Re: Pass the SIGPROF signal on to previously registered signal handler. (issue 11195045)

[will...@chromium.org]

Sorry for the delay, I think I've address the comment now.

https://codereview.chromium.org/11195045/

[da...@chromium.org]

lgtm, I'll land this for you.

https://codereview.chromium.org/11195045/

[da...@chromium.org]

Hold on, although the patch applies cleanly, the call of
SignalSender::CallOldSignalHandler isn't declared before it's use, so
compilation fails. Can you please address this (make sure that "make check
-j15"
runs on your V8 directory) and resubmit the patch?

https://codereview.chromium.org/11195045/

[will...@chromium.org]

OK, I think this should work for real this time.

https://codereview.chromium.org/11195045/

[da...@chromium.org]

lgtm, I'll land this

http://codereview.chromium.org/11195045/

[mstar...@chromium.org]

Two Webkit tests started crashing with this change on both (32 bit and 64
bit)
Webkit Linux integration builders. The name strongly suggests that it is
actually related to this change and not just a flake.

Regressions: Unexpected crashes (2)
inspector/profiler/cpu-profiler-profiling-without-inspector.html
inspector/profiler/cpu-profiler-profiling.html

http://build.chromium.org/p/client.v8/builders/Webkit%20Linux/builds/4988
http://build.chromium.org/p/client.v8/builders/Webkit%20Linux%2064/builds/4975

https://codereview.chromium.org/11195045/

[William Chan (陈智昌)]

That sounds likely. I will return to the office next week and would be happy to look at this then. Feel free to revert this in the meanwhile.

[Michael Starzinger]

OK, I guess in that case we will have to revert the change, because it will prevent us from rolling into Chromium otherwise. I'll do the revert.

Best regards,

Michael