| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
On KVM on my box, this reduces the overhead from an always-accept
seccomp filter from ~130ns to ~17ns. Most of that comes from
avoiding IRET on every syscall when seccomp is enabled.
In extremely approximate hacked-up benchmarking, just bypassing IRET
saves about 80ns, so there's another 43ns of savings here from
simplifying the seccomp path.
The diffstat is also rather nice :)
Signed-off-by: Andy Lutomirski <[email protected]>
Link: http://lkml.kernel.org/r/a3dbd267ee990110478d349f78cccfdac5497a84.1409954077.git.luto@amacapital.net
Signed-off-by: H. Peter Anvin <[email protected]>
|
|
For slowpath syscalls, we initialize regs->ax to -ENOSYS and stick
the syscall number into regs->orig_ax prior to any possible tracing
and syscall execution. This is user-visible ABI used by ptrace
syscall emulation and seccomp.
For fastpath syscalls, there's no good reason not to do the same
thing. It's even slightly simpler than what we're currently doing.
It probably has no measureable performance impact. It should have
no user-visible effect.
The purpose of this patch is to prepare for two-phase syscall
tracing, in which the first phase might modify the saved RAX without
leaving the fast path. This change is just subtle enough that I'm
keeping it separate.
Signed-off-by: Andy Lutomirski <[email protected]>
Link: http://lkml.kernel.org/r/01218b493f12ae2f98034b78c9ae085e38e94350.1409954077.git.luto@amacapital.net
Signed-off-by: H. Peter Anvin <[email protected]>
|
|
This splits syscall_trace_enter into syscall_trace_enter_phase1 and
syscall_trace_enter_phase2. Only phase 2 has full pt_regs, and only
phase 2 is permitted to modify any of pt_regs except for orig_ax.
The intent is that phase 1 can be called from the syscall fast path.
In this implementation, phase1 can handle any combination of
TIF_NOHZ (RCU context tracking), TIF_SECCOMP, and TIF_SYSCALL_AUDIT,
unless seccomp requests a ptrace event, in which case phase2 is
forced.
In principle, this could yield a big speedup for TIF_NOHZ as well as
for TIF_SECCOMP if syscall exit work were similarly split up.
Signed-off-by: Andy Lutomirski <[email protected]>
Link: http://lkml.kernel.org/r/2df320a600020fda055fccf2b668145729dd0c04.1409954077.git.luto@amacapital.net
Signed-off-by: H. Peter Anvin <[email protected]>
|
|
The RCU context tracking code requires that arch code call
user_exit() on any entry into kernel code if TIF_NOHZ is set. This
patch adds a check for TIF_NOHZ and a comment to the syscall entry
tracing code.
The main purpose of this patch is to make the code easier to follow:
one can read the body of user_exit and of every function it calls
without finding any explanation of why it's called for traced
syscalls but not for untraced syscalls. This makes it clear when
user_exit() is necessary.
Cc: Frederic Weisbecker <[email protected]>
Signed-off-by: Andy Lutomirski <[email protected]>
Link: http://lkml.kernel.org/r/0b13e0e24ec0307d67ab7a23b58764f6b1270116.1409954077.git.luto@amacapital.net
Signed-off-by: H. Peter Anvin <[email protected]>
|
|
is_compat_task() is the wrong check for audit arch; the check should
be is_ia32_task(): x32 syscalls should be AUDIT_ARCH_X86_64, not
AUDIT_ARCH_I386.
CONFIG_AUDITSYSCALL is currently incompatible with x32, so this has
no visible effect.
Signed-off-by: Andy Lutomirski <[email protected]>
Link: http://lkml.kernel.org/r/a0138ed8c709882aec06e4acc30bfa9b623b8717.1409954077.git.luto@amacapital.net
Signed-off-by: H. Peter Anvin <[email protected]>
|
|
The description of how archs should implement seccomp filters was
still strictly correct, but it failed to describe the newly
available optimizations.
Signed-off-by: Andy Lutomirski <[email protected]>
Signed-off-by: Kees Cook <[email protected]>
|
|
populate_seccomp_data is expensive: it works by inspecting
task_pt_regs and various other bits to piece together all the
information, and it's does so in multiple partially redundant steps.
Arch-specific code in the syscall entry path can do much better.
Admittedly this adds a bit of additional room for error, but the
speedup should be worth it.
Signed-off-by: Andy Lutomirski <[email protected]>
Signed-off-by: Kees Cook <[email protected]>
|
|
The reason I did this is to add a seccomp API that will be usable
for an x86 fast path. The x86 entry code needs to use a rather
expensive slow path for a syscall that might be visible to things
like ptrace. By splitting seccomp into two phases, we can check
whether we need the slow path and then use the fast path in if the
filter allows the syscall or just returns some errno.
As a side effect, I think the new code is much easier to understand
than the old code.
This has one user-visible effect: the audit record written for
SECCOMP_RET_TRACE is now a simple indication that SECCOMP_RET_TRACE
happened. It used to depend in a complicated way on what the tracer
did. I couldn't make much sense of it.
Signed-off-by: Andy Lutomirski <[email protected]>
Signed-off-by: Kees Cook <[email protected]>
|
|
The secure_computing function took a syscall number parameter, but
it only paid any attention to that parameter if seccomp mode 1 was
enabled. Rather than coming up with a kludge to get the parameter
to work in mode 2, just remove the parameter.
To avoid churn in arches that don't have seccomp filters (and may
not even support syscall_get_nr right now), this leaves the
parameter in secure_computing_strict, which is now a real function.
For ARM, this is a bit ugly due to the fact that ARM conditionally
supports seccomp filters. Fixing that would probably only be a
couple of lines of code, but it should be coordinated with the audit
maintainers.
This will be a slight slowdown on some arches. The right fix is to
pass in all of seccomp_data instead of trying to make just the
syscall nr part be fast.
This is a prerequisite for making two-phase seccomp work cleanly.
Cc: Russell King <[email protected]>
Cc: [email protected]
Cc: Ralf Baechle <[email protected]>
Cc: [email protected]
Cc: Martin Schwidefsky <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: Kees Cook <[email protected]>
Signed-off-by: Andy Lutomirski <[email protected]>
Signed-off-by: Kees Cook <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs
Pull f2fs bug fixes from Jaegeuk Kim:
"This series includes patches to:
- fix recovery routines
- fix bugs related to inline_data/xattr
- fix when casting the dentry names
- handle EIO or ENOMEM correctly
- fix memory leak
- fix lock coverage"
* tag 'for-f2fs-3.17-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (28 commits)
f2fs: reposition unlock_new_inode to prevent accessing invalid inode
f2fs: fix wrong casting for dentry name
f2fs: simplify by using a literal
f2fs: truncate stale block for inline_data
f2fs: use macro for code readability
f2fs: introduce need_do_checkpoint for readability
f2fs: fix incorrect calculation with total/free inode num
f2fs: remove rename and use rename2
f2fs: skip if inline_data was converted already
f2fs: remove rewrite_node_page
f2fs: avoid double lock in truncate_blocks
f2fs: prevent checkpoint during roll-forward
f2fs: add WARN_ON in f2fs_bug_on
f2fs: handle EIO not to break fs consistency
f2fs: check s_dirty under cp_mutex
f2fs: unlock_page when node page is redirtied out
f2fs: introduce f2fs_cp_error for readability
f2fs: give a chance to mount again when encountering errors
f2fs: trigger release_dirty_inode in f2fs_put_super
f2fs: don't skip checkpoint if there is no dirty node pages
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security
Pull key subsystem fixes from James Morris:
"Fixes for the keys subsystem, one of which addresses a use-after-free
bug"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security:
PEFILE: Relax the check on the length of the PKCS#7 cert
KEYS: Fix use-after-free in assoc_array_gc()
KEYS: Fix public_key asymmetric key subtype name
KEYS: Increase root_maxkeys and root_maxbytes sizes
|
|
Structure name and variable name were erroneously interchanged
Signed-off-by: Noam Camus <[email protected]>
Acked-by: Vineet Gupta <[email protected]>
[ Also removed pointless cast from "void *". - Linus ]
Signed-off-by: Linus Torvalds <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull more arm64 fixes from Will Deacon:
"Another handful of arm64 fixes here. They address some issues found
by running smatch on the arch code (ignoring the false positives) and
also stop 32-bit Android from losing track of its stack.
There's one additional irq migration fix in the pipeline, but it came
in after I'd tagged and tested this set.
- a few fixes for real issues found by smatch (after Dan's talk at KS)
- revert the /proc/cpuinfo changes merged during the merge window.
We've opened a can of worms here, so we need to find out where we
stand before we change this interface.
- implement KSTK_ESP for compat tasks, otherwise 32-bit Android gets
confused wondering where its [stack] has gone
- misc fixes (fpsimd context handling, crypto, ...)"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
Revert "arm64: cpuinfo: print info for all CPUs"
arm64: fix bug for reloading FPSIMD state after cpu power off
arm64: report correct stack pointer in KSTK_ESP for compat tasks
arm64: Add brackets around user_stack_pointer()
arm64: perf: don't rely on layout of pt_regs when grabbing sp or pc
arm64: ptrace: fix compat reg getter/setter return values
arm64: ptrace: fix compat hardware watchpoint reporting
arm64: Remove unused variable in head.S
arm64/crypto: remove redundant update of data
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Pull PCI fix from Bjorn Helgaas:
"This fixes an ARM allmodconfig build problem:
Remove module option for ST Microelectronics SPEAr13xx"
* tag 'pci-v3.17-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
PCI: spear: Remove module option
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/cooloney/linux-leds
Pull LED fix from Bryan Wu:
"Hugh, Jiri and many other people found a kernel oops due to a LED
change merged recently. Now the right fix might just revert it and
avoid the kernel oops"
* 'leds-fixes-for-3.17' of git://git.kernel.org/pub/scm/linux/kernel/git/cooloney/linux-leds:
Revert "leds: convert blink timer to workqueue"
|
|
Relax the check on the length of the PKCS#7 cert as it appears that the PE
file wrapper size gets rounded up to the nearest 8.
The debugging output looks like this:
PEFILE: ==> verify_pefile_signature()
PEFILE: ==> pefile_parse_binary()
PEFILE: checksum @ 110
PEFILE: header size = 200
PEFILE: cert = 968 @547be0 [68 09 00 00 00 02 02 00 30 82 09 56 ]
PEFILE: sig wrapper = { 968, 200, 2 }
PEFILE: Signature data not PKCS#7
The wrapper is the first 8 bytes of the hex dump inside []. This indicates a
length of 0x968 bytes, including the wrapper header - so 0x960 bytes of
payload.
The ASN.1 wrapper begins [ ... 30 82 09 56 ]. That indicates an object of size
0x956 - a four byte discrepency, presumably just padding for alignment
purposes.
So we just check that the ASN.1 container is no bigger than the payload and
reduce the recorded size appropriately.
Whilst we're at it, allow shorter PKCS#7 objects that manage to squeeze within
127 or 255 bytes. It's just about conceivable if no X.509 certs are included
in the PKCS#7 message.
Reported-by: Vivek Goyal <[email protected]>
Signed-off-by: David Howells <[email protected]>
Acked-by: Vivek Goyal <[email protected]>
Acked-by: Peter Jones <[email protected]>
Signed-off-by: James Morris <[email protected]>
|
|
An edit script should be considered inaccessible by a function once it has
called assoc_array_apply_edit() or assoc_array_cancel_edit().
However, assoc_array_gc() is accessing the edit script just after the
gc_complete: label.
Reported-by: Andreea-Cristina Bernat <[email protected]>
Signed-off-by: David Howells <[email protected]>
Reviewed-by: Andreea-Cristina Bernat <[email protected]>
cc: [email protected]
cc: [email protected]
Cc: [email protected]
Signed-off-by: James Morris <[email protected]>
|
|
The length of the name of an asymmetric key subtype must be stored in struct
asymmetric_key_subtype::name_len so that it can be matched by a search for
"<subkey_name>:<partial_fingerprint>". Fix the public_key subtype to have
name_len set.
Signed-off-by: David Howells <[email protected]>
Signed-off-by: James Morris <[email protected]>
|
|
Now that NFS client uses the kernel key ring facility to store the NFSv4
id/gid mappings, the defaults for root_maxkeys and root_maxbytes need to be
substantially increased.
These values have been soak tested:
https://bugzilla.redhat.com/show_bug.cgi?id=1033708#c73
Signed-off-by: Steve Dickson <[email protected]>
Signed-off-by: David Howells <[email protected]>
Signed-off-by: James Morris <[email protected]>
|
|
This reverts commit 8b37e1bef5a6b60e949e28a4db3006e4b00bd758.
It's broken as it changes led_blink_set() in a way that it can now sleep
(while synchronously waiting for workqueue to be cancelled). That's a
problem, because it's possible that this function gets called from atomic
context (tpt_trig_timer() takes a readlock and thus disables preemption).
This has been brought up 3 weeks ago already [1] but no proper fix has
materialized, and I keep seeing the problem since 3.17-rc1.
[1] https://lkml.org/lkml/2014/8/16/128
BUG: sleeping function called from invalid context at kernel/workqueue.c:2650
in_atomic(): 1, irqs_disabled(): 0, pid: 2335, name: wpa_supplicant
5 locks held by wpa_supplicant/2335:
#0: (rtnl_mutex){+.+.+.}, at: [<ffffffff814c7c92>] rtnl_lock+0x12/0x20
#1: (&wdev->mtx){+.+.+.}, at: [<ffffffffc06e649c>] cfg80211_mgd_wext_siwessid+0x5c/0x180 [cfg80211]
#2: (&local->mtx){+.+.+.}, at: [<ffffffffc0817dea>] ieee80211_prep_connection+0x17a/0x9a0 [mac80211]
#3: (&local->chanctx_mtx){+.+.+.}, at: [<ffffffffc08081ed>] ieee80211_vif_use_channel+0x5d/0x2a0 [mac80211]
#4: (&trig->leddev_list_lock){.+.+..}, at: [<ffffffffc081e68c>] tpt_trig_timer+0xec/0x170 [mac80211]
CPU: 0 PID: 2335 Comm: wpa_supplicant Not tainted 3.17.0-rc3 #1
Hardware name: LENOVO 7470BN2/7470BN2, BIOS 6DET38WW (2.02 ) 12/19/2008
ffff8800360b5a50 ffff8800751f76d8 ffffffff8159e97f ffff8800360b5a30
ffff8800751f76e8 ffffffff810739a5 ffff8800751f77b0 ffffffff8106862f
ffffffff810685d0 0aa2209200000000 ffff880000000004 ffff8800361c59d0
Call Trace:
[<ffffffff8159e97f>] dump_stack+0x4d/0x66
[<ffffffff810739a5>] __might_sleep+0xe5/0x120
[<ffffffff8106862f>] flush_work+0x5f/0x270
[<ffffffff810685d0>] ? mod_delayed_work_on+0x80/0x80
[<ffffffff810945ca>] ? mark_held_locks+0x6a/0x90
[<ffffffff81068a5f>] ? __cancel_work_timer+0x6f/0x100
[<ffffffff810946ed>] ? trace_hardirqs_on_caller+0xfd/0x1c0
[<ffffffff81068a6b>] __cancel_work_timer+0x7b/0x100
[<ffffffff81068b0e>] cancel_delayed_work_sync+0xe/0x10
[<ffffffff8147cf3b>] led_blink_set+0x1b/0x40
[<ffffffffc081e6b0>] tpt_trig_timer+0x110/0x170 [mac80211]
[<ffffffffc081ecdd>] ieee80211_mod_tpt_led_trig+0x9d/0x160 [mac80211]
[<ffffffffc07e4278>] __ieee80211_recalc_idle+0x98/0x140 [mac80211]
[<ffffffffc07e59ce>] ieee80211_idle_off+0xe/0x10 [mac80211]
[<ffffffffc0804e5b>] ieee80211_add_chanctx+0x3b/0x220 [mac80211]
[<ffffffffc08062e4>] ieee80211_new_chanctx+0x44/0xf0 [mac80211]
[<ffffffffc080838a>] ieee80211_vif_use_channel+0x1fa/0x2a0 [mac80211]
[<ffffffffc0817df8>] ieee80211_prep_connection+0x188/0x9a0 [mac80211]
[<ffffffffc081c246>] ieee80211_mgd_auth+0x256/0x2e0 [mac80211]
[<ffffffffc07eab33>] ieee80211_auth+0x13/0x20 [mac80211]
[<ffffffffc06cb006>] cfg80211_mlme_auth+0x106/0x270 [cfg80211]
[<ffffffffc06ce085>] cfg80211_conn_do_work+0x155/0x3b0 [cfg80211]
[<ffffffffc06cf670>] cfg80211_connect+0x3f0/0x540 [cfg80211]
[<ffffffffc06e6148>] cfg80211_mgd_wext_connect+0x158/0x1f0 [cfg80211]
[<ffffffffc06e651e>] cfg80211_mgd_wext_siwessid+0xde/0x180 [cfg80211]
[<ffffffffc06e36c0>] ? cfg80211_wext_giwessid+0x50/0x50 [cfg80211]
[<ffffffffc06e36dd>] cfg80211_wext_siwessid+0x1d/0x40 [cfg80211]
[<ffffffff81584d0c>] ioctl_standard_iw_point+0x14c/0x3e0
[<ffffffff810946ed>] ? trace_hardirqs_on_caller+0xfd/0x1c0
[<ffffffff8158502a>] ioctl_standard_call+0x8a/0xd0
[<ffffffff81584fa0>] ? ioctl_standard_iw_point+0x3e0/0x3e0
[<ffffffff81584b76>] wireless_process_ioctl.constprop.10+0xb6/0x100
[<ffffffff8158521d>] wext_handle_ioctl+0x5d/0xb0
[<ffffffff814cfb29>] dev_ioctl+0x329/0x620
[<ffffffff810946ed>] ? trace_hardirqs_on_caller+0xfd/0x1c0
[<ffffffff8149c7f2>] sock_ioctl+0x142/0x2e0
[<ffffffff811b0140>] do_vfs_ioctl+0x300/0x520
[<ffffffff815a67fb>] ? sysret_check+0x1b/0x56
[<ffffffff810946ed>] ? trace_hardirqs_on_caller+0xfd/0x1c0
[<ffffffff811b03e1>] SyS_ioctl+0x81/0xa0
[<ffffffff815a67d6>] system_call_fastpath+0x1a/0x1f
wlan0: send auth to 00:0b:6b:3c:8c:e4 (try 1/3)
wlan0: authenticated
wlan0: associate with 00:0b:6b:3c:8c:e4 (try 1/3)
wlan0: RX AssocResp from 00:0b:6b:3c:8c:e4 (capab=0x431 status=0 aid=2)
wlan0: associated
IPv6: ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
cfg80211: Calling CRDA for country: NA
wlan0: Limiting TX power to 27 (27 - 0) dBm as advertised by 00:0b:6b:3c:8c:e4
=================================
[ INFO: inconsistent lock state ]
3.17.0-rc3 #1 Not tainted
---------------------------------
inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage.
swapper/0/0 [HC0[0]:SC1[1]:HE1:SE0] takes:
((&(&led_cdev->blink_work)->work)){+.?...}, at: [<ffffffff810685d0>] flush_work+0x0/0x270
{SOFTIRQ-ON-W} state was registered at:
[<ffffffff81094dbe>] __lock_acquire+0x30e/0x1a30
[<ffffffff81096c81>] lock_acquire+0x91/0x110
[<ffffffff81068608>] flush_work+0x38/0x270
[<ffffffff81068a6b>] __cancel_work_timer+0x7b/0x100
[<ffffffff81068b0e>] cancel_delayed_work_sync+0xe/0x10
[<ffffffff8147cf3b>] led_blink_set+0x1b/0x40
[<ffffffffc081e6b0>] tpt_trig_timer+0x110/0x170 [mac80211]
[<ffffffffc081ecdd>] ieee80211_mod_tpt_led_trig+0x9d/0x160 [mac80211]
[<ffffffffc07e4278>] __ieee80211_recalc_idle+0x98/0x140 [mac80211]
[<ffffffffc07e59ce>] ieee80211_idle_off+0xe/0x10 [mac80211]
[<ffffffffc0804e5b>] ieee80211_add_chanctx+0x3b/0x220 [mac80211]
[<ffffffffc08062e4>] ieee80211_new_chanctx+0x44/0xf0 [mac80211]
[<ffffffffc080838a>] ieee80211_vif_use_channel+0x1fa/0x2a0 [mac80211]
[<ffffffffc0817df8>] ieee80211_prep_connection+0x188/0x9a0 [mac80211]
[<ffffffffc081c246>] ieee80211_mgd_auth+0x256/0x2e0 [mac80211]
[<ffffffffc07eab33>] ieee80211_auth+0x13/0x20 [mac80211]
[<ffffffffc06cb006>] cfg80211_mlme_auth+0x106/0x270 [cfg80211]
[<ffffffffc06ce085>] cfg80211_conn_do_work+0x155/0x3b0 [cfg80211]
[<ffffffffc06cf670>] cfg80211_connect+0x3f0/0x540 [cfg80211]
[<ffffffffc06e6148>] cfg80211_mgd_wext_connect+0x158/0x1f0 [cfg80211]
[<ffffffffc06e651e>] cfg80211_mgd_wext_siwessid+0xde/0x180 [cfg80211]
[<ffffffffc06e36dd>] cfg80211_wext_siwessid+0x1d/0x40 [cfg80211]
[<ffffffff81584d0c>] ioctl_standard_iw_point+0x14c/0x3e0
[<ffffffff8158502a>] ioctl_standard_call+0x8a/0xd0
[<ffffffff81584b76>] wireless_process_ioctl.constprop.10+0xb6/0x100
[<ffffffff8158521d>] wext_handle_ioctl+0x5d/0xb0
[<ffffffff814cfb29>] dev_ioctl+0x329/0x620
[<ffffffff8149c7f2>] sock_ioctl+0x142/0x2e0
[<ffffffff811b0140>] do_vfs_ioctl+0x300/0x520
[<ffffffff811b03e1>] SyS_ioctl+0x81/0xa0
[<ffffffff815a67d6>] system_call_fastpath+0x1a/0x1f
irq event stamp: 493416
hardirqs last enabled at (493416): [<ffffffff81068a5f>] __cancel_work_timer+0x6f/0x100
hardirqs last disabled at (493415): [<ffffffff81067e9f>] try_to_grab_pending+0x1f/0x160
softirqs last enabled at (493408): [<ffffffff81053ced>] _local_bh_enable+0x1d/0x50
softirqs last disabled at (493409): [<ffffffff81054c75>] irq_exit+0xa5/0xb0
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock((&(&led_cdev->blink_work)->work));
<Interrupt>
lock((&(&led_cdev->blink_work)->work));
*** DEADLOCK ***
2 locks held by swapper/0/0:
#0: (((&tpt_trig->timer))){+.-...}, at: [<ffffffff810b4c50>] call_timer_fn+0x0/0x180
#1: (&trig->leddev_list_lock){.+.?..}, at: [<ffffffffc081e68c>] tpt_trig_timer+0xec/0x170 [mac80211]
stack backtrace:
CPU: 0 PID: 0 Comm: swapper/0 Not tainted 3.17.0-rc3 #1
Hardware name: LENOVO 7470BN2/7470BN2, BIOS 6DET38WW (2.02 ) 12/19/2008
ffffffff8246eb30 ffff88007c203b00 ffffffff8159e97f ffffffff81a194c0
ffff88007c203b50 ffffffff81599c29 0000000000000001 ffffffff00000001
ffff880000000000 0000000000000006 ffffffff81a194c0 ffffffff81093ad0
Call Trace:
<IRQ> [<ffffffff8159e97f>] dump_stack+0x4d/0x66
[<ffffffff81599c29>] print_usage_bug+0x1f4/0x205
[<ffffffff81093ad0>] ? check_usage_backwards+0x140/0x140
[<ffffffff810944d3>] mark_lock+0x223/0x2b0
[<ffffffff81094d60>] __lock_acquire+0x2b0/0x1a30
[<ffffffff81096c81>] lock_acquire+0x91/0x110
[<ffffffff810685d0>] ? mod_delayed_work_on+0x80/0x80
[<ffffffffc081e5a0>] ? __ieee80211_get_rx_led_name+0x10/0x10 [mac80211]
[<ffffffff81068608>] flush_work+0x38/0x270
[<ffffffff810685d0>] ? mod_delayed_work_on+0x80/0x80
[<ffffffff810945ca>] ? mark_held_locks+0x6a/0x90
[<ffffffff81068a5f>] ? __cancel_work_timer+0x6f/0x100
[<ffffffffc081e5a0>] ? __ieee80211_get_rx_led_name+0x10/0x10 [mac80211]
[<ffffffff8109469d>] ? trace_hardirqs_on_caller+0xad/0x1c0
[<ffffffffc081e5a0>] ? __ieee80211_get_rx_led_name+0x10/0x10 [mac80211]
[<ffffffff81068a6b>] __cancel_work_timer+0x7b/0x100
[<ffffffff81068b0e>] cancel_delayed_work_sync+0xe/0x10
[<ffffffff8147cf3b>] led_blink_set+0x1b/0x40
[<ffffffffc081e6b0>] tpt_trig_timer+0x110/0x170 [mac80211]
[<ffffffff810b4cc5>] call_timer_fn+0x75/0x180
[<ffffffff810b4c50>] ? process_timeout+0x10/0x10
[<ffffffffc081e5a0>] ? __ieee80211_get_rx_led_name+0x10/0x10 [mac80211]
[<ffffffff810b50ac>] run_timer_softirq+0x1fc/0x2f0
[<ffffffff81054805>] __do_softirq+0x115/0x2e0
[<ffffffff81054c75>] irq_exit+0xa5/0xb0
[<ffffffff810049b3>] do_IRQ+0x53/0xf0
[<ffffffff815a74af>] common_interrupt+0x6f/0x6f
<EOI> [<ffffffff8147b56e>] ? cpuidle_enter_state+0x6e/0x180
[<ffffffff8147b732>] cpuidle_enter+0x12/0x20
[<ffffffff8108bba0>] cpu_startup_entry+0x330/0x360
[<ffffffff8158fb51>] rest_init+0xc1/0xd0
[<ffffffff8158fa90>] ? csum_partial_copy_generic+0x170/0x170
[<ffffffff81af3ff2>] start_kernel+0x44f/0x45a
[<ffffffff81af399c>] ? set_init_arg+0x53/0x53
[<ffffffff81af35ad>] x86_64_start_reservations+0x2a/0x2c
[<ffffffff81af36a0>] x86_64_start_kernel+0xf1/0xf4
Cc: Vincent Donnefort <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Tejun Heo <[email protected]>
Signed-off-by: Jiri Kosina <[email protected]>
Signed-off-by: Bryan Wu <[email protected]>
|
|
As the race condition on the inode cache, following scenario can appear:
[Thread a] [Thread b]
->f2fs_mkdir
->f2fs_add_link
->__f2fs_add_link
->init_inode_metadata failed here
->gc_thread_func
->f2fs_gc
->do_garbage_collect
->gc_data_segment
->f2fs_iget
->iget_locked
->wait_on_inode
->unlock_new_inode
->move_data_page
->make_bad_inode
->iput
When we fail in create/symlink/mkdir/mknod/tmpfile, the new allocated inode
should be set as bad to avoid being accessed by other thread. But in above
scenario, it allows f2fs to access the invalid inode before this inode was set
as bad.
This patch fix the potential problem, and this issue was found by code review.
change log from v1:
o Add condition judgment in gc_data_segment() suggested by Changman Lee.
o use iget_failed to simplify code.
Signed-off-by: Chao Yu <[email protected]>
Signed-off-by: Jaegeuk Kim <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull irq handling fixlet from Thomas Gleixner:
"Just an export for an interrupt flow handler which is now used in gpio
modules"
* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
irq: Export handle_fasteoi_irq
|
|
It turns out that vendors are relying on the format of /proc/cpuinfo,
and we've even spotted out-of-tree hacks attempting to make it look
identical to the format used by arch/arm/. That means we can't afford to
churn this interface in mainline, so revert the recent reformatting of
the file for arm64 pending discussions on the list to find out what
people actually want.
This reverts commit d7a49086f263164a2c4c178eb76412d48cd671d7.
Acked-by: Mark Rutland <[email protected]>
Signed-off-by: Will Deacon <[email protected]>
|
|
Now arm64 defers reloading FPSIMD state, but this optimization also
introduces the bug after cpu resume back from low power mode.
The reason is after the cpu has been powered off, s/w need set the
cpu's fpsimd_last_state to NULL so that it will force to reload
FPSIMD state for the thread, otherwise there has the chance to meet
the condition for both the task's fpsimd_state.cpu field contains the
id of the current cpu, and the cpu's fpsimd_last_state per-cpu variable
points to the task's fpsimd_state, so finally kernel will skip to reload
the context during it return back to userland.
Acked-by: Ard Biesheuvel <[email protected]>
Reviewed-by: Catalin Marinas <[email protected]>
Signed-off-by: Leo Yan <[email protected]>
Signed-off-by: Will Deacon <[email protected]>
|
|
|
|
Pull Xtensa updates from Chris Zankel:
"Xtensa improvements for 3.17:
- support highmem on cores with aliasing data cache. Enable highmem
on kc705 by default
- simplify addition of new core variants (no need to modify Kconfig /
Makefiles)
- improve robustness of unaligned access handler and its interaction
with window overflow/underflow exception handlers
- deprecate atomic and spill registers syscalls
- clean up Kconfig: remove orphan MATH_EMULATION, sort 'select'
statements
- wire up renameat2 syscall.
Various fixes:
- fix address checks in dma_{alloc,free}_coherent (runtime BUG)
- fix access to THREAD_RA/THREAD_SP/THREAD_DS (debug build breakage)
- fix TLBTEMP_BASE_2 region handling in fast_second_level_miss
(runtime unrecoverable exception)
- fix a6 and a7 handling in fast_syscall_xtensa (runtime userspace
register clobbering)
- fix kernel/user jump out of fast_unaligned (potential runtime
unrecoverabl exception)
- replace termios IOCTL code definitions with constants (userspace
build breakage)"
* tag 'xtensa-20140830' of git://github.com/czankel/xtensa-linux: (25 commits)
xtensa: deprecate fast_xtensa and fast_spill_registers syscalls
xtensa: don't allow overflow/underflow on unaligned stack
xtensa: fix a6 and a7 handling in fast_syscall_xtensa
xtensa: allow single-stepping through unaligned load/store
xtensa: move invalid unaligned instruction handler closer to its users
xtensa: make fast_unaligned store restartable
xtensa: add double exception fixup handler for fast_unaligned
xtensa: fix kernel/user jump out of fast_unaligned
xtensa: configure kc705 for highmem
xtensa: support highmem in aliasing cache flushing code
xtensa: support aliasing cache in kmap
xtensa: support aliasing cache in k[un]map_atomic
xtensa: implement clear_user_highpage and copy_user_highpage
xtensa: fix TLBTEMP_BASE_2 region handling in fast_second_level_miss
xtensa: allow fixmap and kmap span more than one page table
xtensa: make fixmap region addressing grow with index
xtensa: fix access to THREAD_RA/THREAD_SP/THREAD_DS
xtensa: add renameat2 syscall
xtensa: fix address checks in dma_{alloc,free}_coherent
xtensa: replace IOCTL code definitions with constants
...
|
|
unicore32 builds fail with
arch/unicore32/kernel/signal.c: In function ‘setup_frame’:
arch/unicore32/kernel/signal.c:257: error: ‘usig’ undeclared (first use in this function)
arch/unicore32/kernel/signal.c:279: error: ‘usig’ undeclared (first use in this function)
arch/unicore32/kernel/signal.c: In function ‘handle_signal’:
arch/unicore32/kernel/signal.c:306: warning: unused variable ‘tsk’
arch/unicore32/kernel/signal.c: In function ‘do_signal’:
arch/unicore32/kernel/signal.c:376: error: implicit declaration of function ‘get_signsl’
make[1]: *** [arch/unicore32/kernel/signal.o] Error 1
make: *** [arch/unicore32/kernel/signal.o] Error 2
Bisect points to commit 649671c90eaf ("unicore32: Use get_signal()
signal_setup_done()").
This code never even compiled. Reverting the patch does not work, since
previously used functions no longer exist, so try to fix it up. Compile
tested only.
Fixes: 649671c90eaf ("unicore32: Use get_signal() signal_setup_done()")
Cc: Richard Weinberger <[email protected]>
Signed-off-by: Guenter Roeck <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Pull ARM fixes from Russell King:
"Various assorted fixes:
- a couple of patches from Mark Rutland to resolve an errata with
Cortex-A15 CPUs.
- fix cpuidle for the CPU part ID changes in the last merge window
- add support for a relocation which ARM binutils is generating in
some circumstances"
* 'fixes' of git://ftp.arm.linux.org.uk/~rmk/linux-arm:
ARM: 8130/1: cpuidle/cpuidle-big_little: fix reading cpu id part number
ARM: 8129/1: errata: work around Cortex-A15 erratum 830321 using dummy strex
ARM: 8128/1: abort: don't clear the exclusive monitors
ARM: 8127/1: module: add support for R_ARM_TARGET1 relocations
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
Pull ARM SoC fixes from Olof Johansson:
"Here's the weekly batch of fixes from arm-soc.
The delta is a largeish negative delta, due to revert of SMP support
for Broadcom's STB SoC -- it was accidentally merged before some
issues had been addressed, so they will make a new attempt for 3.18.
I didn't see a need for a full revert of the whole platform due to
this, we're keeping the rest enabled.
The rest is mostly:
- a handful of DT fixes for i.MX (Hummingboard/Cubox-i in particular)
- some MTD/NAND fixes for OMAP
- minor DT fixes for shmobile
- warning fix for UP builds on vexpress/spc
There's also a couple of patches that wires up hwmod on TI's DRA7 SoC
so it can boot. Drivers and the rest had landed for 3.17, and it's
small and isolated so it made sense to pick up now even if it's not a
bugfix"
* tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (23 commits)
vexpress/spc: fix a build warning on array bounds
ARM: DRA7: hwmod: Add dra74x and dra72x specific ocp interface lists
ARM: DRA7: Add support for soc_is_dra74x() and soc_is_dra72x() variants
MAINTAINERS: catch special Rockchip code locations
ARM: dts: microsom-ar8035: MDIO pad must be set open drain
ARM: dts: omap54xx-clocks: Fix the l3 and l4 clock rates
ARM: brcmstb: revert SMP support
ARM: OMAP2+: hwmod: Rearm wake-up interrupts for DT when MUSB is idled
ARM: dts: Enable UART wake-up events for beagleboard
ARM: dts: Remove twl6030 clk32g "regulator"
ARM: OMAP2+: omap_device: remove warning that clk alias already exists
ARM: OMAP: fix %d confusingly prefixed with 0x in format string
ARM: dts: DRA7: fix interrupt-cells for GPIO
mtd: nand: omap: Fix 1-bit Hamming code scheme, omap_calculate_ecc()
ARM: dts: omap3430-sdp: Revert to using software ECC for NAND
ARM: OMAP2+: GPMC: Support Software ECC scheme via DT
mtd: nand: omap: Revert to using software ECC by default
ARM: dts: hummingboard/cubox-i: change SPDIF output to be more descriptive
ARM: dts: hummingboard/cubox-i: add USB OC pinctrl configuration
ARM: shmobile: r8a7791: add missing 0x0100 for SDCKCR
...
|
|
With ARCH_VEXPRESS_SPC option, kernel build has the following
warning:
arch/arm/mach-vexpress/spc.c: In function ‘ve_spc_clk_init’:
arch/arm/mach-vexpress/spc.c:431:38: warning: array subscript is below array bounds [-Warray-bounds]
struct ve_spc_opp *opps = info->opps[cluster];
^
since 'cluster' maybe '-1' in UP system. This patch does a active
checking to fix this issue.
Signed-off-by: Alex Shi <[email protected]>
Acked-by: Pawel Moll <[email protected]>
Acked-by: Sudeep Holla <[email protected]>
Signed-off-by: Olof Johansson <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/pjw/omap-pending into fixes
Pull "ARM: OMAP2+: DRA72x/DRA74x basic support" from Tony Lindgren:
Add basic subarchitecture support for the DRA72x and DRA74x. These
are OMAP2+ derivative SoCs. This should be low-risk to existing OMAP
platforms.
Basic build, boot, and PM test logs are available here:
http://www.pwsan.com/omap/testlogs/hwmod-a-early-v3.17-rc/20140827194314/
* tag 'for-v3.17-rc/omap-dra72x-d74x-support-a' of git://git.kernel.org/pub/scm/linux/kernel/git/pjw/omap-pending:
ARM: DRA7: hwmod: Add dra74x and dra72x specific ocp interface lists
ARM: DRA7: Add support for soc_is_dra74x() and soc_is_dra72x() variants
Signed-off-by: Olof Johansson <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi
Pull spi bugfixes from Mark Brown:
"A smattering of bug fixes for the SPI subsystem, all in driver code
which has seen active work recently and none of them with any great
global impact.
There's also a new ACPI ID for the pxa2xx driver which required no
code changes and the addition of kerneldoc for some structure fields
that were missing it and generating warnings during documentation
builds as a result"
* tag 'spi-v3.17-rc3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
spi: sh-msiof: Fix transmit-only DMA transfers
spi/rockchip: Avoid accidentally turning off the clock
spi: dw: fix kernel crash due to NULL pointer dereference
spi: dw-pci: fix bug when regs left uninitialized
spi: davinci: fix SPI_NO_CS functionality
spi/rockchip: fixup incorrect dma direction setting
spi/pxa2xx: Add ACPI ID for Intel Braswell
spi: spi-au1550: fix build failure
spi: rspi: Fix leaking of unused DMA descriptors
spi: sh-msiof: Fix leaking of unused DMA descriptors
spi: Add missing kerneldoc bits
spi/omap-mcspi: Fix the spi task hangs waiting dma_rx
|
|
|
|
spi: Bug fixes for v3.17
A smattering of bug fixes for the SPI subsystem, all in driver code
which has seen active work recently and none of them with any great
global impact.
There's also a new ACPI ID for the pxa2xx driver which required no code
changes and the addition of kerneldoc for some structure fields that
were missing it and generating warnings during documentation builds as a
result.
# gpg: Signature made Sun 31 Aug 2014 13:19:12 BST using RSA key ID 7EA229BD
# gpg: Good signature from "Mark Brown <[email protected]>"
# gpg: aka "Mark Brown <[email protected]>"
# gpg: aka "Mark Brown <[email protected]>"
# gpg: aka "Mark Brown <[email protected]>"
# gpg: aka "Mark Brown <[email protected]>"
# gpg: aka "Mark Brown <[email protected]>"
|
|
Fix tx/rx mixup, which broke transmit-only transfers.
Introduced by commit 4240305f7cbdc7782aa8bc40cc702775d9ac0839
("spi: sh-msiof: Fix leaking of unused DMA descriptors").
Reported-by: Laurent Pinchart <[email protected]>
Signed-off-by: Geert Uytterhoeven <[email protected]>
Acked-by: Laurent Pinchart <[email protected]>
Signed-off-by: Mark Brown <[email protected]>
|
|
Pull file locking bugfx from Jeff Layton:
"Just a bugfix for a bug that crept in to v3.15. It's in a rather rare
error path, and I'm not aware of anyone having hit it, but it's worth
fixing for v3.17"
* tag 'locks-v3.17-3' of git://git.samba.org/jlayton/linux:
locks: pass correct "before" pointer to locks_unlink_lock in generic_add_lease
|
|
'spi/fix/doc', 'spi/fix/dw', 'spi/fix/omap-mcspi', 'spi/fix/pxa2xx', 'spi/fix/rockchip' and 'spi/fix/rspi' into spi-linus
|
|
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Peter Anvin:
"One patch to avoid assigning interrupts we don't actually have on
non-PC platforms, and two patches that addresses bugs in the new
IOAPIC assignment code"
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86, irq, PCI: Keep IRQ assignment for runtime power management
x86: irq: Fix bug in setting IOAPIC pin attributes
x86: Fix non-PC platform kernel crash on boot due to NULL dereference
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull ACPI and power management fixes from Rafael Wysocki:
- Fix for an ACPI regression related to the handling of fixed events
that caused netlink routines to be (incorrectly) run in interrupt
context from Lan Tianyu
- Fix for an ACPI EC driver regression on Acer Aspire V5-573G that
caused AC/battery plug/unplug and video brightness change
notifications to be delayed on that machine from Lv Zheng
- Fix for an ACPI device enumeration regression that caused ACPI driver
probe to fail for some devices where it succeeded before (Rafael J
Wysocki)
- intel_pstate driver fix to prevent it from printing an information
message for every CPU in the system on every boot from Andi Kleen
- s5pv210 cpufreq driver fix to remove an __init annotation from a
routine that in fact can be called at any time after init too from
Mark Brown
- New Intel Braswell device ID for the ACPI LPSS (Low-Power Subsystem)
driver from Alan Cox
- New Intel Braswell CPU ID for intel_pstate from Mika Westerberg
* tag 'pm+acpi-3.17-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
cpufreq: s5pv210: Remove spurious __init annotation
cpufreq: intel_pstate: Add CPU ID for Braswell processor
intel_pstate: Turn per cpu printk into pr_debug
ACPI / LPSS: Add ACPI IDs for Intel Braswell
ACPI / EC: Add support to disallow QR_EC to be issued before completing previous QR_EC
ACPI / EC: Add support to disallow QR_EC to be issued when SCI_EVT isn't set
ACPI: Run fixed event device notifications in process context
ACPI / scan: Allow ACPI drivers to bind to PNP device objects
|
|
Merge patches from Andrew Morton:
"22 fixes"
* emailed patches from Andrew Morton <[email protected]>: (22 commits)
kexec: purgatory: add clean-up for purgatory directory
Documentation/kdump/kdump.txt: add ARM description
flush_icache_range: export symbol to fix build errors
tools: selftests: fix build issue with make kselftests target
ocfs2: quorum: add a log for node not fenced
ocfs2: o2net: set tcp user timeout to max value
ocfs2: o2net: don't shutdown connection when idle timeout
ocfs2: do not write error flag to user structure we cannot copy from/to
x86/purgatory: use approprate -m64/-32 build flag for arch/x86/purgatory
drivers/rtc/rtc-s5m.c: re-add support for devices without irq specified
xattr: fix check for simultaneous glibc header inclusion
kexec: remove CONFIG_KEXEC dependency on crypto
kexec: create a new config option CONFIG_KEXEC_FILE for new syscall
x86,mm: fix pte_special versus pte_numa
hugetlb_cgroup: use lockdep_assert_held rather than spin_is_locked
mm/zpool: use prefixed module loading
zram: fix incorrect stat with failed_reads
lib: turn CONFIG_STACKTRACE into an actual option.
mm: actually clear pmd_numa before invalidating
memblock, memhotplug: fix wrong type in memblock_find_in_range_node().
...
|
|
Without this patch the kexec-purgatory.c and purgatory.ro files are not
removed after make mrproper.
Signed-off-by: Michael Welling <[email protected]>
Acked-by: Vivek Goyal <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Add arm specific parts to kdump kernel documentation.
Signed-off-by: Hu Keping <[email protected]>
Acked-by: Vivek Goyal <[email protected]>
Cc: Haren Myneni <[email protected]>
Cc: Rob Landley <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Fix building errors occuring due to a missing export of
flush_icache_range() in
kisskb.ellerman.id.au/kisskb/buildresult/11677809/
ERROR: "flush_icache_range" [drivers/misc/lkdtm.ko] undefined!
Signed-off-by: Pranith Kumar <[email protected]>
Reported-by: Geert Uytterhoeven <[email protected]>
Acked-by: Vineet Gupta <[email protected]> [arc]
Acked-by: Richard Kuo <[email protected]> [hexagon]
Cc: Chris Metcalf <[email protected]>
Cc: Chris Zankel <[email protected]>
Acked-by: Max Filippov <[email protected]> [xtensa]
Cc: Noam Camus <[email protected]>
Cc: Masami Hiramatsu <[email protected]>
Acked-by: Zhigang Lu <[email protected]> [tile]
Cc: Kirill Tkhai <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Fix the typo of ARCH when running 'make kselftests'. Change the 'X86'
to 'x86'. Test by compilation.
Signed-off-by: Phong Tran <[email protected]>
Cc: David Herrmann <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Shuah Khan <[email protected]>
Cc: Sam Ravnborg <[email protected]>
Cc: Michal Marek <[email protected]>
Cc: Shuah Khan <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
For debug use, we can see from the log whether the fence decision is
made and why it is not fenced.
Signed-off-by: Junxiao Bi <[email protected]>
Reviewed-by: Srinivas Eeda <[email protected]>
Reviewed-by: Mark Fasheh <[email protected]>
Cc: Joel Becker <[email protected]>
Cc: Joseph Qi <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
When tcp retransmit timeout(15mins), the connection will be closed.
Pending messages may be lost during this time. So we set tcp user
timeout to override the retransmit timeout to the max value. This is OK
for ocfs2 since we have disk heartbeat, if peer crash, the disk
heartbeat will timeout and it will be evicted, if disk heartbeat not
timeout and connection idle for a long time, then this means the cluster
enters split-brain state, since fence can't happen, we'd better keep the
connection and wait network recover.
Signed-off-by: Junxiao Bi <[email protected]>
Reviewed-by: Srinivas Eeda <[email protected]>
Reviewed-by: Mark Fasheh <[email protected]>
Cc: Joel Becker <[email protected]>
Cc: Joseph Qi <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
This patch series is to fix a possible message lost bug in ocfs2 when
network go bad. This bug will cause ocfs2 hung forever even network
become good again.
The messages may lost in this case. After the tcp connection is
established between two nodes, an idle timer will be set to check its
state periodically, if no messages are received during this time, idle
timer will timeout, it will shutdown the connection and try to
reconnect, so pending messages in tcp queues will be lost. This
messages may be from dlm. Dlm may get hung in this case. This may
cause the whole ocfs2 cluster hung.
This is very possible to happen when network state goes bad. Do the
reconnect is useless, it will fail if network state is still bad. Just
waiting there for network recovering may be a good idea, it will not
lost messages and some node will be fenced until cluster goes into
split-brain state, for this case, Tcp user timeout is used to override
the tcp retransmit timeout. It will timeout after 25 days, user should
have notice this through the provided log and fix the network, if they
don't, ocfs2 will fall back to original reconnect way.
This patch (of 3):
Some messages in the tcp queue maybe lost if we shutdown the connection
and reconnect when idle timeout. If packets lost and reconnect success,
then the ocfs2 cluster maybe hung.
To fix this, we can leave the connection there and do the fence decision
when idle timeout, if network recover before fence dicision is made, the
connection survive without lost any messages.
This bug can be saw when network state go bad. It may cause ocfs2 hung
forever if some packets lost. With this fix, ocfs2 will recover from
hung if network becomes good again.
Signed-off-by: Junxiao Bi <[email protected]>
Reviewed-by: Srinivas Eeda <[email protected]>
Reviewed-by: Mark Fasheh <[email protected]>
Cc: Joel Becker <[email protected]>
Cc: Joseph Qi <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
If we failed to copy from the structure, writing back the flags leaks 31
bits of kernel memory (the rest of the ir_flags field).
In any case, if we cannot copy from/to the structure, why should we
expect putting just the flags to work?
Also make sure ocfs2_info_handle_freeinode() returns the right error
code if the copy_to_user() fails.
Fixes: ddee5cdb70e6 ('Ocfs2: Add new OCFS2_IOC_INFO ioctl for ocfs2 v8.')
Signed-off-by: Ben Hutchings <[email protected]>
Cc: Joel Becker <[email protected]>
Acked-by: Mark Fasheh <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Thomas reported that build of x86_64 kernel was failing for him. He is
using 32bit tool chain.
Problem is that while compiling purgatory, I have not specified -m64
flag. And 32bit tool chain must be assuming -m32 by default.
Following is error message.
(mini) [~/work/linux-2.6] make
scripts/kconfig/conf --silentoldconfig Kconfig
CHK include/config/kernel.release
UPD include/config/kernel.release
CHK include/generated/uapi/linux/version.h
CHK include/generated/utsrelease.h
UPD include/generated/utsrelease.h
CC arch/x86/purgatory/purgatory.o
arch/x86/purgatory/purgatory.c:1:0: error: code model 'large' not supported in
the 32 bit mode
Fix it by explicitly passing appropriate -m64/-m32 build flag for
purgatory.
Reported-by: Thomas Glanzmann <[email protected]>
Tested-by: Thomas Glanzmann <[email protected]>
Suggested-by: H. Peter Anvin <[email protected]>
Signed-off-by: Vivek Goyal <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|