Commits · 93bde9cbf7bd831f6161c91f6fd6ff454a9dfa50 · Roger Ferrer / llvm-epi-0.8

Jan 19, 2006
- Avoid generating a redundant setcc. · 3d2cc7e2
  Evan Cheng authored Jan 19, 2006
```
llvm-svn: 25457
```
  3d2cc7e2
- adc and sbb need an incoming flag to ensure it reads the carry flag · 91007126
  Evan Cheng authored Jan 19, 2006
```
from add / sub.

llvm-svn: 25444
```
  91007126
- Two peepholes: · a7bfbe99
  Evan Cheng authored Jan 19, 2006
```
(or (x >> c) | (y << (32 - c))) ==> (shrd x, y, c)
(or (x << c) | (y >> (32 - c))) ==> (shld x, y, c)

llvm-svn: 25438
```
  a7bfbe99
- Didn't mean to check that in. · 6135a7a5
  Evan Cheng authored Jan 19, 2006
```
llvm-svn: 25436
```
  6135a7a5
- A obvious typo · 267ba596
  Evan Cheng authored Jan 19, 2006
```
llvm-svn: 25435
```
  267ba596
Jan 18, 2006
- SRA shift amount must be in i8 · 621674a1
  Evan Cheng authored Jan 18, 2006
```
llvm-svn: 25416
```
  621674a1
- If a call return type is i1, insert a truncate from X86::AL to i1. · 4b3774e0
  Evan Cheng authored Jan 18, 2006
```
llvm-svn: 25415
```
  4b3774e0
Jan 17, 2006
- Fix lowering of calls which return f32 values. · feaed4d1
  Evan Cheng authored Jan 17, 2006
```
llvm-svn: 25413
```
  feaed4d1
- Zero extending load from i1 to i8. · 14417ed9
  Evan Cheng authored Jan 17, 2006
```
llvm-svn: 25391
```
  14417ed9
- SSE does not support i64 SINT_TO_FP (FP stack doesn't either, but we custom · 0d5b69f7
  Evan Cheng authored Jan 17, 2006
```
expand it), so ask legalizer to expand i32 UINT_TO_FP.

llvm-svn: 25386
```
  0d5b69f7
- Added a FIXME comment about why FST is currently flagged to fpGETRESULT. · 561881f3
  Evan Cheng authored Jan 17, 2006
```
llvm-svn: 25381
```
  561881f3
- Bug fixes: fpGETRESULT should produces a flag result and X86ISD::FST should · bec9d720
  Evan Cheng authored Jan 17, 2006
```
read a flag.

llvm-svn: 25378
```
  bec9d720
- More typo's · c14bb102
  Evan Cheng authored Jan 16, 2006
```
llvm-svn: 25375
```
  c14bb102
Jan 16, 2006
- Some typo's · 64eeed27
  Evan Cheng authored Jan 16, 2006
```
llvm-svn: 25374
```
  64eeed27
- Fix FP_TO_INT**_IN_MEM lowering. · 911c68d7
  Evan Cheng authored Jan 16, 2006
```
llvm-svn: 25368
```
  911c68d7
- transfer some notes from my email to somewhere useful. · b2eacf48
  Chris Lattner authored Jan 16, 2006
```
llvm-svn: 25361
```
  b2eacf48
Jan 15, 2006
- Added patterns for 8-bit multiply · 2494ce49
  Evan Cheng authored Jan 15, 2006
```
llvm-svn: 25338
```
  2494ce49
- Use the default lowering of ISD::DYNAMIC_STACKALLOC, delete now dead code. · 78c358d1
  Chris Lattner authored Jan 15, 2006
```
llvm-svn: 25333
```
  78c358d1
Jan 14, 2006
- silence a warning · 8869c6f7
  Chris Lattner authored Jan 14, 2006
```
llvm-svn: 25322
```
  8869c6f7
- bswap implementation · 2fba8a3a
  Nate Begeman authored Jan 14, 2006
```
llvm-svn: 25312
```
  2fba8a3a
- A typo. · 3bc25e8a
  Evan Cheng authored Jan 14, 2006
```
llvm-svn: 25307
```
  3bc25e8a
Jan 13, 2006
- Add truncstore i1 patterns. · 392c7d27
  Evan Cheng authored Jan 13, 2006
```
llvm-svn: 25296
```
  392c7d27
- Fix a bug in my last X86 checkin, pointed out by cozmic · 5f9c134b
  Chris Lattner authored Jan 13, 2006
```
llvm-svn: 25293
```
  5f9c134b
- LHS = X86ISD::CMOVcc LHS, RHS means LHS = RHS if cc. So the operands must be · dba84bbc
  Evan Cheng authored Jan 13, 2006
```
flipped around.

llvm-svn: 25290
```
  dba84bbc
- Enable X86 support for savestack/restorestack · 1a8d918e
  Chris Lattner authored Jan 13, 2006
```
llvm-svn: 25278
```
  1a8d918e
- expand unsupported stacksave/stackrestore nodes · 8e2f52e6
  Chris Lattner authored Jan 13, 2006
```
llvm-svn: 25272
```
  8e2f52e6
- Minor update. · f00374e4
  Evan Cheng authored Jan 13, 2006
```
llvm-svn: 25263
```
  f00374e4
- More typo's. I need new eye glasses... · d7faa4ba
  Evan Cheng authored Jan 13, 2006
```
llvm-svn: 25261
```
  d7faa4ba
- Oops. Typo. · 731423f3
  Evan Cheng authored Jan 13, 2006
```
llvm-svn: 25260
```
  731423f3
- Fix a SETCC / BRCOND folding bug. · fb22e86c
  Evan Cheng authored Jan 13, 2006
```
llvm-svn: 25259
```
  fb22e86c
Jan 12, 2006
- Fix sint_to_fp (fild*) support. · 6305e50e
  Evan Cheng authored Jan 12, 2006
```
llvm-svn: 25257
```
  6305e50e
- Specify transformation from GlobalAddress to TargetGlobalAddress and · c993d452
  Evan Cheng authored Jan 12, 2006
```
ExternalSymbol to TargetExternalSymbol.

llvm-svn: 25253
```
  c993d452
- X86ISD::SETCC (e.g. SETEr) produces a flag (so multiple SETCC can be · 84dc9b55
  Evan Cheng authored Jan 12, 2006
```
linked together).

llvm-svn: 25247
```
  84dc9b55
- * Materialize GlobalAddress and ExternalSym with MOV32ri rather than · b94db9e9
  Evan Cheng authored Jan 12, 2006
```
  LEA32r.
* Do not lower GlobalAddress to TargetGlobalAddress. Let isel does it.

llvm-svn: 25246
```
  b94db9e9
- Added ROTL and ROTR. · 6d2ab044
  Evan Cheng authored Jan 11, 2006
```
llvm-svn: 25232
```
  6d2ab044
Jan 11, 2006
- Support for MEMCPY and MEMSET. · ae986f1f
  Evan Cheng authored Jan 11, 2006
```
llvm-svn: 25226
```
  ae986f1f
- Select DYNAMIC_STACKALLOC · 2ae799af
  Evan Cheng authored Jan 11, 2006
```
llvm-svn: 25225
```
  2ae799af
- Add bswap, rotl, and rotr nodes · 1b8121b2
  Nate Begeman authored Jan 11, 2006
```
Add dag combiner code to recognize rotl, rotr
Add ppc code to match rotl

Targets should add rotl/rotr patterns if they have them

llvm-svn: 25222
```
  1b8121b2
- * Add special entry code main() (to set x87 to 64-bit precision). · bc7a0f44
  Evan Cheng authored Jan 11, 2006
```
* Allow a register node as SelectAddr() base.
* ExternalSymbol -> TargetExternalSymbol as direct function callee.
* Use X86::ESP register rather than CopyFromReg(X86::ESP) as stack ptr for
  call parmater passing.

llvm-svn: 25207
```
  bc7a0f44
- implement FP_REG_KILL insertion for the dag-dag instruction selector · 7c551268
  Chris Lattner authored Jan 11, 2006
```
llvm-svn: 25192
```
  7c551268