Commits · fc3d7f664b77fd270f800bc3bb3f749b4bfcd9ca · Roger Ferrer / llvm-epi · GitLab

Jan 07, 2011
- Appropriately truncate debug info range in dwarf output. · 6381e158
  Devang Patel authored Jan 07, 2011
  6381e158
- Revert r122955. It seems using movups to lower memcpy can cause massive... · a048c83f
  Evan Cheng authored Jan 07, 2011
  a048c83f
- Try to unbreak the arm buildbot. · 1ec7ecce
  Benjamin Kramer authored Jan 07, 2011
  1ec7ecce
- Add testcases for PR8411 (vget_low and vget_high implemented as shuffles). · 99da75c1
  Bob Wilson authored Jan 07, 2011
  99da75c1
- Add ARM patterns to match EXTRACT_SUBVECTOR nodes. · 8265d566
  Bob Wilson authored Jan 07, 2011
  8265d566
- Fix the other problem reported in PR8582. Testcase and patch by · 61c5708b
  Duncan Sands authored Jan 06, 2011
  61c5708b
- Add a testcase for PR8582, which mysteriously fixed itself, in case the problem · 64b75da0
  Duncan Sands authored Jan 06, 2011
  64b75da0
Jan 06, 2011
- PR8921: LDM/POP do not support interworking prior to v5t. · 914df82a
  Bob Wilson authored Jan 06, 2011
  914df82a
- Use movups to lower memcpy and memset even if it's not fast (like corei7). · 7998b1d6
  Evan Cheng authored Jan 06, 2011
  7998b1d6
- Re-implement r122936 with proper target hooks. Now getMaxStoresPerMemcpy · 3ae2b79a
  Evan Cheng authored Jan 06, 2011
  3ae2b79a
- Revert r122936. I'll re-implement the change. · c052ba7f
  Evan Cheng authored Jan 06, 2011
  c052ba7f
- Fix test to coincide with r122934 change from PR8919. · 2b898548
  Bill Wendling authored Jan 06, 2011
  2b898548
- r105228 reduced the memcpy / memset inline limit to 4 with -Os to avoid blowing · 06536e71
  Evan Cheng authored Jan 06, 2011
  06536e71
- Avoid zero extend bit test operands to pointer type if all the masks fit in · ac730dd2
  Evan Cheng authored Jan 06, 2011
  ac730dd2
- Optimize: · 260acf32
  Evan Cheng authored Jan 05, 2011
  260acf32
Jan 05, 2011
- fix PR8900, a shuffle miscompilation. Patch by Nadav Rotem! · 872908fd
  Chris Lattner authored Jan 05, 2011
  872908fd
Jan 03, 2011
- Use pushq / popq instead of subq $8, %rsp / addq $8, %rsp to adjust stack in · 65089fc6
  Evan Cheng authored Jan 03, 2011
  65089fc6
Jan 02, 2011
- Try to reuse the value when lowering memset. · 25e6e06e
  Benjamin Kramer authored Jan 02, 2011
  25e6e06e
- Lower the i8 extension in memset to a multiply instead of a potentially long... · 2fdea4c8
  Benjamin Kramer authored Jan 02, 2011
  2fdea4c8
Jan 01, 2011
- Fix darwin bots. · abe3eaa4
  Rafael Espindola authored Jan 01, 2011
  abe3eaa4
- Add support for the 'H' modifier. · d606e547
  Rafael Espindola authored Jan 01, 2011
  d606e547
- Update the test · 879be84a
  Anton Korobeynikov authored Jan 01, 2011
  879be84a
- ptx: remove reg-reg addressing mode and st.const · 5451fc91
  Che-Liang Chiou authored Jan 01, 2011
  5451fc91
- ptx: add store instruction · 15e8d2c5
  Che-Liang Chiou authored Jan 01, 2011
  15e8d2c5
Dec 30, 2010
- ptx: add state spaces · 3ee05013
  Che-Liang Chiou authored Dec 30, 2010
  3ee05013
Dec 29, 2010
- test/CodeGen/X86/negative-sin.ll: FileCheck-ize. · dbacfb73
  NAKAMURA Takumi authored Dec 29, 2010
  dbacfb73
- test/CodeGen/X86/fp-in-intregs.ll: FileCheck-ize. · 74835a22
  NAKAMURA Takumi authored Dec 29, 2010
  74835a22
Dec 23, 2010
- Radar 8803471: Fix expansion of ARM BCCi64 pseudo instructions. · 36be00ce
  Bob Wilson authored Dec 23, 2010
  36be00ce
- Fixes PR8823: add-with-overflow-128.ll · 033efdf4
  Andrew Trick authored Dec 23, 2010
  033efdf4
- DAGCombine add (sext i1), X into sub X, (zext i1) if sext from i1 is illegal. ... · 1f4dfbbc
  Benjamin Kramer authored Dec 22, 2010
  1f4dfbbc
- X86: Lower a select directly to a setcc_carry if possible. · 6020ed9d
  Benjamin Kramer authored Dec 22, 2010
  6020ed9d
Dec 22, 2010
- ptx: add ld instruction and test · aaedf8be
  Che-Liang Chiou authored Dec 22, 2010
  aaedf8be
- Fix a bug in ReduceLoadWidth that wasn't handling extending · cafc1e60
  Chris Lattner authored Dec 22, 2010
  cafc1e60
Dec 21, 2010
- Reapply 122353-122355 with fixes. 122354 was wrong; · a94e36bb
  Dale Johannesen authored Dec 21, 2010
  a94e36bb
- Add some x86 specific dagcombines for conditional increments. · f6ddc4a1
  Benjamin Kramer authored Dec 21, 2010
  f6ddc4a1
- Revert 122353-122355 for the moment, they broke stuff. · 87c47499
  Dale Johannesen authored Dec 21, 2010
  87c47499
- Add a new transform to DAGCombiner. · caf42aa6
  Dale Johannesen authored Dec 21, 2010
  caf42aa6
- Get the type of a shift from the shift, not from its shift · fa5dc82f
  Dale Johannesen authored Dec 21, 2010
  fa5dc82f
- Add ARM-specific DAG combining to cast i64 vector element load/stores to f64. · 1a20c2ae
  Bob Wilson authored Dec 21, 2010
  1a20c2ae
Dec 20, 2010
- Cosmetic changes. · 0a291a36
  Dale Johannesen authored Dec 20, 2010
  0a291a36