On 24/09/2013 23:33, Francisco Facioni wrote:
> The actual bug seems to be: http://llvm.org/bugs/show_bug.cgi?id=15429
>
> And this 2 are the patches:
>
> http://llvm.org/viewvc/llvm-project?view=revision&revision=181728
> http://llvm.org/viewvc/llvm-project?view=revision&revision=181750
>
> It
The actual bug seems to be: http://llvm.org/bugs/show_bug.cgi?id=15429
And this 2 are the patches:
http://llvm.org/viewvc/llvm-project?view=revision&revision=181728
http://llvm.org/viewvc/llvm-project?view=revision&revision=181750
It doesn't look hard to backport.
Cheers,
Francisco Facioni
Le 01/06/2013 04:53, Sjors Gielen a écrit :
> Erratum: even though revision 183033 is supposed to fix this bug "on
> architectures for which we don't provide these operations as builtins
> (e.g. ARM)", I've just compiled Clang from the earlier revision 182770 and in
> there, the following test pr
Erratum: even though revision 183033 is supposed to fix this bug "on
architectures for which we don't provide these operations as builtins
(e.g. ARM)", I've just compiled Clang from the earlier revision 182770 and in
there, the following test program already compiles and runs fine:
sjors@baz:~
This bug is fixed in Clang trunk, revision 183033:
http://llvm.org/viewvc/llvm-project?view=revision&revision=183033
This fix will be in Clang 3.4, probably not in Clang 3.3. Considering this bug
makes clang-3.2 unusable on armhf for any serious code, maybe a back-port of
this patch would be so
5 matches
Mail list logo
