tra added inline comments.
Comment at: include/clang/Sema/Sema.h:8599
@@ +8598,3 @@
+CFP_Never, // Invalid caller/callee combination.
+CFP_LastResort, // same as CFP_Never or CFP_Fallback, depending on
+// -fcuda-disable-target-call-checks option
-
tra updated this revision to Diff 33654.
tra marked 2 inline comments as done.
tra added a comment.
addressed eliben@'s comments.
http://reviews.llvm.org/D12453
Files:
include/clang/Basic/Builtins.h
include/clang/Basic/LangOptions.def
include/clang/Driver/CC1Options.td
include/clang/Sem
tra updated the summary for this revision.
tra updated this revision to Diff 33713.
tra added a comment.
Updated the patch to use bitcode linker to perform selective linking and
internalizing.
Removed Internalize+GDCE passes.
http://reviews.llvm.org/D11664
Files:
include/clang/Basic/LangOpti
tra updated the summary for this revision.
tra updated this revision to Diff 33741.
tra marked an inline comment as done.
tra added a comment.
Removed builtin-related changes(http://reviews.llvm.org/D12122). Will commit
them separately.
Added more test cases.
Addressed eliben@'s comments.
http:
tra marked 2 inline comments as done.
Comment at: lib/Sema/SemaCUDA.cpp:106
@@ +105,3 @@
+
+ // (a) Can't call global from global until we support dynamic execution.
+ if (CalleeTarget == CFT_Global &&
eliben wrote:
> Not just global from global. global from dev
tra updated this revision to Diff 34059.
tra marked 2 inline comments as done.
tra added a comment.
Added more test cases to cover constructor/destructor/new/delete.
Refactored code to remove unwanted CUDA functions from Lookup match results.
Disabled HD overloading of H or H destructors to guaran
tra added a comment.
In http://reviews.llvm.org/D9888#257904, @sfantao wrote:
> This diff refactors the original patch and is rebased on top of the latests
> offloading changes inserted for CUDA.
>
> Here I don't touch the CUDA support. I tried, however, to have the
> implementation modular eno
Filipe,
FYI, this change appears to introduce a somewhat subtle problem. clang
compiled with itself starts producing (false positive?) warnings about
uninitialized variables. I didn't get a chance to dig deeper yet.
--Artem
On Fri, Oct 16, 2015 at 8:07 AM, Filipe Cabecinhas via cfe-commits <
cfe
> On Fri, Oct 16, 2015 at 11:52 PM, Artem Belevich via cfe-commits <
> cfe-commits@lists.llvm.org> wrote:
>
>> Filipe,
>>
>> FYI, this change appears to introduce a somewhat subtle problem. clang
>> compiled with itself starts producing (false positive?) warni
tra updated this revision to Diff 37912.
tra added a comment.
Herald added a subscriber: klimek.
Changed header wrapping strategy. Previous version was attempting to
make CUDA headers work for host/device compilations separately. In the
end host and device compilations ended up with different view
tra created this revision.
tra added reviewers: echristo, pcc.
tra added a subscriber: cfe-commits.
This will be used for CUDA where we need to link with vendor-supplied bitcode
library and with a bitcode library pretending to be libm for GPUs.
Linking options for particular file depend on the
tra added a comment.
In http://reviews.llvm.org/D13171#272397, @echristo wrote:
> I'm ignoring the content of the header, but this seems to be a not terrible
> way to do things. I gather that cuda_runtime.h is something that's typically
> included by the driver by nvidia and not the client?
C
tra updated this revision to Diff 38046.
tra added a comment.
Added a comment about format of LinkModules vector.
Added/cleaned up few comments in test files.
http://reviews.llvm.org/D13913
Files:
include/clang/Basic/LangOptions.def
include/clang/CodeGen/CodeGenAction.h
include/clang/Driv
tra marked an inline comment as done.
tra added a comment.
http://reviews.llvm.org/D13913
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
tra created this revision.
tra added reviewers: eliben, jingyue, jpienaar, echristo.
tra added a subscriber: cfe-commits.
Currently -fcuda-disable-target-call-checks option enables parsing of code that
calls across host/device boundary.
However, we don't emit any IR for functions that don't have
tra updated this revision to Diff 38269.
tra added a comment.
Instead of passing AuxTriple around as an argument, store ToolChain info in
Compilation and retrieve it from there.
http://reviews.llvm.org/D13144
Files:
include/clang/Driver/Compilation.h
include/clang/Driver/Driver.h
lib/Dri
Author: tra
Date: Tue Oct 27 12:56:59 2015
New Revision: 251427
URL: http://llvm.org/viewvc/llvm-project?rev=251427&view=rev
Log:
Allow linking multiple bitcode files.
Linking options for particular file depend on the option that specifies the
file.
Currently there are two:
* -mlink-bitcode-fil
This revision was automatically updated to reflect the committed changes.
Closed by commit rL251427: Allow linking multiple bitcode files. (authored by
tra).
Changed prior to commit:
http://reviews.llvm.org/D13913?vs=38046&id=38562#toc
Repository:
rL LLVM
http://reviews.llvm.org/D13913
Fil
tra updated this revision to Diff 38574.
tra added a comment.
Added test cases for force-including of cuda_runtime.h
Tweaked inclusion of one header due to use of default arguments.
http://reviews.llvm.org/D13171
Files:
lib/Driver/ToolChains.cpp
lib/Headers/CMakeLists.txt
lib/Headers/cuda
tra updated this revision to Diff 38576.
tra added a comment.
Updated to reflect latest changes in http://reviews.llvm.org/D13913.
http://reviews.llvm.org/D13170
Files:
include/clang/Driver/Options.td
include/clang/Driver/ToolChain.h
lib/Driver/ToolChain.cpp
lib/Driver/ToolChains.cpp
tra added inline comments.
Comment at: lib/Driver/Driver.cpp:503
@@ -502,3 +502,3 @@
if (TC.getTriple().isOSBinFormatMachO())
-BuildUniversalActions(C->getDefaultToolChain(), C->getArgs(), Inputs,
+BuildUniversalActions(*C, C->getDefaultToolChain(), C->getArgs(), Inputs
tra updated this revision to Diff 38694.
tra added a comment.
Folded some arguments of BuildUniversalActions.
http://reviews.llvm.org/D13144
Files:
include/clang/Driver/Compilation.h
include/clang/Driver/Driver.h
lib/Driver/Compilation.cpp
lib/Driver/Driver.cpp
lib/Driver/Tools.cpp
tra updated this revision to Diff 39017.
tra added a comment.
Moved CUDA toolchain selection to Compilation.
Removed DeviceTriple info from CudaHostActions and CudaDeviceActions.
Removed few now-unnecessary parameters and code that used them.
http://reviews.llvm.org/D13144
Files:
include/clan
tra updated this revision to Diff 39019.
tra added a comment.
Moved CUDA toolchain selection to Driver::BuildCompilation() where Compilation
is set up.
http://reviews.llvm.org/D13144
Files:
include/clang/Driver/Action.h
include/clang/Driver/Compilation.h
include/clang/Driver/Driver.h
l
tra marked an inline comment as done.
tra added a comment.
http://reviews.llvm.org/D13144
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
tra created this revision.
tra added a reviewer: echristo.
tra added a subscriber: cfe-commits.
This is libdevice-related part extracted from D13170
- added detection of libdevice bitcode file and API to find one appropriate
for the GPU we're compiling for.
- pass additional cc1 options
tra updated the summary for this revision.
tra updated this revision to Diff 39860.
tra added a comment.
Moved libdevice detection and linking into a separate patch
http://reviews.llvm.org/D14556.
http://reviews.llvm.org/D13170
Files:
include/clang/Driver/Options.td
include/clang/Driver/To
tra added a comment.
Ping.
http://reviews.llvm.org/D11666
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
tra accepted this revision.
tra added a comment.
This revision is now accepted and ready to land.
LGTM.
https://reviews.llvm.org/D22490
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commit
tra accepted this revision.
tra added a comment.
This revision is now accepted and ready to land.
Looks good.
https://reviews.llvm.org/D22518
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-
tra added a comment.
Looks good.
https://reviews.llvm.org/D22518
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
tra added a comment.
Samuel, the patch breaks CUDA. With the patch clang no longer adds include
paths to standard c++ library during device-side compilation.
if you run "clang++ -### -c -x cuda /dev/null" you will see that host side gets
`"-internal-isystem"
"/usr/lib/gcc/x86_64-linux-gnu/4.8/.
tra created this revision.
tra added a reviewer: jlebar.
tra added a subscriber: cfe-commits.
This makes clang's libdevice selection match that of NVCC as described in
http://docs.nvidia.com/cuda/libdevice-users-guide/basic-usage.html#version-selection
If required libdevice variant is not found,
tra added inline comments.
Comment at: lib/Driver/ToolChains.cpp:4773
@@ +4772,3 @@
+ StringRef GpuArch =
+ DriverArgs.getLastArgValue(options::OPT_march_EQ, "sm_20");
+ std::string LibDeviceFile = CudaInstallation.getLibDeviceFile(GpuArch);
jlebar wrote:
>
tra updated this revision to Diff 66392.
tra added a comment.
Added tests for libdevice mapping correctness.
https://reviews.llvm.org/D23037
Files:
include/clang/Basic/DiagnosticDriverKinds.td
lib/Driver/ToolChains.cpp
test/Driver/Inputs/CUDA_80/usr/local/cuda/nvvm/libdevice/libdevice.co
tra created this revision.
tra added a reviewer: jlebar.
tra added a subscriber: cfe-commits.
It does not make sense and violates enough assumptions in the code to crash
compiler.
https://reviews.llvm.org/D23042
Files:
include/clang/Basic/DiagnosticDriverKinds.td
lib/Driver/ToolChains.cpp
tra added inline comments.
Comment at: lib/Driver/ToolChains.cpp:4834
@@ -4831,2 +4833,3 @@
+getDriver().Diag(diag::err_drv_cuda_nvptx_host);
}
return DAL;
jlebar wrote:
> IRL we talked about putting an assert() here and bailing out earlier. Does
> tha
tra updated this revision to Diff 66505.
tra added a comment.
Herald added a subscriber: klimek.
Abort pipeline constructions early if we detect that NVPTX is used for host
compilation.
Restore assertions for presence of -march flag.
https://reviews.llvm.org/D23042
Files:
include/clang/Basic
tra added a comment.
In https://reviews.llvm.org/D23042#503869, @jlebar wrote:
> > Restore assertions for presence of -march flag.
>
>
> We don't need an explicit assertion in TranslateArgs?
Nope. The action we create for fatbin uses CudaToolChain, but has nullptr
BoundArch and there's no way
tra updated the summary for this revision.
tra updated this revision to Diff 66560.
tra added a comment.
Added a comment describing why we delibrartly error out on use of NVPTX for
host compilation.
https://reviews.llvm.org/D23042
Files:
include/clang/Basic/DiagnosticDriverKinds.td
lib/Dri
tra marked an inline comment as done.
tra added a comment.
https://reviews.llvm.org/D23042
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Author: tra
Date: Tue Aug 2 17:37:47 2016
New Revision: 277537
URL: http://llvm.org/viewvc/llvm-project?rev=277537&view=rev
Log:
[CUDA] Do not allow using NVPTX target for host compilation.
Differential Revision: https://reviews.llvm.org/D23042
Modified:
cfe/trunk/include/clang/Basic/Diagno
This revision was automatically updated to reflect the committed changes.
Closed by commit rL277537: [CUDA] Do not allow using NVPTX target for host
compilation. (authored by tra).
Changed prior to commit:
https://reviews.llvm.org/D23042?vs=66560&id=66578#toc
Repository:
rL LLVM
https://rev
tra updated this revision to Diff 66579.
tra added a comment.
Now that https://reviews.llvm.org/D23042 / r277537 makes sure that
CudaToolchain is only used on device side, we can
remove defaults for -march and restore asserts() ensuring that -march is added
by the driver.
https://reviews.llvm
tra marked an inline comment as done.
tra added a comment.
https://reviews.llvm.org/D23037
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Author: tra
Date: Tue Aug 2 18:12:51 2016
New Revision: 277542
URL: http://llvm.org/viewvc/llvm-project?rev=277542&view=rev
Log:
[CUDA] Fix libdevice selection.
This makes clang's libdevice selection match that of NVCC as described in
http://docs.nvidia.com/cuda/libdevice-users-guide/basic-usage
This revision was automatically updated to reflect the committed changes.
Closed by commit rL277542: [CUDA] Fix libdevice selection. (authored by tra).
Changed prior to commit:
https://reviews.llvm.org/D23037?vs=66579&id=66589#toc
Repository:
rL LLVM
https://reviews.llvm.org/D23037
Files:
> On Tue, Aug 2, 2016 at 4:12 PM, Artem Belevich via cfe-commits
> wrote:
> > Author: tra
> > Date: Tue Aug 2 18:12:51 2016
> > New Revision: 277542
> >
> > URL: http://llvm.org/viewvc/llvm-project?rev=277542&view=rev
> > Log:
> > [CUDA] Fix libde
Author: tra
Date: Tue Aug 2 18:43:04 2016
New Revision: 277552
URL: http://llvm.org/viewvc/llvm-project?rev=277552&view=rev
Log:
[CUDA] Updated CUDA tests that must run w/o CUDA installation.
Fixes test failures after r277542 on systems that don't have
CUDA installed.
Modified:
cfe/trunk/te
;> This broke
>> http://lab.llvm.org:8080/green/job/clang-stage1-cmake-RA-incremental_check/26856
>> ,
>> can you take a look?
>>
>> Thanks,
>>
>> On Tue, Aug 2, 2016 at 4:12 PM, Artem Belevich via cfe-commits
>> wrote:
>> > Author: tr
tra accepted this revision.
tra added a comment.
This revision is now accepted and ready to land.
LGTM.
https://reviews.llvm.org/D23238
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commit
tra added a comment.
I think we need to add `noexcept` for these in c++11.
https://reviews.llvm.org/D23239
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
tra accepted this revision.
tra added a comment.
This revision is now accepted and ready to land.
LGTM.
https://reviews.llvm.org/D23240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commit
tra added inline comments.
Comment at: clang/lib/CodeGen/CodeGenModule.cpp:2886
@@ +2885,3 @@
+ // Check if this function has diagnostics that should be emitted when we
+ // codegen it. If so, don't eit this function definition, but don't emit the
+ // diags just yet. Emittin
tra added a comment.
cuda-detect.cu may be a better place for this test.
Comment at: test/Driver/cuda-simple.cu:27
@@ -16,1 +26,3 @@
+// CHECK-CXXINCLUDE: clang{{.*}} "-cc1" "-triple" "x86_64--linux-gnu"
{{.*}}"-internal-isystem" "{{.+}}/include/c++/4.8"
+// CHECK-CXXINCLUDE: l
tra accepted this revision.
tra added a comment.
This revision is now accepted and ready to land.
Few nits, but looks good otherwise.
Should be add few tests for calling device functions from host-side global
initializers? Perhaps for device->host, too, as there may be unexpected
interplay with
tra accepted this revision.
tra added a comment.
This revision is now accepted and ready to land.
LGTM.
https://reviews.llvm.org/D23239
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commit
tra added inline comments.
Comment at: test/Driver/cuda-detect.cu:75-77
@@ -74,1 +74,5 @@
+// Verify that compiler accepts CUDA syntax with "-x cuda-cpp-output".
+// RUN: %clang -Werror -fsyntax-only -x cuda-cpp-output -c %s
+//
+// Verify that C++ include paths are passed for b
tra accepted this revision.
tra added a comment.
This revision is now accepted and ready to land.
LGTM.
https://reviews.llvm.org/D22946
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commit
tra accepted this revision.
tra added a comment.
This revision is now accepted and ready to land.
LGTM.
Comment at: clang/test/SemaCUDA/global-initializers-host.cu:2
@@ +1,3 @@
+// RUN: %clang_cc1 %s --std=c++11 -triple x86_64-linux-unknown -emit-llvm -o -
-verify
+
+#include "
tra created this revision.
tra added a reviewer: jlebar.
tra added a subscriber: cfe-commits.
This matches the way nvcc encapsulates GPU binaries into host object file.
Now cuobjdump can deal with clang-compiled object files.
https://reviews.llvm.org/D23429
Files:
lib/CodeGen/CGCUDANV.cpp
tra accepted this revision.
tra added a comment.
This revision is now accepted and ready to land.
OK. This should help with PR26966.
https://reviews.llvm.org/D23341
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bi
tra updated this revision to Diff 67851.
tra added a comment.
Reverted argument type to std::string
https://reviews.llvm.org/D23429
Files:
lib/CodeGen/CGCUDANV.cpp
test/CodeGenCUDA/device-stub.cu
Index: test/CodeGenCUDA/device-stub.cu
===
tra marked an inline comment as done.
Comment at: lib/CodeGen/CGCUDANV.cpp:62-69
@@ -60,3 +61,10 @@
llvm::ConstantInt::get(SizeTy, 0)};
auto ConstStr = CGM.GetAddrOfConstantCString(Str, Name.c_str());
+llvm::GlobalVariable *GV =
+ca
This revision was automatically updated to reflect the committed changes.
tra marked an inline comment as done.
Closed by commit rL278549: [CUDA] Place GPU binary into .nv_fatbin section and
align it by 8. (authored by tra).
Changed prior to commit:
https://reviews.llvm.org/D23429?vs=67851&id=6
Author: tra
Date: Fri Aug 12 13:44:01 2016
New Revision: 278549
URL: http://llvm.org/viewvc/llvm-project?rev=278549&view=rev
Log:
[CUDA] Place GPU binary into .nv_fatbin section and align it by 8.
This matches the way nvcc encapsulates GPU binaries into host object file.
Now cuobjdump can deal wi
Richard,
On Fri, Jan 15, 2016 at 5:32 PM, Richard Smith
wrote:
> On Fri, Jan 15, 2016 at 5:29 PM, Richard Smith
> wrote:
> > On Fri, Jan 15, 2016 at 4:22 PM, Artem Belevich wrote:
> >> tra added inline comments.
> >>
> >>
> >> Comment at: lib/CodeGen/CodeGenModule.cpp:2334
> >
tra updated this revision to Diff 46696.
tra marked 8 inline comments as done.
tra added a comment.
Addressed Richard's comments.
Relaxed restrictions a bit to allow constant initializers even those CUDA would
not considered to be empty.
Updated test case accordingly.
http://reviews.llvm.org/D1
Author: tra
Date: Tue Feb 2 16:29:48 2016
New Revision: 259592
URL: http://llvm.org/viewvc/llvm-project?rev=259592&view=rev
Log:
[CUDA] Do not allow dynamic initialization of global device side variables.
In general CUDA does not allow dynamic initialization of
global device-side variables. One
This revision was automatically updated to reflect the committed changes.
Closed by commit rL259592: [CUDA] Do not allow dynamic initialization of global
device side variables. (authored by tra).
Changed prior to commit:
http://reviews.llvm.org/D15305?vs=46696&id=46707#toc
Repository:
rL LLV
tra added inline comments.
Comment at: lib/Sema/SemaCUDA.cpp:429-430
@@ +428,4 @@
+ CXXConstructorDecl *CD) {
+ if (!CD->isDefined() && CD->isTemplateInstantiation())
+InstantiateFunctionDefinition(VarLoc, CD->getFirstDecl());
+
--
tra updated this revision to Diff 46729.
tra added a comment.
Added few more device-side system calls and related wrapper functions.
Added nothrow attributes on malloc/free.
http://reviews.llvm.org/D16638
Files:
lib/Headers/__clang_cuda_runtime_wrapper.h
Index: lib/Headers/__clang_cuda_runti
tra retitled this revision from "[CUDA] Added device-side std::{malloc/free}"
to "[CUDA] Added device-side system call decls and related wrappers.".
tra updated the summary for this revision.
tra updated this revision to Diff 46803.
tra marked 3 inline comments as done.
tra added a comment.
Addre
tra added inline comments.
Comment at: lib/Headers/__clang_cuda_runtime_wrapper.h:232
@@ +231,3 @@
+// Clang will convert printf into vprintf, but we still need
+// device-side declaration for it.
+__device__ int printf(const char *, ...);
jlebar wrote:
> I'd pref
tra updated this revision to Diff 46818.
tra added a comment.
Updated comment.
http://reviews.llvm.org/D16638
Files:
lib/Headers/__clang_cuda_runtime_wrapper.h
Index: lib/Headers/__clang_cuda_runtime_wrapper.h
===
--- lib/Header
Author: tra
Date: Wed Feb 3 14:53:58 2016
New Revision: 259690
URL: http://llvm.org/viewvc/llvm-project?rev=259690&view=rev
Log:
[CUDA] added declarations for device-side system calls
...and std:: wrappers for free/malloc.
Modified:
cfe/trunk/lib/Headers/__clang_cuda_runtime_wrapper.h
Modi
tra closed this revision.
tra added a comment.
Committed in r259690
http://reviews.llvm.org/D16638
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
tra created this revision.
tra added reviewers: jlebar, jingyue, jpienaar, eliben.
tra added a subscriber: cfe-commits.
This is an artefact of split-mode CUDA compilation that we need to
mimic. HD functions are sometimes allowed to call H or D functions. Due
to split compilation mode device-side c
tra added a comment.
When overload set contains h and HD functions that are otherwise equal for
overload resolution, you want to be able to tell which one is better.
http://reviews.llvm.org/D16870
___
cfe-commits mailing list
cfe-commits@lists.llvm.
tra updated this revision to Diff 46927.
tra marked an inline comment as done.
tra added a comment.
Addressed Jingyue's comments.
Fixed function-overload.cu tests to reflect stricter call target checks.
http://reviews.llvm.org/D16870
Files:
include/clang/Sema/Sema.h
lib/Sema/SemaCUDA.cpp
tra marked an inline comment as done.
Comment at: lib/Sema/SemaCUDA.cpp:132-141
@@ -131,12 +131,12 @@
// (d) HostDevice behavior depends on compilation mode.
if (CallerTarget == CFT_HostDevice) {
// Calling a function that matches compilation mode is OK.
// Calling
tra created this revision.
tra added reviewers: jingyue, jlebar.
tra added a subscriber: cfe-commits.
... and pull them into std namespace with using-declaration instead.
http://reviews.llvm.org/D16932
Files:
lib/Headers/__clang_cuda_cmath.h
Index: lib/Headers/__clang_cuda_cmath.h
==
tra added a comment.
I'm not sure what we could test here without CUDA headers.
I've tested out-of-tree by compiling thrust unit tests and the test case in PR.
http://reviews.llvm.org/D16932
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
ht
This revision was automatically updated to reflect the committed changes.
Closed by commit rL259944: [CUDA] Bug 26497 : Remove wrappers for variants
provided by CUDA headers. (authored by tra).
Changed prior to commit:
http://reviews.llvm.org/D16932?vs=47040&id=47058#toc
Repository:
rL LLVM
Author: tra
Date: Fri Feb 5 16:54:05 2016
New Revision: 259944
URL: http://llvm.org/viewvc/llvm-project?rev=259944&view=rev
Log:
[CUDA] Bug 26497 : Remove wrappers for variants provided by CUDA headers.
... and pull global-scope ones into std namespace with using-declaration.
Differential Revis
tra updated the summary for this revision.
tra updated this revision to Diff 47335.
tra marked 3 inline comments as done.
tra added a comment.
Updated the way WrongSide functions are removed from consideration during
overload resolution.
Previous version could provide inconsistent results dependi
tra planned changes to this revision.
tra added a comment.
Previously accepted version was dependent on order of functions in overload set.
In order to make ordering consistent, WrongSide functions are now removed from
the
set before pair-wise comparison of candidates. Please take a look.
tra added inline comments.
Comment at: lib/CodeGen/CodeGenModule.cpp:1880
@@ +1879,3 @@
+// Conservatively, mark all functions in CUDA as convergent (meaning, they
+// may call an intrinsicly convergent op, such as __syncthreads(), and so
+// can't have certain optimiz
tra added a comment.
Erasing an argument would only complicate the problem.
I guess for consistency we need to match clang's behavior for regular C++ code.
For optimized builds it just seems to pass NULL pointer instead.
http://reviews.llvm.org/D17103
_
tra created this revision.
tra added reviewers: jlebar, echristo.
tra added a subscriber: cfe-commits.
ptxas optimizations are disabled if we need to generate debug info
as ptxas does not accept '-g' otherwise.
http://reviews.llvm.org/D17111
Files:
lib/Driver/Tools.cpp
test/Driver/cuda-exter
tra added inline comments.
Comment at: lib/Driver/Tools.cpp:10707
@@ +10706,3 @@
+// ptxas does not accept -g option if optimization is enabled, so we ignore
+// compiler's -O* options if we want debug info.
+CmdArgs.push_back("-g");
hfinkel wrote:
> e
tra added inline comments.
Comment at: lib/Driver/Tools.cpp:10707
@@ +10706,3 @@
+// ptxas does not accept -g option if optimization is enabled, so we ignore
+// compiler's -O* options if we want debug info.
+CmdArgs.push_back("-g");
tra wrote:
> hfink
tra added inline comments.
Comment at: lib/Driver/Tools.cpp:10707
@@ +10706,3 @@
+// ptxas does not accept -g option if optimization is enabled, so we ignore
+// compiler's -O* options if we want debug info.
+CmdArgs.push_back("-g");
hfinkel wrote:
> e
tra retitled this revision from "[CUDA] pass debug options to ptxas." to
"[CUDA] Added --cuda-noopt-device-debug option to control ptxas' debug info
generation.".
tra updated the summary for this revision.
tra updated this revision to Diff 47680.
tra added a comment.
Added --cuda-noopt-device-de
tra requested a review of this revision.
tra added a comment.
This revision is now accepted and ready to land.
@jingyue, @jlebar: can you take a look at the updated version?
http://reviews.llvm.org/D16870
___
cfe-commits mailing list
cfe-commits@lis
tra updated this revision to Diff 47753.
tra marked 14 inline comments as done.
tra added a comment.
Addressed @jlebar's comments.
http://reviews.llvm.org/D16870
Files:
include/clang/Sema/Sema.h
lib/Sema/SemaCUDA.cpp
lib/Sema/SemaOverload.cpp
test/CodeGenCUDA/function-overload.cu
test
Author: tra
Date: Fri Feb 12 12:29:18 2016
New Revision: 260697
URL: http://llvm.org/viewvc/llvm-project?rev=260697&view=rev
Log:
[CUDA] Tweak attribute-based overload resolution to match nvcc behavior.
This is an artefact of split-mode CUDA compilation that we need to
mimic. HD functions are som
This revision was automatically updated to reflect the committed changes.
Closed by commit rL260697: [CUDA] Tweak attribute-based overload resolution to
match nvcc behavior. (authored by tra).
Changed prior to commit:
http://reviews.llvm.org/D16870?vs=47753&id=47819#toc
Repository:
rL LLVM
Author: tra
Date: Fri Feb 12 14:26:43 2016
New Revision: 260719
URL: http://llvm.org/viewvc/llvm-project?rev=260719&view=rev
Log:
Added missing '__'.
Modified:
cfe/trunk/lib/Headers/__clang_cuda_cmath.h
Modified: cfe/trunk/lib/Headers/__clang_cuda_cmath.h
URL:
http://llvm.org/viewvc/llvm-pr
i, Feb 12, 2016, 12:31 PM Artem Belevich via cfe-commits <
> cfe-commits@lists.llvm.org> wrote:
>
>> Author: tra
>> Date: Fri Feb 12 14:26:43 2016
>> New Revision: 260719
>>
>> URL: http://llvm.org/viewvc/llvm-project?rev=260719&view=rev
>>
501 - 600 of 1193 matches
Mail list logo
