danielaustin updated this revision to Diff 43459.
danielaustin added a comment.
Test case added which looks for the existence of multiple trap basic blocks in
generated LLVM-IR with optimization turned on. Flag renamed
(-fmerge-traps/-fno-merge-traps) and placed in a more sensible location
(Cod
danielaustin added a comment.
Test makes sense, will add it with the revisions.
Comment at: include/clang/Basic/LangOptions.h:95
@@ -94,1 +94,3 @@
+ /// \brief Flag controlling whether or not trap calls are merged
+ /// at the end of each function.
samsonov
danielaustin updated this revision to Diff 42328.
danielaustin added a comment.
Patch that removes the debug-mode sanitizer setting and makes the
-fsanitize-merge-traps/-fno-sanitize-merge-traps flags, with the default
setting being to merge traps. Flags tested within the Android build system,
I don't think super-fine grained control would be required, a per-binary
setting should be sufficient for debugging purposes.
If everyone is ok with it, I'll get a diff up later tonight switching the
flag to fsanitize-merge-traps/fnosanitize-merge-traps
On Tue, Dec 8, 2015 at 1:17 PM Evgeniy Step
danielaustin added a comment.
I don't think super-fine grained control would be required, a per-binary
setting should be sufficient for debugging purposes.
If everyone is ok with it, I'll get a diff up later tonight switching the
flag to fsanitize-merge-traps/fnosanitize-merge-traps
Repository:
danielaustin added a comment.
I'm not partial to what its called, or where it is. I'm completely ok with
-fsanitize-merge-checks
Repository:
rL LLVM
http://reviews.llvm.org/D15208
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lis
I'm not partial to what its called, or where it is. I'm completely ok with
-fsanitize-merge-checks
On Tue, Dec 8, 2015 at 1:07 PM Evgeniy Stepanov wrote:
> eugenis added a comment.
>
> Better -fsanitize-merge-checks, and it should apply to non-trap checks as
> well (i.e. SIGILL address should un
danielaustin updated this revision to Diff 41957.
danielaustin added a comment.
Added a flag controlling whether or not the inlining occurs. Tested with ARM 32
bit (shamu) with AOSP. The patched compiler resulted in correct code generation
and no stability issues. The flag is probably not ideall
Makes sense, will add that as well.
On Thu, Dec 3, 2015 at 4:10 PM Alexey Samsonov wrote:
> samsonov added a comment.
>
> I agree with Richard here.
>
>
> Repository:
> rL LLVM
>
> http://reviews.llvm.org/D15208
>
>
>
>
___
cfe-commits mailing list
c
