[Lldb-commits] [llvm] [lldb] [mlir] [mlir] Introduce replaceWithZeroTripCheck in LoopLikeOpInterface (PR #80331)

2024-02-05 Thread Diego Caballero via lldb-commits
@@ -220,6 +220,31 @@ def LoopLikeOpInterface : OpInterface<"LoopLikeOpInterface"> { /*defaultImplementation=*/[{ return ::mlir::failure(); }] +>, +InterfaceMethod<[{ +Add a zero-trip-check around the loop to check if the loop body is ever --

[Lldb-commits] [llvm] [mlir] [lldb] [mlir] Introduce replaceWithZeroTripCheck in LoopLikeOpInterface (PR #80331)

2024-02-05 Thread Diego Caballero via lldb-commits
@@ -220,6 +220,31 @@ def LoopLikeOpInterface : OpInterface<"LoopLikeOpInterface"> { /*defaultImplementation=*/[{ return ::mlir::failure(); }] +>, +InterfaceMethod<[{ +Add a zero-trip-check around the loop to check if the loop body is ever +

[Lldb-commits] [llvm] [lldb] [mlir] [mlir] Introduce replaceWithZeroTripCheck in LoopLikeOpInterface (PR #80331)

2024-02-01 Thread Diego Caballero via lldb-commits
@@ -220,6 +220,28 @@ def LoopLikeOpInterface : OpInterface<"LoopLikeOpInterface"> { /*defaultImplementation=*/[{ return ::mlir::failure(); }] +>, +InterfaceMethod<[{ +Add a zero-trip-check around the loop to check if the loop body is ever +

[Lldb-commits] [mlir] [openmp] [lldb] [flang] [llvm] [mlir][Vector] Add patterns for efficient i4 -> i8 conversion emulation (PR #79494)

2024-01-29 Thread Diego Caballero via lldb-commits
https://github.com/dcaballe closed https://github.com/llvm/llvm-project/pull/79494 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits

[Lldb-commits] [flang] [mlir] [llvm] [openmp] [lldb] [mlir][Vector] Add patterns for efficient i4 -> i8 conversion emulation (PR #79494)

2024-01-26 Thread Diego Caballero via lldb-commits
dcaballe wrote: Thanks for the info! I think making the interleave op at Vector level available to fixed vectors would also make sense. There is a point in knowing that a shuffle is actually implementing an interleave pattern. I guess we should also be fine with this LLVM limitations for now:

[Lldb-commits] [llvm] [openmp] [flang] [mlir] [lldb] [mlir][Vector] Add patterns for efficient i4 -> i8 conversion emulation (PR #79494)

2024-01-25 Thread Diego Caballero via lldb-commits
https://github.com/dcaballe updated https://github.com/llvm/llvm-project/pull/79494 >From b8fb65dd1e65c36cfb2104e5f35179faa6011552 Mon Sep 17 00:00:00 2001 From: Diego Caballero Date: Thu, 25 Jan 2024 02:39:14 + Subject: [PATCH] [mlir][Vector] Add patterns for efficient i4 -> i8 conversion

[Lldb-commits] [lldb] [mlir] Add ContractionOpInterface utility functions for vector matrix multiplication (PR #68945)

2023-10-16 Thread Diego Caballero via lldb-commits
@@ -96,6 +96,79 @@ bool mlir::isRowMajorBatchMatmul(ArrayAttr indexingMaps) { return indexingMaps == maps; } +bool mlir::isVecmat(ArrayAttr indexingMaps) { + if (indexingMaps.size() != 3) +return false; + auto map0 = cast(indexingMaps[0]).getValue(); d

[Lldb-commits] [lldb] [mlir] Add ContractionOpInterface utility functions for vector matrix multiplication (PR #68945)

2023-10-16 Thread Diego Caballero via lldb-commits
https://github.com/dcaballe commented: Awesome! Just a couple of minor comments from my side! https://github.com/llvm/llvm-project/pull/68945 ___ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb

[Lldb-commits] [lldb] [mlir] Add ContractionOpInterface utility functions for vector matrix multiplication (PR #68945)

2023-10-16 Thread Diego Caballero via lldb-commits
@@ -86,6 +86,39 @@ def LinalgContractionOpInterface : OpInterface<"ContractionOpInterface"> { /*methodBody=*/[{ return mlir::isRowMajorBatchMatmul($_op.getIndexingMaps()); }]>, +InterfaceMethod< +/*desc=*/[{ + Returns whether the given op has index