Commits · ed975232bc929251dd1aeda5d609aa40b0372f5f · Roger Ferrer / llvm-epi · GitLab

Jan 24, 2012
- Revert r148686 (and r148694, a fix to it) due to a serious layering · ed975232
  Chandler Carruth authored Jan 24, 2012
  ed975232
Jan 23, 2012
- Fix PR11829. PostRA LICM was too aggressive. · 20948fab
  Jakob Stoklund Olesen authored Jan 23, 2012
  20948fab
- An option to selectively enable parts of ARM EHABI support. · 482cdc4e
  Evgeniy Stepanov authored Jan 23, 2012
  482cdc4e
Jan 22, 2012
- Add fused multiple+add instructions from VFPv4. · 5482b9f5
  Anton Korobeynikov authored Jan 22, 2012
  5482b9f5
Jan 20, 2012
- ARM vector any_extends need to be selected to vmovl. <rdar://problem/10723651 > · 6c7aaec0
  Bob Wilson authored Jan 20, 2012
  6c7aaec0
- VST2 four-register w/ update pseudos for fixed/register update. · 90f5780f
  Jim Grosbach authored Jan 20, 2012
  90f5780f
Jan 19, 2012
- Emit ARM EHABI unwinding instructions for 3 more Thumb instructions. · 4c7eb477
  Evgeniy Stepanov authored Jan 19, 2012
  4c7eb477
Jan 13, 2012
- test/CodeGen/ARM/test-sharedidx.ll: Fix for -Asserts. · 1b845f71
  NAKAMURA Takumi authored Jan 13, 2012
  1b845f71
- DAGCombine's logic for forming pre- and post- indexed loads / stores were being · fa832633
  Evan Cheng authored Jan 13, 2012
  fa832633
Jan 11, 2012
- ARM Ld/St Optimizer fix. · 642f0f6a
  Andrew Trick authored Jan 11, 2012
  642f0f6a
Jan 10, 2012
- ARM updating VST2 pseudo-lowering fixed vs. register update. · 74ac7d50
  Jim Grosbach authored Jan 10, 2012
  74ac7d50
- Allow machine-cse to look across MBB boundary when cse'ing instructions that · 0be4144a
  Evan Cheng authored Jan 10, 2012
  0be4144a
Jan 09, 2012
- Don't print an unused label before .cfi_endproc. · f28213ca
  Rafael Espindola authored Jan 09, 2012
  f28213ca
Jan 07, 2012
- Don't print a label before .cfi_startproc when we don't need to. This makes · 38241203
  Rafael Espindola authored Jan 07, 2012
  38241203
- Use getRegForValue() to materialize the address of ARM globals. · 8cdce7e6
  Jakob Stoklund Olesen authored Jan 07, 2012
  8cdce7e6
- Added a late machine instruction copy propagation pass. This catches · 00b1a3cd
  Evan Cheng authored Jan 07, 2012
  00b1a3cd
- Use movw+movt in ARMFastISel::ARMMaterializeGV. · 68f034ee
  Jakob Stoklund Olesen authored Jan 07, 2012
  68f034ee
Jan 06, 2012
- Enable aligned NEON spilling by default. · 68a922c0
  Jakob Stoklund Olesen authored Jan 06, 2012
  68a922c0
Jan 05, 2012
- Reapply r146997, "Heed spill slot alignment on ARM." · d110e2a8
  Jakob Stoklund Olesen authored Jan 05, 2012
  d110e2a8
Jan 04, 2012
- Fix more places which should be checking for iOS, not darwin. · 801d98b3
  Evan Cheng authored Jan 04, 2012
  801d98b3
Jan 03, 2012
- Revert r146997, "Heed spill slot alignment on ARM." · 1b7f2a76
  Jakob Stoklund Olesen authored Jan 03, 2012
  1b7f2a76
Dec 26, 2011
- Make sure DAGCombiner doesn't introduce multiple loads from the same memory... · e96286cd
  Eli Friedman authored Dec 26, 2011
  e96286cd
Dec 21, 2011
- Fix a couple of copy-n-paste bugs. Noticed by George Russell! · 7248bda5
  Chad Rosier authored Dec 21, 2011
  7248bda5
- Fix a couple of copy-n-paste bugs. Noticed by George Russell. · dc8a1aae
  Evan Cheng authored Dec 21, 2011
  dc8a1aae
Dec 20, 2011
- Heed spill slot alignment on ARM. · b95c102c
  Jakob Stoklund Olesen authored Dec 20, 2011
  b95c102c
- ARM target code clean up. Check for iOS, not Darwin where it makes sense. · 68132d80
  Evan Cheng authored Dec 20, 2011
  68132d80
- Mark ARM eh_sjlj_dispatchsetup as clobbering all registers. Radar 10567930. · 75f12cc3
  Bob Wilson authored Dec 20, 2011
  75f12cc3
- Move tests to FileCheck. · 3bfaefe9
  Evan Cheng authored Dec 19, 2011
  3bfaefe9
Dec 15, 2011
- Do not sink instruction, if it is not profitable. · c2686886
  Devang Patel authored Dec 14, 2011
  c2686886
Dec 14, 2011
- - Add MachineInstrBundle.h and MachineInstrBundle.cpp. This includes a function · 7fae11b2
  Evan Cheng authored Dec 14, 2011
  7fae11b2
- Add newline at EOF. · 4020ae75
  Chad Rosier authored Dec 14, 2011
  4020ae75
Dec 13, 2011
- [fast-isel] Unaligned loads of floats are not supported. Therefore, convert to a regular · 563de603
  Chad Rosier authored Dec 13, 2011
  563de603
Dec 12, 2011
- Implement 'e' and 'f' modifiers for Neon inline asm. <rdar://problem/10551006 > · fadc2c83
  Bob Wilson authored Dec 12, 2011
  fadc2c83
- Manually upgrade the test suite to specify the flag to cttz and ctlz. · 6b0e34c4
  Chandler Carruth authored Dec 12, 2011
  6b0e34c4
Dec 11, 2011
- Fixed bug 9905: Failure in code selection for llvm intrinsics sqrt/exp (fix... · 46837409
  Stepan Dyatkovskiy authored Dec 11, 2011
  46837409
Dec 10, 2011
- Revert associate SelectInsertValue test as well. · 1c468af8
  Chad Rosier authored Dec 10, 2011
  1c468af8
- Revert r146322 to appease buildbots. Original commit message: · 6641294e
  Chad Rosier authored Dec 10, 2011
  6641294e
- Fixed bug 9905: Failure in code selection for llvm intrinsics sqrt/exp (fix... · df0b779e
  Stepan Dyatkovskiy authored Dec 10, 2011
  df0b779e
- Splats can contain undef's; make sure to handle them correctly. PR11526. · 4e36a934
  Eli Friedman authored Dec 09, 2011
  4e36a934
Dec 09, 2011
- [fast-isel] Add support for selecting insertvalue. · dd998ff4
  Chad Rosier authored Dec 09, 2011
  dd998ff4