2013-06-06 |
Arnold Schwaighofer | ARM sched model: Add more ALU and CMP thumb2 instructions
|
commit | commitdiff | tree |
2013-06-05 |
Arnold Schwaighofer | ARM sched model: Add more ALU and CMP instructions
|
commit | commitdiff | tree |
2013-06-05 |
Arnold Schwaighofer | ARM sched model: Add divsion, loads, branches, vfp cvt
|
commit | commitdiff | tree |
2013-06-05 |
Arnold Schwaighofer | ARMInstrInfo: Improve isSwiftFastImmShift
|
commit | commitdiff | tree |
2013-06-05 |
Arnold Schwaighofer | SubtargetEmitter fix
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | Revert series of sched model patches until I figure...
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Add VFP div instruction on Swift
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Add SIMD/VFP load/store instructions...
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Add integer VFP/SIMD instructions...
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Add integer load/store instructions...
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Add integer arithmetic instructions...
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Cortex A9 - More InstRW sched resources
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Add branch thumb instructions
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Add branch thumb2 instructions
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Add branch instructions
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Add preload thumb2 instructions
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Add preload instructions
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Add more ALU and CMP thumb instructions
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Add more ALU and CMP thumb2 instructions
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Add more ALU and CMP instructions
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARM sched model: Add divsion, loads, branches, vfp cvt
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | ARMInstrInfo: Improve isSwiftFastImmShift
|
commit | commitdiff | tree |
2013-06-04 |
Arnold Schwaighofer | SubtargetEmitter fix
|
commit | commitdiff | tree |
2013-05-31 |
Arnold Schwaighofer | LoopVectorize: Change API call to get the backedge...
|
commit | commitdiff | tree |
2013-05-31 |
Arnold Schwaighofer | LoopVectorize: PHIs with only outside users should...
|
commit | commitdiff | tree |
2013-05-22 |
Arnold Schwaighofer | LoopVectorize: Make Value pointers that could be RAUW...
|
commit | commitdiff | tree |
2013-05-18 |
Arnold Schwaighofer | LoopVectorize: Handle single edge PHIs
|
commit | commitdiff | tree |
2013-05-15 |
Arnold Schwaighofer | LoopVectorize: Move call of canHoistAllLoads to canVectorize...
|
commit | commitdiff | tree |
2013-05-15 |
Arnold Schwaighofer | LoopVectorize: Fix comments
|
commit | commitdiff | tree |
2013-05-15 |
Arnold Schwaighofer | LoopVectorize: Hoist conditional loads if possible
|
commit | commitdiff | tree |
2013-05-14 |
Arnold Schwaighofer | ARM ISel: Don't create illegal types during LowerMUL
|
commit | commitdiff | tree |
2013-05-14 |
Arnold Schwaighofer | LoopVectorize: Handle loops with multiple forward inductions
|
commit | commitdiff | tree |
2013-05-11 |
Arnold Schwaighofer | LoopVectorize: Use the widest induction variable type
|
commit | commitdiff | tree |
2013-05-11 |
Arnold Schwaighofer | LoopVectorize: Use variable instead of repeated function...
|
commit | commitdiff | tree |
2013-05-11 |
Arnold Schwaighofer | LoopVectorize: Use IRBuilder interface in more places
|
commit | commitdiff | tree |
2013-05-09 |
Arnold Schwaighofer | LoopVectorizer: Don't assert on the absence of induction...
|
commit | commitdiff | tree |
2013-05-07 |
Arnold Schwaighofer | LoopVectorizer: Improve reduction variable identification
|
commit | commitdiff | tree |
2013-05-07 |
Arnold Schwaighofer | LoopVectorize: getConsecutiveVector must respect signed...
|
commit | commitdiff | tree |
2013-05-05 |
Arnold Schwaighofer | Add missing PatternMatch.cpp to CMakeLists.txt
|
commit | commitdiff | tree |
2013-05-05 |
Arnold Schwaighofer | PatternMatch: Fix documentation - 'function' not 'attribute'
|
commit | commitdiff | tree |
2013-05-05 |
Arnold Schwaighofer | LoopVectorize: Add support for floating point min/max...
|
commit | commitdiff | tree |
2013-05-05 |
Arnold Schwaighofer | PatternMatch: Matcher for (un)ordered floating point...
|
commit | commitdiff | tree |
2013-05-05 |
Arnold Schwaighofer | LoopVectorizer: Cleanup of miminimum/maximum pattern...
|
commit | commitdiff | tree |
2013-05-05 |
Arnold Schwaighofer | LoopVectorize: We don't need an identity element for...
|
commit | commitdiff | tree |
2013-04-29 |
Arnold Schwaighofer | SimplifyCFG: If convert single conditional stores
|
commit | commitdiff | tree |
2013-04-25 |
Arnold Schwaighofer | ARM cost model: Integer div and rem is lowered to a...
|
commit | commitdiff | tree |
2013-04-24 |
Arnold Schwaighofer | LoopVectorizer: Change variable name Stride to ConsecutiveStride
|
commit | commitdiff | tree |
2013-04-24 |
Arnold Schwaighofer | LoopVectorize: Scalarize padded types
|
commit | commitdiff | tree |
2013-04-24 |
Arnold Schwaighofer | LoopVectorizer: Bail out if we don't have datalayout...
|
commit | commitdiff | tree |
2013-04-21 |
Arnold Schwaighofer | Revert "SimplifyCFG: If convert single conditional...
|
commit | commitdiff | tree |
2013-04-20 |
Arnold Schwaighofer | SimplifyCFG: If convert single conditional stores
|
commit | commitdiff | tree |
2013-04-19 |
Arnold Schwaighofer | LoopVectorizer: Use matcher from PatternMatch.h for...
|
commit | commitdiff | tree |
2013-04-18 |
Arnold Schwaighofer | LoopVectorizer: Recognize min/max reductions
|
commit | commitdiff | tree |
2013-04-17 |
Arnold Schwaighofer | X86 cost model: Exit before calling getSimpleVT on...
|
commit | commitdiff | tree |
2013-04-15 |
Arnold Schwaighofer | Fix bit size of v64i8 and v32i16 vector types.
|
commit | commitdiff | tree |
2013-04-12 |
Arnold Schwaighofer | LoopVectorizer: integer division is not a reduction...
|
commit | commitdiff | tree |
2013-04-08 |
Arnold Schwaighofer | X86 cost model: Model cost for uitofp and sitofp on...
|
commit | commitdiff | tree |
2013-04-07 |
Arnold Schwaighofer | TargetLowering: Fix getTypeConversion handling of extended...
|
commit | commitdiff | tree |
2013-04-05 |
Arnold Schwaighofer | ARM scheduler model: Add scheduler info to more instructions...
|
commit | commitdiff | tree |
2013-04-05 |
Arnold Schwaighofer | ARM scheduler model: Swift has varying latencies, uops...
|
commit | commitdiff | tree |
2013-04-04 |
Arnold Schwaighofer | LoopVectorizer: Pass OperandValueKind information to...
|
commit | commitdiff | tree |
2013-04-04 |
Arnold Schwaighofer | X86 cost model: Differentiate cost for vector shifts...
|
commit | commitdiff | tree |
2013-04-04 |
Arnold Schwaighofer | CostModel: Add parameter to instruction cost to further...
|
commit | commitdiff | tree |
2013-04-03 |
Arnold Schwaighofer | X86 cost model: Vector shifts are expensive in most...
|
commit | commitdiff | tree |
2013-04-02 |
Arnold Schwaighofer | DAGCombiner: Merge store/loads when we have extload...
|
commit | commitdiff | tree |
2013-04-01 |
Arnold Schwaighofer | Merge load/store sequences with adresses: base + index...
|
commit | commitdiff | tree |
2013-04-01 |
Arnold Schwaighofer | ARM Scheduler Model: Add resources instructions, map...
|
commit | commitdiff | tree |
2013-03-26 |
Arnold Schwaighofer | BasicAA: Only query twice if the result of the more...
|
commit | commitdiff | tree |
2013-03-26 |
Arnold Schwaighofer | Revert ARM Scheduler Model: Add resources instructions...
|
commit | commitdiff | tree |
2013-03-26 |
Arnold Schwaighofer | ARM Scheduler Model: Add resources instructions, map...
|
commit | commitdiff | tree |
2013-03-26 |
Arnold Schwaighofer | ARM Scheduler Model: Partial implementation of the...
|
commit | commitdiff | tree |
2013-03-18 |
Arnold Schwaighofer | ARM cost model: Make some vector integer to float casts...
|
commit | commitdiff | tree |
2013-03-18 |
Arnold Schwaighofer | ARM cost model: Correct cost for some cheap float to...
|
commit | commitdiff | tree |
2013-03-15 |
Arnold Schwaighofer | ARM cost model: Fix costs for some vector selects
|
commit | commitdiff | tree |
2013-03-15 |
Arnold Schwaighofer | ARM cost model: Fix cost of fptrunc and fpext instructions
|
commit | commitdiff | tree |
2013-03-14 |
Arnold Schwaighofer | LoopVectorizer: Insert some white space to make test...
|
commit | commitdiff | tree |
2013-03-14 |
Arnold Schwaighofer | ARM cost model: Increase cost of some vector selects...
|
commit | commitdiff | tree |
2013-03-14 |
Arnold Schwaighofer | Add missing asserts flag to test - it uses debug flags
|
commit | commitdiff | tree |
2013-03-14 |
Arnold Schwaighofer | LoopVectorize: Invert case when we use a vector cmp...
|
commit | commitdiff | tree |
2013-03-13 |
Arnold Schwaighofer | ARM cost model: Add test case to make sure we would...
|
commit | commitdiff | tree |
2013-03-12 |
Arnold Schwaighofer | ARM cost model: Increase the cost for vector casts...
|
commit | commitdiff | tree |
2013-03-09 |
Arnold Schwaighofer | LoopVectorizer: Ignore all dbg intrinisic
|
commit | commitdiff | tree |
2013-03-09 |
Arnold Schwaighofer | LoopVectorizer: Ignore dbg.value instructions
|
commit | commitdiff | tree |
2013-03-05 |
Arnold Schwaighofer | Use the right number of slashes in comment string
|
commit | commitdiff | tree |
2013-03-05 |
Arnold Schwaighofer | Clarify comment for function getObjectSize
|
commit | commitdiff | tree |
2013-03-02 |
Arnold Schwaighofer | ARM NEON: Fix v2f32 float intrinsics
|
commit | commitdiff | tree |
2013-03-02 |
Arnold Schwaighofer | X86 cost model: Adjust cost for custom lowered vector...
|
commit | commitdiff | tree |
2013-02-20 |
Arnold Schwaighofer | DAGCombiner: Fold pointless truncate, bitcast, buildvector...
|
commit | commitdiff | tree |
2013-02-19 |
Arnold Schwaighofer | ARM NEON: Don't need COPY_TO_REGCLASS in pattern
|
commit | commitdiff | tree |
2013-02-19 |
Arnold Schwaighofer | ARM NEON: Merge a f32 bitcast of a v2i32 extractelt
|
commit | commitdiff | tree |
2013-02-12 |
Arnold Schwaighofer | ARM cost model: Add vector reverse shuffle costs
|
commit | commitdiff | tree |
2013-02-12 |
Arnold Schwaighofer | Cost model: Add check for reverse shuffles to CostModel...
|
commit | commitdiff | tree |
2013-02-12 |
Arnold Schwaighofer | ARM NEON: Handle v16i8 and v8i16 reverse shuffles
|
commit | commitdiff | tree |
2013-02-08 |
Arnold Schwaighofer | ARM cost model: Address computation in vector mem ops...
|
commit | commitdiff | tree |
2013-02-07 |
Arnold Schwaighofer | Loop Vectorizer: Refactor Memory Cost Computation
|
commit | commitdiff | tree |
2013-02-07 |
Arnold Schwaighofer | ARM cost model: Add costs for vector selects
|
commit | commitdiff | tree |
2013-02-05 |
Arnold Schwaighofer | Loop Vectorizer: Refactor code to compute vectorized...
|
commit | commitdiff | tree |
2013-02-05 |
Arnold Schwaighofer | Loop Vectorizer: Handle pointer stores/loads in getWidestType()
|
commit | commitdiff | tree |
2013-02-05 |
Arnold Schwaighofer | ARM cost model: Cost for scalar integer casts and floating...
|
commit | commitdiff | tree |
2013-02-04 |
Arnold Schwaighofer | ARM cost model: Penalize insertelement into D subregisters
|
commit | commitdiff | tree |
|