| Commit message (Expand) | Author | Age | Files | Lines |
* | Fold comparisons against a constant nan, and optimize ORD/UNORD | Chris Lattner | 2007-12-29 | 2 | -17/+22 |
* | this xform is implemented. | Chris Lattner | 2007-12-29 | 1 | -22/+0 |
* | Disable null pointer folding transforms for non-generic address spaces. This ... | Christopher Lamb | 2007-12-29 | 2 | -4/+12 |
* | make sure not to zap volatile stores, thanks a lot to Dale for noticing this! | Chris Lattner | 2007-12-29 | 1 | -1/+2 |
* | Codegen: | Chris Lattner | 2007-12-29 | 1 | -4/+29 |
* | don't fold fp_round(fp_extend(load)) -> fp_round(extload) | Chris Lattner | 2007-12-29 | 1 | -0/+4 |
* | avoid going through a stack slot to convert from fpstack to xmm reg | Chris Lattner | 2007-12-29 | 2 | -29/+39 |
* | Delete a store whose input is a load from the same pointer: | Chris Lattner | 2007-12-29 | 1 | -1/+12 |
* | add a note | Chris Lattner | 2007-12-29 | 1 | -1/+41 |
* | expand note. | Chris Lattner | 2007-12-29 | 1 | -0/+4 |
* | dead calls to llvm.stacksave can be deleted, even though they | Chris Lattner | 2007-12-29 | 1 | -1/+10 |
* | add a note. | Chris Lattner | 2007-12-28 | 1 | -0/+32 |
* | add a note. | Chris Lattner | 2007-12-28 | 1 | -0/+32 |
* | Repair a transform that Chris noticed a bug in. Thanks to Nicholas for point... | Owen Anderson | 2007-12-28 | 1 | -5/+12 |
* | disable this instcombine xform, it miscompiles: | Chris Lattner | 2007-12-28 | 1 | -0/+2 |
* | add a note | Chris Lattner | 2007-12-28 | 1 | -0/+19 |
* | Ignore functions with internal linkages during linking. This snipped mimics the | Anton Korobeynikov | 2007-12-27 | 1 | -0/+3 |
* | Adding bindings for target triple and data layout. | Gordon Henriksen | 2007-12-27 | 1 | -0/+19 |
* | Switch the bindings to use LLVMFoldingBuilder. | Gordon Henriksen | 2007-12-27 | 1 | -1/+1 |
* | Fixing several transforms which would drop the collector attribute | Gordon Henriksen | 2007-12-25 | 4 | -0/+10 |
* | Don't break critical edges for single-bb loops, this helps with PR1877, though | Chris Lattner | 2007-12-25 | 1 | -3/+1 |
* | GC poses hazards to the inliner. Consider: | Gordon Henriksen | 2007-12-25 | 1 | -0/+13 |
* | Noting and enforcing that GC intrinsics are valid only within a | Gordon Henriksen | 2007-12-25 | 1 | -28/+36 |
* | Adjusting verification of "llvm.gc*" intrinsic prototypes to match | Gordon Henriksen | 2007-12-25 | 1 | -26/+33 |
* | Flesh out the Briggs implementation a little bit more, fix a few FIXMEs. | Owen Anderson | 2007-12-24 | 1 | -11/+35 |
* | add a -backedge-hack llc-beta option to codegenprepare. | Chris Lattner | 2007-12-24 | 1 | -2/+10 |
* | add a simple hack | Chris Lattner | 2007-12-24 | 1 | -0/+12 |
* | Cygwin defines uint32_t as unsigned long. Unbreak call to std::min in this case | Anton Korobeynikov | 2007-12-24 | 1 | -1/+1 |
* | Setting GlobalDirective in TargetAsmInfo by default rather than | Gordon Henriksen | 2007-12-23 | 2 | -2/+1 |
* | C and Ocaml bindings for ExecutionEngine (i.e., the JIT compiler). | Gordon Henriksen | 2007-12-23 | 1 | -0/+187 |
* | Sketch out an implementation of Briggs' copy placement algorithm. | Owen Anderson | 2007-12-23 | 1 | -10/+104 |
* | fix some warnings. This code needs to be de-tabified :( | Chris Lattner | 2007-12-22 | 1 | -2/+3 |
* | fix strict-aliasing violation | Chris Lattner | 2007-12-22 | 1 | -3/+1 |
* | initial code for forming an FGETSIGN node. This is disabled until | Chris Lattner | 2007-12-22 | 1 | -0/+26 |
* | improve support for fgetsign | Chris Lattner | 2007-12-22 | 1 | -0/+6 |
* | Tell TargetLoweringOpt whether it is running before | Chris Lattner | 2007-12-22 | 1 | -1/+1 |
* | Add a new FGETSIGN operation, which defaults to expand on all | Chris Lattner | 2007-12-22 | 1 | -1/+5 |
* | Erm, really disable :) | Anton Korobeynikov | 2007-12-22 | 1 | -1/+1 |
* | Disable, until we'll really need it | Anton Korobeynikov | 2007-12-22 | 1 | -0/+3 |
* | Use getIntrinsicID instead of looking up intrinsic prototypes. Also | Gordon Henriksen | 2007-12-22 | 1 | -28/+29 |
* | Unbreak mingw build | Anton Korobeynikov | 2007-12-22 | 1 | -1/+2 |
* | Preliminary PIC JIT support for X86 (32-bit) / Darwin. | Evan Cheng | 2007-12-22 | 5 | -65/+89 |
* | Oops. | Evan Cheng | 2007-12-22 | 1 | -2/+1 |
* | implement InstCombine/shift-trunc-shift.ll. This allows | Chris Lattner | 2007-12-22 | 1 | -0/+44 |
* | Note what still needs doing. | Owen Anderson | 2007-12-22 | 1 | -1/+6 |
* | Remove critical edge breaking. It won't be necessary as long as we are very ... | Owen Anderson | 2007-12-22 | 1 | -37/+0 |
* | Fix JIT code emission of X86::MovePCtoStack. | Evan Cheng | 2007-12-22 | 2 | -5/+7 |
* | If succ has succ itself as one of the predecessors then do | Devang Patel | 2007-12-22 | 1 | -2/+2 |
* | Allow JIT with non-static relocation model. | Evan Cheng | 2007-12-22 | 1 | -1/+2 |
* | Fix silly typo in the FP CEP handling. | Anton Korobeynikov | 2007-12-21 | 1 | -0/+1 |