llvm-svn: 28603

llvm-svn: 28594

llvm-svn: 28582

llvm-svn: 28581

llvm-svn: 28579

llvm-svn: 28578

llvm-svn: 28577

only transform a shufps to pshufd when the first two operands are the same.

llvm-svn: 28575

llvm-svn: 28574

llvm-svn: 28573

llvm-svn: 28565

llvm-svn: 28564

llvm-svn: 28551

Fold c2 in (x << c1) | c2 where (c2 < c1)
e.g.
int test(int x) {
  return (x << 3) + 7;
}

This can be codegen'd as:
leal 7(,%eax,8), %eax

llvm-svn: 28550

llvm-svn: 28548

RET chain, value1, sign1, value2, sign2, ...

llvm-svn: 28510

llvm-svn: 28505

vector arguments, not three.

llvm-svn: 28504

llvm-svn: 28499

llvm-svn: 28498

llvm-svn: 28497

llvm-svn: 28496

llvm-svn: 28493

llvm-svn: 28491

llvm-svn: 28488

llvm-svn: 28483

llvm-svn: 28481

Patch contributed by Anton Korobeynikov!

llvm-svn: 28480

llvm-svn: 28475

the copyto/fromregs instead of making the X86ISD::CALL selection code create
them.

llvm-svn: 28463

llvm-svn: 28459

non-deterministic behavior.

llvm-svn: 28454

by Anton Korobeynikov!  This is a step towards closing PR786.

llvm-svn: 28447

llvm-svn: 28441

llvm-svn: 28440

FORMAL_ARGUMENTS nodes include a token operand.

llvm-svn: 28439

return argument pops the hidden struct pointer if present, not the caller.

For example, in this testcase:

struct X { int D, E, F, G; };
struct X bar() {
  struct X a;
  a.D = 0;
  a.E = 1;
  a.F = 2;
  a.G = 3;
  return a;
}
void foo(struct X *P) {
  *P = bar();
}

We used to emit:

_foo:
        subl $28, %esp
        movl 32(%esp), %eax
        movl %eax, (%esp)
        call _bar
        addl $28, %esp
        ret
_bar:
        movl 4(%esp), %eax
        movl $0, (%eax)
        movl $1, 4(%eax)
        movl $2, 8(%eax)
        movl $3, 12(%eax)
        ret

This is correct on Linux/X86 but not Darwin/X86.  With this patch, we now
emit:

_foo:
        subl $28, %esp
        movl 32(%esp), %eax
        movl %eax, (%esp)
        call _bar
***     addl $24, %esp
        ret
_bar:
        movl 4(%esp), %eax
        movl $0, (%eax)
        movl $1, 4(%eax)
        movl $2, 8(%eax)
        movl $3, 12(%eax)
***     ret $4

For the record, GCC emits (which is functionally equivalent to our new code):

_bar:
        movl    4(%esp), %eax
        movl    $3, 12(%eax)
        movl    $2, 8(%eax)
        movl    $1, 4(%eax)
        movl    $0, (%eax)
        ret     $4
_foo:
        pushl   %esi
        subl    $40, %esp
        movl    48(%esp), %esi
        leal    16(%esp), %eax
        movl    %eax, (%esp)
        call    _bar
        subl    $4, %esp
        movl    16(%esp), %eax
        movl    %eax, (%esi)
        movl    20(%esp), %eax
        movl    %eax, 4(%esi)
        movl    24(%esp), %eax
        movl    %eax, 8(%esi)
        movl    28(%esp), %eax
        movl    %eax, 12(%esi)
        addl    $40, %esp
        popl    %esi
        ret

This fixes SingleSource/Benchmarks/CoyoteBench/fftbench with LLC and the
JIT, and fixes the X86-backend portion of PR729.  The CBE still needs to
be updated.

llvm-svn: 28438

llvm-svn: 28427

llvm-svn: 28425

llvm-svn: 28423