[Mlir-commits] [mlir] 982c815 - [mlir][sparse] fix mismatch between `enter/exitWhileLoop` (#79493)

llvmlistbot at llvm.org llvmlistbot at llvm.org
Thu Jan 25 12:21:51 PST 2024


Author: Peiming Liu
Date: 2024-01-25T12:21:47-08:00
New Revision: 982c815aad97686fdb20e4599efbcc064599747b

URL: https://github.com/llvm/llvm-project/commit/982c815aad97686fdb20e4599efbcc064599747b
DIFF: https://github.com/llvm/llvm-project/commit/982c815aad97686fdb20e4599efbcc064599747b.diff

LOG: [mlir][sparse] fix mismatch between `enter/exitWhileLoop` (#79493)

Added: 
    

Modified: 
    mlir/lib/Dialect/SparseTensor/Transforms/Utils/LoopEmitter.cpp
    mlir/test/Integration/Dialect/SparseTensor/CPU/reshape_dot.mlir

Removed: 
    


################################################################################
diff  --git a/mlir/lib/Dialect/SparseTensor/Transforms/Utils/LoopEmitter.cpp b/mlir/lib/Dialect/SparseTensor/Transforms/Utils/LoopEmitter.cpp
index 0ce6a9efce1c81a..6d832fdc0c2201e 100644
--- a/mlir/lib/Dialect/SparseTensor/Transforms/Utils/LoopEmitter.cpp
+++ b/mlir/lib/Dialect/SparseTensor/Transforms/Utils/LoopEmitter.cpp
@@ -599,7 +599,7 @@ Operation *LoopEmitter::enterCoIterationOverTensorsAtLvls(
 
   // NOTE: we can also prepare for next dim here in advance
   // Pushes the loop into stack.
-  loopStack.emplace_back(tidLvls, l, builder.getInsertionBlock(), iv, loopTag);
+  loopStack.emplace_back(tls, l, builder.getInsertionBlock(), iv, loopTag);
   return l;
 }
 

diff  --git a/mlir/test/Integration/Dialect/SparseTensor/CPU/reshape_dot.mlir b/mlir/test/Integration/Dialect/SparseTensor/CPU/reshape_dot.mlir
index 15427a373b10b97..689428c23f7d789 100644
--- a/mlir/test/Integration/Dialect/SparseTensor/CPU/reshape_dot.mlir
+++ b/mlir/test/Integration/Dialect/SparseTensor/CPU/reshape_dot.mlir
@@ -17,8 +17,6 @@
 // DEFINE: %{env} =
 //--------------------------------------------------------------------------------------------------
 
-// UNSUPPORTED: target={{.*}}
-
 // RUN: %{compile} | %{env} %{run} | FileCheck %s
 //
 // Do the same run, but now with direct IR generation.


        


More information about the Mlir-commits mailing list