This oops was reported recently:
d:mon> e
cpu 0xd: Vector: 300 (Data Access) at [c0000000fd4c7120]
pc: d00000000076f194: .t3_l2t_get+0x44/0x524 [cxgb3]
lr: d000000000b02108: .init_act_open+0x150/0x3d4 [cxgb3i]
sp: c0000000fd4c73a0
msr: 8000000000009032
dar: 0
dsisr: 40000000
current = 0xc0000000fd640d40
paca = 0xc00000000054ff80
pid = 5085, comm = iscsid
d:mon> t
[c0000000fd4c7450] d000000000b02108 .init_act_open+0x150/0x3d4 [cxgb3i]
[c0000000fd4c7500] d000000000e45378 .cxgbi_ep_connect+0x784/0x8e8 [libcxgbi]
[c0000000fd4c7650] d000000000db33f0 .iscsi_if_rx+0x71c/0xb18
[scsi_transport_iscsi2]
[c0000000fd4c7740] c000000000370c9c .netlink_data_ready+0x40/0xa4
[c0000000fd4c77c0] c00000000036f010 .netlink_sendskb+0x4c/0x9c
[c0000000fd4c7850] c000000000370c18 .netlink_sendmsg+0x358/0x39c
[c0000000fd4c7950] c00000000033be24 .sock_sendmsg+0x114/0x1b8
[c0000000fd4c7b50] c00000000033d208 .sys_sendmsg+0x218/0x2ac
[c0000000fd4c7d70] c00000000033f55c .sys_socketcall+0x228/0x27c
[c0000000fd4c7e30] c0000000000086a4 syscall_exit+0x0/0x40
--- Exception: c01 (System Call) at 00000080da560cfc
The root cause was an EEH error, which sent us down the offload_close path in
the cxgb3 driver, which in turn sets cdev->l2opt to NULL, without regard for
upper layer driver (like the cxgbi drivers) which might have execution contexts
in the middle of its use. The result is the oops above, when t3_l2t_get attempts
to dereference L2DATA(cdev)->nentries in arp_hash right after the EEH error handler sets it to NULL.
The fix is to prevent the setting of the NULL pointer until after there are no
further users of it. The t3cdev->l2opt pointer is now converted to be an rcu
pointer and the L2DATA macro is now called under the protection of the
rcu_read_lock(). When the EEH error path:
t3_adapter_error->offload_close->cxgb3_offload_deactivate
Is exectured, setting of that l2opt pointer to NULL, is now gated on an rcu
quiescence point, preventing, allowing L2DATA callers to safely check for a NULL
pointer without concern that the underlying data will be freeded before the
pointer is dereferenced.
This has been tested by the reporter and shown to fix the reproted oops
[nhorman: fix up unitinialised variable reported by Dan Carpenter]
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Reviewed-by: Karen Xie <kxie@chelsio.com>
Cc: stable@kernel.org
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
In an enclosure model where there are chaining expanders to a large body
of storage, it was discovered that libsas, responding to a broadcast
event change, would only revalidate the domain of first child expander
in the list.
The issue is that the pointer value to the discovered source device was
used to break out of the loop, rather than the content of the pointer.
This still remains non-compliant as the revalidate domain code is
supposed to loop through all child expanders, and not stop at the first
one it finds that reports a change count. However, the design of this
routine does not allow multiple device discoveries and that would be a
more complicated set of patches reserved for another day. We are fixing
the glaring bug rather than refactoring the code.
Signed-off-by: Mark Salyzyn <msalyzyn@us.xyratex.com>
Cc: stable@kernel.org
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
xz_dec_run() could incorrectly return XZ_BUF_ERROR if all of the
following was true:
- The caller knows how many bytes of output to expect and only provides
that much output space.
- When the last output bytes are decoded, the caller-provided input
buffer ends right before the LZMA2 end of payload marker. So LZMA2
won't provide more output anymore, but it won't know it yet and thus
won't return XZ_STREAM_END yet.
- A BCJ filter is in use and it hasn't left any unfiltered bytes in the
temp buffer. This can happen with any BCJ filter, but in practice
it's more likely with filters other than the x86 BCJ.
This fixes <https://bugzilla.redhat.com/show_bug.cgi?id=735408> where
Squashfs thinks that a valid file system is corrupt.
This also fixes a similar bug in single-call mode where the uncompressed
size of a block using BCJ + LZMA2 was 0 bytes and caller provided no
output space. Many empty .xz files don't contain any blocks and thus
don't trigger this bug.
This also tweaks a closely related detail: xz_dec_bcj_run() could call
xz_dec_lzma2_run() to decode into temp buffer when it was known to be
useless. This was harmless although it wasted a minuscule number of CPU
cycles.
Signed-off-by: Lasse Collin <lasse.collin@tukaani.org>
Cc: stable <stable@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* git://github.com/davem330/net: (27 commits)
xfrm: Perform a replay check after return from async codepaths
fib:fix BUG_ON in fib_nl_newrule when add new fib rule
ixgbe: fix possible null buffer error
tg3: fix VLAN tagging regression
net: pxa168: Fix build errors by including interrupt.h
netconsole: switch init_netconsole() to late_initcall
gianfar: Fix overflow check and return value for gfar_get_cls_all()
ppp_generic: fix multilink fragment MTU calculation (again)
GRETH: avoid overwrite IP-stack's IP-frags checksum
GRETH: RX/TX bytes were never increased
ipv6: fix a possible double free
b43: Fix beacon problem in ad-hoc mode
Bluetooth: add support for 2011 mac mini
Bluetooth: Add MacBookAir4,1 support
Bluetooth: Fixed BT ST Channel reg order
r8169: do not enable the TBI for anything but the original 8169.
r8169: remove erroneous processing of always set bit.
r8169: fix WOL setting for 8105 and 8111evl
r8169: add MODULE_FIRMWARE for the firmware of 8111evl
r8169: fix the reset setting for 8111evl
...
* 'for-linus' of git://git.kernel.dk/linux-block:
floppy: use del_timer_sync() in init cleanup
blk-cgroup: be able to remove the record of unplugged device
block: Don't check QUEUE_FLAG_SAME_COMP in __blk_complete_request
mm: Add comment explaining task state setting in bdi_forker_thread()
mm: Cleanup clearing of BDI_pending bit in bdi_forker_thread()
block: simplify force plug flush code a little bit
block: change force plug flush call order
block: Fix queue_flag update when rq_affinity goes from 2 to 1
block: separate priority boosting from REQ_META
block: remove READ_META and WRITE_META
xen-blkback: fixed indentation and comments
xen-blkback: Don't disconnect backend until state switched to XenbusStateClosed.
When a malformed loglevel value (for example "${abc}") is passed on the
kernel cmdline, the loglevel itself is being set to 0.
That then suppresses all following messages, including all the errors
and crashes caused by other malformed cmdline options. This could make
debugging process quite tricky.
This patch leaves the previous value of loglevel if the new value is
incorrect and reports an error code in this case.
Signed-off-by: Alexander Sverdlin <alexander.sverdlin@sysgo.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is modeled after the smaps code.
It detects transparent hugepages and then does a single gather_stats()
for the page as a whole. This has two benifits:
1. It is more efficient since it does many pages in a single shot.
2. It does not have to break down the huge page.
Signed-off-by: Dave Hansen <dave@linux.vnet.ibm.com>
Acked-by: Hugh Dickins <hughd@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
gather_pte_stats() does a number of checks on a target page
to see whether it should even be considered for statistics.
This breaks that code out in to a separate function so that
we can use it in the transparent hugepage case in the next
patch.
Signed-off-by: Dave Hansen <dave@linux.vnet.ibm.com>
Acked-by: Hugh Dickins <hughd@google.com>
Reviewed-by: Christoph Lameter <cl@gentwo.org>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We need to teach the numa_maps code about transparent huge pages. The
first step is to teach gather_stats() that the pte it is dealing with
might represent more than one page.
Note that will we use this in a moment for transparent huge pages since
they have use a single pmd_t which _acts_ as a "surrogate" for a bunch
of smaller pte_t's.
I'm a _bit_ unhappy that this interface counts in hugetlbfs page sizes
for hugetlbfs pages and PAGE_SIZE for normal pages. That means that to
figure out how many _bytes_ "dirty=1" means, you must first know the
hugetlbfs page size. That's easier said than done especially if you
don't have visibility in to the mount.
But, that's probably a discussion for another day especially since it
would change behavior to fix it. But, just in case anyone wonders why
this patch only passes a '1' in the hugetlb case...
Signed-off-by: Dave Hansen <dave@linux.vnet.ibm.com>
Acked-by: Hugh Dickins <hughd@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When asyncronous crypto algorithms are used, there might be many
packets that passed the xfrm replay check, but the replay advance
function is not called yet for these packets. So the replay check
function would accept a replay of all of these packets. Also the
system might crash if there are more packets in async processing
than the size of the anti replay window, because the replay advance
function would try to update the replay window beyond the bounds.
This pach adds a second replay check after resuming from the async
processing to fix these issues.
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
add new fib rule can cause BUG_ON happen
the reproduce shell is
ip rule add pref 38
ip rule add pref 38
ip rule add to 192.168.3.0/24 goto 38
ip rule del pref 38
ip rule add to 192.168.3.0/24 goto 38
ip rule add pref 38
then the BUG_ON will happen
del BUG_ON and use (ctarget == NULL) identify whether this rule is unresolved
Signed-off-by: Gao feng <gaofeng@cn.fujitsu.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When no floppy is found the module code can be released while a timer
function is pending or about to be executed.
CPU0 CPU1
floppy_init()
timer_softirq()
spin_lock_irq(&base->lock);
detach_timer();
spin_unlock_irq(&base->lock);
-> Interrupt
del_timer();
return -ENODEV;
module_cleanup();
<- EOI
call_timer_fn();
OOPS
Use del_timer_sync() to prevent this.
Signed-off-by: Carsten Emde <C.Emde@osadl.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The bug is we're not able to remove the device from blkio cgroup's
per-device control files if it gets unplugged.
To reproduce the bug:
# mount -t cgroup -o blkio xxx /cgroup
# cd /cgroup
# echo "8:0 1000" > blkio.throttle.read_bps_device
# unplug the device
# cat blkio.throttle.read_bps_device
8:0 1000
# echo "8:0 0" > blkio.throttle.read_bps_device
-bash: echo: write error: No such device
After patching, the device removal will succeed.
Thanks for the comments of Paul, Zefan, and Vivek.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Paul Menage <paul@paulmenage.org>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
It seems that at least one PPC machine would occasionally give a (valid) 0 as
the return value from dma_map, this caused the ixgbe code to not work
correctly. A fix is pending in the PPC tree to not return 0 from dma map, but
we can also fix the driver to make sure we don't mess up in other arches as
well.
This patch is applicable to all current stable kernels.
Ref: https://bugzilla.redhat.com/show_bug.cgi?id=683611
Reported-by: Neil Horman <nhorman@redhat.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
CC: Alexander Duyck <alexander.h.duyck@intel.com>
CC: stable@kernel.org
Tested-by: Thadeu Lima de Souza Cascardo <cascardo@linux.vnet.ibm.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
commit 92cd3a17ce
tg3: Simplify tx bd assignments
broke VLAN tagging on outbound packets.
It ifdef'ed BCM_KERNEL_SUPPORTS_8021Q, but this
is not set anywhere. So vlan never gets set, and
all packets are sent with vlan=0.
v2: We can just remove the test. vlan_tx_tag_present
is valid regardless of whether the 802.1q module
is built.
Tested on BCM5721 rev 11.
Signed-off-by: Kasper Pedersen <kernel@kasperkp.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
* 'fixes' of git://git.linaro.org/people/arnd/arm-soc:
mach-integrator: fix VGA base regression
arm/dt: Tegra: Update SDHCI nodes to match bindings
ARM: EXYNOS4: fix incorrect pad configuration for keypad row lines
ARM: SAMSUNG: fix to prevent declaring duplicated
ARM: SAMSUNG: fix watchdog reset issue with clk_get()
ARM: S3C64XX: Remove un-used code backlight code on SMDK6410
ARM: EXYNOS4: restart clocksource while system resumes
ARM: EXYNOS4: Fix routing timer interrupt to offline CPU
ARM: EXYNOS4: Fix return type of local_timer_setup()
ARM: EXYNOS4: Fix wrong pll type for vpll
ARM: Dove: fix second SPI initialization call
After commit c5f5c4db39 ("staging: zcache: fix crash on high memory
swap") cleancache crashes on the first successful get. This was caused
by a remaining virt_to_page() call in zcache_pampd_get_data_and_free()
that only gets run in the cleancache path.
The patch converts the virt_to_page() to struct page casting like was
done for other instances in c5f5c4db39.
Signed-off-by: Seth Jennings <sjenning@linux.vnet.ibm.com>
Tested-By: Valdis Kletnieks <valdis.kletnieks@vt.edu>
Acked-by: Dan Magenheimer <dan.magenheimer@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The changes introduced in commit
cc22b4c185
"ARM: set vga memory base at run-time"
Makes the Integrator/AP freeze completely. I appears that
this is due to the VGA base address being assigned at PCI
init time, while this base is needed earlier than that.
Moving the initialization of the base address to the
.map_io function solves this problem.
Cc: Rob Herring <rob.herring@calxeda.com>
Cc: Nicolas Pitre <nicolas.pitre@linaro.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
Acked-by: Rob Herring <rob.herring@calxeda.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
The bindings were recently updated to have separate properties for each
type of GPIO. Update the Device Tree source to match that.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Commit a6b7a40786 removed
linux/interrupt.h from netdevice.h. This fixes below build failure
drivers/net/pxa168_eth.c: In function 'pxa168_eth_collect_events':
drivers/net/pxa168_eth.c:866: error: 'IRQ_NONE' undeclared (first use in this function)
drivers/net/pxa168_eth.c:866: error: (Each undeclared identifier is reported only once
drivers/net/pxa168_eth.c:866: error: for each function it appears in.)
drivers/net/pxa168_eth.c: At top level:
drivers/net/pxa168_eth.c:913: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'pxa168_eth_int_handler'
drivers/net/pxa168_eth.c: In function 'pxa168_eth_open':
drivers/net/pxa168_eth.c:1133: error: implicit declaration of function 'request_irq'
drivers/net/pxa168_eth.c:1133: error: 'pxa168_eth_int_handler' undeclared (first use in this function)
drivers/net/pxa168_eth.c:1134: error: 'IRQF_DISABLED' undeclared (first use in this function)
drivers/net/pxa168_eth.c:1160: error: implicit declaration of function 'free_irq'
Signed-off-by: Tanmay Upadhyay <tanmay.upadhyay@einfochips.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 88491d8(drivers/net: Kconfig & Makefile cleanup) causes a
regression that netconsole does not work if netconsole and network
device driver are build into kernel, because netconsole is linked
before network device driver.
Andrew Morton suggested to fix this with initcall ordering.
Fixes it by switching init_netconsole() to late_initcall.
Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This function may currently fill one entry beyond the end of the
array it is given. It also doesn't return an error code in case
it does detect overflow.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When using MLPPP, the maximum size of a fragment is incorrectly
calculated with an offset of -2.
This patch reverses the changes in the patch found here:
http://marc.info/?l=linux-netdev&m=123541324010539&w=2
The value of hdrlen includes the size of both the 2-byte PPP protocol
field and the 2- or 4-byte multilink header (2+4=6 for long sequence
numbers, 2+2=4 for short sequence numbers). Section 2 of RFC1661 says
that the MRU that is negotiated (i.e., the MTU of the sending system)
includes only the PPP payload but not the protocol field, thus the
correct MTU should be the link's MTU minus the multilink header (mtu -
(hdrlen-2)).
The incorrect calculation causes Linux to fragment packets to a size two
bytes smaller than the allowed MTU. While not technically illegal, this
behaviour confounds MRU-tuning to avoid PPP-layer fragmentation.
Signed-off-by: Henry Wong <henry@stuffedcow.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The GRETH GBIT core does not do checksum offloading for IP
segmentation. This patch adds a check in the xmit function to
determine if the stack has calculated the checksum for us.
Signed-off-by: Daniel Hellstrom <daniel@gaisler.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When calling snmp6_alloc_dev fails, the snmp6 relevant memory
are freed by snmp6_alloc_dev. Calling in6_dev_finish_destroy
will free these memory twice.
Double free will lead that undefined behavior occurs.
Signed-off-by: Roy Li <rongqing.li@windriver.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix a crash/BUG_ON in the clone ioctl due to insufficient reservation. We
need to reserve space for:
- adjusting the old extent (possibly splitting it)
- adding the new extent
- updating the inode
Signed-off-by: Sage Weil <sage@newdream.net>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
* git://git.samba.org/sfrench/cifs-2.6:
cifs: Fix broken sec=ntlmv2/i sec option (try #2)
Fix the conflict between rwpidforward and rw mount options
CIFS: Fix ERR_PTR dereference in cifs_get_root
cifs: fix possible memory corruption in CIFSFindNext
The enable function was using the global timeout variable for local operations.
This resulted in the value of the global variable being corrupted, thus
breaking the code.
Signed-off-by: John Crispin <blogic@openwrt.org>
Signed-off-by: Thomas Langer <thomas.langer@lantiq.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
Cc: linux-watchdog@vger.kernel.org
Cc: linux-mips@linux-mips.org
On platforms with no iCRU support don't print two, (possibly conflicting),
"NMI occurred" messages when the firmware is unable to source the NMI.
Please note that one of the enhancements to the v1.3.0 hpwdt driver is to panic and allow
KDUMP to succeed even on NMIs that are unknown to the platform firmware.
Signed-off-by: Naga Chumbalkar <nagananda.chumbalkar@hp.com>
Reviewed-by: Thomas Mingarelli <thomas.mingarelli@hp.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
Use the passed watchdog_device instead of the static global variable when
testing and setting the status in watchdog_ping, watchdog_start, and
watchdog_stop. Note that the callers of these functions are actually
passing the static global variable.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
Fix sec=ntlmv2/i authentication option during mount of Samba shares.
cifs client was coding ntlmv2 response incorrectly.
All that is needed in temp as specified in MS-NLMP seciton 3.3.2
"Define ComputeResponse(NegFlg, ResponseKeyNT, ResponseKeyLM,
CHALLENGE_MESSAGE.ServerChallenge, ClientChallenge, Time, ServerName)
as
Set temp to ConcatenationOf(Responserversion, HiResponserversion,
Z(6), Time, ClientChallenge, Z(4), ServerName, Z(4)"
is MsvAvNbDomainName.
For sec=ntlmsspi, build_av_pair is not used, a blob is plucked from
type 2 response sent by the server to use in authentication.
I tested sec=ntlmv2/i and sec=ntlmssp/i mount options against
Samba (3.6) and Windows - XP, 2003 Server and 7.
They all worked.
Signed-off-by: Shirish Pargaonkar <shirishpargaonkar@gmail.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
Both these options are started with "rw" - that's why the first one
isn't switched on even if it is specified. Fix this by adding a length
check for "rw" option check.
Cc: <stable@kernel.org>
Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru>
Signed-off-by: Steve French <sfrench@us.ibm.com>
move it to the beginning of the loop.
Signed-off-by: Pavel Shilovsky <piastryyy@gmail.com>
Reviewed-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
The name_len variable in CIFSFindNext is a signed int that gets set to
the resume_name_len in the cifs_search_info. The resume_name_len however
is unsigned and for some infolevels is populated directly from a 32 bit
value sent by the server.
If the server sends a very large value for this, then that value could
look negative when converted to a signed int. That would make that
value pass the PATH_MAX check later in CIFSFindNext. The name_len would
then be used as a length value for a memcpy. It would then be treated
as unsigned again, and the memcpy scribbles over a ton of memory.
Fix this by making the name_len an unsigned value in CIFSFindNext.
Cc: <stable@kernel.org>
Reported-by: Darren Lavender <dcl@hppine99.gbr.hp.com>
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
* 'irq-fixes-for-linus' of git://tesla.tglx.de/git/linux-2.6-tip:
x86, iommu: Mark DMAR IRQ as non-threaded
genirq: Make irq_shutdown() symmetric vs. irq_startup again
* 'for-linus' of git://github.com/chrismason/linux:
Btrfs: only clear the need lookup flag after the dentry is setup
BTRFS: Fix lseek return value for error
Btrfs: don't change inode flag of the dest clone file
Btrfs: don't make a file partly checksummed through file clone
Btrfs: fix pages truncation in btrfs_ioctl_clone()
btrfs: fix d_off in the first dirent
When a xHC host is unable to handle isochronous transfer in the
interval, it reports a Missed Service Error event and skips some tds.
Currently xhci driver handles MSE event in the following ways:
1. When encounter a MSE event, set ep->skip flag, update event ring
dequeue pointer and return.
2. When encounter the next event on this ep, the driver will run the
do-while loop, fetch td from ep's td_list to find the td
corresponding to this event. All tds missed are marked as short
transfer(-EXDEV).
The do-while loop will end in two ways:
1. If the td pointed by the event trb is found;
2. If the ep ring's td_list is empty.
However, if a buggy HW reports some unpredicted event (for example, an
overrun event following a MSE event while the ep ring is actually not
empty), the driver will never find the td, and it will loop until the
td_list is empty.
Unfortunately, the spinlock is dropped when give back a urb in the
do-while loop. During the spinlock released period, the class driver
may still submit urbs and add tds to the td_list. This may cause
disaster, since the td_list will never be empty and the loop never ends,
and the system hangs.
To fix this, count the number of TDs on the ep ring before skipping TDs,
and quit the loop when skipped that number of tds. This guarantees the
do-while loop will end after certain number of cycles, and driver will
not be trapped in an infinite loop.
Signed-off-by: Andiry Xu <andiry.xu@amd.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Sometimes, when a USB 3.0 device is disconnected, the Intel Panther
Point xHCI host controller will report a link state change with the
state set to "SS.Inactive". This causes the xHCI host controller to
issue a warm port reset, which doesn't finish before the USB core times
out while waiting for it to complete.
When the warm port reset does complete, and the xHC gives back a port
status change event, the xHCI driver kicks khubd. However, it fails to
set the bit indicating there is a change event for that port because the
logic in xhci-hub.c doesn't check for the warm port reset bit.
After that, the warm port status change bit is never cleared by the USB
core, and the xHC stops reporting port status change bits. (The xHCI
spec says it shouldn't report more port events until all change bits are
cleared.) This means any port changes when a new device is connected
will never be reported, and the port will seem "dead" until the xHCI
driver is unloaded and reloaded, or the computer is rebooted. Fix this
by making the xHCI driver set the port change bit when a warm port reset
change bit is set.
A better solution would be to make the USB core handle warm port reset
in differently, merging the current code with the standard port reset
code that does an incremental backoff on the timeout, and tries to
complete the port reset two more times before giving up. That more
complicated fix will be merged next window, and this fix will be
backported to stable.
This should be backported to kernels as old as 3.0, since that was the
first kernel with commit a11496ebf3 ("xHCI: warm reset support").
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Cc: stable@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
