aboutsummaryrefslogtreecommitdiff
path: root/llvm/lib/CodeGen/TargetLoweringObjectFileImpl.cpp
diff options
context:
space:
mode:
authorJim Kitchen <jim22k@gmail.com>2022-03-17 12:16:29 -0500
committerJim Kitchen <jim22k@gmail.com>2022-03-17 12:31:09 -0500
commit414ed019acba67e914583fdc51eb3d1328696114 (patch)
tree197a60feeba6b47a7a9b9a8f269a2a7be69752c7 /llvm/lib/CodeGen/TargetLoweringObjectFileImpl.cpp
parentf496330f9719ddce4b9e769c7fa49d0fcad643b8 (diff)
downloadllvm-414ed019acba67e914583fdc51eb3d1328696114.zip
llvm-414ed019acba67e914583fdc51eb3d1328696114.tar.gz
llvm-414ed019acba67e914583fdc51eb3d1328696114.tar.bz2
[mlir][sparse] Introduce new binary and unary op
When the sparse_tensor dialect lowers linalg.generic, it makes inferences about how the operations should affect the looping logic. For example, multiplication is an intersection while addition is a union of two sparse tensors. The new binary and unary op separate the looping logic from the computation by nesting the computation code inside a block which is merged at the appropriate level in the lowered looping code. The binary op can have custom computation code for the overlap, left, and right sparse overlap regions. The unary op can have custom computation code for the present and absent values. Reviewed by: aartbik Differential Revision: https://reviews.llvm.org/D121018
Diffstat (limited to 'llvm/lib/CodeGen/TargetLoweringObjectFileImpl.cpp')
0 files changed, 0 insertions, 0 deletions