Commits · f50b597f6716b44ab553dd3d5470c35ec3a786a5 · Raul Torres / llvm-target-spread · GitLab

Apr 09, 2005
- Fix 64b shifts · f50b597f
  Nate Begeman authored Apr 08, 2005
  f50b597f
Apr 08, 2005
- fix this method for 64-bit constants · 61b6f04a
  Chris Lattner authored Apr 08, 2005
  61b6f04a
- Match Mac OS X 64 bit calling conventions · 705d3c18
  Nate Begeman authored Apr 08, 2005
  705d3c18
- collect a few statistics, factor constants (constant loading and mult), fix... · de5aed3f
  Andrew Lenharth authored Apr 08, 2005
  de5aed3f
- oops · ce9e043c
  Andrew Lenharth authored Apr 08, 2005
  ce9e043c
- added some tests to check stupid pattern matching mistakes · 2e184e25
  Andrew Lenharth authored Apr 08, 2005
  2e184e25
- fix bogus division-by-power-of-2 (was wrong for negative input, adds extr insn) · 41ff5025
  Duraid Madina authored Apr 08, 2005
  41ff5025
- Fix bug: InstCombine/2005-05-07-UDivSelectCrash.ll · 42362619
  Chris Lattner authored Apr 08, 2005
  42362619
- new testcase that crashes the instcombiner. · 9e2b5fc6
  Chris Lattner authored Apr 08, 2005
  9e2b5fc6
Apr 07, 2005
- Optimized code sequences for setcc reg, 0 · b1f66d1a
  Nate Begeman authored Apr 07, 2005
  b1f66d1a
- Alpha zero extends setcc results · 534eebb3
  Andrew Lenharth authored Apr 07, 2005
  534eebb3
- If a target zero or sign extends the result of its setcc, allow folding of · b32d9318
  Chris Lattner authored Apr 07, 2005
  b32d9318
- PowerPC zero extends setcc results · 532ac791
  Chris Lattner authored Apr 07, 2005
  532ac791
- X86 zero extends setcc results · 38fd9708
  Chris Lattner authored Apr 07, 2005
  38fd9708
- Allow targets which produce setcc results in non-MVT::i1 registers to describe · 693e797b
  Chris Lattner authored Apr 07, 2005
  693e797b
- Remove somethign I had for testing · dfed7355
  Chris Lattner authored Apr 07, 2005
  dfed7355
- fix a small optimization opertunity and make gcc happy · 9807ac5d
  Andrew Lenharth authored Apr 07, 2005
  9807ac5d
- This patch does two things. First, it canonicalizes 'X >= C' -> 'X > C-1' · 6b03a0cb
  Chris Lattner authored Apr 07, 2005
  6b03a0cb
- match case change in codegen · 31f5e2f7
  Andrew Lenharth authored Apr 07, 2005
  31f5e2f7
- fixup magic constant making code. tested by thousands of random divisions....... · 6b492bec
  Andrew Lenharth authored Apr 07, 2005
  6b492bec
- lowercase instructions, makes diff happier · d2da7177
  Andrew Lenharth authored Apr 07, 2005
  d2da7177
- Implement the following xforms: · 4706046e
  Chris Lattner authored Apr 07, 2005
  4706046e
- new test · 679c1119
  Chris Lattner authored Apr 07, 2005
  679c1119
- Implement InstCombine/add.ll:test28, transforming C1-(X+C2) --> (C1-C2)-X. · c7f3c1a0
  Chris Lattner authored Apr 07, 2005
  c7f3c1a0
- new testcase · dd83183c
  Chris Lattner authored Apr 07, 2005
  dd83183c
- Transform X-(X+Y) == -Y and X-(Y+X) == -Y · a9be4490
  Chris Lattner authored Apr 07, 2005
  a9be4490
- It wasn't happy about this either · 05e51d92
  Andrew Lenharth authored Apr 07, 2005
  05e51d92
- Yea, it wasn't happy · 85f34a56
  Andrew Lenharth authored Apr 07, 2005
  85f34a56
- teach asmprinter to print s8/s14 operands · a7abda39
  Duraid Madina authored Apr 07, 2005
  a7abda39
- codegen immediate forms of add/sub/shift · 8419da8a
  Duraid Madina authored Apr 07, 2005
  8419da8a
- add immediate forms of add, sub, shift · b484f7c5
  Duraid Madina authored Apr 07, 2005
  b484f7c5
- Fix a really scary bug that Nate found where we weren't deleting the right · 7d13eae2
  Chris Lattner authored Apr 07, 2005
  7d13eae2
- Pattern match bitfield insert, which helps shift long by immediate, among · d20628ff
  Nate Begeman authored Apr 06, 2005
  d20628ff
- Fix some shift bugs · 505f6b76
  Nate Begeman authored Apr 06, 2005
  505f6b76
- Make these 64 bit constants so that this compiles on x86-32 as well. · 7ca09472
  Alkis Evlogimenos authored Apr 06, 2005
  7ca09472
- added sdiv by 2^k and works for neg divisors also · 3ea17c6c
  Andrew Lenharth authored Apr 06, 2005
  3ea17c6c
Apr 06, 2005
- Don't make this require loopsimplify. It works BETTER with loop simplify · 747eee27
  Chris Lattner authored Apr 06, 2005
  747eee27
- Teach ExpandShift how to handle shifts by a constant. This allows targets · b0674925
  Nate Begeman authored Apr 06, 2005
  b0674925
- fix copy/paste errors, and add imm support to SxADDQ and SxSUBQ · df3256aa
  Andrew Lenharth authored Apr 06, 2005
  df3256aa
- Fix SingleSource/Regression/C/2005-05-06-LongLongSignedShift.c, we were not · bd32728a
  Chris Lattner authored Apr 06, 2005
  bd32728a