[Bug 1222] race condition in javac, taking all the CPU.

bugzilla-daemon at icedtea.classpath.org bugzilla-daemon at icedtea.classpath.org
Mon Nov 19 09:09:58 PST 2012


--- Comment #3 from Alexandre Bique <bique.alexandre at gmail.com> ---
Well, to me it looks like the parser is multithreaded and uses a
producer/consumer design pattern.

I have the feeling that this producer/consumer is not very well implemented and
tries to use wait-free and weird stuff, that's why we end up with a race.

When I strace java, I get a lot of sched_yield() and futex() stuff.

I really think that you should try to find some pthread_mutex_try_lock() in the
compiler source code, and review their usage.

Also I think the race comes from the parser and I don't know the java source
code, but I would bet that the parser is shared by the sun vm and this one as
the bug also triggers with the sun jdk.

You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/distro-pkg-dev/attachments/20121119/3209eb5e/attachment.html 

More information about the distro-pkg-dev mailing list