This hunk:
-    } else if (Src->getNumOperands() == 2 && Src->use_size() == 1) {
+    } else if (Src->getNumOperands() == 2) {

Allows GEP folding to be more aggressive, which reduces the number of instructions
and can dramatically speed up BasicAA in some cases.

llvm-svn: 6286

llvm-svn: 6251

llvm-svn: 6248

llvm-svn: 6232

llvm-svn: 6153

llvm-svn: 6030

llvm-svn: 5983

llvm-svn: 5979

llvm-svn: 5972

llvm-svn: 5971

llvm-svn: 5952

the visit function.  The worklist is for when their lattice value changes.

llvm-svn: 5927

  * Spell propagate right
  * Improve performance of phi node handling
  * Delete using directive
  * Other minor changes

llvm-svn: 5920

llvm-svn: 5872

llvm-svn: 5865

llvm-svn: 5863

llvm-svn: 5793

several ways:

 * Load expressions are not PRE'd well.  Alias Analysis should be used to
   get accurate information when computing anticipatibility.
 * The expression collection implementation does not handle PHI nodes properly,
   thus the implementation misses many opportunities to PRE.
 * This code could be sped up quite a bit

Despite these flaws, the code seems to work well, and handles PR's as one
would expect.

llvm-svn: 5759

llvm-svn: 5754

  - Implement simple reassociation: (A|c1)|(B|c2) == (A|B)|(c1|c2)
  - (A & C1)+(B & C2) -> (A & C1)|(B & C2) iff C1&C2 == 0
  - (A & C1)^(B & C2) -> (A & C1)|(B & C2) iff C1&C2 == 0

llvm-svn: 5743

llvm-svn: 5740

llvm-svn: 5738

Generalize not and neg comparison testers to allow constant to be considered not'able and neg'able.  This
allows optimization of this:
int %test4(int %A, int %B) {
        %a = xor int %A, -1
        %c = and int %a, 5    ; 5 = ~c2
        %d = xor int %c, -1
        ret int %d
}

into this:
int %test4(int %A, int %B) {            ; No predecessors!
        %c.demorgan = or int %A, -6             ; <int> [#uses=1]
        ret int %c.demorgan
}

llvm-svn: 5736

llvm-svn: 5733

llvm-svn: 5731

Implement new shift optimizations for shifting the result of a shift.

llvm-svn: 5729

llvm-svn: 5728

llvm-svn: 5704

allows LICM to use access sizes to help alias analysis be more precise.

llvm-svn: 5693

llvm-svn: 5685

llvm-svn: 5675

llvm-svn: 5669

llvm-svn: 5667

llvm-svn: 5664

Ironically the exit block modification code wasn't updating the exit block
information itself.  Fix this.

llvm-svn: 5659

llvm-svn: 5658

  * Remove bogus assertion: there may be a single outside predecessor and
    still need a new loop-preheader if the predecessor has multiple
    successors.  See bug: LICM/2003-02-27-PreheaderProblem.ll

llvm-svn: 5655

   - Now we perform loop exit-block splitting to ensure exit blocks are
     always dominated by the loop header.
   - We now preserve dominance frontier information
   - This fixes bug: LICM/2003-02-26-LoopExitNotDominated.ll

llvm-svn: 5652

Convert to work with new AliasAnalysis interface by conservatively assuming all pointers are arbitrarily large accesses

llvm-svn: 5636

llvm-svn: 5620