summaryrefslogtreecommitdiff
path: root/lib/Target/R600/R600InstrInfo.cpp
Commit message (Expand)AuthorAge
* [C++11] Replace llvm::next and llvm::prior with std::next and std::prev.Benjamin Kramer2014-03-02
* R600: Remove successive JUMP in AnalyzeBranch when AllowModify is trueTom Stellard2014-01-23
* R600: Implement TargetInstrInfo::isLegalToSplitMBBAt()Tom Stellard2013-11-22
* [weak vtables] Remove a bunch of weak vtablesJuergen Ributzka2013-11-19
* Revert r194865 and r194874.Alexey Samsonov2013-11-18
* R600: Make dot_4 instructions predicableVincent Lejeune2013-11-16
* [weak vtables] Remove a bunch of weak vtablesJuergen Ributzka2013-11-15
* R600: Fix scheduling of instructions that use the LDS output queueTom Stellard2013-11-15
* R600/SI: Add support for private address space load/storeTom Stellard2013-11-13
* R600: Simplify handling of private address spaceTom Stellard2013-10-22
* R600: Remove unused InstrInfo::getMovImmInstr() functionTom Stellard2013-10-22
* R600: add a pass that merges clauses.Vincent Lejeune2013-10-01
* R600: Enable -verify-machineinstrs in some tests.Vincent Lejeune2013-10-01
* IfConverter: Use TargetSchedule for instruction latenciesArnold Schwaighofer2013-09-30
* R600: Don't use trans slot for instructions that read LDS source registersTom Stellard2013-09-12
* R600: Use shared op optimization when checking cycle compatibilityVincent Lejeune2013-09-04
* R600: Non vector only instruction can be scheduled on trans unitVincent Lejeune2013-09-04
* R600: Use SchedModel enum for is{Trans,Vector}Only functionsVincent Lejeune2013-09-04
* R600: Add support for i8 and i16 local memory storesTom Stellard2013-08-26
* R600: Add IsExport bit to TableGen instruction definitionsTom Stellard2013-08-16
* R600: Add 64-bit float load/store supportTom Stellard2013-08-01
* Revert "R600: Non vector only instruction can be scheduled on trans unit"Tom Stellard2013-07-31
* Revert "R600: Use SchedModel enum for is{Trans,Vector}Only functions"Tom Stellard2013-07-31
* R600: Avoid more than 4 literals in the same instruction group at schedulingVincent Lejeune2013-07-31
* R600: Non vector only instruction can be scheduled on trans unitVincent Lejeune2013-07-31
* R600: Use SchedModel enum for is{Trans,Vector}Only functionsVincent Lejeune2013-07-31
* R600: Move CONST_ADDRESS folding into AMDGPUDAGToDAGISel::Select()Tom Stellard2013-07-23
* Use SmallVectorImpl& instead of SmallVector to avoid repeating small vector s...Craig Topper2013-07-14
* Replacing an empty switch with its moral equivalent. No functional changes i...Aaron Ballman2013-07-10
* R600: Do not predicated basic block with multiple alu clauseVincent Lejeune2013-07-09
* R600: Fix an unitialized variable in R600InstrInfo.cppVincent Lejeune2013-06-30
* R600: Unbreak GCC build.Benjamin Kramer2013-06-29
* R600: Support schedule and packetization of trans-only instVincent Lejeune2013-06-29
* R600: Bank Swizzle now display SCL equivalentVincent Lejeune2013-06-29
* R600: Add local memory support via LDSTom Stellard2013-06-28
* R600: Add support for GROUP_BARRIER instructionTom Stellard2013-06-28
* R600: Add ALUInst bit to tablegen definitions v2Tom Stellard2013-06-28
* R600: Use new getNamedOperandIdx function generated by TableGenTom Stellard2013-06-25
* R600: PV stores Reg id, not indexVincent Lejeune2013-06-17
* R600: Rework subtarget info and remove AMDILDevice classesTom Stellard2013-06-07
* Don't cache the instruction and register info from the TargetMachine, becauseBill Wendling2013-06-07
* R600: Make sure to schedule AR register uses and defs in the same clauseTom Stellard2013-06-05
* R600: Const/Neg/Abs can be folded to dot4Vincent Lejeune2013-06-04
* Move passes from namespace llvm into anonymous namespaces. Sort includes whil...Benjamin Kramer2013-05-23
* R600: Relax some vector constraints on Dot4.Vincent Lejeune2013-05-17
* R600: Some factorizationVincent Lejeune2013-05-17
* R600: Remove dead code from the CodeEmitter v2Tom Stellard2013-05-06
* R600: Always use texture cache for compute shadersVincent Lejeune2013-04-30
* R600: Packetize instructionsVincent Lejeune2013-04-30
* R600: Rework Scheduling to handle difference between VLIW4 and VLIW5 chipsVincent Lejeune2013-04-30