llvm-svn: 101170

Expand postra machine licm's capability a little more. If an instruction's register operands are all loop invariants, then it's safe to hoist it.

llvm-svn: 101167

Sometimes it is desirable to sink instructions along a critical edge:

x = ...
if (a && b) ...
else use(x);

The 'a && b' condition creates a critical edge to the else block, but we still
want to sink the computation of x into the block. The else block is dominated by
the parent block, so we are not pushing instructions into new code paths.

llvm-svn: 101165

llvm-svn: 101163

Re-apply 101075 and fix it properly. Just reuse the debug info of the branch instruction being optimized. There is no need to --I which can deref off start of the BB.

llvm-svn: 101162

patch by Marius Wachtler!

llvm-svn: 101160

in a nightly tester.

llvm-svn: 101158

Teach postra machine licm to hoist more obvious invariants, e.g. instructions with no source operands.

llvm-svn: 101154

llvm-svn: 101148

MachineBasicBlock::livein_iterator a const_iterator, because
clients shouldn't ever be using the iterator interface to
mutate the livein set.

llvm-svn: 101147

the rest of CodeGen.

llvm-svn: 101146

llvm-svn: 101145

llvm-svn: 101144

llvm-svn: 101143

that one operand is always greater than another.

llvm-svn: 101142

llvm-svn: 101141

llvm-svn: 101139

llvm-svn: 101138

llvm-svn: 101131

We have some code in llvm and clang where a BumpPtrAllocator is declared in a
class but never used in the common case. Stop wasting memory there.

llvm-svn: 101130

llvm-svn: 101125

llvm-svn: 101118

llvm-svn: 101117

patch by 'ether'.

llvm-svn: 101116

recognize all the valid rotated immediates.  This fixes the disassembler
issue and will also help codegen for some unusual constant values.

llvm-svn: 101114

numerator is an induction variable. For example, with code like this:

  for (i=0;i<n;++i)
    x[i%n] = 0;

IndVarSimplify will now recognize that i is always less than n inside
the loop, and eliminate the remainder.

llvm-svn: 101113

Type::getInt{1,8,...}[Ptr]Ty, so code can consistently use the methods on Type
without occasionally needed to call IntegerType::get.

llvm-svn: 101111

llvm-svn: 101110

patch by Sylvere Teissier!

llvm-svn: 101106

after edis.  Really, there ought to be some
mechanism to ensure that PARALLEL_DIRS get built
after DIRS.

llvm-svn: 101095

llvm-svn: 101086

llvm-svn: 101085

they're used a lot by getNodeForGEP, which can be called a lot.
This speeds up -iv-users by around 15% on several testcases.

llvm-svn: 101083

llvm-svn: 101081

llvm-svn: 101079

llvm-svn: 101077

If we have this situation:

    jCC  L1
    jmp  L2
L1:
  ...
L2:
  ...

We can get a small performance boost by emitting this instead:

    jnCC L2
L1:
  ...
L2:
  ...

This testcase shows an example of this:

float func(float x, float y) {
    double product = (double)x * y;
    if (product == 0.0)
        return product;
    return product - 1.0;
}

llvm-svn: 101075

llvm-svn: 101074

libEnhancedDisassembly, so we now build the
static library in all cases (although the shared
library is only built when requested/possible).

Also, fixed a bug where edis wasn't properly
initializing the targets it uses.

llvm-svn: 101072

expression is a UDiv and it doesn't appear that the UDiv came from
the user's source.

ScalarEvolution has recently figured out how to compute a tripcount
expression for the inner loop in
SingleSource/Benchmarks/Shootout/sieve.c, using a udiv. Emitting a
udiv instruction dramatically slows down the enclosing loop.

llvm-svn: 101068