Date: Tue, 29 Nov 2016 09:28:18 +0000 From: bugzilla-noreply@freebsd.org To: freebsd-bugs@FreeBSD.org Subject: [Bug 214923] kqueue hangs with busy loop Message-ID: <bug-214923-8-SdvjheSNmp@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-214923-8@https.bugs.freebsd.org/bugzilla/> References: <bug-214923-8@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D214923 Konstantin Belousov <kib@FreeBSD.org> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |kib@FreeBSD.org --- Comment #1 from Konstantin Belousov <kib@FreeBSD.org> --- (In reply to yjh0502 from comment #0) I do not see a clean indicator that the process 'spins on the kqueue syscal= l'.=20 The wait channel demonstrates that the process waits for some event subsyst= em mutex (the star before the name), but due to the truncated output, I am not sure which mutex is it. It could kqueue lock, or kqueue global lock, for instance. That said, WCPU 100% indicates that there is some other thread in the proce= ss which consumes cycles, which is consistent with the report of four threads = in the process. >From that data, I am not even convinced that this is a kernel bug and not s= ome application issue. For very beginning, look at the kernel backtraces of all threads in your process with procstat -kk. Find and backtrace the lock own= er of a kqueue lock. All that assuming that you believe that the issue is a kernel problem. Otherwise, and I would start with that route, would I debu= g it myself, is to look at the app and see what it is doing (wrong). --=20 You are receiving this mail because: You are the assignee for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-214923-8-SdvjheSNmp>