projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 597fb7f) | patch
[AArch64] Remove G_VECREDUCE_FADD from selectReduction
authorDavid Green <david.green@arm.com>
Thu, 22 Jun 2023 11:46:54 +0000 (12:46 +0100)
committerDavid Green <david.green@arm.com>
Thu, 22 Jun 2023 11:46:54 +0000 (12:46 +0100)
commit68a09c929003bf6af41162ed9e6dc4713d96a997
treea29e23e2642341061ed3cdcab2f0a6564f84170dtree | snapshot
parent597fb7fb4640a85c7e144c1ee2b2d0d5e1da4c84commit | diff
[AArch64] Remove G_VECREDUCE_FADD from selectReduction

I believe that for fp reductions we can use the imported tablegen patterns for
selection, as opposed to going via selectReduction. Integer reductions are more
difficult, as the return types in selection DAG will be promoted to i32.

Differential Revision: https://reviews.llvm.org/D153244
llvm/include/llvm/Target/GlobalISel/SelectionDAGCompat.td diff | blob | history
llvm/lib/Target/AArch64/GISel/AArch64InstructionSelector.cpp diff | blob | history
llvm/test/CodeGen/AArch64/GlobalISel/select-reduce-fadd.mir diff | blob | history
Domain: System / Toolchain;