https://github.com/changpeng closed
https://github.com/llvm/llvm-project/pull/99551
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/changpeng created
https://github.com/llvm/llvm-project/pull/99551
None
>From 31601c539553d7af0efd94722eabf4627f8a387c Mon Sep 17 00:00:00 2001
From: Changpeng Fang
Date: Thu, 18 Jul 2024 11:03:24 -0700
Subject: [PATCH 1/2] AMDGPU: Add back half and bfloat support for
global
@@ -1590,10 +1590,14 @@ let OtherPredicates = [isGFX12Plus] in {
let WaveSizePredicate = isWave32 in {
defm : GlobalFLATLoadPats ;
defm : GlobalFLATLoadPats ;
+defm : GlobalFLATLoadPats ;
+defm : GlobalFLATLoadPats ;
}
let WaveSizePredicate = isWave64 in
https://github.com/changpeng closed
https://github.com/llvm/llvm-project/pull/99540
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/changpeng created
https://github.com/llvm/llvm-project/pull/99540
half and bfloat are common types for 16-bit elements. The support of them was
original there and dropped due to some reasons. This work adds the support of
the float types back.
>From 31601c539553d7af0efd947
https://github.com/changpeng closed
https://github.com/llvm/llvm-project/pull/92612
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -0,0 +1,25 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
UTC_ARGS: --version 4
+//RUN: %clang_cc1 %s -emit-llvm -O1 -o - | FileCheck %s
changpeng wrote:
add "triple spir", same as other tests in the same directory
https://
https://github.com/changpeng updated
https://github.com/llvm/llvm-project/pull/92612
>From 2468a85a47499d90a99610846c632332eb7307b8 Mon Sep 17 00:00:00 2001
From: Changpeng Fang
Date: Fri, 17 May 2024 15:13:07 -0700
Subject: [PATCH 1/3] [OpenCL] Fix an infinite loop in builidng
AddrSpaceQualTy
https://github.com/changpeng edited
https://github.com/llvm/llvm-project/pull/92612
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -3054,6 +3054,13 @@ QualType ASTContext::removeAddrSpaceQualType(QualType T)
const {
if (!T.hasAddressSpace())
return T;
+ // For arrays, strip the qualifier off the element type, then reconstruct the
+ // array type
+ if (T.getTypePtr()->isArrayType()) {
+Qua
@@ -537,8 +537,9 @@ void AggExprEmitter::EmitArrayInit(Address DestPtr,
llvm::ArrayType *AType,
elementType.isTriviallyCopyableType(CGF.getContext())) {
CodeGen::CodeGenModule &CGM = CGF.CGM;
ConstantEmitter Emitter(CGF);
+Qualifiers Quals;
QualType GVAr
https://github.com/changpeng updated
https://github.com/llvm/llvm-project/pull/92612
>From 2468a85a47499d90a99610846c632332eb7307b8 Mon Sep 17 00:00:00 2001
From: Changpeng Fang
Date: Fri, 17 May 2024 15:13:07 -0700
Subject: [PATCH 1/2] [OpenCL] Fix an infinite loop in builidng
AddrSpaceQualTy
@@ -537,8 +537,9 @@ void AggExprEmitter::EmitArrayInit(Address DestPtr,
llvm::ArrayType *AType,
elementType.isTriviallyCopyableType(CGF.getContext())) {
CodeGen::CodeGenModule &CGM = CGF.CGM;
ConstantEmitter Emitter(CGF);
+Qualifiers Quals;
QualType GVAr
@@ -535,20 +535,23 @@ void AggExprEmitter::EmitArrayInit(Address DestPtr,
llvm::ArrayType *AType,
elementType.isTriviallyCopyableType(CGF.getContext())) {
CodeGen::CodeGenModule &CGM = CGF.CGM;
ConstantEmitter Emitter(CGF);
-LangAS AS = ArrayQTy.getAddressSpa
https://github.com/changpeng created
https://github.com/llvm/llvm-project/pull/92612
In building AddrSpaceQualType
(https://github.com/llvm/llvm-project/pull/90048), there is a bug in
removeAddrSpaceQualType() for arrays. Arrays are weird because qualifiers on
the element type also count as
@@ -535,20 +535,23 @@ void AggExprEmitter::EmitArrayInit(Address DestPtr,
llvm::ArrayType *AType,
elementType.isTriviallyCopyableType(CGF.getContext())) {
CodeGen::CodeGenModule &CGM = CGF.CGM;
ConstantEmitter Emitter(CGF);
-LangAS AS = ArrayQTy.getAddressSpa
changpeng wrote:
[test.cl.txt](https://github.com/llvm/llvm-project/files/15355457/test.cl.txt)
https://github.com/llvm/llvm-project/pull/90048
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe
@@ -535,20 +535,23 @@ void AggExprEmitter::EmitArrayInit(Address DestPtr,
llvm::ArrayType *AType,
elementType.isTriviallyCopyableType(CGF.getContext())) {
CodeGen::CodeGenModule &CGM = CGF.CGM;
ConstantEmitter Emitter(CGF);
-LangAS AS = ArrayQTy.getAddressSpa
@@ -535,20 +535,23 @@ void AggExprEmitter::EmitArrayInit(Address DestPtr,
llvm::ArrayType *AType,
elementType.isTriviallyCopyableType(CGF.getContext())) {
CodeGen::CodeGenModule &CGM = CGF.CGM;
ConstantEmitter Emitter(CGF);
-LangAS AS = ArrayQTy.getAddressSpa
@@ -535,20 +535,23 @@ void AggExprEmitter::EmitArrayInit(Address DestPtr,
llvm::ArrayType *AType,
elementType.isTriviallyCopyableType(CGF.getContext())) {
CodeGen::CodeGenModule &CGM = CGF.CGM;
ConstantEmitter Emitter(CGF);
-LangAS AS = ArrayQTy.getAddressSpa
https://github.com/changpeng closed
https://github.com/llvm/llvm-project/pull/86707
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/changpeng updated
https://github.com/llvm/llvm-project/pull/86707
>From 485dff66813104ad73d8eada7cd7d43edf9d093d Mon Sep 17 00:00:00 2001
From: Changpeng Fang
Date: Tue, 26 Mar 2024 11:06:48 -0700
Subject: [PATCH 1/3] AMDGPU: Simplify EmitAMDGPUBuiltinExpr for load
transpose
https://github.com/changpeng updated
https://github.com/llvm/llvm-project/pull/86707
>From 485dff66813104ad73d8eada7cd7d43edf9d093d Mon Sep 17 00:00:00 2001
From: Changpeng Fang
Date: Tue, 26 Mar 2024 11:06:48 -0700
Subject: [PATCH 1/2] AMDGPU: Simplify EmitAMDGPUBuiltinExpr for load
transpose
https://github.com/changpeng created
https://github.com/llvm/llvm-project/pull/86707
We should not manually get the types of the loading data.
Instead, we can get the types from the intrinsics directly.
>From 485dff66813104ad73d8eada7cd7d43edf9d093d Mon Sep 17 00:00:00 2001
From: Changpeng Fa
https://github.com/changpeng edited
https://github.com/llvm/llvm-project/pull/86313
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -18533,51 +18533,35 @@ Value
*CodeGenFunction::EmitAMDGPUBuiltinExpr(unsigned BuiltinID,
}
case AMDGPU::BI__builtin_amdgcn_global_load_tr_b64_i32:
case AMDGPU::BI__builtin_amdgcn_global_load_tr_b64_v2i32:
- case AMDGPU::BI__builtin_amdgcn_global_load_tr_b128_v4bf16:
changpeng wrote:
ping
https://github.com/llvm/llvm-project/pull/86313
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
changpeng wrote:
> There is no issue in changing the names in principle. Curious, what is the
> rationale to use more demangled names?
more user friendly.
https://github.com/llvm/llvm-project/pull/86313
___
cfe-commits mailing list
cfe-commits@lists.
changpeng wrote:
> > global_load_re_b64
>
> Type global_load_re_b64.
Changed! Thanks.
https://github.com/llvm/llvm-project/pull/86313
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/changpeng edited
https://github.com/llvm/llvm-project/pull/86313
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/changpeng created
https://github.com/llvm/llvm-project/pull/86313
Rename the intrinsics to close to the instruction mnemonic names:
Use global_load_re_b64 and global_load_tr_b128 instead of global_load_tr.
This patch also removes f16/bf16 versions of builtins/intrinsics.
changpeng wrote:
I am going to propose to rename intrinsics and remove f16/bf16 versions of
builtins/intrinsics
https://github.com/llvm/llvm-project/pull/86202
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mail
https://github.com/changpeng closed
https://github.com/llvm/llvm-project/pull/86202
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
changpeng wrote:
[AMD Official Use Only - General]
I am fine to remove f16/bf16 versions. Enumerating all possible types could be
very painful. For example we gave up enumerating for B64, and ended up using
v2i32 only. What do others think removing f16/bf16 versions? Thanks
Get Outlook for iO
changpeng wrote:
> I don't think intrinsics are meant for users. Builtins are the user-facing
> front. :-)
Then renaing the intrinsics should be relatively at a lower priority. We may do
it in a separate patch once we have reached an agreement.
https://github.com/llvm/llvm-project/pull/86202
changpeng wrote:
> > > Do you want to rename intrinsics as well? Because now intrinsic names do
> > > not match builtin names.
> >
> >
> > Do we have to match builtins with intrinsics? Renaming intrinsics here
> > means we will have to duplicate the intrinsics.
>
> Is that because of the man
changpeng wrote:
> Do you want to rename intrinsics as well? Because now intrinsic names do not
> match builtin names.
Do we have to match builtins with intrinsics? Renaming intrinsics here means we
will have to duplicate the intrinsics.
https://github.com/llvm/llvm-project/pull/86202
__
https://github.com/changpeng created
https://github.com/llvm/llvm-project/pull/86202
Make the name of a clang builtin as close to the mnemonic instruction name as
possible. The data type suffix may not be enough to tell what instruction the
builtin is going to produce.
This patch also add
https://github.com/changpeng closed
https://github.com/llvm/llvm-project/pull/84248
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/changpeng created
https://github.com/llvm/llvm-project/pull/84248
FeatureDot11Insts (dot11-insts) for:
v_dot4_f32_fp8_fp8, v_dot4_f32_fp8_bf8,
v_dot4_f32_bf8_fp8, v_dot4_f32_bf8_bf8
>From 1bfc1e048d10e57c3d07038f52b072163f3b4ff9 Mon Sep 17 00:00:00 2001
From: Changpeng Fa
https://github.com/changpeng closed
https://github.com/llvm/llvm-project/pull/79394
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/changpeng updated
https://github.com/llvm/llvm-project/pull/79394
>From 0f3af077baeff26d9796db73e7af19b097272fa2 Mon Sep 17 00:00:00 2001
From: Changpeng Fang
Date: Wed, 24 Jan 2024 16:28:23 -0800
Subject: [PATCH] [AMDGPU] Rename AMDGPULoadTr intrinsic class. NFC.
This cla
https://github.com/changpeng closed
https://github.com/llvm/llvm-project/pull/79218
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/changpeng closed
https://github.com/llvm/llvm-project/pull/79104
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/changpeng approved this pull request.
https://github.com/llvm/llvm-project/pull/2
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/changpeng deleted
https://github.com/llvm/llvm-project/pull/2
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -2496,6 +2496,26 @@ def int_amdgcn_flat_atomic_fmax_num :
AMDGPUAtomicRtn;
def int_amdgcn_global_atomic_fmin_num : AMDGPUAtomicRtn;
def int_amdgcn_global_atomic_fmax_num : AMDGPUAtomicRtn;
+class AMDGPUGlobalLoadTr :
+ Intrinsic<
+[data_ty],
+[global_ptr_ty],
+
@@ -18178,6 +18178,51 @@ Value *CodeGenFunction::EmitAMDGPUBuiltinExpr(unsigned
BuiltinID,
llvm::Function *F = CGM.getIntrinsic(IID, {ArgTy});
return Builder.CreateCall(F, {Addr, Val, ZeroI32, ZeroI32, ZeroI1});
}
+ case AMDGPU::BI__builtin_amdgcn_global_load_tr_b64
https://github.com/changpeng closed
https://github.com/llvm/llvm-project/pull/71989
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/changpeng reopened
https://github.com/llvm/llvm-project/pull/71989
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/changpeng closed
https://github.com/llvm/llvm-project/pull/71989
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
changpeng wrote:
> Typo in subject "**Guard** return ..."?
You are right. Thanks.
https://github.com/llvm/llvm-project/pull/71989
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/changpeng edited
https://github.com/llvm/llvm-project/pull/71989
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
changpeng wrote:
> Any tests?
Encountered this issue during a downstream branch testing. No test for trunk
yet but think the issue should be here.
https://github.com/llvm/llvm-project/pull/71989
___
cfe-commits mailing list
cfe-commits@lists.llvm.or
https://github.com/changpeng closed
https://github.com/llvm/llvm-project/pull/70484
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/changpeng edited
https://github.com/llvm/llvm-project/pull/70484
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Author: Changpeng Fang
Date: 2023-08-11T16:37:42-07:00
New Revision: d77c62053c944652846c00a35c921e14b43b1877
URL:
https://github.com/llvm/llvm-project/commit/d77c62053c944652846c00a35c921e14b43b1877
DIFF:
https://github.com/llvm/llvm-project/commit/d77c62053c944652846c00a35c921e14b43b1877.diff
Author: Changpeng Fang
Date: 2023-08-09T18:23:18-07:00
New Revision: 4608686849bcb6e20de827750862d5345cbd
URL:
https://github.com/llvm/llvm-project/commit/4608686849bcb6e20de827750862d5345cbd
DIFF:
https://github.com/llvm/llvm-project/commit/4608686849bcb6e20de827750862d5345cbd.diff
Author: Changpeng Fang
Date: 2022-03-17T14:12:36-07:00
New Revision: dd5895cc39864393f8ca357bc4e23e8d7b5b9723
URL:
https://github.com/llvm/llvm-project/commit/dd5895cc39864393f8ca357bc4e23e8d7b5b9723
DIFF:
https://github.com/llvm/llvm-project/commit/dd5895cc39864393f8ca357bc4e23e8d7b5b9723.diff
Author: chfang
Date: Thu Aug 18 17:04:54 2016
New Revision: 279165
URL: http://llvm.org/viewvc/llvm-project?rev=279165&view=rev
Log:
AMDGPU: Add clang builtin for ds_swizzle.
Summary:
int __builtin_amdgcn_ds_swizzle (int a, int imm);
while imm is a constant.
Differential Revision:
http://rev
60 matches
Mail list logo
