diff options
| author | jacquesguan <Jianjian.Guan@streamcomputing.com> | 2022-07-18 17:37:37 +0800 |
|---|---|---|
| committer | jacquesguan <Jianjian.Guan@streamcomputing.com> | 2022-08-04 15:31:31 +0800 |
| commit | b61cfc91eac83dec7ded1ef1e6d9f9776751d2c0 (patch) | |
| tree | 41c143c1a4d8a83724829fe081d64281e0eb3897 /clang/lib/CodeGen/CodeGenFunction.h | |
| parent | 6f867f9102838ebe314c1f3661fdf95700386e5a (diff) | |
| download | llvm-b61cfc91eac83dec7ded1ef1e6d9f9776751d2c0.zip llvm-b61cfc91eac83dec7ded1ef1e6d9f9776751d2c0.tar.gz llvm-b61cfc91eac83dec7ded1ef1e6d9f9776751d2c0.tar.bz2 | |
[RISCV] Add cost modelling for vector widenning reduction.
In RVV, we use vwredsum.vs and vwredsumu.vs for vecreduce.add(ext(Ty A)) if the result type's width is twice of the input vector's SEW-width. In this situation, the cost of extended add reduction should be same as single-width add reduction. So as the vector float widenning reduction.
Differential Revision: https://reviews.llvm.org/D129994
Diffstat (limited to 'clang/lib/CodeGen/CodeGenFunction.h')
0 files changed, 0 insertions, 0 deletions