Sjoerd Meijer d986ede313 [ARM] Do not fuse VADD and VMUL on the Cortex-M4 and Cortex-M33 ...

A sequence of VMUL and VADD instructions always give the same or better
performance than a fused VMLA instruction on the Cortex-M4 and Cortex-M33.
Executing the VMUL and VADD back-to-back requires the same cycles, but
having separate instructions allows scheduling to avoid the hazard between
these 2 instructions.

Differential Revision: https://reviews.llvm.org/D52289

llvm-svn: 342874

2018-09-24 12:02:50 +00:00

..

2009-07-17-CrossRegClassCopy.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

2009-07-21-ISelBug.ll

[ARM] Generate consistent frame records for Thumb2

2016-08-23 09:19:22 +00:00

2009-07-23-CPIslandBug.ll

…

2009-07-30-PEICrash.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

2009-08-01-WrongLDRBOpc.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

2009-08-02-CoalescerBug.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

2009-08-04-CoalescerAssert.ll

[opaque pointer type] Add textual IR support for explicit type parameter to the call instruction

2015-04-16 23:24:18 +00:00

2009-08-04-CoalescerBug.ll

[opaque pointer type] Add textual IR support for explicit type parameter to the call instruction

2015-04-16 23:24:18 +00:00

2009-08-04-ScavengerAssert.ll

[opaque pointer type] Add textual IR support for explicit type parameter to the call instruction

2015-04-16 23:24:18 +00:00

2009-08-04-SubregLoweringBug2.ll

…

2009-08-04-SubregLoweringBug3.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

2009-08-04-SubregLoweringBug.ll

Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)

2018-01-19 17:13:12 +00:00

2009-08-06-SpDecBug.ll

ARM: When spilling extra registers for alignment, prefer low registers on all Thumb targets.

2015-04-23 20:31:26 +00:00

2009-08-07-CoalescerBug.ll

…

2009-08-07-NeonFPBug.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

2009-08-08-ScavengerAssert.ll

…

2009-08-10-ISelBug.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

2009-08-21-PostRAKill4.ll

[opaque pointer type] Add textual IR support for explicit type parameter to the call instruction

2015-04-16 23:24:18 +00:00

2009-09-01-PostRAProlog.ll

Fix an old memset signature in 2009-09-01-PostRAProlog.ll test causing a buildbot failure

2016-06-23 16:07:10 +00:00

2009-10-15-ITBlockBranch.ll

…

2009-11-01-CopyReg2RegBug.ll

…

2009-11-11-ScavengerAssert.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

2009-11-13-STRDBug.ll

…

2009-12-01-LoopIVUsers.ll

[SCEV] Try to reuse existing value during SCEV expansion

2016-02-04 01:27:38 +00:00

2010-01-06-TailDuplicateLabels.ll

Work around grep vs. CRLF issue in Thumb2 test by matching excess whitespace

2018-09-18 00:04:29 +00:00

2010-01-19-RemovePredicates.ll

[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction

2015-02-27 19:29:02 +00:00

2010-02-11-phi-cycle.ll

ARM: stop emitting blx instructions for most calls on MachO.

2016-05-10 19:17:47 +00:00

2010-02-24-BigStack.ll

…

2010-03-08-addi12-ccout.ll

[opaque pointer type] Add textual IR support for explicit type parameter to gep operator

2015-03-13 18:20:45 +00:00

2010-03-15-AsmCCClobber.ll

ARM: convert ORR instructions to ADD where possible on Thumb.

2018-06-20 12:09:44 +00:00

2010-04-15-DynAllocBug.ll

…

2010-04-26-CopyRegCrash.ll

…

2010-05-24-rsbs.ll

…

2010-06-14-NEONCoalescer.ll

[CodeGen] Don't print "pred:" and "opt:" in -debug output

2018-01-09 17:31:07 +00:00

2010-06-19-ITBlockCrash.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

2010-06-21-TailMergeBug.ll

[opaque pointer type] Add textual IR support for explicit type parameter to gep operator

2015-03-13 18:20:45 +00:00

2010-08-10-VarSizedAllocaBug.ll

[opaque pointer type] Add textual IR support for explicit type parameter to the call instruction

2015-04-16 23:24:18 +00:00

2010-11-22-EpilogueBug.ll

Generalize MergeBlockIntoPredecessor. Replace uses of MergeBasicBlockIntoOnlyPred.

2018-06-20 22:01:04 +00:00

2010-12-03-AddSPNarrowing.ll

…

2011-04-21-FILoweringBug.ll

…

2011-06-07-TwoAddrEarlyClobber.ll

Thumb2: When optimizing for size, do not if-convert branches involving comparisons with zero.

2015-04-23 20:31:30 +00:00

2011-12-16-T2SizeReduceAssert.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

2012-01-13-CBNZBug.ll

Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)

2018-01-19 17:13:12 +00:00

2013-02-19-tail-call-register-hint.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

2013-03-02-vduplane-nonconstant-source-index.ll

Fix a nasty bug in DAGCombine of STORE nodes.

2015-03-19 22:48:57 +00:00

2013-03-06-vector-sext-operand-scalarize.ll

…

aapcs.ll

[ARM] Return true in enableMultipleCopyHints().

2018-02-16 09:51:01 +00:00

aligned-constants.ll

[MC] Use .p2align instead of .align

2016-01-26 00:03:25 +00:00

aligned-spill.ll

[ARM] Generate consistent frame records for Thumb2

2016-08-23 09:19:22 +00:00

bfi.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

bfx.ll

…

bicbfi.ll

[CodeGen] Always use printReg to print registers in both MIR and debug

2017-11-30 16:12:24 +00:00

buildvector-crash.ll

…

carry.ll

Address buildbot fallout from r259065

2016-01-28 18:59:04 +00:00

cbnz.ll

CodeGen: Allow small copyable blocks to "break" the CFG.

2017-01-31 23:48:32 +00:00

cmp-frame.ll

ARM: Don't rewrite add reg, $sp, 0 -> mov reg, $sp if the add defines CPSR.

2018-02-27 19:00:59 +00:00

constant-islands-jump-table.ll

ARM: recommit r237590: allow jump tables to be placed as constant islands.

2015-05-31 19:22:07 +00:00

constant-islands-new-island-padding.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

constant-islands-new-island.ll

[ARM] Make -mcpu=generic schedule for an in-order core (Cortex-A8).

2017-06-28 07:07:03 +00:00

constant-islands.ll

ARM: Do not use llc -march in tests.

2017-08-01 22:20:49 +00:00

cortex-fp.ll

ARM: Do not use llc -march in tests.

2017-08-01 22:20:49 +00:00

crash.ll

[ARM][NEON] Use address space in vld([1234]|[234]lane) and vst([1234]|[234]lane) instructions

2015-09-30 10:56:37 +00:00

cross-rc-coalescing-1.ll

…

cross-rc-coalescing-2.ll

[Thumb2] fix typo in test from r332548

2018-05-17 03:24:25 +00:00

div.ll

[ARM] support for Cortex-R4/R4F

2015-04-09 14:07:28 +00:00

emit-unwinding.ll

ARM: use r7 as the frame-pointer on all MachO targets.

2016-04-11 22:27:40 +00:00

float-cmp.ll

[ARM] Materialise some boolean values to avoid a branch

2018-02-16 09:23:59 +00:00

float-intrinsics-double.ll

ARM: match GCC's behaviour for builtins

2017-01-13 16:25:33 +00:00

float-intrinsics-float.ll

[ARM] Do not fuse VADD and VMUL on the Cortex-M4 and Cortex-M33

2018-09-24 12:02:50 +00:00

float-ops.ll

[Thumb] Teach ISel how to lower compares of AND bitmasks efficiently

2016-12-15 09:38:59 +00:00

frame-pointer.ll

Re-land "[Thumb] Save/restore high registers in Thumb1 pro/epilogues"

2016-10-11 21:14:03 +00:00

frameless2.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

frameless.ll

…

ifcvt-compare.ll

CodeGen: Allow small copyable blocks to "break" the CFG.

2017-01-31 23:48:32 +00:00

ifcvt-neon-deprecated.mir

Followup on Proposal to move MIR physical register namespace to '$' sigil.

2018-01-31 22:04:26 +00:00

ifcvt-no-branch-predictor.ll

[CodeGen] Add a new pass for PostRA sink

2018-03-22 20:06:47 +00:00

ifcvt-rescan-bug-2016-08-22.ll

[CodeGen] Unify MBB reference format in both MIR and debug output

2017-12-04 17:18:51 +00:00

ifcvt-rescan-diamonds.ll

IfConversion: Fix bug introduced by rescanning diamonds.

2016-09-02 18:29:26 +00:00

inflate-regs.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

inlineasm.ll

…

intrinsics-cc.ll

[ARM] Honor -mfloat-abi for libcall calling convention

2017-10-26 21:42:32 +00:00

intrinsics-coprocessor.ll

ARM: Do not use llc -march in tests.

2017-08-01 22:20:49 +00:00

large-call.ll

[opaque pointer type] Add textual IR support for explicit type parameter to the call instruction

2015-04-16 23:24:18 +00:00

large-stack.ll

ARM: Do not use llc -march in tests.

2017-08-01 22:20:49 +00:00

ldr-str-imm12.ll

Elide stores which are overwritten without being observed.

2017-05-16 19:43:56 +00:00

lit.local.cfg

…

longMACt.ll

…

lsr-deficiency.ll

[Thumb] Select (CMPZ X, -C) -> (CMPZ (ADDS X, C), 0)

2016-09-09 12:52:24 +00:00

machine-licm.ll

Make the canonicalisation on shifts benifit to more case.

2016-12-23 02:56:07 +00:00

mul_const.ll

…

pic-load.ll

ARM: Add scheduling information for LDRLIT instructions to swift scheduling model

2015-07-17 23:18:26 +00:00

segmented-stacks.ll

ARM: Do not use llc -march in tests.

2017-08-01 22:20:49 +00:00

setjmp_longjmp.ll

Arm: Don't define a label twice with two setjmps in a function.

2015-07-16 22:34:20 +00:00

stack_guard_remat.ll

Add address space mangling to lifetime intrinsics

2017-04-10 20:18:21 +00:00

t2sizereduction.mir

Followup on Proposal to move MIR physical register namespace to '$' sigil.

2018-01-31 22:04:26 +00:00

tail-call-r9.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

tbb-removeadd.mir

Followup on Proposal to move MIR physical register namespace to '$' sigil.

2018-01-31 22:04:26 +00:00

thumb2-adc.ll

…

thumb2-add2.ll

…

thumb2-add3.ll

…

thumb2-add4.ll

…

thumb2-add5.ll

…

thumb2-add6.ll

…

thumb2-add.ll

…

thumb2-and2.ll

…

thumb2-and.ll

…

thumb2-asr2.ll

…

thumb2-asr.ll

…

thumb2-bcc.ll

…

thumb2-bfc.ll

…

thumb2-bic.ll

…

thumb2-branch.ll

…

thumb2-call-tc.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

thumb2-call.ll

ARM: stop emitting blx instructions for most calls on MachO.

2016-05-10 19:17:47 +00:00

thumb2-cbnz.ll

Codegen: Fix broken assumption in Tail Merge.

2016-06-24 18:16:36 +00:00

thumb2-clz.ll

…

thumb2-cmn2.ll

[ARM] Materialise some boolean values to avoid a branch

2018-02-16 09:23:59 +00:00

thumb2-cmn.ll

[ARM] Materialise some boolean values to avoid a branch

2018-02-16 09:23:59 +00:00

thumb2-cmp.ll

[ARM] Treat cmn immediates as legal in isLegalICmpImmediate.

2018-07-10 23:44:37 +00:00

thumb2-cpsr-liveness.ll

Fix PR26655: Bail out if all regs of an inst BUNDLE have the correct kill flag

2016-05-10 17:57:27 +00:00

thumb2-eor2.ll

…

thumb2-eor.ll

…

thumb2-ifcvt1-tc.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

thumb2-ifcvt1.ll

CodeGen: If Convert blocks that would form a diamond when tail-merged.

2016-08-24 21:34:27 +00:00

thumb2-ifcvt2.ll

MachO: trap unreachable instructions

2018-04-13 22:25:20 +00:00

thumb2-ifcvt3.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

thumb2-jtb.ll

Generalize MergeBlockIntoPredecessor. Replace uses of MergeBasicBlockIntoOnlyPred.

2018-06-20 22:01:04 +00:00

thumb2-ldm.ll

[ARM] Generate consistent frame records for Thumb2

2016-08-23 09:19:22 +00:00

thumb2-ldr_ext.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

thumb2-ldr_post.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

thumb2-ldr_pre.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

thumb2-ldr.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

thumb2-ldrb.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

thumb2-ldrd.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

thumb2-ldrh.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

thumb2-lsl2.ll

…

thumb2-lsl.ll

…

thumb2-lsr2.ll

…

thumb2-lsr3.ll

…

thumb2-lsr.ll

…

thumb2-mla.ll

…

thumb2-mls.ll

…

thumb2-mov.ll

…

thumb2-mul.ll

…

thumb2-mulhi.ll

[ARM] Renaming +t2dsp feature into +dsp, as discussed on llvm-dev

2015-10-23 17:19:19 +00:00

thumb2-mvn2.ll

…

thumb2-mvn.ll

…

thumb2-neg.ll

…

thumb2-orn2.ll

…

thumb2-orn.ll

…

thumb2-orr2.ll

…

thumb2-orr.ll

…

thumb2-pack.ll

[ARM] Remove t2xtpk feature from tests

2017-03-09 15:14:32 +00:00

thumb2-rev16.ll

ARM: Do not use llc -march in tests.

2017-08-01 22:20:49 +00:00

thumb2-rev.ll

[ARM] Remove t2xtpk feature from tests

2017-03-09 15:14:32 +00:00

thumb2-ror.ll

…

thumb2-rsb2.ll

…

thumb2-rsb.ll

…

thumb2-sbc.ll

…

thumb2-select_xform.ll

[ARM] Return true in enableMultipleCopyHints().

2018-02-16 09:51:01 +00:00

thumb2-select.ll

…

thumb2-shifter.ll

…

thumb2-smla.ll

[ARM] Remove t2xtpk feature from tests

2017-03-09 15:14:32 +00:00

thumb2-smul.ll

[ARM] Remove t2xtpk feature from tests

2017-03-09 15:14:32 +00:00

thumb2-spill-q.ll

[Thumb] preserve test intent by removing undef

2018-05-16 22:47:42 +00:00

thumb2-str_post.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

thumb2-str_pre.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

thumb2-str.ll

[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction

2015-02-27 19:29:02 +00:00

thumb2-strb.ll

[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction

2015-02-27 19:29:02 +00:00

thumb2-strh.ll

[opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction

2015-02-27 19:29:02 +00:00

thumb2-sub2.ll

…

thumb2-sub3.ll

…

thumb2-sub4.ll

…

thumb2-sub5.ll

…

thumb2-sub.ll

…

thumb2-sxt_rot.ll

[ARM] Replace HasT2ExtractPack with HasDSP

2017-02-17 15:42:44 +00:00

thumb2-sxt-uxt.ll

[ARM] Replace HasT2ExtractPack with HasDSP

2017-02-17 15:42:44 +00:00

thumb2-tbb.ll

[Thumb-1] Synthesize TBB/TBH instructions to make use of compressed jump tables

2016-11-01 13:37:41 +00:00

thumb2-tbh.ll

[Thumb-1] Synthesize TBB/TBH instructions to make use of compressed jump tables

2016-11-01 13:37:41 +00:00

thumb2-teq2.ll

[ARM] Materialise some boolean values to avoid a branch

2018-02-16 09:23:59 +00:00

thumb2-teq.ll

[ARM] Materialise some boolean values to avoid a branch

2018-02-16 09:23:59 +00:00

thumb2-tst2.ll

[ARM] Materialise some boolean values to avoid a branch

2018-02-16 09:23:59 +00:00

thumb2-tst.ll

[ARM] Materialise some boolean values to avoid a branch

2018-02-16 09:23:59 +00:00

thumb2-uxt_rot.ll

[ARM] Replace HasT2ExtractPack with HasDSP

2017-02-17 15:42:44 +00:00

thumb2-uxtb.ll

ARM: convert ORR instructions to ADD where possible on Thumb.

2018-06-20 12:09:44 +00:00

tls1.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

tls2.ll

Don't print (PLT) on arm.

2016-06-16 16:09:53 +00:00

tpsoft.ll

ARM: When spilling extra registers for alignment, prefer low registers on all Thumb targets.

2015-04-23 20:31:26 +00:00

umulo-64-legalisation-lowering.ll

[SelectionDAG] Improve the legalisation lowering of UMULO.

2018-08-16 18:39:39 +00:00

umulo-128-legalisation-lowering.ll

[SelectionDAG] Improve the legalisation lowering of UMULO.

2018-08-16 18:39:39 +00:00

v8_IT_1.ll

[ARM][NEON] Use address space in vld([1234]|[234]lane) and vst([1234]|[234]lane) instructions

2015-09-30 10:56:37 +00:00

v8_IT_2.ll

[opaque pointer type] Add textual IR support for explicit type parameter to load instruction

2015-02-27 21:17:42 +00:00

v8_IT_3.ll

[arm] Fix Unnecessary reloads from GOT.

2017-11-13 20:45:38 +00:00

v8_IT_4.ll

CodeGen: Allow small copyable blocks to "break" the CFG.

2017-01-31 23:48:32 +00:00

v8_IT_5.ll

[Dominators] Include infinite loops in PostDominatorTree

2017-08-15 18:14:57 +00:00

v8_IT_6.ll

…