grypp wrote:
Can we please not merge anything without review?
https://github.com/llvm/llvm-project/pull/136768
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/grypp approved this pull request.
https://github.com/llvm/llvm-project/pull/127736
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -227,14 +228,14 @@ class NVVMDialectLLVMIRTranslationInterface
} else if (attribute.getName() ==
grypp wrote:
Sure sounds great
https://github.com/llvm/llvm-project/pull/125908
___
cfe-commits mailing list
cfe-
grypp wrote:
Thank you for clarifying! I wasn’t aware that this change also benefits nvcc. A
2% improvement is an excellent result!
>From the MLIR side, the PR looks good to me.
https://github.com/llvm/llvm-project/pull/120806
___
cfe-commits mailing
https://github.com/grypp approved this pull request.
https://github.com/llvm/llvm-project/pull/120806
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
grypp wrote:
In MLIR, we also have other NVVM metadata such as `reqntid` and `maxntid`,
among others. What is the plan for these? Will they remain as metadata, or will
they be expressed differently?
Could you please elaborate on the compile-time improvements?
https://github.com/llvm/llvm-pr
https://github.com/grypp closed https://github.com/llvm/llvm-project/pull/74058
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/grypp updated https://github.com/llvm/llvm-project/pull/74058
>From 9f35504e81246f97a9d8c14a06043685660ae15e Mon Sep 17 00:00:00 2001
From: Guray Ozen
Date: Fri, 1 Dec 2023 11:10:40 +0100
Subject: [PATCH 1/3] [mlir][nvvm] Introduce `fence.mbarrier.init`
This PR introduce `fen
https://github.com/grypp closed https://github.com/llvm/llvm-project/pull/74475
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/grypp updated https://github.com/llvm/llvm-project/pull/74475
>From 2848c9011cb4db5e91754300eb466927738a363a Mon Sep 17 00:00:00 2001
From: Guray Ozen
Date: Tue, 5 Dec 2023 15:16:20 +0100
Subject: [PATCH] [mlir][gpu] Support dynamic_shared_memory Op with vector
dialect
`gpu.
https://github.com/grypp updated https://github.com/llvm/llvm-project/pull/74058
>From 9f35504e81246f97a9d8c14a06043685660ae15e Mon Sep 17 00:00:00 2001
From: Guray Ozen
Date: Fri, 1 Dec 2023 11:10:40 +0100
Subject: [PATCH 1/2] [mlir][nvvm] Introduce `fence.mbarrier.init`
This PR introduce `fen
https://github.com/grypp closed https://github.com/llvm/llvm-project/pull/69648
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/grypp approved this pull request.
This looks good to me! I'm not a SYCL expert, but maybe someone with more
expertise could take a quick look here. @Hardcode84 ?
https://github.com/llvm/llvm-project/pull/69648
___
cfe-commits mailin
@@ -0,0 +1,222 @@
+//===- SyclRuntimeWrappers.cpp - MLIR SYCL wrapper library ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-
@@ -0,0 +1,222 @@
+//===- SyclRuntimeWrappers.cpp - MLIR SYCL wrapper library ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-
@@ -0,0 +1,222 @@
+//===- SyclRuntimeWrappers.cpp - MLIR SYCL wrapper library ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-
@@ -0,0 +1,222 @@
+//===- SyclRuntimeWrappers.cpp - MLIR SYCL wrapper library ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-
@@ -0,0 +1,222 @@
+//===- SyclRuntimeWrappers.cpp - MLIR SYCL wrapper library ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-
https://github.com/grypp edited https://github.com/llvm/llvm-project/pull/69648
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -0,0 +1,222 @@
+//===- SyclRuntimeWrappers.cpp - MLIR SYCL wrapper library ===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-
https://github.com/grypp commented:
Looks good overall to me. I left a few comments
https://github.com/llvm/llvm-project/pull/69648
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -53,6 +55,16 @@ static Value truncToI32(ConversionPatternRewriter &rewriter,
Location loc,
return rewriter.create(loc, rewriter.getI32Type(), value);
}
+/// Returns warp-size as a value.
+static Value getWarpSizeValue(ImplicitLocOpBuilder &b) {
grypp wro
https://github.com/grypp updated https://github.com/llvm/llvm-project/pull/65441
>From 7b71da55fca8fe2a7dbe4982b1959be6a6175fa1 Mon Sep 17 00:00:00 2001
From: Guray Ozen
Date: Thu, 7 Sep 2023 11:52:38 +0200
Subject: [PATCH 1/7] [MLIR][NVGPU] Introduce `nvgpu.warpgroup.mma.store` Op
for Hopper G
https://github.com/grypp updated https://github.com/llvm/llvm-project/pull/65441
>From 7b71da55fca8fe2a7dbe4982b1959be6a6175fa1 Mon Sep 17 00:00:00 2001
From: Guray Ozen
Date: Thu, 7 Sep 2023 11:52:38 +0200
Subject: [PATCH 1/6] [MLIR][NVGPU] Introduce `nvgpu.warpgroup.mma.store` Op
for Hopper G
@@ -811,8 +812,13 @@ LogicalResult
ConvertAllocOpToGpuRuntimeCallPattern::matchAndRewrite(
// descriptor.
Type elementPtrType = this->getElementPtrType(memRefType);
auto stream = adaptor.getAsyncDependencies().front();
+
+ auto isHostShared = rewriter.create(
+ loc
@@ -811,8 +812,13 @@ LogicalResult
ConvertAllocOpToGpuRuntimeCallPattern::matchAndRewrite(
// descriptor.
Type elementPtrType = this->getElementPtrType(memRefType);
auto stream = adaptor.getAsyncDependencies().front();
+
+ auto isHostShared = rewriter.create(
+ loc
@@ -811,8 +812,13 @@ LogicalResult
ConvertAllocOpToGpuRuntimeCallPattern::matchAndRewrite(
// descriptor.
Type elementPtrType = this->getElementPtrType(memRefType);
auto stream = adaptor.getAsyncDependencies().front();
+
+ auto isHostShared = rewriter.create(
+ loc
@@ -811,8 +812,13 @@ LogicalResult
ConvertAllocOpToGpuRuntimeCallPattern::matchAndRewrite(
// descriptor.
Type elementPtrType = this->getElementPtrType(memRefType);
auto stream = adaptor.getAsyncDependencies().front();
+
+ auto isHostShared = rewriter.create(
+ loc
https://github.com/grypp edited https://github.com/llvm/llvm-project/pull/65539
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -811,8 +812,13 @@ LogicalResult
ConvertAllocOpToGpuRuntimeCallPattern::matchAndRewrite(
// descriptor.
Type elementPtrType = this->getElementPtrType(memRefType);
auto stream = adaptor.getAsyncDependencies().front();
+
+ auto isHostShared = rewriter.create(
+ loc
30 matches
Mail list logo
