diff options
| author | Guozhi Wei <carrot@google.com> | 2019-11-01 14:59:08 -0700 |
|---|---|---|
| committer | Guozhi Wei <carrot@google.com> | 2019-11-01 14:59:08 -0700 |
| commit | cecc0d27ad58c0aed8ef9ed99bbf691e137a0f26 (patch) | |
| tree | 34456858eb733680ce3ba8ad2ea8de8e91918c4c /llvm/lib/Target/AMDGPU/R600ControlFlowFinalizer.cpp | |
| parent | 89b7f16204afa25325c28edeaa9973853f2e6796 (diff) | |
| download | bcm5719-llvm-cecc0d27ad58c0aed8ef9ed99bbf691e137a0f26.tar.gz bcm5719-llvm-cecc0d27ad58c0aed8ef9ed99bbf691e137a0f26.zip | |
[NewPM] Add an SROA pass after loop unroll
If there is a small local array accessed in a loop, SROA can't handle memory
accesses with variant offset inside a loop, after the loop is fully unrolled,
all memory accesses to the array are with fixed offset, so now they can be
processed by SROA. But there is no more SROA passes after loop unroll. This
patch add an SROA pass after loop unroll to handle this pattern.
Differential Revision: https://reviews.llvm.org/D68593
Diffstat (limited to 'llvm/lib/Target/AMDGPU/R600ControlFlowFinalizer.cpp')
0 files changed, 0 insertions, 0 deletions
