 +++++++ drivers/scsi/fnic/fnic_res.h | 197 +++ drivers/scsi/fnic/fnic_scsi.c | 1850 +++++++++++++++++++++++++++++ drivers/scsi/fnic/rq_enet_desc.h | 58 + drivers/scsi/fnic/vnic_cq.c | 85 ++ drivers/scsi/fnic/vnic_cq.h | 121 ++ drivers/scsi/fnic/vnic_cq_copy.h | 62 + drivers/scsi/fnic/vnic_dev.c | 690 +++++++++++ drivers/scsi/fnic/vnic_dev.h | 161 +++ drivers/scsi/fnic/vnic_devcmd.h | 281 +++++ drivers/scsi/fnic/vnic_intr.c | 60 + drivers/scsi/fnic/vnic_intr.h | 118 ++ drivers/scsi/fnic/vnic_nic.h | 69 ++ drivers/scsi/fnic/vnic_resource.h | 61 + drivers/scsi/fnic/vnic_rq.c | 196 +++ drivers/scsi/fnic/vnic_rq.h | 235 ++++ drivers/scsi/fnic/vnic_scsi.h | 99 ++ drivers/scsi/fnic/vnic_stats.h | 68 ++ drivers/scsi/fnic/vnic_wq.c | 182 +++ drivers/scsi/fnic/vnic_wq.h | 175 +++ drivers/scsi/fnic/vnic_wq_copy.c | 117 ++ drivers/scsi/fnic/vnic_wq_copy.h | 128 ++ drivers/scsi/fnic/wq_enet_desc.h | 96 ++ 37 files changed, 9199 insertions(+) create mode 100644 drivers/scsi/fnic/Makefile create mode 100644 drivers/scsi/fnic/cq_desc.h create mode 100644 drivers/scsi/fnic/cq_enet_desc.h create mode 100644 drivers/scsi/fnic/cq_exch_desc.h create mode 100644 drivers/scsi/fnic/fcpio.h create mode 100644 drivers/scsi/fnic/fnic.h create mode 100644 drivers/scsi/fnic/fnic_attrs.c create mode 100644 drivers/scsi/fnic/fnic_fcs.c create mode 100644 drivers/scsi/fnic/fnic_io.h create mode 100644 drivers/scsi/fnic/fnic_isr.c create mode 100644 drivers/scsi/fnic/fnic_main.c create mode 100644 drivers/scsi/fnic/fnic_res.c create mode 100644 drivers/scsi/fnic/fnic_res.h create mode 100644 drivers/scsi/fnic/fnic_scsi.c create mode 100644 drivers/scsi/fnic/rq_enet_desc.h create mode 100644 drivers/scsi/fnic/vnic_cq.c create mode 100644 drivers/scsi/fnic/vnic_cq.h create mode 100644 drivers/scsi/fnic/vnic_cq_copy.h create mode 100644 drivers/scsi/fnic/vnic_dev.c create mode 100644 drivers/scsi/fnic/vnic_dev.h create mode 100644 drivers/scsi/fnic/vnic_devcmd.h create mode 100644 drivers/scsi/fnic/vnic_intr.c create mode 100644 drivers/scsi/fnic/vnic_intr.h create mode 100644 drivers/scsi/fnic/vnic_nic.h create mode 100644 drivers/scsi/fnic/vnic_resource.h create mode 100644 drivers/scsi/fnic/vnic_rq.c create mode 100644 drivers/scsi/fnic/vnic_rq.h create mode 100644 drivers/scsi/fnic/vnic_scsi.h create mode 100644 drivers/scsi/fnic/vnic_stats.h create mode 100644 drivers/scsi/fnic/vnic_wq.c create mode 100644 drivers/scsi/fnic/vnic_wq.h create mode 100644 drivers/scsi/fnic/vnic_wq_copy.c create mode 100644 drivers/scsi/fnic/vnic_wq_copy.h create mode 100644 drivers/scsi/fnic/wq_enet_desc.h diff --git a/MAINTAINERS b/MAINTAINERS index 2b349ba4add4..c7bed166ad6f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1431,6 +1431,14 @@ P: Russell King M: linux@arm.linux.org.uk F: include/linux/clk.h +CISCO FCOE HBA DRIVER +P: Abhijeet Joglekar +M: abjoglek@cisco.com +P: Joe Eykholt +M: jeykholt@cisco.com +L: linux-scsi@vger.kernel.org +S: Supported + CODA FILE SYSTEM P: Jan Harkes M: jaharkes@cs.cmu.edu diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig index 8ed2990c826e..fb2740789b68 100644 --- a/drivers/scsi/Kconfig +++ b/drivers/scsi/Kconfig @@ -628,6 +628,17 @@ config FCOE ---help--- Fibre Channel over Ethernet module +config FCOE_FNIC + tristate "Cisco FNIC Driver" + depends on PCI && X86 + select LIBFC + help + This is support for the Cisco PCI-Express FCoE HBA. + + To compile this driver as a module, choose M here and read + . + The module will be called fnic. + config SCSI_DMX3191D tristate "DMX3191D SCSI support" depends on PCI && SCSI diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile index e7c861ac417d..a5049cfb40ed 100644 --- a/drivers/scsi/Makefile +++ b/drivers/scsi/Makefile @@ -39,6 +39,7 @@ obj-$(CONFIG_SCSI_DH) += device_handler/ obj-$(CONFIG_LIBFC) += libfc/ obj-$(CONFIG_LIBFCOE) += fcoe/ obj-$(CONFIG_FCOE) += fcoe/ +obj-$(CONFIG_FCOE_FNIC) += fnic/ obj-$(CONFIG_ISCSI_TCP) += libiscsi.o libiscsi_tcp.o iscsi_tcp.o obj-$(CONFIG_INFINIBAND_ISER) += libiscsi.o obj-$(CONFIG_SCSI_A4000T) += 53c700.o a4000t.o diff --git a/drivers/scsi/fnic/Makefile b/drivers/scsi/fnic/Makefile new file mode 100644 index 000000000000..37c3440bc17c --- /dev/null +++ b/drivers/scsi/fnic/Makefile @@ -0,0 +1,15 @@ +obj-$(CONFIG_FCOE_FNIC) += fnic.o + +fnic-y := \ + fnic_attrs.o \ + fnic_isr.o \ + fnic_main.o \ + fnic_res.o \ + fnic_fcs.o \ + fnic_scsi.o \ + vnic_cq.o \ + vnic_dev.o \ + vnic_intr.o \ + vnic_rq.o \ + vnic_wq_copy.o \ + vnic_wq.o diff --git a/drivers/scsi/fnic/cq_desc.h b/drivers/scsi/fnic/cq_desc.h new file mode 100644 index 000000000000..d1225cf6320e --- /dev/null +++ b/drivers/scsi/fnic/cq_desc.h @@ -0,0 +1,78 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _CQ_DESC_H_ +#define _CQ_DESC_H_ + +/* + * Completion queue descriptor types + */ +enum cq_desc_types { + CQ_DESC_TYPE_WQ_ENET = 0, + CQ_DESC_TYPE_DESC_COPY = 1, + CQ_DESC_TYPE_WQ_EXCH = 2, + CQ_DESC_TYPE_RQ_ENET = 3, + CQ_DESC_TYPE_RQ_FCP = 4, +}; + +/* Completion queue descriptor: 16B + * + * All completion queues have this basic layout. The + * type_specfic area is unique for each completion + * queue type. + */ +struct cq_desc { + __le16 completed_index; + __le16 q_number; + u8 type_specfic[11]; + u8 type_color; +}; + +#define CQ_DESC_TYPE_BITS 4 +#define CQ_DESC_TYPE_MASK ((1 << CQ_DESC_TYPE_BITS) - 1) +#define CQ_DESC_COLOR_MASK 1 +#define CQ_DESC_COLOR_SHIFT 7 +#define CQ_DESC_Q_NUM_BITS 10 +#define CQ_DESC_Q_NUM_MASK ((1 << CQ_DESC_Q_NUM_BITS) - 1) +#define CQ_DESC_COMP_NDX_BITS 12 +#define CQ_DESC_COMP_NDX_MASK ((1 << CQ_DESC_COMP_NDX_BITS) - 1) + +static inline void cq_desc_dec(const struct cq_desc *desc_arg, + u8 *type, u8 *color, u16 *q_number, u16 *completed_index) +{ + const struct cq_desc *desc = desc_arg; + const u8 type_color = desc->type_color; + + *color = (type_color >> CQ_DESC_COLOR_SHIFT) & CQ_DESC_COLOR_MASK; + + /* + * Make sure color bit is read from desc *before* other fields + * are read from desc. Hardware guarantees color bit is last + * bit (byte) written. Adding the rmb() prevents the compiler + * and/or CPU from reordering the reads which would potentially + * result in reading stale values. + */ + + rmb(); + + *type = type_color & CQ_DESC_TYPE_MASK; + *q_number = le16_to_cpu(desc->q_number) & CQ_DESC_Q_NUM_MASK; + *completed_index = le16_to_cpu(desc->completed_index) & + CQ_DESC_COMP_NDX_MASK; +} + +#endif /* _CQ_DESC_H_ */ diff --git a/drivers/scsi/fnic/cq_enet_desc.h b/drivers/scsi/fnic/cq_enet_desc.h new file mode 100644 index 000000000000..a9fa26f82ddd --- /dev/null +++ b/drivers/scsi/fnic/cq_enet_desc.h @@ -0,0 +1,167 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _CQ_ENET_DESC_H_ +#define _CQ_ENET_DESC_H_ + +#include "cq_desc.h" + +/* Ethernet completion queue descriptor: 16B */ +struct cq_enet_wq_desc { + __le16 completed_index; + __le16 q_number; + u8 reserved[11]; + u8 type_color; +}; + +static inline void cq_enet_wq_desc_dec(struct cq_enet_wq_desc *desc, + u8 *type, u8 *color, u16 *q_number, u16 *completed_index) +{ + cq_desc_dec((struct cq_desc *)desc, type, + color, q_number, completed_index); +} + +/* Completion queue descriptor: Ethernet receive queue, 16B */ +struct cq_enet_rq_desc { + __le16 completed_index_flags; + __le16 q_number_rss_type_flags; + __le32 rss_hash; + __le16 bytes_written_flags; + __le16 vlan; + __le16 checksum_fcoe; + u8 flags; + u8 type_color; +}; + +#define CQ_ENET_RQ_DESC_FLAGS_INGRESS_PORT (0x1 << 12) +#define CQ_ENET_RQ_DESC_FLAGS_FCOE (0x1 << 13) +#define CQ_ENET_RQ_DESC_FLAGS_EOP (0x1 << 14) +#define CQ_ENET_RQ_DESC_FLAGS_SOP (0x1 << 15) + +#define CQ_ENET_RQ_DESC_RSS_TYPE_BITS 4 +#define CQ_ENET_RQ_DESC_RSS_TYPE_MASK \ + ((1 << CQ_ENET_RQ_DESC_RSS_TYPE_BITS) - 1) +#define CQ_ENET_RQ_DESC_RSS_TYPE_NONE 0 +#define CQ_ENET_RQ_DESC_RSS_TYPE_IPv4 1 +#define CQ_ENET_RQ_DESC_RSS_TYPE_TCP_IPv4 2 +#define CQ_ENET_RQ_DESC_RSS_TYPE_IPv6 3 +#define CQ_ENET_RQ_DESC_RSS_TYPE_TCP_IPv6 4 +#define CQ_ENET_RQ_DESC_RSS_TYPE_IPv6_EX 5 +#define CQ_ENET_RQ_DESC_RSS_TYPE_TCP_IPv6_EX 6 + +#define CQ_ENET_RQ_DESC_FLAGS_CSUM_NOT_CALC (0x1 << 14) + +#define CQ_ENET_RQ_DESC_BYTES_WRITTEN_BITS 14 +#define CQ_ENET_RQ_DESC_BYTES_WRITTEN_MASK \ + ((1 << CQ_ENET_RQ_DESC_BYTES_WRITTEN_BITS) - 1) +#define CQ_ENET_RQ_DESC_FLAGS_TRUNCATED (0x1 << 14) +#define CQ_ENET_RQ_DESC_FLAGS_VLAN_STRIPPED (0x1 << 15) + +#define CQ_ENET_RQ_DESC_FCOE_SOF_BITS 4 +#define CQ_ENET_RQ_DESC_FCOE_SOF_MASK \ + ((1 << CQ_ENET_RQ_DESC_FCOE_SOF_BITS) - 1) +#define CQ_ENET_RQ_DESC_FCOE_EOF_BITS 8 +#define CQ_ENET_RQ_DESC_FCOE_EOF_MASK \ + ((1 << CQ_ENET_RQ_DESC_FCOE_EOF_BITS) - 1) +#define CQ_ENET_RQ_DESC_FCOE_EOF_SHIFT 8 + +#define CQ_ENET_RQ_DESC_FLAGS_TCP_UDP_CSUM_OK (0x1 << 0) +#define CQ_ENET_RQ_DESC_FCOE_FC_CRC_OK (0x1 << 0) +#define CQ_ENET_RQ_DESC_FLAGS_UDP (0x1 << 1) +#define CQ_ENET_RQ_DESC_FCOE_ENC_ERROR (0x1 << 1) +#define CQ_ENET_RQ_DESC_FLAGS_TCP (0x1 << 2) +#define CQ_ENET_RQ_DESC_FLAGS_IPV4_CSUM_OK (0x1 << 3) +#define CQ_ENET_RQ_DESC_FLAGS_IPV6 (0x1 << 4) +#define CQ_ENET_RQ_DESC_FLAGS_IPV4 (0x1 << 5) +#define CQ_ENET_RQ_DESC_FLAGS_IPV4_FRAGMENT (0x1 << 6) +#define CQ_ENET_RQ_DESC_FLAGS_FCS_OK (0x1 << 7) + +static inline void cq_enet_rq_desc_dec(struct cq_enet_rq_desc *desc, + u8 *type, u8 *color, u16 *q_number, u16 *completed_index, + u8 *ingress_port, u8 *fcoe, u8 *eop, u8 *sop, u8 *rss_type, + u8 *csum_not_calc, u32 *rss_hash, u16 *bytes_written, u8 *packet_error, + u8 *vlan_stripped, u16 *vlan, u16 *checksum, u8 *fcoe_sof, + u8 *fcoe_fc_crc_ok, u8 *fcoe_enc_error, u8 *fcoe_eof, + u8 *tcp_udp_csum_ok, u8 *udp, u8 *tcp, u8 *ipv4_csum_ok, + u8 *ipv6, u8 *ipv4, u8 *ipv4_fragment, u8 *fcs_ok) +{ + u16 completed_index_flags = le16_to_cpu(desc->completed_index_flags); + u16 q_number_rss_type_flags = + le16_to_cpu(desc->q_number_rss_type_flags); + u16 bytes_written_flags = le16_to_cpu(desc->bytes_written_flags); + + cq_desc_dec((struct cq_desc *)desc, type, + color, q_number, completed_index); + + *ingress_port = (completed_index_flags & + CQ_ENET_RQ_DESC_FLAGS_INGRESS_PORT) ? 1 : 0; + *fcoe = (completed_index_flags & CQ_ENET_RQ_DESC_FLAGS_FCOE) ? + 1 : 0; + *eop = (completed_index_flags & CQ_ENET_RQ_DESC_FLAGS_EOP) ? + 1 : 0; + *sop = (completed_index_flags & CQ_ENET_RQ_DESC_FLAGS_SOP) ? + 1 : 0; + + *rss_type = (u8)((q_number_rss_type_flags >> CQ_DESC_Q_NUM_BITS) & + CQ_ENET_RQ_DESC_RSS_TYPE_MASK); + *csum_not_calc = (q_number_rss_type_flags & + CQ_ENET_RQ_DESC_FLAGS_CSUM_NOT_CALC) ? 1 : 0; + + *rss_hash = le32_to_cpu(desc->rss_hash); + + *bytes_written = bytes_written_flags & + CQ_ENET_RQ_DESC_BYTES_WRITTEN_MASK; + *packet_error = (bytes_written_flags & + CQ_ENET_RQ_DESC_FLAGS_TRUNCATED) ? 1 : 0; + *vlan_stripped = (bytes_written_flags & + CQ_ENET_RQ_DESC_FLAGS_VLAN_STRIPPED) ? 1 : 0; + + *vlan = le16_to_cpu(desc->vlan); + + if (*fcoe) { + *fcoe_sof = (u8)(le16_to_cpu(desc->checksum_fcoe) & + CQ_ENET_RQ_DESC_FCOE_SOF_MASK); + *fcoe_fc_crc_ok = (desc->flags & + CQ_ENET_RQ_DESC_FCOE_FC_CRC_OK) ? 1 : 0; + *fcoe_enc_error = (desc->flags & + CQ_ENET_RQ_DESC_FCOE_ENC_ERROR) ? 1 : 0; + *fcoe_eof = (u8)((desc->checksum_fcoe >> + CQ_ENET_RQ_DESC_FCOE_EOF_SHIFT) & + CQ_ENET_RQ_DESC_FCOE_EOF_MASK); + *checksum = 0; + } else { + *fcoe_sof = 0; + *fcoe_fc_crc_ok = 0; + *fcoe_enc_error = 0; + *fcoe_eof = 0; + *checksum = le16_to_cpu(desc->checksum_fcoe); + } + + *tcp_udp_csum_ok = + (desc->flags & CQ_ENET_RQ_DESC_FLAGS_TCP_UDP_CSUM_OK) ? 1 : 0; + *udp = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_UDP) ? 1 : 0; + *tcp = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_TCP) ? 1 : 0; + *ipv4_csum_ok = + (desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV4_CSUM_OK) ? 1 : 0; + *ipv6 = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV6) ? 1 : 0; + *ipv4 = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV4) ? 1 : 0; + *ipv4_fragment = + (desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV4_FRAGMENT) ? 1 : 0; + *fcs_ok = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_FCS_OK) ? 1 : 0; +} + +#endif /* _CQ_ENET_DESC_H_ */ diff --git a/drivers/scsi/fnic/cq_exch_desc.h b/drivers/scsi/fnic/cq_exch_desc.h new file mode 100644 index 000000000000..501660cfe228 --- /dev/null +++ b/drivers/scsi/fnic/cq_exch_desc.h @@ -0,0 +1,182 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _CQ_EXCH_DESC_H_ +#define _CQ_EXCH_DESC_H_ + +#include "cq_desc.h" + +/* Exchange completion queue descriptor: 16B */ +struct cq_exch_wq_desc { + u16 completed_index; + u16 q_number; + u16 exchange_id; + u8 tmpl; + u8 reserved0; + u32 reserved1; + u8 exch_status; + u8 reserved2[2]; + u8 type_color; +}; + +#define CQ_EXCH_WQ_STATUS_BITS 2 +#define CQ_EXCH_WQ_STATUS_MASK ((1 << CQ_EXCH_WQ_STATUS_BITS) - 1) + +enum cq_exch_status_types { + CQ_EXCH_WQ_STATUS_TYPE_COMPLETE = 0, + CQ_EXCH_WQ_STATUS_TYPE_ABORT = 1, + CQ_EXCH_WQ_STATUS_TYPE_SGL_EOF = 2, + CQ_EXCH_WQ_STATUS_TYPE_TMPL_ERR = 3, +}; + +static inline void cq_exch_wq_desc_dec(struct cq_exch_wq_desc *desc_ptr, + u8 *type, + u8 *color, + u16 *q_number, + u16 *completed_index, + u8 *exch_status) +{ + cq_desc_dec((struct cq_desc *)desc_ptr, type, + color, q_number, completed_index); + *exch_status = desc_ptr->exch_status & CQ_EXCH_WQ_STATUS_MASK; +} + +struct cq_fcp_rq_desc { + u16 completed_index_eop_sop_prt; + u16 q_number; + u16 exchange_id; + u16 tmpl; + u16 bytes_written; + u16 vlan; + u8 sof; + u8 eof; + u8 fcs_fer_fck; + u8 type_color; +}; + +#define CQ_FCP_RQ_DESC_FLAGS_SOP (1 << 15) +#define CQ_FCP_RQ_DESC_FLAGS_EOP (1 << 14) +#define CQ_FCP_RQ_DESC_FLAGS_PRT (1 << 12) +#define CQ_FCP_RQ_DESC_TMPL_MASK 0x1f +#define CQ_FCP_RQ_DESC_BYTES_WRITTEN_MASK 0x3fff +#define CQ_FCP_RQ_DESC_PACKET_ERR_SHIFT 14 +#define CQ_FCP_RQ_DESC_PACKET_ERR_MASK (1 << CQ_FCP_RQ_DESC_PACKET_ERR_SHIFT) +#define CQ_FCP_RQ_DESC_VS_STRIPPED_SHIFT 15 +#define CQ_FCP_RQ_DESC_VS_STRIPPED_MASK (1 << CQ_FCP_RQ_DESC_VS_STRIPPED_SHIFT) +#define CQ_FCP_RQ_DESC_FC_CRC_OK_MASK 0x1 +#define CQ_FCP_RQ_DESC_FCOE_ERR_SHIFT 1 +#define CQ_FCP_RQ_DESC_FCOE_ERR_MASK (1 << CQ_FCP_RQ_DESC_FCOE_ERR_SHIFT) +#define CQ_FCP_RQ_DESC_FCS_OK_SHIFT 7 +#define CQ_FCP_RQ_DESC_FCS_OK_MASK (1 << CQ_FCP_RQ_DESC_FCS_OK_SHIFT) + +static inline void cq_fcp_rq_desc_dec(struct cq_fcp_rq_desc *desc_ptr, + u8 *type, + u8 *color, + u16 *q_number, + u16 *completed_index, + u8 *eop, + u8 *sop, + u8 *fck, + u16 *exchange_id, + u16 *tmpl, + u32 *bytes_written, + u8 *sof, + u8 *eof, + u8 *ingress_port, + u8 *packet_err, + u8 *fcoe_err, + u8 *fcs_ok, + u8 *vlan_stripped, + u16 *vlan) +{ + cq_desc_dec((struct cq_desc *)desc_ptr, type, + color, q_number, completed_index); + *eop = (desc_ptr->completed_index_eop_sop_prt & + CQ_FCP_RQ_DESC_FLAGS_EOP) ? 1 : 0; + *sop = (desc_ptr->completed_index_eop_sop_prt & + CQ_FCP_RQ_DESC_FLAGS_SOP) ? 1 : 0; + *ingress_port = + (desc_ptr->completed_index_eop_sop_prt & + CQ_FCP_RQ_DESC_FLAGS_PRT) ? 1 : 0; + *exchange_id = desc_ptr->exchange_id; + *tmpl = desc_ptr->tmpl & CQ_FCP_RQ_DESC_TMPL_MASK; + *bytes_written = + desc_ptr->bytes_written & CQ_FCP_RQ_DESC_BYTES_WRITTEN_MASK; + *packet_err = + (desc_ptr->bytes_written & CQ_FCP_RQ_DESC_PACKET_ERR_MASK) >> + CQ_FCP_RQ_DESC_PACKET_ERR_SHIFT; + *vlan_stripped = + (desc_ptr->bytes_written & CQ_FCP_RQ_DESC_VS_STRIPPED_MASK) >> + CQ_FCP_RQ_DESC_VS_STRIPPED_SHIFT; + *vlan = desc_ptr->vlan; + *sof = desc_ptr->sof; + *fck = desc_ptr->fcs_fer_fck & CQ_FCP_RQ_DESC_FC_CRC_OK_MASK; + *fcoe_err = (desc_ptr->fcs_fer_fck & CQ_FCP_RQ_DESC_FCOE_ERR_MASK) >> + CQ_FCP_RQ_DESC_FCOE_ERR_SHIFT; + *eof = desc_ptr->eof; + *fcs_ok = + (desc_ptr->fcs_fer_fck & CQ_FCP_RQ_DESC_FCS_OK_MASK) >> + CQ_FCP_RQ_DESC_FCS_OK_SHIFT; +} + +struct cq_sgl_desc { + u16 exchange_id; + u16 q_number; + u32 active_burst_offset; + u32 tot_data_bytes; + u16 tmpl; + u8 sgl_err; + u8 type_color; +}; + +enum cq_sgl_err_types { + CQ_SGL_ERR_NO_ERROR = 0, + CQ_SGL_ERR_OVERFLOW, /* data ran beyond end of SGL */ + CQ_SGL_ERR_SGL_LCL_ADDR_ERR, /* sgl access to local vnic addr illegal*/ + CQ_SGL_ERR_ADDR_RSP_ERR, /* sgl address error */ + CQ_SGL_ERR_DATA_RSP_ERR, /* sgl data rsp error */ + CQ_SGL_ERR_CNT_ZERO_ERR, /* SGL count is 0 */ + CQ_SGL_ERR_CNT_MAX_ERR, /* SGL count is larger than supported */ + CQ_SGL_ERR_ORDER_ERR, /* frames recv on both ports, order err */ + CQ_SGL_ERR_DATA_LCL_ADDR_ERR,/* sgl data buf to local vnic addr ill */ + CQ_SGL_ERR_HOST_CQ_ERR, /* host cq entry to local vnic addr ill */ +}; + +#define CQ_SGL_SGL_ERR_MASK 0x1f +#define CQ_SGL_TMPL_MASK 0x1f + +static inline void cq_sgl_desc_dec(struct cq_sgl_desc *desc_ptr, + u8 *type, + u8 *color, + u16 *q_number, + u16 *exchange_id, + u32 *active_burst_offset, + u32 *tot_data_bytes, + u16 *tmpl, + u8 *sgl_err) +{ + /* Cheat a little by assuming exchange_id is the same as completed + index */ + cq_desc_dec((struct cq_desc *)desc_ptr, type, color, q_number, + exchange_id); + *active_burst_offset = desc_ptr->active_burst_offset; + *tot_data_bytes = desc_ptr->tot_data_bytes; + *tmpl = desc_ptr->tmpl & CQ_SGL_TMPL_MASK; + *sgl_err = desc_ptr->sgl_err & CQ_SGL_SGL_ERR_MASK; +} + +#endif /* _CQ_EXCH_DESC_H_ */ diff --git a/drivers/scsi/fnic/fcpio.h b/drivers/scsi/fnic/fcpio.h new file mode 100644 index 000000000000..12d770d885c5 --- /dev/null +++ b/drivers/scsi/fnic/fcpio.h @@ -0,0 +1,780 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _FCPIO_H_ +#define _FCPIO_H_ + +#include + +/* + * This header file includes all of the data structures used for + * communication by the host driver to the fcp firmware. + */ + +/* + * Exchange and sequence id space allocated to the host driver + */ +#define FCPIO_HOST_EXCH_RANGE_START 0x1000 +#define FCPIO_HOST_EXCH_RANGE_END 0x1fff +#define FCPIO_HOST_SEQ_ID_RANGE_START 0x80 +#define FCPIO_HOST_SEQ_ID_RANGE_END 0xff + +/* + * Command entry type + */ +enum fcpio_type { + /* + * Initiator request types + */ + FCPIO_ICMND_16 = 0x1, + FCPIO_ICMND_32, + FCPIO_ICMND_CMPL, + FCPIO_ITMF, + FCPIO_ITMF_CMPL, + + /* + * Target request types + */ + FCPIO_TCMND_16 = 0x11, + FCPIO_TCMND_32, + FCPIO_TDATA, + FCPIO_TXRDY, + FCPIO_TRSP, + FCPIO_TDRSP_CMPL, + FCPIO_TTMF, + FCPIO_TTMF_ACK, + FCPIO_TABORT, + FCPIO_TABORT_CMPL, + + /* + * Misc request types + */ + FCPIO_ACK = 0x20, + FCPIO_RESET, + FCPIO_RESET_CMPL, + FCPIO_FLOGI_REG, + FCPIO_FLOGI_REG_CMPL, + FCPIO_ECHO, + FCPIO_ECHO_CMPL, + FCPIO_LUNMAP_CHNG, + FCPIO_LUNMAP_REQ, + FCPIO_LUNMAP_REQ_CMPL, + FCPIO_FLOGI_FIP_REG, + FCPIO_FLOGI_FIP_REG_CMPL, +}; + +/* + * Header status codes from the firmware + */ +enum fcpio_status { + FCPIO_SUCCESS = 0, /* request was successful */ + + /* + * If a request to the firmware is rejected, the original request + * header will be returned with the status set to one of the following: + */ + FCPIO_INVALID_HEADER, /* header contains invalid data */ + FCPIO_OUT_OF_RESOURCE, /* out of resources to complete request */ + FCPIO_INVALID_PARAM, /* some parameter in request is invalid */ + FCPIO_REQ_NOT_SUPPORTED, /* request type is not supported */ + FCPIO_IO_NOT_FOUND, /* requested I/O was not found */ + + /* + * Once a request is processed, the firmware will usually return + * a cmpl message type. In cases where errors occurred, + * the header status field would be filled in with one of the following: + */ + FCPIO_ABORTED = 0x41, /* request was aborted */ + FCPIO_TIMEOUT, /* request was timed out */ + FCPIO_SGL_INVALID, /* request was aborted due to sgl error */ + FCPIO_MSS_INVALID, /* request was aborted due to mss error */ + FCPIO_DATA_CNT_MISMATCH, /* recv/sent more/less data than exp. */ + FCPIO_FW_ERR, /* request was terminated due to fw error */ + FCPIO_ITMF_REJECTED, /* itmf req was rejected by remote node */ + FCPIO_ITMF_FAILED, /* itmf req was failed by remote node */ + FCPIO_ITMF_INCORRECT_LUN, /* itmf req targeted incorrect LUN */ + FCPIO_CMND_REJECTED, /* request was invalid and rejected */ + FCPIO_NO_PATH_AVAIL, /* no paths to the lun was available */ + FCPIO_PATH_FAILED, /* i/o sent to current path failed */ + FCPIO_LUNMAP_CHNG_PEND, /* i/o rejected due to lunmap change */ +}; + +/* + * The header command tag. All host requests will use the "tag" field + * to mark commands with a unique tag. When the firmware responds to + * a host request, it will copy the tag field into the response. + * + * The only firmware requests that will use the rx_id/ox_id fields instead + * of the tag field will be the target command and target task management + * requests. These two requests do not have corresponding host requests + * since they come directly from the FC initiator on the network. + */ +struct fcpio_tag { + union { + u32 req_id; + struct { + u16 rx_id; + u16 ox_id; + } ex_id; + } u; +}; + +static inline void +fcpio_tag_id_enc(struct fcpio_tag *tag, u32 id) +{ + tag->u.req_id = id; +} + +static inline void +fcpio_tag_id_dec(struct fcpio_tag *tag, u32 *id) +{ + *id = tag->u.req_id; +} + +static inline void +fcpio_tag_exid_enc(struct fcpio_tag *tag, u16 ox_id, u16 rx_id) +{ + tag->u.ex_id.rx_id = rx_id; + tag->u.ex_id.ox_id = ox_id; +} + +static inline void +fcpio_tag_exid_dec(struct fcpio_tag *tag, u16 *ox_id, u16 *rx_id) +{ + *rx_id = tag->u.ex_id.rx_id; + *ox_id = tag->u.ex_id.ox_id; +} + +/* + * The header for an fcpio request, whether from the firmware or from the + * host driver + */ +struct fcpio_header { + u8 type; /* enum fcpio_type */ + u8 status; /* header status entry */ + u16 _resvd; /* reserved */ + struct fcpio_tag tag; /* header tag */ +}; + +static inline void +fcpio_header_enc(struct fcpio_header *hdr, + u8 type, u8 status, + struct fcpio_tag tag) +{ + hdr->type = type; + hdr->status = status; + hdr->_resvd = 0; + hdr->tag = tag; +} + +static inline void +fcpio_header_dec(struct fcpio_header *hdr, + u8 *type, u8 *status, + struct fcpio_tag *tag) +{ + *type = hdr->type; + *status = hdr->status; + *tag = hdr->tag; +} + +#define CDB_16 16 +#define CDB_32 32 +#define LUN_ADDRESS 8 + +/* + * fcpio_icmnd_16: host -> firmware request + * + * used for sending out an initiator SCSI 16-byte command + */ +struct fcpio_icmnd_16 { + u32 lunmap_id; /* index into lunmap table */ + u8 special_req_flags; /* special exchange request flags */ + u8 _resvd0[3]; /* reserved */ + u32 sgl_cnt; /* scatter-gather list count */ + u32 sense_len; /* sense buffer length */ + u64 sgl_addr; /* scatter-gather list addr */ + u64 sense_addr; /* sense buffer address */ + u8 crn; /* SCSI Command Reference No. */ + u8 pri_ta; /* SCSI Priority and Task attribute */ + u8 _resvd1; /* reserved: should be 0 */ + u8 flags; /* command flags */ + u8 scsi_cdb[CDB_16]; /* SCSI Cmnd Descriptor Block */ + u32 data_len; /* length of data expected */ + u8 lun[LUN_ADDRESS]; /* FC vNIC only: LUN address */ + u8 _resvd2; /* reserved */ + u8 d_id[3]; /* FC vNIC only: Target D_ID */ + u16 mss; /* FC vNIC only: max burst */ + u16 _resvd3; /* reserved */ + u32 r_a_tov; /* FC vNIC only: Res. Alloc Timeout */ + u32 e_d_tov; /* FC vNIC only: Err Detect Timeout */ +}; + +/* + * Special request flags + */ +#define FCPIO_ICMND_SRFLAG_RETRY 0x01 /* Enable Retry handling on exchange */ + +/* + * Priority/Task Attribute settings + */ +#define FCPIO_ICMND_PTA_SIMPLE 0 /* simple task attribute */ +#define FCPIO_ICMND_PTA_HEADQ 1 /* head of queue task attribute */ +#define FCPIO_ICMND_PTA_ORDERED 2 /* ordered task attribute */ +#define FCPIO_ICMND_PTA_ACA 4 /* auto contingent allegiance */ +#define FCPIO_ICMND_PRI_SHIFT 3 /* priority field starts in bit 3 */ + +/* + * Command flags + */ +#define FCPIO_ICMND_RDDATA 0x02 /* read data */ +#define FCPIO_ICMND_WRDATA 0x01 /* write data */ + +/* + * fcpio_icmnd_32: host -> firmware request + * + * used for sending out an initiator SCSI 32-byte command + */ +struct fcpio_icmnd_32 { + u32 lunmap_id; /* index into lunmap table */ + u8 special_req_flags; /* special exchange request flags */ + u8 _resvd0[3]; /* reserved */ + u32 sgl_cnt; /* scatter-gather list count */ + u32 sense_len; /* sense buffer length */ + u64 sgl_addr; /* scatter-gather list addr */ + u64 sense_addr; /* sense buffer address */ + u8 crn; /* SCSI Command Reference No. */ + u8 pri_ta; /* SCSI Priority and Task attribute */ + u8 _resvd1; /* reserved: should be 0 */ + u8 flags; /* command flags */ + u8 scsi_cdb[CDB_32]; /* SCSI Cmnd Descriptor Block */ + u32 data_len; /* length of data expected */ + u8 lun[LUN_ADDRESS]; /* FC vNIC only: LUN address */ + u8 _resvd2; /* reserved */ + u8 d_id[3]; /* FC vNIC only: Target D_ID */ + u16 mss; /* FC vNIC only: max burst */ + u16 _resvd3; /* reserved */ + u32 r_a_tov; /* FC vNIC only: Res. Alloc Timeout */ + u32 e_d_tov; /* FC vNIC only: Error Detect Timeout */ +}; + +/* + * fcpio_itmf: host -> firmware request + * + * used for requesting the firmware to abort a request and/or send out + * a task management function + * + * The t_tag field is only needed when the request type is ABT_TASK. + */ +struct fcpio_itmf { + u32 lunmap_id; /* index into lunmap table */ + u32 tm_req; /* SCSI Task Management request */ + u32 t_tag; /* header tag of fcpio to be aborted */ + u32 _resvd; /* _reserved */ + u8 lun[LUN_ADDRESS]; /* FC vNIC only: LUN address */ + u8 _resvd1; /* reserved */ + u8 d_id[3]; /* FC vNIC only: Target D_ID */ + u32 r_a_tov; /* FC vNIC only: R_A_TOV in msec */ + u32 e_d_tov; /* FC vNIC only: E_D_TOV in msec */ +}; + +/* + * Task Management request + */ +enum fcpio_itmf_tm_req_type { + FCPIO_ITMF_ABT_TASK_TERM = 0x01, /* abort task and terminate */ + FCPIO_ITMF_ABT_TASK, /* abort task and issue abts */ + FCPIO_ITMF_ABT_TASK_SET, /* abort task set */ + FCPIO_ITMF_CLR_TASK_SET, /* clear task set */ + FCPIO_ITMF_LUN_RESET, /* logical unit reset task mgmt */ + FCPIO_ITMF_CLR_ACA, /* Clear ACA condition */ +}; + +/* + * fcpio_tdata: host -> firmware request + * + * used for requesting the firmware to send out a read data transfer for a + * target command + */ +struct fcpio_tdata { + u16 rx_id; /* FC rx_id of target command */ + u16 flags; /* command flags */ + u32 rel_offset; /* data sequence relative offset */ + u32 sgl_cnt; /* scatter-gather list count */ + u32 data_len; /* length of data expected to send */ + u64 sgl_addr; /* scatter-gather list address */ +}; + +/* + * Command flags + */ +#define FCPIO_TDATA_SCSI_RSP 0x01 /* send a scsi resp. after last frame */ + +/* + * fcpio_txrdy: host -> firmware request + * + * used for requesting the firmware to send out a write data transfer for a + * target command + */ +struct fcpio_txrdy { + u16 rx_id; /* FC rx_id of target command */ + u16 _resvd0; /* reserved */ + u32 rel_offset; /* data sequence relative offset */ + u32 sgl_cnt; /* scatter-gather list count */ + u32 data_len; /* length of data expected to send */ + u64 sgl_addr; /* scatter-gather list address */ +}; + +/* + * fcpio_trsp: host -> firmware request + * + * used for requesting the firmware to send out a response for a target + * command + */ +struct fcpio_trsp { + u16 rx_id; /* FC rx_id of target command */ + u16 _resvd0; /* reserved */ + u32 sense_len; /* sense data buffer length */ + u64 sense_addr; /* sense data buffer address */ + u16 _resvd1; /* reserved */ + u8 flags; /* response request flags */ + u8 scsi_status; /* SCSI status */ + u32 residual; /* SCSI data residual value of I/O */ +}; + +/* + * resposnse request flags + */ +#define FCPIO_TRSP_RESID_UNDER 0x08 /* residual is valid and is underflow */ +#define FCPIO_TRSP_RESID_OVER 0x04 /* residual is valid and is overflow */ + +/* + * fcpio_ttmf_ack: host -> firmware response + * + * used by the host to indicate to the firmware it has received and processed + * the target tmf request + */ +struct fcpio_ttmf_ack { + u16 rx_id; /* FC rx_id of target command */ + u16 _resvd0; /* reserved */ + u32 tmf_status; /* SCSI task management status */ +}; + +/* + * fcpio_tabort: host -> firmware request + * + * used by the host to request the firmware to abort a target request that was + * received by the firmware + */ +struct fcpio_tabort { + u16 rx_id; /* rx_id of the target request */ +}; + +/* + * fcpio_reset: host -> firmware request + * + * used by the host to signal a reset of the driver to the firmware + * and to request firmware to clean up all outstanding I/O + */ +struct fcpio_reset { + u32 _resvd; +}; + +enum fcpio_flogi_reg_format_type { + FCPIO_FLOGI_REG_DEF_DEST = 0, /* Use the oui | s_id mac format */ + FCPIO_FLOGI_REG_GW_DEST, /* Use the fixed gateway mac */ +}; + +/* + * fcpio_flogi_reg: host -> firmware request + * + * fc vnic only + * used by the host to notify the firmware of the lif's s_id + * and destination mac address format + */ +struct fcpio_flogi_reg { + u8 format; + u8 s_id[3]; /* FC vNIC only: Source S_ID */ + u8 gateway_mac[ETH_ALEN]; /* Destination gateway mac */ + u16 _resvd; + u32 r_a_tov; /* R_A_TOV in msec */ + u32 e_d_tov; /* E_D_TOV in msec */ +}; + +/* + * fcpio_echo: host -> firmware request + * + * sends a heartbeat echo request to the firmware + */ +struct fcpio_echo { + u32 _resvd; +}; + +/* + * fcpio_lunmap_req: host -> firmware request + * + * scsi vnic only + * sends a request to retrieve the lunmap table for scsi vnics + */ +struct fcpio_lunmap_req { + u64 addr; /* address of the buffer */ + u32 len; /* len of the buffer */ +}; + +/* + * fcpio_flogi_fip_reg: host -> firmware request + * + * fc vnic only + * used by the host to notify the firmware of the lif's s_id + * and destination mac address format + */ +struct fcpio_flogi_fip_reg { + u8 _resvd0; + u8 s_id[3]; /* FC vNIC only: Source S_ID */ + u8 fcf_mac[ETH_ALEN]; /* FCF Target destination mac */ + u16 _resvd1; + u32 r_a_tov; /* R_A_TOV in msec */ + u32 e_d_tov; /* E_D_TOV in msec */ + u8 ha_mac[ETH_ALEN]; /* Host adapter source mac */ + u16 _resvd2; +}; + +/* + * Basic structure for all fcpio structures that are sent from the host to the + * firmware. They are 128 bytes per structure. + */ +#define FCPIO_HOST_REQ_LEN 128 /* expected length of host requests */ + +struct fcpio_host_req { + struct fcpio_header hdr; + + union { + /* + * Defines space needed for request + */ + u8 buf[FCPIO_HOST_REQ_LEN - sizeof(struct fcpio_header)]; + + /* + * Initiator host requests + */ + struct fcpio_icmnd_16 icmnd_16; + struct fcpio_icmnd_32 icmnd_32; + struct fcpio_itmf itmf; + + /* + * Target host requests + */ + struct fcpio_tdata tdata; + struct fcpio_txrdy txrdy; + struct fcpio_trsp trsp; + struct fcpio_ttmf_ack ttmf_ack; + struct fcpio_tabort tabort; + + /* + * Misc requests + */ + struct fcpio_reset reset; + struct fcpio_flogi_reg flogi_reg; + struct fcpio_echo echo; + struct fcpio_lunmap_req lunmap_req; + struct fcpio_flogi_fip_reg flogi_fip_reg; + } u; +}; + +/* + * fcpio_icmnd_cmpl: firmware -> host response + * + * used for sending the host a response to an initiator command + */ +struct fcpio_icmnd_cmpl { + u8 _resvd0[6]; /* reserved */ + u8 flags; /* response flags */ + u8 scsi_status; /* SCSI status */ + u32 residual; /* SCSI data residual length */ + u32 sense_len; /* SCSI sense length */ +}; + +/* + * response flags + */ +#define FCPIO_ICMND_CMPL_RESID_UNDER 0x08 /* resid under and valid */ +#define FCPIO_ICMND_CMPL_RESID_OVER 0x04 /* resid over and valid */ + +/* + * fcpio_itmf_cmpl: firmware -> host response + * + * used for sending the host a response for a itmf request + */ +struct fcpio_itmf_cmpl { + u32 _resvd; /* reserved */ +}; + +/* + * fcpio_tcmnd_16: firmware -> host request + * + * used by the firmware to notify the host of an incoming target SCSI 16-Byte + * request + */ +struct fcpio_tcmnd_16 { + u8 lun[LUN_ADDRESS]; /* FC vNIC only: LUN address */ + u8 crn; /* SCSI Command Reference No. */ + u8 pri_ta; /* SCSI Priority and Task attribute */ + u8 _resvd2; /* reserved: should be 0 */ + u8 flags; /* command flags */ + u8 scsi_cdb[CDB_16]; /* SCSI Cmnd Descriptor Block */ + u32 data_len; /* length of data expected */ + u8 _resvd1; /* reserved */ + u8 s_id[3]; /* FC vNIC only: Source S_ID */ +}; + +/* + * Priority/Task Attribute settings + */ +#define FCPIO_TCMND_PTA_SIMPLE 0 /* simple task attribute */ +#define FCPIO_TCMND_PTA_HEADQ 1 /* head of queue task attribute */ +#define FCPIO_TCMND_PTA_ORDERED 2 /* ordered task attribute */ +#define FCPIO_TCMND_PTA_ACA 4 /* auto contingent allegiance */ +#define FCPIO_TCMND_PRI_SHIFT 3 /* priority field starts in bit 3 */ + +/* + * Command flags + */ +#define FCPIO_TCMND_RDDATA 0x02 /* read data */ +#define FCPIO_TCMND_WRDATA 0x01 /* write data */ + +/* + * fcpio_tcmnd_32: firmware -> host request + * + * used by the firmware to notify the host of an incoming target SCSI 32-Byte + * request + */ +struct fcpio_tcmnd_32 { + u8 lun[LUN_ADDRESS]; /* FC vNIC only: LUN address */ + u8 crn; /* SCSI Command Reference No. */ + u8 pri_ta; /* SCSI Priority and Task attribute */ + u8 _resvd2; /* reserved: should be 0 */ + u8 flags; /* command flags */ + u8 scsi_cdb[CDB_32]; /* SCSI Cmnd Descriptor Block */ + u32 data_len; /* length of data expected */ + u8 _resvd0; /* reserved */ + u8 s_id[3]; /* FC vNIC only: Source S_ID */ +}; + +/* + * fcpio_tdrsp_cmpl: firmware -> host response + * + * used by the firmware to notify the host of a response to a host target + * command + */ +struct fcpio_tdrsp_cmpl { + u16 rx_id; /* rx_id of the target request */ + u16 _resvd0; /* reserved */ +}; + +/* + * fcpio_ttmf: firmware -> host request + * + * used by the firmware to notify the host of an incoming task management + * function request + */ +struct fcpio_ttmf { + u8 _resvd0; /* reserved */ + u8 s_id[3]; /* FC vNIC only: Source S_ID */ + u8 lun[LUN_ADDRESS]; /* FC vNIC only: LUN address */ + u8 crn; /* SCSI Command Reference No. */ + u8 _resvd2[3]; /* reserved */ + u32 tmf_type; /* task management request type */ +}; + +/* + * Task Management request + */ +#define FCPIO_TTMF_CLR_ACA 0x40 /* Clear ACA condition */ +#define FCPIO_TTMF_LUN_RESET 0x10 /* logical unit reset task mgmt */ +#define FCPIO_TTMF_CLR_TASK_SET 0x04 /* clear task set */ +#define FCPIO_TTMF_ABT_TASK_SET 0x02 /* abort task set */ +#define FCPIO_TTMF_ABT_TASK 0x01 /* abort task */ + +/* + * fcpio_tabort_cmpl: firmware -> host response + * + * used by the firmware to respond to a host's tabort request + */ +struct fcpio_tabort_cmpl { + u16 rx_id; /* rx_id of the target request */ + u16 _resvd0; /* reserved */ +}; + +/* + * fcpio_ack: firmware -> host response + * + * used by firmware to notify the host of the last work request received + */ +struct fcpio_ack { + u16 request_out; /* last host entry received */ + u16 _resvd; +}; + +/* + * fcpio_reset_cmpl: firmware -> host response + * + * use by firmware to respond to the host's reset request + */ +struct fcpio_reset_cmpl { + u16 vnic_id; +}; + +/* + * fcpio_flogi_reg_cmpl: firmware -> host response + * + * fc vnic only + * response to the fcpio_flogi_reg request + */ +struct fcpio_flogi_reg_cmpl { + u32 _resvd; +}; + +/* + * fcpio_echo_cmpl: firmware -> host response + * + * response to the fcpio_echo request + */ +struct fcpio_echo_cmpl { + u32 _resvd; +}; + +/* + * fcpio_lunmap_chng: firmware -> host notification + * + * scsi vnic only + * notifies the host that the lunmap tables have changed + */ +struct fcpio_lunmap_chng { + u32 _resvd; +}; + +/* + * fcpio_lunmap_req_cmpl: firmware -> host response + * + * scsi vnic only + * response for lunmap table request from the host + */ +struct fcpio_lunmap_req_cmpl { + u32 _resvd; +}; + +/* + * Basic structure for all fcpio structures that are sent from the firmware to + * the host. They are 64 bytes per structure. + */ +#define FCPIO_FW_REQ_LEN 64 /* expected length of fw requests */ +struct fcpio_fw_req { + struct fcpio_header hdr; + + union { + /* + * Defines space needed for request + */ + u8 buf[FCPIO_FW_REQ_LEN - sizeof(struct fcpio_header)]; + + /* + * Initiator firmware responses + */ + struct fcpio_icmnd_cmpl icmnd_cmpl; + struct fcpio_itmf_cmpl itmf_cmpl; + + /* + * Target firmware new requests + */ + struct fcpio_tcmnd_16 tcmnd_16; + struct fcpio_tcmnd_32 tcmnd_32; + + /* + * Target firmware responses + */ + struct fcpio_tdrsp_cmpl tdrsp_cmpl; + struct fcpio_ttmf ttmf; + struct fcpio_tabort_cmpl tabort_cmpl; + + /* + * Firmware response to work received + */ + struct fcpio_ack ack; + + /* + * Misc requests + */ + struct fcpio_reset_cmpl reset_cmpl; + struct fcpio_flogi_reg_cmpl flogi_reg_cmpl; + struct fcpio_echo_cmpl echo_cmpl; + struct fcpio_lunmap_chng lunmap_chng; + struct fcpio_lunmap_req_cmpl lunmap_req_cmpl; + } u; +}; + +/* + * Access routines to encode and decode the color bit, which is the most + * significant bit of the MSB of the structure + */ +static inline void fcpio_color_enc(struct fcpio_fw_req *fw_req, u8 color) +{ + u8 *c = ((u8 *) fw_req) + sizeof(struct fcpio_fw_req) - 1; + + if (color) + *c |= 0x80; + else + *c &= ~0x80; +} + +static inline void fcpio_color_dec(struct fcpio_fw_req *fw_req, u8 *color) +{ + u8 *c = ((u8 *) fw_req) + sizeof(struct fcpio_fw_req) - 1; + + *color = *c >> 7; + + /* + * Make sure color bit is read from desc *before* other fields + * are read from desc. Hardware guarantees color bit is last + * bit (byte) written. Adding the rmb() prevents the compiler + * and/or CPU from reordering the reads which would potentially + * result in reading stale values. + */ + + rmb(); + +} + +/* + * Lunmap table entry for scsi vnics + */ +#define FCPIO_LUNMAP_TABLE_SIZE 256 +#define FCPIO_FLAGS_LUNMAP_VALID 0x80 +#define FCPIO_FLAGS_BOOT 0x01 +struct fcpio_lunmap_entry { + u8 bus; + u8 target; + u8 lun; + u8 path_cnt; + u16 flags; + u16 update_cnt; +}; + +struct fcpio_lunmap_tbl { + u32 update_cnt; + struct fcpio_lunmap_entry lunmaps[FCPIO_LUNMAP_TABLE_SIZE]; +}; + +#endif /* _FCPIO_H_ */ diff --git a/drivers/scsi/fnic/fnic.h b/drivers/scsi/fnic/fnic.h new file mode 100644 index 000000000000..e4c0a3d7d87b --- /dev/null +++ b/drivers/scsi/fnic/fnic.h @@ -0,0 +1,265 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _FNIC_H_ +#define _FNIC_H_ + +#include +#include +#include +#include +#include "fnic_io.h" +#include "fnic_res.h" +#include "vnic_dev.h" +#include "vnic_wq.h" +#include "vnic_rq.h" +#include "vnic_cq.h" +#include "vnic_wq_copy.h" +#include "vnic_intr.h" +#include "vnic_stats.h" +#include "vnic_scsi.h" + +#define DRV_NAME "fnic" +#define DRV_DESCRIPTION "Cisco FCoE HBA Driver" +#define DRV_VERSION "1.0.0.1121" +#define PFX DRV_NAME ": " +#define DFX DRV_NAME "%d: " + +#define DESC_CLEAN_LOW_WATERMARK 8 +#define FNIC_MAX_IO_REQ 2048 /* scsi_cmnd tag map entries */ +#define FNIC_IO_LOCKS 64 /* IO locks: power of 2 */ +#define FNIC_DFLT_QUEUE_DEPTH 32 +#define FNIC_STATS_RATE_LIMIT 4 /* limit rate at which stats are pulled up */ + +/* + * Tag bits used for special requests. + */ +#define BIT(nr) (1UL << (nr)) +#define FNIC_TAG_ABORT BIT(30) /* tag bit indicating abort */ +#define FNIC_TAG_DEV_RST BIT(29) /* indicates device reset */ +#define FNIC_TAG_MASK (BIT(24) - 1) /* mask for lookup */ +#define FNIC_NO_TAG -1 + +/* + * Usage of the scsi_cmnd scratchpad. + * These fields are locked by the hashed io_req_lock. + */ +#define CMD_SP(Cmnd) ((Cmnd)->SCp.ptr) +#define CMD_STATE(Cmnd) ((Cmnd)->SCp.phase) +#define CMD_ABTS_STATUS(Cmnd) ((Cmnd)->SCp.Message) +#define CMD_LR_STATUS(Cmnd) ((Cmnd)->SCp.have_data_in) +#define CMD_TAG(Cmnd) ((Cmnd)->SCp.sent_command) + +#define FCPIO_INVALID_CODE 0x100 /* hdr_status value unused by firmware */ + +#define FNIC_LUN_RESET_TIMEOUT 10000 /* mSec */ +#define FNIC_HOST_RESET_TIMEOUT 10000 /* mSec */ +#define FNIC_RMDEVICE_TIMEOUT 1000 /* mSec */ +#define FNIC_HOST_RESET_SETTLE_TIME 30 /* Sec */ + +#define FNIC_MAX_FCP_TARGET 256 + +extern unsigned int fnic_log_level; + +#define FNIC_MAIN_LOGGING 0x01 +#define FNIC_FCS_LOGGING 0x02 +#define FNIC_SCSI_LOGGING 0x04 +#define FNIC_ISR_LOGGING 0x08 + +#define FNIC_CHECK_LOGGING(LEVEL, CMD) \ +do { \ + if (unlikely(fnic_log_level & LEVEL)) \ + do { \ + CMD; \ + } while (0); \ +} while (0) + +#define FNIC_MAIN_DBG(kern_level, host, fmt, args...) \ + FNIC_CHECK_LOGGING(FNIC_MAIN_LOGGING, \ + shost_printk(kern_level, host, fmt, ##args);) + +#define FNIC_FCS_DBG(kern_level, host, fmt, args...) \ + FNIC_CHECK_LOGGING(FNIC_FCS_LOGGING, \ + shost_printk(kern_level, host, fmt, ##args);) + +#define FNIC_SCSI_DBG(kern_level, host, fmt, args...) \ + FNIC_CHECK_LOGGING(FNIC_SCSI_LOGGING, \ + shost_printk(kern_level, host, fmt, ##args);) + +#define FNIC_ISR_DBG(kern_level, host, fmt, args...) \ + FNIC_CHECK_LOGGING(FNIC_ISR_LOGGING, \ + shost_printk(kern_level, host, fmt, ##args);) + +extern const char *fnic_state_str[]; + +enum fnic_intx_intr_index { + FNIC_INTX_WQ_RQ_COPYWQ, + FNIC_INTX_ERR, + FNIC_INTX_NOTIFY, + FNIC_INTX_INTR_MAX, +}; + +enum fnic_msix_intr_index { + FNIC_MSIX_RQ, + FNIC_MSIX_WQ, + FNIC_MSIX_WQ_COPY, + FNIC_MSIX_ERR_NOTIFY, + FNIC_MSIX_INTR_MAX, +}; + +struct fnic_msix_entry { + int requested; + char devname[IFNAMSIZ]; + irqreturn_t (*isr)(int, void *); + void *devid; +}; + +enum fnic_state { + FNIC_IN_FC_MODE = 0, + FNIC_IN_FC_TRANS_ETH_MODE, + FNIC_IN_ETH_MODE, + FNIC_IN_ETH_TRANS_FC_MODE, +}; + +#define FNIC_WQ_COPY_MAX 1 +#define FNIC_WQ_MAX 1 +#define FNIC_RQ_MAX 1 +#define FNIC_CQ_MAX (FNIC_WQ_COPY_MAX + FNIC_WQ_MAX + FNIC_RQ_MAX) + +struct mempool; + +/* Per-instance private data structure */ +struct fnic { + struct fc_lport *lport; + struct vnic_dev_bar bar0; + + struct msix_entry msix_entry[FNIC_MSIX_INTR_MAX]; + struct fnic_msix_entry msix[FNIC_MSIX_INTR_MAX]; + + struct vnic_stats *stats; + unsigned long stats_time; /* time of stats update */ + struct vnic_nic_cfg *nic_cfg; + char name[IFNAMSIZ]; + struct timer_list notify_timer; /* used for MSI interrupts */ + + unsigned int err_intr_offset; + unsigned int link_intr_offset; + + unsigned int wq_count; + unsigned int cq_count; + + u32 fcoui_mode:1; /* use fcoui address*/ + u32 vlan_hw_insert:1; /* let hw insert the tag */ + u32 in_remove:1; /* fnic device in removal */ + u32 stop_rx_link_events:1; /* stop proc. rx frames, link events */ + + struct completion *remove_wait; /* device remove thread blocks */ + + struct fc_frame *flogi; + struct fc_frame *flogi_resp; + u16 flogi_oxid; + unsigned long s_id; + enum fnic_state state; + spinlock_t fnic_lock; + + u16 vlan_id; /* VLAN tag including priority */ + u8 mac_addr[ETH_ALEN]; + u8 dest_addr[ETH_ALEN]; + u8 data_src_addr[ETH_ALEN]; + u64 fcp_input_bytes; /* internal statistic */ + u64 fcp_output_bytes; /* internal statistic */ + u32 link_down_cnt; + int link_status; + + struct list_head list; + struct pci_dev *pdev; + struct vnic_fc_config config; + struct vnic_dev *vdev; + unsigned int raw_wq_count; + unsigned int wq_copy_count; + unsigned int rq_count; + int fw_ack_index[FNIC_WQ_COPY_MAX]; + unsigned short fw_ack_recd[FNIC_WQ_COPY_MAX]; + unsigned short wq_copy_desc_low[FNIC_WQ_COPY_MAX]; + unsigned int intr_count; + u32 __iomem *legacy_pba; + struct fnic_host_tag *tags; + mempool_t *io_req_pool; + mempool_t *io_sgl_pool[FNIC_SGL_NUM_CACHES]; + spinlock_t io_req_lock[FNIC_IO_LOCKS]; /* locks for scsi cmnds */ + + struct work_struct link_work; + struct work_struct frame_work; + struct sk_buff_head frame_queue; + + /* copy work queue cache line section */ + ____cacheline_aligned struct vnic_wq_copy wq_copy[FNIC_WQ_COPY_MAX]; + /* completion queue cache line section */ + ____cacheline_aligned struct vnic_cq cq[FNIC_CQ_MAX]; + + spinlock_t wq_copy_lock[FNIC_WQ_COPY_MAX]; + + /* work queue cache line section */ + ____cacheline_aligned struct vnic_wq wq[FNIC_WQ_MAX]; + spinlock_t wq_lock[FNIC_WQ_MAX]; + + /* receive queue cache line section */ + ____cacheline_aligned struct vnic_rq rq[FNIC_RQ_MAX]; + + /* interrupt resource cache line section */ + ____cacheline_aligned struct vnic_intr intr[FNIC_MSIX_INTR_MAX]; +}; + +extern struct workqueue_struct *fnic_event_queue; +extern struct device_attribute *fnic_attrs[]; + +void fnic_clear_intr_mode(struct fnic *fnic); +int fnic_set_intr_mode(struct fnic *fnic); +void fnic_free_intr(struct fnic *fnic); +int fnic_request_intr(struct fnic *fnic); + +int fnic_send(struct fc_lport *, struct fc_frame *); +void fnic_free_wq_buf(struct vnic_wq *wq, struct vnic_wq_buf *buf); +void fnic_handle_frame(struct work_struct *work); +void fnic_handle_link(struct work_struct *work); +int fnic_rq_cmpl_handler(struct fnic *fnic, int); +int fnic_alloc_rq_frame(struct vnic_rq *rq); +void fnic_free_rq_buf(struct vnic_rq *rq, struct vnic_rq_buf *buf); +int fnic_send_frame(struct fnic *fnic, struct fc_frame *fp); + +int fnic_queuecommand(struct scsi_cmnd *, void (*done)(struct scsi_cmnd *)); +int fnic_abort_cmd(struct scsi_cmnd *); +int fnic_device_reset(struct scsi_cmnd *); +int fnic_host_reset(struct scsi_cmnd *); +int fnic_reset(struct Scsi_Host *); +void fnic_scsi_cleanup(struct fc_lport *); +void fnic_scsi_abort_io(struct fc_lport *); +void fnic_empty_scsi_cleanup(struct fc_lport *); +void fnic_exch_mgr_reset(struct fc_lport *, u32, u32); +int fnic_wq_copy_cmpl_handler(struct fnic *fnic, int); +int fnic_wq_cmpl_handler(struct fnic *fnic, int); +int fnic_flogi_reg_handler(struct fnic *fnic); +void fnic_wq_copy_cleanup_handler(struct vnic_wq_copy *wq, + struct fcpio_host_req *desc); +int fnic_fw_reset_handler(struct fnic *fnic); +void fnic_terminate_rport_io(struct fc_rport *); +const char *fnic_state_to_str(unsigned int state); + +void fnic_log_q_error(struct fnic *fnic); +void fnic_handle_link_event(struct fnic *fnic); + +#endif /* _FNIC_H_ */ diff --git a/drivers/scsi/fnic/fnic_attrs.c b/drivers/scsi/fnic/fnic_attrs.c new file mode 100644 index 000000000000..aea0c3becfd4 --- /dev/null +++ b/drivers/scsi/fnic/fnic_attrs.c @@ -0,0 +1,56 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#include +#include +#include +#include "fnic.h" + +static ssize_t fnic_show_state(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct fc_lport *lp = shost_priv(class_to_shost(dev)); + struct fnic *fnic = lport_priv(lp); + + return snprintf(buf, PAGE_SIZE, "%s\n", fnic_state_str[fnic->state]); +} + +static ssize_t fnic_show_drv_version(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%s\n", DRV_VERSION); +} + +static ssize_t fnic_show_link_state(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct fc_lport *lp = shost_priv(class_to_shost(dev)); + + return snprintf(buf, PAGE_SIZE, "%s\n", (lp->link_up) + ? "Link Up" : "Link Down"); +} + +static DEVICE_ATTR(fnic_state, S_IRUGO, fnic_show_state, NULL); +static DEVICE_ATTR(drv_version, S_IRUGO, fnic_show_drv_version, NULL); +static DEVICE_ATTR(link_state, S_IRUGO, fnic_show_link_state, NULL); + +struct device_attribute *fnic_attrs[] = { + &dev_attr_fnic_state, + &dev_attr_drv_version, + &dev_attr_link_state, + NULL, +}; diff --git a/drivers/scsi/fnic/fnic_fcs.c b/drivers/scsi/fnic/fnic_fcs.c new file mode 100644 index 000000000000..07e6eedb83ce --- /dev/null +++ b/drivers/scsi/fnic/fnic_fcs.c @@ -0,0 +1,742 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "fnic_io.h" +#include "fnic.h" +#include "cq_enet_desc.h" +#include "cq_exch_desc.h" + +struct workqueue_struct *fnic_event_queue; + +void fnic_handle_link(struct work_struct *work) +{ + struct fnic *fnic = container_of(work, struct fnic, link_work); + unsigned long flags; + int old_link_status; + u32 old_link_down_cnt; + + spin_lock_irqsave(&fnic->fnic_lock, flags); + + if (fnic->stop_rx_link_events) { + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + return; + } + + old_link_down_cnt = fnic->link_down_cnt; + old_link_status = fnic->link_status; + fnic->link_status = vnic_dev_link_status(fnic->vdev); + fnic->link_down_cnt = vnic_dev_link_down_cnt(fnic->vdev); + + if (old_link_status == fnic->link_status) { + if (!fnic->link_status) + /* DOWN -> DOWN */ + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + else { + if (old_link_down_cnt != fnic->link_down_cnt) { + /* UP -> DOWN -> UP */ + fnic->lport->host_stats.link_failure_count++; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host, + "link down\n"); + fc_linkdown(fnic->lport); + FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host, + "link up\n"); + fc_linkup(fnic->lport); + } else + /* UP -> UP */ + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + } + } else if (fnic->link_status) { + /* DOWN -> UP */ + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host, "link up\n"); + fc_linkup(fnic->lport); + } else { + /* UP -> DOWN */ + fnic->lport->host_stats.link_failure_count++; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host, "link down\n"); + fc_linkdown(fnic->lport); + } + +} + +/* + * This function passes incoming fabric frames to libFC + */ +void fnic_handle_frame(struct work_struct *work) +{ + struct fnic *fnic = container_of(work, struct fnic, frame_work); + struct fc_lport *lp = fnic->lport; + unsigned long flags; + struct sk_buff *skb; + struct fc_frame *fp; + + while ((skb = skb_dequeue(&fnic->frame_queue))) { + + spin_lock_irqsave(&fnic->fnic_lock, flags); + if (fnic->stop_rx_link_events) { + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + dev_kfree_skb(skb); + return; + } + fp = (struct fc_frame *)skb; + /* if Flogi resp frame, register the address */ + if (fr_flags(fp)) { + vnic_dev_add_addr(fnic->vdev, + fnic->data_src_addr); + fr_flags(fp) = 0; + } + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + + fc_exch_recv(lp, lp->emp, fp); + } + +} + +static inline void fnic_import_rq_fc_frame(struct sk_buff *skb, + u32 len, u8 sof, u8 eof) +{ + struct fc_frame *fp = (struct fc_frame *)skb; + + skb_trim(skb, len); + fr_eof(fp) = eof; + fr_sof(fp) = sof; +} + + +static inline int fnic_import_rq_eth_pkt(struct sk_buff *skb, u32 len) +{ + struct fc_frame *fp; + struct ethhdr *eh; + struct vlan_ethhdr *vh; + struct fcoe_hdr *fcoe_hdr; + struct fcoe_crc_eof *ft; + u32 transport_len = 0; + + eh = (struct ethhdr *)skb->data; + vh = (struct vlan_ethhdr *)skb->data; + if (vh->h_vlan_proto == htons(ETH_P_8021Q) && + vh->h_vlan_encapsulated_proto == htons(ETH_P_FCOE)) { + skb_pull(skb, sizeof(struct vlan_ethhdr)); + transport_len += sizeof(struct vlan_ethhdr); + } else if (eh->h_proto == htons(ETH_P_FCOE)) { + transport_len += sizeof(struct ethhdr); + skb_pull(skb, sizeof(struct ethhdr)); + } else + return -1; + + fcoe_hdr = (struct fcoe_hdr *)skb->data; + if (FC_FCOE_DECAPS_VER(fcoe_hdr) != FC_FCOE_VER) + return -1; + + fp = (struct fc_frame *)skb; + fc_frame_init(fp); + fr_sof(fp) = fcoe_hdr->fcoe_sof; + skb_pull(skb, sizeof(struct fcoe_hdr)); + transport_len += sizeof(struct fcoe_hdr); + + ft = (struct fcoe_crc_eof *)(skb->data + len - + transport_len - sizeof(*ft)); + fr_eof(fp) = ft->fcoe_eof; + skb_trim(skb, len - transport_len - sizeof(*ft)); + return 0; +} + +static inline int fnic_handle_flogi_resp(struct fnic *fnic, + struct fc_frame *fp) +{ + u8 mac[ETH_ALEN] = FC_FCOE_FLOGI_MAC; + struct ethhdr *eth_hdr; + struct fc_frame_header *fh; + int ret = 0; + unsigned long flags; + struct fc_frame *old_flogi_resp = NULL; + + fh = (struct fc_frame_header *)fr_hdr(fp); + + spin_lock_irqsave(&fnic->fnic_lock, flags); + + if (fnic->state == FNIC_IN_ETH_MODE) { + + /* + * Check if oxid matches on taking the lock. A new Flogi + * issued by libFC might have changed the fnic cached oxid + */ + if (fnic->flogi_oxid != ntohs(fh->fh_ox_id)) { + FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host, + "Flogi response oxid not" + " matching cached oxid, dropping frame" + "\n"); + ret = -1; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + dev_kfree_skb_irq(fp_skb(fp)); + goto handle_flogi_resp_end; + } + + /* Drop older cached flogi response frame, cache this frame */ + old_flogi_resp = fnic->flogi_resp; + fnic->flogi_resp = fp; + fnic->flogi_oxid = FC_XID_UNKNOWN; + + /* + * this frame is part of flogi get the src mac addr from this + * frame if the src mac is fcoui based then we mark the + * address mode flag to use fcoui base for dst mac addr + * otherwise we have to store the fcoe gateway addr + */ + eth_hdr = (struct ethhdr *)skb_mac_header(fp_skb(fp)); + memcpy(mac, eth_hdr->h_source, ETH_ALEN); + + if (ntoh24(mac) == FC_FCOE_OUI) + fnic->fcoui_mode = 1; + else { + fnic->fcoui_mode = 0; + memcpy(fnic->dest_addr, mac, ETH_ALEN); + } + + /* + * Except for Flogi frame, all outbound frames from us have the + * Eth Src address as FC_FCOE_OUI"our_sid". Flogi frame uses + * the vnic MAC address as the Eth Src address + */ + fc_fcoe_set_mac(fnic->data_src_addr, fh->fh_d_id); + + /* We get our s_id from the d_id of the flogi resp frame */ + fnic->s_id = ntoh24(fh->fh_d_id); + + /* Change state to reflect transition from Eth to FC mode */ + fnic->state = FNIC_IN_ETH_TRANS_FC_MODE; + + } else { + FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host, + "Unexpected fnic state %s while" + " processing flogi resp\n", + fnic_state_to_str(fnic->state)); + ret = -1; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + dev_kfree_skb_irq(fp_skb(fp)); + goto handle_flogi_resp_end; + } + + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + + /* Drop older cached frame */ + if (old_flogi_resp) + dev_kfree_skb_irq(fp_skb(old_flogi_resp)); + + /* + * send flogi reg request to firmware, this will put the fnic in + * in FC mode + */ + ret = fnic_flogi_reg_handler(fnic); + + if (ret < 0) { + int free_fp = 1; + spin_lock_irqsave(&fnic->fnic_lock, flags); + /* + * free the frame is some other thread is not + * pointing to it + */ + if (fnic->flogi_resp != fp) + free_fp = 0; + else + fnic->flogi_resp = NULL; + + if (fnic->state == FNIC_IN_ETH_TRANS_FC_MODE) + fnic->state = FNIC_IN_ETH_MODE; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + if (free_fp) + dev_kfree_skb_irq(fp_skb(fp)); + } + + handle_flogi_resp_end: + return ret; +} + +/* Returns 1 for a response that matches cached flogi oxid */ +static inline int is_matching_flogi_resp_frame(struct fnic *fnic, + struct fc_frame *fp) +{ + struct fc_frame_header *fh; + int ret = 0; + u32 f_ctl; + + fh = fc_frame_header_get(fp); + f_ctl = ntoh24(fh->fh_f_ctl); + + if (fnic->flogi_oxid == ntohs(fh->fh_ox_id) && + fh->fh_r_ctl == FC_RCTL_ELS_REP && + (f_ctl & (FC_FC_EX_CTX | FC_FC_SEQ_CTX)) == FC_FC_EX_CTX && + fh->fh_type == FC_TYPE_ELS) + ret = 1; + + return ret; +} + +static void fnic_rq_cmpl_frame_recv(struct vnic_rq *rq, struct cq_desc + *cq_desc, struct vnic_rq_buf *buf, + int skipped __attribute__((unused)), + void *opaque) +{ + struct fnic *fnic = vnic_dev_priv(rq->vdev); + struct sk_buff *skb; + struct fc_frame *fp; + unsigned int eth_hdrs_stripped; + u8 type, color, eop, sop, ingress_port, vlan_stripped; + u8 fcoe = 0, fcoe_sof, fcoe_eof; + u8 fcoe_fc_crc_ok = 1, fcoe_enc_error = 0; + u8 tcp_udp_csum_ok, udp, tcp, ipv4_csum_ok; + u8 ipv6, ipv4, ipv4_fragment, rss_type, csum_not_calc; + u8 fcs_ok = 1, packet_error = 0; + u16 q_number, completed_index, bytes_written = 0, vlan, checksum; + u32 rss_hash; + u16 exchange_id, tmpl; + u8 sof = 0; + u8 eof = 0; + u32 fcp_bytes_written = 0; + unsigned long flags; + + pci_unmap_single(fnic->pdev, buf->dma_addr, buf->len, + PCI_DMA_FROMDEVICE); + skb = buf->os_buf; + buf->os_buf = NULL; + + cq_desc_dec(cq_desc, &type, &color, &q_number, &completed_index); + if (type == CQ_DESC_TYPE_RQ_FCP) { + cq_fcp_rq_desc_dec((struct cq_fcp_rq_desc *)cq_desc, + &type, &color, &q_number, &completed_index, + &eop, &sop, &fcoe_fc_crc_ok, &exchange_id, + &tmpl, &fcp_bytes_written, &sof, &eof, + &ingress_port, &packet_error, + &fcoe_enc_error, &fcs_ok, &vlan_stripped, + &vlan); + eth_hdrs_stripped = 1; + + } else if (type == CQ_DESC_TYPE_RQ_ENET) { + cq_enet_rq_desc_dec((struct cq_enet_rq_desc *)cq_desc, + &type, &color, &q_number, &completed_index, + &ingress_port, &fcoe, &eop, &sop, + &rss_type, &csum_not_calc, &rss_hash, + &bytes_written, &packet_error, + &vlan_stripped, &vlan, &checksum, + &fcoe_sof, &fcoe_fc_crc_ok, + &fcoe_enc_error, &fcoe_eof, + &tcp_udp_csum_ok, &udp, &tcp, + &ipv4_csum_ok, &ipv6, &ipv4, + &ipv4_fragment, &fcs_ok); + eth_hdrs_stripped = 0; + + } else { + /* wrong CQ type*/ + shost_printk(KERN_ERR, fnic->lport->host, + "fnic rq_cmpl wrong cq type x%x\n", type); + goto drop; + } + + if (!fcs_ok || packet_error || !fcoe_fc_crc_ok || fcoe_enc_error) { + FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host, + "fnic rq_cmpl fcoe x%x fcsok x%x" + " pkterr x%x fcoe_fc_crc_ok x%x, fcoe_enc_err" + " x%x\n", + fcoe, fcs_ok, packet_error, + fcoe_fc_crc_ok, fcoe_enc_error); + goto drop; + } + + if (eth_hdrs_stripped) + fnic_import_rq_fc_frame(skb, fcp_bytes_written, sof, eof); + else if (fnic_import_rq_eth_pkt(skb, bytes_written)) + goto drop; + + fp = (struct fc_frame *)skb; + + /* + * If frame is an ELS response that matches the cached FLOGI OX_ID, + * and is accept, issue flogi_reg_request copy wq request to firmware + * to register the S_ID and determine whether FC_OUI mode or GW mode. + */ + if (is_matching_flogi_resp_frame(fnic, fp)) { + if (!eth_hdrs_stripped) { + if (fc_frame_payload_op(fp) == ELS_LS_ACC) { + fnic_handle_flogi_resp(fnic, fp); + return; + } + /* + * Recd. Flogi reject. No point registering + * with fw, but forward to libFC + */ + goto forward; + } + goto drop; + } + if (!eth_hdrs_stripped) + goto drop; + +forward: + spin_lock_irqsave(&fnic->fnic_lock, flags); + if (fnic->stop_rx_link_events) { + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + goto drop; + } + /* Use fr_flags to indicate whether succ. flogi resp or not */ + fr_flags(fp) = 0; + fr_dev(fp) = fnic->lport; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + + skb_queue_tail(&fnic->frame_queue, skb); + queue_work(fnic_event_queue, &fnic->frame_work); + + return; +drop: + dev_kfree_skb_irq(skb); +} + +static int fnic_rq_cmpl_handler_cont(struct vnic_dev *vdev, + struct cq_desc *cq_desc, u8 type, + u16 q_number, u16 completed_index, + void *opaque) +{ + struct fnic *fnic = vnic_dev_priv(vdev); + + vnic_rq_service(&fnic->rq[q_number], cq_desc, completed_index, + VNIC_RQ_RETURN_DESC, fnic_rq_cmpl_frame_recv, + NULL); + return 0; +} + +int fnic_rq_cmpl_handler(struct fnic *fnic, int rq_work_to_do) +{ + unsigned int tot_rq_work_done = 0, cur_work_done; + unsigned int i; + int err; + + for (i = 0; i < fnic->rq_count; i++) { + cur_work_done = vnic_cq_service(&fnic->cq[i], rq_work_to_do, + fnic_rq_cmpl_handler_cont, + NULL); + if (cur_work_done) { + err = vnic_rq_fill(&fnic->rq[i], fnic_alloc_rq_frame); + if (err) + shost_printk(KERN_ERR, fnic->lport->host, + "fnic_alloc_rq_frame cant alloc" + " frame\n"); + } + tot_rq_work_done += cur_work_done; + } + + return tot_rq_work_done; +} + +/* + * This function is called once at init time to allocate and fill RQ + * buffers. Subsequently, it is called in the interrupt context after RQ + * buffer processing to replenish the buffers in the RQ + */ +int fnic_alloc_rq_frame(struct vnic_rq *rq) +{ + struct fnic *fnic = vnic_dev_priv(rq->vdev); + struct sk_buff *skb; + u16 len; + dma_addr_t pa; + + len = FC_FRAME_HEADROOM + FC_MAX_FRAME + FC_FRAME_TAILROOM; + skb = dev_alloc_skb(len); + if (!skb) { + FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host, + "Unable to allocate RQ sk_buff\n"); + return -ENOMEM; + } + skb_reset_mac_header(skb); + skb_reset_transport_header(skb); + skb_reset_network_header(skb); + skb_put(skb, len); + pa = pci_map_single(fnic->pdev, skb->data, len, PCI_DMA_FROMDEVICE); + fnic_queue_rq_desc(rq, skb, pa, len); + return 0; +} + +void fnic_free_rq_buf(struct vnic_rq *rq, struct vnic_rq_buf *buf) +{ + struct fc_frame *fp = buf->os_buf; + struct fnic *fnic = vnic_dev_priv(rq->vdev); + + pci_unmap_single(fnic->pdev, buf->dma_addr, buf->len, + PCI_DMA_FROMDEVICE); + + dev_kfree_skb(fp_skb(fp)); + buf->os_buf = NULL; +} + +static inline int is_flogi_frame(struct fc_frame_header *fh) +{ + return fh->fh_r_ctl == FC_RCTL_ELS_REQ && *(u8 *)(fh + 1) == ELS_FLOGI; +} + +int fnic_send_frame(struct fnic *fnic, struct fc_frame *fp) +{ + struct vnic_wq *wq = &fnic->wq[0]; + struct sk_buff *skb; + dma_addr_t pa; + struct ethhdr *eth_hdr; + struct vlan_ethhdr *vlan_hdr; + struct fcoe_hdr *fcoe_hdr; + struct fc_frame_header *fh; + u32 tot_len, eth_hdr_len; + int ret = 0; + unsigned long flags; + + fh = fc_frame_header_get(fp); + skb = fp_skb(fp); + + if (!fnic->vlan_hw_insert) { + eth_hdr_len = sizeof(*vlan_hdr) + sizeof(*fcoe_hdr); + vlan_hdr = (struct vlan_ethhdr *)skb_push(skb, eth_hdr_len); + eth_hdr = (struct ethhdr *)vlan_hdr; + vlan_hdr->h_vlan_proto = htons(ETH_P_8021Q); + vlan_hdr->h_vlan_encapsulated_proto = htons(ETH_P_FCOE); + vlan_hdr->h_vlan_TCI = htons(fnic->vlan_id); + fcoe_hdr = (struct fcoe_hdr *)(vlan_hdr + 1); + } else { + eth_hdr_len = sizeof(*eth_hdr) + sizeof(*fcoe_hdr); + eth_hdr = (struct ethhdr *)skb_push(skb, eth_hdr_len); + eth_hdr->h_proto = htons(ETH_P_FCOE); + fcoe_hdr = (struct fcoe_hdr *)(eth_hdr + 1); + } + + if (is_flogi_frame(fh)) { + fc_fcoe_set_mac(eth_hdr->h_dest, fh->fh_d_id); + memcpy(eth_hdr->h_source, fnic->mac_addr, ETH_ALEN); + } else { + if (fnic->fcoui_mode) + fc_fcoe_set_mac(eth_hdr->h_dest, fh->fh_d_id); + else + memcpy(eth_hdr->h_dest, fnic->dest_addr, ETH_ALEN); + memcpy(eth_hdr->h_source, fnic->data_src_addr, ETH_ALEN); + } + + tot_len = skb->len; + BUG_ON(tot_len % 4); + + memset(fcoe_hdr, 0, sizeof(*fcoe_hdr)); + fcoe_hdr->fcoe_sof = fr_sof(fp); + if (FC_FCOE_VER) + FC_FCOE_ENCAPS_VER(fcoe_hdr, FC_FCOE_VER); + + pa = pci_map_single(fnic->pdev, eth_hdr, tot_len, PCI_DMA_TODEVICE); + + spin_lock_irqsave(&fnic->wq_lock[0], flags); + + if (!vnic_wq_desc_avail(wq)) { + pci_unmap_single(fnic->pdev, pa, + tot_len, PCI_DMA_TODEVICE); + ret = -1; + goto fnic_send_frame_end; + } + + fnic_queue_wq_desc(wq, skb, pa, tot_len, fr_eof(fp), + fnic->vlan_hw_insert, fnic->vlan_id, 1, 1, 1); +fnic_send_frame_end: + spin_unlock_irqrestore(&fnic->wq_lock[0], flags); + + if (ret) + dev_kfree_skb_any(fp_skb(fp)); + + return ret; +} + +/* + * fnic_send + * Routine to send a raw frame + */ +int fnic_send(struct fc_lport *lp, struct fc_frame *fp) +{ + struct fnic *fnic = lport_priv(lp); + struct fc_frame_header *fh; + int ret = 0; + enum fnic_state old_state; + unsigned long flags; + struct fc_frame *old_flogi = NULL; + struct fc_frame *old_flogi_resp = NULL; + + if (fnic->in_remove) { + dev_kfree_skb(fp_skb(fp)); + ret = -1; + goto fnic_send_end; + } + + fh = fc_frame_header_get(fp); + /* if not an Flogi frame, send it out, this is the common case */ + if (!is_flogi_frame(fh)) + return fnic_send_frame(fnic, fp); + + /* Flogi frame, now enter the state machine */ + + spin_lock_irqsave(&fnic->fnic_lock, flags); +again: + /* Get any old cached frames, free them after dropping lock */ + old_flogi = fnic->flogi; + fnic->flogi = NULL; + old_flogi_resp = fnic->flogi_resp; + fnic->flogi_resp = NULL; + + fnic->flogi_oxid = FC_XID_UNKNOWN; + + old_state = fnic->state; + switch (old_state) { + case FNIC_IN_FC_MODE: + case FNIC_IN_ETH_TRANS_FC_MODE: + default: + fnic->state = FNIC_IN_FC_TRANS_ETH_MODE; + vnic_dev_del_addr(fnic->vdev, fnic->data_src_addr); + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + + if (old_flogi) { + dev_kfree_skb(fp_skb(old_flogi)); + old_flogi = NULL; + } + if (old_flogi_resp) { + dev_kfree_skb(fp_skb(old_flogi_resp)); + old_flogi_resp = NULL; + } + + ret = fnic_fw_reset_handler(fnic); + + spin_lock_irqsave(&fnic->fnic_lock, flags); + if (fnic->state != FNIC_IN_FC_TRANS_ETH_MODE) + goto again; + if (ret) { + fnic->state = old_state; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + dev_kfree_skb(fp_skb(fp)); + goto fnic_send_end; + } + old_flogi = fnic->flogi; + fnic->flogi = fp; + fnic->flogi_oxid = ntohs(fh->fh_ox_id); + old_flogi_resp = fnic->flogi_resp; + fnic->flogi_resp = NULL; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + break; + + case FNIC_IN_FC_TRANS_ETH_MODE: + /* + * A reset is pending with the firmware. Store the flogi + * and its oxid. The transition out of this state happens + * only when Firmware completes the reset, either with + * success or failed. If success, transition to + * FNIC_IN_ETH_MODE, if fail, then transition to + * FNIC_IN_FC_MODE + */ + fnic->flogi = fp; + fnic->flogi_oxid = ntohs(fh->fh_ox_id); + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + break; + + case FNIC_IN_ETH_MODE: + /* + * The fw/hw is already in eth mode. Store the oxid, + * and send the flogi frame out. The transition out of this + * state happens only we receive flogi response from the + * network, and the oxid matches the cached oxid when the + * flogi frame was sent out. If they match, then we issue + * a flogi_reg request and transition to state + * FNIC_IN_ETH_TRANS_FC_MODE + */ + fnic->flogi_oxid = ntohs(fh->fh_ox_id); + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + ret = fnic_send_frame(fnic, fp); + break; + } + +fnic_send_end: + if (old_flogi) + dev_kfree_skb(fp_skb(old_flogi)); + if (old_flogi_resp) + dev_kfree_skb(fp_skb(old_flogi_resp)); + return ret; +} + +static void fnic_wq_complete_frame_send(struct vnic_wq *wq, + struct cq_desc *cq_desc, + struct vnic_wq_buf *buf, void *opaque) +{ + struct sk_buff *skb = buf->os_buf; + struct fc_frame *fp = (struct fc_frame *)skb; + struct fnic *fnic = vnic_dev_priv(wq->vdev); + + pci_unmap_single(fnic->pdev, buf->dma_addr, + buf->len, PCI_DMA_TODEVICE); + dev_kfree_skb_irq(fp_skb(fp)); + buf->os_buf = NULL; +} + +static int fnic_wq_cmpl_handler_cont(struct vnic_dev *vdev, + struct cq_desc *cq_desc, u8 type, + u16 q_number, u16 completed_index, + void *opaque) +{ + struct fnic *fnic = vnic_dev_priv(vdev); + unsigned long flags; + + spin_lock_irqsave(&fnic->wq_lock[q_number], flags); + vnic_wq_service(&fnic->wq[q_number], cq_desc, completed_index, + fnic_wq_complete_frame_send, NULL); + spin_unlock_irqrestore(&fnic->wq_lock[q_number], flags); + + return 0; +} + +int fnic_wq_cmpl_handler(struct fnic *fnic, int work_to_do) +{ + unsigned int wq_work_done = 0; + unsigned int i; + + for (i = 0; i < fnic->raw_wq_count; i++) { + wq_work_done += vnic_cq_service(&fnic->cq[fnic->rq_count+i], + work_to_do, + fnic_wq_cmpl_handler_cont, + NULL); + } + + return wq_work_done; +} + + +void fnic_free_wq_buf(struct vnic_wq *wq, struct vnic_wq_buf *buf) +{ + struct fc_frame *fp = buf->os_buf; + struct fnic *fnic = vnic_dev_priv(wq->vdev); + + pci_unmap_single(fnic->pdev, buf->dma_addr, + buf->len, PCI_DMA_TODEVICE); + + dev_kfree_skb(fp_skb(fp)); + buf->os_buf = NULL; +} diff --git a/drivers/scsi/fnic/fnic_io.h b/drivers/scsi/fnic/fnic_io.h new file mode 100644 index 000000000000..f0b896988cd5 --- /dev/null +++ b/drivers/scsi/fnic/fnic_io.h @@ -0,0 +1,67 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _FNIC_IO_H_ +#define _FNIC_IO_H_ + +#include + +#define FNIC_DFLT_SG_DESC_CNT 32 +#define FNIC_MAX_SG_DESC_CNT 1024 /* Maximum descriptors per sgl */ +#define FNIC_SG_DESC_ALIGN 16 /* Descriptor address alignment */ + +struct host_sg_desc { + __le64 addr; + __le32 len; + u32 _resvd; +}; + +struct fnic_dflt_sgl_list { + struct host_sg_desc sg_desc[FNIC_DFLT_SG_DESC_CNT]; +}; + +struct fnic_sgl_list { + struct host_sg_desc sg_desc[FNIC_MAX_SG_DESC_CNT]; +}; + +enum fnic_sgl_list_type { + FNIC_SGL_CACHE_DFLT = 0, /* cache with default size sgl */ + FNIC_SGL_CACHE_MAX, /* cache with max size sgl */ + FNIC_SGL_NUM_CACHES /* number of sgl caches */ +}; + +enum fnic_ioreq_state { + FNIC_IOREQ_CMD_PENDING = 0, + FNIC_IOREQ_ABTS_PENDING, + FNIC_IOREQ_ABTS_COMPLETE, + FNIC_IOREQ_CMD_COMPLETE, +}; + +struct fnic_io_req { + struct host_sg_desc *sgl_list; /* sgl list */ + void *sgl_list_alloc; /* sgl list address used for free */ + dma_addr_t sense_buf_pa; /* dma address for sense buffer*/ + dma_addr_t sgl_list_pa; /* dma address for sgl list */ + u16 sgl_cnt; + u8 sgl_type; /* device DMA descriptor list type */ + u8 io_completed:1; /* set to 1 when fw completes IO */ + u32 port_id; /* remote port DID */ + struct completion *abts_done; /* completion for abts */ + struct completion *dr_done; /* completion for device reset */ +}; + +#endif /* _FNIC_IO_H_ */ diff --git a/drivers/scsi/fnic/fnic_isr.c b/drivers/scsi/fnic/fnic_isr.c new file mode 100644 index 000000000000..2b3064828aea --- /dev/null +++ b/drivers/scsi/fnic/fnic_isr.c @@ -0,0 +1,332 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#include +#include +#include +#include +#include +#include +#include "vnic_dev.h" +#include "vnic_intr.h" +#include "vnic_stats.h" +#include "fnic_io.h" +#include "fnic.h" + +static irqreturn_t fnic_isr_legacy(int irq, void *data) +{ + struct fnic *fnic = data; + u32 pba; + unsigned long work_done = 0; + + pba = vnic_intr_legacy_pba(fnic->legacy_pba); + if (!pba) + return IRQ_NONE; + + if (pba & (1 << FNIC_INTX_NOTIFY)) { + vnic_intr_return_all_credits(&fnic->intr[FNIC_INTX_NOTIFY]); + fnic_handle_link_event(fnic); + } + + if (pba & (1 << FNIC_INTX_ERR)) { + vnic_intr_return_all_credits(&fnic->intr[FNIC_INTX_ERR]); + fnic_log_q_error(fnic); + } + + if (pba & (1 << FNIC_INTX_WQ_RQ_COPYWQ)) { + work_done += fnic_wq_copy_cmpl_handler(fnic, 8); + work_done += fnic_wq_cmpl_handler(fnic, 4); + work_done += fnic_rq_cmpl_handler(fnic, 4); + + vnic_intr_return_credits(&fnic->intr[FNIC_INTX_WQ_RQ_COPYWQ], + work_done, + 1 /* unmask intr */, + 1 /* reset intr timer */); + } + + return IRQ_HANDLED; +} + +static irqreturn_t fnic_isr_msi(int irq, void *data) +{ + struct fnic *fnic = data; + unsigned long work_done = 0; + + work_done += fnic_wq_copy_cmpl_handler(fnic, 8); + work_done += fnic_wq_cmpl_handler(fnic, 4); + work_done += fnic_rq_cmpl_handler(fnic, 4); + + vnic_intr_return_credits(&fnic->intr[0], + work_done, + 1 /* unmask intr */, + 1 /* reset intr timer */); + + return IRQ_HANDLED; +} + +static irqreturn_t fnic_isr_msix_rq(int irq, void *data) +{ + struct fnic *fnic = data; + unsigned long rq_work_done = 0; + + rq_work_done = fnic_rq_cmpl_handler(fnic, 4); + vnic_intr_return_credits(&fnic->intr[FNIC_MSIX_RQ], + rq_work_done, + 1 /* unmask intr */, + 1 /* reset intr timer */); + + return IRQ_HANDLED; +} + +static irqreturn_t fnic_isr_msix_wq(int irq, void *data) +{ + struct fnic *fnic = data; + unsigned long wq_work_done = 0; + + wq_work_done = fnic_wq_cmpl_handler(fnic, 4); + vnic_intr_return_credits(&fnic->intr[FNIC_MSIX_WQ], + wq_work_done, + 1 /* unmask intr */, + 1 /* reset intr timer */); + return IRQ_HANDLED; +} + +static irqreturn_t fnic_isr_msix_wq_copy(int irq, void *data) +{ + struct fnic *fnic = data; + unsigned long wq_copy_work_done = 0; + + wq_copy_work_done = fnic_wq_copy_cmpl_handler(fnic, 8); + vnic_intr_return_credits(&fnic->intr[FNIC_MSIX_WQ_COPY], + wq_copy_work_done, + 1 /* unmask intr */, + 1 /* reset intr timer */); + return IRQ_HANDLED; +} + +static irqreturn_t fnic_isr_msix_err_notify(int irq, void *data) +{ + struct fnic *fnic = data; + + vnic_intr_return_all_credits(&fnic->intr[FNIC_MSIX_ERR_NOTIFY]); + fnic_log_q_error(fnic); + fnic_handle_link_event(fnic); + + return IRQ_HANDLED; +} + +void fnic_free_intr(struct fnic *fnic) +{ + int i; + + switch (vnic_dev_get_intr_mode(fnic->vdev)) { + case VNIC_DEV_INTR_MODE_INTX: + case VNIC_DEV_INTR_MODE_MSI: + free_irq(fnic->pdev->irq, fnic); + break; + + case VNIC_DEV_INTR_MODE_MSIX: + for (i = 0; i < ARRAY_SIZE(fnic->msix); i++) + if (fnic->msix[i].requested) + free_irq(fnic->msix_entry[i].vector, + fnic->msix[i].devid); + break; + + default: + break; + } +} + +int fnic_request_intr(struct fnic *fnic) +{ + int err = 0; + int i; + + switch (vnic_dev_get_intr_mode(fnic->vdev)) { + + case VNIC_DEV_INTR_MODE_INTX: + err = request_irq(fnic->pdev->irq, &fnic_isr_legacy, + IRQF_SHARED, DRV_NAME, fnic); + break; + + case VNIC_DEV_INTR_MODE_MSI: + err = request_irq(fnic->pdev->irq, &fnic_isr_msi, + 0, fnic->name, fnic); + break; + + case VNIC_DEV_INTR_MODE_MSIX: + + sprintf(fnic->msix[FNIC_MSIX_RQ].devname, + "%.11s-fcs-rq", fnic->name); + fnic->msix[FNIC_MSIX_RQ].isr = fnic_isr_msix_rq; + fnic->msix[FNIC_MSIX_RQ].devid = fnic; + + sprintf(fnic->msix[FNIC_MSIX_WQ].devname, + "%.11s-fcs-wq", fnic->name); + fnic->msix[FNIC_MSIX_WQ].isr = fnic_isr_msix_wq; + fnic->msix[FNIC_MSIX_WQ].devid = fnic; + + sprintf(fnic->msix[FNIC_MSIX_WQ_COPY].devname, + "%.11s-scsi-wq", fnic->name); + fnic->msix[FNIC_MSIX_WQ_COPY].isr = fnic_isr_msix_wq_copy; + fnic->msix[FNIC_MSIX_WQ_COPY].devid = fnic; + + sprintf(fnic->msix[FNIC_MSIX_ERR_NOTIFY].devname, + "%.11s-err-notify", fnic->name); + fnic->msix[FNIC_MSIX_ERR_NOTIFY].isr = + fnic_isr_msix_err_notify; + fnic->msix[FNIC_MSIX_ERR_NOTIFY].devid = fnic; + + for (i = 0; i < ARRAY_SIZE(fnic->msix); i++) { + err = request_irq(fnic->msix_entry[i].vector, + fnic->msix[i].isr, 0, + fnic->msix[i].devname, + fnic->msix[i].devid); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "MSIX: request_irq" + " failed %d\n", err); + fnic_free_intr(fnic); + break; + } + fnic->msix[i].requested = 1; + } + break; + + default: + break; + } + + return err; +} + +int fnic_set_intr_mode(struct fnic *fnic) +{ + unsigned int n = ARRAY_SIZE(fnic->rq); + unsigned int m = ARRAY_SIZE(fnic->wq); + unsigned int o = ARRAY_SIZE(fnic->wq_copy); + unsigned int i; + + /* + * Set interrupt mode (INTx, MSI, MSI-X) depending + * system capabilities. + * + * Try MSI-X first + * + * We need n RQs, m WQs, o Copy WQs, n+m+o CQs, and n+m+o+1 INTRs + * (last INTR is used for WQ/RQ errors and notification area) + */ + + BUG_ON(ARRAY_SIZE(fnic->msix_entry) < n + m + o + 1); + for (i = 0; i < n + m + o + 1; i++) + fnic->msix_entry[i].entry = i; + + if (fnic->rq_count >= n && + fnic->raw_wq_count >= m && + fnic->wq_copy_count >= o && + fnic->cq_count >= n + m + o) { + if (!pci_enable_msix(fnic->pdev, fnic->msix_entry, + n + m + o + 1)) { + fnic->rq_count = n; + fnic->raw_wq_count = m; + fnic->wq_copy_count = o; + fnic->wq_count = m + o; + fnic->cq_count = n + m + o; + fnic->intr_count = n + m + o + 1; + fnic->err_intr_offset = FNIC_MSIX_ERR_NOTIFY; + + FNIC_ISR_DBG(KERN_DEBUG, fnic->lport->host, + "Using MSI-X Interrupts\n"); + vnic_dev_set_intr_mode(fnic->vdev, + VNIC_DEV_INTR_MODE_MSIX); + return 0; + } + } + + /* + * Next try MSI + * We need 1 RQ, 1 WQ, 1 WQ_COPY, 3 CQs, and 1 INTR + */ + if (fnic->rq_count >= 1 && + fnic->raw_wq_count >= 1 && + fnic->wq_copy_count >= 1 && + fnic->cq_count >= 3 && + fnic->intr_count >= 1 && + !pci_enable_msi(fnic->pdev)) { + + fnic->rq_count = 1; + fnic->raw_wq_count = 1; + fnic->wq_copy_count = 1; + fnic->wq_count = 2; + fnic->cq_count = 3; + fnic->intr_count = 1; + fnic->err_intr_offset = 0; + + FNIC_ISR_DBG(KERN_DEBUG, fnic->lport->host, + "Using MSI Interrupts\n"); + vnic_dev_set_intr_mode(fnic->vdev, VNIC_DEV_INTR_MODE_MSI); + + return 0; + } + + /* + * Next try INTx + * We need 1 RQ, 1 WQ, 1 WQ_COPY, 3 CQs, and 3 INTRs + * 1 INTR is used for all 3 queues, 1 INTR for queue errors + * 1 INTR for notification area + */ + + if (fnic->rq_count >= 1 && + fnic->raw_wq_count >= 1 && + fnic->wq_copy_count >= 1 && + fnic->cq_count >= 3 && + fnic->intr_count >= 3) { + + fnic->rq_count = 1; + fnic->raw_wq_count = 1; + fnic->wq_copy_count = 1; + fnic->cq_count = 3; + fnic->intr_count = 3; + + FNIC_ISR_DBG(KERN_DEBUG, fnic->lport->host, + "Using Legacy Interrupts\n"); + vnic_dev_set_intr_mode(fnic->vdev, VNIC_DEV_INTR_MODE_INTX); + + return 0; + } + + vnic_dev_set_intr_mode(fnic->vdev, VNIC_DEV_INTR_MODE_UNKNOWN); + + return -EINVAL; +} + +void fnic_clear_intr_mode(struct fnic *fnic) +{ + switch (vnic_dev_get_intr_mode(fnic->vdev)) { + case VNIC_DEV_INTR_MODE_MSIX: + pci_disable_msix(fnic->pdev); + break; + case VNIC_DEV_INTR_MODE_MSI: + pci_disable_msi(fnic->pdev); + break; + default: + break; + } + + vnic_dev_set_intr_mode(fnic->vdev, VNIC_DEV_INTR_MODE_INTX); +} + diff --git a/drivers/scsi/fnic/fnic_main.c b/drivers/scsi/fnic/fnic_main.c new file mode 100644 index 000000000000..32ef6b87d895 --- /dev/null +++ b/drivers/scsi/fnic/fnic_main.c @@ -0,0 +1,942 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "vnic_dev.h" +#include "vnic_intr.h" +#include "vnic_stats.h" +#include "fnic_io.h" +#include "fnic.h" + +#define PCI_DEVICE_ID_CISCO_FNIC 0x0045 + +/* Timer to poll notification area for events. Used for MSI interrupts */ +#define FNIC_NOTIFY_TIMER_PERIOD (2 * HZ) + +static struct kmem_cache *fnic_sgl_cache[FNIC_SGL_NUM_CACHES]; +static struct kmem_cache *fnic_io_req_cache; +LIST_HEAD(fnic_list); +DEFINE_SPINLOCK(fnic_list_lock); + +/* Supported devices by fnic module */ +static struct pci_device_id fnic_id_table[] = { + { PCI_DEVICE(PCI_VENDOR_ID_CISCO, PCI_DEVICE_ID_CISCO_FNIC) }, + { 0, } +}; + +MODULE_DESCRIPTION(DRV_DESCRIPTION); +MODULE_AUTHOR("Abhijeet Joglekar , " + "Joseph R. Eykholt "); +MODULE_LICENSE("GPL v2"); +MODULE_VERSION(DRV_VERSION); +MODULE_DEVICE_TABLE(pci, fnic_id_table); + +unsigned int fnic_log_level; +module_param(fnic_log_level, int, S_IRUGO|S_IWUSR); +MODULE_PARM_DESC(fnic_log_level, "bit mask of fnic logging levels"); + + +static struct libfc_function_template fnic_transport_template = { + .frame_send = fnic_send, + .fcp_abort_io = fnic_empty_scsi_cleanup, + .fcp_cleanup = fnic_empty_scsi_cleanup, + .exch_mgr_reset = fnic_exch_mgr_reset +}; + +static int fnic_slave_alloc(struct scsi_device *sdev) +{ + struct fc_rport *rport = starget_to_rport(scsi_target(sdev)); + struct fc_lport *lp = shost_priv(sdev->host); + struct fnic *fnic = lport_priv(lp); + + sdev->tagged_supported = 1; + + if (!rport || fc_remote_port_chkready(rport)) + return -ENXIO; + + scsi_activate_tcq(sdev, FNIC_DFLT_QUEUE_DEPTH); + rport->dev_loss_tmo = fnic->config.port_down_timeout / 1000; + + return 0; +} + +static struct scsi_host_template fnic_host_template = { + .module = THIS_MODULE, + .name = DRV_NAME, + .queuecommand = fnic_queuecommand, + .eh_abort_handler = fnic_abort_cmd, + .eh_device_reset_handler = fnic_device_reset, + .eh_host_reset_handler = fnic_host_reset, + .slave_alloc = fnic_slave_alloc, + .change_queue_depth = fc_change_queue_depth, + .change_queue_type = fc_change_queue_type, + .this_id = -1, + .cmd_per_lun = 3, + .can_queue = FNIC_MAX_IO_REQ, + .use_clustering = ENABLE_CLUSTERING, + .sg_tablesize = FNIC_MAX_SG_DESC_CNT, + .max_sectors = 0xffff, + .shost_attrs = fnic_attrs, +}; + +static void fnic_get_host_speed(struct Scsi_Host *shost); +static struct scsi_transport_template *fnic_fc_transport; +static struct fc_host_statistics *fnic_get_stats(struct Scsi_Host *); + +static struct fc_function_template fnic_fc_functions = { + + .show_host_node_name = 1, + .show_host_port_name = 1, + .show_host_supported_classes = 1, + .show_host_supported_fc4s = 1, + .show_host_active_fc4s = 1, + .show_host_maxframe_size = 1, + .show_host_port_id = 1, + .show_host_supported_speeds = 1, + .get_host_speed = fnic_get_host_speed, + .show_host_speed = 1, + .show_host_port_type = 1, + .get_host_port_state = fc_get_host_port_state, + .show_host_port_state = 1, + .show_host_symbolic_name = 1, + .show_rport_maxframe_size = 1, + .show_rport_supported_classes = 1, + .show_host_fabric_name = 1, + .show_starget_node_name = 1, + .show_starget_port_name = 1, + .show_starget_port_id = 1, + .show_rport_dev_loss_tmo = 1, + .issue_fc_host_lip = fnic_reset, + .get_fc_host_stats = fnic_get_stats, + .dd_fcrport_size = sizeof(struct fc_rport_libfc_priv), + .terminate_rport_io = fnic_terminate_rport_io, +}; + +static void fnic_get_host_speed(struct Scsi_Host *shost) +{ + struct fc_lport *lp = shost_priv(shost); + struct fnic *fnic = lport_priv(lp); + u32 port_speed = vnic_dev_port_speed(fnic->vdev); + + /* Add in other values as they get defined in fw */ + switch (port_speed) { + case 10000: + fc_host_speed(shost) = FC_PORTSPEED_10GBIT; + break; + default: + fc_host_speed(shost) = FC_PORTSPEED_10GBIT; + break; + } +} + +static struct fc_host_statistics *fnic_get_stats(struct Scsi_Host *host) +{ + int ret; + struct fc_lport *lp = shost_priv(host); + struct fnic *fnic = lport_priv(lp); + struct fc_host_statistics *stats = &lp->host_stats; + struct vnic_stats *vs; + unsigned long flags; + + if (time_before(jiffies, fnic->stats_time + HZ / FNIC_STATS_RATE_LIMIT)) + return stats; + fnic->stats_time = jiffies; + + spin_lock_irqsave(&fnic->fnic_lock, flags); + ret = vnic_dev_stats_dump(fnic->vdev, &fnic->stats); + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + + if (ret) { + FNIC_MAIN_DBG(KERN_DEBUG, fnic->lport->host, + "fnic: Get vnic stats failed" + " 0x%x", ret); + return stats; + } + vs = fnic->stats; + stats->tx_frames = vs->tx.tx_unicast_frames_ok; + stats->tx_words = vs->tx.tx_unicast_bytes_ok / 4; + stats->rx_frames = vs->rx.rx_unicast_frames_ok; + stats->rx_words = vs->rx.rx_unicast_bytes_ok / 4; + stats->error_frames = vs->tx.tx_errors + vs->rx.rx_errors; + stats->dumped_frames = vs->tx.tx_drops + vs->rx.rx_drop; + stats->invalid_crc_count = vs->rx.rx_crc_errors; + stats->seconds_since_last_reset = (jiffies - lp->boot_time) / HZ; + stats->fcp_input_megabytes = div_u64(fnic->fcp_input_bytes, 1000000); + stats->fcp_output_megabytes = div_u64(fnic->fcp_output_bytes, 1000000); + + return stats; +} + +void fnic_log_q_error(struct fnic *fnic) +{ + unsigned int i; + u32 error_status; + + for (i = 0; i < fnic->raw_wq_count; i++) { + error_status = ioread32(&fnic->wq[i].ctrl->error_status); + if (error_status) + shost_printk(KERN_ERR, fnic->lport->host, + "WQ[%d] error_status" + " %d\n", i, error_status); + } + + for (i = 0; i < fnic->rq_count; i++) { + error_status = ioread32(&fnic->rq[i].ctrl->error_status); + if (error_status) + shost_printk(KERN_ERR, fnic->lport->host, + "RQ[%d] error_status" + " %d\n", i, error_status); + } + + for (i = 0; i < fnic->wq_copy_count; i++) { + error_status = ioread32(&fnic->wq_copy[i].ctrl->error_status); + if (error_status) + shost_printk(KERN_ERR, fnic->lport->host, + "CWQ[%d] error_status" + " %d\n", i, error_status); + } +} + +void fnic_handle_link_event(struct fnic *fnic) +{ + unsigned long flags; + + spin_lock_irqsave(&fnic->fnic_lock, flags); + if (fnic->stop_rx_link_events) { + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + return; + } + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + + queue_work(fnic_event_queue, &fnic->link_work); + +} + +static int fnic_notify_set(struct fnic *fnic) +{ + int err; + + switch (vnic_dev_get_intr_mode(fnic->vdev)) { + case VNIC_DEV_INTR_MODE_INTX: + err = vnic_dev_notify_set(fnic->vdev, FNIC_INTX_NOTIFY); + break; + case VNIC_DEV_INTR_MODE_MSI: + err = vnic_dev_notify_set(fnic->vdev, -1); + break; + case VNIC_DEV_INTR_MODE_MSIX: + err = vnic_dev_notify_set(fnic->vdev, FNIC_MSIX_ERR_NOTIFY); + break; + default: + shost_printk(KERN_ERR, fnic->lport->host, + "Interrupt mode should be set up" + " before devcmd notify set %d\n", + vnic_dev_get_intr_mode(fnic->vdev)); + err = -1; + break; + } + + return err; +} + +static void fnic_notify_timer(unsigned long data) +{ + struct fnic *fnic = (struct fnic *)data; + + fnic_handle_link_event(fnic); + mod_timer(&fnic->notify_timer, + round_jiffies(jiffies + FNIC_NOTIFY_TIMER_PERIOD)); +} + +static void fnic_notify_timer_start(struct fnic *fnic) +{ + switch (vnic_dev_get_intr_mode(fnic->vdev)) { + case VNIC_DEV_INTR_MODE_MSI: + /* + * Schedule first timeout immediately. The driver is + * initiatialized and ready to look for link up notification + */ + mod_timer(&fnic->notify_timer, jiffies); + break; + default: + /* Using intr for notification for INTx/MSI-X */ + break; + }; +} + +static int fnic_dev_wait(struct vnic_dev *vdev, + int (*start)(struct vnic_dev *, int), + int (*finished)(struct vnic_dev *, int *), + int arg) +{ + unsigned long time; + int done; + int err; + + err = start(vdev, arg); + if (err) + return err; + + /* Wait for func to complete...2 seconds max */ + time = jiffies + (HZ * 2); + do { + err = finished(vdev, &done); + if (err) + return err; + if (done) + return 0; + schedule_timeout_uninterruptible(HZ / 10); + } while (time_after(time, jiffies)); + + return -ETIMEDOUT; +} + +static int fnic_cleanup(struct fnic *fnic) +{ + unsigned int i; + int err; + unsigned long flags; + struct fc_frame *flogi = NULL; + struct fc_frame *flogi_resp = NULL; + + vnic_dev_disable(fnic->vdev); + for (i = 0; i < fnic->intr_count; i++) + vnic_intr_mask(&fnic->intr[i]); + + for (i = 0; i < fnic->rq_count; i++) { + err = vnic_rq_disable(&fnic->rq[i]); + if (err) + return err; + } + for (i = 0; i < fnic->raw_wq_count; i++) { + err = vnic_wq_disable(&fnic->wq[i]); + if (err) + return err; + } + for (i = 0; i < fnic->wq_copy_count; i++) { + err = vnic_wq_copy_disable(&fnic->wq_copy[i]); + if (err) + return err; + } + + /* Clean up completed IOs and FCS frames */ + fnic_wq_copy_cmpl_handler(fnic, -1); + fnic_wq_cmpl_handler(fnic, -1); + fnic_rq_cmpl_handler(fnic, -1); + + /* Clean up the IOs and FCS frames that have not completed */ + for (i = 0; i < fnic->raw_wq_count; i++) + vnic_wq_clean(&fnic->wq[i], fnic_free_wq_buf); + for (i = 0; i < fnic->rq_count; i++) + vnic_rq_clean(&fnic->rq[i], fnic_free_rq_buf); + for (i = 0; i < fnic->wq_copy_count; i++) + vnic_wq_copy_clean(&fnic->wq_copy[i], + fnic_wq_copy_cleanup_handler); + + for (i = 0; i < fnic->cq_count; i++) + vnic_cq_clean(&fnic->cq[i]); + for (i = 0; i < fnic->intr_count; i++) + vnic_intr_clean(&fnic->intr[i]); + + /* + * Remove cached flogi and flogi resp frames if any + * These frames are not in any queue, and therefore queue + * cleanup does not clean them. So clean them explicitly + */ + spin_lock_irqsave(&fnic->fnic_lock, flags); + flogi = fnic->flogi; + fnic->flogi = NULL; + flogi_resp = fnic->flogi_resp; + fnic->flogi_resp = NULL; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + + if (flogi) + dev_kfree_skb(fp_skb(flogi)); + + if (flogi_resp) + dev_kfree_skb(fp_skb(flogi_resp)); + + mempool_destroy(fnic->io_req_pool); + for (i = 0; i < FNIC_SGL_NUM_CACHES; i++) + mempool_destroy(fnic->io_sgl_pool[i]); + + return 0; +} + +static void fnic_iounmap(struct fnic *fnic) +{ + if (fnic->bar0.vaddr) + iounmap(fnic->bar0.vaddr); +} + +/* + * Allocate element for mempools requiring GFP_DMA flag. + * Otherwise, checks in kmem_flagcheck() hit BUG_ON(). + */ +static void *fnic_alloc_slab_dma(gfp_t gfp_mask, void *pool_data) +{ + struct kmem_cache *mem = pool_data; + + return kmem_cache_alloc(mem, gfp_mask | GFP_ATOMIC | GFP_DMA); +} + +static int __devinit fnic_probe(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + struct Scsi_Host *host; + struct fc_lport *lp; + struct fnic *fnic; + mempool_t *pool; + int err; + int i; + unsigned long flags; + + /* + * Allocate SCSI Host and set up association between host, + * local port, and fnic + */ + host = scsi_host_alloc(&fnic_host_template, + sizeof(struct fc_lport) + sizeof(struct fnic)); + if (!host) { + printk(KERN_ERR PFX "Unable to alloc SCSI host\n"); + err = -ENOMEM; + goto err_out; + } + lp = shost_priv(host); + lp->host = host; + fnic = lport_priv(lp); + fnic->lport = lp; + + snprintf(fnic->name, sizeof(fnic->name) - 1, "%s%d", DRV_NAME, + host->host_no); + + host->transportt = fnic_fc_transport; + + err = scsi_init_shared_tag_map(host, FNIC_MAX_IO_REQ); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "Unable to alloc shared tag map\n"); + goto err_out_free_hba; + } + + /* Setup PCI resources */ + pci_set_drvdata(pdev, fnic); + + fnic->pdev = pdev; + + err = pci_enable_device(pdev); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "Cannot enable PCI device, aborting.\n"); + goto err_out_free_hba; + } + + err = pci_request_regions(pdev, DRV_NAME); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "Cannot enable PCI resources, aborting\n"); + goto err_out_disable_device; + } + + pci_set_master(pdev); + + /* Query PCI controller on system for DMA addressing + * limitation for the device. Try 40-bit first, and + * fail to 32-bit. + */ + err = pci_set_dma_mask(pdev, DMA_40BIT_MASK); + if (err) { + err = pci_set_dma_mask(pdev, DMA_32BIT_MASK); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "No usable DMA configuration " + "aborting\n"); + goto err_out_release_regions; + } + err = pci_set_consistent_dma_mask(pdev, DMA_32BIT_MASK); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "Unable to obtain 32-bit DMA " + "for consistent allocations, aborting.\n"); + goto err_out_release_regions; + } + } else { + err = pci_set_consistent_dma_mask(pdev, DMA_40BIT_MASK); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "Unable to obtain 40-bit DMA " + "for consistent allocations, aborting.\n"); + goto err_out_release_regions; + } + } + + /* Map vNIC resources from BAR0 */ + if (!(pci_resource_flags(pdev, 0) & IORESOURCE_MEM)) { + shost_printk(KERN_ERR, fnic->lport->host, + "BAR0 not memory-map'able, aborting.\n"); + err = -ENODEV; + goto err_out_release_regions; + } + + fnic->bar0.vaddr = pci_iomap(pdev, 0, 0); + fnic->bar0.bus_addr = pci_resource_start(pdev, 0); + fnic->bar0.len = pci_resource_len(pdev, 0); + + if (!fnic->bar0.vaddr) { + shost_printk(KERN_ERR, fnic->lport->host, + "Cannot memory-map BAR0 res hdr, " + "aborting.\n"); + err = -ENODEV; + goto err_out_release_regions; + } + + fnic->vdev = vnic_dev_register(NULL, fnic, pdev, &fnic->bar0); + if (!fnic->vdev) { + shost_printk(KERN_ERR, fnic->lport->host, + "vNIC registration failed, " + "aborting.\n"); + err = -ENODEV; + goto err_out_iounmap; + } + + err = fnic_dev_wait(fnic->vdev, vnic_dev_open, + vnic_dev_open_done, 0); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "vNIC dev open failed, aborting.\n"); + goto err_out_vnic_unregister; + } + + err = vnic_dev_init(fnic->vdev, 0); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "vNIC dev init failed, aborting.\n"); + goto err_out_dev_close; + } + + err = vnic_dev_mac_addr(fnic->vdev, fnic->mac_addr); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "vNIC get MAC addr failed \n"); + goto err_out_dev_close; + } + + /* Get vNIC configuration */ + err = fnic_get_vnic_config(fnic); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "Get vNIC configuration failed, " + "aborting.\n"); + goto err_out_dev_close; + } + host->max_lun = fnic->config.luns_per_tgt; + host->max_id = FNIC_MAX_FCP_TARGET; + + fnic_get_res_counts(fnic); + + err = fnic_set_intr_mode(fnic); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "Failed to set intr mode, " + "aborting.\n"); + goto err_out_dev_close; + } + + err = fnic_request_intr(fnic); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "Unable to request irq.\n"); + goto err_out_clear_intr; + } + + err = fnic_alloc_vnic_resources(fnic); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "Failed to alloc vNIC resources, " + "aborting.\n"); + goto err_out_free_intr; + } + + + /* initialize all fnic locks */ + spin_lock_init(&fnic->fnic_lock); + + for (i = 0; i < FNIC_WQ_MAX; i++) + spin_lock_init(&fnic->wq_lock[i]); + + for (i = 0; i < FNIC_WQ_COPY_MAX; i++) { + spin_lock_init(&fnic->wq_copy_lock[i]); + fnic->wq_copy_desc_low[i] = DESC_CLEAN_LOW_WATERMARK; + fnic->fw_ack_recd[i] = 0; + fnic->fw_ack_index[i] = -1; + } + + for (i = 0; i < FNIC_IO_LOCKS; i++) + spin_lock_init(&fnic->io_req_lock[i]); + + fnic->io_req_pool = mempool_create_slab_pool(2, fnic_io_req_cache); + if (!fnic->io_req_pool) + goto err_out_free_resources; + + pool = mempool_create(2, fnic_alloc_slab_dma, mempool_free_slab, + fnic_sgl_cache[FNIC_SGL_CACHE_DFLT]); + if (!pool) + goto err_out_free_ioreq_pool; + fnic->io_sgl_pool[FNIC_SGL_CACHE_DFLT] = pool; + + pool = mempool_create(2, fnic_alloc_slab_dma, mempool_free_slab, + fnic_sgl_cache[FNIC_SGL_CACHE_MAX]); + if (!pool) + goto err_out_free_dflt_pool; + fnic->io_sgl_pool[FNIC_SGL_CACHE_MAX] = pool; + + /* setup vlan config, hw inserts vlan header */ + fnic->vlan_hw_insert = 1; + fnic->vlan_id = 0; + + fnic->flogi_oxid = FC_XID_UNKNOWN; + fnic->flogi = NULL; + fnic->flogi_resp = NULL; + fnic->state = FNIC_IN_FC_MODE; + + /* Enable hardware stripping of vlan header on ingress */ + fnic_set_nic_config(fnic, 0, 0, 0, 0, 0, 0, 1); + + /* Setup notification buffer area */ + err = fnic_notify_set(fnic); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "Failed to alloc notify buffer, aborting.\n"); + goto err_out_free_max_pool; + } + + /* Setup notify timer when using MSI interrupts */ + if (vnic_dev_get_intr_mode(fnic->vdev) == VNIC_DEV_INTR_MODE_MSI) + setup_timer(&fnic->notify_timer, + fnic_notify_timer, (unsigned long)fnic); + + /* allocate RQ buffers and post them to RQ*/ + for (i = 0; i < fnic->rq_count; i++) { + err = vnic_rq_fill(&fnic->rq[i], fnic_alloc_rq_frame); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "fnic_alloc_rq_frame can't alloc " + "frame\n"); + goto err_out_free_rq_buf; + } + } + + /* + * Initialization done with PCI system, hardware, firmware. + * Add host to SCSI + */ + err = scsi_add_host(lp->host, &pdev->dev); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "fnic: scsi_add_host failed...exiting\n"); + goto err_out_free_rq_buf; + } + + /* Start local port initiatialization */ + + lp->link_up = 0; + lp->tt = fnic_transport_template; + + lp->emp = fc_exch_mgr_alloc(lp, FC_CLASS_3, + FCPIO_HOST_EXCH_RANGE_START, + FCPIO_HOST_EXCH_RANGE_END); + if (!lp->emp) { + err = -ENOMEM; + goto err_out_remove_scsi_host; + } + + lp->max_retry_count = fnic->config.flogi_retries; + lp->service_params = (FCP_SPPF_INIT_FCN | FCP_SPPF_RD_XRDY_DIS | + FCP_SPPF_CONF_COMPL); + if (fnic->config.flags & VFCF_FCP_SEQ_LVL_ERR) + lp->service_params |= FCP_SPPF_RETRY; + + lp->boot_time = jiffies; + lp->e_d_tov = fnic->config.ed_tov; + lp->r_a_tov = fnic->config.ra_tov; + lp->link_supported_speeds = FC_PORTSPEED_10GBIT; + fc_set_wwnn(lp, fnic->config.node_wwn); + fc_set_wwpn(lp, fnic->config.port_wwn); + + fc_exch_init(lp); + fc_lport_init(lp); + fc_elsct_init(lp); + fc_rport_init(lp); + fc_disc_init(lp); + + fc_lport_config(lp); + + if (fc_set_mfs(lp, fnic->config.maxdatafieldsize + + sizeof(struct fc_frame_header))) { + err = -EINVAL; + goto err_out_free_exch_mgr; + } + fc_host_maxframe_size(lp->host) = lp->mfs; + + sprintf(fc_host_symbolic_name(lp->host), + DRV_NAME " v" DRV_VERSION " over %s", fnic->name); + + spin_lock_irqsave(&fnic_list_lock, flags); + list_add_tail(&fnic->list, &fnic_list); + spin_unlock_irqrestore(&fnic_list_lock, flags); + + INIT_WORK(&fnic->link_work, fnic_handle_link); + INIT_WORK(&fnic->frame_work, fnic_handle_frame); + skb_queue_head_init(&fnic->frame_queue); + + /* Enable all queues */ + for (i = 0; i < fnic->raw_wq_count; i++) + vnic_wq_enable(&fnic->wq[i]); + for (i = 0; i < fnic->rq_count; i++) + vnic_rq_enable(&fnic->rq[i]); + for (i = 0; i < fnic->wq_copy_count; i++) + vnic_wq_copy_enable(&fnic->wq_copy[i]); + + fc_fabric_login(lp); + + vnic_dev_enable(fnic->vdev); + for (i = 0; i < fnic->intr_count; i++) + vnic_intr_unmask(&fnic->intr[i]); + + fnic_notify_timer_start(fnic); + + return 0; + +err_out_free_exch_mgr: + fc_exch_mgr_free(lp->emp); +err_out_remove_scsi_host: + fc_remove_host(fnic->lport->host); + scsi_remove_host(fnic->lport->host); +err_out_free_rq_buf: + for (i = 0; i < fnic->rq_count; i++) + vnic_rq_clean(&fnic->rq[i], fnic_free_rq_buf); + vnic_dev_notify_unset(fnic->vdev); +err_out_free_max_pool: + mempool_destroy(fnic->io_sgl_pool[FNIC_SGL_CACHE_MAX]); +err_out_free_dflt_pool: + mempool_destroy(fnic->io_sgl_pool[FNIC_SGL_CACHE_DFLT]); +err_out_free_ioreq_pool: + mempool_destroy(fnic->io_req_pool); +err_out_free_resources: + fnic_free_vnic_resources(fnic); +err_out_free_intr: + fnic_free_intr(fnic); +err_out_clear_intr: + fnic_clear_intr_mode(fnic); +err_out_dev_close: + vnic_dev_close(fnic->vdev); +err_out_vnic_unregister: + vnic_dev_unregister(fnic->vdev); +err_out_iounmap: + fnic_iounmap(fnic); +err_out_release_regions: + pci_release_regions(pdev); +err_out_disable_device: + pci_disable_device(pdev); +err_out_free_hba: + scsi_host_put(lp->host); +err_out: + return err; +} + +static void __devexit fnic_remove(struct pci_dev *pdev) +{ + struct fnic *fnic = pci_get_drvdata(pdev); + unsigned long flags; + + /* + * Mark state so that the workqueue thread stops forwarding + * received frames and link events to the local port. ISR and + * other threads that can queue work items will also stop + * creating work items on the fnic workqueue + */ + spin_lock_irqsave(&fnic->fnic_lock, flags); + fnic->stop_rx_link_events = 1; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + + if (vnic_dev_get_intr_mode(fnic->vdev) == VNIC_DEV_INTR_MODE_MSI) + del_timer_sync(&fnic->notify_timer); + + /* + * Flush the fnic event queue. After this call, there should + * be no event queued for this fnic device in the workqueue + */ + flush_workqueue(fnic_event_queue); + skb_queue_purge(&fnic->frame_queue); + + /* + * Log off the fabric. This stops all remote ports, dns port, + * logs off the fabric. This flushes all rport, disc, lport work + * before returning + */ + fc_fabric_logoff(fnic->lport); + + spin_lock_irqsave(&fnic->fnic_lock, flags); + fnic->in_remove = 1; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + + fc_lport_destroy(fnic->lport); + + /* + * This stops the fnic device, masks all interrupts. Completed + * CQ entries are drained. Posted WQ/RQ/Copy-WQ entries are + * cleaned up + */ + fnic_cleanup(fnic); + + BUG_ON(!skb_queue_empty(&fnic->frame_queue)); + + spin_lock_irqsave(&fnic_list_lock, flags); + list_del(&fnic->list); + spin_unlock_irqrestore(&fnic_list_lock, flags); + + fc_remove_host(fnic->lport->host); + scsi_remove_host(fnic->lport->host); + fc_exch_mgr_free(fnic->lport->emp); + vnic_dev_notify_unset(fnic->vdev); + fnic_free_vnic_resources(fnic); + fnic_free_intr(fnic); + fnic_clear_intr_mode(fnic); + vnic_dev_close(fnic->vdev); + vnic_dev_unregister(fnic->vdev); + fnic_iounmap(fnic); + pci_release_regions(pdev); + pci_disable_device(pdev); + pci_set_drvdata(pdev, NULL); + scsi_host_put(fnic->lport->host); +} + +static struct pci_driver fnic_driver = { + .name = DRV_NAME, + .id_table = fnic_id_table, + .probe = fnic_probe, + .remove = __devexit_p(fnic_remove), +}; + +static int __init fnic_init_module(void) +{ + size_t len; + int err = 0; + + printk(KERN_INFO PFX "%s, ver %s\n", DRV_DESCRIPTION, DRV_VERSION); + + /* Create a cache for allocation of default size sgls */ + len = sizeof(struct fnic_dflt_sgl_list); + fnic_sgl_cache[FNIC_SGL_CACHE_DFLT] = kmem_cache_create + ("fnic_sgl_dflt", len + FNIC_SG_DESC_ALIGN, FNIC_SG_DESC_ALIGN, + SLAB_HWCACHE_ALIGN | SLAB_CACHE_DMA, + NULL); + if (!fnic_sgl_cache[FNIC_SGL_CACHE_DFLT]) { + printk(KERN_ERR PFX "failed to create fnic dflt sgl slab\n"); + err = -ENOMEM; + goto err_create_fnic_sgl_slab_dflt; + } + + /* Create a cache for allocation of max size sgls*/ + len = sizeof(struct fnic_sgl_list); + fnic_sgl_cache[FNIC_SGL_CACHE_MAX] = kmem_cache_create + ("fnic_sgl_max", len + FNIC_SG_DESC_ALIGN, FNIC_SG_DESC_ALIGN, + SLAB_HWCACHE_ALIGN | SLAB_CACHE_DMA, + NULL); + if (!fnic_sgl_cache[FNIC_SGL_CACHE_MAX]) { + printk(KERN_ERR PFX "failed to create fnic max sgl slab\n"); + err = -ENOMEM; + goto err_create_fnic_sgl_slab_max; + } + + /* Create a cache of io_req structs for use via mempool */ + fnic_io_req_cache = kmem_cache_create("fnic_io_req", + sizeof(struct fnic_io_req), + 0, SLAB_HWCACHE_ALIGN, NULL); + if (!fnic_io_req_cache) { + printk(KERN_ERR PFX "failed to create fnic io_req slab\n"); + err = -ENOMEM; + goto err_create_fnic_ioreq_slab; + } + + fnic_event_queue = create_singlethread_workqueue("fnic_event_wq"); + if (!fnic_event_queue) { + printk(KERN_ERR PFX "fnic work queue create failed\n"); + err = -ENOMEM; + goto err_create_fnic_workq; + } + + spin_lock_init(&fnic_list_lock); + INIT_LIST_HEAD(&fnic_list); + + fnic_fc_transport = fc_attach_transport(&fnic_fc_functions); + if (!fnic_fc_transport) { + printk(KERN_ERR PFX "fc_attach_transport error\n"); + err = -ENOMEM; + goto err_fc_transport; + } + + /* register the driver with PCI system */ + err = pci_register_driver(&fnic_driver); + if (err < 0) { + printk(KERN_ERR PFX "pci register error\n"); + goto err_pci_register; + } + return err; + +err_pci_register: + fc_release_transport(fnic_fc_transport); +err_fc_transport: + destroy_workqueue(fnic_event_queue); +err_create_fnic_workq: + kmem_cache_destroy(fnic_io_req_cache); +err_create_fnic_ioreq_slab: + kmem_cache_destroy(fnic_sgl_cache[FNIC_SGL_CACHE_MAX]); +err_create_fnic_sgl_slab_max: + kmem_cache_destroy(fnic_sgl_cache[FNIC_SGL_CACHE_DFLT]); +err_create_fnic_sgl_slab_dflt: + return err; +} + +static void __exit fnic_cleanup_module(void) +{ + pci_unregister_driver(&fnic_driver); + destroy_workqueue(fnic_event_queue); + kmem_cache_destroy(fnic_sgl_cache[FNIC_SGL_CACHE_MAX]); + kmem_cache_destroy(fnic_sgl_cache[FNIC_SGL_CACHE_DFLT]); + kmem_cache_destroy(fnic_io_req_cache); + fc_release_transport(fnic_fc_transport); +} + +module_init(fnic_init_module); +module_exit(fnic_cleanup_module); + diff --git a/drivers/scsi/fnic/fnic_res.c b/drivers/scsi/fnic/fnic_res.c new file mode 100644 index 000000000000..7ba61ec715d2 --- /dev/null +++ b/drivers/scsi/fnic/fnic_res.c @@ -0,0 +1,444 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#include +#include +#include +#include "wq_enet_desc.h" +#include "rq_enet_desc.h" +#include "cq_enet_desc.h" +#include "vnic_resource.h" +#include "vnic_dev.h" +#include "vnic_wq.h" +#include "vnic_rq.h" +#include "vnic_cq.h" +#include "vnic_intr.h" +#include "vnic_stats.h" +#include "vnic_nic.h" +#include "fnic.h" + +int fnic_get_vnic_config(struct fnic *fnic) +{ + struct vnic_fc_config *c = &fnic->config; + int err; + +#define GET_CONFIG(m) \ + do { \ + err = vnic_dev_spec(fnic->vdev, \ + offsetof(struct vnic_fc_config, m), \ + sizeof(c->m), &c->m); \ + if (err) { \ + shost_printk(KERN_ERR, fnic->lport->host, \ + "Error getting %s, %d\n", #m, \ + err); \ + return err; \ + } \ + } while (0); + + GET_CONFIG(node_wwn); + GET_CONFIG(port_wwn); + GET_CONFIG(wq_enet_desc_count); + GET_CONFIG(wq_copy_desc_count); + GET_CONFIG(rq_desc_count); + GET_CONFIG(maxdatafieldsize); + GET_CONFIG(ed_tov); + GET_CONFIG(ra_tov); + GET_CONFIG(intr_timer); + GET_CONFIG(intr_timer_type); + GET_CONFIG(flags); + GET_CONFIG(flogi_retries); + GET_CONFIG(flogi_timeout); + GET_CONFIG(plogi_retries); + GET_CONFIG(plogi_timeout); + GET_CONFIG(io_throttle_count); + GET_CONFIG(link_down_timeout); + GET_CONFIG(port_down_timeout); + GET_CONFIG(port_down_io_retries); + GET_CONFIG(luns_per_tgt); + + c->wq_enet_desc_count = + min_t(u32, VNIC_FNIC_WQ_DESCS_MAX, + max_t(u32, VNIC_FNIC_WQ_DESCS_MIN, + c->wq_enet_desc_count)); + c->wq_enet_desc_count = ALIGN(c->wq_enet_desc_count, 16); + + c->wq_copy_desc_count = + min_t(u32, VNIC_FNIC_WQ_COPY_DESCS_MAX, + max_t(u32, VNIC_FNIC_WQ_COPY_DESCS_MIN, + c->wq_copy_desc_count)); + c->wq_copy_desc_count = ALIGN(c->wq_copy_desc_count, 16); + + c->rq_desc_count = + min_t(u32, VNIC_FNIC_RQ_DESCS_MAX, + max_t(u32, VNIC_FNIC_RQ_DESCS_MIN, + c->rq_desc_count)); + c->rq_desc_count = ALIGN(c->rq_desc_count, 16); + + c->maxdatafieldsize = + min_t(u16, VNIC_FNIC_MAXDATAFIELDSIZE_MAX, + max_t(u16, VNIC_FNIC_MAXDATAFIELDSIZE_MIN, + c->maxdatafieldsize)); + c->ed_tov = + min_t(u32, VNIC_FNIC_EDTOV_MAX, + max_t(u32, VNIC_FNIC_EDTOV_MIN, + c->ed_tov)); + + c->ra_tov = + min_t(u32, VNIC_FNIC_RATOV_MAX, + max_t(u32, VNIC_FNIC_RATOV_MIN, + c->ra_tov)); + + c->flogi_retries = + min_t(u32, VNIC_FNIC_FLOGI_RETRIES_MAX, c->flogi_retries); + + c->flogi_timeout = + min_t(u32, VNIC_FNIC_FLOGI_TIMEOUT_MAX, + max_t(u32, VNIC_FNIC_FLOGI_TIMEOUT_MIN, + c->flogi_timeout)); + + c->plogi_retries = + min_t(u32, VNIC_FNIC_PLOGI_RETRIES_MAX, c->plogi_retries); + + c->plogi_timeout = + min_t(u32, VNIC_FNIC_PLOGI_TIMEOUT_MAX, + max_t(u32, VNIC_FNIC_PLOGI_TIMEOUT_MIN, + c->plogi_timeout)); + + c->io_throttle_count = + min_t(u32, VNIC_FNIC_IO_THROTTLE_COUNT_MAX, + max_t(u32, VNIC_FNIC_IO_THROTTLE_COUNT_MIN, + c->io_throttle_count)); + + c->link_down_timeout = + min_t(u32, VNIC_FNIC_LINK_DOWN_TIMEOUT_MAX, + c->link_down_timeout); + + c->port_down_timeout = + min_t(u32, VNIC_FNIC_PORT_DOWN_TIMEOUT_MAX, + c->port_down_timeout); + + c->port_down_io_retries = + min_t(u32, VNIC_FNIC_PORT_DOWN_IO_RETRIES_MAX, + c->port_down_io_retries); + + c->luns_per_tgt = + min_t(u32, VNIC_FNIC_LUNS_PER_TARGET_MAX, + max_t(u32, VNIC_FNIC_LUNS_PER_TARGET_MIN, + c->luns_per_tgt)); + + c->intr_timer = min_t(u16, VNIC_INTR_TIMER_MAX, c->intr_timer); + c->intr_timer_type = c->intr_timer_type; + + shost_printk(KERN_INFO, fnic->lport->host, + "vNIC MAC addr %02x:%02x:%02x:%02x:%02x:%02x " + "wq/wq_copy/rq %d/%d/%d\n", + fnic->mac_addr[0], fnic->mac_addr[1], fnic->mac_addr[2], + fnic->mac_addr[3], fnic->mac_addr[4], fnic->mac_addr[5], + c->wq_enet_desc_count, c->wq_copy_desc_count, + c->rq_desc_count); + shost_printk(KERN_INFO, fnic->lport->host, + "vNIC node wwn %llx port wwn %llx\n", + c->node_wwn, c->port_wwn); + shost_printk(KERN_INFO, fnic->lport->host, + "vNIC ed_tov %d ra_tov %d\n", + c->ed_tov, c->ra_tov); + shost_printk(KERN_INFO, fnic->lport->host, + "vNIC mtu %d intr timer %d\n", + c->maxdatafieldsize, c->intr_timer); + shost_printk(KERN_INFO, fnic->lport->host, + "vNIC flags 0x%x luns per tgt %d\n", + c->flags, c->luns_per_tgt); + shost_printk(KERN_INFO, fnic->lport->host, + "vNIC flogi_retries %d flogi timeout %d\n", + c->flogi_retries, c->flogi_timeout); + shost_printk(KERN_INFO, fnic->lport->host, + "vNIC plogi retries %d plogi timeout %d\n", + c->plogi_retries, c->plogi_timeout); + shost_printk(KERN_INFO, fnic->lport->host, + "vNIC io throttle count %d link dn timeout %d\n", + c->io_throttle_count, c->link_down_timeout); + shost_printk(KERN_INFO, fnic->lport->host, + "vNIC port dn io retries %d port dn timeout %d\n", + c->port_down_io_retries, c->port_down_timeout); + + return 0; +} + +int fnic_set_nic_config(struct fnic *fnic, u8 rss_default_cpu, + u8 rss_hash_type, + u8 rss_hash_bits, u8 rss_base_cpu, u8 rss_enable, + u8 tso_ipid_split_en, u8 ig_vlan_strip_en) +{ + u64 a0, a1; + u32 nic_cfg; + int wait = 1000; + + vnic_set_nic_cfg(&nic_cfg, rss_default_cpu, + rss_hash_type, rss_hash_bits, rss_base_cpu, + rss_enable, tso_ipid_split_en, ig_vlan_strip_en); + + a0 = nic_cfg; + a1 = 0; + + return vnic_dev_cmd(fnic->vdev, CMD_NIC_CFG, &a0, &a1, wait); +} + +void fnic_get_res_counts(struct fnic *fnic) +{ + fnic->wq_count = vnic_dev_get_res_count(fnic->vdev, RES_TYPE_WQ); + fnic->raw_wq_count = fnic->wq_count - 1; + fnic->wq_copy_count = fnic->wq_count - fnic->raw_wq_count; + fnic->rq_count = vnic_dev_get_res_count(fnic->vdev, RES_TYPE_RQ); + fnic->cq_count = vnic_dev_get_res_count(fnic->vdev, RES_TYPE_CQ); + fnic->intr_count = vnic_dev_get_res_count(fnic->vdev, + RES_TYPE_INTR_CTRL); +} + +void fnic_free_vnic_resources(struct fnic *fnic) +{ + unsigned int i; + + for (i = 0; i < fnic->raw_wq_count; i++) + vnic_wq_free(&fnic->wq[i]); + + for (i = 0; i < fnic->wq_copy_count; i++) + vnic_wq_copy_free(&fnic->wq_copy[i]); + + for (i = 0; i < fnic->rq_count; i++) + vnic_rq_free(&fnic->rq[i]); + + for (i = 0; i < fnic->cq_count; i++) + vnic_cq_free(&fnic->cq[i]); + + for (i = 0; i < fnic->intr_count; i++) + vnic_intr_free(&fnic->intr[i]); +} + +int fnic_alloc_vnic_resources(struct fnic *fnic) +{ + enum vnic_dev_intr_mode intr_mode; + unsigned int mask_on_assertion; + unsigned int interrupt_offset; + unsigned int error_interrupt_enable; + unsigned int error_interrupt_offset; + unsigned int i, cq_index; + unsigned int wq_copy_cq_desc_count; + int err; + + intr_mode = vnic_dev_get_intr_mode(fnic->vdev); + + shost_printk(KERN_INFO, fnic->lport->host, "vNIC interrupt mode: %s\n", + intr_mode == VNIC_DEV_INTR_MODE_INTX ? "legacy PCI INTx" : + intr_mode == VNIC_DEV_INTR_MODE_MSI ? "MSI" : + intr_mode == VNIC_DEV_INTR_MODE_MSIX ? + "MSI-X" : "unknown"); + + shost_printk(KERN_INFO, fnic->lport->host, "vNIC resources avail: " + "wq %d cp_wq %d raw_wq %d rq %d cq %d intr %d\n", + fnic->wq_count, fnic->wq_copy_count, fnic->raw_wq_count, + fnic->rq_count, fnic->cq_count, fnic->intr_count); + + /* Allocate Raw WQ used for FCS frames */ + for (i = 0; i < fnic->raw_wq_count; i++) { + err = vnic_wq_alloc(fnic->vdev, &fnic->wq[i], i, + fnic->config.wq_enet_desc_count, + sizeof(struct wq_enet_desc)); + if (err) + goto err_out_cleanup; + } + + /* Allocate Copy WQs used for SCSI IOs */ + for (i = 0; i < fnic->wq_copy_count; i++) { + err = vnic_wq_copy_alloc(fnic->vdev, &fnic->wq_copy[i], + (fnic->raw_wq_count + i), + fnic->config.wq_copy_desc_count, + sizeof(struct fcpio_host_req)); + if (err) + goto err_out_cleanup; + } + + /* RQ for receiving FCS frames */ + for (i = 0; i < fnic->rq_count; i++) { + err = vnic_rq_alloc(fnic->vdev, &fnic->rq[i], i, + fnic->config.rq_desc_count, + sizeof(struct rq_enet_desc)); + if (err) + goto err_out_cleanup; + } + + /* CQ for each RQ */ + for (i = 0; i < fnic->rq_count; i++) { + cq_index = i; + err = vnic_cq_alloc(fnic->vdev, + &fnic->cq[cq_index], cq_index, + fnic->config.rq_desc_count, + sizeof(struct cq_enet_rq_desc)); + if (err) + goto err_out_cleanup; + } + + /* CQ for each WQ */ + for (i = 0; i < fnic->raw_wq_count; i++) { + cq_index = fnic->rq_count + i; + err = vnic_cq_alloc(fnic->vdev, &fnic->cq[cq_index], cq_index, + fnic->config.wq_enet_desc_count, + sizeof(struct cq_enet_wq_desc)); + if (err) + goto err_out_cleanup; + } + + /* CQ for each COPY WQ */ + wq_copy_cq_desc_count = (fnic->config.wq_copy_desc_count * 3); + for (i = 0; i < fnic->wq_copy_count; i++) { + cq_index = fnic->raw_wq_count + fnic->rq_count + i; + err = vnic_cq_alloc(fnic->vdev, &fnic->cq[cq_index], + cq_index, + wq_copy_cq_desc_count, + sizeof(struct fcpio_fw_req)); + if (err) + goto err_out_cleanup; + } + + for (i = 0; i < fnic->intr_count; i++) { + err = vnic_intr_alloc(fnic->vdev, &fnic->intr[i], i); + if (err) + goto err_out_cleanup; + } + + fnic->legacy_pba = vnic_dev_get_res(fnic->vdev, + RES_TYPE_INTR_PBA_LEGACY, 0); + + if (!fnic->legacy_pba && intr_mode == VNIC_DEV_INTR_MODE_INTX) { + shost_printk(KERN_ERR, fnic->lport->host, + "Failed to hook legacy pba resource\n"); + err = -ENODEV; + goto err_out_cleanup; + } + + /* + * Init RQ/WQ resources. + * + * RQ[0 to n-1] point to CQ[0 to n-1] + * WQ[0 to m-1] point to CQ[n to n+m-1] + * WQ_COPY[0 to k-1] points to CQ[n+m to n+m+k-1] + * + * Note for copy wq we always initialize with cq_index = 0 + * + * Error interrupt is not enabled for MSI. + */ + + switch (intr_mode) { + case VNIC_DEV_INTR_MODE_INTX: + case VNIC_DEV_INTR_MODE_MSIX: + error_interrupt_enable = 1; + error_interrupt_offset = fnic->err_intr_offset; + break; + default: + error_interrupt_enable = 0; + error_interrupt_offset = 0; + break; + } + + for (i = 0; i < fnic->rq_count; i++) { + cq_index = i; + vnic_rq_init(&fnic->rq[i], + cq_index, + error_interrupt_enable, + error_interrupt_offset); + } + + for (i = 0; i < fnic->raw_wq_count; i++) { + cq_index = i + fnic->rq_count; + vnic_wq_init(&fnic->wq[i], + cq_index, + error_interrupt_enable, + error_interrupt_offset); + } + + for (i = 0; i < fnic->wq_copy_count; i++) { + vnic_wq_copy_init(&fnic->wq_copy[i], + 0 /* cq_index 0 - always */, + error_interrupt_enable, + error_interrupt_offset); + } + + for (i = 0; i < fnic->cq_count; i++) { + + switch (intr_mode) { + case VNIC_DEV_INTR_MODE_MSIX: + interrupt_offset = i; + break; + default: + interrupt_offset = 0; + break; + } + + vnic_cq_init(&fnic->cq[i], + 0 /* flow_control_enable */, + 1 /* color_enable */, + 0 /* cq_head */, + 0 /* cq_tail */, + 1 /* cq_tail_color */, + 1 /* interrupt_enable */, + 1 /* cq_entry_enable */, + 0 /* cq_message_enable */, + interrupt_offset, + 0 /* cq_message_addr */); + } + + /* + * Init INTR resources + * + * mask_on_assertion is not used for INTx due to the level- + * triggered nature of INTx + */ + + switch (intr_mode) { + case VNIC_DEV_INTR_MODE_MSI: + case VNIC_DEV_INTR_MODE_MSIX: + mask_on_assertion = 1; + break; + default: + mask_on_assertion = 0; + break; + } + + for (i = 0; i < fnic->intr_count; i++) { + vnic_intr_init(&fnic->intr[i], + fnic->config.intr_timer, + fnic->config.intr_timer_type, + mask_on_assertion); + } + + /* init the stats memory by making the first call here */ + err = vnic_dev_stats_dump(fnic->vdev, &fnic->stats); + if (err) { + shost_printk(KERN_ERR, fnic->lport->host, + "vnic_dev_stats_dump failed - x%x\n", err); + goto err_out_cleanup; + } + + /* Clear LIF stats */ + vnic_dev_stats_clear(fnic->vdev); + + return 0; + +err_out_cleanup: + fnic_free_vnic_resources(fnic); + + return err; +} diff --git a/drivers/scsi/fnic/fnic_res.h b/drivers/scsi/fnic/fnic_res.h new file mode 100644 index 000000000000..b6f310262534 --- /dev/null +++ b/drivers/scsi/fnic/fnic_res.h @@ -0,0 +1,197 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _FNIC_RES_H_ +#define _FNIC_RES_H_ + +#include "wq_enet_desc.h" +#include "rq_enet_desc.h" +#include "vnic_wq.h" +#include "vnic_rq.h" +#include "fnic_io.h" +#include "fcpio.h" +#include "vnic_wq_copy.h" +#include "vnic_cq_copy.h" + +static inline void fnic_queue_wq_desc(struct vnic_wq *wq, + void *os_buf, dma_addr_t dma_addr, + unsigned int len, unsigned int fc_eof, + int vlan_tag_insert, + unsigned int vlan_tag, + int cq_entry, int sop, int eop) +{ + struct wq_enet_desc *desc = vnic_wq_next_desc(wq); + + wq_enet_desc_enc(desc, + (u64)dma_addr | VNIC_PADDR_TARGET, + (u16)len, + 0, /* mss_or_csum_offset */ + (u16)fc_eof, + 0, /* offload_mode */ + (u8)eop, (u8)cq_entry, + 1, /* fcoe_encap */ + (u8)vlan_tag_insert, + (u16)vlan_tag, + 0 /* loopback */); + + vnic_wq_post(wq, os_buf, dma_addr, len, sop, eop); +} + +static inline void fnic_queue_wq_copy_desc_icmnd_16(struct vnic_wq_copy *wq, + u32 req_id, + u32 lunmap_id, u8 spl_flags, + u32 sgl_cnt, u32 sense_len, + u64 sgl_addr, u64 sns_addr, + u8 crn, u8 pri_ta, + u8 flags, u8 *scsi_cdb, + u32 data_len, u8 *lun, + u32 d_id, u16 mss, + u32 ratov, u32 edtov) +{ + struct fcpio_host_req *desc = vnic_wq_copy_next_desc(wq); + + desc->hdr.type = FCPIO_ICMND_16; /* enum fcpio_type */ + desc->hdr.status = 0; /* header status entry */ + desc->hdr._resvd = 0; /* reserved */ + desc->hdr.tag.u.req_id = req_id; /* id for this request */ + + desc->u.icmnd_16.lunmap_id = lunmap_id; /* index into lunmap table */ + desc->u.icmnd_16.special_req_flags = spl_flags; /* exch req flags */ + desc->u.icmnd_16._resvd0[0] = 0; /* reserved */ + desc->u.icmnd_16._resvd0[1] = 0; /* reserved */ + desc->u.icmnd_16._resvd0[2] = 0; /* reserved */ + desc->u.icmnd_16.sgl_cnt = sgl_cnt; /* scatter-gather list count */ + desc->u.icmnd_16.sense_len = sense_len; /* sense buffer length */ + desc->u.icmnd_16.sgl_addr = sgl_addr; /* scatter-gather list addr */ + desc->u.icmnd_16.sense_addr = sns_addr; /* sense buffer address */ + desc->u.icmnd_16.crn = crn; /* SCSI Command Reference No.*/ + desc->u.icmnd_16.pri_ta = pri_ta; /* SCSI Pri & Task attribute */ + desc->u.icmnd_16._resvd1 = 0; /* reserved: should be 0 */ + desc->u.icmnd_16.flags = flags; /* command flags */ + memcpy(desc->u.icmnd_16.scsi_cdb, scsi_cdb, CDB_16); /* SCSI CDB */ + desc->u.icmnd_16.data_len = data_len; /* length of data expected */ + memcpy(desc->u.icmnd_16.lun, lun, LUN_ADDRESS); /* LUN address */ + desc->u.icmnd_16._resvd2 = 0; /* reserved */ + hton24(desc->u.icmnd_16.d_id, d_id); /* FC vNIC only: Target D_ID */ + desc->u.icmnd_16.mss = mss; /* FC vNIC only: max burst */ + desc->u.icmnd_16.r_a_tov = ratov; /*FC vNIC only: Res. Alloc Timeout */ + desc->u.icmnd_16.e_d_tov = edtov; /*FC vNIC only: Err Detect Timeout */ + + vnic_wq_copy_post(wq); +} + +static inline void fnic_queue_wq_copy_desc_itmf(struct vnic_wq_copy *wq, + u32 req_id, u32 lunmap_id, + u32 tm_req, u32 tm_id, u8 *lun, + u32 d_id, u32 r_a_tov, + u32 e_d_tov) +{ + struct fcpio_host_req *desc = vnic_wq_copy_next_desc(wq); + + desc->hdr.type = FCPIO_ITMF; /* enum fcpio_type */ + desc->hdr.status = 0; /* header status entry */ + desc->hdr._resvd = 0; /* reserved */ + desc->hdr.tag.u.req_id = req_id; /* id for this request */ + + desc->u.itmf.lunmap_id = lunmap_id; /* index into lunmap table */ + desc->u.itmf.tm_req = tm_req; /* SCSI Task Management request */ + desc->u.itmf.t_tag = tm_id; /* tag of fcpio to be aborted */ + desc->u.itmf._resvd = 0; + memcpy(desc->u.itmf.lun, lun, LUN_ADDRESS); /* LUN address */ + desc->u.itmf._resvd1 = 0; + hton24(desc->u.itmf.d_id, d_id); /* FC vNIC only: Target D_ID */ + desc->u.itmf.r_a_tov = r_a_tov; /* FC vNIC only: R_A_TOV in msec */ + desc->u.itmf.e_d_tov = e_d_tov; /* FC vNIC only: E_D_TOV in msec */ + + vnic_wq_copy_post(wq); +} + +static inline void fnic_queue_wq_copy_desc_flogi_reg(struct vnic_wq_copy *wq, + u32 req_id, u8 format, + u32 s_id, u8 *gw_mac) +{ + struct fcpio_host_req *desc = vnic_wq_copy_next_desc(wq); + + desc->hdr.type = FCPIO_FLOGI_REG; /* enum fcpio_type */ + desc->hdr.status = 0; /* header status entry */ + desc->hdr._resvd = 0; /* reserved */ + desc->hdr.tag.u.req_id = req_id; /* id for this request */ + + desc->u.flogi_reg.format = format; + hton24(desc->u.flogi_reg.s_id, s_id); + memcpy(desc->u.flogi_reg.gateway_mac, gw_mac, ETH_ALEN); + + vnic_wq_copy_post(wq); +} + +static inline void fnic_queue_wq_copy_desc_fw_reset(struct vnic_wq_copy *wq, + u32 req_id) +{ + struct fcpio_host_req *desc = vnic_wq_copy_next_desc(wq); + + desc->hdr.type = FCPIO_RESET; /* enum fcpio_type */ + desc->hdr.status = 0; /* header status entry */ + desc->hdr._resvd = 0; /* reserved */ + desc->hdr.tag.u.req_id = req_id; /* id for this request */ + + vnic_wq_copy_post(wq); +} + +static inline void fnic_queue_wq_copy_desc_lunmap(struct vnic_wq_copy *wq, + u32 req_id, u64 lunmap_addr, + u32 lunmap_len) +{ + struct fcpio_host_req *desc = vnic_wq_copy_next_desc(wq); + + desc->hdr.type = FCPIO_LUNMAP_REQ; /* enum fcpio_type */ + desc->hdr.status = 0; /* header status entry */ + desc->hdr._resvd = 0; /* reserved */ + desc->hdr.tag.u.req_id = req_id; /* id for this request */ + + desc->u.lunmap_req.addr = lunmap_addr; /* address of the buffer */ + desc->u.lunmap_req.len = lunmap_len; /* len of the buffer */ + + vnic_wq_copy_post(wq); +} + +static inline void fnic_queue_rq_desc(struct vnic_rq *rq, + void *os_buf, dma_addr_t dma_addr, + u16 len) +{ + struct rq_enet_desc *desc = vnic_rq_next_desc(rq); + + rq_enet_desc_enc(desc, + (u64)dma_addr | VNIC_PADDR_TARGET, + RQ_ENET_TYPE_ONLY_SOP, + (u16)len); + + vnic_rq_post(rq, os_buf, 0, dma_addr, len); +} + + +struct fnic; + +int fnic_get_vnic_config(struct fnic *); +int fnic_alloc_vnic_resources(struct fnic *); +void fnic_free_vnic_resources(struct fnic *); +void fnic_get_res_counts(struct fnic *); +int fnic_set_nic_config(struct fnic *fnic, u8 rss_default_cpu, + u8 rss_hash_type, u8 rss_hash_bits, u8 rss_base_cpu, + u8 rss_enable, u8 tso_ipid_split_en, + u8 ig_vlan_strip_en); + +#endif /* _FNIC_RES_H_ */ diff --git a/drivers/scsi/fnic/fnic_scsi.c b/drivers/scsi/fnic/fnic_scsi.c new file mode 100644 index 000000000000..eabf36502856 --- /dev/null +++ b/drivers/scsi/fnic/fnic_scsi.c @@ -0,0 +1,1850 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "fnic_io.h" +#include "fnic.h" + +const char *fnic_state_str[] = { + [FNIC_IN_FC_MODE] = "FNIC_IN_FC_MODE", + [FNIC_IN_FC_TRANS_ETH_MODE] = "FNIC_IN_FC_TRANS_ETH_MODE", + [FNIC_IN_ETH_MODE] = "FNIC_IN_ETH_MODE", + [FNIC_IN_ETH_TRANS_FC_MODE] = "FNIC_IN_ETH_TRANS_FC_MODE", +}; + +static const char *fnic_ioreq_state_str[] = { + [FNIC_IOREQ_CMD_PENDING] = "FNIC_IOREQ_CMD_PENDING", + [FNIC_IOREQ_ABTS_PENDING] = "FNIC_IOREQ_ABTS_PENDING", + [FNIC_IOREQ_ABTS_COMPLETE] = "FNIC_IOREQ_ABTS_COMPLETE", + [FNIC_IOREQ_CMD_COMPLETE] = "FNIC_IOREQ_CMD_COMPLETE", +}; + +static const char *fcpio_status_str[] = { + [FCPIO_SUCCESS] = "FCPIO_SUCCESS", /*0x0*/ + [FCPIO_INVALID_HEADER] = "FCPIO_INVALID_HEADER", + [FCPIO_OUT_OF_RESOURCE] = "FCPIO_OUT_OF_RESOURCE", + [FCPIO_INVALID_PARAM] = "FCPIO_INVALID_PARAM]", + [FCPIO_REQ_NOT_SUPPORTED] = "FCPIO_REQ_NOT_SUPPORTED", + [FCPIO_IO_NOT_FOUND] = "FCPIO_IO_NOT_FOUND", + [FCPIO_ABORTED] = "FCPIO_ABORTED", /*0x41*/ + [FCPIO_TIMEOUT] = "FCPIO_TIMEOUT", + [FCPIO_SGL_INVALID] = "FCPIO_SGL_INVALID", + [FCPIO_MSS_INVALID] = "FCPIO_MSS_INVALID", + [FCPIO_DATA_CNT_MISMATCH] = "FCPIO_DATA_CNT_MISMATCH", + [FCPIO_FW_ERR] = "FCPIO_FW_ERR", + [FCPIO_ITMF_REJECTED] = "FCPIO_ITMF_REJECTED", + [FCPIO_ITMF_FAILED] = "FCPIO_ITMF_FAILED", + [FCPIO_ITMF_INCORRECT_LUN] = "FCPIO_ITMF_INCORRECT_LUN", + [FCPIO_CMND_REJECTED] = "FCPIO_CMND_REJECTED", + [FCPIO_NO_PATH_AVAIL] = "FCPIO_NO_PATH_AVAIL", + [FCPIO_PATH_FAILED] = "FCPIO_PATH_FAILED", + [FCPIO_LUNMAP_CHNG_PEND] = "FCPIO_LUNHMAP_CHNG_PEND", +}; + +const char *fnic_state_to_str(unsigned int state) +{ + if (state >= ARRAY_SIZE(fnic_state_str) || !fnic_state_str[state]) + return "unknown"; + + return fnic_state_str[state]; +} + +static const char *fnic_ioreq_state_to_str(unsigned int state) +{ + if (state >= ARRAY_SIZE(fnic_ioreq_state_str) || + !fnic_ioreq_state_str[state]) + return "unknown"; + + return fnic_ioreq_state_str[state]; +} + +static const char *fnic_fcpio_status_to_str(unsigned int status) +{ + if (status >= ARRAY_SIZE(fcpio_status_str) || !fcpio_status_str[status]) + return "unknown"; + + return fcpio_status_str[status]; +} + +static void fnic_cleanup_io(struct fnic *fnic, int exclude_id); + +static inline spinlock_t *fnic_io_lock_hash(struct fnic *fnic, + struct scsi_cmnd *sc) +{ + u32 hash = sc->request->tag & (FNIC_IO_LOCKS - 1); + + return &fnic->io_req_lock[hash]; +} + +/* + * Unmap the data buffer and sense buffer for an io_req, + * also unmap and free the device-private scatter/gather list. + */ +static void fnic_release_ioreq_buf(struct fnic *fnic, + struct fnic_io_req *io_req, + struct scsi_cmnd *sc) +{ + if (io_req->sgl_list_pa) + pci_unmap_single(fnic->pdev, io_req->sgl_list_pa, + sizeof(io_req->sgl_list[0]) * io_req->sgl_cnt, + PCI_DMA_TODEVICE); + scsi_dma_unmap(sc); + + if (io_req->sgl_cnt) + mempool_free(io_req->sgl_list_alloc, + fnic->io_sgl_pool[io_req->sgl_type]); + if (io_req->sense_buf_pa) + pci_unmap_single(fnic->pdev, io_req->sense_buf_pa, + SCSI_SENSE_BUFFERSIZE, PCI_DMA_FROMDEVICE); +} + +/* Free up Copy Wq descriptors. Called with copy_wq lock held */ +static int free_wq_copy_descs(struct fnic *fnic, struct vnic_wq_copy *wq) +{ + /* if no Ack received from firmware, then nothing to clean */ + if (!fnic->fw_ack_recd[0]) + return 1; + + /* + * Update desc_available count based on number of freed descriptors + * Account for wraparound + */ + if (wq->to_clean_index <= fnic->fw_ack_index[0]) + wq->ring.desc_avail += (fnic->fw_ack_index[0] + - wq->to_clean_index + 1); + else + wq->ring.desc_avail += (wq->ring.desc_count + - wq->to_clean_index + + fnic->fw_ack_index[0] + 1); + + /* + * just bump clean index to ack_index+1 accounting for wraparound + * this will essentially free up all descriptors between + * to_clean_index and fw_ack_index, both inclusive + */ + wq->to_clean_index = + (fnic->fw_ack_index[0] + 1) % wq->ring.desc_count; + + /* we have processed the acks received so far */ + fnic->fw_ack_recd[0] = 0; + return 0; +} + + +/* + * fnic_fw_reset_handler + * Routine to send reset msg to fw + */ +int fnic_fw_reset_handler(struct fnic *fnic) +{ + struct vnic_wq_copy *wq = &fnic->wq_copy[0]; + int ret = 0; + unsigned long flags; + + spin_lock_irqsave(&fnic->wq_copy_lock[0], flags); + + if (vnic_wq_copy_desc_avail(wq) <= fnic->wq_copy_desc_low[0]) + free_wq_copy_descs(fnic, wq); + + if (!vnic_wq_copy_desc_avail(wq)) + ret = -EAGAIN; + else + fnic_queue_wq_copy_desc_fw_reset(wq, SCSI_NO_TAG); + + spin_unlock_irqrestore(&fnic->wq_copy_lock[0], flags); + + if (!ret) + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "Issued fw reset\n"); + else + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "Failed to issue fw reset\n"); + return ret; +} + + +/* + * fnic_flogi_reg_handler + * Routine to send flogi register msg to fw + */ +int fnic_flogi_reg_handler(struct fnic *fnic) +{ + struct vnic_wq_copy *wq = &fnic->wq_copy[0]; + u8 gw_mac[ETH_ALEN]; + int ret = 0; + unsigned long flags; + + spin_lock_irqsave(&fnic->wq_copy_lock[0], flags); + + if (vnic_wq_copy_desc_avail(wq) <= fnic->wq_copy_desc_low[0]) + free_wq_copy_descs(fnic, wq); + + if (!vnic_wq_copy_desc_avail(wq)) { + ret = -EAGAIN; + goto flogi_reg_ioreq_end; + } + + if (fnic->fcoui_mode) + memset(gw_mac, 0xff, ETH_ALEN); + else + memcpy(gw_mac, fnic->dest_addr, ETH_ALEN); + + fnic_queue_wq_copy_desc_flogi_reg(wq, SCSI_NO_TAG, + FCPIO_FLOGI_REG_GW_DEST, + fnic->s_id, + gw_mac); + +flogi_reg_ioreq_end: + spin_unlock_irqrestore(&fnic->wq_copy_lock[0], flags); + + if (!ret) + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "flog reg issued\n"); + + return ret; +} + +/* + * fnic_queue_wq_copy_desc + * Routine to enqueue a wq copy desc + */ +static inline int fnic_queue_wq_copy_desc(struct fnic *fnic, + struct vnic_wq_copy *wq, + struct fnic_io_req *io_req, + struct scsi_cmnd *sc, + u32 sg_count) +{ + struct scatterlist *sg; + struct fc_rport *rport = starget_to_rport(scsi_target(sc->device)); + struct fc_rport_libfc_priv *rp = rport->dd_data; + struct host_sg_desc *desc; + u8 pri_tag = 0; + unsigned int i; + unsigned long intr_flags; + int flags; + u8 exch_flags; + struct scsi_lun fc_lun; + char msg[2]; + + if (sg_count) { + BUG_ON(sg_count < 0); + BUG_ON(sg_count > FNIC_MAX_SG_DESC_CNT); + + /* For each SGE, create a device desc entry */ + desc = io_req->sgl_list; + for_each_sg(scsi_sglist(sc), sg, sg_count, i) { + desc->addr = cpu_to_le64(sg_dma_address(sg)); + desc->len = cpu_to_le32(sg_dma_len(sg)); + desc->_resvd = 0; + desc++; + } + + io_req->sgl_list_pa = pci_map_single + (fnic->pdev, + io_req->sgl_list, + sizeof(io_req->sgl_list[0]) * sg_count, + PCI_DMA_TODEVICE); + } + + io_req->sense_buf_pa = pci_map_single(fnic->pdev, + sc->sense_buffer, + SCSI_SENSE_BUFFERSIZE, + PCI_DMA_FROMDEVICE); + + int_to_scsilun(sc->device->lun, &fc_lun); + + pri_tag = FCPIO_ICMND_PTA_SIMPLE; + msg[0] = MSG_SIMPLE_TAG; + scsi_populate_tag_msg(sc, msg); + if (msg[0] == MSG_ORDERED_TAG) + pri_tag = FCPIO_ICMND_PTA_ORDERED; + + /* Enqueue the descriptor in the Copy WQ */ + spin_lock_irqsave(&fnic->wq_copy_lock[0], intr_flags); + + if (vnic_wq_copy_desc_avail(wq) <= fnic->wq_copy_desc_low[0]) + free_wq_copy_descs(fnic, wq); + + if (unlikely(!vnic_wq_copy_desc_avail(wq))) { + spin_unlock_irqrestore(&fnic->wq_copy_lock[0], intr_flags); + return SCSI_MLQUEUE_HOST_BUSY; + } + + flags = 0; + if (sc->sc_data_direction == DMA_FROM_DEVICE) + flags = FCPIO_ICMND_RDDATA; + else if (sc->sc_data_direction == DMA_TO_DEVICE) + flags = FCPIO_ICMND_WRDATA; + + exch_flags = 0; + if ((fnic->config.flags & VFCF_FCP_SEQ_LVL_ERR) && + (rp->flags & FC_RP_FLAGS_RETRY)) + exch_flags |= FCPIO_ICMND_SRFLAG_RETRY; + + fnic_queue_wq_copy_desc_icmnd_16(wq, sc->request->tag, + 0, exch_flags, io_req->sgl_cnt, + SCSI_SENSE_BUFFERSIZE, + io_req->sgl_list_pa, + io_req->sense_buf_pa, + 0, /* scsi cmd ref, always 0 */ + pri_tag, /* scsi pri and tag */ + flags, /* command flags */ + sc->cmnd, scsi_bufflen(sc), + fc_lun.scsi_lun, io_req->port_id, + rport->maxframe_size, rp->r_a_tov, + rp->e_d_tov); + + spin_unlock_irqrestore(&fnic->wq_copy_lock[0], intr_flags); + return 0; +} + +/* + * fnic_queuecommand + * Routine to send a scsi cdb + * Called with host_lock held and interrupts disabled. + */ +int fnic_queuecommand(struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *)) +{ + struct fc_lport *lp; + struct fc_rport *rport; + struct fnic_io_req *io_req; + struct fnic *fnic; + struct vnic_wq_copy *wq; + int ret; + u32 sg_count; + unsigned long flags; + unsigned long ptr; + + rport = starget_to_rport(scsi_target(sc->device)); + ret = fc_remote_port_chkready(rport); + if (ret) { + sc->result = ret; + done(sc); + return 0; + } + + lp = shost_priv(sc->device->host); + if (lp->state != LPORT_ST_READY || !(lp->link_up)) + return SCSI_MLQUEUE_HOST_BUSY; + + /* + * Release host lock, use driver resource specific locks from here. + * Don't re-enable interrupts in case they were disabled prior to the + * caller disabling them. + */ + spin_unlock(lp->host->host_lock); + + /* Get a new io_req for this SCSI IO */ + fnic = lport_priv(lp); + + io_req = mempool_alloc(fnic->io_req_pool, GFP_ATOMIC); + if (!io_req) { + ret = SCSI_MLQUEUE_HOST_BUSY; + goto out; + } + memset(io_req, 0, sizeof(*io_req)); + + /* Map the data buffer */ + sg_count = scsi_dma_map(sc); + if (sg_count < 0) { + mempool_free(io_req, fnic->io_req_pool); + goto out; + } + + /* Determine the type of scatter/gather list we need */ + io_req->sgl_cnt = sg_count; + io_req->sgl_type = FNIC_SGL_CACHE_DFLT; + if (sg_count > FNIC_DFLT_SG_DESC_CNT) + io_req->sgl_type = FNIC_SGL_CACHE_MAX; + + if (sg_count) { + io_req->sgl_list = + mempool_alloc(fnic->io_sgl_pool[io_req->sgl_type], + GFP_ATOMIC | GFP_DMA); + if (!io_req->sgl_list) { + ret = SCSI_MLQUEUE_HOST_BUSY; + scsi_dma_unmap(sc); + mempool_free(io_req, fnic->io_req_pool); + goto out; + } + + /* Cache sgl list allocated address before alignment */ + io_req->sgl_list_alloc = io_req->sgl_list; + ptr = (unsigned long) io_req->sgl_list; + if (ptr % FNIC_SG_DESC_ALIGN) { + io_req->sgl_list = (struct host_sg_desc *) + (((unsigned long) ptr + + FNIC_SG_DESC_ALIGN - 1) + & ~(FNIC_SG_DESC_ALIGN - 1)); + } + } + + /* initialize rest of io_req */ + io_req->port_id = rport->port_id; + CMD_STATE(sc) = FNIC_IOREQ_CMD_PENDING; + CMD_SP(sc) = (char *)io_req; + sc->scsi_done = done; + + /* create copy wq desc and enqueue it */ + wq = &fnic->wq_copy[0]; + ret = fnic_queue_wq_copy_desc(fnic, wq, io_req, sc, sg_count); + if (ret) { + /* + * In case another thread cancelled the request, + * refetch the pointer under the lock. + */ + spinlock_t *io_lock = fnic_io_lock_hash(fnic, sc); + + spin_lock_irqsave(io_lock, flags); + io_req = (struct fnic_io_req *)CMD_SP(sc); + CMD_SP(sc) = NULL; + CMD_STATE(sc) = FNIC_IOREQ_CMD_COMPLETE; + spin_unlock_irqrestore(io_lock, flags); + if (io_req) { + fnic_release_ioreq_buf(fnic, io_req, sc); + mempool_free(io_req, fnic->io_req_pool); + } + } +out: + /* acquire host lock before returning to SCSI */ + spin_lock(lp->host->host_lock); + return ret; +} + +/* + * fnic_fcpio_fw_reset_cmpl_handler + * Routine to handle fw reset completion + */ +static int fnic_fcpio_fw_reset_cmpl_handler(struct fnic *fnic, + struct fcpio_fw_req *desc) +{ + u8 type; + u8 hdr_status; + struct fcpio_tag tag; + int ret = 0; + struct fc_frame *flogi; + unsigned long flags; + + fcpio_header_dec(&desc->hdr, &type, &hdr_status, &tag); + + /* Clean up all outstanding io requests */ + fnic_cleanup_io(fnic, SCSI_NO_TAG); + + spin_lock_irqsave(&fnic->fnic_lock, flags); + + flogi = fnic->flogi; + fnic->flogi = NULL; + + /* fnic should be in FC_TRANS_ETH_MODE */ + if (fnic->state == FNIC_IN_FC_TRANS_ETH_MODE) { + /* Check status of reset completion */ + if (!hdr_status) { + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "reset cmpl success\n"); + /* Ready to send flogi out */ + fnic->state = FNIC_IN_ETH_MODE; + } else { + FNIC_SCSI_DBG(KERN_DEBUG, + fnic->lport->host, + "fnic fw_reset : failed %s\n", + fnic_fcpio_status_to_str(hdr_status)); + + /* + * Unable to change to eth mode, cannot send out flogi + * Change state to fc mode, so that subsequent Flogi + * requests from libFC will cause more attempts to + * reset the firmware. Free the cached flogi + */ + fnic->state = FNIC_IN_FC_MODE; + ret = -1; + } + } else { + FNIC_SCSI_DBG(KERN_DEBUG, + fnic->lport->host, + "Unexpected state %s while processing" + " reset cmpl\n", fnic_state_to_str(fnic->state)); + ret = -1; + } + + /* Thread removing device blocks till firmware reset is complete */ + if (fnic->remove_wait) + complete(fnic->remove_wait); + + /* + * If fnic is being removed, or fw reset failed + * free the flogi frame. Else, send it out + */ + if (fnic->remove_wait || ret) { + fnic->flogi_oxid = FC_XID_UNKNOWN; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + if (flogi) + dev_kfree_skb_irq(fp_skb(flogi)); + goto reset_cmpl_handler_end; + } + + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + + if (flogi) + ret = fnic_send_frame(fnic, flogi); + + reset_cmpl_handler_end: + return ret; +} + +/* + * fnic_fcpio_flogi_reg_cmpl_handler + * Routine to handle flogi register completion + */ +static int fnic_fcpio_flogi_reg_cmpl_handler(struct fnic *fnic, + struct fcpio_fw_req *desc) +{ + u8 type; + u8 hdr_status; + struct fcpio_tag tag; + int ret = 0; + struct fc_frame *flogi_resp = NULL; + unsigned long flags; + struct sk_buff *skb; + + fcpio_header_dec(&desc->hdr, &type, &hdr_status, &tag); + + /* Update fnic state based on status of flogi reg completion */ + spin_lock_irqsave(&fnic->fnic_lock, flags); + + flogi_resp = fnic->flogi_resp; + fnic->flogi_resp = NULL; + + if (fnic->state == FNIC_IN_ETH_TRANS_FC_MODE) { + + /* Check flogi registration completion status */ + if (!hdr_status) { + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "flog reg succeeded\n"); + fnic->state = FNIC_IN_FC_MODE; + } else { + FNIC_SCSI_DBG(KERN_DEBUG, + fnic->lport->host, + "fnic flogi reg :failed %s\n", + fnic_fcpio_status_to_str(hdr_status)); + fnic->state = FNIC_IN_ETH_MODE; + ret = -1; + } + } else { + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "Unexpected fnic state %s while" + " processing flogi reg completion\n", + fnic_state_to_str(fnic->state)); + ret = -1; + } + + /* Successful flogi reg cmpl, pass frame to LibFC */ + if (!ret && flogi_resp) { + if (fnic->stop_rx_link_events) { + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + goto reg_cmpl_handler_end; + } + skb = (struct sk_buff *)flogi_resp; + /* Use fr_flags to indicate whether flogi resp or not */ + fr_flags(flogi_resp) = 1; + fr_dev(flogi_resp) = fnic->lport; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + + skb_queue_tail(&fnic->frame_queue, skb); + queue_work(fnic_event_queue, &fnic->frame_work); + + } else { + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + if (flogi_resp) + dev_kfree_skb_irq(fp_skb(flogi_resp)); + } + +reg_cmpl_handler_end: + return ret; +} + +static inline int is_ack_index_in_range(struct vnic_wq_copy *wq, + u16 request_out) +{ + if (wq->to_clean_index <= wq->to_use_index) { + /* out of range, stale request_out index */ + if (request_out < wq->to_clean_index || + request_out >= wq->to_use_index) + return 0; + } else { + /* out of range, stale request_out index */ + if (request_out < wq->to_clean_index && + request_out >= wq->to_use_index) + return 0; + } + /* request_out index is in range */ + return 1; +} + + +/* + * Mark that ack received and store the Ack index. If there are multiple + * acks received before Tx thread cleans it up, the latest value will be + * used which is correct behavior. This state should be in the copy Wq + * instead of in the fnic + */ +static inline void fnic_fcpio_ack_handler(struct fnic *fnic, + unsigned int cq_index, + struct fcpio_fw_req *desc) +{ + struct vnic_wq_copy *wq; + u16 request_out = desc->u.ack.request_out; + unsigned long flags; + + /* mark the ack state */ + wq = &fnic->wq_copy[cq_index - fnic->raw_wq_count - fnic->rq_count]; + spin_lock_irqsave(&fnic->wq_copy_lock[0], flags); + + if (is_ack_index_in_range(wq, request_out)) { + fnic->fw_ack_index[0] = request_out; + fnic->fw_ack_recd[0] = 1; + } + spin_unlock_irqrestore(&fnic->wq_copy_lock[0], flags); +} + +/* + * fnic_fcpio_icmnd_cmpl_handler + * Routine to handle icmnd completions + */ +static void fnic_fcpio_icmnd_cmpl_handler(struct fnic *fnic, + struct fcpio_fw_req *desc) +{ + u8 type; + u8 hdr_status; + struct fcpio_tag tag; + u32 id; + u64 xfer_len = 0; + struct fcpio_icmnd_cmpl *icmnd_cmpl; + struct fnic_io_req *io_req; + struct scsi_cmnd *sc; + unsigned long flags; + spinlock_t *io_lock; + + /* Decode the cmpl description to get the io_req id */ + fcpio_header_dec(&desc->hdr, &type, &hdr_status, &tag); + fcpio_tag_id_dec(&tag, &id); + + if (id >= FNIC_MAX_IO_REQ) + return; + + sc = scsi_host_find_tag(fnic->lport->host, id); + WARN_ON_ONCE(!sc); + if (!sc) + return; + + io_lock = fnic_io_lock_hash(fnic, sc); + spin_lock_irqsave(io_lock, flags); + io_req = (struct fnic_io_req *)CMD_SP(sc); + WARN_ON_ONCE(!io_req); + if (!io_req) { + spin_unlock_irqrestore(io_lock, flags); + return; + } + + /* firmware completed the io */ + io_req->io_completed = 1; + + /* + * if SCSI-ML has already issued abort on this command, + * ignore completion of the IO. The abts path will clean it up + */ + if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) { + spin_unlock_irqrestore(io_lock, flags); + return; + } + + /* Mark the IO as complete */ + CMD_STATE(sc) = FNIC_IOREQ_CMD_COMPLETE; + + icmnd_cmpl = &desc->u.icmnd_cmpl; + + switch (hdr_status) { + case FCPIO_SUCCESS: + sc->result = (DID_OK << 16) | icmnd_cmpl->scsi_status; + xfer_len = scsi_bufflen(sc); + scsi_set_resid(sc, icmnd_cmpl->residual); + + if (icmnd_cmpl->flags & FCPIO_ICMND_CMPL_RESID_UNDER) + xfer_len -= icmnd_cmpl->residual; + + /* + * If queue_full, then try to reduce queue depth for all + * LUNS on the target. Todo: this should be accompanied + * by a periodic queue_depth rampup based on successful + * IO completion. + */ + if (icmnd_cmpl->scsi_status == QUEUE_FULL) { + struct scsi_device *t_sdev; + int qd = 0; + + shost_for_each_device(t_sdev, sc->device->host) { + if (t_sdev->id != sc->device->id) + continue; + + if (t_sdev->queue_depth > 1) { + qd = scsi_track_queue_full + (t_sdev, + t_sdev->queue_depth - 1); + if (qd == -1) + qd = t_sdev->host->cmd_per_lun; + shost_printk(KERN_INFO, + fnic->lport->host, + "scsi[%d:%d:%d:%d" + "] queue full detected," + "new depth = %d\n", + t_sdev->host->host_no, + t_sdev->channel, + t_sdev->id, t_sdev->lun, + t_sdev->queue_depth); + } + } + } + break; + + case FCPIO_TIMEOUT: /* request was timed out */ + sc->result = (DID_TIME_OUT << 16) | icmnd_cmpl->scsi_status; + break; + + case FCPIO_ABORTED: /* request was aborted */ + sc->result = (DID_ERROR << 16) | icmnd_cmpl->scsi_status; + break; + + case FCPIO_DATA_CNT_MISMATCH: /* recv/sent more/less data than exp. */ + scsi_set_resid(sc, icmnd_cmpl->residual); + sc->result = (DID_ERROR << 16) | icmnd_cmpl->scsi_status; + break; + + case FCPIO_OUT_OF_RESOURCE: /* out of resources to complete request */ + sc->result = (DID_REQUEUE << 16) | icmnd_cmpl->scsi_status; + break; + case FCPIO_INVALID_HEADER: /* header contains invalid data */ + case FCPIO_INVALID_PARAM: /* some parameter in request invalid */ + case FCPIO_REQ_NOT_SUPPORTED:/* request type is not supported */ + case FCPIO_IO_NOT_FOUND: /* requested I/O was not found */ + case FCPIO_SGL_INVALID: /* request was aborted due to sgl error */ + case FCPIO_MSS_INVALID: /* request was aborted due to mss error */ + case FCPIO_FW_ERR: /* request was terminated due fw error */ + default: + shost_printk(KERN_ERR, fnic->lport->host, "hdr status = %s\n", + fnic_fcpio_status_to_str(hdr_status)); + sc->result = (DID_ERROR << 16) | icmnd_cmpl->scsi_status; + break; + } + + /* Break link with the SCSI command */ + CMD_SP(sc) = NULL; + + spin_unlock_irqrestore(io_lock, flags); + + fnic_release_ioreq_buf(fnic, io_req, sc); + + mempool_free(io_req, fnic->io_req_pool); + + if (sc->sc_data_direction == DMA_FROM_DEVICE) { + fnic->lport->host_stats.fcp_input_requests++; + fnic->fcp_input_bytes += xfer_len; + } else if (sc->sc_data_direction == DMA_TO_DEVICE) { + fnic->lport->host_stats.fcp_output_requests++; + fnic->fcp_output_bytes += xfer_len; + } else + fnic->lport->host_stats.fcp_control_requests++; + + /* Call SCSI completion function to complete the IO */ + if (sc->scsi_done) + sc->scsi_done(sc); + +} + +/* fnic_fcpio_itmf_cmpl_handler + * Routine to handle itmf completions + */ +static void fnic_fcpio_itmf_cmpl_handler(struct fnic *fnic, + struct fcpio_fw_req *desc) +{ + u8 type; + u8 hdr_status; + struct fcpio_tag tag; + u32 id; + struct scsi_cmnd *sc; + struct fnic_io_req *io_req; + unsigned long flags; + spinlock_t *io_lock; + + fcpio_header_dec(&desc->hdr, &type, &hdr_status, &tag); + fcpio_tag_id_dec(&tag, &id); + + if ((id & FNIC_TAG_MASK) >= FNIC_MAX_IO_REQ) + return; + + sc = scsi_host_find_tag(fnic->lport->host, id & FNIC_TAG_MASK); + WARN_ON_ONCE(!sc); + if (!sc) + return; + + io_lock = fnic_io_lock_hash(fnic, sc); + spin_lock_irqsave(io_lock, flags); + io_req = (struct fnic_io_req *)CMD_SP(sc); + WARN_ON_ONCE(!io_req); + if (!io_req) { + spin_unlock_irqrestore(io_lock, flags); + return; + } + + if (id & FNIC_TAG_ABORT) { + /* Completion of abort cmd */ + if (CMD_STATE(sc) != FNIC_IOREQ_ABTS_PENDING) { + /* This is a late completion. Ignore it */ + spin_unlock_irqrestore(io_lock, flags); + return; + } + CMD_STATE(sc) = FNIC_IOREQ_ABTS_COMPLETE; + CMD_ABTS_STATUS(sc) = hdr_status; + + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "abts cmpl recd. id %d status %s\n", + (int)(id & FNIC_TAG_MASK), + fnic_fcpio_status_to_str(hdr_status)); + + /* + * If scsi_eh thread is blocked waiting for abts to complete, + * signal completion to it. IO will be cleaned in the thread + * else clean it in this context + */ + if (io_req->abts_done) { + complete(io_req->abts_done); + spin_unlock_irqrestore(io_lock, flags); + } else { + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "abts cmpl, completing IO\n"); + CMD_SP(sc) = NULL; + sc->result = (DID_ERROR << 16); + + spin_unlock_irqrestore(io_lock, flags); + + fnic_release_ioreq_buf(fnic, io_req, sc); + mempool_free(io_req, fnic->io_req_pool); + if (sc->scsi_done) + sc->scsi_done(sc); + } + + } else if (id & FNIC_TAG_DEV_RST) { + /* Completion of device reset */ + CMD_LR_STATUS(sc) = hdr_status; + CMD_STATE(sc) = FNIC_IOREQ_CMD_COMPLETE; + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "dev reset cmpl recd. id %d status %s\n", + (int)(id & FNIC_TAG_MASK), + fnic_fcpio_status_to_str(hdr_status)); + if (io_req->dr_done) + complete(io_req->dr_done); + spin_unlock_irqrestore(io_lock, flags); + + } else { + shost_printk(KERN_ERR, fnic->lport->host, + "Unexpected itmf io state %s tag %x\n", + fnic_ioreq_state_to_str(CMD_STATE(sc)), id); + spin_unlock_irqrestore(io_lock, flags); + } + +} + +/* + * fnic_fcpio_cmpl_handler + * Routine to service the cq for wq_copy + */ +static int fnic_fcpio_cmpl_handler(struct vnic_dev *vdev, + unsigned int cq_index, + struct fcpio_fw_req *desc) +{ + struct fnic *fnic = vnic_dev_priv(vdev); + int ret = 0; + + switch (desc->hdr.type) { + case FCPIO_ACK: /* fw copied copy wq desc to its queue */ + fnic_fcpio_ack_handler(fnic, cq_index, desc); + break; + + case FCPIO_ICMND_CMPL: /* fw completed a command */ + fnic_fcpio_icmnd_cmpl_handler(fnic, desc); + break; + + case FCPIO_ITMF_CMPL: /* fw completed itmf (abort cmd, lun reset)*/ + fnic_fcpio_itmf_cmpl_handler(fnic, desc); + break; + + case FCPIO_FLOGI_REG_CMPL: /* fw completed flogi_reg */ + ret = fnic_fcpio_flogi_reg_cmpl_handler(fnic, desc); + break; + + case FCPIO_RESET_CMPL: /* fw completed reset */ + ret = fnic_fcpio_fw_reset_cmpl_handler(fnic, desc); + break; + + default: + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "firmware completion type %d\n", + desc->hdr.type); + break; + } + + return ret; +} + +/* + * fnic_wq_copy_cmpl_handler + * Routine to process wq copy + */ +int fnic_wq_copy_cmpl_handler(struct fnic *fnic, int copy_work_to_do) +{ + unsigned int wq_work_done = 0; + unsigned int i, cq_index; + unsigned int cur_work_done; + + for (i = 0; i < fnic->wq_copy_count; i++) { + cq_index = i + fnic->raw_wq_count + fnic->rq_count; + cur_work_done = vnic_cq_copy_service(&fnic->cq[cq_index], + fnic_fcpio_cmpl_handler, + copy_work_to_do); + wq_work_done += cur_work_done; + } + return wq_work_done; +} + +static void fnic_cleanup_io(struct fnic *fnic, int exclude_id) +{ + unsigned int i; + struct fnic_io_req *io_req; + unsigned long flags = 0; + struct scsi_cmnd *sc; + spinlock_t *io_lock; + + for (i = 0; i < FNIC_MAX_IO_REQ; i++) { + if (i == exclude_id) + continue; + + sc = scsi_host_find_tag(fnic->lport->host, i); + if (!sc) + continue; + + io_lock = fnic_io_lock_hash(fnic, sc); + spin_lock_irqsave(io_lock, flags); + io_req = (struct fnic_io_req *)CMD_SP(sc); + if (!io_req) { + spin_unlock_irqrestore(io_lock, flags); + goto cleanup_scsi_cmd; + } + + CMD_SP(sc) = NULL; + + spin_unlock_irqrestore(io_lock, flags); + + /* + * If there is a scsi_cmnd associated with this io_req, then + * free the corresponding state + */ + fnic_release_ioreq_buf(fnic, io_req, sc); + mempool_free(io_req, fnic->io_req_pool); + +cleanup_scsi_cmd: + sc->result = DID_TRANSPORT_DISRUPTED << 16; + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, "fnic_cleanup_io:" + " DID_TRANSPORT_DISRUPTED\n"); + + /* Complete the command to SCSI */ + if (sc->scsi_done) + sc->scsi_done(sc); + } +} + +void fnic_wq_copy_cleanup_handler(struct vnic_wq_copy *wq, + struct fcpio_host_req *desc) +{ + u32 id; + struct fnic *fnic = vnic_dev_priv(wq->vdev); + struct fnic_io_req *io_req; + struct scsi_cmnd *sc; + unsigned long flags; + spinlock_t *io_lock; + + /* get the tag reference */ + fcpio_tag_id_dec(&desc->hdr.tag, &id); + id &= FNIC_TAG_MASK; + + if (id >= FNIC_MAX_IO_REQ) + return; + + sc = scsi_host_find_tag(fnic->lport->host, id); + if (!sc) + return; + + io_lock = fnic_io_lock_hash(fnic, sc); + spin_lock_irqsave(io_lock, flags); + + /* Get the IO context which this desc refers to */ + io_req = (struct fnic_io_req *)CMD_SP(sc); + + /* fnic interrupts are turned off by now */ + + if (!io_req) { + spin_unlock_irqrestore(io_lock, flags); + goto wq_copy_cleanup_scsi_cmd; + } + + CMD_SP(sc) = NULL; + + spin_unlock_irqrestore(io_lock, flags); + + fnic_release_ioreq_buf(fnic, io_req, sc); + mempool_free(io_req, fnic->io_req_pool); + +wq_copy_cleanup_scsi_cmd: + sc->result = DID_NO_CONNECT << 16; + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, "wq_copy_cleanup_handler:" + " DID_NO_CONNECT\n"); + + if (sc->scsi_done) + sc->scsi_done(sc); +} + +static inline int fnic_queue_abort_io_req(struct fnic *fnic, int tag, + u32 task_req, u8 *fc_lun, + struct fnic_io_req *io_req) +{ + struct vnic_wq_copy *wq = &fnic->wq_copy[0]; + unsigned long flags; + + spin_lock_irqsave(&fnic->wq_copy_lock[0], flags); + + if (vnic_wq_copy_desc_avail(wq) <= fnic->wq_copy_desc_low[0]) + free_wq_copy_descs(fnic, wq); + + if (!vnic_wq_copy_desc_avail(wq)) { + spin_unlock_irqrestore(&fnic->wq_copy_lock[0], flags); + return 1; + } + fnic_queue_wq_copy_desc_itmf(wq, tag | FNIC_TAG_ABORT, + 0, task_req, tag, fc_lun, io_req->port_id, + fnic->config.ra_tov, fnic->config.ed_tov); + + spin_unlock_irqrestore(&fnic->wq_copy_lock[0], flags); + return 0; +} + +void fnic_rport_exch_reset(struct fnic *fnic, u32 port_id) +{ + int tag; + struct fnic_io_req *io_req; + spinlock_t *io_lock; + unsigned long flags; + struct scsi_cmnd *sc; + struct scsi_lun fc_lun; + enum fnic_ioreq_state old_ioreq_state; + + FNIC_SCSI_DBG(KERN_DEBUG, + fnic->lport->host, + "fnic_rport_reset_exch called portid 0x%06x\n", + port_id); + + if (fnic->in_remove) + return; + + for (tag = 0; tag < FNIC_MAX_IO_REQ; tag++) { + sc = scsi_host_find_tag(fnic->lport->host, tag); + if (!sc) + continue; + + io_lock = fnic_io_lock_hash(fnic, sc); + spin_lock_irqsave(io_lock, flags); + + io_req = (struct fnic_io_req *)CMD_SP(sc); + + if (!io_req || io_req->port_id != port_id) { + spin_unlock_irqrestore(io_lock, flags); + continue; + } + + /* + * Found IO that is still pending with firmware and + * belongs to rport that went away + */ + if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) { + spin_unlock_irqrestore(io_lock, flags); + continue; + } + old_ioreq_state = CMD_STATE(sc); + CMD_STATE(sc) = FNIC_IOREQ_ABTS_PENDING; + CMD_ABTS_STATUS(sc) = FCPIO_INVALID_CODE; + + BUG_ON(io_req->abts_done); + + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "fnic_rport_reset_exch: Issuing abts\n"); + + spin_unlock_irqrestore(io_lock, flags); + + /* Now queue the abort command to firmware */ + int_to_scsilun(sc->device->lun, &fc_lun); + + if (fnic_queue_abort_io_req(fnic, tag, + FCPIO_ITMF_ABT_TASK_TERM, + fc_lun.scsi_lun, io_req)) { + /* + * Revert the cmd state back to old state, if + * it hasnt changed in between. This cmd will get + * aborted later by scsi_eh, or cleaned up during + * lun reset + */ + io_lock = fnic_io_lock_hash(fnic, sc); + + spin_lock_irqsave(io_lock, flags); + if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) + CMD_STATE(sc) = old_ioreq_state; + spin_unlock_irqrestore(io_lock, flags); + } + } + +} + +void fnic_terminate_rport_io(struct fc_rport *rport) +{ + int tag; + struct fnic_io_req *io_req; + spinlock_t *io_lock; + unsigned long flags; + struct scsi_cmnd *sc; + struct scsi_lun fc_lun; + struct fc_rport_libfc_priv *rdata = rport->dd_data; + struct fc_lport *lport = rdata->local_port; + struct fnic *fnic = lport_priv(lport); + struct fc_rport *cmd_rport; + enum fnic_ioreq_state old_ioreq_state; + + FNIC_SCSI_DBG(KERN_DEBUG, + fnic->lport->host, "fnic_terminate_rport_io called" + " wwpn 0x%llx, wwnn0x%llx, portid 0x%06x\n", + rport->port_name, rport->node_name, + rport->port_id); + + if (fnic->in_remove) + return; + + for (tag = 0; tag < FNIC_MAX_IO_REQ; tag++) { + sc = scsi_host_find_tag(fnic->lport->host, tag); + if (!sc) + continue; + + cmd_rport = starget_to_rport(scsi_target(sc->device)); + if (rport != cmd_rport) + continue; + + io_lock = fnic_io_lock_hash(fnic, sc); + spin_lock_irqsave(io_lock, flags); + + io_req = (struct fnic_io_req *)CMD_SP(sc); + + if (!io_req || rport != cmd_rport) { + spin_unlock_irqrestore(io_lock, flags); + continue; + } + + /* + * Found IO that is still pending with firmware and + * belongs to rport that went away + */ + if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) { + spin_unlock_irqrestore(io_lock, flags); + continue; + } + old_ioreq_state = CMD_STATE(sc); + CMD_STATE(sc) = FNIC_IOREQ_ABTS_PENDING; + CMD_ABTS_STATUS(sc) = FCPIO_INVALID_CODE; + + BUG_ON(io_req->abts_done); + + FNIC_SCSI_DBG(KERN_DEBUG, + fnic->lport->host, + "fnic_terminate_rport_io: Issuing abts\n"); + + spin_unlock_irqrestore(io_lock, flags); + + /* Now queue the abort command to firmware */ + int_to_scsilun(sc->device->lun, &fc_lun); + + if (fnic_queue_abort_io_req(fnic, tag, + FCPIO_ITMF_ABT_TASK_TERM, + fc_lun.scsi_lun, io_req)) { + /* + * Revert the cmd state back to old state, if + * it hasnt changed in between. This cmd will get + * aborted later by scsi_eh, or cleaned up during + * lun reset + */ + io_lock = fnic_io_lock_hash(fnic, sc); + + spin_lock_irqsave(io_lock, flags); + if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) + CMD_STATE(sc) = old_ioreq_state; + spin_unlock_irqrestore(io_lock, flags); + } + } + +} + +static void fnic_block_error_handler(struct scsi_cmnd *sc) +{ + struct Scsi_Host *shost = sc->device->host; + struct fc_rport *rport = starget_to_rport(scsi_target(sc->device)); + unsigned long flags; + + spin_lock_irqsave(shost->host_lock, flags); + while (rport->port_state == FC_PORTSTATE_BLOCKED) { + spin_unlock_irqrestore(shost->host_lock, flags); + msleep(1000); + spin_lock_irqsave(shost->host_lock, flags); + } + spin_unlock_irqrestore(shost->host_lock, flags); + +} + +/* + * This function is exported to SCSI for sending abort cmnds. + * A SCSI IO is represented by a io_req in the driver. + * The ioreq is linked to the SCSI Cmd, thus a link with the ULP's IO. + */ +int fnic_abort_cmd(struct scsi_cmnd *sc) +{ + struct fc_lport *lp; + struct fnic *fnic; + struct fnic_io_req *io_req; + struct fc_rport *rport; + spinlock_t *io_lock; + unsigned long flags; + int ret = SUCCESS; + u32 task_req; + struct scsi_lun fc_lun; + DECLARE_COMPLETION_ONSTACK(tm_done); + + /* Wait for rport to unblock */ + fnic_block_error_handler(sc); + + /* Get local-port, check ready and link up */ + lp = shost_priv(sc->device->host); + + fnic = lport_priv(lp); + FNIC_SCSI_DBG(KERN_DEBUG, + fnic->lport->host, + "Abort Cmd called FCID 0x%x, LUN 0x%x TAG %d\n", + (starget_to_rport(scsi_target(sc->device)))->port_id, + sc->device->lun, sc->request->tag); + + if (lp->state != LPORT_ST_READY || !(lp->link_up)) { + ret = FAILED; + goto fnic_abort_cmd_end; + } + + /* + * Avoid a race between SCSI issuing the abort and the device + * completing the command. + * + * If the command is already completed by the fw cmpl code, + * we just return SUCCESS from here. This means that the abort + * succeeded. In the SCSI ML, since the timeout for command has + * happened, the completion wont actually complete the command + * and it will be considered as an aborted command + * + * The CMD_SP will not be cleared except while holding io_req_lock. + */ + io_lock = fnic_io_lock_hash(fnic, sc); + spin_lock_irqsave(io_lock, flags); + io_req = (struct fnic_io_req *)CMD_SP(sc); + if (!io_req) { + spin_unlock_irqrestore(io_lock, flags); + goto fnic_abort_cmd_end; + } + + io_req->abts_done = &tm_done; + + if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) { + spin_unlock_irqrestore(io_lock, flags); + goto wait_pending; + } + /* + * Command is still pending, need to abort it + * If the firmware completes the command after this point, + * the completion wont be done till mid-layer, since abort + * has already started. + */ + CMD_STATE(sc) = FNIC_IOREQ_ABTS_PENDING; + CMD_ABTS_STATUS(sc) = FCPIO_INVALID_CODE; + + spin_unlock_irqrestore(io_lock, flags); + + /* + * Check readiness of the remote port. If the path to remote + * port is up, then send abts to the remote port to terminate + * the IO. Else, just locally terminate the IO in the firmware + */ + rport = starget_to_rport(scsi_target(sc->device)); + if (fc_remote_port_chkready(rport) == 0) + task_req = FCPIO_ITMF_ABT_TASK; + else + task_req = FCPIO_ITMF_ABT_TASK_TERM; + + /* Now queue the abort command to firmware */ + int_to_scsilun(sc->device->lun, &fc_lun); + + if (fnic_queue_abort_io_req(fnic, sc->request->tag, task_req, + fc_lun.scsi_lun, io_req)) { + spin_lock_irqsave(io_lock, flags); + io_req = (struct fnic_io_req *)CMD_SP(sc); + if (io_req) + io_req->abts_done = NULL; + spin_unlock_irqrestore(io_lock, flags); + ret = FAILED; + goto fnic_abort_cmd_end; + } + + /* + * We queued an abort IO, wait for its completion. + * Once the firmware completes the abort command, it will + * wake up this thread. + */ + wait_pending: + wait_for_completion_timeout(&tm_done, + msecs_to_jiffies + (2 * fnic->config.ra_tov + + fnic->config.ed_tov)); + + /* Check the abort status */ + spin_lock_irqsave(io_lock, flags); + + io_req = (struct fnic_io_req *)CMD_SP(sc); + if (!io_req) { + spin_unlock_irqrestore(io_lock, flags); + ret = FAILED; + goto fnic_abort_cmd_end; + } + io_req->abts_done = NULL; + + /* fw did not complete abort, timed out */ + if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) { + spin_unlock_irqrestore(io_lock, flags); + ret = FAILED; + goto fnic_abort_cmd_end; + } + + /* + * firmware completed the abort, check the status, + * free the io_req irrespective of failure or success + */ + if (CMD_ABTS_STATUS(sc) != FCPIO_SUCCESS) + ret = FAILED; + + CMD_SP(sc) = NULL; + + spin_unlock_irqrestore(io_lock, flags); + + fnic_release_ioreq_buf(fnic, io_req, sc); + mempool_free(io_req, fnic->io_req_pool); + +fnic_abort_cmd_end: + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "Returning from abort cmd %s\n", + (ret == SUCCESS) ? + "SUCCESS" : "FAILED"); + return ret; +} + +static inline int fnic_queue_dr_io_req(struct fnic *fnic, + struct scsi_cmnd *sc, + struct fnic_io_req *io_req) +{ + struct vnic_wq_copy *wq = &fnic->wq_copy[0]; + struct scsi_lun fc_lun; + int ret = 0; + unsigned long intr_flags; + + spin_lock_irqsave(&fnic->wq_copy_lock[0], intr_flags); + + if (vnic_wq_copy_desc_avail(wq) <= fnic->wq_copy_desc_low[0]) + free_wq_copy_descs(fnic, wq); + + if (!vnic_wq_copy_desc_avail(wq)) { + ret = -EAGAIN; + goto lr_io_req_end; + } + + /* fill in the lun info */ + int_to_scsilun(sc->device->lun, &fc_lun); + + fnic_queue_wq_copy_desc_itmf(wq, sc->request->tag | FNIC_TAG_DEV_RST, + 0, FCPIO_ITMF_LUN_RESET, SCSI_NO_TAG, + fc_lun.scsi_lun, io_req->port_id, + fnic->config.ra_tov, fnic->config.ed_tov); + +lr_io_req_end: + spin_unlock_irqrestore(&fnic->wq_copy_lock[0], intr_flags); + + return ret; +} + +/* + * Clean up any pending aborts on the lun + * For each outstanding IO on this lun, whose abort is not completed by fw, + * issue a local abort. Wait for abort to complete. Return 0 if all commands + * successfully aborted, 1 otherwise + */ +static int fnic_clean_pending_aborts(struct fnic *fnic, + struct scsi_cmnd *lr_sc) +{ + int tag; + struct fnic_io_req *io_req; + spinlock_t *io_lock; + unsigned long flags; + int ret = 0; + struct scsi_cmnd *sc; + struct fc_rport *rport; + struct scsi_lun fc_lun; + struct scsi_device *lun_dev = lr_sc->device; + DECLARE_COMPLETION_ONSTACK(tm_done); + + for (tag = 0; tag < FNIC_MAX_IO_REQ; tag++) { + sc = scsi_host_find_tag(fnic->lport->host, tag); + /* + * ignore this lun reset cmd or cmds that do not belong to + * this lun + */ + if (!sc || sc == lr_sc || sc->device != lun_dev) + continue; + + io_lock = fnic_io_lock_hash(fnic, sc); + spin_lock_irqsave(io_lock, flags); + + io_req = (struct fnic_io_req *)CMD_SP(sc); + + if (!io_req || sc->device != lun_dev) { + spin_unlock_irqrestore(io_lock, flags); + continue; + } + + /* + * Found IO that is still pending with firmware and + * belongs to the LUN that we are resetting + */ + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "Found IO in %s on lun\n", + fnic_ioreq_state_to_str(CMD_STATE(sc))); + + BUG_ON(CMD_STATE(sc) != FNIC_IOREQ_ABTS_PENDING); + + CMD_ABTS_STATUS(sc) = FCPIO_INVALID_CODE; + io_req->abts_done = &tm_done; + spin_unlock_irqrestore(io_lock, flags); + + /* Now queue the abort command to firmware */ + int_to_scsilun(sc->device->lun, &fc_lun); + rport = starget_to_rport(scsi_target(sc->device)); + + if (fnic_queue_abort_io_req(fnic, tag, + FCPIO_ITMF_ABT_TASK_TERM, + fc_lun.scsi_lun, io_req)) { + spin_lock_irqsave(io_lock, flags); + io_req = (struct fnic_io_req *)CMD_SP(sc); + if (io_req) + io_req->abts_done = NULL; + spin_unlock_irqrestore(io_lock, flags); + ret = 1; + goto clean_pending_aborts_end; + } + + wait_for_completion_timeout(&tm_done, + msecs_to_jiffies + (fnic->config.ed_tov)); + + /* Recheck cmd state to check if it is now aborted */ + spin_lock_irqsave(io_lock, flags); + io_req = (struct fnic_io_req *)CMD_SP(sc); + if (!io_req) { + spin_unlock_irqrestore(io_lock, flags); + ret = 1; + goto clean_pending_aborts_end; + } + + io_req->abts_done = NULL; + + /* if abort is still pending with fw, fail */ + if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) { + spin_unlock_irqrestore(io_lock, flags); + ret = 1; + goto clean_pending_aborts_end; + } + CMD_SP(sc) = NULL; + spin_unlock_irqrestore(io_lock, flags); + + fnic_release_ioreq_buf(fnic, io_req, sc); + mempool_free(io_req, fnic->io_req_pool); + } + +clean_pending_aborts_end: + return ret; +} + +/* + * SCSI Eh thread issues a Lun Reset when one or more commands on a LUN + * fail to get aborted. It calls driver's eh_device_reset with a SCSI command + * on the LUN. + */ +int fnic_device_reset(struct scsi_cmnd *sc) +{ + struct fc_lport *lp; + struct fnic *fnic; + struct fnic_io_req *io_req; + struct fc_rport *rport; + int status; + int ret = FAILED; + spinlock_t *io_lock; + unsigned long flags; + DECLARE_COMPLETION_ONSTACK(tm_done); + + /* Wait for rport to unblock */ + fnic_block_error_handler(sc); + + /* Get local-port, check ready and link up */ + lp = shost_priv(sc->device->host); + + fnic = lport_priv(lp); + FNIC_SCSI_DBG(KERN_DEBUG, + fnic->lport->host, + "Device reset called FCID 0x%x, LUN 0x%x\n", + (starget_to_rport(scsi_target(sc->device)))->port_id, + sc->device->lun); + + + if (lp->state != LPORT_ST_READY || !(lp->link_up)) + goto fnic_device_reset_end; + + /* Check if remote port up */ + rport = starget_to_rport(scsi_target(sc->device)); + if (fc_remote_port_chkready(rport)) + goto fnic_device_reset_end; + + io_lock = fnic_io_lock_hash(fnic, sc); + spin_lock_irqsave(io_lock, flags); + io_req = (struct fnic_io_req *)CMD_SP(sc); + + /* + * If there is a io_req attached to this command, then use it, + * else allocate a new one. + */ + if (!io_req) { + io_req = mempool_alloc(fnic->io_req_pool, GFP_ATOMIC); + if (!io_req) { + spin_unlock_irqrestore(io_lock, flags); + goto fnic_device_reset_end; + } + memset(io_req, 0, sizeof(*io_req)); + io_req->port_id = rport->port_id; + CMD_SP(sc) = (char *)io_req; + } + io_req->dr_done = &tm_done; + CMD_STATE(sc) = FNIC_IOREQ_CMD_PENDING; + CMD_LR_STATUS(sc) = FCPIO_INVALID_CODE; + spin_unlock_irqrestore(io_lock, flags); + + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, "TAG %d\n", + sc->request->tag); + + /* + * issue the device reset, if enqueue failed, clean up the ioreq + * and break assoc with scsi cmd + */ + if (fnic_queue_dr_io_req(fnic, sc, io_req)) { + spin_lock_irqsave(io_lock, flags); + io_req = (struct fnic_io_req *)CMD_SP(sc); + if (io_req) + io_req->dr_done = NULL; + goto fnic_device_reset_clean; + } + + /* + * Wait on the local completion for LUN reset. The io_req may be + * freed while we wait since we hold no lock. + */ + wait_for_completion_timeout(&tm_done, + msecs_to_jiffies(FNIC_LUN_RESET_TIMEOUT)); + + spin_lock_irqsave(io_lock, flags); + io_req = (struct fnic_io_req *)CMD_SP(sc); + if (!io_req) { + spin_unlock_irqrestore(io_lock, flags); + goto fnic_device_reset_end; + } + io_req->dr_done = NULL; + + status = CMD_LR_STATUS(sc); + spin_unlock_irqrestore(io_lock, flags); + + /* + * If lun reset not completed, bail out with failed. io_req + * gets cleaned up during higher levels of EH + */ + if (status == FCPIO_INVALID_CODE) { + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "Device reset timed out\n"); + goto fnic_device_reset_end; + } + + /* Completed, but not successful, clean up the io_req, return fail */ + if (status != FCPIO_SUCCESS) { + spin_lock_irqsave(io_lock, flags); + FNIC_SCSI_DBG(KERN_DEBUG, + fnic->lport->host, + "Device reset completed - failed\n"); + io_req = (struct fnic_io_req *)CMD_SP(sc); + goto fnic_device_reset_clean; + } + + /* + * Clean up any aborts on this lun that have still not + * completed. If any of these fail, then LUN reset fails. + * clean_pending_aborts cleans all cmds on this lun except + * the lun reset cmd. If all cmds get cleaned, the lun reset + * succeeds + */ + if (fnic_clean_pending_aborts(fnic, sc)) { + spin_lock_irqsave(io_lock, flags); + io_req = (struct fnic_io_req *)CMD_SP(sc); + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "Device reset failed" + " since could not abort all IOs\n"); + goto fnic_device_reset_clean; + } + + /* Clean lun reset command */ + spin_lock_irqsave(io_lock, flags); + io_req = (struct fnic_io_req *)CMD_SP(sc); + if (io_req) + /* Completed, and successful */ + ret = SUCCESS; + +fnic_device_reset_clean: + if (io_req) + CMD_SP(sc) = NULL; + + spin_unlock_irqrestore(io_lock, flags); + + if (io_req) { + fnic_release_ioreq_buf(fnic, io_req, sc); + mempool_free(io_req, fnic->io_req_pool); + } + +fnic_device_reset_end: + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "Returning from device reset %s\n", + (ret == SUCCESS) ? + "SUCCESS" : "FAILED"); + return ret; +} + +/* Clean up all IOs, clean up libFC local port */ +int fnic_reset(struct Scsi_Host *shost) +{ + struct fc_lport *lp; + struct fnic *fnic; + int ret = SUCCESS; + + lp = shost_priv(shost); + fnic = lport_priv(lp); + + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "fnic_reset called\n"); + + /* + * Reset local port, this will clean up libFC exchanges, + * reset remote port sessions, and if link is up, begin flogi + */ + if (lp->tt.lport_reset(lp)) + ret = FAILED; + + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "Returning from fnic reset %s\n", + (ret == SUCCESS) ? + "SUCCESS" : "FAILED"); + + return ret; +} + +/* + * SCSI Error handling calls driver's eh_host_reset if all prior + * error handling levels return FAILED. If host reset completes + * successfully, and if link is up, then Fabric login begins. + * + * Host Reset is the highest level of error recovery. If this fails, then + * host is offlined by SCSI. + * + */ +int fnic_host_reset(struct scsi_cmnd *sc) +{ + int ret; + unsigned long wait_host_tmo; + struct Scsi_Host *shost = sc->device->host; + struct fc_lport *lp = shost_priv(shost); + + /* + * If fnic_reset is successful, wait for fabric login to complete + * scsi-ml tries to send a TUR to every device if host reset is + * successful, so before returning to scsi, fabric should be up + */ + ret = fnic_reset(shost); + if (ret == SUCCESS) { + wait_host_tmo = jiffies + FNIC_HOST_RESET_SETTLE_TIME * HZ; + ret = FAILED; + while (time_before(jiffies, wait_host_tmo)) { + if ((lp->state == LPORT_ST_READY) && + (lp->link_up)) { + ret = SUCCESS; + break; + } + ssleep(1); + } + } + + return ret; +} + +/* + * This fxn is called from libFC when host is removed + */ +void fnic_scsi_abort_io(struct fc_lport *lp) +{ + int err = 0; + unsigned long flags; + enum fnic_state old_state; + struct fnic *fnic = lport_priv(lp); + DECLARE_COMPLETION_ONSTACK(remove_wait); + + /* Issue firmware reset for fnic, wait for reset to complete */ + spin_lock_irqsave(&fnic->fnic_lock, flags); + fnic->remove_wait = &remove_wait; + old_state = fnic->state; + fnic->state = FNIC_IN_FC_TRANS_ETH_MODE; + vnic_dev_del_addr(fnic->vdev, fnic->data_src_addr); + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + + err = fnic_fw_reset_handler(fnic); + if (err) { + spin_lock_irqsave(&fnic->fnic_lock, flags); + if (fnic->state == FNIC_IN_FC_TRANS_ETH_MODE) + fnic->state = old_state; + fnic->remove_wait = NULL; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + return; + } + + /* Wait for firmware reset to complete */ + wait_for_completion_timeout(&remove_wait, + msecs_to_jiffies(FNIC_RMDEVICE_TIMEOUT)); + + spin_lock_irqsave(&fnic->fnic_lock, flags); + fnic->remove_wait = NULL; + FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, + "fnic_scsi_abort_io %s\n", + (fnic->state == FNIC_IN_ETH_MODE) ? + "SUCCESS" : "FAILED"); + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + +} + +/* + * This fxn called from libFC to clean up driver IO state on link down + */ +void fnic_scsi_cleanup(struct fc_lport *lp) +{ + unsigned long flags; + enum fnic_state old_state; + struct fnic *fnic = lport_priv(lp); + + /* issue fw reset */ + spin_lock_irqsave(&fnic->fnic_lock, flags); + old_state = fnic->state; + fnic->state = FNIC_IN_FC_TRANS_ETH_MODE; + vnic_dev_del_addr(fnic->vdev, fnic->data_src_addr); + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + + if (fnic_fw_reset_handler(fnic)) { + spin_lock_irqsave(&fnic->fnic_lock, flags); + if (fnic->state == FNIC_IN_FC_TRANS_ETH_MODE) + fnic->state = old_state; + spin_unlock_irqrestore(&fnic->fnic_lock, flags); + } + +} + +void fnic_empty_scsi_cleanup(struct fc_lport *lp) +{ +} + +void fnic_exch_mgr_reset(struct fc_lport *lp, u32 sid, u32 did) +{ + struct fnic *fnic = lport_priv(lp); + + /* Non-zero sid, nothing to do */ + if (sid) + goto call_fc_exch_mgr_reset; + + if (did) { + fnic_rport_exch_reset(fnic, did); + goto call_fc_exch_mgr_reset; + } + + /* + * sid = 0, did = 0 + * link down or device being removed + */ + if (!fnic->in_remove) + fnic_scsi_cleanup(lp); + else + fnic_scsi_abort_io(lp); + + /* call libFC exch mgr reset to reset its exchanges */ +call_fc_exch_mgr_reset: + fc_exch_mgr_reset(lp, sid, did); + +} diff --git a/drivers/scsi/fnic/rq_enet_desc.h b/drivers/scsi/fnic/rq_enet_desc.h new file mode 100644 index 000000000000..92e80ae6b725 --- /dev/null +++ b/drivers/scsi/fnic/rq_enet_desc.h @@ -0,0 +1,58 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _RQ_ENET_DESC_H_ +#define _RQ_ENET_DESC_H_ + +/* Ethernet receive queue descriptor: 16B */ +struct rq_enet_desc { + __le64 address; + __le16 length_type; + u8 reserved[6]; +}; + +enum rq_enet_type_types { + RQ_ENET_TYPE_ONLY_SOP = 0, + RQ_ENET_TYPE_NOT_SOP = 1, + RQ_ENET_TYPE_RESV2 = 2, + RQ_ENET_TYPE_RESV3 = 3, +}; + +#define RQ_ENET_ADDR_BITS 64 +#define RQ_ENET_LEN_BITS 14 +#define RQ_ENET_LEN_MASK ((1 << RQ_ENET_LEN_BITS) - 1) +#define RQ_ENET_TYPE_BITS 2 +#define RQ_ENET_TYPE_MASK ((1 << RQ_ENET_TYPE_BITS) - 1) + +static inline void rq_enet_desc_enc(struct rq_enet_desc *desc, + u64 address, u8 type, u16 length) +{ + desc->address = cpu_to_le64(address); + desc->length_type = cpu_to_le16((length & RQ_ENET_LEN_MASK) | + ((type & RQ_ENET_TYPE_MASK) << RQ_ENET_LEN_BITS)); +} + +static inline void rq_enet_desc_dec(struct rq_enet_desc *desc, + u64 *address, u8 *type, u16 *length) +{ + *address = le64_to_cpu(desc->address); + *length = le16_to_cpu(desc->length_type) & RQ_ENET_LEN_MASK; + *type = (u8)((le16_to_cpu(desc->length_type) >> RQ_ENET_LEN_BITS) & + RQ_ENET_TYPE_MASK); +} + +#endif /* _RQ_ENET_DESC_H_ */ diff --git a/drivers/scsi/fnic/vnic_cq.c b/drivers/scsi/fnic/vnic_cq.c new file mode 100644 index 000000000000..c5db32eda5ef --- /dev/null +++ b/drivers/scsi/fnic/vnic_cq.c @@ -0,0 +1,85 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#include +#include +#include +#include "vnic_dev.h" +#include "vnic_cq.h" + +void vnic_cq_free(struct vnic_cq *cq) +{ + vnic_dev_free_desc_ring(cq->vdev, &cq->ring); + + cq->ctrl = NULL; +} + +int vnic_cq_alloc(struct vnic_dev *vdev, struct vnic_cq *cq, unsigned int index, + unsigned int desc_count, unsigned int desc_size) +{ + int err; + + cq->index = index; + cq->vdev = vdev; + + cq->ctrl = vnic_dev_get_res(vdev, RES_TYPE_CQ, index); + if (!cq->ctrl) { + printk(KERN_ERR "Failed to hook CQ[%d] resource\n", index); + return -EINVAL; + } + + err = vnic_dev_alloc_desc_ring(vdev, &cq->ring, desc_count, desc_size); + if (err) + return err; + + return 0; +} + +void vnic_cq_init(struct vnic_cq *cq, unsigned int flow_control_enable, + unsigned int color_enable, unsigned int cq_head, unsigned int cq_tail, + unsigned int cq_tail_color, unsigned int interrupt_enable, + unsigned int cq_entry_enable, unsigned int cq_message_enable, + unsigned int interrupt_offset, u64 cq_message_addr) +{ + u64 paddr; + + paddr = (u64)cq->ring.base_addr | VNIC_PADDR_TARGET; + writeq(paddr, &cq->ctrl->ring_base); + iowrite32(cq->ring.desc_count, &cq->ctrl->ring_size); + iowrite32(flow_control_enable, &cq->ctrl->flow_control_enable); + iowrite32(color_enable, &cq->ctrl->color_enable); + iowrite32(cq_head, &cq->ctrl->cq_head); + iowrite32(cq_tail, &cq->ctrl->cq_tail); + iowrite32(cq_tail_color, &cq->ctrl->cq_tail_color); + iowrite32(interrupt_enable, &cq->ctrl->interrupt_enable); + iowrite32(cq_entry_enable, &cq->ctrl->cq_entry_enable); + iowrite32(cq_message_enable, &cq->ctrl->cq_message_enable); + iowrite32(interrupt_offset, &cq->ctrl->interrupt_offset); + writeq(cq_message_addr, &cq->ctrl->cq_message_addr); +} + +void vnic_cq_clean(struct vnic_cq *cq) +{ + cq->to_clean = 0; + cq->last_color = 0; + + iowrite32(0, &cq->ctrl->cq_head); + iowrite32(0, &cq->ctrl->cq_tail); + iowrite32(1, &cq->ctrl->cq_tail_color); + + vnic_dev_clear_desc_ring(&cq->ring); +} diff --git a/drivers/scsi/fnic/vnic_cq.h b/drivers/scsi/fnic/vnic_cq.h new file mode 100644 index 000000000000..4ede6809fb1e --- /dev/null +++ b/drivers/scsi/fnic/vnic_cq.h @@ -0,0 +1,121 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _VNIC_CQ_H_ +#define _VNIC_CQ_H_ + +#include "cq_desc.h" +#include "vnic_dev.h" + +/* + * These defines avoid symbol clash between fnic and enic (Cisco 10G Eth + * Driver) when both are built with CONFIG options =y + */ +#define vnic_cq_service fnic_cq_service +#define vnic_cq_free fnic_cq_free +#define vnic_cq_alloc fnic_cq_alloc +#define vnic_cq_init fnic_cq_init +#define vnic_cq_clean fnic_cq_clean + +/* Completion queue control */ +struct vnic_cq_ctrl { + u64 ring_base; /* 0x00 */ + u32 ring_size; /* 0x08 */ + u32 pad0; + u32 flow_control_enable; /* 0x10 */ + u32 pad1; + u32 color_enable; /* 0x18 */ + u32 pad2; + u32 cq_head; /* 0x20 */ + u32 pad3; + u32 cq_tail; /* 0x28 */ + u32 pad4; + u32 cq_tail_color; /* 0x30 */ + u32 pad5; + u32 interrupt_enable; /* 0x38 */ + u32 pad6; + u32 cq_entry_enable; /* 0x40 */ + u32 pad7; + u32 cq_message_enable; /* 0x48 */ + u32 pad8; + u32 interrupt_offset; /* 0x50 */ + u32 pad9; + u64 cq_message_addr; /* 0x58 */ + u32 pad10; +}; + +struct vnic_cq { + unsigned int index; + struct vnic_dev *vdev; + struct vnic_cq_ctrl __iomem *ctrl; /* memory-mapped */ + struct vnic_dev_ring ring; + unsigned int to_clean; + unsigned int last_color; +}; + +static inline unsigned int vnic_cq_service(struct vnic_cq *cq, + unsigned int work_to_do, + int (*q_service)(struct vnic_dev *vdev, struct cq_desc *cq_desc, + u8 type, u16 q_number, u16 completed_index, void *opaque), + void *opaque) +{ + struct cq_desc *cq_desc; + unsigned int work_done = 0; + u16 q_number, completed_index; + u8 type, color; + + cq_desc = (struct cq_desc *)((u8 *)cq->ring.descs + + cq->ring.desc_size * cq->to_clean); + cq_desc_dec(cq_desc, &type, &color, + &q_number, &completed_index); + + while (color != cq->last_color) { + + if ((*q_service)(cq->vdev, cq_desc, type, + q_number, completed_index, opaque)) + break; + + cq->to_clean++; + if (cq->to_clean == cq->ring.desc_count) { + cq->to_clean = 0; + cq->last_color = cq->last_color ? 0 : 1; + } + + cq_desc = (struct cq_desc *)((u8 *)cq->ring.descs + + cq->ring.desc_size * cq->to_clean); + cq_desc_dec(cq_desc, &type, &color, + &q_number, &completed_index); + + work_done++; + if (work_done >= work_to_do) + break; + } + + return work_done; +} + +void vnic_cq_free(struct vnic_cq *cq); +int vnic_cq_alloc(struct vnic_dev *vdev, struct vnic_cq *cq, unsigned int index, + unsigned int desc_count, unsigned int desc_size); +void vnic_cq_init(struct vnic_cq *cq, unsigned int flow_control_enable, + unsigned int color_enable, unsigned int cq_head, unsigned int cq_tail, + unsigned int cq_tail_color, unsigned int interrupt_enable, + unsigned int cq_entry_enable, unsigned int message_enable, + unsigned int interrupt_offset, u64 message_addr); +void vnic_cq_clean(struct vnic_cq *cq); + +#endif /* _VNIC_CQ_H_ */ diff --git a/drivers/scsi/fnic/vnic_cq_copy.h b/drivers/scsi/fnic/vnic_cq_copy.h new file mode 100644 index 000000000000..7901ce255a81 --- /dev/null +++ b/drivers/scsi/fnic/vnic_cq_copy.h @@ -0,0 +1,62 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _VNIC_CQ_COPY_H_ +#define _VNIC_CQ_COPY_H_ + +#include "fcpio.h" + +static inline unsigned int vnic_cq_copy_service( + struct vnic_cq *cq, + int (*q_service)(struct vnic_dev *vdev, + unsigned int index, + struct fcpio_fw_req *desc), + unsigned int work_to_do) + +{ + struct fcpio_fw_req *desc; + unsigned int work_done = 0; + u8 color; + + desc = (struct fcpio_fw_req *)((u8 *)cq->ring.descs + + cq->ring.desc_size * cq->to_clean); + fcpio_color_dec(desc, &color); + + while (color != cq->last_color) { + + if ((*q_service)(cq->vdev, cq->index, desc)) + break; + + cq->to_clean++; + if (cq->to_clean == cq->ring.desc_count) { + cq->to_clean = 0; + cq->last_color = cq->last_color ? 0 : 1; + } + + desc = (struct fcpio_fw_req *)((u8 *)cq->ring.descs + + cq->ring.desc_size * cq->to_clean); + fcpio_color_dec(desc, &color); + + work_done++; + if (work_done >= work_to_do) + break; + } + + return work_done; +} + +#endif /* _VNIC_CQ_COPY_H_ */ diff --git a/drivers/scsi/fnic/vnic_dev.c b/drivers/scsi/fnic/vnic_dev.c new file mode 100644 index 000000000000..566770645086 --- /dev/null +++ b/drivers/scsi/fnic/vnic_dev.c @@ -0,0 +1,690 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include +#include +#include +#include +#include "vnic_resource.h" +#include "vnic_devcmd.h" +#include "vnic_dev.h" +#include "vnic_stats.h" + +struct vnic_res { + void __iomem *vaddr; + unsigned int count; +}; + +struct vnic_dev { + void *priv; + struct pci_dev *pdev; + struct vnic_res res[RES_TYPE_MAX]; + enum vnic_dev_intr_mode intr_mode; + struct vnic_devcmd __iomem *devcmd; + struct vnic_devcmd_notify *notify; + struct vnic_devcmd_notify notify_copy; + dma_addr_t notify_pa; + u32 *linkstatus; + dma_addr_t linkstatus_pa; + struct vnic_stats *stats; + dma_addr_t stats_pa; + struct vnic_devcmd_fw_info *fw_info; + dma_addr_t fw_info_pa; +}; + +#define VNIC_MAX_RES_HDR_SIZE \ + (sizeof(struct vnic_resource_header) + \ + sizeof(struct vnic_resource) * RES_TYPE_MAX) +#define VNIC_RES_STRIDE 128 + +void *vnic_dev_priv(struct vnic_dev *vdev) +{ + return vdev->priv; +} + +static int vnic_dev_discover_res(struct vnic_dev *vdev, + struct vnic_dev_bar *bar) +{ + struct vnic_resource_header __iomem *rh; + struct vnic_resource __iomem *r; + u8 type; + + if (bar->len < VNIC_MAX_RES_HDR_SIZE) { + printk(KERN_ERR "vNIC BAR0 res hdr length error\n"); + return -EINVAL; + } + + rh = bar->vaddr; + if (!rh) { + printk(KERN_ERR "vNIC BAR0 res hdr not mem-mapped\n"); + return -EINVAL; + } + + if (ioread32(&rh->magic) != VNIC_RES_MAGIC || + ioread32(&rh->version) != VNIC_RES_VERSION) { + printk(KERN_ERR "vNIC BAR0 res magic/version error " + "exp (%lx/%lx) curr (%x/%x)\n", + VNIC_RES_MAGIC, VNIC_RES_VERSION, + ioread32(&rh->magic), ioread32(&rh->version)); + return -EINVAL; + } + + r = (struct vnic_resource __iomem *)(rh + 1); + + while ((type = ioread8(&r->type)) != RES_TYPE_EOL) { + + u8 bar_num = ioread8(&r->bar); + u32 bar_offset = ioread32(&r->bar_offset); + u32 count = ioread32(&r->count); + u32 len; + + r++; + + if (bar_num != 0) /* only mapping in BAR0 resources */ + continue; + + switch (type) { + case RES_TYPE_WQ: + case RES_TYPE_RQ: + case RES_TYPE_CQ: + case RES_TYPE_INTR_CTRL: + /* each count is stride bytes long */ + len = count * VNIC_RES_STRIDE; + if (len + bar_offset > bar->len) { + printk(KERN_ERR "vNIC BAR0 resource %d " + "out-of-bounds, offset 0x%x + " + "size 0x%x > bar len 0x%lx\n", + type, bar_offset, + len, + bar->len); + return -EINVAL; + } + break; + case RES_TYPE_INTR_PBA_LEGACY: + case RES_TYPE_DEVCMD: + len = count; + break; + default: + continue; + } + + vdev->res[type].count = count; + vdev->res[type].vaddr = (char __iomem *)bar->vaddr + bar_offset; + } + + return 0; +} + +unsigned int vnic_dev_get_res_count(struct vnic_dev *vdev, + enum vnic_res_type type) +{ + return vdev->res[type].count; +} + +void __iomem *vnic_dev_get_res(struct vnic_dev *vdev, enum vnic_res_type type, + unsigned int index) +{ + if (!vdev->res[type].vaddr) + return NULL; + + switch (type) { + case RES_TYPE_WQ: + case RES_TYPE_RQ: + case RES_TYPE_CQ: + case RES_TYPE_INTR_CTRL: + return (char __iomem *)vdev->res[type].vaddr + + index * VNIC_RES_STRIDE; + default: + return (char __iomem *)vdev->res[type].vaddr; + } +} + +unsigned int vnic_dev_desc_ring_size(struct vnic_dev_ring *ring, + unsigned int desc_count, + unsigned int desc_size) +{ + /* The base address of the desc rings must be 512 byte aligned. + * Descriptor count is aligned to groups of 32 descriptors. A + * count of 0 means the maximum 4096 descriptors. Descriptor + * size is aligned to 16 bytes. + */ + + unsigned int count_align = 32; + unsigned int desc_align = 16; + + ring->base_align = 512; + + if (desc_count == 0) + desc_count = 4096; + + ring->desc_count = ALIGN(desc_count, count_align); + + ring->desc_size = ALIGN(desc_size, desc_align); + + ring->size = ring->desc_count * ring->desc_size; + ring->size_unaligned = ring->size + ring->base_align; + + return ring->size_unaligned; +} + +void vnic_dev_clear_desc_ring(struct vnic_dev_ring *ring) +{ + memset(ring->descs, 0, ring->size); +} + +int vnic_dev_alloc_desc_ring(struct vnic_dev *vdev, struct vnic_dev_ring *ring, + unsigned int desc_count, unsigned int desc_size) +{ + vnic_dev_desc_ring_size(ring, desc_count, desc_size); + + ring->descs_unaligned = pci_alloc_consistent(vdev->pdev, + ring->size_unaligned, + &ring->base_addr_unaligned); + + if (!ring->descs_unaligned) { + printk(KERN_ERR + "Failed to allocate ring (size=%d), aborting\n", + (int)ring->size); + return -ENOMEM; + } + + ring->base_addr = ALIGN(ring->base_addr_unaligned, + ring->base_align); + ring->descs = (u8 *)ring->descs_unaligned + + (ring->base_addr - ring->base_addr_unaligned); + + vnic_dev_clear_desc_ring(ring); + + ring->desc_avail = ring->desc_count - 1; + + return 0; +} + +void vnic_dev_free_desc_ring(struct vnic_dev *vdev, struct vnic_dev_ring *ring) +{ + if (ring->descs) { + pci_free_consistent(vdev->pdev, + ring->size_unaligned, + ring->descs_unaligned, + ring->base_addr_unaligned); + ring->descs = NULL; + } +} + +int vnic_dev_cmd(struct vnic_dev *vdev, enum vnic_devcmd_cmd cmd, + u64 *a0, u64 *a1, int wait) +{ + struct vnic_devcmd __iomem *devcmd = vdev->devcmd; + int delay; + u32 status; + int dev_cmd_err[] = { + /* convert from fw's version of error.h to host's version */ + 0, /* ERR_SUCCESS */ + EINVAL, /* ERR_EINVAL */ + EFAULT, /* ERR_EFAULT */ + EPERM, /* ERR_EPERM */ + EBUSY, /* ERR_EBUSY */ + }; + int err; + + status = ioread32(&devcmd->status); + if (status & STAT_BUSY) { + printk(KERN_ERR "Busy devcmd %d\n", _CMD_N(cmd)); + return -EBUSY; + } + + if (_CMD_DIR(cmd) & _CMD_DIR_WRITE) { + writeq(*a0, &devcmd->args[0]); + writeq(*a1, &devcmd->args[1]); + wmb(); + } + + iowrite32(cmd, &devcmd->cmd); + + if ((_CMD_FLAGS(cmd) & _CMD_FLAGS_NOWAIT)) + return 0; + + for (delay = 0; delay < wait; delay++) { + + udelay(100); + + status = ioread32(&devcmd->status); + if (!(status & STAT_BUSY)) { + + if (status & STAT_ERROR) { + err = dev_cmd_err[(int)readq(&devcmd->args[0])]; + printk(KERN_ERR "Error %d devcmd %d\n", + err, _CMD_N(cmd)); + return -err; + } + + if (_CMD_DIR(cmd) & _CMD_DIR_READ) { + rmb(); + *a0 = readq(&devcmd->args[0]); + *a1 = readq(&devcmd->args[1]); + } + + return 0; + } + } + + printk(KERN_ERR "Timedout devcmd %d\n", _CMD_N(cmd)); + return -ETIMEDOUT; +} + +int vnic_dev_fw_info(struct vnic_dev *vdev, + struct vnic_devcmd_fw_info **fw_info) +{ + u64 a0, a1 = 0; + int wait = 1000; + int err = 0; + + if (!vdev->fw_info) { + vdev->fw_info = pci_alloc_consistent(vdev->pdev, + sizeof(struct vnic_devcmd_fw_info), + &vdev->fw_info_pa); + if (!vdev->fw_info) + return -ENOMEM; + + a0 = vdev->fw_info_pa; + + /* only get fw_info once and cache it */ + err = vnic_dev_cmd(vdev, CMD_MCPU_FW_INFO, &a0, &a1, wait); + } + + *fw_info = vdev->fw_info; + + return err; +} + +int vnic_dev_spec(struct vnic_dev *vdev, unsigned int offset, unsigned int size, + void *value) +{ + u64 a0, a1; + int wait = 1000; + int err; + + a0 = offset; + a1 = size; + + err = vnic_dev_cmd(vdev, CMD_DEV_SPEC, &a0, &a1, wait); + + switch (size) { + case 1: + *(u8 *)value = (u8)a0; + break; + case 2: + *(u16 *)value = (u16)a0; + break; + case 4: + *(u32 *)value = (u32)a0; + break; + case 8: + *(u64 *)value = a0; + break; + default: + BUG(); + break; + } + + return err; +} + +int vnic_dev_stats_clear(struct vnic_dev *vdev) +{ + u64 a0 = 0, a1 = 0; + int wait = 1000; + return vnic_dev_cmd(vdev, CMD_STATS_CLEAR, &a0, &a1, wait); +} + +int vnic_dev_stats_dump(struct vnic_dev *vdev, struct vnic_stats **stats) +{ + u64 a0, a1; + int wait = 1000; + + if (!vdev->stats) { + vdev->stats = pci_alloc_consistent(vdev->pdev, + sizeof(struct vnic_stats), &vdev->stats_pa); + if (!vdev->stats) + return -ENOMEM; + } + + *stats = vdev->stats; + a0 = vdev->stats_pa; + a1 = sizeof(struct vnic_stats); + + return vnic_dev_cmd(vdev, CMD_STATS_DUMP, &a0, &a1, wait); +} + +int vnic_dev_close(struct vnic_dev *vdev) +{ + u64 a0 = 0, a1 = 0; + int wait = 1000; + return vnic_dev_cmd(vdev, CMD_CLOSE, &a0, &a1, wait); +} + +int vnic_dev_enable(struct vnic_dev *vdev) +{ + u64 a0 = 0, a1 = 0; + int wait = 1000; + return vnic_dev_cmd(vdev, CMD_ENABLE, &a0, &a1, wait); +} + +int vnic_dev_disable(struct vnic_dev *vdev) +{ + u64 a0 = 0, a1 = 0; + int wait = 1000; + return vnic_dev_cmd(vdev, CMD_DISABLE, &a0, &a1, wait); +} + +int vnic_dev_open(struct vnic_dev *vdev, int arg) +{ + u64 a0 = (u32)arg, a1 = 0; + int wait = 1000; + return vnic_dev_cmd(vdev, CMD_OPEN, &a0, &a1, wait); +} + +int vnic_dev_open_done(struct vnic_dev *vdev, int *done) +{ + u64 a0 = 0, a1 = 0; + int wait = 1000; + int err; + + *done = 0; + + err = vnic_dev_cmd(vdev, CMD_OPEN_STATUS, &a0, &a1, wait); + if (err) + return err; + + *done = (a0 == 0); + + return 0; +} + +int vnic_dev_soft_reset(struct vnic_dev *vdev, int arg) +{ + u64 a0 = (u32)arg, a1 = 0; + int wait = 1000; + return vnic_dev_cmd(vdev, CMD_SOFT_RESET, &a0, &a1, wait); +} + +int vnic_dev_soft_reset_done(struct vnic_dev *vdev, int *done) +{ + u64 a0 = 0, a1 = 0; + int wait = 1000; + int err; + + *done = 0; + + err = vnic_dev_cmd(vdev, CMD_SOFT_RESET_STATUS, &a0, &a1, wait); + if (err) + return err; + + *done = (a0 == 0); + + return 0; +} + +int vnic_dev_hang_notify(struct vnic_dev *vdev) +{ + u64 a0, a1; + int wait = 1000; + return vnic_dev_cmd(vdev, CMD_HANG_NOTIFY, &a0, &a1, wait); +} + +int vnic_dev_mac_addr(struct vnic_dev *vdev, u8 *mac_addr) +{ + u64 a0, a1; + int wait = 1000; + int err, i; + + for (i = 0; i < ETH_ALEN; i++) + mac_addr[i] = 0; + + err = vnic_dev_cmd(vdev, CMD_MAC_ADDR, &a0, &a1, wait); + if (err) + return err; + + for (i = 0; i < ETH_ALEN; i++) + mac_addr[i] = ((u8 *)&a0)[i]; + + return 0; +} + +void vnic_dev_packet_filter(struct vnic_dev *vdev, int directed, int multicast, + int broadcast, int promisc, int allmulti) +{ + u64 a0, a1 = 0; + int wait = 1000; + int err; + + a0 = (directed ? CMD_PFILTER_DIRECTED : 0) | + (multicast ? CMD_PFILTER_MULTICAST : 0) | + (broadcast ? CMD_PFILTER_BROADCAST : 0) | + (promisc ? CMD_PFILTER_PROMISCUOUS : 0) | + (allmulti ? CMD_PFILTER_ALL_MULTICAST : 0); + + err = vnic_dev_cmd(vdev, CMD_PACKET_FILTER, &a0, &a1, wait); + if (err) + printk(KERN_ERR "Can't set packet filter\n"); +} + +void vnic_dev_add_addr(struct vnic_dev *vdev, u8 *addr) +{ + u64 a0 = 0, a1 = 0; + int wait = 1000; + int err; + int i; + + for (i = 0; i < ETH_ALEN; i++) + ((u8 *)&a0)[i] = addr[i]; + + err = vnic_dev_cmd(vdev, CMD_ADDR_ADD, &a0, &a1, wait); + if (err) + printk(KERN_ERR + "Can't add addr [%02x:%02x:%02x:%02x:%02x:%02x], %d\n", + addr[0], addr[1], addr[2], addr[3], addr[4], addr[5], + err); +} + +void vnic_dev_del_addr(struct vnic_dev *vdev, u8 *addr) +{ + u64 a0 = 0, a1 = 0; + int wait = 1000; + int err; + int i; + + for (i = 0; i < ETH_ALEN; i++) + ((u8 *)&a0)[i] = addr[i]; + + err = vnic_dev_cmd(vdev, CMD_ADDR_DEL, &a0, &a1, wait); + if (err) + printk(KERN_ERR + "Can't del addr [%02x:%02x:%02x:%02x:%02x:%02x], %d\n", + addr[0], addr[1], addr[2], addr[3], addr[4], addr[5], + err); +} + +int vnic_dev_notify_set(struct vnic_dev *vdev, u16 intr) +{ + u64 a0, a1; + int wait = 1000; + + if (!vdev->notify) { + vdev->notify = pci_alloc_consistent(vdev->pdev, + sizeof(struct vnic_devcmd_notify), + &vdev->notify_pa); + if (!vdev->notify) + return -ENOMEM; + } + + a0 = vdev->notify_pa; + a1 = ((u64)intr << 32) & 0x0000ffff00000000ULL; + a1 += sizeof(struct vnic_devcmd_notify); + + return vnic_dev_cmd(vdev, CMD_NOTIFY, &a0, &a1, wait); +} + +void vnic_dev_notify_unset(struct vnic_dev *vdev) +{ + u64 a0, a1; + int wait = 1000; + + a0 = 0; /* paddr = 0 to unset notify buffer */ + a1 = 0x0000ffff00000000ULL; /* intr num = -1 to unreg for intr */ + a1 += sizeof(struct vnic_devcmd_notify); + + vnic_dev_cmd(vdev, CMD_NOTIFY, &a0, &a1, wait); +} + +static int vnic_dev_notify_ready(struct vnic_dev *vdev) +{ + u32 *words; + unsigned int nwords = sizeof(struct vnic_devcmd_notify) / 4; + unsigned int i; + u32 csum; + + if (!vdev->notify) + return 0; + + do { + csum = 0; + memcpy(&vdev->notify_copy, vdev->notify, + sizeof(struct vnic_devcmd_notify)); + words = (u32 *)&vdev->notify_copy; + for (i = 1; i < nwords; i++) + csum += words[i]; + } while (csum != words[0]); + + return 1; +} + +int vnic_dev_init(struct vnic_dev *vdev, int arg) +{ + u64 a0 = (u32)arg, a1 = 0; + int wait = 1000; + return vnic_dev_cmd(vdev, CMD_INIT, &a0, &a1, wait); +} + +int vnic_dev_link_status(struct vnic_dev *vdev) +{ + if (vdev->linkstatus) + return *vdev->linkstatus; + + if (!vnic_dev_notify_ready(vdev)) + return 0; + + return vdev->notify_copy.link_state; +} + +u32 vnic_dev_port_speed(struct vnic_dev *vdev) +{ + if (!vnic_dev_notify_ready(vdev)) + return 0; + + return vdev->notify_copy.port_speed; +} + +u32 vnic_dev_msg_lvl(struct vnic_dev *vdev) +{ + if (!vnic_dev_notify_ready(vdev)) + return 0; + + return vdev->notify_copy.msglvl; +} + +u32 vnic_dev_mtu(struct vnic_dev *vdev) +{ + if (!vnic_dev_notify_ready(vdev)) + return 0; + + return vdev->notify_copy.mtu; +} + +u32 vnic_dev_link_down_cnt(struct vnic_dev *vdev) +{ + if (!vnic_dev_notify_ready(vdev)) + return 0; + + return vdev->notify_copy.link_down_cnt; +} + +void vnic_dev_set_intr_mode(struct vnic_dev *vdev, + enum vnic_dev_intr_mode intr_mode) +{ + vdev->intr_mode = intr_mode; +} + +enum vnic_dev_intr_mode vnic_dev_get_intr_mode( + struct vnic_dev *vdev) +{ + return vdev->intr_mode; +} + +void vnic_dev_unregister(struct vnic_dev *vdev) +{ + if (vdev) { + if (vdev->notify) + pci_free_consistent(vdev->pdev, + sizeof(struct vnic_devcmd_notify), + vdev->notify, + vdev->notify_pa); + if (vdev->linkstatus) + pci_free_consistent(vdev->pdev, + sizeof(u32), + vdev->linkstatus, + vdev->linkstatus_pa); + if (vdev->stats) + pci_free_consistent(vdev->pdev, + sizeof(struct vnic_dev), + vdev->stats, vdev->stats_pa); + if (vdev->fw_info) + pci_free_consistent(vdev->pdev, + sizeof(struct vnic_devcmd_fw_info), + vdev->fw_info, vdev->fw_info_pa); + kfree(vdev); + } +} + +struct vnic_dev *vnic_dev_register(struct vnic_dev *vdev, + void *priv, struct pci_dev *pdev, struct vnic_dev_bar *bar) +{ + if (!vdev) { + vdev = kzalloc(sizeof(struct vnic_dev), GFP_KERNEL); + if (!vdev) + return NULL; + } + + vdev->priv = priv; + vdev->pdev = pdev; + + if (vnic_dev_discover_res(vdev, bar)) + goto err_out; + + vdev->devcmd = vnic_dev_get_res(vdev, RES_TYPE_DEVCMD, 0); + if (!vdev->devcmd) + goto err_out; + + return vdev; + +err_out: + vnic_dev_unregister(vdev); + return NULL; +} diff --git a/drivers/scsi/fnic/vnic_dev.h b/drivers/scsi/fnic/vnic_dev.h new file mode 100644 index 000000000000..f9935a8a5a09 --- /dev/null +++ b/drivers/scsi/fnic/vnic_dev.h @@ -0,0 +1,161 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _VNIC_DEV_H_ +#define _VNIC_DEV_H_ + +#include "vnic_resource.h" +#include "vnic_devcmd.h" + +/* + * These defines avoid symbol clash between fnic and enic (Cisco 10G Eth + * Driver) when both are built with CONFIG options =y + */ +#define vnic_dev_priv fnic_dev_priv +#define vnic_dev_get_res_count fnic_dev_get_res_count +#define vnic_dev_get_res fnic_dev_get_res +#define vnic_dev_desc_ring_size fnic_dev_desc_ring_siz +#define vnic_dev_clear_desc_ring fnic_dev_clear_desc_ring +#define vnic_dev_alloc_desc_ring fnic_dev_alloc_desc_ring +#define vnic_dev_free_desc_ring fnic_dev_free_desc_ring +#define vnic_dev_cmd fnic_dev_cmd +#define vnic_dev_fw_info fnic_dev_fw_info +#define vnic_dev_spec fnic_dev_spec +#define vnic_dev_stats_clear fnic_dev_stats_clear +#define vnic_dev_stats_dump fnic_dev_stats_dump +#define vnic_dev_hang_notify fnic_dev_hang_notify +#define vnic_dev_packet_filter fnic_dev_packet_filter +#define vnic_dev_add_addr fnic_dev_add_addr +#define vnic_dev_del_addr fnic_dev_del_addr +#define vnic_dev_mac_addr fnic_dev_mac_addr +#define vnic_dev_notify_set fnic_dev_notify_set +#define vnic_dev_notify_unset fnic_dev_notify_unset +#define vnic_dev_link_status fnic_dev_link_status +#define vnic_dev_port_speed fnic_dev_port_speed +#define vnic_dev_msg_lvl fnic_dev_msg_lvl +#define vnic_dev_mtu fnic_dev_mtu +#define vnic_dev_link_down_cnt fnic_dev_link_down_cnt +#define vnic_dev_close fnic_dev_close +#define vnic_dev_enable fnic_dev_enable +#define vnic_dev_disable fnic_dev_disable +#define vnic_dev_open fnic_dev_open +#define vnic_dev_open_done fnic_dev_open_done +#define vnic_dev_init fnic_dev_init +#define vnic_dev_soft_reset fnic_dev_soft_reset +#define vnic_dev_soft_reset_done fnic_dev_soft_reset_done +#define vnic_dev_set_intr_mode fnic_dev_set_intr_mode +#define vnic_dev_get_intr_mode fnic_dev_get_intr_mode +#define vnic_dev_unregister fnic_dev_unregister +#define vnic_dev_register fnic_dev_register + +#ifndef VNIC_PADDR_TARGET +#define VNIC_PADDR_TARGET 0x0000000000000000ULL +#endif + +#ifndef readq +static inline u64 readq(void __iomem *reg) +{ + return ((u64)readl(reg + 0x4UL) << 32) | (u64)readl(reg); +} + +static inline void writeq(u64 val, void __iomem *reg) +{ + writel(val & 0xffffffff, reg); + writel(val >> 32, reg + 0x4UL); +} +#endif + +enum vnic_dev_intr_mode { + VNIC_DEV_INTR_MODE_UNKNOWN, + VNIC_DEV_INTR_MODE_INTX, + VNIC_DEV_INTR_MODE_MSI, + VNIC_DEV_INTR_MODE_MSIX, +}; + +struct vnic_dev_bar { + void __iomem *vaddr; + dma_addr_t bus_addr; + unsigned long len; +}; + +struct vnic_dev_ring { + void *descs; + size_t size; + dma_addr_t base_addr; + size_t base_align; + void *descs_unaligned; + size_t size_unaligned; + dma_addr_t base_addr_unaligned; + unsigned int desc_size; + unsigned int desc_count; + unsigned int desc_avail; +}; + +struct vnic_dev; +struct vnic_stats; + +void *vnic_dev_priv(struct vnic_dev *vdev); +unsigned int vnic_dev_get_res_count(struct vnic_dev *vdev, + enum vnic_res_type type); +void __iomem *vnic_dev_get_res(struct vnic_dev *vdev, enum vnic_res_type type, + unsigned int index); +unsigned int vnic_dev_desc_ring_size(struct vnic_dev_ring *ring, + unsigned int desc_count, + unsigned int desc_size); +void vnic_dev_clear_desc_ring(struct vnic_dev_ring *ring); +int vnic_dev_alloc_desc_ring(struct vnic_dev *vdev, struct vnic_dev_ring *ring, + unsigned int desc_count, unsigned int desc_size); +void vnic_dev_free_desc_ring(struct vnic_dev *vdev, + struct vnic_dev_ring *ring); +int vnic_dev_cmd(struct vnic_dev *vdev, enum vnic_devcmd_cmd cmd, + u64 *a0, u64 *a1, int wait); +int vnic_dev_fw_info(struct vnic_dev *vdev, + struct vnic_devcmd_fw_info **fw_info); +int vnic_dev_spec(struct vnic_dev *vdev, unsigned int offset, + unsigned int size, void *value); +int vnic_dev_stats_clear(struct vnic_dev *vdev); +int vnic_dev_stats_dump(struct vnic_dev *vdev, struct vnic_stats **stats); +int vnic_dev_hang_notify(struct vnic_dev *vdev); +void vnic_dev_packet_filter(struct vnic_dev *vdev, int directed, int multicast, + int broadcast, int promisc, int allmulti); +void vnic_dev_add_addr(struct vnic_dev *vdev, u8 *addr); +void vnic_dev_del_addr(struct vnic_dev *vdev, u8 *addr); +int vnic_dev_mac_addr(struct vnic_dev *vdev, u8 *mac_addr); +int vnic_dev_notify_set(struct vnic_dev *vdev, u16 intr); +void vnic_dev_notify_unset(struct vnic_dev *vdev); +int vnic_dev_link_status(struct vnic_dev *vdev); +u32 vnic_dev_port_speed(struct vnic_dev *vdev); +u32 vnic_dev_msg_lvl(struct vnic_dev *vdev); +u32 vnic_dev_mtu(struct vnic_dev *vdev); +u32 vnic_dev_link_down_cnt(struct vnic_dev *vdev); +int vnic_dev_close(struct vnic_dev *vdev); +int vnic_dev_enable(struct vnic_dev *vdev); +int vnic_dev_disable(struct vnic_dev *vdev); +int vnic_dev_open(struct vnic_dev *vdev, int arg); +int vnic_dev_open_done(struct vnic_dev *vdev, int *done); +int vnic_dev_init(struct vnic_dev *vdev, int arg); +int vnic_dev_soft_reset(struct vnic_dev *vdev, int arg); +int vnic_dev_soft_reset_done(struct vnic_dev *vdev, int *done); +void vnic_dev_set_intr_mode(struct vnic_dev *vdev, + enum vnic_dev_intr_mode intr_mode); +enum vnic_dev_intr_mode vnic_dev_get_intr_mode(struct vnic_dev *vdev); +void vnic_dev_unregister(struct vnic_dev *vdev); +struct vnic_dev *vnic_dev_register(struct vnic_dev *vdev, + void *priv, struct pci_dev *pdev, + struct vnic_dev_bar *bar); + +#endif /* _VNIC_DEV_H_ */ diff --git a/drivers/scsi/fnic/vnic_devcmd.h b/drivers/scsi/fnic/vnic_devcmd.h new file mode 100644 index 000000000000..d62b9061bf12 --- /dev/null +++ b/drivers/scsi/fnic/vnic_devcmd.h @@ -0,0 +1,281 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _VNIC_DEVCMD_H_ +#define _VNIC_DEVCMD_H_ + +#define _CMD_NBITS 14 +#define _CMD_VTYPEBITS 10 +#define _CMD_FLAGSBITS 6 +#define _CMD_DIRBITS 2 + +#define _CMD_NMASK ((1 << _CMD_NBITS)-1) +#define _CMD_VTYPEMASK ((1 << _CMD_VTYPEBITS)-1) +#define _CMD_FLAGSMASK ((1 << _CMD_FLAGSBITS)-1) +#define _CMD_DIRMASK ((1 << _CMD_DIRBITS)-1) + +#define _CMD_NSHIFT 0 +#define _CMD_VTYPESHIFT (_CMD_NSHIFT+_CMD_NBITS) +#define _CMD_FLAGSSHIFT (_CMD_VTYPESHIFT+_CMD_VTYPEBITS) +#define _CMD_DIRSHIFT (_CMD_FLAGSSHIFT+_CMD_FLAGSBITS) + +/* + * Direction bits (from host perspective). + */ +#define _CMD_DIR_NONE 0U +#define _CMD_DIR_WRITE 1U +#define _CMD_DIR_READ 2U +#define _CMD_DIR_RW (_CMD_DIR_WRITE | _CMD_DIR_READ) + +/* + * Flag bits. + */ +#define _CMD_FLAGS_NONE 0U +#define _CMD_FLAGS_NOWAIT 1U + +/* + * vNIC type bits. + */ +#define _CMD_VTYPE_NONE 0U +#define _CMD_VTYPE_ENET 1U +#define _CMD_VTYPE_FC 2U +#define _CMD_VTYPE_SCSI 4U +#define _CMD_VTYPE_ALL (_CMD_VTYPE_ENET | _CMD_VTYPE_FC | _CMD_VTYPE_SCSI) + +/* + * Used to create cmds.. +*/ +#define _CMDCF(dir, flags, vtype, nr) \ + (((dir) << _CMD_DIRSHIFT) | \ + ((flags) << _CMD_FLAGSSHIFT) | \ + ((vtype) << _CMD_VTYPESHIFT) | \ + ((nr) << _CMD_NSHIFT)) +#define _CMDC(dir, vtype, nr) _CMDCF(dir, 0, vtype, nr) +#define _CMDCNW(dir, vtype, nr) _CMDCF(dir, _CMD_FLAGS_NOWAIT, vtype, nr) + +/* + * Used to decode cmds.. +*/ +#define _CMD_DIR(cmd) (((cmd) >> _CMD_DIRSHIFT) & _CMD_DIRMASK) +#define _CMD_FLAGS(cmd) (((cmd) >> _CMD_FLAGSSHIFT) & _CMD_FLAGSMASK) +#define _CMD_VTYPE(cmd) (((cmd) >> _CMD_VTYPESHIFT) & _CMD_VTYPEMASK) +#define _CMD_N(cmd) (((cmd) >> _CMD_NSHIFT) & _CMD_NMASK) + +enum vnic_devcmd_cmd { + CMD_NONE = _CMDC(_CMD_DIR_NONE, _CMD_VTYPE_NONE, 0), + + /* mcpu fw info in mem: (u64)a0=paddr to struct vnic_devcmd_fw_info */ + CMD_MCPU_FW_INFO = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ALL, 1), + + /* dev-specific block member: + * in: (u16)a0=offset,(u8)a1=size + * out: a0=value */ + CMD_DEV_SPEC = _CMDC(_CMD_DIR_RW, _CMD_VTYPE_ALL, 2), + + /* stats clear */ + CMD_STATS_CLEAR = _CMDCNW(_CMD_DIR_NONE, _CMD_VTYPE_ALL, 3), + + /* stats dump in mem: (u64)a0=paddr to stats area, + * (u16)a1=sizeof stats area */ + CMD_STATS_DUMP = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ALL, 4), + + /* set Rx packet filter: (u32)a0=filters (see CMD_PFILTER_*) */ + CMD_PACKET_FILTER = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 7), + + /* hang detection notification */ + CMD_HANG_NOTIFY = _CMDC(_CMD_DIR_NONE, _CMD_VTYPE_ALL, 8), + + /* MAC address in (u48)a0 */ + CMD_MAC_ADDR = _CMDC(_CMD_DIR_READ, + _CMD_VTYPE_ENET | _CMD_VTYPE_FC, 9), + + /* disable/enable promisc mode: (u8)a0=0/1 */ +/***** XXX DEPRECATED *****/ + CMD_PROMISC_MODE = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 10), + + /* disable/enable all-multi mode: (u8)a0=0/1 */ +/***** XXX DEPRECATED *****/ + CMD_ALLMULTI_MODE = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 11), + + /* add addr from (u48)a0 */ + CMD_ADDR_ADD = _CMDCNW(_CMD_DIR_WRITE, + _CMD_VTYPE_ENET | _CMD_VTYPE_FC, 12), + + /* del addr from (u48)a0 */ + CMD_ADDR_DEL = _CMDCNW(_CMD_DIR_WRITE, + _CMD_VTYPE_ENET | _CMD_VTYPE_FC, 13), + + /* add VLAN id in (u16)a0 */ + CMD_VLAN_ADD = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 14), + + /* del VLAN id in (u16)a0 */ + CMD_VLAN_DEL = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 15), + + /* nic_cfg in (u32)a0 */ + CMD_NIC_CFG = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ALL, 16), + + /* union vnic_rss_key in mem: (u64)a0=paddr, (u16)a1=len */ + CMD_RSS_KEY = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 17), + + /* union vnic_rss_cpu in mem: (u64)a0=paddr, (u16)a1=len */ + CMD_RSS_CPU = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 18), + + /* initiate softreset */ + CMD_SOFT_RESET = _CMDCNW(_CMD_DIR_NONE, _CMD_VTYPE_ALL, 19), + + /* softreset status: + * out: a0=0 reset complete, a0=1 reset in progress */ + CMD_SOFT_RESET_STATUS = _CMDC(_CMD_DIR_READ, _CMD_VTYPE_ALL, 20), + + /* set struct vnic_devcmd_notify buffer in mem: + * in: + * (u64)a0=paddr to notify (set paddr=0 to unset) + * (u32)a1 & 0x00000000ffffffff=sizeof(struct vnic_devcmd_notify) + * (u16)a1 & 0x0000ffff00000000=intr num (-1 for no intr) + * out: + * (u32)a1 = effective size + */ + CMD_NOTIFY = _CMDC(_CMD_DIR_RW, _CMD_VTYPE_ALL, 21), + + /* UNDI API: (u64)a0=paddr to s_PXENV_UNDI_ struct, + * (u8)a1=PXENV_UNDI_xxx */ + CMD_UNDI = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 22), + + /* initiate open sequence (u32)a0=flags (see CMD_OPENF_*) */ + CMD_OPEN = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ALL, 23), + + /* open status: + * out: a0=0 open complete, a0=1 open in progress */ + CMD_OPEN_STATUS = _CMDC(_CMD_DIR_READ, _CMD_VTYPE_ALL, 24), + + /* close vnic */ + CMD_CLOSE = _CMDC(_CMD_DIR_NONE, _CMD_VTYPE_ALL, 25), + + /* initialize virtual link: (u32)a0=flags (see CMD_INITF_*) */ + CMD_INIT = _CMDCNW(_CMD_DIR_READ, _CMD_VTYPE_ALL, 26), + + /* variant of CMD_INIT, with provisioning info + * (u64)a0=paddr of vnic_devcmd_provinfo + * (u32)a1=sizeof provision info */ + CMD_INIT_PROV_INFO = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 27), + + /* enable virtual link */ + CMD_ENABLE = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ALL, 28), + + /* disable virtual link */ + CMD_DISABLE = _CMDC(_CMD_DIR_NONE, _CMD_VTYPE_ALL, 29), + + /* stats dump all vnics on uplink in mem: (u64)a0=paddr (u32)a1=uif */ + CMD_STATS_DUMP_ALL = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ALL, 30), + + /* init status: + * out: a0=0 init complete, a0=1 init in progress + * if a0=0, a1=errno */ + CMD_INIT_STATUS = _CMDC(_CMD_DIR_READ, _CMD_VTYPE_ALL, 31), + + /* INT13 API: (u64)a0=paddr to vnic_int13_params struct + * (u8)a1=INT13_CMD_xxx */ + CMD_INT13 = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_FC, 32), + + /* logical uplink enable/disable: (u64)a0: 0/1=disable/enable */ + CMD_LOGICAL_UPLINK = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 33), + + /* undo initialize of virtual link */ + CMD_DEINIT = _CMDCNW(_CMD_DIR_NONE, _CMD_VTYPE_ALL, 34), +}; + +/* flags for CMD_OPEN */ +#define CMD_OPENF_OPROM 0x1 /* open coming from option rom */ + +/* flags for CMD_INIT */ +#define CMD_INITF_DEFAULT_MAC 0x1 /* init with default mac addr */ + +/* flags for CMD_PACKET_FILTER */ +#define CMD_PFILTER_DIRECTED 0x01 +#define CMD_PFILTER_MULTICAST 0x02 +#define CMD_PFILTER_BROADCAST 0x04 +#define CMD_PFILTER_PROMISCUOUS 0x08 +#define CMD_PFILTER_ALL_MULTICAST 0x10 + +enum vnic_devcmd_status { + STAT_NONE = 0, + STAT_BUSY = 1 << 0, /* cmd in progress */ + STAT_ERROR = 1 << 1, /* last cmd caused error (code in a0) */ +}; + +enum vnic_devcmd_error { + ERR_SUCCESS = 0, + ERR_EINVAL = 1, + ERR_EFAULT = 2, + ERR_EPERM = 3, + ERR_EBUSY = 4, + ERR_ECMDUNKNOWN = 5, + ERR_EBADSTATE = 6, + ERR_ENOMEM = 7, + ERR_ETIMEDOUT = 8, + ERR_ELINKDOWN = 9, +}; + +struct vnic_devcmd_fw_info { + char fw_version[32]; + char fw_build[32]; + char hw_version[32]; + char hw_serial_number[32]; +}; + +struct vnic_devcmd_notify { + u32 csum; /* checksum over following words */ + + u32 link_state; /* link up == 1 */ + u32 port_speed; /* effective port speed (rate limit) */ + u32 mtu; /* MTU */ + u32 msglvl; /* requested driver msg lvl */ + u32 uif; /* uplink interface */ + u32 status; /* status bits (see VNIC_STF_*) */ + u32 error; /* error code (see ERR_*) for first ERR */ + u32 link_down_cnt; /* running count of link down transitions */ +}; +#define VNIC_STF_FATAL_ERR 0x0001 /* fatal fw error */ + +struct vnic_devcmd_provinfo { + u8 oui[3]; + u8 type; + u8 data[0]; +}; + +/* + * Writing cmd register causes STAT_BUSY to get set in status register. + * When cmd completes, STAT_BUSY will be cleared. + * + * If cmd completed successfully STAT_ERROR will be clear + * and args registers contain cmd-specific results. + * + * If cmd error, STAT_ERROR will be set and args[0] contains error code. + * + * status register is read-only. While STAT_BUSY is set, + * all other register contents are read-only. + */ + +/* Make sizeof(vnic_devcmd) a power-of-2 for I/O BAR. */ +#define VNIC_DEVCMD_NARGS 15 +struct vnic_devcmd { + u32 status; /* RO */ + u32 cmd; /* RW */ + u64 args[VNIC_DEVCMD_NARGS]; /* RW cmd args (little-endian) */ +}; + +#endif /* _VNIC_DEVCMD_H_ */ diff --git a/drivers/scsi/fnic/vnic_intr.c b/drivers/scsi/fnic/vnic_intr.c new file mode 100644 index 000000000000..4f4dc8793d23 --- /dev/null +++ b/drivers/scsi/fnic/vnic_intr.c @@ -0,0 +1,60 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include +#include +#include +#include "vnic_dev.h" +#include "vnic_intr.h" + +void vnic_intr_free(struct vnic_intr *intr) +{ + intr->ctrl = NULL; +} + +int vnic_intr_alloc(struct vnic_dev *vdev, struct vnic_intr *intr, + unsigned int index) +{ + intr->index = index; + intr->vdev = vdev; + + intr->ctrl = vnic_dev_get_res(vdev, RES_TYPE_INTR_CTRL, index); + if (!intr->ctrl) { + printk(KERN_ERR "Failed to hook INTR[%d].ctrl resource\n", + index); + return -EINVAL; + } + + return 0; +} + +void vnic_intr_init(struct vnic_intr *intr, unsigned int coalescing_timer, + unsigned int coalescing_type, unsigned int mask_on_assertion) +{ + iowrite32(coalescing_timer, &intr->ctrl->coalescing_timer); + iowrite32(coalescing_type, &intr->ctrl->coalescing_type); + iowrite32(mask_on_assertion, &intr->ctrl->mask_on_assertion); + iowrite32(0, &intr->ctrl->int_credits); +} + +void vnic_intr_clean(struct vnic_intr *intr) +{ + iowrite32(0, &intr->ctrl->int_credits); +} diff --git a/drivers/scsi/fnic/vnic_intr.h b/drivers/scsi/fnic/vnic_intr.h new file mode 100644 index 000000000000..d5fb40e7c98e --- /dev/null +++ b/drivers/scsi/fnic/vnic_intr.h @@ -0,0 +1,118 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _VNIC_INTR_H_ +#define _VNIC_INTR_H_ + +#include +#include "vnic_dev.h" + +/* + * These defines avoid symbol clash between fnic and enic (Cisco 10G Eth + * Driver) when both are built with CONFIG options =y + */ +#define vnic_intr_unmask fnic_intr_unmask +#define vnic_intr_mask fnic_intr_mask +#define vnic_intr_return_credits fnic_intr_return_credits +#define vnic_intr_credits fnic_intr_credits +#define vnic_intr_return_all_credits fnic_intr_return_all_credits +#define vnic_intr_legacy_pba fnic_intr_legacy_pba +#define vnic_intr_free fnic_intr_free +#define vnic_intr_alloc fnic_intr_alloc +#define vnic_intr_init fnic_intr_init +#define vnic_intr_clean fnic_intr_clean + +#define VNIC_INTR_TIMER_MAX 0xffff + +#define VNIC_INTR_TIMER_TYPE_ABS 0 +#define VNIC_INTR_TIMER_TYPE_QUIET 1 + +/* Interrupt control */ +struct vnic_intr_ctrl { + u32 coalescing_timer; /* 0x00 */ + u32 pad0; + u32 coalescing_value; /* 0x08 */ + u32 pad1; + u32 coalescing_type; /* 0x10 */ + u32 pad2; + u32 mask_on_assertion; /* 0x18 */ + u32 pad3; + u32 mask; /* 0x20 */ + u32 pad4; + u32 int_credits; /* 0x28 */ + u32 pad5; + u32 int_credit_return; /* 0x30 */ + u32 pad6; +}; + +struct vnic_intr { + unsigned int index; + struct vnic_dev *vdev; + struct vnic_intr_ctrl __iomem *ctrl; /* memory-mapped */ +}; + +static inline void vnic_intr_unmask(struct vnic_intr *intr) +{ + iowrite32(0, &intr->ctrl->mask); +} + +static inline void vnic_intr_mask(struct vnic_intr *intr) +{ + iowrite32(1, &intr->ctrl->mask); +} + +static inline void vnic_intr_return_credits(struct vnic_intr *intr, + unsigned int credits, int unmask, int reset_timer) +{ +#define VNIC_INTR_UNMASK_SHIFT 16 +#define VNIC_INTR_RESET_TIMER_SHIFT 17 + + u32 int_credit_return = (credits & 0xffff) | + (unmask ? (1 << VNIC_INTR_UNMASK_SHIFT) : 0) | + (reset_timer ? (1 << VNIC_INTR_RESET_TIMER_SHIFT) : 0); + + iowrite32(int_credit_return, &intr->ctrl->int_credit_return); +} + +static inline unsigned int vnic_intr_credits(struct vnic_intr *intr) +{ + return ioread32(&intr->ctrl->int_credits); +} + +static inline void vnic_intr_return_all_credits(struct vnic_intr *intr) +{ + unsigned int credits = vnic_intr_credits(intr); + int unmask = 1; + int reset_timer = 1; + + vnic_intr_return_credits(intr, credits, unmask, reset_timer); +} + +static inline u32 vnic_intr_legacy_pba(u32 __iomem *legacy_pba) +{ + /* read PBA without clearing */ + return ioread32(legacy_pba); +} + +void vnic_intr_free(struct vnic_intr *intr); +int vnic_intr_alloc(struct vnic_dev *vdev, struct vnic_intr *intr, + unsigned int index); +void vnic_intr_init(struct vnic_intr *intr, unsigned int coalescing_timer, + unsigned int coalescing_type, unsigned int mask_on_assertion); +void vnic_intr_clean(struct vnic_intr *intr); + +#endif /* _VNIC_INTR_H_ */ diff --git a/drivers/scsi/fnic/vnic_nic.h b/drivers/scsi/fnic/vnic_nic.h new file mode 100644 index 000000000000..f15b83eeaced --- /dev/null +++ b/drivers/scsi/fnic/vnic_nic.h @@ -0,0 +1,69 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _VNIC_NIC_H_ +#define _VNIC_NIC_H_ + +/* + * These defines avoid symbol clash between fnic and enic (Cisco 10G Eth + * Driver) when both are built with CONFIG options =y + */ +#define vnic_set_nic_cfg fnic_set_nic_cfg + +#define NIC_CFG_RSS_DEFAULT_CPU_MASK_FIELD 0xffUL +#define NIC_CFG_RSS_DEFAULT_CPU_SHIFT 0 +#define NIC_CFG_RSS_HASH_TYPE (0xffUL << 8) +#define NIC_CFG_RSS_HASH_TYPE_MASK_FIELD 0xffUL +#define NIC_CFG_RSS_HASH_TYPE_SHIFT 8 +#define NIC_CFG_RSS_HASH_BITS (7UL << 16) +#define NIC_CFG_RSS_HASH_BITS_MASK_FIELD 7UL +#define NIC_CFG_RSS_HASH_BITS_SHIFT 16 +#define NIC_CFG_RSS_BASE_CPU (7UL << 19) +#define NIC_CFG_RSS_BASE_CPU_MASK_FIELD 7UL +#define NIC_CFG_RSS_BASE_CPU_SHIFT 19 +#define NIC_CFG_RSS_ENABLE (1UL << 22) +#define NIC_CFG_RSS_ENABLE_MASK_FIELD 1UL +#define NIC_CFG_RSS_ENABLE_SHIFT 22 +#define NIC_CFG_TSO_IPID_SPLIT_EN (1UL << 23) +#define NIC_CFG_TSO_IPID_SPLIT_EN_MASK_FIELD 1UL +#define NIC_CFG_TSO_IPID_SPLIT_EN_SHIFT 23 +#define NIC_CFG_IG_VLAN_STRIP_EN (1UL << 24) +#define NIC_CFG_IG_VLAN_STRIP_EN_MASK_FIELD 1UL +#define NIC_CFG_IG_VLAN_STRIP_EN_SHIFT 24 + +static inline void vnic_set_nic_cfg(u32 *nic_cfg, + u8 rss_default_cpu, u8 rss_hash_type, + u8 rss_hash_bits, u8 rss_base_cpu, + u8 rss_enable, u8 tso_ipid_split_en, + u8 ig_vlan_strip_en) +{ + *nic_cfg = (rss_default_cpu & NIC_CFG_RSS_DEFAULT_CPU_MASK_FIELD) | + ((rss_hash_type & NIC_CFG_RSS_HASH_TYPE_MASK_FIELD) + << NIC_CFG_RSS_HASH_TYPE_SHIFT) | + ((rss_hash_bits & NIC_CFG_RSS_HASH_BITS_MASK_FIELD) + << NIC_CFG_RSS_HASH_BITS_SHIFT) | + ((rss_base_cpu & NIC_CFG_RSS_BASE_CPU_MASK_FIELD) + << NIC_CFG_RSS_BASE_CPU_SHIFT) | + ((rss_enable & NIC_CFG_RSS_ENABLE_MASK_FIELD) + << NIC_CFG_RSS_ENABLE_SHIFT) | + ((tso_ipid_split_en & NIC_CFG_TSO_IPID_SPLIT_EN_MASK_FIELD) + << NIC_CFG_TSO_IPID_SPLIT_EN_SHIFT) | + ((ig_vlan_strip_en & NIC_CFG_IG_VLAN_STRIP_EN_MASK_FIELD) + << NIC_CFG_IG_VLAN_STRIP_EN_SHIFT); +} + +#endif /* _VNIC_NIC_H_ */ diff --git a/drivers/scsi/fnic/vnic_resource.h b/drivers/scsi/fnic/vnic_resource.h new file mode 100644 index 000000000000..2d842f79d41a --- /dev/null +++ b/drivers/scsi/fnic/vnic_resource.h @@ -0,0 +1,61 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _VNIC_RESOURCE_H_ +#define _VNIC_RESOURCE_H_ + +#define VNIC_RES_MAGIC 0x766E6963L /* 'vnic' */ +#define VNIC_RES_VERSION 0x00000000L + +/* vNIC resource types */ +enum vnic_res_type { + RES_TYPE_EOL, /* End-of-list */ + RES_TYPE_WQ, /* Work queues */ + RES_TYPE_RQ, /* Receive queues */ + RES_TYPE_CQ, /* Completion queues */ + RES_TYPE_RSVD1, + RES_TYPE_NIC_CFG, /* Enet NIC config registers */ + RES_TYPE_RSVD2, + RES_TYPE_RSVD3, + RES_TYPE_RSVD4, + RES_TYPE_RSVD5, + RES_TYPE_INTR_CTRL, /* Interrupt ctrl table */ + RES_TYPE_INTR_TABLE, /* MSI/MSI-X Interrupt table */ + RES_TYPE_INTR_PBA, /* MSI/MSI-X PBA table */ + RES_TYPE_INTR_PBA_LEGACY, /* Legacy intr status */ + RES_TYPE_RSVD6, + RES_TYPE_RSVD7, + RES_TYPE_DEVCMD, /* Device command region */ + RES_TYPE_PASS_THRU_PAGE, /* Pass-thru page */ + + RES_TYPE_MAX, /* Count of resource types */ +}; + +struct vnic_resource_header { + u32 magic; + u32 version; +}; + +struct vnic_resource { + u8 type; + u8 bar; + u8 pad[2]; + u32 bar_offset; + u32 count; +}; + +#endif /* _VNIC_RESOURCE_H_ */ diff --git a/drivers/scsi/fnic/vnic_rq.c b/drivers/scsi/fnic/vnic_rq.c new file mode 100644 index 000000000000..bedd0d285630 --- /dev/null +++ b/drivers/scsi/fnic/vnic_rq.c @@ -0,0 +1,196 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include +#include +#include "vnic_dev.h" +#include "vnic_rq.h" + +static int vnic_rq_alloc_bufs(struct vnic_rq *rq) +{ + struct vnic_rq_buf *buf; + struct vnic_dev *vdev; + unsigned int i, j, count = rq->ring.desc_count; + unsigned int blks = VNIC_RQ_BUF_BLKS_NEEDED(count); + + vdev = rq->vdev; + + for (i = 0; i < blks; i++) { + rq->bufs[i] = kzalloc(VNIC_RQ_BUF_BLK_SZ, GFP_ATOMIC); + if (!rq->bufs[i]) { + printk(KERN_ERR "Failed to alloc rq_bufs\n"); + return -ENOMEM; + } + } + + for (i = 0; i < blks; i++) { + buf = rq->bufs[i]; + for (j = 0; j < VNIC_RQ_BUF_BLK_ENTRIES; j++) { + buf->index = i * VNIC_RQ_BUF_BLK_ENTRIES + j; + buf->desc = (u8 *)rq->ring.descs + + rq->ring.desc_size * buf->index; + if (buf->index + 1 == count) { + buf->next = rq->bufs[0]; + break; + } else if (j + 1 == VNIC_RQ_BUF_BLK_ENTRIES) { + buf->next = rq->bufs[i + 1]; + } else { + buf->next = buf + 1; + buf++; + } + } + } + + rq->to_use = rq->to_clean = rq->bufs[0]; + rq->buf_index = 0; + + return 0; +} + +void vnic_rq_free(struct vnic_rq *rq) +{ + struct vnic_dev *vdev; + unsigned int i; + + vdev = rq->vdev; + + vnic_dev_free_desc_ring(vdev, &rq->ring); + + for (i = 0; i < VNIC_RQ_BUF_BLKS_MAX; i++) { + kfree(rq->bufs[i]); + rq->bufs[i] = NULL; + } + + rq->ctrl = NULL; +} + +int vnic_rq_alloc(struct vnic_dev *vdev, struct vnic_rq *rq, unsigned int index, + unsigned int desc_count, unsigned int desc_size) +{ + int err; + + rq->index = index; + rq->vdev = vdev; + + rq->ctrl = vnic_dev_get_res(vdev, RES_TYPE_RQ, index); + if (!rq->ctrl) { + printk(KERN_ERR "Failed to hook RQ[%d] resource\n", index); + return -EINVAL; + } + + vnic_rq_disable(rq); + + err = vnic_dev_alloc_desc_ring(vdev, &rq->ring, desc_count, desc_size); + if (err) + return err; + + err = vnic_rq_alloc_bufs(rq); + if (err) { + vnic_rq_free(rq); + return err; + } + + return 0; +} + +void vnic_rq_init(struct vnic_rq *rq, unsigned int cq_index, + unsigned int error_interrupt_enable, + unsigned int error_interrupt_offset) +{ + u64 paddr; + u32 fetch_index; + + paddr = (u64)rq->ring.base_addr | VNIC_PADDR_TARGET; + writeq(paddr, &rq->ctrl->ring_base); + iowrite32(rq->ring.desc_count, &rq->ctrl->ring_size); + iowrite32(cq_index, &rq->ctrl->cq_index); + iowrite32(error_interrupt_enable, &rq->ctrl->error_interrupt_enable); + iowrite32(error_interrupt_offset, &rq->ctrl->error_interrupt_offset); + iowrite32(0, &rq->ctrl->dropped_packet_count); + iowrite32(0, &rq->ctrl->error_status); + + /* Use current fetch_index as the ring starting point */ + fetch_index = ioread32(&rq->ctrl->fetch_index); + rq->to_use = rq->to_clean = + &rq->bufs[fetch_index / VNIC_RQ_BUF_BLK_ENTRIES] + [fetch_index % VNIC_RQ_BUF_BLK_ENTRIES]; + iowrite32(fetch_index, &rq->ctrl->posted_index); + + rq->buf_index = 0; +} + +unsigned int vnic_rq_error_status(struct vnic_rq *rq) +{ + return ioread32(&rq->ctrl->error_status); +} + +void vnic_rq_enable(struct vnic_rq *rq) +{ + iowrite32(1, &rq->ctrl->enable); +} + +int vnic_rq_disable(struct vnic_rq *rq) +{ + unsigned int wait; + + iowrite32(0, &rq->ctrl->enable); + + /* Wait for HW to ACK disable request */ + for (wait = 0; wait < 100; wait++) { + if (!(ioread32(&rq->ctrl->running))) + return 0; + udelay(1); + } + + printk(KERN_ERR "Failed to disable RQ[%d]\n", rq->index); + + return -ETIMEDOUT; +} + +void vnic_rq_clean(struct vnic_rq *rq, + void (*buf_clean)(struct vnic_rq *rq, struct vnic_rq_buf *buf)) +{ + struct vnic_rq_buf *buf; + u32 fetch_index; + + BUG_ON(ioread32(&rq->ctrl->enable)); + + buf = rq->to_clean; + + while (vnic_rq_desc_used(rq) > 0) { + + (*buf_clean)(rq, buf); + + buf = rq->to_clean = buf->next; + rq->ring.desc_avail++; + } + + /* Use current fetch_index as the ring starting point */ + fetch_index = ioread32(&rq->ctrl->fetch_index); + rq->to_use = rq->to_clean = + &rq->bufs[fetch_index / VNIC_RQ_BUF_BLK_ENTRIES] + [fetch_index % VNIC_RQ_BUF_BLK_ENTRIES]; + iowrite32(fetch_index, &rq->ctrl->posted_index); + + rq->buf_index = 0; + + vnic_dev_clear_desc_ring(&rq->ring); +} + diff --git a/drivers/scsi/fnic/vnic_rq.h b/drivers/scsi/fnic/vnic_rq.h new file mode 100644 index 000000000000..aebdfbd6ad3c --- /dev/null +++ b/drivers/scsi/fnic/vnic_rq.h @@ -0,0 +1,235 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _VNIC_RQ_H_ +#define _VNIC_RQ_H_ + +#include +#include "vnic_dev.h" +#include "vnic_cq.h" + +/* + * These defines avoid symbol clash between fnic and enic (Cisco 10G Eth + * Driver) when both are built with CONFIG options =y + */ +#define vnic_rq_desc_avail fnic_rq_desc_avail +#define vnic_rq_desc_used fnic_rq_desc_used +#define vnic_rq_next_desc fnic_rq_next_desc +#define vnic_rq_next_index fnic_rq_next_index +#define vnic_rq_next_buf_index fnic_rq_next_buf_index +#define vnic_rq_post fnic_rq_post +#define vnic_rq_posting_soon fnic_rq_posting_soon +#define vnic_rq_return_descs fnic_rq_return_descs +#define vnic_rq_service fnic_rq_service +#define vnic_rq_fill fnic_rq_fill +#define vnic_rq_free fnic_rq_free +#define vnic_rq_alloc fnic_rq_alloc +#define vnic_rq_init fnic_rq_init +#define vnic_rq_error_status fnic_rq_error_status +#define vnic_rq_enable fnic_rq_enable +#define vnic_rq_disable fnic_rq_disable +#define vnic_rq_clean fnic_rq_clean + +/* Receive queue control */ +struct vnic_rq_ctrl { + u64 ring_base; /* 0x00 */ + u32 ring_size; /* 0x08 */ + u32 pad0; + u32 posted_index; /* 0x10 */ + u32 pad1; + u32 cq_index; /* 0x18 */ + u32 pad2; + u32 enable; /* 0x20 */ + u32 pad3; + u32 running; /* 0x28 */ + u32 pad4; + u32 fetch_index; /* 0x30 */ + u32 pad5; + u32 error_interrupt_enable; /* 0x38 */ + u32 pad6; + u32 error_interrupt_offset; /* 0x40 */ + u32 pad7; + u32 error_status; /* 0x48 */ + u32 pad8; + u32 dropped_packet_count; /* 0x50 */ + u32 pad9; + u32 dropped_packet_count_rc; /* 0x58 */ + u32 pad10; +}; + +/* Break the vnic_rq_buf allocations into blocks of 64 entries */ +#define VNIC_RQ_BUF_BLK_ENTRIES 64 +#define VNIC_RQ_BUF_BLK_SZ \ + (VNIC_RQ_BUF_BLK_ENTRIES * sizeof(struct vnic_rq_buf)) +#define VNIC_RQ_BUF_BLKS_NEEDED(entries) \ + DIV_ROUND_UP(entries, VNIC_RQ_BUF_BLK_ENTRIES) +#define VNIC_RQ_BUF_BLKS_MAX VNIC_RQ_BUF_BLKS_NEEDED(4096) + +struct vnic_rq_buf { + struct vnic_rq_buf *next; + dma_addr_t dma_addr; + void *os_buf; + unsigned int os_buf_index; + unsigned int len; + unsigned int index; + void *desc; +}; + +struct vnic_rq { + unsigned int index; + struct vnic_dev *vdev; + struct vnic_rq_ctrl __iomem *ctrl; /* memory-mapped */ + struct vnic_dev_ring ring; + struct vnic_rq_buf *bufs[VNIC_RQ_BUF_BLKS_MAX]; + struct vnic_rq_buf *to_use; + struct vnic_rq_buf *to_clean; + void *os_buf_head; + unsigned int buf_index; + unsigned int pkts_outstanding; +}; + +static inline unsigned int vnic_rq_desc_avail(struct vnic_rq *rq) +{ + /* how many does SW own? */ + return rq->ring.desc_avail; +} + +static inline unsigned int vnic_rq_desc_used(struct vnic_rq *rq) +{ + /* how many does HW own? */ + return rq->ring.desc_count - rq->ring.desc_avail - 1; +} + +static inline void *vnic_rq_next_desc(struct vnic_rq *rq) +{ + return rq->to_use->desc; +} + +static inline unsigned int vnic_rq_next_index(struct vnic_rq *rq) +{ + return rq->to_use->index; +} + +static inline unsigned int vnic_rq_next_buf_index(struct vnic_rq *rq) +{ + return rq->buf_index++; +} + +static inline void vnic_rq_post(struct vnic_rq *rq, + void *os_buf, unsigned int os_buf_index, + dma_addr_t dma_addr, unsigned int len) +{ + struct vnic_rq_buf *buf = rq->to_use; + + buf->os_buf = os_buf; + buf->os_buf_index = os_buf_index; + buf->dma_addr = dma_addr; + buf->len = len; + + buf = buf->next; + rq->to_use = buf; + rq->ring.desc_avail--; + + /* Move the posted_index every nth descriptor + */ + +#ifndef VNIC_RQ_RETURN_RATE +#define VNIC_RQ_RETURN_RATE 0xf /* keep 2^n - 1 */ +#endif + + if ((buf->index & VNIC_RQ_RETURN_RATE) == 0) { + /* Adding write memory barrier prevents compiler and/or CPU + * reordering, thus avoiding descriptor posting before + * descriptor is initialized. Otherwise, hardware can read + * stale descriptor fields. + */ + wmb(); + iowrite32(buf->index, &rq->ctrl->posted_index); + } +} + +static inline int vnic_rq_posting_soon(struct vnic_rq *rq) +{ + return (rq->to_use->index & VNIC_RQ_RETURN_RATE) == 0; +} + +static inline void vnic_rq_return_descs(struct vnic_rq *rq, unsigned int count) +{ + rq->ring.desc_avail += count; +} + +enum desc_return_options { + VNIC_RQ_RETURN_DESC, + VNIC_RQ_DEFER_RETURN_DESC, +}; + +static inline void vnic_rq_service(struct vnic_rq *rq, + struct cq_desc *cq_desc, u16 completed_index, + int desc_return, void (*buf_service)(struct vnic_rq *rq, + struct cq_desc *cq_desc, struct vnic_rq_buf *buf, + int skipped, void *opaque), void *opaque) +{ + struct vnic_rq_buf *buf; + int skipped; + + buf = rq->to_clean; + while (1) { + + skipped = (buf->index != completed_index); + + (*buf_service)(rq, cq_desc, buf, skipped, opaque); + + if (desc_return == VNIC_RQ_RETURN_DESC) + rq->ring.desc_avail++; + + rq->to_clean = buf->next; + + if (!skipped) + break; + + buf = rq->to_clean; + } +} + +static inline int vnic_rq_fill(struct vnic_rq *rq, + int (*buf_fill)(struct vnic_rq *rq)) +{ + int err; + + while (vnic_rq_desc_avail(rq) > 1) { + + err = (*buf_fill)(rq); + if (err) + return err; + } + + return 0; +} + +void vnic_rq_free(struct vnic_rq *rq); +int vnic_rq_alloc(struct vnic_dev *vdev, struct vnic_rq *rq, unsigned int index, + unsigned int desc_count, unsigned int desc_size); +void vnic_rq_init(struct vnic_rq *rq, unsigned int cq_index, + unsigned int error_interrupt_enable, + unsigned int error_interrupt_offset); +unsigned int vnic_rq_error_status(struct vnic_rq *rq); +void vnic_rq_enable(struct vnic_rq *rq); +int vnic_rq_disable(struct vnic_rq *rq); +void vnic_rq_clean(struct vnic_rq *rq, + void (*buf_clean)(struct vnic_rq *rq, struct vnic_rq_buf *buf)); + +#endif /* _VNIC_RQ_H_ */ diff --git a/drivers/scsi/fnic/vnic_scsi.h b/drivers/scsi/fnic/vnic_scsi.h new file mode 100644 index 000000000000..46baa5254001 --- /dev/null +++ b/drivers/scsi/fnic/vnic_scsi.h @@ -0,0 +1,99 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _VNIC_SCSI_H_ +#define _VNIC_SCSI_H_ + +#define VNIC_FNIC_WQ_COPY_COUNT_MIN 1 +#define VNIC_FNIC_WQ_COPY_COUNT_MAX 1 + +#define VNIC_FNIC_WQ_DESCS_MIN 64 +#define VNIC_FNIC_WQ_DESCS_MAX 128 + +#define VNIC_FNIC_WQ_COPY_DESCS_MIN 64 +#define VNIC_FNIC_WQ_COPY_DESCS_MAX 512 + +#define VNIC_FNIC_RQ_DESCS_MIN 64 +#define VNIC_FNIC_RQ_DESCS_MAX 128 + +#define VNIC_FNIC_EDTOV_MIN 1000 +#define VNIC_FNIC_EDTOV_MAX 255000 +#define VNIC_FNIC_EDTOV_DEF 2000 + +#define VNIC_FNIC_RATOV_MIN 1000 +#define VNIC_FNIC_RATOV_MAX 255000 + +#define VNIC_FNIC_MAXDATAFIELDSIZE_MIN 256 +#define VNIC_FNIC_MAXDATAFIELDSIZE_MAX 2112 + +#define VNIC_FNIC_FLOGI_RETRIES_MIN 0 +#define VNIC_FNIC_FLOGI_RETRIES_MAX 0xffffffff +#define VNIC_FNIC_FLOGI_RETRIES_DEF 0xffffffff + +#define VNIC_FNIC_FLOGI_TIMEOUT_MIN 1000 +#define VNIC_FNIC_FLOGI_TIMEOUT_MAX 255000 + +#define VNIC_FNIC_PLOGI_RETRIES_MIN 0 +#define VNIC_FNIC_PLOGI_RETRIES_MAX 255 +#define VNIC_FNIC_PLOGI_RETRIES_DEF 8 + +#define VNIC_FNIC_PLOGI_TIMEOUT_MIN 1000 +#define VNIC_FNIC_PLOGI_TIMEOUT_MAX 255000 + +#define VNIC_FNIC_IO_THROTTLE_COUNT_MIN 256 +#define VNIC_FNIC_IO_THROTTLE_COUNT_MAX 4096 + +#define VNIC_FNIC_LINK_DOWN_TIMEOUT_MIN 0 +#define VNIC_FNIC_LINK_DOWN_TIMEOUT_MAX 240000 + +#define VNIC_FNIC_PORT_DOWN_TIMEOUT_MIN 0 +#define VNIC_FNIC_PORT_DOWN_TIMEOUT_MAX 240000 + +#define VNIC_FNIC_PORT_DOWN_IO_RETRIES_MIN 0 +#define VNIC_FNIC_PORT_DOWN_IO_RETRIES_MAX 255 + +#define VNIC_FNIC_LUNS_PER_TARGET_MIN 1 +#define VNIC_FNIC_LUNS_PER_TARGET_MAX 1024 + +/* Device-specific region: scsi configuration */ +struct vnic_fc_config { + u64 node_wwn; + u64 port_wwn; + u32 flags; + u32 wq_enet_desc_count; + u32 wq_copy_desc_count; + u32 rq_desc_count; + u32 flogi_retries; + u32 flogi_timeout; + u32 plogi_retries; + u32 plogi_timeout; + u32 io_throttle_count; + u32 link_down_timeout; + u32 port_down_timeout; + u32 port_down_io_retries; + u32 luns_per_tgt; + u16 maxdatafieldsize; + u16 ed_tov; + u16 ra_tov; + u16 intr_timer; + u8 intr_timer_type; +}; + +#define VFCF_FCP_SEQ_LVL_ERR 0x1 /* Enable FCP-2 Error Recovery */ +#define VFCF_PERBI 0x2 /* persistent binding info available */ + +#endif /* _VNIC_SCSI_H_ */ diff --git a/drivers/scsi/fnic/vnic_stats.h b/drivers/scsi/fnic/vnic_stats.h new file mode 100644 index 000000000000..5372e23c1cb3 --- /dev/null +++ b/drivers/scsi/fnic/vnic_stats.h @@ -0,0 +1,68 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _VNIC_STATS_H_ +#define _VNIC_STATS_H_ + +/* Tx statistics */ +struct vnic_tx_stats { + u64 tx_frames_ok; + u64 tx_unicast_frames_ok; + u64 tx_multicast_frames_ok; + u64 tx_broadcast_frames_ok; + u64 tx_bytes_ok; + u64 tx_unicast_bytes_ok; + u64 tx_multicast_bytes_ok; + u64 tx_broadcast_bytes_ok; + u64 tx_drops; + u64 tx_errors; + u64 tx_tso; + u64 rsvd[16]; +}; + +/* Rx statistics */ +struct vnic_rx_stats { + u64 rx_frames_ok; + u64 rx_frames_total; + u64 rx_unicast_frames_ok; + u64 rx_multicast_frames_ok; + u64 rx_broadcast_frames_ok; + u64 rx_bytes_ok; + u64 rx_unicast_bytes_ok; + u64 rx_multicast_bytes_ok; + u64 rx_broadcast_bytes_ok; + u64 rx_drop; + u64 rx_no_bufs; + u64 rx_errors; + u64 rx_rss; + u64 rx_crc_errors; + u64 rx_frames_64; + u64 rx_frames_127; + u64 rx_frames_255; + u64 rx_frames_511; + u64 rx_frames_1023; + u64 rx_frames_1518; + u64 rx_frames_to_max; + u64 rsvd[16]; +}; + +struct vnic_stats { + struct vnic_tx_stats tx; + struct vnic_rx_stats rx; +}; + +#endif /* _VNIC_STATS_H_ */ diff --git a/drivers/scsi/fnic/vnic_wq.c b/drivers/scsi/fnic/vnic_wq.c new file mode 100644 index 000000000000..1f9ea790d130 --- /dev/null +++ b/drivers/scsi/fnic/vnic_wq.c @@ -0,0 +1,182 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include +#include +#include "vnic_dev.h" +#include "vnic_wq.h" + +static int vnic_wq_alloc_bufs(struct vnic_wq *wq) +{ + struct vnic_wq_buf *buf; + struct vnic_dev *vdev; + unsigned int i, j, count = wq->ring.desc_count; + unsigned int blks = VNIC_WQ_BUF_BLKS_NEEDED(count); + + vdev = wq->vdev; + + for (i = 0; i < blks; i++) { + wq->bufs[i] = kzalloc(VNIC_WQ_BUF_BLK_SZ, GFP_ATOMIC); + if (!wq->bufs[i]) { + printk(KERN_ERR "Failed to alloc wq_bufs\n"); + return -ENOMEM; + } + } + + for (i = 0; i < blks; i++) { + buf = wq->bufs[i]; + for (j = 0; j < VNIC_WQ_BUF_BLK_ENTRIES; j++) { + buf->index = i * VNIC_WQ_BUF_BLK_ENTRIES + j; + buf->desc = (u8 *)wq->ring.descs + + wq->ring.desc_size * buf->index; + if (buf->index + 1 == count) { + buf->next = wq->bufs[0]; + break; + } else if (j + 1 == VNIC_WQ_BUF_BLK_ENTRIES) { + buf->next = wq->bufs[i + 1]; + } else { + buf->next = buf + 1; + buf++; + } + } + } + + wq->to_use = wq->to_clean = wq->bufs[0]; + + return 0; +} + +void vnic_wq_free(struct vnic_wq *wq) +{ + struct vnic_dev *vdev; + unsigned int i; + + vdev = wq->vdev; + + vnic_dev_free_desc_ring(vdev, &wq->ring); + + for (i = 0; i < VNIC_WQ_BUF_BLKS_MAX; i++) { + kfree(wq->bufs[i]); + wq->bufs[i] = NULL; + } + + wq->ctrl = NULL; + +} + +int vnic_wq_alloc(struct vnic_dev *vdev, struct vnic_wq *wq, unsigned int index, + unsigned int desc_count, unsigned int desc_size) +{ + int err; + + wq->index = index; + wq->vdev = vdev; + + wq->ctrl = vnic_dev_get_res(vdev, RES_TYPE_WQ, index); + if (!wq->ctrl) { + printk(KERN_ERR "Failed to hook WQ[%d] resource\n", index); + return -EINVAL; + } + + vnic_wq_disable(wq); + + err = vnic_dev_alloc_desc_ring(vdev, &wq->ring, desc_count, desc_size); + if (err) + return err; + + err = vnic_wq_alloc_bufs(wq); + if (err) { + vnic_wq_free(wq); + return err; + } + + return 0; +} + +void vnic_wq_init(struct vnic_wq *wq, unsigned int cq_index, + unsigned int error_interrupt_enable, + unsigned int error_interrupt_offset) +{ + u64 paddr; + + paddr = (u64)wq->ring.base_addr | VNIC_PADDR_TARGET; + writeq(paddr, &wq->ctrl->ring_base); + iowrite32(wq->ring.desc_count, &wq->ctrl->ring_size); + iowrite32(0, &wq->ctrl->fetch_index); + iowrite32(0, &wq->ctrl->posted_index); + iowrite32(cq_index, &wq->ctrl->cq_index); + iowrite32(error_interrupt_enable, &wq->ctrl->error_interrupt_enable); + iowrite32(error_interrupt_offset, &wq->ctrl->error_interrupt_offset); + iowrite32(0, &wq->ctrl->error_status); +} + +unsigned int vnic_wq_error_status(struct vnic_wq *wq) +{ + return ioread32(&wq->ctrl->error_status); +} + +void vnic_wq_enable(struct vnic_wq *wq) +{ + iowrite32(1, &wq->ctrl->enable); +} + +int vnic_wq_disable(struct vnic_wq *wq) +{ + unsigned int wait; + + iowrite32(0, &wq->ctrl->enable); + + /* Wait for HW to ACK disable request */ + for (wait = 0; wait < 100; wait++) { + if (!(ioread32(&wq->ctrl->running))) + return 0; + udelay(1); + } + + printk(KERN_ERR "Failed to disable WQ[%d]\n", wq->index); + + return -ETIMEDOUT; +} + +void vnic_wq_clean(struct vnic_wq *wq, + void (*buf_clean)(struct vnic_wq *wq, struct vnic_wq_buf *buf)) +{ + struct vnic_wq_buf *buf; + + BUG_ON(ioread32(&wq->ctrl->enable)); + + buf = wq->to_clean; + + while (vnic_wq_desc_used(wq) > 0) { + + (*buf_clean)(wq, buf); + + buf = wq->to_clean = buf->next; + wq->ring.desc_avail++; + } + + wq->to_use = wq->to_clean = wq->bufs[0]; + + iowrite32(0, &wq->ctrl->fetch_index); + iowrite32(0, &wq->ctrl->posted_index); + iowrite32(0, &wq->ctrl->error_status); + + vnic_dev_clear_desc_ring(&wq->ring); +} diff --git a/drivers/scsi/fnic/vnic_wq.h b/drivers/scsi/fnic/vnic_wq.h new file mode 100644 index 000000000000..5cd094f79281 --- /dev/null +++ b/drivers/scsi/fnic/vnic_wq.h @@ -0,0 +1,175 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _VNIC_WQ_H_ +#define _VNIC_WQ_H_ + +#include +#include "vnic_dev.h" +#include "vnic_cq.h" + +/* + * These defines avoid symbol clash between fnic and enic (Cisco 10G Eth + * Driver) when both are built with CONFIG options =y + */ +#define vnic_wq_desc_avail fnic_wq_desc_avail +#define vnic_wq_desc_used fnic_wq_desc_used +#define vnic_wq_next_desc fni_cwq_next_desc +#define vnic_wq_post fnic_wq_post +#define vnic_wq_service fnic_wq_service +#define vnic_wq_free fnic_wq_free +#define vnic_wq_alloc fnic_wq_alloc +#define vnic_wq_init fnic_wq_init +#define vnic_wq_error_status fnic_wq_error_status +#define vnic_wq_enable fnic_wq_enable +#define vnic_wq_disable fnic_wq_disable +#define vnic_wq_clean fnic_wq_clean + +/* Work queue control */ +struct vnic_wq_ctrl { + u64 ring_base; /* 0x00 */ + u32 ring_size; /* 0x08 */ + u32 pad0; + u32 posted_index; /* 0x10 */ + u32 pad1; + u32 cq_index; /* 0x18 */ + u32 pad2; + u32 enable; /* 0x20 */ + u32 pad3; + u32 running; /* 0x28 */ + u32 pad4; + u32 fetch_index; /* 0x30 */ + u32 pad5; + u32 dca_value; /* 0x38 */ + u32 pad6; + u32 error_interrupt_enable; /* 0x40 */ + u32 pad7; + u32 error_interrupt_offset; /* 0x48 */ + u32 pad8; + u32 error_status; /* 0x50 */ + u32 pad9; +}; + +struct vnic_wq_buf { + struct vnic_wq_buf *next; + dma_addr_t dma_addr; + void *os_buf; + unsigned int len; + unsigned int index; + int sop; + void *desc; +}; + +/* Break the vnic_wq_buf allocations into blocks of 64 entries */ +#define VNIC_WQ_BUF_BLK_ENTRIES 64 +#define VNIC_WQ_BUF_BLK_SZ \ + (VNIC_WQ_BUF_BLK_ENTRIES * sizeof(struct vnic_wq_buf)) +#define VNIC_WQ_BUF_BLKS_NEEDED(entries) \ + DIV_ROUND_UP(entries, VNIC_WQ_BUF_BLK_ENTRIES) +#define VNIC_WQ_BUF_BLKS_MAX VNIC_WQ_BUF_BLKS_NEEDED(4096) + +struct vnic_wq { + unsigned int index; + struct vnic_dev *vdev; + struct vnic_wq_ctrl __iomem *ctrl; /* memory-mapped */ + struct vnic_dev_ring ring; + struct vnic_wq_buf *bufs[VNIC_WQ_BUF_BLKS_MAX]; + struct vnic_wq_buf *to_use; + struct vnic_wq_buf *to_clean; + unsigned int pkts_outstanding; +}; + +static inline unsigned int vnic_wq_desc_avail(struct vnic_wq *wq) +{ + /* how many does SW own? */ + return wq->ring.desc_avail; +} + +static inline unsigned int vnic_wq_desc_used(struct vnic_wq *wq) +{ + /* how many does HW own? */ + return wq->ring.desc_count - wq->ring.desc_avail - 1; +} + +static inline void *vnic_wq_next_desc(struct vnic_wq *wq) +{ + return wq->to_use->desc; +} + +static inline void vnic_wq_post(struct vnic_wq *wq, + void *os_buf, dma_addr_t dma_addr, + unsigned int len, int sop, int eop) +{ + struct vnic_wq_buf *buf = wq->to_use; + + buf->sop = sop; + buf->os_buf = eop ? os_buf : NULL; + buf->dma_addr = dma_addr; + buf->len = len; + + buf = buf->next; + if (eop) { + /* Adding write memory barrier prevents compiler and/or CPU + * reordering, thus avoiding descriptor posting before + * descriptor is initialized. Otherwise, hardware can read + * stale descriptor fields. + */ + wmb(); + iowrite32(buf->index, &wq->ctrl->posted_index); + } + wq->to_use = buf; + + wq->ring.desc_avail--; +} + +static inline void vnic_wq_service(struct vnic_wq *wq, + struct cq_desc *cq_desc, u16 completed_index, + void (*buf_service)(struct vnic_wq *wq, + struct cq_desc *cq_desc, struct vnic_wq_buf *buf, void *opaque), + void *opaque) +{ + struct vnic_wq_buf *buf; + + buf = wq->to_clean; + while (1) { + + (*buf_service)(wq, cq_desc, buf, opaque); + + wq->ring.desc_avail++; + + wq->to_clean = buf->next; + + if (buf->index == completed_index) + break; + + buf = wq->to_clean; + } +} + +void vnic_wq_free(struct vnic_wq *wq); +int vnic_wq_alloc(struct vnic_dev *vdev, struct vnic_wq *wq, unsigned int index, + unsigned int desc_count, unsigned int desc_size); +void vnic_wq_init(struct vnic_wq *wq, unsigned int cq_index, + unsigned int error_interrupt_enable, + unsigned int error_interrupt_offset); +unsigned int vnic_wq_error_status(struct vnic_wq *wq); +void vnic_wq_enable(struct vnic_wq *wq); +int vnic_wq_disable(struct vnic_wq *wq); +void vnic_wq_clean(struct vnic_wq *wq, + void (*buf_clean)(struct vnic_wq *wq, struct vnic_wq_buf *buf)); + +#endif /* _VNIC_WQ_H_ */ diff --git a/drivers/scsi/fnic/vnic_wq_copy.c b/drivers/scsi/fnic/vnic_wq_copy.c new file mode 100644 index 000000000000..9eab7e7caf38 --- /dev/null +++ b/drivers/scsi/fnic/vnic_wq_copy.c @@ -0,0 +1,117 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include +#include +#include "vnic_wq_copy.h" + +void vnic_wq_copy_enable(struct vnic_wq_copy *wq) +{ + iowrite32(1, &wq->ctrl->enable); +} + +int vnic_wq_copy_disable(struct vnic_wq_copy *wq) +{ + unsigned int wait; + + iowrite32(0, &wq->ctrl->enable); + + /* Wait for HW to ACK disable request */ + for (wait = 0; wait < 100; wait++) { + if (!(ioread32(&wq->ctrl->running))) + return 0; + udelay(1); + } + + printk(KERN_ERR "Failed to disable Copy WQ[%d]," + " fetch index=%d, posted_index=%d\n", + wq->index, ioread32(&wq->ctrl->fetch_index), + ioread32(&wq->ctrl->posted_index)); + + return -ENODEV; +} + +void vnic_wq_copy_clean(struct vnic_wq_copy *wq, + void (*q_clean)(struct vnic_wq_copy *wq, + struct fcpio_host_req *wq_desc)) +{ + BUG_ON(ioread32(&wq->ctrl->enable)); + + if (vnic_wq_copy_desc_in_use(wq)) + vnic_wq_copy_service(wq, -1, q_clean); + + wq->to_use_index = wq->to_clean_index = 0; + + iowrite32(0, &wq->ctrl->fetch_index); + iowrite32(0, &wq->ctrl->posted_index); + iowrite32(0, &wq->ctrl->error_status); + + vnic_dev_clear_desc_ring(&wq->ring); +} + +void vnic_wq_copy_free(struct vnic_wq_copy *wq) +{ + struct vnic_dev *vdev; + + vdev = wq->vdev; + vnic_dev_free_desc_ring(vdev, &wq->ring); + wq->ctrl = NULL; +} + +int vnic_wq_copy_alloc(struct vnic_dev *vdev, struct vnic_wq_copy *wq, + unsigned int index, unsigned int desc_count, + unsigned int desc_size) +{ + int err; + + wq->index = index; + wq->vdev = vdev; + wq->to_use_index = wq->to_clean_index = 0; + wq->ctrl = vnic_dev_get_res(vdev, RES_TYPE_WQ, index); + if (!wq->ctrl) { + printk(KERN_ERR "Failed to hook COPY WQ[%d] resource\n", index); + return -EINVAL; + } + + vnic_wq_copy_disable(wq); + + err = vnic_dev_alloc_desc_ring(vdev, &wq->ring, desc_count, desc_size); + if (err) + return err; + + return 0; +} + +void vnic_wq_copy_init(struct vnic_wq_copy *wq, unsigned int cq_index, + unsigned int error_interrupt_enable, + unsigned int error_interrupt_offset) +{ + u64 paddr; + + paddr = (u64)wq->ring.base_addr | VNIC_PADDR_TARGET; + writeq(paddr, &wq->ctrl->ring_base); + iowrite32(wq->ring.desc_count, &wq->ctrl->ring_size); + iowrite32(0, &wq->ctrl->fetch_index); + iowrite32(0, &wq->ctrl->posted_index); + iowrite32(cq_index, &wq->ctrl->cq_index); + iowrite32(error_interrupt_enable, &wq->ctrl->error_interrupt_enable); + iowrite32(error_interrupt_offset, &wq->ctrl->error_interrupt_offset); +} + diff --git a/drivers/scsi/fnic/vnic_wq_copy.h b/drivers/scsi/fnic/vnic_wq_copy.h new file mode 100644 index 000000000000..6aff9740c3df --- /dev/null +++ b/drivers/scsi/fnic/vnic_wq_copy.h @@ -0,0 +1,128 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _VNIC_WQ_COPY_H_ +#define _VNIC_WQ_COPY_H_ + +#include +#include "vnic_wq.h" +#include "fcpio.h" + +#define VNIC_WQ_COPY_MAX 1 + +struct vnic_wq_copy { + unsigned int index; + struct vnic_dev *vdev; + struct vnic_wq_ctrl __iomem *ctrl; /* memory-mapped */ + struct vnic_dev_ring ring; + unsigned to_use_index; + unsigned to_clean_index; +}; + +static inline unsigned int vnic_wq_copy_desc_avail(struct vnic_wq_copy *wq) +{ + return wq->ring.desc_avail; +} + +static inline unsigned int vnic_wq_copy_desc_in_use(struct vnic_wq_copy *wq) +{ + return wq->ring.desc_count - 1 - wq->ring.desc_avail; +} + +static inline void *vnic_wq_copy_next_desc(struct vnic_wq_copy *wq) +{ + struct fcpio_host_req *desc = wq->ring.descs; + return &desc[wq->to_use_index]; +} + +static inline void vnic_wq_copy_post(struct vnic_wq_copy *wq) +{ + + ((wq->to_use_index + 1) == wq->ring.desc_count) ? + (wq->to_use_index = 0) : (wq->to_use_index++); + wq->ring.desc_avail--; + + /* Adding write memory barrier prevents compiler and/or CPU + * reordering, thus avoiding descriptor posting before + * descriptor is initialized. Otherwise, hardware can read + * stale descriptor fields. + */ + wmb(); + + iowrite32(wq->to_use_index, &wq->ctrl->posted_index); +} + +static inline void vnic_wq_copy_desc_process(struct vnic_wq_copy *wq, u16 index) +{ + unsigned int cnt; + + if (wq->to_clean_index <= index) + cnt = (index - wq->to_clean_index) + 1; + else + cnt = wq->ring.desc_count - wq->to_clean_index + index + 1; + + wq->to_clean_index = ((index + 1) % wq->ring.desc_count); + wq->ring.desc_avail += cnt; + +} + +static inline void vnic_wq_copy_service(struct vnic_wq_copy *wq, + u16 completed_index, + void (*q_service)(struct vnic_wq_copy *wq, + struct fcpio_host_req *wq_desc)) +{ + struct fcpio_host_req *wq_desc = wq->ring.descs; + unsigned int curr_index; + + while (1) { + + if (q_service) + (*q_service)(wq, &wq_desc[wq->to_clean_index]); + + wq->ring.desc_avail++; + + curr_index = wq->to_clean_index; + + /* increment the to-clean index so that we start + * with an unprocessed index next time we enter the loop + */ + ((wq->to_clean_index + 1) == wq->ring.desc_count) ? + (wq->to_clean_index = 0) : (wq->to_clean_index++); + + if (curr_index == completed_index) + break; + + /* we have cleaned all the entries */ + if ((completed_index == (u16)-1) && + (wq->to_clean_index == wq->to_use_index)) + break; + } +} + +void vnic_wq_copy_enable(struct vnic_wq_copy *wq); +int vnic_wq_copy_disable(struct vnic_wq_copy *wq); +void vnic_wq_copy_free(struct vnic_wq_copy *wq); +int vnic_wq_copy_alloc(struct vnic_dev *vdev, struct vnic_wq_copy *wq, + unsigned int index, unsigned int desc_count, unsigned int desc_size); +void vnic_wq_copy_init(struct vnic_wq_copy *wq, unsigned int cq_index, + unsigned int error_interrupt_enable, + unsigned int error_interrupt_offset); +void vnic_wq_copy_clean(struct vnic_wq_copy *wq, + void (*q_clean)(struct vnic_wq_copy *wq, + struct fcpio_host_req *wq_desc)); + +#endif /* _VNIC_WQ_COPY_H_ */ diff --git a/drivers/scsi/fnic/wq_enet_desc.h b/drivers/scsi/fnic/wq_enet_desc.h new file mode 100644 index 000000000000..b121cbad18b8 --- /dev/null +++ b/drivers/scsi/fnic/wq_enet_desc.h @@ -0,0 +1,96 @@ +/* + * Copyright 2008 Cisco Systems, Inc. All rights reserved. + * Copyright 2007 Nuova Systems, Inc. All rights reserved. + * + * This program is free software; you may redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _WQ_ENET_DESC_H_ +#define _WQ_ENET_DESC_H_ + +/* Ethernet work queue descriptor: 16B */ +struct wq_enet_desc { + __le64 address; + __le16 length; + __le16 mss_loopback; + __le16 header_length_flags; + __le16 vlan_tag; +}; + +#define WQ_ENET_ADDR_BITS 64 +#define WQ_ENET_LEN_BITS 14 +#define WQ_ENET_LEN_MASK ((1 << WQ_ENET_LEN_BITS) - 1) +#define WQ_ENET_MSS_BITS 14 +#define WQ_ENET_MSS_MASK ((1 << WQ_ENET_MSS_BITS) - 1) +#define WQ_ENET_MSS_SHIFT 2 +#define WQ_ENET_LOOPBACK_SHIFT 1 +#define WQ_ENET_HDRLEN_BITS 10 +#define WQ_ENET_HDRLEN_MASK ((1 << WQ_ENET_HDRLEN_BITS) - 1) +#define WQ_ENET_FLAGS_OM_BITS 2 +#define WQ_ENET_FLAGS_OM_MASK ((1 << WQ_ENET_FLAGS_OM_BITS) - 1) +#define WQ_ENET_FLAGS_EOP_SHIFT 12 +#define WQ_ENET_FLAGS_CQ_ENTRY_SHIFT 13 +#define WQ_ENET_FLAGS_FCOE_ENCAP_SHIFT 14 +#define WQ_ENET_FLAGS_VLAN_TAG_INSERT_SHIFT 15 + +#define WQ_ENET_OFFLOAD_MODE_CSUM 0 +#define WQ_ENET_OFFLOAD_MODE_RESERVED 1 +#define WQ_ENET_OFFLOAD_MODE_CSUM_L4 2 +#define WQ_ENET_OFFLOAD_MODE_TSO 3 + +static inline void wq_enet_desc_enc(struct wq_enet_desc *desc, + u64 address, u16 length, u16 mss, u16 header_length, + u8 offload_mode, u8 eop, u8 cq_entry, u8 fcoe_encap, + u8 vlan_tag_insert, u16 vlan_tag, u8 loopback) +{ + desc->address = cpu_to_le64(address); + desc->length = cpu_to_le16(length & WQ_ENET_LEN_MASK); + desc->mss_loopback = cpu_to_le16((mss & WQ_ENET_MSS_MASK) << + WQ_ENET_MSS_SHIFT | (loopback & 1) << WQ_ENET_LOOPBACK_SHIFT); + desc->header_length_flags = cpu_to_le16( + (header_length & WQ_ENET_HDRLEN_MASK) | + (offload_mode & WQ_ENET_FLAGS_OM_MASK) << WQ_ENET_HDRLEN_BITS | + (eop & 1) << WQ_ENET_FLAGS_EOP_SHIFT | + (cq_entry & 1) << WQ_ENET_FLAGS_CQ_ENTRY_SHIFT | + (fcoe_encap & 1) << WQ_ENET_FLAGS_FCOE_ENCAP_SHIFT | + (vlan_tag_insert & 1) << WQ_ENET_FLAGS_VLAN_TAG_INSERT_SHIFT); + desc->vlan_tag = cpu_to_le16(vlan_tag); +} + +static inline void wq_enet_desc_dec(struct wq_enet_desc *desc, + u64 *address, u16 *length, u16 *mss, u16 *header_length, + u8 *offload_mode, u8 *eop, u8 *cq_entry, u8 *fcoe_encap, + u8 *vlan_tag_insert, u16 *vlan_tag, u8 *loopback) +{ + *address = le64_to_cpu(desc->address); + *length = le16_to_cpu(desc->length) & WQ_ENET_LEN_MASK; + *mss = (le16_to_cpu(desc->mss_loopback) >> WQ_ENET_MSS_SHIFT) & + WQ_ENET_MSS_MASK; + *loopback = (u8)((le16_to_cpu(desc->mss_loopback) >> + WQ_ENET_LOOPBACK_SHIFT) & 1); + *header_length = le16_to_cpu(desc->header_length_flags) & + WQ_ENET_HDRLEN_MASK; + *offload_mode = (u8)((le16_to_cpu(desc->header_length_flags) >> + WQ_ENET_HDRLEN_BITS) & WQ_ENET_FLAGS_OM_MASK); + *eop = (u8)((le16_to_cpu(desc->header_length_flags) >> + WQ_ENET_FLAGS_EOP_SHIFT) & 1); + *cq_entry = (u8)((le16_to_cpu(desc->header_length_flags) >> + WQ_ENET_FLAGS_CQ_ENTRY_SHIFT) & 1); + *fcoe_encap = (u8)((le16_to_cpu(desc->header_length_flags) >> + WQ_ENET_FLAGS_FCOE_ENCAP_SHIFT) & 1); + *vlan_tag_insert = (u8)((le16_to_cpu(desc->header_length_flags) >> + WQ_ENET_FLAGS_VLAN_TAG_INSERT_SHIFT) & 1); + *vlan_tag = le16_to_cpu(desc->vlan_tag); +} + +#endif /* _WQ_ENET_DESC_H_ */ From fbc97e4c5c31ea198f912196b1379d7493362800 Mon Sep 17 00:00:00 2001 From: Alan Jenkins Date: Mon, 27 Apr 2009 09:23:37 +0200 Subject: [PATCH 029/267] eeepc-laptop: fix wlan rfkill state change during init When an rfkill device is registered, the rfkill core will change its state to the system default. So we need to prepare for state changes *before* we register it. That means installing the eeepc-specific ACPI callback which handles the hotplug of the wireless network adaptor. This problem doesn't occur during normal operation. You have to 1) Boot with wireless enabled. eeepc-laptop should load automatically. 2) modprobe -r eeepc-laptop 3) modprobe eeepc-laptop On boot, the default rfkill state will be set to enabled. With the current core code, step 2) will disable the wireless. Therefore in step 3), the wireless will change state during registration, from disabled to enabled. But without this fix, the PCI device for the wireless adaptor will not appear. Signed-off-by: Alan Jenkins Acked-by: Matthew Garrett Signed-off-by: Corentin Chary Signed-off-by: Len Brown --- drivers/platform/x86/eeepc-laptop.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c index 6f54fd1757cd..e21f7cd72e4e 100644 --- a/drivers/platform/x86/eeepc-laptop.c +++ b/drivers/platform/x86/eeepc-laptop.c @@ -649,6 +649,9 @@ static int eeepc_hotk_add(struct acpi_device *device) if (ACPI_FAILURE(status)) printk(EEEPC_ERR "Error installing notify handler\n"); + eeepc_register_rfkill_notifier("\\_SB.PCI0.P0P6"); + eeepc_register_rfkill_notifier("\\_SB.PCI0.P0P7"); + if (get_acpi(CM_ASL_WLAN) != -1) { ehotk->eeepc_wlan_rfkill = rfkill_allocate(&device->dev, RFKILL_TYPE_WLAN); @@ -704,9 +707,6 @@ static int eeepc_hotk_add(struct acpi_device *device) goto bluetooth_fail; } - eeepc_register_rfkill_notifier("\\_SB.PCI0.P0P6"); - eeepc_register_rfkill_notifier("\\_SB.PCI0.P0P7"); - return 0; bluetooth_fail: From 64b86b6583db832b28bb54575e32b9e2a1a7d84f Mon Sep 17 00:00:00 2001 From: Darren Salt Date: Mon, 27 Apr 2009 09:23:38 +0200 Subject: [PATCH 030/267] eeepc-laptop: report brightness control events via the input layer This maps the brightness control events to one of two keys, either KEY_BRIGHTNESSDOWN or KEY_BRIGHTNESSUP, as needed. Some mapping has to be done due to the fact that the BIOS reports them as + ; the selection is done according to the sign of the change in brightness (if this is 0, no keypress is reported). (Ref. http://lists.alioth.debian.org/pipermail/debian-eeepc-devel/2009-April/002001.html) Signed-off-by: Darren Salt Signed-off-by: Corentin Chary Signed-off-by: Len Brown --- drivers/platform/x86/eeepc-laptop.c | 29 ++++++++++++++++++++++++++--- 1 file changed, 26 insertions(+), 3 deletions(-) diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c index e21f7cd72e4e..f54cfeac5221 100644 --- a/drivers/platform/x86/eeepc-laptop.c +++ b/drivers/platform/x86/eeepc-laptop.c @@ -166,6 +166,8 @@ static struct key_entry eeepc_keymap[] = { {KE_KEY, 0x1b, KEY_ZOOM }, {KE_KEY, 0x1c, KEY_PROG2 }, {KE_KEY, 0x1d, KEY_PROG3 }, + {KE_KEY, NOTIFY_BRN_MIN, KEY_BRIGHTNESSDOWN }, + {KE_KEY, NOTIFY_BRN_MIN + 2, KEY_BRIGHTNESSUP }, {KE_KEY, 0x30, KEY_SWITCHVIDEOMODE }, {KE_KEY, 0x31, KEY_SWITCHVIDEOMODE }, {KE_KEY, 0x32, KEY_SWITCHVIDEOMODE }, @@ -512,11 +514,16 @@ static int eeepc_hotk_check(void) return 0; } -static void notify_brn(void) +static int notify_brn(void) { + /* returns the *previous* brightness, or -1 */ struct backlight_device *bd = eeepc_backlight_device; - if (bd) + if (bd) { + int old = bd->props.brightness; bd->props.brightness = read_brightness(bd); + return old; + } + return -1; } static void eeepc_rfkill_notify(acpi_handle handle, u32 event, void *data) @@ -558,17 +565,33 @@ static void eeepc_hotk_notify(acpi_handle handle, u32 event, void *data) { static struct key_entry *key; u16 count; + int brn = -ENODEV; if (!ehotk) return; if (event >= NOTIFY_BRN_MIN && event <= NOTIFY_BRN_MAX) - notify_brn(); + brn = notify_brn(); count = ehotk->event_count[event % 128]++; acpi_bus_generate_proc_event(ehotk->device, event, count); acpi_bus_generate_netlink_event(ehotk->device->pnp.device_class, dev_name(&ehotk->device->dev), event, count); if (ehotk->inputdev) { + if (brn != -ENODEV) { + /* brightness-change events need special + * handling for conversion to key events + */ + if (brn < 0) + brn = event; + else + brn += NOTIFY_BRN_MIN; + if (event < brn) + event = NOTIFY_BRN_MIN; /* brightness down */ + else if (event > brn) + event = NOTIFY_BRN_MIN + 2; /* ... up */ + else + event = NOTIFY_BRN_MIN + 1; /* ... unchanged */ + } key = eepc_get_entry_by_scancode(event); if (key) { switch (key->type) { From 978605c4fd8e7470f225eec7b5aab69d8796afcc Mon Sep 17 00:00:00 2001 From: Alan Jenkins Date: Mon, 27 Apr 2009 09:23:39 +0200 Subject: [PATCH 031/267] eeepc-laptop: Work around rfkill firmware bug 1) Buggy firmware can change the RFKILL state by itself. This is easily detected. The RFKILL API states that in such cases, we should call rfkill_force_state() to notify the core. I have reported the bug to Asus. I believe this is the right thing to do for robustness, even if this particular firmware bug is fixed. 2) The same bug causes the wireless toggle key to be reported as 0x11 instead of 0x10. 0x11 is otherwise unused, so it should be safe to add this as a new keycode. The bug is triggered by removing the laptop battery while hibernated. On resume, the wireless toggle key causes the firmware to toggle the wireless state itself. (Also, the key is reported as 0x11 when the current wireless state is OFF). This is very poor behaviour because the OS can't predict whether the firmware is controlling the RFKILL state. Without this workaround, the bug means users have to press the wireless toggle key twice to enable, due to the OS/firmware conflict. (Assuming rfkill-input or equivalent is being used). The workaround avoids this. I believe that acpid scripts which toggle the value of the sysfs state file when the toggle key is pressed will be rendered ineffective by the bug, regardless of this workaround. If they simply toggle the state, when the firmware has already toggled it, then you will never see a state change. Tested on "EEEPC 4G" only. Signed-off-by: Alan Jenkins Signed-off-by: Corentin Chary Signed-off-by: Len Brown --- drivers/platform/x86/eeepc-laptop.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c index f54cfeac5221..57f21f0a5655 100644 --- a/drivers/platform/x86/eeepc-laptop.c +++ b/drivers/platform/x86/eeepc-laptop.c @@ -158,6 +158,7 @@ enum { KE_KEY, KE_END }; static struct key_entry eeepc_keymap[] = { /* Sleep already handled via generic ACPI code */ {KE_KEY, 0x10, KEY_WLAN }, + {KE_KEY, 0x11, KEY_WLAN }, {KE_KEY, 0x12, KEY_PROG1 }, {KE_KEY, 0x13, KEY_MUTE }, {KE_KEY, 0x14, KEY_VOLUMEDOWN }, @@ -528,6 +529,7 @@ static int notify_brn(void) static void eeepc_rfkill_notify(acpi_handle handle, u32 event, void *data) { + enum rfkill_state state; struct pci_dev *dev; struct pci_bus *bus = pci_find_bus(0, 1); @@ -539,7 +541,9 @@ static void eeepc_rfkill_notify(acpi_handle handle, u32 event, void *data) return; } - if (get_acpi(CM_ASL_WLAN) == 1) { + eeepc_wlan_rfkill_state(ehotk->eeepc_wlan_rfkill, &state); + + if (state == RFKILL_STATE_UNBLOCKED) { dev = pci_get_slot(bus, 0); if (dev) { /* Device already present */ @@ -559,6 +563,8 @@ static void eeepc_rfkill_notify(acpi_handle handle, u32 event, void *data) pci_dev_put(dev); } } + + rfkill_force_state(ehotk->eeepc_wlan_rfkill, state); } static void eeepc_hotk_notify(acpi_handle handle, u32 event, void *data) From 158ca1d75dd0d6223f3b1dd741d30777da62ab80 Mon Sep 17 00:00:00 2001 From: Grigori Goronzy Date: Mon, 27 Apr 2009 09:23:40 +0200 Subject: [PATCH 032/267] eeepc-laptop: support for super hybrid engine (SHE) The older eeepc-acpi driver allowed to control the SHE performance preset through a ACPI function for just this purpose. SHE underclocks and undervolts the FSB and undervolts the CPU (at preset 2, "powersave"), or slightly overclocks the CPU (at preset 0, "performance"). Preset 1 is the default setting with default clocks and voltage. The new eeepc-laptop driver doesn't support it anymore. The attached patch adds support for it to eeepc-laptop. It's very straight-forward and almost trivial. Signed-off-by: Grigori Goronzy Signed-off-by: Corentin Chary Signed-off-by: Len Brown --- drivers/platform/x86/eeepc-laptop.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c index 57f21f0a5655..7aaf5879f666 100644 --- a/drivers/platform/x86/eeepc-laptop.c +++ b/drivers/platform/x86/eeepc-laptop.c @@ -384,11 +384,13 @@ static ssize_t show_sys_acpi(int cm, char *buf) EEEPC_CREATE_DEVICE_ATTR(camera, CM_ASL_CAMERA); EEEPC_CREATE_DEVICE_ATTR(cardr, CM_ASL_CARDREADER); EEEPC_CREATE_DEVICE_ATTR(disp, CM_ASL_DISPLAYSWITCH); +EEEPC_CREATE_DEVICE_ATTR(cpufv, CM_ASL_CPUFV); static struct attribute *platform_attributes[] = { &dev_attr_camera.attr, &dev_attr_cardr.attr, &dev_attr_disp.attr, + &dev_attr_cpufv.attr, NULL }; From 309f5fbda37d5e8f1233e8b80b8e9de77262e864 Mon Sep 17 00:00:00 2001 From: Corentin Chary Date: Mon, 27 Apr 2009 09:23:42 +0200 Subject: [PATCH 033/267] asus-laptop: fix input keycode KEY_STOP is now KEY_STOPCD It's the correct key to stop a media BTN_EXTRA is now KEY_SCREENLOCK: The laptop manual tells us that this key is for screenlock KEY_TV is now KEY_PROG1 So it can be reported to X server Ref: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/361505 Signed-off-by: Corentin Chary Signed-off-by: Len Brown --- drivers/platform/x86/asus-laptop.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/platform/x86/asus-laptop.c b/drivers/platform/x86/asus-laptop.c index eeafc6c0160d..bfc1a8892a32 100644 --- a/drivers/platform/x86/asus-laptop.c +++ b/drivers/platform/x86/asus-laptop.c @@ -269,16 +269,16 @@ static struct key_entry asus_keymap[] = { {KE_KEY, 0x34, KEY_SWITCHVIDEOMODE}, {KE_KEY, 0x40, KEY_PREVIOUSSONG}, {KE_KEY, 0x41, KEY_NEXTSONG}, - {KE_KEY, 0x43, KEY_STOP}, + {KE_KEY, 0x43, KEY_STOPCD}, {KE_KEY, 0x45, KEY_PLAYPAUSE}, {KE_KEY, 0x50, KEY_EMAIL}, {KE_KEY, 0x51, KEY_WWW}, - {KE_KEY, 0x5C, BTN_EXTRA}, /* Performance */ + {KE_KEY, 0x5C, KEY_SCREENLOCK}, /* Screenlock */ {KE_KEY, 0x5D, KEY_WLAN}, {KE_KEY, 0x61, KEY_SWITCHVIDEOMODE}, {KE_KEY, 0x6B, BTN_TOUCH}, /* Lock Mouse */ {KE_KEY, 0x82, KEY_CAMERA}, - {KE_KEY, 0x8A, KEY_TV}, + {KE_KEY, 0x8A, KEY_PROG1}, {KE_KEY, 0x95, KEY_MEDIA}, {KE_KEY, 0x99, KEY_PHONE}, {KE_END, 0}, From bd32005e126a465deda5d046a62f6bb842f4d9cf Mon Sep 17 00:00:00 2001 From: Corentin Chary Date: Mon, 27 Apr 2009 09:23:43 +0200 Subject: [PATCH 034/267] eeepc-laptop: unregister_rfkill_notifier on failure If there is a failure during eeepc_hotk_add() we need to remove the acpi_notify_handler. Signed-off-by: Corentin Chary Signed-off-by: Len Brown --- drivers/platform/x86/eeepc-laptop.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c index 7aaf5879f666..353a898c3693 100644 --- a/drivers/platform/x86/eeepc-laptop.c +++ b/drivers/platform/x86/eeepc-laptop.c @@ -748,6 +748,8 @@ static int eeepc_hotk_add(struct acpi_device *device) wlan_fail: if (ehotk->eeepc_wlan_rfkill) rfkill_free(ehotk->eeepc_wlan_rfkill); + eeepc_unregister_rfkill_notifier("\\_SB.PCI0.P0P6"); + eeepc_unregister_rfkill_notifier("\\_SB.PCI0.P0P7"); ehotk_fail: kfree(ehotk); ehotk = NULL; From 29321357ac6db54eeb8574da1f6c3e0ce8cfbb60 Mon Sep 17 00:00:00 2001 From: Vladimir Zajac Date: Wed, 6 May 2009 19:34:21 +0200 Subject: [PATCH 035/267] thermal: fix off-by-1 error in trip point trigger condition This patch fixes a regression caused by commit b1569e99c795bf83b4ddf41c4f1c42761ab7f75e "ACPI: move thermal trip handling to generic thermal layer" which accidentally changed trip point trigger condition to temp > trip_temp This patch changes the trigger condition back to temp >= trip_temp Signed-off-by: Vladimir Zajac Acked-by: Zhang Rui Acked-by: Matthew Garrett Signed-off-by: Andrew Morton Signed-off-by: Len Brown --- drivers/thermal/thermal_sys.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/thermal/thermal_sys.c b/drivers/thermal/thermal_sys.c index d0b093b66adc..5e38ba10a3a9 100644 --- a/drivers/thermal/thermal_sys.c +++ b/drivers/thermal/thermal_sys.c @@ -961,7 +961,7 @@ void thermal_zone_device_update(struct thermal_zone_device *tz) switch (trip_type) { case THERMAL_TRIP_CRITICAL: - if (temp > trip_temp) { + if (temp >= trip_temp) { if (tz->ops->notify) ret = tz->ops->notify(tz, count, trip_type); @@ -974,7 +974,7 @@ void thermal_zone_device_update(struct thermal_zone_device *tz) } break; case THERMAL_TRIP_HOT: - if (temp > trip_temp) + if (temp >= trip_temp) if (tz->ops->notify) tz->ops->notify(tz, count, trip_type); break; @@ -986,14 +986,14 @@ void thermal_zone_device_update(struct thermal_zone_device *tz) cdev = instance->cdev; - if (temp > trip_temp) + if (temp >= trip_temp) cdev->ops->set_cur_state(cdev, 1); else cdev->ops->set_cur_state(cdev, 0); } break; case THERMAL_TRIP_PASSIVE: - if (temp > trip_temp || tz->passive) + if (temp >= trip_temp || tz->passive) thermal_zone_device_passive(tz, temp, trip_temp, count); break; From c53a284f8be23735dc6b53929640a987055f2933 Mon Sep 17 00:00:00 2001 From: Edward Goggin Date: Thu, 9 Apr 2009 10:02:22 -0700 Subject: [PATCH 036/267] [SCSI] initialize max_target_blocked in scsi_alloc_target This patch initializes the max_target_blocked field of a scsi target structure so that a queuecommand return value of SCSI_MLQUEUE_TARGET_BUSY will actually result in having the scsi_queue_insert blocking the device queue before requeuing the command and running the queue. Otherwise, can and does cause livelock on single CPU configurations if/when open-iSCSI software initiator's command PDU window fills. Signed-off-by: Ed Goggin Signed-off-by: James Bottomley --- drivers/scsi/scsi_scan.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index 6f51ca485f35..e2b50d8f57a8 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -425,6 +425,7 @@ static struct scsi_target *scsi_alloc_target(struct device *parent, INIT_LIST_HEAD(&starget->devices); starget->state = STARGET_CREATED; starget->scsi_level = SCSI_2; + starget->max_target_blocked = SCSI_DEFAULT_TARGET_BLOCKED; retry: spin_lock_irqsave(shost->host_lock, flags); From 4e5b6d006b61b6b998e15da850d959520f811b4c Mon Sep 17 00:00:00 2001 From: Inaky Perez-Gonzalez Date: Fri, 8 May 2009 14:02:32 -0700 Subject: [PATCH 037/267] wimax/i2400m: fix device crash: fix optimization in _roq_queue_update_ws When the i2400m receives data and the device indicates there has to be reordering, we keep an sliding window implementation to sort the packets before sending them to the network stack. One of the "operations" that the device indicates is "queue a packet and update the window start". When the queue is empty, this is equivalent to "deliver the packet and update the window start". That case was optimized in i2400m_roq_queue_update_ws() so that we would not pointlessly queue and dequeue a packet. However, when the optimization was active, it wasn't updating the window start. That caused the reorder management code to get confused later on with what seemed to be wrong reorder requests from the device. Thus the fix implemented is to do the right thing and update the window start in both cases, when the queue is empty (and the optimization is done) and when not. Signed-off-by: Inaky Perez-Gonzalez --- drivers/net/wimax/i2400m/rx.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/net/wimax/i2400m/rx.c b/drivers/net/wimax/i2400m/rx.c index 02419bfd64b5..f9fc38902322 100644 --- a/drivers/net/wimax/i2400m/rx.c +++ b/drivers/net/wimax/i2400m/rx.c @@ -819,10 +819,9 @@ void i2400m_roq_queue_update_ws(struct i2400m *i2400m, struct i2400m_roq *roq, roq_data = (struct i2400m_roq_data *) &skb->cb; i2400m_net_erx(i2400m, skb, roq_data->cs); } - else { + else __i2400m_roq_queue(i2400m, roq, skb, sn, nsn); - __i2400m_roq_update_ws(i2400m, roq, sn + 1); - } + __i2400m_roq_update_ws(i2400m, roq, sn + 1); i2400m_roq_log_add(i2400m, roq, I2400M_RO_TYPE_PACKET_WS, old_ws, len, sn, nsn, roq->ws); } From ceec1c33c36e4783cba41416f327f4bbc05218c3 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 17 Apr 2009 12:29:22 +0100 Subject: [PATCH 038/267] [ARM] nwfpe: Add decleration for ExtendedCPDO Add header file decleration for 'ExtendedCPDO' in fpa11.h to stop the following sparse warning: extended_cpdo.c:90:14: warning: symbol 'ExtendedCPDO' was not declared. Should it be static? Signed-off-by: Ben Dooks --- arch/arm/nwfpe/fpa11.h | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/arm/nwfpe/fpa11.h b/arch/arm/nwfpe/fpa11.h index 386cbd13eaf4..d3a6f9298e9e 100644 --- a/arch/arm/nwfpe/fpa11.h +++ b/arch/arm/nwfpe/fpa11.h @@ -114,4 +114,8 @@ extern unsigned int SingleCPDO(struct roundingData *roundData, extern unsigned int DoubleCPDO(struct roundingData *roundData, const unsigned int opcode, FPREG * rFd); +/* extneded_cpdo.c */ +extern unsigned int ExtendedCPDO(struct roundingData *roundData, + const unsigned int opcode, FPREG * rFd); + #endif From 3ea385f061d08a9509a884e347ad1007bb6c3c66 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 17 Apr 2009 12:21:56 +0100 Subject: [PATCH 039/267] [ARM] nwfpe: fix 'floatx80_is_nan' sparse warning The symbol 'floatx80_is_nan' prototype was defined locally in fpa11_cprt.c when it was built outside the file in softfloat-specialisze. Move this into softfloat.h to fix the following sparse warning: softfloat-specialize:276:6: warning: symbol 'floatx80_is_nan' was not declared. Should it be static? Signed-off-by: Ben Dooks --- arch/arm/nwfpe/fpa11_cprt.c | 4 ---- arch/arm/nwfpe/softfloat.h | 2 ++ 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/arch/arm/nwfpe/fpa11_cprt.c b/arch/arm/nwfpe/fpa11_cprt.c index 9843dc533047..31c4eeec18b0 100644 --- a/arch/arm/nwfpe/fpa11_cprt.c +++ b/arch/arm/nwfpe/fpa11_cprt.c @@ -27,10 +27,6 @@ #include "fpmodule.inl" #include "softfloat.h" -#ifdef CONFIG_FPE_NWFPE_XP -extern flag floatx80_is_nan(floatx80); -#endif - unsigned int PerformFLT(const unsigned int opcode); unsigned int PerformFIX(const unsigned int opcode); diff --git a/arch/arm/nwfpe/softfloat.h b/arch/arm/nwfpe/softfloat.h index 260fe29d73f5..13e479c5da57 100644 --- a/arch/arm/nwfpe/softfloat.h +++ b/arch/arm/nwfpe/softfloat.h @@ -226,6 +226,8 @@ char floatx80_le_quiet( floatx80, floatx80 ); char floatx80_lt_quiet( floatx80, floatx80 ); char floatx80_is_signaling_nan( floatx80 ); +extern flag floatx80_is_nan(floatx80); + #endif static inline flag extractFloat32Sign(float32 a) From f36dd6e7c0c04a056d3441185a17e5d4e3723192 Mon Sep 17 00:00:00 2001 From: Marek Szyprowski Date: Thu, 7 May 2009 15:07:47 +0200 Subject: [PATCH 040/267] [ARM] S3C64XX: GPIO include cleanup Cleanup arm/plat-s3c64xx/include/plat/gpio-bank-h.h include file. Using shift-left operation with value >32 is a bad habit. Reviewed-by: Kyungmin Park Signed-off-by: Marek Szyprowski Signed-off-by: Ben Dooks --- .../plat-s3c64xx/include/plat/gpio-bank-h.h | 20 +++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/arch/arm/plat-s3c64xx/include/plat/gpio-bank-h.h b/arch/arm/plat-s3c64xx/include/plat/gpio-bank-h.h index 81549516572f..2ba1767512d7 100644 --- a/arch/arm/plat-s3c64xx/include/plat/gpio-bank-h.h +++ b/arch/arm/plat-s3c64xx/include/plat/gpio-bank-h.h @@ -61,14 +61,14 @@ #define S3C64XX_GPH7_ADDR_CF1 (0x06 << 28) #define S3C64XX_GPH7_EINT_G6_7 (0x07 << 28) -#define S3C64XX_GPH8_MMC1_DATA6 (0x02 << 32) -#define S3C64XX_GPH8_MMC2_DATA2 (0x03 << 32) -#define S3C64XX_GPH8_I2S_V40_LRCLK (0x05 << 32) -#define S3C64XX_GPH8_ADDR_CF2 (0x06 << 32) -#define S3C64XX_GPH8_EINT_G6_8 (0x07 << 32) - -#define S3C64XX_GPH9_MMC1_DATA7 (0x02 << 36) -#define S3C64XX_GPH9_MMC2_DATA3 (0x03 << 36) -#define S3C64XX_GPH9_I2S_V40_DI (0x05 << 36) -#define S3C64XX_GPH9_EINT_G6_9 (0x07 << 36) +#define S3C64XX_GPH8_MMC1_DATA6 (0x02 << 0) +#define S3C64XX_GPH8_MMC2_DATA2 (0x03 << 0) +#define S3C64XX_GPH8_I2S_V40_LRCLK (0x05 << 0) +#define S3C64XX_GPH8_ADDR_CF2 (0x06 << 0) +#define S3C64XX_GPH8_EINT_G6_8 (0x07 << 0) +#define S3C64XX_GPH9_OUTPUT (0x01 << 4) +#define S3C64XX_GPH9_MMC1_DATA7 (0x02 << 4) +#define S3C64XX_GPH9_MMC2_DATA3 (0x03 << 4) +#define S3C64XX_GPH9_I2S_V40_DI (0x05 << 4) +#define S3C64XX_GPH9_EINT_G6_9 (0x07 << 4) From beb9f4ed22ee618ffcf5064ff7d5ca4cfc83104b Mon Sep 17 00:00:00 2001 From: Marek Szyprowski Date: Thu, 7 May 2009 15:07:41 +0200 Subject: [PATCH 041/267] [ARM] S3C64XX: fix GPIO debug Fix compilation bug when debug was enabled Reviewed-by: Kyungmin Park Signed-off-by: Marek Szyprowski Signed-off-by: Ben Dooks --- arch/arm/plat-s3c64xx/gpiolib.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/plat-s3c64xx/gpiolib.c b/arch/arm/plat-s3c64xx/gpiolib.c index ee9188add8fb..78ee52cffc9e 100644 --- a/arch/arm/plat-s3c64xx/gpiolib.c +++ b/arch/arm/plat-s3c64xx/gpiolib.c @@ -57,7 +57,7 @@ #if 1 #define gpio_dbg(x...) do { } while(0) #else -#define gpio_dbg(x...) printk(KERN_DEBUG ## x) +#define gpio_dbg(x...) printk(KERN_DEBUG x) #endif /* The s3c64xx_gpiolib_4bit routines are to control the gpio banks where From 871fcd7cf791b853ae044f813db94caefbf66725 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 15 May 2009 14:20:53 +0100 Subject: [PATCH 042/267] [ARM] S3C24XX: Fix unused code warning in arch/arm/plat-s3c24xx/dma.c Fix unused code warning in arch/arm/plat-s3c24xx/dma.c if there is no PM support enabled. The function to_dma_chan() should be marked inline so that the compiler will eliminate it without warning if it isn't used. arch/arm/plat-s3c24xx/dma.c:1239: warning: 'to_dma_chan' defined but not used Signed-off-by: Ben Dooks --- arch/arm/plat-s3c24xx/dma.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/plat-s3c24xx/dma.c b/arch/arm/plat-s3c24xx/dma.c index aee2aeb46c60..07326f632361 100644 --- a/arch/arm/plat-s3c24xx/dma.c +++ b/arch/arm/plat-s3c24xx/dma.c @@ -1235,7 +1235,7 @@ int s3c2410_dma_getposition(unsigned int channel, dma_addr_t *src, dma_addr_t *d EXPORT_SYMBOL(s3c2410_dma_getposition); -static struct s3c2410_dma_chan *to_dma_chan(struct sys_device *dev) +static inline struct s3c2410_dma_chan *to_dma_chan(struct sys_device *dev) { return container_of(dev, struct s3c2410_dma_chan, dev); } From a8af6de00fafed316fea8f39d87c7e8e19ec1ea0 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 15 May 2009 14:57:09 +0100 Subject: [PATCH 043/267] [ARM] S3C2410: mach-bast.c registering i2c data too early The BAST support code is calling s3c_i2c0_set_platdata() from the map_io() entry, instead of the bast_init() code. This causes the registration to fail due to kmalloc() not being available at the time. This fixes the following error: s3c_i2c0_set_platdata: no memory for platform data Signed-off-by: Ben Dooks --- arch/arm/mach-s3c2410/mach-bast.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/arm/mach-s3c2410/mach-bast.c b/arch/arm/mach-s3c2410/mach-bast.c index 4389c160f7d0..8637dea5e150 100644 --- a/arch/arm/mach-s3c2410/mach-bast.c +++ b/arch/arm/mach-s3c2410/mach-bast.c @@ -588,8 +588,6 @@ static void __init bast_map_io(void) s3c_device_nand.dev.platform_data = &bast_nand_info; - s3c_i2c0_set_platdata(&bast_i2c_info); - s3c24xx_init_io(bast_iodesc, ARRAY_SIZE(bast_iodesc)); s3c24xx_init_clocks(0); s3c24xx_init_uarts(bast_uartcfgs, ARRAY_SIZE(bast_uartcfgs)); @@ -602,6 +600,7 @@ static void __init bast_init(void) sysdev_class_register(&bast_pm_sysclass); sysdev_register(&bast_pm_sysdev); + s3c_i2c0_set_platdata(&bast_i2c_info); s3c24xx_fb_set_platdata(&bast_fb_info); platform_add_devices(bast_devices, ARRAY_SIZE(bast_devices)); From 3ac19bb443255579b30f05af5e688b6c73b1bb91 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 15 May 2009 15:21:57 +0100 Subject: [PATCH 044/267] [ARM] S3C: Do not set clk->owner field if unset The s3c24xx_register_clock() function has been doing a test on clk->owner to see if it is NULL, and then setting itself as the owner if clk->owner == NULL. This is not needed, arch/arm/plat-s3c/clock.c cannot be compiled as a module, and even if it was, it should not be playing with this field if it being registered from somewhere else. The best course of action is to remove this bit of code completely. Signed-off-by: Ben Dooks --- arch/arm/plat-s3c/clock.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/arch/arm/plat-s3c/clock.c b/arch/arm/plat-s3c/clock.c index b6be76e2fe51..4d01ef1a25dd 100644 --- a/arch/arm/plat-s3c/clock.c +++ b/arch/arm/plat-s3c/clock.c @@ -306,8 +306,6 @@ struct clk s3c24xx_uclk = { int s3c24xx_register_clock(struct clk *clk) { - clk->owner = THIS_MODULE; - if (clk->enable == NULL) clk->enable = clk_null_enable; From 9a1a69a1f41cbefebf3172761f197db6aba71e68 Mon Sep 17 00:00:00 2001 From: Andrew Vasquez Date: Wed, 29 Apr 2009 13:12:39 -0500 Subject: [PATCH 045/267] [SCSI] fc-transport: Close state transition-window during rport deletion. Andrew Vasquez wrote: > fc-transport: Close state transition-window during rport deletion. > > After an rport's state has transitioned to FC_PORTSTATE_BLOCKED, > but, prior to making the upcall to 'block' the scsi-target > associated with an rport, queued commands can recycle and > ultimately run out of retries causing failures to propagate to > upper-level drivers. Close this transition-window by returning > the non-'retries' modifying DID_IMM_RETRY status for submitted > I/Os. The same can happen for iscsi when transitioning from logged in to failed and blocking the sdevs. This patch converts iscsi and fc's transitions back to use DID_IMM_RETRY instead of DID_TRANSPORT_DISRUPTED which has a limited number of retries that we do not want to use for handling this race. Signed-off-by: Andrew Vasquez [Addition of iscsi and fc port online devloss case conversion by Mike Christie] Signed-off-by: Mike Christie Signed-off-by: James Bottomley --- drivers/scsi/scsi_transport_iscsi.c | 2 +- include/scsi/scsi_transport_fc.h | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index 094795455293..0a2ce7b6325c 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -357,7 +357,7 @@ int iscsi_session_chkready(struct iscsi_cls_session *session) err = 0; break; case ISCSI_SESSION_FAILED: - err = DID_TRANSPORT_DISRUPTED << 16; + err = DID_IMM_RETRY << 16; break; case ISCSI_SESSION_FREE: err = DID_TRANSPORT_FAILFAST << 16; diff --git a/include/scsi/scsi_transport_fc.h b/include/scsi/scsi_transport_fc.h index c9184f756cad..68a8d873bbd9 100644 --- a/include/scsi/scsi_transport_fc.h +++ b/include/scsi/scsi_transport_fc.h @@ -680,7 +680,7 @@ fc_remote_port_chkready(struct fc_rport *rport) if (rport->roles & FC_PORT_ROLE_FCP_TARGET) result = 0; else if (rport->flags & FC_RPORT_DEVLOSS_PENDING) - result = DID_TRANSPORT_DISRUPTED << 16; + result = DID_IMM_RETRY << 16; else result = DID_NO_CONNECT << 16; break; @@ -688,7 +688,7 @@ fc_remote_port_chkready(struct fc_rport *rport) if (rport->flags & FC_RPORT_FAST_FAIL_TIMEDOUT) result = DID_TRANSPORT_FAILFAST << 16; else - result = DID_TRANSPORT_DISRUPTED << 16; + result = DID_IMM_RETRY << 16; break; default: result = DID_NO_CONNECT << 16; From 5e43754fd949193252ecb470d7fb08b547a1e310 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Thu, 30 Apr 2009 19:13:41 -0700 Subject: [PATCH 046/267] [SCSI] ses: fix problems caused by empty SES provided name We use the name provided by SES to name objects. An empty name is legal in SES but causes problems in our generic device hierarchy. Fix this by falling back to a number if the name is either NULL or empty. Also fix a secondary bug spotted in that dev_set_name(dev, name) uses a string format and so would go wrong if name contained a '%'. Signed-off-by: Yinghai Lu Signed-off-by: James Bottomley --- drivers/misc/enclosure.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/misc/enclosure.c b/drivers/misc/enclosure.c index 3cf61ece71d7..348443bdb23b 100644 --- a/drivers/misc/enclosure.c +++ b/drivers/misc/enclosure.c @@ -119,7 +119,7 @@ enclosure_register(struct device *dev, const char *name, int components, edev->edev.class = &enclosure_class; edev->edev.parent = get_device(dev); edev->cb = cb; - dev_set_name(&edev->edev, name); + dev_set_name(&edev->edev, "%s", name); err = device_register(&edev->edev); if (err) goto err; @@ -255,8 +255,8 @@ enclosure_component_register(struct enclosure_device *edev, ecomp->number = number; cdev = &ecomp->cdev; cdev->parent = get_device(&edev->edev); - if (name) - dev_set_name(cdev, name); + if (name && name[0]) + dev_set_name(cdev, "%s", name); else dev_set_name(cdev, "%u", number); From 8454e9888cb0316dd296fd5d47c612248ed5e1d1 Mon Sep 17 00:00:00 2001 From: adam radford Date: Tue, 5 May 2009 11:45:37 -0700 Subject: [PATCH 047/267] [SCSI] 3w-9xxx: scsi_dma_unmap fix This patch fixes the following regression the occurred during the scsi_dma_map()/unmap() changes: 3w-9xxx 0001:45:00.0: DMA-API: device driver tries to free DMA memory it has not allocated [device address=0x0000000000000000] [size=36 bytes] Signed-off-by: Adam Radford Signed-off-by: James Bottomley --- drivers/scsi/3w-9xxx.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/3w-9xxx.c b/drivers/scsi/3w-9xxx.c index 8b7983aba8f7..36c21b19e5d7 100644 --- a/drivers/scsi/3w-9xxx.c +++ b/drivers/scsi/3w-9xxx.c @@ -1978,7 +1978,8 @@ static void twa_unmap_scsi_data(TW_Device_Extension *tw_dev, int request_id) { struct scsi_cmnd *cmd = tw_dev->srb[request_id]; - scsi_dma_unmap(cmd); + if (cmd->SCp.phase == TW_PHASE_SGLIST) + scsi_dma_unmap(cmd); } /* End twa_unmap_scsi_data() */ /* scsi_host_template initializer */ From 7b14f58ad65f9d74e4273fb45360cfea824495aa Mon Sep 17 00:00:00 2001 From: adam radford Date: Mon, 11 May 2009 14:55:55 -0700 Subject: [PATCH 048/267] [SCSI] 3w-xxxx: scsi_dma_unmap fix This patch fixes the following regression that occurred during the scsi_dma_map()/unmap() changes when compiling with CONFIG_DMA_API_DEBUG=y : WARNING: at lib/dma-debug.c:496 check_unmap+0x142/0x542() Hardware name: 3w-xxxx 0000:02:02.0: DMA-API: device driver tries to free DMA memory it has not allocated [device address=0x0000000000000000] [size=36 bytes] Signed-off-by: Adam Radford Signed-off-by: James Bottomley --- drivers/scsi/3w-xxxx.c | 5 +++-- drivers/scsi/3w-xxxx.h | 2 +- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/3w-xxxx.c b/drivers/scsi/3w-xxxx.c index c03f1d2c9e2e..faa0fcfed71e 100644 --- a/drivers/scsi/3w-xxxx.c +++ b/drivers/scsi/3w-xxxx.c @@ -6,7 +6,7 @@ Arnaldo Carvalho de Melo Brad Strand - Copyright (C) 1999-2007 3ware Inc. + Copyright (C) 1999-2009 3ware Inc. Kernel compatiblity By: Andre Hedrick Non-Copyright (C) 2000 Andre Hedrick @@ -1294,7 +1294,8 @@ static void tw_unmap_scsi_data(struct pci_dev *pdev, struct scsi_cmnd *cmd) { dprintk(KERN_WARNING "3w-xxxx: tw_unmap_scsi_data()\n"); - scsi_dma_unmap(cmd); + if (cmd->SCp.phase == TW_PHASE_SGLIST) + scsi_dma_unmap(cmd); } /* End tw_unmap_scsi_data() */ /* This function will reset a device extension */ diff --git a/drivers/scsi/3w-xxxx.h b/drivers/scsi/3w-xxxx.h index 8e71e5e122b3..a5a2ba2561d9 100644 --- a/drivers/scsi/3w-xxxx.h +++ b/drivers/scsi/3w-xxxx.h @@ -6,7 +6,7 @@ Arnaldo Carvalho de Melo Brad Strand - Copyright (C) 1999-2007 3ware Inc. + Copyright (C) 1999-2009 3ware Inc. Kernel compatiblity By: Andre Hedrick Non-Copyright (C) 2000 Andre Hedrick From 88fc86c283d9c3854e67e4155808027bc2519eb6 Mon Sep 17 00:00:00 2001 From: GeunSik Lim Date: Thu, 14 May 2009 17:23:38 +0900 Subject: [PATCH 049/267] tracing: Append prompt in /debug/tracing/README file append prompt in /debug/tracing/README file. This is trivial issue. Fix typo Mini Howto file(README) for ftrace. [ Impact: cleanup ] Signed-off-by: GeunSik Lim Acked-by: Steven Rostedt Cc: williams LKML-Reference: <1242289418.31161.45.camel@centos51> Signed-off-by: Ingo Molnar --- kernel/trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index a884c09006c4..cda81ec58d9f 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2380,7 +2380,7 @@ static const char readme_msg[] = "# echo print-parent > /debug/tracing/trace_options\n" "# echo 1 > /debug/tracing/tracing_enabled\n" "# cat /debug/tracing/trace > /tmp/trace.txt\n" - "echo 0 > /debug/tracing/tracing_enabled\n" + "# echo 0 > /debug/tracing/tracing_enabled\n" ; static ssize_t From b4ecc126991b30fe5f9a59dfacda046aeac124b2 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 13 May 2009 17:16:55 -0700 Subject: [PATCH 050/267] x86: Fix performance regression caused by paravirt_ops on native kernels Xiaohui Xin and some other folks at Intel have been looking into what's behind the performance hit of paravirt_ops when running native. It appears that the hit is entirely due to the paravirtualized spinlocks introduced by: | commit 8efcbab674de2bee45a2e4cdf97de16b8e609ac8 | Date: Mon Jul 7 12:07:51 2008 -0700 | | paravirt: introduce a "lock-byte" spinlock implementation The extra call/return in the spinlock path is somehow causing an increase in the cycles/instruction of somewhere around 2-7% (seems to vary quite a lot from test to test). The working theory is that the CPU's pipeline is getting upset about the call->call->locked-op->return->return, and seems to be failing to speculate (though I haven't seen anything definitive about the precise reasons). This doesn't entirely make sense, because the performance hit is also visible on unlock and other operations which don't involve locked instructions. But spinlock operations clearly swamp all the other pvops operations, even though I can't imagine that they're nearly as common (there's only a .05% increase in instructions executed). If I disable just the pv-spinlock calls, my tests show that pvops is identical to non-pvops performance on native (my measurements show that it is actually about .1% faster, but Xiaohui shows a .05% slowdown). Summary of results, averaging 10 runs of the "mmperf" test, using a no-pvops build as baseline: nopv Pv-nospin Pv-spin CPU cycles 100.00% 99.89% 102.18% instructions 100.00% 100.10% 100.15% CPI 100.00% 99.79% 102.03% cache ref 100.00% 100.84% 100.28% cache miss 100.00% 90.47% 88.56% cache miss rate 100.00% 89.72% 88.31% branches 100.00% 99.93% 100.04% branch miss 100.00% 103.66% 107.72% branch miss rt 100.00% 103.73% 107.67% wallclock 100.00% 99.90% 102.20% The clear effect here is that the 2% increase in CPI is directly reflected in the final wallclock time. (The other interesting effect is that the more ops are out of line calls via pvops, the lower the cache access and miss rates. Not too surprising, but it suggests that the non-pvops kernel is over-inlined. On the flipside, the branch misses go up correspondingly...) So, what's the fix? Paravirt patching turns all the pvops calls into direct calls, so _spin_lock etc do end up having direct calls. For example, the compiler generated code for paravirtualized _spin_lock is: <_spin_lock+0>: mov %gs:0xb4c8,%rax <_spin_lock+9>: incl 0xffffffffffffe044(%rax) <_spin_lock+15>: callq *0xffffffff805a5b30 <_spin_lock+22>: retq The indirect call will get patched to: <_spin_lock+0>: mov %gs:0xb4c8,%rax <_spin_lock+9>: incl 0xffffffffffffe044(%rax) <_spin_lock+15>: callq <__ticket_spin_lock> <_spin_lock+20>: nop; nop /* or whatever 2-byte nop */ <_spin_lock+22>: retq One possibility is to inline _spin_lock, etc, when building an optimised kernel (ie, when there's no spinlock/preempt instrumentation/debugging enabled). That will remove the outer call/return pair, returning the instruction stream to a single call/return, which will presumably execute the same as the non-pvops case. The downsides arel 1) it will replicate the preempt_disable/enable code at eack lock/unlock callsite; this code is fairly small, but not nothing; and 2) the spinlock definitions are already a very heavily tangled mass of #ifdefs and other preprocessor magic, and making any changes will be non-trivial. The other obvious answer is to disable pv-spinlocks. Making them a separate config option is fairly easy, and it would be trivial to enable them only when Xen is enabled (as the only non-default user). But it doesn't really address the common case of a distro build which is going to have Xen support enabled, and leaves the open question of whether the native performance cost of pv-spinlocks is worth the performance improvement on a loaded Xen system (10% saving of overall system CPU when guests block rather than spin). Still it is a reasonable short-term workaround. [ Impact: fix pvops performance regression when running native ] Analysed-by: "Xin Xiaohui" Analysed-by: "Li Xin" Analysed-by: "Nakajima Jun" Signed-off-by: Jeremy Fitzhardinge Acked-by: H. Peter Anvin Cc: Nick Piggin Cc: Xen-devel LKML-Reference: <4A0B62F7.5030802@goop.org> [ fixed the help text ] Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 13 +++++++++++++ arch/x86/include/asm/paravirt.h | 2 +- arch/x86/include/asm/spinlock.h | 4 ++-- arch/x86/kernel/Makefile | 3 ++- arch/x86/kernel/paravirt.c | 2 ++ arch/x86/xen/Makefile | 5 +++-- arch/x86/xen/xen-ops.h | 19 +++++++++++++++---- 7 files changed, 38 insertions(+), 10 deletions(-) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index df9e885eee14..a6efe0a2e9ae 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -498,6 +498,19 @@ config PARAVIRT over full virtualization. However, when run without a hypervisor the kernel is theoretically slower and slightly larger. +config PARAVIRT_SPINLOCKS + bool "Paravirtualization layer for spinlocks" + depends on PARAVIRT && SMP && EXPERIMENTAL + ---help--- + Paravirtualized spinlocks allow a pvops backend to replace the + spinlock implementation with something virtualization-friendly + (for example, block the virtual CPU rather than spinning). + + Unfortunately the downside is an up to 5% performance hit on + native kernels, with various workloads. + + If you are unsure how to answer this question, answer N. + config PARAVIRT_CLOCK bool default n diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index 378e3691c08c..a53da004e08e 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -1443,7 +1443,7 @@ u64 _paravirt_ident_64(u64); #define paravirt_nop ((void *)_paravirt_nop) -#ifdef CONFIG_SMP +#if defined(CONFIG_SMP) && defined(CONFIG_PARAVIRT_SPINLOCKS) static inline int __raw_spin_is_locked(struct raw_spinlock *lock) { diff --git a/arch/x86/include/asm/spinlock.h b/arch/x86/include/asm/spinlock.h index e5e6caffec87..b7e5db876399 100644 --- a/arch/x86/include/asm/spinlock.h +++ b/arch/x86/include/asm/spinlock.h @@ -172,7 +172,7 @@ static inline int __ticket_spin_is_contended(raw_spinlock_t *lock) return (((tmp >> TICKET_SHIFT) - tmp) & ((1 << TICKET_SHIFT) - 1)) > 1; } -#ifndef CONFIG_PARAVIRT +#ifndef CONFIG_PARAVIRT_SPINLOCKS static inline int __raw_spin_is_locked(raw_spinlock_t *lock) { @@ -206,7 +206,7 @@ static __always_inline void __raw_spin_lock_flags(raw_spinlock_t *lock, __raw_spin_lock(lock); } -#endif +#endif /* CONFIG_PARAVIRT_SPINLOCKS */ static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock) { diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 145cce75cda7..88d1bfc847d3 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -89,7 +89,8 @@ obj-$(CONFIG_DEBUG_NX_TEST) += test_nx.o obj-$(CONFIG_VMI) += vmi_32.o vmiclock_32.o obj-$(CONFIG_KVM_GUEST) += kvm.o obj-$(CONFIG_KVM_CLOCK) += kvmclock.o -obj-$(CONFIG_PARAVIRT) += paravirt.o paravirt_patch_$(BITS).o paravirt-spinlocks.o +obj-$(CONFIG_PARAVIRT) += paravirt.o paravirt_patch_$(BITS).o +obj-$(CONFIG_PARAVIRT_SPINLOCKS)+= paravirt-spinlocks.o obj-$(CONFIG_PARAVIRT_CLOCK) += pvclock.o obj-$(CONFIG_PCSPKR_PLATFORM) += pcspeaker.o diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 8e45f4464880..9faf43bea336 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -134,7 +134,9 @@ static void *get_call_destination(u8 type) .pv_irq_ops = pv_irq_ops, .pv_apic_ops = pv_apic_ops, .pv_mmu_ops = pv_mmu_ops, +#ifdef CONFIG_PARAVIRT_SPINLOCKS .pv_lock_ops = pv_lock_ops, +#endif }; return *((void **)&tmpl + type); } diff --git a/arch/x86/xen/Makefile b/arch/x86/xen/Makefile index 3b767d03fd6a..172438f86a02 100644 --- a/arch/x86/xen/Makefile +++ b/arch/x86/xen/Makefile @@ -9,5 +9,6 @@ obj-y := enlighten.o setup.o multicalls.o mmu.o irq.o \ time.o xen-asm.o xen-asm_$(BITS).o \ grant-table.o suspend.o -obj-$(CONFIG_SMP) += smp.o spinlock.o -obj-$(CONFIG_XEN_DEBUG_FS) += debugfs.o \ No newline at end of file +obj-$(CONFIG_SMP) += smp.o +obj-$(CONFIG_PARAVIRT_SPINLOCKS)+= spinlock.o +obj-$(CONFIG_XEN_DEBUG_FS) += debugfs.o diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h index 20139464943c..ca6596b05d53 100644 --- a/arch/x86/xen/xen-ops.h +++ b/arch/x86/xen/xen-ops.h @@ -62,15 +62,26 @@ void xen_setup_vcpu_info_placement(void); #ifdef CONFIG_SMP void xen_smp_init(void); -void __init xen_init_spinlocks(void); -__cpuinit void xen_init_lock_cpu(int cpu); -void xen_uninit_lock_cpu(int cpu); - extern cpumask_var_t xen_cpu_initialized_map; #else static inline void xen_smp_init(void) {} #endif +#ifdef CONFIG_PARAVIRT_SPINLOCKS +void __init xen_init_spinlocks(void); +__cpuinit void xen_init_lock_cpu(int cpu); +void xen_uninit_lock_cpu(int cpu); +#else +static inline void xen_init_spinlocks(void) +{ +} +static inline void xen_init_lock_cpu(int cpu) +{ +} +static inline void xen_uninit_lock_cpu(int cpu) +{ +} +#endif /* Declare an asm function, along with symbols needed to make it inlineable */ From 4484079d517c2b6521621be0b1ea246ccc55c7d7 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Fri, 15 May 2009 23:30:50 +0200 Subject: [PATCH 051/267] PM: check sysdev_suspend(PMSG_FREEZE) return value Check the return value of sysdev_suspend(). I think this was a typo. Without this change, the following "if" check is always false. I also changed the error message so it's distinguishable from the similar message a few lines above. Signed-off-by: Bjorn Helgaas Acked-by: Pavel Machek Signed-off-by: Rafael J. Wysocki --- kernel/power/disk.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/power/disk.c b/kernel/power/disk.c index e71ca9cd81b2..b0dc9e7a0d17 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -241,9 +241,9 @@ static int create_image(int platform_mode) local_irq_disable(); - sysdev_suspend(PMSG_FREEZE); + error = sysdev_suspend(PMSG_FREEZE); if (error) { - printk(KERN_ERR "PM: Some devices failed to power down, " + printk(KERN_ERR "PM: Some system devices failed to power down, " "aborting hibernation\n"); goto Enable_irqs; } From 975b3c474c13d29337eaf7da8f5f5c0299e4943f Mon Sep 17 00:00:00 2001 From: Lin Ming Date: Fri, 15 May 2009 22:27:49 -0400 Subject: [PATCH 052/267] ACPI: resume: re-enable SCI-enable workaround The BIOS bug workaround mistakenly got disabled when we followed the ACPI specification more closely by ignoring OS updates to that bit. (The BIOS is supposed to update SCI_EN, not the OS) http://bugzilla.kernel.org/show_bug.cgi?id=13289 Signed-off-by: Lin Ming Signed-off-by: Len Brown --- drivers/acpi/acpica/aclocal.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/acpi/acpica/aclocal.h b/drivers/acpi/acpica/aclocal.h index 772ee5c4ccca..2ec394a328e9 100644 --- a/drivers/acpi/acpica/aclocal.h +++ b/drivers/acpi/acpica/aclocal.h @@ -787,7 +787,12 @@ struct acpi_bit_register_info { /* For control registers, both ignored and reserved bits must be preserved */ -#define ACPI_PM1_CONTROL_IGNORED_BITS 0x0201 /* Bits 9, 0(SCI_EN) */ +/* + * The ACPI spec says to ignore PM1_CTL.SCI_EN (bit 0) + * but we need to be able to write ACPI_BITREG_SCI_ENABLE directly + * as a BIOS workaround on some machines. + */ +#define ACPI_PM1_CONTROL_IGNORED_BITS 0x0200 /* Bits 9 */ #define ACPI_PM1_CONTROL_RESERVED_BITS 0xC1F8 /* Bits 14-15, 3-8 */ #define ACPI_PM1_CONTROL_PRESERVED_BITS \ (ACPI_PM1_CONTROL_IGNORED_BITS | ACPI_PM1_CONTROL_RESERVED_BITS) From 815ab0fd40579ad2aa42058298073503648762b9 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Thu, 7 May 2009 22:19:45 -0400 Subject: [PATCH 053/267] ACPI: suspend: restore BM_RLD on resume In 2.6.29, 31878dd86b7df9a147f5e6cc6e07092b4308782b "ACPI: remove BM_RLD access from idle entry path" moved BM_RLD initialization to init-time from run time. But we discovered that some BIOS do not restore BM_RLD after suspend, causing device errors on C3 and C4 after resume. So now the kernel restores BM_RLD. http://bugzilla.kernel.org/show_bug.cgi?id=13032 Signed-off-by: Len Brown --- drivers/acpi/processor_idle.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index f7ca8c55956b..c1d59cfdb5fb 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -202,15 +202,38 @@ static void acpi_state_timer_broadcast(struct acpi_processor *pr, * Suspend / resume control */ static int acpi_idle_suspend; +static u32 saved_bm_rld; + +static void acpi_idle_bm_rld_save(void) +{ + acpi_read_bit_register(ACPI_BITREG_BUS_MASTER_RLD, &saved_bm_rld); +} +static void acpi_idle_bm_rld_restore(void) +{ + u32 resumed_bm_rld; + + acpi_read_bit_register(ACPI_BITREG_BUS_MASTER_RLD, &resumed_bm_rld); + + if (resumed_bm_rld != saved_bm_rld) + acpi_write_bit_register(ACPI_BITREG_BUS_MASTER_RLD, saved_bm_rld); +} int acpi_processor_suspend(struct acpi_device * device, pm_message_t state) { + if (acpi_idle_suspend == 1) + return 0; + + acpi_idle_bm_rld_save(); acpi_idle_suspend = 1; return 0; } int acpi_processor_resume(struct acpi_device * device) { + if (acpi_idle_suspend == 0) + return 0; + + acpi_idle_bm_rld_restore(); acpi_idle_suspend = 0; return 0; } From ecb4aed78dcf09e48c8c34c8c2fa7f5c69344be6 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Fri, 24 Apr 2009 11:33:47 -0400 Subject: [PATCH 054/267] ACPI, i915: build fix drivers/built-in.o: In function `intel_opregion_init': (.text+0x9d540): undefined reference to `acpi_video_register' http://bugzilla.kernel.org/show_bug.cgi?id=13165 Signed-off-by: Len Brown --- drivers/gpu/drm/Kconfig | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig index 3a22eb9be378..f33d25279912 100644 --- a/drivers/gpu/drm/Kconfig +++ b/drivers/gpu/drm/Kconfig @@ -83,6 +83,12 @@ config DRM_I915 config DRM_I915_KMS bool "Enable modesetting on intel by default" depends on DRM_I915 + # i915 KMS depends on ACPI_VIDEO when ACPI is enabled + # but for select to work, need to select ACPI_VIDEO's dependencies, ick + select VIDEO_OUTPUT_CONTROL if ACPI + select BACKLIGHT_CLASS_DEVICE if ACPI + select INPUT if ACPI + select ACPI_VIDEO if ACPI help Choose this option if you want kernel modesetting enabled by default, and you have a new enough userspace to support this. Running old From 56c213fa012f2bad9eff908292ff2500f840b020 Mon Sep 17 00:00:00 2001 From: Zhang Rui Date: Mon, 11 May 2009 09:35:57 +0800 Subject: [PATCH 055/267] ACPI processor: introduce module parameter processor.ignore_tpc Introduce module parameter processor.ignore_tpc. Some laptops are shipped with buggy _TPC, this module parameter is used to to disable the buggy support. http://bugzilla.kernel.org/show_bug.cgi?id=13259 Signed-off-by: Zhang Rui Tested-by: James Ettle Signed-off-by: Len Brown --- drivers/acpi/processor_throttling.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/drivers/acpi/processor_throttling.c b/drivers/acpi/processor_throttling.c index d0d1f4d50434..5f09fb8c0e05 100644 --- a/drivers/acpi/processor_throttling.c +++ b/drivers/acpi/processor_throttling.c @@ -45,6 +45,14 @@ #define _COMPONENT ACPI_PROCESSOR_COMPONENT ACPI_MODULE_NAME("processor_throttling"); +/* ignore_tpc: + * 0 -> acpi processor driver doesn't ignore _TPC values + * 1 -> acpi processor driver ignores _TPC values + */ +static int ignore_tpc; +module_param(ignore_tpc, int, 0644); +MODULE_PARM_DESC(ignore_tpc, "Disable broken BIOS _TPC throttling support"); + struct throttling_tstate { unsigned int cpu; /* cpu nr */ int target_state; /* target T-state */ @@ -283,6 +291,10 @@ static int acpi_processor_get_platform_limit(struct acpi_processor *pr) if (!pr) return -EINVAL; + + if (ignore_tpc) + goto end; + status = acpi_evaluate_integer(pr->handle, "_TPC", NULL, &tpc); if (ACPI_FAILURE(status)) { if (status != AE_NOT_FOUND) { @@ -290,6 +302,8 @@ static int acpi_processor_get_platform_limit(struct acpi_processor *pr) } return -ENODEV; } + +end: pr->throttling_platform_limit = (int)tpc; return 0; } @@ -302,6 +316,9 @@ int acpi_processor_tstate_has_changed(struct acpi_processor *pr) struct acpi_processor_limit *limit; int target_state; + if (ignore_tpc) + return 0; + result = acpi_processor_get_platform_limit(pr); if (result) { /* Throttling Limit is unsupported */ From 4973b22aa8c70fe036e3e0039f104cf5bb7fe2b1 Mon Sep 17 00:00:00 2001 From: Zhang Rui Date: Mon, 11 May 2009 09:36:01 +0800 Subject: [PATCH 056/267] ACPI processor: reset the throttling state once it's invalid If the BIOS hands us an invalid throttling state, write a valid state. http://bugzilla.kernel.org/show_bug.cgi?id=13259 Signed-off-by: Zhang Rui Tested-by: James Ettle Signed-off-by: Len Brown --- drivers/acpi/processor_throttling.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/acpi/processor_throttling.c b/drivers/acpi/processor_throttling.c index 5f09fb8c0e05..7f16f5f8e7d3 100644 --- a/drivers/acpi/processor_throttling.c +++ b/drivers/acpi/processor_throttling.c @@ -838,6 +838,14 @@ static int acpi_processor_get_throttling_ptc(struct acpi_processor *pr) ret = acpi_read_throttling_status(pr, &value); if (ret >= 0) { state = acpi_get_throttling_state(pr, value); + if (state == -1) { + ACPI_WARNING((AE_INFO, + "Invalid throttling state, reset\n")); + state = 0; + ret = acpi_processor_set_throttling(pr, state); + if (ret) + return ret; + } pr->throttling.state = state; } From 520daf7217bc1806c02eb4cfa7805447a3da2f66 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Thu, 14 May 2009 17:27:38 -0400 Subject: [PATCH 057/267] ACPI: idle: fix init-time TSC check regression A previous 2.6.30 patch, a71e4917dc0ebbcb5a0ecb7ca3486643c1c9a6e2, (ACPI: idle: mark_tsc_unstable() at init-time, not run-time) erroneously disabled the TSC on systems that did not actually have valid deep C-states. Move the check after the deep-C-states are validated, via new helper, tsc_check_state(), hich replaces tsc_halts_in_c(). Signed-off-by: Len Brown Acked-by: Venkatesh Pallipadi Acked-by: Thomas Gleixner Tested-by: Frans Pop --- drivers/acpi/processor_idle.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index f7ca8c55956b..e39a40a2ceae 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -216,7 +216,7 @@ int acpi_processor_resume(struct acpi_device * device) } #if defined (CONFIG_GENERIC_TIME) && defined (CONFIG_X86) -static int tsc_halts_in_c(int state) +static void tsc_check_state(int state) { switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: @@ -226,13 +226,17 @@ static int tsc_halts_in_c(int state) * C/P/S0/S1 states when this bit is set. */ if (boot_cpu_has(X86_FEATURE_NONSTOP_TSC)) - return 0; + return; /*FALL THROUGH*/ default: - return state > ACPI_STATE_C1; + /* TSC could halt in idle, so notify users */ + if (state > ACPI_STATE_C1) + mark_tsc_unstable("TSC halts in idle"); } } +#else +static void tsc_check_state(int state) { return; } #endif static int acpi_processor_get_power_info_fadt(struct acpi_processor *pr) @@ -581,11 +585,6 @@ static int acpi_processor_power_verify(struct acpi_processor *pr) for (i = 1; i < ACPI_PROCESSOR_MAX_POWER; i++) { struct acpi_processor_cx *cx = &pr->power.states[i]; -#if defined (CONFIG_GENERIC_TIME) && defined (CONFIG_X86) - /* TSC could halt in idle, so notify users */ - if (tsc_halts_in_c(cx->type)) - mark_tsc_unstable("TSC halts in idle");; -#endif switch (cx->type) { case ACPI_STATE_C1: cx->valid = 1; @@ -603,6 +602,8 @@ static int acpi_processor_power_verify(struct acpi_processor *pr) acpi_timer_check_state(i, pr, cx); break; } + if (cx->valid) + tsc_check_state(cx->type); if (cx->valid) working++; From a0bf284bfedd6dc95bbee7ebf5ccf3b5f753a008 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Fri, 15 May 2009 01:29:31 -0400 Subject: [PATCH 058/267] ACPI: Idle C-states disabled by max_cstate should not disable the TSC Processor idle power states C2 and C3 stop the TSC on many machines. Linux recognizes this situation and marks the TSC as unstable: Marking TSC unstable due to TSC halts in idle But if those same machines are booted with "processor.max_cstate=1", then there is no need to validate C2 and C3, and no need to disable the TSC, which can be reliably used as a clocksource. Signed-off-by: Len Brown Acked-by: Thomas Gleixner --- drivers/acpi/processor_idle.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index e39a40a2ceae..e65476fdf40d 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -582,7 +582,7 @@ static int acpi_processor_power_verify(struct acpi_processor *pr) pr->power.timer_broadcast_on_state = INT_MAX; - for (i = 1; i < ACPI_PROCESSOR_MAX_POWER; i++) { + for (i = 1; i < ACPI_PROCESSOR_MAX_POWER && i <= max_cstate; i++) { struct acpi_processor_cx *cx = &pr->power.states[i]; switch (cx->type) { From e102657ed16bbed49820d9c58509220fc8d9289a Mon Sep 17 00:00:00 2001 From: Vikram Pandita Date: Sat, 16 May 2009 08:28:16 -0700 Subject: [PATCH 059/267] ARM: OMAP3: Fix number of GPIO lines for 34xx As per 3430 TRM, there are 6 banks [0 to 191] Signed-off-by: Tom Rix Signed-off-by: Vikram Pandita Signed-off-by: Tony Lindgren --- arch/arm/plat-omap/gpio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/plat-omap/gpio.c b/arch/arm/plat-omap/gpio.c index 17d7afe42b83..ee0b21f5b094 100644 --- a/arch/arm/plat-omap/gpio.c +++ b/arch/arm/plat-omap/gpio.c @@ -307,7 +307,7 @@ static inline int gpio_valid(int gpio) return 0; if (cpu_is_omap24xx() && gpio < 128) return 0; - if (cpu_is_omap34xx() && gpio < 160) + if (cpu_is_omap34xx() && gpio < 192) return 0; return -1; } From 8dbe43930a4e9bede88eb67a9c613773a2747caf Mon Sep 17 00:00:00 2001 From: Kalle Jokiniemi Date: Sat, 16 May 2009 08:28:17 -0700 Subject: [PATCH 060/267] ARM: OMAP3: Fix HW SAVEANDRESTORE shift define The OMAP3430ES2_SAVEANDRESTORE_SHIFT macro is used by powerdomain code in "1 << OMAP3430ES2_SAVEANDRESTORE_SHIFT" manner, but the definition was also (1 << 4), meaning we actually modified bit 16. So the definition needs to be 4. This fixes also a cold reset HW bug in OMAP3430 ES3.x where some of the efuse bits are not isolated during wake-up from off mode. This can cause randomish cold resets with off mode. Enabling the USBTLL hardware SAVEANDRESTORE causes the core power up assert to be delayed in a way that we will not get faulty values when boot ROM is reading the unisolated registers. Signed-off-by: Kalle Jokiniemi Acked-by: Kevin Hilman Acked-by: Paul Walmsley Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/prm-regbits-34xx.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-omap2/prm-regbits-34xx.h b/arch/arm/mach-omap2/prm-regbits-34xx.h index c6a7940f4287..9fd03a2ec95c 100644 --- a/arch/arm/mach-omap2/prm-regbits-34xx.h +++ b/arch/arm/mach-omap2/prm-regbits-34xx.h @@ -409,7 +409,7 @@ /* PM_PREPWSTST_CAM specific bits */ /* PM_PWSTCTRL_USBHOST specific bits */ -#define OMAP3430ES2_SAVEANDRESTORE_SHIFT (1 << 4) +#define OMAP3430ES2_SAVEANDRESTORE_SHIFT 4 /* RM_RSTST_PER specific bits */ From 005187eecaa400b4b43d9f640fbde9fcc50f37c1 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Sat, 16 May 2009 08:28:17 -0700 Subject: [PATCH 061/267] ARM: OMAP2/3: Change omapfb to use clkdev for dispc and rfbi, v2 This makes the framebuffer work on omap3. Also fix the clk_get usage for checkpatch.pl "ERROR: do not use assignment in if condition". Cc: Imre Deak Cc: linux-fbdev-devel@lists.sourceforge.net Acked-by: Krzysztof Helt Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/clock24xx.c | 8 ++++---- arch/arm/mach-omap2/clock34xx.c | 10 +++++----- drivers/video/omap/dispc.c | 14 ++++++++------ drivers/video/omap/rfbi.c | 8 +++++--- 4 files changed, 22 insertions(+), 18 deletions(-) diff --git a/arch/arm/mach-omap2/clock24xx.c b/arch/arm/mach-omap2/clock24xx.c index ff9b4c0b00ae..e4cef333e291 100644 --- a/arch/arm/mach-omap2/clock24xx.c +++ b/arch/arm/mach-omap2/clock24xx.c @@ -103,10 +103,10 @@ static struct omap_clk omap24xx_clks[] = { CLK(NULL, "mdm_ick", &mdm_ick, CK_243X), CLK(NULL, "mdm_osc_ck", &mdm_osc_ck, CK_243X), /* DSS domain clocks */ - CLK(NULL, "dss_ick", &dss_ick, CK_243X | CK_242X), - CLK(NULL, "dss1_fck", &dss1_fck, CK_243X | CK_242X), - CLK(NULL, "dss2_fck", &dss2_fck, CK_243X | CK_242X), - CLK(NULL, "dss_54m_fck", &dss_54m_fck, CK_243X | CK_242X), + CLK("omapfb", "ick", &dss_ick, CK_243X | CK_242X), + CLK("omapfb", "dss1_fck", &dss1_fck, CK_243X | CK_242X), + CLK("omapfb", "dss2_fck", &dss2_fck, CK_243X | CK_242X), + CLK("omapfb", "tv_fck", &dss_54m_fck, CK_243X | CK_242X), /* L3 domain clocks */ CLK(NULL, "core_l3_ck", &core_l3_ck, CK_243X | CK_242X), CLK(NULL, "ssi_fck", &ssi_ssr_sst_fck, CK_243X | CK_242X), diff --git a/arch/arm/mach-omap2/clock34xx.c b/arch/arm/mach-omap2/clock34xx.c index a0575393d0ea..ba05aa42bd8e 100644 --- a/arch/arm/mach-omap2/clock34xx.c +++ b/arch/arm/mach-omap2/clock34xx.c @@ -197,11 +197,11 @@ static struct omap_clk omap34xx_clks[] = { CLK("omap_rng", "ick", &rng_ick, CK_343X), CLK(NULL, "sha11_ick", &sha11_ick, CK_343X), CLK(NULL, "des1_ick", &des1_ick, CK_343X), - CLK(NULL, "dss1_alwon_fck", &dss1_alwon_fck, CK_343X), - CLK(NULL, "dss_tv_fck", &dss_tv_fck, CK_343X), - CLK(NULL, "dss_96m_fck", &dss_96m_fck, CK_343X), - CLK(NULL, "dss2_alwon_fck", &dss2_alwon_fck, CK_343X), - CLK(NULL, "dss_ick", &dss_ick, CK_343X), + CLK("omapfb", "dss1_fck", &dss1_alwon_fck, CK_343X), + CLK("omapfb", "tv_fck", &dss_tv_fck, CK_343X), + CLK("omapfb", "video_fck", &dss_96m_fck, CK_343X), + CLK("omapfb", "dss2_fck", &dss2_alwon_fck, CK_343X), + CLK("omapfb", "ick", &dss_ick, CK_343X), CLK(NULL, "cam_mclk", &cam_mclk, CK_343X), CLK(NULL, "cam_ick", &cam_ick, CK_343X), CLK(NULL, "csi2_96m_fck", &csi2_96m_fck, CK_343X), diff --git a/drivers/video/omap/dispc.c b/drivers/video/omap/dispc.c index dfb72f5e4c96..148cbcc39602 100644 --- a/drivers/video/omap/dispc.c +++ b/drivers/video/omap/dispc.c @@ -880,20 +880,22 @@ static irqreturn_t omap_dispc_irq_handler(int irq, void *dev) static int get_dss_clocks(void) { - if (IS_ERR((dispc.dss_ick = clk_get(dispc.fbdev->dev, "dss_ick")))) { - dev_err(dispc.fbdev->dev, "can't get dss_ick\n"); + dispc.dss_ick = clk_get(dispc.fbdev->dev, "ick"); + if (IS_ERR(dispc.dss_ick)) { + dev_err(dispc.fbdev->dev, "can't get ick\n"); return PTR_ERR(dispc.dss_ick); } - if (IS_ERR((dispc.dss1_fck = clk_get(dispc.fbdev->dev, "dss1_fck")))) { + dispc.dss1_fck = clk_get(dispc.fbdev->dev, "dss1_fck"); + if (IS_ERR(dispc.dss1_fck)) { dev_err(dispc.fbdev->dev, "can't get dss1_fck\n"); clk_put(dispc.dss_ick); return PTR_ERR(dispc.dss1_fck); } - if (IS_ERR((dispc.dss_54m_fck = - clk_get(dispc.fbdev->dev, "dss_54m_fck")))) { - dev_err(dispc.fbdev->dev, "can't get dss_54m_fck\n"); + dispc.dss_54m_fck = clk_get(dispc.fbdev->dev, "tv_fck"); + if (IS_ERR(dispc.dss_54m_fck)) { + dev_err(dispc.fbdev->dev, "can't get tv_fck\n"); clk_put(dispc.dss_ick); clk_put(dispc.dss1_fck); return PTR_ERR(dispc.dss_54m_fck); diff --git a/drivers/video/omap/rfbi.c b/drivers/video/omap/rfbi.c index a13c8dcad2a8..9332d6ca6456 100644 --- a/drivers/video/omap/rfbi.c +++ b/drivers/video/omap/rfbi.c @@ -83,12 +83,14 @@ static inline u32 rfbi_read_reg(int idx) static int rfbi_get_clocks(void) { - if (IS_ERR((rfbi.dss_ick = clk_get(rfbi.fbdev->dev, "dss_ick")))) { - dev_err(rfbi.fbdev->dev, "can't get dss_ick\n"); + rfbi.dss_ick = clk_get(rfbi.fbdev->dev, "ick"); + if (IS_ERR(rfbi.dss_ick)) { + dev_err(rfbi.fbdev->dev, "can't get ick\n"); return PTR_ERR(rfbi.dss_ick); } - if (IS_ERR((rfbi.dss1_fck = clk_get(rfbi.fbdev->dev, "dss1_fck")))) { + rfbi.dss1_fck = clk_get(rfbi.fbdev->dev, "dss1_fck"); + if (IS_ERR(rfbi.dss1_fck)) { dev_err(rfbi.fbdev->dev, "can't get dss1_fck\n"); clk_put(rfbi.dss_ick); return PTR_ERR(rfbi.dss1_fck); From 36999a5af1b6c5379e59f0ddf434db08d03c19e4 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sun, 19 Apr 2009 07:00:41 +0900 Subject: [PATCH 062/267] ide-tape: remove back-to-back REQUEST_SENSE detection Impact: fix an oops which always triggers ide_tape_issue_pc() assumed drive->pc isn't NULL on invocation when checking for back-to-back request sense issues but drive->pc can be NULL and even when it's not NULL, it's not safe to dereference it once the previous command is complete because pc could have been freed or was on stack. Kill back-to-back REQUEST_SENSE detection. Signed-off-by: Tejun Heo Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/ide-tape.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c index cb942a9b580f..3a53e0834cf7 100644 --- a/drivers/ide/ide-tape.c +++ b/drivers/ide/ide-tape.c @@ -614,12 +614,6 @@ static ide_startstop_t ide_tape_issue_pc(ide_drive_t *drive, { idetape_tape_t *tape = drive->driver_data; - if (drive->pc->c[0] == REQUEST_SENSE && - pc->c[0] == REQUEST_SENSE) { - printk(KERN_ERR "ide-tape: possible ide-tape.c bug - " - "Two request sense in serial were issued\n"); - } - if (drive->failed_pc == NULL && pc->c[0] != REQUEST_SENSE) drive->failed_pc = pc; From b019447ce73513432af052f30fa239d6dc93e111 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Sat, 16 May 2009 19:03:36 +0200 Subject: [PATCH 063/267] icside: register second channel of version 6 PCB The second IDE channel of version 6 PCB is not being registered anymore since the commit 48c3c1072651922ed153bcf0a33ea82cf20df390 (ide: add struct ide_host (take 3)). Signed-off-by: Sergei Shtylyov Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/icside.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ide/icside.c b/drivers/ide/icside.c index 4e16ce68b063..36da913cc553 100644 --- a/drivers/ide/icside.c +++ b/drivers/ide/icside.c @@ -466,7 +466,7 @@ icside_register_v6(struct icside_state *state, struct expansion_card *ec) struct ide_host *host; unsigned int sel = 0; int ret; - hw_regs_t hw[2], *hws[] = { &hw[0], NULL, NULL, NULL }; + hw_regs_t hw[2], *hws[] = { &hw[0], &hw[1], NULL, NULL }; struct ide_port_info d = icside_v6_port_info; ioc_base = ecardm_iomap(ec, ECARD_RES_IOCFAST, 0, 0); From d8788298d491ee5026981eb751b0341a996b22e4 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Sat, 16 May 2009 19:03:36 +0200 Subject: [PATCH 064/267] piix: The Sony TZ90 needs the cable type hardcoding The Sony TZ90 needs the cable type hardcoding. See bug #12734 Signed-off-by: Alan Cox Reported-by: Jonathan E. Snow [bart: port it from ata_piix to piix and give reporter the proper credit] Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/piix.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/ide/piix.c b/drivers/ide/piix.c index 2aa699933064..69860dea3820 100644 --- a/drivers/ide/piix.c +++ b/drivers/ide/piix.c @@ -263,6 +263,7 @@ static const struct ich_laptop ich_laptop[] = { { 0x24CA, 0x1025, 0x003d }, /* ICH4 on ACER TM290 */ { 0x266F, 0x1025, 0x0066 }, /* ICH6 on ACER Aspire 1694WLMi */ { 0x2653, 0x1043, 0x82D8 }, /* ICH6M on Asus Eee 701 */ + { 0x27df, 0x104d, 0x900e }, /* ICH7 on Sony TZ-90 */ /* end marker */ { 0, } }; From ff05c0330b9880f9ccbb7fa40f2ed3b5842f5693 Mon Sep 17 00:00:00 2001 From: Hartley Sweeten Date: Thu, 7 May 2009 18:41:47 +0100 Subject: [PATCH 065/267] [ARM] 5509/1: ep93xx: clkdev enable UARTS Fix the clkdev API support for the ep93xx uart clocks. The uarts available in the ep93xx have individual clock controls. The current implementation assumes that the bootloader has enabled the clocks before the kernel has booted. It also assumes that the bootloader has set the UARTBAUD bit indicating that the uarts are running off the 14.7456MHz external crystal. This fixes both issues. It also allows the uart clocks to be stopped when there are no users. Tested-by: Matthias Kaehlcke Cc: Ryan Mallon Signed-off-by: H Hartley Sweeten Signed-off-by: Russell King --- arch/arm/mach-ep93xx/clock.c | 69 ++++++++++++++++--- .../mach-ep93xx/include/mach/ep93xx-regs.h | 5 +- 2 files changed, 65 insertions(+), 9 deletions(-) diff --git a/arch/arm/mach-ep93xx/clock.c b/arch/arm/mach-ep93xx/clock.c index e8ebeaea6c48..b2eede5531c8 100644 --- a/arch/arm/mach-ep93xx/clock.c +++ b/arch/arm/mach-ep93xx/clock.c @@ -21,15 +21,50 @@ #include #include + +/* + * The EP93xx has two external crystal oscillators. To generate the + * required high-frequency clocks, the processor uses two phase-locked- + * loops (PLLs) to multiply the incoming external clock signal to much + * higher frequencies that are then divided down by programmable dividers + * to produce the needed clocks. The PLLs operate independently of one + * another. + */ +#define EP93XX_EXT_CLK_RATE 14745600 +#define EP93XX_EXT_RTC_RATE 32768 + + struct clk { unsigned long rate; int users; + int sw_locked; u32 enable_reg; u32 enable_mask; + + unsigned long (*get_rate)(struct clk *clk); }; -static struct clk clk_uart = { - .rate = 14745600, + +static unsigned long get_uart_rate(struct clk *clk); + + +static struct clk clk_uart1 = { + .sw_locked = 1, + .enable_reg = EP93XX_SYSCON_DEVICE_CONFIG, + .enable_mask = EP93XX_SYSCON_DEVICE_CONFIG_U1EN, + .get_rate = get_uart_rate, +}; +static struct clk clk_uart2 = { + .sw_locked = 1, + .enable_reg = EP93XX_SYSCON_DEVICE_CONFIG, + .enable_mask = EP93XX_SYSCON_DEVICE_CONFIG_U2EN, + .get_rate = get_uart_rate, +}; +static struct clk clk_uart3 = { + .sw_locked = 1, + .enable_reg = EP93XX_SYSCON_DEVICE_CONFIG, + .enable_mask = EP93XX_SYSCON_DEVICE_CONFIG_U3EN, + .get_rate = get_uart_rate, }; static struct clk clk_pll1; static struct clk clk_f; @@ -95,9 +130,9 @@ static struct clk clk_m2m1 = { { .dev_id = dev, .con_id = con, .clk = ck } static struct clk_lookup clocks[] = { - INIT_CK("apb:uart1", NULL, &clk_uart), - INIT_CK("apb:uart2", NULL, &clk_uart), - INIT_CK("apb:uart3", NULL, &clk_uart), + INIT_CK("apb:uart1", NULL, &clk_uart1), + INIT_CK("apb:uart2", NULL, &clk_uart2), + INIT_CK("apb:uart3", NULL, &clk_uart3), INIT_CK(NULL, "pll1", &clk_pll1), INIT_CK(NULL, "fclk", &clk_f), INIT_CK(NULL, "hclk", &clk_h), @@ -125,6 +160,8 @@ int clk_enable(struct clk *clk) u32 value; value = __raw_readl(clk->enable_reg); + if (clk->sw_locked) + __raw_writel(0xaa, EP93XX_SYSCON_SWLOCK); __raw_writel(value | clk->enable_mask, clk->enable_reg); } @@ -138,13 +175,29 @@ void clk_disable(struct clk *clk) u32 value; value = __raw_readl(clk->enable_reg); + if (clk->sw_locked) + __raw_writel(0xaa, EP93XX_SYSCON_SWLOCK); __raw_writel(value & ~clk->enable_mask, clk->enable_reg); } } EXPORT_SYMBOL(clk_disable); +static unsigned long get_uart_rate(struct clk *clk) +{ + u32 value; + + value = __raw_readl(EP93XX_SYSCON_CLOCK_CONTROL); + if (value & EP93XX_SYSCON_CLOCK_UARTBAUD) + return EP93XX_EXT_CLK_RATE; + else + return EP93XX_EXT_CLK_RATE / 2; +} + unsigned long clk_get_rate(struct clk *clk) { + if (clk->get_rate) + return clk->get_rate(clk); + return clk->rate; } EXPORT_SYMBOL(clk_get_rate); @@ -162,7 +215,7 @@ static unsigned long calc_pll_rate(u32 config_word) unsigned long long rate; int i; - rate = 14745600; + rate = EP93XX_EXT_CLK_RATE; rate *= ((config_word >> 11) & 0x1f) + 1; /* X1FBD */ rate *= ((config_word >> 5) & 0x3f) + 1; /* X2FBD */ do_div(rate, (config_word & 0x1f) + 1); /* X2IPD */ @@ -195,7 +248,7 @@ static int __init ep93xx_clock_init(void) value = __raw_readl(EP93XX_SYSCON_CLOCK_SET1); if (!(value & 0x00800000)) { /* PLL1 bypassed? */ - clk_pll1.rate = 14745600; + clk_pll1.rate = EP93XX_EXT_CLK_RATE; } else { clk_pll1.rate = calc_pll_rate(value); } @@ -206,7 +259,7 @@ static int __init ep93xx_clock_init(void) value = __raw_readl(EP93XX_SYSCON_CLOCK_SET2); if (!(value & 0x00080000)) { /* PLL2 bypassed? */ - clk_pll2.rate = 14745600; + clk_pll2.rate = EP93XX_EXT_CLK_RATE; } else if (value & 0x00040000) { /* PLL2 enabled? */ clk_pll2.rate = calc_pll_rate(value); } else { diff --git a/arch/arm/mach-ep93xx/include/mach/ep93xx-regs.h b/arch/arm/mach-ep93xx/include/mach/ep93xx-regs.h index f66be12b856e..1732de7629a5 100644 --- a/arch/arm/mach-ep93xx/include/mach/ep93xx-regs.h +++ b/arch/arm/mach-ep93xx/include/mach/ep93xx-regs.h @@ -159,7 +159,10 @@ #define EP93XX_SYSCON_CLOCK_SET1 EP93XX_SYSCON_REG(0x20) #define EP93XX_SYSCON_CLOCK_SET2 EP93XX_SYSCON_REG(0x24) #define EP93XX_SYSCON_DEVICE_CONFIG EP93XX_SYSCON_REG(0x80) -#define EP93XX_SYSCON_DEVICE_CONFIG_CRUNCH_ENABLE 0x00800000 +#define EP93XX_SYSCON_DEVICE_CONFIG_U3EN (1<<24) +#define EP93XX_SYSCON_DEVICE_CONFIG_CRUNCH_ENABLE (1<<23) +#define EP93XX_SYSCON_DEVICE_CONFIG_U2EN (1<<20) +#define EP93XX_SYSCON_DEVICE_CONFIG_U1EN (1<<18) #define EP93XX_SYSCON_SWLOCK EP93XX_SYSCON_REG(0xc0) #define EP93XX_WATCHDOG_BASE (EP93XX_APB_VIRT_BASE + 0x00140000) From 776abac81764847338a6a02c34609d4b8dfb4918 Mon Sep 17 00:00:00 2001 From: Ricardo Martins Date: Mon, 11 May 2009 00:15:08 +0100 Subject: [PATCH 066/267] [ARM] 5513/1: Eurotech VIPER SBC: fix compilation error Compilation for this board yields the following errors: arch/arm/mach-pxa/viper.c:511: error: 'FFUART' undeclared here (not in a function) arch/arm/mach-pxa/viper.c:520: error: 'BTUART' undeclared here (not in a function) arch/arm/mach-pxa/viper.c:529: error: 'STUART' undeclared here (not in a function) Fix them by including the necessary header. Signed-off-by: Ricardo Martins Signed-off-by: Russell King --- arch/arm/mach-pxa/viper.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/mach-pxa/viper.c b/arch/arm/mach-pxa/viper.c index 0e65344e9f53..dd031cc41847 100644 --- a/arch/arm/mach-pxa/viper.c +++ b/arch/arm/mach-pxa/viper.c @@ -46,6 +46,7 @@ #include #include #include +#include #include #include From 0f6f49a8cd0163fdb1723ed29f01fc65177108dc Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 16 May 2009 13:41:28 -0700 Subject: [PATCH 067/267] Fix caller information for warn_slowpath_null Ian Campbell noticed that since "Eliminate thousands of warnings with gcc 3.2 build" (commit 57adc4d2dbf968fdbe516359688094eef4d46581) all WARN_ON()'s currently appear to come from warn_slowpath_null(), eg: WARNING: at kernel/softirq.c:143 warn_slowpath_null+0x1c/0x20() because now that warn_slowpath_null() is in the call path, the __builtin_return_address(0) returns that, rather than the place that caused the warning. Fix this by splitting up the warn_slowpath_null/fmt cases differently, using a common helper function, and getting the return address in the right place. This also happens to avoid the unnecessary stack usage for the non-stdargs case, and just generally cleans things up. Make the function name printout use %pS while at it. Cc: Ian Campbell Cc: Jesper Nilsson Cc: Johannes Weiner Cc: Arjan van de Ven Cc: Andi Kleen Cc: Hugh Dickins Cc: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/panic.c | 35 ++++++++++++++++++++--------------- 1 file changed, 20 insertions(+), 15 deletions(-) diff --git a/kernel/panic.c b/kernel/panic.c index 874ecf1307ae..984b3ecbd72c 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -340,39 +340,44 @@ void oops_exit(void) } #ifdef WANT_WARN_ON_SLOWPATH -void warn_slowpath_fmt(const char *file, int line, const char *fmt, ...) -{ +struct slowpath_args { + const char *fmt; va_list args; - char function[KSYM_SYMBOL_LEN]; - unsigned long caller = (unsigned long)__builtin_return_address(0); +}; + +static void warn_slowpath_common(const char *file, int line, void *caller, struct slowpath_args *args) +{ const char *board; - sprint_symbol(function, caller); - printk(KERN_WARNING "------------[ cut here ]------------\n"); - printk(KERN_WARNING "WARNING: at %s:%d %s()\n", file, - line, function); + printk(KERN_WARNING "WARNING: at %s:%d %pS()\n", file, line, caller); board = dmi_get_system_info(DMI_PRODUCT_NAME); if (board) printk(KERN_WARNING "Hardware name: %s\n", board); - if (*fmt) { - va_start(args, fmt); - vprintk(fmt, args); - va_end(args); - } + if (args) + vprintk(args->fmt, args->args); print_modules(); dump_stack(); print_oops_end_marker(); add_taint(TAINT_WARN); } + +void warn_slowpath_fmt(const char *file, int line, const char *fmt, ...) +{ + struct slowpath_args args; + + args.fmt = fmt; + va_start(args.args, fmt); + warn_slowpath_common(file, line, __builtin_return_address(0), &args); + va_end(args.args); +} EXPORT_SYMBOL(warn_slowpath_fmt); void warn_slowpath_null(const char *file, int line) { - static const char *empty = ""; - warn_slowpath_fmt(file, line, empty); + warn_slowpath_common(file, line, __builtin_return_address(0), NULL); } EXPORT_SYMBOL(warn_slowpath_null); #endif From 826681043d7184b4d650cab5b007b9a86b628eb5 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 17 May 2009 16:20:18 +0100 Subject: [PATCH 068/267] [ARM] smp: fix cpumask usage in ARM SMP code The ARM SMP code wasn't properly updated for the cpumask changes, which results in smp_timer_broadcast() broadcasting ticks to non-online CPUs. Signed-off-by: Russell King --- arch/arm/common/gic.c | 4 +- arch/arm/include/asm/hardware/gic.h | 2 +- arch/arm/include/asm/smp.h | 12 ++---- arch/arm/kernel/smp.c | 46 ++++++++--------------- arch/arm/mach-realview/include/mach/smp.h | 6 +-- arch/arm/mach-realview/platsmp.c | 4 +- 6 files changed, 28 insertions(+), 46 deletions(-) diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c index c6884ba1d5ed..3e1714c6523f 100644 --- a/arch/arm/common/gic.c +++ b/arch/arm/common/gic.c @@ -253,9 +253,9 @@ void __cpuinit gic_cpu_init(unsigned int gic_nr, void __iomem *base) } #ifdef CONFIG_SMP -void gic_raise_softirq(cpumask_t cpumask, unsigned int irq) +void gic_raise_softirq(const struct cpumask *mask, unsigned int irq) { - unsigned long map = *cpus_addr(cpumask); + unsigned long map = *cpus_addr(*mask); /* this always happens on GIC0 */ writel(map << 16 | irq, gic_data[0].dist_base + GIC_DIST_SOFTINT); diff --git a/arch/arm/include/asm/hardware/gic.h b/arch/arm/include/asm/hardware/gic.h index 4924914af188..7f34333bb545 100644 --- a/arch/arm/include/asm/hardware/gic.h +++ b/arch/arm/include/asm/hardware/gic.h @@ -36,7 +36,7 @@ void gic_dist_init(unsigned int gic_nr, void __iomem *base, unsigned int irq_start); void gic_cpu_init(unsigned int gic_nr, void __iomem *base); void gic_cascade_irq(unsigned int gic_nr, unsigned int irq); -void gic_raise_softirq(cpumask_t cpumask, unsigned int irq); +void gic_raise_softirq(const struct cpumask *mask, unsigned int irq); #endif #endif diff --git a/arch/arm/include/asm/smp.h b/arch/arm/include/asm/smp.h index fad70da5911d..5995935338e1 100644 --- a/arch/arm/include/asm/smp.h +++ b/arch/arm/include/asm/smp.h @@ -53,17 +53,12 @@ extern void smp_store_cpu_info(unsigned int cpuid); /* * Raise an IPI cross call on CPUs in callmap. */ -extern void smp_cross_call(cpumask_t callmap); - -/* - * Broadcast a timer interrupt to the other CPUs. - */ -extern void smp_send_timer(void); +extern void smp_cross_call(const struct cpumask *mask); /* * Broadcast a clock event to other CPUs. */ -extern void smp_timer_broadcast(cpumask_t mask); +extern void smp_timer_broadcast(const struct cpumask *mask); /* * Boot a secondary CPU, and assign it the specified idle task. @@ -102,7 +97,8 @@ extern int platform_cpu_kill(unsigned int cpu); extern void platform_cpu_enable(unsigned int cpu); extern void arch_send_call_function_single_ipi(int cpu); -extern void arch_send_call_function_ipi(cpumask_t mask); +extern void arch_send_call_function_ipi_mask(const struct cpumask *mask); +#define arch_send_call_function_ipi_mask arch_send_call_function_ipi_mask /* * Local timer interrupt handling function (can be IPI'ed). diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c index 7801aac3c043..6014dfd22af4 100644 --- a/arch/arm/kernel/smp.c +++ b/arch/arm/kernel/smp.c @@ -326,14 +326,14 @@ void __init smp_prepare_boot_cpu(void) per_cpu(cpu_data, cpu).idle = current; } -static void send_ipi_message(cpumask_t callmap, enum ipi_msg_type msg) +static void send_ipi_message(const struct cpumask *mask, enum ipi_msg_type msg) { unsigned long flags; unsigned int cpu; local_irq_save(flags); - for_each_cpu_mask(cpu, callmap) { + for_each_cpu(cpu, mask) { struct ipi_data *ipi = &per_cpu(ipi_data, cpu); spin_lock(&ipi->lock); @@ -344,19 +344,19 @@ static void send_ipi_message(cpumask_t callmap, enum ipi_msg_type msg) /* * Call the platform specific cross-CPU call function. */ - smp_cross_call(callmap); + smp_cross_call(mask); local_irq_restore(flags); } -void arch_send_call_function_ipi(cpumask_t mask) +void arch_send_call_function_ipi_mask(const struct cpumask *mask) { send_ipi_message(mask, IPI_CALL_FUNC); } void arch_send_call_function_single_ipi(int cpu) { - send_ipi_message(cpumask_of_cpu(cpu), IPI_CALL_FUNC_SINGLE); + send_ipi_message(cpumask_of(cpu), IPI_CALL_FUNC_SINGLE); } void show_ipi_list(struct seq_file *p) @@ -498,17 +498,10 @@ asmlinkage void __exception do_IPI(struct pt_regs *regs) void smp_send_reschedule(int cpu) { - send_ipi_message(cpumask_of_cpu(cpu), IPI_RESCHEDULE); + send_ipi_message(cpumask_of(cpu), IPI_RESCHEDULE); } -void smp_send_timer(void) -{ - cpumask_t mask = cpu_online_map; - cpu_clear(smp_processor_id(), mask); - send_ipi_message(mask, IPI_TIMER); -} - -void smp_timer_broadcast(cpumask_t mask) +void smp_timer_broadcast(const struct cpumask *mask) { send_ipi_message(mask, IPI_TIMER); } @@ -517,7 +510,7 @@ void smp_send_stop(void) { cpumask_t mask = cpu_online_map; cpu_clear(smp_processor_id(), mask); - send_ipi_message(mask, IPI_CPU_STOP); + send_ipi_message(&mask, IPI_CPU_STOP); } /* @@ -528,20 +521,17 @@ int setup_profiling_timer(unsigned int multiplier) return -EINVAL; } -static int -on_each_cpu_mask(void (*func)(void *), void *info, int wait, cpumask_t mask) +static void +on_each_cpu_mask(void (*func)(void *), void *info, int wait, + const struct cpumask *mask) { - int ret = 0; - preempt_disable(); - ret = smp_call_function_mask(mask, func, info, wait); - if (cpu_isset(smp_processor_id(), mask)) + smp_call_function_many(mask, func, info, wait); + if (cpumask_test_cpu(smp_processor_id(), mask)) func(info); preempt_enable(); - - return ret; } /**********************************************************************/ @@ -602,20 +592,17 @@ void flush_tlb_all(void) void flush_tlb_mm(struct mm_struct *mm) { - cpumask_t mask = mm->cpu_vm_mask; - - on_each_cpu_mask(ipi_flush_tlb_mm, mm, 1, mask); + on_each_cpu_mask(ipi_flush_tlb_mm, mm, 1, &mm->cpu_vm_mask); } void flush_tlb_page(struct vm_area_struct *vma, unsigned long uaddr) { - cpumask_t mask = vma->vm_mm->cpu_vm_mask; struct tlb_args ta; ta.ta_vma = vma; ta.ta_start = uaddr; - on_each_cpu_mask(ipi_flush_tlb_page, &ta, 1, mask); + on_each_cpu_mask(ipi_flush_tlb_page, &ta, 1, &vma->vm_mm->cpu_vm_mask); } void flush_tlb_kernel_page(unsigned long kaddr) @@ -630,14 +617,13 @@ void flush_tlb_kernel_page(unsigned long kaddr) void flush_tlb_range(struct vm_area_struct *vma, unsigned long start, unsigned long end) { - cpumask_t mask = vma->vm_mm->cpu_vm_mask; struct tlb_args ta; ta.ta_vma = vma; ta.ta_start = start; ta.ta_end = end; - on_each_cpu_mask(ipi_flush_tlb_range, &ta, 1, mask); + on_each_cpu_mask(ipi_flush_tlb_range, &ta, 1, &vma->vm_mm->cpu_vm_mask); } void flush_tlb_kernel_range(unsigned long start, unsigned long end) diff --git a/arch/arm/mach-realview/include/mach/smp.h b/arch/arm/mach-realview/include/mach/smp.h index 515819efd046..e2ff53402f8c 100644 --- a/arch/arm/mach-realview/include/mach/smp.h +++ b/arch/arm/mach-realview/include/mach/smp.h @@ -15,15 +15,15 @@ /* * We use IRQ1 as the IPI */ -static inline void smp_cross_call(cpumask_t callmap) +static inline void smp_cross_call(const struct cpumask *mask) { - gic_raise_softirq(callmap, 1); + gic_raise_softirq(mask, 1); } /* * Do nothing on MPcore. */ -static inline void smp_cross_call_done(cpumask_t callmap) +static inline void smp_cross_call_done(const struct cpumask *mask) { } diff --git a/arch/arm/mach-realview/platsmp.c b/arch/arm/mach-realview/platsmp.c index ea3c75595fa9..a5730466dff5 100644 --- a/arch/arm/mach-realview/platsmp.c +++ b/arch/arm/mach-realview/platsmp.c @@ -82,7 +82,7 @@ void __cpuinit platform_secondary_init(unsigned int cpu) * to get this processor out of WFI in the BootMonitor - make * sure that we are no longer being sent this soft interrupt */ - smp_cross_call_done(cpumask_of_cpu(cpu)); + smp_cross_call_done(cpumask_of(cpu)); /* * if any interrupts are already enabled for the primary @@ -136,7 +136,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle) * Use smp_cross_call() for this, since there's little * point duplicating the code here */ - smp_cross_call(cpumask_of_cpu(cpu)); + smp_cross_call(cpumask_of(cpu)); timeout = jiffies + (1 * HZ); while (time_before(jiffies, timeout)) { From 78d236c2b30d4712c1fd8c9768b163c94b39e77d Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 17 May 2009 16:23:45 +0100 Subject: [PATCH 069/267] [ARM] realview: remove useless smp_cross_call_done() smp_cross_call_done() is a no-op for MPCore, and since it's only used by platform code, there's no point in having it unless it's doing something. Signed-off-by: Russell King --- arch/arm/mach-realview/include/mach/smp.h | 7 ------- arch/arm/mach-realview/platsmp.c | 7 ------- 2 files changed, 14 deletions(-) diff --git a/arch/arm/mach-realview/include/mach/smp.h b/arch/arm/mach-realview/include/mach/smp.h index e2ff53402f8c..dd53892d44a7 100644 --- a/arch/arm/mach-realview/include/mach/smp.h +++ b/arch/arm/mach-realview/include/mach/smp.h @@ -20,11 +20,4 @@ static inline void smp_cross_call(const struct cpumask *mask) gic_raise_softirq(mask, 1); } -/* - * Do nothing on MPcore. - */ -static inline void smp_cross_call_done(const struct cpumask *mask) -{ -} - #endif diff --git a/arch/arm/mach-realview/platsmp.c b/arch/arm/mach-realview/platsmp.c index a5730466dff5..b037fd6c82dc 100644 --- a/arch/arm/mach-realview/platsmp.c +++ b/arch/arm/mach-realview/platsmp.c @@ -77,13 +77,6 @@ void __cpuinit platform_secondary_init(unsigned int cpu) { trace_hardirqs_off(); - /* - * the primary core may have used a "cross call" soft interrupt - * to get this processor out of WFI in the BootMonitor - make - * sure that we are no longer being sent this soft interrupt - */ - smp_cross_call_done(cpumask_of(cpu)); - /* * if any interrupts are already enabled for the primary * core (e.g. timer irq), then they will not have been enabled From ee348d5a1d810bc9958cabb7c27302aab235d36e Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 17 May 2009 17:00:47 +0100 Subject: [PATCH 070/267] [ARM] realview: fix broadcast tick support Having discussed broadcast tick support with Thomas Glexiner, the broadcast tick devices should be registered with a higher rating than the global tick device, and it should have the ONESHOT and PERIODIC feature flags set. Signed-off-by: Russell King Acked-by: Thomas Glexiner --- arch/arm/mach-realview/core.c | 8 -------- arch/arm/mach-realview/localtimer.c | 6 ++++-- arch/arm/mach-realview/platsmp.c | 6 ++---- 3 files changed, 6 insertions(+), 14 deletions(-) diff --git a/arch/arm/mach-realview/core.c b/arch/arm/mach-realview/core.c index 942e1a7eb9b2..076acbc50706 100644 --- a/arch/arm/mach-realview/core.c +++ b/arch/arm/mach-realview/core.c @@ -750,14 +750,6 @@ void __init realview_timer_init(unsigned int timer_irq) { u32 val; -#ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST - /* - * The dummy clock device has to be registered before the main device - * so that the latter will broadcast the clock events - */ - local_timer_setup(); -#endif - /* * set clock frequency: * REALVIEW_REFCLK is 32KHz diff --git a/arch/arm/mach-realview/localtimer.c b/arch/arm/mach-realview/localtimer.c index d0d39adf6407..1c01d13460f0 100644 --- a/arch/arm/mach-realview/localtimer.c +++ b/arch/arm/mach-realview/localtimer.c @@ -189,8 +189,10 @@ void __cpuinit local_timer_setup(void) struct clock_event_device *clk = &per_cpu(local_clockevent, cpu); clk->name = "dummy_timer"; - clk->features = CLOCK_EVT_FEAT_DUMMY; - clk->rating = 200; + clk->features = CLOCK_EVT_FEAT_ONESHOT | + CLOCK_EVT_FEAT_PERIODIC | + CLOCK_EVT_FEAT_DUMMY; + clk->rating = 400; clk->mult = 1; clk->set_mode = dummy_timer_set_mode; clk->broadcast = smp_timer_broadcast; diff --git a/arch/arm/mach-realview/platsmp.c b/arch/arm/mach-realview/platsmp.c index b037fd6c82dc..30a9c68591f6 100644 --- a/arch/arm/mach-realview/platsmp.c +++ b/arch/arm/mach-realview/platsmp.c @@ -217,11 +217,9 @@ void __init smp_prepare_cpus(unsigned int max_cpus) if (max_cpus > ncores) max_cpus = ncores; -#ifdef CONFIG_LOCAL_TIMERS +#if defined(CONFIG_LOCAL_TIMERS) || defined(CONFIG_GENERIC_CLOCKEVENTS_BROADCAST) /* - * Enable the local timer for primary CPU. If the device is - * dummy (!CONFIG_LOCAL_TIMERS), it was already registers in - * realview_timer_init + * Enable the local timer or broadcast device for the boot CPU. */ local_timer_setup(); #endif From 12abb35a03e32c97235fcefdcf2d851be9f82dc2 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Sun, 17 May 2009 01:02:01 -0400 Subject: [PATCH 071/267] reiserfs: clean up ifdefs With xattr cleanup even with xattrs disabled, much of the initial setup is still performed. Some #ifdefs are just not needed since the options they protect wouldn't be available anyway. This cleans those up. Signed-off-by: Jeff Mahoney Signed-off-by: Linus Torvalds --- fs/reiserfs/xattr.c | 45 ++++++++++++++++++++++----------------------- 1 file changed, 22 insertions(+), 23 deletions(-) diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index 2237e10c7c7c..cf949646dd57 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -685,20 +685,6 @@ out: return err; } -/* Actual operations that are exported to VFS-land */ -struct xattr_handler *reiserfs_xattr_handlers[] = { - &reiserfs_xattr_user_handler, - &reiserfs_xattr_trusted_handler, -#ifdef CONFIG_REISERFS_FS_SECURITY - &reiserfs_xattr_security_handler, -#endif -#ifdef CONFIG_REISERFS_FS_POSIX_ACL - &reiserfs_posix_acl_access_handler, - &reiserfs_posix_acl_default_handler, -#endif - NULL -}; - /* * In order to implement different sets of xattr operations for each xattr * prefix with the generic xattr API, a filesystem should create a @@ -922,6 +908,28 @@ static int create_privroot(struct dentry *dentry) return 0; } +#else +int __init reiserfs_xattr_register_handlers(void) { return 0; } +void reiserfs_xattr_unregister_handlers(void) {} +static int create_privroot(struct dentry *dentry) { return 0; } +#endif + +/* Actual operations that are exported to VFS-land */ +struct xattr_handler *reiserfs_xattr_handlers[] = { +#ifdef CONFIG_REISERFS_FS_XATTR + &reiserfs_xattr_user_handler, + &reiserfs_xattr_trusted_handler, +#endif +#ifdef CONFIG_REISERFS_FS_SECURITY + &reiserfs_xattr_security_handler, +#endif +#ifdef CONFIG_REISERFS_FS_POSIX_ACL + &reiserfs_posix_acl_access_handler, + &reiserfs_posix_acl_default_handler, +#endif + NULL +}; + static int xattr_mount_check(struct super_block *s) { /* We need generation numbers to ensure that the oid mapping is correct @@ -941,11 +949,6 @@ static int xattr_mount_check(struct super_block *s) return 0; } -#else -int __init reiserfs_xattr_register_handlers(void) { return 0; } -void reiserfs_xattr_unregister_handlers(void) {} -#endif - /* This will catch lookups from the fs root to .reiserfs_priv */ static int xattr_lookup_poison(struct dentry *dentry, struct qstr *q1, struct qstr *name) @@ -992,7 +995,6 @@ int reiserfs_xattr_init(struct super_block *s, int mount_flags) int err = 0; struct dentry *privroot = REISERFS_SB(s)->priv_root; -#ifdef CONFIG_REISERFS_FS_XATTR err = xattr_mount_check(s); if (err) goto error; @@ -1023,14 +1025,11 @@ error: clear_bit(REISERFS_XATTRS_USER, &(REISERFS_SB(s)->s_mount_opt)); clear_bit(REISERFS_POSIXACL, &(REISERFS_SB(s)->s_mount_opt)); } -#endif /* The super_block MS_POSIXACL must mirror the (no)acl mount option. */ -#ifdef CONFIG_REISERFS_FS_POSIX_ACL if (reiserfs_posixacl(s)) s->s_flags |= MS_POSIXACL; else -#endif s->s_flags &= ~MS_POSIXACL; return err; From ceb5edc457f07956c82dccfb54ca8ae7e2a399f0 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Sun, 17 May 2009 01:02:02 -0400 Subject: [PATCH 072/267] reiserfs: deal with NULL xattr root w/ xattrs disabled This avoids an Oops in open_xa_root that can occur when deleting a file with xattrs disabled. It assumes that the xattr root will be there, and that is not guaranteed. Signed-off-by: Jeff Mahoney Signed-off-by: Linus Torvalds --- fs/reiserfs/super.c | 2 -- fs/reiserfs/xattr.c | 4 +++- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 1215a4f50cd2..3567fb9e3fb1 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -448,13 +448,11 @@ int remove_save_link(struct inode *inode, int truncate) static void reiserfs_kill_sb(struct super_block *s) { if (REISERFS_SB(s)) { -#ifdef CONFIG_REISERFS_FS_XATTR if (REISERFS_SB(s)->xattr_root) { d_invalidate(REISERFS_SB(s)->xattr_root); dput(REISERFS_SB(s)->xattr_root); REISERFS_SB(s)->xattr_root = NULL; } -#endif if (REISERFS_SB(s)->priv_root) { d_invalidate(REISERFS_SB(s)->priv_root); dput(REISERFS_SB(s)->priv_root); diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index cf949646dd57..628075ca82c1 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -123,7 +123,9 @@ static struct dentry *open_xa_root(struct super_block *sb, int flags) mutex_lock_nested(&privroot->d_inode->i_mutex, I_MUTEX_XATTR); xaroot = dget(REISERFS_SB(sb)->xattr_root); - if (!xaroot->d_inode) { + if (!xaroot) + xaroot = ERR_PTR(-ENODATA); + else if (!xaroot->d_inode) { int err = -ENODATA; if (xattr_may_create(flags)) err = xattr_mkdir(privroot->d_inode, xaroot, 0700); From b83674c0da6558e357c6b482ccf299eeea77d8ef Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Sun, 17 May 2009 01:02:03 -0400 Subject: [PATCH 073/267] reiserfs: fixup perms when xattrs are disabled This adds CONFIG_REISERFS_FS_XATTR protection from reiserfs_permission. This is needed to avoid warnings during file deletions and chowns with xattrs disabled. Signed-off-by: Jeff Mahoney Signed-off-by: Linus Torvalds --- fs/reiserfs/xattr.c | 36 ++++++++++++++++++---------------- include/linux/reiserfs_xattr.h | 4 +--- 2 files changed, 20 insertions(+), 20 deletions(-) diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index 628075ca82c1..8e7deb0e6964 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -871,23 +871,6 @@ static int reiserfs_check_acl(struct inode *inode, int mask) return error; } -int reiserfs_permission(struct inode *inode, int mask) -{ - /* - * We don't do permission checks on the internal objects. - * Permissions are determined by the "owning" object. - */ - if (IS_PRIVATE(inode)) - return 0; - /* - * Stat data v1 doesn't support ACLs. - */ - if (get_inode_sd_version(inode) == STAT_DATA_V1) - return generic_permission(inode, mask, NULL); - else - return generic_permission(inode, mask, reiserfs_check_acl); -} - static int create_privroot(struct dentry *dentry) { int err; @@ -951,6 +934,25 @@ static int xattr_mount_check(struct super_block *s) return 0; } +int reiserfs_permission(struct inode *inode, int mask) +{ + /* + * We don't do permission checks on the internal objects. + * Permissions are determined by the "owning" object. + */ + if (IS_PRIVATE(inode)) + return 0; + +#ifdef CONFIG_REISERFS_FS_XATTR + /* + * Stat data v1 doesn't support ACLs. + */ + if (get_inode_sd_version(inode) != STAT_DATA_V1) + return generic_permission(inode, mask, reiserfs_check_acl); +#endif + return generic_permission(inode, mask, NULL); +} + /* This will catch lookups from the fs root to .reiserfs_priv */ static int xattr_lookup_poison(struct dentry *dentry, struct qstr *q1, struct qstr *name) diff --git a/include/linux/reiserfs_xattr.h b/include/linux/reiserfs_xattr.h index cdedc01036e4..99928dce37ea 100644 --- a/include/linux/reiserfs_xattr.h +++ b/include/linux/reiserfs_xattr.h @@ -41,6 +41,7 @@ int reiserfs_xattr_init(struct super_block *sb, int mount_flags); int reiserfs_lookup_privroot(struct super_block *sb); int reiserfs_delete_xattrs(struct inode *inode); int reiserfs_chown_xattrs(struct inode *inode, struct iattr *attrs); +int reiserfs_permission(struct inode *inode, int mask); #ifdef CONFIG_REISERFS_FS_XATTR #define has_xattr_dir(inode) (REISERFS_I(inode)->i_flags & i_has_xattr_dir) @@ -50,7 +51,6 @@ int reiserfs_setxattr(struct dentry *dentry, const char *name, const void *value, size_t size, int flags); ssize_t reiserfs_listxattr(struct dentry *dentry, char *buffer, size_t size); int reiserfs_removexattr(struct dentry *dentry, const char *name); -int reiserfs_permission(struct inode *inode, int mask); int reiserfs_xattr_get(struct inode *, const char *, void *, size_t); int reiserfs_xattr_set(struct inode *, const char *, const void *, size_t, int); @@ -117,8 +117,6 @@ static inline void reiserfs_init_xattr_rwsem(struct inode *inode) #define reiserfs_listxattr NULL #define reiserfs_removexattr NULL -#define reiserfs_permission NULL - static inline void reiserfs_init_xattr_rwsem(struct inode *inode) { } From 22ef37eed673587ac984965dc88ba94c68873291 Mon Sep 17 00:00:00 2001 From: Toshiyuki Okajima Date: Sat, 16 May 2009 22:56:28 -0700 Subject: [PATCH 074/267] page-writeback: fix the calculation of the oldest_jif in wb_kupdate() wb_kupdate() function has a bug on linux-2.6.30-rc5. This bug causes generic_sync_sb_inodes() to start to write inodes back much earlier than our expectations because it miscalculates oldest_jif in wb_kupdate(). This bug was introduced in 704503d836042d4a4c7685b7036e7de0418fbc0f ('mm: fix proc_dointvec_userhz_jiffies "breakage"'). Signed-off-by: Toshiyuki Okajima Cc: Alexey Dobriyan Cc: Peter Zijlstra Cc: Nick Piggin Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page-writeback.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 30351f0063ac..bb553c3e955d 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -94,12 +94,12 @@ unsigned long vm_dirty_bytes; /* * The interval between `kupdate'-style writebacks */ -unsigned int dirty_writeback_interval = 5 * 100; /* sentiseconds */ +unsigned int dirty_writeback_interval = 5 * 100; /* centiseconds */ /* * The longest time for which data is allowed to remain dirty */ -unsigned int dirty_expire_interval = 30 * 100; /* sentiseconds */ +unsigned int dirty_expire_interval = 30 * 100; /* centiseconds */ /* * Flag that makes the machine dump writes/reads and block dirtyings. @@ -770,7 +770,7 @@ static void wb_kupdate(unsigned long arg) sync_supers(); - oldest_jif = jiffies - msecs_to_jiffies(dirty_expire_interval); + oldest_jif = jiffies - msecs_to_jiffies(dirty_expire_interval * 10); start_jif = jiffies; next_jif = start_jif + msecs_to_jiffies(dirty_writeback_interval * 10); nr_to_write = global_page_state(NR_FILE_DIRTY) + From 5e392739d6ab72f7c35040aa07f4097904bce6e7 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Mon, 11 May 2009 00:36:35 +0000 Subject: [PATCH 075/267] netpoll: don't dereference NULL dev from np It looks like the dev in netpoll_poll can be NULL - at lease it's checked at the function beginning. Thus the dev->netde_ops dereference looks dangerous. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller --- net/core/netpoll.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/net/core/netpoll.c b/net/core/netpoll.c index b5873bdff612..64f51eec6576 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -175,9 +175,13 @@ static void service_arp_queue(struct netpoll_info *npi) void netpoll_poll(struct netpoll *np) { struct net_device *dev = np->dev; - const struct net_device_ops *ops = dev->netdev_ops; + const struct net_device_ops *ops; - if (!dev || !netif_running(dev) || !ops->ndo_poll_controller) + if (!dev || !netif_running(dev)) + return; + + ops = dev->netdev_ops; + if (!ops->ndo_poll_controller) return; /* Process pending work on NIC */ From 2513dfb83fc775364fe85803d3a84d7ebe5763a5 Mon Sep 17 00:00:00 2001 From: Chris Friesen Date: Sun, 17 May 2009 20:39:33 -0700 Subject: [PATCH 076/267] ipconfig: handle case of delayed DHCP server If a DHCP server is delayed, it's possible for the client to receive the DHCPOFFER after it has already sent out a new DHCPDISCOVER message from a second interface. The client then sends out a DHCPREQUEST from the second interface, but the server doesn't recognize the device and rejects the request. This patch simply tracks the current device being configured and throws away the OFFER if it is not intended for the current device. A more sophisticated approach would be to put the OFFER information into the struct ic_device rather than storing it globally. Signed-off-by: Chris Friesen Signed-off-by: David S. Miller --- net/ipv4/ipconfig.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index 90d22ae0a419..88bf051d0cbb 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -139,6 +139,8 @@ __be32 ic_servaddr = NONE; /* Boot server IP address */ __be32 root_server_addr = NONE; /* Address of NFS server */ u8 root_server_path[256] = { 0, }; /* Path to mount as root */ +u32 ic_dev_xid; /* Device under configuration */ + /* vendor class identifier */ static char vendor_class_identifier[253] __initdata; @@ -932,6 +934,13 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str goto drop_unlock; } + /* Is it a reply for the device we are configuring? */ + if (b->xid != ic_dev_xid) { + if (net_ratelimit()) + printk(KERN_ERR "DHCP/BOOTP: Ignoring delayed packet \n"); + goto drop_unlock; + } + /* Parse extensions */ if (ext_len >= 4 && !memcmp(b->exten, ic_bootp_cookie, 4)) { /* Check magic cookie */ @@ -1115,6 +1124,9 @@ static int __init ic_dynamic(void) get_random_bytes(&timeout, sizeof(timeout)); timeout = CONF_BASE_TIMEOUT + (timeout % (unsigned) CONF_TIMEOUT_RANDOM); for (;;) { + /* Track the device we are configuring */ + ic_dev_xid = d->xid; + #ifdef IPCONFIG_BOOTP if (do_bootp && (d->able & IC_BOOTP)) ic_bootp_send_if(d, jiffies - start_jiffies); From 72876a603422d3767273ffb3918033fa1bfda0f3 Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Sun, 17 May 2009 20:48:59 -0700 Subject: [PATCH 077/267] mlx4_en: Fix not deleted napi structures Napi structures are being created each time we open a port, but when the port is closed the napi structure is only disabled but not removed. This bug caused hang while removing the driver. Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- drivers/net/mlx4/en_cq.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/mlx4/en_cq.c b/drivers/net/mlx4/en_cq.c index 91f50de84be9..a276125b709b 100644 --- a/drivers/net/mlx4/en_cq.c +++ b/drivers/net/mlx4/en_cq.c @@ -125,8 +125,10 @@ void mlx4_en_deactivate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq) if (cq->is_tx) del_timer(&cq->timer); - else + else { napi_disable(&cq->napi); + netif_napi_del(&cq->napi); + } mlx4_cq_free(mdev->dev, &cq->mcq); } From a8f492c6ac6072d6c5bce4b2f13dc44612991051 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sat, 16 May 2009 01:21:58 +0000 Subject: [PATCH 078/267] NET: Meth: Fix unsafe mix of irq and non-irq spinlocks. Mixing of normal and irq spinlocks results in the following lockdep messages on bootup on IP32: [...] Sending DHCP requests . ====================================================== [ INFO: SOFTIRQ-safe -> SOFTIRQ-unsafe lock order detected ] 2.6.30-rc5-00164-g41baeef #30 ------------------------------------------------------ swapper/1 [HC0[0]:SC0[1]:HE0:SE0] is trying to acquire: (&priv->meth_lock){+.+...}, at: [] meth_tx+0x48/0x43c and this task is already holding: (_xmit_ETHER#2){+.-...}, at: [] __qdisc_run+0x118/0x30c which would create a new lock dependency: (_xmit_ETHER#2){+.-...} -> (&priv->meth_lock){+.+...} but this new dependency connects a SOFTIRQ-irq-safe lock: (_xmit_ETHER#2){+.-...} ... which became SOFTIRQ-irq-safe at: [] __lock_acquire+0x784/0x1a14 [] lock_acquire+0xf8/0x150 [] _spin_lock+0x30/0x44 [] dev_watchdog+0x70/0x398 [] run_timer_softirq+0x1a8/0x248 [] __do_softirq+0xec/0x208 [] do_softirq+0x60/0xe4 [] irq_exit+0x54/0x9c [] ret_from_irq+0x0/0x4 [] r4k_wait+0x20/0x40 [] cpu_idle+0x30/0x60 [] start_kernel+0x3ec/0x404 to a SOFTIRQ-irq-unsafe lock: (&priv->meth_lock){+.+...} ... which became SOFTIRQ-irq-unsafe at: ... [] __lock_acquire+0x824/0x1a14 [] lock_acquire+0xf8/0x150 [] _spin_lock+0x30/0x44 [] meth_reset+0x118/0x2d8 [] meth_open+0x28/0x140 [] dev_open+0xe0/0x18c [] dev_change_flags+0xd8/0x1d4 [] ip_auto_config+0x1d4/0xf28 [] do_one_initcall+0x58/0x170 [] kernel_init+0x98/0x104 [] kernel_thread_helper+0x10/0x18 other info that might help us debug this: 2 locks held by swapper/1: #0: (rcu_read_lock){.+.+..}, at: [] dev_queue_xmit+0x1e0/0x4b0 #1: (_xmit_ETHER#2){+.-...}, at: [] __qdisc_run+0x118/0x30c the SOFTIRQ-irq-safe lock's dependencies: -> (_xmit_ETHER#2){+.-...} ops: 0 { HARDIRQ-ON-W at: [] __lock_acquire+0x7fc/0x1a14 [] lock_acquire+0xf8/0x150 [] _spin_lock+0x30/0x44 [] dev_watchdog+0x70/0x398 [] run_timer_softirq+0x1a8/0x248 [] __do_softirq+0xec/0x208 [] do_softirq+0x60/0xe4 [] irq_exit+0x54/0x9c [] ret_from_irq+0x0/0x4 [] r4k_wait+0x20/0x40 [] cpu_idle+0x30/0x60 [] start_kernel+0x3ec/0x404 IN-SOFTIRQ-W at: [] __lock_acquire+0x784/0x1a14 [] lock_acquire+0xf8/0x150 [] _spin_lock+0x30/0x44 [] dev_watchdog+0x70/0x398 [] run_timer_softirq+0x1a8/0x248 [] __do_softirq+0xec/0x208 [] do_softirq+0x60/0xe4 [] irq_exit+0x54/0x9c [] ret_from_irq+0x0/0x4 [] r4k_wait+0x20/0x40 [] cpu_idle+0x30/0x60 [] start_kernel+0x3ec/0x404 INITIAL USE at: [] __lock_acquire+0x89c/0x1a14 [] lock_acquire+0xf8/0x150 [] _spin_lock+0x30/0x44 [] dev_watchdog+0x70/0x398 [] run_timer_softirq+0x1a8/0x248 [] __do_softirq+0xec/0x208 [] do_softirq+0x60/0xe4 [] irq_exit+0x54/0x9c [] ret_from_irq+0x0/0x4 [] r4k_wait+0x20/0x40 [] cpu_idle+0x30/0x60 [] start_kernel+0x3ec/0x404 } ... key at: [] netdev_xmit_lock_key+0x8/0x1c8 the SOFTIRQ-irq-unsafe lock's dependencies: -> (&priv->meth_lock){+.+...} ops: 0 { HARDIRQ-ON-W at: [] __lock_acquire+0x7fc/0x1a14 [] lock_acquire+0xf8/0x150 [] _spin_lock+0x30/0x44 [] meth_reset+0x118/0x2d8 [] meth_open+0x28/0x140 [] dev_open+0xe0/0x18c [] dev_change_flags+0xd8/0x1d4 [] ip_auto_config+0x1d4/0xf28 [] do_one_initcall+0x58/0x170 [] kernel_init+0x98/0x104 [] kernel_thread_helper+0x10/0x18 SOFTIRQ-ON-W at: [] __lock_acquire+0x824/0x1a14 [] lock_acquire+0xf8/0x150 [] _spin_lock+0x30/0x44 [] meth_reset+0x118/0x2d8 [] meth_open+0x28/0x140 [] dev_open+0xe0/0x18c [] dev_change_flags+0xd8/0x1d4 [] ip_auto_config+0x1d4/0xf28 [] do_one_initcall+0x58/0x170 [] kernel_init+0x98/0x104 [] kernel_thread_helper+0x10/0x18 INITIAL USE at: [] __lock_acquire+0x89c/0x1a14 [] lock_acquire+0xf8/0x150 [] _spin_lock+0x30/0x44 [] meth_reset+0x118/0x2d8 [] meth_open+0x28/0x140 [] dev_open+0xe0/0x18c [] dev_change_flags+0xd8/0x1d4 [] ip_auto_config+0x1d4/0xf28 [] do_one_initcall+0x58/0x170 [] kernel_init+0x98/0x104 [] kernel_thread_helper+0x10/0x18 } ... key at: [] __key.32424+0x0/0x8 stack backtrace: Call Trace: [] dump_stack+0x8/0x34 [] check_usage+0x470/0x4a0 [] check_irq_usage+0x90/0x130 [] __lock_acquire+0x12a4/0x1a14 [] lock_acquire+0xf8/0x150 [] _spin_lock_irqsave+0x60/0x84 [] meth_tx+0x48/0x43c [] __qdisc_run+0x150/0x30c [] dev_queue_xmit+0x334/0x4b0 [] ip_auto_config+0x8d0/0xf28 [] do_one_initcall+0x58/0x170 [] kernel_init+0x98/0x104 [] kernel_thread_helper+0x10/0x18 ..... timed out! IP-Config: Retrying forever (NFS root)... Sending DHCP requests ., OK [...] Fixed by converting all locks to irq locks. Signed-off-by: Ralf Baechle Tested-by: Andrew Randrianasulu Signed-off-by: David S. Miller --- drivers/net/meth.c | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-) diff --git a/drivers/net/meth.c b/drivers/net/meth.c index aa08987f6e81..dbd3436912b8 100644 --- a/drivers/net/meth.c +++ b/drivers/net/meth.c @@ -127,11 +127,11 @@ static unsigned long mdio_read(struct meth_private *priv, unsigned long phyreg) static int mdio_probe(struct meth_private *priv) { int i; - unsigned long p2, p3; + unsigned long p2, p3, flags; /* check if phy is detected already */ if(priv->phy_addr>=0&&priv->phy_addr<32) return 0; - spin_lock(&priv->meth_lock); + spin_lock_irqsave(&priv->meth_lock, flags); for (i=0;i<32;++i){ priv->phy_addr=i; p2=mdio_read(priv,2); @@ -157,7 +157,7 @@ static int mdio_probe(struct meth_private *priv) break; } } - spin_unlock(&priv->meth_lock); + spin_unlock_irqrestore(&priv->meth_lock, flags); if(priv->phy_addr<32) { return 0; } @@ -373,14 +373,14 @@ static int meth_release(struct net_device *dev) static void meth_rx(struct net_device* dev, unsigned long int_status) { struct sk_buff *skb; - unsigned long status; + unsigned long status, flags; struct meth_private *priv = netdev_priv(dev); unsigned long fifo_rptr = (int_status & METH_INT_RX_RPTR_MASK) >> 8; - spin_lock(&priv->meth_lock); + spin_lock_irqsave(&priv->meth_lock, flags); priv->dma_ctrl &= ~METH_DMA_RX_INT_EN; mace->eth.dma_ctrl = priv->dma_ctrl; - spin_unlock(&priv->meth_lock); + spin_unlock_irqrestore(&priv->meth_lock, flags); if (int_status & METH_INT_RX_UNDERFLOW) { fifo_rptr = (fifo_rptr - 1) & 0x0f; @@ -452,12 +452,12 @@ static void meth_rx(struct net_device* dev, unsigned long int_status) mace->eth.rx_fifo = priv->rx_ring_dmas[priv->rx_write]; ADVANCE_RX_PTR(priv->rx_write); } - spin_lock(&priv->meth_lock); + spin_lock_irqsave(&priv->meth_lock, flags); /* In case there was underflow, and Rx DMA was disabled */ priv->dma_ctrl |= METH_DMA_RX_INT_EN | METH_DMA_RX_EN; mace->eth.dma_ctrl = priv->dma_ctrl; mace->eth.int_stat = METH_INT_RX_THRESHOLD; - spin_unlock(&priv->meth_lock); + spin_unlock_irqrestore(&priv->meth_lock, flags); } static int meth_tx_full(struct net_device *dev) @@ -470,11 +470,11 @@ static int meth_tx_full(struct net_device *dev) static void meth_tx_cleanup(struct net_device* dev, unsigned long int_status) { struct meth_private *priv = netdev_priv(dev); - unsigned long status; + unsigned long status, flags; struct sk_buff *skb; unsigned long rptr = (int_status&TX_INFO_RPTR) >> 16; - spin_lock(&priv->meth_lock); + spin_lock_irqsave(&priv->meth_lock, flags); /* Stop DMA notification */ priv->dma_ctrl &= ~(METH_DMA_TX_INT_EN); @@ -527,12 +527,13 @@ static void meth_tx_cleanup(struct net_device* dev, unsigned long int_status) } mace->eth.int_stat = METH_INT_TX_EMPTY | METH_INT_TX_PKT; - spin_unlock(&priv->meth_lock); + spin_unlock_irqrestore(&priv->meth_lock, flags); } static void meth_error(struct net_device* dev, unsigned status) { struct meth_private *priv = netdev_priv(dev); + unsigned long flags; printk(KERN_WARNING "meth: error status: 0x%08x\n",status); /* check for errors too... */ @@ -547,7 +548,7 @@ static void meth_error(struct net_device* dev, unsigned status) printk(KERN_WARNING "meth: Rx overflow\n"); if (status & (METH_INT_RX_UNDERFLOW)) { printk(KERN_WARNING "meth: Rx underflow\n"); - spin_lock(&priv->meth_lock); + spin_lock_irqsave(&priv->meth_lock, flags); mace->eth.int_stat = METH_INT_RX_UNDERFLOW; /* more underflow interrupts will be delivered, * effectively throwing us into an infinite loop. @@ -555,7 +556,7 @@ static void meth_error(struct net_device* dev, unsigned status) priv->dma_ctrl &= ~METH_DMA_RX_EN; mace->eth.dma_ctrl = priv->dma_ctrl; DPRINTK("Disabled meth Rx DMA temporarily\n"); - spin_unlock(&priv->meth_lock); + spin_unlock_irqrestore(&priv->meth_lock, flags); } mace->eth.int_stat = METH_INT_ERROR; } From a598f6aebea2481531b0757ed90cfb0d8cf1d8f5 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 15 May 2009 06:10:13 +0000 Subject: [PATCH 079/267] bridge: relay bridge multicast pkgs if !STP Currently the bridge catches all STP packets; even if STP is turned off. This prevents other systems (which do have STP turned on) from being able to detect loops in the network. With this patch, if STP is off, then any packet sent to the STP multicast group address is forwarded to all ports. Based on earlier patch by Joakim Tjernlund with changes to go through forwarding (not local chain), and optimization that only last octet needs to be checked. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/bridge/br_input.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 30b88777c3df..5ee1a3682bf2 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -134,6 +134,10 @@ struct sk_buff *br_handle_frame(struct net_bridge_port *p, struct sk_buff *skb) if (skb->protocol == htons(ETH_P_PAUSE)) goto drop; + /* If STP is turned off, then forward */ + if (p->br->stp_enabled == BR_NO_STP && dest[5] == 0) + goto forward; + if (NF_HOOK(PF_BRIDGE, NF_BR_LOCAL_IN, skb, skb->dev, NULL, br_handle_local_finish)) return NULL; /* frame consumed by filter */ @@ -141,6 +145,7 @@ struct sk_buff *br_handle_frame(struct net_bridge_port *p, struct sk_buff *skb) return skb; /* continue processing */ } +forward: switch (p->state) { case BR_STATE_FORWARDING: rhook = rcu_dereference(br_should_route_hook); From 4f0611af47e25807cf18cd2b4d4e94206c75b29e Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 15 May 2009 06:11:58 +0000 Subject: [PATCH 080/267] bridge: fix initial packet flood if !STP If bridge is configured with no STP and forwarding delay of 0 (which is typical for virtualization) then when link starts it will flood all packets for the first 20 seconds. This bug was introduced by a combination of earlier changes: * forwarding database uses hold time of zero to indicate user wants to always flood packets * optimzation of the case of forwarding delay of 0 avoids the initial timer tick The fix is to just skip all the topology change detection code if kernel STP is not being used. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/bridge/br_stp.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/bridge/br_stp.c b/net/bridge/br_stp.c index 6e63ec3f1fcf..0660515f3992 100644 --- a/net/bridge/br_stp.c +++ b/net/bridge/br_stp.c @@ -297,6 +297,9 @@ void br_topology_change_detection(struct net_bridge *br) { int isroot = br_is_root_bridge(br); + if (br->stp_enabled != BR_KERNEL_STP) + return; + pr_info("%s: topology change detected, %s\n", br->dev->name, isroot ? "propagating" : "sending tcn bpdu"); From 4cd6fe1c6483cde93e2ec91f58b7af9c9eea51ad Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 15 May 2009 08:44:32 +0000 Subject: [PATCH 081/267] bonding: fix link down handling in 802.3ad mode One of the purposes of bonding is to allow for redundant links, and failover correctly if the cable is pulled. If all the members of a bonded device have no carrier present, the bonded device itself needs to report no carrier present to user space so management tools (like routing daemons) can respond. Bonding in 802.3ad mode does not work correctly for this because it incorrectly chooses a link that is down as a possible aggregator. Signed-off-by: Stephen Hemminger Signed-off-by: Jay Vosburgh Signed-off-by: David S. Miller --- drivers/net/bonding/bond_3ad.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c index 8c2e5ab51f08..faf094abef7f 100644 --- a/drivers/net/bonding/bond_3ad.c +++ b/drivers/net/bonding/bond_3ad.c @@ -1465,6 +1465,12 @@ static struct aggregator *ad_agg_selection_test(struct aggregator *best, return best; } +static int agg_device_up(const struct aggregator *agg) +{ + return (netif_running(agg->slave->dev) && + netif_carrier_ok(agg->slave->dev)); +} + /** * ad_agg_selection_logic - select an aggregation group for a team * @aggregator: the aggregator we're looking at @@ -1496,14 +1502,13 @@ static void ad_agg_selection_logic(struct aggregator *agg) struct port *port; origin = agg; - active = __get_active_agg(agg); - best = active; + best = (active && agg_device_up(active)) ? active : NULL; do { agg->is_active = 0; - if (agg->num_of_ports) + if (agg->num_of_ports && agg_device_up(agg)) best = ad_agg_selection_test(best, agg); } while ((agg = __get_next_agg(agg))); From eb0519b5a1cf07d5a1b919cffa0304cdaaa97a7e Mon Sep 17 00:00:00 2001 From: Gabriel Paubert Date: Sun, 17 May 2009 21:16:47 -0700 Subject: [PATCH 082/267] mv643xx_eth: fix PPC DMA breakage After 2.6.29, PPC no more admits passing NULL to the dev parameter of the DMA API. The result is a BUG followed by solid lock-up when the mv643xx_eth driver brings an interface up. The following patch makes the driver work on my Pegasos again; it is mostly a search and replace of NULL by mp->dev->dev.parent in dma allocation/freeing/mapping/unmapping functions. Signed-off-by: Gabriel Paubert Acked-by: Lennert Buytenhek Signed-off-by: David S. Miller --- drivers/net/mv643xx_eth.c | 41 +++++++++++++++++++++------------------ 1 file changed, 22 insertions(+), 19 deletions(-) diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index a400d7115f78..6bb5af35eda6 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -569,7 +569,7 @@ static int rxq_process(struct rx_queue *rxq, int budget) if (rxq->rx_curr_desc == rxq->rx_ring_size) rxq->rx_curr_desc = 0; - dma_unmap_single(NULL, rx_desc->buf_ptr, + dma_unmap_single(mp->dev->dev.parent, rx_desc->buf_ptr, rx_desc->buf_size, DMA_FROM_DEVICE); rxq->rx_desc_count--; rx++; @@ -678,8 +678,9 @@ static int rxq_refill(struct rx_queue *rxq, int budget) rx_desc = rxq->rx_desc_area + rx; - rx_desc->buf_ptr = dma_map_single(NULL, skb->data, - mp->skb_size, DMA_FROM_DEVICE); + rx_desc->buf_ptr = dma_map_single(mp->dev->dev.parent, + skb->data, mp->skb_size, + DMA_FROM_DEVICE); rx_desc->buf_size = mp->skb_size; rxq->rx_skb[rx] = skb; wmb(); @@ -718,6 +719,7 @@ static inline unsigned int has_tiny_unaligned_frags(struct sk_buff *skb) static void txq_submit_frag_skb(struct tx_queue *txq, struct sk_buff *skb) { + struct mv643xx_eth_private *mp = txq_to_mp(txq); int nr_frags = skb_shinfo(skb)->nr_frags; int frag; @@ -746,10 +748,10 @@ static void txq_submit_frag_skb(struct tx_queue *txq, struct sk_buff *skb) desc->l4i_chk = 0; desc->byte_cnt = this_frag->size; - desc->buf_ptr = dma_map_page(NULL, this_frag->page, - this_frag->page_offset, - this_frag->size, - DMA_TO_DEVICE); + desc->buf_ptr = dma_map_page(mp->dev->dev.parent, + this_frag->page, + this_frag->page_offset, + this_frag->size, DMA_TO_DEVICE); } } @@ -826,7 +828,8 @@ no_csum: desc->l4i_chk = l4i_chk; desc->byte_cnt = length; - desc->buf_ptr = dma_map_single(NULL, skb->data, length, DMA_TO_DEVICE); + desc->buf_ptr = dma_map_single(mp->dev->dev.parent, skb->data, + length, DMA_TO_DEVICE); __skb_queue_tail(&txq->tx_skb, skb); @@ -956,10 +959,10 @@ static int txq_reclaim(struct tx_queue *txq, int budget, int force) } if (cmd_sts & TX_FIRST_DESC) { - dma_unmap_single(NULL, desc->buf_ptr, + dma_unmap_single(mp->dev->dev.parent, desc->buf_ptr, desc->byte_cnt, DMA_TO_DEVICE); } else { - dma_unmap_page(NULL, desc->buf_ptr, + dma_unmap_page(mp->dev->dev.parent, desc->buf_ptr, desc->byte_cnt, DMA_TO_DEVICE); } @@ -1894,9 +1897,9 @@ static int rxq_init(struct mv643xx_eth_private *mp, int index) mp->rx_desc_sram_size); rxq->rx_desc_dma = mp->rx_desc_sram_addr; } else { - rxq->rx_desc_area = dma_alloc_coherent(NULL, size, - &rxq->rx_desc_dma, - GFP_KERNEL); + rxq->rx_desc_area = dma_alloc_coherent(mp->dev->dev.parent, + size, &rxq->rx_desc_dma, + GFP_KERNEL); } if (rxq->rx_desc_area == NULL) { @@ -1947,7 +1950,7 @@ out_free: if (index == 0 && size <= mp->rx_desc_sram_size) iounmap(rxq->rx_desc_area); else - dma_free_coherent(NULL, size, + dma_free_coherent(mp->dev->dev.parent, size, rxq->rx_desc_area, rxq->rx_desc_dma); @@ -1979,7 +1982,7 @@ static void rxq_deinit(struct rx_queue *rxq) rxq->rx_desc_area_size <= mp->rx_desc_sram_size) iounmap(rxq->rx_desc_area); else - dma_free_coherent(NULL, rxq->rx_desc_area_size, + dma_free_coherent(mp->dev->dev.parent, rxq->rx_desc_area_size, rxq->rx_desc_area, rxq->rx_desc_dma); kfree(rxq->rx_skb); @@ -2007,9 +2010,9 @@ static int txq_init(struct mv643xx_eth_private *mp, int index) mp->tx_desc_sram_size); txq->tx_desc_dma = mp->tx_desc_sram_addr; } else { - txq->tx_desc_area = dma_alloc_coherent(NULL, size, - &txq->tx_desc_dma, - GFP_KERNEL); + txq->tx_desc_area = dma_alloc_coherent(mp->dev->dev.parent, + size, &txq->tx_desc_dma, + GFP_KERNEL); } if (txq->tx_desc_area == NULL) { @@ -2053,7 +2056,7 @@ static void txq_deinit(struct tx_queue *txq) txq->tx_desc_area_size <= mp->tx_desc_sram_size) iounmap(txq->tx_desc_area); else - dma_free_coherent(NULL, txq->tx_desc_area_size, + dma_free_coherent(mp->dev->dev.parent, txq->tx_desc_area_size, txq->tx_desc_area, txq->tx_desc_dma); } From d77dd8d27e73a9277096453eb901aae0bfd4ca3c Mon Sep 17 00:00:00 2001 From: roel kluin Date: Fri, 15 May 2009 10:19:51 +0000 Subject: [PATCH 083/267] Neterion: *FIFO1_DMA_ERR set twice, should 2nd be *FIFO2_DMA_ERR? FIFO1_DMA_ERR is set twice, the second should be FIFO2_DMA_ERR. Signed-off-by: Roel Kluin Acked-by: Ram Vepa Signed-off-by: David S. Miller --- drivers/net/vxge/vxge-traffic.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/vxge/vxge-traffic.c b/drivers/net/vxge/vxge-traffic.c index 7be0ae10d69b..c2eeac4125f3 100644 --- a/drivers/net/vxge/vxge-traffic.c +++ b/drivers/net/vxge/vxge-traffic.c @@ -115,7 +115,7 @@ enum vxge_hw_status vxge_hw_vpath_intr_enable(struct __vxge_hw_vpath_handle *vp) VXGE_HW_KDFCCTL_ERRORS_REG_KDFCCTL_FIFO1_POISON| VXGE_HW_KDFCCTL_ERRORS_REG_KDFCCTL_FIFO2_POISON| VXGE_HW_KDFCCTL_ERRORS_REG_KDFCCTL_FIFO1_DMA_ERR| - VXGE_HW_KDFCCTL_ERRORS_REG_KDFCCTL_FIFO1_DMA_ERR), 0, 32), + VXGE_HW_KDFCCTL_ERRORS_REG_KDFCCTL_FIFO2_DMA_ERR), 0, 32), &vp_reg->kdfcctl_errors_mask); __vxge_hw_pio_mem_write32_upper(0, &vp_reg->vpath_ppif_int_mask); From 705efc3b03cbee449e4d83b230423894152f7982 Mon Sep 17 00:00:00 2001 From: Wang Tinggong Date: Thu, 14 May 2009 22:49:36 +0000 Subject: [PATCH 084/267] Doc: fixed descriptions on /proc/sys/net/core/* and /proc/sys/net/unix/* Signed-off-by: Wang Tinggong Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index ec5de02f543f..b121c5db707f 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -1266,13 +1266,22 @@ sctp_rmem - vector of 3 INTEGERs: min, default, max sctp_wmem - vector of 3 INTEGERs: min, default, max See tcp_wmem for a description. -UNDOCUMENTED: /proc/sys/net/core/* - dev_weight FIXME +dev_weight - INTEGER + The maximum number of packets that kernel can handle on a NAPI + interrupt, it's a Per-CPU variable. + + Default: 64 /proc/sys/net/unix/* - max_dgram_qlen FIXME +max_dgram_qlen - INTEGER + The maximum length of dgram socket receive queue + + Default: 10 + + +UNDOCUMENTED: /proc/sys/net/irda/* fast_poll_increase FIXME From af3e4aca47d2e05a545a5e10ba5c7193e0b665e0 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 30 Apr 2009 10:59:19 +0000 Subject: [PATCH 085/267] powerpc: Do not assert pte_locked for hugepage PTE entries With CONFIG_DEBUG_VM, an assertion is made when changing the protection flags of a PTE that the PTE is locked. Huge pages use a different pagetable format and the assertion is bogus and will always trigger with a bug looking something like Unable to handle kernel paging request for data at address 0xf1a00235800006f8 Faulting instruction address: 0xc000000000034a80 Oops: Kernel access of bad area, sig: 11 [#1] SMP NR_CPUS=32 NUMA Maple Modules linked in: dm_snapshot dm_mirror dm_region_hash dm_log dm_mod loop evdev ext3 jbd mbcache sg sd_mod ide_pci_generic pata_amd ata_generic ipr libata tg3 libphy scsi_mod windfarm_pid windfarm_smu_sat windfarm_max6690_sensor windfarm_lm75_sensor windfarm_cpufreq_clamp windfarm_core i2c_powermac NIP: c000000000034a80 LR: c000000000034b18 CTR: 0000000000000003 REGS: c000000003037600 TRAP: 0300 Not tainted (2.6.30-rc3-autokern1) MSR: 9000000000009032 CR: 28002484 XER: 200fffff DAR: f1a00235800006f8, DSISR: 0000000040010000 TASK = c0000002e54cc740[2960] 'map_high_trunca' THREAD: c000000003034000 CPU: 2 GPR00: 4000000000000000 c000000003037880 c000000000895d30 c0000002e5a2e500 GPR04: 00000000a0000000 c0000002edc40880 0000005700000393 0000000000000001 GPR08: f000000011ac0000 01a00235800006e8 00000000000000f5 f1a00235800006e8 GPR12: 0000000028000484 c0000000008dd780 0000000000001000 0000000000000000 GPR16: fffffffffffff000 0000000000000000 00000000a0000000 c000000003037a20 GPR20: c0000002e5f4ece8 0000000000001000 c0000002edc40880 0000000000000000 GPR24: c0000002e5f4ece8 0000000000000000 00000000a0000000 c0000002e5f4ece8 GPR28: 0000005700000393 c0000002e5a2e500 00000000a0000000 c000000003037880 NIP [c000000000034a80] .assert_pte_locked+0xa4/0xd0 LR [c000000000034b18] .ptep_set_access_flags+0x6c/0xb4 Call Trace: [c000000003037880] [c000000003037990] 0xc000000003037990 (unreliable) [c000000003037910] [c000000000034b18] .ptep_set_access_flags+0x6c/0xb4 [c0000000030379b0] [c00000000014bef8] .hugetlb_cow+0x124/0x674 [c000000003037b00] [c00000000014c930] .hugetlb_fault+0x4e8/0x6f8 [c000000003037c00] [c00000000013443c] .handle_mm_fault+0xac/0x828 [c000000003037cf0] [c0000000000340a8] .do_page_fault+0x39c/0x584 [c000000003037e30] [c0000000000057b0] handle_page_fault+0x20/0x5c Instruction dump: 7d29582a 7d200074 7800d182 0b000000 3c004000 3960ffff 780007c6 796b00c4 7d290214 7929a302 1d290068 7d6b4a14 <800b0010> 7c000074 7800d182 0b000000 This patch fixes the problem by not asseting the PTE is locked for VMAs backed by huge pages. Signed-off-by: Mel Gorman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/mm/pgtable.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/mm/pgtable.c b/arch/powerpc/mm/pgtable.c index f5c6fd42265c..ae1d67cc090c 100644 --- a/arch/powerpc/mm/pgtable.c +++ b/arch/powerpc/mm/pgtable.c @@ -219,7 +219,8 @@ int ptep_set_access_flags(struct vm_area_struct *vma, unsigned long address, entry = do_dcache_icache_coherency(entry); changed = !pte_same(*(ptep), entry); if (changed) { - assert_pte_locked(vma->vm_mm, address); + if (!(vma->vm_flags & VM_HUGETLB)) + assert_pte_locked(vma->vm_mm, address); __ptep_set_access_flags(ptep, entry); flush_tlb_page_nohash(vma, address); } From 021376a3b655364c92c10be544a3319946a792e8 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Wed, 13 May 2009 20:30:24 +0000 Subject: [PATCH 086/267] powerpc/ftrace: Use pr_devel() in ftrace.c pr_debug() can now result in code being generated even when #DEBUG is not defined. That's not really desirable in the ftrace code which we want to be snappy. With CONFIG_DYNAMIC_DEBUG=y: size before: text data bss dec hex filename 3334 672 4 4010 faa arch/powerpc/kernel/ftrace.o size after: text data bss dec hex filename 2616 360 4 2980 ba4 arch/powerpc/kernel/ftrace.o Signed-off-by: Michael Ellerman Acked-by: Steven Rostedt Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/ftrace.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/arch/powerpc/kernel/ftrace.c b/arch/powerpc/kernel/ftrace.c index 70e2a736be1f..5b078ee391fc 100644 --- a/arch/powerpc/kernel/ftrace.c +++ b/arch/powerpc/kernel/ftrace.c @@ -157,7 +157,7 @@ __ftrace_make_nop(struct module *mod, * 0xe8, 0x4c, 0x00, 0x28, ld r2,40(r12) */ - pr_debug("ip:%lx jumps to %lx r2: %lx", ip, tramp, mod->arch.toc); + pr_devel("ip:%lx jumps to %lx r2: %lx", ip, tramp, mod->arch.toc); /* Find where the trampoline jumps to */ if (probe_kernel_read(jmp, (void *)tramp, sizeof(jmp))) { @@ -165,7 +165,7 @@ __ftrace_make_nop(struct module *mod, return -EFAULT; } - pr_debug(" %08x %08x", jmp[0], jmp[1]); + pr_devel(" %08x %08x", jmp[0], jmp[1]); /* verify that this is what we expect it to be */ if (((jmp[0] & 0xffff0000) != 0x3d820000) || @@ -181,18 +181,18 @@ __ftrace_make_nop(struct module *mod, offset = ((unsigned)((unsigned short)jmp[0]) << 16) + (int)((short)jmp[1]); - pr_debug(" %x ", offset); + pr_devel(" %x ", offset); /* get the address this jumps too */ tramp = mod->arch.toc + offset + 32; - pr_debug("toc: %lx", tramp); + pr_devel("toc: %lx", tramp); if (probe_kernel_read(jmp, (void *)tramp, 8)) { printk(KERN_ERR "Failed to read %lx\n", tramp); return -EFAULT; } - pr_debug(" %08x %08x\n", jmp[0], jmp[1]); + pr_devel(" %08x %08x\n", jmp[0], jmp[1]); ptr = ((unsigned long)jmp[0] << 32) + jmp[1]; @@ -269,7 +269,7 @@ __ftrace_make_nop(struct module *mod, * 0x4e, 0x80, 0x04, 0x20 bctr */ - pr_debug("ip:%lx jumps to %lx", ip, tramp); + pr_devel("ip:%lx jumps to %lx", ip, tramp); /* Find where the trampoline jumps to */ if (probe_kernel_read(jmp, (void *)tramp, sizeof(jmp))) { @@ -277,7 +277,7 @@ __ftrace_make_nop(struct module *mod, return -EFAULT; } - pr_debug(" %08x %08x ", jmp[0], jmp[1]); + pr_devel(" %08x %08x ", jmp[0], jmp[1]); /* verify that this is what we expect it to be */ if (((jmp[0] & 0xffff0000) != 0x3d600000) || @@ -293,7 +293,7 @@ __ftrace_make_nop(struct module *mod, if (tramp & 0x8000) tramp -= 0x10000; - pr_debug(" %lx ", tramp); + pr_devel(" %lx ", tramp); if (tramp != addr) { printk(KERN_ERR @@ -402,7 +402,7 @@ __ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) /* ld r2,40(r1) */ op[1] = 0xe8410028; - pr_debug("write to %lx\n", rec->ip); + pr_devel("write to %lx\n", rec->ip); if (probe_kernel_write((void *)ip, op, MCOUNT_INSN_SIZE * 2)) return -EPERM; @@ -442,7 +442,7 @@ __ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) return -EINVAL; } - pr_debug("write to %lx\n", rec->ip); + pr_devel("write to %lx\n", rec->ip); if (probe_kernel_write((void *)ip, &op, MCOUNT_INSN_SIZE)) return -EPERM; From c3cf8667ed7db58c1960958cbb0a9098d513cc60 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 15 May 2009 04:33:54 +0000 Subject: [PATCH 087/267] powerpc/ftrace: Fix constraint to be early clobber After upgrading my distcc boxes from gcc 4.2.2 to 4.4.0, the function graph tracer broke. This was discovered on my x86 boxes. The issue is that gcc used the same register for an output as it did for an input in an asm statement. I first thought this was a bug in gcc and reported it. I was notified that gcc was correct and that the output had to be flagged as an "early clobber". I noticed that powerpc had the same issue and this patch fixes it. Signed-off-by: Steven Rostedt Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/ftrace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/powerpc/kernel/ftrace.c b/arch/powerpc/kernel/ftrace.c index 5b078ee391fc..2d182f119d1d 100644 --- a/arch/powerpc/kernel/ftrace.c +++ b/arch/powerpc/kernel/ftrace.c @@ -594,7 +594,7 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr) PPC_LONG "2b,4b\n" ".previous" - : [old] "=r" (old), [faulted] "=r" (faulted) + : [old] "=&r" (old), [faulted] "=r" (faulted) : [parent] "r" (parent), [return_hooker] "r" (return_hooker) : "memory" ); From dc892288f42661a140124ecbf9d44850a95de222 Mon Sep 17 00:00:00 2001 From: Geoff Levand Date: Fri, 15 May 2009 08:01:59 +0000 Subject: [PATCH 088/267] powerpc/ps3: Update ps3_defconfig Refresh and set these options: CONFIG_SYSFS_DEPRECATED_V2: y -> n CONFIG_INPUT_JOYSTICK: y -> n CONFIG_HID_SONY: n -> m CONFIG_RTC_DRV_PS3: - -> m Signed-off-by: Geoff Levand Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/configs/ps3_defconfig | 105 +++++++++++++++++------------ 1 file changed, 62 insertions(+), 43 deletions(-) diff --git a/arch/powerpc/configs/ps3_defconfig b/arch/powerpc/configs/ps3_defconfig index ac14f5245d2a..e28e65e7a0e1 100644 --- a/arch/powerpc/configs/ps3_defconfig +++ b/arch/powerpc/configs/ps3_defconfig @@ -1,13 +1,14 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.29-rc8 -# Fri Mar 13 09:28:45 2009 +# Linux kernel version: 2.6.30-rc5 +# Fri May 15 10:37:00 2009 # CONFIG_PPC64=y # # Processor support # +CONFIG_PPC_BOOK3S=y # CONFIG_POWER4_ONLY is not set CONFIG_POWER3=y CONFIG_POWER4=y @@ -55,9 +56,11 @@ CONFIG_OF=y # CONFIG_GENERIC_TBSYNC is not set CONFIG_AUDIT_ARCH=y CONFIG_GENERIC_BUG=y +CONFIG_DTC=y # CONFIG_DEFAULT_UIMAGE is not set # CONFIG_PPC_DCR_NATIVE is not set # CONFIG_PPC_DCR_MMIO is not set +CONFIG_ARCH_SUPPORTS_DEBUG_PAGEALLOC=y CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" # @@ -72,6 +75,7 @@ CONFIG_SWAP=y CONFIG_SYSVIPC=y CONFIG_SYSVIPC_SYSCTL=y CONFIG_POSIX_MQUEUE=y +CONFIG_POSIX_MQUEUE_SYSCTL=y # CONFIG_BSD_PROCESS_ACCT is not set # CONFIG_TASKSTATS is not set # CONFIG_AUDIT is not set @@ -88,8 +92,7 @@ CONFIG_CLASSIC_RCU=y CONFIG_LOG_BUF_SHIFT=17 # CONFIG_GROUP_SCHED is not set # CONFIG_CGROUPS is not set -CONFIG_SYSFS_DEPRECATED=y -CONFIG_SYSFS_DEPRECATED_V2=y +# CONFIG_SYSFS_DEPRECATED_V2 is not set # CONFIG_RELAY is not set CONFIG_NAMESPACES=y # CONFIG_UTS_NS is not set @@ -99,6 +102,9 @@ CONFIG_NAMESPACES=y # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y CONFIG_INITRAMFS_SOURCE="" +CONFIG_RD_GZIP=y +# CONFIG_RD_BZIP2 is not set +# CONFIG_RD_LZMA is not set CONFIG_CC_OPTIMIZE_FOR_SIZE=y CONFIG_SYSCTL=y CONFIG_ANON_INODES=y @@ -107,6 +113,7 @@ CONFIG_SYSCTL_SYSCALL=y CONFIG_KALLSYMS=y CONFIG_KALLSYMS_ALL=y CONFIG_KALLSYMS_EXTRA_PASS=y +# CONFIG_STRIP_ASM_SYMS is not set CONFIG_HOTPLUG=y CONFIG_PRINTK=y CONFIG_BUG=y @@ -138,6 +145,7 @@ CONFIG_HAVE_KRETPROBES=y CONFIG_HAVE_ARCH_TRACEHOOK=y CONFIG_HAVE_DMA_ATTRS=y CONFIG_USE_GENERIC_SMP_HELPERS=y +# CONFIG_SLOW_WORK is not set # CONFIG_HAVE_GENERIC_DMA_COHERENT is not set CONFIG_SLABINFO=y CONFIG_RT_MUTEXES=y @@ -150,7 +158,6 @@ CONFIG_MODULE_UNLOAD=y # CONFIG_MODULE_SRCVERSION_ALL is not set CONFIG_STOP_MACHINE=y CONFIG_BLOCK=y -# CONFIG_BLK_DEV_IO_TRACE is not set CONFIG_BLK_DEV_BSG=y # CONFIG_BLK_DEV_INTEGRITY is not set CONFIG_BLOCK_COMPAT=y @@ -172,7 +179,6 @@ CONFIG_DEFAULT_IOSCHED="anticipatory" # # Platform support # -CONFIG_PPC_MULTIPLATFORM=y # CONFIG_PPC_PSERIES is not set # CONFIG_PPC_ISERIES is not set # CONFIG_PPC_PMAC is not set @@ -209,6 +215,7 @@ CONFIG_SPU_FS_64K_LS=y # CONFIG_SPU_TRACE is not set CONFIG_SPU_BASE=y # CONFIG_PQ2ADS is not set +# CONFIG_PPC_OF_BOOT_TRAMPOLINE is not set # CONFIG_IPIC is not set # CONFIG_MPIC is not set # CONFIG_MPIC_WEIRD is not set @@ -279,11 +286,14 @@ CONFIG_PHYS_ADDR_T_64BIT=y CONFIG_ZONE_DMA_FLAG=1 CONFIG_BOUNCE=y CONFIG_UNEVICTABLE_LRU=y +CONFIG_HAVE_MLOCK=y +CONFIG_HAVE_MLOCKED_PAGE_BIT=y CONFIG_ARCH_MEMORY_PROBE=y CONFIG_PPC_HAS_HASH_64K=y CONFIG_PPC_4K_PAGES=y # CONFIG_PPC_16K_PAGES is not set # CONFIG_PPC_64K_PAGES is not set +# CONFIG_PPC_256K_PAGES is not set CONFIG_FORCE_MAX_ZONEORDER=13 CONFIG_SCHED_SMT=y CONFIG_PROC_DEVICETREE=y @@ -316,7 +326,6 @@ CONFIG_NET=y # # Networking options # -CONFIG_COMPAT_NET_DEV_OPS=y CONFIG_PACKET=y CONFIG_PACKET_MMAP=y CONFIG_UNIX=y @@ -389,6 +398,7 @@ CONFIG_IPV6_NDISC_NODETYPE=y # CONFIG_LAPB is not set # CONFIG_ECONET is not set # CONFIG_WAN_ROUTER is not set +# CONFIG_PHONET is not set # CONFIG_NET_SCHED is not set # CONFIG_DCB is not set @@ -396,6 +406,7 @@ CONFIG_IPV6_NDISC_NODETYPE=y # Network testing # # CONFIG_NET_PKTGEN is not set +# CONFIG_NET_DROP_MONITOR is not set # CONFIG_HAMRADIO is not set # CONFIG_CAN is not set # CONFIG_IRDA is not set @@ -419,11 +430,9 @@ CONFIG_BT_HCIBTUSB=m # CONFIG_BT_HCIBFUSB is not set # CONFIG_BT_HCIVHCI is not set # CONFIG_AF_RXRPC is not set -# CONFIG_PHONET is not set CONFIG_WIRELESS=y CONFIG_CFG80211=m # CONFIG_CFG80211_REG_DEBUG is not set -CONFIG_NL80211=y # CONFIG_WIRELESS_OLD_REGULATORY is not set CONFIG_WIRELESS_EXT=y # CONFIG_WIRELESS_EXT_SYSFS is not set @@ -602,6 +611,7 @@ CONFIG_SCSI_WAIT_SCAN=m # CONFIG_SCSI_SRP_ATTRS is not set # CONFIG_SCSI_LOWLEVEL is not set # CONFIG_SCSI_DH is not set +# CONFIG_SCSI_OSD_INITIATOR is not set # CONFIG_ATA is not set CONFIG_MD=y # CONFIG_BLK_DEV_MD is not set @@ -616,6 +626,7 @@ CONFIG_BLK_DEV_DM=m # CONFIG_DM_UEVENT is not set # CONFIG_MACINTOSH_DRIVERS is not set CONFIG_NETDEVICES=y +CONFIG_COMPAT_NET_DEV_OPS=y # CONFIG_DUMMY is not set # CONFIG_BONDING is not set # CONFIG_MACVLAN is not set @@ -625,6 +636,8 @@ CONFIG_NETDEVICES=y # CONFIG_PHYLIB is not set CONFIG_NET_ETHERNET=y CONFIG_MII=m +# CONFIG_ETHOC is not set +# CONFIG_DNET is not set # CONFIG_IBM_NEW_EMAC_ZMII is not set # CONFIG_IBM_NEW_EMAC_RGMII is not set # CONFIG_IBM_NEW_EMAC_TAH is not set @@ -646,12 +659,13 @@ CONFIG_GELIC_WIRELESS_OLD_PSK_INTERFACE=y CONFIG_WLAN_80211=y # CONFIG_LIBERTAS is not set # CONFIG_LIBERTAS_THINFIRM is not set +# CONFIG_AT76C50X_USB is not set # CONFIG_USB_ZD1201 is not set # CONFIG_USB_NET_RNDIS_WLAN is not set # CONFIG_RTL8187 is not set # CONFIG_MAC80211_HWSIM is not set # CONFIG_P54_COMMON is not set -# CONFIG_IWLWIFI_LEDS is not set +# CONFIG_AR9170_USB is not set # CONFIG_HOSTAP is not set # CONFIG_B43 is not set # CONFIG_B43LEGACY is not set @@ -673,6 +687,7 @@ CONFIG_USB_PEGASUS=m CONFIG_USB_USBNET=m CONFIG_USB_NET_AX8817X=m # CONFIG_USB_NET_CDCETHER is not set +# CONFIG_USB_NET_CDC_EEM is not set # CONFIG_USB_NET_DM9601 is not set # CONFIG_USB_NET_SMSC95XX is not set # CONFIG_USB_NET_GL620A is not set @@ -724,28 +739,7 @@ CONFIG_INPUT_EVDEV=m # # CONFIG_INPUT_KEYBOARD is not set # CONFIG_INPUT_MOUSE is not set -CONFIG_INPUT_JOYSTICK=y -# CONFIG_JOYSTICK_ANALOG is not set -# CONFIG_JOYSTICK_A3D is not set -# CONFIG_JOYSTICK_ADI is not set -# CONFIG_JOYSTICK_COBRA is not set -# CONFIG_JOYSTICK_GF2K is not set -# CONFIG_JOYSTICK_GRIP is not set -# CONFIG_JOYSTICK_GRIP_MP is not set -# CONFIG_JOYSTICK_GUILLEMOT is not set -# CONFIG_JOYSTICK_INTERACT is not set -# CONFIG_JOYSTICK_SIDEWINDER is not set -# CONFIG_JOYSTICK_TMDC is not set -# CONFIG_JOYSTICK_IFORCE is not set -# CONFIG_JOYSTICK_WARRIOR is not set -# CONFIG_JOYSTICK_MAGELLAN is not set -# CONFIG_JOYSTICK_SPACEORB is not set -# CONFIG_JOYSTICK_SPACEBALL is not set -# CONFIG_JOYSTICK_STINGER is not set -# CONFIG_JOYSTICK_TWIDJOY is not set -# CONFIG_JOYSTICK_ZHENHUA is not set -# CONFIG_JOYSTICK_JOYDUMP is not set -# CONFIG_JOYSTICK_XPAD is not set +# CONFIG_INPUT_JOYSTICK is not set # CONFIG_INPUT_TABLET is not set # CONFIG_INPUT_TOUCHSCREEN is not set # CONFIG_INPUT_MISC is not set @@ -864,6 +858,7 @@ CONFIG_FB_PS3_DEFAULT_SIZE_M=9 # CONFIG_FB_VIRTUAL is not set # CONFIG_FB_METRONOME is not set # CONFIG_FB_MB862XX is not set +# CONFIG_FB_BROADSHEET is not set # CONFIG_BACKLIGHT_LCD_SUPPORT is not set # @@ -934,15 +929,17 @@ CONFIG_USB_HIDDEV=y # # Special HID drivers # -# CONFIG_HID_COMPAT is not set # CONFIG_HID_A4TECH is not set # CONFIG_HID_APPLE is not set # CONFIG_HID_BELKIN is not set # CONFIG_HID_CHERRY is not set # CONFIG_HID_CHICONY is not set # CONFIG_HID_CYPRESS is not set +# CONFIG_DRAGONRISE_FF is not set # CONFIG_HID_EZKEY is not set +# CONFIG_HID_KYE is not set # CONFIG_HID_GYRATION is not set +# CONFIG_HID_KENSINGTON is not set # CONFIG_HID_LOGITECH is not set # CONFIG_HID_MICROSOFT is not set # CONFIG_HID_MONTEREY is not set @@ -950,7 +947,7 @@ CONFIG_USB_HIDDEV=y # CONFIG_HID_PANTHERLORD is not set # CONFIG_HID_PETALYNX is not set # CONFIG_HID_SAMSUNG is not set -# CONFIG_HID_SONY is not set +CONFIG_HID_SONY=m # CONFIG_HID_SUNPLUS is not set # CONFIG_GREENASIA_FF is not set # CONFIG_HID_TOPSEED is not set @@ -1012,11 +1009,11 @@ CONFIG_USB_OHCI_LITTLE_ENDIAN=y # CONFIG_USB_TMC is not set # -# NOTE: USB_STORAGE depends on SCSI but BLK_DEV_SD may also be needed; +# NOTE: USB_STORAGE depends on SCSI but BLK_DEV_SD may # # -# see USB_STORAGE Help for more information +# also be needed; see USB_STORAGE Help for more info # CONFIG_USB_STORAGE=m # CONFIG_USB_STORAGE_DEBUG is not set @@ -1058,7 +1055,6 @@ CONFIG_USB_STORAGE=m # CONFIG_USB_LED is not set # CONFIG_USB_CYPRESS_CY7C63 is not set # CONFIG_USB_CYTHERM is not set -# CONFIG_USB_PHIDGET is not set # CONFIG_USB_IDMOUSE is not set # CONFIG_USB_FTDI_ELAN is not set # CONFIG_USB_APPLEDISPLAY is not set @@ -1074,6 +1070,7 @@ CONFIG_USB_STORAGE=m # # OTG and related infrastructure # +# CONFIG_NOP_USB_XCEIV is not set # CONFIG_MMC is not set # CONFIG_MEMSTICK is not set # CONFIG_NEW_LEDS is not set @@ -1113,8 +1110,10 @@ CONFIG_RTC_INTF_DEV=y # # on-CPU RTC drivers # -CONFIG_RTC_DRV_PPC=m +# CONFIG_RTC_DRV_GENERIC is not set +CONFIG_RTC_DRV_PS3=m # CONFIG_DMADEVICES is not set +# CONFIG_AUXDISPLAY is not set # CONFIG_UIO is not set # CONFIG_STAGING is not set @@ -1125,6 +1124,7 @@ CONFIG_EXT2_FS=m # CONFIG_EXT2_FS_XATTR is not set # CONFIG_EXT2_FS_XIP is not set CONFIG_EXT3_FS=m +# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set CONFIG_EXT3_FS_XATTR=y # CONFIG_EXT3_FS_POSIX_ACL is not set # CONFIG_EXT3_FS_SECURITY is not set @@ -1160,6 +1160,11 @@ CONFIG_AUTOFS_FS=m CONFIG_AUTOFS4_FS=m # CONFIG_FUSE_FS is not set +# +# Caches +# +# CONFIG_FSCACHE is not set + # # CD-ROM/DVD Filesystems # @@ -1211,6 +1216,7 @@ CONFIG_MISC_FILESYSTEMS=y # CONFIG_ROMFS_FS is not set # CONFIG_SYSV_FS is not set # CONFIG_UFS_FS is not set +# CONFIG_NILFS2_FS is not set CONFIG_NETWORK_FILESYSTEMS=y CONFIG_NFS_FS=y CONFIG_NFS_V3=y @@ -1223,7 +1229,6 @@ CONFIG_LOCKD_V4=y CONFIG_NFS_COMMON=y CONFIG_SUNRPC=y CONFIG_SUNRPC_GSS=y -# CONFIG_SUNRPC_REGISTER_V4 is not set CONFIG_RPCSEC_GSS_KRB5=y # CONFIG_RPCSEC_GSS_SPKM3 is not set # CONFIG_SMB_FS is not set @@ -1283,6 +1288,7 @@ CONFIG_NLS_ISO8859_1=y # CONFIG_NLS_KOI8_U is not set # CONFIG_NLS_UTF8 is not set # CONFIG_DLM is not set +CONFIG_BINARY_PRINTF=y # # Library routines @@ -1296,15 +1302,16 @@ CONFIG_CRC_ITU_T=m CONFIG_CRC32=y # CONFIG_CRC7 is not set # CONFIG_LIBCRC32C is not set -CONFIG_ZLIB_INFLATE=m +CONFIG_ZLIB_INFLATE=y CONFIG_ZLIB_DEFLATE=m CONFIG_LZO_COMPRESS=m CONFIG_LZO_DECOMPRESS=m -CONFIG_PLIST=y +CONFIG_DECOMPRESS_GZIP=y CONFIG_HAS_IOMEM=y CONFIG_HAS_IOPORT=y CONFIG_HAS_DMA=y CONFIG_HAVE_LMB=y +CONFIG_NLATTR=y # # Kernel hacking @@ -1322,6 +1329,9 @@ CONFIG_DEBUG_KERNEL=y CONFIG_DETECT_SOFTLOCKUP=y # CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC is not set CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC_VALUE=0 +CONFIG_DETECT_HUNG_TASK=y +# CONFIG_BOOTPARAM_HUNG_TASK_PANIC is not set +CONFIG_BOOTPARAM_HUNG_TASK_PANIC_VALUE=0 CONFIG_SCHED_DEBUG=y # CONFIG_SCHEDSTATS is not set # CONFIG_TIMER_STATS is not set @@ -1357,12 +1367,15 @@ CONFIG_DEBUG_LIST=y # CONFIG_FAULT_INJECTION is not set # CONFIG_LATENCYTOP is not set CONFIG_SYSCTL_SYSCALL_CHECK=y +# CONFIG_DEBUG_PAGEALLOC is not set CONFIG_NOP_TRACER=y CONFIG_HAVE_FUNCTION_TRACER=y +CONFIG_HAVE_FUNCTION_GRAPH_TRACER=y CONFIG_HAVE_DYNAMIC_FTRACE=y CONFIG_HAVE_FTRACE_MCOUNT_RECORD=y CONFIG_RING_BUFFER=y CONFIG_TRACING=y +CONFIG_TRACING_SUPPORT=y # # Tracers @@ -1371,18 +1384,21 @@ CONFIG_TRACING=y # CONFIG_IRQSOFF_TRACER is not set # CONFIG_SCHED_TRACER is not set # CONFIG_CONTEXT_SWITCH_TRACER is not set +# CONFIG_EVENT_TRACER is not set # CONFIG_BOOT_TRACER is not set # CONFIG_TRACE_BRANCH_PROFILING is not set # CONFIG_STACK_TRACER is not set +# CONFIG_KMEMTRACE is not set +# CONFIG_WORKQUEUE_TRACER is not set +# CONFIG_BLK_DEV_IO_TRACE is not set # CONFIG_FTRACE_STARTUP_TEST is not set -# CONFIG_DYNAMIC_PRINTK_DEBUG is not set +# CONFIG_DYNAMIC_DEBUG is not set # CONFIG_SAMPLES is not set CONFIG_HAVE_ARCH_KGDB=y # CONFIG_KGDB is not set CONFIG_PRINT_STACK_DEPTH=64 CONFIG_DEBUG_STACKOVERFLOW=y # CONFIG_DEBUG_STACK_USAGE is not set -# CONFIG_DEBUG_PAGEALLOC is not set # CONFIG_CODE_PATCHING_SELFTEST is not set # CONFIG_FTR_FIXUP_SELFTEST is not set # CONFIG_MSI_BITMAP_SELFTEST is not set @@ -1415,10 +1431,12 @@ CONFIG_CRYPTO_HASH=y CONFIG_CRYPTO_HASH2=y CONFIG_CRYPTO_RNG=m CONFIG_CRYPTO_RNG2=y +CONFIG_CRYPTO_PCOMP=y CONFIG_CRYPTO_MANAGER=y CONFIG_CRYPTO_MANAGER2=y CONFIG_CRYPTO_GF128MUL=m # CONFIG_CRYPTO_NULL is not set +CONFIG_CRYPTO_WORKQUEUE=y # CONFIG_CRYPTO_CRYPTD is not set # CONFIG_CRYPTO_AUTHENC is not set # CONFIG_CRYPTO_TEST is not set @@ -1487,6 +1505,7 @@ CONFIG_CRYPTO_SALSA20=m # Compression # # CONFIG_CRYPTO_DEFLATE is not set +# CONFIG_CRYPTO_ZLIB is not set CONFIG_CRYPTO_LZO=m # From 0e337b42d620ca7c45fe64e64dd71957c56216c9 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sun, 17 May 2009 18:29:03 +0000 Subject: [PATCH 089/267] powerpc: Explicit alignment for .data.cacheline_aligned I don't think anything guarantees that the objects in data.page_aligned are a multiple of PAGE_SIZE, thus the section may end on any boundary. So the following section, .data.cacheline_aligned needs an explicit alignment. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/vmlinux.lds.S | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S index a047a6cfca4d..8ef8a14abc95 100644 --- a/arch/powerpc/kernel/vmlinux.lds.S +++ b/arch/powerpc/kernel/vmlinux.lds.S @@ -264,6 +264,7 @@ SECTIONS *(.data.page_aligned) } + . = ALIGN(L1_CACHE_BYTES); .data.cacheline_aligned : AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET) { *(.data.cacheline_aligned) } From 5b4662f098b47f68d7fcea9b065d1513547fef12 Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Fri, 15 May 2009 14:50:33 -0400 Subject: [PATCH 090/267] regulator: da903x: add missing __devexit_p() The remove function uses __devexit, so the .remove assignment needs __devexit_p() to fix a build error with hotplug disabled. Signed-off-by: Mike Frysinger CC: Liam Girdwood CC: Mike Rapoport CC: Eric Miao Acked-by: Eric Miao Signed-off-by: Liam Girdwood --- drivers/regulator/da903x.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/regulator/da903x.c b/drivers/regulator/da903x.c index 72b15495183c..c6628f5a0af7 100644 --- a/drivers/regulator/da903x.c +++ b/drivers/regulator/da903x.c @@ -497,7 +497,7 @@ static struct platform_driver da903x_regulator_driver = { .owner = THIS_MODULE, }, .probe = da903x_regulator_probe, - .remove = da903x_regulator_remove, + .remove = __devexit_p(da903x_regulator_remove), }; static int __init da903x_regulator_init(void) From eb33575cf67d3f35fa2510210ef92631266e2465 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 13 May 2009 17:34:48 +0100 Subject: [PATCH 091/267] [ARM] Double check memmap is actually valid with a memmap has unexpected holes V2 pfn_valid() is meant to be able to tell if a given PFN has valid memmap associated with it or not. In FLATMEM, it is expected that holes always have valid memmap as long as there is valid PFNs either side of the hole. In SPARSEMEM, it is assumed that a valid section has a memmap for the entire section. However, ARM and maybe other embedded architectures in the future free memmap backing holes to save memory on the assumption the memmap is never used. The page_zone linkages are then broken even though pfn_valid() returns true. A walker of the full memmap must then do this additional check to ensure the memmap they are looking at is sane by making sure the zone and PFN linkages are still valid. This is expensive, but walkers of the full memmap are extremely rare. This was caught before for FLATMEM and hacked around but it hits again for SPARSEMEM because the page_zone linkages can look ok where the PFN linkages are totally screwed. This looks like a hatchet job but the reality is that any clean solution would end up consumning all the memory saved by punching these unexpected holes in the memmap. For example, we tried marking the memmap within the section invalid but the section size exceeds the size of the hole in most cases so pfn_valid() starts returning false where valid memmap exists. Shrinking the size of the section would increase memory consumption offsetting the gains. This patch identifies when an architecture is punching unexpected holes in the memmap that the memory model cannot automatically detect and sets ARCH_HAS_HOLES_MEMORYMODEL. At the moment, this is restricted to EP93xx which is the model sub-architecture this has been reported on but may expand later. When set, walkers of the full memmap must call memmap_valid_within() for each PFN and passing in what it expects the page and zone to be for that PFN. If it finds the linkages to be broken, it assumes the memmap is invalid for that PFN. Signed-off-by: Mel Gorman Signed-off-by: Russell King --- arch/arm/Kconfig | 6 +++--- include/linux/mmzone.h | 26 ++++++++++++++++++++++++++ mm/mmzone.c | 15 +++++++++++++++ mm/vmstat.c | 19 ++++--------------- 4 files changed, 48 insertions(+), 18 deletions(-) diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index e60ec54df334..9d02cdb15b23 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -273,6 +273,7 @@ config ARCH_EP93XX select HAVE_CLK select COMMON_CLKDEV select ARCH_REQUIRE_GPIOLIB + select ARCH_HAS_HOLES_MEMORYMODEL help This enables support for the Cirrus EP93xx series of CPUs. @@ -976,10 +977,9 @@ config OABI_COMPAT UNPREDICTABLE (in fact it can be predicted that it won't work at all). If in doubt say Y. -config ARCH_FLATMEM_HAS_HOLES +config ARCH_HAS_HOLES_MEMORYMODEL bool - default y - depends on FLATMEM + default n # Discontigmem is deprecated config ARCH_DISCONTIGMEM_ENABLE diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 186ec6ab334d..a47c879e1304 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -1097,6 +1097,32 @@ unsigned long __init node_memmap_size_bytes(int, unsigned long, unsigned long); #define pfn_valid_within(pfn) (1) #endif +#ifdef CONFIG_ARCH_HAS_HOLES_MEMORYMODEL +/* + * pfn_valid() is meant to be able to tell if a given PFN has valid memmap + * associated with it or not. In FLATMEM, it is expected that holes always + * have valid memmap as long as there is valid PFNs either side of the hole. + * In SPARSEMEM, it is assumed that a valid section has a memmap for the + * entire section. + * + * However, an ARM, and maybe other embedded architectures in the future + * free memmap backing holes to save memory on the assumption the memmap is + * never used. The page_zone linkages are then broken even though pfn_valid() + * returns true. A walker of the full memmap must then do this additional + * check to ensure the memmap they are looking at is sane by making sure + * the zone and PFN linkages are still valid. This is expensive, but walkers + * of the full memmap are extremely rare. + */ +int memmap_valid_within(unsigned long pfn, + struct page *page, struct zone *zone); +#else +static inline int memmap_valid_within(unsigned long pfn, + struct page *page, struct zone *zone) +{ + return 1; +} +#endif /* CONFIG_ARCH_HAS_HOLES_MEMORYMODEL */ + #endif /* !__GENERATING_BOUNDS.H */ #endif /* !__ASSEMBLY__ */ #endif /* _LINUX_MMZONE_H */ diff --git a/mm/mmzone.c b/mm/mmzone.c index 16ce8b955dcf..f5b7d1760213 100644 --- a/mm/mmzone.c +++ b/mm/mmzone.c @@ -6,6 +6,7 @@ #include +#include #include #include @@ -72,3 +73,17 @@ struct zoneref *next_zones_zonelist(struct zoneref *z, *zone = zonelist_zone(z); return z; } + +#ifdef CONFIG_ARCH_HAS_HOLES_MEMORYMODEL +int memmap_valid_within(unsigned long pfn, + struct page *page, struct zone *zone) +{ + if (page_to_pfn(page) != pfn) + return 0; + + if (page_zone(page) != zone) + return 0; + + return 1; +} +#endif /* CONFIG_ARCH_HAS_HOLES_MEMORYMODEL */ diff --git a/mm/vmstat.c b/mm/vmstat.c index 66f6130976cb..74d66dba0cbe 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -509,22 +509,11 @@ static void pagetypeinfo_showblockcount_print(struct seq_file *m, continue; page = pfn_to_page(pfn); -#ifdef CONFIG_ARCH_FLATMEM_HAS_HOLES - /* - * Ordinarily, memory holes in flatmem still have a valid - * memmap for the PFN range. However, an architecture for - * embedded systems (e.g. ARM) can free up the memmap backing - * holes to save memory on the assumption the memmap is - * never used. The page_zone linkages are then broken even - * though pfn_valid() returns true. Skip the page if the - * linkages are broken. Even if this test passed, the impact - * is that the counters for the movable type are off but - * fragmentation monitoring is likely meaningless on small - * systems. - */ - if (page_zone(page) != zone) + + /* Watch for unexpected holes punched in the memmap */ + if (!memmap_valid_within(pfn, page, zone)) continue; -#endif + mtype = get_pageblock_migratetype(page); if (mtype < MIGRATE_TYPES) From 8190b37f6483fedc530110d8c7d710d1c0317ba4 Mon Sep 17 00:00:00 2001 From: Pavel Roskin Date: Tue, 12 May 2009 13:36:46 -0700 Subject: [PATCH 092/267] [ARM] mach-l7200: fix spelling of SYS_CLOCK_OFF Signed-off-by: Pavel Roskin Signed-off-by: Andrew Morton Signed-off-by: Russell King --- arch/arm/mach-l7200/include/mach/sys-clock.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-l7200/include/mach/sys-clock.h b/arch/arm/mach-l7200/include/mach/sys-clock.h index 2d7722be60ea..e9729a35751d 100644 --- a/arch/arm/mach-l7200/include/mach/sys-clock.h +++ b/arch/arm/mach-l7200/include/mach/sys-clock.h @@ -18,7 +18,7 @@ /* IO_START and IO_BASE are defined in hardware.h */ -#define SYS_CLOCK_START (IO_START + SYS_CLCOK_OFF) /* Physical address */ +#define SYS_CLOCK_START (IO_START + SYS_CLOCK_OFF) /* Physical address */ #define SYS_CLOCK_BASE (IO_BASE + SYS_CLOCK_OFF) /* Virtual address */ /* Define the interface to the SYS_CLOCK */ From 3026589c1bc17b0c389b95afec127e92e2a745e2 Mon Sep 17 00:00:00 2001 From: Michal Simek Date: Mon, 11 May 2009 09:24:47 +0200 Subject: [PATCH 093/267] microblaze: Update Microblaze defconfig Signed-off-by: Michal Simek --- arch/microblaze/configs/nommu_defconfig | 50 ++++++++++++++++--------- 1 file changed, 32 insertions(+), 18 deletions(-) diff --git a/arch/microblaze/configs/nommu_defconfig b/arch/microblaze/configs/nommu_defconfig index beb7ecd72793..4ef6af0a8f31 100644 --- a/arch/microblaze/configs/nommu_defconfig +++ b/arch/microblaze/configs/nommu_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.29 -# Tue Mar 24 10:23:20 2009 +# Linux kernel version: 2.6.30-rc5 +# Mon May 11 09:01:02 2009 # CONFIG_MICROBLAZE=y # CONFIG_SWAP is not set @@ -32,6 +32,7 @@ CONFIG_LOCALVERSION_AUTO=y CONFIG_SYSVIPC=y CONFIG_SYSVIPC_SYSCTL=y CONFIG_POSIX_MQUEUE=y +CONFIG_POSIX_MQUEUE_SYSCTL=y CONFIG_BSD_PROCESS_ACCT=y CONFIG_BSD_PROCESS_ACCT_V3=y # CONFIG_TASKSTATS is not set @@ -63,6 +64,7 @@ CONFIG_SYSCTL_SYSCALL=y CONFIG_KALLSYMS=y CONFIG_KALLSYMS_ALL=y CONFIG_KALLSYMS_EXTRA_PASS=y +# CONFIG_STRIP_ASM_SYMS is not set # CONFIG_HOTPLUG is not set CONFIG_PRINTK=y CONFIG_BUG=y @@ -80,6 +82,8 @@ CONFIG_SLAB=y # CONFIG_SLUB is not set # CONFIG_SLOB is not set # CONFIG_PROFILING is not set +# CONFIG_MARKERS is not set +# CONFIG_SLOW_WORK is not set # CONFIG_HAVE_GENERIC_DMA_COHERENT is not set CONFIG_SLABINFO=y CONFIG_RT_MUTEXES=y @@ -92,7 +96,6 @@ CONFIG_MODULE_UNLOAD=y # CONFIG_MODULE_SRCVERSION_ALL is not set CONFIG_BLOCK=y # CONFIG_LBD is not set -# CONFIG_BLK_DEV_IO_TRACE is not set # CONFIG_BLK_DEV_BSG is not set # CONFIG_BLK_DEV_INTEGRITY is not set @@ -166,6 +169,8 @@ CONFIG_SPLIT_PTLOCK_CPUS=4 # CONFIG_PHYS_ADDR_T_64BIT is not set CONFIG_ZONE_DMA_FLAG=0 CONFIG_VIRT_TO_BUS=y +CONFIG_UNEVICTABLE_LRU=y +CONFIG_NOMMU_INITIAL_TRIM_EXCESS=1 # # Exectuable file formats @@ -180,7 +185,6 @@ CONFIG_NET=y # # Networking options # -CONFIG_COMPAT_NET_DEV_OPS=y CONFIG_PACKET=y # CONFIG_PACKET_MMAP is not set CONFIG_UNIX=y @@ -232,6 +236,7 @@ CONFIG_DEFAULT_TCP_CONG="cubic" # CONFIG_LAPB is not set # CONFIG_ECONET is not set # CONFIG_WAN_ROUTER is not set +# CONFIG_PHONET is not set # CONFIG_NET_SCHED is not set # CONFIG_DCB is not set @@ -244,7 +249,6 @@ CONFIG_DEFAULT_TCP_CONG="cubic" # CONFIG_IRDA is not set # CONFIG_BT is not set # CONFIG_AF_RXRPC is not set -# CONFIG_PHONET is not set CONFIG_WIRELESS=y # CONFIG_CFG80211 is not set CONFIG_WIRELESS_OLD_REGULATORY=y @@ -379,6 +383,7 @@ CONFIG_MISC_DEVICES=y # CONFIG_ATA is not set # CONFIG_MD is not set CONFIG_NETDEVICES=y +CONFIG_COMPAT_NET_DEV_OPS=y # CONFIG_DUMMY is not set # CONFIG_BONDING is not set # CONFIG_MACVLAN is not set @@ -388,6 +393,7 @@ CONFIG_NETDEVICES=y # CONFIG_PHYLIB is not set CONFIG_NET_ETHERNET=y # CONFIG_MII is not set +# CONFIG_ETHOC is not set # CONFIG_DNET is not set # CONFIG_IBM_NEW_EMAC_ZMII is not set # CONFIG_IBM_NEW_EMAC_RGMII is not set @@ -405,7 +411,6 @@ CONFIG_NETDEV_10000=y # # CONFIG_WLAN_PRE80211 is not set # CONFIG_WLAN_80211 is not set -# CONFIG_IWLWIFI_LEDS is not set # # Enable WiMAX (Networking options) to see the WiMAX drivers @@ -455,6 +460,7 @@ CONFIG_LEGACY_PTYS=y CONFIG_LEGACY_PTY_COUNT=256 # CONFIG_IPMI_HANDLER is not set CONFIG_HW_RANDOM=y +# CONFIG_HW_RANDOM_TIMERIOMEM is not set # CONFIG_RTC is not set # CONFIG_GEN_RTC is not set # CONFIG_R3964 is not set @@ -525,7 +531,7 @@ CONFIG_USB_SUPPORT=y # # -# NOTE: USB_STORAGE depends on SCSI but BLK_DEV_SD may also be needed; +# NOTE: USB_STORAGE depends on SCSI but BLK_DEV_SD may # # CONFIG_USB_GADGET is not set @@ -538,6 +544,7 @@ CONFIG_USB_SUPPORT=y # CONFIG_ACCESSIBILITY is not set # CONFIG_RTC_CLASS is not set # CONFIG_DMADEVICES is not set +# CONFIG_AUXDISPLAY is not set # CONFIG_UIO is not set # CONFIG_STAGING is not set @@ -562,6 +569,11 @@ CONFIG_FILE_LOCKING=y # CONFIG_AUTOFS4_FS is not set # CONFIG_FUSE_FS is not set +# +# Caches +# +# CONFIG_FSCACHE is not set + # # CD-ROM/DVD Filesystems # @@ -601,8 +613,13 @@ CONFIG_CRAMFS=y # CONFIG_HPFS_FS is not set # CONFIG_QNX4FS_FS is not set CONFIG_ROMFS_FS=y +CONFIG_ROMFS_BACKED_BY_BLOCK=y +# CONFIG_ROMFS_BACKED_BY_MTD is not set +# CONFIG_ROMFS_BACKED_BY_BOTH is not set +CONFIG_ROMFS_ON_BLOCK=y # CONFIG_SYSV_FS is not set # CONFIG_UFS_FS is not set +# CONFIG_NILFS2_FS is not set CONFIG_NETWORK_FILESYSTEMS=y CONFIG_NFS_FS=y CONFIG_NFS_V3=y @@ -614,7 +631,6 @@ CONFIG_LOCKD_V4=y CONFIG_NFS_ACL_SUPPORT=y CONFIG_NFS_COMMON=y CONFIG_SUNRPC=y -# CONFIG_SUNRPC_REGISTER_V4 is not set # CONFIG_RPCSEC_GSS_KRB5 is not set # CONFIG_RPCSEC_GSS_SPKM3 is not set # CONFIG_SMB_FS is not set @@ -647,6 +663,9 @@ CONFIG_DEBUG_SHIRQ=y CONFIG_DETECT_SOFTLOCKUP=y CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC=y CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC_VALUE=1 +CONFIG_DETECT_HUNG_TASK=y +# CONFIG_BOOTPARAM_HUNG_TASK_PANIC is not set +CONFIG_BOOTPARAM_HUNG_TASK_PANIC_VALUE=0 CONFIG_SCHED_DEBUG=y CONFIG_SCHEDSTATS=y CONFIG_TIMER_STATS=y @@ -678,15 +697,8 @@ CONFIG_DEBUG_SG=y # CONFIG_DEBUG_BLOCK_EXT_DEVT is not set # CONFIG_FAULT_INJECTION is not set CONFIG_SYSCTL_SYSCALL_CHECK=y - -# -# Tracers -# -# CONFIG_SCHED_TRACER is not set -# CONFIG_CONTEXT_SWITCH_TRACER is not set -# CONFIG_BOOT_TRACER is not set -# CONFIG_TRACE_BRANCH_PROFILING is not set -# CONFIG_DYNAMIC_PRINTK_DEBUG is not set +# CONFIG_PAGE_POISONING is not set +# CONFIG_DYNAMIC_DEBUG is not set # CONFIG_SAMPLES is not set CONFIG_EARLY_PRINTK=y CONFIG_HEART_BEAT=y @@ -777,6 +789,7 @@ CONFIG_CRYPTO=y # Compression # # CONFIG_CRYPTO_DEFLATE is not set +# CONFIG_CRYPTO_ZLIB is not set # CONFIG_CRYPTO_LZO is not set # @@ -784,6 +797,7 @@ CONFIG_CRYPTO=y # # CONFIG_CRYPTO_ANSI_CPRNG is not set CONFIG_CRYPTO_HW=y +# CONFIG_BINARY_PRINTF is not set # # Library routines @@ -797,8 +811,8 @@ CONFIG_GENERIC_FIND_LAST_BIT=y # CONFIG_CRC7 is not set # CONFIG_LIBCRC32C is not set CONFIG_ZLIB_INFLATE=y -CONFIG_PLIST=y CONFIG_HAS_IOMEM=y CONFIG_HAS_IOPORT=y CONFIG_HAS_DMA=y CONFIG_HAVE_LMB=y +CONFIG_NLATTR=y From 7b7210d7a9d260becef4dba98b9075e2c9b41f93 Mon Sep 17 00:00:00 2001 From: Michal Simek Date: Thu, 14 May 2009 13:35:52 +0200 Subject: [PATCH 094/267] microblaze: Fix kind-of-intr checking against number of interrupts + Fix typographic fault. Signed-off-by: Michal Simek --- arch/microblaze/kernel/intc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/microblaze/kernel/intc.c b/arch/microblaze/kernel/intc.c index a69d3e3c2fd4..b15605299a57 100644 --- a/arch/microblaze/kernel/intc.c +++ b/arch/microblaze/kernel/intc.c @@ -137,8 +137,8 @@ void __init init_IRQ(void) intr_type = *(int *) of_get_property(intc, "xlnx,kind-of-intr", NULL); - if (intr_type >= (1 << nr_irq)) - printk(KERN_INFO " ERROR: Mishmash in king-of-intr param\n"); + if (intr_type >= (1 << (nr_irq + 1))) + printk(KERN_INFO " ERROR: Mismatch in kind-of-intr param\n"); #ifdef CONFIG_SELFMOD_INTC selfmod_function((int *) arr_func, intc_baseaddr); From bac9caf016bf147af7d3afbe7580a7f773cb1566 Mon Sep 17 00:00:00 2001 From: Roel Kluin Date: Sun, 17 May 2009 18:18:58 -0700 Subject: [PATCH 095/267] asm-generic: fix local_add_unless macro `local_add_unless(x, y, z)' will be expanded to `(&(x)->y, (y), (x))', but `&(x)->y' should be `&(x)->a' Signed-off-by: Roel Kluin Cc: Mathieu Desnoyers Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-generic/local.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/asm-generic/local.h b/include/asm-generic/local.h index dbd6150763e9..fc218444e315 100644 --- a/include/asm-generic/local.h +++ b/include/asm-generic/local.h @@ -42,7 +42,7 @@ typedef struct #define local_cmpxchg(l, o, n) atomic_long_cmpxchg((&(l)->a), (o), (n)) #define local_xchg(l, n) atomic_long_xchg((&(l)->a), (n)) -#define local_add_unless(l, a, u) atomic_long_add_unless((&(l)->a), (a), (u)) +#define local_add_unless(l, _a, u) atomic_long_add_unless((&(l)->a), (_a), (u)) #define local_inc_not_zero(l) atomic_long_inc_not_zero(&(l)->a) /* Non-atomic variants, ie. preemption disabled and won't be touched From dbf8c11f821b6ff83302c34f2403b4f7231f50ae Mon Sep 17 00:00:00 2001 From: Peter Korsgaard Date: Mon, 18 May 2009 11:13:54 +0100 Subject: [PATCH 096/267] mtd_dataflash: unbreak erase support Commit 5b7f3a50 (fix dataflash 64-bit divisions) unfortunately introduced a typo. Erase addr and len were swapped in the pageaddr calculation, causing the wrong sectors to get erased. Signed-off-by: Peter Korsgaard Acked-by: Artem Bityutskiy Signed-off-by: David Woodhouse Signed-off-by: Linus Torvalds --- drivers/mtd/devices/mtd_dataflash.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mtd/devices/mtd_dataflash.c b/drivers/mtd/devices/mtd_dataflash.c index 62dee54af0a5..43976aa4dbb1 100644 --- a/drivers/mtd/devices/mtd_dataflash.c +++ b/drivers/mtd/devices/mtd_dataflash.c @@ -178,7 +178,7 @@ static int dataflash_erase(struct mtd_info *mtd, struct erase_info *instr) /* Calculate flash page address; use block erase (for speed) if * we're at a block boundary and need to erase the whole block. */ - pageaddr = div_u64(instr->len, priv->page_size); + pageaddr = div_u64(instr->addr, priv->page_size); do_block = (pageaddr & 0x7) == 0 && instr->len >= blocksize; pageaddr = pageaddr << priv->page_offset; From 80193195f87ebca6d7417516d6edeb3969631c15 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Mon, 18 May 2009 13:07:35 +0100 Subject: [PATCH 097/267] Fix oops on close of hot-unplugged FTDI serial converter Commit c45d6320 ("fix reference counting of ftdi_private") stopped ftdi_sio_port_remove() from directly freeing the port-private data, with the intention if the port was still open, it would be freed when ftdi_close() is eventually called and releases the last refcount on the structure. That's all very well, but ftdi_sio_port_remove() still contains a call to usb_set_serial_port_data(port, NULL) -- so by the time we get to ftdi_close() for the port which was unplugged, it _still_ oopses on dereferencing that NULL pointer, as it did before (and does in 2.6.29). The fix is just not to clear the private data in ftdi_sio_port_remove(). Then the refcount is properly reduced to zero when the final kref_put() happens in ftdi_close(). Remove a bogus comment too, while we're at it. And stop doing things inside "if (priv)" -- it must _always_ be there. Based loosely on an earlier patch by Daniel Mack, and suggestions by Alan Stern. Signed-off-by: David Woodhouse Tested-by: Daniel Mack Cc: stable@kernel.org Signed-off-by: Linus Torvalds --- drivers/usb/serial/ftdi_sio.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c index 0ab8474b00cb..d9fcdaedf389 100644 --- a/drivers/usb/serial/ftdi_sio.c +++ b/drivers/usb/serial/ftdi_sio.c @@ -1487,14 +1487,7 @@ static int ftdi_sio_port_remove(struct usb_serial_port *port) remove_sysfs_attrs(port); - /* all open ports are closed at this point - * (by usbserial.c:__serial_close, which calls ftdi_close) - */ - - if (priv) { - usb_set_serial_port_data(port, NULL); - kref_put(&priv->kref, ftdi_sio_priv_release); - } + kref_put(&priv->kref, ftdi_sio_priv_release); return 0; } From 982db66352d31892f624390cfb64a1cea5df765a Mon Sep 17 00:00:00 2001 From: Rabin Vincent Date: Mon, 18 May 2009 17:29:30 +0100 Subject: [PATCH 098/267] [ARM] 5518/1: versatile: don't put clock lookups in __initdata Remove the __initdata annotation for the clock lookups, since they will be needed when loading modules which use clk_get(). Signed-off-by: Rabin Vincent Signed-off-by: Russell King --- arch/arm/mach-versatile/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c index 1f929c391af7..b3bebcc5623b 100644 --- a/arch/arm/mach-versatile/core.c +++ b/arch/arm/mach-versatile/core.c @@ -413,7 +413,7 @@ static struct clk ref24_clk = { .rate = 24000000, }; -static struct clk_lookup lookups[] __initdata = { +static struct clk_lookup lookups[] = { { /* UART0 */ .dev_id = "dev:f1", .clk = &ref24_clk, From a93ea9b357a4d4fce9a1f65bf9c152fb67c30716 Mon Sep 17 00:00:00 2001 From: Rabin Vincent Date: Mon, 18 May 2009 17:26:08 +0100 Subject: [PATCH 099/267] [ARM] 5517/1: integrator: don't put clock lookups in __initdata Remove the __initdata annotation for the clock lookups, since they will be needed when loading modules which use clk_get(). Signed-off-by: Rabin Vincent Signed-off-by: Russell King --- arch/arm/mach-integrator/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-integrator/core.c b/arch/arm/mach-integrator/core.c index 6f8872913073..a0f60e55da6a 100644 --- a/arch/arm/mach-integrator/core.c +++ b/arch/arm/mach-integrator/core.c @@ -121,7 +121,7 @@ static struct clk uartclk = { .rate = 14745600, }; -static struct clk_lookup lookups[] __initdata = { +static struct clk_lookup lookups[] = { { /* UART0 */ .dev_id = "mb:16", .clk = &uartclk, From 2b69a8a2b6e5f5d26a038a6494a88a1a776ac88f Mon Sep 17 00:00:00 2001 From: Eric Moore Date: Mon, 18 May 2009 12:57:24 -0600 Subject: [PATCH 100/267] [SCSI] mpt2sas: fix driver version inconsistency In Commit commit 3b8b5c9b1f08660583e5dfe095c24170df62f1d2 Author: Eric Moore Date: Tue Apr 21 15:44:27 2009 -0600 [SCSI] mpt2sas : bump driver version to 01.100.02.00 The MPT2SAS_MAJOR_VERSION didn't get bumped from 00 to 01 so applications will see it incorrectly as 00.100.02.00 driver instead of 01.100.02.00. Fix by making MPT2SAS_MAJOR_VERSION match the major number in MPT2SAS_DRIVER_VERSION Signed-off-by: Eric Moore Signed-off-by: James Bottomley --- drivers/scsi/mpt2sas/mpt2sas_base.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/mpt2sas/mpt2sas_base.h b/drivers/scsi/mpt2sas/mpt2sas_base.h index babd4cc0cb25..36b1d1052ba1 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_base.h +++ b/drivers/scsi/mpt2sas/mpt2sas_base.h @@ -69,7 +69,7 @@ #define MPT2SAS_AUTHOR "LSI Corporation " #define MPT2SAS_DESCRIPTION "LSI MPT Fusion SAS 2.0 Device Driver" #define MPT2SAS_DRIVER_VERSION "01.100.02.00" -#define MPT2SAS_MAJOR_VERSION 00 +#define MPT2SAS_MAJOR_VERSION 01 #define MPT2SAS_MINOR_VERSION 100 #define MPT2SAS_BUILD_VERSION 02 #define MPT2SAS_RELEASE_VERSION 00 From 775273131810caa41dfc7f9e552ea5d8508caf40 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= Date: Sun, 10 May 2009 20:32:34 +0000 Subject: [PATCH 101/267] tcp: fix MSG_PEEK race check MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 518a09ef11 (tcp: Fix recvmsg MSG_PEEK influence of blocking behavior) lets the loop run longer than the race check did previously expect, so we need to be more careful with this check and consider the work we have been doing. I tried my best to deal with urg hole madness too which happens here: if (!sock_flag(sk, SOCK_URGINLINE)) { ++*seq; ... by using additional offset by one but I certainly have very little interest in testing that part. Signed-off-by: Ilpo Järvinen Tested-by: Frans Pop Tested-by: Ian Zimmermann Signed-off-by: David S. Miller --- net/ipv4/tcp.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 1d7f49c6f0ca..7a0f0b27bf1f 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1321,6 +1321,7 @@ int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, struct task_struct *user_recv = NULL; int copied_early = 0; struct sk_buff *skb; + u32 urg_hole = 0; lock_sock(sk); @@ -1532,7 +1533,8 @@ do_prequeue: } } } - if ((flags & MSG_PEEK) && peek_seq != tp->copied_seq) { + if ((flags & MSG_PEEK) && + (peek_seq - copied - urg_hole != tp->copied_seq)) { if (net_ratelimit()) printk(KERN_DEBUG "TCP(%s:%d): Application bug, race in MSG_PEEK.\n", current->comm, task_pid_nr(current)); @@ -1553,6 +1555,7 @@ do_prequeue: if (!urg_offset) { if (!sock_flag(sk, SOCK_URGINLINE)) { ++*seq; + urg_hole++; offset++; used--; if (!used) From c0f84d0d4be3f7d818b4ffb04d27f9bae64397f0 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 18 May 2009 15:12:31 -0700 Subject: [PATCH 102/267] sch_teql: should not dereference skb after ndo_start_xmit() It is illegal to dereference a skb after a successful ndo_start_xmit() call. We must store skb length in a local variable instead. Bug was introduced in 2.6.27 by commit 0abf77e55a2459aa9905be4b226e4729d5b4f0cb (net_sched: Add accessor function for packet length for qdiscs) Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/sched/sch_teql.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c index ec697cebb63b..3b6418297231 100644 --- a/net/sched/sch_teql.c +++ b/net/sched/sch_teql.c @@ -303,6 +303,8 @@ restart: switch (teql_resolve(skb, skb_res, slave)) { case 0: if (__netif_tx_trylock(slave_txq)) { + unsigned int length = qdisc_pkt_len(skb); + if (!netif_tx_queue_stopped(slave_txq) && !netif_tx_queue_frozen(slave_txq) && slave_ops->ndo_start_xmit(skb, slave) == 0) { @@ -310,8 +312,7 @@ restart: master->slaves = NEXT_SLAVE(q); netif_wake_queue(dev); master->stats.tx_packets++; - master->stats.tx_bytes += - qdisc_pkt_len(skb); + master->stats.tx_bytes += length; return 0; } __netif_tx_unlock(slave_txq); From c4ca2374312b4de819dd700e72a68395eddb5fcb Mon Sep 17 00:00:00 2001 From: Ajit Khaparde Date: Mon, 18 May 2009 15:38:55 -0700 Subject: [PATCH 103/267] be2net: add two new pci device ids to pci device table Signed-off-by: Ajit Khaparde Signed-off-by: David S. Miller --- drivers/net/benet/be.h | 14 ++++++++++++++ drivers/net/benet/be_main.c | 10 +++++----- 2 files changed, 19 insertions(+), 5 deletions(-) diff --git a/drivers/net/benet/be.h b/drivers/net/benet/be.h index c49ddd08b2aa..b4bb06fdf307 100644 --- a/drivers/net/benet/be.h +++ b/drivers/net/benet/be.h @@ -35,8 +35,22 @@ #define DRV_VER "2.0.348" #define DRV_NAME "be2net" #define BE_NAME "ServerEngines BladeEngine2 10Gbps NIC" +#define OC_NAME "Emulex OneConnect 10Gbps NIC" #define DRV_DESC BE_NAME "Driver" +#define BE_VENDOR_ID 0x19a2 +#define BE_DEVICE_ID1 0x211 +#define OC_DEVICE_ID1 0x700 +#define OC_DEVICE_ID2 0x701 + +static inline char *nic_name(struct pci_dev *pdev) +{ + if (pdev->device == OC_DEVICE_ID1 || pdev->device == OC_DEVICE_ID2) + return OC_NAME; + else + return BE_NAME; +} + /* Number of bytes of an RX frame that are copied to skb->data */ #define BE_HDR_LEN 64 #define BE_MAX_JUMBO_FRAME_SIZE 9018 diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c index 30d0c81c989e..5c378b5e8e41 100644 --- a/drivers/net/benet/be_main.c +++ b/drivers/net/benet/be_main.c @@ -28,10 +28,10 @@ static unsigned int rx_frag_size = 2048; module_param(rx_frag_size, uint, S_IRUGO); MODULE_PARM_DESC(rx_frag_size, "Size of a fragment that holds rcvd data."); -#define BE_VENDOR_ID 0x19a2 -#define BE2_DEVICE_ID_1 0x0211 static DEFINE_PCI_DEVICE_TABLE(be_dev_ids) = { - { PCI_DEVICE(BE_VENDOR_ID, BE2_DEVICE_ID_1) }, + { PCI_DEVICE(BE_VENDOR_ID, BE_DEVICE_ID1) }, + { PCI_DEVICE(BE_VENDOR_ID, OC_DEVICE_ID1) }, + { PCI_DEVICE(BE_VENDOR_ID, OC_DEVICE_ID2) }, { 0 } }; MODULE_DEVICE_TABLE(pci, be_dev_ids); @@ -1859,7 +1859,7 @@ static int __devinit be_probe(struct pci_dev *pdev, if (status != 0) goto stats_clean; - dev_info(&pdev->dev, BE_NAME " port %d\n", adapter->port_num); + dev_info(&pdev->dev, "%s port %d\n", nic_name(pdev), adapter->port_num); return 0; stats_clean: @@ -1873,7 +1873,7 @@ rel_reg: disable_dev: pci_disable_device(pdev); do_none: - dev_warn(&pdev->dev, BE_NAME " initialization failed\n"); + dev_err(&pdev->dev, "%s initialization failed\n", nic_name(pdev)); return status; } From 511e11e396dc596825ce04d53d7f6d579404bc01 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 18 May 2009 19:26:37 -0700 Subject: [PATCH 104/267] pkt_sched: gen_estimator: use 64 bit intermediate counters for bps gen_estimator can overflow bps (bytes per second) with Gb links, while it was designed with a u32 API, with a theorical limit of 34360Mbit (2^32 bytes) Using 64 bit intermediate avbps/brate counters can allow us to reach this theorical limit. Signed-off-by: Eric Dumazet Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller --- net/core/gen_estimator.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/net/core/gen_estimator.c b/net/core/gen_estimator.c index 9cc9f95b109e..6d62d4618cfc 100644 --- a/net/core/gen_estimator.c +++ b/net/core/gen_estimator.c @@ -66,9 +66,9 @@ NOTES. - * The stored value for avbps is scaled by 2^5, so that maximal - rate is ~1Gbit, avpps is scaled by 2^10. - + * avbps is scaled by 2^5, avpps is scaled by 2^10. + * both values are reported as 32 bit unsigned values. bps can + overflow for fast links : max speed being 34360Mbit/sec * Minimal interval is HZ/4=250msec (it is the greatest common divisor for HZ=100 and HZ=1024 8)), maximal interval is (HZ*2^EST_MAX_INTERVAL)/4 = 8sec. Shorter intervals @@ -86,9 +86,9 @@ struct gen_estimator spinlock_t *stats_lock; int ewma_log; u64 last_bytes; + u64 avbps; u32 last_packets; u32 avpps; - u32 avbps; struct rcu_head e_rcu; struct rb_node node; }; @@ -115,6 +115,7 @@ static void est_timer(unsigned long arg) rcu_read_lock(); list_for_each_entry_rcu(e, &elist[idx].list, list) { u64 nbytes; + u64 brate; u32 npackets; u32 rate; @@ -125,9 +126,9 @@ static void est_timer(unsigned long arg) nbytes = e->bstats->bytes; npackets = e->bstats->packets; - rate = (nbytes - e->last_bytes)<<(7 - idx); + brate = (nbytes - e->last_bytes)<<(7 - idx); e->last_bytes = nbytes; - e->avbps += ((long)rate - (long)e->avbps) >> e->ewma_log; + e->avbps += ((s64)(brate - e->avbps)) >> e->ewma_log; e->rate_est->bps = (e->avbps+0xF)>>5; rate = (npackets - e->last_packets)<<(12 - idx); From 7ee2cb7f32b299c2b06a31fde155457203e4b7dd Mon Sep 17 00:00:00 2001 From: Frank Filz Date: Mon, 18 May 2009 17:41:40 -0400 Subject: [PATCH 105/267] nfs: Fix NFS v4 client handling of MAY_EXEC in nfs_permission. The problem is that permission checking is skipped if atomic open is possible, but when exec opens a file, it just opens it O_READONLY which means EXEC permission will not be checked at that time. This problem is observed by the following sequence (executed as root): mount -t nfs4 server:/ /mnt4 echo "ls" >/mnt4/foo chmod 744 /mnt4/foo su guest -c "mnt4/foo" Signed-off-by: Frank Filz Signed-off-by: Trond Myklebust Cc: stable@kernel.org Tested-by: Eugene Teo Signed-off-by: Linus Torvalds --- fs/nfs/dir.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 370b190a09d1..89f98e9a024b 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1943,7 +1943,8 @@ int nfs_permission(struct inode *inode, int mask) case S_IFREG: /* NFSv4 has atomic_open... */ if (nfs_server_capable(inode, NFS_CAP_ATOMIC_OPEN) - && (mask & MAY_OPEN)) + && (mask & MAY_OPEN) + && !(mask & MAY_EXEC)) goto out; break; case S_IFDIR: From 995b337952cdf7e05d288eede580257b632a8343 Mon Sep 17 00:00:00 2001 From: Thomas Chenault Date: Mon, 18 May 2009 21:43:27 -0700 Subject: [PATCH 106/267] net: fix skb_seq_read returning wrong offset/length for page frag data When called with a consumed value that is less than skb_headlen(skb) bytes into a page frag, skb_seq_read() incorrectly returns an offset/length relative to skb->data. Ensure that data which should come from a page frag does. Signed-off-by: Thomas Chenault Tested-by: Shyam Iyer Signed-off-by: David S. Miller --- net/core/skbuff.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index d152394b2611..e505b5392e1e 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2288,7 +2288,7 @@ unsigned int skb_seq_read(unsigned int consumed, const u8 **data, next_skb: block_limit = skb_headlen(st->cur_skb) + st->stepped_offset; - if (abs_offset < block_limit) { + if (abs_offset < block_limit && !st->frag_data) { *data = st->cur_skb->data + (abs_offset - st->stepped_offset); return block_limit - abs_offset; } From bc8a5397433e4effbaddfa7e462d10b3c060cabb Mon Sep 17 00:00:00 2001 From: Frans Pop Date: Mon, 18 May 2009 21:48:38 -0700 Subject: [PATCH 107/267] ipv4: make default for INET_LRO consistent with help text Commit e81963b1 ("ipv4: Make INET_LRO a bool instead of tristate.") changed this config from tristate to bool. Add default so that it is consistent with the help text. Signed-off-by: Frans Pop Signed-off-by: David S. Miller --- net/ipv4/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index 9d26a3da37e5..5b919f7b45db 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -408,7 +408,7 @@ config INET_XFRM_MODE_BEET config INET_LRO bool "Large Receive Offload (ipv4/tcp)" - + default y ---help--- Support for Large Receive Offload (ipv4/tcp). From 28afea5b2ffaa48f4f43d22ae8edcc384c05df80 Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Tue, 19 May 2009 08:25:48 +0200 Subject: [PATCH 108/267] xen/blkfront: allow xenbus state transition to Closing->Closed when not Connected This situation can occur when attempting to attach a block device whose backend is an empty physical CD-ROM driver. The backend in this case will go directly from the Initialising state to Closing->Closed. Previously this would result in a NULL pointer deref on info->gd (xenbus_dev_fatal does not return as a1a15ac5 seems to expect) Cc: stable@kernel.org Signed-off-by: Ian Campbell Acked-by: Jeremy Fitzhardinge Signed-off-by: Jens Axboe --- drivers/block/xen-blkfront.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index 8f905089b72b..6971a12e8b9b 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -977,8 +977,10 @@ static void backend_changed(struct xenbus_device *dev, break; case XenbusStateClosing: - if (info->gd == NULL) - xenbus_dev_fatal(dev, -ENODEV, "gd is NULL"); + if (info->gd == NULL) { + xenbus_frontend_closed(dev); + break; + } bd = bdget_disk(info->gd, 0); if (bd == NULL) xenbus_dev_fatal(dev, -ENODEV, "bdget failed"); From 31a14400e8f365e5d4d1a0b2a1ef4d5405d6091e Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Tue, 19 May 2009 08:27:42 +0200 Subject: [PATCH 109/267] xen/blkfront: fix warning when deleting gendisk on unplug/shutdown Currently blkfront gives a warning when hot unplugging due to calling del_gendisk() with interrupts disabled (due to blkif_io_lock). WARNING: at kernel/softirq.c:124 local_bh_enable+0x36/0x84() Modules linked in: xenfs xen_netfront ext3 jbd mbcache xen_blkfront Pid: 13, comm: xenwatch Not tainted 2.6.29-xs5.5.0.13 #3 Call Trace: [] warn_slowpath+0x80/0xb6 [] xen_sched_clock+0x16/0x63 [] xen_force_evtchn_callback+0xc/0x10 [] check_events+0x8/0xe [] xen_restore_fl_direct_end+0x0/0x1 [] xen_mc_flush+0x10a/0x13f [] __switch_to+0x114/0x14e [] dequeue_task+0x62/0x70 [] finish_task_switch+0x2b/0x84 [] schedule+0x66d/0x6e7 [] xen_force_evtchn_callback+0xc/0x10 [] xen_force_evtchn_callback+0xc/0x10 [] local_bh_enable+0x36/0x84 [] sk_filter+0x57/0x5c [] netlink_broadcast+0x1d5/0x315 [] kobject_uevent_env+0x28d/0x331 [] device_del+0x10f/0x120 [] device_unregister+0x8/0x10 [] bdi_unregister+0x2d/0x39 [] unlink_gendisk+0x23/0x3e [] del_gendisk+0x7b/0xe7 [] blkfront_closing+0x28/0x6e [xen_blkfront] [] backend_changed+0x3ad/0x41d [xen_blkfront] We can fix this by calling del_gendisk() later in blkfront_closing, after releasing blkif_io_lock. Since the queue is stopped during the interrupts disabled phase I don't think there is any danger of an event occuring between releasing the blkif_io_lock and deleting the disk. Signed-off-by: Ian Campbell Cc: Jeremy Fitzhardinge Signed-off-by: Jens Axboe --- drivers/block/xen-blkfront.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index 6971a12e8b9b..a6cbf7b808e6 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -934,8 +934,6 @@ static void blkfront_closing(struct xenbus_device *dev) spin_lock_irqsave(&blkif_io_lock, flags); - del_gendisk(info->gd); - /* No more blkif_request(). */ blk_stop_queue(info->rq); @@ -949,6 +947,8 @@ static void blkfront_closing(struct xenbus_device *dev) blk_cleanup_queue(info->rq); info->rq = NULL; + del_gendisk(info->gd); + out: xenbus_frontend_closed(dev); } From 8b6427a2a8f7dd43e9208fb33a3b116d66db4979 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Tue, 19 May 2009 09:57:03 -0400 Subject: [PATCH 110/267] cifs: fix pointer initialization and checks in cifs_follow_symlink (try #4) This is the third respin of the patch posted yesterday to fix the error handling in cifs_follow_symlink. It also includes a fix for a bogus NULL pointer check in CIFSSMBQueryUnixSymLink that Jeff Moyer spotted. It's possible for CIFSSMBQueryUnixSymLink to return without setting target_path to a valid pointer. If that happens then the current value to which we're initializing this pointer could cause an oops when it's kfree'd. This patch is a little more comprehensive than the last patches. It reorganizes cifs_follow_link a bit for (hopefully) better readability. It should also eliminate the uneeded allocation of full_path on servers without unix extensions (assuming they can get to this point anyway, of which I'm not convinced). On a side note, I'm not sure I agree with the logic of enabling this query even when unix extensions are disabled on the client. It seems like that should disable this as well. But, changing that is outside the scope of this fix, so I've left it alone for now. Reported-by: Jeff Moyer Signed-off-by: Jeff Layton Reviewed-by: Jeff Moyer Reviewed-by: Christoph Hellwig Signed-off-by: Steve French --- fs/cifs/cifssmb.c | 2 +- fs/cifs/link.c | 52 +++++++++++++++++++++++------------------------ 2 files changed, 27 insertions(+), 27 deletions(-) diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 5759ba53dc96..d06260251c30 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -2475,7 +2475,7 @@ querySymLinkRetry: /* BB FIXME investigate remapping reserved chars here */ *symlinkinfo = cifs_strndup_from_ucs(data_start, count, is_unicode, nls_codepage); - if (!symlinkinfo) + if (!*symlinkinfo) rc = -ENOMEM; } } diff --git a/fs/cifs/link.c b/fs/cifs/link.c index ea9d11e3dcbb..cd83c53fcbb5 100644 --- a/fs/cifs/link.c +++ b/fs/cifs/link.c @@ -107,48 +107,48 @@ void * cifs_follow_link(struct dentry *direntry, struct nameidata *nd) { struct inode *inode = direntry->d_inode; - int rc = -EACCES; + int rc = -ENOMEM; int xid; char *full_path = NULL; - char *target_path = ERR_PTR(-ENOMEM); - struct cifs_sb_info *cifs_sb; - struct cifsTconInfo *pTcon; + char *target_path = NULL; + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifsTconInfo *tcon = cifs_sb->tcon; xid = GetXid(); - full_path = build_path_from_dentry(direntry); + /* + * For now, we just handle symlinks with unix extensions enabled. + * Eventually we should handle NTFS reparse points, and MacOS + * symlink support. For instance... + * + * rc = CIFSSMBQueryReparseLinkInfo(...) + * + * For now, just return -EACCES when the server doesn't support posix + * extensions. Note that we still allow querying symlinks when posix + * extensions are manually disabled. We could disable these as well + * but there doesn't seem to be any harm in allowing the client to + * read them. + */ + if (!(tcon->ses->capabilities & CAP_UNIX)) { + rc = -EACCES; + goto out; + } + full_path = build_path_from_dentry(direntry); if (!full_path) goto out; cFYI(1, ("Full path: %s inode = 0x%p", full_path, inode)); - cifs_sb = CIFS_SB(inode->i_sb); - pTcon = cifs_sb->tcon; - - /* We could change this to: - if (pTcon->unix_ext) - but there does not seem any point in refusing to - get symlink info if we can, even if unix extensions - turned off for this mount */ - - if (pTcon->ses->capabilities & CAP_UNIX) - rc = CIFSSMBUnixQuerySymLink(xid, pTcon, full_path, - &target_path, - cifs_sb->local_nls); - else { - /* BB add read reparse point symlink code here */ - /* rc = CIFSSMBQueryReparseLinkInfo */ - /* BB Add code to Query ReparsePoint info */ - /* BB Add MAC style xsymlink check here if enabled */ - } + rc = CIFSSMBUnixQuerySymLink(xid, tcon, full_path, &target_path, + cifs_sb->local_nls); + kfree(full_path); +out: if (rc != 0) { kfree(target_path); target_path = ERR_PTR(rc); } - kfree(full_path); -out: FreeXid(xid); nd_set_link(nd, target_path); return NULL; From 8e7d2b2c6ecd3c21a54b877eae3d5be48292e6b5 Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Fri, 8 May 2009 16:13:25 -0700 Subject: [PATCH 111/267] drm/i915: allocate large pointer arrays with vmalloc For awhile now, many of the GEM code paths have allocated page or object arrays with the slab allocator. This is nice and fast, but won't work well if memory is fragmented, since the slab allocator works with physically contiguous memory (i.e. order > 2 allocations are likely to fail fairly early after booting and doing some work). This patch works around the issue by falling back to vmalloc for >PAGE_SIZE allocations. This is ugly, but much less work than chaining a bunch of pages together by hand (suprisingly there's not a bunch of generic kernel helpers for this yet afaik). vmalloc space is somewhat precious on 32 bit kernels, but our allocations shouldn't be big enough to cause problems, though they're routinely more than a page. Note that this patch doesn't address the unchecked alloc-based-on-ioctl-args in GEM; that needs to be fixed in a separate patch. Also, I've deliberately ignored the DRM's "area" junk. I don't think anyone actually uses it anymore and I'm hoping it gets ripped out soon. [Updated: removed size arg to new free function. We could unify the free functions as well once the DRM mem tracking is ripped out.] fd.o bug #20152 (part 1/3) Signed-off-by: Jesse Barnes Signed-off-by: Eric Anholt --- drivers/gpu/drm/i915/i915_gem.c | 38 +++++++++++++-------------------- include/drm/drmP.h | 24 +++++++++++++++++++++ 2 files changed, 39 insertions(+), 23 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index b189b49c7602..4a24c90fb940 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -349,7 +349,7 @@ i915_gem_shmem_pread_slow(struct drm_device *dev, struct drm_gem_object *obj, last_data_page = (data_ptr + args->size - 1) / PAGE_SIZE; num_pages = last_data_page - first_data_page + 1; - user_pages = kcalloc(num_pages, sizeof(struct page *), GFP_KERNEL); + user_pages = drm_calloc_large(num_pages, sizeof(struct page *)); if (user_pages == NULL) return -ENOMEM; @@ -429,7 +429,7 @@ fail_put_user_pages: SetPageDirty(user_pages[i]); page_cache_release(user_pages[i]); } - kfree(user_pages); + drm_free_large(user_pages); return ret; } @@ -649,7 +649,7 @@ i915_gem_gtt_pwrite_slow(struct drm_device *dev, struct drm_gem_object *obj, last_data_page = (data_ptr + args->size - 1) / PAGE_SIZE; num_pages = last_data_page - first_data_page + 1; - user_pages = kcalloc(num_pages, sizeof(struct page *), GFP_KERNEL); + user_pages = drm_calloc_large(num_pages, sizeof(struct page *)); if (user_pages == NULL) return -ENOMEM; @@ -719,7 +719,7 @@ out_unlock: out_unpin_pages: for (i = 0; i < pinned_pages; i++) page_cache_release(user_pages[i]); - kfree(user_pages); + drm_free_large(user_pages); return ret; } @@ -824,7 +824,7 @@ i915_gem_shmem_pwrite_slow(struct drm_device *dev, struct drm_gem_object *obj, last_data_page = (data_ptr + args->size - 1) / PAGE_SIZE; num_pages = last_data_page - first_data_page + 1; - user_pages = kcalloc(num_pages, sizeof(struct page *), GFP_KERNEL); + user_pages = drm_calloc_large(num_pages, sizeof(struct page *)); if (user_pages == NULL) return -ENOMEM; @@ -902,7 +902,7 @@ fail_unlock: fail_put_user_pages: for (i = 0; i < pinned_pages; i++) page_cache_release(user_pages[i]); - kfree(user_pages); + drm_free_large(user_pages); return ret; } @@ -1408,9 +1408,7 @@ i915_gem_object_put_pages(struct drm_gem_object *obj) } obj_priv->dirty = 0; - drm_free(obj_priv->pages, - page_count * sizeof(struct page *), - DRM_MEM_DRIVER); + drm_free_large(obj_priv->pages); obj_priv->pages = NULL; } @@ -2024,8 +2022,7 @@ i915_gem_object_get_pages(struct drm_gem_object *obj) */ page_count = obj->size / PAGE_SIZE; BUG_ON(obj_priv->pages != NULL); - obj_priv->pages = drm_calloc(page_count, sizeof(struct page *), - DRM_MEM_DRIVER); + obj_priv->pages = drm_calloc_large(page_count, sizeof(struct page *)); if (obj_priv->pages == NULL) { DRM_ERROR("Faled to allocate page list\n"); obj_priv->pages_refcount--; @@ -3111,7 +3108,7 @@ i915_gem_get_relocs_from_user(struct drm_i915_gem_exec_object *exec_list, reloc_count += exec_list[i].relocation_count; } - *relocs = drm_calloc(reloc_count, sizeof(**relocs), DRM_MEM_DRIVER); + *relocs = drm_calloc_large(reloc_count, sizeof(**relocs)); if (*relocs == NULL) return -ENOMEM; @@ -3125,8 +3122,7 @@ i915_gem_get_relocs_from_user(struct drm_i915_gem_exec_object *exec_list, exec_list[i].relocation_count * sizeof(**relocs)); if (ret != 0) { - drm_free(*relocs, reloc_count * sizeof(**relocs), - DRM_MEM_DRIVER); + drm_free_large(*relocs); *relocs = NULL; return -EFAULT; } @@ -3165,7 +3161,7 @@ i915_gem_put_relocs_to_user(struct drm_i915_gem_exec_object *exec_list, } err: - drm_free(relocs, reloc_count * sizeof(*relocs), DRM_MEM_DRIVER); + drm_free_large(relocs); return ret; } @@ -3198,10 +3194,8 @@ i915_gem_execbuffer(struct drm_device *dev, void *data, return -EINVAL; } /* Copy in the exec list from userland */ - exec_list = drm_calloc(sizeof(*exec_list), args->buffer_count, - DRM_MEM_DRIVER); - object_list = drm_calloc(sizeof(*object_list), args->buffer_count, - DRM_MEM_DRIVER); + exec_list = drm_calloc_large(sizeof(*exec_list), args->buffer_count); + object_list = drm_calloc_large(sizeof(*object_list), args->buffer_count); if (exec_list == NULL || object_list == NULL) { DRM_ERROR("Failed to allocate exec or object list " "for %d buffers\n", @@ -3462,10 +3456,8 @@ err: } pre_mutex_err: - drm_free(object_list, sizeof(*object_list) * args->buffer_count, - DRM_MEM_DRIVER); - drm_free(exec_list, sizeof(*exec_list) * args->buffer_count, - DRM_MEM_DRIVER); + drm_free_large(object_list); + drm_free_large(exec_list); drm_free(cliprects, sizeof(*cliprects) * args->num_cliprects, DRM_MEM_DRIVER); diff --git a/include/drm/drmP.h b/include/drm/drmP.h index c8c422151431..b84d8ae35e6f 100644 --- a/include/drm/drmP.h +++ b/include/drm/drmP.h @@ -1519,6 +1519,30 @@ static __inline__ void *drm_calloc(size_t nmemb, size_t size, int area) { return kcalloc(nmemb, size, GFP_KERNEL); } + +static __inline__ void *drm_calloc_large(size_t nmemb, size_t size) +{ + u8 *addr; + + if (size <= PAGE_SIZE) + return kcalloc(nmemb, size, GFP_KERNEL); + + addr = vmalloc(nmemb * size); + if (!addr) + return NULL; + + memset(addr, 0, nmemb * size); + + return addr; +} + +static __inline void drm_free_large(void *ptr) +{ + if (!is_vmalloc_addr(ptr)) + return kfree(ptr); + + vfree(ptr); +} #else extern void *drm_alloc(size_t size, int area); extern void drm_free(void *pt, size_t size, int area); From 26a9a418237c0b06528941bca693c49c8d97edbe Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 19 May 2009 11:25:35 -0700 Subject: [PATCH 112/267] Avoid ICE in get_random_int() with gcc-3.4.5 Martin Knoblauch reports that trying to build 2.6.30-rc6-git3 with RHEL4.3 userspace (gcc (GCC) 3.4.5 20051201 (Red Hat 3.4.5-2)) causes an internal compiler error (ICE): drivers/char/random.c: In function `get_random_int': drivers/char/random.c:1672: error: unrecognizable insn: (insn 202 148 150 0 /scratch/build/linux-2.6.30-rc6-git3/arch/x86/include/asm/tsc.h:23 (set (reg:SI 0 ax [91]) (subreg:SI (plus:DI (plus:DI (reg:DI 0 ax [88]) (subreg:DI (reg:SI 6 bp) 0)) (const_int -4 [0xfffffffffffffffc])) 0)) -1 (nil) (nil)) drivers/char/random.c:1672: internal compiler error: in extract_insn, at recog.c:2083 and after some debugging it turns out that it's due to the code trying to figure out the rough value of the current stack pointer by taking an address of an uninitialized variable and casting that to an integer. This is clearly a compiler bug, but it's not worth fighting - while the current stack kernel pointer might be somewhat hard to predict in user space, it's also not generally going to change for a lot of the call chains for a particular process. So just drop it, and mumble some incoherent curses at the compiler. Tested-by: Martin Knoblauch Cc: Matt Mackall Cc: Ingo Molnar Signed-off-by: Linus Torvalds --- drivers/char/random.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/char/random.c b/drivers/char/random.c index b2ced39d76b2..8c7444857a4b 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -1673,7 +1673,7 @@ unsigned int get_random_int(void) int ret; keyptr = get_keyptr(); - hash[0] += current->pid + jiffies + get_cycles() + (int)(long)&ret; + hash[0] += current->pid + jiffies + get_cycles(); ret = half_md4_transform(hash, keyptr->secret); put_cpu_var(get_random_int_hash); From b126d113484886d900179b4ae55ea7fabe15f936 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Thu, 30 Apr 2009 14:48:36 +0100 Subject: [PATCH 113/267] mfd: Keep a cache of WM8350 volatile values Due to the way that the WM8350 audio driver handles CODEC_ENA many of the WM8350 audio registers are marked as volatile when they aren't actually so. Allow the audio driver to see a cache of these values for inspection during interrupt context. To do this we need to stop satisfying any bits from volatile registers from cache - there's no real benefit from doing so anyway, we did the read already. Signed-off-by: Mark Brown Signed-off-by: Samuel Ortiz --- drivers/mfd/wm8350-core.c | 8 -------- 1 file changed, 8 deletions(-) diff --git a/drivers/mfd/wm8350-core.c b/drivers/mfd/wm8350-core.c index c2be3088e2e1..fe24079387c5 100644 --- a/drivers/mfd/wm8350-core.c +++ b/drivers/mfd/wm8350-core.c @@ -79,10 +79,6 @@ static int wm8350_phys_read(struct wm8350 *wm8350, u8 reg, int num_regs, /* Cache is CPU endian */ dest[i - reg] = be16_to_cpu(dest[i - reg]); - /* Satisfy non-volatile bits from cache */ - dest[i - reg] &= wm8350_reg_io_map[i].vol; - dest[i - reg] |= wm8350->reg_cache[i]; - /* Mask out non-readable bits */ dest[i - reg] &= wm8350_reg_io_map[i].readable; } @@ -182,9 +178,6 @@ static int wm8350_write(struct wm8350 *wm8350, u8 reg, int num_regs, u16 *src) (wm8350->reg_cache[i] & ~wm8350_reg_io_map[i].writable) | src[i - reg]; - /* Don't store volatile bits */ - wm8350->reg_cache[i] &= ~wm8350_reg_io_map[i].vol; - src[i - reg] = cpu_to_be16(src[i - reg]); } @@ -1261,7 +1254,6 @@ static int wm8350_create_cache(struct wm8350 *wm8350, int type, int mode) (i < WM8350_CLOCK_CONTROL_1 || i > WM8350_AIF_TEST)) { value = be16_to_cpu(wm8350->reg_cache[i]); value &= wm8350_reg_io_map[i].readable; - value &= ~wm8350_reg_io_map[i].vol; wm8350->reg_cache[i] = value; } else wm8350->reg_cache[i] = reg_map[i]; From f43ab901005c2bb3c5440e91b6efae9f5db02e7c Mon Sep 17 00:00:00 2001 From: Nelson Castillo Date: Tue, 12 May 2009 13:26:47 -0700 Subject: [PATCH 114/267] mfd: pcf50633: fix unsafe disable_irq() Without this change Openmoko Freerunner (GTA02) bootstrap will deadlock. As pointed out in other patches this issue is in the wild since the merge of: : commit 3aa551c9b4c40018f0e261a178e3d25478dc04a9 : Author: Thomas Gleixner : Date: Mon Mar 23 18:28:15 2009 +0100 : : genirq: add threaded interrupt handler support : : Add support for threaded interrupt handlers Signed-off-by: Nelson Castillo Cc: Ingo Molnar Cc: Cc: Thomas Gleixner Signed-off-by: Andrew Morton Signed-off-by: Samuel Ortiz --- drivers/mfd/pcf50633-core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mfd/pcf50633-core.c b/drivers/mfd/pcf50633-core.c index 7793932a513b..11a6248cc1c1 100644 --- a/drivers/mfd/pcf50633-core.c +++ b/drivers/mfd/pcf50633-core.c @@ -443,7 +443,7 @@ static irqreturn_t pcf50633_irq(int irq, void *data) dev_dbg(pcf->dev, "pcf50633_irq\n"); get_device(pcf->dev); - disable_irq(pcf->irq); + disable_irq_nosync(pcf->irq); schedule_work(&pcf->irq_work); return IRQ_HANDLED; From 33252572e727ccdcc54efdb67157e7ab3d6942db Mon Sep 17 00:00:00 2001 From: Nico Schottelius Date: Sat, 16 May 2009 14:00:56 +0200 Subject: [PATCH 115/267] Fix scripts/setlocalversion with tagged git commit Produce correct output for - tagged commit (v2.6.30-rc6) - past tagged commit (v2.6.30-rc5-299-g7c7327d) - no tag Signed-off-by: Linus Torvalds --- scripts/setlocalversion | 30 +++++++++++++++++++++++------- 1 file changed, 23 insertions(+), 7 deletions(-) diff --git a/scripts/setlocalversion b/scripts/setlocalversion index 32c8554f3946..00790472f641 100755 --- a/scripts/setlocalversion +++ b/scripts/setlocalversion @@ -1,5 +1,13 @@ #!/bin/sh -# Print additional version information for non-release trees. +# +# This scripts adds local version information from the version +# control systems git, mercurial (hg) and subversion (svn). +# +# If something goes wrong, send a mail the kernel build mailinglist +# (see MAINTAINERS) and CC Nico Schottelius +# . +# +# usage() { echo "Usage: $0 [srctree]" >&2 @@ -10,12 +18,20 @@ cd "${1:-.}" || usage # Check for git and a git repo. if head=`git rev-parse --verify --short HEAD 2>/dev/null`; then - # Do we have an untagged tag? - if atag=`git describe 2>/dev/null`; then - echo "$atag" | awk -F- '{printf("-%05d-%s", $(NF-1),$(NF))}' - # add -g${head}, if there is no usable tag - else - printf '%s%s' -g $head + + # If we are at a tagged commit (like "v2.6.30-rc6"), we ignore it, + # because this version is defined in the top level Makefile. + if [ -z "`git describe --exact-match 2>/dev/null`" ]; then + + # If we are past a tagged commit (like "v2.6.30-rc5-302-g72357d5"), + # we pretty print it. + if atag="`git describe 2>/dev/null`"; then + echo "$atag" | awk -F- '{printf("-%05d-%s", $(NF-1),$(NF))}' + + # If we don't have a tag at all we print -g{commitish}. + else + printf '%s%s' -g $head + fi fi # Is this git on svn? From 64d1304a64477629cb16b75491a77bafe6f86963 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 18 May 2009 21:20:10 +0200 Subject: [PATCH 116/267] futex: setup writeable mapping for futex ops which modify user space data The futex code installs a read only mapping via get_user_pages_fast() even if the futex op function has to modify user space data. The eventual fault was fixed up by futex_handle_fault() which walked the VMA with mmap_sem held. After the cleanup patches which removed the mmap_sem dependency of the futex code commit 4dc5b7a36a49eff97050894cf1b3a9a02523717 (futex: clean up fault logic) removed the private VMA walk logic from the futex code. This change results in a stale RO mapping which is not fixed up. Instead of reintroducing the previous fault logic we set up the mapping in get_user_pages_fast() read/write for all operations which modify user space data. Also handle private futexes in the same way and make the current unconditional access_ok(VERIFY_WRITE) depend on the futex op. Reported-by: Andreas Schwab Signed-off-by: Thomas Gleixner CC: stable@kernel.org --- kernel/futex.c | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) diff --git a/kernel/futex.c b/kernel/futex.c index eef8cd26b5e5..d546b2d53a62 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -193,6 +193,7 @@ static void drop_futex_key_refs(union futex_key *key) * @uaddr: virtual address of the futex * @fshared: 0 for a PROCESS_PRIVATE futex, 1 for PROCESS_SHARED * @key: address where result is stored. + * @rw: mapping needs to be read/write (values: VERIFY_READ, VERIFY_WRITE) * * Returns a negative error code or 0 * The key words are stored in *key on success. @@ -203,7 +204,8 @@ static void drop_futex_key_refs(union futex_key *key) * * lock_page() might sleep, the caller should not hold a spinlock. */ -static int get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key) +static int +get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key, int rw) { unsigned long address = (unsigned long)uaddr; struct mm_struct *mm = current->mm; @@ -226,7 +228,7 @@ static int get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key) * but access_ok() should be faster than find_vma() */ if (!fshared) { - if (unlikely(!access_ok(VERIFY_WRITE, uaddr, sizeof(u32)))) + if (unlikely(!access_ok(rw, uaddr, sizeof(u32)))) return -EFAULT; key->private.mm = mm; key->private.address = address; @@ -235,7 +237,7 @@ static int get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key) } again: - err = get_user_pages_fast(address, 1, 0, &page); + err = get_user_pages_fast(address, 1, rw == VERIFY_WRITE, &page); if (err < 0) return err; @@ -677,7 +679,7 @@ static int futex_wake(u32 __user *uaddr, int fshared, int nr_wake, u32 bitset) if (!bitset) return -EINVAL; - ret = get_futex_key(uaddr, fshared, &key); + ret = get_futex_key(uaddr, fshared, &key, VERIFY_READ); if (unlikely(ret != 0)) goto out; @@ -723,10 +725,10 @@ futex_wake_op(u32 __user *uaddr1, int fshared, u32 __user *uaddr2, int ret, op_ret; retry: - ret = get_futex_key(uaddr1, fshared, &key1); + ret = get_futex_key(uaddr1, fshared, &key1, VERIFY_READ); if (unlikely(ret != 0)) goto out; - ret = get_futex_key(uaddr2, fshared, &key2); + ret = get_futex_key(uaddr2, fshared, &key2, VERIFY_WRITE); if (unlikely(ret != 0)) goto out_put_key1; @@ -814,10 +816,10 @@ static int futex_requeue(u32 __user *uaddr1, int fshared, u32 __user *uaddr2, int ret, drop_count = 0; retry: - ret = get_futex_key(uaddr1, fshared, &key1); + ret = get_futex_key(uaddr1, fshared, &key1, VERIFY_READ); if (unlikely(ret != 0)) goto out; - ret = get_futex_key(uaddr2, fshared, &key2); + ret = get_futex_key(uaddr2, fshared, &key2, VERIFY_READ); if (unlikely(ret != 0)) goto out_put_key1; @@ -1140,7 +1142,7 @@ static int futex_wait(u32 __user *uaddr, int fshared, q.bitset = bitset; retry: q.key = FUTEX_KEY_INIT; - ret = get_futex_key(uaddr, fshared, &q.key); + ret = get_futex_key(uaddr, fshared, &q.key, VERIFY_READ); if (unlikely(ret != 0)) goto out; @@ -1330,7 +1332,7 @@ static int futex_lock_pi(u32 __user *uaddr, int fshared, q.pi_state = NULL; retry: q.key = FUTEX_KEY_INIT; - ret = get_futex_key(uaddr, fshared, &q.key); + ret = get_futex_key(uaddr, fshared, &q.key, VERIFY_WRITE); if (unlikely(ret != 0)) goto out; @@ -1594,7 +1596,7 @@ retry: if ((uval & FUTEX_TID_MASK) != task_pid_vnr(current)) return -EPERM; - ret = get_futex_key(uaddr, fshared, &key); + ret = get_futex_key(uaddr, fshared, &key, VERIFY_WRITE); if (unlikely(ret != 0)) goto out; From b3bad72e494fb2ff0c81be4ca2ddb94adf6a47c2 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 17 May 2009 20:17:06 +0200 Subject: [PATCH 117/267] PCI PM: Fix initialization and kexec breakage for some devices Recent PCI PM changes introduced a bug that causes some devices to be mishandled after kexec and during early initialization. The failure scenario in the kexec case is the following: * Assume a PCI device is not power-manageable by the platform and has PCI_PM_CTRL_NO_SOFT_RESET set in PMCSR. * The device is put into D3 before kexec (using the native PCI PM). * After kexec, pci_setup_device() sets the device's power state to PCI_UNKNOWN. * pci_set_power_state(dev, PCI_D0) is called by the device's driver. * __pci_start_power_transition(dev, PCI_D0) is called and since the device is not power-manageable by the platform, it causes pci_update_current_state(dev, PCI_D0) to be called. As a result the device's current_state field is updated to PCI_D3, in accordance with the contents of its PCI PM registers. * pci_raw_set_power_state() is called and it changes the device power state to D0. *However*, it should also call pci_restore_bars() to reinitialize the device, but it doesn't, because the device's current_state field has been modified earlier. To prevent this from happening, modify pci_platform_power_transition() so that it doesn't use pci_update_current_state() to update the current_state field for devices that aren't power-manageable by the platform. Instead, this field should be updated directly for devices that don't support the native PCI PM. Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes --- drivers/pci/pci.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 34bf0fdf5047..1a91bf9687af 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -557,7 +557,8 @@ static int pci_platform_power_transition(struct pci_dev *dev, pci_power_t state) } else { error = -ENODEV; /* Fall back to PCI_D0 if native PM is not supported */ - pci_update_current_state(dev, PCI_D0); + if (!dev->pm_cap) + dev->current_state = PCI_D0; } return error; From fbaa58696cef848de818768783ef185bd3f05158 Mon Sep 17 00:00:00 2001 From: Eric Paris Date: Wed, 13 May 2009 12:50:40 -0400 Subject: [PATCH 118/267] TPM: get_event_name stack corruption get_event_name uses sprintf to fill a buffer declared on the stack. It fills the buffer 2 bytes at a time. What the code doesn't take into account is that sprintf(buf, "%02x", data) actually writes 3 bytes. 2 bytes for the data and then it nul terminates the string. Since we declare buf to be 40 characters long and then we write 40 bytes of data into buf sprintf is going to write 41 characters. The fix is to leave room in buf for the nul terminator. Signed-off-by: Eric Paris Signed-off-by: James Morris --- drivers/char/tpm/tpm_bios.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/char/tpm/tpm_bios.c b/drivers/char/tpm/tpm_bios.c index ed306eb1057f..0c2f55a38b95 100644 --- a/drivers/char/tpm/tpm_bios.c +++ b/drivers/char/tpm/tpm_bios.c @@ -212,7 +212,8 @@ static int get_event_name(char *dest, struct tcpa_event *event, unsigned char * event_entry) { const char *name = ""; - char data[40] = ""; + /* 41 so there is room for 40 data and 1 nul */ + char data[41] = ""; int i, n_len = 0, d_len = 0; struct tcpa_pc_event *pc_event; From b674137755bbe2750f997a2a1264db3cdf8abcb3 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Mon, 18 May 2009 11:56:16 +1000 Subject: [PATCH 119/267] drm: Round size of SHM maps to PAGE_SIZE Currently, userspace can fail to obtain the SAREA mapping (among other reasons) if it passes SAREA_MAX to drmAddMap without aligning it to the page size. This breaks for example on PowerPC with 64K pages and radeon despite the kernel radeon actually doing the right rouding in the first place. The way SAREA_MAX is defined with a bunch of ifdef's and duplicated between libdrm and the X server is gross, ultimately it should be retrieved by userspace from the kernel, but in the meantime, we have plenty of existing userspace built with bad values that need to work. This patch works around broken userspace by rounding the requested size in drm_addmap_core() of any SHM map to the page size. Since the backing memory for SHM maps is also allocated within addmap_core, there is no danger of adjacent memory being exposed due to the increased map size. The only side effect is that drivers that previously tried to create or access SHM maps using a size < PAGE_SIZE and failed (getting -EINVAL), will now succeed at the cost of a little bit more memory used if that happens to be when the map is created. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Jesse Barnes --- drivers/gpu/drm/drm_bufs.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/drm_bufs.c b/drivers/gpu/drm/drm_bufs.c index 6d80d17f1e96..0411d912d82a 100644 --- a/drivers/gpu/drm/drm_bufs.c +++ b/drivers/gpu/drm/drm_bufs.c @@ -170,6 +170,14 @@ static int drm_addmap_core(struct drm_device * dev, resource_size_t offset, } DRM_DEBUG("offset = 0x%08llx, size = 0x%08lx, type = %d\n", (unsigned long long)map->offset, map->size, map->type); + + /* page-align _DRM_SHM maps. They are allocated here so there is no security + * hole created by that and it works around various broken drivers that use + * a non-aligned quantity to map the SAREA. --BenH + */ + if (map->type == _DRM_SHM) + map->size = PAGE_ALIGN(map->size); + if ((map->offset & (~(resource_size_t)PAGE_MASK)) || (map->size & (~PAGE_MASK))) { drm_free(map, sizeof(*map), DRM_MEM_MAPS); return -EINVAL; From d3375ea7213d09284ee3683d8bc075b9d3786a03 Mon Sep 17 00:00:00 2001 From: Roel Kluin Date: Wed, 20 May 2009 08:42:06 +0200 Subject: [PATCH 120/267] cdrom: beyond ARRAY_SIZE of viocd_diskinfo Do not go beyond ARRAY_SIZE of viocd_diskinfo Signed-off-by: Roel Kluin Signed-off-by: Jens Axboe --- drivers/cdrom/viocd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/cdrom/viocd.c b/drivers/cdrom/viocd.c index 13929356135c..9b1624e0ddeb 100644 --- a/drivers/cdrom/viocd.c +++ b/drivers/cdrom/viocd.c @@ -587,7 +587,7 @@ static int viocd_probe(struct vio_dev *vdev, const struct vio_device_id *id) struct device_node *node = vdev->dev.archdata.of_node; deviceno = vdev->unit_address; - if (deviceno > VIOCD_MAX_CD) + if (deviceno >= VIOCD_MAX_CD) return -ENODEV; if (!node) return -ENODEV; From b3b778b387ed3849ebc4a51baf8617be90df6625 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 20 May 2009 17:05:52 +0200 Subject: [PATCH 121/267] ALSA: pcsp - fix printk format warning again The commit 5a641bcd6398841cc4606b0a732d41a09256fd94 changed the printk format to '%lu', but the value passed seems to be dependent on the architecture. On x86-64, I got a new warning now because an int value is passed actaully. As a workaround, just cast the value always to unsigned long. Signed-off-by: Takashi Iwai --- sound/drivers/pcsp/pcsp_mixer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/drivers/pcsp/pcsp_mixer.c b/sound/drivers/pcsp/pcsp_mixer.c index 771955a9be71..199b03377142 100644 --- a/sound/drivers/pcsp/pcsp_mixer.c +++ b/sound/drivers/pcsp/pcsp_mixer.c @@ -51,7 +51,7 @@ static int pcsp_treble_info(struct snd_kcontrol *kcontrol, if (uinfo->value.enumerated.item > chip->max_treble) uinfo->value.enumerated.item = chip->max_treble; sprintf(uinfo->value.enumerated.name, "%lu", - PCSP_CALC_RATE(uinfo->value.enumerated.item)); + (unsigned long)PCSP_CALC_RATE(uinfo->value.enumerated.item)); return 0; } From 950312ce22696ddfa42a957bdadaa9c24151e679 Mon Sep 17 00:00:00 2001 From: Andrew Randrianasulu Date: Thu, 14 May 2009 11:29:27 -0700 Subject: [PATCH 122/267] MIPS: IP32: Fix hang on shutdown in power button interrupt handler. The hang was caused by the use of disable_irq() from the interrupt handler itself. Fixed by the use of disable_irq_nosync(). The issue was triggered by: commit 3aa551c9b4c40018f0e261a178e3d25478dc04a9 Author: Thomas Gleixner Date: Mon Mar 23 18:28:15 2009 +0100 genirq: add threaded interrupt handler support Signed-off-by: Ralf Baechle --- arch/mips/sgi-ip32/ip32-reset.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/sgi-ip32/ip32-reset.c b/arch/mips/sgi-ip32/ip32-reset.c index b6cab089561e..667da932b7b2 100644 --- a/arch/mips/sgi-ip32/ip32-reset.c +++ b/arch/mips/sgi-ip32/ip32-reset.c @@ -145,7 +145,7 @@ static irqreturn_t ip32_rtc_int(int irq, void *dev_id) "%s: RTC IRQ without RTC_IRQF\n", __func__); } /* Wait until interrupt goes away */ - disable_irq(MACEISA_RTC_IRQ); + disable_irq_nosync(MACEISA_RTC_IRQ); init_timer(&debounce_timer); debounce_timer.function = debounce; debounce_timer.expires = jiffies + 50; From 7e9e05cad94217498e4d9bd6ef7137b4e9e7ed64 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sat, 16 May 2009 12:23:45 +0100 Subject: [PATCH 123/267] MIPS: IP22: Fix hang in power button interrupt handler The hang was caused by the use of disable_irq() from the interrupt handler itself. Fixed by the use of disable_irq_nosync(). The issue was triggered by: commit 3aa551c9b4c40018f0e261a178e3d25478dc04a9 Author: Thomas Gleixner Date: Mon Mar 23 18:28:15 2009 +0100 genirq: add threaded interrupt handler support Signed-off-by: Ralf Baechle --- arch/mips/sgi-ip22/ip22-reset.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/sgi-ip22/ip22-reset.c b/arch/mips/sgi-ip22/ip22-reset.c index 4ad5c3393fd3..45b6694c2079 100644 --- a/arch/mips/sgi-ip22/ip22-reset.c +++ b/arch/mips/sgi-ip22/ip22-reset.c @@ -148,7 +148,7 @@ static irqreturn_t panel_int(int irq, void *dev_id) if (sgint->istat1 & SGINT_ISTAT1_PWR) { /* Wait until interrupt goes away */ - disable_irq(SGI_PANEL_IRQ); + disable_irq_nosync(SGI_PANEL_IRQ); init_timer(&debounce_timer); debounce_timer.function = debounce; debounce_timer.expires = jiffies + 5; From 195d1a96ae5fdfbedb8dc4b97afee578921fa99e Mon Sep 17 00:00:00 2001 From: peter fuerst Date: Sun, 17 May 2009 23:49:45 +0200 Subject: [PATCH 124/267] MIPS: IP28: Change to build with -mr10k-cache-barrier=store Richard Sandiford's new code for inserting the cache-barriers, for GCC 4.3 and above and already incorporated in the current GCC-release, uses a slightly different option-syntax. Signed-off-by: peter fuerst Signed-off-by: Ralf Baechle --- arch/mips/Makefile | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/mips/Makefile b/arch/mips/Makefile index 26947ab85260..c4cae9e6b802 100644 --- a/arch/mips/Makefile +++ b/arch/mips/Makefile @@ -473,12 +473,12 @@ endif # Simplified: what IP22 does at 128MB+ in ksegN, IP28 does at 512MB+ in xkphys # ifdef CONFIG_SGI_IP28 - ifeq ($(call cc-option-yn,-mr10k-cache-barrier=1), n) - $(error gcc doesn't support needed option -mr10k-cache-barrier=1) + ifeq ($(call cc-option-yn,-mr10k-cache-barrier=store), n) + $(error gcc doesn't support needed option -mr10k-cache-barrier=store) endif endif core-$(CONFIG_SGI_IP28) += arch/mips/sgi-ip22/ -cflags-$(CONFIG_SGI_IP28) += -mr10k-cache-barrier=1 -I$(srctree)/arch/mips/include/asm/mach-ip28 +cflags-$(CONFIG_SGI_IP28) += -mr10k-cache-barrier=store -I$(srctree)/arch/mips/include/asm/mach-ip28 load-$(CONFIG_SGI_IP28) += 0xa800000020004000 # From a5e696e5d0f1377ff6beb10d2f40edb6a3d1de18 Mon Sep 17 00:00:00 2001 From: Greg Ungerer Date: Wed, 20 May 2009 16:12:32 +1000 Subject: [PATCH 125/267] MIPS: 64-bit: Fix system lockup. The address range size calculation inside local_flush_tlb_kernel_range() is being truncated by a too small size variable holder on 64-bit systems. The truncated size can result in an erroneous tlbsize check that means we sit spinning inside a loop trying to flush a hige number of TLB entries. This is for all intents and purposes a system hang. Fix by using an appropriately sized valiable to hold the size. [Ralf: Greg's original patch submission identified the issue and fixed one instance in tlb-r4k.c but there there were several more. For consistency I also modified tlb-r3k.c even though that file is only used on 32-bit.] Signed-off-by: Greg Ungerer Signed-off-by: Ralf Baechle --- arch/mips/mm/tlb-r3k.c | 6 ++---- arch/mips/mm/tlb-r4k.c | 6 ++---- arch/mips/mm/tlb-r8k.c | 3 +-- 3 files changed, 5 insertions(+), 10 deletions(-) diff --git a/arch/mips/mm/tlb-r3k.c b/arch/mips/mm/tlb-r3k.c index f0cf46adb978..1c0048a6f5cf 100644 --- a/arch/mips/mm/tlb-r3k.c +++ b/arch/mips/mm/tlb-r3k.c @@ -82,8 +82,7 @@ void local_flush_tlb_range(struct vm_area_struct *vma, unsigned long start, int cpu = smp_processor_id(); if (cpu_context(cpu, mm) != 0) { - unsigned long flags; - int size; + unsigned long size, flags; #ifdef DEBUG_TLB printk("[tlbrange<%lu,0x%08lx,0x%08lx>]", @@ -121,8 +120,7 @@ void local_flush_tlb_range(struct vm_area_struct *vma, unsigned long start, void local_flush_tlb_kernel_range(unsigned long start, unsigned long end) { - unsigned long flags; - int size; + unsigned long size, flags; #ifdef DEBUG_TLB printk("[tlbrange<%lu,0x%08lx,0x%08lx>]", start, end); diff --git a/arch/mips/mm/tlb-r4k.c b/arch/mips/mm/tlb-r4k.c index 9619f66e531e..892be426787c 100644 --- a/arch/mips/mm/tlb-r4k.c +++ b/arch/mips/mm/tlb-r4k.c @@ -117,8 +117,7 @@ void local_flush_tlb_range(struct vm_area_struct *vma, unsigned long start, int cpu = smp_processor_id(); if (cpu_context(cpu, mm) != 0) { - unsigned long flags; - int size; + unsigned long size, flags; ENTER_CRITICAL(flags); size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT; @@ -160,8 +159,7 @@ void local_flush_tlb_range(struct vm_area_struct *vma, unsigned long start, void local_flush_tlb_kernel_range(unsigned long start, unsigned long end) { - unsigned long flags; - int size; + unsigned long size, flags; ENTER_CRITICAL(flags); size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT; diff --git a/arch/mips/mm/tlb-r8k.c b/arch/mips/mm/tlb-r8k.c index 4f01a3be215c..4ec95cc2df2f 100644 --- a/arch/mips/mm/tlb-r8k.c +++ b/arch/mips/mm/tlb-r8k.c @@ -111,8 +111,7 @@ out_restore: /* Usable for KV1 addresses only! */ void local_flush_tlb_kernel_range(unsigned long start, unsigned long end) { - unsigned long flags; - int size; + unsigned long size, flags; size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT; size = (size + 1) >> 1; From 2b611cb6eed04062d0a9861c82248e02c844ba3f Mon Sep 17 00:00:00 2001 From: Pavel Roskin Date: Fri, 27 Mar 2009 17:47:27 -0400 Subject: [PATCH 126/267] ath5k: fix scanning in AR2424 AR5K_PHY_PLL_40MHZ_5413 should not be ORed with AR5K_PHY_MODE_RAD_RF5112 for 5 GHz channels. The incorrect PLL value breaks scanning in the countries where 5 GHz channels are allowed. Signed-off-by: Pavel Roskin Acked-by: Nick Kossifidis Signed-off-by: John W. Linville --- drivers/net/wireless/ath5k/reset.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/ath5k/reset.c b/drivers/net/wireless/ath5k/reset.c index 7a17d31b2fd9..faede828c8ff 100644 --- a/drivers/net/wireless/ath5k/reset.c +++ b/drivers/net/wireless/ath5k/reset.c @@ -359,7 +359,7 @@ int ath5k_hw_nic_wakeup(struct ath5k_hw *ah, int flags, bool initial) mode |= AR5K_PHY_MODE_FREQ_5GHZ; if (ah->ah_radio == AR5K_RF5413) - clock |= AR5K_PHY_PLL_40MHZ_5413; + clock = AR5K_PHY_PLL_40MHZ_5413; else clock |= AR5K_PHY_PLL_40MHZ; From 88f16db7a2fa63b9242e8a0fbc40d51722f2e2f9 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 13 May 2009 12:04:30 +0200 Subject: [PATCH 127/267] wext: verify buffer size for SIOCSIWENCODEEXT Another design flaw in wireless extensions (is anybody surprised?) in the way it handles the iw_encode_ext structure: The structure is part of the 'extra' memory but contains the key length explicitly, instead of it just being the length of the extra buffer - size of the struct and using the explicit key length only for the get operation (which only writes it). Therefore, we have this layout: extra: +-------------------------+ | struct iw_encode_ext { | | ... | | u16 key_len; | | u8 key[0]; | | }; | +-------------------------+ | key material | +-------------------------+ Now, all drivers I checked use ext->key_len without checking that both key_len and the struct fit into the extra buffer that has been copied from userspace. This leads to a buffer overrun while reading that buffer, depending on the driver it may be possible to specify arbitrary key_len or it may need to be a proper length for the key algorithm specified. Thankfully, this is only exploitable by root, but root can actually cause a segfault or use kernel memory as a key (which you can even get back with siocgiwencode or siocgiwencodeext from the key buffer). Fix this by verifying that key_len fits into the buffer along with struct iw_encode_ext. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- net/wireless/wext.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/net/wireless/wext.c b/net/wireless/wext.c index cb6a5bb85d80..0e59f9ae9b81 100644 --- a/net/wireless/wext.c +++ b/net/wireless/wext.c @@ -786,6 +786,13 @@ static int ioctl_standard_iw_point(struct iw_point *iwp, unsigned int cmd, err = -EFAULT; goto out; } + + if (cmd == SIOCSIWENCODEEXT) { + struct iw_encode_ext *ee = (void *) extra; + + if (iwp->length < sizeof(*ee) + ee->key_len) + return -EFAULT; + } } err = handler(dev, info, (union iwreq_data *) iwp, extra); From a54be5d43aa2d6febc5a4f8dd3b87b9429b60437 Mon Sep 17 00:00:00 2001 From: Forrest Zhang Date: Wed, 13 May 2009 11:14:39 -0400 Subject: [PATCH 128/267] ath5k: fix exp off-by-one when computing OFDM delta slope Commit e8f055f0c3b ("ath5k: Update reset code") subtly changed the code that computes floating point values for the PHY3_TIMING register such that the exponent is off by a decimal point, which can cause problems with OFDM channel operation. get_bitmask_order() actually returns the highest bit set plus one, whereas the previous code wanted the highest bit set. Instead, use ilog2 which is what this code is really calculating. Also check coef_scaled to handle the (invalid) case where we need log2(0). Signed-off-by: Bob Copeland Signed-off-by: John W. Linville --- drivers/net/wireless/ath5k/reset.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/ath5k/reset.c b/drivers/net/wireless/ath5k/reset.c index faede828c8ff..5f72c111c2e8 100644 --- a/drivers/net/wireless/ath5k/reset.c +++ b/drivers/net/wireless/ath5k/reset.c @@ -26,7 +26,7 @@ \*****************************/ #include /* To determine if a card is pci-e */ -#include /* For get_bitmask_order */ +#include #include "ath5k.h" #include "reg.h" #include "base.h" @@ -69,10 +69,10 @@ static inline int ath5k_hw_write_ofdm_timings(struct ath5k_hw *ah, /* Get exponent * ALGO: coef_exp = 14 - highest set bit position */ - coef_exp = get_bitmask_order(coef_scaled); + coef_exp = ilog2(coef_scaled); /* Doesn't make sense if it's zero*/ - if (!coef_exp) + if (!coef_scaled || !coef_exp) return -EINVAL; /* Note: we've shifted coef_scaled by 24 */ From fbc9f97bbf5e1eaee562eba93dc60faaff3f3bfa Mon Sep 17 00:00:00 2001 From: Reinette Chatre Date: Fri, 15 May 2009 16:13:46 -0700 Subject: [PATCH 129/267] iwlwifi: do not cancel delayed work inside spin_lock_irqsave Calling cancel_delayed_work() from inside spin_lock_irqsave, introduces a potential deadlock. As explained by Johannes Berg A - lock T - timer phase CPU 1 CPU 2 --------------------------------------------- some place that calls cancel_timer_sync() (which is the | code) lock-irq(A) | "lock-irq"(T) | "unlock"(T) | wait(T) unlock(A) timer softirq "lock"(T) run(T) "unlock"(T) irq handler lock(A) unlock(A) Now all that again, interleaved, leading to deadlock: lock-irq(A) "lock"(T) run(T) IRQ during or maybe before run(T) --> lock(A) "lock-irq"(T) wait(T) We fix this by moving the call to cancel_delayed_work() into workqueue. There are cases where the work may not actually be queued or running at the time we are trying to cancel it, but cancel_delayed_work() is able to deal with this. Also cleanup iwl_set_mode related to this call. This function (iwl_set_mode) is only called when bringing interface up and there will thus not be any scanning done. No need to try to cancel scanning. Fixes http://bugzilla.kernel.org/show_bug.cgi?id=13224, which was also reported at http://marc.info/?l=linux-wireless&m=124081921903223&w=2 . Tested-by: Miles Lane Signed-off-by: Reinette Chatre Acked-by: Zhu Yi Signed-off-by: John W. Linville --- drivers/net/wireless/iwlwifi/iwl-agn.c | 7 ------- drivers/net/wireless/iwlwifi/iwl-scan.c | 7 ++++--- drivers/net/wireless/iwlwifi/iwl3945-base.c | 9 ++------- 3 files changed, 6 insertions(+), 17 deletions(-) diff --git a/drivers/net/wireless/iwlwifi/iwl-agn.c b/drivers/net/wireless/iwlwifi/iwl-agn.c index 3bb28db4a40f..f46ba2475776 100644 --- a/drivers/net/wireless/iwlwifi/iwl-agn.c +++ b/drivers/net/wireless/iwlwifi/iwl-agn.c @@ -669,13 +669,6 @@ static int iwl_set_mode(struct iwl_priv *priv, int mode) if (!iwl_is_ready_rf(priv)) return -EAGAIN; - cancel_delayed_work(&priv->scan_check); - if (iwl_scan_cancel_timeout(priv, 100)) { - IWL_WARN(priv, "Aborted scan still in progress after 100ms\n"); - IWL_DEBUG_MAC80211(priv, "leaving - scan abort failed.\n"); - return -EAGAIN; - } - iwl_commit_rxon(priv); return 0; diff --git a/drivers/net/wireless/iwlwifi/iwl-scan.c b/drivers/net/wireless/iwlwifi/iwl-scan.c index e7c65c4f741b..6330b91e37ce 100644 --- a/drivers/net/wireless/iwlwifi/iwl-scan.c +++ b/drivers/net/wireless/iwlwifi/iwl-scan.c @@ -227,9 +227,6 @@ static void iwl_rx_scan_complete_notif(struct iwl_priv *priv, /* The HW is no longer scanning */ clear_bit(STATUS_SCAN_HW, &priv->status); - /* The scan completion notification came in, so kill that timer... */ - cancel_delayed_work(&priv->scan_check); - IWL_DEBUG_INFO(priv, "Scan pass on %sGHz took %dms\n", (priv->scan_bands & BIT(IEEE80211_BAND_2GHZ)) ? "2.4" : "5.2", @@ -712,6 +709,8 @@ static void iwl_bg_request_scan(struct work_struct *data) mutex_lock(&priv->mutex); + cancel_delayed_work(&priv->scan_check); + if (!iwl_is_ready(priv)) { IWL_WARN(priv, "request scan called when driver not ready.\n"); goto done; @@ -925,6 +924,8 @@ void iwl_bg_scan_completed(struct work_struct *work) IWL_DEBUG_SCAN(priv, "SCAN complete scan\n"); + cancel_delayed_work(&priv->scan_check); + ieee80211_scan_completed(priv->hw, false); if (test_bit(STATUS_EXIT_PENDING, &priv->status)) diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c index 4cce66133500..ff4d0e41d7c4 100644 --- a/drivers/net/wireless/iwlwifi/iwl3945-base.c +++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c @@ -782,13 +782,6 @@ static int iwl3945_set_mode(struct iwl_priv *priv, int mode) if (!iwl_is_ready_rf(priv)) return -EAGAIN; - cancel_delayed_work(&priv->scan_check); - if (iwl_scan_cancel_timeout(priv, 100)) { - IWL_WARN(priv, "Aborted scan still in progress after 100ms\n"); - IWL_DEBUG_MAC80211(priv, "leaving - scan abort failed.\n"); - return -EAGAIN; - } - iwl3945_commit_rxon(priv); return 0; @@ -3298,6 +3291,8 @@ static void iwl3945_bg_request_scan(struct work_struct *data) mutex_lock(&priv->mutex); + cancel_delayed_work(&priv->scan_check); + if (!iwl_is_ready(priv)) { IWL_WARN(priv, "request scan called when driver not ready.\n"); goto done; From 875690c378d64d9ee2de15cad8206d3f11ae5096 Mon Sep 17 00:00:00 2001 From: Fabio Rossi Date: Wed, 1 Apr 2009 20:37:50 +0200 Subject: [PATCH 130/267] ath5k: fix interpolation with equal power levels When the EEPROM contains weird values for the power levels we have to fix the interpolation process. Signed-off-by: Fabio Rossi Acked-by: Nick Kossifidis Signed-off-by: John W. Linville --- drivers/net/wireless/ath5k/phy.c | 43 +++++++++++++++++++------------- 1 file changed, 25 insertions(+), 18 deletions(-) diff --git a/drivers/net/wireless/ath5k/phy.c b/drivers/net/wireless/ath5k/phy.c index 9e2faae5ae94..b48b29dca3d2 100644 --- a/drivers/net/wireless/ath5k/phy.c +++ b/drivers/net/wireless/ath5k/phy.c @@ -1487,28 +1487,35 @@ ath5k_get_linear_pcdac_min(const u8 *stepL, const u8 *stepR, { s8 tmp; s16 min_pwrL, min_pwrR; - s16 pwr_i = pwrL[0]; + s16 pwr_i; - do { - pwr_i--; - tmp = (s8) ath5k_get_interpolated_value(pwr_i, - pwrL[0], pwrL[1], - stepL[0], stepL[1]); + if (pwrL[0] == pwrL[1]) + min_pwrL = pwrL[0]; + else { + pwr_i = pwrL[0]; + do { + pwr_i--; + tmp = (s8) ath5k_get_interpolated_value(pwr_i, + pwrL[0], pwrL[1], + stepL[0], stepL[1]); + } while (tmp > 1); - } while (tmp > 1); + min_pwrL = pwr_i; + } - min_pwrL = pwr_i; + if (pwrR[0] == pwrR[1]) + min_pwrR = pwrR[0]; + else { + pwr_i = pwrR[0]; + do { + pwr_i--; + tmp = (s8) ath5k_get_interpolated_value(pwr_i, + pwrR[0], pwrR[1], + stepR[0], stepR[1]); + } while (tmp > 1); - pwr_i = pwrR[0]; - do { - pwr_i--; - tmp = (s8) ath5k_get_interpolated_value(pwr_i, - pwrR[0], pwrR[1], - stepR[0], stepR[1]); - - } while (tmp > 1); - - min_pwrR = pwr_i; + min_pwrR = pwr_i; + } /* Keep the right boundary so that it works for both curves */ return max(min_pwrL, min_pwrR); From 267d493b322b05984048aef8ea9b5b213490bbe0 Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Wed, 20 May 2009 10:51:41 -0400 Subject: [PATCH 131/267] airo: fix airo_get_encode{,ext} buffer overflow like I mean it... "airo: airo_get_encode{,ext} potential buffer overflow" was actually a no-op, due to an unrecognized type overflow in an assignment. Oddly, gcc only seems to tell me about it when using -Wextra...grrr... Signed-off-by: John W. Linville --- drivers/net/wireless/airo.c | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-) diff --git a/drivers/net/wireless/airo.c b/drivers/net/wireless/airo.c index d73475739127..9eabf4d1f2e7 100644 --- a/drivers/net/wireless/airo.c +++ b/drivers/net/wireless/airo.c @@ -6467,6 +6467,7 @@ static int airo_get_encode(struct net_device *dev, { struct airo_info *local = dev->ml_priv; int index = (dwrq->flags & IW_ENCODE_INDEX) - 1; + int wep_key_len; u8 buf[16]; if (!local->wep_capable) @@ -6500,11 +6501,13 @@ static int airo_get_encode(struct net_device *dev, dwrq->flags |= index + 1; /* Copy the key to the user buffer */ - dwrq->length = get_wep_key(local, index, &buf[0], sizeof(buf)); - if (dwrq->length != -1) - memcpy(extra, buf, dwrq->length); - else + wep_key_len = get_wep_key(local, index, &buf[0], sizeof(buf)); + if (wep_key_len < 0) { dwrq->length = 0; + } else { + dwrq->length = wep_key_len; + memcpy(extra, buf, dwrq->length); + } return 0; } @@ -6617,7 +6620,7 @@ static int airo_get_encodeext(struct net_device *dev, struct airo_info *local = dev->ml_priv; struct iw_point *encoding = &wrqu->encoding; struct iw_encode_ext *ext = (struct iw_encode_ext *)extra; - int idx, max_key_len; + int idx, max_key_len, wep_key_len; u8 buf[16]; if (!local->wep_capable) @@ -6661,11 +6664,13 @@ static int airo_get_encodeext(struct net_device *dev, memset(extra, 0, 16); /* Copy the key to the user buffer */ - ext->key_len = get_wep_key(local, idx, &buf[0], sizeof(buf)); - if (ext->key_len != -1) - memcpy(extra, buf, ext->key_len); - else + wep_key_len = get_wep_key(local, idx, &buf[0], sizeof(buf)); + if (wep_key_len < 0) { ext->key_len = 0; + } else { + ext->key_len = wep_key_len; + memcpy(extra, buf, ext->key_len); + } return 0; } From 5078b2e32ad4b1f753b1c837c15892202f753c97 Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Wed, 13 May 2009 17:04:42 -0400 Subject: [PATCH 132/267] cfg80211: fix race between core hint and driver's custom apply Its possible for cfg80211 to have scheduled the work and for the global workqueue to not have kicked in prior to a cfg80211 driver's regulatory hint or wiphy_apply_custom_regulatory(). Although this is very unlikely its possible and should fix this race. When this race would happen you are expected to have hit a null pointer dereference panic. Cc: stable@kernel.org Signed-off-by: Luis R. Rodriguez Tested-by: Alan Jenkins Signed-off-by: John W. Linville --- net/wireless/reg.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/net/wireless/reg.c b/net/wireless/reg.c index 08265ca15785..487cb627ddba 100644 --- a/net/wireless/reg.c +++ b/net/wireless/reg.c @@ -1551,6 +1551,13 @@ static int regulatory_hint_core(const char *alpha2) queue_regulatory_request(request); + /* + * This ensures last_request is populated once modules + * come swinging in and calling regulatory hints and + * wiphy_apply_custom_regulatory(). + */ + flush_scheduled_work(); + return 0; } From c9d2fbf36df5e04efa226614093bb1bacc6fe131 Mon Sep 17 00:00:00 2001 From: Jay Sternberg Date: Tue, 19 May 2009 14:56:36 -0700 Subject: [PATCH 133/267] iwlwifi: update 5000 ucode support to version 2 of API enable iwl driver to support 5000 ucode having version 2 of API Signed-off-by: Jay Sternberg Signed-off-by: Reinette Chatre Signed-off-by: John W. Linville --- drivers/net/wireless/iwlwifi/iwl-5000.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/iwlwifi/iwl-5000.c b/drivers/net/wireless/iwlwifi/iwl-5000.c index e5ca2511a81a..9452461ce864 100644 --- a/drivers/net/wireless/iwlwifi/iwl-5000.c +++ b/drivers/net/wireless/iwlwifi/iwl-5000.c @@ -46,7 +46,7 @@ #include "iwl-6000-hw.h" /* Highest firmware API version supported */ -#define IWL5000_UCODE_API_MAX 1 +#define IWL5000_UCODE_API_MAX 2 #define IWL5150_UCODE_API_MAX 2 /* Lowest firmware API version supported */ From a6c67339784db5763d6f20ae1881aeebe8c5a9f4 Mon Sep 17 00:00:00 2001 From: Roel Kluin Date: Wed, 20 May 2009 02:12:56 +0200 Subject: [PATCH 134/267] wireless: beyond ARRAY_SIZE of intf->crypto_stats Do not go beyond ARRAY_SIZE of intf->crypto_stats Signed-off-by: Roel Kluin Acked-by: Ivo van Doorn Signed-off-by: John W. Linville --- drivers/net/wireless/rt2x00/rt2x00debug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/rt2x00/rt2x00debug.c b/drivers/net/wireless/rt2x00/rt2x00debug.c index 07d378ef0b46..7b3ee8c2eaef 100644 --- a/drivers/net/wireless/rt2x00/rt2x00debug.c +++ b/drivers/net/wireless/rt2x00/rt2x00debug.c @@ -138,7 +138,7 @@ void rt2x00debug_update_crypto(struct rt2x00_dev *rt2x00dev, if (cipher == CIPHER_TKIP_NO_MIC) cipher = CIPHER_TKIP; - if (cipher == CIPHER_NONE || cipher > CIPHER_MAX) + if (cipher == CIPHER_NONE || cipher >= CIPHER_MAX) return; /* Remove CIPHER_NONE index */ From 9b6fe313bfce27d4a261257da70196be0ac2bef5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michel=20D=C3=A4nzer?= Date: Wed, 20 May 2009 13:32:00 +0200 Subject: [PATCH 135/267] drm: Copy back ioctl data to userspace regardless of return code. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fixes a regression from commit 9d5b3ffc42f7820e8ee07705496955e4c2c38dd9 ('drm: fixup some of the ioctl function exit paths'): The vblank ioctl needs to update the userspace parameters when interrupted by a signal, which was prevented by the return code check. This could cause the X server to hang in drmWaitVBlank(). Signed-off-by: Michel Dänzer Signed-off-by: Jesse Barnes --- drivers/gpu/drm/drm_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c index f01def16a669..019b7c578236 100644 --- a/drivers/gpu/drm/drm_drv.c +++ b/drivers/gpu/drm/drm_drv.c @@ -481,7 +481,7 @@ int drm_ioctl(struct inode *inode, struct file *filp, } retcode = func(dev, kdata, file_priv); - if ((retcode == 0) && (cmd & IOC_OUT)) { + if (cmd & IOC_OUT) { if (copy_to_user((void __user *)arg, kdata, _IOC_SIZE(cmd)) != 0) retcode = -EFAULT; From 03fbdb15c14e9746c63168e3ff2c64b9c8336d33 Mon Sep 17 00:00:00 2001 From: Alessandro Rubini Date: Wed, 20 May 2009 22:39:08 +0100 Subject: [PATCH 136/267] [ARM] 5519/1: amba probe: pass "struct amba_id *" instead of void * The second argument of the probe method points to the amba_id structure, so it's better passed with the correct type. None of the current in-tree drivers uses the pointer, so they have only been checked for a clean compile. Change suggested by Russell King. Signed-off-by: Alessandro Rubini Signed-off-by: Russell King --- drivers/input/serio/ambakmi.c | 2 +- drivers/mmc/host/mmci.c | 2 +- drivers/rtc/rtc-pl030.c | 2 +- drivers/rtc/rtc-pl031.c | 2 +- drivers/serial/amba-pl010.c | 2 +- drivers/serial/amba-pl011.c | 2 +- drivers/video/amba-clcd.c | 2 +- include/linux/amba/bus.h | 2 +- sound/arm/aaci.c | 2 +- 9 files changed, 9 insertions(+), 9 deletions(-) diff --git a/drivers/input/serio/ambakmi.c b/drivers/input/serio/ambakmi.c index e29cdc13a199..a28c06d686e1 100644 --- a/drivers/input/serio/ambakmi.c +++ b/drivers/input/serio/ambakmi.c @@ -107,7 +107,7 @@ static void amba_kmi_close(struct serio *io) clk_disable(kmi->clk); } -static int amba_kmi_probe(struct amba_device *dev, void *id) +static int amba_kmi_probe(struct amba_device *dev, struct amba_id *id) { struct amba_kmi_port *kmi; struct serio *io; diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c index 36875dcfa492..7d4febdab286 100644 --- a/drivers/mmc/host/mmci.c +++ b/drivers/mmc/host/mmci.c @@ -490,7 +490,7 @@ static void mmci_check_status(unsigned long data) mod_timer(&host->timer, jiffies + HZ); } -static int __devinit mmci_probe(struct amba_device *dev, void *id) +static int __devinit mmci_probe(struct amba_device *dev, struct amba_id *id) { struct mmc_platform_data *plat = dev->dev.platform_data; struct mmci_host *host; diff --git a/drivers/rtc/rtc-pl030.c b/drivers/rtc/rtc-pl030.c index 826153552157..aaf1f75fa293 100644 --- a/drivers/rtc/rtc-pl030.c +++ b/drivers/rtc/rtc-pl030.c @@ -102,7 +102,7 @@ static const struct rtc_class_ops pl030_ops = { .set_alarm = pl030_set_alarm, }; -static int pl030_probe(struct amba_device *dev, void *id) +static int pl030_probe(struct amba_device *dev, struct amba_id *id) { struct pl030_rtc *rtc; int ret; diff --git a/drivers/rtc/rtc-pl031.c b/drivers/rtc/rtc-pl031.c index 333eec689d2f..451fc13784d1 100644 --- a/drivers/rtc/rtc-pl031.c +++ b/drivers/rtc/rtc-pl031.c @@ -127,7 +127,7 @@ static int pl031_remove(struct amba_device *adev) return 0; } -static int pl031_probe(struct amba_device *adev, void *id) +static int pl031_probe(struct amba_device *adev, struct amba_id *id) { int ret; struct pl031_local *ldata; diff --git a/drivers/serial/amba-pl010.c b/drivers/serial/amba-pl010.c index e3a5ad5ef1d6..cdc049d4350f 100644 --- a/drivers/serial/amba-pl010.c +++ b/drivers/serial/amba-pl010.c @@ -665,7 +665,7 @@ static struct uart_driver amba_reg = { .cons = AMBA_CONSOLE, }; -static int pl010_probe(struct amba_device *dev, void *id) +static int pl010_probe(struct amba_device *dev, struct amba_id *id) { struct uart_amba_port *uap; void __iomem *base; diff --git a/drivers/serial/amba-pl011.c b/drivers/serial/amba-pl011.c index 8b2b9700f3e4..88fdac51b6c5 100644 --- a/drivers/serial/amba-pl011.c +++ b/drivers/serial/amba-pl011.c @@ -729,7 +729,7 @@ static struct uart_driver amba_reg = { .cons = AMBA_CONSOLE, }; -static int pl011_probe(struct amba_device *dev, void *id) +static int pl011_probe(struct amba_device *dev, struct amba_id *id) { struct uart_amba_port *uap; void __iomem *base; diff --git a/drivers/video/amba-clcd.c b/drivers/video/amba-clcd.c index 61050ab14128..d1f80bac54f0 100644 --- a/drivers/video/amba-clcd.c +++ b/drivers/video/amba-clcd.c @@ -437,7 +437,7 @@ static int clcdfb_register(struct clcd_fb *fb) return ret; } -static int clcdfb_probe(struct amba_device *dev, void *id) +static int clcdfb_probe(struct amba_device *dev, struct amba_id *id) { struct clcd_board *board = dev->dev.platform_data; struct clcd_fb *fb; diff --git a/include/linux/amba/bus.h b/include/linux/amba/bus.h index 51e6e54b2aa1..9b93cafa82a0 100644 --- a/include/linux/amba/bus.h +++ b/include/linux/amba/bus.h @@ -28,7 +28,7 @@ struct amba_id { struct amba_driver { struct device_driver drv; - int (*probe)(struct amba_device *, void *); + int (*probe)(struct amba_device *, struct amba_id *); int (*remove)(struct amba_device *); void (*shutdown)(struct amba_device *); int (*suspend)(struct amba_device *, pm_message_t); diff --git a/sound/arm/aaci.c b/sound/arm/aaci.c index 7fbd68fab944..5c48e36038f2 100644 --- a/sound/arm/aaci.c +++ b/sound/arm/aaci.c @@ -1074,7 +1074,7 @@ static unsigned int __devinit aaci_size_fifo(struct aaci *aaci) return i; } -static int __devinit aaci_probe(struct amba_device *dev, void *id) +static int __devinit aaci_probe(struct amba_device *dev, struct amba_id *id) { struct aaci *aaci; int ret, i; From cf8da764fc6959b7efb482f375dfef9830e98205 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 19 May 2009 18:54:22 +0000 Subject: [PATCH 137/267] net: fix length computation in rt_check_expire() rt_check_expire() computes average and standard deviation of chain lengths, but not correclty reset length to 0 at beginning of each chain. This probably gives overflows for sum2 (and sum) on loaded machines instead of meaningful results. Signed-off-by: Eric Dumazet Acked-by: Neil Horman Signed-off-by: David S. Miller --- net/ipv4/route.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index c4c60e9f068a..869cf1c44b78 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -785,7 +785,7 @@ static void rt_check_expire(void) static unsigned int rover; unsigned int i = rover, goal; struct rtable *rth, **rthp; - unsigned long length = 0, samples = 0; + unsigned long samples = 0; unsigned long sum = 0, sum2 = 0; u64 mult; @@ -795,9 +795,9 @@ static void rt_check_expire(void) goal = (unsigned int)mult; if (goal > rt_hash_mask) goal = rt_hash_mask + 1; - length = 0; for (; goal > 0; goal--) { unsigned long tmo = ip_rt_gc_timeout; + unsigned long length; i = (i + 1) & rt_hash_mask; rthp = &rt_hash_table[i].chain; @@ -809,6 +809,7 @@ static void rt_check_expire(void) if (*rthp == NULL) continue; + length = 0; spin_lock_bh(rt_hash_lock_addr(i)); while ((rth = *rthp) != NULL) { if (rt_is_expired(rth)) { From 1ddbcb005c395518c2cd0df504cff3d4b5c85853 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 19 May 2009 20:14:28 +0000 Subject: [PATCH 138/267] net: fix rtable leak in net/ipv4/route.c Alexander V. Lukyanov found a regression in 2.6.29 and made a complete analysis found in http://bugzilla.kernel.org/show_bug.cgi?id=13339 Quoted here because its a perfect one : begin_of_quotation 2.6.29 patch has introduced flexible route cache rebuilding. Unfortunately the patch has at least one critical flaw, and another problem. rt_intern_hash calculates rthi pointer, which is later used for new entry insertion. The same loop calculates cand pointer which is used to clean the list. If the pointers are the same, rtable leak occurs, as first the cand is removed then the new entry is appended to it. This leak leads to unregister_netdevice problem (usage count > 0). Another problem of the patch is that it tries to insert the entries in certain order, to facilitate counting of entries distinct by all but QoS parameters. Unfortunately, referencing an existing rtable entry moves it to list beginning, to speed up further lookups, so the carefully built order is destroyed. For the first problem the simplest patch it to set rthi=0 when rthi==cand, but it will also destroy the ordering. end_of_quotation Problematic commit is 1080d709fb9d8cd4392f93476ee46a9d6ea05a5b (net: implement emergency route cache rebulds when gc_elasticity is exceeded) Trying to keep dst_entries ordered is too complex and breaks the fact that order should depend on the frequency of use for garbage collection. A possible fix is to make rt_intern_hash() simpler, and only makes rt_check_expire() a litle bit smarter, being able to cope with an arbitrary entries order. The added loop is running on cache hot data, while cpu is prefetching next object, so should be unnoticied. Reported-and-analyzed-by: Alexander V. Lukyanov Signed-off-by: Eric Dumazet Acked-by: Neil Horman Signed-off-by: David S. Miller --- net/ipv4/route.c | 55 +++++++++++++++--------------------------------- 1 file changed, 17 insertions(+), 38 deletions(-) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 869cf1c44b78..28205e5bfa9b 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -784,7 +784,7 @@ static void rt_check_expire(void) { static unsigned int rover; unsigned int i = rover, goal; - struct rtable *rth, **rthp; + struct rtable *rth, *aux, **rthp; unsigned long samples = 0; unsigned long sum = 0, sum2 = 0; u64 mult; @@ -812,6 +812,7 @@ static void rt_check_expire(void) length = 0; spin_lock_bh(rt_hash_lock_addr(i)); while ((rth = *rthp) != NULL) { + prefetch(rth->u.dst.rt_next); if (rt_is_expired(rth)) { *rthp = rth->u.dst.rt_next; rt_free(rth); @@ -820,33 +821,30 @@ static void rt_check_expire(void) if (rth->u.dst.expires) { /* Entry is expired even if it is in use */ if (time_before_eq(jiffies, rth->u.dst.expires)) { +nofree: tmo >>= 1; rthp = &rth->u.dst.rt_next; /* - * Only bump our length if the hash - * inputs on entries n and n+1 are not - * the same, we only count entries on + * We only count entries on * a chain with equal hash inputs once * so that entries for different QOS * levels, and other non-hash input * attributes don't unfairly skew * the length computation */ - if ((*rthp == NULL) || - !compare_hash_inputs(&(*rthp)->fl, - &rth->fl)) - length += ONE; + for (aux = rt_hash_table[i].chain;;) { + if (aux == rth) { + length += ONE; + break; + } + if (compare_hash_inputs(&aux->fl, &rth->fl)) + break; + aux = aux->u.dst.rt_next; + } continue; } - } else if (!rt_may_expire(rth, tmo, ip_rt_gc_timeout)) { - tmo >>= 1; - rthp = &rth->u.dst.rt_next; - if ((*rthp == NULL) || - !compare_hash_inputs(&(*rthp)->fl, - &rth->fl)) - length += ONE; - continue; - } + } else if (!rt_may_expire(rth, tmo, ip_rt_gc_timeout)) + goto nofree; /* Cleanup aged off entries. */ *rthp = rth->u.dst.rt_next; @@ -1069,7 +1067,6 @@ out: return 0; static int rt_intern_hash(unsigned hash, struct rtable *rt, struct rtable **rp) { struct rtable *rth, **rthp; - struct rtable *rthi; unsigned long now; struct rtable *cand, **candp; u32 min_score; @@ -1089,7 +1086,6 @@ restart: } rthp = &rt_hash_table[hash].chain; - rthi = NULL; spin_lock_bh(rt_hash_lock_addr(hash)); while ((rth = *rthp) != NULL) { @@ -1135,17 +1131,6 @@ restart: chain_length++; rthp = &rth->u.dst.rt_next; - - /* - * check to see if the next entry in the chain - * contains the same hash input values as rt. If it does - * This is where we will insert into the list, instead of - * at the head. This groups entries that differ by aspects not - * relvant to the hash function together, which we use to adjust - * our chain length - */ - if (*rthp && compare_hash_inputs(&(*rthp)->fl, &rt->fl)) - rthi = rth; } if (cand) { @@ -1206,10 +1191,7 @@ restart: } } - if (rthi) - rt->u.dst.rt_next = rthi->u.dst.rt_next; - else - rt->u.dst.rt_next = rt_hash_table[hash].chain; + rt->u.dst.rt_next = rt_hash_table[hash].chain; #if RT_CACHE_DEBUG >= 2 if (rt->u.dst.rt_next) { @@ -1225,10 +1207,7 @@ restart: * previous writes to rt are comitted to memory * before making rt visible to other CPUS. */ - if (rthi) - rcu_assign_pointer(rthi->u.dst.rt_next, rt); - else - rcu_assign_pointer(rt_hash_table[hash].chain, rt); + rcu_assign_pointer(rt_hash_table[hash].chain, rt); spin_unlock_bh(rt_hash_lock_addr(hash)); *rp = rt; From 4f72427998b105392e60bae7a6798a0c96fe4f0a Mon Sep 17 00:00:00 2001 From: Jean-Mickael Guerin Date: Wed, 20 May 2009 17:38:59 -0700 Subject: [PATCH 139/267] IPv6: set RTPROT_KERNEL to initial route The use of unspecified protocol in IPv6 initial route prevents quagga to install IPv6 default route: # show ipv6 route S ::/0 [1/0] via fe80::1, eth1_0 K>* ::/0 is directly connected, lo, rej C>* ::1/128 is directly connected, lo C>* fe80::/64 is directly connected, eth1_0 # ip -6 route fe80::/64 dev eth1_0 proto kernel metric 256 mtu 1500 advmss 1440 hoplimit -1 ff00::/8 dev eth1_0 metric 256 mtu 1500 advmss 1440 hoplimit -1 unreachable default dev lo proto none metric -1 error -101 hoplimit 255 The attached patch ensures RTPROT_KERNEL to the default initial route and fixes the problem for quagga. This is similar to "ipv6: protocol for address routes" f410a1fba7afa79d2992620e874a343fdba28332. # show ipv6 route S>* ::/0 [1/0] via fe80::1, eth1_0 C>* ::1/128 is directly connected, lo C>* fe80::/64 is directly connected, eth1_0 # ip -6 route fe80::/64 dev eth1_0 proto kernel metric 256 mtu 1500 advmss 1440 hoplimit -1 fe80::/64 dev eth1_0 proto kernel metric 256 mtu 1500 advmss 1440 hoplimit -1 ff00::/8 dev eth1_0 metric 256 mtu 1500 advmss 1440 hoplimit -1 default via fe80::1 dev eth1_0 proto zebra metric 1024 mtu 1500 advmss 1440 hoplimit -1 unreachable default dev lo proto kernel metric -1 error -101 hoplimit 255 Signed-off-by: Jean-Mickael Guerin Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/ipv6/route.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 1394ddb6e35c..032a5ec391c5 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -137,6 +137,7 @@ static struct rt6_info ip6_null_entry_template = { } }, .rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP), + .rt6i_protocol = RTPROT_KERNEL, .rt6i_metric = ~(u32) 0, .rt6i_ref = ATOMIC_INIT(1), }; @@ -159,6 +160,7 @@ static struct rt6_info ip6_prohibit_entry_template = { } }, .rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP), + .rt6i_protocol = RTPROT_KERNEL, .rt6i_metric = ~(u32) 0, .rt6i_ref = ATOMIC_INIT(1), }; @@ -176,6 +178,7 @@ static struct rt6_info ip6_blk_hole_entry_template = { } }, .rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP), + .rt6i_protocol = RTPROT_KERNEL, .rt6i_metric = ~(u32) 0, .rt6i_ref = ATOMIC_INIT(1), }; From 98f32602d42951e61a059685f842aa7d778ffab0 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Thu, 21 May 2009 20:33:58 +0100 Subject: [PATCH 140/267] hugh: update email address My old address will shut down in a few days time: remove it from the tree, and add a tmpfs (shmem filesystem) maintainer entry with the new address. Signed-off-by: Hugh Dickins Signed-off-by: Hugh Dickins Signed-off-by: Linus Torvalds --- Documentation/filesystems/tmpfs.txt | 2 +- MAINTAINERS | 8 ++++++++ mm/rmap.c | 2 +- 3 files changed, 10 insertions(+), 2 deletions(-) diff --git a/Documentation/filesystems/tmpfs.txt b/Documentation/filesystems/tmpfs.txt index 222437efd75a..3015da0c6b2a 100644 --- a/Documentation/filesystems/tmpfs.txt +++ b/Documentation/filesystems/tmpfs.txt @@ -133,4 +133,4 @@ RAM/SWAP in 10240 inodes and it is only accessible by root. Author: Christoph Rohland , 1.12.01 Updated: - Hugh Dickins , 4 June 2007 + Hugh Dickins, 4 June 2007 diff --git a/MAINTAINERS b/MAINTAINERS index 2b349ba4add4..64ea80e45e3b 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5579,6 +5579,14 @@ M: ian@mnementh.co.uk S: Maintained F: drivers/mmc/host/tmio_mmc.* +TMPFS (SHMEM FILESYSTEM) +P: Hugh Dickins +M: hugh.dickins@tiscali.co.uk +L: linux-mm@kvack.org +S: Maintained +F: include/linux/shmem_fs.h +F: mm/shmem.c + TPM DEVICE DRIVER P: Debora Velarde M: debora@linux.vnet.ibm.com diff --git a/mm/rmap.c b/mm/rmap.c index 16521664010d..23122af32611 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -14,7 +14,7 @@ * Original design by Rik van Riel 2001 * File methods by Dave McCracken 2003, 2004 * Anonymous methods by Andrea Arcangeli 2004 - * Contributions by Hugh Dickins 2003, 2004 + * Contributions by Hugh Dickins 2003, 2004 */ /* From c40499e04b2005e61f989824251f9343b55f96bb Mon Sep 17 00:00:00 2001 From: Roel Kluin Date: Thu, 21 May 2009 15:04:15 -0700 Subject: [PATCH 141/267] gigaset: beyond ARRAY_SIZE of iwb->data Signed-off-by: Roel Kluin Signed-off-by: David S. Miller --- drivers/isdn/gigaset/isocdata.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/isdn/gigaset/isocdata.c b/drivers/isdn/gigaset/isocdata.c index b171e75cb52e..29808c4fb1cb 100644 --- a/drivers/isdn/gigaset/isocdata.c +++ b/drivers/isdn/gigaset/isocdata.c @@ -175,7 +175,7 @@ int gigaset_isowbuf_getbytes(struct isowbuf_t *iwb, int size) return -EINVAL; } src = iwb->read; - if (unlikely(limit > BAS_OUTBUFSIZE + BAS_OUTBUFPAD || + if (unlikely(limit >= BAS_OUTBUFSIZE + BAS_OUTBUFPAD || (read < src && limit >= src))) { pr_err("isoc write buffer frame reservation violated\n"); return -EFAULT; From 5b5f792a6a9a2f9ae812d151ed621f72e99b1725 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 21 May 2009 15:07:12 -0700 Subject: [PATCH 142/267] pktgen: do not access flows[] beyond its length typo -- pkt_dev->nflows is for stats only, the number of concurrent flows is stored in cflows. Reported-By: Vladimir Ivashchenko Signed-off-by: Florian Westphal Signed-off-by: David S. Miller --- net/core/pktgen.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/pktgen.c b/net/core/pktgen.c index 3779c1438c11..0666a827bc62 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -2447,7 +2447,7 @@ static inline void free_SAs(struct pktgen_dev *pkt_dev) if (pkt_dev->cflows) { /* let go of the SAs if we have them */ int i = 0; - for (; i < pkt_dev->nflows; i++){ + for (; i < pkt_dev->cflows; i++) { struct xfrm_state *x = pkt_dev->flows[i].x; if (x) { xfrm_state_put(x); From 3ed18d76d959e5cbfa5d70c8f7ba95476582a556 Mon Sep 17 00:00:00 2001 From: Robert Olsson Date: Thu, 21 May 2009 15:20:59 -0700 Subject: [PATCH 143/267] ipv4: Fix oops with FIB_TRIE It seems we can fix this by disabling preemption while we re-balance the trie. This is with the CONFIG_CLASSIC_RCU. It's been stress-tested at high loads continuesly taking a full BGP table up/down via iproute -batch. Note. fib_trie is not updated for CONFIG_PREEMPT_RCU Reported-by: Andrei Popa Signed-off-by: Robert Olsson Signed-off-by: David S. Miller --- net/ipv4/fib_trie.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index ec0ae490f0b6..33c7c85dfe40 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -986,9 +986,12 @@ fib_find_node(struct trie *t, u32 key) static struct node *trie_rebalance(struct trie *t, struct tnode *tn) { int wasfull; - t_key cindex, key = tn->key; + t_key cindex, key; struct tnode *tp; + preempt_disable(); + key = tn->key; + while (tn != NULL && (tp = node_parent((struct node *)tn)) != NULL) { cindex = tkey_extract_bits(key, tp->pos, tp->bits); wasfull = tnode_full(tp, tnode_get_child(tp, cindex)); @@ -1007,6 +1010,7 @@ static struct node *trie_rebalance(struct trie *t, struct tnode *tn) if (IS_TNODE(tn)) tn = (struct tnode *)resize(t, (struct tnode *)tn); + preempt_enable(); return (struct node *)tn; } From 0975ecba3b670df7c488a5e0e6fe9f1f370a8ad8 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 21 May 2009 15:22:02 -0700 Subject: [PATCH 144/267] RxRPC: Error handling for rxrpc_alloc_connection() rxrpc_alloc_connection() doesn't return an error code on failure, it just returns NULL. IS_ERR(NULL) is false. Signed-off-by: Dan Carpenter Signed-off-by: David Howells Signed-off-by: David S. Miller --- net/rxrpc/ar-connection.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/net/rxrpc/ar-connection.c b/net/rxrpc/ar-connection.c index 0f1218b8d289..67e38a056240 100644 --- a/net/rxrpc/ar-connection.c +++ b/net/rxrpc/ar-connection.c @@ -343,9 +343,9 @@ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx, /* not yet present - create a candidate for a new connection * and then redo the check */ conn = rxrpc_alloc_connection(gfp); - if (IS_ERR(conn)) { - _leave(" = %ld", PTR_ERR(conn)); - return PTR_ERR(conn); + if (!conn) { + _leave(" = -ENOMEM"); + return -ENOMEM; } conn->trans = trans; @@ -508,9 +508,9 @@ int rxrpc_connect_call(struct rxrpc_sock *rx, /* not yet present - create a candidate for a new connection and then * redo the check */ candidate = rxrpc_alloc_connection(gfp); - if (IS_ERR(candidate)) { - _leave(" = %ld", PTR_ERR(candidate)); - return PTR_ERR(candidate); + if (!candidate) { + _leave(" = -ENOMEM"); + return -ENOMEM; } candidate->trans = trans; From 703a3b8e5c01cf6fb33c6d8dc99905f889a4e992 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 21 May 2009 22:21:53 +0000 Subject: [PATCH 145/267] [CIFS] fix posix open regression Posix open code was not properly adding the file to the list of open files. Fix allocating cifsFileInfo more than once, and adding twice to flist and tlist. Also fix mode setting to be done in one place in these paths. Signed-off-by: Steve French Reviewed-by: Shirish Pargaonkar Tested-by: Jeff Layton Tested-by: Luca Tettamanti --- fs/cifs/dir.c | 14 +++++------ fs/cifs/file.c | 66 ++++++++++++++++++++++++++++---------------------- 2 files changed, 44 insertions(+), 36 deletions(-) diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 11431ed72a7f..f49d684edd96 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -225,6 +225,7 @@ int cifs_posix_open(char *full_path, struct inode **pinode, if (!(oflags & FMODE_READ)) write_only = true; + mode &= ~current_umask(); rc = CIFSPOSIXCreate(xid, cifs_sb->tcon, posix_flags, mode, pnetfid, presp_data, &oplock, full_path, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & @@ -310,7 +311,6 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, return -ENOMEM; } - mode &= ~current_umask(); if (oplockEnabled) oplock = REQ_OPLOCK; @@ -336,7 +336,7 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, else /* success, no need to query */ goto cifs_create_set_dentry; } else if ((rc != -EIO) && (rc != -EREMOTE) && - (rc != -EOPNOTSUPP)) /* path not found or net err */ + (rc != -EOPNOTSUPP) && (rc != -EINVAL)) goto cifs_create_out; /* else fallthrough to retry, using older open call, this is case where server does not support this SMB level, and @@ -609,7 +609,6 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, int xid; int rc = 0; /* to get around spurious gcc warning, set to zero here */ int oplock = 0; - int mode; __u16 fileHandle = 0; bool posix_open = false; struct cifs_sb_info *cifs_sb; @@ -660,13 +659,12 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, if (pTcon->unix_ext) { if (!(nd->flags & (LOOKUP_PARENT | LOOKUP_DIRECTORY)) && - (nd->flags & LOOKUP_OPEN)) { + (nd->flags & LOOKUP_OPEN) && !pTcon->broken_posix_open) { if (!((nd->intent.open.flags & O_CREAT) && (nd->intent.open.flags & O_EXCL))) { - mode = nd->intent.open.create_mode & - ~current_umask(); rc = cifs_posix_open(full_path, &newInode, - parent_dir_inode->i_sb, mode, + parent_dir_inode->i_sb, + nd->intent.open.create_mode, nd->intent.open.flags, &oplock, &fileHandle, xid); /* @@ -681,6 +679,8 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, */ if ((rc != -EINVAL) && (rc != -EOPNOTSUPP)) posix_open = true; + else + pTcon->broken_posix_open = true; } } if (!posix_open) diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 38c06f826575..302ea15f02e6 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -130,10 +130,6 @@ static inline int cifs_posix_open_inode_helper(struct inode *inode, struct cifsFileInfo *pCifsFile, int oplock, u16 netfid) { - file->private_data = kmalloc(sizeof(struct cifsFileInfo), GFP_KERNEL); - if (file->private_data == NULL) - return -ENOMEM; - pCifsFile = cifs_init_private(file->private_data, inode, file, netfid); write_lock(&GlobalSMBSeslock); pCifsInode = CIFS_I(file->f_path.dentry->d_inode); @@ -184,6 +180,38 @@ psx_client_can_cache: return 0; } +static struct cifsFileInfo * +cifs_fill_filedata(struct file *file) +{ + struct list_head *tmp; + struct cifsFileInfo *pCifsFile = NULL; + struct cifsInodeInfo *pCifsInode = NULL; + + /* search inode for this file and fill in file->private_data */ + pCifsInode = CIFS_I(file->f_path.dentry->d_inode); + read_lock(&GlobalSMBSeslock); + list_for_each(tmp, &pCifsInode->openFileList) { + pCifsFile = list_entry(tmp, struct cifsFileInfo, flist); + if ((pCifsFile->pfile == NULL) && + (pCifsFile->pid == current->tgid)) { + /* mode set in cifs_create */ + + /* needed for writepage */ + pCifsFile->pfile = file; + file->private_data = pCifsFile; + break; + } + } + read_unlock(&GlobalSMBSeslock); + + if (file->private_data != NULL) { + return pCifsFile; + } else if ((file->f_flags & O_CREAT) && (file->f_flags & O_EXCL)) + cERROR(1, ("could not find file instance for " + "new file %p", file)); + return NULL; +} + /* all arguments to this function must be checked for validity in caller */ static inline int cifs_open_inode_helper(struct inode *inode, struct file *file, struct cifsInodeInfo *pCifsInode, struct cifsFileInfo *pCifsFile, @@ -258,7 +286,6 @@ int cifs_open(struct inode *inode, struct file *file) struct cifsTconInfo *tcon; struct cifsFileInfo *pCifsFile; struct cifsInodeInfo *pCifsInode; - struct list_head *tmp; char *full_path = NULL; int desiredAccess; int disposition; @@ -270,32 +297,12 @@ int cifs_open(struct inode *inode, struct file *file) cifs_sb = CIFS_SB(inode->i_sb); tcon = cifs_sb->tcon; - /* search inode for this file and fill in file->private_data */ pCifsInode = CIFS_I(file->f_path.dentry->d_inode); - read_lock(&GlobalSMBSeslock); - list_for_each(tmp, &pCifsInode->openFileList) { - pCifsFile = list_entry(tmp, struct cifsFileInfo, - flist); - if ((pCifsFile->pfile == NULL) && - (pCifsFile->pid == current->tgid)) { - /* mode set in cifs_create */ - - /* needed for writepage */ - pCifsFile->pfile = file; - - file->private_data = pCifsFile; - break; - } - } - read_unlock(&GlobalSMBSeslock); - - if (file->private_data != NULL) { - rc = 0; + pCifsFile = cifs_fill_filedata(file); + if (pCifsFile) { FreeXid(xid); - return rc; - } else if ((file->f_flags & O_CREAT) && (file->f_flags & O_EXCL)) - cERROR(1, ("could not find file instance for " - "new file %p", file)); + return 0; + } full_path = build_path_from_dentry(file->f_path.dentry); if (full_path == NULL) { @@ -325,6 +332,7 @@ int cifs_open(struct inode *inode, struct file *file) /* no need for special case handling of setting mode on read only files needed here */ + pCifsFile = cifs_fill_filedata(file); cifs_posix_open_inode_helper(inode, file, pCifsInode, pCifsFile, oplock, netfid); goto out; From 37869fa2da1a952ed736858f8b9e45c0b9131ae9 Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Wed, 20 May 2009 14:30:06 +0000 Subject: [PATCH 146/267] sh: ap325 camera without i2c driver fix This patch fixes the ap325rxa ncm03j camera code to handle the case where no i2c driver is present. Without this fix i2c_transfer() may be passed NULL as adapter which results in a crash. Triggered when i2c-sh_mobile.c failed to probe() due to missing MSTP clocks. Signed-off-by: Magnus Damm Signed-off-by: Paul Mundt --- arch/sh/boards/board-ap325rxa.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/sh/boards/board-ap325rxa.c b/arch/sh/boards/board-ap325rxa.c index 39e46919df14..f2a29641b6a3 100644 --- a/arch/sh/boards/board-ap325rxa.c +++ b/arch/sh/boards/board-ap325rxa.c @@ -263,6 +263,9 @@ static int camera_probe(void) struct i2c_msg msg; int ret; + if (!a) + return -ENODEV; + camera_power(1); msg.addr = 0x6e; msg.buf = camera_ncm03j_magic; From 8e9bb19ef97d6594e735bee64b6d72103e350854 Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Wed, 20 May 2009 14:34:43 +0000 Subject: [PATCH 147/267] video: stop sh_mobile_lcdcfb only if started This patch fixes the LCDC driver to avoid calling the function sh_mobile_lcdc_start_stop(priv, 0) unless the same function has been called before to start the LCDC hardware. Triggered when sh_mobile_lcdcfb.c failed to probe() due to missing MSTP clocks. Signed-off-by: Magnus Damm Signed-off-by: Paul Mundt --- drivers/video/sh_mobile_lcdcfb.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/video/sh_mobile_lcdcfb.c b/drivers/video/sh_mobile_lcdcfb.c index 92ea0ab44ce2..f10d2fbeda06 100644 --- a/drivers/video/sh_mobile_lcdcfb.c +++ b/drivers/video/sh_mobile_lcdcfb.c @@ -47,6 +47,7 @@ struct sh_mobile_lcdc_priv { #endif unsigned long lddckr; struct sh_mobile_lcdc_chan ch[2]; + int started; }; /* shared registers */ @@ -451,6 +452,7 @@ static int sh_mobile_lcdc_start(struct sh_mobile_lcdc_priv *priv) /* start the lcdc */ sh_mobile_lcdc_start_stop(priv, 1); + priv->started = 1; /* tell the board code to enable the panel */ for (k = 0; k < ARRAY_SIZE(priv->ch); k++) { @@ -493,7 +495,10 @@ static void sh_mobile_lcdc_stop(struct sh_mobile_lcdc_priv *priv) } /* stop the lcdc */ - sh_mobile_lcdc_start_stop(priv, 0); + if (priv->started) { + sh_mobile_lcdc_start_stop(priv, 0); + priv->started = 0; + } /* stop clocks */ for (k = 0; k < ARRAY_SIZE(priv->ch); k++) From 6eb0ac03899a1363ba176abe0830a9e6698c0503 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Thu, 21 May 2009 19:10:23 +0000 Subject: [PATCH 148/267] powerpc/maple: Add a quirk to disable MSI for IPR on Bimini Something in the HW or FW setup is busted and MSIs aren't working with IPR on Bimini, so until we figure out exaxtly what's up, we quirk them out Signed-off-by: Michael Ellerman Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/maple/pci.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/arch/powerpc/platforms/maple/pci.c b/arch/powerpc/platforms/maple/pci.c index 301855263b81..04296ffff8bf 100644 --- a/arch/powerpc/platforms/maple/pci.c +++ b/arch/powerpc/platforms/maple/pci.c @@ -592,3 +592,17 @@ int maple_pci_get_legacy_ide_irq(struct pci_dev *pdev, int channel) } return irq; } + +static void __devinit quirk_ipr_msi(struct pci_dev *dev) +{ + /* Something prevents MSIs from the IPR from working on Bimini, + * and the driver has no smarts to recover. So disable MSI + * on it for now. */ + + if (machine_is(maple)) { + dev->no_msi = 1; + dev_info(&dev->dev, "Quirk disabled MSI\n"); + } +} +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_OBSIDIAN, + quirk_ipr_msi); From 87488957a68293357a94c8142de7d0ae17914912 Mon Sep 17 00:00:00 2001 From: Adam Williamson Date: Thu, 21 May 2009 18:32:59 -0400 Subject: [PATCH 149/267] ALSA: hda - fix audio on HP TX25xx series notebooks Fixes https://bugtrack.alsa-project.org/alsa-bug/view.php?id=4121 Taken from https://bugzilla.redhat.com/show_bug.cgi?id=498060 Signed-off-by: Adam Williamson Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index b8a0d3e79272..bcbb736f94f0 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -12058,6 +12058,7 @@ static struct snd_pci_quirk alc268_cfg_tbl[] = { SND_PCI_QUIRK(0x1028, 0x0253, "Dell OEM", ALC268_DELL), SND_PCI_QUIRK(0x1028, 0x02b0, "Dell Inspiron Mini9", ALC268_DELL), SND_PCI_QUIRK(0x103c, 0x30cc, "TOSHIBA", ALC268_TOSHIBA), + SND_PCI_QUIRK(0x103c, 0x30f1, "HP TX25xx series", ALC268_TOSHIBA), SND_PCI_QUIRK(0x1043, 0x1205, "ASUS W7J", ALC268_3ST), SND_PCI_QUIRK(0x1179, 0xff10, "TOSHIBA A205", ALC268_TOSHIBA), SND_PCI_QUIRK(0x1179, 0xff50, "TOSHIBA A305", ALC268_TOSHIBA), From 88dff4936c0a5fa53080cca68dc963a8a2a674b0 Mon Sep 17 00:00:00 2001 From: Zhang Rui Date: Fri, 22 May 2009 11:35:50 +0800 Subject: [PATCH 150/267] x86: DMI match for the Sony VGN-Z540N as it needs BIOS reboot x86: DMI match for the Sony VGN-Z540N as it needs BIOS reboot, see: http://bugzilla.kernel.org/show_bug.cgi?id=12901 [ Impact: fix hung reboot on certain systems ] Signed-off-by: Zhang Rui Cc: Len Brown LKML-Reference: <1242963350.32574.53.camel@rzhang-dt> Signed-off-by: Ingo Molnar --- arch/x86/kernel/reboot.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 1340dad417f4..667188e0b5a0 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -232,6 +232,14 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "Dell DXP061"), }, }, + { /* Handle problems with rebooting on Sony VGN-Z540N */ + .callback = set_bios_reboot, + .ident = "Sony VGN-Z540N", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Sony Corporation"), + DMI_MATCH(DMI_PRODUCT_NAME, "VGN-Z540N"), + }, + }, { } }; From e069c0cf7c169ae5a8bfdc8d083a5d66fbef73d8 Mon Sep 17 00:00:00 2001 From: Inaky Perez-Gonzalez Date: Fri, 8 May 2009 15:51:44 -0700 Subject: [PATCH 151/267] wimax/i2400m: usb: fix device reset on autosuspend while not yet idle When the i2400m is connected to a network, the host interface (USB) cannot be suspended. For that to happen, the device has to have negotiated with the basestation to put the link on IDLE state. If the host tries to put the device in standby while it is connected but not idle, the device resets, as the driver should not do that. To avoid triggering that, when the USB susbsytem requires the driver to autosuspend the device, the driver checks if the device is not yet idle. If it is not, the request is rejected (will be retried again later on after the autosuspend timeout). At some point the device will enter idle and the request will succeed (unless of course, there is network traffic, but at that point, there is no idle neither in the link or the host interface). Signed-off-by: Inaky Perez-Gonzalez --- drivers/net/wimax/i2400m/usb.c | 35 +++++++++++++++++++++++++++++----- 1 file changed, 30 insertions(+), 5 deletions(-) diff --git a/drivers/net/wimax/i2400m/usb.c b/drivers/net/wimax/i2400m/usb.c index ca4151a9e222..17851321b7fd 100644 --- a/drivers/net/wimax/i2400m/usb.c +++ b/drivers/net/wimax/i2400m/usb.c @@ -505,27 +505,52 @@ int i2400mu_suspend(struct usb_interface *iface, pm_message_t pm_msg) #ifdef CONFIG_PM struct usb_device *usb_dev = i2400mu->usb_dev; #endif + unsigned is_autosuspend = 0; struct i2400m *i2400m = &i2400mu->i2400m; +#ifdef CONFIG_PM + if (usb_dev->auto_pm > 0) + is_autosuspend = 1; +#endif + d_fnstart(3, dev, "(iface %p pm_msg %u)\n", iface, pm_msg.event); if (i2400m->updown == 0) goto no_firmware; - d_printf(1, dev, "fw up, requesting standby\n"); + if (i2400m->state == I2400M_SS_DATA_PATH_CONNECTED && is_autosuspend) { + /* ugh -- the device is connected and this suspend + * request is an autosuspend one (not a system standby + * / hibernate). + * + * The only way the device can go to standby is if the + * link with the base station is in IDLE mode; that + * were the case, we'd be in status + * I2400M_SS_CONNECTED_IDLE. But we are not. + * + * If we *tell* him to go power save now, it'll reset + * as a precautionary measure, so if this is an + * autosuspend thing, say no and it'll come back + * later, when the link is IDLE + */ + result = -EBADF; + d_printf(1, dev, "fw up, link up, not-idle, autosuspend: " + "not entering powersave\n"); + goto error_not_now; + } + d_printf(1, dev, "fw up: entering powersave\n"); atomic_dec(&i2400mu->do_autopm); result = i2400m_cmd_enter_powersave(i2400m); atomic_inc(&i2400mu->do_autopm); -#ifdef CONFIG_PM - if (result < 0 && usb_dev->auto_pm == 0) { + if (result < 0 && !is_autosuspend) { /* System suspend, can't fail */ dev_err(dev, "failed to suspend, will reset on resume\n"); result = 0; } -#endif if (result < 0) goto error_enter_powersave; i2400mu_notification_release(i2400mu); - d_printf(1, dev, "fw up, got standby\n"); + d_printf(1, dev, "powersave requested\n"); error_enter_powersave: +error_not_now: no_firmware: d_fnend(3, dev, "(iface %p pm_msg %u) = %d\n", iface, pm_msg.event, result); From d5046853634a8d73f28bad3cf68d182c4a99035d Mon Sep 17 00:00:00 2001 From: Ryusuke Konishi Date: Fri, 22 May 2009 20:36:21 +0900 Subject: [PATCH 152/267] nilfs2: fix memory leak in nilfs_ioctl_clean_segments This fixes a new memory leak problem in garbage collection. The problem was brought by the bugfix patch ("nilfs2: fix lock order reversal in nilfs_clean_segments ioctl"). Thanks to Kentaro Suzuki for finding this problem. Reported-by: Kentaro Suzuki Signed-off-by: Ryusuke Konishi --- fs/nilfs2/ioctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c index 50ff3f2cdf24..d6759b92006f 100644 --- a/fs/nilfs2/ioctl.c +++ b/fs/nilfs2/ioctl.c @@ -576,7 +576,7 @@ static int nilfs_ioctl_clean_segments(struct inode *inode, struct file *filp, ret = nilfs_clean_segments(inode->i_sb, argv, kbufs); out_free: - while (--n > 0) + while (--n >= 0) vfree(kbufs[n]); kfree(kbufs[4]); return ret; From 63d3892379f93b73ef905fb3449f4e4438a53b40 Mon Sep 17 00:00:00 2001 From: Wu Zhangjin Date: Thu, 21 May 2009 05:50:01 +0800 Subject: [PATCH 153/267] MIPS: Fix sparse warning in incompatiable argument type of clear_user. The type of the second argument of access_ok should be (void __user *). The unnecessary conversion of the clear_user address argument was causing sparse to emit warnings on the __chk_user_ptr check. Signed-off-by: Wu Zhangjin Signed-off-by: Ralf Baechle --- arch/mips/include/asm/uaccess.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/include/asm/uaccess.h b/arch/mips/include/asm/uaccess.h index 8de858f5449f..c2d53c18fd36 100644 --- a/arch/mips/include/asm/uaccess.h +++ b/arch/mips/include/asm/uaccess.h @@ -956,7 +956,7 @@ __clear_user(void __user *addr, __kernel_size_t size) void __user * __cl_addr = (addr); \ unsigned long __cl_size = (n); \ if (__cl_size && access_ok(VERIFY_WRITE, \ - ((unsigned long)(__cl_addr)), __cl_size)) \ + __cl_addr, __cl_size)) \ __cl_size = __clear_user(__cl_addr, __cl_size); \ __cl_size; \ }) From 63c901c7e6fb878805cd2f8f14fa3eee8c03ee84 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Fri, 22 May 2009 10:48:17 +0100 Subject: [PATCH 154/267] MIPS: IP32: Fix build error due to uninitialized variable. CC arch/mips/sgi-ip32/ip32-reset.o cc1: warnings being treated as errors arch/mips/sgi-ip32/ip32-reset.c: In function 'debounce': arch/mips/sgi-ip32/ip32-reset.c:97: error: 'reg_a' is used uninitialized in this function The issues is old but due to the volatile keyword gcc older than 4.4 did not warn about this obvious bug. Signed-off-by: Ralf Baechle --- arch/mips/sgi-ip32/ip32-reset.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/mips/sgi-ip32/ip32-reset.c b/arch/mips/sgi-ip32/ip32-reset.c index 667da932b7b2..cc549a9a99e8 100644 --- a/arch/mips/sgi-ip32/ip32-reset.c +++ b/arch/mips/sgi-ip32/ip32-reset.c @@ -94,6 +94,7 @@ static void debounce(unsigned long data) volatile unsigned char reg_a, reg_c, xctrl_a; reg_c = CMOS_READ(RTC_INTR_FLAGS); + reg_a = CMOS_READ(RTC_REG_A); CMOS_WRITE(reg_a | DS_REGA_DV0, RTC_REG_A); wbflush(); xctrl_a = CMOS_READ(DS_B1_XCTRL4A); From d2f82c2f70d56ba4623de25edb383fec01f43b89 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Fri, 22 May 2009 10:58:43 +0100 Subject: [PATCH 155/267] MIPS: IP32: Remove unnecessary if not even harmful volatile keywords. They are unneeded and as the issue fixed in lmo commit 63f7ec59053e3f850ab67a9938e631bcba64c6ce shows even harmful. Signed-off-by: Ralf Baechle --- arch/mips/sgi-ip32/ip32-reset.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/mips/sgi-ip32/ip32-reset.c b/arch/mips/sgi-ip32/ip32-reset.c index cc549a9a99e8..9b95d80ebc6e 100644 --- a/arch/mips/sgi-ip32/ip32-reset.c +++ b/arch/mips/sgi-ip32/ip32-reset.c @@ -53,7 +53,7 @@ static inline void ip32_machine_halt(void) static void ip32_machine_power_off(void) { - volatile unsigned char reg_a, xctrl_a, xctrl_b; + unsigned char reg_a, xctrl_a, xctrl_b; disable_irq(MACEISA_RTC_IRQ); reg_a = CMOS_READ(RTC_REG_A); @@ -91,7 +91,7 @@ static void blink_timeout(unsigned long data) static void debounce(unsigned long data) { - volatile unsigned char reg_a, reg_c, xctrl_a; + unsigned char reg_a, reg_c, xctrl_a; reg_c = CMOS_READ(RTC_INTR_FLAGS); reg_a = CMOS_READ(RTC_REG_A); @@ -138,7 +138,7 @@ static inline void ip32_power_button(void) static irqreturn_t ip32_rtc_int(int irq, void *dev_id) { - volatile unsigned char reg_c; + unsigned char reg_c; reg_c = CMOS_READ(RTC_INTR_FLAGS); if (!(reg_c & RTC_IRQF)) { From 8369d5fa63260cc54464b4687aa6a0f78402d98e Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Fri, 22 May 2009 16:23:36 +0200 Subject: [PATCH 156/267] ide: fix 40-wire cable detection for TSST SH-S202* ATAPI devices (v2) Since 2.6.26 we support UDMA66 on ATAPI devices requiring IVB quirk: commit 8588a2b732928b343233af9b1855705b8286bed4 ("ide: add SH-S202J to ivb_list[]") We also later added support for more such devices in: commit e97564f362a93f8c248246c19828895950341252 ("ide: More TSST drives with broken cable detection") and in: commit 3ced5c49bd2d1f2c7f769e3a54385883de63a652 ("ide: add TSSTcorp CDDVDW SH-S202H to ivb_list[]") It turns out that such devices lack cable detection altogether (which in turn results in incorrect detection of 40-wire cables by our current cable detection strategy) so always handle them by trusting host-side cable detection only. v2: Model detection fixup from Martin. Reported-and-tested-by: Martin Lottermoser Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/ide-iops.c | 21 +++++++++++++++++---- 1 file changed, 17 insertions(+), 4 deletions(-) diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c index c19a221b1e18..06fe002116ec 100644 --- a/drivers/ide/ide-iops.c +++ b/drivers/ide/ide-iops.c @@ -206,8 +206,6 @@ EXPORT_SYMBOL_GPL(ide_in_drive_list); /* * Early UDMA66 devices don't set bit14 to 1, only bit13 is valid. - * We list them here and depend on the device side cable detection for them. - * * Some optical devices with the buggy firmwares have the same problem. */ static const struct drive_list_entry ivb_list[] = { @@ -251,10 +249,25 @@ u8 eighty_ninty_three(ide_drive_t *drive) * - force bit13 (80c cable present) check also for !ivb devices * (unless the slave device is pre-ATA3) */ - if ((id[ATA_ID_HW_CONFIG] & 0x4000) || - (ivb && (id[ATA_ID_HW_CONFIG] & 0x2000))) + if (id[ATA_ID_HW_CONFIG] & 0x4000) return 1; + if (ivb) { + const char *model = (char *)&id[ATA_ID_PROD]; + + if (strstr(model, "TSSTcorp CDDVDW SH-S202")) { + /* + * These ATAPI devices always report 80c cable + * so we have to depend on the host in this case. + */ + if (hwif->cbl == ATA_CBL_PATA80) + return 1; + } else { + /* Depend on the device side cable detection. */ + if (id[ATA_ID_HW_CONFIG] & 0x2000) + return 1; + } + } no_80w: if (drive->dev_flags & IDE_DFLAG_UDMA33_WARNED) return 0; From e3b29f05124b07303088795396ff858811d2acb8 Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Fri, 22 May 2009 16:23:37 +0200 Subject: [PATCH 157/267] ide: fix OOPS during ide-cd error recovery On Tuesday 19 May 2009 20:29:28 Martin Lottermoser wrote: > hdc: cdrom_decode_status: error=0x40 <3>{ LastFailedSense=0x04 } > ide: failed opcode was: unknown > hdc: DMA disabled > ------------[ cut here ]------------ > kernel BUG at drivers/ide/ide-io.c:872! It is possible for ide-cd to ignore ide_error()'s return value under some circumstances. Workaround it in ide_intr() and ide_timer_expiry() by checking if there is a device/port reset pending currently. Fixes bug #13345: http://bugzilla.kernel.org/show_bug.cgi?id=13345 Reported-by: Martin Lottermoser Reported-and-tested-by: Modestas Vainius Cc: Borislav Petkov Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/ide-io.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c index 35dc38d3b2c5..6415a2e2ba87 100644 --- a/drivers/ide/ide-io.c +++ b/drivers/ide/ide-io.c @@ -696,7 +696,7 @@ void ide_timer_expiry (unsigned long data) } spin_lock_irq(&hwif->lock); enable_irq(hwif->irq); - if (startstop == ide_stopped) { + if (startstop == ide_stopped && hwif->polling == 0) { ide_unlock_port(hwif); plug_device = 1; } @@ -868,7 +868,7 @@ irqreturn_t ide_intr (int irq, void *dev_id) * same irq as is currently being serviced here, and Linux * won't allow another of the same (on any CPU) until we return. */ - if (startstop == ide_stopped) { + if (startstop == ide_stopped && hwif->polling == 0) { BUG_ON(hwif->handler); ide_unlock_port(hwif); plug_device = 1; From 26bfcf21e25fa090f099fa0ccf201424989cbd7b Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Fri, 22 May 2009 16:23:37 +0200 Subject: [PATCH 158/267] ide: fix printk() levels in ide_dump_ata[pi]_error() Fixes "<3>" in error messages like this one: hdc: cdrom_decode_status: error=0x40 <3>{ LastFailedSense=0x04 } Reported-by: Martin Lottermoser Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/ide-lib.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c index 56ff8c46c7d1..85b9bae111f6 100644 --- a/drivers/ide/ide-lib.c +++ b/drivers/ide/ide-lib.c @@ -91,7 +91,7 @@ static void ide_dump_sector(ide_drive_t *drive) static void ide_dump_ata_error(ide_drive_t *drive, u8 err) { - printk(KERN_ERR "{ "); + printk(KERN_CONT "{ "); if (err & ATA_ABORTED) printk(KERN_CONT "DriveStatusError "); if (err & ATA_ICRC) @@ -121,7 +121,7 @@ static void ide_dump_ata_error(ide_drive_t *drive, u8 err) static void ide_dump_atapi_error(ide_drive_t *drive, u8 err) { - printk(KERN_ERR "{ "); + printk(KERN_CONT "{ "); if (err & ATAPI_ILI) printk(KERN_CONT "IllegalLengthIndication "); if (err & ATAPI_EOM) From cc30137a221372c67a943ad9ea68121a2bd57a6e Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Fri, 22 May 2009 16:23:38 +0200 Subject: [PATCH 159/267] ide: improve failed opcode reporting Nowadays we (almost) always store the currently executing command in hwif->cmd so we can use it for the failed opcode reporting. Cc: Martin Lottermoser Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/ide-lib.c | 23 ++++------------------- 1 file changed, 4 insertions(+), 19 deletions(-) diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c index 85b9bae111f6..2148df836ce7 100644 --- a/drivers/ide/ide-lib.c +++ b/drivers/ide/ide-lib.c @@ -31,24 +31,6 @@ void ide_toggle_bounce(ide_drive_t *drive, int on) blk_queue_bounce_limit(drive->queue, addr); } -static void ide_dump_opcode(ide_drive_t *drive) -{ - struct request *rq = drive->hwif->rq; - struct ide_cmd *cmd = NULL; - - if (!rq) - return; - - if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) - cmd = rq->special; - - printk(KERN_ERR "ide: failed opcode was: "); - if (cmd == NULL) - printk(KERN_CONT "unknown\n"); - else - printk(KERN_CONT "0x%02x\n", cmd->tf.command); -} - u64 ide_get_lba_addr(struct ide_cmd *cmd, int lba48) { struct ide_taskfile *tf = &cmd->tf; @@ -179,7 +161,10 @@ u8 ide_dump_status(ide_drive_t *drive, const char *msg, u8 stat) else ide_dump_atapi_error(drive, err); } - ide_dump_opcode(drive); + + printk(KERN_ERR "%s: possibly failed opcode: 0x%02x\n", + drive->name, drive->hwif->cmd.tf.command); + return err; } EXPORT_SYMBOL(ide_dump_status); From 28ee9bc5cc42776e0364399b401a64906ac1ac8e Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Fri, 22 May 2009 16:23:38 +0200 Subject: [PATCH 160/267] ide: report timeouts in ide_busy_sleep() * change 'hwif' argument to 'drive' * report an error on timeout Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/ide-probe.c | 9 ++++++--- include/linux/ide.h | 2 +- 2 files changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c index 7f264ed1141b..c895ed52b2e8 100644 --- a/drivers/ide/ide-probe.c +++ b/drivers/ide/ide-probe.c @@ -295,7 +295,7 @@ int ide_dev_read_id(ide_drive_t *drive, u8 cmd, u16 *id) timeout = ((cmd == ATA_CMD_ID_ATA) ? WAIT_WORSTCASE : WAIT_PIDENTIFY) / 2; - if (ide_busy_sleep(hwif, timeout, use_altstatus)) + if (ide_busy_sleep(drive, timeout, use_altstatus)) return 1; /* wait for IRQ and ATA_DRQ */ @@ -316,8 +316,9 @@ int ide_dev_read_id(ide_drive_t *drive, u8 cmd, u16 *id) return rc; } -int ide_busy_sleep(ide_hwif_t *hwif, unsigned long timeout, int altstatus) +int ide_busy_sleep(ide_drive_t *drive, unsigned long timeout, int altstatus) { + ide_hwif_t *hwif = drive->hwif; u8 stat; timeout += jiffies; @@ -330,6 +331,8 @@ int ide_busy_sleep(ide_hwif_t *hwif, unsigned long timeout, int altstatus) return 0; } while (time_before(jiffies, timeout)); + printk(KERN_ERR "%s: timeout in %s\n", drive->name, __func__); + return 1; /* drive timed-out */ } @@ -420,7 +423,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd) tp_ops->dev_select(drive); msleep(50); tp_ops->exec_command(hwif, ATA_CMD_DEV_RESET); - (void)ide_busy_sleep(hwif, WAIT_WORSTCASE, 0); + (void)ide_busy_sleep(drive, WAIT_WORSTCASE, 0); rc = ide_dev_read_id(drive, cmd, id); } diff --git a/include/linux/ide.h b/include/linux/ide.h index ff65fffb078f..9fed365a598b 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -1109,7 +1109,7 @@ void ide_fix_driveid(u16 *); extern void ide_fixstring(u8 *, const int, const int); -int ide_busy_sleep(ide_hwif_t *, unsigned long, int); +int ide_busy_sleep(ide_drive_t *, unsigned long, int); int ide_wait_stat(ide_startstop_t *, ide_drive_t *, u8, u8, unsigned long); From 5993856e53fbc4b4f28e2d481deaebeb715b1267 Mon Sep 17 00:00:00 2001 From: Harald Welte Date: Fri, 22 May 2009 16:23:39 +0200 Subject: [PATCH 161/267] via82cxxx: Add VIA VX855 PCI Device ID This patch adds the PCI Device ID 0xc409 to the PCI ID table of via82cxxx.c, as well as the 0x8409 south bridge ID. This is required to make the IDE driver work on the VX855/VX875 integrated chipset. Signed-off-by: Harald Welte Cc: Joseph Chan Cc: Bruce Chang Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/via82cxxx.c | 2 ++ include/linux/pci_ids.h | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/ide/via82cxxx.c b/drivers/ide/via82cxxx.c index 3ff7231e4858..028de26a25fe 100644 --- a/drivers/ide/via82cxxx.c +++ b/drivers/ide/via82cxxx.c @@ -67,6 +67,7 @@ static struct via_isa_bridge { u8 udma_mask; u8 flags; } via_isa_bridges[] = { + { "vx855", PCI_DEVICE_ID_VIA_VX855, 0x00, 0x2f, ATA_UDMA6, VIA_BAD_AST }, { "vx800", PCI_DEVICE_ID_VIA_VX800, 0x00, 0x2f, ATA_UDMA6, VIA_BAD_AST }, { "cx700", PCI_DEVICE_ID_VIA_CX700, 0x00, 0x2f, ATA_UDMA6, VIA_BAD_AST }, { "vt8237s", PCI_DEVICE_ID_VIA_8237S, 0x00, 0x2f, ATA_UDMA6, VIA_BAD_AST }, @@ -474,6 +475,7 @@ static const struct pci_device_id via_pci_tbl[] = { { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_82C576_1), 0 }, { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_82C586_1), 0 }, { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_CX700_IDE), 0 }, + { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_VX855_IDE), 0 }, { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_6410), 1 }, { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_SATA_EIDE), 1 }, { 0, }, diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 06ba90c211a5..0f71812d67d3 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1406,7 +1406,7 @@ #define PCI_DEVICE_ID_VIA_82C598_1 0x8598 #define PCI_DEVICE_ID_VIA_838X_1 0xB188 #define PCI_DEVICE_ID_VIA_83_87XX_1 0xB198 -#define PCI_DEVICE_ID_VIA_C409_IDE 0XC409 +#define PCI_DEVICE_ID_VIA_VX855_IDE 0xC409 #define PCI_DEVICE_ID_VIA_ANON 0xFFFF #define PCI_VENDOR_ID_SIEMENS 0x110A From 9a2845c453d170e4e9b1437fa671dbf39b0e7bd8 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Wed, 20 May 2009 13:36:17 -0500 Subject: [PATCH 162/267] ipmi: fix ipmi_si modprobe hang Instead of queuing IPMB messages before channel initialization, just throw them away. Nobody will be listening for them at this point, anyway, and they will clog up the queue and nothing will be delivered if we queue them. Also set the current channel to the number of channels, as this value is used to tell if the channel information has been initialized. Signed-off-by: Corey Minyard Cc: Ferenc Wagner Cc: Dan Frazier Signed-off-by: Linus Torvalds --- drivers/char/ipmi/ipmi_msghandler.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index aa83a0865ec1..09050797c76a 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -2856,6 +2856,7 @@ int ipmi_register_smi(struct ipmi_smi_handlers *handlers, /* Assume a single IPMB channel at zero. */ intf->channels[0].medium = IPMI_CHANNEL_MEDIUM_IPMB; intf->channels[0].protocol = IPMI_CHANNEL_PROTOCOL_IPMB; + intf->curr_channel = IPMI_MAX_CHANNELS; } if (rv == 0) @@ -3648,13 +3649,13 @@ static int handle_new_recv_msg(ipmi_smi_t intf, } /* - ** We need to make sure the channels have been initialized. - ** The channel_handler routine will set the "curr_channel" - ** equal to or greater than IPMI_MAX_CHANNELS when all the - ** channels for this interface have been initialized. - */ + * We need to make sure the channels have been initialized. + * The channel_handler routine will set the "curr_channel" + * equal to or greater than IPMI_MAX_CHANNELS when all the + * channels for this interface have been initialized. + */ if (intf->curr_channel < IPMI_MAX_CHANNELS) { - requeue = 1; /* Just put the message back for now */ + requeue = 0; /* Throw the message away */ goto out; } From afe6d7e3c4a9aba020637f4ae15527a89ba31f21 Mon Sep 17 00:00:00 2001 From: Andreas Mohr Date: Fri, 22 May 2009 17:48:58 +0200 Subject: [PATCH 163/267] ALSA: Kill truncate warning by shortening Sigmatel-specific AC97 control name ALSA sound/core/control.c:232: Control name 'Sigmatel Surround Phase Inversion Playback Switch' truncated to 'Sigmatel Surround Phase Inversion Playback ' bootup message by omitting weird Sigmatel prefix in this case; also fix up the related ca0106 mixer control removal part by using identical naming there. Signed-off-by: Andreas Mohr Signed-off-by: Takashi Iwai --- sound/pci/ac97/ac97_patch.c | 7 +++++-- sound/pci/ca0106/ca0106_mixer.c | 2 +- 2 files changed, 6 insertions(+), 3 deletions(-) diff --git a/sound/pci/ac97/ac97_patch.c b/sound/pci/ac97/ac97_patch.c index 81bc93e5f1e3..7337abdbe4e3 100644 --- a/sound/pci/ac97/ac97_patch.c +++ b/sound/pci/ac97/ac97_patch.c @@ -958,10 +958,13 @@ static int patch_sigmatel_stac9708_3d(struct snd_ac97 * ac97) } static const struct snd_kcontrol_new snd_ac97_sigmatel_4speaker = -AC97_SINGLE("Sigmatel 4-Speaker Stereo Playback Switch", AC97_SIGMATEL_DAC2INVERT, 2, 1, 0); +AC97_SINGLE("Sigmatel 4-Speaker Stereo Playback Switch", + AC97_SIGMATEL_DAC2INVERT, 2, 1, 0); +/* "Sigmatel " removed due to excessive name length: */ static const struct snd_kcontrol_new snd_ac97_sigmatel_phaseinvert = -AC97_SINGLE("Sigmatel Surround Phase Inversion Playback Switch", AC97_SIGMATEL_DAC2INVERT, 3, 1, 0); +AC97_SINGLE("Surround Phase Inversion Playback Switch", + AC97_SIGMATEL_DAC2INVERT, 3, 1, 0); static const struct snd_kcontrol_new snd_ac97_sigmatel_controls[] = { AC97_SINGLE("Sigmatel DAC 6dB Attenuate", AC97_SIGMATEL_ANALOG, 1, 1, 0), diff --git a/sound/pci/ca0106/ca0106_mixer.c b/sound/pci/ca0106/ca0106_mixer.c index ad2888705d2a..c111efe61c3c 100644 --- a/sound/pci/ca0106/ca0106_mixer.c +++ b/sound/pci/ca0106/ca0106_mixer.c @@ -800,7 +800,7 @@ int __devinit snd_ca0106_mixer(struct snd_ca0106 *emu) "Capture Volume", "External Amplifier", "Sigmatel 4-Speaker Stereo Playback Switch", - "Sigmatel Surround Phase Inversion Playback ", + "Surround Phase Inversion Playback Switch", NULL }; static char *ca0106_rename_ctls[] = { From bca23dba760d6705c013f89113c46570378fb626 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Thu, 21 May 2009 11:46:16 -0700 Subject: [PATCH 164/267] x86, setup: revert ACPI 3 E820 extended attributes support Remove ACPI 3 E820 extended memory attributes support. At least one vendor actively set all the flags to zero, but left ECX on return at 24. This bug may be present in other BIOSes. The breakage functionally means the ACPI 3 flags are probably completely useless, and that no OS any time soon is going to rely on their existence. Therefore, drop support completely. We may want to revisit this question in the future, if we find ourselves actually needing the flags. This reverts all or part of the following checkins: cd670599b7b00d9263f6f11a05c0edeb9cbedaf3 c549e71d073a6e9a4847497344db28a784061455 However, retain the part from the latter commit that copies e820 into a temporary buffer; that is an unrelated BIOS workaround. Put in a comment to explain that part. See https://bugzilla.redhat.com/show_bug.cgi?id=499396 for some additional information. [ Impact: detect all memory on affected machines ] Reported-by: Thomas J. Baker Signed-off-by: H. Peter Anvin Acked-by: Len Brown Cc: Chuck Ebbert Cc: Kyle McMartin Cc: Matt Domsch --- arch/x86/boot/memory.c | 29 +++++++++++++---------------- 1 file changed, 13 insertions(+), 16 deletions(-) diff --git a/arch/x86/boot/memory.c b/arch/x86/boot/memory.c index 5054c2ddd1a0..74b3d2ba84e9 100644 --- a/arch/x86/boot/memory.c +++ b/arch/x86/boot/memory.c @@ -17,11 +17,6 @@ #define SMAP 0x534d4150 /* ASCII "SMAP" */ -struct e820_ext_entry { - struct e820entry std; - u32 ext_flags; -} __attribute__((packed)); - static int detect_memory_e820(void) { int count = 0; @@ -29,13 +24,21 @@ static int detect_memory_e820(void) u32 size, id, edi; u8 err; struct e820entry *desc = boot_params.e820_map; - static struct e820_ext_entry buf; /* static so it is zeroed */ + static struct e820entry buf; /* static so it is zeroed */ /* - * Set this here so that if the BIOS doesn't change this field - * but still doesn't change %ecx, we're still okay... + * Note: at least one BIOS is known which assumes that the + * buffer pointed to by one e820 call is the same one as + * the previous call, and only changes modified fields. Therefore, + * we use a temporary buffer and copy the results entry by entry. + * + * This routine deliberately does not try to account for + * ACPI 3+ extended attributes. This is because there are + * BIOSes in the field which report zero for the valid bit for + * all ranges, and we don't currently make any use of the + * other attribute bits. Revisit this if we see the extended + * attribute bits deployed in a meaningful way in the future. */ - buf.ext_flags = 1; do { size = sizeof buf; @@ -66,13 +69,7 @@ static int detect_memory_e820(void) break; } - /* ACPI 3.0 added the extended flags support. If bit 0 - in the extended flags is zero, we're supposed to simply - ignore the entry -- a backwards incompatible change! */ - if (size > 20 && !(buf.ext_flags & 1)) - continue; - - *desc++ = buf.std; + *desc++ = buf; count++; } while (next && count < ARRAY_SIZE(boot_params.e820_map)); From 14b60391587ab9b2207c4fb6281763a93ae85e0f Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Wed, 20 May 2009 16:47:08 -0400 Subject: [PATCH 165/267] i915: support 8xx desktop cursors MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit For some reason we never added 8xx desktop cursor support to the kernel. This patch fixes that. [krh: Also set the size on pre-i915 hw.] Tested-by: Kristian Høgsberg Signed-off-by: Jesse Barnes Signed-off-by: Eric Anholt --- drivers/gpu/drm/i915/i915_gem.c | 4 ++-- drivers/gpu/drm/i915/i915_reg.h | 17 +++++++++++++++++ drivers/gpu/drm/i915/intel_display.c | 26 ++++++++++++++++++++------ 3 files changed, 39 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 4a24c90fb940..717b6a854bcd 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1145,7 +1145,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) mutex_unlock(&dev->struct_mutex); return VM_FAULT_SIGBUS; } - list_add(&obj_priv->list, &dev_priv->mm.inactive_list); + list_add_tail(&obj_priv->list, &dev_priv->mm.inactive_list); } /* Need a new fence register? */ @@ -1375,7 +1375,7 @@ i915_gem_mmap_gtt_ioctl(struct drm_device *dev, void *data, mutex_unlock(&dev->struct_mutex); return ret; } - list_add(&obj_priv->list, &dev_priv->mm.inactive_list); + list_add_tail(&obj_priv->list, &dev_priv->mm.inactive_list); } drm_gem_object_unreference(obj); diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 15da44cf21b1..9668cc0d7f4e 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -1410,9 +1410,25 @@ /* Cursor A & B regs */ #define CURACNTR 0x70080 +/* Old style CUR*CNTR flags (desktop 8xx) */ +#define CURSOR_ENABLE 0x80000000 +#define CURSOR_GAMMA_ENABLE 0x40000000 +#define CURSOR_STRIDE_MASK 0x30000000 +#define CURSOR_FORMAT_SHIFT 24 +#define CURSOR_FORMAT_MASK (0x07 << CURSOR_FORMAT_SHIFT) +#define CURSOR_FORMAT_2C (0x00 << CURSOR_FORMAT_SHIFT) +#define CURSOR_FORMAT_3C (0x01 << CURSOR_FORMAT_SHIFT) +#define CURSOR_FORMAT_4C (0x02 << CURSOR_FORMAT_SHIFT) +#define CURSOR_FORMAT_ARGB (0x04 << CURSOR_FORMAT_SHIFT) +#define CURSOR_FORMAT_XRGB (0x05 << CURSOR_FORMAT_SHIFT) +/* New style CUR*CNTR flags */ +#define CURSOR_MODE 0x27 #define CURSOR_MODE_DISABLE 0x00 #define CURSOR_MODE_64_32B_AX 0x07 #define CURSOR_MODE_64_ARGB_AX ((1 << 5) | CURSOR_MODE_64_32B_AX) +#define MCURSOR_PIPE_SELECT (1 << 28) +#define MCURSOR_PIPE_A 0x00 +#define MCURSOR_PIPE_B (1 << 28) #define MCURSOR_GAMMA_ENABLE (1 << 26) #define CURABASE 0x70084 #define CURAPOS 0x70088 @@ -1420,6 +1436,7 @@ #define CURSOR_POS_SIGN 0x8000 #define CURSOR_X_SHIFT 0 #define CURSOR_Y_SHIFT 16 +#define CURSIZE 0x700a0 #define CURBCNTR 0x700c0 #define CURBBASE 0x700c4 #define CURBPOS 0x700c8 diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 3387cf32f385..c9d6f10ba92e 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -1357,7 +1357,7 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc, int pipe = intel_crtc->pipe; uint32_t control = (pipe == 0) ? CURACNTR : CURBCNTR; uint32_t base = (pipe == 0) ? CURABASE : CURBBASE; - uint32_t temp; + uint32_t temp = I915_READ(control); size_t addr; int ret; @@ -1366,7 +1366,12 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc, /* if we want to turn off the cursor ignore width and height */ if (!handle) { DRM_DEBUG("cursor off\n"); - temp = CURSOR_MODE_DISABLE; + if (IS_MOBILE(dev) || IS_I9XX(dev)) { + temp &= ~(CURSOR_MODE | MCURSOR_GAMMA_ENABLE); + temp |= CURSOR_MODE_DISABLE; + } else { + temp &= ~(CURSOR_ENABLE | CURSOR_GAMMA_ENABLE); + } addr = 0; bo = NULL; mutex_lock(&dev->struct_mutex); @@ -1409,10 +1414,19 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc, addr = obj_priv->phys_obj->handle->busaddr; } - temp = 0; - /* set the pipe for the cursor */ - temp |= (pipe << 28); - temp |= CURSOR_MODE_64_ARGB_AX | MCURSOR_GAMMA_ENABLE; + if (!IS_I9XX(dev)) + I915_WRITE(CURSIZE, (height << 12) | width); + + /* Hooray for CUR*CNTR differences */ + if (IS_MOBILE(dev) || IS_I9XX(dev)) { + temp &= ~(CURSOR_MODE | MCURSOR_PIPE_SELECT); + temp |= CURSOR_MODE_64_ARGB_AX | MCURSOR_GAMMA_ENABLE; + temp |= (pipe << 28); /* Connect to correct pipe */ + } else { + temp &= ~(CURSOR_FORMAT_MASK); + temp |= CURSOR_ENABLE; + temp |= CURSOR_FORMAT_ARGB | CURSOR_GAMMA_ENABLE; + } finish: I915_WRITE(control, temp); From 8863170628da4b0b461eb96bf797df1dca0bd03e Mon Sep 17 00:00:00 2001 From: Ma Ling Date: Wed, 13 May 2009 11:19:55 +0800 Subject: [PATCH 166/267] drm/i915: Fetch SDVO LVDS mode lines from VBT, then reserve them Signed-off-by: Ma Ling Signed-off-by: Eric Anholt --- drivers/gpu/drm/i915/i915_drv.h | 3 +- drivers/gpu/drm/i915/intel_bios.c | 104 +++++++++++++++++++++--------- drivers/gpu/drm/i915/intel_bios.h | 17 +++++ drivers/gpu/drm/i915/intel_lvds.c | 4 +- 4 files changed, 94 insertions(+), 34 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 9b149fe824c3..c431fa54bbb5 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -180,7 +180,8 @@ typedef struct drm_i915_private { int backlight_duty_cycle; /* restore backlight to this value */ bool panel_wants_dither; struct drm_display_mode *panel_fixed_mode; - struct drm_display_mode *vbt_mode; /* if any */ + struct drm_display_mode *lfp_lvds_vbt_mode; /* if any */ + struct drm_display_mode *sdvo_lvds_vbt_mode; /* if any */ /* Feature bits from the VBIOS */ unsigned int int_tv_support:1; diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c index fc28e2bbd542..9d78cff33b24 100644 --- a/drivers/gpu/drm/i915/intel_bios.c +++ b/drivers/gpu/drm/i915/intel_bios.c @@ -57,9 +57,43 @@ find_section(struct bdb_header *bdb, int section_id) return NULL; } -/* Try to find panel data */ static void -parse_panel_data(struct drm_i915_private *dev_priv, struct bdb_header *bdb) +fill_detail_timing_data(struct drm_display_mode *panel_fixed_mode, + struct lvds_dvo_timing *dvo_timing) +{ + panel_fixed_mode->hdisplay = (dvo_timing->hactive_hi << 8) | + dvo_timing->hactive_lo; + panel_fixed_mode->hsync_start = panel_fixed_mode->hdisplay + + ((dvo_timing->hsync_off_hi << 8) | dvo_timing->hsync_off_lo); + panel_fixed_mode->hsync_end = panel_fixed_mode->hsync_start + + dvo_timing->hsync_pulse_width; + panel_fixed_mode->htotal = panel_fixed_mode->hdisplay + + ((dvo_timing->hblank_hi << 8) | dvo_timing->hblank_lo); + + panel_fixed_mode->vdisplay = (dvo_timing->vactive_hi << 8) | + dvo_timing->vactive_lo; + panel_fixed_mode->vsync_start = panel_fixed_mode->vdisplay + + dvo_timing->vsync_off; + panel_fixed_mode->vsync_end = panel_fixed_mode->vsync_start + + dvo_timing->vsync_pulse_width; + panel_fixed_mode->vtotal = panel_fixed_mode->vdisplay + + ((dvo_timing->vblank_hi << 8) | dvo_timing->vblank_lo); + panel_fixed_mode->clock = dvo_timing->clock * 10; + panel_fixed_mode->type = DRM_MODE_TYPE_PREFERRED; + + /* Some VBTs have bogus h/vtotal values */ + if (panel_fixed_mode->hsync_end > panel_fixed_mode->htotal) + panel_fixed_mode->htotal = panel_fixed_mode->hsync_end + 1; + if (panel_fixed_mode->vsync_end > panel_fixed_mode->vtotal) + panel_fixed_mode->vtotal = panel_fixed_mode->vsync_end + 1; + + drm_mode_set_name(panel_fixed_mode); +} + +/* Try to find integrated panel data */ +static void +parse_lfp_panel_data(struct drm_i915_private *dev_priv, + struct bdb_header *bdb) { struct bdb_lvds_options *lvds_options; struct bdb_lvds_lfp_data *lvds_lfp_data; @@ -91,35 +125,9 @@ parse_panel_data(struct drm_i915_private *dev_priv, struct bdb_header *bdb) panel_fixed_mode = drm_calloc(1, sizeof(*panel_fixed_mode), DRM_MEM_DRIVER); - panel_fixed_mode->hdisplay = (dvo_timing->hactive_hi << 8) | - dvo_timing->hactive_lo; - panel_fixed_mode->hsync_start = panel_fixed_mode->hdisplay + - ((dvo_timing->hsync_off_hi << 8) | dvo_timing->hsync_off_lo); - panel_fixed_mode->hsync_end = panel_fixed_mode->hsync_start + - dvo_timing->hsync_pulse_width; - panel_fixed_mode->htotal = panel_fixed_mode->hdisplay + - ((dvo_timing->hblank_hi << 8) | dvo_timing->hblank_lo); + fill_detail_timing_data(panel_fixed_mode, dvo_timing); - panel_fixed_mode->vdisplay = (dvo_timing->vactive_hi << 8) | - dvo_timing->vactive_lo; - panel_fixed_mode->vsync_start = panel_fixed_mode->vdisplay + - dvo_timing->vsync_off; - panel_fixed_mode->vsync_end = panel_fixed_mode->vsync_start + - dvo_timing->vsync_pulse_width; - panel_fixed_mode->vtotal = panel_fixed_mode->vdisplay + - ((dvo_timing->vblank_hi << 8) | dvo_timing->vblank_lo); - panel_fixed_mode->clock = dvo_timing->clock * 10; - panel_fixed_mode->type = DRM_MODE_TYPE_PREFERRED; - - /* Some VBTs have bogus h/vtotal values */ - if (panel_fixed_mode->hsync_end > panel_fixed_mode->htotal) - panel_fixed_mode->htotal = panel_fixed_mode->hsync_end + 1; - if (panel_fixed_mode->vsync_end > panel_fixed_mode->vtotal) - panel_fixed_mode->vtotal = panel_fixed_mode->vsync_end + 1; - - drm_mode_set_name(panel_fixed_mode); - - dev_priv->vbt_mode = panel_fixed_mode; + dev_priv->lfp_lvds_vbt_mode = panel_fixed_mode; DRM_DEBUG("Found panel mode in BIOS VBT tables:\n"); drm_mode_debug_printmodeline(panel_fixed_mode); @@ -127,6 +135,39 @@ parse_panel_data(struct drm_i915_private *dev_priv, struct bdb_header *bdb) return; } +/* Try to find sdvo panel data */ +static void +parse_sdvo_panel_data(struct drm_i915_private *dev_priv, + struct bdb_header *bdb) +{ + struct bdb_sdvo_lvds_options *sdvo_lvds_options; + struct lvds_dvo_timing *dvo_timing; + struct drm_display_mode *panel_fixed_mode; + + dev_priv->sdvo_lvds_vbt_mode = NULL; + + sdvo_lvds_options = find_section(bdb, BDB_SDVO_LVDS_OPTIONS); + if (!sdvo_lvds_options) + return; + + dvo_timing = find_section(bdb, BDB_SDVO_PANEL_DTDS); + if (!dvo_timing) + return; + + panel_fixed_mode = drm_calloc(1, sizeof(*panel_fixed_mode), + DRM_MEM_DRIVER); + + if (!panel_fixed_mode) + return; + + fill_detail_timing_data(panel_fixed_mode, + dvo_timing + sdvo_lvds_options->panel_type); + + dev_priv->sdvo_lvds_vbt_mode = panel_fixed_mode; + + return; +} + static void parse_general_features(struct drm_i915_private *dev_priv, struct bdb_header *bdb) @@ -199,7 +240,8 @@ intel_init_bios(struct drm_device *dev) /* Grab useful general definitions */ parse_general_features(dev_priv, bdb); - parse_panel_data(dev_priv, bdb); + parse_lfp_panel_data(dev_priv, bdb); + parse_sdvo_panel_data(dev_priv, bdb); pci_unmap_rom(pdev, bios); diff --git a/drivers/gpu/drm/i915/intel_bios.h b/drivers/gpu/drm/i915/intel_bios.h index de621aad85b5..8ca2cde15804 100644 --- a/drivers/gpu/drm/i915/intel_bios.h +++ b/drivers/gpu/drm/i915/intel_bios.h @@ -279,6 +279,23 @@ struct vch_bdb_22 { struct vch_panel_data panels[16]; } __attribute__((packed)); +struct bdb_sdvo_lvds_options { + u8 panel_backlight; + u8 h40_set_panel_type; + u8 panel_type; + u8 ssc_clk_freq; + u16 als_low_trip; + u16 als_high_trip; + u8 sclalarcoeff_tab_row_num; + u8 sclalarcoeff_tab_row_size; + u8 coefficient[8]; + u8 panel_misc_bits_1; + u8 panel_misc_bits_2; + u8 panel_misc_bits_3; + u8 panel_misc_bits_4; +} __attribute__((packed)); + + bool intel_init_bios(struct drm_device *dev); /* diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c index 439a86514993..53731f0ffcb5 100644 --- a/drivers/gpu/drm/i915/intel_lvds.c +++ b/drivers/gpu/drm/i915/intel_lvds.c @@ -511,10 +511,10 @@ void intel_lvds_init(struct drm_device *dev) } /* Failed to get EDID, what about VBT? */ - if (dev_priv->vbt_mode) { + if (dev_priv->lfp_lvds_vbt_mode) { mutex_lock(&dev->mode_config.mutex); dev_priv->panel_fixed_mode = - drm_mode_duplicate(dev, dev_priv->vbt_mode); + drm_mode_duplicate(dev, dev_priv->lfp_lvds_vbt_mode); mutex_unlock(&dev->mode_config.mutex); if (dev_priv->panel_fixed_mode) { dev_priv->panel_fixed_mode->type |= From 7086c87fb1446ceb37918ffa0941359a7c2ec6cf Mon Sep 17 00:00:00 2001 From: Ma Ling Date: Wed, 13 May 2009 11:20:06 +0800 Subject: [PATCH 167/267] drm/i915: Return SDVO LVDS VBT mode if no EDID modes are detected. Some new SDVO LVDS hardware doesn't have DDC available, and this should fix the display on it. Signed-off-by: Ma Ling Signed-off-by: Eric Anholt --- drivers/gpu/drm/i915/intel_sdvo.c | 41 +++++++++++++++++++++++++++++++ 1 file changed, 41 insertions(+) diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index 9913651c1e17..f79ebf4de63c 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -69,6 +69,10 @@ struct intel_sdvo_priv { * This is set if we treat the device as HDMI, instead of DVI. */ bool is_hdmi; + /** + * This is set if we detect output of sdvo device as LVDS. + */ + bool is_lvds; /** * Returned SDTV resolutions allowed for the current format, if the @@ -1543,6 +1547,37 @@ static void intel_sdvo_get_tv_modes(struct drm_connector *connector) } } +static void intel_sdvo_get_lvds_modes(struct drm_connector *connector) +{ + struct intel_output *intel_output = to_intel_output(connector); + struct intel_sdvo_priv *sdvo_priv = intel_output->dev_priv; + struct drm_i915_private *dev_priv = connector->dev->dev_private; + + /* + * Attempt to get the mode list from DDC. + * Assume that the preferred modes are + * arranged in priority order. + */ + /* set the bus switch and get the modes */ + intel_sdvo_set_control_bus_switch(intel_output, sdvo_priv->ddc_bus); + intel_ddc_get_modes(intel_output); + if (list_empty(&connector->probed_modes) == false) + return; + + /* Fetch modes from VBT */ + if (dev_priv->sdvo_lvds_vbt_mode != NULL) { + struct drm_display_mode *newmode; + newmode = drm_mode_duplicate(connector->dev, + dev_priv->sdvo_lvds_vbt_mode); + if (newmode != NULL) { + /* Guarantee the mode is preferred */ + newmode->type = (DRM_MODE_TYPE_PREFERRED | + DRM_MODE_TYPE_DRIVER); + drm_mode_probed_add(connector, newmode); + } + } +} + static int intel_sdvo_get_modes(struct drm_connector *connector) { struct intel_output *output = to_intel_output(connector); @@ -1550,6 +1585,8 @@ static int intel_sdvo_get_modes(struct drm_connector *connector) if (sdvo_priv->is_tv) intel_sdvo_get_tv_modes(connector); + else if (sdvo_priv->is_lvds == true) + intel_sdvo_get_lvds_modes(connector); else intel_sdvo_get_ddc_modes(connector); @@ -1720,6 +1757,8 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device) } } + /* In defaut case sdvo lvds is false */ + sdvo_priv->is_lvds = false; intel_sdvo_get_capabilities(intel_output, &sdvo_priv->caps); if (sdvo_priv->caps.output_flags & @@ -1773,6 +1812,7 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device) connector->display_info.subpixel_order = SubPixelHorizontalRGB; encoder_type = DRM_MODE_ENCODER_LVDS; connector_type = DRM_MODE_CONNECTOR_LVDS; + sdvo_priv->is_lvds = true; } else if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_LVDS1) { @@ -1780,6 +1820,7 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device) connector->display_info.subpixel_order = SubPixelHorizontalRGB; encoder_type = DRM_MODE_ENCODER_LVDS; connector_type = DRM_MODE_CONNECTOR_LVDS; + sdvo_priv->is_lvds = true; } else { From ad5b2a6db3eddc41358d8a73f5cfe1c38e7e3a19 Mon Sep 17 00:00:00 2001 From: Jonas Bonn Date: Fri, 15 May 2009 09:10:41 +0200 Subject: [PATCH 168/267] drm/i915: Determine type before initialising connector drm_connector_init sets both the connector type and the connector type_id on the newly initialised connector. As the connector type_id is coupled to the connector type, the connector type cannot simply be modified on an initialised connector. This patch changes the order of operations on intel_sdvo_init so that the type is determined before the connector is intialised. This fixes a bug whereby the name card0-VGA-1 would be allocted to both a CRT and an SDVO connector since the SDVO connector would be initialised with type 'unknown' and hence have its type_id assigned from the wrong pool. Signed-off-by: Jonas Bonn Signed-off-by: Eric Anholt --- drivers/gpu/drm/i915/intel_sdvo.c | 29 ++++++++++------------------- 1 file changed, 10 insertions(+), 19 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index f79ebf4de63c..ded122c1ae2d 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -1713,17 +1713,9 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device) return false; } - connector = &intel_output->base; - - drm_connector_init(dev, connector, &intel_sdvo_connector_funcs, - DRM_MODE_CONNECTOR_Unknown); - drm_connector_helper_add(connector, &intel_sdvo_connector_helper_funcs); sdvo_priv = (struct intel_sdvo_priv *)(intel_output + 1); intel_output->type = INTEL_OUTPUT_SDVO; - connector->interlace_allowed = 0; - connector->doublescan_allowed = 0; - /* setup the DDC bus. */ if (output_device == SDVOB) i2cbus = intel_i2c_create(dev, GPIOE, "SDVOCTRL_E for SDVOB"); @@ -1731,7 +1723,7 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device) i2cbus = intel_i2c_create(dev, GPIOE, "SDVOCTRL_E for SDVOC"); if (!i2cbus) - goto err_connector; + goto err_inteloutput; sdvo_priv->i2c_bus = i2cbus; @@ -1747,7 +1739,6 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device) intel_output->i2c_bus = i2cbus; intel_output->dev_priv = sdvo_priv; - /* Read the regs to test if we can talk to the device */ for (i = 0; i < 0x40; i++) { if (!intel_sdvo_read_byte(intel_output, i, &ch[i])) { @@ -1768,7 +1759,6 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device) else sdvo_priv->controlled_output = SDVO_OUTPUT_TMDS1; - connector->display_info.subpixel_order = SubPixelHorizontalRGB; encoder_type = DRM_MODE_ENCODER_TMDS; connector_type = DRM_MODE_CONNECTOR_DVID; @@ -1786,7 +1776,6 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device) else if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_SVID0) { sdvo_priv->controlled_output = SDVO_OUTPUT_SVID0; - connector->display_info.subpixel_order = SubPixelHorizontalRGB; encoder_type = DRM_MODE_ENCODER_TVDAC; connector_type = DRM_MODE_CONNECTOR_SVIDEO; sdvo_priv->is_tv = true; @@ -1795,21 +1784,18 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device) else if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_RGB0) { sdvo_priv->controlled_output = SDVO_OUTPUT_RGB0; - connector->display_info.subpixel_order = SubPixelHorizontalRGB; encoder_type = DRM_MODE_ENCODER_DAC; connector_type = DRM_MODE_CONNECTOR_VGA; } else if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_RGB1) { sdvo_priv->controlled_output = SDVO_OUTPUT_RGB1; - connector->display_info.subpixel_order = SubPixelHorizontalRGB; encoder_type = DRM_MODE_ENCODER_DAC; connector_type = DRM_MODE_CONNECTOR_VGA; } else if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_LVDS0) { sdvo_priv->controlled_output = SDVO_OUTPUT_LVDS0; - connector->display_info.subpixel_order = SubPixelHorizontalRGB; encoder_type = DRM_MODE_ENCODER_LVDS; connector_type = DRM_MODE_CONNECTOR_LVDS; sdvo_priv->is_lvds = true; @@ -1817,7 +1803,6 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device) else if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_LVDS1) { sdvo_priv->controlled_output = SDVO_OUTPUT_LVDS1; - connector->display_info.subpixel_order = SubPixelHorizontalRGB; encoder_type = DRM_MODE_ENCODER_LVDS; connector_type = DRM_MODE_CONNECTOR_LVDS; sdvo_priv->is_lvds = true; @@ -1836,9 +1821,16 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device) goto err_i2c; } + connector = &intel_output->base; + drm_connector_init(dev, connector, &intel_sdvo_connector_funcs, + connector_type); + drm_connector_helper_add(connector, &intel_sdvo_connector_helper_funcs); + connector->interlace_allowed = 0; + connector->doublescan_allowed = 0; + connector->display_info.subpixel_order = SubPixelHorizontalRGB; + drm_encoder_init(dev, &intel_output->enc, &intel_sdvo_enc_funcs, encoder_type); drm_encoder_helper_add(&intel_output->enc, &intel_sdvo_helper_funcs); - connector->connector_type = connector_type; drm_mode_connector_attach_encoder(&intel_output->base, &intel_output->enc); drm_sysfs_connector_add(connector); @@ -1876,8 +1868,7 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device) err_i2c: intel_i2c_destroy(intel_output->i2c_bus); -err_connector: - drm_connector_cleanup(connector); +err_inteloutput: kfree(intel_output); return false; From 0c752a93353d9b17dbe148312d732fbe06d235e1 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Fri, 22 May 2009 12:17:45 -0700 Subject: [PATCH 169/267] x86: introduce noxsave boot parameter Introduce "noxsave" boot parameter which will disable the cpu's xsave/xrstor capabilities. Useful for debugging and working around xsave related issues. [ Impact: make it possible to debug problems in the field ] Signed-off-by: Suresh Siddha Signed-off-by: H. Peter Anvin --- Documentation/kernel-parameters.txt | 4 ++++ arch/x86/kernel/cpu/common.c | 7 +++++++ 2 files changed, 11 insertions(+) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index e87bdbfbcc75..fd5cac013037 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1535,6 +1535,10 @@ and is between 256 and 4096 characters. It is defined in the file register save and restore. The kernel will only save legacy floating-point registers on task switch. + noxsave [BUGS=X86] Disables x86 extended register state save + and restore using xsave. The kernel will fallback to + enabling legacy floating-point and sse state. + nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or wfi(ARM) instruction doesn't work correctly and not to use it. This is also useful when using JTAG debugger. diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index c1caefc82e62..77848d9fca68 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -114,6 +114,13 @@ DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = { } }; EXPORT_PER_CPU_SYMBOL_GPL(gdt_page); +static int __init x86_xsave_setup(char *s) +{ + setup_clear_cpu_cap(X86_FEATURE_XSAVE); + return 1; +} +__setup("noxsave", x86_xsave_setup); + #ifdef CONFIG_X86_32 static int cachesize_override __cpuinitdata = -1; static int disable_x86_serial_nr __cpuinitdata = 1; From 619ac3b75a1e9b2df66857f6a0fb466f1da5fa9e Mon Sep 17 00:00:00 2001 From: Ma Ling Date: Mon, 18 May 2009 16:12:46 +0800 Subject: [PATCH 170/267] drm/i915: Use an I2C algo to do the flip to SDVO DDC bus. Previously, we would set the control bus switch before calls were made to request EDID information over DDC. But recently the DDC code started doing multiple I2C transfers to get the EDID extensions as well. This tripped up SDVO, because the control bus switch is only in effect until the next STOP after a START. By doing our own algo, we can wrap each i2c transaction on the DDC I2C bus with the control bus switch it requires. freedesktop.org bug #21042 Signed-off-by: Ma Ling [anholt: Hand application for conflict, fixed error path] Signed-off-by: Eric Anholt --- drivers/gpu/drm/i915/intel_sdvo.c | 67 +++++++++++++++++++++++++++++-- 1 file changed, 63 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index ded122c1ae2d..f3ef6bfd8ffc 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -1402,10 +1402,8 @@ static enum drm_connector_status intel_sdvo_detect(struct drm_connector *connect static void intel_sdvo_get_ddc_modes(struct drm_connector *connector) { struct intel_output *intel_output = to_intel_output(connector); - struct intel_sdvo_priv *sdvo_priv = intel_output->dev_priv; /* set the bus switch and get the modes */ - intel_sdvo_set_control_bus_switch(intel_output, sdvo_priv->ddc_bus); intel_ddc_get_modes(intel_output); #if 0 @@ -1601,6 +1599,9 @@ static void intel_sdvo_destroy(struct drm_connector *connector) if (intel_output->i2c_bus) intel_i2c_destroy(intel_output->i2c_bus); + if (intel_output->ddc_bus) + intel_i2c_destroy(intel_output->ddc_bus); + drm_sysfs_connector_remove(connector); drm_connector_cleanup(connector); kfree(intel_output); @@ -1697,12 +1698,56 @@ intel_sdvo_get_digital_encoding_mode(struct intel_output *output) return true; } +static struct intel_output * +intel_sdvo_chan_to_intel_output(struct intel_i2c_chan *chan) +{ + struct drm_device *dev = chan->drm_dev; + struct drm_connector *connector; + struct intel_output *intel_output = NULL; + + list_for_each_entry(connector, + &dev->mode_config.connector_list, head) { + if (to_intel_output(connector)->ddc_bus == chan) { + intel_output = to_intel_output(connector); + break; + } + } + return intel_output; +} + +static int intel_sdvo_master_xfer(struct i2c_adapter *i2c_adap, + struct i2c_msg msgs[], int num) +{ + struct intel_output *intel_output; + struct intel_sdvo_priv *sdvo_priv; + struct i2c_algo_bit_data *algo_data; + struct i2c_algorithm *algo; + + algo_data = (struct i2c_algo_bit_data *)i2c_adap->algo_data; + intel_output = + intel_sdvo_chan_to_intel_output( + (struct intel_i2c_chan *)(algo_data->data)); + if (intel_output == NULL) + return -EINVAL; + + sdvo_priv = intel_output->dev_priv; + algo = (struct i2c_algorithm *)intel_output->i2c_bus->adapter.algo; + + intel_sdvo_set_control_bus_switch(intel_output, sdvo_priv->ddc_bus); + return algo->master_xfer(i2c_adap, msgs, num); +} + +static struct i2c_algorithm intel_sdvo_i2c_bit_algo = { + .master_xfer = intel_sdvo_master_xfer, +}; + bool intel_sdvo_init(struct drm_device *dev, int output_device) { struct drm_connector *connector; struct intel_output *intel_output; struct intel_sdvo_priv *sdvo_priv; struct intel_i2c_chan *i2cbus = NULL; + struct intel_i2c_chan *ddcbus = NULL; int connector_type; u8 ch[0x40]; int i; @@ -1748,6 +1793,20 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device) } } + /* setup the DDC bus. */ + if (output_device == SDVOB) + ddcbus = intel_i2c_create(dev, GPIOE, "SDVOB DDC BUS"); + else + ddcbus = intel_i2c_create(dev, GPIOE, "SDVOC DDC BUS"); + + if (ddcbus == NULL) + goto err_i2c; + + intel_sdvo_i2c_bit_algo.functionality = + intel_output->i2c_bus->adapter.algo->functionality; + ddcbus->adapter.algo = &intel_sdvo_i2c_bit_algo; + intel_output->ddc_bus = ddcbus; + /* In defaut case sdvo lvds is false */ sdvo_priv->is_lvds = false; intel_sdvo_get_capabilities(intel_output, &sdvo_priv->caps); @@ -1862,11 +1921,11 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device) sdvo_priv->caps.output_flags & (SDVO_OUTPUT_TMDS1 | SDVO_OUTPUT_RGB1) ? 'Y' : 'N'); - intel_output->ddc_bus = i2cbus; - return true; err_i2c: + if (ddcbus != NULL) + intel_i2c_destroy(intel_output->ddc_bus); intel_i2c_destroy(intel_output->i2c_bus); err_inteloutput: kfree(intel_output); From 0b827537e339c084ac9384df588969d400be9e0d Mon Sep 17 00:00:00 2001 From: "venkatesh.pallipadi@intel.com" Date: Fri, 22 May 2009 13:23:37 -0700 Subject: [PATCH 171/267] x86: bugfix wbinvd() model check instead of family check wbinvd is supported on all CPUs 486 or later. But, pageattr.c is checking x86_model >= 4 before wbinvd(), which looks like an oversight bug. It was first introduced at one place by changeset d7c8f21a8cad0228c7c5ce2bb6dbd95d1ee49d13 and got copied over to second place in the same file later. [ Impact: fix missing cache flush on early-model CPUs, potential data corruption ] Signed-off-by: Venkatesh Pallipadi Signed-off-by: H. Peter Anvin --- arch/x86/mm/pageattr.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 797f9f107cb6..2cc019a3f71b 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -153,7 +153,7 @@ static void __cpa_flush_all(void *arg) */ __flush_tlb_all(); - if (cache && boot_cpu_data.x86_model >= 4) + if (cache && boot_cpu_data.x86 >= 4) wbinvd(); } @@ -218,7 +218,7 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache, /* 4M threshold */ if (numpages >= 1024) { - if (boot_cpu_data.x86_model >= 4) + if (boot_cpu_data.x86 >= 4) wbinvd(); return; } From 0af48f42df15b97080b450d24219dd95db7b929a Mon Sep 17 00:00:00 2001 From: "venkatesh.pallipadi@intel.com" Date: Fri, 22 May 2009 13:23:38 -0700 Subject: [PATCH 172/267] x86: cpa_flush_array wbinvd should be done on all CPUs cpa_flush_array seems to prefer wbinvd() over clflush at 4M threshold. clflush needs to be done on only one CPU as per instruction definition. wbinvd() however, should be done on all CPUs. [ Impact: fix missing flush which could cause data corruption ] Signed-off-by: Venkatesh Pallipadi Signed-off-by: Suresh Siddha Signed-off-by: H. Peter Anvin --- arch/x86/mm/pageattr.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 2cc019a3f71b..0f9052bcec4b 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -204,6 +204,11 @@ static void cpa_flush_range(unsigned long start, int numpages, int cache) } } +static void wbinvd_local(void *unused) +{ + wbinvd(); +} + static void cpa_flush_array(unsigned long *start, int numpages, int cache, int in_flags, struct page **pages) { @@ -219,7 +224,8 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache, /* 4M threshold */ if (numpages >= 1024) { if (boot_cpu_data.x86 >= 4) - wbinvd(); + on_each_cpu(wbinvd_local, NULL, 1); + return; } /* From 95caa0a9bdaf93607bd0cc8932f53112496f2f22 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Fri, 22 May 2009 21:30:39 -0300 Subject: [PATCH 173/267] icom: fix rmmod crash Actually the icom driver is crashing when is being removed because the driver is kfreeing the adapter structure before calling pci_release_regions(), which result in the following error: Unable to handle kernel paging request for data at address 0x6b6b6b6b6b6b6d33 Faulting instruction address: 0xc000000000246b80 Oops: Kernel access of bad area, sig: 11 [#1] .... [c000000012d436a0] [c0000000001002d0] .kfree+0x120/0x34c (unreliable) [c000000012d43730] [c000000000246d60] .pci_release_selected_regions+0x3c/0x68 [c000000012d437c0] [d000000002d54700] .icom_kref_release+0xf4/0x118 [icom] [c000000012d43850] [c000000000232e50] .kref_put+0x74/0x94 [c000000012d438d0] [d000000002d56c58] .icom_remove+0x40/0xa4 [icom] [c000000012d43960] [c000000000249e48] .pci_device_remove+0x50/0x90 [c000000012d439e0] [c0000000002d68d8] .__device_release_driver+0x94/0xd4 [c000000012d43a70] [c0000000002d7104] .driver_detach+0xf8/0x12c [c000000012d43b00] [c0000000002d549c] .bus_remove_driver+0xbc/0x11c [c000000012d43b90] [c0000000002d71dc] .driver_unregister+0x60/0x80 [c000000012d43c20] [c00000000024a07c] .pci_unregister_driver+0x44/0xe8 [c000000012d43cb0] [d000000002d56bf4] .icom_exit+0x1c/0x40 [icom] [c000000012d43d30] [c000000000095fa8] .SyS_delete_module+0x214/0x2a8 [c000000012d43e30] [c00000000000852c] syscall_exit+0x0/0x40 Signed-off-by: Breno Leitao Cc: stable@kernel.org Cc: Alan Cox Signed-off-by: Linus Torvalds --- drivers/serial/icom.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/serial/icom.c b/drivers/serial/icom.c index 6579e2be1dd1..a461b3b2c72d 100644 --- a/drivers/serial/icom.c +++ b/drivers/serial/icom.c @@ -1472,8 +1472,8 @@ static void icom_remove_adapter(struct icom_adapter *icom_adapter) free_irq(icom_adapter->pci_dev->irq, (void *) icom_adapter); iounmap(icom_adapter->base_addr); - icom_free_adapter(icom_adapter); pci_release_regions(icom_adapter->pci_dev); + icom_free_adapter(icom_adapter); } static void icom_kref_release(struct kref *kref) From df391e0eda1e678add56a8e34226edf05d89af6a Mon Sep 17 00:00:00 2001 From: Henrik Rydberg Date: Sat, 23 May 2009 09:51:20 -0700 Subject: [PATCH 174/267] Input: multitouch - add tracking ID to the protocol MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There are a few multi-touch devices that support finger tracking well in hardware, Stantum being the prime example. By exposing the tracking ID in the MT protocol, evdev bandwidth and cpu usage in user space can be reduced. This patch adds the ABS_MT_TRACKING_ID to the MT protocol. Signed-off-by: Henrik Rydberg Tested-by: Stéphane Chatty Signed-off-by: Dmitry Torokhov --- drivers/input/input.c | 1 + include/linux/input.h | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/input/input.c b/drivers/input/input.c index e54e002665b0..5d445f48789b 100644 --- a/drivers/input/input.c +++ b/drivers/input/input.c @@ -42,6 +42,7 @@ static unsigned int input_abs_bypass_init_data[] __initdata = { ABS_MT_POSITION_Y, ABS_MT_TOOL_TYPE, ABS_MT_BLOB_ID, + ABS_MT_TRACKING_ID, 0 }; static unsigned long input_abs_bypass[BITS_TO_LONGS(ABS_CNT)]; diff --git a/include/linux/input.h b/include/linux/input.h index 0e6ff5de3588..6fed4f6a9c9e 100644 --- a/include/linux/input.h +++ b/include/linux/input.h @@ -656,6 +656,7 @@ struct input_absinfo { #define ABS_MT_POSITION_Y 0x36 /* Center Y ellipse position */ #define ABS_MT_TOOL_TYPE 0x37 /* Type of touching device */ #define ABS_MT_BLOB_ID 0x38 /* Group a set of packets as a blob */ +#define ABS_MT_TRACKING_ID 0x39 /* Unique ID of initiated contact */ #define ABS_MAX 0x3f #define ABS_CNT (ABS_MAX+1) From f9fcfc3b4627a1ec9b50411060f1b384926d6610 Mon Sep 17 00:00:00 2001 From: Henrik Rydberg Date: Sat, 23 May 2009 09:51:21 -0700 Subject: [PATCH 175/267] Input: multitouch - augment event semantics documentation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Through the collaboration to adapt the N-trig and Stantum HID drivers to the MT protocol, some semantic clarifications to the protocol have been made. This patch adds them to the MT documentation. Signed-off-by: Henrik Rydberg Tested-by: Stéphane Chatty Signed-off-by: Dmitry Torokhov --- Documentation/input/multi-touch-protocol.txt | 103 ++++++++++++++----- 1 file changed, 79 insertions(+), 24 deletions(-) diff --git a/Documentation/input/multi-touch-protocol.txt b/Documentation/input/multi-touch-protocol.txt index 9f09557aea39..a12ea3b586e6 100644 --- a/Documentation/input/multi-touch-protocol.txt +++ b/Documentation/input/multi-touch-protocol.txt @@ -18,8 +18,12 @@ Usage Anonymous finger details are sent sequentially as separate packets of ABS events. Only the ABS_MT events are recognized as part of a finger packet. The end of a packet is marked by calling the input_mt_sync() -function, which generates a SYN_MT_REPORT event. The end of multi-touch -transfer is marked by calling the usual input_sync() function. +function, which generates a SYN_MT_REPORT event. This instructs the +receiver to accept the data for the current finger and prepare to receive +another. The end of a multi-touch transfer is marked by calling the usual +input_sync() function. This instructs the receiver to act upon events +accumulated since last EV_SYN/SYN_REPORT and prepare to receive a new +set of events/packets. A set of ABS_MT events with the desired properties is defined. The events are divided into categories, to allow for partial implementation. The @@ -27,11 +31,26 @@ minimum set consists of ABS_MT_TOUCH_MAJOR, ABS_MT_POSITION_X and ABS_MT_POSITION_Y, which allows for multiple fingers to be tracked. If the device supports it, the ABS_MT_WIDTH_MAJOR may be used to provide the size of the approaching finger. Anisotropy and direction may be specified with -ABS_MT_TOUCH_MINOR, ABS_MT_WIDTH_MINOR and ABS_MT_ORIENTATION. Devices with -more granular information may specify general shapes as blobs, i.e., as a -sequence of rectangular shapes grouped together by an -ABS_MT_BLOB_ID. Finally, the ABS_MT_TOOL_TYPE may be used to specify -whether the touching tool is a finger or a pen or something else. +ABS_MT_TOUCH_MINOR, ABS_MT_WIDTH_MINOR and ABS_MT_ORIENTATION. The +ABS_MT_TOOL_TYPE may be used to specify whether the touching tool is a +finger or a pen or something else. Devices with more granular information +may specify general shapes as blobs, i.e., as a sequence of rectangular +shapes grouped together by an ABS_MT_BLOB_ID. Finally, for the few devices +that currently support it, the ABS_MT_TRACKING_ID event may be used to +report finger tracking from hardware [5]. + +Here is what a minimal event sequence for a two-finger touch would look +like: + + ABS_MT_TOUCH_MAJOR + ABS_MT_POSITION_X + ABS_MT_POSITION_Y + SYN_MT_REPORT + ABS_MT_TOUCH_MAJOR + ABS_MT_POSITION_X + ABS_MT_POSITION_Y + SYN_MT_REPORT + SYN_REPORT Event Semantics @@ -44,24 +63,24 @@ ABS_MT_TOUCH_MAJOR The length of the major axis of the contact. The length should be given in surface units. If the surface has an X times Y resolution, the largest -possible value of ABS_MT_TOUCH_MAJOR is sqrt(X^2 + Y^2), the diagonal. +possible value of ABS_MT_TOUCH_MAJOR is sqrt(X^2 + Y^2), the diagonal [4]. ABS_MT_TOUCH_MINOR The length, in surface units, of the minor axis of the contact. If the -contact is circular, this event can be omitted. +contact is circular, this event can be omitted [4]. ABS_MT_WIDTH_MAJOR The length, in surface units, of the major axis of the approaching tool. This should be understood as the size of the tool itself. The orientation of the contact and the approaching tool are assumed to be the -same. +same [4]. ABS_MT_WIDTH_MINOR The length, in surface units, of the minor axis of the approaching -tool. Omit if circular. +tool. Omit if circular [4]. The above four values can be used to derive additional information about the contact. The ratio ABS_MT_TOUCH_MAJOR / ABS_MT_WIDTH_MAJOR approximates @@ -70,14 +89,17 @@ different characteristic widths [1]. ABS_MT_ORIENTATION -The orientation of the ellipse. The value should describe half a revolution -clockwise around the touch center. The scale of the value is arbitrary, but -zero should be returned for an ellipse aligned along the Y axis of the -surface. As an example, an index finger placed straight onto the axis could -return zero orientation, something negative when twisted to the left, and -something positive when twisted to the right. This value can be omitted if -the touching object is circular, or if the information is not available in -the kernel driver. +The orientation of the ellipse. The value should describe a signed quarter +of a revolution clockwise around the touch center. The signed value range +is arbitrary, but zero should be returned for a finger aligned along the Y +axis of the surface, a negative value when finger is turned to the left, and +a positive value when finger turned to the right. When completely aligned with +the X axis, the range max should be returned. Orientation can be omitted +if the touching object is circular, or if the information is not available +in the kernel driver. Partial orientation support is possible if the device +can distinguish between the two axis, but not (uniquely) any values in +between. In such cases, the range of ABS_MT_ORIENTATION should be [0, 1] +[4]. ABS_MT_POSITION_X @@ -98,8 +120,35 @@ ABS_MT_BLOB_ID The BLOB_ID groups several packets together into one arbitrarily shaped contact. This is a low-level anonymous grouping, and should not be confused -with the high-level contactID, explained below. Most kernel drivers will -not have this capability, and can safely omit the event. +with the high-level trackingID [5]. Most kernel drivers will not have blob +capability, and can safely omit the event. + +ABS_MT_TRACKING_ID + +The TRACKING_ID identifies an initiated contact throughout its life cycle +[5]. There are currently only a few devices that support it, so this event +should normally be omitted. + + +Event Computation +----------------- + +The flora of different hardware unavoidably leads to some devices fitting +better to the MT protocol than others. To simplify and unify the mapping, +this section gives recipes for how to compute certain events. + +For devices reporting contacts as rectangular shapes, signed orientation +cannot be obtained. Assuming X and Y are the lengths of the sides of the +touching rectangle, here is a simple formula that retains the most +information possible: + + ABS_MT_TOUCH_MAJOR := max(X, Y) + ABS_MT_TOUCH_MINOR := min(X, Y) + ABS_MT_ORIENTATION := bool(X > Y) + +The range of ABS_MT_ORIENTATION should be set to [0, 1], to indicate that +the device can distinguish between a finger along the Y axis (0) and a +finger along the X axis (1). Finger Tracking @@ -109,14 +158,18 @@ The kernel driver should generate an arbitrary enumeration of the set of anonymous contacts currently on the surface. The order in which the packets appear in the event stream is not important. -The process of finger tracking, i.e., to assign a unique contactID to each +The process of finger tracking, i.e., to assign a unique trackingID to each initiated contact on the surface, is left to user space; preferably the -multi-touch X driver [3]. In that driver, the contactID stays the same and +multi-touch X driver [3]. In that driver, the trackingID stays the same and unique until the contact vanishes (when the finger leaves the surface). The problem of assigning a set of anonymous fingers to a set of identified fingers is a euclidian bipartite matching problem at each event update, and relies on a sufficiently rapid update rate. +There are a few devices that support trackingID in hardware. User space can +make use of these native identifiers to reduce bandwidth and cpu usage. + + Notes ----- @@ -136,5 +189,7 @@ could be used to derive tilt. time of writing (April 2009), the MT protocol is not yet merged, and the prototype implements finger matching, basic mouse support and two-finger scrolling. The project aims at improving the quality of current multi-touch -functionality available in the synaptics X driver, and in addition +functionality available in the Synaptics X driver, and in addition implement more advanced gestures. +[4] See the section on event computation. +[5] See the section on finger tracking. From 8db14ca12569fe885694bd3d5ff84c2d973d3cb0 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 23 May 2009 18:57:25 +0000 Subject: [PATCH 176/267] [CIFS] Avoid open on possible directories since Samba now rejects them Small change (mostly formatting) to limit lookup based open calls to file create only. After discussion yesteday on samba-technical about the posix lookup regression, and looking at a problem with cifs posix open to one particular Samba version, Jeff and JRA realized that Samba server's behavior changed in this area (posix open behavior on files vs. directories). To make this behavior consistent, JRA just made a fix to Samba server to alter how it handles open of directories (now returning the equivalent of EISDIR instead of success). Since we don't know at lookup time whether the inode is a directory or file (and thus whether posix open will succeed with most current Samba server), this change avoids the posix open code on lookup open (just issues posix open on creates). This gets the semantic benefits we want (atomicity, posix byte range locks, improved write semantics on newly created files) and file create still is fast, and we avoid the problem that Jeff noticed yesterday with "openat" (and some open directory calls) of non-cached directories to one version of Samba server, and will work with future Samba versions (which include the fix jra just pushed into Samba server). I confirmed this approach with jra yesterday and with Shirish today. Posix open is only called (at lookup time) for file create now. For opens (rather than creates), because we do not know if it is a file or directory yet, and current Samba no longer allows us to do posix open on dirs, we could end up wasting an open call on what turns out to be a dir. For file opens, we wait to call posix open till cifs_open. It could be added here (lookup) in the future but the performance tradeoff of the extra network request when EISDIR or EACCES is returned would have to be weighed against the 50% reduction in network traffic in the other paths. Reviewed-by: Shirish Pargaonkar Tested-by: Jeff Layton CC: Jeremy Allison Signed-off-by: Steve French --- fs/cifs/dir.c | 43 ++++++++++++++++++++++++------------------- 1 file changed, 24 insertions(+), 19 deletions(-) diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index f49d684edd96..3758965d73d5 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -657,31 +657,36 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, } cFYI(1, ("Full path: %s inode = 0x%p", full_path, direntry->d_inode)); + /* Posix open is only called (at lookup time) for file create now. + * For opens (rather than creates), because we do not know if it + * is a file or directory yet, and current Samba no longer allows + * us to do posix open on dirs, we could end up wasting an open call + * on what turns out to be a dir. For file opens, we wait to call posix + * open till cifs_open. It could be added here (lookup) in the future + * but the performance tradeoff of the extra network request when EISDIR + * or EACCES is returned would have to be weighed against the 50% + * reduction in network traffic in the other paths. + */ if (pTcon->unix_ext) { if (!(nd->flags & (LOOKUP_PARENT | LOOKUP_DIRECTORY)) && - (nd->flags & LOOKUP_OPEN) && !pTcon->broken_posix_open) { - if (!((nd->intent.open.flags & O_CREAT) && - (nd->intent.open.flags & O_EXCL))) { - rc = cifs_posix_open(full_path, &newInode, + (nd->flags & LOOKUP_OPEN) && !pTcon->broken_posix_open && + (nd->intent.open.flags & O_CREAT)) { + rc = cifs_posix_open(full_path, &newInode, parent_dir_inode->i_sb, nd->intent.open.create_mode, nd->intent.open.flags, &oplock, &fileHandle, xid); - /* - * This code works around a bug in - * samba posix open in samba versions 3.3.1 - * and earlier where create works - * but open fails with invalid parameter. - * If either of these error codes are - * returned, follow the normal lookup. - * Otherwise, the error during posix open - * is handled. - */ - if ((rc != -EINVAL) && (rc != -EOPNOTSUPP)) - posix_open = true; - else - pTcon->broken_posix_open = true; - } + /* + * The check below works around a bug in POSIX + * open in samba versions 3.3.1 and earlier where + * open could incorrectly fail with invalid parameter. + * If either that or op not supported returned, follow + * the normal lookup. + */ + if ((rc == 0) || (rc == -ENOENT)) + posix_open = true; + else if ((rc == -EINVAL) || (rc != -EOPNOTSUPP)) + pTcon->broken_posix_open = true; } if (!posix_open) rc = cifs_get_inode_info_unix(&newInode, full_path, From 59a3759d0fe8d969888c741bb33f4946e4d3750d Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 23 May 2009 14:47:00 -0700 Subject: [PATCH 177/267] Linux 2.6.30-rc7 --- Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Makefile b/Makefile index b57e1f539e83..739fd34a72a2 100644 --- a/Makefile +++ b/Makefile @@ -1,8 +1,8 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 30 -EXTRAVERSION = -rc6 -NAME = Vindictive Armadillo +EXTRAVERSION = -rc7 +NAME = Man-Eating Seals of Antiquity # *DOCUMENTATION* # To see a list of typical targets execute "make help" From 93574844bc3906941b89d6b6f72e01e87413f3c4 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ozan=20=C3=87a=C4=9Flayan?= Date: Sat, 23 May 2009 15:00:04 +0300 Subject: [PATCH 178/267] ALSA: hda - Add forced codec-slots for ASUS W5Fm MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit ASUS W5Fm needs the fixed codec-slots to probe to override the BIOS problem like W5F. Tested-by: Alp Kılıç Signed-off-by: Ozan Çağlayan Signed-off-by: Takashi Iwai --- sound/pci/hda/hda_intel.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index 21e99cfa8c49..3128e1a6bc65 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -2141,6 +2141,7 @@ static struct snd_pci_quirk probe_mask_list[] __devinitdata = { /* including bogus ALC268 in slot#2 that conflicts with ALC888 */ SND_PCI_QUIRK(0x17c0, 0x4085, "Medion MD96630", 0x01), /* forced codec slots */ + SND_PCI_QUIRK(0x1043, 0x1262, "ASUS W5Fm", 0x103), SND_PCI_QUIRK(0x1046, 0x1262, "ASUS W5F", 0x103), {} }; From 679d92ed1403b6cf9a19aa42ec62b81cae1aa017 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Sun, 24 May 2009 19:00:08 +0200 Subject: [PATCH 179/267] ALSA: hda - Add 5stack-no-fp model for STAC927x The recent fix for the headphone volume control on IDT/STAC codecs resulted in the removal of invalid "Side" volume eventually. But, if the front panel doesn't exist, this setup could be regarded as a sort of regression, as reported in kernel bug #13250. Now as a workaround, a new model 5stack-no-fp is added so that the user without the front panel can choose this one explicitly. Reference: bko#13250 http://bugzilla.kernel.org/show_bug.cgi?id=13250 Signed-off-by: Takashi Iwai --- Documentation/sound/alsa/HD-Audio-Models.txt | 1 + sound/pci/hda/patch_sigmatel.c | 10 ++++++++++ 2 files changed, 11 insertions(+) diff --git a/Documentation/sound/alsa/HD-Audio-Models.txt b/Documentation/sound/alsa/HD-Audio-Models.txt index 8eec05bc079e..322869fc8a9e 100644 --- a/Documentation/sound/alsa/HD-Audio-Models.txt +++ b/Documentation/sound/alsa/HD-Audio-Models.txt @@ -334,6 +334,7 @@ STAC9227/9228/9229/927x ref-no-jd Reference board without HP/Mic jack detection 3stack D965 3stack 5stack D965 5stack + SPDIF + 5stack-no-fp D965 5stack without front panel dell-3stack Dell Dimension E520 dell-bios Fixes with Dell BIOS setup auto BIOS setup (default) diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index 03b3646018a1..d2fd8ef6aef8 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -150,6 +150,7 @@ enum { STAC_D965_REF, STAC_D965_3ST, STAC_D965_5ST, + STAC_D965_5ST_NO_FP, STAC_DELL_3ST, STAC_DELL_BIOS, STAC_927X_MODELS @@ -2154,6 +2155,13 @@ static unsigned int d965_5st_pin_configs[14] = { 0x40000100, 0x40000100 }; +static unsigned int d965_5st_no_fp_pin_configs[14] = { + 0x40000100, 0x40000100, 0x0181304e, 0x01014010, + 0x01a19040, 0x01011012, 0x01016011, 0x40000100, + 0x40000100, 0x40000100, 0x40000100, 0x01442070, + 0x40000100, 0x40000100 +}; + static unsigned int dell_3st_pin_configs[14] = { 0x02211230, 0x02a11220, 0x01a19040, 0x01114210, 0x01111212, 0x01116211, 0x01813050, 0x01112214, @@ -2166,6 +2174,7 @@ static unsigned int *stac927x_brd_tbl[STAC_927X_MODELS] = { [STAC_D965_REF] = ref927x_pin_configs, [STAC_D965_3ST] = d965_3st_pin_configs, [STAC_D965_5ST] = d965_5st_pin_configs, + [STAC_D965_5ST_NO_FP] = d965_5st_no_fp_pin_configs, [STAC_DELL_3ST] = dell_3st_pin_configs, [STAC_DELL_BIOS] = NULL, }; @@ -2176,6 +2185,7 @@ static const char *stac927x_models[STAC_927X_MODELS] = { [STAC_D965_REF] = "ref", [STAC_D965_3ST] = "3stack", [STAC_D965_5ST] = "5stack", + [STAC_D965_5ST_NO_FP] = "5stack-no-fp", [STAC_DELL_3ST] = "dell-3stack", [STAC_DELL_BIOS] = "dell-bios", }; From 657cafa6b0f5296424d6f43f6f6eeb4a3222117e Mon Sep 17 00:00:00 2001 From: Alex Riesen Date: Sun, 24 May 2009 15:30:48 +0200 Subject: [PATCH 180/267] Use a format for linux_banner There is no format specifiers left in the linux_banner, and gcc-4.3 complains seeing the printk. Signed-off-by: Alex Riesen Signed-off-by: Linus Torvalds --- init/main.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/init/main.c b/init/main.c index 3bbf93be744c..d721dad05dd7 100644 --- a/init/main.c +++ b/init/main.c @@ -566,8 +566,7 @@ asmlinkage void __init start_kernel(void) tick_init(); boot_cpu_init(); page_address_init(); - printk(KERN_NOTICE); - printk(linux_banner); + printk(KERN_NOTICE "%s", linux_banner); setup_arch(&command_line); mm_init_owner(&init_mm, &init_task); setup_command_line(command_line); From 32bdfac5462d777f35b00838893c4f87baf23efe Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 24 May 2009 21:15:07 +0200 Subject: [PATCH 181/267] PM: Do not hold dpm_list_mtx while disabling/enabling nonboot CPUs We shouldn't hold dpm_list_mtx while executing [disable|enable]_nonboot_cpus(), because theoretically this may lead to a deadlock as shown by the following example (provided by Johannes Berg): CPU 3 CPU 2 CPU 1 suspend/hibernate something: rtnl_lock() device_pm_lock() -> mutex_lock(&dpm_list_mtx) mutex_lock(&dpm_list_mtx) linkwatch_work -> rtnl_lock() disable_nonboot_cpus() -> flush CPU 3 workqueue Fortunately, device drivers are supposed to stop any activities that might lead to the registration of new device objects way before disable_nonboot_cpus() is called, so it shouldn't be necessary to hold dpm_list_mtx over the entire late part of device suspend and early part of device resume. Thus, during the late suspend and the early resume of devices acquire dpm_list_mtx only when dpm_list is going to be traversed and release it right after that. This patch is reported to fix the regressions tracked as http://bugzilla.kernel.org/show_bug.cgi?id=13245. Signed-off-by: Rafael J. Wysocki Acked-by: Alan Stern Reported-by: Miles Lane Tested-by: Ming Lei --- drivers/base/power/main.c | 4 ++++ kernel/kexec.c | 2 -- kernel/power/disk.c | 21 +++------------------ kernel/power/main.c | 7 +------ 4 files changed, 8 insertions(+), 26 deletions(-) diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 69b4ddb7de3b..3e4bc699bc0f 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -357,6 +357,7 @@ static void dpm_power_up(pm_message_t state) { struct device *dev; + mutex_lock(&dpm_list_mtx); list_for_each_entry(dev, &dpm_list, power.entry) if (dev->power.status > DPM_OFF) { int error; @@ -366,6 +367,7 @@ static void dpm_power_up(pm_message_t state) if (error) pm_dev_err(dev, state, " early", error); } + mutex_unlock(&dpm_list_mtx); } /** @@ -614,6 +616,7 @@ int device_power_down(pm_message_t state) int error = 0; suspend_device_irqs(); + mutex_lock(&dpm_list_mtx); list_for_each_entry_reverse(dev, &dpm_list, power.entry) { error = suspend_device_noirq(dev, state); if (error) { @@ -622,6 +625,7 @@ int device_power_down(pm_message_t state) } dev->power.status = DPM_OFF_IRQ; } + mutex_unlock(&dpm_list_mtx); if (error) device_power_up(resume_event(state)); return error; diff --git a/kernel/kexec.c b/kernel/kexec.c index 5a758c6e4950..e4983770913b 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -1451,7 +1451,6 @@ int kernel_kexec(void) error = device_suspend(PMSG_FREEZE); if (error) goto Resume_console; - device_pm_lock(); /* At this point, device_suspend() has been called, * but *not* device_power_down(). We *must* * device_power_down() now. Otherwise, drivers for @@ -1489,7 +1488,6 @@ int kernel_kexec(void) enable_nonboot_cpus(); device_power_up(PMSG_RESTORE); Resume_devices: - device_pm_unlock(); device_resume(PMSG_RESTORE); Resume_console: resume_console(); diff --git a/kernel/power/disk.c b/kernel/power/disk.c index b0dc9e7a0d17..5cb080e7eebd 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -215,8 +215,6 @@ static int create_image(int platform_mode) if (error) return error; - device_pm_lock(); - /* At this point, device_suspend() has been called, but *not* * device_power_down(). We *must* call device_power_down() now. * Otherwise, drivers for some devices (e.g. interrupt controllers) @@ -227,7 +225,7 @@ static int create_image(int platform_mode) if (error) { printk(KERN_ERR "PM: Some devices failed to power down, " "aborting hibernation\n"); - goto Unlock; + return error; } error = platform_pre_snapshot(platform_mode); @@ -280,9 +278,6 @@ static int create_image(int platform_mode) device_power_up(in_suspend ? (error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE); - Unlock: - device_pm_unlock(); - return error; } @@ -344,13 +339,11 @@ static int resume_target_kernel(bool platform_mode) { int error; - device_pm_lock(); - error = device_power_down(PMSG_QUIESCE); if (error) { printk(KERN_ERR "PM: Some devices failed to power down, " "aborting resume\n"); - goto Unlock; + return error; } error = platform_pre_restore(platform_mode); @@ -403,9 +396,6 @@ static int resume_target_kernel(bool platform_mode) device_power_up(PMSG_RECOVER); - Unlock: - device_pm_unlock(); - return error; } @@ -464,11 +454,9 @@ int hibernation_platform_enter(void) goto Resume_devices; } - device_pm_lock(); - error = device_power_down(PMSG_HIBERNATE); if (error) - goto Unlock; + goto Resume_devices; error = hibernation_ops->prepare(); if (error) @@ -493,9 +481,6 @@ int hibernation_platform_enter(void) device_power_up(PMSG_RESTORE); - Unlock: - device_pm_unlock(); - Resume_devices: entering_platform_hibernation = false; device_resume(PMSG_RESTORE); diff --git a/kernel/power/main.c b/kernel/power/main.c index f99ed6a75eac..868028280d13 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -289,12 +289,10 @@ static int suspend_enter(suspend_state_t state) { int error; - device_pm_lock(); - if (suspend_ops->prepare) { error = suspend_ops->prepare(); if (error) - goto Done; + return error; } error = device_power_down(PMSG_SUSPEND); @@ -343,9 +341,6 @@ static int suspend_enter(suspend_state_t state) if (suspend_ops->finish) suspend_ops->finish(); - Done: - device_pm_unlock(); - return error; } From d5a877e8dd409d8c702986d06485c374b705d340 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Sun, 24 May 2009 13:03:43 -0700 Subject: [PATCH 182/267] async: make sure independent async domains can't accidentally entangle The problem occurs when async_synchronize_full_domain() is called when the async_pending list is not empty. This will cause lowest_running() to return the cookie of the first entry on the async_pending list, which might be nothing at all to do with the domain being asked for and thus cause the domain synchronization to wait for an unrelated domain. This can cause a deadlock if domain synchronization is used from one domain to wait for another. Fix by running over the async_pending list to see if any pending items actually belong to our domain (and return their cookies if they do). Signed-off-by: James Bottomley Signed-off-by: Arjan van de Ven Signed-off-by: Linus Torvalds --- kernel/async.c | 20 ++++++++++++-------- 1 file changed, 12 insertions(+), 8 deletions(-) diff --git a/kernel/async.c b/kernel/async.c index 968ef9457d4e..50540301ed0f 100644 --- a/kernel/async.c +++ b/kernel/async.c @@ -92,19 +92,23 @@ extern int initcall_debug; static async_cookie_t __lowest_in_progress(struct list_head *running) { struct async_entry *entry; + async_cookie_t ret = next_cookie; /* begin with "infinity" value */ + if (!list_empty(running)) { entry = list_first_entry(running, struct async_entry, list); - return entry->cookie; - } else if (!list_empty(&async_pending)) { - entry = list_first_entry(&async_pending, - struct async_entry, list); - return entry->cookie; - } else { - /* nothing in progress... next_cookie is "infinity" */ - return next_cookie; + ret = entry->cookie; } + if (!list_empty(&async_pending)) { + list_for_each_entry(entry, &async_pending, list) + if (entry->running == running) { + ret = entry->cookie; + break; + } + } + + return ret; } static async_cookie_t lowest_in_progress(struct list_head *running) From 71c9d8b68b299bef614afc7907393564a9f1476f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 25 May 2009 12:01:59 +0900 Subject: [PATCH 183/267] x86: Remove remap percpu allocator for the time being Remap percpu allocator has subtle bug when combined with page attribute changing. Remap percpu allocator aliases PMD pages for the first chunk and as pageattr doesn't know about the alias it ends up updating page attributes of the original mapping thus leaving the alises in inconsistent state which might lead to subtle data corruption. Please read the following threads for more information: http://thread.gmane.org/gmane.linux.kernel/835783 The following is the proposed fix which teaches pageattr about percpu aliases. http://thread.gmane.org/gmane.linux.kernel/837157 However, the above changes are deemed too pervasive for upstream inclusion for 2.6.30 release, so this patch essentially disables the remap allocator for the time being. Signed-off-by: Tejun Heo LKML-Reference: <4A1A0A27.4050301@kernel.org> Signed-off-by: Ingo Molnar --- arch/x86/kernel/setup_percpu.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index 3a97a4cf1872..8f0e13be36b3 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -160,8 +160,10 @@ static ssize_t __init setup_pcpu_remap(size_t static_size) /* * If large page isn't supported, there's no benefit in doing * this. Also, on non-NUMA, embedding is better. + * + * NOTE: disabled for now. */ - if (!cpu_has_pse || !pcpu_need_numa()) + if (true || !cpu_has_pse || !pcpu_need_numa()) return -EINVAL; /* From 461c6c3a0a23a8fac1a4b636e93ff5dfe599a241 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Mon, 25 May 2009 08:06:02 +0200 Subject: [PATCH 184/267] ALSA: hda - Add missing check of pin vref 50 and others in Realtek codecs Some Realtek codecs like ALC861 seem to support only VREF50 while the current driver assumes it's only VREF80. Check other VREF bits to set the correct value. Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index bcbb736f94f0..0fd258eba3a5 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -776,6 +776,12 @@ static void alc_set_input_pin(struct hda_codec *codec, hda_nid_t nid, pincap = (pincap & AC_PINCAP_VREF) >> AC_PINCAP_VREF_SHIFT; if (pincap & AC_PINCAP_VREF_80) val = PIN_VREF80; + else if (pincap & AC_PINCAP_VREF_50) + val = PIN_VREF50; + else if (pincap & AC_PINCAP_VREF_100) + val = PIN_VREF100; + else if (pincap & AC_PINCAP_VREF_GRD) + val = PIN_VREFGRD; } snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, val); } From 4e2fd555199977c5994d1a4d2d3b8761b20ca4c7 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Mon, 25 May 2009 00:42:34 -0700 Subject: [PATCH 185/267] gianfar: fix BUG under load after introduction of skb recycling Since commit 0fd56bb5be6455d0d42241e65aed057244665e5e ("gianfar: Add support for skb recycling"), gianfar puts skbuffs that are in the rx ring back onto the recycle list as-is in case there was a receive error, but this breaks the following invariant: that all skbuffs on the recycle list have skb->data = skb->head + NET_SKB_PAD. The RXBUF_ALIGNMENT realignment done in gfar_new_skb() will be done twice on skbuffs recycled in this way, causing there not to be enough room in the skb anymore to receive a full packet, eventually leading to an skb_over_panic from gfar_clean_rx_ring() -> skb_put(). Resetting the skb->data pointer to skb->head + NET_SKB_PAD before putting the skb back onto the recycle list restores the mentioned invariant, and should fix this issue. Reported-by: Michael Guntsche Tested-by: Michael Guntsche Signed-off-by: Lennert Buytenhek Signed-off-by: David S. Miller --- drivers/net/gianfar.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c index b2c49679bba7..a0519184e54e 100644 --- a/drivers/net/gianfar.c +++ b/drivers/net/gianfar.c @@ -1885,8 +1885,17 @@ int gfar_clean_rx_ring(struct net_device *dev, int rx_work_limit) if (unlikely(!newskb)) newskb = skb; - else if (skb) + else if (skb) { + /* + * We need to reset ->data to what it + * was before gfar_new_skb() re-aligned + * it to an RXBUF_ALIGNMENT boundary + * before we put the skb back on the + * recycle list. + */ + skb->data = skb->head + NET_SKB_PAD; __skb_queue_head(&priv->rx_recycle, skb); + } } else { /* Increment the number of packets */ dev->stats.rx_packets++; From a8cd0244e9cebcf9b358d24c7e7410062f3665cb Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Sun, 24 May 2009 22:15:25 +0300 Subject: [PATCH 186/267] KVM: Make paravirt tlb flush also reload the PAE PDPTRs The paravirt tlb flush may be used not only to flush TLBs, but also to reload the four page-directory-pointer-table entries, as it is used as a replacement for reloading CR3. Change the code to do the entire CR3 reloading dance instead of simply flushing the TLB. Cc: stable@kernel.org Signed-off-by: Avi Kivity --- arch/x86/kvm/mmu.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index b6caf1329b1b..32cf11e5728a 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -2897,8 +2897,7 @@ static int kvm_pv_mmu_write(struct kvm_vcpu *vcpu, static int kvm_pv_mmu_flush_tlb(struct kvm_vcpu *vcpu) { - kvm_x86_ops->tlb_flush(vcpu); - set_bit(KVM_REQ_MMU_SYNC, &vcpu->requests); + kvm_set_cr3(vcpu, vcpu->arch.cr3); return 1; } From a2edf57f510cce6a389cc14e58c6ad0a4296d6f9 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Sun, 24 May 2009 22:19:00 +0300 Subject: [PATCH 187/267] KVM: Fix PDPTR reloading on CR4 writes The processor is documented to reload the PDPTRs while in PAE mode if any of the CR4 bits PSE, PGE, or PAE change. Linux relies on this behaviour when zapping the low mappings of PAE kernels during boot. The code already handled changes to CR4.PAE; augment it to also notice changes to PSE and PGE. This triggered while booting an F11 PAE kernel; the futex initialization code runs before any CR3 reloads and writes to a NULL pointer; the futex subsystem ended up uninitialized, killing PI futexes and pulseaudio which uses them. Cc: stable@kernel.org Signed-off-by: Avi Kivity --- arch/x86/kvm/x86.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 49079a46687b..3944e917e794 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -338,6 +338,9 @@ EXPORT_SYMBOL_GPL(kvm_lmsw); void kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) { + unsigned long old_cr4 = vcpu->arch.cr4; + unsigned long pdptr_bits = X86_CR4_PGE | X86_CR4_PSE | X86_CR4_PAE; + if (cr4 & CR4_RESERVED_BITS) { printk(KERN_DEBUG "set_cr4: #GP, reserved bits\n"); kvm_inject_gp(vcpu, 0); @@ -351,7 +354,8 @@ void kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) kvm_inject_gp(vcpu, 0); return; } - } else if (is_paging(vcpu) && !is_pae(vcpu) && (cr4 & X86_CR4_PAE) + } else if (is_paging(vcpu) && (cr4 & X86_CR4_PAE) + && ((cr4 ^ old_cr4) & pdptr_bits) && !load_pdptrs(vcpu, vcpu->arch.cr3)) { printk(KERN_DEBUG "set_cr4: #GP, pdptrs reserved bits\n"); kvm_inject_gp(vcpu, 0); From 62e1e389f87a8839ad83b08c44691d1df8320846 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 26 May 2009 09:40:59 +1000 Subject: [PATCH 188/267] md: always update level / chunk_size / layout when writing v1.x metadata. We previously didn't update these fields when writing the metadata because they could never change. They can now, so we better write them. v0.90 metadata always updated these fields. Signed-off-by: NeilBrown --- drivers/md/md.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/md/md.c b/drivers/md/md.c index fccc8343a250..aa79d55875f0 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -1375,6 +1375,9 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev) sb->raid_disks = cpu_to_le32(mddev->raid_disks); sb->size = cpu_to_le64(mddev->dev_sectors); + sb->chunksize = cpu_to_le32(mddev->chunk_size >> 9); + sb->level = cpu_to_le32(mddev->level); + sb->layout = cpu_to_le32(mddev->layout); if (mddev->bitmap && mddev->bitmap_file == NULL) { sb->bitmap_offset = cpu_to_le32((__u32)mddev->bitmap_offset); From 2b69c83924396ad1eda36fdd267c9d2f360f5555 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 26 May 2009 09:41:17 +1000 Subject: [PATCH 189/267] md: improve errno return when setting array_size Instead of always returns EINVAL if anything goes wrong when setting the array size, add the option of E2BIG if the size requested is too large. This makes it easier for user-space to be sure what went wrong. Signed-off-by: NeilBrown --- drivers/md/md.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index aa79d55875f0..58e0b02a74c2 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -3683,7 +3683,7 @@ array_size_store(mddev_t *mddev, const char *buf, size_t len) if (strict_blocks_to_sectors(buf, §ors) < 0) return -EINVAL; if (mddev->pers && mddev->pers->size(mddev, 0, 0) < sectors) - return -EINVAL; + return -E2BIG; mddev->external_size = 1; } From be512691036cc989c11d0f418187efbbf14468e6 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 26 May 2009 09:41:17 +1000 Subject: [PATCH 190/267] md: bitmap: improve bitmap maintenance code. The code for checking which bits in the bitmap can be cleared has 2 problems: 1/ it repeatedly takes and drops a spinlock, where it would make more sense to just hold on to it most of the time. 2/ it doesn't make use of some opportunities to skip large sections of the bitmap This patch fixes those. It will only affect CPU consumption, not correctness. Signed-off-by: NeilBrown --- drivers/md/bitmap.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index 47c68bc75a17..56df1cee8fb3 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -1097,14 +1097,12 @@ void bitmap_daemon_work(struct bitmap *bitmap) } bitmap->allclean = 1; + spin_lock_irqsave(&bitmap->lock, flags); for (j = 0; j < bitmap->chunks; j++) { bitmap_counter_t *bmc; - spin_lock_irqsave(&bitmap->lock, flags); - if (!bitmap->filemap) { + if (!bitmap->filemap) /* error or shutdown */ - spin_unlock_irqrestore(&bitmap->lock, flags); break; - } page = filemap_get_page(bitmap, j); @@ -1121,6 +1119,8 @@ void bitmap_daemon_work(struct bitmap *bitmap) write_page(bitmap, page, 0); bitmap->allclean = 0; } + spin_lock_irqsave(&bitmap->lock, flags); + j |= (PAGE_BITS - 1); continue; } @@ -1181,9 +1181,10 @@ void bitmap_daemon_work(struct bitmap *bitmap) ext2_clear_bit(file_page_offset(j), paddr); kunmap_atomic(paddr, KM_USER0); } - } - spin_unlock_irqrestore(&bitmap->lock, flags); + } else + j |= PAGE_COUNTER_MASK; } + spin_unlock_irqrestore(&bitmap->lock, flags); /* now sync the final page */ if (lastpage != NULL) { From b6a9ce688f613e2ee5f15e6720e0bb8520efc36e Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 26 May 2009 09:41:17 +1000 Subject: [PATCH 191/267] md: export 'frozen' resync state through sysfs The md resync engine has a 'frozen' state which ensures that no resync/recovery. This is used to avoid races. Export this state through the 'sync_action' sysfs attribute so that user-space can benefit and also avoid some races. Signed-off-by: NeilBrown --- drivers/md/md.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index 58e0b02a74c2..384e4f0904c8 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -3306,7 +3306,9 @@ static ssize_t action_show(mddev_t *mddev, char *page) { char *type = "idle"; - if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) || + if (test_bit(MD_RECOVERY_FROZEN, &mddev->recovery)) + type = "frozen"; + else if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) || (!mddev->ro && test_bit(MD_RECOVERY_NEEDED, &mddev->recovery))) { if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery)) type = "reshape"; @@ -3329,7 +3331,12 @@ action_store(mddev_t *mddev, const char *page, size_t len) if (!mddev->pers || !mddev->pers->sync_request) return -EINVAL; - if (cmd_match(page, "idle")) { + if (cmd_match(page, "frozen")) + set_bit(MD_RECOVERY_FROZEN, &mddev->recovery); + else + clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery); + + if (cmd_match(page, "idle") || cmd_match(page, "frozen")) { if (mddev->sync_thread) { set_bit(MD_RECOVERY_INTR, &mddev->recovery); md_unregister_thread(mddev->sync_thread); From 848b3182365fdf5a05bcd5ed949071cac2c894b3 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 26 May 2009 12:41:08 +1000 Subject: [PATCH 192/267] md: raid5: avoid sector values going negative when testing reshape progress. As sector_t in unsigned, we cannot afford to let 'safepos' etc go negative. So replace a -= b; by a -= min(b,a); Signed-off-by: NeilBrown --- drivers/md/raid5.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 4616bc3a6e71..3c3626d2a1f9 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -3811,13 +3811,13 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped safepos = conf->reshape_safe; sector_div(safepos, data_disks); if (mddev->delta_disks < 0) { - writepos -= reshape_sectors; + writepos -= min(reshape_sectors, writepos); readpos += reshape_sectors; safepos += reshape_sectors; } else { writepos += reshape_sectors; - readpos -= reshape_sectors; - safepos -= reshape_sectors; + readpos -= min(reshape_sectors, readpos); + safepos -= min(reshape_sectors, safepos); } /* 'writepos' is the most advanced device address we might write. From 7a91ee1f628ef6bfe3f13067c0ddf9db520cb86b Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 26 May 2009 12:57:21 +1000 Subject: [PATCH 193/267] md: don't update curr_resync_completed without also updating reshape_position. In order for the metadata to always be consistent, we mustn't updated curr_resync_completed without also updating reshape_position. The reshape code updates both at the same time. However since commit 97e4f42d62badb0f9fbc27c013e89bc1336a03bc the common md_do_sync will sometimes update curr_resync_completed but is not in a position to update reshape_position. So if MD_RECOVERY_RESHAPE is set (indicating that a reshape is happening, so reshape_position might change), don't update curr_resync_completed in md_do_sync, leave it to the per-personality reshape code. Signed-off-by: NeilBrown --- drivers/md/md.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index 384e4f0904c8..954456532acc 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -6362,12 +6362,13 @@ void md_do_sync(mddev_t *mddev) skipped = 0; - if ((mddev->curr_resync > mddev->curr_resync_completed && - (mddev->curr_resync - mddev->curr_resync_completed) - > (max_sectors >> 4)) || - (j - mddev->curr_resync_completed)*2 - >= mddev->resync_max - mddev->curr_resync_completed - ) { + if (!test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) && + ((mddev->curr_resync > mddev->curr_resync_completed && + (mddev->curr_resync - mddev->curr_resync_completed) + > (max_sectors >> 4)) || + (j - mddev->curr_resync_completed)*2 + >= mddev->resync_max - mddev->curr_resync_completed + )) { /* time to update curr_resync_completed */ blk_unplug(mddev->queue); wait_event(mddev->recovery_wait, From b492b852cd8c99505708152c29a5e09a787af9de Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 26 May 2009 12:57:36 +1000 Subject: [PATCH 194/267] md: don't use locked_ioctl. md has no need for the BKL - it does its own locking. So md_ioctl doesn't need to be a locked_ioctl. Signed-off-by: NeilBrown --- drivers/md/md.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index 954456532acc..641b211fe3fe 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -5567,7 +5567,7 @@ static struct block_device_operations md_fops = .owner = THIS_MODULE, .open = md_open, .release = md_release, - .locked_ioctl = md_ioctl, + .ioctl = md_ioctl, .getgeo = md_getgeo, .media_changed = md_media_changed, .revalidate_disk= md_revalidate, From 8e35961b57da14cb64cb0e4e1b7e3aabda6396fe Mon Sep 17 00:00:00 2001 From: Hideo Saito Date: Sun, 24 May 2009 15:33:34 +0000 Subject: [PATCH 195/267] powerpc/mm: Fix broken MMU PID stealing on !SMP The recent rework of the MMU PID handling for non-hash CPUs has a subtle bug in the !SMP "optimized" variant of the PID stealing function. It clears the PID in the mm context before it calls local_flush_tlb_mm(). However, the later will not flush anything if the PID in the context is clear... Signed-off-by: Hideo Saito Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/mm/mmu_context_nohash.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/powerpc/mm/mmu_context_nohash.c b/arch/powerpc/mm/mmu_context_nohash.c index a70e311bd457..030d0005b4d2 100644 --- a/arch/powerpc/mm/mmu_context_nohash.c +++ b/arch/powerpc/mm/mmu_context_nohash.c @@ -127,12 +127,12 @@ static unsigned int steal_context_up(unsigned int id) pr_debug("[%d] steal context %d from mm @%p\n", cpu, id, mm); - /* Mark this mm has having no context anymore */ - mm->context.id = MMU_NO_CONTEXT; - /* Flush the TLB for that context */ local_flush_tlb_mm(mm); + /* Mark this mm has having no context anymore */ + mm->context.id = MMU_NO_CONTEXT; + /* XXX This clear should ultimately be part of local_flush_tlb_mm */ __clear_bit(id, stale_map[cpu]); From 217cbfa856dc1cbc2890781626c4032d9e3ec59f Mon Sep 17 00:00:00 2001 From: Finn Thain Date: Mon, 25 May 2009 22:43:49 -0700 Subject: [PATCH 196/267] mac8390: fix regression caused during net_device_ops conversion Changeset ca17584bf2ad1b1e37a5c0e4386728cc5fc9dabc ("mac8390: update to net_device_ops") broke mac8390 by adding 8390.o to the link. That meant that lib8390.c was included twice, once in mac8390.c and once in 8390.c, subject to different macros. This patch reverts that by avoiding the wrappers in 8390.c. They seem to be of no value since COMPAT_NET_DEV_OPS is going away soon. Tested with a Kinetics EtherPort card. Signed-off-by: Finn Thain Signed-off-by: David S. Miller --- drivers/net/Makefile | 2 +- drivers/net/mac8390.c | 10 +++++----- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/net/Makefile b/drivers/net/Makefile index 1fc4602a6ff2..a1c25cb4669f 100644 --- a/drivers/net/Makefile +++ b/drivers/net/Makefile @@ -102,7 +102,7 @@ obj-$(CONFIG_HAMACHI) += hamachi.o obj-$(CONFIG_NET) += Space.o loopback.o obj-$(CONFIG_SEEQ8005) += seeq8005.o obj-$(CONFIG_NET_SB1000) += sb1000.o -obj-$(CONFIG_MAC8390) += mac8390.o 8390.o +obj-$(CONFIG_MAC8390) += mac8390.o obj-$(CONFIG_APNE) += apne.o 8390.o obj-$(CONFIG_PCMCIA_PCNET) += 8390.o obj-$(CONFIG_HP100) += hp100.o diff --git a/drivers/net/mac8390.c b/drivers/net/mac8390.c index 8e884869a05b..f26667d5eaae 100644 --- a/drivers/net/mac8390.c +++ b/drivers/net/mac8390.c @@ -304,7 +304,7 @@ struct net_device * __init mac8390_probe(int unit) if (!MACH_IS_MAC) return ERR_PTR(-ENODEV); - dev = alloc_ei_netdev(); + dev = ____alloc_ei_netdev(0); if (!dev) return ERR_PTR(-ENOMEM); @@ -481,10 +481,10 @@ void cleanup_module(void) static const struct net_device_ops mac8390_netdev_ops = { .ndo_open = mac8390_open, .ndo_stop = mac8390_close, - .ndo_start_xmit = ei_start_xmit, - .ndo_tx_timeout = ei_tx_timeout, - .ndo_get_stats = ei_get_stats, - .ndo_set_multicast_list = ei_set_multicast_list, + .ndo_start_xmit = __ei_start_xmit, + .ndo_tx_timeout = __ei_tx_timeout, + .ndo_get_stats = __ei_get_stats, + .ndo_set_multicast_list = __ei_set_multicast_list, .ndo_validate_addr = eth_validate_addr, .ndo_set_mac_address = eth_mac_addr, .ndo_change_mtu = eth_change_mtu, From c80a5cdfc5ca6533cb893154f546370da1fdb8f0 Mon Sep 17 00:00:00 2001 From: Doug Leith Date: Mon, 25 May 2009 22:44:59 -0700 Subject: [PATCH 197/267] tcp: tcp_vegas ssthresh bugfix This patch fixes ssthresh accounting issues in tcp_vegas when cwnd decreases Signed-off-by: Doug Leith Signed-off-by: David S. Miller --- net/ipv4/tcp_vegas.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/net/ipv4/tcp_vegas.c b/net/ipv4/tcp_vegas.c index a453aac91bd3..c6743eec9b7d 100644 --- a/net/ipv4/tcp_vegas.c +++ b/net/ipv4/tcp_vegas.c @@ -158,6 +158,11 @@ void tcp_vegas_cwnd_event(struct sock *sk, enum tcp_ca_event event) } EXPORT_SYMBOL_GPL(tcp_vegas_cwnd_event); +static inline u32 tcp_vegas_ssthresh(struct tcp_sock *tp) +{ + return min(tp->snd_ssthresh, tp->snd_cwnd-1); +} + static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 in_flight) { struct tcp_sock *tp = tcp_sk(sk); @@ -221,11 +226,10 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 in_flight) */ diff = tp->snd_cwnd * (rtt-vegas->baseRTT) / vegas->baseRTT; - if (diff > gamma && tp->snd_ssthresh > 2 ) { + if (diff > gamma && tp->snd_cwnd <= tp->snd_ssthresh) { /* Going too fast. Time to slow down * and switch to congestion avoidance. */ - tp->snd_ssthresh = 2; /* Set cwnd to match the actual rate * exactly: @@ -235,6 +239,7 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 in_flight) * utilization. */ tp->snd_cwnd = min(tp->snd_cwnd, (u32)target_cwnd+1); + tp->snd_ssthresh = tcp_vegas_ssthresh(tp); } else if (tp->snd_cwnd <= tp->snd_ssthresh) { /* Slow start. */ @@ -250,6 +255,8 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 in_flight) * we slow down. */ tp->snd_cwnd--; + tp->snd_ssthresh + = tcp_vegas_ssthresh(tp); } else if (diff < alpha) { /* We don't have enough extra packets * in the network, so speed up. From 46176b4f6bac19454b7b5c35f68594b85850a600 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 26 May 2009 14:42:40 +0900 Subject: [PATCH 198/267] x86, relocs: ignore R_386_NONE in kernel relocation entries For relocatable 32bit kernels, boot/compressed/relocs.c processes relocation entries in the kernel image and appends it to the kernel image such that boot/compressed/head_32.S can relocate the kernel. The kernel image is one statically linked object and only uses two relocation types - R_386_PC32 and R_386_32, of the two only the latter needs massaging during kernel relocation and thus handled by relocs. R_386_PC32 is ignored and all other relocation types are considered error. When the target of a relocation resides in a discarded section, binutils doesn't throw away the relocation record but nullifies it by changing it to R_386_NONE, which unfortunately makes relocs fail. The problem was triggered by yet out-of-tree x86 stack unwind patches but given the binutils behavior, ignoring R_386_NONE is the right thing to do. The problem has been tracked down to binutils behavior by Jan Beulich. [ Impact: fix build with certain binutils by ignoring R_386_NONE ] Signed-off-by: Tejun Heo Cc: Jan Beulich Cc: Ingo Molnar LKML-Reference: <4A1B8150.40702@kernel.org> Signed-off-by: H. Peter Anvin --- arch/x86/boot/compressed/relocs.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/arch/x86/boot/compressed/relocs.c b/arch/x86/boot/compressed/relocs.c index 857e492c571e..bbeb0c3fbd90 100644 --- a/arch/x86/boot/compressed/relocs.c +++ b/arch/x86/boot/compressed/relocs.c @@ -504,8 +504,11 @@ static void walk_relocs(void (*visit)(Elf32_Rel *rel, Elf32_Sym *sym)) if (sym->st_shndx == SHN_ABS) { continue; } - if (r_type == R_386_PC32) { - /* PC relative relocations don't need to be adjusted */ + if (r_type == R_386_NONE || r_type == R_386_PC32) { + /* + * NONE can be ignored and and PC relative + * relocations don't need to be adjusted. + */ } else if (r_type == R_386_32) { /* Visit relocations that need to be adjusted */ From f11a377b3f4e897d11f0e8d1fc688667e2f19708 Mon Sep 17 00:00:00 2001 From: David Dillow Date: Fri, 22 May 2009 15:29:34 +0000 Subject: [PATCH 199/267] r8169: avoid losing MSI interrupts The 8169 chip only generates MSI interrupts when all enabled event sources are quiescent and one or more sources transition to active. If not all of the active events are acknowledged, or a new event becomes active while the existing ones are cleared in the handler, we will not see a new interrupt. The current interrupt handler masks off the Rx and Tx events once the NAPI handler has been scheduled, which opens a race window in which we can get another Rx or Tx event and never ACK'ing it, stopping all activity until the link is reset (ifconfig down/up). Fix this by always ACK'ing all event sources, and loop in the handler until we have all sources quiescent. Signed-off-by: David Dillow Tested-by: Michael Buesch Signed-off-by: David S. Miller --- drivers/net/r8169.c | 112 ++++++++++++++++++++++++-------------------- 1 file changed, 62 insertions(+), 50 deletions(-) diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c index 0b6e8c896835..8247a945a1d9 100644 --- a/drivers/net/r8169.c +++ b/drivers/net/r8169.c @@ -3554,54 +3554,64 @@ static irqreturn_t rtl8169_interrupt(int irq, void *dev_instance) int handled = 0; int status; + /* loop handling interrupts until we have no new ones or + * we hit a invalid/hotplug case. + */ status = RTL_R16(IntrStatus); + while (status && status != 0xffff) { + handled = 1; - /* hotplug/major error/no more work/shared irq */ - if ((status == 0xffff) || !status) - goto out; - - handled = 1; - - if (unlikely(!netif_running(dev))) { - rtl8169_asic_down(ioaddr); - goto out; - } - - status &= tp->intr_mask; - RTL_W16(IntrStatus, - (status & RxFIFOOver) ? (status | RxOverflow) : status); - - if (!(status & tp->intr_event)) - goto out; - - /* Work around for rx fifo overflow */ - if (unlikely(status & RxFIFOOver) && - (tp->mac_version == RTL_GIGA_MAC_VER_11)) { - netif_stop_queue(dev); - rtl8169_tx_timeout(dev); - goto out; - } - - if (unlikely(status & SYSErr)) { - rtl8169_pcierr_interrupt(dev); - goto out; - } - - if (status & LinkChg) - rtl8169_check_link_status(dev, tp, ioaddr); - - if (status & tp->napi_event) { - RTL_W16(IntrMask, tp->intr_event & ~tp->napi_event); - tp->intr_mask = ~tp->napi_event; - - if (likely(napi_schedule_prep(&tp->napi))) - __napi_schedule(&tp->napi); - else if (netif_msg_intr(tp)) { - printk(KERN_INFO "%s: interrupt %04x in poll\n", - dev->name, status); + /* Handle all of the error cases first. These will reset + * the chip, so just exit the loop. + */ + if (unlikely(!netif_running(dev))) { + rtl8169_asic_down(ioaddr); + break; } + + /* Work around for rx fifo overflow */ + if (unlikely(status & RxFIFOOver) && + (tp->mac_version == RTL_GIGA_MAC_VER_11)) { + netif_stop_queue(dev); + rtl8169_tx_timeout(dev); + break; + } + + if (unlikely(status & SYSErr)) { + rtl8169_pcierr_interrupt(dev); + break; + } + + if (status & LinkChg) + rtl8169_check_link_status(dev, tp, ioaddr); + + /* We need to see the lastest version of tp->intr_mask to + * avoid ignoring an MSI interrupt and having to wait for + * another event which may never come. + */ + smp_rmb(); + if (status & tp->intr_mask & tp->napi_event) { + RTL_W16(IntrMask, tp->intr_event & ~tp->napi_event); + tp->intr_mask = ~tp->napi_event; + + if (likely(napi_schedule_prep(&tp->napi))) + __napi_schedule(&tp->napi); + else if (netif_msg_intr(tp)) { + printk(KERN_INFO "%s: interrupt %04x in poll\n", + dev->name, status); + } + } + + /* We only get a new MSI interrupt when all active irq + * sources on the chip have been acknowledged. So, ack + * everything we've seen and check if new sources have become + * active to avoid blocking all interrupts from the chip. + */ + RTL_W16(IntrStatus, + (status & RxFIFOOver) ? (status | RxOverflow) : status); + status = RTL_R16(IntrStatus); } -out: + return IRQ_RETVAL(handled); } @@ -3617,13 +3627,15 @@ static int rtl8169_poll(struct napi_struct *napi, int budget) if (work_done < budget) { napi_complete(napi); - tp->intr_mask = 0xffff; - /* - * 20040426: the barrier is not strictly required but the - * behavior of the irq handler could be less predictable - * without it. Btw, the lack of flush for the posted pci - * write is safe - FR + + /* We need for force the visibility of tp->intr_mask + * for other CPUs, as we can loose an MSI interrupt + * and potentially wait for a retransmit timeout if we don't. + * The posted write to IntrMask is safe, as it will + * eventually make it to the chip and we won't loose anything + * until it does. */ + tp->intr_mask = 0xffff; smp_wmb(); RTL_W16(IntrMask, tp->intr_event); } From 4319503779060120fa5de9b8fde056603bb6e0fd Mon Sep 17 00:00:00 2001 From: Jarod Wilson Date: Fri, 6 Mar 2009 20:24:57 +0000 Subject: [PATCH 200/267] [CPUFREQ] add atom family to p4-clockmod Some atom procs don't do freq scaling (such as the atom 330 on my own littlefalls2 board). By adding the atom family here, we at least get the benefit of passive cooling in a thermal emergency. Not sure how to see that its actually helping any, but the driver does bind and claim its functioning on my atom 330. Signed-off-by: Jarod Wilson Signed-off-by: Dave Jones --- arch/x86/kernel/cpu/cpufreq/p4-clockmod.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c index 6ac55bd341ae..869615193720 100644 --- a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c +++ b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c @@ -168,6 +168,7 @@ static unsigned int cpufreq_p4_get_frequency(struct cpuinfo_x86 *c) case 0x0E: /* Core */ case 0x0F: /* Core Duo */ case 0x16: /* Celeron Core */ + case 0x1C: /* Atom */ p4clockmod_driver.flags |= CPUFREQ_CONST_LOOPS; return speedstep_get_frequency(SPEEDSTEP_CPU_PCORE); case 0x0D: /* Pentium M (Dothan) */ From d38e73e8dad454a5916f446b0d3523c1161ae95a Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Thu, 23 Apr 2009 13:36:12 -0400 Subject: [PATCH 201/267] [CPUFREQ] powernow-k7 build fix when ACPI=n MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit arch/x86/kernel/cpu/cpufreq/powernow-k7.c:172: warning: 'invalidate_entry' defined but not used Reported-by: Toralf Förster Signed-off-by: Dave Jones --- arch/x86/kernel/cpu/cpufreq/powernow-k7.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c index 3c28ccd49742..a8363e5be4ef 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c @@ -168,10 +168,12 @@ static int check_powernow(void) return 1; } +#ifdef CONFIG_X86_POWERNOW_K7_ACPI static void invalidate_entry(unsigned int entry) { powernow_table[entry].frequency = CPUFREQ_ENTRY_INVALID; } +#endif static int get_ranges(unsigned char *pst) { From 42a06f2166f2f6f7bf04f32b4e823eacdceafdc9 Mon Sep 17 00:00:00 2001 From: Mathieu Desnoyers Date: Sun, 17 May 2009 10:23:52 -0400 Subject: [PATCH 202/267] [CPUFREQ] remove rwsem lock from CPUFREQ_GOV_STOP call * Rafael J. Wysocki (rjw@sisk.pl) wrote: > This message has been generated automatically as a part of a report > of regressions introduced between 2.6.28 and 2.6.29. > > The following bug entry is on the current list of known regressions > introduced between 2.6.28 and 2.6.29. Please verify if it still should > be listed and let me know (either way). > > > Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=13186 > Subject : cpufreq timer teardown problem > Submitter : Mathieu Desnoyers > Date : 2009-04-23 14:00 (24 days old) > References : http://marc.info/?l=linux-kernel&m=124049523515036&w=4 > Handled-By : Mathieu Desnoyers > Patch : http://patchwork.kernel.org/patch/19754/ > http://patchwork.kernel.org/patch/19753/ The patches linked above depend on the following patch to remove circular locking dependency : cpufreq: remove rwsem lock from CPUFREQ_GOV_STOP call (the following issue was faced when using cancel_delayed_work_sync() in the timer teardown (which fixes a race). * KOSAKI Motohiro (kosaki.motohiro@jp.fujitsu.com) wrote: > Hi > > my box output following warnings. > it seems regression by commit 7ccc7608b836e58fbacf65ee4f8eefa288e86fac. > > A: work -> do_dbs_timer() -> cpu_policy_rwsem > B: store() -> cpu_policy_rwsem -> cpufreq_governor_dbs() -> work > > Hrm, I think it must be due to my attempt to fix the timer teardown race in ondemand governor mixed with new locking behavior in 2.6.30-rc. The rwlock seems to be taken around the whole call to cpufreq_governor_dbs(), when it should be only taken around accesses to the locked data, and especially *not* around the call to dbs_timer_exit(). Reverting my fix attempt would put the teardown race back in place (replacing the cancel_delayed_work_sync by cancel_delayed_work). Instead, a proper fix would imply modifying this critical section : cpufreq.c: __cpufreq_remove_dev() ... if (cpufreq_driver->target) __cpufreq_governor(data, CPUFREQ_GOV_STOP); unlock_policy_rwsem_write(cpu); To make sure the __cpufreq_governor() callback is not called with rwsem held. This would allow execution of cancel_delayed_work_sync() without being nested within the rwsem. Applies on top of the 2.6.30-rc5 tree. Required to remove circular dep in teardown of both conservative and ondemande governors so they can use cancel_delayed_work_sync(). CPUFREQ_GOV_STOP does not modify the policy, therefore this locking seemed unneeded. Signed-off-by: Mathieu Desnoyers CC: KOSAKI Motohiro Cc: Greg KH CC: Ingo Molnar CC: "Rafael J. Wysocki" CC: Ben Slusky CC: Chris Wright CC: Andrew Morton Signed-off-by: Dave Jones --- drivers/cpufreq/cpufreq.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index d270e8eb3e67..47d2ad0ae079 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1070,11 +1070,11 @@ static int __cpufreq_remove_dev(struct sys_device *sys_dev) spin_unlock_irqrestore(&cpufreq_driver_lock, flags); #endif + unlock_policy_rwsem_write(cpu); + if (cpufreq_driver->target) __cpufreq_governor(data, CPUFREQ_GOV_STOP); - unlock_policy_rwsem_write(cpu); - kobject_put(&data->kobj); /* we need to make sure that the underlying kobj is actually From b253d2b2d28ead6fed012feb54694b3d0562839a Mon Sep 17 00:00:00 2001 From: Mathieu Desnoyers Date: Sun, 17 May 2009 10:29:33 -0400 Subject: [PATCH 203/267] [CPUFREQ] fix timer teardown in conservative governor * Rafael J. Wysocki (rjw@sisk.pl) wrote: > This message has been generated automatically as a part of a report > of regressions introduced between 2.6.28 and 2.6.29. > > The following bug entry is on the current list of known regressions > introduced between 2.6.28 and 2.6.29. Please verify if it still should > be listed and let me know (either way). > > > Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=13186 > Subject : cpufreq timer teardown problem > Submitter : Mathieu Desnoyers > Date : 2009-04-23 14:00 (24 days old) > References : http://marc.info/?l=linux-kernel&m=124049523515036&w=4 > Handled-By : Mathieu Desnoyers > Patch : http://patchwork.kernel.org/patch/19754/ > http://patchwork.kernel.org/patch/19753/ > (re-send with updated changelog) cpufreq fix timer teardown in conservative governor The problem is that dbs_timer_exit() uses cancel_delayed_work() when it should use cancel_delayed_work_sync(). cancel_delayed_work() does not wait for the workqueue handler to exit. The ondemand governor does not seem to be affected because the "if (!dbs_info->enable)" check at the beginning of the workqueue handler returns immediately without rescheduling the work. The conservative governor in 2.6.30-rc has the same check as the ondemand governor, which makes things usually run smoothly. However, if the governor is quickly stopped and then started, this could lead to the following race : dbs_enable could be reenabled and multiple do_dbs_timer handlers would run. This is why a synchronized teardown is required. Depends on patch cpufreq: remove rwsem lock from CPUFREQ_GOV_STOP call The following patch applies to 2.6.30-rc2. Stable kernels have a similar issue which should also be fixed, but the code changed between 2.6.29 and 2.6.30, so this patch only applies to 2.6.30-rc. Signed-off-by: Mathieu Desnoyers CC: Andrew Morton CC: gregkh@suse.de CC: stable@kernel.org CC: cpufreq@vger.kernel.org CC: Ingo Molnar CC: rjw@sisk.pl CC: Ben Slusky Signed-off-by: Dave Jones --- drivers/cpufreq/cpufreq_conservative.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/cpufreq/cpufreq_conservative.c b/drivers/cpufreq/cpufreq_conservative.c index 2ecd95e4ab1a..7a74d175287b 100644 --- a/drivers/cpufreq/cpufreq_conservative.c +++ b/drivers/cpufreq/cpufreq_conservative.c @@ -91,6 +91,9 @@ static unsigned int dbs_enable; /* number of CPUs using this policy */ * (like __cpufreq_driver_target()) is being called with dbs_mutex taken, then * cpu_hotplug lock should be taken before that. Note that cpu_hotplug lock * is recursive for the same process. -Venki + * DEADLOCK ALERT! (2) : do_dbs_timer() must not take the dbs_mutex, because it + * would deadlock with cancel_delayed_work_sync(), which is needed for proper + * raceless workqueue teardown. */ static DEFINE_MUTEX(dbs_mutex); @@ -542,7 +545,7 @@ static inline void dbs_timer_init(struct cpu_dbs_info_s *dbs_info) static inline void dbs_timer_exit(struct cpu_dbs_info_s *dbs_info) { dbs_info->enable = 0; - cancel_delayed_work(&dbs_info->work); + cancel_delayed_work_sync(&dbs_info->work); } static int cpufreq_governor_dbs(struct cpufreq_policy *policy, From b14893a62c73af0eca414cfed505b8c09efc613c Mon Sep 17 00:00:00 2001 From: Mathieu Desnoyers Date: Sun, 17 May 2009 10:30:45 -0400 Subject: [PATCH 204/267] [CPUFREQ] fix timer teardown in ondemand governor * Rafael J. Wysocki (rjw@sisk.pl) wrote: > This message has been generated automatically as a part of a report > of regressions introduced between 2.6.28 and 2.6.29. > > The following bug entry is on the current list of known regressions > introduced between 2.6.28 and 2.6.29. Please verify if it still should > be listed and let me know (either way). > > > Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=13186 > Subject : cpufreq timer teardown problem > Submitter : Mathieu Desnoyers > Date : 2009-04-23 14:00 (24 days old) > References : http://marc.info/?l=linux-kernel&m=124049523515036&w=4 > Handled-By : Mathieu Desnoyers > Patch : http://patchwork.kernel.org/patch/19754/ > http://patchwork.kernel.org/patch/19753/ > (updated changelog) cpufreq fix timer teardown in ondemand governor The problem is that dbs_timer_exit() uses cancel_delayed_work() when it should use cancel_delayed_work_sync(). cancel_delayed_work() does not wait for the workqueue handler to exit. The ondemand governor does not seem to be affected because the "if (!dbs_info->enable)" check at the beginning of the workqueue handler returns immediately without rescheduling the work. The conservative governor in 2.6.30-rc has the same check as the ondemand governor, which makes things usually run smoothly. However, if the governor is quickly stopped and then started, this could lead to the following race : dbs_enable could be reenabled and multiple do_dbs_timer handlers would run. This is why a synchronized teardown is required. The following patch applies to, at least, 2.6.28.x, 2.6.29.1, 2.6.30-rc2. Depends on patch cpufreq: remove rwsem lock from CPUFREQ_GOV_STOP call Signed-off-by: Mathieu Desnoyers CC: Andrew Morton CC: gregkh@suse.de CC: stable@kernel.org CC: cpufreq@vger.kernel.org CC: Ingo Molnar CC: rjw@sisk.pl CC: Ben Slusky Signed-off-by: Dave Jones --- drivers/cpufreq/cpufreq_ondemand.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c index 338f428a15b7..e741c339df76 100644 --- a/drivers/cpufreq/cpufreq_ondemand.c +++ b/drivers/cpufreq/cpufreq_ondemand.c @@ -98,6 +98,9 @@ static unsigned int dbs_enable; /* number of CPUs using this policy */ * (like __cpufreq_driver_target()) is being called with dbs_mutex taken, then * cpu_hotplug lock should be taken before that. Note that cpu_hotplug lock * is recursive for the same process. -Venki + * DEADLOCK ALERT! (2) : do_dbs_timer() must not take the dbs_mutex, because it + * would deadlock with cancel_delayed_work_sync(), which is needed for proper + * raceless workqueue teardown. */ static DEFINE_MUTEX(dbs_mutex); @@ -562,7 +565,7 @@ static inline void dbs_timer_init(struct cpu_dbs_info_s *dbs_info) static inline void dbs_timer_exit(struct cpu_dbs_info_s *dbs_info) { dbs_info->enable = 0; - cancel_delayed_work(&dbs_info->work); + cancel_delayed_work_sync(&dbs_info->work); } static int cpufreq_governor_dbs(struct cpufreq_policy *policy, From df1829770db415dc5a5ed5ada3bd70176c6f0a01 Mon Sep 17 00:00:00 2001 From: Thomas Renninger Date: Wed, 22 Apr 2009 13:48:32 +0200 Subject: [PATCH 205/267] [CPUFREQ] powernow-k8 cleanup msg if BIOS does not export ACPI _PSS cpufreq data - Make the message shorter and easier to grep for - Use printk_once instead of WARN_ONCE (functionality of these was mixed) Signed-off-by: Thomas Renninger Cc: Langsdorf, Mark Signed-off-by: Dave Jones --- arch/x86/kernel/cpu/cpufreq/powernow-k8.c | 19 +++++-------------- 1 file changed, 5 insertions(+), 14 deletions(-) diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c index 4709ead2db52..feef10c085a1 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c @@ -1215,13 +1215,16 @@ static int powernowk8_verify(struct cpufreq_policy *pol) return cpufreq_frequency_table_verify(pol, data->powernow_table); } +static const char ACPI_PSS_BIOS_BUG_MSG[] = + KERN_ERR FW_BUG PFX "No compatible ACPI _PSS objects found.\n" + KERN_ERR FW_BUG PFX "Try again with latest BIOS.\n"; + /* per CPU init entry point to the driver */ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) { struct powernow_k8_data *data; cpumask_t oldmask; int rc; - static int print_once; if (!cpu_online(pol->cpu)) return -ENODEV; @@ -1244,19 +1247,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) * an UP version, and is deprecated by AMD. */ if (num_online_cpus() != 1) { - /* - * Replace this one with print_once as soon as such a - * thing gets introduced - */ - if (!print_once) { - WARN_ONCE(1, KERN_ERR FW_BUG PFX "Your BIOS " - "does not provide ACPI _PSS objects " - "in a way that Linux understands. " - "Please report this to the Linux ACPI" - " maintainers and complain to your " - "BIOS vendor.\n"); - print_once++; - } + printk_once(ACPI_PSS_BIOS_BUG_MSG); goto err_out; } if (pol->cpu != 0) { From ca446d06351992e4f1a7c1e5e99870ab4ec5188f Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Wed, 22 Apr 2009 13:48:33 +0200 Subject: [PATCH 206/267] [CPUFREQ] powernow-k8: determine exact CPU frequency for HW Pstates Slightly modified by trenn@suse.de -> only do this on fam 10h and fam 11h. Currently powernow-k8 determines CPU frequency from ACPI PSS objects, but according to AMD family 11h BKDG this frequency is just a rounded value: "CoreFreq (MHz) = The CPU COF specified by MSRC001_00[6B:64][CpuFid] rounded to the nearest 100 Mhz." As a consequnce powernow-k8 reports wrong CPU frequency on some systems, e.g. on Turion X2 Ultra: powernow-k8: Found 1 AMD Turion(tm)X2 Ultra DualCore Mobile ZM-82 processors (2 cpu cores) (version 2.20.00) powernow-k8: 0 : pstate 0 (2200 MHz) powernow-k8: 1 : pstate 1 (1100 MHz) powernow-k8: 2 : pstate 2 (600 MHz) But this is wrong as frequency for Pstate2 is 550 MHz. x86info reports it correctly: #x86info -a |grep Pstate ... Pstate-0: fid=e, did=0, vid=24 (2200MHz) Pstate-1: fid=e, did=1, vid=30 (1100MHz) Pstate-2: fid=e, did=2, vid=3c (550MHz) (current) Solution is to determine the frequency directly from Pstate MSRs instead of using rounded values from ACPI table. Signed-off-by: Andreas Herrmann Signed-off-by: Thomas Renninger Signed-off-by: Dave Jones --- arch/x86/kernel/cpu/cpufreq/powernow-k8.c | 23 +++++++++++++++++++++-- 1 file changed, 21 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c index feef10c085a1..f6b32d112357 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c @@ -649,6 +649,20 @@ static void print_basics(struct powernow_k8_data *data) data->batps); } +static u32 freq_from_fid_did(u32 fid, u32 did) +{ + u32 mhz = 0; + + if (boot_cpu_data.x86 == 0x10) + mhz = (100 * (fid + 0x10)) >> did; + else if (boot_cpu_data.x86 == 0x11) + mhz = (100 * (fid + 8)) >> did; + else + BUG(); + + return mhz * 1000; +} + static int fill_powernow_table(struct powernow_k8_data *data, struct pst_s *pst, u8 maxvid) { @@ -923,8 +937,13 @@ static int fill_powernow_table_pstate(struct powernow_k8_data *data, powernow_table[i].index = index; - powernow_table[i].frequency = - data->acpi_data.states[i].core_frequency * 1000; + /* Frequency may be rounded for these */ + if (boot_cpu_data.x86 == 0x10 || boot_cpu_data.x86 == 0x11) { + powernow_table[i].frequency = + freq_from_fid_did(lo & 0x3f, (lo >> 6) & 7); + } else + powernow_table[i].frequency = + data->acpi_data.states[i].core_frequency * 1000; } return 0; } From e4a5d54f924ea5ce2913d9d0687d034004816465 Mon Sep 17 00:00:00 2001 From: Ma Ling Date: Tue, 26 May 2009 11:31:00 +0800 Subject: [PATCH 207/267] drm/i915: Add support for VGA load detection (pre-945). Two approaches for VGA detections: hot plug detection for 945G onwards and load pipe detection for Pre-945G. Load pipe detection will get one free pipe, set border color as red and blue, then check CRT status by swf register. This is a sync-up with the 2D driver. Signed-off-by: Ma Ling Signed-off-by: Eric Anholt --- drivers/gpu/drm/i915/intel_crt.c | 149 ++++++++++++++++++++++++++++++- 1 file changed, 147 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c index 19148c3df637..640f5158effc 100644 --- a/drivers/gpu/drm/i915/intel_crt.c +++ b/drivers/gpu/drm/i915/intel_crt.c @@ -198,9 +198,142 @@ static bool intel_crt_detect_ddc(struct drm_connector *connector) return intel_ddc_probe(intel_output); } +static enum drm_connector_status +intel_crt_load_detect(struct drm_crtc *crtc, struct intel_output *intel_output) +{ + struct drm_encoder *encoder = &intel_output->enc; + struct drm_device *dev = encoder->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + uint32_t pipe = intel_crtc->pipe; + uint32_t save_bclrpat; + uint32_t save_vtotal; + uint32_t vtotal, vactive; + uint32_t vsample; + uint32_t vblank, vblank_start, vblank_end; + uint32_t dsl; + uint32_t bclrpat_reg; + uint32_t vtotal_reg; + uint32_t vblank_reg; + uint32_t vsync_reg; + uint32_t pipeconf_reg; + uint32_t pipe_dsl_reg; + uint8_t st00; + enum drm_connector_status status; + + if (pipe == 0) { + bclrpat_reg = BCLRPAT_A; + vtotal_reg = VTOTAL_A; + vblank_reg = VBLANK_A; + vsync_reg = VSYNC_A; + pipeconf_reg = PIPEACONF; + pipe_dsl_reg = PIPEADSL; + } else { + bclrpat_reg = BCLRPAT_B; + vtotal_reg = VTOTAL_B; + vblank_reg = VBLANK_B; + vsync_reg = VSYNC_B; + pipeconf_reg = PIPEBCONF; + pipe_dsl_reg = PIPEBDSL; + } + + save_bclrpat = I915_READ(bclrpat_reg); + save_vtotal = I915_READ(vtotal_reg); + vblank = I915_READ(vblank_reg); + + vtotal = ((save_vtotal >> 16) & 0xfff) + 1; + vactive = (save_vtotal & 0x7ff) + 1; + + vblank_start = (vblank & 0xfff) + 1; + vblank_end = ((vblank >> 16) & 0xfff) + 1; + + /* Set the border color to purple. */ + I915_WRITE(bclrpat_reg, 0x500050); + + if (IS_I9XX(dev)) { + uint32_t pipeconf = I915_READ(pipeconf_reg); + I915_WRITE(pipeconf_reg, pipeconf | PIPECONF_FORCE_BORDER); + /* Wait for next Vblank to substitue + * border color for Color info */ + intel_wait_for_vblank(dev); + st00 = I915_READ8(VGA_MSR_WRITE); + status = ((st00 & (1 << 4)) != 0) ? + connector_status_connected : + connector_status_disconnected; + + I915_WRITE(pipeconf_reg, pipeconf); + } else { + bool restore_vblank = false; + int count, detect; + + /* + * If there isn't any border, add some. + * Yes, this will flicker + */ + if (vblank_start <= vactive && vblank_end >= vtotal) { + uint32_t vsync = I915_READ(vsync_reg); + uint32_t vsync_start = (vsync & 0xffff) + 1; + + vblank_start = vsync_start; + I915_WRITE(vblank_reg, + (vblank_start - 1) | + ((vblank_end - 1) << 16)); + restore_vblank = true; + } + /* sample in the vertical border, selecting the larger one */ + if (vblank_start - vactive >= vtotal - vblank_end) + vsample = (vblank_start + vactive) >> 1; + else + vsample = (vtotal + vblank_end) >> 1; + + /* + * Wait for the border to be displayed + */ + while (I915_READ(pipe_dsl_reg) >= vactive) + ; + while ((dsl = I915_READ(pipe_dsl_reg)) <= vsample) + ; + /* + * Watch ST00 for an entire scanline + */ + detect = 0; + count = 0; + do { + count++; + /* Read the ST00 VGA status register */ + st00 = I915_READ8(VGA_MSR_WRITE); + if (st00 & (1 << 4)) + detect++; + } while ((I915_READ(pipe_dsl_reg) == dsl)); + + /* restore vblank if necessary */ + if (restore_vblank) + I915_WRITE(vblank_reg, vblank); + /* + * If more than 3/4 of the scanline detected a monitor, + * then it is assumed to be present. This works even on i830, + * where there isn't any way to force the border color across + * the screen + */ + status = detect * 4 > count * 3 ? + connector_status_connected : + connector_status_disconnected; + } + + /* Restore previous settings */ + I915_WRITE(bclrpat_reg, save_bclrpat); + + return status; +} + static enum drm_connector_status intel_crt_detect(struct drm_connector *connector) { struct drm_device *dev = connector->dev; + struct intel_output *intel_output = to_intel_output(connector); + struct drm_encoder *encoder = &intel_output->enc; + struct drm_crtc *crtc; + int dpms_mode; + enum drm_connector_status status; if (IS_I9XX(dev) && !IS_I915G(dev) && !IS_I915GM(dev)) { if (intel_crt_detect_hotplug(connector)) @@ -212,8 +345,20 @@ static enum drm_connector_status intel_crt_detect(struct drm_connector *connecto if (intel_crt_detect_ddc(connector)) return connector_status_connected; - /* TODO use load detect */ - return connector_status_unknown; + /* for pre-945g platforms use load detect */ + if (encoder->crtc && encoder->crtc->enabled) { + status = intel_crt_load_detect(encoder->crtc, intel_output); + } else { + crtc = intel_get_load_detect_pipe(intel_output, + NULL, &dpms_mode); + if (crtc) { + status = intel_crt_load_detect(crtc, intel_output); + intel_release_load_detect_pipe(intel_output, dpms_mode); + } else + status = connector_status_unknown; + } + + return status; } static void intel_crt_destroy(struct drm_connector *connector) From 68743082b560067e3e93eab8b2568f238e486865 Mon Sep 17 00:00:00 2001 From: Vu Pham Date: Tue, 26 May 2009 14:51:00 -0400 Subject: [PATCH 208/267] XPRTRDMA: fix client rpcrdma FRMR registration on mlx4 devices mlx4/connectX FRMR requires local write enable together with remote rdma write enable. This fixes NFS/RDMA operation over the ConnectX Infiniband HCA in the default memreg mode. Signed-off-by: Vu Pham Signed-off-by: Tom Talpey Signed-off-by: Trond Myklebust --- net/sunrpc/xprtrdma/verbs.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c index 3b21e0cc5e69..465aafc2007f 100644 --- a/net/sunrpc/xprtrdma/verbs.c +++ b/net/sunrpc/xprtrdma/verbs.c @@ -1495,7 +1495,8 @@ rpcrdma_register_frmr_external(struct rpcrdma_mr_seg *seg, frmr_wr.wr.fast_reg.page_shift = PAGE_SHIFT; frmr_wr.wr.fast_reg.length = i << PAGE_SHIFT; frmr_wr.wr.fast_reg.access_flags = (writing ? - IB_ACCESS_REMOTE_WRITE : IB_ACCESS_REMOTE_READ); + IB_ACCESS_REMOTE_WRITE | IB_ACCESS_LOCAL_WRITE : + IB_ACCESS_REMOTE_READ); frmr_wr.wr.fast_reg.rkey = seg1->mr_chunk.rl_mw->r.frmr.fr_mr->rkey; DECR_CQCOUNT(&r_xprt->rx_ep); From d0367a508af9cf97beb202935bb9ad8883d30cd1 Mon Sep 17 00:00:00 2001 From: Sam Ravnborg Date: Tue, 26 May 2009 14:51:00 -0400 Subject: [PATCH 209/267] nfs: fix build error in nfsroot with initconst fix build error with latest kbuild adjustments to initconst. The commit a447c0932445f92ce6f4c1bd020f62c5097a7842 ("vfs: Use const for kernel parser table") changed: static match_table_t __initdata tokens = { to static match_table_t __initconst tokens = { But the missing const causes popwerpc to fail with latest updates to __initconst like this: fs/nfs/nfsroot.c:400: error: __setup_str_nfs_root_setup causes a section type conflict fs/nfs/nfsroot.c:400: error: __setup_str_nfs_root_setup causes a section type conflict The bug is only present with kbuild-next. Following patch has been build tested. Signed-off-by: Sam Ravnborg Cc: Steven Whitehouse Cc: Stephen Rothwell Acked-by: Jan Beulich Signed-off-by: Trond Myklebust --- fs/nfs/nfsroot.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c index d9ef602fbc5a..e3ed5908820b 100644 --- a/fs/nfs/nfsroot.c +++ b/fs/nfs/nfsroot.c @@ -129,7 +129,7 @@ enum { Opt_err }; -static match_table_t __initconst tokens = { +static const match_table_t tokens __initconst = { {Opt_port, "port=%u"}, {Opt_rsize, "rsize=%u"}, {Opt_wsize, "wsize=%u"}, From 95baa25c7321eb8613246acbf61b97911cc748d3 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 26 May 2009 14:51:00 -0400 Subject: [PATCH 210/267] NFSv4: Fix the case where NFSv4 renewal fails If the asynchronous lease renewal fails (usually due to a soft timeout), then we _must_ schedule state recovery in order to ensure that we don't lose the lease unnecessarily or, if the lease is already lost, that we recover the locking state promptly... Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index a4d242680299..4674f8092da8 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -2594,12 +2594,9 @@ static void nfs4_renew_done(struct rpc_task *task, void *data) unsigned long timestamp = (unsigned long)data; if (task->tk_status < 0) { - switch (task->tk_status) { - case -NFS4ERR_STALE_CLIENTID: - case -NFS4ERR_EXPIRED: - case -NFS4ERR_CB_PATH_DOWN: - nfs4_schedule_state_recovery(clp); - } + /* Unless we're shutting down, schedule state recovery! */ + if (test_bit(NFS_CS_RENEWD, &clp->cl_res_state) != 0) + nfs4_schedule_state_recovery(clp); return; } spin_lock(&clp->cl_lock); From ab2b7ebaad16226c9a5e85c5f384d19fa58a7459 Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Tue, 26 May 2009 09:11:03 +0100 Subject: [PATCH 211/267] kmod: Release sub_info on cred allocation failure. call_usermodehelper_setup() forgot to kfree(sub_info) when prepare_usermodehelper_creds() failed. Signed-off-by: Tetsuo Handa Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- kernel/kmod.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/kernel/kmod.c b/kernel/kmod.c index b750675251e5..7e95bedb2bfc 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -370,8 +370,10 @@ struct subprocess_info *call_usermodehelper_setup(char *path, char **argv, sub_info->argv = argv; sub_info->envp = envp; sub_info->cred = prepare_usermodehelper_creds(); - if (!sub_info->cred) + if (!sub_info->cred) { + kfree(sub_info); return NULL; + } out: return sub_info; From 564346224daaa8f7222d7a92cdbb7bafde59ae6e Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Tue, 26 May 2009 20:54:41 +0930 Subject: [PATCH 212/267] lguest: fix on Intel when KVM loaded (unhandled trap 13) When KVM is loaded, and hence VT set up, the vmcall instruction in an lguest guest causes a #GP, not #UD. Signed-off-by: Rusty Russell Signed-off-by: Linus Torvalds --- drivers/lguest/x86/core.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) diff --git a/drivers/lguest/x86/core.c b/drivers/lguest/x86/core.c index 1a83910f674f..eaf722fe309a 100644 --- a/drivers/lguest/x86/core.c +++ b/drivers/lguest/x86/core.c @@ -358,6 +358,16 @@ void lguest_arch_handle_trap(struct lg_cpu *cpu) if (emulate_insn(cpu)) return; } + /* If KVM is active, the vmcall instruction triggers a + * General Protection Fault. Normally it triggers an + * invalid opcode fault (6): */ + case 6: + /* We need to check if ring == GUEST_PL and + * faulting instruction == vmcall. */ + if (is_hypercall(cpu)) { + rewrite_hypercall(cpu); + return; + } break; case 14: /* We've intercepted a Page Fault. */ /* The Guest accessed a virtual address that wasn't mapped. @@ -403,15 +413,6 @@ void lguest_arch_handle_trap(struct lg_cpu *cpu) * up the pointer now to indicate a hypercall is pending. */ cpu->hcall = (struct hcall_args *)cpu->regs; return; - case 6: - /* kvm hypercalls trigger an invalid opcode fault (6). - * We need to check if ring == GUEST_PL and - * faulting instruction == vmcall. */ - if (is_hypercall(cpu)) { - rewrite_hypercall(cpu); - return; - } - break; } /* We didn't handle the trap, so it needs to go to the Guest. */ From 2171787be2e71ff71159857bfeb21398b61eb615 Mon Sep 17 00:00:00 2001 From: "Pallipadi, Venkatesh" Date: Tue, 26 May 2009 10:33:35 -0700 Subject: [PATCH 213/267] x86: avoid back to back on_each_cpu in cpa_flush_array Cleanup cpa_flush_array() to avoid back to back on_each_cpu() calls. [ Impact: optimizes fix 0af48f42df15b97080b450d24219dd95db7b929a ] Signed-off-by: Venkatesh Pallipadi Signed-off-by: H. Peter Anvin --- arch/x86/mm/pageattr.c | 17 +++-------------- 1 file changed, 3 insertions(+), 14 deletions(-) diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 0f9052bcec4b..e17efed088c5 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -204,30 +204,19 @@ static void cpa_flush_range(unsigned long start, int numpages, int cache) } } -static void wbinvd_local(void *unused) -{ - wbinvd(); -} - static void cpa_flush_array(unsigned long *start, int numpages, int cache, int in_flags, struct page **pages) { unsigned int i, level; + unsigned long do_wbinvd = cache && numpages >= 1024; /* 4M threshold */ BUG_ON(irqs_disabled()); - on_each_cpu(__cpa_flush_range, NULL, 1); + on_each_cpu(__cpa_flush_all, (void *) do_wbinvd, 1); - if (!cache) + if (!cache || do_wbinvd) return; - /* 4M threshold */ - if (numpages >= 1024) { - if (boot_cpu_data.x86 >= 4) - on_each_cpu(wbinvd_local, NULL, 1); - - return; - } /* * We only need to flush on one CPU, * clflush is a MESI-coherent instruction that From b1338d199dda6681d9af0297928af0a7eb9cba7b Mon Sep 17 00:00:00 2001 From: Herton Ronaldo Krzesinski Date: Tue, 26 May 2009 12:15:53 +0900 Subject: [PATCH 214/267] tomoyo: add missing call to cap_bprm_set_creds cap_bprm_set_creds() has to be called from security_bprm_set_creds(). TOMOYO forgot to call cap_bprm_set_creds() from tomoyo_bprm_set_creds() and suid executables were not being working. Make sure we call cap_bprm_set_creds() with TOMOYO, to set credentials properly inside tomoyo_bprm_set_creds(). Signed-off-by: Herton Ronaldo Krzesinski Acked-by: Tetsuo Handa Signed-off-by: James Morris --- security/tomoyo/tomoyo.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/security/tomoyo/tomoyo.c b/security/tomoyo/tomoyo.c index 5b481912752a..e42be5c4f055 100644 --- a/security/tomoyo/tomoyo.c +++ b/security/tomoyo/tomoyo.c @@ -27,6 +27,12 @@ static int tomoyo_cred_prepare(struct cred *new, const struct cred *old, static int tomoyo_bprm_set_creds(struct linux_binprm *bprm) { + int rc; + + rc = cap_bprm_set_creds(bprm); + if (rc) + return rc; + /* * Do only if this function is called for the first time of an execve * operation. From e76a16deb8785317a23cca7204331af053e0fb4e Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Tue, 26 May 2009 17:44:56 -0700 Subject: [PATCH 215/267] drm/i915: Fix tiling pitch handling on 8xx. The pitch field is an exponent on pre-965, so we were rejecting buffers on 8xx that we shouldn't have. 915 got lucky in that the largest legal value happened to match (8KB / 512 = 0x10), but 8xx has a smaller tile width. Additionally, we programmed that bad value into the register on 8xx, so the only pitch that would work correctly was 4096 (512-1023 pixels), while others would probably give bad rendering or hangs. Signed-off-by: Eric Anholt fd.o bug #20473. --- drivers/gpu/drm/i915/i915_gem.c | 6 ++++-- drivers/gpu/drm/i915/i915_gem_tiling.c | 14 +++++++++++--- drivers/gpu/drm/i915/i915_reg.h | 3 ++- 3 files changed, 17 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 717b6a854bcd..e4408daf8cef 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -2128,8 +2128,10 @@ static void i830_write_fence_reg(struct drm_i915_fence_reg *reg) return; } - pitch_val = (obj_priv->stride / 128) - 1; - WARN_ON(pitch_val & ~0x0000000f); + pitch_val = obj_priv->stride / 128; + pitch_val = ffs(pitch_val) - 1; + WARN_ON(pitch_val > I830_FENCE_MAX_PITCH_VAL); + val = obj_priv->gtt_offset; if (obj_priv->tiling_mode == I915_TILING_Y) val |= 1 << I830_FENCE_TILING_Y_SHIFT; diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c index 52a059354e83..540dd336e6ec 100644 --- a/drivers/gpu/drm/i915/i915_gem_tiling.c +++ b/drivers/gpu/drm/i915/i915_gem_tiling.c @@ -213,7 +213,8 @@ i915_tiling_ok(struct drm_device *dev, int stride, int size, int tiling_mode) if (tiling_mode == I915_TILING_NONE) return true; - if (tiling_mode == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev)) + if (!IS_I9XX(dev) || + (tiling_mode == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev))) tile_width = 128; else tile_width = 512; @@ -225,11 +226,18 @@ i915_tiling_ok(struct drm_device *dev, int stride, int size, int tiling_mode) if (stride / 128 > I965_FENCE_MAX_PITCH_VAL) return false; } else if (IS_I9XX(dev)) { - if (stride / tile_width > I830_FENCE_MAX_PITCH_VAL || + uint32_t pitch_val = ffs(stride / tile_width) - 1; + + /* XXX: For Y tiling, FENCE_MAX_PITCH_VAL is actually 6 (8KB) + * instead of 4 (2KB) on 945s. + */ + if (pitch_val > I915_FENCE_MAX_PITCH_VAL || size > (I830_FENCE_MAX_SIZE_VAL << 20)) return false; } else { - if (stride / 128 > I830_FENCE_MAX_PITCH_VAL || + uint32_t pitch_val = ffs(stride / tile_width) - 1; + + if (pitch_val > I830_FENCE_MAX_PITCH_VAL || size > (I830_FENCE_MAX_SIZE_VAL << 19)) return false; } diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 9668cc0d7f4e..375569d01d01 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -190,7 +190,8 @@ #define I830_FENCE_SIZE_BITS(size) ((ffs((size) >> 19) - 1) << 8) #define I830_FENCE_PITCH_SHIFT 4 #define I830_FENCE_REG_VALID (1<<0) -#define I830_FENCE_MAX_PITCH_VAL 0x10 +#define I915_FENCE_MAX_PITCH_VAL 0x10 +#define I830_FENCE_MAX_PITCH_VAL 6 #define I830_FENCE_MAX_SIZE_VAL (1<<8) #define I915_FENCE_START_MASK 0x0ff00000 From cfa16a0de5392c54db553ec2233a7110e4b4da7a Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Tue, 26 May 2009 18:46:16 -0700 Subject: [PATCH 216/267] drm/i915: Apply a big hammer to 865 GEM object CPU cache flushing. On the 865, but not the 855, the clflush we do appears to not actually make it out to the hardware all the time. An easy way to safely reproduce was X -retro, which would show that some of the blits involved in drawing the lovely root weave didn't make it out to the hardware. Those blits are 32 bytes each, and 1-2 would be missing at various points around the screen. Other experimentation (doing more clflush, doing more AGP chipset flush, poking at some more device registers to maybe trigger more flushing) didn't help. krh came up with the wbinvd as a way to successfully get all those blits to appear. Signed-off-by: Eric Anholt --- drivers/gpu/drm/i915/i915_gem.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index e4408daf8cef..e2421869a40c 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -2423,6 +2423,16 @@ i915_gem_clflush_object(struct drm_gem_object *obj) if (obj_priv->pages == NULL) return; + /* XXX: The 865 in particular appears to be weird in how it handles + * cache flushing. We haven't figured it out, but the + * clflush+agp_chipset_flush doesn't appear to successfully get the + * data visible to the PGU, while wbinvd + agp_chipset_flush does. + */ + if (IS_I865G(obj->dev)) { + wbinvd(); + return; + } + drm_clflush_pages(obj_priv->pages, obj->size / PAGE_SIZE); } From 84532a0fc3d5811dca8e3726fe4d372ea87bd7c6 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 27 May 2009 13:33:14 +1000 Subject: [PATCH 217/267] Revert "powerpc: Rework dma-noncoherent to use generic vmalloc layer" This reverts commit 33f00dcedb0e22cdb156a23632814fc580fcfcf8. While it was a good idea to try to use the mm/vmalloc.c allocator instead of our own (in fact, ours is itself a dup on an old variant of the vmalloc one), unfortunately, the approach is terminally busted since dma_alloc_coherent() can be called at interrupt time or in atomic contexts and there's little chances we'll make the code in mm/vmalloc.c cope with\ that :-( Until we can get the generic code to forbid that idiocy and fix all drivers abusing it, we pretty much have no choice but revert to our custom virtual space allocator. There's also a problem with SMP safety since freeing such mapping would require an IPI which cannot be done at interrupt time. However, right now, I don't think we support any platform that is both SMP and has non-coherent DMA (don't laugh, I know such things do exist !) so we can sort that out later. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/Kconfig | 25 +++ arch/powerpc/lib/dma-noncoherent.c | 299 +++++++++++++++++++++++------ 2 files changed, 269 insertions(+), 55 deletions(-) diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index a0d1146a0578..3bb43adce44d 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -868,6 +868,31 @@ config TASK_SIZE default "0x80000000" if PPC_PREP || PPC_8xx default "0xc0000000" +config CONSISTENT_START_BOOL + bool "Set custom consistent memory pool address" + depends on ADVANCED_OPTIONS && NOT_COHERENT_CACHE + help + This option allows you to set the base virtual address + of the consistent memory pool. This pool of virtual + memory is used to make consistent memory allocations. + +config CONSISTENT_START + hex "Base virtual address of consistent memory pool" if CONSISTENT_START_BOOL + default "0xfd000000" if (NOT_COHERENT_CACHE && 8xx) + default "0xff100000" if NOT_COHERENT_CACHE + +config CONSISTENT_SIZE_BOOL + bool "Set custom consistent memory pool size" + depends on ADVANCED_OPTIONS && NOT_COHERENT_CACHE + help + This option allows you to set the size of the + consistent memory pool. This pool of virtual memory + is used to make consistent memory allocations. + +config CONSISTENT_SIZE + hex "Size of consistent memory pool" if CONSISTENT_SIZE_BOOL + default "0x00200000" if NOT_COHERENT_CACHE + config PIN_TLB bool "Pinned Kernel TLBs (860 ONLY)" depends on ADVANCED_OPTIONS && 8xx diff --git a/arch/powerpc/lib/dma-noncoherent.c b/arch/powerpc/lib/dma-noncoherent.c index 005a28d380af..b7dc4c19f582 100644 --- a/arch/powerpc/lib/dma-noncoherent.c +++ b/arch/powerpc/lib/dma-noncoherent.c @@ -29,10 +29,120 @@ #include #include #include -#include #include +/* + * This address range defaults to a value that is safe for all + * platforms which currently set CONFIG_NOT_COHERENT_CACHE. It + * can be further configured for specific applications under + * the "Advanced Setup" menu. -Matt + */ +#define CONSISTENT_BASE (CONFIG_CONSISTENT_START) +#define CONSISTENT_END (CONFIG_CONSISTENT_START + CONFIG_CONSISTENT_SIZE) +#define CONSISTENT_OFFSET(x) (((unsigned long)(x) - CONSISTENT_BASE) >> PAGE_SHIFT) + +/* + * This is the page table (2MB) covering uncached, DMA consistent allocations + */ +static pte_t *consistent_pte; +static DEFINE_SPINLOCK(consistent_lock); + +/* + * VM region handling support. + * + * This should become something generic, handling VM region allocations for + * vmalloc and similar (ioremap, module space, etc). + * + * I envisage vmalloc()'s supporting vm_struct becoming: + * + * struct vm_struct { + * struct vm_region region; + * unsigned long flags; + * struct page **pages; + * unsigned int nr_pages; + * unsigned long phys_addr; + * }; + * + * get_vm_area() would then call vm_region_alloc with an appropriate + * struct vm_region head (eg): + * + * struct vm_region vmalloc_head = { + * .vm_list = LIST_HEAD_INIT(vmalloc_head.vm_list), + * .vm_start = VMALLOC_START, + * .vm_end = VMALLOC_END, + * }; + * + * However, vmalloc_head.vm_start is variable (typically, it is dependent on + * the amount of RAM found at boot time.) I would imagine that get_vm_area() + * would have to initialise this each time prior to calling vm_region_alloc(). + */ +struct ppc_vm_region { + struct list_head vm_list; + unsigned long vm_start; + unsigned long vm_end; +}; + +static struct ppc_vm_region consistent_head = { + .vm_list = LIST_HEAD_INIT(consistent_head.vm_list), + .vm_start = CONSISTENT_BASE, + .vm_end = CONSISTENT_END, +}; + +static struct ppc_vm_region * +ppc_vm_region_alloc(struct ppc_vm_region *head, size_t size, gfp_t gfp) +{ + unsigned long addr = head->vm_start, end = head->vm_end - size; + unsigned long flags; + struct ppc_vm_region *c, *new; + + new = kmalloc(sizeof(struct ppc_vm_region), gfp); + if (!new) + goto out; + + spin_lock_irqsave(&consistent_lock, flags); + + list_for_each_entry(c, &head->vm_list, vm_list) { + if ((addr + size) < addr) + goto nospc; + if ((addr + size) <= c->vm_start) + goto found; + addr = c->vm_end; + if (addr > end) + goto nospc; + } + + found: + /* + * Insert this entry _before_ the one we found. + */ + list_add_tail(&new->vm_list, &c->vm_list); + new->vm_start = addr; + new->vm_end = addr + size; + + spin_unlock_irqrestore(&consistent_lock, flags); + return new; + + nospc: + spin_unlock_irqrestore(&consistent_lock, flags); + kfree(new); + out: + return NULL; +} + +static struct ppc_vm_region *ppc_vm_region_find(struct ppc_vm_region *head, unsigned long addr) +{ + struct ppc_vm_region *c; + + list_for_each_entry(c, &head->vm_list, vm_list) { + if (c->vm_start == addr) + goto out; + } + c = NULL; + out: + return c; +} + /* * Allocate DMA-coherent memory space and return both the kernel remapped * virtual and bus address for that space. @@ -41,21 +151,21 @@ void * __dma_alloc_coherent(size_t size, dma_addr_t *handle, gfp_t gfp) { struct page *page; + struct ppc_vm_region *c; unsigned long order; - int i; - unsigned int nr_pages = PAGE_ALIGN(size)>>PAGE_SHIFT; - unsigned int array_size = nr_pages * sizeof(struct page *); - struct page **pages; - struct page *end; u64 mask = 0x00ffffff, limit; /* ISA default */ - struct vm_struct *area; - BUG_ON(!mem_init_done); + if (!consistent_pte) { + printk(KERN_ERR "%s: not initialised\n", __func__); + dump_stack(); + return NULL; + } + size = PAGE_ALIGN(size); limit = (mask + 1) & ~mask; - if (limit && size >= limit) { - printk(KERN_WARNING "coherent allocation too big (requested " - "%#x mask %#Lx)\n", size, mask); + if ((limit && size >= limit) || size >= (CONSISTENT_END - CONSISTENT_BASE)) { + printk(KERN_WARNING "coherent allocation too big (requested %#x mask %#Lx)\n", + size, mask); return NULL; } @@ -68,8 +178,6 @@ __dma_alloc_coherent(size_t size, dma_addr_t *handle, gfp_t gfp) if (!page) goto no_page; - end = page + (1 << order); - /* * Invalidate any data that might be lurking in the * kernel direct-mapped region for device DMA. @@ -80,59 +188,48 @@ __dma_alloc_coherent(size_t size, dma_addr_t *handle, gfp_t gfp) flush_dcache_range(kaddr, kaddr + size); } - split_page(page, order); - /* - * Set the "dma handle" + * Allocate a virtual address in the consistent mapping region. */ - *handle = page_to_phys(page); + c = ppc_vm_region_alloc(&consistent_head, size, + gfp & ~(__GFP_DMA | __GFP_HIGHMEM)); + if (c) { + unsigned long vaddr = c->vm_start; + pte_t *pte = consistent_pte + CONSISTENT_OFFSET(vaddr); + struct page *end = page + (1 << order); - area = get_vm_area_caller(size, VM_IOREMAP, - __builtin_return_address(1)); - if (!area) - goto out_free_pages; + split_page(page, order); - if (array_size > PAGE_SIZE) { - pages = vmalloc(array_size); - area->flags |= VM_VPAGES; - } else { - pages = kmalloc(array_size, GFP_KERNEL); - } - if (!pages) - goto out_free_area; + /* + * Set the "dma handle" + */ + *handle = page_to_phys(page); - area->pages = pages; - area->nr_pages = nr_pages; + do { + BUG_ON(!pte_none(*pte)); - for (i = 0; i < nr_pages; i++) - pages[i] = page + i; + SetPageReserved(page); + set_pte_at(&init_mm, vaddr, + pte, mk_pte(page, pgprot_noncached(PAGE_KERNEL))); + page++; + pte++; + vaddr += PAGE_SIZE; + } while (size -= PAGE_SIZE); - if (map_vm_area(area, pgprot_noncached(PAGE_KERNEL), &pages)) - goto out_unmap; + /* + * Free the otherwise unused pages. + */ + while (page < end) { + __free_page(page); + page++; + } - /* - * Free the otherwise unused pages. - */ - page += nr_pages; - while (page < end) { - __free_page(page); - page++; + return (void *)c->vm_start; } - return area->addr; -out_unmap: - vunmap(area->addr); - if (array_size > PAGE_SIZE) - vfree(pages); - else - kfree(pages); - goto out_free_pages; -out_free_area: - free_vm_area(area); -out_free_pages: if (page) __free_pages(page, order); -no_page: + no_page: return NULL; } EXPORT_SYMBOL(__dma_alloc_coherent); @@ -142,11 +239,103 @@ EXPORT_SYMBOL(__dma_alloc_coherent); */ void __dma_free_coherent(size_t size, void *vaddr) { - vfree(vaddr); + struct ppc_vm_region *c; + unsigned long flags, addr; + pte_t *ptep; + size = PAGE_ALIGN(size); + + spin_lock_irqsave(&consistent_lock, flags); + + c = ppc_vm_region_find(&consistent_head, (unsigned long)vaddr); + if (!c) + goto no_area; + + if ((c->vm_end - c->vm_start) != size) { + printk(KERN_ERR "%s: freeing wrong coherent size (%ld != %d)\n", + __func__, c->vm_end - c->vm_start, size); + dump_stack(); + size = c->vm_end - c->vm_start; + } + + ptep = consistent_pte + CONSISTENT_OFFSET(c->vm_start); + addr = c->vm_start; + do { + pte_t pte = ptep_get_and_clear(&init_mm, addr, ptep); + unsigned long pfn; + + ptep++; + addr += PAGE_SIZE; + + if (!pte_none(pte) && pte_present(pte)) { + pfn = pte_pfn(pte); + + if (pfn_valid(pfn)) { + struct page *page = pfn_to_page(pfn); + ClearPageReserved(page); + + __free_page(page); + continue; + } + } + + printk(KERN_CRIT "%s: bad page in kernel page table\n", + __func__); + } while (size -= PAGE_SIZE); + + flush_tlb_kernel_range(c->vm_start, c->vm_end); + + list_del(&c->vm_list); + + spin_unlock_irqrestore(&consistent_lock, flags); + + kfree(c); + return; + + no_area: + spin_unlock_irqrestore(&consistent_lock, flags); + printk(KERN_ERR "%s: trying to free invalid coherent area: %p\n", + __func__, vaddr); + dump_stack(); } EXPORT_SYMBOL(__dma_free_coherent); +/* + * Initialise the consistent memory allocation. + */ +static int __init dma_alloc_init(void) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + int ret = 0; + + do { + pgd = pgd_offset(&init_mm, CONSISTENT_BASE); + pud = pud_alloc(&init_mm, pgd, CONSISTENT_BASE); + pmd = pmd_alloc(&init_mm, pud, CONSISTENT_BASE); + if (!pmd) { + printk(KERN_ERR "%s: no pmd tables\n", __func__); + ret = -ENOMEM; + break; + } + + pte = pte_alloc_kernel(pmd, CONSISTENT_BASE); + if (!pte) { + printk(KERN_ERR "%s: no pte tables\n", __func__); + ret = -ENOMEM; + break; + } + + consistent_pte = pte; + } while (0); + + return ret; +} + +core_initcall(dma_alloc_init); + /* * make an area consistent. */ From 7a1450fdf4c69961f3926352fd8bc4ea19676756 Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Tue, 26 May 2009 04:55:38 -0400 Subject: [PATCH 218/267] Blackfin: hook up preadv/pwritev syscalls Signed-off-by: Mike Frysinger --- arch/blackfin/include/asm/unistd.h | 4 +++- arch/blackfin/mach-common/entry.S | 2 ++ 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/blackfin/include/asm/unistd.h b/arch/blackfin/include/asm/unistd.h index 1e57b636e0bc..cf5066d3efd2 100644 --- a/arch/blackfin/include/asm/unistd.h +++ b/arch/blackfin/include/asm/unistd.h @@ -378,8 +378,10 @@ #define __NR_dup3 363 #define __NR_pipe2 364 #define __NR_inotify_init1 365 +#define __NR_preadv 366 +#define __NR_pwritev 367 -#define __NR_syscall 366 +#define __NR_syscall 368 #define NR_syscalls __NR_syscall /* Old optional stuff no one actually uses */ diff --git a/arch/blackfin/mach-common/entry.S b/arch/blackfin/mach-common/entry.S index 21e65a339a22..a063a434f7e3 100644 --- a/arch/blackfin/mach-common/entry.S +++ b/arch/blackfin/mach-common/entry.S @@ -1581,6 +1581,8 @@ ENTRY(_sys_call_table) .long _sys_dup3 .long _sys_pipe2 .long _sys_inotify_init1 /* 365 */ + .long _sys_preadv + .long _sys_pwritev .rept NR_syscalls-(.-_sys_call_table)/4 .long _sys_ni_syscall From 6c83429a1c32c914dfb81939cc2ddece97e48294 Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Sun, 24 May 2009 02:13:15 -0400 Subject: [PATCH 219/267] MAINTAINERS: update Blackfin items With Bryan Wu having moved on to another job, push the slack onto some other ADI lackeys. Signed-off-by: Mike Frysinger --- MAINTAINERS | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 77cbfb1a696c..7dd34750264f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1132,16 +1132,16 @@ F: fs/bfs/ F: include/linux/bfs_fs.h BLACKFIN ARCHITECTURE -P: Bryan Wu -M: cooloney@kernel.org +P: Mike Frysinger +M: vapier@gentoo.org L: uclinux-dist-devel@blackfin.uclinux.org W: http://blackfin.uclinux.org S: Supported F: arch/blackfin/ BLACKFIN EMAC DRIVER -P: Bryan Wu -M: cooloney@kernel.org +P: Michael Hennerich +M: michael.hennerich@analog.com L: uclinux-dist-devel@blackfin.uclinux.org (subscribers-only) W: http://blackfin.uclinux.org S: Supported From 49afa60948f859e71d68a74c1af6ccd7b5b94d82 Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Mon, 18 May 2009 04:33:07 -0400 Subject: [PATCH 220/267] MAINTAINERS: drop (subscribers-only) markings on Blackfin lists All of the Blackfin lists are transparently moderated for non-subscribers. i.e. there are no annoying notices and people get whitelisted after first their posting. Signed-off-by: Mike Frysinger --- MAINTAINERS | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 7dd34750264f..5ee166e27b9f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1142,7 +1142,7 @@ F: arch/blackfin/ BLACKFIN EMAC DRIVER P: Michael Hennerich M: michael.hennerich@analog.com -L: uclinux-dist-devel@blackfin.uclinux.org (subscribers-only) +L: uclinux-dist-devel@blackfin.uclinux.org W: http://blackfin.uclinux.org S: Supported F: drivers/net/bfin_mac.* @@ -1150,7 +1150,7 @@ F: drivers/net/bfin_mac.* BLACKFIN RTC DRIVER P: Mike Frysinger M: vapier.adi@gmail.com -L: uclinux-dist-devel@blackfin.uclinux.org (subscribers-only) +L: uclinux-dist-devel@blackfin.uclinux.org W: http://blackfin.uclinux.org S: Supported F: drivers/rtc/rtc-bfin.c @@ -1158,7 +1158,7 @@ F: drivers/rtc/rtc-bfin.c BLACKFIN SERIAL DRIVER P: Sonic Zhang M: sonic.zhang@analog.com -L: uclinux-dist-devel@blackfin.uclinux.org (subscribers-only) +L: uclinux-dist-devel@blackfin.uclinux.org W: http://blackfin.uclinux.org S: Supported F: drivers/serial/bfin_5xx.c @@ -1166,7 +1166,7 @@ F: drivers/serial/bfin_5xx.c BLACKFIN WATCHDOG DRIVER P: Mike Frysinger M: vapier.adi@gmail.com -L: uclinux-dist-devel@blackfin.uclinux.org (subscribers-only) +L: uclinux-dist-devel@blackfin.uclinux.org W: http://blackfin.uclinux.org S: Supported F: drivers/watchdog/bfin_wdt.c @@ -1174,7 +1174,7 @@ F: drivers/watchdog/bfin_wdt.c BLACKFIN I2C TWI DRIVER P: Sonic Zhang M: sonic.zhang@analog.com -L: uclinux-dist-devel@blackfin.uclinux.org (subscribers-only) +L: uclinux-dist-devel@blackfin.uclinux.org W: http://blackfin.uclinux.org/ S: Supported F: drivers/i2c/busses/i2c-bfin-twi.c From 6b50520b2fd9bf521f9c947b5f6999bad273a51d Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Tue, 19 May 2009 10:03:22 -0400 Subject: [PATCH 221/267] Blackfin: ignore generated vmlinux.lds Signed-off-by: Mike Frysinger --- arch/blackfin/kernel/.gitignore | 1 + 1 file changed, 1 insertion(+) create mode 100644 arch/blackfin/kernel/.gitignore diff --git a/arch/blackfin/kernel/.gitignore b/arch/blackfin/kernel/.gitignore new file mode 100644 index 000000000000..c5f676c3c224 --- /dev/null +++ b/arch/blackfin/kernel/.gitignore @@ -0,0 +1 @@ +vmlinux.lds From 2ec10ea91bf3688013b00638f29df4f8f6b5c18b Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Wed, 20 May 2009 19:45:39 -0400 Subject: [PATCH 222/267] Blackfin: drop unneeded asm/.gitignore We don't create a include/asm/mach/ symlink anymore, so we don't need the .gitignore for it. Signed-off-by: Mike Frysinger --- arch/blackfin/include/asm/.gitignore | 1 - 1 file changed, 1 deletion(-) delete mode 100644 arch/blackfin/include/asm/.gitignore diff --git a/arch/blackfin/include/asm/.gitignore b/arch/blackfin/include/asm/.gitignore deleted file mode 100644 index 7858564a4466..000000000000 --- a/arch/blackfin/include/asm/.gitignore +++ /dev/null @@ -1 +0,0 @@ -+mach From add8a5050a52f1bd1be6b97be86fdd1cfbea2d1d Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Tue, 26 May 2009 05:03:52 -0400 Subject: [PATCH 223/267] Blackfin: fix strncmp.o build error MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix some more fallout of the string changes: CC arch/blackfin/lib/strncmp.o In file included from include/linux/bitmap.h:9, from include/linux/nodemask.h:90, from include/linux/mmzone.h:17, from include/linux/gfp.h:5, from include/linux/kmod.h:23, from include/linux/module.h:14, from arch/blackfin/lib/strncmp.c:14: include/linux/string.h: In function ‘strstarts’: include/linux/string.h:132: error: implicit declaration of function ‘strncmp’ make[1]: *** [arch/blackfin/lib/strncmp.o] Error 1 Signed-off-by: Mike Frysinger CC: Rusty Russell --- arch/blackfin/lib/strncmp.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/blackfin/lib/strncmp.c b/arch/blackfin/lib/strncmp.c index 2aaae78a68e0..46518b1d2983 100644 --- a/arch/blackfin/lib/strncmp.c +++ b/arch/blackfin/lib/strncmp.c @@ -8,9 +8,8 @@ #define strncmp __inline_strncmp #include -#undef strncmp - #include +#undef strncmp int strncmp(const char *cs, const char *ct, size_t count) { From b16e7766d6436835f473ba823ad04fbdfe5e9cbd Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 27 May 2009 13:36:10 +1000 Subject: [PATCH 224/267] powerpc: Move dma-noncoherent.c from arch/powerpc/lib to arch/powerpc/mm (pre-requisite to make the next patches more palatable) Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/lib/Makefile | 1 - arch/powerpc/mm/Makefile | 1 + arch/powerpc/{lib => mm}/dma-noncoherent.c | 0 3 files changed, 1 insertion(+), 1 deletion(-) rename arch/powerpc/{lib => mm}/dma-noncoherent.c (100%) diff --git a/arch/powerpc/lib/Makefile b/arch/powerpc/lib/Makefile index 8db35278a4b4..29b742b90f1f 100644 --- a/arch/powerpc/lib/Makefile +++ b/arch/powerpc/lib/Makefile @@ -18,7 +18,6 @@ obj-$(CONFIG_PPC64) += copypage_64.o copyuser_64.o \ memcpy_64.o usercopy_64.o mem_64.o string.o obj-$(CONFIG_XMON) += sstep.o obj-$(CONFIG_KPROBES) += sstep.o -obj-$(CONFIG_NOT_COHERENT_CACHE) += dma-noncoherent.o ifeq ($(CONFIG_PPC64),y) obj-$(CONFIG_SMP) += locks.o diff --git a/arch/powerpc/mm/Makefile b/arch/powerpc/mm/Makefile index 17290bcedc5e..b746f4ca4209 100644 --- a/arch/powerpc/mm/Makefile +++ b/arch/powerpc/mm/Makefile @@ -26,3 +26,4 @@ obj-$(CONFIG_NEED_MULTIPLE_NODES) += numa.o obj-$(CONFIG_PPC_MM_SLICES) += slice.o obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o obj-$(CONFIG_PPC_SUBPAGE_PROT) += subpage-prot.o +obj-$(CONFIG_NOT_COHERENT_CACHE) += dma-noncoherent.o diff --git a/arch/powerpc/lib/dma-noncoherent.c b/arch/powerpc/mm/dma-noncoherent.c similarity index 100% rename from arch/powerpc/lib/dma-noncoherent.c rename to arch/powerpc/mm/dma-noncoherent.c From f637a49e507c88354ab32b5d914e06acfb7ee00d Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 27 May 2009 13:44:50 +1000 Subject: [PATCH 225/267] powerpc: Minor cleanups of kernel virt address space definitions Make FIXADDR_TOP a compile time constant and cleanup a couple of definitions relative to the layout of the kernel address space on ppc32. We also print out that layout at boot time for debugging purposes. This is a pre-requisite for properly fixing non-coherent DMA allocactions. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/fixmap.h | 4 ++-- arch/powerpc/include/asm/pgtable-ppc32.h | 22 ++++++++++++++++++++-- arch/powerpc/mm/init_32.c | 8 ++------ arch/powerpc/mm/mem.c | 13 +++++++++++++ arch/powerpc/mm/pgtable_32.c | 2 -- 5 files changed, 37 insertions(+), 12 deletions(-) diff --git a/arch/powerpc/include/asm/fixmap.h b/arch/powerpc/include/asm/fixmap.h index d60fd18f428c..f1f4e23a84e9 100644 --- a/arch/powerpc/include/asm/fixmap.h +++ b/arch/powerpc/include/asm/fixmap.h @@ -14,8 +14,6 @@ #ifndef _ASM_FIXMAP_H #define _ASM_FIXMAP_H -extern unsigned long FIXADDR_TOP; - #ifndef __ASSEMBLY__ #include #include @@ -24,6 +22,8 @@ extern unsigned long FIXADDR_TOP; #include #endif +#define FIXADDR_TOP ((unsigned long)(-PAGE_SIZE)) + /* * Here we define all the compile-time 'special' virtual * addresses. The point is to have a constant address at diff --git a/arch/powerpc/include/asm/pgtable-ppc32.h b/arch/powerpc/include/asm/pgtable-ppc32.h index ba45c997830f..28fe9d4bae35 100644 --- a/arch/powerpc/include/asm/pgtable-ppc32.h +++ b/arch/powerpc/include/asm/pgtable-ppc32.h @@ -10,7 +10,7 @@ extern unsigned long va_to_phys(unsigned long address); extern pte_t *va_to_pte(unsigned long address); -extern unsigned long ioremap_bot, ioremap_base; +extern unsigned long ioremap_bot; #ifdef CONFIG_44x extern int icache_44x_need_flush; @@ -55,9 +55,27 @@ extern int icache_44x_need_flush; #define pgd_ERROR(e) \ printk("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e)) +/* + * This is the bottom of the PKMAP area with HIGHMEM or an arbitrary + * value (for now) on others, from where we can start layout kernel + * virtual space that goes below PKMAP and FIXMAP + */ +#ifdef CONFIG_HIGHMEM +#define KVIRT_TOP PKMAP_BASE +#else +#define KVIRT_TOP (0xfe000000UL) /* for now, could be FIXMAP_BASE ? */ +#endif + +/* + * ioremap_bot starts at that address. Early ioremaps move down from there, + * until mem_init() at which point this becomes the top of the vmalloc + * and ioremap space + */ +#define IOREMAP_TOP KVIRT_TOP + /* * Just any arbitrary offset to the start of the vmalloc VM area: the - * current 64MB value just means that there will be a 64MB "hole" after the + * current 16MB value just means that there will be a 64MB "hole" after the * physical memory until the kernel virtual memory starts. That means that * any out-of-bounds memory accesses will hopefully be caught. * The vmalloc() routines leaves a hole of 4kB between each vmalloced diff --git a/arch/powerpc/mm/init_32.c b/arch/powerpc/mm/init_32.c index 666a5e8a5be1..3de6a0d93824 100644 --- a/arch/powerpc/mm/init_32.c +++ b/arch/powerpc/mm/init_32.c @@ -168,12 +168,8 @@ void __init MMU_init(void) ppc_md.progress("MMU:mapin", 0x301); mapin_ram(); -#ifdef CONFIG_HIGHMEM - ioremap_base = PKMAP_BASE; -#else - ioremap_base = 0xfe000000UL; /* for now, could be 0xfffff000 */ -#endif /* CONFIG_HIGHMEM */ - ioremap_bot = ioremap_base; + /* Initialize early top-down ioremap allocator */ + ioremap_bot = IOREMAP_TOP; /* Map in I/O resources */ if (ppc_md.progress) diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c index d0602a76bf7f..d3a4e67561fa 100644 --- a/arch/powerpc/mm/mem.c +++ b/arch/powerpc/mm/mem.c @@ -380,6 +380,19 @@ void __init mem_init(void) bsssize >> 10, initsize >> 10); +#ifdef CONFIG_PPC32 + pr_info("Kernel virtual memory layout:\n"); + pr_info(" * 0x%08lx..0x%08lx : fixmap\n", FIXADDR_START, FIXADDR_TOP); +#ifdef CONFIG_HIGHMEM + pr_info(" * 0x%08lx..0x%08lx : highmem PTEs\n", + PKMAP_BASE, PKMAP_ADDR(LAST_PKMAP)); +#endif /* CONFIG_HIGHMEM */ + pr_info(" * 0x%08lx..0x%08lx : early ioremap\n", + ioremap_bot, IOREMAP_TOP); + pr_info(" * 0x%08lx..0x%08lx : vmalloc & ioremap\n", + VMALLOC_START, VMALLOC_END); +#endif /* CONFIG_PPC32 */ + mem_init_done = 1; } diff --git a/arch/powerpc/mm/pgtable_32.c b/arch/powerpc/mm/pgtable_32.c index 430d0908fa50..5422169626ba 100644 --- a/arch/powerpc/mm/pgtable_32.c +++ b/arch/powerpc/mm/pgtable_32.c @@ -399,8 +399,6 @@ void kernel_map_pages(struct page *page, int numpages, int enable) #endif /* CONFIG_DEBUG_PAGEALLOC */ static int fixmaps; -unsigned long FIXADDR_TOP = (-PAGE_SIZE); -EXPORT_SYMBOL(FIXADDR_TOP); void __set_fixmap (enum fixed_addresses idx, phys_addr_t phys, pgprot_t flags) { From 8b31e49d1d75729c1da9009664ba52abd1adc628 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 27 May 2009 13:50:33 +1000 Subject: [PATCH 226/267] powerpc: Fix up dma_alloc_coherent() on platforms without cache coherency. The implementation we just revived has issues, such as using a Kconfig-defined virtual address area in kernel space that nothing actually carves out (and thus will overlap whatever is there), or having some dependencies on being self contained in a single PTE page which adds unnecessary constraints on the kernel virtual address space. This fixes it by using more classic PTE accessors and automatically locating the area for consistent memory, carving an appropriate hole in the kernel virtual address space, leaving only the size of that area as a Kconfig option. It also brings some dma-mask related fixes from the ARM implementation which was almost identical initially but grew its own fixes. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/Kconfig | 13 --- arch/powerpc/include/asm/dma-mapping.h | 6 +- arch/powerpc/include/asm/pgtable-ppc32.h | 4 + arch/powerpc/kernel/dma.c | 2 +- arch/powerpc/mm/dma-noncoherent.c | 108 +++++++++-------------- arch/powerpc/mm/mem.c | 4 + 6 files changed, 54 insertions(+), 83 deletions(-) diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 3bb43adce44d..cdc9a6ff4be8 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -868,19 +868,6 @@ config TASK_SIZE default "0x80000000" if PPC_PREP || PPC_8xx default "0xc0000000" -config CONSISTENT_START_BOOL - bool "Set custom consistent memory pool address" - depends on ADVANCED_OPTIONS && NOT_COHERENT_CACHE - help - This option allows you to set the base virtual address - of the consistent memory pool. This pool of virtual - memory is used to make consistent memory allocations. - -config CONSISTENT_START - hex "Base virtual address of consistent memory pool" if CONSISTENT_START_BOOL - default "0xfd000000" if (NOT_COHERENT_CACHE && 8xx) - default "0xff100000" if NOT_COHERENT_CACHE - config CONSISTENT_SIZE_BOOL bool "Set custom consistent memory pool size" depends on ADVANCED_OPTIONS && NOT_COHERENT_CACHE diff --git a/arch/powerpc/include/asm/dma-mapping.h b/arch/powerpc/include/asm/dma-mapping.h index c69f2b5f0cc4..cb448d68452c 100644 --- a/arch/powerpc/include/asm/dma-mapping.h +++ b/arch/powerpc/include/asm/dma-mapping.h @@ -26,7 +26,9 @@ * allocate the space "normally" and use the cache management functions * to ensure it is consistent. */ -extern void *__dma_alloc_coherent(size_t size, dma_addr_t *handle, gfp_t gfp); +struct device; +extern void *__dma_alloc_coherent(struct device *dev, size_t size, + dma_addr_t *handle, gfp_t gfp); extern void __dma_free_coherent(size_t size, void *vaddr); extern void __dma_sync(void *vaddr, size_t size, int direction); extern void __dma_sync_page(struct page *page, unsigned long offset, @@ -37,7 +39,7 @@ extern void __dma_sync_page(struct page *page, unsigned long offset, * Cache coherent cores. */ -#define __dma_alloc_coherent(gfp, size, handle) NULL +#define __dma_alloc_coherent(dev, gfp, size, handle) NULL #define __dma_free_coherent(size, addr) ((void)0) #define __dma_sync(addr, size, rw) ((void)0) #define __dma_sync_page(pg, off, sz, rw) ((void)0) diff --git a/arch/powerpc/include/asm/pgtable-ppc32.h b/arch/powerpc/include/asm/pgtable-ppc32.h index 28fe9d4bae35..c9ff9d75990e 100644 --- a/arch/powerpc/include/asm/pgtable-ppc32.h +++ b/arch/powerpc/include/asm/pgtable-ppc32.h @@ -71,7 +71,11 @@ extern int icache_44x_need_flush; * until mem_init() at which point this becomes the top of the vmalloc * and ioremap space */ +#ifdef CONFIG_NOT_COHERENT_CACHE +#define IOREMAP_TOP ((KVIRT_TOP - CONFIG_CONSISTENT_SIZE) & PAGE_MASK) +#else #define IOREMAP_TOP KVIRT_TOP +#endif /* * Just any arbitrary offset to the start of the vmalloc VM area: the diff --git a/arch/powerpc/kernel/dma.c b/arch/powerpc/kernel/dma.c index 53c7788cba78..6b02793dc75b 100644 --- a/arch/powerpc/kernel/dma.c +++ b/arch/powerpc/kernel/dma.c @@ -32,7 +32,7 @@ void *dma_direct_alloc_coherent(struct device *dev, size_t size, { void *ret; #ifdef CONFIG_NOT_COHERENT_CACHE - ret = __dma_alloc_coherent(size, dma_handle, flag); + ret = __dma_alloc_coherent(dev, size, dma_handle, flag); if (ret == NULL) return NULL; *dma_handle += get_dma_direct_offset(dev); diff --git a/arch/powerpc/mm/dma-noncoherent.c b/arch/powerpc/mm/dma-noncoherent.c index b7dc4c19f582..36692f5c9a76 100644 --- a/arch/powerpc/mm/dma-noncoherent.c +++ b/arch/powerpc/mm/dma-noncoherent.c @@ -32,20 +32,21 @@ #include +#include "mmu_decl.h" + /* * This address range defaults to a value that is safe for all * platforms which currently set CONFIG_NOT_COHERENT_CACHE. It * can be further configured for specific applications under * the "Advanced Setup" menu. -Matt */ -#define CONSISTENT_BASE (CONFIG_CONSISTENT_START) -#define CONSISTENT_END (CONFIG_CONSISTENT_START + CONFIG_CONSISTENT_SIZE) +#define CONSISTENT_BASE (IOREMAP_TOP) +#define CONSISTENT_END (CONSISTENT_BASE + CONFIG_CONSISTENT_SIZE) #define CONSISTENT_OFFSET(x) (((unsigned long)(x) - CONSISTENT_BASE) >> PAGE_SHIFT) /* * This is the page table (2MB) covering uncached, DMA consistent allocations */ -static pte_t *consistent_pte; static DEFINE_SPINLOCK(consistent_lock); /* @@ -148,22 +149,38 @@ static struct ppc_vm_region *ppc_vm_region_find(struct ppc_vm_region *head, unsi * virtual and bus address for that space. */ void * -__dma_alloc_coherent(size_t size, dma_addr_t *handle, gfp_t gfp) +__dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *handle, gfp_t gfp) { struct page *page; struct ppc_vm_region *c; unsigned long order; - u64 mask = 0x00ffffff, limit; /* ISA default */ + u64 mask = ISA_DMA_THRESHOLD, limit; - if (!consistent_pte) { - printk(KERN_ERR "%s: not initialised\n", __func__); - dump_stack(); - return NULL; + if (dev) { + mask = dev->coherent_dma_mask; + + /* + * Sanity check the DMA mask - it must be non-zero, and + * must be able to be satisfied by a DMA allocation. + */ + if (mask == 0) { + dev_warn(dev, "coherent DMA mask is unset\n"); + goto no_page; + } + + if ((~mask) & ISA_DMA_THRESHOLD) { + dev_warn(dev, "coherent DMA mask %#llx is smaller " + "than system GFP_DMA mask %#llx\n", + mask, (unsigned long long)ISA_DMA_THRESHOLD); + goto no_page; + } } + size = PAGE_ALIGN(size); limit = (mask + 1) & ~mask; - if ((limit && size >= limit) || size >= (CONSISTENT_END - CONSISTENT_BASE)) { + if ((limit && size >= limit) || + size >= (CONSISTENT_END - CONSISTENT_BASE)) { printk(KERN_WARNING "coherent allocation too big (requested %#x mask %#Lx)\n", size, mask); return NULL; @@ -171,6 +188,7 @@ __dma_alloc_coherent(size_t size, dma_addr_t *handle, gfp_t gfp) order = get_order(size); + /* Might be useful if we ever have a real legacy DMA zone... */ if (mask != 0xffffffff) gfp |= GFP_DMA; @@ -195,7 +213,6 @@ __dma_alloc_coherent(size_t size, dma_addr_t *handle, gfp_t gfp) gfp & ~(__GFP_DMA | __GFP_HIGHMEM)); if (c) { unsigned long vaddr = c->vm_start; - pte_t *pte = consistent_pte + CONSISTENT_OFFSET(vaddr); struct page *end = page + (1 << order); split_page(page, order); @@ -206,13 +223,10 @@ __dma_alloc_coherent(size_t size, dma_addr_t *handle, gfp_t gfp) *handle = page_to_phys(page); do { - BUG_ON(!pte_none(*pte)); - SetPageReserved(page); - set_pte_at(&init_mm, vaddr, - pte, mk_pte(page, pgprot_noncached(PAGE_KERNEL))); + map_page(vaddr, page_to_phys(page), + pgprot_noncached(PAGE_KERNEL)); page++; - pte++; vaddr += PAGE_SIZE; } while (size -= PAGE_SIZE); @@ -241,8 +255,7 @@ void __dma_free_coherent(size_t size, void *vaddr) { struct ppc_vm_region *c; unsigned long flags, addr; - pte_t *ptep; - + size = PAGE_ALIGN(size); spin_lock_irqsave(&consistent_lock, flags); @@ -258,29 +271,26 @@ void __dma_free_coherent(size_t size, void *vaddr) size = c->vm_end - c->vm_start; } - ptep = consistent_pte + CONSISTENT_OFFSET(c->vm_start); addr = c->vm_start; do { - pte_t pte = ptep_get_and_clear(&init_mm, addr, ptep); + pte_t *ptep; unsigned long pfn; - ptep++; - addr += PAGE_SIZE; - - if (!pte_none(pte) && pte_present(pte)) { - pfn = pte_pfn(pte); - + ptep = pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k(addr), + addr), + addr), + addr); + if (!pte_none(*ptep) && pte_present(*ptep)) { + pfn = pte_pfn(*ptep); + pte_clear(&init_mm, addr, ptep); if (pfn_valid(pfn)) { struct page *page = pfn_to_page(pfn); - ClearPageReserved(page); + ClearPageReserved(page); __free_page(page); - continue; } } - - printk(KERN_CRIT "%s: bad page in kernel page table\n", - __func__); + addr += PAGE_SIZE; } while (size -= PAGE_SIZE); flush_tlb_kernel_range(c->vm_start, c->vm_end); @@ -300,42 +310,6 @@ void __dma_free_coherent(size_t size, void *vaddr) } EXPORT_SYMBOL(__dma_free_coherent); -/* - * Initialise the consistent memory allocation. - */ -static int __init dma_alloc_init(void) -{ - pgd_t *pgd; - pud_t *pud; - pmd_t *pmd; - pte_t *pte; - int ret = 0; - - do { - pgd = pgd_offset(&init_mm, CONSISTENT_BASE); - pud = pud_alloc(&init_mm, pgd, CONSISTENT_BASE); - pmd = pmd_alloc(&init_mm, pud, CONSISTENT_BASE); - if (!pmd) { - printk(KERN_ERR "%s: no pmd tables\n", __func__); - ret = -ENOMEM; - break; - } - - pte = pte_alloc_kernel(pmd, CONSISTENT_BASE); - if (!pte) { - printk(KERN_ERR "%s: no pte tables\n", __func__); - ret = -ENOMEM; - break; - } - - consistent_pte = pte; - } while (0); - - return ret; -} - -core_initcall(dma_alloc_init); - /* * make an area consistent. */ diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c index d3a4e67561fa..579382c163a9 100644 --- a/arch/powerpc/mm/mem.c +++ b/arch/powerpc/mm/mem.c @@ -387,6 +387,10 @@ void __init mem_init(void) pr_info(" * 0x%08lx..0x%08lx : highmem PTEs\n", PKMAP_BASE, PKMAP_ADDR(LAST_PKMAP)); #endif /* CONFIG_HIGHMEM */ +#ifdef CONFIG_NOT_COHERENT_CACHE + pr_info(" * 0x%08lx..0x%08lx : consistent mem\n", + IOREMAP_TOP, IOREMAP_TOP + CONFIG_CONSISTENT_SIZE); +#endif /* CONFIG_NOT_COHERENT_CACHE */ pr_info(" * 0x%08lx..0x%08lx : early ioremap\n", ioremap_bot, IOREMAP_TOP); pr_info(" * 0x%08lx..0x%08lx : vmalloc & ioremap\n", From 6af3fb72d2437239e5eb13a59e95dc43ccab3e8f Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 27 May 2009 10:49:26 +0200 Subject: [PATCH 227/267] ALSA: Fix invalid jiffies check after pause The hw_ptr_jiffies has to be reset properly to avoid the invalid check of jiffies delta in snd_pcm_update_hw_ptr*() functions. Especailly this patch fixes the bogus jiffies check after the puase and resume. This patch is a modified version of the original patch by Jaroslav. Signed-off-by: Takashi Iwai --- sound/core/pcm_lib.c | 1 - sound/core/pcm_native.c | 6 ++++++ 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/sound/core/pcm_lib.c b/sound/core/pcm_lib.c index a2a792c18c40..3eea98a4e65a 100644 --- a/sound/core/pcm_lib.c +++ b/sound/core/pcm_lib.c @@ -1478,7 +1478,6 @@ static int snd_pcm_lib_ioctl_reset(struct snd_pcm_substream *substream, runtime->status->hw_ptr %= runtime->buffer_size; else runtime->status->hw_ptr = 0; - runtime->hw_ptr_jiffies = jiffies; snd_pcm_stream_unlock_irqrestore(substream, flags); return 0; } diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c index fc6f98e257df..b5da656d1ece 100644 --- a/sound/core/pcm_native.c +++ b/sound/core/pcm_native.c @@ -848,6 +848,7 @@ static void snd_pcm_post_start(struct snd_pcm_substream *substream, int state) { struct snd_pcm_runtime *runtime = substream->runtime; snd_pcm_trigger_tstamp(substream); + runtime->hw_ptr_jiffies = jiffies; runtime->status->state = state; if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK && runtime->silence_size > 0) @@ -961,6 +962,11 @@ static int snd_pcm_do_pause(struct snd_pcm_substream *substream, int push) { if (substream->runtime->trigger_master != substream) return 0; + /* The jiffies check in snd_pcm_update_hw_ptr*() is done by + * a delta betwen the current jiffies, this gives a large enough + * delta, effectively to skip the check once. + */ + substream->runtime->hw_ptr_jiffies = jiffies - HZ * 1000; return substream->ops->trigger(substream, push ? SNDRV_PCM_TRIGGER_PAUSE_PUSH : SNDRV_PCM_TRIGGER_PAUSE_RELEASE); From c87d9732004b3f8fd82d729f12ccfb96c0df279e Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 27 May 2009 10:53:33 +0200 Subject: [PATCH 228/267] ALSA: Enable PCM hw_ptr_jiffies check only in xrun_debug mode The PCM hw_ptr jiffies check results sometimes in problems when a hardware doesn't give smooth hw_ptr updates. So far, au88x0 and some other drivers appear not working due to this strict check. However, this check is a nice debug tool, and the capability should be still kept. Hence, we disable this check now as default unless the user enables it by setting the xrun_debug mode to the specific stream via a proc file. Signed-off-by: Takashi Iwai --- Documentation/sound/alsa/Procfile.txt | 5 +++++ sound/core/pcm_lib.c | 9 ++++++++- 2 files changed, 13 insertions(+), 1 deletion(-) diff --git a/Documentation/sound/alsa/Procfile.txt b/Documentation/sound/alsa/Procfile.txt index bba2dbb79d81..cfac20cf9e33 100644 --- a/Documentation/sound/alsa/Procfile.txt +++ b/Documentation/sound/alsa/Procfile.txt @@ -104,6 +104,11 @@ card*/pcm*/xrun_debug When this value is greater than 1, the driver will show the stack trace additionally. This may help the debugging. + Since 2.6.30, this option also enables the hwptr check using + jiffies. This detects spontaneous invalid pointer callback + values, but can be lead to too much corrections for a (mostly + buggy) hardware that doesn't give smooth pointer updates. + card*/pcm*/sub*/info The general information of this PCM sub-stream. diff --git a/sound/core/pcm_lib.c b/sound/core/pcm_lib.c index 3eea98a4e65a..d659995ac3ac 100644 --- a/sound/core/pcm_lib.c +++ b/sound/core/pcm_lib.c @@ -249,6 +249,11 @@ static int snd_pcm_update_hw_ptr_interrupt(struct snd_pcm_substream *substream) new_hw_ptr = hw_base + pos; } } + + /* Do jiffies check only in xrun_debug mode */ + if (!xrun_debug(substream)) + goto no_jiffies_check; + /* Skip the jiffies check for hardwares with BATCH flag. * Such hardware usually just increases the position at each IRQ, * thus it can't give any strange position. @@ -336,7 +341,9 @@ int snd_pcm_update_hw_ptr(struct snd_pcm_substream *substream) hw_base = 0; new_hw_ptr = hw_base + pos; } - if (((delta * HZ) / runtime->rate) > jdelta + HZ/100) { + /* Do jiffies check only in xrun_debug mode */ + if (xrun_debug(substream) && + ((delta * HZ) / runtime->rate) > jdelta + HZ/100) { hw_ptr_error(substream, "hw_ptr skipping! " "(pos=%ld, delta=%ld, period=%ld, jdelta=%lu/%lu)\n", From 55de5ef970c680d8d75f2a9aa7e4f172140dbd9c Mon Sep 17 00:00:00 2001 From: Clemens Ladisch Date: Wed, 27 May 2009 10:49:30 +0200 Subject: [PATCH 229/267] sound: usb-audio: make the MotU Fastlane work again Kernel 2.6.18 broke the MotU Fastlane, which uses duplicate endpoint numbers in a manner that is not only illegal but also confuses the kernel's endpoint descriptor caching mechanism. To work around this, we have to add a separate usb_set_interface() call to guide the USB core to the correct descriptors. Signed-off-by: Clemens Ladisch Reported-and-tested-by: David Fries Cc: Signed-off-by: Takashi Iwai --- sound/usb/usbaudio.c | 2 +- sound/usb/usbaudio.h | 2 +- sound/usb/usbmidi.c | 12 +++++++++++- sound/usb/usbquirks.h | 2 +- 4 files changed, 14 insertions(+), 4 deletions(-) diff --git a/sound/usb/usbaudio.c b/sound/usb/usbaudio.c index 823296d7d578..a6b88482637b 100644 --- a/sound/usb/usbaudio.c +++ b/sound/usb/usbaudio.c @@ -3347,7 +3347,7 @@ static int snd_usb_create_quirk(struct snd_usb_audio *chip, [QUIRK_MIDI_YAMAHA] = snd_usb_create_midi_interface, [QUIRK_MIDI_MIDIMAN] = snd_usb_create_midi_interface, [QUIRK_MIDI_NOVATION] = snd_usb_create_midi_interface, - [QUIRK_MIDI_RAW] = snd_usb_create_midi_interface, + [QUIRK_MIDI_FASTLANE] = snd_usb_create_midi_interface, [QUIRK_MIDI_EMAGIC] = snd_usb_create_midi_interface, [QUIRK_MIDI_CME] = snd_usb_create_midi_interface, [QUIRK_AUDIO_STANDARD_INTERFACE] = create_standard_audio_quirk, diff --git a/sound/usb/usbaudio.h b/sound/usb/usbaudio.h index 36e4f7a29adc..8e7f78941ba6 100644 --- a/sound/usb/usbaudio.h +++ b/sound/usb/usbaudio.h @@ -153,7 +153,7 @@ enum quirk_type { QUIRK_MIDI_YAMAHA, QUIRK_MIDI_MIDIMAN, QUIRK_MIDI_NOVATION, - QUIRK_MIDI_RAW, + QUIRK_MIDI_FASTLANE, QUIRK_MIDI_EMAGIC, QUIRK_MIDI_CME, QUIRK_MIDI_US122L, diff --git a/sound/usb/usbmidi.c b/sound/usb/usbmidi.c index 26bad373fe65..2fb35cc22a30 100644 --- a/sound/usb/usbmidi.c +++ b/sound/usb/usbmidi.c @@ -1778,8 +1778,18 @@ int snd_usb_create_midi_interface(struct snd_usb_audio* chip, umidi->usb_protocol_ops = &snd_usbmidi_novation_ops; err = snd_usbmidi_detect_per_port_endpoints(umidi, endpoints); break; - case QUIRK_MIDI_RAW: + case QUIRK_MIDI_FASTLANE: umidi->usb_protocol_ops = &snd_usbmidi_raw_ops; + /* + * Interface 1 contains isochronous endpoints, but with the same + * numbers as in interface 0. Since it is interface 1 that the + * USB core has most recently seen, these descriptors are now + * associated with the endpoint numbers. This will foul up our + * attempts to submit bulk/interrupt URBs to the endpoints in + * interface 0, so we have to make sure that the USB core looks + * again at interface 0 by calling usb_set_interface() on it. + */ + usb_set_interface(umidi->chip->dev, 0, 0); err = snd_usbmidi_detect_per_port_endpoints(umidi, endpoints); break; case QUIRK_MIDI_EMAGIC: diff --git a/sound/usb/usbquirks.h b/sound/usb/usbquirks.h index 647ef5029651..5d955aaad85f 100644 --- a/sound/usb/usbquirks.h +++ b/sound/usb/usbquirks.h @@ -1868,7 +1868,7 @@ YAMAHA_DEVICE(0x7010, "UB99"), .data = & (const struct snd_usb_audio_quirk[]) { { .ifnum = 0, - .type = QUIRK_MIDI_RAW + .type = QUIRK_MIDI_FASTLANE }, { .ifnum = 1, From ed37d83e6aa218192fb28bb6b82498d2a8c74070 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Wed, 27 May 2009 21:39:05 +1000 Subject: [PATCH 230/267] md: raid5: change incorrect usage of 'min' macro to 'min_t' A recent patch to raid5.c use min on an int and a sector_t. This isn't allowed. So change it to min_t(sector_t,x,y). Signed-off-by: NeilBrown --- drivers/md/raid5.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 3c3626d2a1f9..5d400aef8d9b 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -3811,13 +3811,13 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped safepos = conf->reshape_safe; sector_div(safepos, data_disks); if (mddev->delta_disks < 0) { - writepos -= min(reshape_sectors, writepos); + writepos -= min_t(sector_t, reshape_sectors, writepos); readpos += reshape_sectors; safepos += reshape_sectors; } else { writepos += reshape_sectors; - readpos -= min(reshape_sectors, readpos); - safepos -= min(reshape_sectors, safepos); + readpos -= min_t(sector_t, reshape_sectors, readpos); + safepos -= min_t(sector_t, reshape_sectors, safepos); } /* 'writepos' is the most advanced device address we might write. From 346a850e3c3a20159cef2b79235e6d34aa497c65 Mon Sep 17 00:00:00 2001 From: Manuel Traut Date: Wed, 27 May 2009 06:20:05 -0700 Subject: [PATCH 231/267] Input: usb1400_ts - fix access to "device data" in resume function platform_data != driver_data driver data is actually the "correct" place of the struct however it is not placed there due to the need of the ac97 struct. This is broken since d9105c2b01 aka "[ARM] 5184/1: Split ucb1400_ts into core and touchscreen" Signed-off-by: Manuel Traut Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/ucb1400_ts.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/touchscreen/ucb1400_ts.c b/drivers/input/touchscreen/ucb1400_ts.c index f100c7f4c1db..6954f5500108 100644 --- a/drivers/input/touchscreen/ucb1400_ts.c +++ b/drivers/input/touchscreen/ucb1400_ts.c @@ -419,7 +419,7 @@ static int ucb1400_ts_remove(struct platform_device *dev) #ifdef CONFIG_PM static int ucb1400_ts_resume(struct platform_device *dev) { - struct ucb1400_ts *ucb = platform_get_drvdata(dev); + struct ucb1400_ts *ucb = dev->dev.platform_data; if (ucb->ts_task) { /* From 348ca1029e8bae6e0c49097ad25439b17c5326f4 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 27 May 2009 15:46:50 +0100 Subject: [PATCH 232/267] FS-Cache: Fixup renamed filenames in comments in internal.h Fix up renamed filenames in comments in fs/fscache/internal.h. Originally, the files were all called fsc-xxx.c, but they got renamed to just xxx.c. Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- fs/fscache/internal.h | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/fs/fscache/internal.h b/fs/fscache/internal.h index e0cbd16f6dc9..1c341304621f 100644 --- a/fs/fscache/internal.h +++ b/fs/fscache/internal.h @@ -28,7 +28,7 @@ #define FSCACHE_MAX_THREADS 32 /* - * fsc-cache.c + * cache.c */ extern struct list_head fscache_cache_list; extern struct rw_semaphore fscache_addremove_sem; @@ -37,7 +37,7 @@ extern struct fscache_cache *fscache_select_cache_for_object( struct fscache_cookie *); /* - * fsc-cookie.c + * cookie.c */ extern struct kmem_cache *fscache_cookie_jar; @@ -45,13 +45,13 @@ extern void fscache_cookie_init_once(void *); extern void __fscache_cookie_put(struct fscache_cookie *); /* - * fsc-fsdef.c + * fsdef.c */ extern struct fscache_cookie fscache_fsdef_index; extern struct fscache_cookie_def fscache_fsdef_netfs_def; /* - * fsc-histogram.c + * histogram.c */ #ifdef CONFIG_FSCACHE_HISTOGRAM extern atomic_t fscache_obj_instantiate_histogram[HZ]; @@ -75,7 +75,7 @@ extern const struct file_operations fscache_histogram_fops; #endif /* - * fsc-main.c + * main.c */ extern unsigned fscache_defer_lookup; extern unsigned fscache_defer_create; @@ -86,14 +86,14 @@ extern int fscache_wait_bit(void *); extern int fscache_wait_bit_interruptible(void *); /* - * fsc-object.c + * object.c */ extern void fscache_withdrawing_object(struct fscache_cache *, struct fscache_object *); extern void fscache_enqueue_object(struct fscache_object *); /* - * fsc-operation.c + * operation.c */ extern int fscache_submit_exclusive_op(struct fscache_object *, struct fscache_operation *); @@ -104,7 +104,7 @@ extern void fscache_start_operations(struct fscache_object *); extern void fscache_operation_gc(struct work_struct *); /* - * fsc-proc.c + * proc.c */ #ifdef CONFIG_PROC_FS extern int __init fscache_proc_init(void); @@ -115,7 +115,7 @@ extern void fscache_proc_cleanup(void); #endif /* - * fsc-stats.c + * stats.c */ #ifdef CONFIG_FSCACHE_STATS extern atomic_t fscache_n_ops_processed[FSCACHE_MAX_THREADS]; From 911e690e70540f009125bacd16c017eb1a7b1916 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 27 May 2009 15:46:55 +0100 Subject: [PATCH 233/267] CacheFiles: Fixup renamed filenames in comments in internal.h Fix up renamed filenames in comments in fs/cachefiles/internal.h. Originally, the files were all called cf-xxx.c, but they got renamed to just xxx.c. Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- fs/cachefiles/internal.h | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/fs/cachefiles/internal.h b/fs/cachefiles/internal.h index 19218e1463d6..f7c255f9c624 100644 --- a/fs/cachefiles/internal.h +++ b/fs/cachefiles/internal.h @@ -122,13 +122,13 @@ static inline void cachefiles_state_changed(struct cachefiles_cache *cache) } /* - * cf-bind.c + * bind.c */ extern int cachefiles_daemon_bind(struct cachefiles_cache *cache, char *args); extern void cachefiles_daemon_unbind(struct cachefiles_cache *cache); /* - * cf-daemon.c + * daemon.c */ extern const struct file_operations cachefiles_daemon_fops; @@ -136,17 +136,17 @@ extern int cachefiles_has_space(struct cachefiles_cache *cache, unsigned fnr, unsigned bnr); /* - * cf-interface.c + * interface.c */ extern const struct fscache_cache_ops cachefiles_cache_ops; /* - * cf-key.c + * key.c */ extern char *cachefiles_cook_key(const u8 *raw, int keylen, uint8_t type); /* - * cf-namei.c + * namei.c */ extern int cachefiles_delete_object(struct cachefiles_cache *cache, struct cachefiles_object *object); @@ -165,7 +165,7 @@ extern int cachefiles_check_in_use(struct cachefiles_cache *cache, struct dentry *dir, char *filename); /* - * cf-proc.c + * proc.c */ #ifdef CONFIG_CACHEFILES_HISTOGRAM extern atomic_t cachefiles_lookup_histogram[HZ]; @@ -190,7 +190,7 @@ void cachefiles_hist(atomic_t histogram[], unsigned long start_jif) #endif /* - * cf-rdwr.c + * rdwr.c */ extern int cachefiles_read_or_alloc_page(struct fscache_retrieval *, struct page *, gfp_t); @@ -205,7 +205,7 @@ extern int cachefiles_write_page(struct fscache_storage *, struct page *); extern void cachefiles_uncache_page(struct fscache_object *, struct page *); /* - * cf-security.c + * security.c */ extern int cachefiles_get_security_ID(struct cachefiles_cache *cache); extern int cachefiles_determine_cache_security(struct cachefiles_cache *cache, @@ -225,7 +225,7 @@ static inline void cachefiles_end_secure(struct cachefiles_cache *cache, } /* - * cf-xattr.c + * xattr.c */ extern int cachefiles_check_object_type(struct cachefiles_object *object); extern int cachefiles_set_object_xattr(struct cachefiles_object *object, From 07f4f3e8a24138ca2f3650723d670df25687cd05 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Kristian=20H=C3=B8gsberg?= Date: Wed, 27 May 2009 14:37:28 -0400 Subject: [PATCH 234/267] i915: Set object to gtt domain when faulting it back in MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When a GEM object is evicted from the GTT we set it to the CPU domain, as it might get swapped in and out or ever mmapped regularly. If the object is mmapped through the GTT it can still get evicted in this way by other objects requiring GTT space. When the GTT mapping is touched again we fault it back into the GTT, but fail to set it back to the GTT domain. This means we fail to flush any cached CPU writes to the pages backing the object which will then happen "eventually", typically after we write to the page through the uncached GTT mapping. [anholt: Note that userland does do a set_domain(GTT, GTT) when starting to access the GTT mapping. That covers getting the existing mapping of the object synchronized if it's bound to the GTT. But set_domain(GTT, GTT) doesn't do anything if the object is currently unbound. This fix covers the transition to being bound for GTT mapping.] Fixes glyph and other pixmap corruption during swapping. fd.o bug #21790 Signed-off-by: Kristian Høgsberg Signed-off-by: Eric Anholt --- drivers/gpu/drm/i915/i915_gem.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index e2421869a40c..670d12881468 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1145,6 +1145,13 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) mutex_unlock(&dev->struct_mutex); return VM_FAULT_SIGBUS; } + + ret = i915_gem_object_set_to_gtt_domain(obj, write); + if (ret) { + mutex_unlock(&dev->struct_mutex); + return VM_FAULT_SIGBUS; + } + list_add_tail(&obj_priv->list, &dev_priv->mm.inactive_list); } From a0d24b295aed7a9daf4ca36bd4784e4d40f82303 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 19 May 2009 12:03:15 +0800 Subject: [PATCH 235/267] nfsd: fix hung up of nfs client while sync write data to nfs server Commit 'Short write in nfsd becomes a full write to the client' (31dec2538e45e9fff2007ea1f4c6bae9f78db724) broken the sync write. With the following commands to reproduce: $ mount -t nfs -o sync 192.168.0.21:/nfsroot /mnt $ cd /mnt $ echo aaaa > temp.txt Then nfs client is hung up. In SYNC mode the server alaways return the write count 0 to the client. This is because the value of host_err in nfsd_vfs_write() will be overwrite in SYNC mode by 'host_err=nfsd_sync(file);', and then we return host_err(which is now 0) as write count. This patch fixed the problem. Signed-off-by: Wei Yongjun Signed-off-by: J. Bruce Fields --- fs/nfsd/vfs.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 6c68ffd6b4bb..b660435978d2 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1015,6 +1015,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, host_err = vfs_writev(file, (struct iovec __user *)vec, vlen, &offset); set_fs(oldfs); if (host_err >= 0) { + *cnt = host_err; nfsdstats.io_write += host_err; fsnotify_modify(file->f_path.dentry); } @@ -1060,10 +1061,9 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, } dprintk("nfsd: write complete host_err=%d\n", host_err); - if (host_err >= 0) { + if (host_err >= 0) err = 0; - *cnt = host_err; - } else + else err = nfserrno(host_err); out: return err; From 7f4218354fe312b327af06c3d8c95ed5f214c8ca Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 27 May 2009 18:51:06 -0400 Subject: [PATCH 236/267] nfsd: Revert "svcrpc: take advantage of tcp autotuning" This reverts commit 47a14ef1af48c696b214ac168f056ddc79793d0e "svcrpc: take advantage of tcp autotuning", which uncovered some further problems in the server rpc code, causing significant performance regressions in common cases. We will likely reinstate this patch after releasing 2.6.30 and applying some work on the underlying fixes to the problem (developed by Trond). Reported-by: Jeff Moyer Cc: Olga Kornievskaia Cc: Jim Rees Cc: Trond Myklebust Signed-off-by: J. Bruce Fields --- net/sunrpc/svcsock.c | 35 ++++++++++++++++++++++++++++------- 1 file changed, 28 insertions(+), 7 deletions(-) diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index af3198814c15..9d504234af4a 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -345,6 +345,7 @@ static void svc_sock_setbufsize(struct socket *sock, unsigned int snd, lock_sock(sock->sk); sock->sk->sk_sndbuf = snd * 2; sock->sk->sk_rcvbuf = rcv * 2; + sock->sk->sk_userlocks |= SOCK_SNDBUF_LOCK|SOCK_RCVBUF_LOCK; release_sock(sock->sk); #endif } @@ -796,6 +797,23 @@ static int svc_tcp_recvfrom(struct svc_rqst *rqstp) test_bit(XPT_CONN, &svsk->sk_xprt.xpt_flags), test_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags)); + if (test_and_clear_bit(XPT_CHNGBUF, &svsk->sk_xprt.xpt_flags)) + /* sndbuf needs to have room for one request + * per thread, otherwise we can stall even when the + * network isn't a bottleneck. + * + * We count all threads rather than threads in a + * particular pool, which provides an upper bound + * on the number of threads which will access the socket. + * + * rcvbuf just needs to be able to hold a few requests. + * Normally they will be removed from the queue + * as soon a a complete request arrives. + */ + svc_sock_setbufsize(svsk->sk_sock, + (serv->sv_nrthreads+3) * serv->sv_max_mesg, + 3 * serv->sv_max_mesg); + clear_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags); /* Receive data. If we haven't got the record length yet, get @@ -1043,6 +1061,15 @@ static void svc_tcp_init(struct svc_sock *svsk, struct svc_serv *serv) tcp_sk(sk)->nonagle |= TCP_NAGLE_OFF; + /* initialise setting must have enough space to + * receive and respond to one request. + * svc_tcp_recvfrom will re-adjust if necessary + */ + svc_sock_setbufsize(svsk->sk_sock, + 3 * svsk->sk_xprt.xpt_server->sv_max_mesg, + 3 * svsk->sk_xprt.xpt_server->sv_max_mesg); + + set_bit(XPT_CHNGBUF, &svsk->sk_xprt.xpt_flags); set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags); if (sk->sk_state != TCP_ESTABLISHED) set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags); @@ -1112,14 +1139,8 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, /* Initialize the socket */ if (sock->type == SOCK_DGRAM) svc_udp_init(svsk, serv); - else { - /* initialise setting must have enough space to - * receive and respond to one request. - */ - svc_sock_setbufsize(svsk->sk_sock, 4 * serv->sv_max_mesg, - 4 * serv->sv_max_mesg); + else svc_tcp_init(svsk, serv); - } dprintk("svc: svc_setup_socket created %p (inet %p)\n", svsk, svsk->sk_sk); From 98779be861a05c4cb75bed916df72ec0cba8b53d Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Thu, 14 May 2009 16:34:28 -0500 Subject: [PATCH 237/267] svcrdma: dma unmap the correct length for the RPCRDMA header page. The svcrdma module was incorrectly unmapping the RPCRDMA header page. On IBM pserver systems this causes a resource leak that results in running out of bus address space (10 cthon iterations will reproduce it). The code was mapping the full page but only unmapping the actual header length. The fix is to only map the header length. I also cleaned up the use of ib_dma_map_page() calls since the unmap logic always uses ib_dma_unmap_single(). I made these symmetrical. Signed-off-by: Steve Wise Signed-off-by: Tom Tucker Signed-off-by: J. Bruce Fields --- net/sunrpc/xprtrdma/svc_rdma_sendto.c | 12 ++++++------ net/sunrpc/xprtrdma/svc_rdma_transport.c | 10 +++++----- 2 files changed, 11 insertions(+), 11 deletions(-) diff --git a/net/sunrpc/xprtrdma/svc_rdma_sendto.c b/net/sunrpc/xprtrdma/svc_rdma_sendto.c index 8b510c5e8777..f11be72a1a80 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_sendto.c +++ b/net/sunrpc/xprtrdma/svc_rdma_sendto.c @@ -128,7 +128,8 @@ static int fast_reg_xdr(struct svcxprt_rdma *xprt, page_bytes -= sge_bytes; frmr->page_list->page_list[page_no] = - ib_dma_map_page(xprt->sc_cm_id->device, page, 0, + ib_dma_map_single(xprt->sc_cm_id->device, + page_address(page), PAGE_SIZE, DMA_TO_DEVICE); if (ib_dma_mapping_error(xprt->sc_cm_id->device, frmr->page_list->page_list[page_no])) @@ -532,18 +533,17 @@ static int send_reply(struct svcxprt_rdma *rdma, clear_bit(RDMACTXT_F_FAST_UNREG, &ctxt->flags); /* Prepare the SGE for the RPCRDMA Header */ + ctxt->sge[0].lkey = rdma->sc_dma_lkey; + ctxt->sge[0].length = svc_rdma_xdr_get_reply_hdr_len(rdma_resp); ctxt->sge[0].addr = - ib_dma_map_page(rdma->sc_cm_id->device, - page, 0, PAGE_SIZE, DMA_TO_DEVICE); + ib_dma_map_single(rdma->sc_cm_id->device, page_address(page), + ctxt->sge[0].length, DMA_TO_DEVICE); if (ib_dma_mapping_error(rdma->sc_cm_id->device, ctxt->sge[0].addr)) goto err; atomic_inc(&rdma->sc_dma_used); ctxt->direction = DMA_TO_DEVICE; - ctxt->sge[0].length = svc_rdma_xdr_get_reply_hdr_len(rdma_resp); - ctxt->sge[0].lkey = rdma->sc_dma_lkey; - /* Determine how many of our SGE are to be transmitted */ for (sge_no = 1; byte_count && sge_no < vec->count; sge_no++) { sge_bytes = min_t(size_t, vec->sge[sge_no].iov_len, byte_count); diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c index 4b0c2fa15e0b..5151f9f6c573 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_transport.c +++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c @@ -500,8 +500,8 @@ int svc_rdma_post_recv(struct svcxprt_rdma *xprt) BUG_ON(sge_no >= xprt->sc_max_sge); page = svc_rdma_get_page(); ctxt->pages[sge_no] = page; - pa = ib_dma_map_page(xprt->sc_cm_id->device, - page, 0, PAGE_SIZE, + pa = ib_dma_map_single(xprt->sc_cm_id->device, + page_address(page), PAGE_SIZE, DMA_FROM_DEVICE); if (ib_dma_mapping_error(xprt->sc_cm_id->device, pa)) goto err_put_ctxt; @@ -1315,8 +1315,8 @@ void svc_rdma_send_error(struct svcxprt_rdma *xprt, struct rpcrdma_msg *rmsgp, length = svc_rdma_xdr_encode_error(xprt, rmsgp, err, va); /* Prepare SGE for local address */ - sge.addr = ib_dma_map_page(xprt->sc_cm_id->device, - p, 0, PAGE_SIZE, DMA_FROM_DEVICE); + sge.addr = ib_dma_map_single(xprt->sc_cm_id->device, + page_address(p), PAGE_SIZE, DMA_FROM_DEVICE); if (ib_dma_mapping_error(xprt->sc_cm_id->device, sge.addr)) { put_page(p); return; @@ -1343,7 +1343,7 @@ void svc_rdma_send_error(struct svcxprt_rdma *xprt, struct rpcrdma_msg *rmsgp, if (ret) { dprintk("svcrdma: Error %d posting send for protocol error\n", ret); - ib_dma_unmap_page(xprt->sc_cm_id->device, + ib_dma_unmap_single(xprt->sc_cm_id->device, sge.addr, PAGE_SIZE, DMA_FROM_DEVICE); svc_rdma_put_context(ctxt, 1); From 1812e67c7410c8d0d57f14a3dc81a99bd5b30e3e Mon Sep 17 00:00:00 2001 From: Tony Vroon Date: Wed, 27 May 2009 21:00:41 +0100 Subject: [PATCH 238/267] ALSA: hda - Compaq Presario CQ60 patching for Conexant A docking mic control is shown by default. The Compaq Presario CQ60 laptop has no docking connector, so designate it as a CXT5051_HP model. This makes the phantom mixer slider disappear. Signed-off-by: Tony Vroon Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_conexant.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/pci/hda/patch_conexant.c b/sound/pci/hda/patch_conexant.c index 56ce19e68cb5..4fcbe21829ab 100644 --- a/sound/pci/hda/patch_conexant.c +++ b/sound/pci/hda/patch_conexant.c @@ -1848,6 +1848,7 @@ static const char *cxt5051_models[CXT5051_MODELS] = { static struct snd_pci_quirk cxt5051_cfg_tbl[] = { SND_PCI_QUIRK(0x103c, 0x30cf, "HP DV6736", CXT5051_HP_DV6736), + SND_PCI_QUIRK(0x103c, 0x360b, "Compaq Presario CQ60", CXT5051_HP), SND_PCI_QUIRK(0x14f1, 0x0101, "Conexant Reference board", CXT5051_LAPTOP), SND_PCI_QUIRK(0x14f1, 0x5051, "HP Spartan 1.1", CXT5051_HP), From a3ce6ea46cc0d6397d1b92b1a5983bb2935306ed Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Thu, 28 May 2009 09:51:31 -0700 Subject: [PATCH 239/267] Input: libps2 - better handle bad scheduler decisions Sometimes devices send us their responses in time but due to unfortunate scheduling decisions the receiving thread does not get scheduled till much later and we erroneously decide that device timed out. Work around this problem by checking whether we received the data we needed instead of checking timeout condition. Tested-by: Sitsofe Wheeler Signed-off-by: Dmitry Torokhov --- drivers/input/serio/libps2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/serio/libps2.c b/drivers/input/serio/libps2.c index 67248c31e19a..be5bbbb8ae4e 100644 --- a/drivers/input/serio/libps2.c +++ b/drivers/input/serio/libps2.c @@ -210,7 +210,7 @@ int ps2_command(struct ps2dev *ps2dev, unsigned char *param, int command) timeout = wait_event_timeout(ps2dev->wait, !(ps2dev->flags & PS2_FLAG_CMD1), timeout); - if (ps2dev->cmdcnt && timeout > 0) { + if (ps2dev->cmdcnt && !(ps2dev->flags & PS2_FLAG_CMD1)) { timeout = ps2_adjust_timeout(ps2dev, command, timeout); wait_event_timeout(ps2dev->wait, From f42706c90470851fd2e97eda7a4109e8949bde8a Mon Sep 17 00:00:00 2001 From: Martin Fuzzey Date: Fri, 1 May 2009 17:21:11 +0200 Subject: [PATCH 240/267] USB: atmel-usba-udc : fix control out requests. usbtest #14 was failing with "udc: ep0: TXCOMP: Invalid endpoint state 2, halting endpoint..." This occured since ep0 is bidirectional and ep->is_in is not valid (must always use ep->state) Signed-off-by: Martin Fuzzey Acked-by: David Brownell Acked-by: Haavard Skinnemoen Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/atmel_usba_udc.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/usb/gadget/atmel_usba_udc.c b/drivers/usb/gadget/atmel_usba_udc.c index 563d57275448..5644897c1c5d 100644 --- a/drivers/usb/gadget/atmel_usba_udc.c +++ b/drivers/usb/gadget/atmel_usba_udc.c @@ -794,7 +794,8 @@ usba_ep_queue(struct usb_ep *_ep, struct usb_request *_req, gfp_t gfp_flags) if (ep->desc) { list_add_tail(&req->queue, &ep->queue); - if (ep->is_in || (ep_is_control(ep) + if ((!ep_is_control(ep) && ep->is_in) || + (ep_is_control(ep) && (ep->state == DATA_STAGE_IN || ep->state == STATUS_STAGE_IN))) usba_ep_writel(ep, CTL_ENB, USBA_TX_PK_RDY); From fe92c9e481a147a9e1e14f55870f32903b967777 Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Mon, 4 May 2009 17:22:43 -0700 Subject: [PATCH 241/267] USB: atmel_usb_udc: Use kzalloc() to allocate ep structures This ensures that all fields are properly initialized. Signed-off-by: Haavard Skinnemoen Acked-by: David Brownell Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/atmel_usba_udc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/gadget/atmel_usba_udc.c b/drivers/usb/gadget/atmel_usba_udc.c index 5644897c1c5d..05c913cc3658 100644 --- a/drivers/usb/gadget/atmel_usba_udc.c +++ b/drivers/usb/gadget/atmel_usba_udc.c @@ -1941,7 +1941,7 @@ static int __init usba_udc_probe(struct platform_device *pdev) usba_writel(udc, CTRL, USBA_DISABLE_MASK); clk_disable(pclk); - usba_ep = kmalloc(sizeof(struct usba_ep) * pdata->num_ep, + usba_ep = kzalloc(sizeof(struct usba_ep) * pdata->num_ep, GFP_KERNEL); if (!usba_ep) goto err_alloc_ep; From cab98a0a349829b145d924c0649a2d30cd6a9e3d Mon Sep 17 00:00:00 2001 From: Xiao Kaijian Date: Fri, 8 May 2009 00:48:23 +0800 Subject: [PATCH 242/267] USB: Yet another Conexant Clone to add to cdc-acm.c This patch adds another quirky Conexant USB Modem Clone to usb cdc-acm.c Signed-off-by: Xiao Kaijian Signed-off-by: Greg Kroah-Hartman --- drivers/usb/class/cdc-acm.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index 0a69c0977e3f..7a1164dd1d37 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -1375,6 +1375,9 @@ static struct usb_device_id acm_ids[] = { { USB_DEVICE(0x0572, 0x1324), /* Conexant USB MODEM RD02-D400 */ .driver_info = NO_UNION_NORMAL, /* has no union descriptor */ }, + { USB_DEVICE(0x0572, 0x1328), /* Shiro / Aztech USB MODEM UM-3100 */ + .driver_info = NO_UNION_NORMAL, /* has no union descriptor */ + }, { USB_DEVICE(0x22b8, 0x6425), /* Motorola MOTOMAGX phones */ }, { USB_DEVICE(0x0572, 0x1329), /* Hummingbird huc56s (Conexant) */ From 0afb20e00b5053170c85298fed842b32d20b4ea9 Mon Sep 17 00:00:00 2001 From: Warren Free Date: Fri, 8 May 2009 10:27:08 +0200 Subject: [PATCH 243/267] USB: isp1760: urb_dequeue doesn't always find the urbs The option driver (and presumably others) allocates several URBs when it opens and tries to free them when it closes. The isp1760_urb_dequeue function gets called, but the packet being dequeued is not necessarily at the front of one of the 32 queues. If not, the isp1760_urb_done function doesn't get called for the URB and the process trying to free it hangs forever on a wait_queue. This patch does two things. If the URB being dequeued has others queued behind it, it re-queues them. And it searches the queues looking for the URB being dequeued rather than just looking at the one at the front of the queue. [bigeasy@linutronix] whitespace fixes, reformating Cc: stable Signed-off-by: Warren Free Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/isp1760-hcd.c | 24 ++++++++++++++++++++++-- 1 file changed, 22 insertions(+), 2 deletions(-) diff --git a/drivers/usb/host/isp1760-hcd.c b/drivers/usb/host/isp1760-hcd.c index cd07ea3f0c63..15438469f21a 100644 --- a/drivers/usb/host/isp1760-hcd.c +++ b/drivers/usb/host/isp1760-hcd.c @@ -1658,6 +1658,7 @@ static int isp1760_urb_dequeue(struct usb_hcd *hcd, struct urb *urb, u32 reg_base, or_reg, skip_reg; unsigned long flags; struct ptd ptd; + packet_enqueue *pe; switch (usb_pipetype(urb->pipe)) { case PIPE_ISOCHRONOUS: @@ -1669,6 +1670,7 @@ static int isp1760_urb_dequeue(struct usb_hcd *hcd, struct urb *urb, reg_base = INT_REGS_OFFSET; or_reg = HC_INT_IRQ_MASK_OR_REG; skip_reg = HC_INT_PTD_SKIPMAP_REG; + pe = enqueue_an_INT_packet; break; default: @@ -1676,6 +1678,7 @@ static int isp1760_urb_dequeue(struct usb_hcd *hcd, struct urb *urb, reg_base = ATL_REGS_OFFSET; or_reg = HC_ATL_IRQ_MASK_OR_REG; skip_reg = HC_ATL_PTD_SKIPMAP_REG; + pe = enqueue_an_ATL_packet; break; } @@ -1687,6 +1690,7 @@ static int isp1760_urb_dequeue(struct usb_hcd *hcd, struct urb *urb, u32 skip_map; u32 or_map; struct isp1760_qtd *qtd; + struct isp1760_qh *qh = ints->qh; skip_map = isp1760_readl(hcd->regs + skip_reg); skip_map |= 1 << i; @@ -1699,8 +1703,7 @@ static int isp1760_urb_dequeue(struct usb_hcd *hcd, struct urb *urb, priv_write_copy(priv, (u32 *)&ptd, hcd->regs + reg_base + i * sizeof(ptd), sizeof(ptd)); qtd = ints->qtd; - - clean_up_qtdlist(qtd); + qtd = clean_up_qtdlist(qtd); free_mem(priv, ints->payload); @@ -1711,7 +1714,24 @@ static int isp1760_urb_dequeue(struct usb_hcd *hcd, struct urb *urb, ints->payload = 0; isp1760_urb_done(priv, urb, status); + if (qtd) + pe(hcd, qh, qtd); break; + + } else if (ints->qtd) { + struct isp1760_qtd *qtd, *prev_qtd = ints->qtd; + + for (qtd = ints->qtd->hw_next; qtd; qtd = qtd->hw_next) { + if (qtd->urb == urb) { + prev_qtd->hw_next = clean_up_qtdlist(qtd); + isp1760_urb_done(priv, urb, status); + break; + } + prev_qtd = qtd; + } + /* we found the urb before the end of the list */ + if (qtd) + break; } ints++; } From 0a3c8549ea7e94d74a41096d42bc6cdf43d183bf Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Wed, 27 May 2009 11:25:52 -0400 Subject: [PATCH 244/267] usb-serial: fix crash when sub-driver updates firmware This patch (as1244) fixes a crash in usb-serial that occurs when a sub-driver returns a positive value from its attach method, indicating that new firmware was loaded and the device will disconnect and reconnect. The usb-serial core then skips the step of registering the port devices; when the disconnect occurs, the attempt to unregister the ports fails dramatically. This problem shows up with Keyspan devices and it might affect others as well. When the attach method returns a positive value, the patch sets num_ports to 0. This tells usb_serial_disconnect() not to try unregistering any of the ports; instead they are cleaned up by destroy_serial(). Signed-off-by: Alan Stern Tested-by: Benjamin Herrenschmidt Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/serial/usb-serial.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c index 0a566eea49c0..f331e2bde88a 100644 --- a/drivers/usb/serial/usb-serial.c +++ b/drivers/usb/serial/usb-serial.c @@ -974,6 +974,7 @@ int usb_serial_probe(struct usb_interface *interface, if (retval > 0) { /* quietly accept this device, but don't bind to a serial port as it's about to disappear */ + serial->num_ports = 0; goto exit; } } From 29868b281f6d057b4cbe348f4483f1717c021c5c Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Wed, 20 May 2009 08:49:48 -0400 Subject: [PATCH 245/267] Revert "USB: Correct Makefile to make isp1760 buildable" This reverts commit 26e1287594864169577327fef233befc9739be3b. A larger patch (f7e7aa585) a few days after this one added the same line to the Makefile, but in a different place. While it'd be more correct to revert that one, it's easier to revert this one because this is a one-liner. Signed-off-by: Mike Frysinger CC: Greg Kroah-Hartman CC: linux-usb@vger.kernel.org Signed-off-by: Greg Kroah-Hartman --- drivers/usb/Makefile | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/usb/Makefile b/drivers/usb/Makefile index 0716cdb44cd8..0a3dc5ece634 100644 --- a/drivers/usb/Makefile +++ b/drivers/usb/Makefile @@ -11,7 +11,6 @@ obj-$(CONFIG_USB_MON) += mon/ obj-$(CONFIG_PCI) += host/ obj-$(CONFIG_USB_EHCI_HCD) += host/ obj-$(CONFIG_USB_ISP116X_HCD) += host/ -obj-$(CONFIG_USB_ISP1760_HCD) += host/ obj-$(CONFIG_USB_OHCI_HCD) += host/ obj-$(CONFIG_USB_UHCI_HCD) += host/ obj-$(CONFIG_USB_FHCI_HCD) += host/ From 086a377edc969aea6c761176a7e4ff68f264d6fe Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 7 May 2009 12:36:53 -0700 Subject: [PATCH 246/267] sysfs: file.c: use create_singlethread_workqueue() We don't need a kernel thread per CPU for this application. Acked-by: Alex Chiang Cc: Lai Jiangshan Cc: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- fs/sysfs/file.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index b1606e07b7a3..561a9c050cef 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -723,7 +723,7 @@ int sysfs_schedule_callback(struct kobject *kobj, void (*func)(void *), mutex_unlock(&sysfs_workq_mutex); if (sysfs_workqueue == NULL) { - sysfs_workqueue = create_workqueue("sysfsd"); + sysfs_workqueue = create_singlethread_workqueue("sysfsd"); if (sysfs_workqueue == NULL) { module_put(owner); return -ENOMEM; From 5c8563d773c0e9f0ac2a552e84806decd98ce732 Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Thu, 28 May 2009 14:24:07 -0700 Subject: [PATCH 247/267] Driver Core: do not oops when driver_unregister() is called for unregistered drivers MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We also fix a problem with cleaning up properly when initializing drivers and devices, so checks like this will work successfully. Portions of the patch by Linus and Greg and Ingo. Reported-by: Ozan Çağlayan Signed-off-by: Kay Sievers Cc: Linus Torvalds Cc: Ingo Molnar Signed-off-by: Greg Kroah-Hartman --- drivers/base/bus.c | 4 +++- drivers/base/core.c | 5 ++++- drivers/base/driver.c | 4 ++++ 3 files changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/base/bus.c b/drivers/base/bus.c index dc030f1f00f1..c6599618523e 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -700,8 +700,10 @@ int bus_add_driver(struct device_driver *drv) } kobject_uevent(&priv->kobj, KOBJ_ADD); - return error; + return 0; out_unregister: + kfree(drv->p); + drv->p = NULL; kobject_put(&priv->kobj); out_put_bus: bus_put(bus); diff --git a/drivers/base/core.c b/drivers/base/core.c index 4aa527b8a913..1977d4beb89e 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -879,7 +879,7 @@ int device_add(struct device *dev) } if (!dev_name(dev)) - goto done; + goto name_error; pr_debug("device: '%s': %s\n", dev_name(dev), __func__); @@ -978,6 +978,9 @@ done: cleanup_device_parent(dev); if (parent) put_device(parent); +name_error: + kfree(dev->p); + dev->p = NULL; goto done; } diff --git a/drivers/base/driver.c b/drivers/base/driver.c index c51f11bb29ae..8ae0f63602e0 100644 --- a/drivers/base/driver.c +++ b/drivers/base/driver.c @@ -257,6 +257,10 @@ EXPORT_SYMBOL_GPL(driver_register); */ void driver_unregister(struct device_driver *drv) { + if (!drv || !drv->p) { + WARN(1, "Unexpected driver unregister!\n"); + return; + } driver_remove_groups(drv, drv->groups); bus_remove_driver(drv); } From 8541c1180a355c4da283fc6b03a92c0233823c1b Mon Sep 17 00:00:00 2001 From: Vladimir Barinov Date: Thu, 23 Apr 2009 15:47:22 +0400 Subject: [PATCH 248/267] mtd: MXC NAND driver fixes (v5) The following patch fixes: - re-initialization of host->col_addr which is used as byte index between the successive READID flash commands. - compile error when CONFIG_PM is enabled - pass on the error code from clk_get() - return -ENOMEM in case of failed ioremap() - pass on the return value of platform_driver_probe() directly - remove excessive printk - let command line partition table parsing with mxc_nand name. The cmd_line parsing is done via name that differs from mxc_nand by default and looks like "NAND 256MiB 1,8V 8-bit" Signed-off-by: Vladimir Barinov Signed-off-by: Lothar Wassmann Acked-by: Sascha Hauer Signed-off-by: Artem Bityutskiy Signed-off-by: David Woodhouse --- drivers/mtd/nand/mxc_nand.c | 43 ++++++++++++++++++++----------------- 1 file changed, 23 insertions(+), 20 deletions(-) diff --git a/drivers/mtd/nand/mxc_nand.c b/drivers/mtd/nand/mxc_nand.c index f3548d048014..40c26080ecda 100644 --- a/drivers/mtd/nand/mxc_nand.c +++ b/drivers/mtd/nand/mxc_nand.c @@ -831,6 +831,7 @@ static void mxc_nand_command(struct mtd_info *mtd, unsigned command, break; case NAND_CMD_READID: + host->col_addr = 0; send_read_id(host); break; @@ -867,6 +868,7 @@ static int __init mxcnd_probe(struct platform_device *pdev) mtd->priv = this; mtd->owner = THIS_MODULE; mtd->dev.parent = &pdev->dev; + mtd->name = "mxc_nand"; /* 50 us command delay time */ this->chip_delay = 5; @@ -882,8 +884,10 @@ static int __init mxcnd_probe(struct platform_device *pdev) this->verify_buf = mxc_nand_verify_buf; host->clk = clk_get(&pdev->dev, "nfc"); - if (IS_ERR(host->clk)) + if (IS_ERR(host->clk)) { + err = PTR_ERR(host->clk); goto eclk; + } clk_enable(host->clk); host->clk_act = 1; @@ -896,7 +900,7 @@ static int __init mxcnd_probe(struct platform_device *pdev) host->regs = ioremap(res->start, res->end - res->start + 1); if (!host->regs) { - err = -EIO; + err = -ENOMEM; goto eres; } @@ -1011,30 +1015,35 @@ static int __devexit mxcnd_remove(struct platform_device *pdev) #ifdef CONFIG_PM static int mxcnd_suspend(struct platform_device *pdev, pm_message_t state) { - struct mtd_info *info = platform_get_drvdata(pdev); + struct mtd_info *mtd = platform_get_drvdata(pdev); + struct nand_chip *nand_chip = mtd->priv; + struct mxc_nand_host *host = nand_chip->priv; int ret = 0; DEBUG(MTD_DEBUG_LEVEL0, "MXC_ND : NAND suspend\n"); - if (info) - ret = info->suspend(info); - - /* Disable the NFC clock */ - clk_disable(nfc_clk); /* FIXME */ + if (mtd) { + ret = mtd->suspend(mtd); + /* Disable the NFC clock */ + clk_disable(host->clk); + } return ret; } static int mxcnd_resume(struct platform_device *pdev) { - struct mtd_info *info = platform_get_drvdata(pdev); + struct mtd_info *mtd = platform_get_drvdata(pdev); + struct nand_chip *nand_chip = mtd->priv; + struct mxc_nand_host *host = nand_chip->priv; int ret = 0; DEBUG(MTD_DEBUG_LEVEL0, "MXC_ND : NAND resume\n"); - /* Enable the NFC clock */ - clk_enable(nfc_clk); /* FIXME */ - if (info) - info->resume(info); + if (mtd) { + /* Enable the NFC clock */ + clk_enable(host->clk); + mtd->resume(mtd); + } return ret; } @@ -1055,13 +1064,7 @@ static struct platform_driver mxcnd_driver = { static int __init mxc_nd_init(void) { - /* Register the device driver structure. */ - pr_info("MXC MTD nand Driver\n"); - if (platform_driver_probe(&mxcnd_driver, mxcnd_probe) != 0) { - printk(KERN_ERR "Driver register failed for mxcnd_driver\n"); - return -ENODEV; - } - return 0; + return platform_driver_probe(&mxcnd_driver, mxcnd_probe); } static void __exit mxc_nd_cleanup(void) From 81e2962801bbb4e740c501ca687d5cb857929c04 Mon Sep 17 00:00:00 2001 From: Joakim Tjernlund Date: Thu, 28 May 2009 17:43:59 +0200 Subject: [PATCH 249/267] jffs2: Fix corruption when flash erase/write failure Erase errors such as: "Newly-erased block contained word 0xa4ef223e at offset 0x0296a014" and failure to write the clean marker, moves the offending erase block to erasing list before calling jffs2_erase_failed(). This is bad as jffs2_erase_failed() will also move the block to the bad_list, but is now moving the wrong block, causing FS corruption. Signed-off-by: Joakim Tjernlund Signed-off-by: David Woodhouse --- fs/jffs2/erase.c | 7 ------- 1 file changed, 7 deletions(-) diff --git a/fs/jffs2/erase.c b/fs/jffs2/erase.c index c32b4a1ad6cf..a0244740b75a 100644 --- a/fs/jffs2/erase.c +++ b/fs/jffs2/erase.c @@ -480,13 +480,6 @@ static void jffs2_mark_erased_block(struct jffs2_sb_info *c, struct jffs2_eraseb return; filebad: - mutex_lock(&c->erase_free_sem); - spin_lock(&c->erase_completion_lock); - /* Stick it on a list (any list) so erase_failed can take it - right off again. Silly, but shouldn't happen often. */ - list_move(&jeb->list, &c->erasing_list); - spin_unlock(&c->erase_completion_lock); - mutex_unlock(&c->erase_free_sem); jffs2_erase_failed(c, jeb, bad_offset); return; From 681a1b4032d72f4ad6d4beed751bc65574572746 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Thu, 28 May 2009 14:34:18 -0700 Subject: [PATCH 250/267] MAINTAINERS: pair EDAC-E752X P: and M: entries Entries should be P: name then M: email address. Signed-off-by: Joe Perches Acked-by: Doug Thompson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 5ee166e27b9f..ea4eced29a5c 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1971,8 +1971,8 @@ F: include/linux/edac.h EDAC-E752X P: Mark Gross -P: Doug Thompson M: mark.gross@intel.com +P: Doug Thompson M: dougthompson@xmission.com L: bluesmoke-devel@lists.sourceforge.net (moderated for non-subscribers) W: bluesmoke.sourceforge.net From 6d2661ede5f20f968422e790af3334908c3bc857 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Thu, 28 May 2009 14:34:19 -0700 Subject: [PATCH 251/267] oom: fix possible oom_dump_tasks NULL pointer When /proc/sys/vm/oom_dump_tasks is enabled, it is possible to get a NULL pointer for tasks that have detached mm's since task_lock() is not held during the tasklist scan. Add the task_lock(). Acked-by: Nick Piggin Acked-by: Mel Gorman Cc: Rik van Riel Signed-off-by: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/oom_kill.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) diff --git a/mm/oom_kill.c b/mm/oom_kill.c index 92bcf1db16b2..a7b2460e922b 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -284,22 +284,28 @@ static void dump_tasks(const struct mem_cgroup *mem) printk(KERN_INFO "[ pid ] uid tgid total_vm rss cpu oom_adj " "name\n"); do_each_thread(g, p) { - /* - * total_vm and rss sizes do not exist for tasks with a - * detached mm so there's no need to report them. - */ - if (!p->mm) - continue; + struct mm_struct *mm; + if (mem && !task_in_mem_cgroup(p, mem)) continue; if (!thread_group_leader(p)) continue; task_lock(p); + mm = p->mm; + if (!mm) { + /* + * total_vm and rss sizes do not exist for tasks with no + * mm so there's no need to report them; they can't be + * oom killed anyway. + */ + task_unlock(p); + continue; + } printk(KERN_INFO "[%5d] %5d %5d %8lu %8lu %3d %3d %s\n", - p->pid, __task_cred(p)->uid, p->tgid, - p->mm->total_vm, get_mm_rss(p->mm), (int)task_cpu(p), - p->oomkilladj, p->comm); + p->pid, __task_cred(p)->uid, p->tgid, mm->total_vm, + get_mm_rss(mm), (int)task_cpu(p), p->oomkilladj, + p->comm); task_unlock(p); } while_each_thread(g, p); } From b2e1feaf0af6b8a826b86748a19ddc2013ab7dbd Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 28 May 2009 14:34:20 -0700 Subject: [PATCH 252/267] cred: #include init.h in cred.h linux/cred.h can't be included as first header (alphabetical order) because it uses __init which is enough to break compilation on some archs. Signed-off-by: Alexey Dobriyan Acked-by: James Morris Acked-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cred.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/linux/cred.h b/include/linux/cred.h index 3282ee4318e7..4fa999696310 100644 --- a/include/linux/cred.h +++ b/include/linux/cred.h @@ -13,6 +13,7 @@ #define _LINUX_CRED_H #include +#include #include #include From bd6daba909d8484bd2ccf6017db4028d7a420927 Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Thu, 28 May 2009 14:34:21 -0700 Subject: [PATCH 253/267] procfs: make errno values consistent when open pident vs exit(2) race occurs proc_pident_instantiate() has following call flow. proc_pident_lookup() proc_pident_instantiate() proc_pid_make_inode() And, proc_pident_lookup() has following error handling. const struct pid_entry *p, *last; error = ERR_PTR(-ENOENT); if (!task) goto out_no_task; Then, proc_pident_instantiate should return ENOENT too when racing against exit(2) occur. EINAL has two bad reason. - it implies caller is wrong. bad the race isn't caller's mistake. - man 2 open don't explain EINVAL. user often don't handle it. Note: Other proc_pid_make_inode() caller already use ENOENT properly. Acked-by: Eric W. Biederman Cc: Alexey Dobriyan Signed-off-by: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/base.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/proc/base.c b/fs/proc/base.c index fb45615943c2..3326bbf9ab95 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1956,7 +1956,7 @@ static struct dentry *proc_pident_instantiate(struct inode *dir, const struct pid_entry *p = ptr; struct inode *inode; struct proc_inode *ei; - struct dentry *error = ERR_PTR(-EINVAL); + struct dentry *error = ERR_PTR(-ENOENT); inode = proc_pid_make_inode(dir->i_sb, task); if (!inode) From e767e0561d7fd2333df1921f1ab4176211f9036b Mon Sep 17 00:00:00 2001 From: Daisuke Nishimura Date: Thu, 28 May 2009 14:34:28 -0700 Subject: [PATCH 254/267] memcg: fix deadlock between lock_page_cgroup and mapping tree_lock mapping->tree_lock can be acquired from interrupt context. Then, following dead lock can occur. Assume "A" as a page. CPU0: lock_page_cgroup(A) interrupted -> take mapping->tree_lock. CPU1: take mapping->tree_lock -> lock_page_cgroup(A) This patch tries to fix above deadlock by moving memcg's hook to out of mapping->tree_lock. charge/uncharge of pagecache/swapcache is protected by page lock, not tree_lock. After this patch, lock_page_cgroup() is not called under mapping->tree_lock. Signed-off-by: KAMEZAWA Hiroyuki Signed-off-by: Daisuke Nishimura Cc: Balbir Singh Cc: Daisuke Nishimura Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/swap.h | 5 +++++ mm/filemap.c | 6 +++--- mm/memcontrol.c | 4 +++- mm/swap_state.c | 4 +--- mm/truncate.c | 1 + mm/vmscan.c | 2 ++ 6 files changed, 15 insertions(+), 7 deletions(-) diff --git a/include/linux/swap.h b/include/linux/swap.h index 62d81435347a..d476aad3ff57 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -437,6 +437,11 @@ static inline int mem_cgroup_cache_charge_swapin(struct page *page, return 0; } +static inline void +mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent) +{ +} + #endif /* CONFIG_SWAP */ #endif /* __KERNEL__*/ #endif /* _LINUX_SWAP_H */ diff --git a/mm/filemap.c b/mm/filemap.c index 379ff0bcbf6e..1b60f30cebfa 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -121,7 +121,6 @@ void __remove_from_page_cache(struct page *page) mapping->nrpages--; __dec_zone_page_state(page, NR_FILE_PAGES); BUG_ON(page_mapped(page)); - mem_cgroup_uncharge_cache_page(page); /* * Some filesystems seem to re-dirty the page even after @@ -145,6 +144,7 @@ void remove_from_page_cache(struct page *page) spin_lock_irq(&mapping->tree_lock); __remove_from_page_cache(page); spin_unlock_irq(&mapping->tree_lock); + mem_cgroup_uncharge_cache_page(page); } static int sync_page(void *word) @@ -476,13 +476,13 @@ int add_to_page_cache_locked(struct page *page, struct address_space *mapping, if (likely(!error)) { mapping->nrpages++; __inc_zone_page_state(page, NR_FILE_PAGES); + spin_unlock_irq(&mapping->tree_lock); } else { page->mapping = NULL; + spin_unlock_irq(&mapping->tree_lock); mem_cgroup_uncharge_cache_page(page); page_cache_release(page); } - - spin_unlock_irq(&mapping->tree_lock); radix_tree_preload_end(); } else mem_cgroup_uncharge_cache_page(page); diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 01c2d8f14685..4a747a27a22f 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1488,8 +1488,9 @@ void mem_cgroup_uncharge_cache_page(struct page *page) __mem_cgroup_uncharge_common(page, MEM_CGROUP_CHARGE_TYPE_CACHE); } +#ifdef CONFIG_SWAP /* - * called from __delete_from_swap_cache() and drop "page" account. + * called after __delete_from_swap_cache() and drop "page" account. * memcg information is recorded to swap_cgroup of "ent" */ void mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent) @@ -1506,6 +1507,7 @@ void mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent) if (memcg) css_put(&memcg->css); } +#endif #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP /* diff --git a/mm/swap_state.c b/mm/swap_state.c index 3ecea98ecb45..1416e7e9e02d 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -109,8 +109,6 @@ int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp_mask) */ void __delete_from_swap_cache(struct page *page) { - swp_entry_t ent = {.val = page_private(page)}; - VM_BUG_ON(!PageLocked(page)); VM_BUG_ON(!PageSwapCache(page)); VM_BUG_ON(PageWriteback(page)); @@ -121,7 +119,6 @@ void __delete_from_swap_cache(struct page *page) total_swapcache_pages--; __dec_zone_page_state(page, NR_FILE_PAGES); INC_CACHE_INFO(del_total); - mem_cgroup_uncharge_swapcache(page, ent); } /** @@ -191,6 +188,7 @@ void delete_from_swap_cache(struct page *page) __delete_from_swap_cache(page); spin_unlock_irq(&swapper_space.tree_lock); + mem_cgroup_uncharge_swapcache(page, entry); swap_free(entry); page_cache_release(page); } diff --git a/mm/truncate.c b/mm/truncate.c index 55206fab7b99..12e1579f9165 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -359,6 +359,7 @@ invalidate_complete_page2(struct address_space *mapping, struct page *page) BUG_ON(page_has_private(page)); __remove_from_page_cache(page); spin_unlock_irq(&mapping->tree_lock); + mem_cgroup_uncharge_cache_page(page); page_cache_release(page); /* pagecache ref */ return 1; failed: diff --git a/mm/vmscan.c b/mm/vmscan.c index 5fa3eda1f03f..d254306562cd 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -470,10 +470,12 @@ static int __remove_mapping(struct address_space *mapping, struct page *page) swp_entry_t swap = { .val = page_private(page) }; __delete_from_swap_cache(page); spin_unlock_irq(&mapping->tree_lock); + mem_cgroup_uncharge_swapcache(page, swap); swap_free(swap); } else { __remove_from_page_cache(page); spin_unlock_irq(&mapping->tree_lock); + mem_cgroup_uncharge_cache_page(page); } return 1; From b898f4f869da5b9d41f297fff87aca4cd42d80b3 Mon Sep 17 00:00:00 2001 From: Roel Kluin Date: Thu, 28 May 2009 14:34:29 -0700 Subject: [PATCH 255/267] drivers/serial/mpc52xx_uart.c: fix array overindexing check The check for an overindexing of mpc52xx_uart_{ports,nodes} has an off-by-one. Signed-off-by: Roel Kluin Acked-by: Wolfram Sang Acked-by: Grant Likely Cc: Benjamin Herrenschmidt Cc: Alan Cox Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/serial/mpc52xx_uart.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/serial/mpc52xx_uart.c b/drivers/serial/mpc52xx_uart.c index 7f72f8ceaa6f..b3feb6198d57 100644 --- a/drivers/serial/mpc52xx_uart.c +++ b/drivers/serial/mpc52xx_uart.c @@ -988,7 +988,7 @@ mpc52xx_console_setup(struct console *co, char *options) pr_debug("mpc52xx_console_setup co=%p, co->index=%i, options=%s\n", co, co->index, options); - if ((co->index < 0) || (co->index > MPC52xx_PSC_MAXNUM)) { + if ((co->index < 0) || (co->index >= MPC52xx_PSC_MAXNUM)) { pr_debug("PSC%x out of range\n", co->index); return -EINVAL; } From ba9447198bdd945666a9bac5e556632a7acb235d Mon Sep 17 00:00:00 2001 From: Thomas Dahlmann Date: Thu, 28 May 2009 14:34:30 -0700 Subject: [PATCH 256/267] MAINTAINERS: change email address for Thomas Dahlmann Signed-off-by: Thomas Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index ea4eced29a5c..41c6605feb0a 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -434,7 +434,7 @@ F: arch/alpha/ AMD GEODE CS5536 USB DEVICE CONTROLLER DRIVER P: Thomas Dahlmann -M: thomas.dahlmann@amd.com +M: dahlmann.thomas@arcor.de L: linux-geode@lists.infradead.org (moderated for non-subscribers) S: Supported F: drivers/usb/gadget/amd5536udc.* From c3dc5bec05a2ae03a72ef82e321d77fb549d951c Mon Sep 17 00:00:00 2001 From: Oskar Schirmer Date: Thu, 28 May 2009 14:34:31 -0700 Subject: [PATCH 257/267] flat: fix data sections alignment The flat loader uses an architecture's flat_stack_align() to align the stack but assumes word-alignment is enough for the data sections. However, on the Xtensa S6000 we have registers up to 128bit width which can be used from userspace and therefor need userspace stack and data-section alignment of at least this size. This patch drops flat_stack_align() and uses the same alignment that is required for slab caches, ARCH_SLAB_MINALIGN, or wordsize if it's not defined by the architecture. It also fixes m32r which was obviously kaput, aligning an uninitialized stack entry instead of the stack pointer. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Oskar Schirmer Cc: David Howells Cc: Russell King Cc: Bryan Wu Cc: Geert Uytterhoeven Acked-by: Paul Mundt Cc: Greg Ungerer Signed-off-by: Johannes Weiner Acked-by: Mike Frysinger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/include/asm/flat.h | 3 --- arch/blackfin/include/asm/flat.h | 1 - arch/h8300/include/asm/flat.h | 1 - arch/m32r/include/asm/flat.h | 1 - arch/m68k/include/asm/flat.h | 1 - arch/sh/include/asm/flat.h | 1 - fs/binfmt_flat.c | 46 +++++++++++++++++++++----------- 7 files changed, 31 insertions(+), 23 deletions(-) diff --git a/arch/arm/include/asm/flat.h b/arch/arm/include/asm/flat.h index 1d77e51907f6..59426a4595c9 100644 --- a/arch/arm/include/asm/flat.h +++ b/arch/arm/include/asm/flat.h @@ -5,9 +5,6 @@ #ifndef __ARM_FLAT_H__ #define __ARM_FLAT_H__ -/* An odd number of words will be pushed after this alignment, so - deliberately misalign the value. */ -#define flat_stack_align(sp) sp = (void *)(((unsigned long)(sp) - 4) | 4) #define flat_argvp_envp_on_stack() 1 #define flat_old_ram_flag(flags) (flags) #define flat_reloc_valid(reloc, size) ((reloc) <= (size)) diff --git a/arch/blackfin/include/asm/flat.h b/arch/blackfin/include/asm/flat.h index e70074e05f4e..733a178d782d 100644 --- a/arch/blackfin/include/asm/flat.h +++ b/arch/blackfin/include/asm/flat.h @@ -10,7 +10,6 @@ #include -#define flat_stack_align(sp) /* nothing needed */ #define flat_argvp_envp_on_stack() 0 #define flat_old_ram_flag(flags) (flags) diff --git a/arch/h8300/include/asm/flat.h b/arch/h8300/include/asm/flat.h index 2a873508a9a1..bd12b31b90e6 100644 --- a/arch/h8300/include/asm/flat.h +++ b/arch/h8300/include/asm/flat.h @@ -5,7 +5,6 @@ #ifndef __H8300_FLAT_H__ #define __H8300_FLAT_H__ -#define flat_stack_align(sp) /* nothing needed */ #define flat_argvp_envp_on_stack() 1 #define flat_old_ram_flag(flags) 1 #define flat_reloc_valid(reloc, size) ((reloc) <= (size)) diff --git a/arch/m32r/include/asm/flat.h b/arch/m32r/include/asm/flat.h index d851cf0c4aa5..5d711c4688fb 100644 --- a/arch/m32r/include/asm/flat.h +++ b/arch/m32r/include/asm/flat.h @@ -12,7 +12,6 @@ #ifndef __ASM_M32R_FLAT_H #define __ASM_M32R_FLAT_H -#define flat_stack_align(sp) (*sp += (*sp & 3 ? (4 - (*sp & 3)): 0)) #define flat_argvp_envp_on_stack() 0 #define flat_old_ram_flag(flags) (flags) #define flat_set_persistent(relval, p) 0 diff --git a/arch/m68k/include/asm/flat.h b/arch/m68k/include/asm/flat.h index 814b5174a8e0..a0e290793978 100644 --- a/arch/m68k/include/asm/flat.h +++ b/arch/m68k/include/asm/flat.h @@ -5,7 +5,6 @@ #ifndef __M68KNOMMU_FLAT_H__ #define __M68KNOMMU_FLAT_H__ -#define flat_stack_align(sp) /* nothing needed */ #define flat_argvp_envp_on_stack() 1 #define flat_old_ram_flag(flags) (flags) #define flat_reloc_valid(reloc, size) ((reloc) <= (size)) diff --git a/arch/sh/include/asm/flat.h b/arch/sh/include/asm/flat.h index d3b2b4f109e3..5d84df5e27f6 100644 --- a/arch/sh/include/asm/flat.h +++ b/arch/sh/include/asm/flat.h @@ -12,7 +12,6 @@ #ifndef __ASM_SH_FLAT_H #define __ASM_SH_FLAT_H -#define flat_stack_align(sp) /* nothing needed */ #define flat_argvp_envp_on_stack() 0 #define flat_old_ram_flag(flags) (flags) #define flat_reloc_valid(reloc, size) ((reloc) <= (size)) diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index 5cebf0b37798..697f6b5f1313 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -41,6 +41,7 @@ #include #include #include +#include /****************************************************************************/ @@ -54,6 +55,18 @@ #define DBG_FLT(a...) #endif +/* + * User data (stack, data section and bss) needs to be aligned + * for the same reasons as SLAB memory is, and to the same amount. + * Avoid duplicating architecture specific code by using the same + * macro as with SLAB allocation: + */ +#ifdef ARCH_SLAB_MINALIGN +#define FLAT_DATA_ALIGN (ARCH_SLAB_MINALIGN) +#else +#define FLAT_DATA_ALIGN (sizeof(void *)) +#endif + #define RELOC_FAILED 0xff00ff01 /* Relocation incorrect somewhere */ #define UNLOADED_LIB 0x7ff000ff /* Placeholder for unused library */ @@ -114,20 +127,18 @@ static unsigned long create_flat_tables( int envc = bprm->envc; char uninitialized_var(dummy); - sp = (unsigned long *) ((-(unsigned long)sizeof(char *))&(unsigned long) p); + sp = (unsigned long *)p; + sp -= (envc + argc + 2) + 1 + (flat_argvp_envp_on_stack() ? 2 : 0); + sp = (unsigned long *) ((unsigned long)sp & -FLAT_DATA_ALIGN); + argv = sp + 1 + (flat_argvp_envp_on_stack() ? 2 : 0); + envp = argv + (argc + 1); - sp -= envc+1; - envp = sp; - sp -= argc+1; - argv = sp; - - flat_stack_align(sp); if (flat_argvp_envp_on_stack()) { - --sp; put_user((unsigned long) envp, sp); - --sp; put_user((unsigned long) argv, sp); + put_user((unsigned long) envp, sp + 2); + put_user((unsigned long) argv, sp + 1); } - put_user(argc,--sp); + put_user(argc, sp); current->mm->arg_start = (unsigned long) p; while (argc-->0) { put_user((unsigned long) p, argv++); @@ -558,7 +569,9 @@ static int load_flat_file(struct linux_binprm * bprm, ret = realdatastart; goto err; } - datapos = realdatastart + MAX_SHARED_LIBS * sizeof(unsigned long); + datapos = ALIGN(realdatastart + + MAX_SHARED_LIBS * sizeof(unsigned long), + FLAT_DATA_ALIGN); DBG_FLT("BINFMT_FLAT: Allocated data+bss+stack (%d bytes): %x\n", (int)(data_len + bss_len + stack_len), (int)datapos); @@ -604,9 +617,12 @@ static int load_flat_file(struct linux_binprm * bprm, } realdatastart = textpos + ntohl(hdr->data_start); - datapos = realdatastart + MAX_SHARED_LIBS * sizeof(unsigned long); - reloc = (unsigned long *) (textpos + ntohl(hdr->reloc_start) + - MAX_SHARED_LIBS * sizeof(unsigned long)); + datapos = ALIGN(realdatastart + + MAX_SHARED_LIBS * sizeof(unsigned long), + FLAT_DATA_ALIGN); + + reloc = (unsigned long *) + (datapos + (ntohl(hdr->reloc_start) - text_len)); memp = textpos; memp_size = len; #ifdef CONFIG_BINFMT_ZFLAT @@ -854,7 +870,7 @@ static int load_flat_binary(struct linux_binprm * bprm, struct pt_regs * regs) stack_len = TOP_OF_ARGS - bprm->p; /* the strings */ stack_len += (bprm->argc + 1) * sizeof(char *); /* the argv array */ stack_len += (bprm->envc + 1) * sizeof(char *); /* the envp array */ - + stack_len += FLAT_DATA_ALIGN - 1; /* reserve for upcoming alignment */ res = load_flat_file(bprm, &libinfo, 0, &stack_len); if (res > (unsigned long)-4096) From b5d598b41aebee67bf95802b68b888e98a449687 Mon Sep 17 00:00:00 2001 From: Alexander Beregalov Date: Thu, 28 May 2009 14:34:33 -0700 Subject: [PATCH 258/267] parport_gsc: fix printk format error drivers/parport/parport_gsc.c:356: warning: format '%lx' expects type 'long unsigned int', but argument 2 has type 'resource_size_t' [akpm@linux-foundation.org: fix it to handle u64's] Signed-off-by: Alexander Beregalov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/parport/parport_gsc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/parport/parport_gsc.c b/drivers/parport/parport_gsc.c index e6a7e847ee80..ea31a452b153 100644 --- a/drivers/parport/parport_gsc.c +++ b/drivers/parport/parport_gsc.c @@ -352,8 +352,8 @@ static int __devinit parport_init_chip(struct parisc_device *dev) unsigned long port; if (!dev->irq) { - printk(KERN_WARNING "IRQ not found for parallel device at 0x%lx\n", - dev->hpa.start); + printk(KERN_WARNING "IRQ not found for parallel device at 0x%llx\n", + (unsigned long long)dev->hpa.start); return -ENODEV; } From 8e8e8267f0a08c2415d5f51bc9a9fde6d5400619 Mon Sep 17 00:00:00 2001 From: Alexander Beregalov Date: Thu, 28 May 2009 14:34:34 -0700 Subject: [PATCH 259/267] serial: 8250_gsc: fix printk format error drivers/serial/8250_gsc.c:44: warning: format '%lx' expects type 'long unsigned int', but argument 2 has type 'resource_size_t' [akpm@linux-foundation.org: fix it to handle u64's] Signed-off-by: Alexander Beregalov Cc: Alan Cox Cc: Kyle McMartin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/serial/8250_gsc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/serial/8250_gsc.c b/drivers/serial/8250_gsc.c index 418b4fe9a0a1..33149d982e82 100644 --- a/drivers/serial/8250_gsc.c +++ b/drivers/serial/8250_gsc.c @@ -39,9 +39,9 @@ static int __init serial_init_chip(struct parisc_device *dev) */ if (parisc_parent(dev)->id.hw_type != HPHW_IOA) printk(KERN_INFO - "Serial: device 0x%lx not configured.\n" + "Serial: device 0x%llx not configured.\n" "Enable support for Wax, Lasi, Asp or Dino.\n", - dev->hpa.start); + (unsigned long long)dev->hpa.start); return -ENODEV; } From 17663e59704bea838a9236f299104e30909a43b1 Mon Sep 17 00:00:00 2001 From: Marek Szyprowski Date: Thu, 28 May 2009 14:34:35 -0700 Subject: [PATCH 260/267] S3C-fb: PM fix Correctly restore the FrameBuffer register state in the resume function. Reviewed-by: Kyungmin Park Signed-off-by: Marek Szyprowski Cc: Ben Dooks Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/s3c-fb.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/drivers/video/s3c-fb.c b/drivers/video/s3c-fb.c index 5e9c6302433b..d3a568e6b169 100644 --- a/drivers/video/s3c-fb.c +++ b/drivers/video/s3c-fb.c @@ -947,7 +947,8 @@ static int __devexit s3c_fb_remove(struct platform_device *pdev) int win; for (win = 0; win <= S3C_FB_MAX_WIN; win++) - s3c_fb_release_win(sfb, sfb->windows[win]); + if (sfb->windows[win]) + s3c_fb_release_win(sfb, sfb->windows[win]); iounmap(sfb->regs); @@ -985,11 +986,20 @@ static int s3c_fb_suspend(struct platform_device *pdev, pm_message_t state) static int s3c_fb_resume(struct platform_device *pdev) { struct s3c_fb *sfb = platform_get_drvdata(pdev); + struct s3c_fb_platdata *pd = sfb->pdata; struct s3c_fb_win *win; int win_no; clk_enable(sfb->bus_clk); + /* setup registers */ + writel(pd->vidcon1, sfb->regs + VIDCON1); + + /* zero all windows before we do anything */ + for (win_no = 0; win_no < S3C_FB_MAX_WIN; win_no++) + s3c_fb_clear_win(sfb, win_no); + + /* restore framebuffers */ for (win_no = 0; win_no < S3C_FB_MAX_WIN; win_no++) { win = sfb->windows[win_no]; if (!win) From 53b7479bbdaedcc7846c66fd608fe66f1b5aa35b Mon Sep 17 00:00:00 2001 From: Nicolas Ferre Date: Thu, 28 May 2009 14:34:36 -0700 Subject: [PATCH 261/267] atmel_lcdfb: correct fifo size for some products Remove wrong fifo size definition for some AT91 products. Due to a misunderstanding of some AT91 datasheets, a fifo size of 2048 (words) has been introduced by mistake. In fact, all products (AT91/AT32) are sharing the same fifo size of 512 words. Signed-off-by: Nicolas Ferre Cc: Andrew Victor Acked-by: Haavard Skinnemoen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/atmel_lcdfb.c | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) diff --git a/drivers/video/atmel_lcdfb.c b/drivers/video/atmel_lcdfb.c index 9a577a800db5..2fb63f6ea2f1 100644 --- a/drivers/video/atmel_lcdfb.c +++ b/drivers/video/atmel_lcdfb.c @@ -29,14 +29,8 @@ /* configurable parameters */ #define ATMEL_LCDC_CVAL_DEFAULT 0xc8 -#define ATMEL_LCDC_DMA_BURST_LEN 8 - -#if defined(CONFIG_ARCH_AT91SAM9263) || defined(CONFIG_ARCH_AT91CAP9) || \ - defined(CONFIG_ARCH_AT91SAM9RL) -#define ATMEL_LCDC_FIFO_SIZE 2048 -#else -#define ATMEL_LCDC_FIFO_SIZE 512 -#endif +#define ATMEL_LCDC_DMA_BURST_LEN 8 /* words */ +#define ATMEL_LCDC_FIFO_SIZE 512 /* words */ #if defined(CONFIG_ARCH_AT91) #define ATMEL_LCDFB_FBINFO_DEFAULT (FBINFO_DEFAULT \ From 32b154c0b0bae2879bf4e549d861caf1759a3546 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 28 May 2009 14:34:37 -0700 Subject: [PATCH 262/267] x86: ignore VM_LOCKED when determining if hugetlb-backed page tables can be shared or not Addresses http://bugzilla.kernel.org/show_bug.cgi?id=13302 On x86 and x86-64, it is possible that page tables are shared beween shared mappings backed by hugetlbfs. As part of this, page_table_shareable() checks a pair of vma->vm_flags and they must match if they are to be shared. All VMA flags are taken into account, including VM_LOCKED. The problem is that VM_LOCKED is cleared on fork(). When a process with a shared memory segment forks() to exec() a helper, there will be shared VMAs with different flags. The impact is that the shared segment is sometimes considered shareable and other times not, depending on what process is checking. What happens is that the segment page tables are being shared but the count is inaccurate depending on the ordering of events. As the page tables are freed with put_page(), bad pmd's are found when some of the children exit. The hugepage counters also get corrupted and the Total and Free count will no longer match even when all the hugepage-backed regions are freed. This requires a reboot of the machine to "fix". This patch addresses the problem by comparing all flags except VM_LOCKED when deciding if pagetables should be shared or not for hugetlbfs-backed mapping. Signed-off-by: Mel Gorman Acked-by: Hugh Dickins Cc: Ingo Molnar Cc: Cc: Lee Schermerhorn Cc: KOSAKI Motohiro Cc: Cc: Eric B Munson Cc: Adam Litke Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/mm/hugetlbpage.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c index 8f307d914c2e..f46c340727b8 100644 --- a/arch/x86/mm/hugetlbpage.c +++ b/arch/x86/mm/hugetlbpage.c @@ -26,12 +26,16 @@ static unsigned long page_table_shareable(struct vm_area_struct *svma, unsigned long sbase = saddr & PUD_MASK; unsigned long s_end = sbase + PUD_SIZE; + /* Allow segments to share if only one is marked locked */ + unsigned long vm_flags = vma->vm_flags & ~VM_LOCKED; + unsigned long svm_flags = svma->vm_flags & ~VM_LOCKED; + /* * match the virtual addresses, permission and the alignment of the * page table page. */ if (pmd_index(addr) != pmd_index(saddr) || - vma->vm_flags != svma->vm_flags || + vm_flags != svm_flags || sbase < svma->vm_start || svma->vm_end < s_end) return 0; From f83a275dbc5ca1721143698e844243fcadfabf6a Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 28 May 2009 14:34:40 -0700 Subject: [PATCH 263/267] mm: account for MAP_SHARED mappings using VM_MAYSHARE and not VM_SHARED in hugetlbfs Addresses http://bugzilla.kernel.org/show_bug.cgi?id=13302 hugetlbfs reserves huge pages but does not fault them at mmap() time to ensure that future faults succeed. The reservation behaviour differs depending on whether the mapping was mapped MAP_SHARED or MAP_PRIVATE. For MAP_SHARED mappings, hugepages are reserved when mmap() is first called and are tracked based on information associated with the inode. Other processes mapping MAP_SHARED use the same reservation. MAP_PRIVATE track the reservations based on the VMA created as part of the mmap() operation. Each process mapping MAP_PRIVATE must make its own reservation. hugetlbfs currently checks if a VMA is MAP_SHARED with the VM_SHARED flag and not VM_MAYSHARE. For file-backed mappings, such as hugetlbfs, VM_SHARED is set only if the mapping is MAP_SHARED and the file was opened read-write. If a shared memory mapping was mapped shared-read-write for populating of data and mapped shared-read-only by other processes, then hugetlbfs would account for the mapping as if it was MAP_PRIVATE. This causes processes to fail to map the file MAP_SHARED even though it should succeed as the reservation is there. This patch alters mm/hugetlb.c and replaces VM_SHARED with VM_MAYSHARE when the intent of the code was to check whether the VMA was mapped MAP_SHARED or MAP_PRIVATE. Signed-off-by: Mel Gorman Cc: Hugh Dickins Cc: Ingo Molnar Cc: Cc: Lee Schermerhorn Cc: KOSAKI Motohiro Cc: Cc: Eric B Munson Cc: Adam Litke Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 28c655ba9353..e83ad2c9228c 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -316,7 +316,7 @@ static void resv_map_release(struct kref *ref) static struct resv_map *vma_resv_map(struct vm_area_struct *vma) { VM_BUG_ON(!is_vm_hugetlb_page(vma)); - if (!(vma->vm_flags & VM_SHARED)) + if (!(vma->vm_flags & VM_MAYSHARE)) return (struct resv_map *)(get_vma_private_data(vma) & ~HPAGE_RESV_MASK); return NULL; @@ -325,7 +325,7 @@ static struct resv_map *vma_resv_map(struct vm_area_struct *vma) static void set_vma_resv_map(struct vm_area_struct *vma, struct resv_map *map) { VM_BUG_ON(!is_vm_hugetlb_page(vma)); - VM_BUG_ON(vma->vm_flags & VM_SHARED); + VM_BUG_ON(vma->vm_flags & VM_MAYSHARE); set_vma_private_data(vma, (get_vma_private_data(vma) & HPAGE_RESV_MASK) | (unsigned long)map); @@ -334,7 +334,7 @@ static void set_vma_resv_map(struct vm_area_struct *vma, struct resv_map *map) static void set_vma_resv_flags(struct vm_area_struct *vma, unsigned long flags) { VM_BUG_ON(!is_vm_hugetlb_page(vma)); - VM_BUG_ON(vma->vm_flags & VM_SHARED); + VM_BUG_ON(vma->vm_flags & VM_MAYSHARE); set_vma_private_data(vma, get_vma_private_data(vma) | flags); } @@ -353,7 +353,7 @@ static void decrement_hugepage_resv_vma(struct hstate *h, if (vma->vm_flags & VM_NORESERVE) return; - if (vma->vm_flags & VM_SHARED) { + if (vma->vm_flags & VM_MAYSHARE) { /* Shared mappings always use reserves */ h->resv_huge_pages--; } else if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) { @@ -369,14 +369,14 @@ static void decrement_hugepage_resv_vma(struct hstate *h, void reset_vma_resv_huge_pages(struct vm_area_struct *vma) { VM_BUG_ON(!is_vm_hugetlb_page(vma)); - if (!(vma->vm_flags & VM_SHARED)) + if (!(vma->vm_flags & VM_MAYSHARE)) vma->vm_private_data = (void *)0; } /* Returns true if the VMA has associated reserve pages */ static int vma_has_reserves(struct vm_area_struct *vma) { - if (vma->vm_flags & VM_SHARED) + if (vma->vm_flags & VM_MAYSHARE) return 1; if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) return 1; @@ -924,7 +924,7 @@ static long vma_needs_reservation(struct hstate *h, struct address_space *mapping = vma->vm_file->f_mapping; struct inode *inode = mapping->host; - if (vma->vm_flags & VM_SHARED) { + if (vma->vm_flags & VM_MAYSHARE) { pgoff_t idx = vma_hugecache_offset(h, vma, addr); return region_chg(&inode->i_mapping->private_list, idx, idx + 1); @@ -949,7 +949,7 @@ static void vma_commit_reservation(struct hstate *h, struct address_space *mapping = vma->vm_file->f_mapping; struct inode *inode = mapping->host; - if (vma->vm_flags & VM_SHARED) { + if (vma->vm_flags & VM_MAYSHARE) { pgoff_t idx = vma_hugecache_offset(h, vma, addr); region_add(&inode->i_mapping->private_list, idx, idx + 1); @@ -1893,7 +1893,7 @@ retry_avoidcopy: * at the time of fork() could consume its reserves on COW instead * of the full address range. */ - if (!(vma->vm_flags & VM_SHARED) && + if (!(vma->vm_flags & VM_MAYSHARE) && is_vma_resv_set(vma, HPAGE_RESV_OWNER) && old_page != pagecache_page) outside_reserve = 1; @@ -2000,7 +2000,7 @@ retry: clear_huge_page(page, address, huge_page_size(h)); __SetPageUptodate(page); - if (vma->vm_flags & VM_SHARED) { + if (vma->vm_flags & VM_MAYSHARE) { int err; struct inode *inode = mapping->host; @@ -2104,7 +2104,7 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, goto out_mutex; } - if (!(vma->vm_flags & VM_SHARED)) + if (!(vma->vm_flags & VM_MAYSHARE)) pagecache_page = hugetlbfs_pagecache_page(h, vma, address); } @@ -2289,7 +2289,7 @@ int hugetlb_reserve_pages(struct inode *inode, * to reserve the full area even if read-only as mprotect() may be * called to make the mapping read-write. Assume !vma is a shm mapping */ - if (!vma || vma->vm_flags & VM_SHARED) + if (!vma || vma->vm_flags & VM_MAYSHARE) chg = region_chg(&inode->i_mapping->private_list, from, to); else { struct resv_map *resv_map = resv_map_alloc(); @@ -2330,7 +2330,7 @@ int hugetlb_reserve_pages(struct inode *inode, * consumed reservations are stored in the map. Hence, nothing * else has to be done for private mappings here */ - if (!vma || vma->vm_flags & VM_SHARED) + if (!vma || vma->vm_flags & VM_MAYSHARE) region_add(&inode->i_mapping->private_list, from, to); return 0; } From 46f7e602fb32e02145ef14f8c0ca6d399f0a96b9 Mon Sep 17 00:00:00 2001 From: Nikanth Karthikesan Date: Thu, 28 May 2009 14:34:41 -0700 Subject: [PATCH 264/267] memcg: fix build warning and avoid checking for mem != null again and again Fix build warning, "mem_cgroup_is_obsolete defined but not used" when CONFIG_DEBUG_VM is not set. Also avoid checking for !mem again and again. Signed-off-by: Nikanth Karthikesan Acked-by: Pekka Enberg Acked-by: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 10 +--------- 1 file changed, 1 insertion(+), 9 deletions(-) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 4a747a27a22f..78eb8552818b 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -314,14 +314,6 @@ static struct mem_cgroup *try_get_mem_cgroup_from_mm(struct mm_struct *mm) return mem; } -static bool mem_cgroup_is_obsolete(struct mem_cgroup *mem) -{ - if (!mem) - return true; - return css_is_removed(&mem->css); -} - - /* * Call callback function against all cgroup under hierarchy tree. */ @@ -932,7 +924,7 @@ static int __mem_cgroup_try_charge(struct mm_struct *mm, if (unlikely(!mem)) return 0; - VM_BUG_ON(!mem || mem_cgroup_is_obsolete(mem)); + VM_BUG_ON(css_is_removed(&mem->css)); while (1) { int ret; From 56ec0c7b88c6eb17733e5015f31302f6312511ed Mon Sep 17 00:00:00 2001 From: Harry Ciao Date: Thu, 28 May 2009 14:34:42 -0700 Subject: [PATCH 265/267] edac: AMD8111 & AMD8131 use dev_name() The "bus_id" member in the device structure has been obsolete, use dev_name() instead. Signed-off-by: Harry Ciao Cc: Doug Thompson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/edac/amd8111_edac.c | 4 ++-- drivers/edac/amd8131_edac.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/edac/amd8111_edac.c b/drivers/edac/amd8111_edac.c index 614692181120..2cb58ef743e0 100644 --- a/drivers/edac/amd8111_edac.c +++ b/drivers/edac/amd8111_edac.c @@ -389,7 +389,7 @@ static int amd8111_dev_probe(struct pci_dev *dev, dev_info->edac_dev->dev = &dev_info->dev->dev; dev_info->edac_dev->mod_name = AMD8111_EDAC_MOD_STR; dev_info->edac_dev->ctl_name = dev_info->ctl_name; - dev_info->edac_dev->dev_name = dev_info->dev->dev.bus_id; + dev_info->edac_dev->dev_name = dev_name(&dev_info->dev->dev); if (edac_op_state == EDAC_OPSTATE_POLL) dev_info->edac_dev->edac_check = dev_info->check; @@ -473,7 +473,7 @@ static int amd8111_pci_probe(struct pci_dev *dev, pci_info->edac_dev->dev = &pci_info->dev->dev; pci_info->edac_dev->mod_name = AMD8111_EDAC_MOD_STR; pci_info->edac_dev->ctl_name = pci_info->ctl_name; - pci_info->edac_dev->dev_name = pci_info->dev->dev.bus_id; + pci_info->edac_dev->dev_name = dev_name(&pci_info->dev->dev); if (edac_op_state == EDAC_OPSTATE_POLL) pci_info->edac_dev->edac_check = pci_info->check; diff --git a/drivers/edac/amd8131_edac.c b/drivers/edac/amd8131_edac.c index c083b31cac5a..b432d60c622a 100644 --- a/drivers/edac/amd8131_edac.c +++ b/drivers/edac/amd8131_edac.c @@ -287,7 +287,7 @@ static int amd8131_probe(struct pci_dev *dev, const struct pci_device_id *id) dev_info->edac_dev->dev = &dev_info->dev->dev; dev_info->edac_dev->mod_name = AMD8131_EDAC_MOD_STR; dev_info->edac_dev->ctl_name = dev_info->ctl_name; - dev_info->edac_dev->dev_name = dev_info->dev->dev.bus_id; + dev_info->edac_dev->dev_name = dev_name(&dev_info->dev->dev); if (edac_op_state == EDAC_OPSTATE_POLL) dev_info->edac_dev->edac_check = amd8131_chipset.check; From 715fe7af9fd7328af661742bfadc195e665a837f Mon Sep 17 00:00:00 2001 From: Harry Ciao Date: Thu, 28 May 2009 14:34:43 -0700 Subject: [PATCH 266/267] edac: AMD8111 & AMD8131 Kconfig fixup The amd8111_edac.c driver will fail allmodconfig on architectures other than PPC, introduce Kconfig dependency to avoid this, since both AMD8111 and AMD8131 chips are only adopted on Maple so far. Signed-off-by: Harry Ciao Cc: Doug Thompson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/edac/Kconfig | 8 ++++++-- drivers/edac/Makefile | 2 ++ 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/edac/Kconfig b/drivers/edac/Kconfig index e5f5c5a8ba6c..956982f8739b 100644 --- a/drivers/edac/Kconfig +++ b/drivers/edac/Kconfig @@ -192,16 +192,20 @@ config EDAC_PPC4XX config EDAC_AMD8131 tristate "AMD8131 HyperTransport PCI-X Tunnel" - depends on EDAC_MM_EDAC && PCI + depends on EDAC_MM_EDAC && PCI && PPC_MAPLE help Support for error detection and correction on the AMD8131 HyperTransport PCI-X Tunnel chip. + Note, add more Kconfig dependency if it's adopted + on some machine other than Maple. config EDAC_AMD8111 tristate "AMD8111 HyperTransport I/O Hub" - depends on EDAC_MM_EDAC && PCI + depends on EDAC_MM_EDAC && PCI && PPC_MAPLE help Support for error detection and correction on the AMD8111 HyperTransport I/O Hub chip. + Note, add more Kconfig dependency if it's adopted + on some machine other than Maple. endif # EDAC diff --git a/drivers/edac/Makefile b/drivers/edac/Makefile index a5fdcf02f591..59076819135d 100644 --- a/drivers/edac/Makefile +++ b/drivers/edac/Makefile @@ -35,3 +35,5 @@ obj-$(CONFIG_EDAC_MPC85XX) += mpc85xx_edac.o obj-$(CONFIG_EDAC_MV64X60) += mv64x60_edac.o obj-$(CONFIG_EDAC_CELL) += cell_edac.o obj-$(CONFIG_EDAC_PPC4XX) += ppc4xx_edac.o +obj-$(CONFIG_EDAC_AMD8111) += amd8111_edac.o +obj-$(CONFIG_EDAC_AMD8131) += amd8131_edac.o From b8e7e40abeac49644fec4a4f52ffe74c7b05eca0 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Thu, 28 May 2009 14:01:35 +0100 Subject: [PATCH 267/267] 8250: Fix oops from setserial If you setserial a port which has never been initialised we change the type but don't update the I/O method pointers. The same problem is true if you change the io type of a port - but nobody ever does that so nobody noticed! Remember the old type and when attaching if the type has changed reload the port accessor pointers. We can't do it blindly as some 8250 drivers load custom accessors and we must not stomp those. Tested-by: Victor Seryodkin Closes-bug: #13367 Signed-off-by: Alan Cox Signed-off-by: Linus Torvalds --- drivers/serial/8250.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c index b4b39811b445..a0127e93ade0 100644 --- a/drivers/serial/8250.c +++ b/drivers/serial/8250.c @@ -137,6 +137,7 @@ struct uart_8250_port { unsigned char mcr; unsigned char mcr_mask; /* mask of user bits */ unsigned char mcr_force; /* mask of forced bits */ + unsigned char cur_iotype; /* Running I/O type */ /* * Some bits in registers are cleared on a read, so they must @@ -471,6 +472,7 @@ static void io_serial_out(struct uart_port *p, int offset, int value) static void set_io_from_upio(struct uart_port *p) { + struct uart_8250_port *up = (struct uart_8250_port *)p; switch (p->iotype) { case UPIO_HUB6: p->serial_in = hub6_serial_in; @@ -509,6 +511,8 @@ static void set_io_from_upio(struct uart_port *p) p->serial_out = io_serial_out; break; } + /* Remember loaded iotype */ + up->cur_iotype = p->iotype; } static void @@ -1937,6 +1941,9 @@ static int serial8250_startup(struct uart_port *port) up->capabilities = uart_config[up->port.type].flags; up->mcr = 0; + if (up->port.iotype != up->cur_iotype) + set_io_from_upio(port); + if (up->port.type == PORT_16C950) { /* Wake up and initialize UART */ up->acr = 0; @@ -2563,6 +2570,9 @@ static void serial8250_config_port(struct uart_port *port, int flags) if (ret < 0) probeflags &= ~PROBE_RSA; + if (up->port.iotype != up->cur_iotype) + set_io_from_upio(port); + if (flags & UART_CONFIG_TYPE) autoconfig(up, probeflags); if (up->port.type != PORT_UNKNOWN && flags & UART_CONFIG_IRQ) @@ -2671,6 +2681,11 @@ serial8250_register_ports(struct uart_driver *drv, struct device *dev) { int i; + for (i = 0; i < nr_uarts; i++) { + struct uart_8250_port *up = &serial8250_ports[i]; + up->cur_iotype = 0xFF; + } + serial8250_isa_init_ports(); for (i = 0; i < nr_uarts; i++) {