2015-06-02 |
Asaf Badouh | revert 238809 |
blob | commitdiff | raw |
2015-06-02 |
Asaf Badouh | AVX-512: Implemented GETEXP instruction for KNL and SKX |
blob | commitdiff | raw | diff to current |
2015-06-01 |
Elena Demikhovsky | AVX-512: Optimized vector shuffle for v16f32 and v16i32... |
blob | commitdiff | raw |
2015-06-01 |
Elena Demikhovsky | AVX-512: Implemented VRANGEPD and VRANGEPD instructions... |
blob | commitdiff | raw | diff to current |
2015-06-01 |
Elena Demikhovsky | AVX-512: Implemented vector shuffle lowering for v8i64... |
blob | commitdiff | raw | diff to current |
2015-06-01 |
Elena Demikhovsky | AVX-512: Implemented VFIXUPIMMPD and VFIXUPIMMPS instru... |
blob | commitdiff | raw | diff to current |
2015-06-01 |
Elena Demikhovsky | AVX-512: Fixed a bug in compress and expand intrinsics. |
blob | commitdiff | raw | diff to current |
2015-06-01 |
Matt Arsenault | Add address space argument to isLegalAddressingMode |
blob | commitdiff | raw | diff to current |
2015-05-30 |
Simon Pilgrim | Stripped trailing whitespace. NFC. |
blob | commitdiff | raw | diff to current |
2015-05-30 |
Chandler Carruth | [x86] Unify the horizontal adding used for popcount... |
blob | commitdiff | raw | diff to current |
2015-05-30 |
Chandler Carruth | [x86] Split out the horizontal byte sum lowering compon... |
blob | commitdiff | raw | diff to current |
2015-05-30 |
Chandler Carruth | [x86] Replace the long spelling of getting a bitcast... |
blob | commitdiff | raw | diff to current |
2015-05-30 |
Chandler Carruth | [x86] Replace the long spelling of getting a bitcast... |
blob | commitdiff | raw | diff to current |
2015-05-30 |
Chandler Carruth | [sdag] Add the helper I most want to the DAG -- buildin... |
blob | commitdiff | raw | diff to current |
2015-05-30 |
Chandler Carruth | [x86] Restore the bitcasts I removed when refactoring... |
blob | commitdiff | raw | diff to current |
2015-05-30 |
Chandler Carruth | [x86] Implement a faster vector population count based... |
blob | commitdiff | raw | diff to current |
2015-05-30 |
Chandler Carruth | [x86] Restructure the parallel bitmath lowering of... |
blob | commitdiff | raw | diff to current |
2015-05-30 |
Jim Grosbach | MC: Clean up MCExpr naming. NFC. |
blob | commitdiff | raw | diff to current |
2015-05-28 |
Reid Kleckner | Remove debug prints from r238487 |
blob | commitdiff | raw | diff to current |
2015-05-28 |
Reid Kleckner | Disable x86 tail call optimizations that jump through GOT |
blob | commitdiff | raw | diff to current |
2015-05-28 |
Reid Kleckner | [WinEH] Remove debugging dump() call |
blob | commitdiff | raw | diff to current |
2015-05-27 |
Elena Demikhovsky | AVX-512: Implemented all forms of sign-extend and zero... |
blob | commitdiff | raw | diff to current |
2015-05-26 |
Elena Demikhovsky | AVX-512: fixed a bug in arithmetic operations lowering... |
blob | commitdiff | raw | diff to current |
2015-05-26 |
Elena Demikhovsky | AVX-512: fixed a bug in lowering VSELECT for 512-bit... |
blob | commitdiff | raw | diff to current |
2015-05-25 |
Simon Pilgrim | [X86][AVX2] Vectorized i16 shift operators |
blob | commitdiff | raw | diff to current |
2015-05-25 |
Elena Demikhovsky | Added promotion to EXTRACT_SUBVECTOR operand. |
blob | commitdiff | raw | diff to current |
2015-05-23 |
Rafael Espindola | Revert "make reciprocal estimate code generation more... |
blob | commitdiff | raw | diff to current |
2015-05-22 |
Sanjay Patel | make reciprocal estimate code generation more flexible... |
blob | commitdiff | raw | diff to current |
2015-05-21 |
Simon Pilgrim | Fixed unused variable warning in non-assert builds... |
blob | commitdiff | raw | diff to current |
2015-05-21 |
Simon Pilgrim | [X86][SSE] Improve support for 128-bit vector sign... |
blob | commitdiff | raw | diff to current |
2015-05-20 |
Reid Kleckner | [WinEH] Store pointers to the LSDA in the exception... |
blob | commitdiff | raw | diff to current |
2015-05-20 |
Hans Wennborg | Revert r237828 "[X86] Remove unused node after morphing... |
blob | commitdiff | raw | diff to current |
2015-05-20 |
Benjamin Kramer | [X86] Remove unused node after morphing it from shr... |
blob | commitdiff | raw | diff to current |
2015-05-20 |
Elena Demikhovsky | AVX-512: fixed algorithm of building vectors of i1... |
blob | commitdiff | raw | diff to current |
2015-05-20 |
David Majnemer | [X86] Implement the local-exec TLS model for Windows... |
blob | commitdiff | raw | diff to current |
2015-05-18 |
Reid Kleckner | Re-land r237175: [X86] Always return the sret parameter... |
blob | commitdiff | raw | diff to current |
2015-05-18 |
David Blaikie | Simplify IRBuilder::CreateCall* by using ArrayRef+initi... |
blob | commitdiff | raw | diff to current |
2015-05-18 |
Elena Demikhovsky | AVX-512: Added intrinsics for ADDSS/D, MULSS/D, SUBSS... |
blob | commitdiff | raw | diff to current |
2015-05-18 |
Elena Demikhovsky | fixed compilation warning/error |
blob | commitdiff | raw | diff to current |
2015-05-17 |
Elena Demikhovsky | AVX-512: fixed extended load to 512-bit register |
blob | commitdiff | raw | diff to current |
2015-05-15 |
David Majnemer | [X86] Use a better sentinel offset for the FrameAddr... |
blob | commitdiff | raw | diff to current |
2015-05-14 |
Elena Demikhovsky | AVX-512: Added i1 type handling for calling conventions. |
blob | commitdiff | raw | diff to current |
2015-05-12 |
Chandler Carruth | Revert r237175: [X86] Always return the sret parameter... |
blob | commitdiff | raw | diff to current |
2015-05-12 |
Reid Kleckner | [X86] Always return the sret parameter in eax/rax,... |
blob | commitdiff | raw | diff to current |
2015-05-12 |
Sanjay Patel | use 'auto' to improve readability; NFC |
blob | commitdiff | raw | diff to current |
2015-05-12 |
Elena Demikhovsky | AVX-512, X86: Added lowering for shift operations for... |
blob | commitdiff | raw | diff to current |
2015-05-12 |
Andrea Di Biagio | [X86] Remove useless target specific combine on TRUNCAT... |
blob | commitdiff | raw | diff to current |
2015-05-12 |
Elena Demikhovsky | AVX-512: select operation for i1 vectors |
blob | commitdiff | raw | diff to current |
2015-05-12 |
Michael Kuperstein | [X86] DAGCombine should not assume arbitrary vector... |
blob | commitdiff | raw | diff to current |
2015-05-12 |
Eric Christopher | Migrate existing backends that care about software... |
blob | commitdiff | raw | diff to current |
2015-05-11 |
Pirama Arumuga Nainar | [X86] Updates to X86 backend for f16 promotion |
blob | commitdiff | raw | diff to current |
2015-05-11 |
Elena Demikhovsky | Fixed compilation warning, NFC. |
blob | commitdiff | raw | diff to current |
2015-05-10 |
Elena Demikhovsky | AVX-512: fixed UINT_TO_FP operation for 512-bit types. |
blob | commitdiff | raw | diff to current |
2015-05-10 |
Elena Demikhovsky | AVX-512: fixed a bug in i1 vectors lowering |
blob | commitdiff | raw | diff to current |
2015-05-08 |
Arnold Schwaighofer | ScheduleDAGInstrs: In functions with tail calls PseudoS... |
blob | commitdiff | raw | diff to current |
2015-05-08 |
Pat Gavlin | Extend the statepoint intrinsic to allow statepoints... |
blob | commitdiff | raw | diff to current |
2015-05-08 |
Andrea Di Biagio | [X86] Teach 'getTargetShuffleMask' how to look through... |
blob | commitdiff | raw | diff to current |
2015-05-08 |
Denis Protivensky | Fix gcc warning of different enum and non-enum types... |
blob | commitdiff | raw | diff to current |
2015-05-07 |
Matthias Braun | Change getTargetNodeName() to produce compiler warnings... |
blob | commitdiff | raw | diff to current |
2015-05-07 |
Elena Demikhovsky | AVX-512: Added all forms of FP compare instructions... |
blob | commitdiff | raw | diff to current |
2015-05-06 |
NAKAMURA Takumi | Revert r236546, "propagate IR-level fast-math-flags... |
blob | commitdiff | raw | diff to current |
2015-05-05 |
Sanjay Patel | propagate IR-level fast-math-flags to DAG nodes (NFC) |
blob | commitdiff | raw | diff to current |
2015-05-04 |
Elena Demikhovsky | AVX-512: added calling convention for i1 vectors in... |
blob | commitdiff | raw | diff to current |
2015-05-03 |
Elena Demikhovsky | Masked gather and scatter intrinsics - enabled codegen... |
blob | commitdiff | raw | diff to current |
2015-05-02 |
Simon Pilgrim | [SSE2] Minor tidyup of v16i8 SHL lowering. NFC. |
blob | commitdiff | raw | diff to current |
2015-04-30 |
Simon Pilgrim | [SSE] Fix for MUL v16i8 on pre-SSE41 targets (PR23369). |
blob | commitdiff | raw | diff to current |
2015-04-29 |
Elena Demikhovsky | fixed 80-chars; NFC |
blob | commitdiff | raw | diff to current |
2015-04-28 |
Eric Christopher | Reuse a lookup in an assert. |
blob | commitdiff | raw | diff to current |
2015-04-28 |
Sanjay Patel | move IR-level optimization flags into their own struct |
blob | commitdiff | raw | diff to current |
2015-04-28 |
Elena Demikhovsky | Fixed crash of variable shift inst on AVX2 |
blob | commitdiff | raw | diff to current |
2015-04-28 |
Sergey Dmitrouk | Reapply r235977 "[DebugInfo] Add debug locations to... |
blob | commitdiff | raw | diff to current |
2015-04-28 |
Daniel Jasper | Revert "[DebugInfo] Add debug locations to constant... |
blob | commitdiff | raw | diff to current |
2015-04-28 |
Sergey Dmitrouk | [DebugInfo] Add debug locations to constant SD nodes |
blob | commitdiff | raw | diff to current |
2015-04-27 |
Elena Demikhovsky | AVX-512: added calling conventions for i1 vectors. |
blob | commitdiff | raw | diff to current |
2015-04-27 |
Elena Demikhovsky | AVX-512: Extend/Truncate operations for SKX, |
blob | commitdiff | raw | diff to current |
2015-04-27 |
Simon Pilgrim | [X86][SSE] Add v16i8/v32i8 multiplication support |
blob | commitdiff | raw | diff to current |
2015-04-22 |
Sanjay Patel | [x86] allow 64-bit extracted vector element integer... |
blob | commitdiff | raw | diff to current |
2015-04-21 |
Matthias Braun | X86: Match for X86ISD nodes in LowerBUILD_VECTOR instea... |
blob | commitdiff | raw | diff to current |
2015-04-21 |
Elena Demikhovsky | AVX-512: Added logical and arithmetic instructions... |
blob | commitdiff | raw | diff to current |
2015-04-21 |
Matthias Braun | X86: Do not select X86 custom vector nodes if operand... |
blob | commitdiff | raw | diff to current |
2015-04-19 |
Simon Pilgrim | [X86][SSE] Fix for getScalarValueForVectorElement to... |
blob | commitdiff | raw | diff to current |
2015-04-15 |
Sanjay Patel | [x86] Implement combineRepeatedFPDivisors |
blob | commitdiff | raw | diff to current |
2015-04-13 |
Krzysztof Parzyszek | Allow memory intrinsics to be tail calls |
blob | commitdiff | raw | diff to current |
2015-04-10 |
Benjamin Kramer | Reduce dyn_cast<> to isa<> or cast<> where possible. |
blob | commitdiff | raw | diff to current |
2015-04-06 |
Simon Pilgrim | [X86][SSE] Use (V)PINSRB for direct byte insertion... |
blob | commitdiff | raw | diff to current |
2015-04-03 |
David Majnemer | [WinEH] Sink UnwindHelp completely out of IR |
blob | commitdiff | raw | diff to current |
2015-04-02 |
Sanjay Patel | [AVX] Improve insertion of i8 or i16 into low element... |
blob | commitdiff | raw | diff to current |
2015-03-31 |
David Majnemer | [WinEH] Generate .xdata for catch handlers |
blob | commitdiff | raw | diff to current |
2015-03-31 |
Sanjay Patel | [X86, AVX] try to lowerVectorShuffleAsElementInsertion... |
blob | commitdiff | raw | diff to current |
2015-03-30 |
Sanjay Patel | more space; NFC |
blob | commitdiff | raw | diff to current |
2015-03-26 |
Sanjay Patel | comment cleanup; NFC |
blob | commitdiff | raw | diff to current |
2015-03-26 |
Sanjay Patel | Use SDValue bool checks; NFC intended |
blob | commitdiff | raw | diff to current |
2015-03-25 |
Sanjay Patel | [X86, AVX] improve insertion into zero element of 256... |
blob | commitdiff | raw | diff to current |
2015-03-24 |
Sanjay Patel | [X86, AVX] recognize shufflevector with zero input... |
blob | commitdiff | raw | diff to current |
2015-03-20 |
Sanjay Patel | [X86] Prefer blendps over insertps codegen for one... |
blob | commitdiff | raw | diff to current |
2015-03-20 |
Simon Pilgrim | Stripped trailing whitespace. NFC. |
blob | commitdiff | raw | diff to current |
2015-03-19 |
Sanjay Patel | move insert, extract, concat helper functions closer... |
blob | commitdiff | raw | diff to current |
2015-03-19 |
Sanjay Patel | [X86, AVX] use blends instead of insert128 with index 0 |
blob | commitdiff | raw | diff to current |
2015-03-18 |
Simon Pilgrim | [X86][SSE] Avoid scalarization of v2i64 vector shifts... |
blob | commitdiff | raw | diff to current |
2015-03-18 |
Eric Christopher | Revert "[X86][SSE] Avoid scalarization of v2i64 vector... |
blob | commitdiff | raw | diff to current |
next |