Mikulas Patocka [Fri, 21 Dec 2012 20:23:40 +0000 (20:23 +0000)]
dm raid1: dont use map_context
Don't use map_info any more in dm-raid1.
map_info was used for writes to hold the region number. For this purpose
we add a new field dm_bio_details to dm_raid1_bio_record.
map_info was used for reads to hold a pointer to dm_raid1_bio_record (if
the pointer was non-NULL, bio details were saved; if the pointer was
NULL, bio details were not saved). We use
dm_raid1_bio_record.details->bi_bdev for this purpose. If bi_bdev is
NULL, details were not saved, if bi_bdev is non-NULL, details were
saved.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Mikulas Patocka [Fri, 21 Dec 2012 20:23:38 +0000 (20:23 +0000)]
dm: introduce per_bio_data
Introduce a field per_bio_data_size in struct dm_target.
Targets can set this field in the constructor. If a target sets this
field to a non-zero value, "per_bio_data_size" bytes of auxiliary data
are allocated for each bio submitted to the target. These data can be
used for any purpose by the target and help us improve performance by
removing some per-target mempools.
Per-bio data is accessed with dm_per_bio_data. The
argument data_size must be the same as the value per_bio_data_size in
dm_target.
If the target has a pointer to per_bio_data, it can get a pointer to
the bio with dm_bio_from_per_bio_data() function (data_size must be the
same as the value passed to dm_per_bio_data).
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Mike Snitzer [Fri, 21 Dec 2012 20:23:37 +0000 (20:23 +0000)]
dm kcopyd: add WRITE SAME support to dm_kcopyd_zero
Add WRITE SAME support to dm-io and make it accessible to
dm_kcopyd_zero(). dm_kcopyd_zero() provides an asynchronous interface
whereas the blkdev_issue_write_same() interface is synchronous.
WRITE SAME is a SCSI command that can be leveraged for more efficient
zeroing of a specified logical extent of a device which supports it.
Only a single zeroed logical block is transfered to the target for each
WRITE SAME and the target then writes that same block across the
specified extent.
The dm thin target uses this.
Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Mike Snitzer [Fri, 21 Dec 2012 20:23:37 +0000 (20:23 +0000)]
dm: add WRITE SAME support
WRITE SAME bios have a payload that contain a single page. When
cloning WRITE SAME bios DM has no need to modify the bi_io_vec
attributes (and doing so would be detrimental). DM need only alter the
start and end of the WRITE SAME bio accordingly.
Rather than duplicate __clone_and_map_discard, factor out a common
function that is also used by __clone_and_map_write_same.
Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Mikulas Patocka [Fri, 21 Dec 2012 20:23:36 +0000 (20:23 +0000)]
dm ioctl: use kmalloc if possible
If the parameter buffer is small enough, try to allocate it with kmalloc()
rather than vmalloc().
vmalloc is noticeably slower than kmalloc because it has to manipulate
page tables.
In my tests, on PA-RISC this patch speeds up activation 13 times.
On Opteron this patch speeds up activation by 5%.
This patch introduces a new function free_params() to free the
parameters and this uses new flags that record whether or not vmalloc()
was used and whether or not the input buffer must be wiped after use.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Mike Snitzer [Fri, 21 Dec 2012 20:23:34 +0000 (20:23 +0000)]
dm block manager: reinstate message when validator fails
Reinstate a useful error message when the block manager buffer validator fails.
This was mistakenly eliminated when the block manager was converted to use
dm-bufio.
Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Jonathan Brassow [Fri, 21 Dec 2012 20:23:33 +0000 (20:23 +0000)]
dm raid: round region_size to power of two
If the user does not supply a bitmap region_size to the dm raid target,
a reasonable size is computed automatically. If this is not a power of 2,
the md code will report an error later.
This patch catches the problem early and rounds the region_size to the
next power of two.
Signed-off-by: Jonathan Brassow <jbrassow@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Mikulas Patocka [Fri, 21 Dec 2012 20:23:33 +0000 (20:23 +0000)]
dm snapshot: optimize track_chunk
track_chunk is always called with interrupts enabled. Consequently, we
do not need to save and restore interrupt state in "flags" variable.
This patch changes spin_lock_irqsave to spin_lock_irq and
spin_unlock_irqrestore to spin_unlock_irq.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Mike Snitzer [Fri, 21 Dec 2012 20:23:32 +0000 (20:23 +0000)]
dm thin: emit ignore_discard in status when discards disabled
If "ignore_discard" is specified when creating the thin pool device then
discard support is disabled for that device. The pool device's status
should reflect this fact rather than stating "no_discard_passdown"
(which implies discards are enabled but passdown is disabled).
Reported-by: Zdenek Kabelac <zkabelac@redhat.com> Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Joe Thornber [Fri, 21 Dec 2012 20:23:32 +0000 (20:23 +0000)]
dm persistent data: fix nested btree deletion
When deleting nested btrees, the code forgets to delete the innermost
btree. The thin-metadata code serendipitously compensates for this by
claiming there is one extra layer in the tree.
This patch corrects both problems.
Signed-off-by: Joe Thornber <ejt@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Joe Thornber [Fri, 21 Dec 2012 20:23:31 +0000 (20:23 +0000)]
dm thin: wake worker when discard is prepared
When discards are prepared it is best to directly wake the worker that
will process them. The worker will be woken anyway, via periodic
commit, but there is no reason to not wake_worker here.
Signed-off-by: Joe Thornber <ejt@redhat.com> Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Joe Thornber [Fri, 21 Dec 2012 20:23:31 +0000 (20:23 +0000)]
dm thin: fix race between simultaneous io and discards to same block
There is a race when discard bios and non-discard bios are issued
simultaneously to the same block.
Discard support is expensive for all thin devices precisely because you
have to be careful to quiesce the area you're discarding. DM thin must
handle this conflicting IO pattern (simultaneous non-discard vs discard)
even though a sane application shouldn't be issuing such IO.
The race manifests as follows:
1. A non-discard bio is mapped in thin_bio_map.
This doesn't lock out parallel activity to the same block.
2. A discard bio is issued to the same block as the non-discard bio.
3. The discard bio is locked in a dm_bio_prison_cell in process_discard
to lock out parallel activity against the same block.
4. The non-discard bio's mapping continues and its all_io_entry is
incremented so the bio is accounted for in the thin pool's all_io_ds
which is a dm_deferred_set used to track time locality of non-discard IO.
5. The non-discard bio is finally locked in a dm_bio_prison_cell in
process_bio.
The race can result in deadlock, leaving the block layer hanging waiting
for completion of a discard bio that never completes, e.g.:
The thinp-test-suite's test_discard_random_sectors reliably hits this
deadlock on fast SSD storage.
The fix for this race is that the all_io_entry for a bio must be
incremented whilst the dm_bio_prison_cell is held for the bio's
associated virtual and physical blocks. That cell locking wasn't
occurring early enough in thin_bio_map. This patch fixes this.
Care is taken to always call the new function inc_all_io_entry() with
the relevant cells locked, but they are generally unlocked before
calling issue() to try to avoid holding the cells locked across
generic_submit_request.
Also, now that thin_bio_map may lock bios in a cell, process_bio() is no
longer the only thread that will do so. Because of this we must be sure
to use cell_defer_except() to release all non-holder entries, that
were added by the other thread, because they must be deferred.
This patch depends on "dm thin: replace dm_cell_release_singleton with
cell_defer_except".
Signed-off-by: Joe Thornber <ejt@redhat.com> Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com> Cc: stable@vger.kernel.org
Joe Thornber [Fri, 21 Dec 2012 20:23:31 +0000 (20:23 +0000)]
dm thin: replace dm_cell_release_singleton with cell_defer_except
Change existing users of the function dm_cell_release_singleton to share
cell_defer_except instead, and then remove the now-unused function.
Everywhere that calls dm_cell_release_singleton, the bio in question
is the holder of the cell.
If there are no non-holder entries in the cell then cell_defer_except
behaves exactly like dm_cell_release_singleton. Conversely, if there
*are* non-holder entries then dm_cell_release_singleton must not be used
because those entries would need to be deferred.
Consequently, it is safe to replace use of dm_cell_release_singleton
with cell_defer_except.
This patch is a pre-requisite for "dm thin: fix race between
simultaneous io and discards to same block".
Signed-off-by: Joe Thornber <ejt@redhat.com> Signed-off-by: Mike Snitzer <snitzer@redhat.com> Cc: stable@vger.kernel.org Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Mike Snitzer [Fri, 21 Dec 2012 20:23:30 +0000 (20:23 +0000)]
dm: disable WRITE SAME
WRITE SAME bios are not yet handled correctly by device-mapper so
disable their use on device-mapper devices by setting
max_write_same_sectors to zero.
As an example, a ciphertext device is incompatible because the data
gets changed according to the location at which it written and so the
dm crypt target cannot support it.
Signed-off-by: Mike Snitzer <snitzer@redhat.com> Cc: stable@vger.kernel.org Cc: Milan Broz <mbroz@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
dm ioctl: prevent unsafe change to dm_ioctl data_size
Abort dm ioctl processing if userspace changes the data_size parameter
after we validated it but before we finished copying the data buffer
from userspace.
The dm ioctl parameters are processed in the following sequence:
1. ctl_ioctl() calls copy_params();
2. copy_params() makes a first copy of the fixed-sized portion of the
userspace parameters into the local variable "tmp";
3. copy_params() then validates tmp.data_size and allocates a new
structure big enough to hold the complete data and copies the whole
userspace buffer there;
4. ctl_ioctl() reads userspace data the second time and copies the whole
buffer into the pointer "param";
5. ctl_ioctl() reads param->data_size without any validation and stores it
in the variable "input_param_size";
6. "input_param_size" is further used as the authoritative size of the
kernel buffer.
The problem is that userspace code could change the contents of user
memory between steps 2 and 4. In particular, the data_size parameter
can be changed to an invalid value after the kernel has validated it.
This lets userspace force the kernel to access invalid kernel memory.
The fix is to ensure that the size has not changed at step 4.
This patch shouldn't have a security impact because CAP_SYS_ADMIN is
required to run this code, but it should be fixed anyway.
Mikulas Patocka [Fri, 21 Dec 2012 20:23:30 +0000 (20:23 +0000)]
dm persistent data: rename node to btree_node
This patch fixes a compilation failure on sparc32 by renaming struct node.
struct node is already defined in include/linux/node.h. On sparc32, it
happens to be included through other dependencies and persistent-data
doesn't compile because of conflicting declarations.
Florian Fainelli [Mon, 10 Dec 2012 20:25:32 +0000 (12:25 -0800)]
Input: matrix-keymap - provide proper module license
The matrix-keymap module is currently lacking a proper module license,
add one so we don't have this module tainting the entire kernel. This
issue has been present since commit 1932811f426f ("Input: matrix-keymap
- uninline and prepare for device tree support")
1) Netlink socket dumping had several missing verifications and checks.
In particular, address comparisons in the request byte code
interpreter could access past the end of the address in the
inet_request_sock.
Also, address family and address prefix lengths were not validated
properly at all.
This means arbitrary applications can read past the end of certain
kernel data structures.
Fixes from Neal Cardwell.
2) ip_check_defrag() operates in contexts where we're in the process
of, or about to, input the packet into the real protocols
(specifically macvlan and AF_PACKET snooping).
Unfortunately, it does a pskb_may_pull() which can modify the
backing packet data which is not legal if the SKB is shared. It
very much can be shared in this context.
Deal with the possibility that the SKB is segmented by using
skb_copy_bits().
Fix from Johannes Berg based upon a report by Eric Leblond.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net:
ipv4: ip_check_defrag must not modify skb before unsharing
inet_diag: validate port comparison byte code to prevent unsafe reads
inet_diag: avoid unsafe and nonsensical prefix matches in inet_diag_bc_run()
inet_diag: validate byte code to prevent oops in inet_diag_bc_run()
inet_diag: fix oops for IPv4 AF_INET6 TCP SYN-RECV state
This is a revert of a revert of a revert. In addition, it reverts the
even older i915 change to stop using the __GFP_NO_KSWAPD flag due to the
original commits in linux-next.
It turns out that the original patch really was bogus, and that the
original revert was the correct thing to do after all. We thought we
had fixed the problem, and then reverted the revert, but the problem
really is fundamental: waking up kswapd simply isn't the right thing to
do, and direct reclaim sometimes simply _is_ the right thing to do.
When certain allocations fail, we simply should try some direct reclaim,
and if that fails, fail the allocation. That's the right thing to do
for THP allocations, which can easily fail, and the GPU allocations want
to do that too.
So starting kswapd is sometimes simply wrong, and removing the flag that
said "don't start kswapd" was a mistake. Let's hope we never revisit
this mistake again - and certainly not this many times ;)
Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Rik van Riel <riel@redhat.com> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Johannes Berg [Sun, 9 Dec 2012 23:41:06 +0000 (23:41 +0000)]
ipv4: ip_check_defrag must not modify skb before unsharing
ip_check_defrag() might be called from af_packet within the
RX path where shared SKBs are used, so it must not modify
the input SKB before it has unshared it for defragmentation.
Use skb_copy_bits() to get the IP header and only pull in
everything later.
The same is true for the other caller in macvlan as it is
called from dev->rx_handler which can also get a shared SKB.
Reported-by: Eric Leblond <eric@regit.org> Cc: stable@vger.kernel.org Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
We are going to reinstate the __GFP_NO_KSWAPD flag that has been
removed, the removal reverted, and then removed again. Making this
commit a pointless fixup for a problem that was caused by the removal of
__GFP_NO_KSWAPD flag.
The thing is, we really don't want to wake up kswapd for THP allocations
(because they fail quite commonly under any kind of memory pressure,
including when there is tons of memory free), and these patches were
just trying to fix up the underlying bug: the original removal of
__GFP_NO_KSWAPD in commit c654345924f7 ("mm: remove __GFP_NO_KSWAPD")
was simply bogus.
Neal Cardwell [Sun, 9 Dec 2012 11:09:54 +0000 (11:09 +0000)]
inet_diag: validate port comparison byte code to prevent unsafe reads
Add logic to verify that a port comparison byte code operation
actually has the second inet_diag_bc_op from which we read the port
for such operations.
Previously the code blindly referenced op[1] without first checking
whether a second inet_diag_bc_op struct could fit there. So a
malicious user could make the kernel read 4 bytes beyond the end of
the bytecode array by claiming to have a whole port comparison byte
code (2 inet_diag_bc_op structs) when in fact the bytecode was not
long enough to hold both.
Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Neal Cardwell [Sat, 8 Dec 2012 19:43:23 +0000 (19:43 +0000)]
inet_diag: avoid unsafe and nonsensical prefix matches in inet_diag_bc_run()
Add logic to check the address family of the user-supplied conditional
and the address family of the connection entry. We now do not do
prefix matching of addresses from different address families (AF_INET
vs AF_INET6), except for the previously existing support for having an
IPv4 prefix match an IPv4-mapped IPv6 address (which this commit
maintains as-is).
This change is needed for two reasons:
(1) The addresses are different lengths, so comparing a 128-bit IPv6
prefix match condition to a 32-bit IPv4 connection address can cause
us to unwittingly walk off the end of the IPv4 address and read
garbage or oops.
(2) The IPv4 and IPv6 address spaces are semantically distinct, so a
simple bit-wise comparison of the prefixes is not meaningful, and
would lead to bogus results (except for the IPv4-mapped IPv6 case,
which this commit maintains).
Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Neal Cardwell [Sat, 8 Dec 2012 19:43:22 +0000 (19:43 +0000)]
inet_diag: validate byte code to prevent oops in inet_diag_bc_run()
Add logic to validate INET_DIAG_BC_S_COND and INET_DIAG_BC_D_COND
operations.
Previously we did not validate the inet_diag_hostcond, address family,
address length, and prefix length. So a malicious user could make the
kernel read beyond the end of the bytecode array by claiming to have a
whole inet_diag_hostcond when the bytecode was not long enough to
contain a whole inet_diag_hostcond of the given address family. Or
they could make the kernel read up to about 27 bytes beyond the end of
a connection address by passing a prefix length that exceeded the
length of addresses of the given family.
Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Neal Cardwell [Sat, 8 Dec 2012 19:43:21 +0000 (19:43 +0000)]
inet_diag: fix oops for IPv4 AF_INET6 TCP SYN-RECV state
Fix inet_diag to be aware of the fact that AF_INET6 TCP connections
instantiated for IPv4 traffic and in the SYN-RECV state were actually
created with inet_reqsk_alloc(), instead of inet6_reqsk_alloc(). This
means that for such connections inet6_rsk(req) returns a pointer to a
random spot in memory up to roughly 64KB beyond the end of the
request_sock.
With this bug, for a server using AF_INET6 TCP sockets and serving
IPv4 traffic, an inet_diag user like `ss state SYN-RECV` would lead to
inet_diag_fill_req() causing an oops or the export to user space of 16
bytes of kernel memory as a garbage IPv6 address, depending on where
the garbage inet6_rsk(req) pointed.
Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Johannes Weiner [Thu, 6 Dec 2012 20:23:25 +0000 (15:23 -0500)]
mm: vmscan: fix inappropriate zone congestion clearing
commit c702418f8a2f ("mm: vmscan: do not keep kswapd looping forever due
to individual uncompactable zones") removed zone watermark checks from
the compaction code in kswapd but left in the zone congestion clearing,
which now happens unconditionally on higher order reclaim.
This messes up the reclaim throttling logic for zones with
dirty/writeback pages, where zones should only lose their congestion
status when their watermarks have been restored.
Remove the clearing from the zone compaction section entirely. The
preliminary zone check and the reclaim loop in kswapd will clear it if
the zone is considered balanced.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Rik van Riel <riel@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds [Sat, 8 Dec 2012 00:48:39 +0000 (16:48 -0800)]
vfs: fix O_DIRECT read past end of block device
The direct-IO write path already had the i_size checks in mm/filemap.c,
but it turns out the read path did not, and removing the block size
checks in fs/block_dev.c (commit bbec0270bdd8: "blkdev_max_block: make
private to fs/buffer.c") removed the magic "shrink IO to past the end of
the device" code there.
Fix it by truncating the IO to the size of the block device, like the
write path already does.
NOTE! I suspect the write path would be *much* better off doing it this
way in fs/block_dev.c, rather than hidden deep in mm/filemap.c. The
mm/filemap.c code is extremely hard to follow, and has various
conditionals on the target being a block device (ie the flag passed in
to 'generic_write_checks()', along with a conditional update of the
inode timestamp etc).
It is also quite possible that we should treat this whole block device
size as a "s_maxbytes" issue, and try to make the logic even more
generic. However, in the meantime this is the fairly minimal targeted
fix.
Noted by Milan Broz thanks to a regression test for the cryptsetup
reencrypt tool.
Reported-and-tested-by: Milan Broz <mbroz@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull networking fixes from David Miller:
"Two stragglers:
1) The new code that adds new flushing semantics to GRO can cause SKB
pointer list corruption, manage the lists differently to avoid the
OOPS. Fix from Eric Dumazet.
2) When TCP fast open does a retransmit of data in a SYN-ACK or
similar, we update retransmit state that we shouldn't triggering a
WARN_ON later. Fix from Yuchung Cheng."
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net:
net: gro: fix possible panic in skb_gro_receive()
tcp: bug fix Fast Open client retransmission
Eric Dumazet [Thu, 6 Dec 2012 13:54:59 +0000 (13:54 +0000)]
net: gro: fix possible panic in skb_gro_receive()
commit 2e71a6f8084e (net: gro: selective flush of packets) added
a bug for skbs using frag_list. This part of the GRO stack is rarely
used, as it needs skb not using a page fragment for their skb->head.
Most drivers do use a page fragment, but some of them use GFP_KERNEL
allocations for the initial fill of their RX ring buffer.
napi_gro_flush() overwrite skb->prev that was used for these skb to
point to the last skb in frag_list.
Fix this using a separate field in struct napi_gro_cb to point to the
last fragment.
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Yuchung Cheng [Thu, 6 Dec 2012 08:45:32 +0000 (08:45 +0000)]
tcp: bug fix Fast Open client retransmission
If SYN-ACK partially acks SYN-data, the client retransmits the
remaining data by tcp_retransmit_skb(). This increments lost recovery
state variables like tp->retrans_out in Open state. If loss recovery
happens before the retransmission is acked, it triggers the WARN_ON
check in tcp_fastretrans_alert(). For example: the client sends
SYN-data, gets SYN-ACK acking only ISN, retransmits data, sends
another 4 data packets and get 3 dupacks.
Since the retransmission is not caused by network drop it should not
update the recovery state variables. Further the server may return a
smaller MSS than the cached MSS used for SYN-data, so the retranmission
needs a loop. Otherwise some data will not be retransmitted until timeout
or other loss recovery events.
Signed-off-by: Yuchung Cheng <ycheng@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Mel Gorman [Wed, 5 Dec 2012 22:01:41 +0000 (14:01 -0800)]
tmpfs: fix shared mempolicy leak
This fixes a regression in 3.7-rc, which has since gone into stable.
Commit 00442ad04a5e ("mempolicy: fix a memory corruption by refcount
imbalance in alloc_pages_vma()") changed get_vma_policy() to raise the
refcount on a shmem shared mempolicy; whereas shmem_alloc_page() went
on expecting alloc_page_vma() to drop the refcount it had acquired.
This deserves a rework: but for now fix the leak in shmem_alloc_page().
Hugh: shmem_swapin() did not need a fix, but surely it's clearer to use
the same refcounting there as in shmem_alloc_page(), delete its onstack
mempolicy, and the strange mpol_cond_copy() and __mpol_cond_copy() -
those were invented to let swapin_readahead() make an unknown number of
calls to alloc_pages_vma() with one mempolicy; but since 00442ad04a5e,
alloc_pages_vma() has kept refcount in balance, so now no problem.
Johannes Weiner [Tue, 4 Dec 2012 16:11:31 +0000 (11:11 -0500)]
mm: vmscan: do not keep kswapd looping forever due to individual uncompactable zones
When a zone meets its high watermark and is compactable in case of
higher order allocations, it contributes to the percentage of the node's
memory that is considered balanced.
This requirement, that a node be only partially balanced, came about
when kswapd was desparately trying to balance tiny zones when all bigger
zones in the node had plenty of free memory. Arguably, the same should
apply to compaction: if a significant part of the node is balanced
enough to run compaction, do not get hung up on that tiny zone that
might never get in shape.
When the compaction logic in kswapd is reached, we know that at least
25% of the node's memory is balanced properly for compaction (see
zone_balanced and pgdat_balanced). Remove the individual zone checks
that restart the kswapd cycle.
Otherwise, we may observe more endless looping in kswapd where the
compaction code loops back to reclaim because of a single zone and
reclaim does nothing because the node is considered balanced overall.
Mel Gorman [Thu, 6 Dec 2012 19:01:14 +0000 (19:01 +0000)]
mm: compaction: validate pfn range passed to isolate_freepages_block
Commit 0bf380bc70ec ("mm: compaction: check pfn_valid when entering a
new MAX_ORDER_NR_PAGES block during isolation for migration") added a
check for pfn_valid() when isolating pages for migration as the scanner
does not necessarily start pageblock-aligned.
Since commit c89511ab2f8f ("mm: compaction: Restart compaction from near
where it left off"), the free scanner has the same problem. This patch
makes sure that the pfn range passed to isolate_freepages_block() is
within the same block so that pfn_valid() checks are unnecessary.
In answer to Henrik's wondering why others have not reported this:
reproducing this requires a large enough hole with the right aligment to
have compaction walk into a PFN range with no memmap. Size and
alignment depends in the memory model - 4M for FLATMEM and 128M for
SPARSEMEM on x86. It needs a "lucky" machine.
mmc: sh-mmcif: avoid oops on spurious interrupts (second try)
On some systems, e.g., kzm9g, MMCIF interfaces can produce spurious
interrupts without any active request. To prevent the Oops, that results
in such cases, don't dereference the mmc request pointer until we make
sure, that we are indeed processing such a request.
Heiko Stübner [Sun, 18 Nov 2012 18:50:05 +0000 (19:50 +0100)]
mmc: sdhci-s3c: fix missing clock for gpio card-detect
2abeb5c5ded2 ("Add clk_(enable/disable) in runtime suspend/resume")
added the capability to stop the clocks when the device is runtime
suspended, but forgot to handle the case of the card-detect using
an external gpio.
Therefore in the case that runtime-pm is enabled, start the io-clock
when a card is inserted and stop it again once it is removed.
Signed-off-by: Heiko Stuebner <heiko@sntech.de> Signed-off-by: Chris Ball <cjb@laptop.org>
Linus Torvalds [Thu, 6 Dec 2012 16:42:13 +0000 (08:42 -0800)]
Merge branch 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus
Pull MIPS fixes from Ralf Baechle:
"These are the fixes for the N32 syscall bugs found by Al, an
extraneous break that broke detection for R3000 and R3081 processors,
an endless loop processing signals for kernel task (x86 received the
same fix a while ago) and a fix for transparent huge page which took
ages to track down because it was so hard to come up with a workable
test case."
* 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus:
MIPS: Fix endless loop when processing signals for kernel tasks
MIPS: R3000/R3081: Fix CPU detection.
MIPS: N32: Fix signalfd4 syscall entry point
MIPS: N32: Fix preadv(2) and pwritev(2) entry points.
MIPS: Avoid mcheck by flushing page range in huge_ptep_set_access_flags()
Linus Torvalds [Thu, 6 Dec 2012 16:39:57 +0000 (08:39 -0800)]
Merge branch 'more-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux
Pull build fix from Rusty Russell:
"Tim Gardner <tim.gardner@canonical.com> writes:
> It is $(obj)/oid_registry.o that is dependent on $(obj)/oid_registry_data.c.
> The object file cannot be built until $(obj)/oid_registry_data.c has been
> generated.
>
> A periodic and hard to reproduce parallel build failure is due to
> this incorrect lib/Makefile dependency. The compile error is completely
> disingenuous.
>
> GEN lib/oid_registry_data.c
> Compiling 49 OIDs
> CC lib/oid_registry.o
> gcc: error: lib/oid_registry.c: No such file or directory
> gcc: fatal error: no input files
> compilation terminated.
> make[3]: *** [lib/oid_registry.o] Error 4
I can't reproduce it either. It's completely weird; nothing ever
removes lib/oid_registry.c, so either gcc is giving the wrong message
or it's a weird fs with a very odd race.
But your version is definitely more correct than the previous one,
so..."
* 'more-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux:
lib/Makefile: Fix oid_registry build dependency
Linus Torvalds [Thu, 6 Dec 2012 16:29:08 +0000 (08:29 -0800)]
Merge branch 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux
Pull module signing fixes from Rusty Russell:
"David gave me these a month ago, during my git workflow churn :("
* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux:
ASN.1: Fix an indefinite length skip error
MODSIGN: Don't use enum-type bitfields in module signature info block
Tim Gardner [Tue, 4 Dec 2012 19:52:28 +0000 (12:52 -0700)]
lib/Makefile: Fix oid_registry build dependency
It is $(obj)/oid_registry.o that is dependent on $(obj)/oid_registry_data.c.
The object file cannot be built until $(obj)/oid_registry_data.c has been
generated.
A periodic and hard to reproduce parallel build failure is due to
this incorrect lib/Makefile dependency. The compile error is completely
disingenuous.
GEN lib/oid_registry_data.c
Compiling 49 OIDs
CC lib/oid_registry.o
gcc: error: lib/oid_registry.c: No such file or directory
gcc: fatal error: no input files
compilation terminated.
make[3]: *** [lib/oid_registry.o] Error 4
Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Akinobu Mita <akinobu.mita@gmail.com> Cc: Michel Lespinasse <walken@google.com> Cc: David Howells <dhowells@redhat.com> Cc: "David S. Miller" <davem@davemloft.net> Signed-off-by: Tim Gardner <tim.gardner@canonical.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
MIPS: Fix endless loop when processing signals for kernel tasks
The problem occurs [1] when a kernel-mode task returns from a system
call with a pending signal.
A real-life scenario is a child of 'khelper' returning from a failed
kernel_execve() in ____call_usermodehelper() [ kernel/kmod.c ].
kernel_execve() fails due to a pending SIGKILL, which is the result of
"kill -9 -1" (at least, busybox's init does it upon reboot).
The loop is as follows:
* syscall_exit_work:
- work_pending: // start_of_the_loop
- work_notifysig:
- do_notify_resume()
- do_signal()
- if (!user_mode(regs)) return;
- resume_userspace // TIF_SIGPENDING is still set
- work_pending // so we call work_pending => goto
// start_of_the_loop
More information can be found in another LKML thread:
http://www.serverphorums.com/read.php?12,457826
[1] The problem was also reproduced on !CONFIG_VM86 x86, and the
following fix was accepted.
David Howells [Mon, 22 Oct 2012 14:05:55 +0000 (15:05 +0100)]
ASN.1: Fix an indefinite length skip error
Fix an error in asn1_find_indefinite_length() whereby small definite length
elements of size 0x7f are incorrecly classified as non-small. Without this
fix, an error will be given as the length of the length will be perceived as
being very much greater than the maximum supported size.
Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
David Howells [Mon, 22 Oct 2012 14:05:48 +0000 (15:05 +0100)]
MODSIGN: Don't use enum-type bitfields in module signature info block
Don't use enum-type bitfields in the module signature info block as we can't be
certain how the compiler will handle them. As I understand it, it is arch
dependent, and it is possible for the compiler to rearrange them based on
endianness and to insert a byte of padding to pad the three enums out to four
bytes.
Instead use u8 fields for these, which the compiler should emit in the right
order without padding.
Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Thomas Gleixner [Tue, 4 Dec 2012 17:59:34 +0000 (18:59 +0100)]
watchdog: Fix CPU hotplug regression
Norbert reported:
"3.7-rc6 booted with nmi_watchdog=0 fails to suspend to RAM or
offline CPUs. It's reproducable with a KVM guest and physical
system."
The reason is that commit bcd951cf(watchdog: Use hotplug thread
infrastructure) missed to take this into account. So the cpu offline
code gets stuck in the teardown function because it accesses non
initialized data structures.
Add a check for watchdog_enabled into that path to cure the issue.
Linus Torvalds [Tue, 4 Dec 2012 17:32:12 +0000 (09:32 -0800)]
Merge branch 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux
Pull module fixes from Rusty Russell:
"Module signing build fixes for blackfin and metag"
* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux:
modsign: add symbol prefix to certificate list
linux/kernel.h: define SYMBOL_PREFIX
Linus Torvalds [Tue, 4 Dec 2012 17:15:51 +0000 (09:15 -0800)]
Merge tag 'upstream-3.7-rc9' of git://git.infradead.org/linux-ubi
Pull UBI changes from Artem Bityutskiy:
"Fixes for 2 brown-paperbag bugs introduced this merge window by the
fastmap code:
1. The UBI background thread got stuck when a bit-flip happened
because free LEBs was not removed from the "free" tree when we
started using it.
2. I/O debugging checks did not work because we called a sleeping
function in atomic context."
* tag 'upstream-3.7-rc9' of git://git.infradead.org/linux-ubi:
UBI: dont call ubi_self_check_all_ff() in __wl_get_peb()
UBI: remove PEB from free tree in get_peb_for_wl()
Linus Torvalds [Tue, 4 Dec 2012 17:02:45 +0000 (09:02 -0800)]
Merge branch 'for-3.7-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq
Pull workqueue fixes from Tejun Heo:
"So, safe fixes my ass.
Commit 8852aac25e79 ("workqueue: mod_delayed_work_on() shouldn't queue
timer on 0 delay") had the side-effect of performing delayed_work
sanity checks even when @delay is 0, which should be fine for any sane
use cases.
Unfortunately, megaraid was being overly ingenious. It seemingly
wanted to use cancel_delayed_work_sync() before cancel_work_sync() was
introduced, but didn't want to waste the space for full delayed_work
as it was only going to use 0 @delay. So, it only allocated space for
struct work_struct and then cast it to struct delayed_work and passed
it into delayed_work functions - truly awesome engineering tradeoff to
save some bytes.
Xiaotian fixed it by making megraid allocate full delayed_work for
now. It should be converted to use work_struct and cancel_work_sync()
but I think we better do that after 3.7.
I added another commit to change BUG_ON()s in __queue_delayed_work()
to WARN_ON_ONCE()s so that the kernel doesn't crash even if there are
more such abuses."
* 'for-3.7-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
workqueue: convert BUG_ON()s in __queue_delayed_work() to WARN_ON_ONCE()s
megaraid: fix BUG_ON() from incorrect use of delayed work
Ralf Baechle [Tue, 4 Dec 2012 16:40:44 +0000 (17:40 +0100)]
MIPS: N32: Fix preadv(2) and pwritev(2) entry points.
By using the native syscall entry point the kernel was also expecting
64-bit iovec structures.
This is broken since ddd9e91b71072b8ebe89311c3a44b077defa1756 [preadv/
pwritev: MIPS: Add preadv(2) and pwritev(2) syscalls.] which originally
added these two syscalls. I walked through piles of code, including
libc and couldn't find anything that would have worked around the issue
so this change the API to what it should always have been.
Pull sparc fixes from David Miller:
"Two small fixes for Sparc, nobody uses sparc, so these are low risk :-)
1) Piggyback is too picky about the symbol types that _start and _end
have in the final kernel image, and it thus breaks with newer
binutils. Future proof by getting rid of the symbol type checks.
2) exit_group() should kill register windows on sparc64 the same way
we do for plain exit(). Thanks to Al Viro for spotting this."
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc:
sparc: Fix piggyback with newer binutils.
sparc64: exit_group should kill register windows just like plain exit.
Linus Torvalds [Tue, 4 Dec 2012 16:25:11 +0000 (08:25 -0800)]
vfs: avoid "attempt to access beyond end of device" warnings
The block device access simplification that avoided accessing the (racy)
block size information (commit bbec0270bdd8: "blkdev_max_block: make
private to fs/buffer.c") no longer checks the maximum block size in the
block mapping path.
That was _almost_ as simple as just removing the code entirely, because
the readers and writers all check the size of the device anyway, so
under normal circumstances it "just worked".
However, the block size may be such that the end of the device may
straddle one single buffer_head. At which point we may still want to
access the end of the device, but the buffer we use to access it
partially extends past the end.
The 'bd_set_size()' function intentionally sets the block size to avoid
this, but mounting the device - or setting the block size by hand to
some other value - can modify that block size.
So instead, teach 'submit_bh()' about the special case of the buffer
head straddling the end of the device, and turning such an access into a
smaller IO access, avoiding the problem.
This, btw, also means that unlike before, we can now access the whole
device regardless of device block size setting. So now, even if the
device size is only 512-byte aligned, we can read and write even the
last sector even when having a much bigger block size for accessing the
rest of the device.
So with this, we could now get rid of the 'bd_set_size()' block size
code entirely - resulting in faster IO for the common case - but that
would be a separate patch.
Reported-and-tested-by: Romain Francoise <romain@orebokech.com> Reporeted-and-tested-by: Meelis Roos <mroos@linux.ee> Reported-by: Tony Luck <tony.luck@intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Tejun Heo [Tue, 4 Dec 2012 15:40:39 +0000 (07:40 -0800)]
workqueue: convert BUG_ON()s in __queue_delayed_work() to WARN_ON_ONCE()s
8852aac25e ("workqueue: mod_delayed_work_on() shouldn't queue timer on
0 delay") unexpectedly uncovered a very nasty abuse of delayed_work in
megaraid - it allocated work_struct, casted it to delayed_work and
then pass that into queue_delayed_work().
Previously, this was okay because 0 @delay short-circuited to
queue_work() before doing anything with delayed_work. 8852aac25e
moved 0 @delay test into __queue_delayed_work() after sanity check on
delayed_work making megaraid trigger BUG_ON().
Although megaraid is already fixed by c1d390d8e6 ("megaraid: fix
BUG_ON() from incorrect use of delayed work"), this patch converts
BUG_ON()s in __queue_delayed_work() to WARN_ON_ONCE()s so that such
abusers, if there are more, trigger warning but don't crash the
machine.
David Daney [Mon, 3 Dec 2012 20:44:26 +0000 (12:44 -0800)]
MIPS: Avoid mcheck by flushing page range in huge_ptep_set_access_flags()
Problem:
1) Huge page mapping of anonymous memory is initially invalid. Will be
faulted in by copy-on-write mechanism.
2) Userspace attempts store at the end of the huge mapping.
3) TLB Refill exception handler fill TLB with a normal (4K sized)
invalid page at the end of the huge mapping virtual address range.
4) Userspace restarted, and re-attempts the store at the end of the
huge mapping.
5) Page from #3 is invalid, we get a fault and go to the hugepage
fault handler. This tries to map a huge page and calls
huge_ptep_set_access_flags() to install the mapping.
6) We just call the generic ptep_set_access_flags() to set up the page
tables, but the flush there assumes a normal (4K sized) page and
only tries to flush the first part of the huge page virtual address
out of the TLB, since the existing entry from step #3 doesn't
conflict, nothing is flushed.
7) We attempt to load the mapping into the TLB, but because it
conflicts with the entry from step #3, we get a Machine Check
exception.
The fix: Flush the entire rage covered by the huge page in
huge_ptep_set_access_flags(), and remove the optimization in
local_flush_tlb_range() so that the flush actually does the correct
thing.
Xiaotian Feng [Tue, 4 Dec 2012 11:33:54 +0000 (19:33 +0800)]
megaraid: fix BUG_ON() from incorrect use of delayed work
megaraid use INIT_WORK to declare a hotplug_work, but cast the
hotplug_work from work_struct to delayed_work and
schedule_delayed_work on it. This is very dangerous, as other part of
delayed_work might be kernel memories allocated by others.
With commit 8852aac ("workqueue: mod_delayed_work_on() shouldn't queue
timer on 0 delay"), schedule_delayed_work() will check dwork->timer
before queue_work even when @delay is 0, this causes megaraid code to
hit the BUG_ON() in workqueue code. Change megaraid code to use
delayed work.
UBI: dont call ubi_self_check_all_ff() in __wl_get_peb()
As ubi_self_check_all_ff() might sleep we are not allowed
to call it from atomic context.
For now we call it only from ubi_wl_get_peb().
There are some code paths where it would also make sense,
but these paths are currently atomic and only enabled
when fastmap is used.
Signed-off-by: Richard Weinberger <richard@nod.at> Signed-off-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
Linus Torvalds [Mon, 3 Dec 2012 19:13:32 +0000 (11:13 -0800)]
Merge branch 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
Pull media fixes from Mauro Carvalho Chehab:
"Some driver fixes for s5p/exynos (mostly race fixes)"
* 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media:
[media] s5p-mfc: Handle multi-frame input buffer
[media] s5p-mfc: Bug fix of timestamp/timecode copy mechanism
[media] exynos-gsc: Add missing video device vfl_dir flag initialization
[media] exynos-gsc: Fix settings for input and output image RGB type
[media] exynos-gsc: Don't use mutex_lock_interruptible() in device release()
[media] fimc-lite: Don't use mutex_lock_interruptible() in device release()
[media] s5p-fimc: Don't use mutex_lock_interruptible() in device release()
[media] s5p-fimc: Prevent race conditions during subdevs registration
Al Viro [Mon, 3 Dec 2012 18:15:30 +0000 (18:15 +0000)]
[parisc] open(2) compat bug
In commit 9d73fc2d641f ("open*(2) compat fixes (s390, arm64)") I said:
>
> The usual rules for open()/openat()/open_by_handle_at() are
> 1) native 32bit - don't force O_LARGEFILE in flags
> 2) native 64bit - force O_LARGEFILE in flags
> 3) compat on 64bit host - as for native 32bit
> 4) native 32bit ABI for 64bit system (mips/n32, x86/x32) - as for native 64bit
>
> There are only two exceptions - s390 compat has open() forcing O_LARGEFILE and
> arm64 compat has open_by_handle_at() doing the same thing. The same binaries
> on native host (s390/31 and arm resp.) will *not* force O_LARGEFILE, so IMO
> both are emulation bugs.
Three exceptions, actually - parisc open() is another case like that.
Native 32bit won't force O_LARGEFILE, the same binary on parisc64 will.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Between commits 8323f26ce342 ("sched: Fix race in task_group()") and 800d4d30c8f2 ("sched, autogroup: Stop going ahead if autogroup is
disabled"), autogroup is a wreck.
With both applied, all you have to do to crash a box is disable
autogroup during boot up, then reboot.. boom, NULL pointer dereference
due to commit 800d4d30c8f2 not allowing autogroup to move things, and
commit 8323f26ce342 making that the only way to switch runqueues:
Linus Torvalds [Mon, 3 Dec 2012 18:53:25 +0000 (10:53 -0800)]
Merge branch 'block-dev'
Merge 'block-dev' branch.
I was going to just mark everything here for stable and leave it to the
3.8 merge window, but having decided on doing another -rc, I migth as
well merge it now.
This removes the bd_block_size_semaphore semaphore that was added in
this release to fix a race condition between block size changes and
block IO, and replaces it with atomicity guaratees in fs/buffer.c
instead, along with simplifying fs/block-dev.c.
This removes more lines than it adds, makes the code generally simpler,
and avoids the latency/rt issues that the block size semaphore
introduced for mount.
I'm not happy with the timing, but it wouldn't be much better doing this
during the merge window and then having some delayed back-port of it
into stable.
* block-dev:
blkdev_max_block: make private to fs/buffer.c
direct-io: don't read inode->i_blkbits multiple times
blockdev: remove bd_block_size_semaphore again
fs/buffer.c: make block-size be per-page and protected by the page lock
James Hogan [Fri, 23 Nov 2012 12:08:16 +0000 (12:08 +0000)]
modsign: add symbol prefix to certificate list
Add the arch symbol prefix (if applicable) to the asm definition of
modsign_certificate_list and modsign_certificate_list_end. This uses the
recently defined SYMBOL_PREFIX which is derived from
CONFIG_SYMBOL_PREFIX.
This fixes the build of module signing on the blackfin and metag
architectures.
Signed-off-by: James Hogan <james.hogan@imgtec.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: David Howells <dhowells@redhat.com> Cc: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
James Hogan [Fri, 23 Nov 2012 12:08:15 +0000 (12:08 +0000)]
linux/kernel.h: define SYMBOL_PREFIX
Define SYMBOL_PREFIX to be the same as CONFIG_SYMBOL_PREFIX if set by
the architecture, or "" otherwise. This avoids the need for ugly #ifdefs
whenever symbols are referenced in asm blocks.
Signed-off-by: James Hogan <james.hogan@imgtec.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Joe Perches <joe@perches.com> Cc: Paul Gortmaker <paul.gortmaker@windriver.com> Cc: Jean Delvare <khali@linux-fr.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Al Viro [Sun, 2 Dec 2012 17:55:07 +0000 (17:55 +0000)]
open*(2) compat fixes (s390, arm64)
The usual rules for open()/openat()/open_by_handle_at() are
1) native 32bit - don't force O_LARGEFILE in flags
2) native 64bit - force O_LARGEFILE in flags
3) compat on 64bit host - as for native 32bit
4) native 32bit ABI for 64bit system (mips/n32, x86/x32) - as for
native 64bit
There are only two exceptions - s390 compat has open() forcing
O_LARGEFILE and arm64 compat has open_by_handle_at() doing the same
thing. The same binaries on native host (s390/31 and arm resp.) will
*not* force O_LARGEFILE, so IMO both are emulation bugs.
Objections? The fix is obvious...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds [Sun, 2 Dec 2012 01:55:13 +0000 (17:55 -0800)]
Merge branch 'for-3.7-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq
Pull late workqueue fixes from Tejun Heo:
"Unfortunately, I have two really late fixes. One was for a
long-standing bug and queued for 3.8 but I found out about a
regression introduced during 3.7-rc1 two days ago, so I'm sending out
the two fixes together.
The first (long-standing) one is rescuer_thread() entering exit path
w/ TASK_INTERRUPTIBLE. It only triggers on workqueue destructions
which isn't very frequent and the exit path can usually survive being
called with TASK_INTERRUPT, so it was hidden pretty well. Apparently,
if you're reiserfs, this could lead to the exiting kthread sleeping
indefinitely holding a mutex, which is never good.
The fix is simple - restoring TASK_RUNNING before returning from the
kthread function.
The second one is introduced by the new mod_delayed_work().
mod_delayed_work() was missing special case handling for 0 delay.
Instead of queueing the work item immediately, it queued the timer
which expires on the closest next tick. Some users of the new
function converted from "[__]cancel_delayed_work() +
queue_delayed_work()" combination became unhappy with the extra delay.
Block unplugging led to noticeably higher number of context switches
and intel 6250 wireless failed to associate with WPA-Enterprise
network. The fix, again, is fairly simple. The 0 delay special case
logic from queue_delayed_work_on() should be moved to
__queue_delayed_work() which is shared by both queue_delayed_work_on()
and mod_delayed_work_on().
The first one is difficult to trigger and the failure mode for the
latter isn't completely catastrophic, so missing these two for 3.7
wouldn't make it a disastrous release, but both bugs are nasty and the
fixes are fairly safe"
* 'for-3.7-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
workqueue: mod_delayed_work_on() shouldn't queue timer on 0 delay
workqueue: exit rescuer_thread() as TASK_RUNNING
Eric Dumazet [Sat, 1 Dec 2012 13:07:02 +0000 (13:07 +0000)]
tcp: fix crashes in do_tcp_sendpages()
Recent network changes allowed high order pages being used
for skb fragments.
This uncovered a bug in do_tcp_sendpages() which was assuming its caller
provided an array of order-0 page pointers.
We only have to deal with a single page in this function, and its order
is irrelevant.
Reported-by: Willy Tarreau <w@1wt.eu> Tested-by: Willy Tarreau <w@1wt.eu> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Tejun Heo [Sun, 2 Dec 2012 00:23:42 +0000 (16:23 -0800)]
workqueue: mod_delayed_work_on() shouldn't queue timer on 0 delay
8376fe22c7 ("workqueue: implement mod_delayed_work[_on]()")
implemented mod_delayed_work[_on]() using the improved
try_to_grab_pending(). The function is later used, among others, to
replace [__]candel_delayed_work() + queue_delayed_work() combinations.
Unfortunately, a delayed_work item w/ zero @delay is handled slightly
differently by mod_delayed_work_on() compared to
queue_delayed_work_on(). The latter skips timer altogether and
directly queues it using queue_work_on() while the former schedules
timer which will expire on the closest tick. This means, when @delay
is zero, that [__]cancel_delayed_work() + queue_delayed_work_on()
makes the target item immediately executable while
mod_delayed_work_on() may induce delay of upto a full tick.
This somewhat subtle difference breaks some of the converted users.
e.g. block queue plugging uses delayed_work for deferred processing
and uses mod_delayed_work_on() when the queue needs to be immediately
unplugged. The above problem manifested as noticeably higher number
of context switches under certain circumstances.
The difference in behavior was caused by missing special case handling
for 0 delay in mod_delayed_work_on() compared to
queue_delayed_work_on(). Joonsoo Kim posted a patch to add it -
("workqueue: optimize mod_delayed_work_on() when @delay == 0")[1].
The patch was queued for 3.8 but it was described as optimization and
I missed that it was a correctness issue.
As both queue_delayed_work_on() and mod_delayed_work_on() use
__queue_delayed_work() for queueing, it seems that the better approach
is to move the 0 delay special handling to the function instead of
duplicating it in mod_delayed_work_on().
Fix the problem by moving 0 delay special case handling from
queue_delayed_work_on() to __queue_delayed_work(). This replaces
Joonsoo's patch.
Mike Galbraith [Wed, 28 Nov 2012 06:17:18 +0000 (07:17 +0100)]
workqueue: exit rescuer_thread() as TASK_RUNNING
A rescue thread exiting TASK_INTERRUPTIBLE can lead to a task scheduling
off, never to be seen again. In the case where this occurred, an exiting
thread hit reiserfs homebrew conditional resched while holding a mutex,
bringing the box to its knees.
Linus Torvalds [Sat, 1 Dec 2012 21:29:55 +0000 (13:29 -0800)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull vfs fixes from Al Viro:
"A bunch of fixes; the last one is this cycle regression, the rest are
-stable fodder."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
fix off-by-one in argument passed by iterate_fd() to callbacks
lookup_one_len: don't accept . and ..
cifs: get rid of blind d_drop() in readdir
nfs_lookup_revalidate(): fix a leak
don't do blind d_drop() in nfs_prime_dcache()
Linus Torvalds [Sat, 1 Dec 2012 21:08:36 +0000 (13:08 -0800)]
Merge branch 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull RCU fix from Ingo Molnar:
"Fix leaking RCU extended quiescent state, which might trigger warnings
and mess up the extended quiescent state tracking logic into thinking
that we are in "RCU user mode" while we aren't."
* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
rcu: Fix unrecovered RCU user mode in syscall_trace_leave()
Linus Torvalds [Sat, 1 Dec 2012 21:07:48 +0000 (13:07 -0800)]
Merge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf fixes from Ingo Molnar:
"This is mostly about unbreaking architectures that took the UAPI
changes in the v3.7 cycle, plus misc fixes."
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf kvm: Fix building perf kvm on non x86 arches
perf kvm: Rename perf_kvm to perf_kvm_stat
perf: Make perf build for x86 with UAPI disintegration applied
perf powerpc: Use uapi/unistd.h to fix build error
tools: Pass the target in descend
tools: Honour the O= flag when tool build called from a higher Makefile
tools: Define a Makefile function to do subdir processing
x86: Export asm/{svm.h,vmx.h,perf_regs.h}
perf tools: Fix strbuf_addf() when the buffer needs to grow
perf header: Fix numa topology printing
perf, powerpc: Fix hw breakpoints returning -ENOSPC
projects / mirror_ubuntu-eoan-kernel.git / log