[PATCH] Memory fencing problem in pthread cancellation

[Jeff Law]

On 01/14/2013 01:50 PM, Roland McGrath wrote:
> I'd like to see comments clarifying what the memory clobber is actually
> accomplishing and why it's necessary.
I suspect the asm memory clobber is historical, probably introduced to 
work around issues with pthread_cancel_init being inlined and the 
compiler moving things around too aggressively.  One could make an 
argument it should be removed.


> The other introductions of barriers
> that have no comments need comments.
Sure.  Happy to add them.

jeff