http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48076
--- Comment #7 from torvald at gcc dot gnu.org 2012-11-28 14:29:47 UTC ---
(In reply to comment #6)
> There seems to be a similar bug in code generated for function static
> variables.
> The fast-path load is a plain load rather than atomic
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48076
--- Comment #6 from Dmitry Vyukov 2012-11-28
03:56:50 UTC ---
There seems to be a similar bug in code generated for function static
variables.
The fast-path load is a plain load rather than atomic acquire load.
On Sat, Nov 24, 2012 at 3
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48076
Dmitry Vyukov changed:
What|Removed |Added
CC||dvyukov at google dot com
--- C
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48076
--- Comment #4 from Richard Henderson 2012-11-27
22:05:34 UTC ---
Created attachment 28798
--> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28798
proposed patch
This should fix the race that I eventually saw (read), as well as one th
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48076
Richard Henderson changed:
What|Removed |Added
Status|UNCONFIRMED |ASSIGNED
Last reconfirmed|
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48076
--- Comment #2 from Richard Henderson 2012-11-27
18:01:17 UTC ---
Are you sure this isn't a false-positive?
The way I read this code, it is certainly possible for the optimizer
(or the processor) to prefetch emutls_key before the load of
