RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-24 19:52:54 +01:00
Code Issues Projects Releases Wiki Activity
61,574 Commits 4 Branches 5 Tags 704 MiB
7953fd5824
Commit Graph

61574 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Dan Gohman
7953fd5824 Optimize the "bit test" code path for switch lowering in the 
case where the bit mask has exactly one bit.

llvm-svn: 106716
 2010-06-24 02:06:24 +00:00
Jakob Stoklund Olesen
1c9d50ed92 Revert "Replace a big gob of old coalescer logic with the new CoalescerPair class." 
Whiny buildbots.

llvm-svn: 106710
 2010-06-24 00:52:22 +00:00
Gabor Greif
39ff38468c use getNumArgOperands 
llvm-svn: 106709
 2010-06-24 00:48:48 +00:00
Gabor Greif
1d1526999b use ArgOperand API 
llvm-svn: 106707
 2010-06-24 00:44:01 +00:00
Devang Patel
0c927a80fe Use ValueMap instead of DenseMap. 
The ValueMapper used by various cloning utility maps MDNodes also.

llvm-svn: 106706
 2010-06-24 00:33:28 +00:00
Bruno Cardoso Lopes
633f345ba9 Add AVX CMP{SS,SD}{rr,rm} instructions and encoding testcases 
llvm-svn: 106705
 2010-06-24 00:32:06 +00:00
Bruno Cardoso Lopes
5aea735b46 Move SSE and AVX shuffle, unpack and compare code to more appropriate places 
llvm-svn: 106702
 2010-06-24 00:15:50 +00:00
Jakob Stoklund Olesen
19abbf4387 Replace a big gob of old coalescer logic with the new CoalescerPair class. 
CoalescerPair can determine if a copy can be coalesced, and which register gets
merged away. The old logic in SimpleRegisterCoalescing had evolved into
something a bit too convoluted.

llvm-svn: 106701
 2010-06-24 00:12:39 +00:00
Devang Patel
ad744f3fa3 Use available typedef for " DenseMap<const Value*, Value*>". 
llvm-svn: 106699
 2010-06-24 00:00:42 +00:00
Devang Patel
373b10cbaa Cosmetic change. 
Do not use "ValueMap" as a name for a local variable or an argument.

llvm-svn: 106698
 2010-06-23 23:55:51 +00:00
Gabor Greif
b86c04f740 use ArgOperand accessors 
llvm-svn: 106697
 2010-06-23 23:38:07 +00:00
Bill Wendling
777c025ad8 We are missing opportunites to use ldm. Take code like this: 
void t(int *cp0, int *cp1, int *dp, int fmd) {
  int c0, c1, d0, d1, d2, d3;
  c0 = (*cp0++ & 0xffff) | ((*cp1++ << 16) & 0xffff0000);
  c1 = (*cp0++ & 0xffff) | ((*cp1++ << 16) & 0xffff0000);
  /* ... */
}

It code gens into something pretty bad. But with this change (analogous to the
X86 back-end), it will use ldm and generate few instructions.

llvm-svn: 106693
 2010-06-23 23:00:16 +00:00
Gabor Greif
bfb7386690 use the new isFreeCall API and ArgOperand accessors 
llvm-svn: 106692
 2010-06-23 22:48:06 +00:00
Gabor Greif
4615fc1c93 minor enhancement to llvm::isFreeCall API: return CallInst; no functional change 
llvm-svn: 106686
 2010-06-23 21:51:12 +00:00
Gabor Greif
0b348ae475 use ArgOperand getters 
llvm-svn: 106685
 2010-06-23 21:41:47 +00:00
Bruno Cardoso Lopes
b1bfbacead Add AVX MOVMSK{PS,PD}rr instructions 
llvm-svn: 106683
 2010-06-23 21:30:27 +00:00
Bruno Cardoso Lopes
c6ac26123d Add tests for different AVX cmp opcodes, also teach the x86 asm parser to understand the vcmp instruction 
llvm-svn: 106678
 2010-06-23 21:10:57 +00:00
Eric Christopher
436ff8863f Update according to feedback. 
llvm-svn: 106677
 2010-06-23 20:49:35 +00:00
Rafael Espindola
ea22615ccd Add an extra-library-path option to the plugin. This is used to support 
having a library both as bitcode and native code. We want to use the
bitcode first, but if codegen produces new undefined references we have to use
the native code to satisfy those references.

Gold has no notion of bitcode and native search directories, so instead it has
an API where the plugin can instruct it to look for the libraries it is passing
to it. This patch uses that API.

llvm-svn: 106674
 2010-06-23 20:20:59 +00:00
Bruno Cardoso Lopes
8cfdcf7691 Add AVX SHUF{PS,PD}{rr,rm} instructions 
llvm-svn: 106672
 2010-06-23 20:07:15 +00:00
Nico Weber
04606293a5 Add support for the x86 instructions "pusha" and "popa". 
llvm-svn: 106671
 2010-06-23 20:00:58 +00:00
Bruno Cardoso Lopes
fe5b207577 Fix a tblgen bug. 
Given the pattern below as an example:
list<dag> Pattern = [(set RC:$dst, (v4f32 (shufp:src3 RC:$src1,
                            (mem_frag addr:$src2))))];

The right reference resolving should lead to:
list<dag> Pattern = [(set VR128:$dst, (v4f32 (shufp:src3 VR128:$src1,
                            (mem_frag addr:$src2))))];
But was yielding:
list<dag> Pattern = [(set VR128:$dst, (v4f32 (shufp VR128:$src1,
                            (mem_frag addr:$src2))))];

Fix this by passing the right name when creating a new DagInit node.

llvm-svn: 106670
 2010-06-23 19:50:39 +00:00
Dale Johannesen
f0763e6300 Reinstate correct test, remove the real invalidated test. 
llvm-svn: 106664
 2010-06-23 18:56:06 +00:00
Dale Johannesen
4e632b4237 Remove tests invalidated by previous checkin. 
llvm-svn: 106663
 2010-06-23 18:53:12 +00:00
Dale Johannesen
e618e80a13 Do not do tail calls to external symbols. If the 
branch turns out to be ARM-to-Thumb or vice versa
the linker cannot resolve this.  8120438.

If this optimization is going to be useful we probably
need a compiler flag "assume callees are same architecture"
or something like that.

llvm-svn: 106662
 2010-06-23 18:52:34 +00:00
Bill Wendling
f92cb857aa MorphNodeTo doesn't preserve the memory operands. Because we're morphing a node 
into the same node, but with different non-memory operands, we need to replace
the memory operands after it's finished morphing.

llvm-svn: 106643
 2010-06-23 18:16:24 +00:00
Daniel Dunbar
a83dc9a464 tests: Tweak lit.cfg to fix breakage with out-of-dir lookup. 
llvm-svn: 106638
 2010-06-23 18:06:16 +00:00
Daniel Dunbar
be50ef88bd Revert r106263, "Fold the ShrinkDemandedOps pass into the regular DAGCombiner pass,"... it was causing both 'file' (with clang) and 176.gcc (with llvm-gcc) to be miscompiled. 
llvm-svn: 106634
 2010-06-23 17:09:26 +00:00
Jim Grosbach
ef4e0249a0 The generic DAG combiner can now fold atomic fences when needed, so switch 
to using that.

llvm-svn: 106633
 2010-06-23 16:25:07 +00:00
Jim Grosbach
414eb48a14 When using libcall expansions for the atomic intrinsics, the explicit 
MEMBARRIER fences aren't necessary for ARM. Tell the combiner to fold them
away.

llvm-svn: 106631
 2010-06-23 16:08:49 +00:00
Jim Grosbach
12afda541e Some targets don't require the fencing MEMBARRIER instructions surrounding 
atomic intrinsics, either because the use locking instructions for the
atomics, or because they perform the locking directly. Add support in the
DAG combiner to fold away the fences.

llvm-svn: 106630
 2010-06-23 16:07:42 +00:00
Jim Grosbach
395b0a671c remove trailing whitespace 
llvm-svn: 106628
 2010-06-23 15:49:18 +00:00
Jakob Stoklund Olesen
ac7e537231 Add a few VNInfo data structure checks. 
llvm-svn: 106627
 2010-06-23 15:34:36 +00:00
Gabor Greif
6c86715183 use ArgOperand accessors 
llvm-svn: 106626
 2010-06-23 13:56:57 +00:00
Gabor Greif
992ef6fa9c use ArgOperand accessors 
llvm-svn: 106623
 2010-06-23 13:09:06 +00:00
Gabor Greif
dedbd07ed6 use helper to neatly access arguments 
llvm-svn: 106622
 2010-06-23 08:45:32 +00:00
Eric Christopher
c6382036ef Update uses, defs, and comments for darwin tls patterns. 
llvm-svn: 106621
 2010-06-23 08:01:49 +00:00
Nick Lewycky
4d160cb75c Don't link against libm and libpthread which don't exist in BeOS/Haiku. Also, 
Haiku like Linux provides <regex.h>, so use it. Patch by Paul Davey!

llvm-svn: 106620
 2010-06-23 06:48:34 +00:00
Daniel Dunbar
e6ee7409b9 Revert r106066, "Create a more targeted fix for not sinking instructions into a range where it"... it causes bzip2 to be miscompiled by Clang. 
Conflicts:

	lib/CodeGen/MachineSink.cpp

llvm-svn: 106614
 2010-06-23 00:48:25 +00:00
Stuart Hastings
bf77774e2f Less incorrect handling of zero-length bitfields. Radars 7992077 and 8093043. 
llvm-svn: 106611
 2010-06-23 00:31:14 +00:00
Eric Christopher
0c4230b33c Get the addend correct for i386 pic. 
Thanks Daniel!

llvm-svn: 106608
 2010-06-22 23:51:47 +00:00
Bill Wendling
fb956a0e47 Generate DWARF information during Apple-style build. They'll be stripped out 
later on. But we need them saved in the symbols directory.

llvm-svn: 106604
 2010-06-22 23:44:15 +00:00
Dan Gohman
f44783ab4e Replace ScalarEvolution's private copy of getLoopPredecessor 
with LoopInfo's public copy.

llvm-svn: 106603
 2010-06-22 23:43:28 +00:00
Bruno Cardoso Lopes
db9027d95d Add AVX compare packed instructions 
llvm-svn: 106600
 2010-06-22 23:37:59 +00:00
Dan Gohman
c33529255a Loosen up this test so that it doesn't depend as much on register 
allocation details.

llvm-svn: 106599
 2010-06-22 23:32:47 +00:00
Devang Patel
1967a51efc Revert 106592 for now. It causes clang-selfhost build failure. 
llvm-svn: 106598
 2010-06-22 23:29:55 +00:00
Dan Gohman
2ac2a33a1b Fix OptimizeMax to handle an odd case where one of the max operands 
is another max which folds. This fixes PR7454.

llvm-svn: 106594
 2010-06-22 23:07:13 +00:00
Bruno Cardoso Lopes
424b206ad4 Reapply support for AVX unpack and interleave instructions, with 
testcases this time.

llvm-svn: 106593
 2010-06-22 23:02:38 +00:00
Devang Patel
a24bf5907c If a metadata operand is seeded in value map and the metadata should also be seeded in value map. This is not limited to function local metadata. 
Failure to seed metdata in such cases causes troubles when in a cloned module, metadata from a new module refers to values in old module. Usually this results in mysterious bugpoint crashes. For example,

 Checking to see if we can delete global inits: Unknown constant!
 UNREACHABLE executed at /d/g/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp:904!

llvm-svn: 106592
 2010-06-22 22:53:21 +00:00
Devang Patel
dca60347e5 While cloning a module, clone metadata attached with instructions. 
llvm-svn: 106591
 2010-06-22 22:50:42 +00:00