Commits · 64634c36dd8770ca85bf09137ce3a453dcd39b1f · Roger Ferrer / llvm-epi-0.8 · GitLab

This is an archived project. Repository and other project resources are read-only.

Nov 03, 2010
- Two sets of changes. Sorry they are intermingled. · debf9c50
  Evan Cheng authored Nov 03, 2010
  debf9c50
Oct 29, 2010
- Avoiding overly aggressive latency scheduling. If the two nodes share an · 6c1414f9
  Evan Cheng authored Oct 29, 2010
  6c1414f9
Oct 28, 2010
- Re-commit 117518 and 117519 now that ARM MC test failures are out of the way. · ff310737
  Evan Cheng authored Oct 28, 2010
  ff310737
- Revert 117518 and 117519 for now. They changed scheduling and cause MC tests to fail. Ugh. · e2c211c1
  Evan Cheng authored Oct 28, 2010
  e2c211c1
- Fix a major bug in operand latency computation. The use index must be adjusted · 523fa3a2
  Evan Cheng authored Oct 28, 2010
  523fa3a2
Oct 06, 2010
- - Add TargetInstrInfo::getOperandLatency() to compute operand latencies. This · 49d4c0bd
  Evan Cheng authored Oct 06, 2010
  49d4c0bd
Sep 30, 2010
- Model Cortex-a9 load to SUB, RSB, ADD, ADC, SBC, RSC, CMN, MVN, or CMP · 4a010fd1
  Evan Cheng authored Sep 29, 2010
  4a010fd1
Sep 10, 2010
- Teach if-converter to be more careful with predicating instructions that would · bf407075
  Evan Cheng authored Sep 10, 2010
  bf407075
Aug 10, 2010
- Add missing null check reported by Amaury Pouly. · 23ef8290
  Evan Cheng authored Aug 10, 2010
  23ef8290
Jul 11, 2010
- Fix a bug in the code which re-inserts DBG_VALUE nodes after scheduling; · a64a3235
  Dan Gohman authored Jul 10, 2010
  a64a3235
Jul 10, 2010
- Reapply bottom-up fast-isel, with several fixes for x86-32: · d7b5ce33
  Dan Gohman authored Jul 10, 2010
  d7b5ce33
Jul 09, 2010
- --- Reverse-merging r107947 into '.': · 6586e9b2
  Bob Wilson authored Jul 09, 2010
  6586e9b2
- Re-apply bottom-up fast-isel, with fixes. Be very careful to avoid emitting · 0b5aa1cd
  Dan Gohman authored Jul 09, 2010
  0b5aa1cd
Jun 30, 2010
- grammar tweak in comment. · caf9b3ab
  Jim Grosbach authored Jun 30, 2010
  caf9b3ab
Jun 29, 2010
- Add a VT argument to getMinimalPhysRegClass and replace the copy related uses · 38a7d7cb
  Rafael Espindola authored Jun 29, 2010
  38a7d7cb
Jun 25, 2010
- Remove variables which are assigned to but for which the value · 2dc70bea
  Duncan Sands authored Jun 25, 2010
  2dc70bea
- It's possible that a flag is added to the SDNode that points back to the · 2d3c4900
  Bill Wendling authored Jun 24, 2010
  2d3c4900
Jun 23, 2010
- MorphNodeTo doesn't preserve the memory operands. Because we're morphing a node · a136521a
  Bill Wendling authored Jun 23, 2010
  a136521a
Jun 21, 2010
- Use A.append(...) instead of A.insert(A.end(), ...) when A is a · dd41bba5
  Dan Gohman authored Jun 21, 2010
  dd41bba5
Jun 10, 2010
- Code refactoring, no functionality changes. · 38f65604
  Evan Cheng authored Jun 10, 2010
  38f65604
May 29, 2010
- Fix some latency computation bugs: if the use is not a machine opcode do not just return zero. · cc2efe11
  Evan Cheng authored May 28, 2010
  cc2efe11
May 21, 2010
- Allow targets more controls on what nodes are scheduled by reg pressure, what... · 4401f887
  Evan Cheng authored May 20, 2010
  4401f887
May 20, 2010
- Add a hybrid bottom up scheduler that reduce register usage while avoiding · bdd062da
  Evan Cheng authored May 20, 2010
  bdd062da
- Code clean up. · 70e506e1
  Evan Cheng authored May 19, 2010
  70e506e1
May 01, 2010
- Get rid of the EdgeMapping map. Instead, just check for BasicBlock · 25c16537
  Dan Gohman authored May 01, 2010
  25c16537
Apr 30, 2010
- EmitDbgValue doesn't need its EdgeMapping argument. · 8acc8f7d
  Dan Gohman authored Apr 30, 2010
  8acc8f7d
Apr 26, 2010
- Add DBG_VALUE handling for byval parameters; this · e098352e
  Dale Johannesen authored Apr 26, 2010
  e098352e
- - Move TargetLowering::EmitTargetCodeForFrameDebugValue to TargetInstrInfo and... · ed69b382
  Evan Cheng authored Apr 26, 2010
  ed69b382
Apr 17, 2010
- Fix -Wcast-qual warnings. · 1f0f2142
  Dan Gohman authored Apr 17, 2010
  1f0f2142
Mar 25, 2010
- Scheduler assumes SDDbgValue nodes are in source order. That's true currently.... · 1889440b
  Evan Cheng authored Mar 25, 2010
  1889440b
- Remove a fixme that doesn't make sense any more. · 08b3364c
  Evan Cheng authored Mar 25, 2010
  08b3364c
- Change how dbg_value sdnodes are converted into machine instructions. Their... · 563fe3cc
  Evan Cheng authored Mar 25, 2010
  563fe3cc
Mar 14, 2010
- Rename SDDbgValue.h to SDNodeDbgValue.h for consistency. · 00fd0b67
  Evan Cheng authored Mar 14, 2010
  00fd0b67
Mar 10, 2010
- Progress towards shepherding debug info through SelectionDAG. · 49de0607
  Dale Johannesen authored Mar 10, 2010
  49de0607
Feb 24, 2010
- Change the scheduler from adding nodes in allnodes order · df8a8a8c
  Chris Lattner authored Feb 24, 2010
  df8a8a8c
Jan 23, 2010
- Enable pre-regalloc scheduling load clustering by default. · c22893a3
  Evan Cheng authored Jan 22, 2010
  c22893a3
Jan 22, 2010
- Teach pre-regalloc scheduler to schedule loads from nearby addresses. It may... · 9d92aaab
  Evan Cheng authored Jan 22, 2010
  9d92aaab
Jan 05, 2010
- Change errs() to dbgs(). · 4eb5bed6
  David Greene authored Jan 05, 2010
  4eb5bed6
Dec 19, 2009
- Changes from review: · 022d18fa
  Bill Wendling authored Dec 18, 2009
  022d18fa
Dec 16, 2009
- Reapply r91392, it was only unmasking the bug, and since TOT is still broken... · b827e526
  Daniel Dunbar authored Dec 16, 2009
  b827e526