2014-08-05 |
Chandler Carruth | [x86] Reformat some code I moved around in a prior... |
blob | commitdiff | raw |
2014-08-05 |
Chandler Carruth | [x86] Fix a crash and wrong-code bug in the new vector... |
blob | commitdiff | raw | diff to current |
2014-08-05 |
Eric Christopher | Have MachineFunction cache a pointer to the subtarget... |
blob | commitdiff | raw | diff to current |
2014-08-04 |
Eric Christopher | Remove the TargetMachine forwards for TargetSubtargetIn... |
blob | commitdiff | raw | diff to current |
2014-08-04 |
Chandler Carruth | [x86] Just unilaterally prefer SSSE3-style PSHUFB lower... |
blob | commitdiff | raw | diff to current |
2014-08-04 |
Chandler Carruth | [x86] Implement more aggressive use of PACKUS chains... |
blob | commitdiff | raw | diff to current |
2014-08-04 |
Chandler Carruth | [x86] Handle single input shuffles in the SSSE3 case... |
blob | commitdiff | raw | diff to current |
2014-08-02 |
Chandler Carruth | [x86] Remove the FIXME that was implemented in r214628... |
blob | commitdiff | raw | diff to current |
2014-08-02 |
Chandler Carruth | [x86] Largely complete the use of PSHUFB in the new... |
blob | commitdiff | raw | diff to current |
2014-08-02 |
Chandler Carruth | [x86] Fix a few typos in my comments spotted in passing. |
blob | commitdiff | raw | diff to current |
2014-08-02 |
Chandler Carruth | [x86] Teach the target shuffle mask extraction to recog... |
blob | commitdiff | raw | diff to current |
2014-08-01 |
Akira Hatanaka | [X86] Simplify X87 stackifier pass. |
blob | commitdiff | raw | diff to current |
2014-07-31 |
Louis Gerbarg | Make sure no loads resulting from load->switch DAGCombi... |
blob | commitdiff | raw | diff to current |
2014-07-27 |
Matt Arsenault | Add alignment value to allowsUnalignedMemoryAccess |
blob | commitdiff | raw | diff to current |
2014-07-27 |
Chandler Carruth | [x86] Sink a variable only used by asserts into the... |
blob | commitdiff | raw | diff to current |
2014-07-27 |
Chandler Carruth | [x86] Add a much more powerful framework for combining... |
blob | commitdiff | raw | diff to current |
2014-07-26 |
Chandler Carruth | [x86] Fix PR20355 (for real). There are many layers... |
blob | commitdiff | raw | diff to current |
2014-07-26 |
Chandler Carruth | [x86] Revert r214007: Fix PR20355 ... |
blob | commitdiff | raw | diff to current |
2014-07-26 |
Chandler Carruth | [x86] Fix PR20355 (and dups) by not using unsigned... |
blob | commitdiff | raw | diff to current |
2014-07-25 |
Akira Hatanaka | [stack protector] Fix a potential security bug in stack... |
blob | commitdiff | raw | diff to current |
2014-07-25 |
Chandler Carruth | [SDAG] Enable the new assert for out-of-range result... |
blob | commitdiff | raw | diff to current |
2014-07-24 |
Chandler Carruth | [x86] Make vector legalization of extloads work more... |
blob | commitdiff | raw | diff to current |
2014-07-24 |
Reid Kleckner | Replace an assertion with a fatal error |
blob | commitdiff | raw | diff to current |
2014-07-24 |
Saleem Abdulrasool | X86: silence sign comparison warning |
blob | commitdiff | raw | diff to current |
2014-07-24 |
Filipe Cabecinhas | Fixed PR20411 - bug in getINSERTPS() |
blob | commitdiff | raw | diff to current |
2014-07-23 |
Jim Grosbach | [X86,AArch64] Extend vcmp w/ unary op combine to work... |
blob | commitdiff | raw | diff to current |
2014-07-23 |
Jim Grosbach | X86: restrict combine to when type sizes are safe. |
blob | commitdiff | raw | diff to current |
2014-07-23 |
Robert Khasanov | [SKX] Enabling mask instructions: encoding, lowering |
blob | commitdiff | raw | diff to current |
2014-07-23 |
Andrea Di Biagio | Revert r211771. It was: "[X86] Improve the selection... |
blob | commitdiff | raw | diff to current |
2014-07-21 |
Andrea Di Biagio | [DAG] Refactor some logic. No functional change. |
blob | commitdiff | raw | diff to current |
2014-07-18 |
Tim Northover | X86: support fpext/fptrunc operations to and from 16... |
blob | commitdiff | raw | diff to current |
2014-07-18 |
Jim Grosbach | X86: Constant fold converting vector setcc results... |
blob | commitdiff | raw | diff to current |
2014-07-17 |
Tim Northover | CodeGen: generate single libcall for fptrunc -> f16... |
blob | commitdiff | raw | diff to current |
2014-07-17 |
Tim Northover | X86: support double extension of f16 type. |
blob | commitdiff | raw | diff to current |
2014-07-17 |
Tim Northover | CodeGen: extend f16 conversions to permit types > float. |
blob | commitdiff | raw | diff to current |
2014-07-16 |
Tim Northover | CodeGen: don't form illegail EXTLOAD operations. |
blob | commitdiff | raw | diff to current |
2014-07-16 |
Andrea Di Biagio | [X86] Add a check for 'isMOVHLPSMask' within method... |
blob | commitdiff | raw | diff to current |
2014-07-15 |
David Majnemer | Fix typo in comment |
blob | commitdiff | raw | diff to current |
2014-07-11 |
Quentin Colombet | [X86] Fix the inversion of low and high bits for the... |
blob | commitdiff | raw | diff to current |
2014-07-10 |
Chandler Carruth | [x86] Add another combine that is particularly useful... |
blob | commitdiff | raw | diff to current |
2014-07-10 |
Chandler Carruth | [x86] Expand the target DAG combining for PSHUFD nodes... |
blob | commitdiff | raw | diff to current |
2014-07-10 |
Chandler Carruth | [x86] Tweak the v16i8 single input special case lowerin... |
blob | commitdiff | raw | diff to current |
2014-07-10 |
Chandler Carruth | [x86] Initial improvements to the new shuffle lowering... |
blob | commitdiff | raw | diff to current |
2014-07-10 |
Chandler Carruth | [x86] Refactor some of the new code for lowering v16i8... |
blob | commitdiff | raw | diff to current |
2014-07-09 |
Chandler Carruth | [SDAG] Make the new zext-vector-inreg node default... |
blob | commitdiff | raw | diff to current |
2014-07-09 |
Benjamin Kramer | X86: When lowering v8i32 himuls use the correct shuffle... |
blob | commitdiff | raw | diff to current |
2014-07-09 |
Chandler Carruth | [x86] Add a ZERO_EXTEND_VECTOR_INREG DAG node and use... |
blob | commitdiff | raw | diff to current |
2014-07-09 |
Chandler Carruth | [x86] Initialize a pointer to null to fix a bug in... |
blob | commitdiff | raw | diff to current |
2014-07-09 |
Chandler Carruth | [x86] Re-apply a variant of the x86 side of r212324... |
blob | commitdiff | raw | diff to current |
2014-07-08 |
Chandler Carruth | [x86,SDAG] Sink the logic for folding shuffles of splat... |
blob | commitdiff | raw | diff to current |
2014-07-07 |
Andrea Di Biagio | [x86] Fix assertion failure caused by a wrong combine... |
blob | commitdiff | raw | diff to current |
2014-07-07 |
Chandler Carruth | [x86] Revert r212324 which was too aggressive w.r.t... |
blob | commitdiff | raw | diff to current |
2014-07-07 |
Chandler Carruth | [x86] Teach the new vector shuffle lowering code to... |
blob | commitdiff | raw | diff to current |
2014-07-04 |
Chandler Carruth | [x86] Generalize BuildVectorSDNode::getConstantSplatVal... |
blob | commitdiff | raw | diff to current |
2014-07-03 |
Chandler Carruth | [x86] Clarify that this lowering only applies to vector... |
blob | commitdiff | raw | diff to current |
2014-07-03 |
Andrea Di Biagio | [X86] Add ISel patterns to select 'f32_to_f16' and... |
blob | commitdiff | raw | diff to current |
2014-07-03 |
Chandler Carruth | [x86] Fix crashes in lowering bitcast instructions... |
blob | commitdiff | raw | diff to current |
2014-07-03 |
Chandler Carruth | [x86] Based on a long conversation between myself,... |
blob | commitdiff | raw | diff to current |
2014-07-02 |
Benjamin Kramer | X86: When combining shuffles just remove shuffles that... |
blob | commitdiff | raw | diff to current |
2014-07-01 |
Juergen Ributzka | [DAG] Pass the argument list to the CallLoweringInfo... |
blob | commitdiff | raw | diff to current |
2014-07-01 |
Tim Northover | X86: delegate expanding atomic libcalls to generic... |
blob | commitdiff | raw | diff to current |
2014-07-01 |
Tim Northover | X86: expand atomics in IR instead of as MachineInstrs. |
blob | commitdiff | raw | diff to current |
2014-06-30 |
Andrea Di Biagio | [X86] Add support for builtin to read performance monit... |
blob | commitdiff | raw | diff to current |
2014-06-28 |
Chandler Carruth | [x86] Fix a bug in the v8i16 shuffling exposed by the... |
blob | commitdiff | raw | diff to current |
2014-06-28 |
Chandler Carruth | [x86] Add handling for splat-like widenings of v16i8... |
blob | commitdiff | raw | diff to current |
2014-06-27 |
Chandler Carruth | [x86] Fix another bug hit when bootstrapping with the... |
blob | commitdiff | raw | diff to current |
2014-06-27 |
Chandler Carruth | [x86] Fix a miscompile in the new shuffle lowering... |
blob | commitdiff | raw | diff to current |
2014-06-27 |
Alexander Kornienko | Clean up unused variable warning in release build. |
blob | commitdiff | raw | diff to current |
2014-06-27 |
Chandler Carruth | [x86] Clean up some unused variables, especially in... |
blob | commitdiff | raw | diff to current |
2014-06-27 |
Chandler Carruth | [x86] Teach the target combine step to aggressively... |
blob | commitdiff | raw | diff to current |
2014-06-27 |
Chandler Carruth | [x86] Teach the target-specific combining how to aggres... |
blob | commitdiff | raw | diff to current |
2014-06-27 |
Chandler Carruth | [x86] Teach the X86 backend to DAG-combine SSE2 shuffle... |
blob | commitdiff | raw | diff to current |
2014-06-27 |
Chandler Carruth | [x86] Begin a significant overhaul of how vector loweri... |
blob | commitdiff | raw | diff to current |
2014-06-26 |
Andrea Di Biagio | Silence a warning due to a comparison between signed... |
blob | commitdiff | raw | diff to current |
2014-06-26 |
Andrea Di Biagio | [X86] Improve the selection of SSE3/AVX addsub instruct... |
blob | commitdiff | raw | diff to current |
2014-06-25 |
Andrea Di Biagio | [X86] Always prefer to lower a VECTOR_SHUFFLE into... |
blob | commitdiff | raw | diff to current |
2014-06-25 |
Chandler Carruth | [x86] Add intrinsics for the pshufd, pshuflw, and pshuf... |
blob | commitdiff | raw | diff to current |
2014-06-25 |
NAKAMURA Takumi | Re-apply r211399, "Generate native unwind info on Win64... |
blob | commitdiff | raw | diff to current |
2014-06-25 |
Andrea Di Biagio | [X86] Add target combine rule to select ADDSUB instruct... |
blob | commitdiff | raw | diff to current |
2014-06-24 |
Robert Khasanov | vpblend intrinsics combines as shifts intrinsics due... |
blob | commitdiff | raw | diff to current |
2014-06-22 |
NAKAMURA Takumi | Revert r211399, "Generate native unwind info on Win64" |
blob | commitdiff | raw | diff to current |
2014-06-22 |
Filipe Cabecinhas | Fix PR20087 by using the source index when changing... |
blob | commitdiff | raw | diff to current |
2014-06-20 |
Reid Kleckner | Generate native unwind info on Win64 |
blob | commitdiff | raw | diff to current |
2014-06-20 |
Chandler Carruth | [x86] Make the x86 PACKSSWB, PACKSSDW, PACKUSWB, and... |
blob | commitdiff | raw | diff to current |
2014-06-19 |
Andrea Di Biagio | [X86] Teach how to combine horizontal binop even in... |
blob | commitdiff | raw | diff to current |
2014-06-16 |
Cameron McInally | Hook up vector int_ctlz for AVX512. |
blob | commitdiff | raw | diff to current |
2014-06-13 |
Tim Northover | X86: lower ATOMIC_CMP_SWAP_WITH_SUCCESS directly |
blob | commitdiff | raw | diff to current |
2014-06-13 |
Tim Northover | IR: add "cmpxchg weak" variant to support permitted... |
blob | commitdiff | raw | diff to current |
2014-06-12 |
Andrea Di Biagio | [X86] Teach how to dump the name of target node RDTSCP_DAG. |
blob | commitdiff | raw | diff to current |
2014-06-12 |
Andrea Di Biagio | [X86] Teach how to combine AVX and AVX2 horizontal... |
blob | commitdiff | raw | diff to current |
2014-06-11 |
Tim Northover | X86: add stringy name for X86ISD::LCMPXCHG16_DAG |
blob | commitdiff | raw | diff to current |
2014-06-11 |
Andrea Di Biagio | [X86] Refactor the logic to select horizontal adds... |
blob | commitdiff | raw | diff to current |
2014-06-10 |
Eric Christopher | Use the TargetMachine on the DAG or the MachineFunction... |
blob | commitdiff | raw | diff to current |
2014-06-10 |
Eric Christopher | Add a FIXME. |
blob | commitdiff | raw | diff to current |
2014-06-10 |
Andrea Di Biagio | [X86] Improved target combine rules for selecting horiz... |
blob | commitdiff | raw | diff to current |
2014-06-10 |
Tom Stellard | SelectionDAG: Don't use MVT::Other to determine legalit... |
blob | commitdiff | raw | diff to current |
2014-06-10 |
Tim Northover | Revert "X86: elide comparisons after cmpxchg instructions." |
blob | commitdiff | raw | diff to current |
2014-06-10 |
Tim Northover | X86: elide comparisons after cmpxchg instructions. |
blob | commitdiff | raw | diff to current |
2014-06-09 |
Eric Christopher | Move all of the x86 subtarget initialized variables... |
blob | commitdiff | raw | diff to current |
2014-06-09 |
Andrea Di Biagio | [X86] Add target combine rules for horizontal add/sub. |
blob | commitdiff | raw | diff to current |
next |