Dave Estes f657899174 [AArch64] Refines the Cortex-A57 Machine Model ...

Primarily refines all of the instructions with accurate latency
and micro-op information. Refinements largely focus on the NEON
instructions.

Additionally, a few advanced features are modeled, including
forwarding for MAC instructions and hazards for floating point SQRT
and DIV.

Lastly, the issue-width is reduced to three so that the scheduler
will better accommodate the narrower decode and dispatch width.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218627 91177308-0d34-0410-b5e6-96231b3b80d8

2014-09-29 21:27:36 +00:00

..

AsmParser

AArch64: allow constant expressions for shifted reg literals

2014-09-23 22:16:02 +00:00

Disassembler

Fix left shifts of negative integers in AArch64 InstPrinter/Disassembler

2014-09-02 16:19:41 +00:00

InstPrinter

Add override to overriden virtual methods, remove virtual keywords.

2014-09-03 11:41:21 +00:00

MCTargetDesc

Fix segfault in AArch64 backend with -g and -mbig-endian

2014-09-23 15:38:11 +00:00

TargetInfo

AArch64: stop trying to take control of all UnknownArch triples.

2014-08-08 08:27:44 +00:00

Utils

[AArch 64] Use a constant pool load for weak symbol references when

2014-09-10 13:54:38 +00:00

AArch64.h

[AArch64] Add experimental PBQP support

2014-09-10 14:06:10 +00:00

AArch64.td

…

AArch64A57FPLoadBalancing.cpp

[A57FPLoadBalancing] Modify r217689 - actually we do need to check defs

2014-09-14 18:24:26 +00:00

AArch64AddressTypePromotion.cpp

Fix typos in comments

2014-08-15 22:17:28 +00:00

AArch64AdvSIMDScalarPass.cpp

Remove the TargetMachine forwards for TargetSubtargetInfo based

2014-08-04 21:25:23 +00:00

AArch64AsmPrinter.cpp

Delete unused argument in AArch64MCInstLower constructor: it doesn't

2014-08-19 21:51:08 +00:00

AArch64BranchRelaxation.cpp

Testing commit access.

2014-08-14 16:20:50 +00:00

AArch64CallingConvention.td

Teach the AArch64 backend about v4f16 and v8f16

2014-08-27 16:16:04 +00:00

AArch64CleanupLocalDynamicTLSPass.cpp

Remove the TargetMachine forwards for TargetSubtargetInfo based

2014-08-04 21:25:23 +00:00

AArch64CollectLOH.cpp

AArch64: use std::fill instead of memset

2014-08-26 03:33:26 +00:00

AArch64ConditionalCompares.cpp

Change MCSchedModel to be a struct of statically initialized data.

2014-09-02 17:43:54 +00:00

AArch64ConditionOptimizer.cpp

[AArch64] Add pass to enable additional comparison optimizations by CSE.

2014-09-05 02:55:24 +00:00

AArch64DeadRegisterDefinitionsPass.cpp

Remove 'virtual' keyword from methods markedwith 'override' keyword.

2014-08-30 16:48:34 +00:00

AArch64ExpandPseudoInsts.cpp

Have MachineFunction cache a pointer to the subtarget to make lookups

2014-08-05 02:39:49 +00:00

AArch64FastISel.cpp

[FastISel][AArch64] Also allow folding of sign-/zero-extend and shift-left for booleans (i1).

2014-09-22 21:08:53 +00:00

AArch64FrameLowering.cpp

Fix typos:

2014-08-11 18:04:46 +00:00

AArch64FrameLowering.h

Canonicalize header guards into a common format.

2014-08-13 16:26:38 +00:00

AArch64InstrAtomics.td

Make use of isAtLeastRelease/Acquire in the ARM/AArch64 backends

2014-08-18 16:48:58 +00:00

AArch64InstrFormats.td

Teach the AArch64 backend about v4f16 and v8f16

2014-08-27 16:16:04 +00:00

AArch64InstrInfo.cpp

[AArch64] Improve AA to remove unneeded edges in the AA MI scheduling graph.

2014-09-08 14:43:48 +00:00

AArch64InstrInfo.h

[AArch64] Improve AA to remove unneeded edges in the AA MI scheduling graph.

2014-09-08 14:43:48 +00:00

AArch64InstrInfo.td

AArch64: fix vector-immediate BIC/ORR on big-endian devices.

2014-09-04 15:05:24 +00:00

AArch64ISelDAGToDAG.cpp

Merge Extend and Shift into a UBFX

2014-09-02 09:33:56 +00:00

AArch64ISelLowering.cpp

[X86] Use the generic AtomicExpandPass instead of X86AtomicExpandPass

2014-09-17 00:06:58 +00:00

AArch64ISelLowering.h

[X86] Use the generic AtomicExpandPass instead of X86AtomicExpandPass

2014-09-17 00:06:58 +00:00

AArch64LoadStoreOptimizer.cpp

Add missing closing namespace comment.

2014-08-11 22:42:31 +00:00

AArch64MachineCombinerPattern.h

Canonicalize header guards into a common format.

2014-08-13 16:26:38 +00:00

AArch64MachineFunctionInfo.h

Canonicalize header guards into a common format.

2014-08-13 16:26:38 +00:00

AArch64MCInstLower.cpp

Delete unused argument in AArch64MCInstLower constructor: it doesn't

2014-08-19 21:51:08 +00:00

AArch64MCInstLower.h

Delete unused argument in AArch64MCInstLower constructor: it doesn't

2014-08-19 21:51:08 +00:00

AArch64PBQPRegAlloc.cpp

Fix gcc -Wpedantic.

2014-09-12 12:32:08 +00:00

AArch64PerfectShuffle.h

Canonicalize header guards into a common format.

2014-08-13 16:26:38 +00:00

AArch64PromoteConstant.cpp

Run sort_includes.py on the AArch64 backend.

2014-07-25 11:42:14 +00:00

AArch64RegisterInfo.cpp

Have MachineFunction cache a pointer to the subtarget to make lookups

2014-08-05 02:39:49 +00:00

AArch64RegisterInfo.h

Canonicalize header guards into a common format.

2014-08-13 16:26:38 +00:00

AArch64RegisterInfo.td

Teach the AArch64 backend about v4f16 and v8f16

2014-08-27 16:16:04 +00:00

AArch64SchedA53.td

…

AArch64SchedA57.td

[AArch64] Refines the Cortex-A57 Machine Model

2014-09-29 21:27:36 +00:00

AArch64SchedA57WriteRes.td

[AArch64] Refines the Cortex-A57 Machine Model

2014-09-29 21:27:36 +00:00

AArch64SchedCyclone.td

…

AArch64Schedule.td

…

AArch64SelectionDAGInfo.cpp

Remove the TargetMachine forwards for TargetSubtargetInfo based

2014-08-04 21:25:23 +00:00

AArch64SelectionDAGInfo.h

Canonicalize header guards into a common format.

2014-08-13 16:26:38 +00:00

AArch64StorePairSuppress.cpp

Change MCSchedModel to be a struct of statically initialized data.

2014-09-02 17:43:54 +00:00

AArch64Subtarget.cpp

[AArch 64] Use a constant pool load for weak symbol references when

2014-09-10 13:54:38 +00:00

AArch64Subtarget.h

[AArch64] Enable post-RA MI scheduler.

2014-09-12 17:40:39 +00:00

AArch64TargetMachine.cpp

[AArch64] Don't enable the post-RA MI scheduler at OptNone.

2014-09-12 22:17:28 +00:00

AArch64TargetMachine.h

Reverting NFC changes from r218050. Instead, the warning was disabled for GCC in r218059, so these changes are no longer required.

2014-09-18 17:34:23 +00:00

AArch64TargetObjectFile.cpp

…

AArch64TargetObjectFile.h

Canonicalize header guards into a common format.

2014-08-13 16:26:38 +00:00

AArch64TargetTransformInfo.cpp

[AArch64] Improve cost model to handle sdiv by a pow-of-two.

2014-09-29 13:59:31 +00:00

CMakeLists.txt

[AArch64] Add experimental PBQP support

2014-09-10 14:06:10 +00:00

LLVMBuild.txt

Prune redundant libdeps.

2014-07-24 11:45:27 +00:00

Makefile

…