| Commit message (Expand) | Author | Age | Files | Lines |
* | Update aosp/master llvm for rebase to r233350 | Pirama Arumuga Nainar | 2015-04-09 | 1 | -30/+25 |
* | Update aosp/master LLVM for rebase to r230699. | Stephen Hines | 2015-03-23 | 1 | -419/+1014 |
* | Update aosp/master LLVM for rebase to r222494. | Stephen Hines | 2014-12-02 | 1 | -132/+332 |
* | Update LLVM for rebase to r212749. | Stephen Hines | 2014-07-21 | 1 | -63/+73 |
* | Update LLVM for 3.5 rebase (r209712). | Stephen Hines | 2014-05-29 | 1 | -96/+166 |
* | Update to LLVM 3.5a. | Stephen Hines | 2014-04-24 | 1 | -199/+207 |
* | Merging r195094: | Bill Wendling | 2013-11-19 | 1 | -8/+6 |
* | [weak vtables] Remove a bunch of weak vtables | Juergen Ributzka | 2013-11-19 | 1 | -1/+4 |
* | Revert r194865 and r194874. | Alexey Samsonov | 2013-11-18 | 1 | -4/+1 |
* | Added a size field to the stack map record to handle subregister spills. | Andrew Trick | 2013-11-17 | 1 | -1/+12 |
* | During folding for patchpoint/stackmap instructions, defer creation of new MIs | Lang Hames | 2013-11-15 | 1 | -4/+5 |
* | [weak vtables] Remove a bunch of weak vtables | Juergen Ributzka | 2013-11-15 | 1 | -1/+4 |
* | AVX-512: Handled extractelement from mask vector; | Elena Demikhovsky | 2013-11-14 | 1 | -2/+4 |
* | Cleanup the stackmap operand folding code and fix a corner case. | Andrew Trick | 2013-11-12 | 1 | -6/+12 |
* | Simplify operand folding when rematerializing a load. | Andrew Trick | 2013-11-12 | 1 | -1/+6 |
* | Fix the recently added anyregcc convention to handle spilled operands. | Andrew Trick | 2013-11-11 | 1 | -1/+10 |
* | [Stackmap] Add AnyReg calling convention support for patchpoint intrinsic. | Juergen Ributzka | 2013-11-08 | 1 | -2/+8 |
* | Add support for stack map generation in the X86 backend. | Andrew Trick | 2013-10-31 | 1 | -4/+39 |
* | Replace (V)MOVZDI2PDIrr/rm instructions with patterns that select (V)MOVDI2PD... | Craig Topper | 2013-10-22 | 1 | -2/+0 |
* | Fix the ExecutionDepsFix pass to handle AVX instructions. | Andrew Trick | 2013-10-14 | 1 | -14/+67 |
* | whitespace | Andrew Trick | 2013-10-14 | 1 | -1/+1 |
* | Remove FsMOVAPSrr and friends. They have no patterns and are no longer select... | Craig Topper | 2013-10-07 | 1 | -8/+0 |
* | X86: Don't fold spills into SSE operations if the stack is unaligned. | Benjamin Kramer | 2013-10-06 | 1 | -0/+4 |
* | AVX-512: added scalar convert instructions and intrinsics. | Elena Demikhovsky | 2013-10-06 | 1 | -4/+5 |
* | Add TBM instructions to loading folding tables. | Craig Topper | 2013-10-05 | 1 | -1/+21 |
* | AVX-512: fixed a bug in getLoadStoreRegOpcode() for AVX-512 target | Elena Demikhovsky | 2013-10-02 | 1 | -6/+3 |
* | Add AES and SHA instructions to the load folding tables. | Craig Topper | 2013-09-17 | 1 | -0/+25 |
* | Fix column alignment. No functional change. | Craig Topper | 2013-09-17 | 1 | -4/+4 |
* | AVX-512: updated the list of high-latency instructions. | Elena Demikhovsky | 2013-09-02 | 1 | -1/+23 |
* | AVX-512: gather-scatter tests; added foldable instructions; | Elena Demikhovsky | 2013-09-02 | 1 | -1/+45 |
* | AVX-512: added UNPACK instructions and tests for all-zero/all-ones vectors | Elena Demikhovsky | 2013-08-25 | 1 | -1/+5 |
* | AVX-512: Added VMOVD, VMOVQ, VMOVSS, VMOVSD instructions. | Elena Demikhovsky | 2013-08-18 | 1 | -12/+36 |
* | AVX-512: Added VPERM* instructons and MOV* zmm-to-zmm instructions. | Elena Demikhovsky | 2013-08-11 | 1 | -7/+42 |
* | Add MI-Sched support for x86 macro fusion. | Andrew Trick | 2013-06-23 | 1 | -0/+161 |
* | DebugInfo: remove target-specific Frame Index handling for DBG_VALUE MachineI... | David Blaikie | 2013-06-16 | 1 | -13/+0 |
* | X86: Stop LEA64_32r doing unspeakable things to its arguments. | Tim Northover | 2013-06-10 | 1 | -49/+163 |
* | Don't cache the instruction and register info from the TargetMachine, because | Bill Wendling | 2013-06-07 | 1 | -1/+1 |
* | Revert r183069: "TMP: LEA64_32r fixing" | Tim Northover | 2013-06-01 | 1 | -110/+33 |
* | TMP: LEA64_32r fixing | Tim Northover | 2013-06-01 | 1 | -33/+110 |
* | X86: use sub-register sequences for MOV*r0 operations | Tim Northover | 2013-05-30 | 1 | -50/+15 |
* | X86: change zext moves to use sub-register infrastructure. | Tim Northover | 2013-05-30 | 1 | -10/+1 |
* | Track IR ordering of SelectionDAG nodes 2/4. | Andrew Trick | 2013-05-25 | 1 | -1/+1 |
* | X86: Remove test instructions proceeding shift by immediate instructions | David Majnemer | 2013-05-22 | 1 | -17/+53 |
* | X86: Bad peephole interaction between adc, MOV32r0 | David Majnemer | 2013-05-18 | 1 | -3/+18 |
* | X86: Remove redundant test instructions | David Majnemer | 2013-05-15 | 1 | -7/+39 |
* | ArrayRefize getMachineNode(). No functionality change. | Michael Liao | 2013-04-19 | 1 | -5/+3 |
* | This patch follows is a follow up to r178171, which uses the register | Preston Gurd | 2013-03-27 | 1 | -0/+9 |
* | Move all of the header files which are involved in modelling the LLVM IR | Chandler Carruth | 2013-01-02 | 1 | -2/+2 |
* | Remove the Function::getFnAttributes method in favor of using the AttributeSet | Bill Wendling | 2012-12-30 | 1 | -4/+4 |
* | Remove intrinsic specific instructions for (V)SQRTPS/PD. Instead lower to tar... | Craig Topper | 2012-12-29 | 1 | -15/+0 |