Commit Graph

18403 Commits

Author SHA1 Message Date
Benjamin Kramer
6e35e4c8c1 Remove noisy semicolons.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131724 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-20 09:20:25 +00:00
Akira Hatanaka
cd0f90f831 Fix bug in which nodes that write to argument registers do not get glued with the JALR node. Patch by Sasa Stankovic
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131714 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-20 02:30:51 +00:00
Akira Hatanaka
d992f6c666 Remove code that creates unnecessary frame objects.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131711 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-20 01:45:06 +00:00
Akira Hatanaka
da0a3571d6 Define variables and functions in MipsFunctionInfo.
This is the first of a series of patches that attempt to simplify handling of 
stack frame objects. 


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131710 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-20 01:17:58 +00:00
Chad Rosier
a166089f36 Don't attempt to tail call optimize for Win64.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131709 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-20 00:59:28 +00:00
Evan Cheng
2e6496026f Revert r131664 and fix it in instcombine instead. rdar://9467055
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131708 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-20 00:54:37 +00:00
Eli Friedman
dc51575a5f Add fast-isel support for zeroext and signext ret instructions on x86.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131689 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 22:16:13 +00:00
Eric Christopher
2bbecd8f6d Oddly people want to use the 'r' constraint for fp constants on x86.
Fixes rdar://9218925
Fixes PR9601


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131682 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 21:33:47 +00:00
Jason W Kim
861b9c6a39 This fixes one divergence between LLVM and binutils for ARM in the
text section.

Assume the following bit of annotated assembly:

.section	.data.rel.ro,"aw",%progbits
.align	2
.LAlpha:
.long	startval(GOTOFF)

.text
.align	2
.type	main,%function
.align	4

main: ;;; assume "main" starts at offset 0x20
0x0	push	{r11, lr}
0x4	movw	r0, :lower16:(.LAlpha-(.LBeta+8))
;;; ==> (.AddrOf(.LAlpha) - ((.AddrOf(.LBeta) - .AddrOf(".")) + 8)
;;; ==> (??? - ((16-4) + 8) = -20
0x8	movt	r0, :upper16:(.LAlpha-(.LBeta+8))
;;; ==> (.AddrOf(.LAlpha) - ((.AddrOf(.LBeta) - .AddrOf(".")) + 8)
;;; ==> (??? - ((16-8) + 8) = -16
0xc	... blah

.LBeta:
0x10	add	r0, pc, r0
0x14	... blah

.LGamma:
0x18	add	r1, pc, r1

Above snippet results in the following relocs in the .o file for the
first pair of movw/movt instructions

00000024 R_ARM_MOVW_PREL_NC .LAlpha
00000028 R_ARM_MOVT_PREL .LAlpha

And the encoded instructions in the .o file for main: must be

00000020 <main>:
20:	e92d4800 push	{fp, lr}
24:	e30f0fec movw	r0, #65516	; 0xffec i.e. -20
28:	e34f0ff0 movt	r0, #65520	; 0xfff0 i.e. -16

However, llc (prior to this commit) generates the following sequence

00000020 <main>:
20:	e92d4800 push	{fp, lr}
24:	e30f0fec movw	r0, #65516	; 0xffec - i.e. -20
28:	e34f0fff movt	r0, #65535	; 0xffff - i.e. -1

What has to happen in the ArmAsmBackend is that if the relocation is PC
relative, the 16 bits encoded as part of movw and movt must be both addends,
not addresses. It makes sense to encode addresses by right shifting the value
by 16, but the result is incorrect for PIC.
i.e., the right shift by 16 for movt is ONLY valid for the NON-PCRel case.

This change agrees with what GNU as does, and makes the PIC code run.

MC/ARM/elf-movt.s covers this case.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131674 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 20:55:25 +00:00
Rafael Espindola
a3bff99f0a ADD64ri32 sign extends its argument, so we need to use a R_X86_64_32S.
Fixes PR9934.

We really need to start tblgening the relocation info :-(

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131669 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 20:32:34 +00:00
Akira Hatanaka
a1a7ba8382 Align i64 arguments to 64 bit boundaries.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131668 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 20:29:48 +00:00
Evan Cheng
0efaa5e6a1 crc32 with 64-bit output zeros upper 32-bits. rdar://9467055
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131664 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 18:57:12 +00:00
Akira Hatanaka
59d266029c Increase number of available registers when target is MIPS32.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131660 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 18:25:03 +00:00
Akira Hatanaka
95b8ae190e Simplify CC_MipsO32 and merge it with CC_MipsO32_VarArgs. Patch by Sasa Stankovic.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131657 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 18:06:05 +00:00
Stuart Hastings
d22f036c2a Reverting 131641 to investigate 'bot complaint.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131654 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 17:54:42 +00:00
Jim Grosbach
6635b04a43 80 columns.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131649 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 17:34:53 +00:00
Akira Hatanaka
d48cfaec35 Fix data layout string. i64 is aligned to 64 bit boundaries.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131642 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 17:21:09 +00:00
Stuart Hastings
b6dcf3c514 Revise MOVSX16rr8/MOVZX16rr8 (and rm variants) to no longer be
pseudos.  rdar://problem/8614450


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131641 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 16:59:50 +00:00
Cameron Zwarich
055cdfc541 Use the correct register class for Cell varargs spilling. This fixes all of the
verifier failures in the CodeGen/CellSPU tests.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131631 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 04:44:19 +00:00
Mon P Wang
28e2b1d29a Fixed sdiv and udiv for <4 x i16>. The test from r125402 still applies for this change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131630 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 04:15:07 +00:00
Cameron Zwarich
d76773a221 Make CodeGen/PowerPC/2007-09-11-RegCoalescerAssert.ll pass with the verifier.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131627 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 03:11:06 +00:00
Cameron Zwarich
0113e4e3f2 Fix PR8828 by removing the explicit def in MovePCToLR as well as the pointless
piclabel operand. The operand in the tablegen definition doesn't actually turn
into an MI operand, so it just confuses anything checking the TargetInstrDesc
for the number of operands. It suffices to just have an implicit def of LR.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131626 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 02:56:28 +00:00
Cameron Zwarich
462b6dc6bf Reuse the TargetInstrDesc.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131625 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 02:56:23 +00:00
Cameron Zwarich
21803721d5 Correctly constrain a register class when computing frame offsets, as the Thumb2
add instruction takes an rGPR. This fixes the last of PR8825.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131619 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-19 02:18:27 +00:00
Eli Friedman
22486c9aba Revert unintentional commit.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131597 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 23:13:10 +00:00
Eli Friedman
107ffd58f7 More instcombine simplifications towards better debug locations.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131596 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 23:11:30 +00:00
Cameron Zwarich
631fcbf810 Add missing mayLoad / mayStore flags to instruction definitions without patterns,
which fixes all of the CodeGen/MBlaze verifier failures.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131595 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 23:03:10 +00:00
Cameron Zwarich
e4c6445405 Reserve the segment registers on x86 to fix verifier failures in any code that
uses them.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131591 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 22:24:48 +00:00
Cameron Zwarich
20a41cb2b2 Reserve r29 on Alpha. This fixes all verifier failures in CodeGen/Alpha.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131587 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 21:54:32 +00:00
Tanya Lattner
db28247522 Handle perfect shuffle case that generates a vrev for vectors of floats.
Add test case.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131582 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 21:44:54 +00:00
Cameron Zwarich
955db42568 Fix more of PR8825. Now all of CodeGen/ARM passes with VerifyCoalescing turned on.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131578 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 21:25:14 +00:00
Johnny Chen
a96581f4f7 Disassembly of tBcc was wrongly adding 4 to the SignExtend'ed imm8:'0' immediate operand.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131565 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 20:32:41 +00:00
Chad Rosier
871f664093 Enables vararg functions that pass all arguments via registers to be optimized into tail-calls when possible.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131560 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 19:59:50 +00:00
Evan Cheng
b936e3006f Revise r131553. Just use the type of the input node and forgo the bitcast. rdar://9449159.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131555 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 18:59:17 +00:00
Evan Cheng
d48fda46f5 Fix an ARMTargetLowering::LowerSELECT bug: legalized result must have same type as input. Sorry test cases only trigger when dag combine is disabled. rdar://9449178
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131553 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 18:47:27 +00:00
Justin Holewinski
657d1bed23 PTX: add flag to disable mad/fma selection
Patch by Dan Bailey

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131537 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 15:42:23 +00:00
Tanya Lattner
2a8eb722c7 In r131488 I misunderstood how VREV works. It splits the vector in half and splits each half. Therefore, the real problem was that we were using a VREV64 for a 4xi16, when we should have been using a VREV32.
Updated test case and reverted change to the PerfectShuffle Table.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131529 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 06:42:21 +00:00
Cameron Zwarich
141ec63962 Fix typo.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131519 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 02:29:50 +00:00
Cameron Zwarich
7d336c0c68 Fix more of PR8825 by correctly using rGPR registers when lowering atomic
compare-and-swap intrinsics.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131518 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-18 02:20:07 +00:00
Cameron Zwarich
d6ffcd88ba Actually, the address operand of the Thumb2 LDREX / STREX instructions *can*
take r13, so we can just make it a GPR. This fixes PR8825.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131507 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-17 23:26:20 +00:00
Cameron Zwarich
3c60ff4f3d Correct a minor problem with the Thumb2 LDREX and STREX instruction encodings. They
were marked as taking a tGPR when in reality they take an rGPR.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131506 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-17 23:11:12 +00:00
Tanya Lattner
c81c9709ef vrev is incorrectly defined in the perfect shuffle table. The ordering is backwards (should be 0x3210 versus 0x1032) which exposed a bug when doing a shuffle on a 4xi16. I've attached a test case.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131488 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-17 20:48:40 +00:00
Mon P Wang
fee22869ca Enable autodetect of popcnt
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131476 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-17 18:33:37 +00:00
Eli Friedman
19515b4e52 Add x86 fast-isel for calls returning first-class aggregates. rdar://9435872.
This is r131438 with a couple small fixes.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131474 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-17 18:29:03 +00:00
Eli Friedman
b8e0d3412c Clean up the mess created by r131467+r131469.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131471 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-17 18:02:22 +00:00
Stuart Hastings
6db2c2fe21 Revert 131467 due to buildbot complaint.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131469 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-17 16:59:46 +00:00
Stuart Hastings
504421e327 Fix an obscure issue in X86_64 parameter passing: if a tiny byval is
passed as the fifth parameter, insure it's passed correctly (in R9).
rdar://problem/6920088


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131467 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-17 16:45:55 +00:00
Nadav Rotem
86694297ff Fix a bug in PerformEXTRACT_VECTOR_ELTCombine. The code created an ADD SDNode
with two different types, in cases where the index and the ptr had different
types.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131461 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-17 08:31:57 +00:00
Eric Christopher
6abb7bae10 Update comment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131459 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-17 08:16:14 +00:00
Eric Christopher
c324f72ab7 Support XOR and AND optimization with no return value.
Finishes off rdar://8470697


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131458 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-17 08:10:18 +00:00