The problems lie in the types used for some inotify interfaces, both at the kernel level and at the glibc level. This mail addresses the kernel problem. I will follow up with some suggestions for glibc changes.
For the sys_inotify_rm_watch() interface, the type of the 'wd' argument is
currently 'u32', it should be '__s32' . That is Robert's suggestion, and
is consistent with the other declarations of watch descriptors in the
kernel source, in particular, the inotify_event structure in
include/linux/inotify.h:
struct inotify_event {
__s32 wd; /* watch descriptor */
__u32 mask; /* watch mask */
__u32 cookie; /* cookie to synchronize two events */
__u32 len; /* length (including nulls) of name */
char name[0]; /* stub for possible name */
};
The patch makes the changes needed for inotify_rm_watch().
Signed-off-by: Michael Kerrisk <mtk.manpages@googlemail.com> Cc: Robert Love <rlove@google.com> Cc: Vegard Nossum <vegard.nossum@gmail.com> Cc: Ulrich Drepper <drepper@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
_boot_cpu_pda is the actual pda used by the boot cpu and is definitely
not "__read_mostly" and ended up polluting the read mostly section with
writes. This bug caused regression of about 8-10% on certain syscall
intensive workloads.
A kernel for a 64bit x86 system should always contain the swiotlb code
in case it is booted on a machine without any hardware IOMMU supported
by the kernel and more than 4GB of RAM. This patch changes Kconfig to
always compile swiotlb into the kernel for x86_64.
this driver can't handle (of course) any brdige class devices. So we
now are just active on one specific bridge which should be only the
isp1761 chip behind a PLX bridge.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Tested-by: Karl Bongers <kblists08@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
After the last "package" of firmware data is sent to the device, we dereference
NULL pointer (on access to rec->addr). Fix it.
Reported--by: David Flatz <david@upcs.at> Tested-by: David Flatz <david@upcs.at> Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com> Cc: David Woodhouse <dwmw2@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
5b7dba4: sched_clock: prevent scd->clock from moving backwards
which had to be reverted due to s2ram hangs:
ca7e716: Revert "sched_clock: prevent scd->clock from moving backwards"
... this time with resume restoring GTOD later in the sequence
taken into account as well.
The "timekeeping_suspended" flag is not very nice but we cannot call into
GTOD before it has been properly resumed and the scheduler will run very
early in the resume sequence.
Previously we allocate Rx SKB with GFP_ATOMIC flag. This is because we need
to hold a spinlock to protect the two rx_used and rx_free lists operation
in the rxq.
After spliting the rx_used delete and rx_free insert into two operations,
we don't require the skb allocation in an atomic context any more (the
function itself is scheduled in a workqueue).
This patch should fix the "iwlagn: Can not allocate SKB buffers" warning
we see recently.
Signed-off-by: Zhu Yi <yi.zhu@intel.com> Acked-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
In the multiple device case we need to re-arm the completion and protect
against concurrent self-tests. The printk from the test callback is
removed as it can arbitrarily delay completion of the test.
Cc: Maciej Sosnowski <maciej.sosnowski@intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Always increase the error count when I/O on a leg of a mirror fails.
The error count is used to decide whether to select an alternative
mirror leg. If the target doesn't use the "handle_errors" feature, the
error count is not updated and the bio can get requeued forever by the
read callback.
Fix it by increasing error_count before the handle_errors feature
checking.
Signed-off-by: Milan Broz <mbroz@redhat.com> Signed-off-by: Jonathan Brassow <jbrassow@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
In create_log_context function, dm_io_client_destroy function needs
to be called, when memory allocation of disk_header, sync_bits and
recovering_bits failed, but dm_io_client_destroy is not called.
Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Acked-by: Jonathan Brassow <jbrassow@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The devcgroup_inode_permission() hook in the devices whitelist cgroup has
always bypassed access checks on fifos. But the mknod hook did not. The
devices whitelist is only about block and char devices, and fifos can't
even be added to the whitelist, so fifos can't be created at all except by
tasks which have 'a' in their whitelist (meaning they have access to all
devices).
Fix the behavior by bypassing access checks to mkfifo.
Signed-off-by: Serge E. Hallyn <serue@us.ibm.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Pavel Emelyanov <xemul@openvz.org> Cc: Paul Menage <menage@google.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: James Morris <jmorris@namei.org> Reported-by: Daniel Lezcano <dlezcano@fr.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The race is calling cgroup_clone() while umounting the ns cgroup subsys,
and thus cgroup_clone() might access invalid cgroup_fs, or kill_sb() is
called after cgroup_clone() created a new dir in it.
The BUG I triggered is BUG_ON(root->number_of_cgroups != 1);
This is pointer list and if we dereference an uninitialized pointer
later this results in a kernel crash at boot. Happens typically after
3-5 hours of rebooting.
Under special circumstances the IOMMU does not reset the head and tail
pointer of its command ringbuffer to zero when the command base is
written. This causes the IOMMU to fetch random memory and executes it as
an command. Since these commands are likely illegal IOMMU stops fetching
further commands including IOTLB flushes. This leads to completion wait
errors at boot and in some cases to data corruption and kernel crashes.
starfire napi ->poll() handler can return work == weight after calling
netif_rx_complete() (if there is no more work). It is illegal and this
patch fixes it.
Reported-by: Alexander Huemer <alexander.huemer@sbg.ac.at> Tested-by: Alexander Huemer <alexander.huemer@sbg.ac.at> Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Cc: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
flush_tlb_mm's "optimized" uniprocessor case of allocating a new
context for userspace is exposing a race where we can suddely return
to a syscall with the protection id and space id out of sync, trapping
on the next userspace access.
Fix problem that deleting multiple logical drives could cause a panic.
It fixes a panic which can be easily reproduced in the following way: Just
create several "arrays," each with multiple logical drives via hpacucli,
then delete the first array, and it will blow up in deregister_disk(), in
the call to get_host() when it tries to dig the hba pointer out of a NULL
queue pointer.
The problem has been present since my code to make rebuild_lun_table
behave better went in.
Signed-off-by: Stephen M. Cameron <scameron@beardog.cca.cpqcorp.net> Signed-off-by: Jens Axboe <jens.axboe@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
but fail with tons of "aac_srb: aac_fib_send failed with status: 8195"
instead. This patch disables the quirk introduced in the change cited
above for those two controllers again.
[thenzl: added 2120S Controller] Signed-off-by: Gernot Hillier <gernot.hillier@siemens.com> Signed-off-by: Tomas Henzl <thenzl@redhat.com> Acked-by: Matt Domsch <Matt_Domsch@dell.com> Cc: AACRAID list <aacraid@adaptec.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Calling crq_queue_create could lead to the creation of a rport. We
need to set up everything before creating a rport. This moves
crq_queue_create to the end of initialization to avoid a race which
causes an oops if lost.
Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Reported-by: Olaf Hering <olh@suse.de> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
There will be a Oops or frequent underrun messages when playing music with
omap soc driver, this is because a data region is incorretly sized, other data
region will be overwriten when writing to this data region.
Signed-off-by: Stanley Miao <stanley.miao@windriver.com> Acked-by: Jarkko Nikula <jarkko.nikula@nokia.com> Signed-off-by: Takashi Iwai <tiwai@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When we read the write-intent-bitmap off the device, we currently
read a whole number of pages.
When PAGE_SIZE is 4K, this works due to the alignment we enforce
on the superblock and bitmap.
When PAGE_SIZE is 64K, this case read past the end-of-device
which causes an error.
When we write the superblock, we ensure to clip the last page
to just be the required size. Copy that code into the read path
to just read the required number of sectors.
Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
As part of the ioat_dma self-test it performs a printk from a completion
callback. Depending on the system console configuration this output can
take longer than a millisecond causing the self-test to fail. Introduce a
completion with a generous timeout to mitigate this failure.
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
There is a possibility to have two devices registered with the same id.
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Mapping the destination multiple times is a misuse of the dma-api.
Since the destination may be reused as a source, ensure that it is only
mapped once and that it is mapped bidirectionally. This appears to add
ugliness on the unmap side in that it always reads back the destination
address from the descriptor, but gcc can determine that dma_unmap is a
nop and not emit the code that calculates its arguments.
Cc: Saeed Bishara <saeed@marvell.com> Acked-by: Yuri Tikhonov <yur@emcraft.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This BUG_ON really shouldn't trigger, but if it does, as on my machine,
it leaves you wondering what happened because you won't see it. Let's
instead leak a bit of state and memory and at least make it possible to
report it to the kerneloops project to track it.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Cc: François Valenduc <francois.valenduc@tvcablenet.be> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
There is an error in rh_alloc_fixed() of the Remote Heap code:
If there is at least one free block blk won't be NULL at the end of the
search loop, so -ENOMEM won't be returned and the else branch of
"if (bs == s || be == e)" will be taken, corrupting the management
structures.
Signed-off-by: Guillaume Knispel <gknispel@proformatique.com> Acked-by: Timur Tabi <timur@freescale.com> Signed-off-by: Kumar Gala <galak@kernel.crashing.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2.6.26(.x, cannot remember) could handle the microSD card in my Nokia
3109c attached via USB as mass storage, 2.6.27(.x, up to and included
2.6.27.8) cannot. Please find the attached patch which fixes this
regression, and a copy of /proc/bus/usb/devices with my phone plugged in
running with this patch on Frugalware.
Fix a bug specific to highspeed mode in the recently updated RNDIS
support: it wasn't setting up the high speed notification endpoint,
which prevented high speed RNDIS links from working.
This replaces zone->lru_lock in setup_per_zone_pages_min() with zone->lock.
There seems to be no need for the lru_lock anymore, but there is a need for
zone->lock instead, because that function may call move_freepages() via
setup_zone_migrate_reserve().
Don't overflow the 16-character fb_fix_screeninfo id string (fixes some
console erasing and blanking artifacts). Have the ID default to "Unknown"
on machines with no built-in video and no nubus devices. Check for
fb_alloc_cmap failure.
On "/etc/init.d/capiutils stop", this oops happened.
The oops happens on reading /proc/capi/controllers because
capi_ctrl->procinfo is called for the wrongly not unregistered
controller, which points to b1isa_procinfo(), which was removed on
module unload.
b1isa_exit() did not call b1isa_remove() for its controllers because
io[0] == 0 on module unload despite having been 0x340 on module load.
Besides, just removing the controllers that where added on module
load time and not those that were added later via b1isa_add_card() is
wrong too - the place where all added cards are found is isa_dev[].
relevant dmesg lines:
[ 0.000000] Linux version 2.6.27.4 (w@shubashi) (gcc version 4.3.2 (Debian 4.3.2-1) ) #3 Thu Oct 30 16:49:03 CET 2008
[ 67.403555] CAPI Subsystem Rev 1.1.2.8
[ 68.529154] capifs: Rev 1.1.2.3
[ 68.563292] capi20: Rev 1.1.2.7: started up with major 68 (middleware+capifs)
[ 77.026936] b1: revision 1.1.2.2
[ 77.049992] b1isa: revision 1.1.2.3
[ 77.722655] kcapi: Controller [001]: b1isa-340 attached
[ 77.722671] b1isa: AVM B1 ISA at i/o 0x340, irq 5, revision 255
[ 81.272669] b1isa-340: card 1 "B1" ready.
[ 81.272683] b1isa-340: card 1 Protocol: DSS1
[ 81.272689] b1isa-340: card 1 Linetype: point to multipoint
[ 81.272695] b1isa-340: B1-card (3.11-03) now active
[ 81.272702] kcapi: card [001] "b1isa-340" ready.
Fix a regression reported by Max Kellermann whereby kernel profiling
showed that his clients were spending 45% of their time in
rpcauth_lookup_credcache.
It turns out that although his processes had identical uid/gid/groups,
generic_match() was failing to detect this, because the task->group_info
pointers were not shared. This again lead to the creation of a huge number
of identical credentials at the RPC layer.
The regression is fixed by comparing the contents of task->group_info
if the actual pointers are not identical.
According to http://bugzilla.kernel.org/show_bug.cgi?id=12206, Freecom
FireWire Hard Drive 1TB reports max_rom=2 but returns garbage if block
read requests are used to read the config ROM. Force max_rom=0 to limit
them to quadlet read requests.
Reported-by: Christian Mueller <cm1@mumac.de> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
There is a DMA map/ unmap imbalance whenever a block write request
packet is sent and then dequeued with ohci_cancel_packet. The latter
may happen frequently if the AR resp tasklet is executed before the AT
req tasklet for the same transaction.
Add the missing dma_unmap_single. This fixes
https://bugzilla.redhat.com/show_bug.cgi?id=475156
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Cc: Kadianakis George <desnacked@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
So I dug deeper into the DMA problems I had with iwlagn and a kind soul
helped me in that he said something about pci-e alignment and mentioned
the iwl_rx_allocate function to check for crossing 4KB boundaries. Since
there's 8KB A-MPDU support, crossing 4k boundaries didn't seem like
something the device would fail with, but when I looked into the
function for a minute anyway I stumbled over this little gem:
Clearly, that is a totally bogus check, one would hope the compiler
removes it entirely. (Think about it)
After fixing it, I obviously ran into it, nothing guarantees the
alignment the way you want it, because of the way skbs and their
headroom are allocated. I won't explain that here nor double-check that
I'm right, that goes beyond what most of the CC'ed people care about.
So then I came up with the patch below, and so far my system has
survived minutes with 64K pages, when it would previously fail in
seconds. And I haven't seen a single instance of the TX bug either. But
when you see the patch it'll be pretty obvious to you why.
This should fix the following reported kernel bugs:
I haven't checked if there are any elsewhere, but I suppose RHBZ will
have a few instances too...
I'd like to ask anyone who is CC'ed (those are people I know ran into
the bug) to try this patch.
I am convinced that this patch is correct in spirit, but I haven't
understood why, for example, there are so many unmap calls. I'm not
entirely convinced that this is the only bug leading to the TX reply
errors.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
For the console, there is a 1:1 mapping of glyphs which cannot be found
in the current font. This seems to be meant as a kind of 'emergency
fallback' for fonts without unicode mapping which otherwise would
display nothing readable on the screen.
At the moment it affects all chars for which no substitution character
is defined. In particular this means that for all chars (>= 128) where
there is no iso88591-1/unicode character (e.g. control character area)
you'll get the very strange 1:1 mapping of the (cp437) graphics card
glyphs.
I'm pretty sure that the 1:1 mapping should only affect strict ASCII
code characters, i.e. chars < 128.
The patch limits the mapping as it probably was meant anyway.
There is a major bug in the cp437 to unicode translation table. Char
0x7c is mapped to U+00a5 which is the Yen sign and wrong. The right
mapping is U+00a6 (broken bar).
Furthermore, a mapping for U+00b4 (a widely used character) is missing
even though easily possible.
The patch fixes these, as well as it provides a few other useful
mappings.
The changes are as follows:
0x0f (enhancement) enables a sort of currency symbol
0x27 (bug) enables a sort of acute accent which is a widely used character
0x44 (enhancement) enables a sort of icelandic capital letter eth
0x7c (major bug) corrects mapping
0xeb (enhancement) enables a sort of icelandic small letter eth
0xee (enhancement) enables a sort of math 'element of'
Signed-off-by: Ingo Brueckl <ib@wupperonline.de> Acked-by: H. Peter Anvin <hpa@zytor.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The bridge device always causes a warning because when it is first created
it has the no checksum flag set along with all the segmentation/fragmentation
offload bits. The code in register_netdevice incorrectly checks for only
hardware checksum bit and ignores no checksum bit.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Cc: David Miller <davem@davemloft.net> Cc: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch cleans uCode key table bit map iwl_clear_stations_table
since all stations are cleared also the key table must be.
Since the keys are not removed properly on suspend by mac80211
this may result in exhausting key table on resume leading
to memory corruption during removal
This patch also fixes a memory corruption problem reported in
http://marc.info/?l=linux-wireless&m=122641417231586&w=2 and tracked in
http://bugzilla.kernel.org/show_bug.cgi?id=12040.
When the key is removed a second time the offset is set to 255 - this
index is not valid for the ucode_key_table and corrupts the eeprom pointer
(which is 255 bits from ucode_key_table).
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Reported-by: Carlos R. Mafra <crmafra2@gmail.com> Reported-by: Lukas Hejtmanek <xhejtman@ics.muni.cz> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
So we need to omit the check of these filter lists when receiving RTR
CAN frames.
Signed-off-by: Oliver Hartkopp <oliver@hartkopp.net> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Due to a wrong safety check in af_can.c it was not possible to filter
for SFF frames with a specific CAN identifier without getting the
same selected CAN identifier from a received EFF frame also.
This fix has a minimum (but user visible) impact on the CAN filter
API and therefore the CAN version is set to a new date.
Indeed the 'old' API is still working as-is. But when now setting
CAN_(EFF|RTR)_FLAG in can_filter.can_mask you might get less traffic
than before - but still the stuff that you expected to get for your
defined filter ...
Thanks to Kurt Van Dijck for pointing at this issue and for the review.
Signed-off-by: Oliver Hartkopp <oliver@hartkopp.net> Acked-by: Kurt Van Dijck <kurt.van.dijck@eia.be> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
During a reset, releasing the swflag after it failed to be acquired would
cause a double unlock of the mutex. Instead, test whether acquisition of
the swflag was successful and if not, do not release the swflag. The reset
must still be done to bring the device to a quiescent state.
This resolves [BUG 12200] BUG: bad unlock balance detected! e1000e
http://bugzilla.kernel.org/show_bug.cgi?id=12200
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The last patch to lib/idr.c caused a bug if idr_get_new_above() was
called on an empty idr.
Usually, nodes stay on the same layer. New layers are added to the top
of the tree.
The exception is idr_get_new_above() on an empty tree: In this case, the
new root node is first added on layer 0, then moved upwards. p->layer
was not updated.
As usual: You shall never rely on the source code comments, they will
only mislead you.
VMI initialiation can relocate the fixmap, causing early_ioremap to
malfunction if it is initialized before the relocation. To fix this,
VMI activation is split into two phases; the detection, which must
happen before setting up ioremap, and the activation, which must happen
after parsing early boot parameters.
This fixes a crash on boot when VMI is enabled under VMware.
Impact: makes device isolation the default for AMD IOMMU
Some device drivers showed double-free bugs of DMA memory while testing
them with AMD IOMMU. If all devices share the same protection domain
this can lead to data corruption and data loss. Prevent this by putting
each device into its own protection domain per default.
This fixes Bug 11399:
if ibwdt_set_heartbeat(int t) is called with value 30 then
the check "if ((t < 0) || (t > 30))" in ibwdt_set_heartbeat
is not going to fail because t == 30, but in the loop, the
check wd_times[i] > t is never going to be true because
none of the wd_times are greater than the value of t (i.e. 30).
So we are exiting the loop with i == -1 and therefore setting
wd_margin to -1 which is wrong.
When project quota is active and is being used for directory tree
quota control, we disallow rename outside the current directory
tree. This requires a check to be made after all the inodes
involved in the rename are locked. We fail to unlock the inodes
correctly if we disallow the rename when the target is outside the
current directory tree. This results in a hang on the next access
to the inodes involved in failed rename.
We used to assume that even numbered threads were the primary
threads, ie those that would be listed and started as a cpu from
open firmware. Replace a left over is even (% 2) check with a check
for it being a primary thread and update the comments.
Tested with a debug print on pseries, identical code found for cell.
Signed-off-by: Milton Miller <miltonm@bga.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Several cifs patches were added to 2.6.27.8 to fix some races in the
mount/umount codepath. When this was done, a couple of prerequisite
patches were missed causing a minor regression.
When the last cifs mount to a server goes away, the kthread that manages
the socket is supposed to come down. The patches that went into 2.6.27.8
removed the kthread_stop calls that used to take down these threads, but
left the thread function expecting them. This made the thread stay up
even after the last mount was gone.
This patch should fix up this regression and also prevent a possible
race where a dead task could be signalled.
Signed-off-by: Jeff Layton <jlayton@redhat.com> Cc: Suresh Jayaraman <sjayaraman@suse.de> Acked-by: Steve French <smfrench@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Len Brown [Mon, 8 Dec 2008 21:03:07 +0000 (16:03 -0500)]
ACPI: delete OSI(Linux) DMI dmesg spam
In 2.6.28 a6e0887f21bbab337ee32d9c0a84d7c0b6e9141b removed this code
because the linux-acpi community no longer needs the feedback
that these console messages solicit.
here in .stable, we apply a simpler version of that patch,
but for the exact same reasons.
Signed-off-by: Len Brown <len.brown@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The 440x5 core in the Virtex5 uses the 440A type machine check
(ie, they have MCSRR0/MCSRR1). They thus need to call the
appropriate fixup function to hook the right variant of the
exception.
Without this, all machine checks become fatal due to loss
of context when entering the exception handler.
The patch 6341c39 "tracehook: exec" introduced a small regression in
2.6.27 regarding binfmt_misc exec event reporting. Since the reporting
is now done in the common search_binary_handler() function, an exec
of a misc binary will result in two (or possibly multiple) exec events
being reported, instead of just a single one, because the misc handler
contains a recursive call to search_binary_handler.
To add to the confusion, if PTRACE_O_TRACEEXEC is not active, the multiple
SIGTRAP signals will in fact cause only a single ptrace intercept, as the
signals are not queued. However, if PTRACE_O_TRACEEXEC is on, the debugger
will actually see multiple ptrace intercepts (PTRACE_EVENT_EXEC).
The test program included below demonstrates the problem.
This change fixes the bug by calling tracehook_report_exec() only in the
outermost search_binary_handler() call (bprm->recursion_depth == 0).
The additional change to restore bprm->recursion_depth after each binfmt
load_binary call is actually superfluous for this bug, since we test the
value saved on entry to search_binary_handler(). But it keeps the use of
of the depth count to its most obvious expected meaning. Depending on what
binfmt handlers do in certain cases, there could have been false-positive
tests for recursion limits before this change.
/* Test program using PTRACE_O_TRACEEXEC.
This forks and exec's the first argument with the rest of the arguments,
while ptrace'ing. It expects to see one PTRACE_EVENT_EXEC stop and
then a successful exit, with no other signals or events in between.
Test for kernel doing two PTRACE_EVENT_EXEC stops for a binfmt_misc exec:
$ gcc -g traceexec.c -o traceexec
$ sudo sh -c 'echo :test:M::foobar::/bin/cat: > /proc/sys/fs/binfmt_misc/register'
$ echo 'foobar test' > ./foobar
$ chmod +x ./foobar
$ ./traceexec ./foobar; echo $?
==> good <==
foobar test
0
$
==> bad <==
foobar test
unexpected status 0x4057f != 0
3
$
This patch introduces field 'recursion_depth' into struct linux_binprm to
track recursion level in binfmt_misc and binfmt_script. If recursion
level more then BINPRM_MAX_RECURSION it generates -ENOEXEC.
[akpm@linux-foundation.org: make linux_binprm.recursion_depth a uint] Signed-off-by: Kirill A. Shutemov <kirill@shutemov.name> Cc: Pavel Emelyanov <xemul@openvz.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Lee Schermerhorn noticed yesterday that I broke the mapping_writably_mapped
test in 2.6.7! Bad bad bug, good good find.
The i_mmap_writable count must be incremented for VM_SHARED (just as
i_writecount is for VM_DENYWRITE, but while holding the i_mmap_lock)
when dup_mmap() copies the vma for fork: it has its own more optimal
version of __vma_link_file(), and I missed this out. So the count
was later going down to 0 (dangerous) when one end unmapped, then
wrapping negative (inefficient) when the other end unmapped.
The only impact on x86 would have been that setting a mandatory lock on
a file which has at some time been opened O_RDWR and mapped MAP_SHARED
(but not necessarily PROT_WRITE) across a fork, might fail with -EAGAIN
when it should succeed, or succeed when it should fail.
But those architectures which rely on flush_dcache_page() to flush
userspace modifications back into the page before the kernel reads it,
may in some cases have skipped the flush after such a fork - though any
repetitive test will soon wrap the count negative, in which case it will
flush_dcache_page() unnecessarily.
Fix would be a two-liner, but mapping variable added, and comment moved.
The large pages fix from bcf8039ed45 broke 32-bit pagemap by pulling the
pagemap entry code out into a function with the wrong return type.
Pagemap entries are 64 bits on all systems and unsigned long is only 32
bits on 32-bit systems.
Signed-off-by: Matt Mackall <mpm@selenic.com> Reported-by: Doug Graham <dgraham@nortel.com> Cc: Alexey Dobriyan <adobriyan@gmail.com> Cc: Dave Hansen <dave@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
mconsole_init() passed 256 bytes as length in os_create_unix_socket, while
the sizeof UNIX_PATH_MAX is 108. This patch fixes that problem and avoids
a big overrun bug reported on UML bootup.
sockaddr_un.sun_path is UNIX_PATH_MAX long which causes the problem. Reported-by: Vikas K Managutte <vikki.km@gmail.com> Reported-by: Sarvesh Kumar Lal Das <skldas@gmail.com> Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com> Reviewed-by: Pekka Enberg <penberg@cs.helsinki.fi> Reviewed-by: WANG Cong <wangcong@zeuux.org> Cc: Jeff Dike <jdike@addtoit.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Makes a Compaq 6735s boot reliably again. It used to hang in the loop
on some boots. Give the link one second to train, otherwise break out
of the loop and reset the previously set clock bits.
Signed-off-by: Thomas Renninger <trenn@suse.de> Signed-off-by: Shaohua Li <shaohua.li@intel.com> Signed-off-by: Matthew Garrett <mjg59@srcf.ucam.org> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
In pci_create_slot(), the local variable 'slot_name' is allocated by
make_slot_name(), but never freed. We never use it after passing it to
the kobject core, so we should free it upon function exit.
Impact: fix possible deadlock in CPU hot-remove path
This patch fixes a possible deadlock scenario in the CPU remove path.
migration_call grabs rq->lock, then wakes up everything on rq->migration_queue
with the lock held. Then one of the tasks on the migration queue ends up
calling tg_shares_up which then also tries to acquire the same rq->lock.
Gets rid of dmesg spam created during physical memory hot-add which
will very likely confuse users. The change removes what appears to
be debugging code which I assume was unintentionally included in:
| I just did some experiments on a desktop for memory hotplug and this bug
| triggered a crash in my test.
|
| Yinghai's suggestion also fixed the bug.
We don't need to round it, just remove that extra -1
The generic lro code checks TCP flags/options.
Remove duplicate tests done in the driver.
Signed-off-by: Divy Le Ray <divy@chelsio.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Cc: Hannes Reinecke <hare@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
A SGE queue set timer might access registers while in EEH recovery,
triggering an EEH error loop. Stop all timers early in EEH process.
Signed-off-by: Divy Le Ray <divy@chelsio.com> Signed-off-by: David S. Miller <davem@davemloft.net> Cc: Karsten Keil <kkeil@novell.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
STAC/IDT driver creates "Headphone as Line-Out" switch even if there
is no line-out pins on the machine. For devices only with headpohnes
and speaker-outs, this switch shouldn't be created.
Added a QUIRK to patch_analog.c for the HP Elitebook 8530p
(IDs 0x103c:0x30e7) to use AD1884A model 'laptop' by default.
Playback and Capture confirmed working.
Use model=lenovo-ms7195-dig for MEDION MD96630 laptop (17c0:4085)
with ALC888 codec.
Reference: Novell bnc#412548
https://bugzilla.novell.com/show_bug.cgi?id=412528
Added a quirk for another Acer Aspier laptop (1025:0090) with ALC883
codec. Reported in Novell bnc#426935:
https://bugzilla.novell.com/show_bug.cgi?id=426935
Some machines have broken BIOS resume that doesn't restore the default
pin configuration properly, which results in a wrong detection of HP
pin. This causes a silent speaker output due to missing HP detection.
Related bug: Novell bug#406101
https://bugzilla.novell.com/show_bug.cgi?id=406101
This patch fixes the issue by saving/restoring the default pin configs
by the driver itself.
projects / karo-tx-linux.git / log