[MLIR][GPUToNVVM] Remove typed pointer support #70861

Dinistro · 2023-10-31T21:33:39Z

This commit removes the support for lowering GPU to NVVM dialect with typed pointers. Typed pointers have been deprecated for a while now and it's planned to soon remove them from the LLVM dialect.

Related PSA: https://discourse.llvm.org/t/psa-removal-of-typed-pointers-from-the-llvm-dialect/74502

This commit removes the support for lowering GPU to NVVM dialect with typed pointers. Typed pointers have been deprecated for a while now and it's planned to soon remove them from the LLVM dialect. Related PSA: https://discourse.llvm.org/t/psa-removal-of-typed-pointers-from-the-llvm-dialect/74502

llvmbot · 2023-10-31T21:34:27Z

@llvm/pr-subscribers-mlir

@llvm/pr-subscribers-mlir-gpu

Author: Christian Ulmann (Dinistro)

Changes

This commit removes the support for lowering GPU to NVVM dialect with typed pointers. Typed pointers have been deprecated for a while now and it's planned to soon remove them from the LLVM dialect.

Related PSA: https://discourse.llvm.org/t/psa-removal-of-typed-pointers-from-the-llvm-dialect/74502

Full diff: https://github.com/llvm/llvm-project/pull/70861.diff

11 Files Affected:

(modified) mlir/include/mlir/Conversion/Passes.td (+1-4)
(modified) mlir/lib/Conversion/GPUToNVVM/LowerGpuOpsToNVVMOps.cpp (-1)
(removed) mlir/test/Conversion/GPUCommon/lower-memory-space-attrs-typed-pointers.mlir (-15)
(modified) mlir/test/Conversion/GPUCommon/lower-memory-space-attrs.mlir (+1-1)
(modified) mlir/test/Conversion/GPUCommon/memory-attrbution.mlir (+1-1)
(modified) mlir/test/Conversion/GPUCommon/memref-arg-attrs.mlir (+1-1)
(modified) mlir/test/Conversion/GPUCommon/memref-arg-noalias-attrs.mlir (+1-1)
(modified) mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm-32b.mlir (+1-1)
(modified) mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm.mlir (+1-1)
(removed) mlir/test/Conversion/GPUToNVVM/typed-pointers.mlir (-39)
(modified) mlir/test/Conversion/GPUToNVVM/wmma-ops-to-nvvm.mlir (+2-2)

diff --git a/mlir/include/mlir/Conversion/Passes.td b/mlir/include/mlir/Conversion/Passes.td
index 036c9b0039779ab..aae364ad02e8e7b 100644
--- a/mlir/include/mlir/Conversion/Passes.td
+++ b/mlir/include/mlir/Conversion/Passes.td
@@ -501,10 +501,7 @@ def ConvertGpuOpsToNVVMOps : Pass<"convert-gpu-to-nvvm", "gpu::GPUModuleOp"> {
     Option<"useBarePtrCallConv", "use-bare-ptr-memref-call-conv", "bool",
            /*default=*/"false",
            "Replace memref arguments in GPU functions with bare pointers. "
-           "All memrefs must have static shape.">,
-    Option<"useOpaquePointers", "use-opaque-pointers", "bool",
-                   /*default=*/"true", "Generate LLVM IR using opaque pointers "
-                   "instead of typed pointers">,
+           "All memrefs must have static shape.">
   ];
 }
 
diff --git a/mlir/lib/Conversion/GPUToNVVM/LowerGpuOpsToNVVMOps.cpp b/mlir/lib/Conversion/GPUToNVVM/LowerGpuOpsToNVVMOps.cpp
index 764b6a779b98c7a..935e3d2a4095003 100644
--- a/mlir/lib/Conversion/GPUToNVVM/LowerGpuOpsToNVVMOps.cpp
+++ b/mlir/lib/Conversion/GPUToNVVM/LowerGpuOpsToNVVMOps.cpp
@@ -227,7 +227,6 @@ struct LowerGpuOpsToNVVMOpsPass
         DataLayout(cast<DataLayoutOpInterface>(m.getOperation())));
     if (indexBitwidth != kDeriveIndexBitwidthFromDataLayout)
       options.overrideIndexBitwidth(indexBitwidth);
-    options.useOpaquePointers = useOpaquePointers;
     options.useBarePtrCallConv = useBarePtrCallConv;
 
     // Apply in-dialect lowering. In-dialect lowering will replace
diff --git a/mlir/test/Conversion/GPUCommon/lower-memory-space-attrs-typed-pointers.mlir b/mlir/test/Conversion/GPUCommon/lower-memory-space-attrs-typed-pointers.mlir
deleted file mode 100644
index ebf1f309b0e44e4..000000000000000
--- a/mlir/test/Conversion/GPUCommon/lower-memory-space-attrs-typed-pointers.mlir
+++ /dev/null
@@ -1,15 +0,0 @@
-// RUN: mlir-opt %s -split-input-file -convert-gpu-to-rocdl='use-opaque-pointers=0' | FileCheck %s --check-prefixes=CHECK,ROCDL
-// RUN: mlir-opt %s -split-input-file -convert-gpu-to-nvvm='use-opaque-pointers=0' | FileCheck %s --check-prefixes=CHECK,NVVM
-
-gpu.module @kernel {
-  gpu.func @private(%arg0: f32) private(%arg1: memref<4xf32, #gpu.address_space<private>>) {
-    %c0 = arith.constant 0 : index
-    memref.store %arg0, %arg1[%c0] : memref<4xf32, #gpu.address_space<private>>
-    gpu.return
-  }
-}
-
-// CHECK-LABEL:  llvm.func @private
-//      CHECK:  llvm.store
-// ROCDL-SAME:   : !llvm.ptr<f32, 5>
-//  NVVM-SAME:   : !llvm.ptr<f32>
diff --git a/mlir/test/Conversion/GPUCommon/lower-memory-space-attrs.mlir b/mlir/test/Conversion/GPUCommon/lower-memory-space-attrs.mlir
index cde11a73397f4de..2d7e90dd1f5cb77 100644
--- a/mlir/test/Conversion/GPUCommon/lower-memory-space-attrs.mlir
+++ b/mlir/test/Conversion/GPUCommon/lower-memory-space-attrs.mlir
@@ -1,5 +1,5 @@
 // RUN: mlir-opt %s -split-input-file -convert-gpu-to-rocdl='use-opaque-pointers=1' | FileCheck %s --check-prefixes=CHECK,ROCDL
-// RUN: mlir-opt %s -split-input-file -convert-gpu-to-nvvm='use-opaque-pointers=1' | FileCheck %s --check-prefixes=CHECK,NVVM
+// RUN: mlir-opt %s -split-input-file -convert-gpu-to-nvvm | FileCheck %s --check-prefixes=CHECK,NVVM
 
 gpu.module @kernel {
   gpu.func @private(%arg0: f32) private(%arg1: memref<4xf32, #gpu.address_space<private>>) {
diff --git a/mlir/test/Conversion/GPUCommon/memory-attrbution.mlir b/mlir/test/Conversion/GPUCommon/memory-attrbution.mlir
index 2457cbcf3251cb3..0231cf6f9806659 100644
--- a/mlir/test/Conversion/GPUCommon/memory-attrbution.mlir
+++ b/mlir/test/Conversion/GPUCommon/memory-attrbution.mlir
@@ -1,4 +1,4 @@
-// RUN: mlir-opt -allow-unregistered-dialect --convert-gpu-to-nvvm='use-opaque-pointers=1' --split-input-file %s | FileCheck --check-prefix=NVVM %s
+// RUN: mlir-opt -allow-unregistered-dialect --convert-gpu-to-nvvm --split-input-file %s | FileCheck --check-prefix=NVVM %s
 // RUN: mlir-opt -allow-unregistered-dialect --convert-gpu-to-rocdl='use-opaque-pointers=1' --split-input-file %s | FileCheck --check-prefix=ROCDL %s
 
 gpu.module @kernel {
diff --git a/mlir/test/Conversion/GPUCommon/memref-arg-attrs.mlir b/mlir/test/Conversion/GPUCommon/memref-arg-attrs.mlir
index 71ae5fbf154adf8..21c4d5696c42151 100644
--- a/mlir/test/Conversion/GPUCommon/memref-arg-attrs.mlir
+++ b/mlir/test/Conversion/GPUCommon/memref-arg-attrs.mlir
@@ -1,5 +1,5 @@
 // RUN: mlir-opt %s -split-input-file -convert-gpu-to-rocdl='use-opaque-pointers=1 use-bare-ptr-memref-call-conv=0' | FileCheck %s --check-prefixes=CHECK,ROCDL
-// RUN: mlir-opt %s -split-input-file -convert-gpu-to-nvvm='use-opaque-pointers=1 use-bare-ptr-memref-call-conv=0' | FileCheck %s --check-prefixes=CHECK,NVVM
+// RUN: mlir-opt %s -split-input-file -convert-gpu-to-nvvm='use-bare-ptr-memref-call-conv=0' | FileCheck %s --check-prefixes=CHECK,NVVM
 
 gpu.module @kernel {
   gpu.func @test_func_readonly(%arg0 : memref<f32> {llvm.readonly} ) {
diff --git a/mlir/test/Conversion/GPUCommon/memref-arg-noalias-attrs.mlir b/mlir/test/Conversion/GPUCommon/memref-arg-noalias-attrs.mlir
index 64c2cfdbcc5ef29..81babac5502c759 100644
--- a/mlir/test/Conversion/GPUCommon/memref-arg-noalias-attrs.mlir
+++ b/mlir/test/Conversion/GPUCommon/memref-arg-noalias-attrs.mlir
@@ -1,5 +1,5 @@
 // RUN: mlir-opt %s -split-input-file -convert-gpu-to-rocdl='use-opaque-pointers=1 use-bare-ptr-memref-call-conv=1' | FileCheck %s --check-prefixes=CHECK,ROCDL
-// RUN: mlir-opt %s -split-input-file -convert-gpu-to-nvvm='use-opaque-pointers=1 use-bare-ptr-memref-call-conv=1' | FileCheck %s --check-prefixes=CHECK,NVVM
+// RUN: mlir-opt %s -split-input-file -convert-gpu-to-nvvm='use-bare-ptr-memref-call-conv=1' | FileCheck %s --check-prefixes=CHECK,NVVM
 
 gpu.module @kernel {
   gpu.func @func_with_noalias_attr(%arg0 : memref<f32> {llvm.noalias} ) {
diff --git a/mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm-32b.mlir b/mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm-32b.mlir
index 9bdce31261064e5..7b873463a5f98f4 100644
--- a/mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm-32b.mlir
+++ b/mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm-32b.mlir
@@ -1,4 +1,4 @@
-// RUN: mlir-opt %s -convert-gpu-to-nvvm='index-bitwidth=32 use-opaque-pointers=1' -split-input-file | FileCheck %s
+// RUN: mlir-opt %s -convert-gpu-to-nvvm='index-bitwidth=32' -split-input-file | FileCheck %s
 
 // RUN: mlir-opt %s -transform-interpreter | FileCheck %s
 
diff --git a/mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm.mlir b/mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm.mlir
index 92d85a473d2318f..c18bb423a6e6001 100644
--- a/mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm.mlir
+++ b/mlir/test/Conversion/GPUToNVVM/gpu-to-nvvm.mlir
@@ -1,4 +1,4 @@
-// RUN: mlir-opt %s -convert-gpu-to-nvvm='has-redux=1 use-opaque-pointers=1' -split-input-file | FileCheck %s
+// RUN: mlir-opt %s -convert-gpu-to-nvvm='has-redux=1' -split-input-file | FileCheck %s
 // RUN: mlir-opt %s -transform-interpreter | FileCheck %s
 
 gpu.module @test_module_0 {
diff --git a/mlir/test/Conversion/GPUToNVVM/typed-pointers.mlir b/mlir/test/Conversion/GPUToNVVM/typed-pointers.mlir
deleted file mode 100644
index de0a638432e57bd..000000000000000
--- a/mlir/test/Conversion/GPUToNVVM/typed-pointers.mlir
+++ /dev/null
@@ -1,39 +0,0 @@
-// RUN: mlir-opt --convert-gpu-to-nvvm="use-opaque-pointers=0" --split-input-file %s | FileCheck %s
-// RUN: mlir-opt --convert-gpu-to-nvvm="index-bitwidth=32 use-opaque-pointers=0" --split-input-file %s | FileCheck --check-prefix=CHECK32 %s
-
-gpu.module @test_module {
-
-  // CHECK-LABEL: func @gpu_wmma_load_op() ->
-  // CHECK-SAME: !llvm.struct<(vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>)>
-  // CHECK32-LABEL: func @gpu_wmma_load_op() ->
-  func.func @gpu_wmma_load_op() -> (!gpu.mma_matrix<16x16xf16, "AOp">) {
-    %wg = memref.alloca() {alignment = 32} : memref<32x32xf16, 3>
-    %i = arith.constant 16 : index
-    %j = arith.constant 16 : index
-    %0 = gpu.subgroup_mma_load_matrix %wg[%i, %j] {leadDimension = 32 : index, transpose} : memref<32x32xf16, 3> -> !gpu.mma_matrix<16x16xf16, "AOp">
-    // CHECK:  %[[INX:.*]] = llvm.mlir.constant(16 : index) : i64
-    // CHECK: %{{.*}} = llvm.insertvalue %{{.*}}, %{{.*}}[{{.*}}, {{.*}}]
-    // CHECK:  %[[BASE:.*]] = llvm.extractvalue %{{.*}}[1] : !llvm.struct<(ptr<f16, 3>, ptr<f16, 3>, i64, array<2 x i64>, array<2 x i64>)>
-    // CHECK:  %[[LDM:.*]] = llvm.mlir.constant(32 : index) : i64
-    // CHECK:  %[[LI:.*]] = llvm.mul %[[INX]], %[[LDM]]  : i64
-    // CHECK:  %[[LIJ:.*]] = llvm.add %[[LI]], %[[INX]]  : i64
-    // CHECK:  %[[ADDRESS:.*]] = llvm.getelementptr %[[BASE]][%[[LIJ]]] : (!llvm.ptr<f16, 3>, i64) -> !llvm.ptr<f16, 3>
-    // CHECK:  %[[LDM32:.*]] = llvm.mlir.constant(32 : index) : i32
-    // CHECK:  %[[FRAG:.*]] = nvvm.wmma.load %[[ADDRESS]], %[[LDM32]]
-    // CHECK-SAME: {eltype = #nvvm.mma_type<f16>, frag = #nvvm.mma_frag<a>, k = 16 : i32, layout = #nvvm.mma_layout<col>, m = 16 : i32, n = 16 : i32}  : (!llvm.ptr<f16, 3>) -> !llvm.struct<(vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>)>
-    // CHECK:  llvm.return %[[FRAG]] : !llvm.struct<(vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>)>
-
-    // CHECK32:  %[[INX:.*]] = llvm.mlir.constant(16 : index) : i32
-    // CHECK32: %{{.*}} = llvm.insertvalue %{{.*}}, %{{.*}}[{{.*}}, {{.*}}]
-    // CHECK32:  %[[BASE:.*]] = llvm.extractvalue %{{.*}}[1] : !llvm.struct<(ptr<f16, 3>, ptr<f16, 3>, i32, array<2 x i32>, array<2 x i32>)>
-    // CHECK32:  %[[LDM:.*]] = llvm.mlir.constant(32 : index) : i32
-    // CHECK32:  %[[LI:.*]] = llvm.mul %[[INX]], %[[LDM]]  : i32
-    // CHECK32:  %[[LIJ:.*]] = llvm.add %[[LI]], %[[INX]]  : i32
-    // CHECK32:  %[[ADDRESS:.*]] = llvm.getelementptr %[[BASE]][%[[LIJ]]] : (!llvm.ptr<f16, 3>, i32) -> !llvm.ptr<f16, 3>
-    // CHECK32:  %[[LDM32:.*]] = llvm.mlir.constant(32 : index) : i32
-    // CHECK32:  %[[FRAG:.*]] = nvvm.wmma.load %[[ADDRESS]], %[[LDM32]]
-    // CHECK32-SAME: {eltype = #nvvm.mma_type<f16>, frag = #nvvm.mma_frag<a>, k = 16 : i32, layout = #nvvm.mma_layout<col>, m = 16 : i32, n = 16 : i32}  : (!llvm.ptr<f16, 3>) -> !llvm.struct<(vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>)>
-    // CHECK32:  llvm.return %[[FRAG]] : !llvm.struct<(vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>, vector<2xf16>)>
-    return %0 : !gpu.mma_matrix<16x16xf16, "AOp">
-  }
-}
diff --git a/mlir/test/Conversion/GPUToNVVM/wmma-ops-to-nvvm.mlir b/mlir/test/Conversion/GPUToNVVM/wmma-ops-to-nvvm.mlir
index 520d6294bc1c16a..9dec666bf4b3d8f 100644
--- a/mlir/test/Conversion/GPUToNVVM/wmma-ops-to-nvvm.mlir
+++ b/mlir/test/Conversion/GPUToNVVM/wmma-ops-to-nvvm.mlir
@@ -1,5 +1,5 @@
-// RUN: mlir-opt --convert-gpu-to-nvvm='use-opaque-pointers=1' --split-input-file %s | FileCheck %s
-// RUN: mlir-opt --convert-gpu-to-nvvm="index-bitwidth=32 use-opaque-pointers=1" --split-input-file %s | FileCheck --check-prefix=CHECK32 %s
+// RUN: mlir-opt --convert-gpu-to-nvvm --split-input-file %s | FileCheck %s
+// RUN: mlir-opt --convert-gpu-to-nvvm="index-bitwidth=32" --split-input-file %s | FileCheck --check-prefix=CHECK32 %s
 
 gpu.module @test_module {

zero9178

LGTM

grypp

Thanks, looks good!

Dinistro requested review from grypp, zero9178 and nicolasvasilache October 31, 2023 21:33

Dinistro added the mlir:gpu label Oct 31, 2023

llvmbot added the mlir label Oct 31, 2023

zero9178 approved these changes Oct 31, 2023

View reviewed changes

grypp approved these changes Nov 1, 2023

View reviewed changes

Dinistro merged commit 02307a1 into llvm:main Nov 1, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[MLIR][GPUToNVVM] Remove typed pointer support #70861

[MLIR][GPUToNVVM] Remove typed pointer support #70861

Uh oh!

Dinistro commented Oct 31, 2023

Uh oh!

llvmbot commented Oct 31, 2023 •

edited

Loading

Uh oh!

zero9178 left a comment

Uh oh!

grypp left a comment

Uh oh!

Uh oh!

[MLIR][GPUToNVVM] Remove typed pointer support #70861

[MLIR][GPUToNVVM] Remove typed pointer support #70861

Uh oh!

Conversation

Dinistro commented Oct 31, 2023

Uh oh!

llvmbot commented Oct 31, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

zero9178 left a comment

Choose a reason for hiding this comment

Uh oh!

grypp left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

llvmbot commented Oct 31, 2023 •

edited

Loading