llvm-svn: 26045

llvm-svn: 26040

is just as efficient as MVIZ and is also more general.

Fix a few minor bugs introduced in recent patches

llvm-svn: 26036

mask.  This allows the code to be simpler and more efficient.

Also, generalize some of the cases in MVIZ a bit, making it slightly more aggressive.

llvm-svn: 26035

llvm-svn: 26034

'demanded bits', inspired by Nate's work in the dag combiner.  This isn't
complete, but needs to unrelated instcombiner changes to continue.

llvm-svn: 26033

Turn A / (C1 << N), where C1 is "1<<C2" into A >> (N+C2) [udiv only].

Tested with: rem.ll:test5, div.ll:test10

llvm-svn: 26003

#LLVM LOC, and auto-cse's cast instructions.

llvm-svn: 25974

1. When rewriting code in outer loops, sometimes we would insert code into
   inner loops that is invariant in that loop.
2. Notice that 4*(2+x) is 8+4*x and use that to simplify expressions.

This is a performance neutral change.

llvm-svn: 25964

llvm-svn: 25661

llvm-svn: 25587

llvm-svn: 25514

need the float->double part.

llvm-svn: 25452

llvm-svn: 25406

llvm-svn: 25363

llvm-svn: 25349

llvm-svn: 25315

llvm-svn: 25299

llvm-svn: 25294

llvm-svn: 25292

Patch written by Daniel Berlin!

llvm-svn: 25202

Patch written by Daniel Berlin!

llvm-svn: 25201

llvm-svn: 25180

llvm-svn: 25137

llvm-svn: 25130

the shifts.

This allows us to fold this (which is the 'integer add a constant' sequence
from cozmic's scheme compmiler):

int %x(uint %anf-temporary776) {
        %anf-temporary777 = shr uint %anf-temporary776, ubyte 1
        %anf-temporary800 = cast uint %anf-temporary777 to int
        %anf-temporary804 = shl int %anf-temporary800, ubyte 1
        %anf-temporary805 = add int %anf-temporary804, -2
        %anf-temporary806 = or int %anf-temporary805, 1
        ret int %anf-temporary806
}

into this:

int %x(uint %anf-temporary776) {
        %anf-temporary776 = cast uint %anf-temporary776 to int
        %anf-temporary776.mask1 = add int %anf-temporary776, -2
        %anf-temporary805 = or int %anf-temporary776.mask1, 1
        ret int %anf-temporary805
}

note that instcombine already knew how to eliminate the AND that the two
shifts fold into.  This is tested by InstCombine/shift.ll:test26

-Chris

llvm-svn: 25128

llvm-svn: 25126

functionality changes.

llvm-svn: 25125

llvm-svn: 25021

behavior in 126.gcc on big-endian systems.

llvm-svn: 24708

186.crafty by about 16% (from 15.109s to 13.045s) on my system.

This turns allocas with unions/casts into scalars.  For example crafty has
something like this:

    union doub {
      unsigned short i[4];
      long long d;
    };
int f(long long a) {
  return ((union doub){.d=a}).i[1];
}

Instead of generating loads and stores to an alloca, we now promote the
whole thing to a scalar long value.

This implements: Transforms/ScalarRepl/AggregatePromote.ll

llvm-svn: 24667

know that small negative values fit into the immediate field of addressing
modes.

llvm-svn: 24608

Transforms/DeadStoreElimination/2005-11-30-vaarg.ll

llvm-svn: 24545

llvm-svn: 24491

llvm-svn: 24488

llvm-svn: 24487

half the problem.

llvm-svn: 24414

compiling mysql reported by Ted Kremenek.

llvm-svn: 24402

llvm-svn: 24288

llvm-svn: 24270