Handle inconsistent PR_SET_TIMERSLACK_PID values on Android.

Handle inconsistent PR_SET_TIMERSLACK_PID values on Android.

This prctl is custom to Android, and depending on the Android kernel
version it has several different values. By random luck, the other
prctl numbers they clobbered do not look super scary (unless there is a
bug involving Intel mpx), so I opted to allow them all instead of
relying on runtime kernel version detection.

BUG=611916
Committed: https://crrev.com/41db13c68491ee1ab9149ba1bef307b71a19bb39
Cr-Commit-Position: refs/heads/master@{#395737}

[rickyz (no longer on Chrome), 2016-05-24 07:56:51]

rickyz@chromium.org changed reviewers:
+ rsesek@chromium.org

[Robert Sesek, 2016-05-24 14:51:40]

https://codereview.chromium.org/2005183002/diff/1/sandbox/linux/seccomp-bpf-h...
File sandbox/linux/seccomp-bpf-helpers/syscall_parameters_restrictions.cc
(right):

https://codereview.chromium.org/2005183002/diff/1/sandbox/linux/seccomp-bpf-h...
sandbox/linux/seccomp-bpf-helpers/syscall_parameters_restrictions.cc:160: #else
I think we need to permit all three unconditionally, otherwise if this is
compiled against one version of Bionic that includes PR_SET_TIMERSLACK_PID, it
will fail on other Androids using a different kernel version.

[rickyz (no longer on Chrome), 2016-05-24 21:04:06]

https://codereview.chromium.org/2005183002/diff/1/sandbox/linux/seccomp-bpf-h...
File sandbox/linux/seccomp-bpf-helpers/syscall_parameters_restrictions.cc
(right):

https://codereview.chromium.org/2005183002/diff/1/sandbox/linux/seccomp-bpf-h...
sandbox/linux/seccomp-bpf-helpers/syscall_parameters_restrictions.cc:160: #else
On 2016/05/24 at 14:51:40, Robert Sesek wrote:
> I think we need to permit all three unconditionally, otherwise if this is
compiled against one version of Bionic that includes PR_SET_TIMERSLACK_PID, it
will fail on other Androids using a different kernel version.

Oops, I don't know what I was thinking - thanks for the catch!

[Robert Sesek, 2016-05-24 23:14:02]

lgtm

[rickyz (no longer on Chrome), 2016-05-24 23:31:31]

The CQ bit was checked by rickyz@chromium.org

[commit-bot: I haz the power, 2016-05-24 23:32:32]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2005183002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/2005183002/20001

[commit-bot: I haz the power, 2016-05-24 23:37:22]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-05-24 23:40:30]

Description was changed from

==========
Handle inconsistent PR_SET_TIMERSLACK_PID values on Android.

This prctl is custom to Android, and depending on the Android kernel
version it has several different values. By random luck, the other
prctl numbers they clobbered do not look super scary (unless there is a
bug involving Intel mpx), so I opted to allow them all instead of
relying on runtime kernel version detection.

BUG=611916
==========

to

==========
Handle inconsistent PR_SET_TIMERSLACK_PID values on Android.

This prctl is custom to Android, and depending on the Android kernel
version it has several different values. By random luck, the other
prctl numbers they clobbered do not look super scary (unless there is a
bug involving Intel mpx), so I opted to allow them all instead of
relying on runtime kernel version detection.

BUG=611916

Committed: https://crrev.com/41db13c68491ee1ab9149ba1bef307b71a19bb39
Cr-Commit-Position: refs/heads/master@{#395737}
==========

[commit-bot: I haz the power, 2016-05-24 23:40:31]

Patchset 2 (id:??) landed as
https://crrev.com/41db13c68491ee1ab9149ba1bef307b71a19bb39
Cr-Commit-Position: refs/heads/master@{#395737}