llvm-svn: 94123

llvm-svn: 94114

llvm-svn: 94113

llvm-svn: 94112

DbgInfoIntrinsics no longer appear in an instruction's use list; so clean up looking for them in use iterations and remove OnlyUsedByDbgInfoIntrinsics()

llvm-svn: 94111

loop-variant components, adds must be inserted after the increment.
Keep track of the increment position for this case, and insert
these adds in the correct location.

llvm-svn: 94110

llvm-svn: 94108

operands exceeds the number of registers used in the initial
solution, as that wouldn't lead to a profitable solution anyway.

llvm-svn: 94107

llvm-svn: 94104

ValueMapper.cpp ends up calling an out of line 
__ZNK4llvm12PATypeHolder3getEv, which is a template and llvm-config
determines arbitrarily to use the one in libipo.  This sucks, but
keeping the #include is a reasonable workaround.

llvm-svn: 94103

llvm-svn: 94102

llvm-svn: 94101

llvm-svn: 94092

llvm-svn: 94091

This new version is much more aggressive about doing "full" reduction in
cases where it reduces register pressure, and also more aggressive about
rewriting induction variables to count down (or up) to zero when doing so
reduces register pressure.

It currently uses fairly simplistic algorithms for finding reuse
opportunities, but it introduces a new framework allows it to combine
multiple strategies at once to form hybrid solutions, instead of doing
all full-reduction or all base+index.

llvm-svn: 94061

answers.  This will update as object size checking gets better information.

llvm-svn: 94059

llvm-svn: 94048

No functional change except the forgotten test for
InlineLimit.getNumOccurrences() == 0 in the CurrentThreshold2 calculation.

llvm-svn: 94007

llvm-svn: 93989

Map operands of all function-local metadata, not just metadata passed to llvm.dbg.declare intrinsics

llvm-svn: 93979

than the scaled register. This makes it more likely that subsequent
AddrModeMatcher queries will match the new address the same way as the
old, instead of accidentally matching what had been the base register
as the new scaled register, and then failing to match the scaled register.
This fixes some problems with address-mode sinking multiple muls into a
block, which will be a lot more common with some upcoming
LoopStrengthReduction changes.

llvm-svn: 93935

Evans!

llvm-svn: 93884

are the same.  I had already fixed a similar problem where the source and
destination were different bitcasts derived from the same alloca, but the
previous fix still did not handle the case where both operands are exactly
the same value.  Radar 7552893.

llvm-svn: 93848

llvm-svn: 93831

aggressive changed the canonical form from sext(trunc(x)) to ashr(lshr(x)),
make sure to transform a couple more things into that canonical form,
and catch a case where we missed turning zext/shl/ashr into a single sext.

llvm-svn: 93787

because it points to an alloca instruction through metadata.

llvm-svn: 93757

llvm-svn: 93693

llvm-svn: 93679

llvm-svn: 93560

added to the FSub version. However, the original version of this xform guarded
against doing this for floating point (!Op0->getType()->isFPOrFPVector()).

This is causing LLVM to perform incorrect xforms for code like:

void func(double *rhi, double *rlo, double xh, double xl, double yh, double yl){
  double mh, ml;
  double c = 134217729.0;
  double up, u1, u2, vp, v1, v2;
        
  up = xh*c;
  u1 = (xh - up) + up;
  u2 = xh - u1;
        
  vp = yh*c;
  v1 = (yh - vp) + vp;
  v2 = yh - v1;
        
  mh = xh*yh;
  ml = (((u1*v1 - mh) + (u1*v2)) + (u2*v1)) + (u2*v2);
  ml += xh*yl + xl*yh;
        
  *rhi = mh + ml;
  *rlo = (mh - (*rhi)) + ml;
}

The last line was optimized away, but rl is intended to be the difference
between the infinitely precise result of mh + ml and after it has been rounded
to double precision.

llvm-svn: 93369

in JT.

2) When cloning blocks for PHI or xor conditions, use
instsimplify to simplify the code as we go.  This allows us to 
squish common cases early in JT which opens up opportunities for
subsequent iterations, and allows it to completely simplify the
testcase.

llvm-svn: 93253

llvm-svn: 93251

llvm-svn: 93222

condition is a xor with a phi node.  This eliminates nonsense
like this from 176.gcc in several places:

 LBB166_84:
        testl   %eax, %eax
-       setne   %al
-       xorb    %cl, %al
-       notb    %al
-       testb   $1, %al
-       je      LBB166_85
+       je      LBB166_69
+       jmp     LBB166_85

This is rdar://7391699

llvm-svn: 93221

on branches by renaming it and checking for a branch at the call site.

llvm-svn: 93208

codegen has no apparent problem with the trunc version of this, because it turns
into a simple subreg idiom

llvm-svn: 93202

trunc has multiple uses.  Codegen is not able to coalesce the subreg case 
correctly and so this leads to higher register pressure and spilling (see PR5997).

This speeds up 256.bzip2 from 8.60 -> 8.04s on my machine, ~7%.

llvm-svn: 93200

good code on PR4216:

_test_bitfield:                                             ## @test_bitfield
	orl	$32962, %edi
	movl	$4294941946, %eax
	andq	%rdi, %rax
	ret

instead of:

_test_bitfield:
        movl    $4294941696, %ecx
        movl    %edi, %eax
        orl     $194, %edi
        orl     $32768, %eax
        andq    $250, %rdi
        andq    %rax, %rcx
        movq    %rdi, %rax
        orq     %rcx, %rax
        ret

Evan is looking into the remaining andq+imm -> andl optimization.

llvm-svn: 93147

BitsToClear case.  This allows it to promote expressions which have an
and/or/xor after the lshr, promoting cases like test2 (from PR4216) 
and test3 (random extample extracted from a spec benchmark).

clang now compiles the code in PR4216 into:

_test_bitfield:                                             ## @test_bitfield
	movl	%edi, %eax
	orl	$194, %eax
	movl	$4294902010, %ecx
	andq	%rax, %rcx
	orl	$32768, %edi
	andq	$39936, %rdi
	movq	%rdi, %rax
	orq	%rcx, %rax
	ret

instead of:

_test_bitfield:                                             ## @test_bitfield
	movl	%edi, %eax
	orl	$194, %eax
	movl	$4294902010, %ecx
	andq	%rax, %rcx
	shrl	$8, %edi
	orl	$128, %edi
	shlq	$8, %rdi
	andq	$39936, %rdi
	movq	%rdi, %rax
	orq	%rcx, %rax
	ret

which is still not great, but is progress.

llvm-svn: 93145

new BitsToClear result which allows us to start promoting
expressions that end with a lshr-by-constant.  This is
conservatively correct and better than what we had before
(see testcases) but still needs to be extended further.

llvm-svn: 93144