Generate better code for v16i8 shuffles on SSE2 (avoids stack)
Generate pshufb for v8i16 and v16i8 shuffles on SSSE3 where it is fewer uops.
Document the shuffle matching logic and add some FIXMEs for later further
  cleanups.
New tests that test the above.

Examples:

New:
_shuf2:
	pextrw	$7, %xmm0, %eax
	punpcklqdq	%xmm1, %xmm0
	pshuflw	$128, %xmm0, %xmm0
	pinsrw	$2, %eax, %xmm0

Old:
_shuf2:
	pextrw	$2, %xmm0, %eax
	pextrw	$7, %xmm0, %ecx
	pinsrw	$2, %ecx, %xmm0
	pinsrw	$3, %eax, %xmm0
	movd	%xmm1, %eax
	pinsrw	$4, %eax, %xmm0
	ret

=========

New:
_shuf4:
	punpcklqdq	%xmm1, %xmm0
	pshufb	LCPI1_0, %xmm0

Old:
_shuf4:
	pextrw	$3, %xmm0, %eax
	movsd	%xmm1, %xmm0
	pextrw	$3, %xmm1, %ecx
	pinsrw	$4, %ecx, %xmm0
	pinsrw	$5, %eax, %xmm0

========

New:
_shuf1:
	pushl	%ebx
	pushl	%edi
	pushl	%esi
	pextrw	$1, %xmm0, %eax
	rolw	$8, %ax
	movd	%xmm0, %ecx
	rolw	$8, %cx
	pextrw	$5, %xmm0, %edx
	pextrw	$4, %xmm0, %esi
	pextrw	$3, %xmm0, %edi
	pextrw	$2, %xmm0, %ebx
	movaps	%xmm0, %xmm1
	pinsrw	$0, %ecx, %xmm1
	pinsrw	$1, %eax, %xmm1
	rolw	$8, %bx
	pinsrw	$2, %ebx, %xmm1
	rolw	$8, %di
	pinsrw	$3, %edi, %xmm1
	rolw	$8, %si
	pinsrw	$4, %esi, %xmm1
	rolw	$8, %dx
	pinsrw	$5, %edx, %xmm1
	pextrw	$7, %xmm0, %eax
	rolw	$8, %ax
	movaps	%xmm1, %xmm0
	pinsrw	$7, %eax, %xmm0
	popl	%esi
	popl	%edi
	popl	%ebx
	ret

Old:
_shuf1:
	subl	$252, %esp
	movaps	%xmm0, (%esp)
	movaps	%xmm0, 16(%esp)
	movaps	%xmm0, 32(%esp)
	movaps	%xmm0, 48(%esp)
	movaps	%xmm0, 64(%esp)
	movaps	%xmm0, 80(%esp)
	movaps	%xmm0, 96(%esp)
	movaps	%xmm0, 224(%esp)
	movaps	%xmm0, 208(%esp)
	movaps	%xmm0, 192(%esp)
	movaps	%xmm0, 176(%esp)
	movaps	%xmm0, 160(%esp)
	movaps	%xmm0, 144(%esp)
	movaps	%xmm0, 128(%esp)
	movaps	%xmm0, 112(%esp)
	movzbl	14(%esp), %eax
	movd	%eax, %xmm1
	movzbl	22(%esp), %eax
	movd	%eax, %xmm2
	punpcklbw	%xmm1, %xmm2
	movzbl	42(%esp), %eax
	movd	%eax, %xmm1
	movzbl	50(%esp), %eax
	movd	%eax, %xmm3
	punpcklbw	%xmm1, %xmm3
	punpcklbw	%xmm2, %xmm3
	movzbl	77(%esp), %eax
	movd	%eax, %xmm1
	movzbl	84(%esp), %eax
	movd	%eax, %xmm2
	punpcklbw	%xmm1, %xmm2
	movzbl	104(%esp), %eax
	movd	%eax, %xmm1
	punpcklbw	%xmm1, %xmm0
	punpcklbw	%xmm2, %xmm0
	movaps	%xmm0, %xmm1
	punpcklbw	%xmm3, %xmm1
	movzbl	127(%esp), %eax
	movd	%eax, %xmm0
	movzbl	135(%esp), %eax
	movd	%eax, %xmm2
	punpcklbw	%xmm0, %xmm2
	movzbl	155(%esp), %eax
	movd	%eax, %xmm0
	movzbl	163(%esp), %eax
	movd	%eax, %xmm3
	punpcklbw	%xmm0, %xmm3
	punpcklbw	%xmm2, %xmm3
	movzbl	188(%esp), %eax
	movd	%eax, %xmm0
	movzbl	197(%esp), %eax
	movd	%eax, %xmm2
	punpcklbw	%xmm0, %xmm2
	movzbl	217(%esp), %eax
	movd	%eax, %xmm4
	movzbl	225(%esp), %eax
	movd	%eax, %xmm0
	punpcklbw	%xmm4, %xmm0
	punpcklbw	%xmm2, %xmm0
	punpcklbw	%xmm3, %xmm0
	punpcklbw	%xmm1, %xmm0
	addl	$252, %esp
	ret

llvm-svn: 65311

instruction. The class also consolidates the code for detecting constant
splats that's shared across PowerPC and the CellSPU backends (and might be
useful for other backends.) Also introduces SelectionDAG::getBUID_VECTOR() for
generating new BUILD_VECTOR nodes.

llvm-svn: 65296

llvm-svn: 65274

llvm-svn: 65152

(Note: Eventually, commits like this will be handled via a pre-commit hook that
 does this automagically, as well as expand tabs to spaces and look for 80-col
 violations.)

llvm-svn: 64827

llvm-svn: 64496

in inline asm as signed (what gcc does).  Add partial support
for x86-specific "e" and "Z" constraints, with appropriate
signedness for printing.

llvm-svn: 64400

No functional change.

llvm-svn: 64026

ScheduleDAG's TLI member to use const.

llvm-svn: 64018

Many targets build placeholder nodes for special operands, e.g.
GlobalBaseReg on X86 and PPC for the PIC base.  There's no
sensible way to associate debug info with these.  I've left
them built with getNode calls with explicit DebugLoc::getUnknownLoc operands. 
I'm not too happy about this but don't see a good improvement;
I considered adding a getPseudoOperand or something, but it
seems to me that'll just make it harder to read.

llvm-svn: 63992

getCALLSEQ_{END,START} to permit passing no DebugLoc
there.  UNDEF doesn't logically have DebugLoc; add
getUNDEF to encapsulate this.

llvm-svn: 63978

llvm-svn: 63969

its corresponding getTargetNode.  Lots of
caller changes.

llvm-svn: 63904

Adjust the many callers of those versions.

llvm-svn: 63767

llvm-svn: 63740

of undefs and incorrectly determining if we have punpckldq.

llvm-svn: 63702

llvm-svn: 63693

have it yet.  More coming.

llvm-svn: 63673

llvm-svn: 63656

llvm-svn: 63650

llvm-svn: 63509

llvm-svn: 63506

Teach LowerBRCOND to recognize (xor (setcc x), 1). The xor inverts the condition. It's normally transformed by the dag combiner, unless the condition is set by a arithmetic op with overflow.

llvm-svn: 63505

var-args, and don't allow FP return values

llvm-svn: 63495

crashes or wrong code with codegen of large integers:
eliminate the legacy getIntegerVTBitMask and
getIntegerVTSignBit methods, which returned their
value as a uint64_t, so couldn't handle huge types.

llvm-svn: 63494

argument.  Adjust all callers and overloaded versions.

llvm-svn: 63444

llvm-svn: 63442

an illegal type.

llvm-svn: 63380

dagcombines that help it match in several more cases. Add
several more cases to test/CodeGen/X86/bt.ll. This doesn't
yet include matching for BT with an immediate operand, it
just covers more register+register cases.

llvm-svn: 63266

llvm-svn: 63252

llvm-svn: 63193

tidy up SDUse and related code.
 - Replace the operator= member functions with a set method, like
   LLVM Use has, and variants setInitial and setNode, which take
   care up updating use lists, like LLVM Use's does. This simplifies
   code that calls these functions.
 - getSDValue() is renamed to get(), as in LLVM Use, though most
   places can either use the implicit conversion to SDValue or the
   convenience functions instead.
 - Fix some more node vs. value terminology issues.

Also, eliminate the one remaining use of SDOperandPtr, and
SDOperandPtr itself.

llvm-svn: 62995

llvm-svn: 62988

llvm-svn: 62979

corresponding to the "not" and "vnot" PatFrags.  Use the new method
in some places where it seems appropriate.

llvm-svn: 62768

Minor tweak to LowerUINT_TO_FP_i32. Bias (after scalar_to_vector) has two uses so we should make it the second source operand of ISD::OR so 2-address pass won't have to be smart about commuting.

%reg1024<def> = MOVSDrm %reg0, 1, %reg0, <cp#0>, Mem:LD(8,8) [ConstantPool + 0]
%reg1025<def> = MOVSD2PDrr %reg1024
%reg1026<def> = MOVDI2PDIrm <fi#-1>, 1, %reg0, 0, Mem:LD(4,16) [FixedStack-1 + 0]
%reg1027<def> = ORPSrr %reg1025<kill>, %reg1026<kill>
%reg1028<def> = MOVPD2SDrr %reg1027<kill>
%reg1029<def> = SUBSDrr %reg1028<kill>, %reg1024<kill>
%reg1030<def> = CVTSD2SSrr %reg1029<kill>
MOVSSmr <fi#0>, 1, %reg0, 0, %reg1030<kill>, Mem:ST(4,4) [FixedStack0 + 0]
%reg1031<def> = LD_Fp32m80 <fi#0>, 1, %reg0, 0, Mem:LD(4,16) [FixedStack0 + 0]
RET %reg1031<kill>, %ST0<imp-use,kill>

The reason 2-addr pass isn't smart enough to commute the ORPSrr is because it can't look pass the MOVSD2PDrr instruction.

llvm-svn: 62505

optimize it to a SINT_TO_FP when the sign bit is known zero. X86 isel should perform the optimization itself.

llvm-svn: 62504

llvm-svn: 62415

llvm-svn: 62405

X86. This code:

void f() {
  uint32_t x;
  float y = (float)x;
}

used to be:

     movl     %eax, -8(%ebp)
     movl     [2^52 double], -4(%ebp)
     movsd    -8(%ebp), %xmm0
     subsd    [2^52 double], %xmm0
     cvtsd2ss %xmm0, %xmm0

Is now:

   movsd        [2^52 double], %xmm0
   movsd        %xmm0, %xmm1
   movd         %ecx, %xmm2
   orps         %xmm2, %xmm1
   subsd        %xmm0, %xmm1
   cvtsd2ss     %xmm1, %xmm0

This is faster on X86. Note that there's an extra load of %xmm0 into %xmm1. That
will be fixed in a later coalescer fix.

llvm-svn: 62404