x86/asm/entry: Check for syscall exit work with IRQs disabled
authorAndy Lutomirski <luto@kernel.org>
Mon, 23 Mar 2015 19:32:54 +0000 (12:32 -0700)
committerIngo Molnar <mingo@kernel.org>
Tue, 24 Mar 2015 20:08:28 +0000 (21:08 +0100)
commitb3494a4ab20f6bdf74cdf2badf7918bb65ee8a00
tree2875f99133dd8b019b6aee8d0e81c79428630748
parentbc465aa9d045feb0e13b4a8f32cc33c1943f62d6
x86/asm/entry: Check for syscall exit work with IRQs disabled

We currently have a race: if we're preempted during syscall
exit, we can fail to process syscall return work that is queued
up while we're preempted in ret_from_sys_call after checking
ti.flags.

Fix it by disabling interrupts before checking ti.flags.

Reported-by: Stefan Seyfried <stefan.seyfried@googlemail.com>
Reported-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Acked-by: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Jiri Kosina <jkosina@suse.cz>
Cc: Tejun Heo <tj@kernel.org>
Fixes: 96b6352c1271 ("x86_64, entry: Remove the syscall exit audit")
Link: http://lkml.kernel.org/r/189320d42b4d671df78c10555976bb10af1ffc75.1427137498.git.luto@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
arch/x86/kernel/entry_64.S