RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-22 12:33:33 +02:00
Code Issues Projects Releases Wiki Activity
26a5894f0f
llvm-mirror/test/Transforms
History
Philip Reames e6c29ed949 [CGP] Duplicate addressing computation in cold paths if required to sink addressing mode 
This patch teaches CGP to duplicate addressing mode computations into cold paths (detected via explicit cold attribute on calls) if required to let addressing mode be safely sunk into the basic block containing each load and store.

In general, duplicating code into cold blocks may result in code growth, but should not effect performance. In this case, it's better to duplicate some code than to put extra pressure on the register allocator by making it keep the address through the entirely of the fast path.

This patch only handles addressing computations, but in principal, we could implement a more general cold cold scheduling heuristic which tries to reduce register pressure in the fast path by duplicating code into the cold path. Getting the profitability of the general case right seemed likely to be challenging, so I stuck to the existing case (addressing computation) we already had.

Differential Revision: http://reviews.llvm.org/D17652

llvm-svn: 263074
2016-03-09 23:13:12 +00:00
..
ADCE
AddDiscriminators
AlignmentFromAssumptions
ArgumentPromotion
AtomicExpand ARM: sink atomic release barrier as far as possible into cmpxchg. 2016-02-22 20:55:50 +00:00
BBVectorize
BDCE
BranchFolding
CodeExtractor
CodeGenPrepare [CGP] Duplicate addressing computation in cold paths if required to sink addressing mode 2016-03-09 23:13:12 +00:00
ConstantHoisting
ConstantMerge
ConstProp
CorrelatedValuePropagation [LVI] Fix a bug which prevented use of !range metadata within a query 2016-03-04 22:27:39 +00:00
CrossDSOCFI [cfi] Cross-DSO CFI diagnostic mode (LLVM part). 2016-01-25 23:35:03 +00:00
DeadArgElim
DeadStoreElimination [BasicAliasAnalysis] Take into account operand bundles in the getModRefInfo function 2016-01-16 12:15:53 +00:00
EarlyCSE
EliminateAvailableExternally
Float2Int
ForcedFunctionAttrs
FunctionAttrs Revert "[attrs] Handle convergent CallSites." 2016-02-22 18:24:43 +00:00
FunctionImport FunctionImport: add a progressive heuristic to limit importing too deep in the callgraph 2016-02-10 23:31:45 +00:00
GCOVProfiling [GCOV] Avoid emitting profile arcs for module and skeleton CUs 2016-01-21 17:04:42 +00:00
GlobalDCE
GlobalOpt Add another test for the GlobalOpt change in r212079. 2016-03-02 20:02:25 +00:00
GVN Fix tests that used CHECK-NEXT-NOT and CHECK-DAG-NOT. 2016-02-26 19:40:34 +00:00
IndVarSimplify [SCEV] Try to reuse existing value during SCEV expansion 2016-02-04 01:27:38 +00:00
InferFunctionAttrs [BasicAA] Remove special casing of memset_pattern16 in favor of generic attribute inference 2016-01-06 04:53:16 +00:00
Inline Revert revisions 262636, 262643, 262679, and 262682. 2016-03-08 00:36:35 +00:00
InstCombine [InstCombine] (icmp sgt smin(PosA, B) 0) -> (icmp sgt B 0) 2016-03-09 21:05:07 +00:00
InstMerge
InstSimplify [InstSimplify] Restore fsub 0.0, (fsub 0.0, X) ==> X optzn 2016-02-29 12:18:25 +00:00
Internalize
IPConstantProp [SCCP] Don't violate the lattice invariants 2016-01-07 21:36:16 +00:00
IRCE [SCEV] Try to reuse existing value during SCEV expansion 2016-02-04 01:27:38 +00:00
JumpThreading [JumpThreading] Simplify Instructions first in ComputeValueKnownInPredecessors() 2016-02-26 06:06:04 +00:00
LCSSA
LICM [LICM] Store promotion when memory is thread local 2016-03-09 22:59:30 +00:00
LoadCombine
LoopDataPrefetch/PowerPC [LoopDataPrefetch] Make it testable with opt 2016-02-22 21:41:22 +00:00
LoopDeletion
LoopDistribute
LoopIdiom [LIR] Avoid turning non-temporal stores into memset 2016-02-17 21:00:06 +00:00
LoopInterchange
LoopLoadElim [LLE] Add missing check for unit stride 2016-03-09 20:47:55 +00:00
LoopReroll Allow setting MaxRerollIterations above 16 2016-02-22 09:38:28 +00:00
LoopRotate [LoopRotate] Don't perform loop rotation if the loop header calls a convergent function. 2016-02-12 21:01:33 +00:00
LoopSimplify
LoopSimplifyCFG Add LoopSimplifyCFG pass 2016-01-29 22:35:36 +00:00
LoopStrengthReduce [SCEV] Try to reuse existing value during SCEV expansion 2016-02-04 01:27:38 +00:00
LoopUnroll [LoopUnrolling] Fix a bug introduced in r259869 (PR26688). 2016-02-22 21:21:45 +00:00
LoopUnswitch
LoopVectorize [x86] fix cost model inaccuracy for vector memory ops 2016-03-09 22:23:33 +00:00
LoopVersioning [LoopVersioning] Expose loop versioning as a pass too 2016-02-03 00:06:10 +00:00
LoopVersioningLICM Corrected tests for Loop Versioning LICM, by adding “REQUIRES: asserts”. 2016-02-06 12:34:41 +00:00
LowerAtomic
LowerBitSets
LowerExpectIntrinsic
LowerInvoke
LowerSwitch
Mem2Reg
MemCpyOpt Fix PR26051: Memcpy optimization should introduce a call to memcpy before the store destination position 2016-01-06 23:50:22 +00:00
MergeFunc
MetaRenamer
NaryReassociate
ObjCARC [ObjCARC] Handle ARCInstKind::ClaimRV in OptimizeIndividualCalls. 2016-02-17 18:51:27 +00:00
PartiallyInlineLibCalls
PGOProfile [PGO] Indirect-call profile annotation in IR level profiling 2016-02-10 18:24:45 +00:00
PhaseOrdering
PlaceSafepoints [PlaceSafepoints] Clamp NoStatepoints to true 2016-01-28 21:51:14 +00:00
PruneEH [PruneEH] Don't try to insert a terminator after another terminator 2016-01-23 06:00:44 +00:00
Reassociate Reassociate: Reprocess RedoInsts after each inst 2016-01-26 18:42:36 +00:00
Reg2Mem
RewriteStatepointsForGC [RS4GC] "Constant fold" the rs4gc-split-vector-values flag 2016-02-22 21:01:28 +00:00
SafeStack [safestack] Make sure the unsafe stack pointer is popped in all cases 2016-02-02 01:03:11 +00:00
SampleProfile Perform InstructioinCombiningPass before SampleProfile pass. 2016-03-01 22:53:02 +00:00
Scalarizer
ScalarRepl
SCCP
SeparateConstOffsetFromGEP [ValueTracking] Remove dead code from an old experiment 2016-03-03 19:44:06 +00:00
SimplifyCFG [SimplifyCFG] Do not blindly remove unreachable blocks 2016-02-24 10:02:16 +00:00
Sink
SLPVectorizer Fix tests that used CHECK-NEXT-NOT and CHECK-DAG-NOT. 2016-02-26 19:40:34 +00:00
SpeculativeExecution
SROA [SROA] Also insert a bit piece expression if only one piece is needed 2016-01-14 20:06:34 +00:00
StraightLineStrengthReduce
StripDeadPrototypes
StripSymbols
StructurizeCFG AMDGPU: Remove some old intrinsic uses from tests 2016-02-11 06:02:01 +00:00
TailCallElim Push isDereferenceableAndAlignedPointer down into isSafeToLoadUnconditionally 2016-01-17 12:35:29 +00:00
TailDup
Util Make check line consistent 2016-02-07 20:57:46 +00:00
WholeProgramDevirt WholeProgramDevirt: introduce. 2016-02-09 22:50:34 +00:00