diff options
author | Yingwei Zheng <dtcxzyw2333@gmail.com> | 2023-05-02 00:06:26 +0800 |
---|---|---|
committer | Yingwei Zheng <dtcxzyw2333@gmail.com> | 2023-05-02 00:28:39 +0800 |
commit | 6d667d4b261e81f325756fdfd5bb43b3b3d2451d (patch) | |
tree | 14a89995cdc74115460abc68323a498025a8e544 /llvm/lib/ProfileData/SampleProfReader.cpp | |
parent | da0089c99ba1507b876cf3d2a205ba108aad65ff (diff) | |
download | llvm-6d667d4b261e81f325756fdfd5bb43b3b3d2451d.zip llvm-6d667d4b261e81f325756fdfd5bb43b3b3d2451d.tar.gz llvm-6d667d4b261e81f325756fdfd5bb43b3b3d2451d.tar.bz2 |
[InstCombine] Combine const GEP chains
This patch reverts rGae739aefd7473517d3f08b5c8d08a66c7f469198 to address performance regressions reported by our [CI](https://github.com/dtcxzyw/llvm-ci/issues/137) after rG2ec1d0f427c7822540352c0c14d057e7bfe4f77b.
For example:
```
define ptr @const_gep_chain(ptr %p, i64 %a) {
%p1 = getelementptr inbounds i8, ptr %p, i64 %a
%p2 = getelementptr inbounds i8, ptr %p1, i64 1
%p3 = getelementptr inbounds i8, ptr %p2, i64 2
%p4 = getelementptr inbounds i8, ptr %p3, i64 3
ret ptr %p4
}
```
The last three GEPs will not be folded since rG2ec1d0f427c7822540352c0c14d057e7bfe4f77b.
I think it is appropriate to remove this code because there is no compile-time regression reported in our benchmarks.
Reviewed By: nikic
Differential Revision: https://reviews.llvm.org/D149240
Diffstat (limited to 'llvm/lib/ProfileData/SampleProfReader.cpp')
0 files changed, 0 insertions, 0 deletions