2010-09-01 |
Bruno Cardoso Lopes | Use movlps, movlpd, movss and movsd specific nodes... |
tree | commitdiff |
2010-09-01 |
Bruno Cardoso Lopes | minor change, simplify some logic |
tree | commitdiff |
2010-09-01 |
Bruno Cardoso Lopes | Move some functions around so they can be used for... |
tree | commitdiff |
2010-08-31 |
Bill Wendling | We have a chance for an optimization. Consider this... |
tree | commitdiff |
2010-08-31 |
Bruno Cardoso Lopes | Use x86 specific MOVSLDUP node, add more patterns to... |
tree | commitdiff |
2010-08-31 |
Bruno Cardoso Lopes | Use x86 specific MOVSHDUP node and add more patterns... |
tree | commitdiff |
2010-08-31 |
Bill Wendling | And ANDS pattern to match the t2ANDS pattern. |
tree | commitdiff |
2010-08-31 |
Jakob Stoklund Olesen | Make %EFLAGS unallocatable. |
tree | commitdiff |
2010-08-31 |
Bruno Cardoso Lopes | Use MOVHLPS node instead of matching using movhlps... |
tree | commitdiff |
2010-08-31 |
Bruno Cardoso Lopes | Use MOVLHPS and MOVHLPS x86 nodes whenever possible... |
tree | commitdiff |
2010-08-31 |
Jim Grosbach | SP relative offsets need to be adjusted by the local... |
tree | commitdiff |
2010-08-31 |
Jim Grosbach | this assert should just be a condition, since this... |
tree | commitdiff |
2010-08-31 |
Bill Wendling | - Cleanup some whitespaces. |
tree | commitdiff |
2010-08-31 |
Bruno Cardoso Lopes | Use X86ISD::MOVSS and MOVSD to represent the movl mask... |
tree | commitdiff |
2010-08-31 |
Eric Christopher | Rewrite slightly so we can expand for floating point... |
tree | commitdiff |
2010-08-30 |
Eric Christopher | If we have an unhandled type then assert, we shouldn... |
tree | commitdiff |
2010-08-30 |
Anton Korobeynikov | Expand MOVi32imm in ARM mode after regalloc. This provides |
tree | commitdiff |
2010-08-30 |
Bill Wendling | Use the existing T2I_bin_s_irs pattern instead of creat... |
tree | commitdiff |
2010-08-30 |
Jakob Stoklund Olesen | Remember to clear the shadow kill flag at the same... |
tree | commitdiff |
2010-08-30 |
Bob Wilson | Remove NEON vmovn intrinsic, replacing it with vector... |
tree | commitdiff |
2010-08-30 |
Jim Grosbach | Make ARM add rN, sp, #imm instructions rematerializable... |
tree | commitdiff |
2010-08-30 |
Bob Wilson | When expanding NEON VST pseudo instructions, if the... |
tree | commitdiff |
2010-08-30 |
Bill Wendling | Create Thumb2sI_cpsr and T2sI_cpsr. These new classes... |
tree | commitdiff |
2010-08-29 |
Kalle Raiskila | Fix lowering of INSERT_VECTOR_ELT in SPU. |
tree | commitdiff |
2010-08-29 |
Bill Wendling | Fix whitespaces. No functionality changes. |
tree | commitdiff |
2010-08-29 |
Bob Wilson | Remove NEON vaddl, vaddw, vsubl, and vsubw intrinsics... |
tree | commitdiff |
2010-08-29 |
Eli Friedman | A couple of small missed optimizations. |
tree | commitdiff |
2010-08-29 |
Bill Wendling | - Add a parameter to T2I_bin_irs for those patterns... |
tree | commitdiff |
2010-08-29 |
Chris Lattner | add a bunch more common shuffles to the instprinter. |
tree | commitdiff |
2010-08-29 |
Bill Wendling | Name ANDflag to ANDS, which is less stupid. |
tree | commitdiff |
2010-08-29 |
Bill Wendling | File missing from last commit. |
tree | commitdiff |
2010-08-29 |
Bill Wendling | Create an ARMISD::AND node. This node is exactly like... |
tree | commitdiff |
2010-08-28 |
Chris Lattner | I have manually decoded the imm field of an insertps... |
tree | commitdiff |
2010-08-28 |
Chris Lattner | fix the buildvector->insertp[sd] logic to not always... |
tree | commitdiff |
2010-08-28 |
Chris Lattner | fix the BuildVector -> unpcklps logic to not do pointle... |
tree | commitdiff |
2010-08-28 |
Chris Lattner | improve comments in the unpcklps generating logic,... |
tree | commitdiff |
2010-08-28 |
Chris Lattner | remove the MSIL backend. It isn't maintained, is buggy... |
tree | commitdiff |
2010-08-28 |
Bob Wilson | Use pseudo instructions for VST1 and VST2. |
tree | commitdiff |
2010-08-28 |
Chris Lattner | remove unions from LLVM IR. They are severely buggy... |
tree | commitdiff |
2010-08-28 |
Bruno Cardoso Lopes | Clean up the logic of vector shuffles -> vector shifts. |
tree | commitdiff |
2010-08-28 |
Bob Wilson | We don't need to custom-select VLDMQ and VSTMQ anymore. |
tree | commitdiff |
2010-08-27 |
Bob Wilson | When merging Thumb2 loads/stores, do not give up when... |
tree | commitdiff |
2010-08-27 |
Bob Wilson | Change ARM VFP VLDM/VSTM instructions to use addressing... |
tree | commitdiff |
2010-08-27 |
Bob Wilson | Unsigned value cannot be < 0. |
tree | commitdiff |
2010-08-27 |
Anton Korobeynikov | Properly handle passing of FP stuff to varargs function... |
tree | commitdiff |
2010-08-27 |
Daniel Dunbar | X86: Fix an encoding issue with LOCK_ADD64mr, which... |
tree | commitdiff |
2010-08-26 |
Jim Grosbach | Simplify eliminateFrameIndex() interface back down... |
tree | commitdiff |
2010-08-26 |
Jim Grosbach | tidy up a bit. no functional change. |
tree | commitdiff |
2010-08-26 |
Jim Grosbach | Turn off the scavenging based frame reg reuse briefly... |
tree | commitdiff |
2010-08-26 |
Bruno Cardoso Lopes | zap the now unused MVT::getIntVectorWithNumElements |
tree | commitdiff |
2010-08-26 |
Bob Wilson | Use pseudo instructions for VST3. |
tree | commitdiff |
2010-08-26 |
Bill Wendling | Reapply r112176 without removing the other CMN patterns... |
tree | commitdiff |
2010-08-26 |
Bob Wilson | Fix comment typos. |
tree | commitdiff |
2010-08-26 |
Jim Grosbach | Restrict the register to tGPR to make sure the str... |
tree | commitdiff |
2010-08-26 |
Dan Gohman | Revert r112176; it broke test/CodeGen/Thumb2/thumb2... |
tree | commitdiff |
2010-08-26 |
Dan Gohman | Reapply r112091 and r111922, support for metadata linki... |
tree | commitdiff |
2010-08-26 |
Bill Wendling | There seems to be a (potential) hardware bug with the... |
tree | commitdiff |
2010-08-26 |
Chris Lattner | implement SplitVecOp_CONCAT_VECTORS, fixing the include... |
tree | commitdiff |
2010-08-26 |
Bob Wilson | Use pseudo instructions for VST1d64Q. |
tree | commitdiff |
2010-08-26 |
Chris Lattner | fix sse1 only codegen in x86-64 mode, which is something we |
tree | commitdiff |
2010-08-26 |
Chris Lattner | remove dead proto |
tree | commitdiff |
2010-08-26 |
Bruno Cardoso Lopes | Fix PR7748 without using microsoft extensions |
tree | commitdiff |
2010-08-26 |
Jim Grosbach | Enable pre-RA virtual frame base register allocation... |
tree | commitdiff |
2010-08-26 |
Bob Wilson | Revert svn 107892 (with changes to work with trunk... |
tree | commitdiff |
2010-08-25 |
Chris Lattner | we should pattern match the SSE complex arithmetic... |
tree | commitdiff |
2010-08-25 |
Bob Wilson | Start converting NEON load/stores to use pseudo instruc... |
tree | commitdiff |
2010-08-25 |
Bruno Cardoso Lopes | Revert this for now, PUNPCKLDQ dont operate on v4f32 |
tree | commitdiff |
2010-08-25 |
Daniel Dunbar | X86: Fix misencode of RI64mi8. This fixes OpenSSL ... |
tree | commitdiff |
2010-08-25 |
Jim Grosbach | Don't override the var from the enclosing scope. |
tree | commitdiff |
2010-08-25 |
Chris Lattner | zap dead code |
tree | commitdiff |
2010-08-25 |
Benjamin Kramer | Remove dead recursive function. Yay for clang -Wunused... |
tree | commitdiff |
2010-08-25 |
Daniel Dunbar | ARM/Thumb2: Fix a misselect in getARMCmp, when attempti... |
tree | commitdiff |
2010-08-25 |
Eric Christopher | Do type checks before we bother to do everything else. |
tree | commitdiff |
2010-08-25 |
Anton Korobeynikov | Fix nasty mingw32 bug, which e.g. prevented llvm-gcc... |
tree | commitdiff |
2010-08-25 |
Eric Christopher | Reorganize load mechanisms. Handle types in a little... |
tree | commitdiff |
2010-08-25 |
Bruno Cardoso Lopes | PUNPCKLDQ should also be used for v4f32 |
tree | commitdiff |
2010-08-25 |
Bruno Cardoso Lopes | teach lowering to get target specific nodes for pshufd... |
tree | commitdiff |
2010-08-24 |
Eric Christopher | Fix predicate and add a comment. |
tree | commitdiff |
2010-08-24 |
Eric Christopher | Rework braindead conditionals I put in yesterday. |
tree | commitdiff |
2010-08-24 |
Eric Christopher | Fix thumb2 mode loads to have the correct operand order... |
tree | commitdiff |
2010-08-24 |
Jim Grosbach | Add ARM heuristic for when to allocate a virtual base... |
tree | commitdiff |
2010-08-24 |
Daniel Dunbar | MC/X86: Tweak imul recognition, previous hack only... |
tree | commitdiff |
2010-08-24 |
Daniel Dunbar | MC/X86: Add custom hack for recognizing "imul $12,... |
tree | commitdiff |
2010-08-24 |
Daniel Dunbar | MC/X86: Warn on scale factors > 1 without index registe... |
tree | commitdiff |
2010-08-24 |
Jim Grosbach | Move enabling the local stack allocation pass into... |
tree | commitdiff |
2010-08-24 |
Jim Grosbach | add ARM cmd line option to force always using virtual... |
tree | commitdiff |
2010-08-24 |
Dan Gohman | Fix X86's isLegalAddressingMode to recognize that stati... |
tree | commitdiff |
2010-08-24 |
Kalle Raiskila | Fix SPU BE to use all the available return registers. |
tree | commitdiff |
2010-08-24 |
Kalle Raiskila | Remove some dead code from SPU BE that remained |
tree | commitdiff |
2010-08-24 |
Bruno Cardoso Lopes | Use pshufhw and pshuflw in more cases and fix getTarget... |
tree | commitdiff |
2010-08-24 |
Bill Wendling | Add comments for what the condition code symbols mean. |
tree | commitdiff |
2010-08-24 |
Eric Christopher | Update comment. |
tree | commitdiff |
2010-08-24 |
Eric Christopher | Fix the opcode and the operands for the load instruction. |
tree | commitdiff |
2010-08-24 |
Eric Christopher | Add register class hack that needs to go away, but... |
tree | commitdiff |
2010-08-24 |
Eric Christopher | Add some more debugging code, make it more obvious... |
tree | commitdiff |
2010-08-23 |
Eric Christopher | Don't need the extra register here. |
tree | commitdiff |
2010-08-23 |
Eric Christopher | Add some more "get address into register" code and... |
tree | commitdiff |
2010-08-23 |
Eric Christopher | Add an ARMFunctionInfo member and use it. |
tree | commitdiff |
2010-08-23 |
Eric Christopher | Start getting ARM loads/address computation going. |
tree | commitdiff |
2010-08-23 |
Bruno Cardoso Lopes | Start using target speficic nodes for shuffles: pshufhw... |
tree | commitdiff |
next |