load.  rdar://problem/6373334

llvm-svn: 132696

rdar://problem/5993888

llvm-svn: 132606

Part of rdar://9119939

llvm-svn: 132510

llvm-svn: 132479

rdar://problem/6373334

llvm-svn: 132458

llvm-svn: 132424

llvm-svn: 132419

floating-point comparison, generate a mask of 0s or 1s, and generally
DTRT with NaNs.  Only profitable when the user wants a materialized 0
or 1 at runtime.  rdar://problem/5993888

llvm-svn: 132404

patch to TargetLowering.cpp.  rdar://problem/5660695

llvm-svn: 132388

llvm-svn: 132108

rdar://problem/6920088

llvm-svn: 132105

  non-zero.
- Teach X86 cmov optimization to eliminate the cmov from ctlz, cttz extension
  when the source of X86ISD::BSR / X86ISD::BSF is proven to be non-zero.

rdar://9490949

llvm-svn: 131948

llvm-svn: 131709

llvm-svn: 131708

Fixes rdar://9218925
Fixes PR9601

llvm-svn: 131682

llvm-svn: 131664

Enables vararg functions that pass all arguments via registers to be optimized into tail-calls when possible.

llvm-svn: 131560

llvm-svn: 131471

llvm-svn: 131469

passed as the fifth parameter, insure it's passed correctly (in R9).
rdar://problem/6920088

llvm-svn: 131467

Fix a bug in PerformEXTRACT_VECTOR_ELTCombine. The code created an ADD SDNode
with two different types, in cases where the index and the ptr had different
types.

llvm-svn: 131461

llvm-svn: 131424

Add custom lowering of X86 vector SRA/SRL/SHL when the shift amount is a splat vector.

llvm-svn: 131179

llvm-svn: 131012

triple component.

llvm-svn: 129838

llvm-svn: 129813

Luis Felipe Strano Moraes!

llvm-svn: 129558

llvm-svn: 128586

llvm-svn: 128338

FIXME: Some cleanups would be needed.
llvm-svn: 128206

I'm backing this out for the second time. It was supposed to be fixed by r128164, but the mingw self-host must be defeating the fix.

llvm-svn: 128181

(target-specific branchless method for double-width relational comparisons on x86)

llvm-svn: 128175

Re-apply r127953 with fixes: eliminate empty return block if it has no predecessors; update dominator tree if cfg is modified.

llvm-svn: 127981

to canonicalize IR", it broke a lot of things.

llvm-svn: 127954

to have single return block (at least getting there) for optimizations. This
is general goodness but it would prevent some tailcall optimizations.
One specific case is code like this:
int f1(void);
int f2(void);
int f3(void);
int f4(void);
int f5(void);
int f6(void);
int foo(int x) {
  switch(x) {
  case 1: return f1();
  case 2: return f2();
  case 3: return f3();
  case 4: return f4();
  case 5: return f5();
  case 6: return f6();
  }
}

=>
LBB0_2:                                 ## %sw.bb
  callq   _f1
  popq    %rbp
  ret
LBB0_3:                                 ## %sw.bb1
  callq   _f2
  popq    %rbp
  ret
LBB0_4:                                 ## %sw.bb3
  callq   _f3
  popq    %rbp
  ret

This patch teaches codegenprep to duplicate returns when the return value
is a phi and where the phi operands are produced by tail calls followed by
an unconditional branch:

sw.bb7:                                           ; preds = %entry
  %call8 = tail call i32 @f5() nounwind
  br label %return
sw.bb9:                                           ; preds = %entry
  %call10 = tail call i32 @f6() nounwind
  br label %return
return:
  %retval.0 = phi i32 [ %call10, %sw.bb9 ], [ %call8, %sw.bb7 ], ... [ 0, %entry ]
  ret i32 %retval.0

This allows codegen to generate better code like this:

LBB0_2:                                 ## %sw.bb
        jmp     _f1                     ## TAILCALL
LBB0_3:                                 ## %sw.bb1
        jmp     _f2                     ## TAILCALL
LBB0_4:                                 ## %sw.bb3
        jmp     _f3                     ## TAILCALL

rdar://9147433

llvm-svn: 127953

not have native support for this operation (such as X86).
The legalized code uses two vector INT_TO_FP operations and is faster
than scalarizing.

llvm-svn: 127951

llvm-svn: 127909

comparisons on x86.  Essentially, the way this works is that SUB+SBB sets
the relevant flags the same way a double-width CMP would.

This is a substantial improvement over the generic lowering in LLVM. The output
is also shorter than the gcc-generated output; I haven't done any detailed
benchmarking, though.

llvm-svn: 127852

llvm-svn: 127809

llvm-svn: 127807