Commits · aef55c8801ef2b6ab81ac73d1499e98eb86947bf · Roger Ferrer / llvm-epi-0.8

Apr 15, 2011
- Fix lines that have incorrect indentation or exceed 80 columns. There is no... · aef55c88
  Akira Hatanaka authored Apr 15, 2011
```
Fix lines that have incorrect indentation or exceed 80 columns. There is no change in functionality.

llvm-svn: 129606
```
  aef55c88
- The AND instruction leaves the V flag unmodified, so it falls victim to the same · 0829b306
  Cameron Zwarich authored Apr 15, 2011
```
problem as all of the other instructions we fold with CMPs.

llvm-svn: 129602
```
  0829b306
- Fix cmake build. · 7583dbdc
  Rafael Espindola authored Apr 15, 2011
```
llvm-svn: 129601
```
  7583dbdc
- Add missing register forms of instructions to the ARM CMP-folding code. This · 93eae157
  Cameron Zwarich authored Apr 15, 2011
```
fixes <rdar://problem/9287901>.

llvm-svn: 129599
```
  93eae157
- Add pass that expands pseudo instructions into target instructions after... · 27916977
  Akira Hatanaka authored Apr 15, 2011
```
Add pass that expands pseudo instructions into target instructions after register allocation. Define pseudos that get expanded into mtc1 or mfc1 instructions.

llvm-svn: 129594
```
  27916977
- Increase SubtargetFeatureKV Value and Implies fields to 64 bits since some... · a2e61292
  Evan Cheng authored Apr 15, 2011
```
Increase SubtargetFeatureKV Value and Implies fields to 64 bits since some targets are getting very close to 32 subtarget features. Also teach tablegen to error when there are more than 64 features to guard against undefined behavior. rdar://9282332

llvm-svn: 129590
```
  a2e61292
- Add 129518 back with a fix for when we are producing eh just because of debug info. · a01cdb0e
  Rafael Espindola authored Apr 15, 2011
```
Change ELF systems to use CFI for producing the EH tables. This reduces the
size of the clang binary in Debug builds from 690MB to 679MB.

llvm-svn: 129571
```
  a01cdb0e
- Fix a ton of comment typos found by codespell. Patch by · 0ab5e2cd
  Chris Lattner authored Apr 15, 2011
```
Luis Felipe Strano Moraes!

llvm-svn: 129558
```
  0ab5e2cd
- Revert r129518, "Change ELF systems to use CFI for producing the EH tables. This reduces the" · b5e3e9dd
  NAKAMURA Takumi authored Apr 15, 2011
```
It broke several builds.

llvm-svn: 129557
```
  b5e3e9dd
- Fix another fcopysign lowering bug. If src is f64 and destination is f32, don't · 12bb05b7
  Evan Cheng authored Apr 15, 2011
```
forget to right shift the source by 32 first. rdar://9287902

llvm-svn: 129556
```
  12bb05b7
- For t2BFI, both Inst{26} and Inst{5} "should" be 0. · 681fef59
  Johnny Chen authored Apr 15, 2011
```
Ref: I.1 Instruction encoding diagrams and pseudocode
llvm-svn: 129552
```
  681fef59
- Add 3DNow! intrinsics. · 30088ba1
  Michael J. Spencer authored Apr 15, 2011
```
llvm-svn: 129551
```
  30088ba1
- The ARM disassembler did not handle the alignment correctly for VLD*DUP* instructions · 42131617
  Johnny Chen authored Apr 15, 2011
```
(single element or n-element structure to all lanes).

llvm-svn: 129550
```
  42131617
- Follow up on r127913. Fix Thumb revsh isel. rdar://9286766 · 44887f9c
  Evan Cheng authored Apr 14, 2011
```
llvm-svn: 129548
```
  44887f9c
Apr 14, 2011
- Add sanity checkings for Thumb2 Load/Store Register Exclusive family of operations. · 4251b151
  Johnny Chen authored Apr 14, 2011
```
llvm-svn: 129531
```
  4251b151
- move PR9661 out to here. · 6f195469
  Chris Lattner authored Apr 14, 2011
```
llvm-svn: 129527
```
  6f195469
- Change ELF systems to use CFI for producing the EH tables. This reduces the · aa2a7cd8
  Rafael Espindola authored Apr 14, 2011
```
size of the clang binary in Debug builds from 690MB to 679MB.

llvm-svn: 129518
```
  aa2a7cd8
- Fix whitespace and tabs. · b88784c1
  Michael J. Spencer authored Apr 14, 2011
```
llvm-svn: 129517
```
  b88784c1
- add a minor missed dag combine that is blocking mid-level optimization · 1d313c6f
  Chris Lattner authored Apr 14, 2011
```
improvements, that will lead to fixing PR6627.

llvm-svn: 129504
```
  1d313c6f
- As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias · 410ec4aa
  Bill Wendling authored Apr 14, 2011
```
(movzx/movsx) because they give more information. Revert that part of the patch.

llvm-svn: 129498
```
  410ec4aa
- Have the X86 back-end emit the alias instead of what's being aliased. In most · 7e07d6fb
  Bill Wendling authored Apr 14, 2011
```
cases, it's much nicer and more informative reading the alias.

llvm-svn: 129497
```
  7e07d6fb
- Add an option to not print the alias of an instruction. It defaults to "print · 6dd69d92
  Bill Wendling authored Apr 13, 2011
```
the alias".

llvm-svn: 129485
```
  6dd69d92
Apr 13, 2011
- Thumb disassembler did not handle tBRIND (indirect branch) properly. · d0fb04f4
  Johnny Chen authored Apr 13, 2011
```
rdar://problem/9280370

llvm-svn: 129480
```
  d0fb04f4
- Check for unallocated instruction encodings when disassembling Thumb Branch... · b6a37bff
  Johnny Chen authored Apr 13, 2011
```
Check for unallocated instruction encodings when disassembling Thumb Branch instructions (tBcc and t2Bcc).

rdar://problem/9280470

llvm-svn: 129471
```
  b6a37bff
- The LDR*T/STR*T (unpriviledged load/store) operations don't take SP or PC as Rt. · ffa6378f
  Johnny Chen authored Apr 13, 2011
```
rdar://problem/9279440

llvm-svn: 129469
```
  ffa6378f
- Fix a typo in an ARM-specific DAG combine. This fixes <rdar://problem/9278274>. · 415b5e83
  Cameron Zwarich authored Apr 13, 2011
```
llvm-svn: 129468
```
  415b5e83
- Fix a regression caused by r102515 where explicit alignment on globals is · 9398197e
  Cameron Zwarich authored Apr 13, 2011
```
ignored. There was a test to catch this, but it was just blindly updated in
a large change. This fixes another part of <rdar://problem/9275290>.

llvm-svn: 129466
```
  9398197e
- Check the corner cases for t2LDRSHi12 correctly and mark invalid encodings as such. · 70591cbc
  Johnny Chen authored Apr 13, 2011
```
rdar://problem/9276651

llvm-svn: 129462
```
  70591cbc
- Fix a bug where for t2MOVCCi disassembly, the TIED_TO register operand was not properly handled. · 0d306a78
  Johnny Chen authored Apr 13, 2011
```
rdar://problem/9276427

llvm-svn: 129456
```
  0d306a78
- Forgot to add this change for http://llvm.org/viewvc/llvm-project?view=rev&revision=129387. · b2f9fa1f
  Johnny Chen authored Apr 13, 2011
```
llvm-svn: 129451
```
  b2f9fa1f
- Fix an obvious problem with an alignment computation. AsmPrinter actually does · 70be27e9
  Cameron Zwarich authored Apr 13, 2011
```
the max itself, so it is not easy to write a test case for this, but I added a
test case that would fail if the code in AsmPrinter were removed.

llvm-svn: 129432
```
  70be27e9
- Fix a typo. · 8001850e
  Cameron Zwarich authored Apr 13, 2011
```
llvm-svn: 129429
```
  8001850e
- If a global variable has a specified alignment that is less than the preferred · cdf59f70
  Cameron Zwarich authored Apr 13, 2011
```
alignment for its type, use the minimum of the specified alignment and the ABI
alignment. This fixes <rdar://problem/9275290>.

llvm-svn: 129428
```
  cdf59f70
- Reapply r129401 with patch for clang. · b902f1dd
  Bill Wendling authored Apr 13, 2011
```
llvm-svn: 129419
```
  b902f1dd
- Add sanity check for Ld/St Dual forms of Thumb2 instructions. · 3c2f74c9
  Johnny Chen authored Apr 12, 2011
```
rdar://problem/9273947

llvm-svn: 129411
```
  3c2f74c9
- Add @earlyclobber constraints to the writeback register of all ARM store instructions. · 98716404
  Jakob Stoklund Olesen authored Apr 12, 2011
```
The ARMARM specifies these instructions as unpredictable when storing the
writeback register. This shouldn't affect code generation much since storing a
pointer to itself is quite rare.

llvm-svn: 129409
```
  98716404
- Revert r129401 for now. Clang is using the old way of doing things. · dbfde424
  Bill Wendling authored Apr 12, 2011
```
llvm-svn: 129403
```
  dbfde424
- Remove the unaligned load intrinsics in favor of using native unaligned loads. · 47c24875
  Bill Wendling authored Apr 12, 2011
```
Now that we have a first-class way to represent unaligned loads, the unaligned
load intrinsics are superfluous.

First part of <rdar://problem/8460511>.

llvm-svn: 129401
```
  47c24875
Apr 12, 2011
- The Thumb2 RFE instructions need to have their second halfword fully specified. · 960eef3d
  Johnny Chen authored Apr 12, 2011
```
In addition, the base register is not rGPR, but GPR with th exception that:

    if n == 15 then UNPREDICTABLE

rdar://problem/9273836

llvm-svn: 129391
```
  960eef3d
- Add bad register checks for Thumb2 Ld/St instructions. · 01637b9a
  Johnny Chen authored Apr 12, 2011
```
rdar://problem/9269047

llvm-svn: 129387
```
  01637b9a