Which reports that we take mems_allowed_seq with interrupts enabled. A
little digging found that this can only be from
cpuset_change_task_nodemask().
This is an actual deadlock because an interrupt doing an allocation will
hit get_mems_allowed()->...->__read_seqcount_begin(), which will spin
forever waiting for the write side to complete.
Cc: John Stultz <john.stultz@linaro.org> Cc: Mel Gorman <mgorman@suse.de> Reported-by: Juri Lelli <juri.lelli@gmail.com> Signed-off-by: Peter Zijlstra <peterz@infradead.org> Tested-by: Juri Lelli <juri.lelli@gmail.com> Acked-by: Li Zefan <lizefan@huawei.com> Acked-by: Mel Gorman <mgorman@suse.de> Signed-off-by: Tejun Heo <tj@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If a cgroup file implements either read_map() or read_seq_string(),
such file is served using seq_file by overriding file->f_op to
cgroup_seqfile_operations, which also overrides the release method to
single_release() from cgroup_file_release().
Because cgroup_file_open() didn't use to acquire any resources, this
used to be fine, but since f7d58818ba42 ("cgroup: pin
cgroup_subsys_state when opening a cgroupfs file"), cgroup_file_open()
pins the css (cgroup_subsys_state) which is put by
cgroup_file_release(). The patch forgot to update the release path
for seq_files and each open/release cycle leaks a css reference.
Fix it by updating cgroup_file_release() to also handle seq_files and
using it for seq_file release path too.
Since be44562613851 ("cgroup: remove synchronize_rcu() from
cgroup_diput()"), cgroup destruction path makes use of workqueue. css
freeing is performed from a work item from that point on and a later
commit, ea15f8ccdb430 ("cgroup: split cgroup destruction into two
steps"), moves css offlining to workqueue too.
As cgroup destruction isn't depended upon for memory reclaim, the
destruction work items were put on the system_wq; unfortunately, some
controller may block in the destruction path for considerable duration
while holding cgroup_mutex. As large part of destruction path is
synchronized through cgroup_mutex, when combined with high rate of
cgroup removals, this has potential to fill up system_wq's max_active
of 256.
Also, it turns out that memcg's css destruction path ends up queueing
and waiting for work items on system_wq through work_on_cpu(). If
such operation happens while system_wq is fully occupied by cgroup
destruction work items, work_on_cpu() can't make forward progress
because system_wq is full and other destruction work items on
system_wq can't make forward progress because the work item waiting
for work_on_cpu() is holding cgroup_mutex, leading to deadlock.
This can be fixed by queueing destruction work items on a separate
workqueue. This patch creates a dedicated workqueue -
cgroup_destroy_wq - for this purpose. As these work items shouldn't
have inter-dependencies and mostly serialized by cgroup_mutex anyway,
giving high concurrency level doesn't buy anything and the workqueue's
@max_active is set to 1 so that destruction work items are executed
one by one on each CPU.
Hugh Dickins: Because cgroup_init() is run before init_workqueues(),
cgroup_destroy_wq can't be allocated from cgroup_init(). Do it from a
separate core_initcall(). In the future, we probably want to reorder
so that workqueue init happens before cgroup_init().
An ordered workqueue implements execution ordering by using single
pool_workqueue with max_active == 1. On a given pool_workqueue, work
items are processed in FIFO order and limiting max_active to 1
enforces the queued work items to be processed one by one.
Unfortunately, 4c16bd327c ("workqueue: implement NUMA affinity for
unbound workqueues") accidentally broke this guarantee by applying
NUMA affinity to ordered workqueues too. On NUMA setups, an ordered
workqueue would end up with separate pool_workqueues for different
nodes. Each pool_workqueue still limits max_active to 1 but multiple
work items may be executed concurrently and out of order depending on
which node they are queued to.
Fix it by using dedicated ordered_wq_attrs[] when creating ordered
workqueues. The new attrs match the unbound ones except that no_numa
is always set thus forcing all NUMA nodes to share the default
pool_workqueue.
While at it, add sanity check in workqueue creation path which
verifies that an ordered workqueues has only the default
pool_workqueue.
When translating a user space address, the address must be checked against
the ASCE limit of the process. If the address is larger than the maximum
address that is reachable with the ASCE, an ASCE type exception must be
generated.
The current code simply ignored the higher order bits. This resulted in an
address wrap around in user space instead of an exception in user space.
PTE_PROT_NONE means that a pte is present but does not have any
read/write attributes. However, setting the memory type like
pgprot_writecombine() is allowed and such bits overlap with
PTE_PROT_NONE. This causes mmap/munmap issues in drivers that change the
vma->vm_pg_prot on PROT_NONE mappings.
This patch reverts the PTE_FILE/PTE_PROT_NONE shift in commit 59911ca4325d (ARM64: mm: Move PTE_PROT_NONE bit) and moves PTE_PROT_NONE
together with the other software bits.
Signed-off-by: Steve Capper <steve.capper@linaro.org> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Tested-by: Steve Capper <steve.capper@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In a recent patch:
commit c13f20ac48328b05cd3b8c19e31ed6c132b44b42
Author: Michael Neuling <mikey@neuling.org>
powerpc/signals: Mark VSX not saved with small contexts
We fixed an issue but an improved solution was later discussed after the patch
was merged.
Firstly, this patch doesn't handle the 64bit signals case, which could also hit
this issue (but has never been reported).
Secondly, the original patch isn't clear what MSR VSX should be set to. The
new approach below always clears the MSR VSX bit (to indicate no VSX is in the
context) and sets it only in the specific case where VSX is available (ie. when
VSX has been used and the signal context passed has space to provide the
state).
This reverts the original patch and replaces it with the improved solution. It
also adds a 64 bit version.
Signed-off-by: Michael Neuling <mikey@neuling.org> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
UHID allows short writes so user-space can omit unused fields. We
automatically set them to 0 in the kernel. However, the 64/32 bit
compat-handler didn't do that in the UHID_CREATE fallback. This will
reveal random kernel heap data (of random size, even) to user-space.
Fixes: befde0226a59 ('HID: uhid: make creating devices work on 64/32 systems') Reported-by: Ben Hutchings <ben@decadent.org.uk> Signed-off-by: David Herrmann <dh.herrmann@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 7a0a5355cbc71efa md: Don't test all of mddev->flags at once.
made most tests on mddev->flags safer, but missed one.
When
commit 260fa034ef7a4ff8b7306 md: avoid deadlock when dirty buffers during md_stop.
added MD_STILL_CLOSED, this caused md_check_recovery to misbehave.
It can think there is something to do but find nothing. This can
lead to the md thread spinning during array shutdown.
https://bugzilla.kernel.org/show_bug.cgi?id=65721
Reported-and-tested-by: Richard W.M. Jones <rjones@redhat.com> Fixes: 260fa034ef7a4ff8b7306 Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This can happen if there are some stripes left, fewer than MAX_STRIPE_BATCH.
A worker is queued to handle them.
But before calling raid5_do_work, raid5d handles those
stripes making conf->active_stripe = 0.
So mddev_suspend() can return.
We might then free old worker resources before the queued
raid5_do_work() handled them. When it runs, it crashes.
The various ->run routines of md personalities assume that the 'queue'
has been initialised by the blk_set_stacking_limits() call in
md_alloc().
However when the level is changed (by level_store()) the ->run routine
for the new level is called for an array which has already had the
stacking limits modified. This can result in incorrect final
settings.
So call blk_set_stacking_limits() before ->run in level_store().
A specific consequence of this bug is that it causes
discard_granularity to be set incorrectly when reshaping a RAID4 to a
RAID0.
This is suitable for any -stable kernel since 3.3 in which
blk_set_stacking_limits() was introduced.
When release_stripe() is called in grow_one_stripe(), the
mddev->thread is null. So it will omit one wakeup this thread to
release stripe.
For this condition, use slow_path to release stripe.
setfacl over cifs mounts can remove the default ACL when setting the
(non-default part of) the ACL and vice versa (we were leaving at 0
rather than setting to -1 the count field for the unaffected
half of the ACL. For example notice the setfacl removed
the default ACL in this sequence:
steven@steven-GA-970A-DS3:~/cifs-2.6$ getfacl /mnt/test-dir
getfacl: Removing leading '/' from absolute path names
user::rwx
user:test:rwx
group::r-x
mask::rwx
other::r-x
Signed-off-by: Steve French <smfrench@gmail.com> Acked-by: Jeremy Allison <jra@samba.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The analog-stick vertical axes are inverted. Fix that! Otherwise, games
and other gamepad applications need to carry their own fixups (which they
thankfully haven't done, yet).
Reported-by: Rafael Brune <mail@rbrune.de> Tested-by: Rafael Brune <mail@rbrune.de> Signed-off-by: David Herrmann <dh.herrmann@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch avoids a duplicate iscsit_increment_maxcmdsn() call for
ISER_IB_RDMA_WRITE within isert_map_rdma() + isert_reg_rdma_frwr(),
which will already be occuring once during isert_put_datain() ->
iscsit_build_rsp_pdu() operation.
It also removes the local conn->stat_sn assignment + increment,
and changes the third parameter to iscsit_build_rsp_pdu() to
signal this should be done by iscsi_target_mode code.
Tested-by: Moussa Ba <moussaba@micron.com> Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
GPU with low amount of ram can fails at pinning new framebuffer before
unpinning old one. On such failure, retry with unpinning old one before
pinning new one allowing to work around the issue. This is somewhat
ugly but only affect those old GPU we care about.
Fixes crashes when handling atif events due to the lack of a
callback being registered.
Signed-off-by: Samuel Li <samuel.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To make this work both for resets and for BIOS takeover just add the
forcewake clearing call back to intel_uncore_early_sanitize.
We need to clear the forcewake in early sanitize so that the forcewak
dance in intel_uncore_init (to figure out whether we have mt or legacy
forcewake on ivb) works. That cleanup fits in nicely with the general
topic of early_sanitize to prepare for the very first mmio ops.
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Reported-by: Jörg Otte <jrg.otte@gmail.com> Cc: Jörg Otte <jrg.otte@gmail.com>
References: https://lkml.org/lkml/2013/11/16/40 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Apparently they need the same treatment as primary planes. This fixes
modesetting failures because of stuck cursors (!) on Thomas' i830M
machine.
I've figured while at it I'll also roll it out for the ivb 3 pipe
version of this function. I didn't do this for i845/i865 since Bspec
says the update mechanism works differently, and there's some
additional rules about what can be updated in which order.
Tested-by: Thomas Richter <thor@math.tu-berlin.de> Cc: Thomas Richter <thor@math.tu-berlin.de> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The ns2501 controller seems to need the dpll and dvo port to accept
the timing update commands. Quick testing on my x30 here seems to
indicate that other dvo controllers don't mind. So let's move the
->mode_set callback to a place where we have the port up and running
already.
Tested-by: Chris Wilson <chris@chris-wilson.co.uk> Tested-by: Thomas Richter <thor@math.tu-berlin.de> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
All error paths will want to keep the mm node, so handle this at the
function exit. This fixes an ioremap failure error path.
Also add some comments to make the function a bit easier to understand.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Also check the busy placements before deciding to move a buffer object.
Failing to do this may result in a completely unneccessary move within a
single memory type.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> Reviewed-by: Jakob Bornecrantz <jakob@vmware.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 8c4f3c3fa9681 "ftrace: Check module functions being traced on reload"
fixed module loading and unloading with respect to function tracing, but
it missed the function graph tracer. If you perform the following
The above mentioned commit didn't go far enough. Well, it covered the
function tracer by adding checks in __register_ftrace_function(). The
problem is that the function graph tracer circumvents that (for a slight
efficiency gain when function graph trace is running with a function
tracer. The gain was not worth this).
The problem came with ftrace_startup() which should always be called after
__register_ftrace_function(), if you want this bug to be completely fixed.
Anyway, this solution moves __register_ftrace_function() inside of
ftrace_startup() and removes the need to call them both.
Reported-by: Dave Wysochanski <dwysocha@redhat.com> Fixes: ed926f9b35cd ("ftrace: Use counters to enable functions to trace") Signed-off-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Follow-up to commit 294d31e8227c ("sony-laptop: don't change keyboard
backlight settings"): avoid messing up the state on resume. Leave it to
what was before suspending as it's anyway likely that we still don't
know what value we should write to the EC registers. This fix is also
required in 3.12
prior to week 08 of 2013 Freescale misprogrammed between 1 and 3% of
PFUZE1000 parts with a ID=0x8 instead of the expected ID=0x0
Signed-off-by: Tim Harvey <tharvey@gateworks.com> Signed-off-by: Mark Brown <broonie@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Userspace uses the netdev devtype for stuff like device naming and type
detection. Be nice and set it. Remove the pointless #if/#endif around
SET_NETDEV_DEV too.
Signed-off-by: Dan Williams <dcbw@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A departing reader must restart a flush_to_ldisc() worker _before_
the next reader enters the read loop; this is to avoid the new reader
concluding no more i/o is available and prematurely exiting, when the
old reader simply hasn't re-started the worker yet.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A common security idiom is to hangup the current tty (via vhangup())
after forking but before execing a root shell. This hangs up any
existing opens which other processes may have and ensures subsequent
opens have the necessary permissions to open the root shell tty/pty.
Reset the TTY_HUPPED state after the driver has successfully
returned the opened tty (perform the reset while the tty is locked
to avoid racing with concurrent hangups).
Commit cbfd0340ae1993378fd47179db949e050e16e697,
'n_tty: Process echoes in blocks', introduced an error when
consuming the echo buffer tail to prevent buffer overrun, where
the incorrect operation code byte is checked to determine how
far to advance the tail to the next echo byte.
Check the correct byte for the echo operation code byte.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Although the maximum allowable canonical line is specified to
be 255 bytes (MAX_CANON), the practical limit has actually been
the size of the line discipline read buffer (N_TTY_BUF_SIZE == 4096).
Commit 32f13521ca68bc624ff6effc77f308a52b038bf0,
n_tty: Line copy to user buffer in canonical mode, limited the
line copy to 4095 bytes. With a completely full line discipline
read buffer and a userspace buffer > 4095, _no_ data was copied,
and the read() syscall returned 0, indicating EOF.
Fix the interval arithmetic to compute the correct number of bytes
to copy to userspace in the range [1..4096].
Signed-off-by: Peter Hurley <peter@hurleysoftware.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch fixes following error (for big kernels):
---8<---
arch/avr32/boot/u-boot/head.o: In function `no_tag_table':
(.init.text+0x44): relocation truncated to fit: R_AVR32_22H_PCREL against symbol `panic' defined in .text.unlikely section in kernel/built-in.o
arch/avr32/kernel/built-in.o: In function `bad_return':
(.ex.text+0x236): relocation truncated to fit: R_AVR32_22H_PCREL against symbol `panic' defined in .text.unlikely section in kernel/built-in.o
--->8---
It comes up when the kernel increases and 'panic()' is too far away to fit in
the +/- 2MiB range. Which in turn issues from the 21-bit displacement in
'br{cond4}' mnemonic which is one of the two ways to do jumps (rjmp has just
10-bit displacement and therefore a way smaller range). This fact was stated
before in 8d29b7b9f81d6b83d869ff054e6c189d6da73f1f.
One solution to solve this is to add a local storage for the symbol address
and just load the $pc with that value.
The SELinux/NetLabel glue code has a locking bug that affects systems
with NetLabel enabled, see the kernel error message below. This patch
corrects this problem by converting the bottom half socket lock to a
more conventional, and correct for this call-path, lock_sock() call.
The PCI host bridge scan handler installs its own notify handler,
handle_hotplug_event_root(), by itself. Nevertheless, the ACPI
hotplug framework also installs the common notify handler,
acpi_hotplug_notify_cb(), for PCI root bridges. This causes
acpi_hotplug_notify_cb() to call _OST method with unsupported
error as hotplug.enabled is not set.
To address this issue, introduce hotplug.ignore flag, which
indicates that the scan handler installs its own notify handler by
itself. The ACPI hotplug framework does not install the common
notify handler when this flag is set.
Signed-off-by: Toshi Kani <toshi.kani@hp.com>
[rjw: Changed the name of the new flag] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The pcie_portdrv .probe() method calls pci_enable_device() once, in
pcie_port_device_register(), but the .remove() method calls
pci_disable_device() twice, in pcie_port_device_remove() and in
pcie_portdrv_remove().
That causes a "disabling already-disabled device" warning when removing a
PCIe port device. This happens all the time when removing Thunderbolt
devices, but is also easy to reproduce with, e.g.,
"echo 0000:00:1c.3 > /sys/bus/pci/drivers/pcieport/unbind"
This patch removes the disable from pcie_portdrv_remove().
[bhelgaas: changelog, tag for stable] Reported-by: David Bulkow <David.Bulkow@stratus.com> Reported-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
...to make it clear what the intent behind each record's operation was.
In many cases you can infer this, based on the context of the syscall
and the result. In other cases it's not so obvious. For instance, in
the case where you have a file being renamed over another, you'll have
two different records with the same filename but different inode info.
By logging this information we can clearly tell which one was created
and which was deleted.
This fixes what was broken in commit bfcec708.
Commit 79f6530c should also be backported to stable v3.7+.
Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Eric Paris <eparis@redhat.com> Signed-off-by: Richard Guy Briggs <rgb@redhat.com> Signed-off-by: Eric Paris <eparis@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Historically, when a syscall that creates a dentry fails, you get an audit
record that looks something like this (when trying to create a file named
"new" in "/tmp/tmp.SxiLnCcv63"):
This record makes no sense since it's associating the inode information for
"/tmp/tmp.SxiLnCcv63" with the path "/tmp/tmp.SxiLnCcv63/new". The recent
patch I posted to fix the audit_inode call in do_last fixes this, by making it
look more like this:
While this is more correct, if the creation of the file fails, then we
have no record of the filename that the user tried to create.
This patch adds a call to audit_inode_child to may_create. This creates
an AUDIT_TYPE_CHILD_CREATE record that will sit in place until the
create succeeds. When and if the create does succeed, then this record
will be updated with the correct inode info from the create.
This fixes what was broken in commit bfcec708.
Commit 79f6530c should also be backported to stable v3.7+.
Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Eric Paris <eparis@redhat.com> Signed-off-by: Richard Guy Briggs <rgb@redhat.com> Signed-off-by: Eric Paris <eparis@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We leak 4 bytes of kernel stack in response to an AUDIT_GET request as
we miss to initialize the mask member of status_set. Fix that.
Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Eric Paris <eparis@redhat.com> Signed-off-by: Mathias Krause <minipli@googlemail.com> Signed-off-by: Richard Guy Briggs <rgb@redhat.com> Signed-off-by: Eric Paris <eparis@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Using the nlmsg_len member of the netlink header to test if the message
is valid is wrong as it includes the size of the netlink header itself.
Thereby allowing to send short netlink messages that pass those checks.
Use nlmsg_len() instead to test for the right message length. The result
of nlmsg_len() is guaranteed to be non-negative as the netlink message
already passed the checks of nlmsg_ok().
Also switch to min_t() to please checkpatch.pl.
Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Eric Paris <eparis@redhat.com> Signed-off-by: Mathias Krause <minipli@googlemail.com> Signed-off-by: Richard Guy Briggs <rgb@redhat.com> Signed-off-by: Eric Paris <eparis@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When the audit=1 kernel parameter is absent and auditd is not running,
AUDIT_USER_AVC messages are being silently discarded.
AUDIT_USER_AVC messages should be sent to userspace using printk(), as
mentioned in the commit message of 4a4cd633 ("AUDIT: Optimise the
audit-disabled case for discarding user messages").
When audit_enabled is 0, audit_receive_msg() discards all user messages
except for AUDIT_USER_AVC messages. However, audit_log_common_recv_msg()
refuses to allocate an audit_buffer if audit_enabled is 0. The fix is to
special case AUDIT_USER_AVC messages in both functions.
It looks like commit 50397bd1 ("[AUDIT] clean up audit_receive_msg()")
introduced this bug.
Signed-off-by: Tyler Hicks <tyhicks@canonical.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Eric Paris <eparis@redhat.com> Cc: linux-audit@redhat.com Acked-by: Kees Cook <keescook@chromium.org> Signed-off-by: Richard Guy Briggs <rgb@redhat.com> Signed-off-by: Eric Paris <eparis@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
While receiving a packet on SDIO interface, we allocate skb with
size multiple of SDIO block size. We need to resize this skb
after RX using packet length from RX header.
Signed-off-by: Avinash Patil <patila@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Now we treat any reparse point as a symbolic link and map it to a Unix
one that is not true in a common case due to many reparse point types
supported by SMB servers.
Distinguish reparse point types into two groups:
1) that can be accessed directly through a reparse point
(junctions, deduplicated files, NFS symlinks);
2) that need to be processed manually (Windows symbolic links, DFS);
and map only Windows symbolic links to Unix ones.
Acked-by: Jeff Layton <jlayton@redhat.com> Reported-and-tested-by: Joao Correia <joaomiguelcorreia@gmail.com> Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru> Signed-off-by: Steve French <smfrench@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Dirty data accounting wasn't quite right - firstly, we were adding the key we're
inserting after it could have merged with another dirty key already in the
btree, and secondly we could sometimes pass the wrong offset to
bcache_dev_sectors_dirty_add() for dirty data we were overwriting - which is
important when tracking dirty data by stripe.
Signed-off-by: Kent Overstreet <kmo@daterainc.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
I have received a report about the BUG_ON() in free_basic_memory_bitmaps()
triggering mysteriously during an aborted s2disk hibernation attempt.
The only way I can explain that is that /dev/snapshot was first
opened for writing (resume mode), then closed and then opened again
for reading and closed again without freezing tasks. In that case
the first invocation of snapshot_open() would set the free_bitmaps
flag in snapshot_state, which is a static variable. That flag
wouldn't be cleared later and the second invocation of snapshot_open()
would just leave it like that, so the subsequent snapshot_release()
would see data->frozen set and free_basic_memory_bitmaps() would be
called unnecessarily.
To prevent that from happening clear data->free_bitmaps in
snapshot_open() when the file is being opened for reading (hibernate
mode).
In addition to that, replace the BUG_ON() in free_basic_memory_bitmaps()
with a WARN_ON() as the kernel can continue just fine if the condition
checked by that macro occurs.
Fixes: aab172891542 (PM / hibernate: Fix user space driven resume regression) Reported-by: Oliver Lorenz <olli@olorenz.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit fa180eb448fa (PM / Runtime: Idle devices asynchronously after
probe|release) modified __device_release_driver() to call
pm_runtime_put(dev) instead of pm_runtime_put_sync(dev) before
detaching the driver from the device. However, that was a mistake,
because pm_runtime_put(dev) causes rpm_idle() to be queued up and
the driver may be gone already when that function is executed.
That breaks the assumptions the drivers have the right to make
about the core's behavior on the basis of the existing documentation
and actually causes problems to happen, so revert that part of
commit fa180eb448fa and restore the previous behavior of
__device_release_driver().
Reported-by: Tomi Valkeinen <tomi.valkeinen@ti.com> Fixes: fa180eb448fa (PM / Runtime: Idle devices asynchronously after probe|release) Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by: Kevin Hilman <khilman@linaro.org> Acked-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When system has a lot of highmem (e.g. 16GiB using a 32 bits kernel),
the code to calculate how much memory we need to preallocate in
normal zone may cause overflow. As Leon has analysed:
It looks that during computing 'alloc' variable there is overflow:
alloc = (3943404 - 1970542) - 1978280 = -5418 (signed)
And this function goes to err_out.
Fix this by avoiding that overflow.
References: https://bugzilla.kernel.org/show_bug.cgi?id=60817 Reported-and-tested-by: Leon Drugi <eyak@wp.pl> Signed-off-by: Aaron Lu <aaron.lu@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This fix should be backported to stable kernels starting with 2.6.37. Note
that in the kernels prior to 3.5 the affected code is different, but the
bug is still there - bdi_init is called and bdi_destroy isn't.
This patch fixes a bug in delayed Task Aborted Status (TAS) handling,
where transport_send_task_abort() was not returning for the case
when the se_tfo->write_pending() callback indicated that last fabric
specific WRITE PDU had not yet been received.
It also adds an explicit cmd->scsi_status = SAM_STAT_TASK_ABORTED
assignment within transport_check_aborted_status() to avoid the case
where se_tfo->queue_status() is called when the SAM_STAT_TASK_ABORTED
assignment + ->queue_status() in transport_send_task_abort() does not
occur once SCF_SENT_DELAYED_TAS has been set.
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch changes isert_reg_rdma_frwr() to not use FRMR for single
dma entry requests from small I/Os, in order to avoid the associated
memory registration overhead.
Using DMA MR is sufficient here for the single dma entry requests,
and addresses a >= v3.12 performance regression.
Signed-off-by: Vu Pham <vu@mellanox.com> Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When performing continuations there are implied sources that need to be
added to the source count. Quoting dma_set_maxpq:
/* dma_maxpq - reduce maxpq in the face of continued operations
* @dma - dma device with PQ capability
* @flags - to check if DMA_PREP_CONTINUE and DMA_PREP_PQ_DISABLE_P are set
*
* When an engine does not support native continuation we need 3 extra
* source slots to reuse P and Q with the following coefficients:
* 1/ {00} * P : remove P from Q', but use it as a source for P'
* 2/ {01} * Q : use Q to continue Q' calculation
* 3/ {00} * Q : subtract Q from P' to cancel (2)
*
* In the case where P is disabled we only need 1 extra source:
* 1/ {01} * Q : use Q to continue Q' calculation
*/
...fix the selection of the 16 source path to take these implied sources
into account.
Note this also kills the BUG_ON(src_cnt < 9) check in
__ioat3_prep_pq16_lock(). Besides not accounting for implied sources
the check is redundant given we already made the path selection.
Cc: Dave Jiang <dave.jiang@intel.com> Acked-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The array to lookup the sed pool based on the number of sources
(pq16_idx_to_sedi) is 16 entries and expects a max source index.
However, we pass the total source count which runs off the end of the
array when src_cnt == 16. The minimal fix is to just pass src_cnt-1,
but given we know the source count is > 8 we can just calculate the sed
pool by (src_cnt - 2) >> 3.
Cc: Dave Jiang <dave.jiang@intel.com> Acked-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 48a9db4 (3.11) removed the memset op in the xor selftest for ioatdma.
The issue is that with the removal of that op, it never replaced the memset
with a CPU memset. The memory being operated on is expected to be zeroes but
was not. This is causing the xor selftest to fail.
Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
dm-mpath and dm-thin must process messages even if some device is
suspended, so we allocate argv buffer with GFP_NOIO. These messages have
a small fixed number of arguments.
On the other hand, dm-switch needs to process bulk data using messages
so excessive use of GFP_NOIO could cause trouble.
The patch also lowers the default number of arguments from 64 to 8, so
that there is smaller load on GFP_NOIO allocations.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Acked-by: Alasdair G Kergon <agk@redhat.com> Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The code that was trying to do this was inadequate. The postsuspend
method (in ioctl context), needs to wait for the worker thread to
acknowledge the request to quiesce. Otherwise the migration count may
drop to zero temporarily before the worker thread realises we're
quiescing. In this case the target will be taken down, but the worker
thread may have issued a new migration, which will cause an oops when
it completes.
Signed-off-by: Joe Thornber <ejt@redhat.com> Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Whenever multipath_dtr() is happening we must prevent queueing any
further path activation work. Implement this by adding a new
'pg_init_disabled' flag to the multipath structure that denotes future
path activation work should be skipped if it is set. By disabling
pg_init and then re-enabling in flush_multipath_work() we also avoid the
potential for pg_init to be initiated while suspending an mpath device.
Without this patch a race condition exists that may result in a kernel
panic:
1) If after pg_init_done() decrements pg_init_in_progress to 0, a call
to wait_for_pg_init_completion() assumes there are no more pending path
management commands.
2) If pg_init_required is set by pg_init_done(), due to retryable
mode_select errors, then process_queued_ios() will again queue the
path activation work.
3) If free_multipath() completes before activate_path() work is called a
NULL pointer dereference like the following can be seen when
accessing members of the recently destructed multipath:
When a software timeout occurs, the transfer is not stopped. In DMA case,
it causes DMA channel to be stuck because the transfer is still active
causing following transfers to be queued but not computed.
Signed-off-by: Ludovic Desroches <ludovic.desroches@atmel.com> Reported-by: Alexander Morozov <etesial@gmail.com> Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com> Signed-off-by: Chris Ball <cjb@laptop.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
thread 0: reclaim entry x (get refcount, but not call zswap_get_swap_cache_page)
thread 1: call zswap_frontswap_invalidate_page to invalidate entry x.
finished, entry x and its zbud is not freed as its refcount != 0
now, the swap_map[x] = 0
thread 0: now call zswap_get_swap_cache_page
swapcache_prepare return -ENOENT because entry x is not used any more
zswap_get_swap_cache_page return ZSWAP_SWAPCACHE_NOMEM
zswap_writeback_entry do nothing except put refcount
Now, the memory of zswap_entry x and its zpage leak.
Modify:
- check the refcount in fail path, free memory if it is not referenced.
- use ZSWAP_SWAPCACHE_FAIL instead of ZSWAP_SWAPCACHE_NOMEM as the fail path
can be not only caused by nomem but also by invalidate.
Signed-off-by: Weijie Yang <weijie.yang@samsung.com> Reviewed-by: Bob Liu <bob.liu@oracle.com> Reviewed-by: Minchan Kim <minchan@kernel.org> Acked-by: Seth Jennings <sjenning@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@linuxfoundation.org>
This patch fixes the problem that get_unmapped_area() can return illegal
address and result in failing mmap(2) etc.
In case that the address higher than PAGE_SIZE is set to
/proc/sys/vm/mmap_min_addr, the address lower than mmap_min_addr can be
returned by get_unmapped_area(), even if you do not pass any virtual
address hint (i.e. the second argument).
This is because the current get_unmapped_area() code does not take into
account mmap_min_addr.
This leads to two actual problems as follows:
1. mmap(2) can fail with EPERM on the process without CAP_SYS_RAWIO,
although any illegal parameter is not passed.
2. The bottom-up search path after the top-down search might not work in
arch_get_unmapped_area_topdown().
Note: The first and third chunk of my patch, which changes "len" check,
are for more precise check using mmap_min_addr, and not for solving the
above problem.
ath5k_tx_frame_completed saves the intended per-rate retry counts before
they are cleared by ieee80211_tx_info_clear_status, however at this
point the information in info->status.rates is incomplete.
This causes significant throughput degradation and excessive packet loss
on links where high bit rates don't work properly.
Move the copy from bf->rates a few lines up to ensure that the saved
retry counts are updated, and that they are really cleared in
info->status.rates after the call to ieee80211_tx_info_clear_status.
Cc: Thomas Huehn <thomas@net.t-labs.tu-berlin.de> Cc: Benjamin Vahl <bvahl@net.t-labs.tu-berlin.de> Reported-by: Ben West <ben@gowasabi.net> Signed-off-by: Felix Fietkau <nbd@openwrt.org> Acked-by: Thomas Huehn <thomas@net.t-labs.tu-berlin.de> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The routine that processes received frames was returning the RSSI value for the
signal strength; however, that value is available only for associated APs. As
a result, the strength was the absurd value of 10 dBm. As a result, scans
return incorrect values for the strength, which causes unwanted attempts to roam.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The routine that processes received frames was returning the RSSI value for the
signal strength; however, that value is available only for associated APs. As
a result, the strength was the absurd value of 10 dBm. As a result, scans
return incorrect values for the strength, which causes unwanted attempts to roam.
This patch fixes https://bugzilla.kernel.org/show_bug.cgi?id=63881.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Reported-by: Matthieu Baerts <matttbe@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The routine that processes received frames was returning the RSSI value for the
signal strength; however, that value is available only for associated APs. As
a result, the strength was the absurd value of 10 dBm. As a result, scans
return incorrect values for the strength, which causes unwanted attempts to roam.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If the permission check fails, we drop a reference to the blkif without
having taken it in the first place. The bug was introduced in commit 604c499cbbcc3d5fe5fb8d53306aa0fae1990109 (xen/blkback: Check device
permissions before allowing OP_DISCARD).
Cc: Jan Beulich <JBeulich@suse.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The imx23 board will check the fingerprint, so it will call the
mx23_check_transcription_stamp. This function will use @chip->buffers->databuf
as its buffer which is allocated in the nand_scan_tail().
Unfortunately, the mx23_check_transcription_stamp is called before the
nand_scan_tail(). So we will meet a NULL pointer bug:
This patch rearrange the init procedure:
Set the NAND_SKIP_BBTSCAN to skip the nand scan firstly, and after we
set the proper settings, we will call the chip->scan_bbt() manually.
[1] The gpmi uses the nand_command_lp to issue the commands to NAND chips.
The gpmi issues a DMA operation with gpmi_cmd_ctrl when it handles
a NAND_CMD_NONE control command. So when we read a page(NAND_CMD_READ0)
from the NAND, we may send two DMA operations back-to-back.
If we do not serialize the two DMA operations, we will meet a bug when
1.1) we enable CONFIG_DMA_API_DEBUG, CONFIG_DMADEVICES_DEBUG,
and CONFIG_DEBUG_SG.
1.2) Use the following commands in an UART console and a SSH console:
cmd 1: while true;do dd if=/dev/mtd0 of=/dev/null;done
cmd 1: while true;do dd if=/dev/mmcblk0 of=/dev/null;done
The kernel log shows below:
-----------------------------------------------------------------
kernel BUG at lib/scatterlist.c:28!
Unable to handle kernel NULL pointer dereference at virtual address 00000000
.........................
[<80044a0c>] (__bug+0x18/0x24) from [<80249b74>] (sg_next+0x48/0x4c)
[<80249b74>] (sg_next+0x48/0x4c) from [<80255398>] (debug_dma_unmap_sg+0x170/0x1a4)
[<80255398>] (debug_dma_unmap_sg+0x170/0x1a4) from [<8004af58>] (dma_unmap_sg+0x14/0x6c)
[<8004af58>] (dma_unmap_sg+0x14/0x6c) from [<8027e594>] (mxs_dma_tasklet+0x18/0x1c)
[<8027e594>] (mxs_dma_tasklet+0x18/0x1c) from [<8007d444>] (tasklet_action+0x114/0x164)
-----------------------------------------------------------------
1.3) Assume the two DMA operations is X (first) and Y (second).
The root cause of the bug:
Assume process P issues DMA X, and sleep on the completion
@this->dma_done. X's tasklet callback is dma_irq_callback. It firstly
wake up the process sleeping on the completion @this->dma_done,
and then trid to unmap the scatterlist S. The waked process P will
issue Y in another ARM core. Y initializes S->sg_magic to zero
with sg_init_one(), while dma_irq_callback is unmapping S at the same
time.
See the diagram:
ARM core 0 | ARM core 1
-------------------------------------------------------------
(P issues DMA X, then sleep) --> |
|
(X's tasklet wakes P) --> |
|
| <-- (P begin to issue DMA Y)
|
(X's tasklet unmap the |
scatterlist S with dma_unmap_sg) --> | <-- (Y calls sg_init_one() to init
| scatterlist S)
|
[2] This patch serialize both the X and Y in the following way:
Unmap the DMA scatterlist S firstly, and wake up the process at the end
of the DMA callback, in such a way, Y will be executed after X.
After this patch:
ARM core 0 | ARM core 1
-------------------------------------------------------------
(P issues DMA X, then sleep) --> |
|
(X's tasklet unmap the |
scatterlist S with dma_unmap_sg) --> |
|
(X's tasklet wakes P) --> |
|
| <-- (P begin to issue DMA Y)
|
| <-- (Y calls sg_init_one() to init
| scatterlist S)
|
If no nand flash detected, the driver probe function will goto
err_nand_ioremap label.
Then platform_driver_unregister() will be called. It will get the
lock of atmel_nand device since it is parent of nfc_device. The
problem is the lock is already hold by atmel_nand_probe itself.
So system will be in a dead lock.
This patch just simply removed to platform_driver_unregister() call.
When atmel_nand driver is quit the platform_driver_unregister() will
be called in atmel_nand_remove().
[Brian: the NAND platform probe really has no business
registering/unregistering another driver; this fixes the deadlock, but
we should follow up the likely racy behavior here with a better
architecture]
Hardware:
CPU: XLP832,the 64-bit OS
NOR Flash:S29GL128S 128M
Software:
Kernel:2.6.32.41
Filesystem:JFFS2
When writing files, errors appear:
Write len 182 but return retlen 180
Write of 182 bytes at 0x072c815c failed. returned -5, retlen 180
Write len 186 but return retlen 184
Write of 186 bytes at 0x072caff4 failed. returned -5, retlen 184
These errors exist only in 64-bit systems,not in 32-bit systems. After analysis, we
found that the left shift operation is wrong in map_word_load_partial. For instance:
unsigned char buf[3] ={0x9e,0x3a,0xea};
map_bankwidth(map) is 4;
for (i=0; i < 3; i++) {
int bitpos;
bitpos = (map_bankwidth(map)-1-i)*8;
orig.x[0] &= ~(0xff << bitpos);
orig.x[0] |= buf[i] << bitpos;
}
The value of orig.x[0] is expected to be 0x9e3aeaff, but in this situation(64-bit
System) we'll get the wrong value of 0xffffffff9e3aeaff due to the 64-bit sign
extension:
buf[i] is defined as "unsigned char" and the left-shift operation will convert it
to the type of "signed int", so when left-shift buf[i] by 24 bits, the final result
will get the wrong value: 0xffffffff9e3aeaff.
If the left-shift bits are less than 24, then sign extension will not occur. Whereas
the bankwidth of the nor flash we used is 4, therefore this BUG emerges.
Signed-off-by: Pang Xunlei <pang.xunlei@zte.com.cn> Signed-off-by: Zhang Yi <zhang.yi20@zte.com.cn> Signed-off-by: Lu Zhongjun <lu.zhongjun@zte.com.cn> Signed-off-by: Brian Norris <computersforpeace@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
1. During a probe failure (in mtd_device_parse_register?) the command
buffer would not be freed.
2. The command buffer's size is determined based on the 'fast_read'
boolean, but the assignment of fast_read is made after this
allocation. Thus, the buffer may be allocated "too small".
To fix the first, just switch to the devres version of kzalloc.
To fix the second, increase MAX_CMD_SIZE unconditionally. It's not worth
saving a byte to fiddle around with the conditions here.
This problem was reported by Yuhang Wang a while back.
Signed-off-by: Brian Norris <computersforpeace@gmail.com> Reported-by: Yuhang Wang <wangyuhang2014@gmail.com> Reviewed-by: Sourav Poddar <sourav.poddar@ti.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some bright specification writers decided to write this in the ONFI spec
(from ONFI 3.0, Section 3.1):
"The number of blocks and number of pages per block is not required to
be a power of two. In the case where one of these values is not a
power of two, the corresponding address shall be rounded to an
integral number of bits such that it addresses a range up to the
subsequent power of two value. The host shall not access upper
addresses in a range that is shown as not supported."
This breaks every assumption MTD makes about NAND block/chip-size
dimensions -- they *must* be a power of two!
And of course, an enterprising manufacturer has made use of this lovely
freedom. Exhibit A: Micron MT29F32G08CBADAWP
"- Plane size: 2 planes x 1064 blocks per plane
- Device size: 32Gb: 2128 blockss [sic]"
This quickly hits a BUG() in nand_base.c, since the extra dimensions
overflow so we think it's a second chip (on my single-chip setup):
ONFI param page 0 valid
ONFI flash detected
NAND device: Manufacturer ID: 0x2c, Chip ID: 0x44 (Micron MT29F32G08CBADAWP), 4256MiB, page size: 8192, OOB size: 744
------------[ cut here ]------------
kernel BUG at drivers/mtd/nand/nand_base.c:203!
Internal error: Oops - BUG: 0 [#1] SMP ARM
[... trim ...]
[<c02cf3e4>] (nand_select_chip+0x18/0x2c) from [<c02d25c0>] (nand_do_read_ops+0x90/0x424)
[<c02d25c0>] (nand_do_read_ops+0x90/0x424) from [<c02d2dd8>] (nand_read+0x54/0x78)
[<c02d2dd8>] (nand_read+0x54/0x78) from [<c02ad2c8>] (mtd_read+0x84/0xbc)
[<c02ad2c8>] (mtd_read+0x84/0xbc) from [<c02d4b28>] (scan_read.clone.4+0x4c/0x64)
[<c02d4b28>] (scan_read.clone.4+0x4c/0x64) from [<c02d4c88>] (search_bbt+0x148/0x290)
[<c02d4c88>] (search_bbt+0x148/0x290) from [<c02d4ea4>] (nand_scan_bbt+0xd4/0x5c0)
[... trim ...]
---[ end trace 0c9363860d865ff2 ]---
So to fix this, just truncate these dimensions down to the greatest
power-of-2 dimension that is less than or equal to the specified
dimension.
Signed-off-by: Brian Norris <computersforpeace@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When the loop module is loaded, it creates 8 loop devices /dev/loop[0-7].
The devices have no request routine and thus, when they are used without
being assigned, a crash happens.
For example, these commands cause crash (assuming there are no used loop
devices):
If blk_alloc_queue fails, loop_add cleans up, but it doesn't clean up the
identifier allocated with idr_alloc. That causes crash on module unload in
idr_for_each(&loop_index_idr, &loop_exit_cb, NULL); where we attempt to
remove non-existed device with that id.
The IB spec does not guarantee that the opcode is available in error
completions. Hence do not rely on it. See also commit 948d1e889e5b
("IB/srp: Introduce srp_handle_qp_err()").
Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Roland Dreier <roland@purestorage.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If SCSI commands are submitted with a SCSI request timeout that is
lower than the the IB RC timeout, it can happen that the SCSI error
handler has already started device recovery before transport layer
error handling starts. So it can happen that the SCSI error handler
tries to abort a SCSI command after it has been reset by
srp_rport_reconnect().
Tell the SCSI error handler that such commands have finished and that
it is not necessary to continue its recovery strategy for commands
that have been reset by srp_rport_reconnect().
Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Roland Dreier <roland@purestorage.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove an SRP target from the SRP target list before invoking the last
scsi_host_put() call. This change is necessary because that last put
frees the memory that holds the srp_target_port structure.
Commit 7fac33014f54("IB/qib: checkpatch fixes") was overzealous in
removing a simple_strtoul for a parse routine, setup_txselect(). That
routine is required to handle a multi-value string.
Unwind that aspect of the fix.
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by: Roland Dreier <roland@purestorage.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
qib_user_sdma_queue_pkts() gets called with mmap_sem held for
writing. Except for get_user_pages() deep down in
qib_user_sdma_pin_pages() we don't seem to need mmap_sem at all. Even
more interestingly the function qib_user_sdma_queue_pkts() (and also
qib_user_sdma_coalesce() called somewhat later) call copy_from_user()
which can hit a page fault and we deadlock on trying to get mmap_sem
when handling that fault.
So just make qib_user_sdma_pin_pages() use get_user_pages_fast() and
leave mmap_sem locking for mm.
This deadlock has actually been observed in the wild when the node
is under memory pressure.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Roland Dreier <roland@purestorage.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ipath_user_sdma_queue_pkts() gets called with mmap_sem held for
writing. Except for get_user_pages() deep down in
ipath_user_sdma_pin_pages() we don't seem to need mmap_sem at all.
Even more interestingly the function ipath_user_sdma_queue_pkts() (and
also ipath_user_sdma_coalesce() called somewhat later) call
copy_from_user() which can hit a page fault and we deadlock on trying
to get mmap_sem when handling that fault. So just make
ipath_user_sdma_pin_pages() use get_user_pages_fast() and leave
mmap_sem locking for mm.
This deadlock has actually been observed in the wild when the node
is under memory pressure.
Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
[ Merged in fix for call to get_user_pages_fast from Tetsuo Handa
<penguin-kernel@I-love.SAKURA.ne.jp>. - Roland ] Signed-off-by: Roland Dreier <roland@purestorage.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In iSCSI negotiations with initiator CHAP enabled, usernames with
trailing garbage are permitted, because the string comparison only
checks the strlen of the configured username.
e.g. "usernameXXXXX" will be permitted to match "username".
Just check one more byte so the trailing null char is also matched.
Signed-off-by: Eric Seppanen <eric@purestorage.com> Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
projects / people / ms / linux.git / log
