Commits

Branch selector

 master

User selector

rampitec

Datepicker

All time

Commit History

Commits on Oct 6, 2017

• [AMDGPU] New 64 bit div/rem expansion

  Show description for 287608c
  rampitec

  committed
  287608c

Commits on Oct 2, 2017

• Eliminate ftrunc if source is know to be rounded

  Show description for 319e857
  rampitec

  committed
  319e857

Commits on Sep 29, 2017

• [AMDGPU] Set fast-math flags on functions given the options

  Show description for 8b431a9
  rampitec

  committed
  8b431a9

Commits on Sep 20, 2017

• [AMDGPU] Fixed memory leak with inliner replaced

  Show description for b5a9104
  rampitec

  committed
  b5a9104

• [AMDGPU] Fix regression in test clang/test/CodeGen/backend-unsupported-error.ll

  Show description for 2e5d75b
  rampitec

  committed
  2e5d75b

• [AMDGPU] Port of HSAIL inliner

  Show description for fbf0e16
  rampitec

  committed
  fbf0e16

Commits on Sep 19, 2017

• [AMDGPU] Prevent post-RA scheduler from breaking memory clauses

  Show description for 60873e2
  rampitec

  committed
  60873e2

Commits on Sep 13, 2017

• Allow target to decide when to cluster loads/stores in misched

  Show description for 63c545d
  rampitec

  committed
  63c545d

Commits on Sep 11, 2017

• [AMDGPU] Produce madak and madmk from the two-address pass

  Show description for 46582be
  rampitec

  committed
  46582be

Commits on Sep 6, 2017

• [AMDGPU] Use v_pk_max_f16 for fcanonicalize

  Show description for 6148c30
  rampitec

  committed
  6148c30

• [AMDGPU] Fixed encoding of v_pk_mul_f16 in fcanonicalize

  Show description for 953b703
  rampitec

  committed
  953b703

• [AMDGPU] Fix shouldClusterMemOps to process flat loads

  Show description for 651c4ef
  rampitec

  committed
  651c4ef

Commits on Sep 1, 2017

• [AMDGPU] Testcase for computeKnownBits recursion. NFC.

  Show description for 9910650
  rampitec

  committed
  9910650

• [AMDGPU] Prevent infinite recursion in DAG.computeKnownBits()

  Show description for f3b5f2a
  rampitec

  committed
  f3b5f2a

Commits on Aug 30, 2017

• [AMDGPU] Use v_max_f* for fcanonicalize

  Show description for 0a4c4f2
  rampitec

  committed
  0a4c4f2

Commits on Aug 28, 2017

• [AMDGPU] Fix regression in AMDGPULibCalls allowing native for doubles

  Show description for 9324a77
  rampitec

  committed
  9324a77

• [AMDGPU] computeKnownBitsForTargetNode for 24 bit mul

  Show description for f4dd1bd
  rampitec

  committed
  f4dd1bd

Commits on Aug 16, 2017

• [AMDGPU] Eliminate no effect instructions before s_endpgm

  Show description for 2df3faf
  rampitec

  committed
  2df3faf

Commits on Aug 11, 2017

• [AMDGPU] Fix santizer error after last commit

  Show description for 911c1e4
  rampitec

  committed
  911c1e4

• [AMDGPU] Ported and adopted AMDLibCalls pass

  Show description for 9643e6b
  rampitec

  committed
  9643e6b

Commits on Aug 4, 2017

• [AMDGPU] Preserve inverted bit in SI_IF in presence of SI_KILL

  Show description for 6069fa8
  rampitec

  committed
  6069fa8

Commits on Aug 2, 2017

• [AMDGPU] Fix asan error after last commit

  Show description for dbd873d
  rampitec

  committed
  dbd873d

Commits on Aug 1, 2017

• [AMDGPU] Turn s_and_saveexec_b64 into s_and_b64 if result is unused

  Show description for 3499304
  rampitec

  committed
  3499304

• [AMDGPU] Collapse adjacent SI_END_CF

  Show description for 5b53ac9
  rampitec

  committed
  5b53ac9

Commits on Jul 26, 2017

• [AMDGPU] Optimize SI_IF lowering for simple if regions

  Show description for ddb10d2
  rampitec

  committed
  ddb10d2

Commits on Jul 13, 2017

• [AMDGPU] fcaninicalize optimization for GFX9+

  Show description for 9fc15af
  rampitec

  committed
  9fc15af

Commits on Jul 12, 2017

• [AMDGPU] fcanonicalize elimination optimization

  Show description for 16be511
  rampitec

  committed
  16be511

Commits on Jul 6, 2017

• [AMDGPU] Always use rcp + mul with fast math

  Show description for 71b4fe4
  rampitec

  committed
  71b4fe4

Commits on Jun 28, 2017

• Fold fneg and fabs like multiplications

  Show description for a143b4a
  rampitec

  committed
  a143b4a

• [AMDGPU] Add pattern for v_alignbit_b32 with immediate

  Show description for 8b38a13
  rampitec

  committed
  8b38a13

• Allow to truncate left shift with non-constant shift amount

  Show description for a5e3faf
  rampitec

  committed
  a5e3faf

Commits on Jun 27, 2017

• [AMDGPU] Add 2 new alignbit patterns

  Show description for 040f338
  rampitec

  committed
  040f338

• [AMDGPU] Simplify setcc (sext from i1 b), -1|0, cc

  Show description for e764e24
  rampitec

  committed
  e764e24

• [AMDGPU] Combine and x, (sext cc from i1) => select cc, x, 0

  Show description for e2d9355
  rampitec

  committed
  e2d9355

Commits on Jun 21, 2017

• [AMDGPU] Add FP_CLASS to the add/setcc combine

  Show description for 91cd127
  rampitec

  committed
  91cd127

Pagination

PreviousNext