Commits

Branch selector

 master

User selector

alex-t

Datepicker

All time

Commit History

Commits on Dec 1, 2017

• [AMDGPU] SiFixSGPRCopies should not modify non-divergent PHI

  Show description for b0efc4f
  alex-t

  committed
  b0efc4f

Commits on Nov 10, 2017

• [AMDGPU] Prevent Machine Copy Propagation from replacing live copy with the dead one

  Show description for 23476c4
  alex-t

  committed
  23476c4

Commits on Oct 16, 2017

• [AMDGPU] : revert r315908

  Show description for f6eb7ff
  alex-t

  committed
  f6eb7ff

• [AMDGPU] Prevent Machine Copy Propagation from replacing live copy with the dead one

  Show description for 7811640
  alex-t

  committed
  7811640

Commits on Oct 3, 2017

• [AMDGPU] Avoid predicated execution of the basic blocks containing scalar

  Show description for 673f1dd
  alex-t

  committed
  673f1dd

Commits on Jul 4, 2017

• [AMDGPU] Switch scalarize global loads ON by default

  Show description for f9e9586
  alex-t

  committed
  f9e9586

Commits on Jul 3, 2017

• [AMDGPU] Switch scalarize global loads ON by default

  Show description for 0f9ec97
  alex-t

  committed
  0f9ec97

Commits on Jun 15, 2017

• DivergencyAnalysis patch for review

  Show description for 7807f69
  alex-t

  committed
  7807f69

Commits on Jun 2, 2017

• AMDGPUAnnotateUniformValue should always treat volatile loads as divergent

  Show description for dfdb788
  alex-t

  committed
  dfdb788

Commits on Feb 14, 2017

• Revert "[AMDGPU] Fix for SIMachineScheduler crash. SI Scheduler should track"

  Show description for 23db8ab
  alex-t

  committed
  23db8ab

Commits on Feb 7, 2017

• [AMDGPU] Fix for SIMachineScheduler crash. SI Scheduler should track

  Show description for ce06d9c
  alex-t

  committed
  ce06d9c

Commits on Dec 15, 2016

• Fix for regression after Global Load Scalarization patch

  Show description for 82c3914
  alex-t

  committed
  82c3914

Commits on Dec 8, 2016

• [AMDGPU] Scalarization of global uniform loads.

  Show description for 9ae2b48
  alex-t

  committed
  9ae2b48

Commits on Nov 3, 2016

• [AMDGPU][CodeGen] To improve CGEMM performance: combine LDS reads.

  Show description for d767830
  alex-t

  committed
  d767830