kosarev added a comment.
Responded here: https://bugs.llvm.org/show_bug.cgi?id=26603
Repository:
rL LLVM
https://reviews.llvm.org/D39455
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-co
aheejin added a comment.
Thank you for confirming and reverting!
Repository:
rL LLVM
https://reviews.llvm.org/D39455
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
hfinkel added a comment.
In https://reviews.llvm.org/D39455#943182, @aheejin wrote:
> Once you confirm the bug, could you possibly revert the patch?
I agree. We should revert this. The relevant part of the test case is:
short *q;
p->u.vec[i] = 0;
q = &p->u.vec[16];
*q = 1;
return p->
aheejin added a comment.
Once you confirm the bug, could you possibly revert the patch?
Repository:
rL LLVM
https://reviews.llvm.org/D39455
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe
aheejin added a comment.
After this patch, a couple of gcc torture tests started to fail. Could you
possibly look at this? https://bugs.llvm.org/show_bug.cgi?id=35503
Repository:
rL LLVM
https://reviews.llvm.org/D39455
___
cfe-commits mailing li
This revision was automatically updated to reflect the committed changes.
Closed by commit rL319413: [CodeGen] Add initial support for union members in
TBAA (authored by kosarev).
Changed prior to commit:
https://reviews.llvm.org/D39455?vs=124726&id=124883#toc
Repository:
rL LLVM
https://re
kosarev added a comment.
Thanks Hal. Will commit it tomorrow.
https://reviews.llvm.org/D39455
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
kosarev updated this revision to Diff 124726.
kosarev added a comment.
- Added tests with nested unions.
- Rebased.
https://reviews.llvm.org/D39455
Files:
lib/CodeGen/CGExpr.cpp
lib/CodeGen/CodeGenModule.h
lib/CodeGen/CodeGenTBAA.cpp
lib/CodeGen/CodeGenTBAA.h
test/CodeGen/tbaa-union.c
hfinkel accepted this revision.
hfinkel added a comment.
This revision is now accepted and ready to land.
I think this looks fine (and I agree that the underlying premise makes sense).
Before you commit, could you also please add some tests with nested union
members?
Repository:
rL LLVM
htt
kosarev added a comment.
Thanks John.
Hal, do you see any problems with this patch?
Repository:
rL LLVM
https://reviews.llvm.org/D39455
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-co
rjmccall added a comment.
I think the implementation looks fine. The metadata design also sounds good to
me, but I'd like Hal to weigh in before you commit.
Repository:
rL LLVM
https://reviews.llvm.org/D39455
___
cfe-commits mailing list
cfe-co
kosarev added a comment.
Colleagues, please let me know if there are any concerns about this patch. It
was specifically designed to not cause any substantial changes, but I may be
missing something important. Thanks.
Repository:
rL LLVM
https://reviews.llvm.org/D39455
___
kosarev created this revision.
kosarev added a project: clang.
The basic idea behind this patch is that since in strict aliasing mode all
accesses to union members require their outermost enclosing union objects to be
specified explicitly, then for a couple given accesses to union members of the
13 matches
Mail list logo