New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Profiling timer always armed on initialization #409
Comments
Reported by - _Attachment: [gperftools_prof_signals.patch](https://storage.googleapis.com/google-code-attachments/gperftools/issue-406/comment-1/gperftools_prof_signals.patch)_ |
Reported by |
Reported by |
Reported by - _Attachment: [gperftools_prof_signals.patch](https://storage.googleapis.com/google-code-attachments/gperftools/issue-406/comment-4/gperftools_prof_signals.patch)_ |
Reported by |
Reported by |
Reported by
|
Reported by |
Reported by |
Reported by |
Reported by |
Reported by |
Reported by - _Attachment: [gperftools_prof_signals.patch](https://storage.googleapis.com/google-code-attachments/gperftools/issue-406/comment-13/gperftools_prof_signals.patch)_ |
Reported by |
Reported by |
Does anybody know what the current status on this is? I would like this issue fixed, and would be happy to take over on getting it merged. I think @saittam is the original reporter. Any chance you're still interested in working on this, and if not, do you mind if I start with what you did? |
I'm the original reporter indeed. Unfortunately, this has lingered in the hardly-will-ever-get-to portion of my TODO list since forever. I originally got interested in this because the continuously firing signals tended to confuse gdb (which is a different bug). I've since carried the patch that's attached to the issue in my working trees, but there was nothing else that provided a strong motivation to invest time into pushing for a resolution. At this point, I can't tell whether the problem still exists, and whether the patch still applies... sorry. The original reproduction instructions should still be valid though, so if you're seeing SIGPROF firing in a single-threaded test program after calling ProfileHandler::RegisterThread, the issue is probably still valid. My lack of spare time to spend is only going to get worse in the upcoming months, so I'd be most delighted if you were to pick this issue up and resolve it. |
It causes a noticeable performance hit and can sometimes confuse GDB. Tested with and without CPUPROFILE_PER_THREAD_TIMERS=1, and on a linuxthreads system (CPUPROFILE_PER_THREAD_TIMERS=1 isn't supported there) (I couldn't get all the tests to run there before, but profile_handler_unittest and profiler_unittest.sh passed). Originally written by mnissler@google.com. I mostly had to deal with integrating with CPUPROFILE_PER_THREAD_TIMERS. Fixes gperftools#409
It causes a noticeable performance hit and can sometimes confuse GDB. Tested with and without CPUPROFILE_PER_THREAD_TIMERS=1, and on a linuxthreads system (CPUPROFILE_PER_THREAD_TIMERS=1 isn't supported there) (I couldn't get all the tests to run there before, but profile_handler_unittest and profiler_unittest.sh passed). Originally written by mnissler@google.com. I mostly had to deal with integrating with CPUPROFILE_PER_THREAD_TIMERS. Fixes gperftools#409
Fixed by #762. Thanks to Brian Silverman's great effort. |
Originally reported on Google Code with ID 406
Reported by
mnissler@google.com
on 2012-02-21 20:50:07The text was updated successfully, but these errors were encountered: