This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug target/52080] Stores to bitfields introduce a store-data-race on adjacent data


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52080

--- Comment #11 from Petr Tesarik <gcc-bugs at tesarici dot cz> 2012-02-02 12:39:25 UTC ---
(In reply to comment #10)
> IA64 uses
> 
>         .mmi
>         ld8.acq r14 = [r32]
>         ;;
>         nop 0
>         dep r14 = r15, r14, 32, 1
>         ;;
>         .mib
>         st8.rel [r32] = r14
> 
> which seems to be an attempt to work around this issue (albeit a
> possibly very slow one).

Are you referring to the ".acq" and ".rel" forms? That doesn't change the
situation at all. All it does is ensure correct memory ordering with respect to
external visibility, but it does nothing to avoid the race condition.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]