Commit Graph

12867 Commits

Author SHA1 Message Date
Benjamin Kramer
b20a8fc8a6 X86: Try to use a smaller encoding by transforming (X << C1) & C2 into (X & (C2 >> C1)) & C1. (Part of PR5039)
This tends to happen a lot with bitfield code generated by clang. A simple example for x86_64 is
uint64_t foo(uint64_t x) { return (x&1) << 42; }
which used to compile into bloated code:
	shlq	$42, %rdi               ## encoding: [0x48,0xc1,0xe7,0x2a]
	movabsq	$4398046511104, %rax    ## encoding: [0x48,0xb8,0x00,0x00,0x00,0x00,0x00,0x04,0x00,0x00]
	andq	%rdi, %rax              ## encoding: [0x48,0x21,0xf8]
	ret                             ## encoding: [0xc3]

with this patch we can fold the immediate into the and:
	andq	$1, %rdi                ## encoding: [0x48,0x83,0xe7,0x01]
	movq	%rdi, %rax              ## encoding: [0x48,0x89,0xf8]
	shlq	$42, %rax               ## encoding: [0x48,0xc1,0xe0,0x2a]
	ret                             ## encoding: [0xc3]

It's possible to save another byte by using 'andl' instead of 'andq' but I currently see no way of doing
that without making this code even more complicated. See the TODOs in the code.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129990 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-22 15:30:40 +00:00
Evan Cheng
db6cbe1ff1 In Thumb2 mode, lower frame indix references to:
add <rd>, sp, #<imm8>
ldr <rd>, [sp, #<imm8>]
When the offset from sp is multiple of 4 and in range of 0-1020.
This saves code size by utilizing 16-bit instructions.

rdar://9321541


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129971 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-22 01:42:52 +00:00
Devang Patel
71f3f1146f Fix DWARF description of Q registers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129952 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-21 23:22:35 +00:00
Devang Patel
27f5acb7d4 Fix DWARF description of S registers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129947 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-21 22:48:26 +00:00
Devang Patel
8859df5911 Test case for r129922
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129934 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-21 20:16:43 +00:00
Rafael Espindola
1ac7fe0f4d Fix relative relocations. This is sufficient for running the rust testsuite with
MC :-)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129923 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-21 18:36:50 +00:00
Daniel Dunbar
63c21deee1 Revert r1296656, "Fix rdar://9289512 - not folding load into compare at -O0...",
which broke a couple GCC test suite tests at -O0.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129914 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-21 16:14:46 +00:00
Che-Liang Chiou
5efde18dd2 ptx: fix parameter ordering
This patch depends on the prior fix r129908 that changes to use std::find,
rather than std::binary_search, on unordered array.

Patch by Dan Bailey


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129909 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-21 10:56:58 +00:00
Evan Cheng
c8578948c9 Remove -use-divmod-libcall. Let targets opt in when they are available.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129884 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-20 22:20:12 +00:00
Cameron Zwarich
344731c018 Fix another case of <rdar://problem/9184212> that only occurs with code
generated by llvm-gcc, since llvm-gcc uses 2 i64s for passing a 4 x float
vector on ARM rather than an i64 array like Clang.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129878 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-20 21:48:38 +00:00
Stuart Hastings
2575a9c606 Un-XFAIL this test for ARM. <rdar://problem/7662569>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129875 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-20 21:47:45 +00:00
Justin Holewinski
e1fee48cd0 PTX: Add intrinsics to list of built-in intrinsics, which allows them to be
used by Clang.  To help Clang integration, the PTX target has been split
     into two targets: ptx32 and ptx64, depending on the desired pointer size.

- Add GCCBuiltin class to all intrinsics
- Split PTX target into ptx32 and ptx64

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129851 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-20 15:37:17 +00:00
Rafael Espindola
3660a847f1 Behave like gnu as when a relocation crosses sections.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129850 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-20 14:01:45 +00:00
Eric Christopher
abbbfbd672 Rewrite the expander for umulo/smulo to remember to sign extend the input
manually and pass all (now) 4 arguments to the mul libcall. Add a new
ExpandLibCall for just this (copied gratuitously from type legalization).

Fixes rdar://9292577


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129842 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-20 01:19:45 +00:00
Daniel Dunbar
d285139e0e llc: Eliminate a use of getDarwinMajorNumber().
- As before, there is a minor semantic change here (evidenced by the test
   change) for Darwin triples that have no version component. I debated changing
   the default behavior of isOSVersionLT, but decided it made more sense for
   triples to be explicit.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129805 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 20:46:13 +00:00
Daniel Dunbar
ebc5066b9b CodeGen: Eliminate a use of getDarwinMajorNumber().
- There is a minor semantic change here (evidenced by the test change) for
   Darwin triples that have no version component. I debated changing the default
   behavior of isOSVersionLT, but decided it made more sense for triples to be
   explicit.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129802 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 20:32:39 +00:00
Bob Wilson
84c5eed15b This patch combines several changes from Evan Cheng for rdar://8659675.
Making use of VFP / NEON floating point multiply-accumulate / subtraction is
difficult on current ARM implementations for a few reasons.
1. Even though a single vmla has latency that is one cycle shorter than a pair
   of vmul + vadd, a RAW hazard during the first (4? on Cortex-a8) can cause
   additional pipeline stall. So it's frequently better to single codegen
   vmul + vadd.
2. A vmla folowed by a vmul, vmadd, or vsub causes the second fp instruction to
   stall for 4 cycles. We need to schedule them apart.
3. A vmla followed vmla is a special case. Obvious issuing back to back RAW
   vmla + vmla is very bad. But this isn't ideal either:
     vmul
     vadd
     vmla
   Instead, we want to expand the second vmla:
     vmla
     vmul
     vadd
   Even with the 4 cycle vmul stall, the second sequence is still 2 cycles
   faster.

Up to now, isel simply avoid codegen'ing fp vmla / vmls. This works well enough
but it isn't the optimial solution. This patch attempts to make it possible to
use vmla / vmls in cases where it is profitable.

A. Add missing isel predicates which cause vmla to be codegen'ed.
B. Make sure the fmul in (fadd (fmul)) has a single use. We don't want to
   compute a fmul and a fmla.
C. Add additional isel checks for vmla, avoid cases where vmla is feeding into
   fp instructions (except for the #3 exceptional case).
D. Add ARM hazard recognizer to model the vmla / vmls hazards.
E. Add a special pre-regalloc case to expand vmla / vmls when it's likely the
   vmla / vmls will trigger one of the special hazards.

Enable these fp vmlx codegen changes for Cortex-A9.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129775 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 18:11:57 +00:00
Bob Wilson
cd70496ad1 Add -mcpu=cortex-a9-mp. It's cortex-a9 with MP extension. rdar://8648637.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129774 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 18:11:52 +00:00
Bob Wilson
5dde893c2b Avoid some 's' 16-bit instruction which partially update CPSR
(and add false dependency) when it isn't dependent on last CPSR defining
instruction. rdar://8928208

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129773 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 18:11:49 +00:00
Bob Wilson
f6a4d3c2f3 Avoid write-after-write issue hazards for Cortex-A9.
Add a avoidWriteAfterWrite() target hook to identify register classes that
suffer from write-after-write hazards. For those register classes, try to avoid
writing the same register in two consecutive instructions.

This is currently disabled by default.  We should not spill to avoid hazards!
The command line flag -avoid-waw-hazard can be used to enable waw avoidance.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129772 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 18:11:45 +00:00
Eli Friedman
3762046dbf Add support for FastISel'ing varargs calls.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129765 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 17:22:22 +00:00
Jakob Stoklund Olesen
ea70a06fa7 Tighten test case a bit.
Ideally, we would match an S-register to its containing D-register, but that
requires arithmetic (divide by 2).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129756 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 06:14:45 +00:00
Chris Lattner
832e494359 Implement support for x86 fastisel of small fixed-sized memcpys, which are generated
en-mass for C++ PODs.  On my c++ test file, this cuts the fast isel rejects by 10x 
and shrinks the generated .s file by 5%


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129755 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 05:52:03 +00:00
Chris Lattner
b44101c140 Implement support for fast isel of calls of i1 arguments, even though they are illegal,
when they are a truncate from something else.  This eliminates fully half of all the 
fastisel rejections on a test c++ file I'm working with, which should make a substantial
improvement for -O0 compile of c++ code.

This fixed rdar://9297003 - fast isel bails out on all functions taking bools


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129752 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 05:09:50 +00:00
Chris Lattner
e03b8d3162 Handle i1/i8/i16 constant integer arguments to calls by prepromoting them.
Before we would bail out on i1 arguments all together, now we just bail on
non-constant ones.  Also, we used to emit extraneous code.  e.g. test12 was:

	movb	$0, %al
	movzbl	%al, %edi
	callq	_test12

and test13 was:
	movb	$0, %al
	xorl	%edi, %edi
	movb	%al, 7(%rsp)
	callq	_test13f

Now we get:

	movl	$0, %edi
	callq	_test12
and:
	movl	$0, %edi
	callq	_test13f



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129751 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 04:42:38 +00:00
Chris Lattner
c76d121807 be layout aware, to produce:
testb	$1, %al
	je	LBB0_2
## BB#1:                                ## %if.then
	movb	$0, %al

instead of:

	testb	$1, %al
	jne	LBB0_1
	jmp	LBB0_2
LBB0_1:                                 ## %if.then
	movb	$0, %al

how 'bout that.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129749 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 04:26:32 +00:00
Chris Lattner
90cb88a9b4 fix rdar://9297006 - fast isel bails out on trunc to i1 -> bools cry,
a common cause of fast isel rejects on c++ code.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129748 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 04:22:17 +00:00
Jakob Stoklund Olesen
775c3e5824 Make tests register allocation independent again.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129739 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 00:14:43 +00:00
Evan Cheng
b58a340fa2 Do not lose mem_operands while lowering VLD / VST intrinsics.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129738 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 00:04:03 +00:00
Devang Patel
1fc7a24c25 Remove test to check line numbers. There are other numerous tests in our test harness to check line number information.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129725 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-18 22:27:20 +00:00
Eric Christopher
d574bb5a6e Fix a bug where we were counting the alias sets as completely used
registers for fast allocation a different way. This has us updating
used registers only when we're using that exact register.

Fixes rdar://9207598



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129711 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-18 19:26:25 +00:00
Chris Lattner
f051c1a29d while we're at it, handle 'sdiv exact' of a power of 2 also,
this fixes a few rejects on c++ iterator loops.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129694 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-18 07:00:40 +00:00
Chris Lattner
090ca9108b fix rdar://9297011 - udiv by power of two causing fast-isel rejects
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129693 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-18 06:55:51 +00:00
Chris Lattner
1518afddea Implement major new fastisel functionality: the matcher can now handle immediates with
value constraints on them (when defined as ImmLeaf's).  This is particularly important
for X86-64, where almost all reg/imm instructions take a i64immSExt32 immediate operand,
which has a value constraint.  Before this patch we ended up iseling the examples into
such amazing code as:

	movabsq	$7, %rax
	imulq	%rax, %rdi
	movq	%rdi, %rax
	ret

now we produce:

	imulq	$7, %rdi, %rax
	ret

This dramatically shrinks the generated code at -O0 on x86-64.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129691 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-18 06:22:33 +00:00
Chris Lattner
1023643d50 relax this test to just check that the lock prefix is encoded properly,
and to not rely on the register allocator's arbitrary operand choices.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129690 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-18 06:15:35 +00:00
Chris Lattner
602fc06817 1. merge fast-isel-shift-imm.ll into fast-isel-x86-64.ll
2. implement rdar://9289501 - fast isel should fold trivial multiplies to shifts
3. teach tblgen to handle shift immediates that are different sizes than the 
   shifted operands, eliminating some code from the X86 fast isel backend.
4. Have FastISel::SelectBinaryOp use (the poorly named) FastEmit_ri_ function
   instead of FastEmit_ri to simplify code.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129666 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-17 20:23:29 +00:00
Chris Lattner
0a1c997c27 fix an x86 fast isel issue where we'd completely give up on folding an address
when we have a global variable base an an index.  Instead, just give up on
folding the global variable.

Before we'd geenrate:

_test:                                  ## @test
## BB#0:
	movq	_rtx_length@GOTPCREL(%rip), %rax
	leaq	(%rax), %rax
	addq	%rdi, %rax
	movzbl	(%rax), %eax
	ret

now we generate:

_test:                                  ## @test
## BB#0:
	movq	_rtx_length@GOTPCREL(%rip), %rax
	movzbl	(%rax,%rdi), %eax
	ret

The difference is even more significant when there is a scale
involved.

This fixes rdar://9289558 - total fail with addr mode formation at -O0/x86-64


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129664 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-17 17:47:38 +00:00
Chris Lattner
685090f598 fix an oversight which caused us to compile the testcase (and other
less trivial things) into a dummy lea.  Before we generated:

_test:                                  ## @test
	movq	_G@GOTPCREL(%rip), %rax
	leaq	(%rax), %rax
	ret

now we produce:

_test:                                  ## @test
	movq	_G@GOTPCREL(%rip), %rax
	ret

This is part of rdar://9289558



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129662 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-17 17:12:08 +00:00
Chris Lattner
fd3f635103 Fix rdar://9289512 - not folding load into compare at -O0
The basic issue here is that bottom-up isel is matching the branch
and compare, and was failing to fold the load into the branch/compare
combo.  Fixing this (by allowing folding into any instruction of a
sequence that is selected) allows us to produce things like:


cmpb    $0, 52(%rax)
je      LBB4_2

instead of:

movb    52(%rax), %cl
cmpb    $0, %cl
je      LBB4_2

This makes the generated -O0 code run a bit faster, but also speeds up
compile time by putting less pressure on the register allocator and 
generating less code.

This was one of the biggest classes of missing load folding.  Implementing
this shrinks 176.gcc's c-decl.s (as a random example) by about 4% in (verbose-asm)
line count.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129656 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-17 06:35:44 +00:00
Eli Friedman
2f108f81c1 Remove working entry from README.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129654 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-17 02:36:27 +00:00
Chris Lattner
fff65b354f fix rdar://9289583 - fast isel should handle non-canonical commutative binops
allowing us to fold the immediate into the 'and' in this case:

int test1(int i) {
  return 8&i;
}



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129653 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-17 01:16:47 +00:00
Eli Friedman
e545d38a28 PR9055: extend the fix to PR4050 (r70179) to apply to zext and anyext.
Returning a new node makes the code try to replace the old node, which
in the included testcase is killed by CSE.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129650 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-16 23:25:34 +00:00
Frits van Bommel
338af5e749 Add test cases for Jay's r129641 and fix a 32-bit-centric testcase in a file with a 64-bit datalayout.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129643 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-16 14:31:50 +00:00
Evan Cheng
65279cb9bd Fix divmod libcall lowering. Convert to {S|U}DIVREM first and then expand the node to a libcall. rdar://9280991
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129633 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-16 03:08:26 +00:00
Johnny Chen
3a96122c4a Thumb2 BFC was insufficiently encoded.
rdar://problem/9292717


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129619 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 22:52:15 +00:00
Johnny Chen
a704bc9354 A8.6.315 VLD3 (single 3-element structure to all lanes)
The a bit must be encoded as 0.

rdar://problem/9292625


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129618 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 22:49:08 +00:00
Akira Hatanaka
fc693036a4 Re-enable test o32_cc_vararg.ll.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129616 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 22:23:09 +00:00
Cameron Zwarich
0cb11ac32f Add ORR and EOR to the CMP peephole optimizer. It's hard to get isel to generate
a case involving EOR, so I only added a test for ORR.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129610 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 21:24:38 +00:00
Rafael Espindola
27c4ba16ae Add this test back for Darwin.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129607 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 21:06:27 +00:00
Cameron Zwarich
b485de5d8c The AND instruction leaves the V flag unmodified, so it falls victim to the same
problem as all of the other instructions we fold with CMPs.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129602 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 20:45:00 +00:00
Cameron Zwarich
ca3f6a3925 Add missing register forms of instructions to the ARM CMP-folding code. This
fixes <rdar://problem/9287901>.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129599 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 20:28:28 +00:00
Akira Hatanaka
99a2e98edd Add pass that expands pseudo instructions into target instructions after register allocation. Define pseudos that get expanded into mtc1 or mfc1 instructions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129594 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 19:52:08 +00:00
Joerg Sonnenberger
c25e8d8cea Add encoding tests for flds/filds
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129589 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 19:25:31 +00:00
Rafael Espindola
f0adba9a7e Add 129518 back with a fix for when we are producing eh just because of debug info.
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.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129571 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 15:11:06 +00:00
Chris Lattner
7a2bdde0a0 Fix a ton of comment typos found by codespell. Patch by
Luis Felipe Strano Moraes!



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129558 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 05:18:47 +00:00
NAKAMURA Takumi
bcb8c6d09e Revert r129518, "Change ELF systems to use CFI for producing the EH tables. This reduces the"
It broke several builds.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129557 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 03:35:57 +00:00
Evan Cheng
9eec66e604 Fix another fcopysign lowering bug. If src is f64 and destination is f32, don't
forget to right shift the source by 32 first. rdar://9287902


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129556 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 01:31:00 +00:00
Michael J. Spencer
4babeeeeed Add 3DNow! intrinsics.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129551 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 00:32:41 +00:00
Johnny Chen
de29a52940 The ARM disassembler did not handle the alignment correctly for VLD*DUP* instructions
(single element or n-element structure to all lanes).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129550 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15 00:10:45 +00:00
Evan Cheng
06b2a60ef9 Follow up on r127913. Fix Thumb revsh isel. rdar://9286766
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129548 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-14 23:27:44 +00:00
Eli Friedman
3ec01b7dac Add an instcombine for constructs like a | -(b != c); a select is more
canonical, and generally leads to better code.  Found while looking at
an article about saturating arithmetic.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129545 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-14 22:41:27 +00:00
Owen Anderson
f6832bbda0 Fix an infinite alternation in JumpThreading where two transforms would repeatedly undo each other. The solution is to perform more aggressive constant folding to make one of the edges just folded away rather than trying to thread it.
Fixes <rdar://problem/9284786>.

Discovered with CSmith.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129538 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-14 21:35:50 +00:00
Johnny Chen
cd695fdac1 Add sanity checkings for Thumb2 Load/Store Register Exclusive family of operations.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129531 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-14 19:13:28 +00:00
Daniel Dunbar
2cb2aa610e tests: Remove a FrontendC test which is no longer valid.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129519 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-14 15:21:16 +00:00
Rafael Espindola
3dae6e7333 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.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129518 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-14 15:18:53 +00:00
Andrew Trick
12f0dc6bb5 In the pre-RA scheduler, maintain cmp+br proximity.
This is done by pushing physical register definitions close to their
use, which happens to handle flag definitions if they're not glued to
the branch. This seems to be generally a good thing though, so I
didn't need to add a target hook yet.

The primary motivation is to generate code closer to what people
expect and rule out missed opportunity from enabling macro-op
fusion. As a side benefit, we get several 2-5% gains on x86
benchmarks. There is one regression:
SingleSource/Benchmarks/Shootout/lists slows down be -10%. But this is
an independent scheduler bug that will be tracked separately.
See rdar://problem/9283108.

Incidentally, pre-RA scheduling is only half the solution. Fixing the
later passes is tracked by:
<rdar://problem/8932804> [pre-RA-sched] on x86, attempt to schedule CMP/TEST adjacent with condition jump

Fixes:
<rdar://problem/9262453> Scheduler unnecessary break of cmp/jump fusion


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129508 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-14 05:15:06 +00:00
Bill Wendling
d336de318e As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
(movzx/movsx) because they give more information. Revert that part of the patch.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129498 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-14 01:46:37 +00:00
Bill Wendling
c6df9883da Have the X86 back-end emit the alias instead of what's being aliased. In most
cases, it's much nicer and more informative reading the alias.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129497 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-14 01:11:51 +00:00
Johnny Chen
e8d087ad35 Thumb disassembler did not handle tBRIND (indirect branch) properly.
rdar://problem/9280370


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129480 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 21:59:01 +00:00
Mon P Wang
be0761c820 Vectors with different number of elements of the same element type can have
the same allocation size but different primitive sizes(e.g., <3xi32> and
<4xi32>).  When ScalarRepl promotes them, it can't use a bit cast but
should use a shuffle vector instead.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129472 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 21:40:02 +00:00
Johnny Chen
6c7e4147dc Check for unallocated instruction encodings when disassembling Thumb Branch instructions (tBcc and t2Bcc).
rdar://problem/9280470


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129471 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 21:35:49 +00:00
Johnny Chen
471d73d5d3 The LDR*T/STR*T (unpriviledged load/store) operations don't take SP or PC as Rt.
rdar://problem/9279440


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129469 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 21:04:32 +00:00
Cameron Zwarich
5af60ce2a8 Fix a typo in an ARM-specific DAG combine. This fixes <rdar://problem/9278274>.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129468 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 21:01:19 +00:00
Cameron Zwarich
1335022e19 Fix a regression caused by r102515 where explicit alignment on globals is
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>.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129466 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 20:36:04 +00:00
Johnny Chen
9bb386a933 Check the corner cases for t2LDRSHi12 correctly and mark invalid encodings as such.
rdar://problem/9276651


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129462 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 19:46:05 +00:00
Johnny Chen
119af20c7b Fix a bug where for t2MOVCCi disassembly, the TIED_TO register operand was not properly handled.
rdar://problem/9276427


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129456 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 17:51:02 +00:00
Cameron Zwarich
eb04a33dc7 Fix an obvious problem with an alignment computation. AsmPrinter actually does
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.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129432 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 09:02:43 +00:00
Cameron Zwarich
d8b88d8558 If a global variable has a specified alignment that is less than the preferred
alignment for its type, use the minimum of the specified alignment and the ABI
alignment. This fixes <rdar://problem/9275290>.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129428 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 06:03:16 +00:00
Andrew Trick
87896d9368 Recommit r129383. PreRA scheduler heuristic fixes: VRegCycle, TokenFactor latency.
Additional fixes:
Do something reasonable for subtargets with generic
itineraries by handle node latency the same as for an empty
itinerary. Now nodes default to unit latency unless an itinerary
explicitly specifies a zero cycle stage or it is a TokenFactor chain.

Original fixes:
UnitsSharePred was a source of randomness in the scheduler: node
priority depended on the queue data structure. I rewrote the recent
VRegCycle heuristics to completely replace the old heuristic without
any randomness. To make the ndoe latency adjustments work, I also
needed to do something a little more reasonable with TokenFactor. I
gave it zero latency to its consumers and always schedule it as low as
possible.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129421 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 00:38:32 +00:00
Bill Wendling
f93f7b2446 Reapply r129401 with patch for clang.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129419 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 00:36:11 +00:00
Eric Christopher
d31df87f41 Temporarily revert r129408 to see if it brings the bots back.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129417 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 00:20:59 +00:00
Johnny Chen
55e6419b12 Add sanity check for Ld/St Dual forms of Thumb2 instructions.
rdar://problem/9273947


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129411 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 23:31:00 +00:00
Eric Christopher
8e98de9979 Fix a bug where we were counting the alias sets as completely used
registers for fast allocation.

Fixes rdar://9207598


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129408 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 23:23:14 +00:00
Bill Wendling
f9b2dc66c8 Revert r129401 for now. Clang is using the old way of doing things.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129403 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 22:59:27 +00:00
Bill Wendling
d5f323d70b Remove the unaligned load intrinsics in favor of using native unaligned loads.
Now that we have a first-class way to represent unaligned loads, the unaligned
load intrinsics are superfluous.

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


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129401 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 22:46:31 +00:00
Oscar Fuentes
2a2bc5ccee Fix compiler command line used by lit.py when working with NMake
generators. It may improve robustness when testing from VS too.

Based on a patch by David Neto!

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129398 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 22:10:38 +00:00
Johnny Chen
ec51a6225c The Thumb2 RFE instructions need to have their second halfword fully specified.
In addition, the base register is not rGPR, but GPR with th exception that:

    if n == 15 then UNPREDICTABLE

rdar://problem/9273836


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129391 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 21:41:51 +00:00
Johnny Chen
32cefad4b3 Add bad register checks for Thumb2 Ld/St instructions.
rdar://problem/9269047


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129387 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 21:17:51 +00:00
Andrew Trick
c558bf3972 Revert 129383. It causes some targets to hit a scheduler assert.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129385 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 20:14:07 +00:00
Andrew Trick
3eb4319313 PreRA scheduler heuristic fixes: VRegCycle, TokenFactor latency.
UnitsSharePred was a source of randomness in the scheduler: node
priority depended on the queue data structure. I rewrote the recent
VRegCycle heuristics to completely replace the old heuristic without
any randomness. To make these heuristic adjustments to node latency work,
I also needed to do something a little more reasonable with TokenFactor. I
gave it zero latency to its consumers and always schedule it as low as
possible.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129383 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 19:54:36 +00:00
Johnny Chen
f9ce2cba42 The Thumb2 Ld, St, and Preload instructions with the i12 forms should have its Inst{23}
be specified as '1' (add = TRUE).

Also add a utility function for Thumb2.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129377 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 18:48:00 +00:00
Johnny Chen
49fdfe3ce5 Print out a debug message when the reglist fails the sanity check for Thumb Ld/St Multiple.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129365 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 17:09:04 +00:00
Rafael Espindola
25f492e778 Fix the case of a .cfi_rel_offset before any .cfi_def_cfa_offset.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129362 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 16:12:03 +00:00
Rafael Espindola
c57543964d Implement .cfi_same_value.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129361 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 15:31:05 +00:00
Cameron Zwarich
d0aacbcc2e Split a store of a VMOVDRR into two integer stores to avoid mixing NEON and ARM
stores of arguments in the same cache line. This fixes the second half of
<rdar://problem/8674845>.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129345 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 02:24:17 +00:00
Johnny Chen
163b6eaf25 Add one test case (svc).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129327 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 00:21:48 +00:00
Eric Christopher
05f9e4e8bd Match case for invalid constant error messages and add a new
test for invalid hexadecimals.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129326 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 00:18:03 +00:00
Johnny Chen
e77f72d7d2 A8.6.16 B
Encoding T1 (tBcc)
if cond == '1110' then UNDEFINED;

rdar://problem/9268681


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129325 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 00:14:49 +00:00
Dan Gohman
dac5dbadeb Fix reassociate to use a worklist instead of recursing when new
reassociation opportunities are exposed. This fixes a bug where
the nested reassociation expects to be the IR to be consistent,
but it isn't, because the outer reassociation has disconnected
some of the operands.  rdar://9167457


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129324 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 00:11:56 +00:00
Eric Christopher
164254d77c Test for invalid constant expr addition - bad octal constant.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129323 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 00:03:38 +00:00