llvm-svn: 27993

llvm-svn: 27992

(VAND, VADD, etc.). Legalizer will assert otherwise.

llvm-svn: 27991

llvm-svn: 27989

llvm-svn: 27988

PR748.

llvm-svn: 27987

- Fixed vararg support.

llvm-svn: 27985

nondeterminism being bad) could cause some trivial missed optimizations (dead
phi nodes being left around for later passes to clean up).

With this, llvm-gcc4 now bootstraps and correctly compares.  I don't know
why I never tried to do it before... :)

llvm-svn: 27984

recognize some simple affine IV's.

llvm-svn: 27982

llvm-svn: 27980

llvm-svn: 27975

llvm-svn: 27974

failres.

llvm-svn: 27973

llvm-svn: 27972

llvm-svn: 27971

llvm-svn: 27970

llvm-svn: 27968

llvm-svn: 27967

llvm-svn: 27966

result). Otherwise tblgen will pick the default (v16i8 for 128-bit vector).

llvm-svn: 27965

a temporary workaround for the 2-wide vector_shuffle problem (i.e. its mask
would have type v2i32 which is not legal).

llvm-svn: 27964

llvm-svn: 27963

llvm-svn: 27961

llvm-svn: 27960

llvm-svn: 27959

llvm-svn: 27958

llvm-svn: 27955

llvm-svn: 27954

llvm-svn: 27953

the jump table's range check block.  This re-enables 100% dense jump tables
by default on PPC & x86

llvm-svn: 27952

these.

llvm-svn: 27950

updating the machine CFG.

llvm-svn: 27949

llvm-svn: 27948

x86 and ppc for 100% dense switch statements when relocations are non-PIC.
This support will be extended and enhanced in the coming days to support
PIC, and less dense forms of jump tables.

llvm-svn: 27947

Don't do all the lowering stuff for 2-wide build_vector's. Also, minor optimization for shuffle of undef.

llvm-svn: 27946

Fix a performance regression. Use {p}shuf* when there are only two distinct elements in a build_vector.

llvm-svn: 27945

llvm-svn: 27943

This allows Prolangs-C++/city and probably a bunch of other stuff to work
well with the new front-end

llvm-svn: 27941

movd always clear the top 96 bits and movss does so when it's loading the
value from memory.
The net result is codegen for 4-wide shuffles is much improved. It is near
optimal if one or more elements is a zero. e.g.

__m128i test(int a, int b) {
  return _mm_set_epi32(0, 0, b, a);
}

compiles to

_test:
	movd 8(%esp), %xmm1
	movd 4(%esp), %xmm0
	punpckldq %xmm1, %xmm0
	ret

compare to gcc:

_test:
	subl	$12, %esp
	movd	20(%esp), %xmm0
	movd	16(%esp), %xmm1
	punpckldq	%xmm0, %xmm1
	movq	%xmm1, %xmm0
	movhps	LC0, %xmm0
	addl	$12, %esp
	ret

or icc:

_test:
        movd      4(%esp), %xmm0                                #5.10
        movd      8(%esp), %xmm3                                #5.10
        xorl      %eax, %eax                                    #5.10
        movd      %eax, %xmm1                                   #5.10
        punpckldq %xmm1, %xmm0                                  #5.10
        movd      %eax, %xmm2                                   #5.10
        punpckldq %xmm2, %xmm3                                  #5.10
        punpckldq %xmm3, %xmm0                                  #5.10
        ret                                                     #5.10

There are still room for improvement, for example the FP variant of the above example:

__m128 test(float a, float b) {
  return _mm_set_ps(0.0, 0.0, b, a);
}

_test:
	movss 8(%esp), %xmm1
	movss 4(%esp), %xmm0
	unpcklps %xmm1, %xmm0
	xorps %xmm1, %xmm1
	movlhps %xmm1, %xmm0
	ret

The xorps and movlhps are unnecessary. This will require post legalizer optimization to handle.

llvm-svn: 27939

llvm-svn: 27938