Jose M Monsalve Diaz 5b7208da36 [OpenMP] Folding threadLimit and numThreads when single value in kernels ... The device runtime contains several calls to `__kmpc_get_hardware_num_threads_in_block` and `__kmpc_get_hardware_num_blocks`. If the thread_limit and the num_teams are constant, these calls can be folded to the constant value. In this patch we use the already introduced `AAFoldRuntimeCall` and the `NumTeams` and `NumThreads` kernel attributes (to be introduced in a different patch) to fold these functions. The code checks all the kernels, and if their attributes match, the functions are folded. In the future we will explore specializing for multiple values of NumThreads and NumTeams. Depends on D106390 Reviewed By: jdoerfert, JonChesterfield Differential Revision: https://reviews.llvm.org/D106033		2021-07-27 21:47:12 -04:00
..
AlwaysInliner.cpp
Annotation2Metadata.cpp
ArgumentPromotion.cpp	[OpaquePtr] Use byval type more	2021-07-13 09:34:34 -07:00
Attributor.cpp	Reapply "[Attributor] Disable simplification AAs if a callback is present""	2021-07-27 19:14:50 -05:00
AttributorAttributes.cpp	Reapply "[Attributor] Disable simplification AAs if a callback is present""	2021-07-27 19:14:50 -05:00
BarrierNoopPass.cpp
BlockExtractor.cpp	Internalize some cl::opt global variables or move them under namespace llvm	2021-05-07 11:15:43 -07:00
CalledValuePropagation.cpp
CMakeLists.txt	Move Function Specialization to its correct location. NFC.	2021-06-11 15:00:10 +01:00
ConstantMerge.cpp	[ConstantMerge] Don't merge thread_local constants with non-thread_local constants	2021-04-28 23:44:20 +01:00
CrossDSOCFI.cpp
DeadArgumentElimination.cpp	[Attributes] Clean up handling of UB implying attributes (NFC)	2021-07-25 18:21:13 +02:00
ElimAvailExtern.cpp
ExtractGV.cpp	[Transforms] Use range-based for loops (NFC)	2021-02-08 22:33:53 -08:00
ForceFunctionAttrs.cpp	[Attributes] Determine attribute properties from TableGen data	2021-07-12 22:13:38 +02:00
FunctionAttrs.cpp	Revert "[NewPM] Only invalidate modified functions' analyses in CGSCC passes"	2021-05-21 16:38:03 -07:00
FunctionImport.cpp	[HIP] support ThinLTO	2021-05-22 10:48:34 -04:00
FunctionSpecialization.cpp	[FuncSpec] Add an option to specializing literal constant	2021-06-30 11:26:44 +08:00
GlobalDCE.cpp
GlobalOpt.cpp	This patch extends the OptimizeGlobalAddressOfMalloc to handle the null check of global pointer variables. It is disabled with https://reviews.llvm.org/rGb7cd291c1542aee12c9e9fde6c411314a163a8ea. This PR is to reenable it while fixing the original problem reported. The fix is to set the store value correctly when creating store for the new created global init bool symbol.	2021-07-20 12:27:26 -04:00
GlobalSplit.cpp
HotColdSplitting.cpp	Calculate getTerminator only when necessary	2021-06-13 20:16:07 -07:00
InferFunctionAttrs.cpp	[inferattrs] Don't infer lib func attributes for nobuiltin functions	2021-04-16 15:36:15 -07:00
Inliner.cpp	[NewPM][Inliner] Check if deleted function is in current SCC	2021-07-21 08:47:45 -07:00
InlineSimple.cpp
Internalize.cpp	[IR] Rename comdat noduplicates to comdat nodeduplicate	2021-07-20 12:47:10 -07:00
IPO.cpp	Function Specialization Pass	2021-06-11 09:11:29 +01:00
IROutliner.cpp	[IROutliner] Adding DebugInfo handling for IR Outlined Functions	2021-06-15 10:57:08 -05:00
LoopExtractor.cpp
LowerTypeTests.cpp	[LTT] Handle merged llvm.assume when dropping type tests	2021-05-25 17:02:13 -07:00
MergeFunctions.cpp	[OpaquePtr] Use GlobalValue::getValueType() more	2021-07-13 09:34:34 -07:00
OpenMPOpt.cpp	[OpenMP] Folding threadLimit and numThreads when single value in kernels	2021-07-27 21:47:12 -04:00
PartialInlining.cpp	NFC: Migrate PartialInlining to work on InstructionCost	2021-03-30 11:59:45 +01:00
PassManagerBuilder.cpp	Add jump-threading optimization for deterministic finite automata	2021-07-27 14:34:04 -04:00
PruneEH.cpp	[Local] Do not introduce a new llvm.trap before unreachable	2021-07-26 23:33:36 -05:00
SampleContextTracker.cpp	[CSSPGO] Fix a typo in SampleContextTracker	2021-07-22 16:44:50 -07:00
SampleProfile.cpp	[CSSPGO] Tweak ICP threshold in top-down inliner	2021-07-26 21:49:20 -07:00
SampleProfileProbe.cpp	[CSSPGO] Undoing the concept of dangling pseudo probe	2021-06-18 15:14:11 -07:00
SCCP.cpp	Function Specialization Pass	2021-06-11 09:11:29 +01:00
StripDeadPrototypes.cpp
StripSymbols.cpp	[Transforms] Use range-based for loops (NFC)	2021-02-08 22:33:53 -08:00
SyntheticCountsPropagation.cpp	Internalize some cl::opt global variables or move them under namespace llvm	2021-05-07 11:15:43 -07:00
ThinLTOBitcodeWriter.cpp	Revert "ThinLTO: Fix inline assembly references to static functions with CFI"	2021-07-20 13:59:46 -07:00
WholeProgramDevirt.cpp	[WPD] Don't optimize calls more than once	2021-06-24 13:28:09 -07:00