shenhan added a comment.
Thanks for pointing out the relevant CLs. I agree that's the clearer and better
solution.
(The only concern is that it has been sitting there for 2 months. I'll check
back.)
In https://reviews.llvm.org/D35849#820473, @vsk wrote:
> This won't do the right thing if more
mikerice created this revision.
When an omp for loop is canceled the constructed objects are being destructed
twice.
It looks like the desired code is:
{
Obj o;
If (cancelled) branch-through-cleanups to cancel.exit.
}
[cleanups]
cancel.exit:
__kmpc_for_static_fini
br cancel.cont (*)
b-sumner added a comment.
In https://reviews.llvm.org/D28691#820595, @rjmccall wrote:
> In https://reviews.llvm.org/D28691#820541, @b-sumner wrote:
>
> > There are other languages for heterogeneous compute that have scopes,
> > although not exposed quite as explicitly as OpenCL. For example AMD
rjmccall added a comment.
In https://reviews.llvm.org/D28691#820641, @b-sumner wrote:
> In https://reviews.llvm.org/D28691#820595, @rjmccall wrote:
>
> > In https://reviews.llvm.org/D28691#820541, @b-sumner wrote:
> >
> > > There are other languages for heterogeneous compute that have scopes,
>
Author: szepet
Date: Tue Jul 25 14:54:58 2017
New Revision: 309036
URL: http://llvm.org/viewvc/llvm-project?rev=309036&view=rev
Log:
[StaticAnalyzer] LoopUnrolling - Attempt to fix a crash in r309006.
Modified:
cfe/trunk/lib/StaticAnalyzer/Core/LoopUnrolling.cpp
Modified: cfe/trunk/lib/Stat
majnemer updated this revision to Diff 108166.
majnemer added a comment.
- Address review comments
https://reviews.llvm.org/D35824
Files:
include/clang/AST/Type.h
lib/AST/Type.cpp
lib/CodeGen/CodeGenTBAA.cpp
test/CodeGenCXX/std-byte.cpp
Index: test/CodeGenCXX/std-byte.cpp
danielcdh updated this revision to Diff 108174.
danielcdh added a comment.
update
https://reviews.llvm.org/D35746
Files:
lib/CodeGen/BackendUtil.cpp
Index: lib/CodeGen/BackendUtil.cpp
===
--- lib/CodeGen/BackendUtil.cpp
+++ lib
t-tye added inline comments.
Comment at: include/clang/Basic/Builtins.def:717
+ATOMIC_BUILTIN(__opencl_atomic_fetch_max, "v.", "t")
+
#undef ATOMIC_BUILTIN
Will the OpenCL 2.0 memory fences also be supported which also have a memory
order and memory scope?
==
mgrang created this revision.
Herald added subscribers: kristof.beyls, javed.absar, aemerson.
This fixes compiling with headers from the Windows SDK for ARM64.
https://reviews.llvm.org/D35862
Files:
lib/Headers/armintr.h
Index: lib/Headers/armintr.h
=
Quuxplusone created this revision.
`pointer_traits::to_pointer(r)` is not required to return a deallocatable
pointer; indeed generally it *cannot* determine a deallocatable representation
without help from the allocator. Therefore, we must not rely on representations
derived from `to_pointer` w
mgrang added a comment.
Refer
http://infocenter.arm.com/help/topic/com.arm.doc.ihi0053c/IHI0053C_acle_2_0.pdf
for memory barriers.
https://reviews.llvm.org/D35862
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bi
This revision was automatically updated to reflect the committed changes.
Closed by commit rL309049: [clang] Add abi-breaking-checks support to clang
(authored by mgrang).
Changed prior to commit:
https://reviews.llvm.org/D35426?vs=106671&id=108180#toc
Repository:
rL LLVM
https://reviews.ll
Author: mgrang
Date: Tue Jul 25 16:00:02 2017
New Revision: 309049
URL: http://llvm.org/viewvc/llvm-project?rev=309049&view=rev
Log:
[clang] Add abi-breaking-checks support to clang
Summary: You can now use REQUIRES:abi-breaking-checks in clang too
Reviewers: chapuni, probinson, ddunbar, jroelof
Looks like one more failure (this time under ubsan) remains in this code
http://lab.llvm.org:8011/builders/sanitizer-x86_64-linux-fast/builds/6708/steps/check-clang%20ubsan/logs/stdio
Please fix asap.
/mnt/b/sanitizer-buildbot3/sanitizer-x86_64-linux-fast/build/llvm/tools/clang/lib/StaticAnalyzer/
Author: chapuni
Date: Tue Jul 25 16:23:17 2017
New Revision: 309052
URL: http://llvm.org/viewvc/llvm-project?rev=309052&view=rev
Log:
clang/StaticAnalyzer/Core/PathSensitive/LoopUnrolling.h: Add a forward decl
AnalysisManager, to unbreak modules build.
Modified:
cfe/trunk/include/clang/Stati
rsmith accepted this revision.
rsmith added a comment.
This revision is now accepted and ready to land.
LGTM with a tweaked comment. Thanks!
Comment at: lib/CodeGen/CodeGenTBAA.cpp:88-90
+ // C++1z [intro.object]p3: If a complete object is created in storage
+ // associated w
Author: rsmith
Date: Tue Jul 25 16:31:42 2017
New Revision: 309054
URL: http://llvm.org/viewvc/llvm-project?rev=309054&view=rev
Log:
Reorder tests to match latest SD-6 draft.
Modified:
cfe/trunk/test/Lexer/cxx-features.cpp
Modified: cfe/trunk/test/Lexer/cxx-features.cpp
URL:
http://llvm.org
Author: majnemer
Date: Tue Jul 25 16:33:58 2017
New Revision: 309058
URL: http://llvm.org/viewvc/llvm-project?rev=309058&view=rev
Log:
[CodeGen] Correctly model std::byte's aliasing properties
std::byte, when defined as an enum, needs to be given special treatment
with regards to its aliasing pro
This revision was automatically updated to reflect the committed changes.
Closed by commit rL309058: [CodeGen] Correctly model std::byte's aliasing
properties (authored by majnemer).
Changed prior to commit:
https://reviews.llvm.org/D35824?vs=108166&id=108185#toc
Repository:
rL LLVM
https:/
Should this go into 5.0 ?
On Tue, Jul 25, 2017 at 4:33 PM, David Majnemer via cfe-commits <
cfe-commits@lists.llvm.org> wrote:
> Author: majnemer
> Date: Tue Jul 25 16:33:58 2017
> New Revision: 309058
>
> URL: http://llvm.org/viewvc/llvm-project?rev=309058&view=rev
> Log:
> [CodeGen] Correctly m
Author: szepet
Date: Tue Jul 25 16:49:16 2017
New Revision: 309061
URL: http://llvm.org/viewvc/llvm-project?rev=309061&view=rev
Log:
[StaticAnalyzer] LoopUnrolling - Attempt #2 to fix a crash in r309006.
Modified:
cfe/trunk/lib/StaticAnalyzer/Core/LoopUnrolling.cpp
Modified: cfe/trunk/lib/St
Author: flx
Date: Tue Jul 25 17:45:41 2017
New Revision: 309067
URL: http://llvm.org/viewvc/llvm-project?rev=309067&view=rev
Log:
[clang-tidy] Do not issue fixit for explicit template specializations
Summary:
Do not issue fixit in UnnecessaryValueParamCheck if the function is an explicit
templa
This revision was automatically updated to reflect the committed changes.
Closed by commit rL309067: [clang-tidy] Do not issue fixit for explicit
template specializations (authored by flx).
Changed prior to commit:
https://reviews.llvm.org/D35718?vs=107786&id=108193#toc
Repository:
rL LLVM
chandlerc added inline comments.
Comment at: lib/CodeGen/BackendUtil.cpp:843
PGOOptions PGOOpt;
Maybe make this an optional to avoid the big predicate below?
Comment at: lib/CodeGen/BackendUtil.cpp:847-859
if (PGOOpt.RunProfileGen)
mstorsjo added a comment.
The Microsoft version of `intrin.h` includes a header named `arm64intr.h`
(IIRC), I think it'd be better for compatibility if we also provided a file
with the same name instead of reusing this one.
https://reviews.llvm.org/D35862
___
compnerd accepted this revision.
compnerd added a comment.
This revision is now accepted and ready to land.
Please split this out into `arm64intr.h` and include it from `intrin.h` under a
`#if defined(_M_ARM64)` case before committing.
https://reviews.llvm.org/D35862
compnerd added a comment.
I think that the approach is reasonable. However, the test needs tweaking.
The function and the call both are ambiguous in the match. Can you match the
call sequence? Or explicitly the local label and then the call.
https://reviews.llvm.org/D35538
_
mgrang updated this revision to Diff 108208.
Herald added a subscriber: mgorny.
https://reviews.llvm.org/D35862
Files:
lib/Headers/CMakeLists.txt
lib/Headers/arm64intr.h
lib/Headers/intrin.h
Index: lib/Headers/intrin.h
===
--
mgrang added a comment.
Patch updated as per comments.
https://reviews.llvm.org/D35862
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
majnemer added inline comments.
Comment at: lib/CodeGen/TargetInfo.cpp:5625
+ // The Run-time ABI for the ARM Architecture section 4.1.2 requires
+ // AEABI-complying FP helper functions to use the base AAPCS
+ // These AEABI functions are expanded in the ARM llvm backend, all
Author: mgrang
Date: Tue Jul 25 22:29:40 2017
New Revision: 309081
URL: http://llvm.org/viewvc/llvm-project?rev=309081&view=rev
Log:
[clang] Add ARM64 support to armintr.h for MSVC compatibility
Summary: This fixes compiling with headers from the Windows SDK for ARM64.
Reviewers: compnerd, ruiu,
This revision was automatically updated to reflect the committed changes.
Closed by commit rL309081: [clang] Add ARM64 support to armintr.h for MSVC
compatibility (authored by mgrang).
Changed prior to commit:
https://reviews.llvm.org/D35862?vs=108208&id=108222#toc
Repository:
rL LLVM
https
teemperor added a comment.
If you can't reproduce, you should try running a debug build through valgrind.
It points out this issue:
==29522== Invalid read of size 4
==29522==at 0x16EBED0: clang::LocationContext::getCurrentStackFrame()
const (in /opt/clang/build/bin/clang-6.0)
==29522=
101 - 133 of 133 matches
Mail list logo
