| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | handle combining A / (B << N) into A >>u (log2(B)+N) when B is a power of 2 | Nate Begeman | 2006-02-05 | 1 | -2/+13 |
| * | Add a framework for eliminating instructions that produces undemanded bits. | Nate Begeman | 2006-02-03 | 1 | -10/+30 |
| * | Add common code for reassociating ops in the dag combiner | Nate Begeman | 2006-02-03 | 1 | -50/+55 |
| * | Turn any_extend nodes into zero_extend nodes when it allows us to remove an | Chris Lattner | 2006-02-02 | 1 | -0/+13 |
| * | add two dag combines: | Chris Lattner | 2006-02-02 | 1 | -8/+25 |
| * | Fix some of the stuff in the PPC README file, and clean up legalization | Nate Begeman | 2006-02-01 | 1 | -3/+49 |
| * | Move MaskedValueIsZero from the DAGCombiner to the TargetLowering interface,m... | Chris Lattner | 2006-01-30 | 1 | -120/+21 |
| * | pass the address of MaskedValueIsZero into isMaskedValueZeroForTargetNode, | Chris Lattner | 2006-01-30 | 1 | -1/+1 |
| * | eliminate uses of SelectionDAG::getBR2Way_CC | Chris Lattner | 2006-01-29 | 1 | -3/+10 |
| * | Add a missing case to the dag combiner. | Nate Begeman | 2006-01-28 | 1 | -2/+3 |
| * | Add explicit #includes of <iostream> | Chris Lattner | 2006-01-22 | 1 | -0/+1 |
| * | Get rid of code in the DAGCombiner that is duplicated in SelectionDAG.cpp | Nate Begeman | 2006-01-18 | 1 | -74/+80 |
| * | Fix a backwards conditional that caused an inf loop in some cases. This | Chris Lattner | 2006-01-18 | 1 | -1/+1 |
| * | Disable two transformations that contribute to bus errors on SparcV8. | Chris Lattner | 2006-01-15 | 1 | -2/+6 |
| * | Add a simple missing fold to produce this: | Chris Lattner | 2006-01-12 | 1 | -0/+8 |
| * | Don't create rotate instructions in unsupported types, because we don't have | Chris Lattner | 2006-01-12 | 1 | -2/+2 |
| * | Add bswap, rotl, and rotr nodes | Nate Begeman | 2006-01-11 | 1 | -2/+36 |
| * | Revert the previous check-in. Leave shl x, 1 along for target to deal with. | Evan Cheng | 2006-01-06 | 1 | -3/+0 |
| * | fold (shl x, 1) -> (add x, x) | Evan Cheng | 2006-01-06 | 1 | -0/+3 |
| * | Added initial support for DEBUG_LABEL allowing debug specific labels to be | Jim Laskey | 2006-01-05 | 1 | -2/+1 |
| * | Add unique id to debug location for debug label use (work in progress.) | Jim Laskey | 2006-01-04 | 1 | -1/+2 |
| * | Remove redundant debug locations. | Jim Laskey | 2005-12-23 | 1 | -0/+34 |
| * | Simplify store(bitconv(x)) to store(x). This allows us to compile this: | Chris Lattner | 2005-12-23 | 1 | -0/+5 |
| * | fold (conv (load x)) -> (load (conv*)x). | Chris Lattner | 2005-12-23 | 1 | -0/+10 |
| * | Fold bitconv(bitconv(x)) -> x. We now compile this: | Chris Lattner | 2005-12-23 | 1 | -0/+3 |
| * | constant fold bits_convert in getNode and in the dag combiner for fp<->int | Chris Lattner | 2005-12-23 | 1 | -0/+15 |
| * | * Fix a GlobalAddress lowering bug. | Evan Cheng | 2005-12-21 | 1 | -1/+4 |
| * | Don't create SEXTLOAD/ZEXTLOAD instructions that the target doesn't support | Chris Lattner | 2005-12-15 | 1 | -2/+4 |
| * | When folding loads into ops, immediately replace uses of the op with the | Chris Lattner | 2005-12-14 | 1 | -6/+6 |
| * | Fix the (zext (zextload)) case to trigger, similarly for sign extends. | Chris Lattner | 2005-12-14 | 1 | -7/+28 |
| * | Fix a miscompilation in crafty due to a recent patch | Chris Lattner | 2005-12-14 | 1 | -1/+1 |
| * | Fold (zext (load x) to (zextload x). | Evan Cheng | 2005-12-14 | 1 | -1/+18 |
| * | Only transform (sext (truncate x)) -> (sextinreg x) if before legalize or | Chris Lattner | 2005-12-07 | 1 | -1/+3 |
| * | Teach the dag combiner to turn a truncate/sign_extend pair into a sextinreg | Chris Lattner | 2005-12-07 | 1 | -0/+4 |
| * | Fix operator precedence bug caught by VC++. | Jeff Cohen | 2005-11-12 | 1 | -2/+2 |
| * | Switch the allnodes list from a vector of pointers to an ilist of nodes.This ... | Chris Lattner | 2005-11-09 | 1 | -1/+3 |
| * | Fix a crash that Andrew noticed, and add a pair of braces to unfconfuse | Nate Begeman | 2005-11-02 | 1 | -5/+5 |
| * | Fix a source of undefined behavior when dealing with 64-bit types. This | Chris Lattner | 2005-11-02 | 1 | -1/+1 |
| * | Codegen mul by negative power of two with a shift and negate. | Chris Lattner | 2005-10-30 | 1 | -3/+13 |
| * | Fix DSE to not nuke dead stores unless they redundant store is the same | Chris Lattner | 2005-10-27 | 1 | -1/+4 |
| * | Add a simple xform that is useful for bitfield operations. | Chris Lattner | 2005-10-27 | 1 | -0/+9 |
| * | Clear a bit in this file that was causing a miscompilation of 178.galgel. | Chris Lattner | 2005-10-25 | 1 | -1/+1 |
| * | BuildSDIV and BuildUDIV only work for i32/i64, but they don't check that | Chris Lattner | 2005-10-22 | 1 | -10/+20 |
| * | add a case missing from the dag combiner that exposed the failure on | Chris Lattner | 2005-10-21 | 1 | -0/+3 |
| * | Fix a typo in the dag combiner, so that this can work on i64 targets | Nate Begeman | 2005-10-21 | 1 | -3/+2 |
| * | Invert the TargetLowering flag that controls divide by consant expansion. | Nate Begeman | 2005-10-21 | 1 | -10/+37 |
| * | Fix a couple bugs in the const div stuff where we'd generate MULHS/MULHU | Nate Begeman | 2005-10-20 | 1 | -3/+4 |
| * | don't use llabs with apparently VC++ doesn't have | Chris Lattner | 2005-10-20 | 1 | -1/+1 |
| * | Move the target constant divide optimization up into the dag combiner, so | Nate Begeman | 2005-10-20 | 1 | -0/+263 |
| * | Fold (select C, load A, load B) -> load (select C, A, B). This happens quite | Chris Lattner | 2005-10-18 | 1 | -7/+82 |
