Barry K. Nathan reported the following lockdep warning:
[ 197.343948] BUG: warning at kernel/lockdep.c:1856/trace_hardirqs_on()
[ 197.345928] [<c010329b>] show_trace_log_lvl+0x5b/0x105
[ 197.346359] [<c0103896>] show_trace+0x1b/0x20
[ 197.346759] [<c01038ed>] dump_stack+0x1f/0x24
[ 197.347159] [<c012efa2>] trace_hardirqs_on+0xfb/0x185
[ 197.348873] [<c029b009>] _spin_unlock_irq+0x24/0x2d
[ 197.350620] [<e09034e8>] do_tx_done+0x171/0x179 [ns83820]
[ 197.350895] [<e090445c>] ns83820_irq+0x149/0x20b [ns83820]
[ 197.351166] [<c013b4b8>] handle_IRQ_event+0x1d/0x52
[ 197.353216] [<c013c6c2>] handle_level_irq+0x97/0xe1
[ 197.355157] [<c01048c3>] do_IRQ+0x8b/0xac
[ 197.355612] [<c0102d9d>] common_interrupt+0x25/0x2c
this is caused because the ns83820 driver re-enables irq flags
in hardirq context.
While legal in theory, in practice it should only be done if the
hardware is really old and has some very high overhead in its ISR.
(such as PIO IDE)
For modern hardware, running ISRs with irqs enabled is discouraged,
because 1) new hardware is fast enough to not cause latency problems
2) allowing the nesting of hardware interrupts only 'spreads out'
the handling of the current ISR, causing extra cachemisses that would
otherwise not happen. Furthermore, on architectures where ISRs share
the kernel stacks, enabling interrupts in ISRs introduces a much
higher kernel-stack-nesting and thus kernel-stack-overflow risk.
3) not managing irq-flags via the _irqsave / _irqrestore variants
is dangerous: it's easy to forget whether one function nests inside
another, and irq flags might be mismanaged.
In the few cases where re-enabling interrupts in an ISR is considered
useful (and unavoidable), it has to be taught to the lock validator
explicitly (because the lock validator needs the "no ISR ever enables
hardirqs" artificial simplification to keep the IRQ/softirq locking
dependencies manageable).
This teaching is done via the explicit use local_irq_enable_in_hardirq().
On a stock kernel this maps to local_irq_enable(). If the lock validator
is enabled then this does not enable interrupts.
Now, the analysis of drivers/net/ns83820.c's irq flags use: the
irq-enabling in irq context seems intentional, but i dont think it's
justified. Furthermore, the driver suffers from problem #3 above too,
in ns83820_tx_timeout() it disables irqs via local_irq_save(), but
then it calls do_tx_done() which does a spin_unlock_irq(),
re-enabling for a function that does not expect it! While currently
this bug seems harmless (only some debug printout seems to be
affected by it), it's nevertheless something to be fixed.
So this patch makes the ns83820 ISR irq-flags-safe, and cleans up
do_tx_done() use and locking to avoid the ns83820_tx_timeout() bug.
From: Arjan van de Ven <arjan@linux.intel.com>
ns83820_mib_isr takes the misc_lock in IRQ context. All other places that
do this in the ISR already use _irqsave versions, make this consistent at
least. At some point in the future someone should audit the driver to see
if all _irqsave's in the ISR can go away, this is generally an iffy/fragile
proposition though; for now get it safe, simple and consistent.
From: Arjan van de Ven <arjan@linux.intel.com>
ok this is a real driver deadlock:
The ns83820 driver enabled interrupts (by unlocking the misc_lock with
_irq) while still holding the rx_info.lock, which is required to be irq
safe since it's used in the ISR like this:
writel(1, dev->base + IER);
spin_unlock_irq(&dev->misc_lock);
kick_rx(ndev);
spin_unlock_irq(&dev->rx_info.lock);
This is can cause a deadlock if an irq was pending at the first
spin_unlock_irq already, or if one would hit during kick_rx().
Simply remove the first _irq solves this
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
More cleanup to pcnet32_loopback_test to release receive buffers if
device is not up. Created common routine to free rx buffers.
Tested ia32 and ppc64
Signed-off-by: Don Fry <brazilnut@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Suspend the chip if possible rather than stop and discard all tx and rx
frames, when changing the mcast list or entering/leaving promiscuous
mode. Created common pcnet32_suspend routine.
Tested ia32 and ppc64
Signed-off-by: Don Fry <brazilnut@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Fix pcnet32_set_ringparam to handle memory allocation errors without
leaving the adapter in an inoperative state and null pointers waiting to
be dereferenced.
Tested ia32 and ppc64.
Signed-off-by: Don Fry <brazilnut@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
On 2006-03-08 Eric Sesterhenn wrote:
converts drivers/net to kzalloc usage.
Don Fry modified it to use netif_msg_drv. Tested ia32 and ppc64.
Signed-off-by: Eric Sesterhenn <snakebyte@gmx.de>
Signed-off-by: Don Fry <brazilnut@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Jon Mason wrote on Thu, 12 Jan 2006 17:07:49 -0600:
This patch adds the PCI_DEVICE macro to the pcnet32 driver.
This has been tested on my opteron with my "trident" adapter.
Don Fry modified it slightly and tested on ia32 and ppc64.
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Don Fry <brazilnut@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Add support for the Cicada 8201 PHY, a.k.a Vitesse VSC8201. This PHY is present on the MPC8349mITX.
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
There are 60+ USB wifi adapters available on the market based on the ZyDAS
ZD1211 chip.
Unlike the predecessor (ZD1201), ZD1211 does not have a hardware MAC, so most
data operations are coordinated by the device driver. The ZD1211 chip sits
alongside an RF transceiver which is also controlled by the driver. Our driver
currently supports 2 RF types, we know of one other available in a few marketed
products which we will be supporting soon.
Our driver also supports the newer revision of ZD1211, called ZD1211B. The
initialization and RF operations are slightly different for the new revision,
but the main difference is 802.11e support. Our driver does not support the
QoS features yet, but we think we know how to use them.
This driver is based on ZyDAS's own GPL driver available from www.zydas.com.tw.
ZyDAS engineers have been responsive and supportive of our efforts, so thumbs
up to them. Additionally, the firmware is redistributable and they have
provided device specs.
This driver has been written primarily by Ulrich Kunitz and myself. Graham
Gower, Greg KH, Remco and Bryan Rittmeyer have also contributed. The
developers of ieee80211 and softmac have made our lives so much easier- thanks!
We maintain a small info-page: http://zd1211.ath.cx/wiki/DriverRewrite
If there is enough time for review, we would like to aim for inclusion in
2.6.18. The driver works nicely as a STA, and can connect to both open and
encrypted networks (we are using software-based encryption for now). We will
work towards supporting more advanced features in the future (ad-hoc, master
mode, 802.11a, ...).
Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
WARNING: /lib/modules/2.6.17-mm2/kernel/net/ieee80211/ieee80211.ko
needs unknown symbol wireless_spy_update
Someone removed the `#ifdef CONFIG_WIRELESS_EXT' from around the callsite
in net/ieee80211/ieee80211_rx.c and didn't update Kconfig appropriately.
The offending patchset seems to be 35c14b855f
which is tittled
[PATCH] ieee80211: remove unnecessary CONFIG_WIRELESS_EXT checking
After a quick look it seems that wireless_spy_update() lives in
net/core/wirless.c, and that file is only compiled if
CONFIG_WIRELESS_EXT is set. Perhaps this is Kconig work, but
in the mean time here is a reversal of the recent change.
Signed-Off-By: Horms <horms@verge.net.au>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
The ieee80211softmac_call_events function, when called with event type
IEEE80211SOFTMAC_EVENT_ASSOCIATE_TIMEOUT should pass the network as the
third parameter. This patch does that.
Signed-off-by: Joseph Jezak <josejx@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
This patch addresses the "No queue exists" messages commonly seen during
authentication and associating. These appear due to scheduling multiple
authentication attempts on the same network. To prevent this, I added a
flag to stop multiple authentication attempts by the association layer.
I also added a check to the wx handler to see if we're connecting to a
different network than the one already in progress. This scenario was
causing multiple requests on the same network because the network BSSID
was not being updated despite the fact that the ESSID changed.
Signed-off-by: Joseph Jezak <josejx@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
I recently patched softmac to enable shared key authentication. This small patch
will enable crazy or unfortunate bcm43xx users to use this new capability.
Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
In commit ba9b28d19a, routine
ieee80211softmac_capabilities was added to ieee80211softmac_io.c. As
denoted by its name, it completes the capabilities IE that is
needed in the associate and reassociate requests sent to the
AP. For at least one AP, the Linksys WRT54G V5, the capabilities
field must set the 'short preamble' bit or the AP refuses to
associate. In the commit noted above, there is a call to the
new routine from ieee80211softmac_reassoc_req, but not from
ieee80211softmac_assoc_req. This patch fixes that oversight.
As noted in the subject, v2.6.17 is affected. My bcm43xx card had been
unable to associate since I was forced to buy a new AP. I finally was
able to get a packet dump and traced the problem to the capabilities
info. Although I had heard that a patch was "floating around", I had
not seen it before 2.6.17 was released. As this bug does not affect
security and I seem to have the only AP affected by it, there should
be no problem in leaving it for 2.6.18.
Signed-Off-By: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
The current version of bcm43xx-softmac uses local routines to check
if a channel is valid. As noted in the comments, these routines do
not take any regulatory information into account. This patch converts
the code to use the equivalent routine in ieee80211, which is being
converted to know about regulatory information.
Signed-Off-By: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
this patch fixes coverity id #913. ieee80211_monitor_rx() passes the skb
to netif_rx() and we should not reference it any longer.
Signed-off-by: Eric Sesterhenn <snakebyte@gmx.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
We should preallocate IV+ICV space when encrypting the frame.
Currently no problem shows up just because dev_alloc_skb aligns the
data len to SMP_CACHE_BYTES which can be used for ICV.
Signed-off-by: Hong Liu <hong.liu@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Place the Init-vs-IRQ workaround before any card register
access, because we might not have the wireless core mapped
at all times in init. So this will result in a Machine Check
caused by a bus error.
Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Use Softmac-suggested TX ratecode:
ieee80211softmac_suggest_txrate()
Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: Michael Buesch <mb@bu3sch.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
This patch adds device IDs for Symbol LA-4123 and Global Sun Tech
GL24110P to the HostAP PLX driver.
This is not tested with real hardware, but there is no reason why it
shouldn't work.
Please test.
Signed-off-by: Faidon Liambotis <faidon@cube.gr>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
For messages prior to register_netdev(), prefer dev_printk() because
that prints out both our driver name and our [PCI | whatever] bus id.
Updates: 8139{cp,too}, b44, bnx2, cassini, {eepro,epic}100, fealnx,
hamachi, ne2k-pci, ns83820, pci-skeleton, r8169.
Signed-off-by: Jeff Garzik <jeff@garzik.org>
- const-ify pci_device_id table
- clean up pci_device_id table with PCI_DEVICE()
- don't store internal pointer in pci_device_id table,
use pci_device_id::driver_data as an integer index
- use dev_printk() for messages where eth%d prefix is unavailable
- formatting fixes
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Becker-derived drivers often have the 'io_size' member in their chip
info struct, indicating the minimum required size of the I/O resource
(usually a PCI BAR). For many situations, this number is either
constant or irrelevant (due to pci_iomap convenience behavior).
This change removes the io_size invariant member, and replaces it with a
compile-time constant.
Drivers updated: fealnx, gt96100eth, winbond-840, yellowfin
Additionally,
- gt96100eth: unused 'drv_flags' removed from gt96100eth
- winbond-840: unused struct match_info removed
- winbond-840: mark pci_id_tbl[] const, __devinitdata
Signed-off-by: Jeff Garzik <jeff@garzik.org>
When in-kernel net drivers branched from Donald Becker's vanilla driver
set, in the days before BitKeeper and git, a driver changelog was
maintained in the driver source code. These days, the kernel's
changelog is far superior and much more accurate, so the in-driver
changelogs are removed.
Another relic of the Becker/kernel split was version numbering, using
"foo-LKx.y.z" notation, resulting in weird version numbers like
"1.17b-LK1.1.9". These drivers are for older hardware, and see few
changes these days, so the version numbers were all bumped to something
more simple.
Finally, in xircom_tulip_cb specifically, an additional cleanup removes
the always-enabled CARDBUS cpp macro.
Signed-off-by: Jeff Garzik <jeff@garzik.org>
- Remove in-source changelog, it's in the global kernel history.
- convert silly and useless version to useful one
- replace invariant pci_id_tbl[]::io_size uses with EPIC_TOTAL_SIZE
- remove now-unused io_size member from pci_id_tbl[]
- current kernel style prefers dev_printk() for the rare ethernet driver
messages that cannot print an 'eth%d' prefix.
Signed-off-by: Jeff Garzik <jeff@garzik.org>
* master.kernel.org:/pub/scm/linux/kernel/git/davej/cpufreq:
Move workqueue exports to where the functions are defined.
[CPUFREQ] Misc cleanups in ondemand.
[CPUFREQ] Make ondemand sampling per CPU and remove the mutex usage in sampling path.
[CPUFREQ] Add queue_delayed_work_on() interface for workqueues.
[CPUFREQ] Remove slowdown from ondemand sampling path.
There is a code sequence where the locking is substream->self_group.lock
-> ins->scbs[index].lock
substream->self_group.lock is interrupt safe, and taken from irq context
as well (trace is snipped for brevity)
so what can happen is
cpu 0 cpu 1
user context user context
take ins->scbs[index].lock without disabling interrupts
get substream->self_group.lock (irqsafe)
try to get ins->scbs[index].lock (spins)
interrupt happens
try to get substream->self_group.lock (spins)
which is an obvious AB-BA deadlock
fix is to just take the lock with _irqsafe
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Jaroslav Kysela <perex@suse.cz>
Acked-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
mthca: initialize send and receive queue locks separately
lockdep identifies a lock by the call site of its initialization. By
initializing the send and receive queue locks in mthca_wq_init() we confuse
lockdep. It warns that that the ordered acquiry of both locks in
mthca_modify_qp() is recursive acquiry of one lock:
=============================================
[ INFO: possible recursive locking detected ]
---------------------------------------------
modprobe/1192 is trying to acquire lock:
(&wq->lock){....}, at: [<f892b4db>] mthca_modify_qp+0x60/0xa7b [ib_mthca]
but task is already holding lock:
(&wq->lock){....}, at: [<f892b4ce>] mthca_modify_qp+0x53/0xa7b [ib_mthca]
Initializing the locks separately in mthca_alloc_qp_common() stops the
warning and will let lockdep enforce proper ordering on paths that acquire
both locks.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
Cc: Roland Dreier <rolandd@cisco.com>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
I found a bug in memory hot-add code for ia64.
IA64's code has copies of pgdat's array on each node to reduce memory
access over crossing node. This array is used by NODE_DATA() macro. When
new node is hot-added, this pgdat's array should be updated and copied on
new node too.
However, I used for_each_online_node() in scatter_node_data() to copy
it. This meant its array is not copied on new node.
Because initialization of structures for new node was halfway,
so online_node_map couldn't be set at this time.
To copy arrays on new node, I changed it to check value of pgdat_list[]
which is source array of copies. I tested this patch with my Memory Hotadd
emulation on Tiger4. This patch is for 2.6.17-git20.
Signed-off-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* 'devel' of master.kernel.org:/home/rmk/linux-2.6-serial:
[SERIAL] Ensure 8250_pci quirks are not marked __devinit
[SERIAL] Convert fifosize to an unsigned int
* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6: (27 commits)
[Bluetooth] Add RFCOMM role switch support
[Bluetooth] Allow disabling of credit based flow control
[Bluetooth] Small cleanup of the L2CAP source code
[Bluetooth] Use real devices for host controllers
[Bluetooth] Add platform device for virtual and serial devices
[Bluetooth] Add automatic sniff mode support
[Bluetooth] Correct SCO buffer size on request
[Bluetooth] Add suspend/resume support to the HCI USB driver
[Bluetooth] Use raw mode for the Frontline sniffer device
[BRIDGE]: br_dump_ifinfo index fix
[ATM]: add+use poison defines
[NET]: add+use poison defines
[IOAT]: fix kernel-doc in source files
[IOAT]: fix header file kernel-doc
[TG3]: Add ipv6 TSO feature
[IPV6]: Fix ipv6 GSO payload length
[TIPC] Fixed sk_buff panic caused by tipc_link_bundle_buf (REVISED)
[NET]: Verify gso_type too in gso_segment
[IPVS]: Add sysctl documentation
[ROSE]: Try all routes when establishing a ROSE connections.
...
Jiri reports that the stop_machin kthread conversion caused his machine to
hang when suspending. Hyperthreading is apparently involved.
I don't see why that would be and I can't reproduce it. Revert to the 2.6.17
code.
Cc: "Serge E. Hallyn" <serue@us.ibm.com>
Cc: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
With recent change, if CONFIG_HOTPLUG_CPU is disabled,
register_cpu_notifier() is not exported. And it breaked moduler msr/cpuid
(msr.c was already fixed).
We need to use register_hotcpu_notifier() now in module, instead of
register_cpu_notifier().
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
