Commits · 009ea54262b03b034fd474f58f9ed3f2b5700fa0 · Roger Ferrer / llvm-epi-0.8

Mar 15, 2007
- Multiplication support for MMX. · e3103412
  Bill Wendling authored Mar 15, 2007
```
llvm-svn: 35118
```
  e3103412
Mar 14, 2007
- Under X86-64 large code model, do not emit 32-bit pc relative calls. · a1779b97
  Evan Cheng authored Mar 14, 2007
```
llvm-svn: 35108
```
  a1779b97
- Notes about codegen issues. · 71face25
  Evan Cheng authored Mar 14, 2007
```
llvm-svn: 35107
```
  71face25
- Clean up. · e278f30a
  Evan Cheng authored Mar 14, 2007
```
llvm-svn: 35105
```
  e278f30a
- Oops. · eb76f7c6
  Evan Cheng authored Mar 14, 2007
```
llvm-svn: 35104
```
  eb76f7c6
- X86-64 JIT is in large code model. Need stubs for direct calls. · dda4c32d
  Evan Cheng authored Mar 14, 2007
```
llvm-svn: 35097
```
  dda4c32d
- x86-64 JIT stub codegen. · 386dfc92
  Evan Cheng authored Mar 14, 2007
```
llvm-svn: 35096
```
  386dfc92
- Preliminary support for X86-64 JIT stub codegen. · e1a95206
  Evan Cheng authored Mar 14, 2007
```
llvm-svn: 35095
```
  e1a95206
Mar 13, 2007
- More flexible TargetLowering LSR hooks for testing whether an immediate is · 3ab7ea79
  Evan Cheng authored Mar 12, 2007
```
 a legal target address immediate or scale.

llvm-svn: 35073
```
  3ab7ea79
Mar 12, 2007
- Stupid bug: SSE2 supports v2i64 add / sub. · 57f261b1
  Evan Cheng authored Mar 12, 2007
```
llvm-svn: 35070
```
  57f261b1
Mar 10, 2007
- Adding more arithmetic operators to MMX. This is an almost exact copy of · e9b81f53
  Bill Wendling authored Mar 10, 2007
```
the addition. Please let me know if you have suggestions.

llvm-svn: 35055
```
  e9b81f53
Mar 08, 2007
- Added "padd*" support for MMX. Added MMX move stuff to X86InstrInfo so that · 6092ce25
  Bill Wendling authored Mar 08, 2007
```
moves, loads, etc. are recognized.

llvm-svn: 35031
```
  6092ce25
- Putting more constants which do not contain relocations into .literal{4|8|16} · ddf08208
  Evan Cheng authored Mar 08, 2007
```
llvm-svn: 35026
```
  ddf08208
- Put constant data to .const, .const_data, .literal{4|8|16} sections. · 13a1d969
  Evan Cheng authored Mar 08, 2007
```
llvm-svn: 35016
```
  13a1d969
Mar 07, 2007
- Remove useless pattern fragments. · 6d8211c0
  Bill Wendling authored Mar 07, 2007
```
llvm-svn: 35009
```
  6d8211c0
- Refactoring of formal parameter flags. Enable properly use of · ed4b303c
  Anton Korobeynikov authored Mar 07, 2007
```
zext/sext/aext stuff.

llvm-svn: 35008
```
  ed4b303c
- Properly support v8i8 and v4i16 types. It now converts them to v2i32 for · 97905b40
  Bill Wendling authored Mar 07, 2007
```
load and stores.

llvm-svn: 35002
```
  97905b40
- Fix DWARF debugging information on x86/Linux and (hopefully) · 942fda02
  Anton Korobeynikov authored Mar 07, 2007
```
Mingw32/Cygwin targets. This fixes PR978

llvm-svn: 35000
```
  942fda02
Mar 06, 2007
- Add LOAD/STORE support for MMX. · bbd25984
  Bill Wendling authored Mar 06, 2007
```
llvm-svn: 34978
```
  bbd25984
- Use new SDIselParamAttr enumeration. This removes "magick" constants · e7ec3bc7
  Anton Korobeynikov authored Mar 06, 2007
```
from formal attributes' flags processing.

llvm-svn: 34963
```
  e7ec3bc7
- Add the emms intrinsic for MMX support. · b1c86b49
  Bill Wendling authored Mar 05, 2007
```
llvm-svn: 34938
```
  b1c86b49
Mar 04, 2007
- add missing braces · d647f92a
  Chris Lattner authored Mar 04, 2007
```
llvm-svn: 34905
```
  d647f92a
Mar 03, 2007
- X86-64 VACOPY needs custom expansion. va_list is a struct { i32, i32, i8*, i8* }. · deaea25e
  Evan Cheng authored Mar 02, 2007
```
llvm-svn: 34857
```
  deaea25e
Mar 02, 2007
- Simplify things · 57af2a4f
  Anton Korobeynikov authored Mar 02, 2007
```
llvm-svn: 34849
```
  57af2a4f
- argument lowering should copy from the vreg shadows of live-in arguments · 9c7e5e36
  Chris Lattner authored Mar 02, 2007
```
passed in registers, not directly from the pregs themselves.

llvm-svn: 34838
```
  9c7e5e36
- add a note · fc2f5212
  Chris Lattner authored Mar 02, 2007
```
llvm-svn: 34837
```
  fc2f5212
Mar 01, 2007
- Ensure that fastcall'ed function is correctly mangled & stack is · af8be445
  Anton Korobeynikov authored Mar 01, 2007
```
properly aligned

llvm-svn: 34788
```
  af8be445
Feb 28, 2007
- remove dead option · 7373f3a3
  Chris Lattner authored Feb 28, 2007
```
llvm-svn: 34754
```
  7373f3a3
- bugfix: fastcall does not require the first two params to be marked 'inreg', · eb09e28c
  Chris Lattner authored Feb 28, 2007
```
they always get registers.

llvm-svn: 34748
```
  eb09e28c
- use high-level functions in CCState · 152bfa10
  Chris Lattner authored Feb 28, 2007
```
llvm-svn: 34739
```
  152bfa10
- make use of helper functions in CCState for analyzing formals and calls. · 227b6c5d
  Chris Lattner authored Feb 28, 2007
```
llvm-svn: 34737
```
  227b6c5d
- switch LowerFastCCCallTo over to using the new fastcall description. · d439e860
  Chris Lattner authored Feb 28, 2007
```
llvm-svn: 34734
```
  d439e860
- switch LowerFastCCArguments over to using the autogenerated Fastcall description. · 66e1d1dd
  Chris Lattner authored Feb 28, 2007
```
llvm-svn: 34733
```
  66e1d1dd
- add new CC_X86_32_FastCall calling conv, which describes fastcall on win32. · 0bc74fd6
  Chris Lattner authored Feb 28, 2007
```
Factor out a CC_X86_32_Common convention, which is the part shared between
ccc, stdcall and fastcall

llvm-svn: 34732
```
  0bc74fd6
- rearrange code · 3066becc
  Chris Lattner authored Feb 28, 2007
```
llvm-svn: 34731
```
  3066becc
- remove fastcc (not fastcall) support · 3ed3be3b
  Chris Lattner authored Feb 28, 2007
```
llvm-svn: 34730
```
  3ed3be3b
- switch LowerCCCArguments over to using autogenerated CC. · b9db2250
  Chris Lattner authored Feb 28, 2007
```
llvm-svn: 34729
```
  b9db2250
- simplify sret handling · 5958b176
  Chris Lattner authored Feb 28, 2007
```
llvm-svn: 34728
```
  5958b176
- switch LowerCCCCallTo over to using an autogenerated callingconv · be799595
  Chris Lattner authored Feb 28, 2007
```
llvm-svn: 34727
```
  be799595
- rename stuff · 0c0a7b28
  Chris Lattner authored Feb 28, 2007
```
llvm-svn: 34726
```
  0c0a7b28