linux/drivers/block/osdblk.c

700 lines
16 KiB
C
Raw Normal View History

/*
osdblk.c -- Export a single SCSI OSD object as a Linux block device
Copyright 2009 Red Hat, Inc.
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; see the file COPYING. If not, write to
the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
Instructions for use
--------------------
1) Map a Linux block device to an existing OSD object.
In this example, we will use partition id 1234, object id 5678,
OSD device /dev/osd1.
$ echo "1234 5678 /dev/osd1" > /sys/class/osdblk/add
2) List all active blkdev<->object mappings.
In this example, we have performed step #1 twice, creating two blkdevs,
mapped to two separate OSD objects.
$ cat /sys/class/osdblk/list
0 174 1234 5678 /dev/osd1
1 179 1994 897123 /dev/osd0
The columns, in order, are:
- blkdev unique id
- blkdev assigned major
- OSD object partition id
- OSD object id
- OSD device
3) Remove an active blkdev<->object mapping.
In this example, we remove the mapping with blkdev unique id 1.
$ echo 1 > /sys/class/osdblk/remove
NOTE: The actual creation and deletion of OSD objects is outside the scope
of this driver.
*/
#include <linux/kernel.h>
#include <linux/device.h>
#include <linux/module.h>
#include <linux/fs.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/slab.h>
#include <scsi/osd_initiator.h>
#include <scsi/osd_attributes.h>
#include <scsi/osd_sec.h>
#include <scsi/scsi_device.h>
#define DRV_NAME "osdblk"
#define PFX DRV_NAME ": "
/* #define _OSDBLK_DEBUG */
#ifdef _OSDBLK_DEBUG
#define OSDBLK_DEBUG(fmt, a...) \
printk(KERN_NOTICE "osdblk @%s:%d: " fmt, __func__, __LINE__, ##a)
#else
#define OSDBLK_DEBUG(fmt, a...) \
do { if (0) printk(fmt, ##a); } while (0)
#endif
MODULE_AUTHOR("Jeff Garzik <jeff@garzik.org>");
MODULE_DESCRIPTION("block device inside an OSD object osdblk.ko");
MODULE_LICENSE("GPL");
struct osdblk_device;
enum {
OSDBLK_MINORS_PER_MAJOR = 256, /* max minors per blkdev */
OSDBLK_MAX_REQ = 32, /* max parallel requests */
OSDBLK_OP_TIMEOUT = 4 * 60, /* sync OSD req timeout */
};
struct osdblk_request {
struct request *rq; /* blk layer request */
struct bio *bio; /* cloned bio */
struct osdblk_device *osdev; /* associated blkdev */
};
struct osdblk_device {
int id; /* blkdev unique id */
int major; /* blkdev assigned major */
struct gendisk *disk; /* blkdev's gendisk and rq */
struct request_queue *q;
struct osd_dev *osd; /* associated OSD */
char name[32]; /* blkdev name, e.g. osdblk34 */
spinlock_t lock; /* queue lock */
struct osd_obj_id obj; /* OSD partition, obj id */
uint8_t obj_cred[OSD_CAP_LEN]; /* OSD cred */
struct osdblk_request req[OSDBLK_MAX_REQ]; /* request table */
struct list_head node;
char osd_path[0]; /* OSD device path */
};
static struct class *class_osdblk; /* /sys/class/osdblk */
static DEFINE_MUTEX(ctl_mutex); /* Serialize open/close/setup/teardown */
static LIST_HEAD(osdblkdev_list);
static const struct block_device_operations osdblk_bd_ops = {
.owner = THIS_MODULE,
};
static const struct osd_attr g_attr_logical_length = ATTR_DEF(
OSD_APAGE_OBJECT_INFORMATION, OSD_ATTR_OI_LOGICAL_LENGTH, 8);
static void osdblk_make_credential(u8 cred_a[OSD_CAP_LEN],
const struct osd_obj_id *obj)
{
osd_sec_init_nosec_doall_caps(cred_a, obj, false, true);
}
/* copied from exofs; move to libosd? */
/*
* Perform a synchronous OSD operation. copied from exofs; move to libosd?
*/
static int osd_sync_op(struct osd_request *or, int timeout, uint8_t *credential)
{
int ret;
or->timeout = timeout;
ret = osd_finalize_request(or, 0, credential, NULL);
if (ret)
return ret;
ret = osd_execute_request(or);
/* osd_req_decode_sense(or, ret); */
return ret;
}
/*
* Perform an asynchronous OSD operation. copied from exofs; move to libosd?
*/
static int osd_async_op(struct osd_request *or, osd_req_done_fn *async_done,
void *caller_context, u8 *cred)
{
int ret;
ret = osd_finalize_request(or, 0, cred, NULL);
if (ret)
return ret;
ret = osd_execute_request_async(or, async_done, caller_context);
return ret;
}
/* copied from exofs; move to libosd? */
static int extract_attr_from_req(struct osd_request *or, struct osd_attr *attr)
{
struct osd_attr cur_attr = {.attr_page = 0}; /* start with zeros */
void *iter = NULL;
int nelem;
do {
nelem = 1;
osd_req_decode_get_attr_list(or, &cur_attr, &nelem, &iter);
if ((cur_attr.attr_page == attr->attr_page) &&
(cur_attr.attr_id == attr->attr_id)) {
attr->len = cur_attr.len;
attr->val_ptr = cur_attr.val_ptr;
return 0;
}
} while (iter);
return -EIO;
}
static int osdblk_get_obj_size(struct osdblk_device *osdev, u64 *size_out)
{
struct osd_request *or;
struct osd_attr attr;
int ret;
/* start request */
or = osd_start_request(osdev->osd, GFP_KERNEL);
if (!or)
return -ENOMEM;
/* create a get-attributes(length) request */
osd_req_get_attributes(or, &osdev->obj);
osd_req_add_get_attr_list(or, &g_attr_logical_length, 1);
/* execute op synchronously */
ret = osd_sync_op(or, OSDBLK_OP_TIMEOUT, osdev->obj_cred);
if (ret)
goto out;
/* extract length from returned attribute info */
attr = g_attr_logical_length;
ret = extract_attr_from_req(or, &attr);
if (ret)
goto out;
*size_out = get_unaligned_be64(attr.val_ptr);
out:
osd_end_request(or);
return ret;
}
static void osdblk_osd_complete(struct osd_request *or, void *private)
{
struct osdblk_request *orq = private;
struct osd_sense_info osi;
int ret = osd_req_decode_sense(or, &osi);
if (ret) {
ret = -EIO;
OSDBLK_DEBUG("osdblk_osd_complete with err=%d\n", ret);
}
/* complete OSD request */
osd_end_request(or);
/* complete request passed to osdblk by block layer */
__blk_end_request_all(orq->rq, ret);
}
static void bio_chain_put(struct bio *chain)
{
struct bio *tmp;
while (chain) {
tmp = chain;
chain = chain->bi_next;
bio_put(tmp);
}
}
static struct bio *bio_chain_clone(struct bio *old_chain, gfp_t gfpmask)
{
struct bio *tmp, *new_chain = NULL, *tail = NULL;
while (old_chain) {
tmp = bio_clone_kmalloc(old_chain, gfpmask);
if (!tmp)
goto err_out;
tmp->bi_bdev = NULL;
mm, page_alloc: distinguish between being unable to sleep, unwilling to sleep and avoiding waking kswapd __GFP_WAIT has been used to identify atomic context in callers that hold spinlocks or are in interrupts. They are expected to be high priority and have access one of two watermarks lower than "min" which can be referred to as the "atomic reserve". __GFP_HIGH users get access to the first lower watermark and can be called the "high priority reserve". Over time, callers had a requirement to not block when fallback options were available. Some have abused __GFP_WAIT leading to a situation where an optimisitic allocation with a fallback option can access atomic reserves. This patch uses __GFP_ATOMIC to identify callers that are truely atomic, cannot sleep and have no alternative. High priority users continue to use __GFP_HIGH. __GFP_DIRECT_RECLAIM identifies callers that can sleep and are willing to enter direct reclaim. __GFP_KSWAPD_RECLAIM to identify callers that want to wake kswapd for background reclaim. __GFP_WAIT is redefined as a caller that is willing to enter direct reclaim and wake kswapd for background reclaim. This patch then converts a number of sites o __GFP_ATOMIC is used by callers that are high priority and have memory pools for those requests. GFP_ATOMIC uses this flag. o Callers that have a limited mempool to guarantee forward progress clear __GFP_DIRECT_RECLAIM but keep __GFP_KSWAPD_RECLAIM. bio allocations fall into this category where kswapd will still be woken but atomic reserves are not used as there is a one-entry mempool to guarantee progress. o Callers that are checking if they are non-blocking should use the helper gfpflags_allow_blocking() where possible. This is because checking for __GFP_WAIT as was done historically now can trigger false positives. Some exceptions like dm-crypt.c exist where the code intent is clearer if __GFP_DIRECT_RECLAIM is used instead of the helper due to flag manipulations. o Callers that built their own GFP flags instead of starting with GFP_KERNEL and friends now also need to specify __GFP_KSWAPD_RECLAIM. The first key hazard to watch out for is callers that removed __GFP_WAIT and was depending on access to atomic reserves for inconspicuous reasons. In some cases it may be appropriate for them to use __GFP_HIGH. The second key hazard is callers that assembled their own combination of GFP flags instead of starting with something like GFP_KERNEL. They may now wish to specify __GFP_KSWAPD_RECLAIM. It's almost certainly harmless if it's missed in most cases as other activity will wake kswapd. Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Vlastimil Babka <vbabka@suse.cz> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Christoph Lameter <cl@linux.com> Cc: David Rientjes <rientjes@google.com> Cc: Vitaly Wool <vitalywool@gmail.com> Cc: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-11-07 00:28:21 +00:00
gfpmask &= ~__GFP_DIRECT_RECLAIM;
tmp->bi_next = NULL;
if (!new_chain)
new_chain = tail = tmp;
else {
tail->bi_next = tmp;
tail = tmp;
}
old_chain = old_chain->bi_next;
}
return new_chain;
err_out:
OSDBLK_DEBUG("bio_chain_clone with err\n");
bio_chain_put(new_chain);
return NULL;
}
static void osdblk_rq_fn(struct request_queue *q)
{
struct osdblk_device *osdev = q->queuedata;
while (1) {
struct request *rq;
struct osdblk_request *orq;
struct osd_request *or;
struct bio *bio;
bool do_write, do_flush;
/* peek at request from block layer */
rq = blk_fetch_request(q);
if (!rq)
break;
/* filter out block requests we don't understand */
if (rq->cmd_type != REQ_TYPE_FS) {
blk_end_request_all(rq, 0);
continue;
}
/* deduce our operation (read, write, flush) */
/* I wish the block layer simplified cmd_type/cmd_flags/cmd[]
* into a clearly defined set of RPC commands:
* read, write, flush, scsi command, power mgmt req,
* driver-specific, etc.
*/
do_flush = rq->cmd_flags & REQ_FLUSH;
do_write = (rq_data_dir(rq) == WRITE);
if (!do_flush) { /* osd_flush does not use a bio */
/* a bio clone to be passed down to OSD request */
bio = bio_chain_clone(rq->bio, GFP_ATOMIC);
if (!bio)
break;
} else
bio = NULL;
/* alloc internal OSD request, for OSD command execution */
or = osd_start_request(osdev->osd, GFP_ATOMIC);
if (!or) {
bio_chain_put(bio);
OSDBLK_DEBUG("osd_start_request with err\n");
break;
}
orq = &osdev->req[rq->tag];
orq->rq = rq;
orq->bio = bio;
orq->osdev = osdev;
/* init OSD command: flush, write or read */
if (do_flush)
osd_req_flush_object(or, &osdev->obj,
OSD_CDB_FLUSH_ALL, 0, 0);
else if (do_write)
osd_req_write(or, &osdev->obj, blk_rq_pos(rq) * 512ULL,
bio, blk_rq_bytes(rq));
else
osd_req_read(or, &osdev->obj, blk_rq_pos(rq) * 512ULL,
bio, blk_rq_bytes(rq));
OSDBLK_DEBUG("%s 0x%x bytes at 0x%llx\n",
do_flush ? "flush" : do_write ?
"write" : "read", blk_rq_bytes(rq),
blk_rq_pos(rq) * 512ULL);
/* begin OSD command execution */
if (osd_async_op(or, osdblk_osd_complete, orq,
osdev->obj_cred)) {
osd_end_request(or);
blk_requeue_request(q, rq);
bio_chain_put(bio);
OSDBLK_DEBUG("osd_execute_request_async with err\n");
break;
}
/* remove the special 'flush' marker, now that the command
* is executing
*/
rq->special = NULL;
}
}
static void osdblk_free_disk(struct osdblk_device *osdev)
{
struct gendisk *disk = osdev->disk;
if (!disk)
return;
if (disk->flags & GENHD_FL_UP)
del_gendisk(disk);
if (disk->queue)
blk_cleanup_queue(disk->queue);
put_disk(disk);
}
static int osdblk_init_disk(struct osdblk_device *osdev)
{
struct gendisk *disk;
struct request_queue *q;
int rc;
u64 obj_size = 0;
/* contact OSD, request size info about the object being mapped */
rc = osdblk_get_obj_size(osdev, &obj_size);
if (rc)
return rc;
/* create gendisk info */
disk = alloc_disk(OSDBLK_MINORS_PER_MAJOR);
if (!disk)
return -ENOMEM;
sprintf(disk->disk_name, DRV_NAME "%d", osdev->id);
disk->major = osdev->major;
disk->first_minor = 0;
disk->fops = &osdblk_bd_ops;
disk->private_data = osdev;
/* init rq */
q = blk_init_queue(osdblk_rq_fn, &osdev->lock);
if (!q) {
put_disk(disk);
return -ENOMEM;
}
/* switch queue to TCQ mode; allocate tag map */
rc = blk_queue_init_tags(q, OSDBLK_MAX_REQ, NULL, BLK_TAG_ALLOC_FIFO);
if (rc) {
blk_cleanup_queue(q);
put_disk(disk);
return rc;
}
/* Set our limits to the lower device limits, because osdblk cannot
* sleep when allocating a lower-request and therefore cannot be
* bouncing.
*/
blk_queue_stack_limits(q, osd_request_queue(osdev->osd));
blk_queue_prep_rq(q, blk_queue_start_tag);
blk_queue_flush(q, REQ_FLUSH);
disk->queue = q;
q->queuedata = osdev;
osdev->disk = disk;
osdev->q = q;
/* finally, announce the disk to the world */
set_capacity(disk, obj_size / 512ULL);
add_disk(disk);
printk(KERN_INFO "%s: Added of size 0x%llx\n",
disk->disk_name, (unsigned long long)obj_size);
return 0;
}
/********************************************************************
* /sys/class/osdblk/
* add map OSD object to blkdev
* remove unmap OSD object
* list show mappings
*******************************************************************/
static void class_osdblk_release(struct class *cls)
{
kfree(cls);
}
static ssize_t class_osdblk_list(struct class *c,
struct class_attribute *attr,
char *data)
{
int n = 0;
struct list_head *tmp;
mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
list_for_each(tmp, &osdblkdev_list) {
struct osdblk_device *osdev;
osdev = list_entry(tmp, struct osdblk_device, node);
n += sprintf(data+n, "%d %d %llu %llu %s\n",
osdev->id,
osdev->major,
osdev->obj.partition,
osdev->obj.id,
osdev->osd_path);
}
mutex_unlock(&ctl_mutex);
return n;
}
static ssize_t class_osdblk_add(struct class *c,
struct class_attribute *attr,
const char *buf, size_t count)
{
struct osdblk_device *osdev;
ssize_t rc;
int irc, new_id = 0;
struct list_head *tmp;
if (!try_module_get(THIS_MODULE))
return -ENODEV;
/* new osdblk_device object */
osdev = kzalloc(sizeof(*osdev) + strlen(buf) + 1, GFP_KERNEL);
if (!osdev) {
rc = -ENOMEM;
goto err_out_mod;
}
/* static osdblk_device initialization */
spin_lock_init(&osdev->lock);
INIT_LIST_HEAD(&osdev->node);
/* generate unique id: find highest unique id, add one */
mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
list_for_each(tmp, &osdblkdev_list) {
struct osdblk_device *osdev;
osdev = list_entry(tmp, struct osdblk_device, node);
if (osdev->id > new_id)
new_id = osdev->id + 1;
}
osdev->id = new_id;
/* add to global list */
list_add_tail(&osdev->node, &osdblkdev_list);
mutex_unlock(&ctl_mutex);
/* parse add command */
if (sscanf(buf, "%llu %llu %s", &osdev->obj.partition, &osdev->obj.id,
osdev->osd_path) != 3) {
rc = -EINVAL;
goto err_out_slot;
}
/* initialize rest of new object */
sprintf(osdev->name, DRV_NAME "%d", osdev->id);
/* contact requested OSD */
osdev->osd = osduld_path_lookup(osdev->osd_path);
if (IS_ERR(osdev->osd)) {
rc = PTR_ERR(osdev->osd);
goto err_out_slot;
}
/* build OSD credential */
osdblk_make_credential(osdev->obj_cred, &osdev->obj);
/* register our block device */
irc = register_blkdev(0, osdev->name);
if (irc < 0) {
rc = irc;
goto err_out_osd;
}
osdev->major = irc;
/* set up and announce blkdev mapping */
rc = osdblk_init_disk(osdev);
if (rc)
goto err_out_blkdev;
return count;
err_out_blkdev:
unregister_blkdev(osdev->major, osdev->name);
err_out_osd:
osduld_put_device(osdev->osd);
err_out_slot:
mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
list_del_init(&osdev->node);
mutex_unlock(&ctl_mutex);
kfree(osdev);
err_out_mod:
OSDBLK_DEBUG("Error adding device %s\n", buf);
module_put(THIS_MODULE);
return rc;
}
static ssize_t class_osdblk_remove(struct class *c,
struct class_attribute *attr,
const char *buf,
size_t count)
{
struct osdblk_device *osdev = NULL;
int target_id, rc;
unsigned long ul;
struct list_head *tmp;
rc = kstrtoul(buf, 10, &ul);
if (rc)
return rc;
/* convert to int; abort if we lost anything in the conversion */
target_id = (int) ul;
if (target_id != ul)
return -EINVAL;
/* remove object from list immediately */
mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
list_for_each(tmp, &osdblkdev_list) {
osdev = list_entry(tmp, struct osdblk_device, node);
if (osdev->id == target_id) {
list_del_init(&osdev->node);
break;
}
osdev = NULL;
}
mutex_unlock(&ctl_mutex);
if (!osdev)
return -ENOENT;
/* clean up and free blkdev and associated OSD connection */
osdblk_free_disk(osdev);
unregister_blkdev(osdev->major, osdev->name);
osduld_put_device(osdev->osd);
kfree(osdev);
/* release module ref */
module_put(THIS_MODULE);
return count;
}
static struct class_attribute class_osdblk_attrs[] = {
__ATTR(add, 0200, NULL, class_osdblk_add),
__ATTR(remove, 0200, NULL, class_osdblk_remove),
__ATTR(list, 0444, class_osdblk_list, NULL),
__ATTR_NULL
};
static int osdblk_sysfs_init(void)
{
int ret = 0;
/*
* create control files in sysfs
* /sys/class/osdblk/...
*/
class_osdblk = kzalloc(sizeof(*class_osdblk), GFP_KERNEL);
if (!class_osdblk)
return -ENOMEM;
class_osdblk->name = DRV_NAME;
class_osdblk->owner = THIS_MODULE;
class_osdblk->class_release = class_osdblk_release;
class_osdblk->class_attrs = class_osdblk_attrs;
ret = class_register(class_osdblk);
if (ret) {
kfree(class_osdblk);
class_osdblk = NULL;
printk(PFX "failed to create class osdblk\n");
return ret;
}
return 0;
}
static void osdblk_sysfs_cleanup(void)
{
if (class_osdblk)
class_destroy(class_osdblk);
class_osdblk = NULL;
}
static int __init osdblk_init(void)
{
int rc;
rc = osdblk_sysfs_init();
if (rc)
return rc;
return 0;
}
static void __exit osdblk_exit(void)
{
osdblk_sysfs_cleanup();
}
module_init(osdblk_init);
module_exit(osdblk_exit);