diff options
| author | Russell King <rmk+kernel@armlinux.org.uk> | 2016-05-30 23:14:56 +0100 |
|---|---|---|
| committer | Sasha Levin <sasha.levin@oracle.com> | 2016-06-17 15:37:18 -0400 |
| commit | 5642859e6a68f8c8c1e0c5163fad29ad4497f6f5 (patch) | |
| tree | 5617c63680d361173bbd9c733cc3f2d9c5fafd65 | |
| parent | 34e6c73e0bebcec0ff4788170b1abd51cbf683ad (diff) | |
| download | linux-stable-5642859e6a68f8c8c1e0c5163fad29ad4497f6f5.tar.gz linux-stable-5642859e6a68f8c8c1e0c5163fad29ad4497f6f5.tar.bz2 linux-stable-5642859e6a68f8c8c1e0c5163fad29ad4497f6f5.zip | |
ARM: fix PTRACE_SETVFPREGS on SMP systems
[ Upstream commit e2dfb4b880146bfd4b6aa8e138c0205407cebbaf ]
PTRACE_SETVFPREGS fails to properly mark the VFP register set to be
reloaded, because it undoes one of the effects of vfp_flush_hwstate().
Specifically vfp_flush_hwstate() sets thread->vfpstate.hard.cpu to
an invalid CPU number, but vfp_set() overwrites this with the original
CPU number, thereby rendering the hardware state as apparently "valid",
even though the software state is more recent.
Fix this by reverting the previous change.
Cc: <stable@vger.kernel.org>
Fixes: 8130b9d7b9d8 ("ARM: 7308/1: vfp: flush thread hwstate before copying ptrace registers")
Acked-by: Will Deacon <will.deacon@arm.com>
Tested-by: Simon Marchi <simon.marchi@ericsson.com>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
| -rw-r--r-- | arch/arm/kernel/ptrace.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/arch/arm/kernel/ptrace.c b/arch/arm/kernel/ptrace.c index ef9119f7462e..4d9375814b53 100644 --- a/arch/arm/kernel/ptrace.c +++ b/arch/arm/kernel/ptrace.c @@ -733,8 +733,8 @@ static int vfp_set(struct task_struct *target, if (ret) return ret; - vfp_flush_hwstate(thread); thread->vfpstate.hard = new_vfp; + vfp_flush_hwstate(thread); return 0; } |