llvm/lib/Transforms
Alexey Bataev d6db829b03 [SLP] Fix for PR32036: Vectorized horizontal reduction returning wrong result
Summary:
If the same value is used several times as an extra value, SLP
vectorizer takes it into account only once instead of actual number of
using.
For example:
```
int val = 1;
for (int y = 0; y < 8; y++) {
  for (int x = 0; x < 8; x++) {
    val = val + input[y * 8 + x] + 3;
  }
}
```
We have 2 extra rguments: `1` - initial value of horizontal reduction
and `3`, which is added 8*8 times to the reduction. Before the patch we
added `1` to the reduction value and added once `3`, though it must be
added 64 times.

Reviewers: mkuper, mzolotukhin

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D30262

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@295868 91177308-0d34-0410-b5e6-96231b3b80d8
2017-02-22 20:06:40 +00:00
..
Coroutines
Hello
InstCombine AMDGPU: Add cvt.pkrtz intrinsic 2017-02-22 00:27:34 +00:00
Instrumentation AddressSanitizer: don't track swifterror memory addresses 2017-02-15 20:43:43 +00:00
IPO Increases full-unroll threshold. 2017-02-18 03:46:51 +00:00
ObjCARC
Scalar [LoopUnroll] Enable PGO-based loop peeling by default. 2017-02-22 00:27:34 +00:00
Utils Use const-ref in range-loop for to avoid copying pairs of std::string 2017-02-22 06:34:04 +00:00
Vectorize [SLP] Fix for PR32036: Vectorized horizontal reduction returning wrong result 2017-02-22 20:06:40 +00:00
CMakeLists.txt
LLVMBuild.txt