commit 66018506e1 (ip: Router Alert RCU conversion) introduced RCU
lookups to ip_call_ra_chain(). It missed proper deinit phase :
When ip_ra_control() deletes an ip_ra_chain, it should make sure
ip_call_ra_chain() users can not start to use socket during the rcu
grace period. It should also delay the sock_put() after the grace
period, or we risk a premature socket freeing and corruptions, as
raw sockets are not rcu protected yet.
This delay avoids using expensive atomic_inc_not_zero() in
ip_call_ra_chain().
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use call_rcu rather than synchronize_rcu.
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- dev_get_by_flags() changed to dev_get_by_flags_rcu()
- ipv6_sock_ac_join() dont touch dev & idev refcounts
- ipv6_sock_ac_drop() dont touch dev & idev refcounts
- ipv6_sock_ac_close() dont touch dev & idev refcounts
- ipv6_dev_ac_dec() dount touch idev refcount
- ipv6_chk_acast_addr() dont touch idev refcount
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Straightforward conversion to RCU.
One rwlock becomes a spinlock, and is static.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch address a serious performance issue in reading the
TCP sockets table (/proc/net/tcp).
Reading the full table is done by a number of sequential read
operations. At each read operation, a seek is done to find the
last socket that was previously read. This seek operation requires
that the sockets in the table need to be counted up to the current
file position, and to count each of these requires taking a lock for
each non-empty bucket. The whole algorithm is O(n^2).
The fix is to cache the last bucket value, offset within the bucket,
and the file position returned by the last read operation. On the
next sequential read, the bucket and offset are used to find the
last read socket immediately without needing ot scan the previous
buckets the table. This algorithm t read the whole table is O(n).
The improvement offered by this patch is easily show by performing
cat'ing /proc/net/tcp on a machine with a lot of connections. With
about 182K connections in the table, I see the following:
- Without patch
time cat /proc/net/tcp > /dev/null
real 1m56.729s
user 0m0.214s
sys 1m56.344s
- With patch
time cat /proc/net/tcp > /dev/null
real 0m0.894s
user 0m0.290s
sys 0m0.594s
Signed-off-by: Tom Herbert <therbert@google.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
v2: changed bonding module version, modified to apply on top of changes
from previous patch in series, and updated documentation to elaborate on
multiqueue awareness that now exists in bonding driver.
This patch give the user the ability to control the output slave for
round-robin and active-backup bonding. Similar functionality was
discussed in the past, but Jay Vosburgh indicated he would rather see a
feature like this added to existing modes rather than creating a
completely new mode. Jay's thoughts as well as Neil's input surrounding
some of the issues with the first implementation pushed us toward a
design that relied on the queue_mapping rather than skb marks.
Round-robin and active-backup modes were chosen as the first users of
this slave selection as they seemed like the most logical choices when
considering a multi-switch environment.
Round-robin mode works without any modification, but active-backup does
require inclusion of the first patch in this series and setting
the 'all_slaves_active' flag. This will allow reception of unicast traffic on
any of the backup interfaces.
This was tested with IPv4-based filters as well as VLAN-based filters
with good results.
More information as well as a configuration example is available in the
patch to Documentation/networking/bonding.txt.
Signed-off-by: Andy Gospodarek <andy@greyhouse.net>
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We can avoid an unecessary cache miss by checking if the skb is non-linear
before accessing gso_size/gso_type in skb_warn_if_lro, the same can also be
done to avoid a cache miss on nr_frags if data_len is 0.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
xfrm triggers a warning if dst_pop() drops a refcount
on a noref dst. This patch changes dst_pop() to
skb_dst_pop(). skb_dst_pop() drops the refcnt only
on a refcounted dst. Also we don't clone the child
dst_entry, so it is not refcounted and we can use
skb_dst_set_noref() in xfrm_output_one().
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
cleanup patch.
Use new __packed annotation in net/ and include/
(except netfilter)
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Christoph Lameter mentioned that packets could be dropped in input path
because of rp_filter settings, without any SNMP counter being
incremented. System administrator can have a hard time to track the
problem.
This patch introduces a new counter, LINUX_MIB_IPRPFILTER, incremented
each time we drop a packet because Reverse Path Filter triggers.
(We receive an IPv4 datagram on a given interface, and find the route to
send an answer would use another interface)
netstat -s | grep IPReversePathFilter
IPReversePathFilter: 21714
Reported-by: Christoph Lameter <cl@linux-foundation.org>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
What this patch does is it removes two receive frame hooks (for bridge and for
macvlan) from __netif_receive_skb. These are replaced them with a single
hook for both. It only supports one hook per device because it makes no
sense to do bridging and macvlan on the same device.
Then a network driver (of virtual netdev like macvlan or bridge) can register
an rx_handler for needed net device.
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are more than a dozen occurrences of following code in the
IPv6 stack:
if (opt && opt->srcrt) {
struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt;
ipv6_addr_copy(&final, &fl.fl6_dst);
ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
final_p = &final;
}
Replace those with a helper. Note that the helper overrides final_p
in all cases. This is ok as final_p was previously initialized to
NULL when declared.
Signed-off-by: Arnaud Ebalard <arno@natisbad.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds a setting, PACKET_TIMESTAMP, to specify the packet
timestamp source that is exported to capture utilities like tcpdump by
packet_mmap.
PACKET_TIMESTAMP accepts the same integer bit field as
SO_TIMESTAMPING. However, only the SOF_TIMESTAMPING_SYS_HARDWARE and
SOF_TIMESTAMPING_RAW_HARDWARE values are currently recognized by
PACKET_TIMESTAMP. SOF_TIMESTAMPING_SYS_HARDWARE takes precedence over
SOF_TIMESTAMPING_RAW_HARDWARE if both bits are set.
If PACKET_TIMESTAMP is not set, a software timestamp generated inside
the networking stack is used (the behavior before this setting was
added).
Signed-off-by: Scott McMillan <scott.a.mcmillan@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use read_pnet() and write_pnet() to reduce number of ifdef CONFIG_NET_NS
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When many cpus compete for sending frames on a given qdisc, the qdisc
spinlock suffers from very high contention.
The cpu owning __QDISC_STATE_RUNNING bit has same priority to acquire
the lock, and cannot dequeue packets fast enough, since it must wait for
this lock for each dequeued packet.
One solution to this problem is to force all cpus spinning on a second
lock before trying to get the main lock, when/if they see
__QDISC_STATE_RUNNING already set.
The owning cpu then compete with at most one other cpu for the main
lock, allowing for higher dequeueing rate.
Based on a previous patch from Alexander Duyck. I added the heuristic to
avoid the atomic in fast path, and put the new lock far away from the
cache line used by the dequeue worker. Also try to release the busylock
lock as late as possible.
Tests with following script gave a boost from ~50.000 pps to ~600.000
pps on a dual quad core machine (E5450 @3.00GHz), tg3 driver.
(A single netperf flow can reach ~800.000 pps on this platform)
for j in `seq 0 3`; do
for i in `seq 0 7`; do
netperf -H 192.168.0.1 -t UDP_STREAM -l 60 -N -T $i -- -m 6 &
done
done
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
__QDISC_STATE_RUNNING is always changed while qdisc lock is held.
We can avoid two atomic operations in xmit path, if we move this bit in
a new __state container.
Location of this __state container is carefully chosen so that fast path
only dirties one qdisc cache line.
THROTTLED bit could later be moved into this __state location too, to
avoid dirtying first qdisc cache line.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Define three helpers to manipulate QDISC_STATE_RUNNIG flag, that a
second patch will move on another location.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Correct sk_forward_alloc handling for error_queue would need to use a
backlog of frames that softirq handler could not deliver because socket
is owned by user thread. Or extend backlog processing to be able to
process normal and error packets.
Another possibility is to not use mem charge for error queue, this is
what I implemented in this patch.
Note: this reverts commit 29030374
(net: fix sk_forward_alloc corruptions), since we dont need to lock
socket anymore.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
commit f3c5c1bfd4 (netfilter: xtables: make ip_tables reentrant)
introduced a performance regression, because stackptr array is shared by
all cpus, adding cache line ping pongs. (16 cpus share a 64 bytes cache
line)
Fix this using alloc_percpu()
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-By: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Currently such notifications are only generated when the device comes up or the
address changes. However one use case for these notifications is to enable
faster network recovery after a virtual machine migration (by causing switches
to relearn their MAC tables). A migration appears to the network stack as a
temporary loss of carrier and therefore does not trigger either of the current
conditions. Rather than adding carrier up as a trigger (which can cause issues
when interfaces a flapping) simply add an interface which the driver can use
to explicitly trigger the notification.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Stephen Hemminger <shemminger@linux-foundation.org>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Cc: stable@kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
* 'slub/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg/slab-2.6:
SLUB: Allow full duplication of kmalloc array for 390
slub: move kmem_cache_node into it's own cacheline
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
mm: export generic_pipe_buf_*() to modules
fuse: support splice() reading from fuse device
fuse: allow splice to move pages
mm: export remove_from_page_cache() to modules
mm: export lru_cache_add_*() to modules
fuse: support splice() writing to fuse device
fuse: get page reference for readpages
fuse: use get_user_pages_fast()
fuse: remove unneeded variable
* 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/sameo/mfd-2.6: (47 commits)
mfd: Rename twl5031 sih modules
mfd: Storage class for timberdale should be before const qualifier
mfd: Remove unneeded and dangerous clearing of clientdata
mfd: New AB8500 driver
gpio: Fix inverted rdc321x gpio data out registers
mfd: Change rdc321x resources flags to IORESOURCE_IO
mfd: Move pcf50633 irq related functions to its own file.
mfd: Use threaded irq for pcf50633
mfd: pcf50633-adc: Fix potential race in pcf50633_adc_sync_read
mfd: Fix pcf50633 bitfield logic in interrupt handler
gpio: rdc321x needs to select MFD_CORE
mfd: Use menuconfig for quicker config editing
ARM: AB3550 board configuration and irq for U300
mfd: AB3550 core driver
mfd: AB3100 register access change to abx500 API
mfd: Renamed ab3100.h to abx500.h
gpio: Add TC35892 GPIO driver
mfd: Add Toshiba's TC35892 MFD core
mfd: Delay to mask tsc irq in max8925
mfd: Remove incorrect wm8350 kfree
...
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs-2.6:
quota: Convert quota statistics to generic percpu_counter
ext3 uses rb_node = NULL; to zero rb_root.
quota: Fixup dquot_transfer
reiserfs: Fix resuming of quotas on remount read-write
pohmelfs: Remove dead quota code
ufs: Remove dead quota code
udf: Remove dead quota code
quota: rename default quotactl methods to dquot_
quota: explicitly set ->dq_op and ->s_qcop
quota: drop remount argument to ->quota_on and ->quota_off
quota: move unmount handling into the filesystem
quota: kill the vfs_dq_off and vfs_dq_quota_on_remount wrappers
quota: move remount handling into the filesystem
ocfs2: Fix use after free on remount read-only
Fix up conflicts in fs/ext4/super.c and fs/ufs/file.c
Fix a bunch of new rapidio kernel-doc warnings:
Warning(include/linux/rio.h:123): No description found for parameter 'comp_tag'
Warning(include/linux/rio.h:123): No description found for parameter 'phys_efptr'
Warning(include/linux/rio.h:123): No description found for parameter 'em_efptr'
Warning(include/linux/rio.h:123): No description found for parameter 'pwcback'
Warning(include/linux/rio.h:247): No description found for parameter 'set_domain'
Warning(include/linux/rio.h:247): No description found for parameter 'get_domain'
Warning(drivers/rapidio/rio-scan.c:1133): No description found for parameter 'rdev'
Warning(drivers/rapidio/rio-scan.c:1133): Excess function parameter 'port' description in 'rio_init_em'
Warning(drivers/rapidio/rio.c:349): No description found for parameter 'rdev'
Warning(drivers/rapidio/rio.c:349): Excess function parameter 'mport' description in 'rio_request_inb_pwrite'
Warning(drivers/rapidio/rio.c:393): No description found for parameter 'port'
Warning(drivers/rapidio/rio.c:393): No description found for parameter 'local'
Warning(drivers/rapidio/rio.c:393): No description found for parameter 'destid'
Warning(drivers/rapidio/rio.c:393): No description found for parameter 'hopcount'
Warning(drivers/rapidio/rio.c:393): Excess function parameter 'rdev' description in 'rio_mport_get_physefb'
Warning(drivers/rapidio/rio.c:845): Excess function parameter 'local' description in 'rio_std_route_clr_table'
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Alexandre Bounine <alexandre.bounine@idt.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This reverts commit 0ac0c0d0f8, which
caused cross-architecture build problems for all the wrong reasons.
IA64 already added its own version of __node_random(), but the fact is,
there is nothing architectural about the function, and the original
commit was just badly done. Revert it, since no fix is forthcoming.
Requested-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
ceph: clean up on forwarded aborted mds request
ceph: fix leak of osd authorizer
ceph: close out mds, osd connections before stopping auth
ceph: make lease code DN specific
fs/ceph: Use ERR_CAST
ceph: renew auth tickets before they expire
ceph: do not resend mon requests on auth ticket renewal
ceph: removed duplicated #includes
ceph: avoid possible null dereference
ceph: make mds requests killable, not interruptible
sched: add wait_for_completion_killable_timeout
Commit 756dee7587 ("SLUB: Get rid of dynamic DMA
kmalloc cache allocation") makes S390 run out of kmalloc caches. Increase the
number of kmalloc caches to a safe size.
Cc: <stable@kernel.org> [ .33 and .34 ]
Reported-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Tested-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Christoph Lameter <cl@linux-foundation.org>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6: (26 commits)
ALSA: snd-usb-caiaq: Bump version number to 1.3.21
ALSA: Revert "ALSA: snd-usb-caiaq: Set default input mode of A4DJ"
ALSA: snd-usb-caiaq: Simplify single case to an 'if'
ALSA: snd-usb-caiaq: Restore 'Control vinyl' input mode on A4DJ
ALSA: hda: Use LPIB for a Shuttle device
ALSA: hda: Add support for another Lenovo ThinkPad Edge in conexant codec
ALSA: hda: Use LPIB for Sony VPCS11V9E
ALSA: usb-audio: fix feature unit parser for UAC2
ALSA: asihpi - Minor code cleanup
ALSA: asihpi - Add support for new ASI8800 family
ALSA: asihpi - Fix bug preventing outstream_write preload from happening
ALSA: asihpi - Fix imbalanced lock path in hw_message
ALSA: asihpi - Remove support for old ASI8800 family
ALSA: asihpi - Add hd radio blend functions
ALSA: asihpi - Remove unused io map functions
ALSA: usb-audio: add support for UAC2 pitch control
ALSA: usb-audio: parse UAC2 endpoint descriptors correctly
ALSA: usb-audio: fix return values
ALSA: usb-audio: parse more format descriptors with structs
sound: Add missing spin_unlock
...
Add missing _killable_timeout variant for wait_for_completion that will
return when a timeout expires or the task is killed.
CC: Ingo Molnar <mingo@elte.hu>
CC: Andreas Herrmann <andreas.herrmann3@amd.com>
CC: Thomas Gleixner <tglx@linutronix.de>
CC: Mike Galbraith <efault@gmx.de>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Sage Weil <sage@newdream.net>
* 'idle-release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-idle-2.6:
intel_idle: native hardware cpuidle driver for latest Intel processors
ACPI: acpi_idle: touch TS_POLLING only in the non-MWAIT case
acpi_pad: uses MONITOR/MWAIT, so it doesn't need to clear TS_POLLING
sched: clarify commment for TS_POLLING
ACPI: allow a native cpuidle driver to displace ACPI
cpuidle: make cpuidle_curr_driver static
cpuidle: add cpuidle_unregister_driver() error check
cpuidle: fail to register if !CONFIG_CPU_IDLE
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (22 commits)
netlink: bug fix: wrong size was calculated for vfinfo list blob
netlink: bug fix: don't overrun skbs on vf_port dump
xt_tee: use skb_dst_drop()
netdev/fec: fix ifconfig eth0 down hang issue
cnic: Fix context memory init. on 5709.
drivers/net: Eliminate a NULL pointer dereference
drivers/net/hamradio: Eliminate a NULL pointer dereference
be2net: Patch removes redundant while statement in loop.
ipv6: Add GSO support on forwarding path
net: fix __neigh_event_send()
vhost: fix the memory leak which will happen when memory_access_ok fails
vhost-net: fix to check the return value of copy_to/from_user() correctly
vhost: fix to check the return value of copy_to/from_user() correctly
vhost: Fix host panic if ioctl called with wrong index
net: fix lock_sock_bh/unlock_sock_bh
net/iucv: Add missing spin_unlock
net: ll_temac: fix checksum offload logic
net: ll_temac: fix interrupt bug when interrupt 0 is used
sctp: dubious bitfields in sctp_transport
ipmr: off by one in __ipmr_fill_mroute()
...
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:
remove detritus left by "mm: make read_cache_page synchronous"
fix fs/sysv s_dirt handling
fat: convert to use the new truncate convention.
ext2: convert to use the new truncate convention.
tmpfs: convert to use the new truncate convention
fs: convert simple fs to new truncate
kill spurious reference to vmtruncate
fs: introduce new truncate sequence
fs/super: fix kernel-doc warning
fs/minix: bugfix, number of indirect block ptrs per block depends on block size
rename the generic fsync implementations
drop unused dentry argument to ->fsync
fs: Add missing mutex_unlock
Fix racy use of anon_inode_getfd() in perf_event.c
get rid of the magic around f_count in aio
VFS: fix recent breakage of FS_REVAL_DOT
Revert "anon_inode: set S_IFREG on the anon_inode"
Introduce a new truncate calling sequence into fs/mm subsystems. Rather than
setattr > vmtruncate > truncate, have filesystems call their truncate sequence
from ->setattr if filesystem specific operations are required. vmtruncate is
deprecated, and truncate_pagecache and inode_newsize_ok helpers introduced
previously should be used.
simple_setattr is introduced for simple in-ram filesystems to implement
the new truncate sequence. Eventually all filesystems should be converted
to implement a setattr, and the default code in notify_change should go
away.
simple_setsize is also introduced to perform just the ATTR_SIZE portion
of simple_setattr (ie. changing i_size and trimming pagecache).
To implement the new truncate sequence:
- filesystem specific manipulations (eg freeing blocks) must be done in
the setattr method rather than ->truncate.
- vmtruncate can not be used by core code to trim blocks past i_size in
the event of write failure after allocation, so this must be performed
in the fs code.
- convert usage of helpers block_write_begin, nobh_write_begin,
cont_write_begin, and *blockdev_direct_IO* to use _newtrunc postfixed
variants. These avoid calling vmtruncate to trim blocks (see previous).
- inode_setattr should not be used. generic_setattr is a new function
to be used to copy simple attributes into the generic inode.
- make use of the better opportunity to handle errors with the new sequence.
Big problem with the previous calling sequence: the filesystem is not called
until i_size has already changed. This means it is not allowed to fail the
call, and also it does not know what the previous i_size was. Also, generic
code calling vmtruncate to truncate allocated blocks in case of error had
no good way to return a meaningful error (or, for example, atomically handle
block deallocation).
Cc: Christoph Hellwig <hch@lst.de>
Acked-by: Jan Kara <jack@suse.cz>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
We don't name our generic fsync implementations very well currently.
The no-op implementation for in-memory filesystems currently is called
simple_sync_file which doesn't make too much sense to start with,
the the generic one for simple filesystems is called simple_fsync
which can lead to some confusion.
This patch renames the generic file fsync method to generic_file_fsync
to match the other generic_file_* routines it is supposed to be used
with, and the no-op implementation to noop_fsync to make it obvious
what to expect. In addition add some documentation for both methods.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
__aio_put_req() plays sick games with file refcount. What
it wants is fput() from atomic context; it's almost always
done with f_count > 1, so they only have to deal with delayed
work in rare cases when their reference happens to be the
last one. Current code decrements f_count and if it hasn't
hit 0, everything is fine. Otherwise it keeps a pointer
to struct file (with zero f_count!) around and has delayed
work do __fput() on it.
Better way to do it: use atomic_long_add_unless( , -1, 1)
instead of !atomic_long_dec_and_test(). IOW, decrement it
only if it's not the last reference, leave refcount alone
if it was. And use normal fput() in delayed work.
I've made that atomic_long_add_unless call a new helper -
fput_atomic(). Drops a reference to file if it's safe to
do in atomic (i.e. if that's not the last one), tells if
it had been able to do that. aio.c converted to it, __fput()
use is gone. req->ki_file *always* contributes to refcount
now. And __fput() became static.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev:
libata: implement dump_id force param
libata: disable ATAPI AN by default
libata-sff: make BMDMA optional
libata-sff: kill dummy BMDMA ops from sata_qstor and pata_octeon_cf
libata-sff: separate out BMDMA init
libata-sff: separate out BMDMA irq handler
libata-sff: ata_sff_irq_clear() is BMDMA specific
sata_mv: drop unncessary EH callback resetting
Eric Dumazet
Jiri Pirko
Tom Herbert
David S. Miller
Andy Gospodarek
Alexander Duyck
Steffen Klassert
Arnaud Ebalard
Scott McMillan
Ian Campbell
Linus Torvalds
Randy Dunlap
Christoph Lameter
Sage Weil
Changli Gao
Len Brown
npiggin@suse.de
Christoph Hellwig
Al Viro