Commit Graph

85 Commits

Author SHA1 Message Date
Eli Friedman
7675040979 Misc encoding fixes; reported on llvmdev.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@75142 91177308-0d34-0410-b5e6-96231b3b80d8
2009-07-09 16:49:25 +00:00
Bill Wendling
5c324d77d9 "The MMX_MASKMOVQ and MMX_MASKMOVQ64 instructions are labeled as MRMDestMem
instructions, which implies that there is an explicit memory operand.  There is
(however) no explicit memory operand; although this is a store, the only memory
operand is implicit, indicated by DS:EDI.  This causes the table-generation code
for the disassembler to report an error."

Patch by Sean Callanan!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@73989 91177308-0d34-0410-b5e6-96231b3b80d8
2009-06-23 19:52:59 +00:00
Eli Friedman
1041553424 Get rid of some bogus patterns for X86vzmovl. Don't create VZEXT_MOVL
nodes for vectors with an i16 element type.  Add an optimization for 
building a vector which is all zeros/undef except for the bottom 
element, where the bottom element is an i8 or i16.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@72988 91177308-0d34-0410-b5e6-96231b3b80d8
2009-06-06 06:05:10 +00:00
Eli Friedman
1762c1439e Get rid of a bogus pattern that interferes with optimization.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@72985 91177308-0d34-0410-b5e6-96231b3b80d8
2009-06-06 04:17:04 +00:00
Stuart Hastings
d58902a19b Evan says it's wrong; back out 72808.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@72817 91177308-0d34-0410-b5e6-96231b3b80d8
2009-06-03 22:59:34 +00:00
Stuart Hastings
77648cfd75 Recognize another euphemism for MOVDQ2Q.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@72808 91177308-0d34-0410-b5e6-96231b3b80d8
2009-06-03 21:39:14 +00:00
Bill Wendling
3b1259bb9f "The instructions MMX_PSADBWrm and MMX_PSADBWrr have opcode 0b11100000 (e0), but
the Intel manual (screenshot) says it should be 0b11110110 (f6).  The existing
encoding causes a disassembly conflict with MMX_PAVGBrm, which really should be
0f e0."

Patch by Sean Callanan!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@72508 91177308-0d34-0410-b5e6-96231b3b80d8
2009-05-28 02:04:00 +00:00
Nate Begeman
9008ca6b6b 2nd attempt, fixing SSE4.1 issues and implementing feedback from duncan.
PR2957

ISD::VECTOR_SHUFFLE now stores an array of integers representing the shuffle
mask internal to the node, rather than taking a BUILD_VECTOR of ConstantSDNodes
as the shuffle mask.  A value of -1 represents UNDEF.

In addition to eliminating the creation of illegal BUILD_VECTORS just to 
represent shuffle masks, we are better about canonicalizing the shuffle mask,
resulting in substantially better code for some classes of shuffles.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@70225 91177308-0d34-0410-b5e6-96231b3b80d8
2009-04-27 18:41:29 +00:00
Rafael Espindola
15684b2955 Revert 69952. Causes testsuite failures on linux x86-64.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@69967 91177308-0d34-0410-b5e6-96231b3b80d8
2009-04-24 12:40:33 +00:00
Nate Begeman
b706d29f9c PR2957
ISD::VECTOR_SHUFFLE now stores an array of integers representing the shuffle
mask internal to the node, rather than taking a BUILD_VECTOR of ConstantSDNodes
as the shuffle mask.  A value of -1 represents UNDEF.

In addition to eliminating the creation of illegal BUILD_VECTORS just to 
represent shuffle masks, we are better about canonicalizing the shuffle mask,
resulting in substantially better code for some classes of shuffles.

A clean up of x86 shuffle code, and some canonicalizing in DAGCombiner is next.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@69952 91177308-0d34-0410-b5e6-96231b3b80d8
2009-04-24 03:42:54 +00:00
Evan Cheng
242b38bae5 Only v1i16 (i.e. _m64) is returned via RAX / RDX.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65313 91177308-0d34-0410-b5e6-96231b3b80d8
2009-02-23 09:03:22 +00:00
Mon P Wang
9e5ecb8f8f Added support for SELECT v8i8 v4i16 for X86 (MMX)
Added support for TRUNC v8i16 to v8i8 for X86 (MMX)


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@60916 91177308-0d34-0410-b5e6-96231b3b80d8
2008-12-12 01:25:51 +00:00
Evan Cheng
62fb4f2e1c Use mmx (punpckldq VR64, (mmx_v_set0)) to clear high 32-bits of a VR64 register.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@60499 91177308-0d34-0410-b5e6-96231b3b80d8
2008-12-03 19:38:05 +00:00
Dan Gohman
15511cf166 Rename isSimpleLoad to canFoldAsLoad, to better reflect its meaning.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@60487 91177308-0d34-0410-b5e6-96231b3b80d8
2008-12-03 18:15:48 +00:00
Evan Cheng
b35ed9247a Add more vector move low and zero-extend patterns.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@58752 91177308-0d34-0410-b5e6-96231b3b80d8
2008-11-05 06:04:51 +00:00
Bill Wendling
1dd0086b06 Make "movdq2q" and "movq2dq" dependent upon having SSE2 because they use the
SSE2 registers as well as the MMX registers.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@55436 91177308-0d34-0410-b5e6-96231b3b80d8
2008-08-27 21:32:04 +00:00
Bill Wendling
cb3c51a4ae Nevermind. This broke the bootstrap (?!).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@55318 91177308-0d34-0410-b5e6-96231b3b80d8
2008-08-25 18:32:39 +00:00
Bill Wendling
51e05e7a99 MOVQ2DQ and MOVQ2DQ use SSE2. We should conditionalize the use of these
instructions on having SSE2.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@55317 91177308-0d34-0410-b5e6-96231b3b80d8
2008-08-25 18:20:52 +00:00
Anton Korobeynikov
017c260944 Provide a 64 bit variant of mmx.maskmovq intrinsic lowering.
Is there way to avoid explicit target check?

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@55238 91177308-0d34-0410-b5e6-96231b3b80d8
2008-08-23 15:53:19 +00:00
Nate Begeman
5f6ae30de5 Remove dead PatLeaf; there are a number of issues around MMX movl that need to be fixed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54026 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-25 17:25:04 +00:00
Dale Johannesen
a68f9013f2 Add v2f32 (MMX) type to X86. Support is primitive:
load,store,call,return,bitcast.  This is enough to
make call and return work.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52691 91177308-0d34-0410-b5e6-96231b3b80d8
2008-06-24 22:01:44 +00:00
Evan Cheng
f26ffe987c Implement vector shift up / down and insert zero with ps{rl}lq / ps{rl}ldq.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@51667 91177308-0d34-0410-b5e6-96231b3b80d8
2008-05-29 08:22:04 +00:00
Evan Cheng
d880b97257 Handle a few more cases of folding load i64 into xmm and zero top bits.
Note, some of the code will be moved into target independent part of DAG combiner in a subsequent patch.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@50918 91177308-0d34-0410-b5e6-96231b3b80d8
2008-05-09 21:53:03 +00:00
Evan Cheng
7e2ff77ef0 Handle vector move / load which zero the destination register top bits (i.e. movd, movq, movss (addr), movsd (addr)) with X86 specific dag combine.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@50838 91177308-0d34-0410-b5e6-96231b3b80d8
2008-05-08 00:57:18 +00:00
Evan Cheng
22b942aa4d Add separate intrinsics for MMX / SSE shifts with i32 integer operands. This allow us to simplify the horribly complicated matching code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@50601 91177308-0d34-0410-b5e6-96231b3b80d8
2008-05-03 00:52:09 +00:00
Evan Cheng
082948df9b Fix illegal MMX_MOVDQ2Qrr pattern. vector_extract result must be a scalar value.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@50291 91177308-0d34-0410-b5e6-96231b3b80d8
2008-04-25 20:12:46 +00:00
Evan Cheng
10e864276b Special handling for MMX values being passed in either GPR64 or lower 64-bits of XMM registers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@50289 91177308-0d34-0410-b5e6-96231b3b80d8
2008-04-25 19:11:04 +00:00
Evan Cheng
80f5404e18 Fix MMX_MOVQ2DQrr pattern. It's illegal to do a bitconvert from a smaller type to a larger one.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@50278 91177308-0d34-0410-b5e6-96231b3b80d8
2008-04-25 18:19:54 +00:00
Dan Gohman
0416b0a26d Fix the encoding of the MMX movd that moves from MMX to 64-bit GPR.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@50053 91177308-0d34-0410-b5e6-96231b3b80d8
2008-04-21 19:52:29 +00:00
Dan Gohman
a630f4ed91 Add movd instructions to move from MMX registers
to 64-bit GPR registers on x86-64.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@49757 91177308-0d34-0410-b5e6-96231b3b80d8
2008-04-15 23:55:07 +00:00
Evan Cheng
fa5a91a71e Undo 48570. Correctly match mmx shift instructions with an immediate operand.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48627 91177308-0d34-0410-b5e6-96231b3b80d8
2008-03-21 00:40:09 +00:00
Evan Cheng
dff1dcac6c Add intrinsics to match mmx shift builtin's with immediate operand.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48569 91177308-0d34-0410-b5e6-96231b3b80d8
2008-03-19 23:38:52 +00:00
Evan Cheng
da47e6e0d0 Replace all target specific implicit def instructions with a target independent one: TargetInstrInfo::IMPLICIT_DEF.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48380 91177308-0d34-0410-b5e6-96231b3b80d8
2008-03-15 00:03:38 +00:00
Evan Cheng
c8e3b147ee Clean up my own mess.
X86 lowering normalize vector 0 to v4i32. However DAGCombine can fold (sub x, x) -> 0 after legalization. It can create a zero vector of a type that's not expected (e.g. v8i16). We don't want to disable the optimization since leaving a (sub x, x) is really bad. Add isel patterns for other types of vector 0 to ensure correctness. It's highly unlikely to happen other than in bugpoint reduced test cases.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48279 91177308-0d34-0410-b5e6-96231b3b80d8
2008-03-12 07:02:50 +00:00
Anders Carlsson
b26947e967 Use the correct instruction encodings for the 64-bit MMX movd.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47740 91177308-0d34-0410-b5e6-96231b3b80d8
2008-02-29 01:35:12 +00:00
Evan Cheng
efec751a1b - When DAG combiner is folding a bit convert into a BUILD_VECTOR, it should check if it's essentially a SCALAR_TO_VECTOR. Avoid turning (v8i16) <10, u, u, u> to <10, 0, u, u, u, u, u, u>. Instead, simply convert it to a SCALAR_TO_VECTOR of the proper type.
- X86 now normalize SCALAR_TO_VECTOR to (BIT_CONVERT (v4i32 SCALAR_TO_VECTOR)). Get rid of X86ISD::S2VEC.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@47290 91177308-0d34-0410-b5e6-96231b3b80d8
2008-02-18 23:04:32 +00:00
Chris Lattner
ba7e756c22 Start inferring side effect information more aggressively, and fix many bugs in the
x86 backend where instructions were not marked maystore/mayload, and perf issues where
instructions were not marked neverHasSideEffects.  It would be really nice if we could
write patterns for copy instructions.

I have audited all the x86 instructions down to MOVDQAmr.  The flags on others and on
other targets are probably not right in all cases, but no clients currently use this
info that are enabled by default.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45829 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-10 07:59:24 +00:00
Chris Lattner
dd41527a7d remove explicit sets of 'neverHasSideEffects' that can now be
inferred from the instr patterns.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45824 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-10 05:45:39 +00:00
Chris Lattner
834f1ce031 rename isLoad -> isSimpleLoad due to evan's desire to have such a predicate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45667 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-06 23:38:27 +00:00
Chris Lattner
4ee451de36 Remove attribution from file headers, per discussion on llvmdev.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45418 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-29 20:36:04 +00:00
Bill Wendling
627c00b663 Add "mayHaveSideEffects" and "neverHasSideEffects" flags to some instructions. I
based what flag to set on whether it was already marked as
"isRematerializable". If there was a further check to determine if it's "really"
rematerializable, then I marked it as "mayHaveSideEffects" and created a check
in the X86 back-end similar to the remat one.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45132 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-17 23:07:56 +00:00
Evan Cheng
6e141fd048 Implicit def instructions, e.g. X86::IMPLICIT_DEF_GR32, are always re-materializable and they should not be spilled.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44960 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-12 23:12:09 +00:00
Chris Lattner
8a594489bf Fix a long standing deficiency in the X86 backend: we would
sometimes emit "zero" and "all one" vectors multiple times,
for example:

_test2:
	pcmpeqd	%mm0, %mm0
	movq	%mm0, _M1
	pcmpeqd	%mm0, %mm0
	movq	%mm0, _M2
	ret

instead of:

_test2:
	pcmpeqd	%mm0, %mm0
	movq	%mm0, _M1
	movq	%mm0, _M2
	ret

This patch fixes this by always arranging for zero/one vectors
to be defined as v4i32 or v2i32 (SSE/MMX) instead of letting them be
any random type.  This ensures they get trivially CSE'd on the dag.
This fix is also important for LegalizeDAGTypes, as it gets unhappy
when the x86 backend wants BUILD_VECTOR(i64 0) to be legal even when
'i64' isn't legal.

This patch makes the following changes:

1) X86TargetLowering::LowerBUILD_VECTOR now lowers 0/1 vectors into
   their canonical types.
2) The now-dead patterns are removed from the SSE/MMX .td files.
3) All the patterns in the .td file that referred to immAllOnesV or
   immAllZerosV in the wrong form now use *_bc to match them with a
   bitcast wrapped around them.
4) X86DAGToDAGISel::SelectScalarSSELoad is generalized to handle 
   bitcast'd zero vectors, which simplifies the code actually.
5) getShuffleVectorZeroOrUndef is updated to generate a shuffle that
   is legal, instead of generating one that is illegal and expecting
   a later legalize pass to clean it up.
6) isZeroShuffle is generalized to handle bitcast of zeros.
7) several other minor tweaks.

This patch is definite goodness, but has the potential to cause random
code quality regressions.  Please be on the lookout for these and let 
me know if they happen.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@44310 91177308-0d34-0410-b5e6-96231b3b80d8
2007-11-25 00:24:49 +00:00
Evan Cheng
071a279e94 Remove (somewhat confusing) Imp<> helper, use let Defs = [], Uses = [] instead.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41863 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-11 19:55:27 +00:00
Evan Cheng
2f39426ec9 Mark load instructions with isLoad = 1.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41595 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-30 05:49:43 +00:00
Dan Gohman
73a902b228 Mark the SSE and MMX load instructions that
X86InstrInfo::isReallyTriviallyReMaterializable knows how to handle
with the isReMaterializable flag so that it is given a chance to handle
them. Without hoisting constant-pool loads from loops this isn't very
visible, though it does keep CodeGen/X86/constant-pool-remat-0.ll from
making a copy of the constant pool on the stack.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40736 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-02 14:27:55 +00:00
Dan Gohman
b1576f56c8 Change the x86 assembly output to use tab characters to separate the
mnemonics from their operands instead of single spaces. This makes the
assembly output a little more consistent with various other compilers
(f.e. GCC), and slightly easier to read. Also, update the regression
tests accordingly.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40648 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-31 20:11:57 +00:00
Evan Cheng
c64a1a921c Redo and generalize previously removed opt for pinsrw: (vextract (v4i32 bc (v4f32 s2v (f32 load ))), 0) -> (i32 load )
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40628 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-31 08:04:03 +00:00
Evan Cheng
64d80e3387 Change instruction description to split OperandList into OutOperandList and
InOperandList. This gives one piece of important information: # of results
produced by an instruction.
An example of the change:
def ADD32rr  : I<0x01, MRMDestReg, (ops GR32:$dst, GR32:$src1, GR32:$src2),
                 "add{l} {$src2, $dst|$dst, $src2}",
                 [(set GR32:$dst, (add GR32:$src1, GR32:$src2))]>;
=>
def ADD32rr  : I<0x01, MRMDestReg, (outs GR32:$dst), (ins GR32:$src1, GR32:$src2),
                 "add{l} {$src2, $dst|$dst, $src2}",
                 [(set GR32:$dst, (add GR32:$src1, GR32:$src2))]>;


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40033 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-19 01:14:50 +00:00
Bill Wendling
10404c47d1 Support generation of GR64 to MMX code in the JIT.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@37866 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-04 01:29:22 +00:00