Commits · c2266463377a1727f0627fe4db9a88a1b33a9f4d · llvm-doe / llvm-project

This project is mirrored from https://github.com/llvm-doe-org/llvm-project.git. Pull mirroring updated Jan 29, 2023.

24 Jan, 2020 1 commit

Resubmit: [DA][TTI][AMDGPU] Add option to select GPUDA with TTI · c2266463

Austin Kerbow authored Jan 20, 2020

Summary:
Enable the new diveregence analysis by default for AMDGPU.

Resubmit with test updates since GPUDA was causing failures on Windows.

Reviewers: rampitec, nhaehnle, arsenm, thakis

Subscribers: kzhuravl, jvesely, wdng, yaxunl, dstuttard, tpr, t-tye, hiraditya, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D73315

c2266463

30 Aug, 2018 1 commit

[NFC] Rename the DivergenceAnalysis to LegacyDivergenceAnalysis · 35617ed4

Nicolai Haehnle authored Aug 30, 2018

Summary:
This is patch 1 of the new DivergenceAnalysis (https://reviews.llvm.org/D50433).

The purpose of this patch is to free up the name DivergenceAnalysis for the new generic
implementation. The generic implementation class will be shared by specialized
divergence analysis classes.

Patch by: Simon Moll

Reviewed By: nhaehnle

Subscribers: jvesely, jholewinski, arsenm, nhaehnle, mgorny, jfb, llvm-commits

Differential Revision: https://reviews.llvm.org/D50434

Change-Id: Ie8146b11be2c50d5312f30e11c7a3036a15b48cb
llvm-svn: 341071

35617ed4

30 Mar, 2017 1 commit
- AMDGPU: Add all atomicrmw fields to atomic.inc/dec · 79f837c2
  Matt Arsenault authored Mar 30, 2017
```
Add scope, order, isVolatile

llvm-svn: 299122
```
  79f837c2
30 Jan, 2017 1 commit
- AMDGPU: Fix atomic_inc/atomic_dec + ds_swizzle not being divergent · 41c14995
  Matt Arsenault authored Jan 30, 2017
```
llvm-svn: 293504
```
  41c14995
13 Apr, 2016 1 commit
- AMDGPU: Remove leftover ShaderType attributes in tests · b34eea9c
  Matt Arsenault authored Apr 13, 2016
```
llvm-svn: 266155
```
  b34eea9c
17 Mar, 2016 1 commit

AMDGPU: mark atomic instructions as sources of divergence · 79cad857

Nicolai Haehnle authored Mar 17, 2016

Summary:
As explained by the comment, threads will typically see different values
returned by atomic instructions even if the arguments are equal.

Reviewers: arsenm, tstellarAMD

Subscribers: arsenm, llvm-commits

Differential Revision: http://reviews.llvm.org/D18156

llvm-svn: 263719

79cad857