| Commit message (Expand) | Author | Age | Files | Lines |
* | DAGCombine add (sext i1), X into sub X, (zext i1) if sext from i1 is illegal.... | Benjamin Kramer | 2010-12-22 | 1 | -0/+9 |
* | Fix a bug in ReduceLoadWidth that wasn't handling extending | Chris Lattner | 2010-12-22 | 1 | -1/+4 |
* | more cleanups, move a check for "roundedness" earlier to reject | Chris Lattner | 2010-12-22 | 1 | -14/+20 |
* | reduce indentation and improve comments, no functionality change. | Chris Lattner | 2010-12-22 | 1 | -51/+53 |
* | Reapply 122353-122355 with fixes. 122354 was wrong; | Dale Johannesen | 2010-12-21 | 1 | -4/+31 |
* | Revert 122353-122355 for the moment, they broke stuff. | Dale Johannesen | 2010-12-21 | 1 | -29/+3 |
* | Add a new transform to DAGCombiner. | Dale Johannesen | 2010-12-21 | 1 | -0/+26 |
* | Get the type of a shift from the shift, not from its shift | Dale Johannesen | 2010-12-21 | 1 | -1/+1 |
* | Shift by the word size is invalid IR; don't create it. | Dale Johannesen | 2010-12-21 | 1 | -2/+2 |
* | fix some typos | Chris Lattner | 2010-12-21 | 1 | -2/+1 |
* | rename MVT::Flag to MVT::Glue. "Flag" is a terrible name for | Chris Lattner | 2010-12-21 | 1 | -3/+3 |
* | Cosmetic changes. | Dale Johannesen | 2010-12-20 | 1 | -2/+2 |
* | Fix a DAGCombiner crash when folding binary vector operations with constant | Bob Wilson | 2010-12-17 | 1 | -16/+9 |
* | Add a transform to DAG Combiner. This improves the | Dale Johannesen | 2010-12-17 | 1 | -0/+20 |
* | take care of some todos, transforming [us]mul_lohi into | Chris Lattner | 2010-12-15 | 1 | -2/+46 |
* | when transforming a MULHS into a wider MUL, there is no need to SRA the | Chris Lattner | 2010-12-15 | 1 | -1/+1 |
* | Add a couple dag combines to transform mulhi/mullo into a wider multiply | Chris Lattner | 2010-12-13 | 1 | -0/+36 |
* | 80-col fixups. | Eric Christopher | 2010-12-09 | 1 | -9/+15 |
* | PR5207: Change APInt methods trunc(), sext(), zext(), sextOrTrunc() and | Jay Foad | 2010-12-07 | 1 | -11/+11 |
* | Fix a comment typo. | Bob Wilson | 2010-11-28 | 1 | -1/+1 |
* | Renaming ISD::BIT_CONVERT to ISD::BITCAST to better reflect the LLVM IR concept. | Wesley Peck | 2010-11-23 | 1 | -105/+105 |
* | Fix thinko: we must turn select(anyext, sext) into sext(select) | Duncan Sands | 2010-11-18 | 1 | -1/+2 |
* | The DAGCombiner was threading select over pairs of extending loads even | Duncan Sands | 2010-11-18 | 1 | -0/+5 |
* | Fix DAGCombiner to avoid folding a sext-in-reg or similar through a shl | Dan Gohman | 2010-11-09 | 1 | -0/+1 |
* | Just return undef for invalid masks or elts, and since we're doing that, | Eric Christopher | 2010-11-03 | 1 | -5/+6 |
* | If we have an undef mask our Elt will be -1 for our access, handle | Eric Christopher | 2010-11-03 | 1 | -3/+7 |
* | Fix DAGCombiner to avoid going into an infinite loop when it | Dan Gohman | 2010-11-03 | 1 | -1/+36 |
* | Remove DAG combiner patch to fold vector splats. Instcombiner does it now. | Bob Wilson | 2010-10-29 | 1 | -5/+0 |
* | Teach the DAG combiner to fold a splat of a splat. Radar 8597790. | Bob Wilson | 2010-10-28 | 1 | -24/+28 |
* | Make CodeGen TBAA-aware. | Dan Gohman | 2010-10-20 | 1 | -8/+21 |
* | This DAG combine BRCOND transformation can look pass truncate of the operand: | Evan Cheng | 2010-10-04 | 1 | -13/+18 |
* | fix rdar://8494845 + PR8244 - a miscompile exposed by my patch in r101350 | Chris Lattner | 2010-10-01 | 1 | -0/+9 |
* | A select between a constant and zero, when fed by a bit test, can be efficiently | Owen Anderson | 2010-09-22 | 1 | -0/+29 |
* | Reimplement r114460 in target-independent DAGCombine rather than target-depen... | Owen Anderson | 2010-09-21 | 1 | -0/+14 |
* | update a bunch of code to use the MachinePointerInfo version of getStore. | Chris Lattner | 2010-09-21 | 1 | -14/+8 |
* | eliminate an old SelectionDAG::getTruncStore method, propagating | Chris Lattner | 2010-09-21 | 1 | -7/+4 |
* | propagate MachinePointerInfo through various uses of the old | Chris Lattner | 2010-09-21 | 1 | -28/+22 |
* | convert dagcombine off the old form of getLoad. This fixes several bugs | Chris Lattner | 2010-09-21 | 1 | -16/+18 |
* | simplify DAGCombiner::SimplifySelectOps step #2/2. | Chris Lattner | 2010-09-21 | 1 | -55/+53 |
* | substantially reduce indentation and simplify DAGCombiner::SimplifySelectOps. | Chris Lattner | 2010-09-21 | 1 | -85/+76 |
* | a few more trivial updates. This fixes PerformInsertVectorEltInMemory to not | Chris Lattner | 2010-09-21 | 1 | -1/+1 |
* | When TCO is turned on, it is possible to end up with aliasing FrameIndex's. ... | Owen Anderson | 2010-09-20 | 1 | -2/+13 |
* | Revert r114312 while I sort out some issues. | Owen Anderson | 2010-09-19 | 1 | -1/+1 |
* | Tentatively enabled DAGCombiner Alias Analysis by default. As far as I know, | Owen Anderson | 2010-09-19 | 1 | -1/+1 |
* | Don't narrow the load and store in a load+twiddle+store sequence unless | Dan Gohman | 2010-09-02 | 1 | -1/+2 |
* | Fix a crash in the dag combiner caused by ConstantFoldBIT_CONVERTofBUILD_VECT... | Nate Begeman | 2010-07-27 | 1 | -2/+10 |
* | Remove r108639 now that it is handled by InstCombine instead. | Owen Anderson | 2010-07-19 | 1 | -13/+0 |
* | Add a DAGCombine xform to fold away redundant float->double->float conversion... | Owen Anderson | 2010-07-18 | 1 | -0/+13 |
* | Convert some tab stops into spaces. | Duncan Sands | 2010-07-12 | 1 | -28/+28 |
* | Reenable DAG combining for vector shuffles. It looks like it was temporarily | Bob Wilson | 2010-07-09 | 1 | -2/+0 |