Files
clang-p2996/llvm/lib/Target/AMDGPU/AMDGPUPerfHintAnalysis.cpp
Stanislav Mekhanoshin a397c1c82f [AMDGPU] Tune perfhint analysis to account access width
A function with less memory instructions but wider access
is the same as a function with more but narrower accesses
in terms of memory boundness. In fact the pass would give
different answers before and after vectorization without
this change.

Differential Revision: https://reviews.llvm.org/D105651
2021-07-21 12:46:10 -07:00

12 KiB