Enable use_sigaltstack=1 for ASan builds. This will ease the stack overflow detection.

Enable use_sigaltstack=1 for ASan builds. This will ease the stack overflow detection.
Also whitelist __NR_sigaltstack in seccomp-bpf under ASan.

BUG=302040
R=jln@chromium.org
TBR=thakis@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=267220

[Alexander Potapenko, 2014-04-29 09:07:21]

Julien, can you please take a look?

[jln (very slow on Chromium), 2014-04-29 17:49:54]

On 2014/04/29 09:07:21, Alexander Potapenko wrote:
> Julien, can you please take a look?

Could you change baseline_policy.cc (EvaluateSyscallImpl) instead?

Just put it at the top, above the "if (IsBaselinePolicyAllowed(sysno)) ..."
statement.

Thanks!

[jln (very slow on Chromium), 2014-04-30 05:52:24]

On 2014/04/29 17:49:54, jln wrote:
> On 2014/04/29 09:07:21, Alexander Potapenko wrote:
> > Julien, can you please take a look?
> 
> Could you change baseline_policy.cc (EvaluateSyscallImpl) instead?
> 
> Just put it at the top, above the "if (IsBaselinePolicyAllowed(sysno)) ..."
> statement.
> 
> Thanks!

lgtm in advance if you move it to baseline_policy.cc::EvaluateSyscallImpl(), so
that we don't loose another day due to timezone differences.

[Alexander Potapenko, 2014-04-30 10:54:31]

The CQ bit was checked by glider@chromium.org

[commit-bot: I haz the power, 2014-04-30 10:54:38]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/glider@chromium.org/253753003/40001

[commit-bot: I haz the power, 2014-04-30 10:58:24]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-04-30 10:58:24]

Try jobs failed on following builders:
  linux_chromium_clang_dbg on tryserver.chromium

[Alexander Potapenko, 2014-04-30 12:00:39]

The CQ bit was checked by glider@chromium.org

[commit-bot: I haz the power, 2014-04-30 12:01:16]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/glider@chromium.org/253753003/40001

[commit-bot: I haz the power, 2014-04-30 12:04:37]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-04-30 12:04:37]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium

[Alexander Potapenko, 2014-04-30 13:36:07]

The CQ bit was checked by glider@chromium.org

[commit-bot: I haz the power, 2014-04-30 13:36:40]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/glider@chromium.org/253753003/40001

[commit-bot: I haz the power, 2014-04-30 13:49:44]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-04-30 13:49:45]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium

[Alexander Potapenko, 2014-04-30 14:15:49]

On 2014/04/30 13:49:45, I haz the power (commit-bot) wrote:
> Try jobs failed on following builders:
>   chromium_presubmit on tryserver.chromium

TBRing Nico for base/debug ownership.
Nico, care if I add myself as a per-file owner of sanitizer_options.cc?

[Alexander Potapenko, 2014-04-30 14:16:33]

TBRing Nico for base/debug ownership.
Nico, care if I add myself as a per-file owner of sanitizer_options.cc?

[Alexander Potapenko, 2014-04-30 14:16:38]

The CQ bit was checked by glider@chromium.org

[commit-bot: I haz the power, 2014-04-30 14:17:08]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/glider@chromium.org/253753003/40001

[commit-bot: I haz the power, 2014-04-30 15:18:01]

Change committed as 267220

[Nico, 2014-04-30 15:23:38]

base lgtm

The file owner suggestions sounds good!

https://codereview.chromium.org/253753003/diff/40001/base/debug/sanitizer_opt...
File base/debug/sanitizer_options.cc (right):

https://codereview.chromium.org/253753003/diff/40001/base/debug/sanitizer_opt...
base/debug/sanitizer_options.cc:39: //   use_sigaltstack=1 - handle signals on
an alternate signal stack. Useful
Typo use_sigalstack (missing n)

…oh, below too, so I suppose use_sigaltstack is the correct name. That seems
like an unfortunate name, as it looks like a typo for use_signalstack which
means the same thing and doesn't look like a typo :-/

[jln (very slow on Chromium), 2014-04-30 16:05:08]

https://codereview.chromium.org/253753003/diff/40001/sandbox/linux/seccomp-bp...
File sandbox/linux/seccomp-bpf-helpers/syscall_sets.cc (left):

https://codereview.chromium.org/253753003/diff/40001/sandbox/linux/seccomp-bp...
sandbox/linux/seccomp-bpf-helpers/syscall_sets.cc:302: case __NR_sigaltstack:
FYI you should dnot have removed this, the syscall sets must form a partition of
all the syscalls. I'll fix it.

[Alexander Potapenko, 2014-04-30 16:08:15]

https://codereview.chromium.org/253753003/diff/40001/base/debug/sanitizer_opt...
File base/debug/sanitizer_options.cc (right):

https://codereview.chromium.org/253753003/diff/40001/base/debug/sanitizer_opt...
base/debug/sanitizer_options.cc:39: //   use_sigaltstack=1 - handle signals on
an alternate signal stack. Useful
That's named after the sigaltstack() function :)