[Python-Dev] Possible GIL/threading issue involving subprocess and PyMem_MALLOC...

[Trent Nelson]

On Fri, Dec 21, 2012 at 01:43:11AM -0800, Antoine Pitrou wrote:
> Le Fri, 21 Dec 2012 09:31:44 +0000,
> Kristján Valur Jónsson <kristjan at ccpgames.com> a écrit :
> > I ran into this the other day.  I had put in hooks in the
> > PyMem_MALLOC to track memory per tasklet, and it crashed in those
> > cases because it was being called without the GIL.  My local patch
> > was simply to _not_ release the GIL. Clearly, calling PyMem_MALLOC
> > without the GIL is an API violation.
> 
> Indeed, this deserves fixing.
> (it would be better to still release the GIL around the low-level I/O
> call, of course)

    Created http://bugs.python.org/issue16742 to capture the issue for
    now.  I want to make some more progress on the parallel stuff first
    so if somebody wants to tackle it in the meantime, be my guest.

> Thanks Trent for finding this!

    Unexpected (but handy) side-effect of the parallel context work :-)
    (I wonder if that's the only thread-safe issue in our code base...)

> Antoine.

        Trent.