Commit Graph

78431 Commits

Author SHA1 Message Date
Jakub Staszak 43d7cf6b82 Add link to builtin_expect in Release Notes.
llvm-svn: 145979
2011-12-06 22:31:27 +00:00
Bill Wendling efdd2f8fef Re-enable compact unwind. It seems to work now. <rdar://problem/10441838>
llvm-svn: 145977
2011-12-06 22:18:12 +00:00
Bill Wendling 67a70c995a Explicitly check for the different SUB instructions.
llvm-svn: 145976
2011-12-06 22:14:27 +00:00
Evan Cheng 2a81dd4a3c First chunk of MachineInstr bundle support.
1. Added opcode BUNDLE
2. Taught MachineInstr class to deal with bundled MIs
3. Changed MachineBasicBlock iterator to skip over bundled MIs; added an iterator to walk all the MIs
4. Taught MachineBasicBlock methods about bundled MIs

llvm-svn: 145975
2011-12-06 22:12:01 +00:00
Jim Grosbach 1f5c5aa209 Tidy up. Fix naming convention stuff for some internal functions.
llvm-svn: 145974
2011-12-06 22:07:02 +00:00
Jakob Stoklund Olesen 33fe130e12 Use conservative size estimate for tBR_JTr.
This pseudo-instruction contains a .align directive in its expansion, so
the total size may vary by 2 bytes.

It is too difficult to accurately keep track of this alignment
directive, just use the worst-case size instead.

llvm-svn: 145971
2011-12-06 21:55:39 +00:00
Jakob Stoklund Olesen 2fa7448f31 Remove alignment from deserted constant islands.
ARMConstantIslandPass may sometimes leave empty constant islands behind
(it really shouldn't). Remove the alignment from the empty islands so
the size calculations are still correct.

This should fix the many Thumb1 assembler errors in the nightly test
suite.

The reduced test case for this problem is way too big. That is to be
expected for ARMConstantIslandPass bugs.

<rdar://problem/10534709>

llvm-svn: 145970
2011-12-06 21:55:35 +00:00
Bill Wendling 5a173cd367 Encode the total stack if there isn't a frame.
llvm-svn: 145969
2011-12-06 21:34:01 +00:00
Bill Wendling a73c0c99ea * Add a macro to remove a magic number.
* Rename variables to reflect what they're actually used for.

llvm-svn: 145968
2011-12-06 21:23:42 +00:00
Jakob Stoklund Olesen 2a2b37ea4a Pretty-print basic block alignment.
llvm-svn: 145965
2011-12-06 21:08:39 +00:00
Jakub Staszak cf9af0fc3b Add link to llvm.expect in Release Notes.
llvm-svn: 145964
2011-12-06 20:56:36 +00:00
Hal Finkel 0fc34bc2d3 delaying restore-cr changed assigned registers in some tests
llvm-svn: 145963
2011-12-06 20:55:46 +00:00
Hal Finkel 0702bc1b28 add a test case that uses RESTORE_CR
llvm-svn: 145962
2011-12-06 20:55:41 +00:00
Hal Finkel bde7f8ffe2 add RESTORE_CR and support CR unspills
llvm-svn: 145961
2011-12-06 20:55:36 +00:00
Hal Finkel 4ec02b02ac remove old FIXME
llvm-svn: 145960
2011-12-06 20:52:56 +00:00
Bill Wendling 87571b6392 Check the correct value for small stack sizes. Also modify some comments.
llvm-svn: 145954
2011-12-06 19:16:17 +00:00
Bill Wendling a4e87944a8 For a small sized stack, we encode that value directly with no "stack adjust" value.
llvm-svn: 145952
2011-12-06 19:09:06 +00:00
Justin Holewinski 04424665c3 PTX: Continue to fix up the register mess.
llvm-svn: 145947
2011-12-06 17:39:48 +00:00
Justin Holewinski 3063ac87aa PTX: Encode registers as unsigned values in the MC asm printer instead of using external symbols
llvm-svn: 145946
2011-12-06 17:39:46 +00:00
Sebastian Pop ac35a4d0f7 use space star instead of star space
llvm-svn: 145944
2011-12-06 17:34:16 +00:00
Sebastian Pop 9aa6137d97 add missing point at the end of sentences
llvm-svn: 145943
2011-12-06 17:34:11 +00:00
Benjamin Kramer b5188f163a Simplify common predecessor finding.
- Walking over pred_begin/pred_end is an expensive operation.
- PHINodes contain a value for each predecessor anyway.
- While it may look like we used to save a few iterations with the set,
  be aware that getIncomingValueForBlock does a linear search on
  the values of the phi node.
- Another -5% on ARMDisassembler.cpp (Release build). This was the last
  entry in the profile that was obviously wasting time.

llvm-svn: 145937
2011-12-06 16:14:29 +00:00
Benjamin Kramer b3bd019cd7 Push StringRefs through the metadata interface.
llvm-svn: 145934
2011-12-06 11:50:26 +00:00
Craig Topper 83320e03e6 Add X86ISD::HADD/HSUB to getTargetNodeName
llvm-svn: 145929
2011-12-06 09:31:36 +00:00
Craig Topper 6572e0f203 Fix a bunch of SSE/AVX patterns to use v2i64/v4i64 loads since all other integer vector loads are promoted to those.
llvm-svn: 145927
2011-12-06 09:04:59 +00:00
Craig Topper 8d4ba198d6 Merge floating point and integer UNPCK X86ISD node types.
llvm-svn: 145926
2011-12-06 08:21:25 +00:00
NAKAMURA Takumi 51416d5f00 test/MC: Introduce MC/MachO/ARM, and relocate relax-thumb2-branches.s into it.
FIXME: Restore more other arch-dependent MachO tests. (eg. r126401 and r133856)
llvm-svn: 145925
2011-12-06 06:48:26 +00:00
Craig Topper 3cb802c775 Clean up some of the shuffle decoding code for UNPCK instructions. Add instruction commenting for AVX/AVX2 forms for integer UNPCKs.
llvm-svn: 145924
2011-12-06 05:31:16 +00:00
Jim Grosbach e303e24d77 ARM mode 'mul' operand ordering tweak.
Same as r145922, just for ARM mode.

llvm-svn: 145923
2011-12-06 05:28:00 +00:00
Jim Grosbach 5f143be8c5 Thumb2: MUL two-operand form encoding operand order fix.
Fix the alias to encode 'mul r5, r6' as if it were 'mul r5, r6, r5' so we
match gas.

rdar://10532439

llvm-svn: 145922
2011-12-06 05:03:45 +00:00
Craig Topper bf41eb3a98 Merge isSHUFPMask and isCommutedSHUFPMask into single function that can do both. Do the same for the 256-bit version. Use loops to reduce size of isVSHUFPYMask. Fix test cases that were incorrectly passing due to isCommutedSHUFPMask not checking for the vector being 128-bit. This caused some 256-bit shuffles to be incorrectly commuted.
llvm-svn: 145921
2011-12-06 04:59:07 +00:00
Jim Grosbach 175c7d0da5 Thumb2 encoding choice correction for PLD.
Using encoding T1 for offset of #0 and encoding T2 for #-0.

rdar://10532413

llvm-svn: 145919
2011-12-06 04:49:29 +00:00
NAKAMURA Takumi 5bdc0fbabd test/MC: Move relax-thumb2-branches.s from MC/MachO/ to MC/ARM.
MC/MachO assumes x86.

llvm-svn: 145916
2011-12-06 03:56:05 +00:00
Dan Gohman 32772f7790 Fix a subtle semantic issue with poison values that came up in
recent discussions. Poison can't make every value that depends on
it act in maximally undefined ways, because the optimizer may still
hoist code following the usual rules for undef. Make Poison invoke
its full undefined behavior only when it reaches an instruction with
externally visible side effects.

llvm-svn: 145913
2011-12-06 03:35:58 +00:00
Bruno Cardoso Lopes 0c24d8a406 Use branches instead of jumps + variable cleanup. Testcase coming next. Patch by Jack Carter
llvm-svn: 145912
2011-12-06 03:34:48 +00:00
Bruno Cardoso Lopes 87cfffe149 Explicit symbols for gnu mimicing relocations. Patch by Jack Carter
llvm-svn: 145911
2011-12-06 03:34:42 +00:00
Bruno Cardoso Lopes 1b1a122b4c Add register HWR29 numbering. Patch by Jack Carter
llvm-svn: 145910
2011-12-06 03:34:36 +00:00
Dan Gohman 5f115a78cb Line up the comments in a code example.
llvm-svn: 145908
2011-12-06 03:31:14 +00:00
Dan Gohman 9a2a0933ed Rename "Trap Values" to "Poison Values", to better reflect their
purpose, and to avoid ambiguity with other uses of the word "trap"
in LangRef.

llvm-svn: 145907
2011-12-06 03:18:47 +00:00
Andrew Trick 5df9096584 LSR: prune undesirable formulae early.
It's always good to prune early, but formulae that are unsatisfactory
in their own right need to be removed before running any other pruning
heuristics. We easily avoid generating such formulae, but we need them
as an intermediate basis for forming other good formulae.

llvm-svn: 145906
2011-12-06 03:13:31 +00:00
Evan Cheng c1610bede1 Mix some minor misuse of MachineBasicBlock iterator.
llvm-svn: 145903
2011-12-06 02:49:06 +00:00
Pete Cooper d2971264c6 Removed isWinToJoinCrossClass from the register coalescer.
The new register allocator is much more able to split back up ranges too constrained by register classes.

Fixes <rdar://problem/10466609>

llvm-svn: 145899
2011-12-06 02:06:50 +00:00
Chris Lattner 654d054705 allow TinyPtrVector to implicitly convert to ArrayRef.
llvm-svn: 145898
2011-12-06 02:00:33 +00:00
Lang Hames 52f24d7a32 Kill off the LoopSplitter. It's not being used or maintained.
llvm-svn: 145897
2011-12-06 01:57:59 +00:00
Bill Wendling 4e87e850a2 Add a comment.
llvm-svn: 145896
2011-12-06 01:57:48 +00:00
Jim Grosbach 425e180ce8 Tidy up value checking.
llvm-svn: 145895
2011-12-06 01:53:17 +00:00
NAKAMURA Takumi d3002490bf MipsAsmBackend.cpp, PPCAsmBackend.cpp: Fix -Asserts build to appease msvc.
llvm-svn: 145894
2011-12-06 01:48:32 +00:00
Lang Hames b13b6a04d0 Update PBQP's analysis usage to reflect the requirements of the inline spiller.
llvm-svn: 145893
2011-12-06 01:45:57 +00:00
Chad Rosier c77830d21e [arm-fast-isel] Doublewords only require word-alignment.
rdar://10528060

llvm-svn: 145891
2011-12-06 01:44:17 +00:00
Jakob Stoklund Olesen 2e05db2fa0 Align ARM constant pool islands via their basic block.
Previously, all ARM::CONSTPOOL_ENTRY instructions had a hardwired
alignment of 4 bytes emitted by ARMAsmPrinter.  Now the same alignment
is set on the basic block.

This is in preparation of supporting ARM constant pool islands with
different alignments.

llvm-svn: 145890
2011-12-06 01:43:02 +00:00