2010/11/28 Brian Harring <email address hidden>:
> Additional note... it *looks* like the deadlock potential is there
> already in 0.13, it's just heavily exacerbated by this patch- out of
> about 600 builds I've seen 2 lockup in the same fashion (rate was far
> higher with the patch on this ticket).
Thanks for the testing. I had a nasty feeling this wouldn't
be the only problem, which is why I didn't propose that patch
as a real fix.
(I think that running multithreaded programs under user-mode
emulation is effectively hitting a lot of the same locking issues
that you would get for emulating an MP core in multiple threads.)
2010/11/28 Brian Harring <email address hidden>:
> Additional note... it *looks* like the deadlock potential is there
> already in 0.13, it's just heavily exacerbated by this patch- out of
> about 600 builds I've seen 2 lockup in the same fashion (rate was far
> higher with the patch on this ticket).
Thanks for the testing. I had a nasty feeling this wouldn't
be the only problem, which is why I didn't propose that patch
as a real fix.
(I think that running multithreaded programs under user-mode
emulation is effectively hitting a lot of the same locking issues
that you would get for emulating an MP core in multiple threads.)
-- PMM