[Mlir-commits] [mlir] [mlir][memref] Fix emulate narrow types for strided memref offset (PR #68181)

Kunwar Grover llvmlistbot at llvm.org
Wed Oct 4 22:18:31 PDT 2023


https://github.com/Groverkss updated https://github.com/llvm/llvm-project/pull/68181

>From 4697178ab883acb96c65d7dedff1a507f24613ec Mon Sep 17 00:00:00 2001
From: Groverkss <groverkss at gmail.com>
Date: Wed, 4 Oct 2023 08:56:54 +0530
Subject: [PATCH 1/4] [mlir][memref] Fix emulate narrow types for strided
 memref offset

---
 .../MemRef/Transforms/EmulateNarrowType.cpp     | 17 +++++++++++++++--
 1 file changed, 15 insertions(+), 2 deletions(-)

diff --git a/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp b/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
index 2a524ceb9db887b..2a9b27debaece3f 100644
--- a/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
+++ b/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
@@ -271,9 +271,22 @@ void memref::populateMemRefNarrowTypeEmulationConversions(
           return std::nullopt;
 
         StridedLayoutAttr layoutAttr;
+        // If the offset is 0, we do not need a strided layout as the stride is
+        // 1, so we only use the strided layout if the offset is not 0.
         if (offset != 0) {
-          layoutAttr = StridedLayoutAttr::get(ty.getContext(), offset,
-                                              ArrayRef<int64_t>{1});
+          if (offset == ShapedType::kDynamic) {
+            layoutAttr = StridedLayoutAttr::get(ty.getContext(), offset,
+                                                ArrayRef<int64_t>{1});
+          } else {
+            // Check if the number of bytes are a multiple of the loadStoreWidth
+            // and if so, divide it by the loadStoreWidth to get the offset.
+            if ((offset * width) % loadStoreWidth != 0)
+              return std::nullopt;
+            offset = (offset * width) / loadStoreWidth;
+
+            layoutAttr = StridedLayoutAttr::get(ty.getContext(), offset,
+                                                ArrayRef<int64_t>{1});
+          }
         }
 
         return MemRefType::get(getLinearizedShape(ty, width, loadStoreWidth),

>From f9b25612e2389980166523edb774a807c2228c3a Mon Sep 17 00:00:00 2001
From: Groverkss <groverkss at gmail.com>
Date: Wed, 4 Oct 2023 11:19:55 +0530
Subject: [PATCH 2/4] Add tests and memref.subview support in emulate narrow
 types

---
 .../MemRef/Transforms/EmulateNarrowType.cpp   | 74 ++++++++++++++++++-
 .../Dialect/MemRef/emulate-narrow-type.mlir   | 19 +++++
 2 files changed, 90 insertions(+), 3 deletions(-)

diff --git a/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp b/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
index 2a9b27debaece3f..453a18ff3c254e6 100644
--- a/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
+++ b/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
@@ -209,6 +209,74 @@ struct ConvertMemRefLoad final : OpConversionPattern<memref::LoadOp> {
     return success();
   }
 };
+
+//===----------------------------------------------------------------------===//
+// ConvertMemRefAssumeAlignment
+//===----------------------------------------------------------------------===//
+
+struct ConvertMemRefSubview final : OpConversionPattern<memref::SubViewOp> {
+  using OpConversionPattern::OpConversionPattern;
+
+  LogicalResult
+  matchAndRewrite(memref::SubViewOp op, OpAdaptor adaptor,
+                  ConversionPatternRewriter &rewriter) const override {
+    auto convertedType =
+        cast<MemRefType>(getTypeConverter()->convertType(op.getSourceType()));
+    auto convertedElementType = convertedType.getElementType();
+    auto oldElementType = op.getSourceType().getElementType();
+    int srcBits = oldElementType.getIntOrFloatBitWidth();
+    int dstBits = convertedElementType.getIntOrFloatBitWidth();
+    if (dstBits % srcBits != 0) {
+      return rewriter.notifyMatchFailure(
+          op, "only dstBits % srcBits == 0 supported");
+    }
+
+    MemRefType newTy =
+        cast<MemRefType>(getTypeConverter()->convertType(op.getType()));
+    if (!newTy) {
+      return rewriter.notifyMatchFailure(
+          op->getLoc(),
+          llvm::formatv("failed to convert memref type: {0}", op.getType()));
+    }
+
+    // Only support offset for 1-D subview.
+    if (op.getType().getRank() != 1) {
+      return rewriter.notifyMatchFailure(
+          op->getLoc(), "subview with rank > 1 is not supported");
+    }
+
+    // Only support stride of 1.
+    if (op.getStaticStride(0) != 1) {
+      return rewriter.notifyMatchFailure(
+          op->getLoc(), "subview with stride != 1 is not supported");
+    }
+
+    auto size = op.getStaticSize(0);
+    auto offset = op.getStaticOffset(0);
+    // Only support static sizes and offsets.
+    if (size == ShapedType::kDynamic || offset == ShapedType::kDynamic) {
+      return rewriter.notifyMatchFailure(
+          op->getLoc(), "subview with dynamic size or offset is not supported");
+    }
+
+    int elementsPerByte = dstBits / srcBits;
+    if (size % elementsPerByte != 0 || offset % elementsPerByte != 0) {
+      return rewriter.notifyMatchFailure(
+          op->getLoc(),
+          "subview with size or offset not multiple of elementsPerByte is not "
+          "supported");
+    }
+
+    size = size / elementsPerByte;
+    offset = offset / elementsPerByte;
+
+    rewriter.replaceOpWithNewOp<memref::SubViewOp>(
+        op, newTy, *adaptor.getODSOperands(0).begin(), offset, size,
+        op.getStaticStrides());
+    return success();
+  }
+};
+
 } // end anonymous namespace
 
 //===----------------------------------------------------------------------===//
@@ -220,9 +288,9 @@ void memref::populateMemRefNarrowTypeEmulationPatterns(
     RewritePatternSet &patterns) {
 
   // Populate `memref.*` conversion patterns.
-  patterns
-      .add<ConvertMemRefAlloc, ConvertMemRefLoad, ConvertMemRefAssumeAlignment>(
-          typeConverter, patterns.getContext());
+  patterns.add<ConvertMemRefAlloc, ConvertMemRefLoad,
+               ConvertMemRefAssumeAlignment, ConvertMemRefSubview>(
+      typeConverter, patterns.getContext());
   memref::populateResolveExtractStridedMetadataPatterns(patterns);
 }
 
diff --git a/mlir/test/Dialect/MemRef/emulate-narrow-type.mlir b/mlir/test/Dialect/MemRef/emulate-narrow-type.mlir
index c0050d8c510d53f..6ed97f05aa7cff2 100644
--- a/mlir/test/Dialect/MemRef/emulate-narrow-type.mlir
+++ b/mlir/test/Dialect/MemRef/emulate-narrow-type.mlir
@@ -155,3 +155,22 @@ func.func @rank_zero_memref() -> i4 {
 //       CHECK32:   %[[LOAD:.+]] = memref.load %[[ALLOC]][] : memref<i32>
 //       CHECK32:   %[[TRUNC:.+]] = arith.trunci %[[LOAD]] : i32 to i4
 //       CHECK32:   return %[[TRUNC]]
+
+// -----
+
+func.func @memref_strided_i4(%idx : index) -> i4 {
+  %arr = memref.alloc() : memref<128xi4>
+  %subview = memref.subview %arr[32] [32] [1] : memref<128xi4> to memref<32xi4, strided<[1], offset:32>>
+  %1 = memref.load %subview[%idx] : memref<32xi4, strided<[1], offset:32>>
+  return %1 : i4
+}
+
+// CHECK-LABEL: func @memref_strided_i4
+//       CHECK:   %[[ALLOC:.+]] = memref.alloc() : memref<64xi8>
+//       CHECK:   %[[SUBVIEW:.+]] = memref.subview %[[ALLOC]][16] [16] [1] : memref<64xi8> to memref<16xi8, strided<[1], offset: 16>>
+//       CHECK:   %[[LOAD:.+]] = memref.load %[[SUBVIEW]]
+
+// CHECK32-LABEL: func @memref_strided_i4
+//       CHECK32:   %[[ALLOC:.+]] = memref.alloc() : memref<16xi32>
+//       CHECK32:   %[[SUBVIEW:.+]] = memref.subview %[[ALLOC]][4] [4] [1] : memref<16xi32> to memref<4xi32, strided<[1], offset: 4>>
+//       CHECK32:   %[[LOAD:.+]] = memref.load %[[SUBVIEW]]

>From 3c3cd037829a6ed79375ae6b7d76a47dc919a58f Mon Sep 17 00:00:00 2001
From: Groverkss <groverkss at gmail.com>
Date: Wed, 4 Oct 2023 11:28:53 +0530
Subject: [PATCH 3/4] Fix doc

---
 mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp b/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
index 453a18ff3c254e6..c98dda27f6c0b5a 100644
--- a/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
+++ b/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
@@ -211,7 +211,7 @@ struct ConvertMemRefLoad final : OpConversionPattern<memref::LoadOp> {
 };
 
 //===----------------------------------------------------------------------===//
-// ConvertMemRefAssumeAlignment
+// ConvertMemRefSubview
 //===----------------------------------------------------------------------===//
 
 struct ConvertMemRefSubview final : OpConversionPattern<memref::SubViewOp> {

>From 0b1caf93c97e08fcac6bbf6e95680d6adc7a2050 Mon Sep 17 00:00:00 2001
From: Groverkss <groverkss at gmail.com>
Date: Thu, 5 Oct 2023 10:47:56 +0530
Subject: [PATCH 4/4] Address Mahesh's comments

---
 .../MemRef/Transforms/EmulateNarrowType.cpp   | 37 ++++++++++---------
 1 file changed, 20 insertions(+), 17 deletions(-)

diff --git a/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp b/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
index c98dda27f6c0b5a..9f58e9055acadbb 100644
--- a/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
+++ b/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
@@ -17,6 +17,7 @@
 #include "mlir/Dialect/MemRef/Transforms/Transforms.h"
 #include "mlir/Dialect/MemRef/Utils/MemRefUtils.h"
 #include "mlir/Dialect/Vector/IR/VectorOps.h"
+#include "mlir/Support/MathExtras.h"
 #include "mlir/Transforms/DialectConversion.h"
 #include "llvm/Support/FormatVariadic.h"
 #include "llvm/Support/MathExtras.h"
@@ -214,31 +215,33 @@ struct ConvertMemRefLoad final : OpConversionPattern<memref::LoadOp> {
 // ConvertMemRefSubview
 //===----------------------------------------------------------------------===//
 
+/// Emulating narrow ints on subview have limited support, supporting only
+/// static offset and size and stride of 1. Ideally, the subview should be
+/// folded away before running narrow type emulation, and this pattern would
+/// never run. This pattern is mostly used for testing pruposes.
 struct ConvertMemRefSubview final : OpConversionPattern<memref::SubViewOp> {
   using OpConversionPattern::OpConversionPattern;
 
   LogicalResult
   matchAndRewrite(memref::SubViewOp op, OpAdaptor adaptor,
                   ConversionPatternRewriter &rewriter) const override {
-    auto convertedType =
-        cast<MemRefType>(getTypeConverter()->convertType(op.getSourceType()));
-    auto convertedElementType = convertedType.getElementType();
-    auto oldElementType = op.getSourceType().getElementType();
-    int srcBits = oldElementType.getIntOrFloatBitWidth();
-    int dstBits = convertedElementType.getIntOrFloatBitWidth();
-    if (dstBits % srcBits != 0) {
-      return rewriter.notifyMatchFailure(
-          op, "only dstBits % srcBits == 0 supported");
-    }
-
     MemRefType newTy =
-        cast<MemRefType>(getTypeConverter()->convertType(op.getType()));
+        dyn_cast<MemRefType>(getTypeConverter()->convertType(op.getType()));
     if (!newTy) {
       return rewriter.notifyMatchFailure(
           op->getLoc(),
           llvm::formatv("failed to convert memref type: {0}", op.getType()));
     }
 
+    auto convertedElementType = newTy.getElementType();
+    auto oldElementType = op.getType().getElementType();
+    int srcBits = oldElementType.getIntOrFloatBitWidth();
+    int dstBits = convertedElementType.getIntOrFloatBitWidth();
+    if (dstBits % srcBits != 0) {
+      return rewriter.notifyMatchFailure(
+          op, "only dstBits % srcBits == 0 supported");
+    }
+
     // Only support offset for 1-D subview.
     if (op.getType().getRank() != 1) {
       return rewriter.notifyMatchFailure(
@@ -251,8 +254,8 @@ struct ConvertMemRefSubview final : OpConversionPattern<memref::SubViewOp> {
           op->getLoc(), "subview with stride != 1 is not supported");
     }
 
-    auto size = op.getStaticSize(0);
-    auto offset = op.getStaticOffset(0);
+    int64_t size = op.getStaticSize(0);
+    int64_t offset = op.getStaticOffset(0);
     // Only support static sizes and offsets.
     if (size == ShapedType::kDynamic || offset == ShapedType::kDynamic) {
       return rewriter.notifyMatchFailure(
@@ -260,14 +263,14 @@ struct ConvertMemRefSubview final : OpConversionPattern<memref::SubViewOp> {
     }
 
     int elementsPerByte = dstBits / srcBits;
-    if (size % elementsPerByte != 0 || offset % elementsPerByte != 0) {
+    if (offset % elementsPerByte != 0) {
       return rewriter.notifyMatchFailure(
           op->getLoc(),
-          "subview with size or offset not multiple of elementsPerByte is not "
+          "subview with offset not multiple of elementsPerByte is not "
           "supported");
     }
 
-    size = size / elementsPerByte;
+    size = ceilDiv(size, elementsPerByte);
     offset = offset / elementsPerByte;
 
     rewriter.replaceOpWithNewOp<memref::SubViewOp>(



More information about the Mlir-commits mailing list