2010-11-09 |
Dan Gohman | Fix DAGCombiner to avoid folding a sext-in-reg or simil... |
tree | commitdiff |
2010-11-09 |
Dale Johannesen | Fix an inline asm pasto from 117667; was preventing |
tree | commitdiff |
2010-11-05 |
Duncan Sands | When passing a parameter using the 'byval' mechanism... |
tree | commitdiff |
2010-11-03 |
Eric Christopher | Just return undef for invalid masks or elts, and since... |
tree | commitdiff |
2010-11-03 |
Duncan Sands | Simplify uses of MVT and EVT. An MVT can be compared... |
tree | commitdiff |
2010-11-03 |
Duncan Sands | Inside the calling convention logic LocVT is always... |
tree | commitdiff |
2010-11-03 |
Eric Christopher | If we have an undef mask our Elt will be -1 for our... |
tree | commitdiff |
2010-11-03 |
Dan Gohman | Fix DAGCombiner to avoid going into an infinite loop... |
tree | commitdiff |
2010-11-03 |
Evan Cheng | Two sets of changes. Sorry they are intermingled. |
tree | commitdiff |
2010-11-02 |
Devang Patel | If value map does not have register for an argument... |
tree | commitdiff |
2010-11-02 |
Devang Patel | Use frameindex, if available, as a last resort to emit... |
tree | commitdiff |
2010-10-29 |
Bob Wilson | Remove DAG combiner patch to fold vector splats. Instc... |
tree | commitdiff |
2010-10-29 |
Evan Cheng | Avoiding overly aggressive latency scheduling. If the... |
tree | commitdiff |
2010-10-29 |
John Thompson | Inline asm multiple alternative constraints development... |
tree | commitdiff |
2010-10-28 |
Bob Wilson | Teach the DAG combiner to fold a splat of a splat.... |
tree | commitdiff |
2010-10-28 |
Evan Cheng | Re-commit 117518 and 117519 now that ARM MC test failur... |
tree | commitdiff |
2010-10-28 |
Evan Cheng | Revert 117518 and 117519 for now. They changed scheduli... |
tree | commitdiff |
2010-10-28 |
Evan Cheng | Fix a major bug in operand latency computation. The... |
tree | commitdiff |
2010-10-26 |
Dale Johannesen | Use a MemIntrinsicSDNode for ISD::PREFETCH, which touches |
tree | commitdiff |
2010-10-26 |
Devang Patel | Assign source ordering to nodes created for StoreInst. |
tree | commitdiff |
2010-10-26 |
Nick Lewycky | For statistics that are only used in functions declared... |
tree | commitdiff |
2010-10-25 |
Devang Patel | Simplify. |
tree | commitdiff |
2010-10-25 |
Devang Patel | Add counters to count basic blocks and machine basic... |
tree | commitdiff |
2010-10-23 |
Chandler Carruth | Move the remaining attribute macros to systematic names... |
tree | commitdiff |
2010-10-21 |
Michael J. Spencer | X86: Base _fltused on the FunctionType of the called... |
tree | commitdiff |
2010-10-21 |
Michael J. Spencer | CodeGen-Windows: Only emit _fltused if a VarArg functio... |
tree | commitdiff |
2010-10-20 |
Dale Johannesen | Remove Synthesizable from the Type system; as MMX vector |
tree | commitdiff |
2010-10-20 |
Dan Gohman | Make CodeGen TBAA-aware. |
tree | commitdiff |
2010-10-19 |
Jim Grosbach | Add a pre-dispatch SjLj EH hook on the unwind edge... |
tree | commitdiff |
2010-10-19 |
Owen Anderson | Get rid of static constructors for pass registration... |
tree | commitdiff |
2010-10-16 |
Michael J. Spencer | X86-Windows: Emit an undefined global __fltused symbol... |
tree | commitdiff |
2010-10-16 |
Michael J. Spencer | Whitespace! |
tree | commitdiff |
2010-10-10 |
Chris Lattner | fix the default va_arg expansion (in the realignment... |
tree | commitdiff |
2010-10-06 |
Dan Gohman | ComputeLinearIndex doesn't need its TLI argument. |
tree | commitdiff |
2010-10-06 |
Evan Cheng | - Add TargetInstrInfo::getOperandLatency() to compute... |
tree | commitdiff |
2010-10-05 |
Owen Anderson | Use a more efficient lowering of uint64_t --> float... |
tree | commitdiff |
2010-10-04 |
Evan Cheng | This DAG combine BRCOND transformation can look pass... |
tree | commitdiff |
2010-10-01 |
Devang Patel | Fix code gen crash reported in PR 8235. We still lose... |
tree | commitdiff |
2010-10-01 |
Gabor Greif | typo |
tree | commitdiff |
2010-10-01 |
Chris Lattner | fix typo |
tree | commitdiff |
2010-10-01 |
Chris Lattner | fix rdar://8494845 + PR8244 - a miscompile exposed... |
tree | commitdiff |
2010-09-30 |
Dale Johannesen | Massive rewrite of MMX: |
tree | commitdiff |
2010-09-30 |
Jakob Stoklund Olesen | When isel is emitting instructions for an x86 target... |
tree | commitdiff |
2010-09-29 |
Evan Cheng | Model Cortex-a9 load to SUB, RSB, ADD, ADC, SBC, RSC... |
tree | commitdiff |
2010-09-28 |
Oscar Fuentes | Removed a bunch of unnecessary target_link_libraries. |
tree | commitdiff |
2010-09-27 |
Dale Johannesen | Don't try to make a vector of x86mmx; this won't work, |
tree | commitdiff |
2010-09-24 |
John Thompson | Fix for test/CodeGen/PowerPC/2008-10-17-AsmMatchingOper... |
tree | commitdiff |
2010-09-24 |
Michael J. Spencer | Get rid of pop_macro warnings on MSVC. |
tree | commitdiff |
2010-09-23 |
Evan Cheng | Revert 114634 for now since buildbot claim it broke... |
tree | commitdiff |
2010-09-23 |
Oscar Fuentes | Fix VS 2010 build. |
tree | commitdiff |
2010-09-23 |
Evan Cheng | Follow up to r114630. Do not optimize away unconditiona... |
tree | commitdiff |
2010-09-23 |
Evan Cheng | SDISel should not optimize a unconditional branch follo... |
tree | commitdiff |
2010-09-22 |
Owen Anderson | A select between a constant and zero, when fed by a... |
tree | commitdiff |
2010-09-21 |
John Thompson | Fixed pr20314-2.c failure, added E, F, p constraint... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | Rework passing parent pointers into complexpatterns... |
tree | commitdiff |
2010-09-21 |
Devang Patel | If only user of a vreg is an copy instruction to export... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | correct this logic. |
tree | commitdiff |
2010-09-21 |
Owen Anderson | Reimplement r114460 in target-independent DAGCombine... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | just like they can opt into getting the root of the... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | finish pushing MachinePointerInfo through selectiondags... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | update a bunch of code to use the MachinePointerInfo... |
tree | commitdiff |
2010-09-21 |
Bob Wilson | Define the TargetLowering::getTgtMemIntrinsic hook... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | eliminate an old SelectionDAG::getTruncStore method... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | eliminate last SelectionDAG::getLoad old entrypoint... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | fix the code that infers SV info to be correct when... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | propagate MachinePointerInfo through various uses of... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | continue MachinePointerInfo'izing, eliminating use... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | convert dagcombine off the old form of getLoad. This... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | simplify DAGCombiner::SimplifySelectOps step #2/2. |
tree | commitdiff |
2010-09-21 |
Chris Lattner | substantially reduce indentation and simplify DAGCombin... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | a few more trivial updates. This fixes PerformInsertVe... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | reimplement memcpy/memmove/memset lowering to use Machi... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | add overloads for SelectionDAG::getLoad, getStore,... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | simplify interface to SelectionDAG::getMemIntrinsicNode... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | chagne interface to SelectionDAG::getAtomic to take... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | don't implicitly drop the offset of a machinememoperand... |
tree | commitdiff |
2010-09-21 |
Chris Lattner | force clients of MachineFunction::getMachineMemOperand... |
tree | commitdiff |
2010-09-20 |
Owen Anderson | When TCO is turned on, it is possible to end up with... |
tree | commitdiff |
2010-09-19 |
Owen Anderson | Revert r114312 while I sort out some issues. |
tree | commitdiff |
2010-09-19 |
Owen Anderson | Tentatively enabled DAGCombiner Alias Analysis by defau... |
tree | commitdiff |
2010-09-18 |
Owen Anderson | Invert the logic of reachesChainWithoutSideEffects... |
tree | commitdiff |
2010-09-15 |
Devang Patel | Check bb to ensure that alloca is in separate basic... |
tree | commitdiff |
2010-09-15 |
Devang Patel | If dbg.declare from non-entry block is using alloca... |
tree | commitdiff |
2010-09-14 |
Devang Patel | Use frame index, if available for byval argument while... |
tree | commitdiff |
2010-09-13 |
Michael J. Spencer | Revert "CMake: Get rid of LLVMLibDeps.cmake and export... |
tree | commitdiff |
2010-09-13 |
Eric Christopher | Silence more warnings. Two more unused variables. |
tree | commitdiff |
2010-09-13 |
John Thompson | Added skeleton for inline asm multiple alternative... |
tree | commitdiff |
2010-09-10 |
Michael J. Spencer | CMake: Get rid of LLVMLibDeps.cmake and export the... |
tree | commitdiff |
2010-09-10 |
Devang Patel | Add DEBUG message. |
tree | commitdiff |
2010-09-10 |
Evan Cheng | Teach if-converter to be more careful with predicating... |
tree | commitdiff |
2010-09-05 |
Chris Lattner | implement rdar://6653118 - fastisel should fold loads... |
tree | commitdiff |
2010-09-03 |
Bob Wilson | Add a missing check when legalizing a vector extending... |
tree | commitdiff |
2010-09-02 |
Devang Patel | Detect undef value early and save unnecessary NodeMap... |
tree | commitdiff |
2010-09-02 |
Dan Gohman | Don't narrow the load and store in a load+twiddle+store... |
tree | commitdiff |
2010-09-02 |
Devang Patel | Tidy up. |
tree | commitdiff |
2010-08-31 |
Devang Patel | Reapply r112623. Included additional check for unused... |
tree | commitdiff |
2010-08-31 |
Devang Patel | Revert r112623. It is causing self host build failures. |
tree | commitdiff |
2010-08-31 |
Devang Patel | Remember byval argument's frame index during argument... |
tree | commitdiff |
2010-08-31 |
Devang Patel | Offset is not always unsigned number. |
tree | commitdiff |
2010-08-31 |
Bruno Cardoso Lopes | zap unused method. x86 is the only user and already... |
tree | commitdiff |
next |