Add VCVTPD2PS, VCVTPS2DQ, VCVTPS2PDY, VCVTTPD2DQY, VCVTTPS2DQ and VCVTPD2DQ 256-bit conversion intrinsics

llvm-svn: 110608

Add patterns to AVX conversions instructions. Do that instead of declaring more intructions whenever is possible, more coming

llvm-svn: 110605

Next time the build is broken due to wrong library dependencies, just
try building again (if you are on some Unix and are building all LLVM
targets) or ask someone to commit the regenerated LLVMLibDeps.cmake.

llvm-svn: 110593

llvm-svn: 110582

llvm-svn: 110580

form of CMPSD (etc.)  Matching a 128-bit memory
operand is wrong, the instruction uses only 64 bits
(same as ADDSD etc.)  8193553.

llvm-svn: 110491

llvm-svn: 110480

llvm-svn: 110468

llvm-svn: 110460

llvm-svn: 110427

llvm-svn: 110425

llvm-svn: 110410

llvm-svn: 110404

ID member as the sole unique type identifier.  Clean up APIs related to this change.

llvm-svn: 110396

Support very basic (doesn't include ABI support in the front-end, varags, ...) 256-bit argument passing and return for AVX

llvm-svn: 110394

llvm-svn: 110371

llvm-svn: 110369

instructions.

llvm-svn: 110360

llvm-svn: 110359

uses.

llvm-svn: 110274

llvm-svn: 110268

llvm-svn: 110224

llvm-svn: 110200

- The COFF backend doesn't support MingW/Cygwin at the moment, it'll report an
  error, but it's still much better than random assertions from the MachO backend.
- We want to make ELF the default eventually, it's what the majority of targets use.

llvm-svn: 110197

llvm-svn: 110164

llvm-svn: 109949

llvm-svn: 109947

declared during the addition of the assembler support, the additional
changes are:
- Add missing intrinsics
- Move all SSE conversion instructions in X86InstInfo64.td to the SSE.td file.
- Duplicate some patterns to AVX mode.
- Step into PCMPEST/PCMPIST custom inserter and add AVX versions.

llvm-svn: 109878

llvm-svn: 109877

We do sometimes load from a too small stack slot when dealing with x86 arguments
(varargs and smaller-than-32-bit args). It looks like we know what we are doing
in those cases, so I am going to remove the assert instead of artifically
enlarging stack slot sizes.

The assert in storeRegToStackSlot stays in. We don't want to write beyond the
bounds of a stack slot.

llvm-svn: 109764

The size of this object isn't used for anything - technically it is of variable
size.

This avoids a false positive from the assert in
X86InstrInfo::loadRegFromStackSlot, and fixes PR7735.

llvm-svn: 109652

This is about 4x faster and smaller than the existing scalarization.

llvm-svn: 109566

~40% faster vector shl <4 x i32> on SSE 4.1  Larger improvements for smaller types coming in future patches.

For:

define <2 x i64> @shl(<4 x i32> %r, <4 x i32> %a) nounwind readnone ssp {
entry:
  %shl = shl <4 x i32> %r, %a                     ; <<4 x i32>> [#uses=1]
  %tmp2 = bitcast <4 x i32> %shl to <2 x i64>     ; <<2 x i64>> [#uses=1]
  ret <2 x i64> %tmp2
}

We get:

_shl:                                   ## @shl
	pslld	$23, %xmm1
	paddd	LCPI0_0, %xmm1
	cvttps2dq	%xmm1, %xmm1
	pmulld	%xmm1, %xmm0
	ret

Instead of:

_shl:                                   ## @shl
	pshufd	$3, %xmm0, %xmm2
	movd	%xmm2, %eax
	pshufd	$3, %xmm1, %xmm2
	movd	%xmm2, %ecx
	shll	%cl, %eax
	movd	%eax, %xmm2
	pshufd	$1, %xmm0, %xmm3
	movd	%xmm3, %eax
	pshufd	$1, %xmm1, %xmm3
	movd	%xmm3, %ecx
	shll	%cl, %eax
	movd	%eax, %xmm3
	punpckldq	%xmm2, %xmm3
	movd	%xmm0, %eax
	movd	%xmm1, %ecx
	shll	%cl, %eax
	movd	%eax, %xmm2
	movhlps	%xmm0, %xmm0
	movd	%xmm0, %eax
	movhlps	%xmm1, %xmm1
	movd	%xmm1, %ecx
	shll	%cl, %eax
	movd	%eax, %xmm0
	punpckldq	%xmm0, %xmm2
	movdqa	%xmm2, %xmm0
	punpckldq	%xmm3, %xmm0
	ret

llvm-svn: 109549

llvm-svn: 109494

subregister operands like this:

%reg1040:sub_32bit<def> = MOV32rm <fi#-2>, 1, %reg0, 0, %reg0, %reg1040<imp-def>; mem:LD4[FixedStack-2](align=8)

Make them return false when subreg operands are present. VirtRegRewriter is
making bad assumptions otherwise.

This fixes PR7713.

llvm-svn: 109489

with a too-big register class.

llvm-svn: 109488

llvm-svn: 109450

we are using AVX and no AVX version of the desired intruction is present,
this is better for incremental dev (without fallbacks it's easier to spot
what's missing). Not sure this is the best hack thought (we can also disable
all HasSSE* predicates by dinamically marking them 'false' if AVX is present)

llvm-svn: 109434

appropriate for targets without detailed instruction iterineries.
The scheduler schedules for increased instruction level parallelism in
low register pressure situation; it schedules to reduce register pressure
when the register pressure becomes high.

On x86_64, this is a win for all tests in CFP2000. It also sped up 256.bzip2
by 16%.

llvm-svn: 109300

llvm-svn: 109295