Page MenuHomePhabricator

hliao (Michael Liao)
User

Projects

User does not belong to any projects.

User Details

User Since
Aug 7 2014, 12:01 PM (295 w, 4 d)

Recent Activity

Today

hliao committed rGc97be2c37785: [hip] Remove `hip_pinned_shadow`. (authored by hliao).
[hip] Remove `hip_pinned_shadow`.
Tue, Apr 7, 7:02 AM
hliao closed D77583: [hip] Remove `hip_pinned_shadow`..
Tue, Apr 7, 7:02 AM · Restricted Project

Yesterday

hliao added a comment to D77583: [hip] Remove `hip_pinned_shadow`..

Is the runtime and HIP directed test change in place?

Mon, Apr 6, 7:06 PM · Restricted Project
hliao created D77583: [hip] Remove `hip_pinned_shadow`..
Mon, Apr 6, 1:37 PM · Restricted Project

Sun, Apr 5

hliao updated the diff for D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

Rebase to the latest trunk.

Sun, Apr 5, 10:25 PM · Restricted Project

Fri, Apr 3

hliao committed rGb952d799cacd: [cuda][hip] Fix `RegisterVar` function prototype. (authored by hliao).
[cuda][hip] Fix `RegisterVar` function prototype.
Fri, Apr 3, 10:16 AM
hliao closed D77398: [cuda][hip] Fix `RegisterVar` function prototype..
Fri, Apr 3, 10:16 AM · Restricted Project
hliao created D77398: [cuda][hip] Fix `RegisterVar` function prototype..
Fri, Apr 3, 7:29 AM · Restricted Project

Thu, Apr 2

hliao added a comment to D77245: [AMDGPU] Fix crash in SILoadStoreOptimizer.

My fix has been hanging around for a little while now - so my memory of it is a little hazy.

However, the issue I saw (and attempted to fix) was similar to yours. Agreed that I think this is a rare occurrence, so perhaps your approach makes more sense (and is simpler/cheaper).
One question - the reason I went with the solution I did was that a previous iteration of the loop doing the optimisation was invalidating the iterator stored in a later list - which caused a crash - is your fix working around this by avoiding creating those potentially dependent lists in the first place?? (I wasn't 100% sure about the logic).

I guess one potential advantage with my approach is that you still get the potential optimisations at the expense of having to re-compute the lists for dependent instructions.

Thu, Apr 2, 8:38 AM · Restricted Project
hliao added a comment to D59321: AMDGPU: Teach toolchain to link rocm device libs.

Do we have a better way to avoid adding those empty bitcode files?

No, we need the files to exist for tests. This is what existing bitcode link tests do

Thu, Apr 2, 7:33 AM

Wed, Apr 1

hliao added a comment to D77238: [CUDA][NFC] Split math.h functions out of __clang_cuda_device_functions.h.

this patch breaks the test clang/test/Headers/nvptx_device_cmath_functions.cpp, could you please fix it ASAP?

Wed, Apr 1, 10:18 PM · Restricted Project
hliao added a comment to D59321: AMDGPU: Teach toolchain to link rocm device libs.

Do we have a better way to avoid adding those empty bitcode files?

Wed, Apr 1, 11:49 AM

Tue, Mar 31

hliao added a comment to D77154: [cmake] Add LLVM_USE_GDB_INDEX option for gold/lld linkers.

This seems fine. I have -DCMAKE_EXE_LINKER_FLAGS=-Wl,--gdb-index -DCMAKE_SHARED_LINKER_FLAGS=-Wl,--gdb-index -DLLVM_USE_SPLIT_DWARF=On in my long cmake command line. I am just wondering whether this new option will be discoverable.

Some people don't use -DLLVM_USE_LINKER=lld, but use -DLLVM_ENABLE_LLD=On.

Tue, Mar 31, 2:52 PM · Restricted Project
hliao added a comment to D77154: [cmake] Add LLVM_USE_GDB_INDEX option for gold/lld linkers.

The same patch posted previously, https://reviews.llvm.org/D58628

Tue, Mar 31, 2:52 PM · Restricted Project

Sat, Mar 28

hliao committed rGcb6389360b05: Fix GCC warning on enum class bitfield. NFC. (authored by hliao).
Fix GCC warning on enum class bitfield. NFC.
Sat, Mar 28, 7:32 AM
hliao committed rGd2dd0fac483c: Fix `-Wsign-compare` warning. NFC. (authored by hliao).
Fix `-Wsign-compare` warning. NFC.
Sat, Mar 28, 7:32 AM

Fri, Mar 27

hliao added a comment to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
In D76365#1947103, @tra wrote:

The new revision is accepted, right? Just want to confirm as it seems you accept it before I posted the new change.

The approval was for the old version. I didn't undo it when I reopened the review. The diff looks OK, though the last variant still leaves open the question of what's the meaning of these attributes and what are the restrictions on their use.

So what's the reasonable thing to do if I write something like this:

__attribute__((device_builtin_surface_type)) int foo; // Ignore? Warn? Error? Do something sensible?

For such case, NVCC reports the following error:

kernel.cu(3): error: attribute "device_builtin_surface_type" does not apply here

1 error detected in the compilation of "kernel.cpp1.ii"

That error is generated after nvcc --keep -g -c kernel.cu from this sample code (kernel.cu)

#include <cuda.h>

__attribute__((device_builtin_surface_type)) int foo;

int f() {
  return foo;
}

I changed that sample code a little bit to this one

#include <cuda.h>

#if 1
typedef __attribute__((device_builtin_surface_type)) int dev_texsurf_int_t;
dev_texsurf_int_t foo;
#else
__attribute__((device_builtin_surface_type)) int foo;
#endif

int f() {
  return foo;
}

It triggers a crash in NVCC with the same compilation command line.

We may enhance clang to report an error instead of a warning only so far.

Fri, Mar 27, 9:28 PM · Restricted Project
hliao added a comment to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
In D76365#1947103, @tra wrote:

The new revision is accepted, right? Just want to confirm as it seems you accept it before I posted the new change.

The approval was for the old version. I didn't undo it when I reopened the review. The diff looks OK, though the last variant still leaves open the question of what's the meaning of these attributes and what are the restrictions on their use.

So what's the reasonable thing to do if I write something like this:

__attribute__((device_builtin_surface_type)) int foo; // Ignore? Warn? Error? Do something sensible?
Fri, Mar 27, 8:55 PM · Restricted Project
hliao added a comment to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
In D76365#1947103, @tra wrote:

The new revision is accepted, right? Just want to confirm as it seems you accept it before I posted the new change.

The approval was for the old version. I didn't undo it when I reopened the review. The diff looks OK, though the last variant still leaves open the question of what's the meaning of these attributes and what are the restrictions on their use.

So what's the reasonable thing to do if I write something like this:

__attribute__((device_builtin_surface_type)) int foo; // Ignore? Warn? Error? Do something sensible?
Fri, Mar 27, 5:05 PM · Restricted Project
hliao committed rG5be9b8cbe2b2: [cuda][hip] Add CUDA builtin surface/texture reference support. (authored by hliao).
[cuda][hip] Add CUDA builtin surface/texture reference support.
Fri, Mar 27, 2:21 PM
hliao closed D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
Fri, Mar 27, 2:20 PM · Restricted Project
hliao added a comment to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..

Fix Sema checks on partial template specialization.

  • Revise Sema checks on the template class.
Fri, Mar 27, 2:17 PM · Restricted Project
hliao abandoned D76948: [cuda][hip] Add CUDA builtin surface/texture reference support..
Fri, Mar 27, 2:17 PM · Restricted Project
hliao updated the diff for D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..

Fix Sema checks on partial template specialization.

Fri, Mar 27, 2:17 PM · Restricted Project
hliao added inline comments to D76948: [cuda][hip] Add CUDA builtin surface/texture reference support..
Fri, Mar 27, 2:16 PM · Restricted Project
hliao added a comment to D76948: [cuda][hip] Add CUDA builtin surface/texture reference support..
In D76948#1946861, @tra wrote:

Would it be possible to update the old review with the new diff? It would make it easier to see the incremental changes you've made. If the old review can be reopened that would be great as it would keep all relevant info in one place, but I'm fine doing the review here, too, if phabricator does not let you do it.

Fri, Mar 27, 1:43 PM · Restricted Project
hliao added a comment to D76948: [cuda][hip] Add CUDA builtin surface/texture reference support..
In D76948#1946861, @tra wrote:

Would it be possible to update the old review with the new diff? It would make it easier to see the incremental changes you've made. If the old review can be reopened that would be great as it would keep all relevant info in one place, but I'm fine doing the review here, too, if phabricator does not let you do it.

Fri, Mar 27, 1:43 PM · Restricted Project
hliao added a comment to D76948: [cuda][hip] Add CUDA builtin surface/texture reference support..

This's revised change from https://reviews.llvm.org/D76365 after fixing Sema checks on the template partial specialization. With this change, I could compile the following sample code using surface reference.

Fri, Mar 27, 1:10 PM · Restricted Project
hliao created D76948: [cuda][hip] Add CUDA builtin surface/texture reference support..
Fri, Mar 27, 1:10 PM · Restricted Project
hliao added a comment to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
In D76365#1946407, @tra wrote:
In D76365#1946345, @tra wrote:

Looks like the change breaks compilation for us:

In file included from <built-in>:1:
In file included from llvm_unstable/toolchain/lib/clang/google3-trunk/include/__clang_cuda_runtime_wrapper.h:104:
In file included from cuda/include/cuda_runtime.h:116: cuda/include/cuda_surface_types.h:91:42: error: illegal device builtin surface reference type 'surface<void, dim>' declared here
struct  __device_builtin_surface_type__  surface<void, dim> : public surfaceReference
                                         ^
cuda/include/cuda_surface_types.h:91:42: note: 'surface<void, dim>' needs to be instantiated from a class template with the 2nd template argument as an integral value
1 error generated when compiling for sm_60.

I'm investigating, but we may need to roll back this patch. Stay tuned.

It appears that the assumptions of what types the attributes can apply to are not valid. In CUDA headers they are also used on non-templated classes/structs. E.g in cuda/include/cuda_surface_types.h:74

struct __attribute__((device_builtin_surface_type)) surface : public surfaceReference
{
...
};

I'll undo this patch until we can make it work.

Fri, Mar 27, 10:16 AM · Restricted Project
hliao added a comment to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
In D76365#1946345, @tra wrote:

Looks like the change breaks compilation for us:

In file included from <built-in>:1:
In file included from llvm_unstable/toolchain/lib/clang/google3-trunk/include/__clang_cuda_runtime_wrapper.h:104:
In file included from cuda/include/cuda_runtime.h:116: cuda/include/cuda_surface_types.h:91:42: error: illegal device builtin surface reference type 'surface<void, dim>' declared here
struct  __device_builtin_surface_type__  surface<void, dim> : public surfaceReference
                                         ^
cuda/include/cuda_surface_types.h:91:42: note: 'surface<void, dim>' needs to be instantiated from a class template with the 2nd template argument as an integral value
1 error generated when compiling for sm_60.

I'm investigating, but we may need to roll back this patch. Stay tuned.

Fri, Mar 27, 10:16 AM · Restricted Project

Thu, Mar 26

hliao committed rG6a9ad5f3f4ac: [cuda][hip] Add CUDA builtin surface/texture reference support. (authored by hliao).
[cuda][hip] Add CUDA builtin surface/texture reference support.
Thu, Mar 26, 11:56 AM
hliao closed D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
Thu, Mar 26, 11:55 AM · Restricted Project
hliao added a comment to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
In D76365#1944272, @tra wrote:

LGTM. Next step is to figure out what various __nv_tex_surf_handler(<string>...) maps to for various strings (there are ~110 of them in CUDA-10.2) and implement its replacement. I think we should be able to do it in the wrapper file.

Thu, Mar 26, 11:55 AM · Restricted Project
hliao updated the diff for D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..

Rebase to the master code

Thu, Mar 26, 7:00 AM · Restricted Project
hliao updated the diff for D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..

Fix windows build and revise header change.

Thu, Mar 26, 12:30 AM · Restricted Project
hliao added inline comments to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
Thu, Mar 26, 12:30 AM · Restricted Project

Wed, Mar 25

hliao committed rGd264f02c6f50: Fix `-Wreturn-type` warning. NFC. (authored by hliao).
Fix `-Wreturn-type` warning. NFC.
Wed, Mar 25, 10:12 PM
hliao added inline comments to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
Wed, Mar 25, 8:04 AM · Restricted Project

Tue, Mar 24

hliao added inline comments to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
Tue, Mar 24, 7:09 PM · Restricted Project
hliao updated the diff for D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..

Add Sema checks on CUDA device builtin surface/texture attributes.

Tue, Mar 24, 6:36 PM · Restricted Project
hliao committed rG4f4e68799fd5: [test][clang][driver] Add required features. (authored by hliao).
[test][clang][driver] Add required features.
Tue, Mar 24, 2:34 PM

Fri, Mar 20

hliao updated the diff for D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..

Minor revising following reviewer's comment. Work on Sema checks and upload another review.

Fri, Mar 20, 3:12 PM · Restricted Project
hliao added inline comments to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
Fri, Mar 20, 3:12 PM · Restricted Project
hliao added a comment to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
In D76365#1932517, @tra wrote:

That's a magic. I could not figure out how it works. From its use, e.g. tex2D on texture<T, cudaTextureType2D, cudaReadModeElementType>,

__nv_tex_surf_handler("__tex2D_v2", (typename __nv_tex_rmet_cast<T>::type) &temp, t, x, y);

__tex2D_v2 is a string literal. However, it's more likely a underly function name for the real implementation. Hardly imagine that that string literal is checked directly instead used for constructing the real function name. If that's the case, we also need to find that where that underlying functions are defined as the device bitcode library has no such definition.

Most likely it's a compiler built-in with no implementation we could reuse and we'll need to implement our own. It should be fairly straightforward to figure out what it does by compiling all variants used by CUDA headers and observing generated PTX. The first 'meta' argument may be tricky, but we should be able to retrieve the constant string value in the front-end and map it to appropriate intrinsic or generate necessary glue.

Fri, Mar 20, 7:32 AM · Restricted Project
hliao committed rGa4edea29be2a: Fix `-Wunused-variable` warning. NFC. (authored by hliao).
Fix `-Wunused-variable` warning. NFC.
Fri, Mar 20, 7:00 AM

Thu, Mar 19

hliao added a comment to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
In D76365#1932398, @tra wrote:
Thu, Mar 19, 5:00 PM · Restricted Project
hliao added a comment to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
In D76365#1932392, @tra wrote:

Note that, clang-based one needs defining texture fetch functions as they could not be reused from CUDA SDK. That part is enclosed with #if defined(clang).

What prevents clang to compile the texture functions in the CUDA headers? It looks like we'll need to implement the __nv_tex_surf_handler() builtin, but other than that it should work.

Thu, Mar 19, 5:00 PM · Restricted Project
hliao added a comment to D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..

With this revision, the following sample could be compiled with CUDA SDK and almost the same PTX code is generated.

Thu, Mar 19, 3:24 PM · Restricted Project
hliao updated the diff for D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..

More refinement to compile sample code with CUDA headers.

Thu, Mar 19, 3:22 PM · Restricted Project

Wed, Mar 18

hliao updated the diff for D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..

Revise one part of the logic to reduce condition evaluation overhead.

Wed, Mar 18, 7:32 PM · Restricted Project
hliao updated the diff for D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

Fix more clang-tidy warnings.

Wed, Mar 18, 4:51 PM · Restricted Project
hliao updated the diff for D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

Fix warnings from clang-tidy.

Wed, Mar 18, 2:40 PM · Restricted Project
hliao updated the diff for D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..

Fix warnings from clang-tidy.

Wed, Mar 18, 10:20 AM · Restricted Project
hliao updated the diff for D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..

Reformatting with clang-format.

Wed, Mar 18, 9:14 AM · Restricted Project
hliao created D76365: [cuda][hip] Add CUDA builtin surface/texture reference support..
Wed, Mar 18, 8:08 AM · Restricted Project
hliao committed rG4cf01ed75e35: [hip] Revise `GlobalDecl` constructors. NFC. (authored by hliao).
[hip] Revise `GlobalDecl` constructors. NFC.
Wed, Mar 18, 7:03 AM
hliao closed D76344: [hip] Revise `GlobalDecl` constructors. NFC..
Wed, Mar 18, 7:03 AM · Restricted Project

Tue, Mar 17

hliao created D76344: [hip] Revise `GlobalDecl` constructors. NFC..
Tue, Mar 17, 11:13 PM · Restricted Project
hliao updated the diff for D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

Rebase to the latest trunk.

Tue, Mar 17, 7:27 PM · Restricted Project
hliao committed rGf2f8bdc2b166: Fix `-Wunused-variable` warning. NFC. (authored by hliao).
Fix `-Wunused-variable` warning. NFC.
Tue, Mar 17, 5:19 PM
hliao committed rGa2920c4ea997: [codegen] Fix one more case where `getGlobalDecl` should be used. NFC. (authored by hliao).
[codegen] Fix one more case where `getGlobalDecl` should be used. NFC.
Tue, Mar 17, 3:09 PM

Mon, Mar 16

hliao committed rGd00d6a19dda9: Fix `-Wpedantic` warning. NFC. (authored by hliao).
Fix `-Wpedantic` warning. NFC.
Mon, Mar 16, 7:08 PM

Fri, Mar 13

hliao committed rG7c504548cae7: Fix `-Wunused-variable`. NFC. (authored by hliao).
Fix `-Wunused-variable`. NFC.
Fri, Mar 13, 6:22 PM

Mar 8 2020

hliao committed rG073dbaae3972: Fix GCC warnings. NFC. (authored by hliao).
Fix GCC warnings. NFC.
Mar 8 2020, 10:08 AM

Mar 2 2020

hliao added a comment to D75402: [HIP] Make sure, unused hip-pinned-shadow global var is kept within device code.

BTW, why that variable cannot have an initializer? Suppose that initializer is a trivial one, initializing to 0, would that cause any issue in the compilation?

Mar 2 2020, 9:55 AM · Restricted Project
hliao added inline comments to D75402: [HIP] Make sure, unused hip-pinned-shadow global var is kept within device code.
Mar 2 2020, 9:55 AM · Restricted Project
hliao updated the diff for D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

Remove unncessary condition checking.

Mar 2 2020, 9:30 AM · Restricted Project
hliao added a comment to D75402: [HIP] Make sure, unused hip-pinned-shadow global var is kept within device code.

the assertion in addUsedGlobal should not be removed. that will remove the guard for potential bugs

Mar 2 2020, 9:29 AM · Restricted Project
hliao added a comment to D75423: [OpenCL] Mark pointers to constant address space as invariant.

invariant checking already takes account of loading from constant address space or memory (AA::pointsToConstantMemory), that's almost equivalent to adding invariant attributes. Why do we mark these constant loads with additional attributes?

Mar 2 2020, 8:41 AM

Feb 28 2020

hliao committed rG1a52d3d47b84: [cmake] Fix LLVM_USE_SPLIT_DWARF (authored by hliao).
[cmake] Fix LLVM_USE_SPLIT_DWARF
Feb 28 2020, 8:08 PM
hliao closed D75328: [cmake] Fix LLVM_USE_SPLIT_DWARF.
Feb 28 2020, 8:08 PM · Restricted Project
hliao updated the diff for D75328: [cmake] Fix LLVM_USE_SPLIT_DWARF.

That makes sense to me. At least, in release build, we won't generate
unnecessary files.

Feb 28 2020, 12:21 PM · Restricted Project
hliao updated the diff for D75328: [cmake] Fix LLVM_USE_SPLIT_DWARF.

Think again. There's no need to tell the languages for adding that compile option.

Feb 28 2020, 11:12 AM · Restricted Project
hliao updated the diff for D75328: [cmake] Fix LLVM_USE_SPLIT_DWARF.

Apply to objc and objcxx.

Feb 28 2020, 11:01 AM · Restricted Project
hliao updated the diff for D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

Fix pre-merge checks.

Feb 28 2020, 6:39 AM · Restricted Project

Feb 27 2020

hliao created D75328: [cmake] Fix LLVM_USE_SPLIT_DWARF.
Feb 27 2020, 9:46 PM · Restricted Project
Michael LIAO <michaelhongbin.liao@amd.com> committed rG5392566a34f6: CP 243ebfba17da72566ba29a891193e4814cbc4ef3 and revert unnecessary changes. (authored by hliao).
CP 243ebfba17da72566ba29a891193e4814cbc4ef3 and revert unnecessary changes.
Feb 27 2020, 9:28 AM
Michael LIAO <michaelhongbin.liao@amd.com> committed rG4e7ccb37582e: Merge [AMDGPU] Allow using integral non-type template parameters (authored by hliao).
Merge [AMDGPU] Allow using integral non-type template parameters
Feb 27 2020, 12:57 AM

Feb 26 2020

hliao added a comment to D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

@rjmccall @rsmith @tra, could you review on this revision?

Feb 26 2020, 3:06 PM · Restricted Project
hliao updated the diff for D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

Rebase to the latest trunk.

Feb 26 2020, 2:02 PM · Restricted Project

Feb 25 2020

hliao committed rG61f538d37b80: Add missing dependency to fix shared library build. (authored by hliao).
Add missing dependency to fix shared library build.
Feb 25 2020, 11:03 PM
hliao added a comment to D75135: Revert "[WebAssembly] MC: Mark more function aliases as functions".

Sounds reasonable. We can looking into other solutions.

My understand is that the IR in question is perfectly valid, and can be compiled on other targets, can't it? Its only WebAssembly that can't compile this. We can't ask the producer so not produce this valid bitcode when building for WebAssembly can we?

Feb 25 2020, 10:34 PM · Restricted Project
hliao added a comment to D75135: Revert "[WebAssembly] MC: Mark more function aliases as functions".

It seems incorrect for me as the input IR has an alias casting function pointer to data pointer. That symbol itself should be regarded as a function symbol any more. If that IR needs to pass compilation, it should correct its generation by replacing that casting with one casting between function pointers. The casting from a function pointer to data pointer could be done within the accessor function.

Feb 25 2020, 12:52 PM · Restricted Project
hliao created D75135: Revert "[WebAssembly] MC: Mark more function aliases as functions".
Feb 25 2020, 12:49 PM · Restricted Project

Feb 24 2020

hliao added a comment to D75014: [InstrEmitter, SystemZ] Copy Access registers with the correct register class..

Removed from reviewer list as the new patch takes all target-specific approach.

Feb 24 2020, 3:06 PM · Restricted Project
hliao removed a reviewer for D75014: [InstrEmitter, SystemZ] Copy Access registers with the correct register class.: hliao.
Feb 24 2020, 3:06 PM · Restricted Project
hliao updated the diff for D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

Rebase to the trunk.

Feb 24 2020, 7:35 AM · Restricted Project

Feb 21 2020

hliao updated the diff for D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

Rebase to the trunk.

Feb 21 2020, 8:13 AM · Restricted Project

Feb 20 2020

hliao updated the diff for D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

Revise following reviewer comments.

Feb 20 2020, 2:30 PM · Restricted Project
hliao updated the diff for D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

Rebase the code to the latest trunk.

Feb 20 2020, 2:02 PM · Restricted Project
hliao added inline comments to D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..
Feb 20 2020, 10:02 AM · Restricted Project
hliao updated the diff for D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

Skip non-function or non-TU context so far as more cases need considering.

Feb 20 2020, 10:02 AM · Restricted Project

Feb 19 2020

hliao updated the diff for D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

Rebase to the latest trunk code.

Feb 19 2020, 10:40 AM · Restricted Project

Feb 17 2020

hliao added a reviewer for D71227: [cuda][hip] Fix function overload resolution in the global initiailizer.: rjmccall.
Feb 17 2020, 8:47 AM · Restricted Project
hliao added a comment to D71227: [cuda][hip] Fix function overload resolution in the global initiailizer..

+ @rjmccall

Feb 17 2020, 8:47 AM · Restricted Project

Feb 16 2020

hliao committed rG487fcc8d3de4: Fix `-Wpedantic` warning. NFC. (authored by hliao).
Fix `-Wpedantic` warning. NFC.
Feb 16 2020, 9:19 PM
hliao added a comment to D70172: [CUDA][HIP][OpenMP] Emit deferred diagnostics by a post-parsing AST travese.

one header is missing and breaks the build

Feb 16 2020, 8:05 PM · Restricted Project

Feb 12 2020

hliao committed rGf6a3ac150b8d: Fix `-Wunused-variable` warning. NFC. (authored by hliao).
Fix `-Wunused-variable` warning. NFC.
Feb 12 2020, 9:47 AM

Feb 9 2020

hliao committed rGab3da5dd66c0: Fix `-Wparentheses` warning. NFC. (authored by hliao).
Fix `-Wparentheses` warning. NFC.
Feb 9 2020, 9:51 PM