RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-22 20:43:44 +02:00
Code Issues Projects Releases Wiki Activity
59,933 Commits 4 Branches 5 Tags 704 MiB
a324da99ae
Commit Graph

38050 Commits

Author SHA1 Message Date
Evan Cheng
a324da99ae Do not try to optimize a copy that has already been marked for deletion. 
llvm-svn: 102027
 2010-04-21 20:57:54 +00:00
Devang Patel
293567822a Add command line option to disable debug info printing in .s file. This option does not impact debug info generation and preservation through earlier compile starges. 
llvm-svn: 102012
 2010-04-21 19:08:53 +00:00
Bob Wilson
7716410af6 Fix a performance problem with the new SSAUpdater. This showed up in the 
GCCAS time for MultiSource/Benchmarks/ASCI_Purple/SMG2000.

llvm-svn: 102009
 2010-04-21 18:39:03 +00:00
Johnny Chen
c23127cb7e Modified some assert() msg strings; no other functionality change. 
llvm-svn: 102008
 2010-04-21 18:37:48 +00:00
Jakob Stoklund Olesen
8cf10fe9e4 Add fast register allocator, enabled with -regalloc=fast. 
So far this is just a clone of -regalloc=local that has been lobotomized to run
25% faster. It drops the least-recently-used calculations, and is just plain
stupid when it runs out of registers.

The plan is to make this go even faster for -O0 by taking advantage of the short
live intervals in unoptimized code. It should not be necessary to calculate
liveness when most virtual registers are killed 2-3 instructions after they are
born.

llvm-svn: 102006
 2010-04-21 18:02:42 +00:00
Devang Patel
158b496cd1 Identify when a lexical scope is split in to multiple instruction ranges. Emit such ranges using DW_AT_ranges. 
llvm-svn: 102004
 2010-04-21 16:32:19 +00:00
Dan Gohman
4ca879bebb Make ScalarEvolution::getConstant support pointer types, for consistency 
with ScalarEvolution's overall approach to pointer types.

llvm-svn: 102003
 2010-04-21 16:04:04 +00:00
Evan Cheng
dbfb7dc438 Implement -disable-non-leaf-fp-elim which disable frame pointer elimination 
optimization for non-leaf functions. This will be hooked up to gcc's
-momit-leaf-frame-pointer option. rdar://7886181

llvm-svn: 101984
 2010-04-21 03:18:23 +00:00
Evan Cheng
0f4671b0dd isel (i32 anyext i16) as insert_subreg when 16-bit ops are being promoted. 
llvm-svn: 101979
 2010-04-21 01:47:12 +00:00
Evan Cheng
e67a37c4ce Trim include. 
llvm-svn: 101978
 2010-04-21 01:39:06 +00:00
Dan Gohman
4ff2d817a3 Add more const qualifiers on TargetMachine and friends. 
llvm-svn: 101977
 2010-04-21 01:34:56 +00:00
Dan Gohman
6bde42d9f5 Update CMakeLists.txt. 
llvm-svn: 101976
 2010-04-21 01:32:29 +00:00
Dan Gohman
1d3532d925 Move several SelectionDAG-independent utility functions out of the 
SelectionDAG directory and into a new Analysis.cpp file.

llvm-svn: 101975
 2010-04-21 01:22:34 +00:00
Johnny Chen
6e4b1607ee Thumb instructions which have reglist operands at the end and predicate operands 
before reglist were not properly handled with respect to IT Block.  Fix that by
creating a new method ARMBasicMCBuilder::DoPredicateOperands() used by those
instructions for disassembly.  Add a test case.

llvm-svn: 101974
 2010-04-21 01:01:19 +00:00
Chris Lattner
c840cfe5c9 Implement (but don't enable) PR6724 and rdar://6295824. In short, 
we have RefreshCallGraph detect when a function pass devirtualizes
a call, and have CGSCCPassMgr iterate (up to a count) when this 
happens.  This allows (in the example) GVN to devirtualize the 
call in foo, then the inliner to inline it away.

This is not currently enabled because I haven't done any analysis
on the (potentially substantial) code size or performance impact of
doing this, and guess what, it exposes callgraph updating bugs in
various passes.  This is progress though, and you can play with it
by passing -max-cg-scc-iterations=5 to opt.

llvm-svn: 101973
 2010-04-21 00:47:40 +00:00
Evan Cheng
a0c4b2952f - Clean up some crappy code which deals with coalescing of copies which look at 
extract_subreg / insert_subreg, etc.
- Add support for more aggressive insert_subreg coalescing.

llvm-svn: 101971
 2010-04-21 00:44:22 +00:00
Dan Gohman
4d1724c3e8 Revert r101471. For tight recursive functions which have multiple 
recursive callsites, inlining can reduce the number of calls by
exponential factors, as it does in
MultiSource/Benchmarks/Olden/treeadd. More involved heuristics
will be needed.

llvm-svn: 101969
 2010-04-21 00:43:30 +00:00
Bill Wendling
9bf4b878e0 Handle a displacement location in 64-bit as an RIP-relative displacement. It 
fixes a bug (<rdar://problem/7880900>) in the JIT. This code wouldn't work:

target triple = "x86_64-apple-darwin"

define double @func(double %a) {
  %tmp1 = fmul double %a, 5.000000e-01            ; <double> [#uses=1]
  ret double %tmp1
}

define i32 @main() nounwind {
  %1 = call double @func(double 4.770000e-04) ; <i64> [#uses=0]
  ret i32 0
}

llvm-svn: 101965
 2010-04-21 00:34:04 +00:00
Evan Cheng
69142a2a6a Rewrite machine cse to avoid recursion. 
llvm-svn: 101964
 2010-04-21 00:21:07 +00:00
Dan Gohman
570b621976 Add another variant of this test which found a place where 
CodeGen's ComputeMaskedBits was being over-conservative when computing
bits for an ADD.

llvm-svn: 101963
 2010-04-21 00:19:28 +00:00
Chris Lattner
6db0f451a7 teach the x86 address matching stuff to handle 
(shl (or x,c), 3) the same as (shl (add x, c), 3)
when x doesn't have any bits from c set.

This finishes off PR1135.  Before we compiled the block to:
to:

LBB0_3:                                 ## %bb
	cmpb	$4, %dl
	sete	%dl
	addb	%dl, %cl
	movb	%cl, %dl
	shlb	$2, %dl
	addb	%r8b, %dl
	shlb	$2, %dl
	movzbl	%dl, %edx
	movl	%esi, (%rdi,%rdx,4)
	leaq	2(%rdx), %r9
	movl	%esi, (%rdi,%r9,4)
	leaq	1(%rdx), %r9
	movl	%esi, (%rdi,%r9,4)
	addq	$3, %rdx
	movl	%esi, (%rdi,%rdx,4)
	incb	%r8b
	decb	%al
	movb	%r8b, %dl
	jne	LBB0_1

Now we produce:

LBB0_3:                                 ## %bb
	cmpb	$4, %dl
	sete	%dl
	addb	%dl, %cl
	movb	%cl, %dl
	shlb	$2, %dl
	addb	%r8b, %dl
	shlb	$2, %dl
	movzbl	%dl, %edx
	movl	%esi, (%rdi,%rdx,4)
	movl	%esi, 8(%rdi,%rdx,4)
	movl	%esi, 4(%rdi,%rdx,4)
	movl	%esi, 12(%rdi,%rdx,4)
	incb	%r8b
	decb	%al
	movb	%r8b, %dl
	jne	LBB0_1

llvm-svn: 101958
 2010-04-20 23:18:40 +00:00
Dale Johannesen
510282d54b Because of the EMMS problem, right now we have to support 
user-defined operations that use MMX register types, but
the compiler shouldn't generate them on its own.  This adds
a Synthesizable abstraction to represent this, and changes
the vector widening computation so it won't produce MMX types.
(The motivation is to remove noise from the ABI compatibility
part of the gcc test suite, which has some breakage right now.)

llvm-svn: 101951
 2010-04-20 22:34:09 +00:00
Devang Patel
ef601352ff Rename ValueMapTy as ValueToValueMapTy to clearly indicate that this has no replationship with ADT/ValueMap. 
llvm-svn: 101950
 2010-04-20 22:24:18 +00:00
Devang Patel
788913a131 There is no need to install ValueMapper.h header. 
llvm-svn: 101949
 2010-04-20 22:18:31 +00:00
Johnny Chen
d7ffae797f Better error-handling of getBitFieldInvMask() where msb < lsb (encoding error), 
instead of just asserting.

llvm-svn: 101942
 2010-04-20 21:29:28 +00:00
Jakob Stoklund Olesen
1ba4c9d2c2 When MachineLICM is hoisting a physical register after regalloc, make sure the 
register is not killed in the loop.

This fixes 188.ammp on ARM where the post-ra scheduler would grab a register
that looked available but wasn't.

A testcase would be huge and fragile, sorry.

llvm-svn: 101930
 2010-04-20 18:45:47 +00:00
Johnny Chen
bd5bf58029 For t2LDRT, t2LDRBT, t2LDRHT, t2LDRSBT, and t2LDRSHT, if Rn(Inst{19-16})=='1111', 
transform the Opcode to the corresponding t2LDR*pci counterpart.

Ref: A8.6.86 LDRT, A8.6.65 LDRBT, A8.6.77 LDRHT, A8.6.81 LDRSBT, A8.6.85 LDRSHT
llvm-svn: 101915
 2010-04-20 17:28:50 +00:00
Evan Cheng
ebe1fbf676 Typo. 
llvm-svn: 101914
 2010-04-20 17:27:38 +00:00
Dan Gohman
44ab9ed9b8 Sink the CopyToExportRegsIfNeeded calls out of SelectionDAGISel 
into SelectionDAGBuilder. This avoids a separate pass over the
instructions, and has the side effect of providing debug location
information to the copy.

llvm-svn: 101906
 2010-04-20 15:03:56 +00:00
Dan Gohman
3915c2cc7c Don't send PHI nodes down to SelectionDAGBuilder of FastISel, since 
they end up doing nothing.

llvm-svn: 101904
 2010-04-20 15:00:41 +00:00
Dan Gohman
0c1f92afe1 Sink this use_empty() check into isUsedOutsideOfDefiningBlock. 
llvm-svn: 101902
 2010-04-20 14:50:13 +00:00
Dan Gohman
c276def9e9 If a PHI node somehow has debug info, propogate it to the MachineInstr PHI. 
llvm-svn: 101901
 2010-04-20 14:48:02 +00:00
Dan Gohman
47859e6848 Don't iterate through the whole block just to find the PHI nodes. 
llvm-svn: 101900
 2010-04-20 14:46:25 +00:00
Gabor Greif
2eaf8afcff use abstract accessors to CallInst 
llvm-svn: 101899
 2010-04-20 13:13:04 +00:00
Benjamin Kramer
b61f85894f PR6880: Don't dereference CallsExternalNode if it's NULL. 
llvm-svn: 101897
 2010-04-20 12:16:50 +00:00
Chris Lattner
b66b0c36cd Bill's change in r95336 broke empty aggregates embedded 
in other types.  fix this by only bumping zero-byte globals
up to a single byte if the *entire global* is zero size,
fixing PR6340.

This also fixes empty arrays etc to be handled correctly,
and only does this on subsection-via-symbols targets (aka
darwin) which is the only place where this matters.

llvm-svn: 101879
 2010-04-20 06:20:21 +00:00
Chris Lattner
04fb51984f teach cellspu how to return i8 and i16 from calls, 
patch by Kalle Raiskila!

llvm-svn: 101875
 2010-04-20 05:36:09 +00:00
Chris Lattner
5041463692 remove a bunch of ad-hoc code to simplify instructions from 
loop unswitch, and use inst simplify instead.  It is more
powerful and less duplication.

llvm-svn: 101874
 2010-04-20 05:33:18 +00:00
Chris Lattner
b6b95141ec move some select simplifications out out instcombine into 
inst simplify.  No functionality change.

llvm-svn: 101873
 2010-04-20 05:32:14 +00:00
Chris Lattner
e5a995a834 RewriteLoopBodyWithConditionConstant can end up rewriting the 
condition we're unswitching on.  In this case, don't try to
simplify the second copy of the loop which may be dead or not,
but is probably a constant now.  This fixes PR6879

llvm-svn: 101870
 2010-04-20 05:09:16 +00:00
Chris Lattner
7a2ce4afd8 reapply 'reject forward references to functions whose type don't match' 
now that the testsuite has been updated.

llvm-svn: 101866
 2010-04-20 04:49:11 +00:00
Dan Gohman
5f8489396e Delete a redundant return statement. 
llvm-svn: 101860
 2010-04-20 01:58:20 +00:00
Bill Wendling
887dac2aa6 The visitXOR method can return the same SDNode. If so, we don't want to delete 
it as it's not dead.

llvm-svn: 101855
 2010-04-20 01:25:01 +00:00
Chris Lattner
339a3cd59b disable optimizations in this directory for MSVC9. This avoids 
an optimizer infinite loop on the file, PR6866.

llvm-svn: 101854
 2010-04-20 01:11:32 +00:00
Johnny Chen
4f388a02d6 Better error-handling for DisassembleThumb2DPModImm() with 2-reg operands where 
d==15 is considered illegal.  Return false instead of assert().

llvm-svn: 101852
 2010-04-20 01:01:57 +00:00
Eric Christopher
53e7e0fcfb Remove the palignr intrinsics now that we lower them to vector shuffles, 
shifts and null vectors. Autoupgrade these to what we'd lower them to.

Add a testcase to exercise this.

llvm-svn: 101851
 2010-04-20 00:59:54 +00:00
Dan Gohman
5d47f180a2 Remove this debug output; it isn't that useful, and it's incomplete 
in the case where a basic block is split.

llvm-svn: 101850
 2010-04-20 00:56:44 +00:00
Dan Gohman
e8387b1250 Sink DebugLoc handling out of SelectionDAGISel into FastISel and 
SelectionDAGBuilder, where it doesn't have to be as complicated.

llvm-svn: 101848
 2010-04-20 00:48:35 +00:00
Chris Lattner
2f4001f3a8 make CallGraphNode dtor abort if a node is deleted when there are still 
references to it.

llvm-svn: 101847
 2010-04-20 00:47:34 +00:00
Chris Lattner
92df9f46f0 make the inliner do less work for leaf functions. 
llvm-svn: 101846
 2010-04-20 00:47:08 +00:00