6502/llvm-6502
1
0
Fork 0
mirror of https://github.com/c64scene-ar/llvm-6502.git synced 2025-02-23 05:29:23 +00:00
Code Issues Projects Releases Wiki Activity
llvm-6502/test/CodeGen/X86
History
Chandler Carruth f103b3d1b9 Teach the X86 instruction selection to do some heroic transforms to 
detect a pattern which can be implemented with a small 'shl' embedded in
the addressing mode scale. This happens in real code as follows:

  unsigned x = my_accelerator_table[input >> 11];

Here we have some lookup table that we look into using the high bits of
'input'. Each entity in the table is 4-bytes, which means this
implicitly gets turned into (once lowered out of a GEP):

  *(unsigned*)((char*)my_accelerator_table + ((input >> 11) << 2));

The shift right followed by a shift left is canonicalized to a smaller
shift right and masking off the low bits. That hides the shift right
which x86 has an addressing mode designed to support. We now detect
masks of this form, and produce the longer shift right followed by the
proper addressing mode. In addition to saving a (rather large)
instruction, this also reduces stalls in Intel chips on benchmarks I've
measured.

In order for all of this to work, one part of the DAG needs to be
canonicalized *still further* than it currently is. This involves
removing pointless 'trunc' nodes between a zextload and a zext. Without
that, we end up generating spurious masks and hiding the pattern.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147936 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-11 08:41:08 +00:00
..
GC
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
3addr-16bit.ll
…
3addr-or.ll
…
3dnow-intrinsics.ll
Add 3DNow! intrinsics.
2011-04-15 00:32:41 +00:00
4char-promote.ll
This test was failing on X86 machines which do not have SSE4. Fixed the test by
2011-06-15 12:26:53 +00:00
2003-08-03-CallArgLiveRanges.ll
…
2003-08-23-DeadBlockTest.ll
…
2003-11-03-GlobalBool.ll
…
2004-02-13-FrameReturnAddress.ll
…
2004-02-14-InefficientStackPointer.ll
…
2004-02-22-Casts.ll
…
2004-03-30-Select-Max.ll
…
2004-04-13-FPCMOV-Crash.ll
…
2004-06-10-StackifierCrash.ll
…
2004-10-08-SelectSetCCFold.ll
…
2005-01-17-CycleInDAG.ll
…
2005-02-14-IllegalAssembler.ll
…
2005-05-08-FPStackifierPHI.ll
…
2006-01-19-ISelFoldingBug.ll
…
2006-03-01-InstrSchedBug.ll
…
2006-03-02-InstrSchedBug.ll
…
2006-04-04-CrossBlockCrash.ll
…
2006-04-27-ISelFoldingBug.ll
…
2006-05-01-SchedCausingSpills.ll
Change the x86 32-bit scheduler to register pressure and fix up the
2011-03-11 01:05:58 +00:00
2006-05-02-InstrSched1.ll
…
2006-05-02-InstrSched2.ll
…
2006-05-08-CoalesceSubRegClass.ll
…
2006-05-08-InstrSched.ll
…
2006-05-11-InstrSched.ll
Use a bigger hammer to fix PR11314 by disabling the "forcing two-address
2011-11-10 07:43:16 +00:00
2006-05-17-VectorArg.ll
…
2006-05-22-FPSetEQ.ll
A forthcoming SSE patch will break this test; since the test is also
2011-06-01 16:13:09 +00:00
2006-05-25-CycleInDAG.ll
…
2006-07-10-InlineAsmAConstraint.ll
…
2006-07-12-InlineAsmQConstraint.ll
…
2006-07-20-InlineAsm.ll
…
2006-07-28-AsmPrint-Long-As-Pointer.ll
…
2006-07-31-SingleRegClass.ll
…
2006-08-07-CycleInDAG.ll
…
2006-08-16-CycleInDAG.ll
…
2006-08-21-ExtraMovInst.ll
…
2006-09-01-CycleInDAG.ll
…
2006-10-02-BoolRetCrash.ll
…
2006-10-07-ScalarSSEMiscompile.ll
…
2006-10-09-CycleInDAG.ll
…
2006-10-10-FindModifiedNodeSlotBug.ll
…
2006-10-12-CycleInDAG.ll
…
2006-10-13-CycleInDAG.ll
…
2006-10-19-SwitchUnnecessaryBranching.ll
…
2006-11-12-CSRetCC.ll
FileCheck-ize
2011-07-02 20:43:08 +00:00
2006-11-17-IllegalMove.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
2006-11-27-SelectLegalize.ll
…
2006-12-16-InlineAsmCrash.ll
…
2006-12-19-IntelSyntax.ll
…
2007-01-08-InstrSched.ll
…
2007-01-08-X86-64-Pointer.ll
test/CodeGen/X86: FileCheck-ize and add actions for x86_64-linux and x86_64-win32.
2011-03-16 13:53:07 +00:00
2007-01-13-StackPtrIndex.ll
…
2007-01-29-InlineAsm-ir.ll
…
2007-02-04-OrAddrMode.ll
FileCheck-ize and tighten up assertions to only check the relevant sections.
2011-07-02 20:43:04 +00:00
2007-02-16-BranchFold.ll
…
2007-02-19-LiveIntervalAssert.ll
…
2007-02-23-DAGCombine-Miscompile.ll
FileCheck-ize and cleanup IR.
2011-07-02 20:43:01 +00:00
2007-02-25-FastCCStack.ll
…
2007-03-01-SpillerCrash.ll
…
2007-03-15-GEP-Idx-Sink.ll
…
2007-03-16-InlineAsm.ll
Remove support for using "foo" as symbols instead of %"foo". This is ancient
2011-06-17 06:36:20 +00:00
2007-03-18-LiveIntervalAssert.ll
…
2007-03-24-InlineAsmMultiRegConstraint.ll
…
2007-03-24-InlineAsmPModifier.ll
…
2007-03-24-InlineAsmVectorOp.ll
…
2007-03-24-InlineAsmXConstraint.ll
FileCheck-ize
2011-07-02 20:42:59 +00:00
2007-03-26-CoalescerBug.ll
…
2007-04-08-InlineAsmCrash.ll
…
2007-04-11-InlineAsmVectorResult.ll
…
2007-04-17-LiveIntervalAssert.ll
…
2007-04-24-Huge-Stack.ll
…
2007-04-24-VectorCrash.ll
…
2007-04-25-MMX-PADDQ.ll
Allow folded spills in test.
2011-04-30 08:00:50 +00:00
2007-04-27-InlineAsm-IntMemInput.ll
…
2007-05-05-Personality.ll
Convert to the new EH model.
2011-11-08 00:17:28 +00:00
2007-05-05-VecCastExpand.ll
…
2007-05-07-InvokeSRet.ll
Update more tests to the new EH scheme.
2011-08-31 21:39:05 +00:00
2007-05-14-LiveIntervalAssert.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
2007-05-15-maskmovq.ll
…
2007-05-17-ShuffleISelBug.ll
…
2007-06-04-X86-64-CtorAsmBugs.ll
…
2007-06-15-IntToMMX.ll
…
2007-06-28-X86-64-isel.ll
…
2007-06-29-DAGCombinerBug.ll
…
2007-06-29-VecFPConstantCSEBug.ll
…
2007-07-03-GR64ToVR64.ll
Fix register-dependent X86 tests.
2011-04-05 00:32:44 +00:00
2007-07-10-StackerAssert.ll
…
2007-07-18-Vector-Extract.ll
test/CodeGen/X86: FileCheck-ize and add actions for x86_64-linux and x86_64-win32.
2011-03-16 13:53:07 +00:00
2007-08-01-LiveVariablesBug.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
2007-08-09-IllegalX86-64Asm.ll
…
2007-08-10-SignExtSubreg.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
2007-08-13-AppendingLinkage.ll
…
2007-09-05-InvalidAsm.ll
…
2007-09-06-ExtWeakAliasee.ll
…
2007-09-17-ObjcFrameEH.ll
Remove a grep that is already checked with FileCheck.
2011-07-02 20:42:56 +00:00
2007-09-18-ShuffleXformBug.ll
…
2007-09-27-LDIntrinsics.ll
Heuristic: If the number of operands in the alias are more than the number of
2011-06-14 03:17:20 +00:00
2007-10-04-AvoidEFLAGSCopy.ll
…
2007-10-12-CoalesceExtSubReg.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
2007-10-12-SpillerUnfold1.ll
…
2007-10-12-SpillerUnfold2.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
2007-10-14-CoalescerCrash.ll
…
2007-10-15-CoalescerCrash.ll
Stop accepting and ignoring attributes in function types. Attributes are applied
2011-06-17 17:37:13 +00:00
2007-10-16-CoalescerCrash.ll
…
2007-10-17-IllegalAsm.ll
…
2007-10-19-SpillerUnfold.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
2007-10-28-inlineasm-q-modifier.ll
…
2007-10-29-ExtendSetCC.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
2007-10-30-LSRCrash.ll
…
2007-10-31-extractelement-i64.ll
…
2007-11-01-ISelCrash.ll
…
2007-11-03-x86-64-q-constraint.ll
…
2007-11-04-LiveIntervalCrash.ll
…
2007-11-04-LiveVariablesBug.ll
…
2007-11-04-rip-immediate-constant.ll
…
2007-11-06-InstrSched.ll
…
2007-11-07-MulBy4.ll
…
2007-11-30-LoadFolding-Bug.ll
…
2007-12-16-BURRSchedCrash.ll
…
2007-12-18-LoadCSEBug.ll
…
2008-01-08-IllegalCMP.ll
…
2008-01-08-SchedulerCrash.ll
This test depends on cmov being available.
2011-09-01 18:40:01 +00:00
2008-01-09-LongDoubleSin.ll
…
2008-01-16-FPStackifierAssert.ll
…
2008-01-16-InvalidDAGCombineXform.ll
…
2008-02-05-ISelCrash.ll
…
2008-02-06-LoadFoldingBug.ll
…
2008-02-08-LoadFoldingBug.ll
…
2008-02-14-BitMiscompile.ll
…
2008-02-18-TailMergingBug.ll
…
2008-02-20-InlineAsmClobber.ll
…
2008-02-22-LocalRegAllocBug.ll
…
2008-02-25-InlineAsmBug.ll
…
2008-02-25-X86-64-CoalescerBug.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
2008-02-26-AsmDirectMemOp.ll
…
2008-02-27-DeadSlotElimBug.ll
…
2008-02-27-PEICrash.ll
…
2008-03-06-frem-fpstack.ll
…
2008-03-07-APIntBug.ll
…
2008-03-10-RegAllocInfLoop.ll
…
2008-03-12-ThreadLocalAlias.ll
…
2008-03-13-TwoAddrPassCrash.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
2008-03-14-SpillerCrash.ll
…
2008-03-19-DAGCombinerBug.ll
…
2008-03-23-DarwinAsmComments.ll
…
2008-03-25-TwoAddrPassBug.ll
…
2008-03-31-SpillerFoldingBug.ll
…
2008-04-02-unnamedEH.ll
Add r130623 back now that ELF has been fixed to work with -fno-dwarf2-cfi-asm.
2011-05-01 15:44:13 +00:00
2008-04-08-CoalescerCrash.ll
…
2008-04-09-BranchFolding.ll
…
2008-04-15-LiveVariableBug.ll
…
2008-04-16-CoalescerBug.ll
…
2008-04-16-ReMatBug.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
2008-04-17-CoalescerBug.ll
Update more tests to the new EH scheme.
2011-08-31 21:39:05 +00:00
2008-04-24-MemCpyBug.ll
…
2008-04-24-pblendw-fold-crash.ll
…
2008-04-26-Asm-Optimize-Imm.ll
FileCheck-ize
2011-07-02 20:42:53 +00:00
2008-04-28-CoalescerBug.ll
…
2008-04-28-CyclicSchedUnit.ll
…
2008-05-01-InvalidOrdCompare.ll
…
2008-05-09-PHIElimBug.ll
…
2008-05-09-ShuffleLoweringBug.ll
…
2008-05-12-tailmerge-5.ll
…
2008-05-21-CoalescerBug.ll
Switch a couple -O0 tests to RABasic.
2011-11-12 20:11:04 +00:00
2008-05-22-FoldUnalignedLoad.ll
…
2008-05-28-CoalescerBug.ll
…
2008-05-28-LocalRegAllocBug.ll
Update more tests to the new EH scheme.
2011-08-31 21:39:05 +00:00
2008-06-13-NotVolatileLoadStore.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
2008-06-13-VolatileLoadStore.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
2008-06-16-SubregsBug.ll
…
2008-06-18-BadShuffle.ll
…
2008-06-25-VecISelBug.ll
…
2008-07-07-DanglingDeadInsts.ll
…
2008-07-09-ELFSectionAttributes.ll
…
2008-07-11-SHLBy1.ll
…
2008-07-16-CoalescerCrash.ll
…
2008-07-19-movups-spills.ll
Turn on list-ilp scheduling by default on x86 and x86-64, fix up
2011-03-08 02:42:25 +00:00
2008-07-22-CombinerCrash.ll
…
2008-07-23-VSetCC.ll
…
2008-08-06-CmpStride.ll
…
2008-08-06-RewriterBug.ll
…
2008-08-17-UComiCodeGenBug.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
2008-08-19-SubAndFetch.ll
Convert a bunch more tests over to the new atomic instructions.
2011-09-26 23:15:09 +00:00
2008-08-23-64Bit-maskmovq.ll
…
2008-08-25-AsmRegTypeMismatch.ll
Make X86::FsFLD0SS / FsFLD0SD real pseudo-instructions.
2011-11-29 22:27:25 +00:00
2008-08-31-EH_RETURN32.ll
Use the dwarf->llvm mapping to print register names in the cfi
2011-05-30 20:20:15 +00:00
2008-08-31-EH_RETURN64.ll
Use the dwarf->llvm mapping to print register names in the cfi
2011-05-30 20:20:15 +00:00
2008-09-05-sinttofp-2xi32.ll
Enable element promotion type legalization by deafault.
2011-10-16 20:31:33 +00:00
2008-09-09-LinearScanBug.ll
…
2008-09-11-CoalescerBug2.ll
…
2008-09-11-CoalescerBug.ll
…
2008-09-17-inline-asm-1.ll
Do not model all INLINEASM instructions as having unmodelled side effects.
2011-01-07 23:50:32 +00:00
2008-09-18-inline-asm-2.ll
Linear scan is going away.
2011-11-12 22:39:34 +00:00
2008-09-19-RegAllocBug.ll
…
2008-09-25-sseregparm-1.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
2008-09-26-FrameAddrBug.ll
…
2008-09-29-ReMatBug.ll
…
2008-09-29-VolatileBug.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
2008-10-06-MMXISelBug.ll
…
2008-10-06-x87ld-nan-1.ll
…
2008-10-06-x87ld-nan-2.ll
…
2008-10-07-SSEISelBug.ll
…
2008-10-11-CallCrash.ll
…
2008-10-13-CoalescerBug.ll
…
2008-10-16-VecUnaryOp.ll
…
2008-10-17-Asm64bitRConstraint.ll
…
2008-10-20-AsmDoubleInI32.ll
…
2008-10-24-FlippedCompare.ll
…
2008-10-27-CoalescerBug.ll
…
2008-10-27-StackRealignment.ll
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is
2011-06-18 06:05:24 +00:00
2008-10-29-ExpandVAARG.ll
…
2008-11-03-F80VAARG.ll
…
2008-11-06-testb.ll
…
2008-11-13-inlineasm-3.ll
…
2008-11-29-ULT-Sign.ll
…
2008-12-01-loop-iv-used-outside-loop.ll
…
2008-12-01-SpillerAssert.ll
…
2008-12-02-dagcombine-1.ll
…
2008-12-02-dagcombine-2.ll
…
2008-12-02-dagcombine-3.ll
…
2008-12-02-IllegalResultType.ll
…
2008-12-12-PrivateEHSymbol.ll
Add r130623 back now that ELF has been fixed to work with -fno-dwarf2-cfi-asm.
2011-05-01 15:44:13 +00:00
2008-12-16-BadShift.ll
…
2008-12-16-dagcombine-4.ll
…
2008-12-19-EarlyClobberBug.ll
Prepare remaining tests for -join-physreg going away.
2011-05-04 23:54:59 +00:00
2008-12-22-dagcombine-5.ll
…
2008-12-23-crazy-address.ll
…
2008-12-23-dagcombine-6.ll
…
2009-01-13-DoubleUpdate.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
2009-01-16-SchedulerBug.ll
…
2009-01-16-UIntToFP.ll
…
2009-01-18-ConstantExprCrash.ll
…
2009-01-25-NoSSE.ll
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is
2011-06-18 06:05:24 +00:00
2009-01-26-WrongCheck.ll
…
2009-01-27-NullStrings.ll
Update tests.
2011-01-16 18:02:57 +00:00
2009-01-31-BigShift2.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
2009-01-31-BigShift3.ll
…
2009-01-31-BigShift.ll
…
2009-02-01-LargeMask.ll
…
2009-02-03-AnalyzedTwice.ll
…
2009-02-04-sext-i64-gep.ll
…
2009-02-08-CoalescerBug.ll
…
2009-02-09-ivs-different-sizes.ll
…
2009-02-11-codegenprepare-reuse.ll
…
2009-02-12-DebugInfoVLA.ll
…
2009-02-12-InlineAsm-nieZ-constraints.ll
FileCheck-ize and modernize IR.
2011-07-02 20:42:50 +00:00
2009-02-12-SpillerBug.ll
…
2009-02-21-ExtWeakInitializer.ll
…
2009-02-25-CommuteBug.ll
…
2009-02-26-MachineLICMBug.ll
…
2009-03-03-BitcastLongDouble.ll
…
2009-03-03-BTHang.ll
…
2009-03-05-burr-list-crash.ll
…
2009-03-07-FPConstSelect.ll
…
2009-03-09-APIntCrash.ll
…
2009-03-09-SpillerBug.ll
…
2009-03-10-CoalescerBug.ll
…
2009-03-12-CPAlignBug.ll
…
2009-03-13-PHIElimBug.ll
Update more tests to the new EH scheme.
2011-08-31 21:39:05 +00:00
2009-03-16-PHIElimInLPad.ll
Update more tests to the new EH scheme.
2011-08-31 21:39:05 +00:00
2009-03-23-i80-fp80.ll
…
2009-03-23-LinearScanBug.ll
…
2009-03-23-MultiUseSched.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
2009-03-25-TestBug.ll
…
2009-03-26-NoImplicitFPBug.ll
…
2009-04-12-FastIselOverflowCrash.ll
Remove support for parsing the "type i32" syntax for defining a numbered
2011-06-19 00:03:46 +00:00
2009-04-12-picrel.ll
…
2009-04-13-2AddrAssert-2.ll
Remove support for parsing the "type i32" syntax for defining a numbered
2011-06-19 00:03:46 +00:00
2009-04-13-2AddrAssert.ll
…
2009-04-14-IllegalRegs.ll
…
2009-04-16-SpillerUnfold.ll
…
2009-04-21-NoReloadImpDef.ll
Linear scan is going away.
2011-11-12 22:39:34 +00:00
2009-04-24.ll
Fix mistyped CHECK lines.
2011-03-09 22:07:31 +00:00
2009-04-25-CoalescerBug.ll
…
2009-04-27-CoalescerAssert.ll
…
2009-04-27-LiveIntervalsAssert2.ll
…
2009-04-27-LiveIntervalsAssert.ll
…
2009-04-29-IndirectDestOperands.ll
…
2009-04-29-LinearScanBug.ll
Update test.
2011-06-19 12:14:34 +00:00
2009-04-29-RegAllocAssert.ll
Remove support for parsing the "type i32" syntax for defining a numbered
2011-06-19 00:03:46 +00:00
2009-04-scale.ll
…
2009-05-08-InlineAsmIOffset.ll
…
2009-05-11-tailmerge-crash.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
2009-05-19-SingleElementExtractElement.ll
…
2009-05-23-available_externally.ll
…
2009-05-23-dagcombine-shifts.ll
…
2009-05-28-DAGCombineCrash.ll
…
2009-05-30-ISelBug.ll
…
2009-06-02-RewriterBug.ll
…
2009-06-03-Win64DisableRedZone.ll
Triple::MinGW64 is deprecated and removed. We can use Triple::MinGW32 generally.
2011-02-17 12:24:17 +00:00
2009-06-03-Win64SpillXMM.ll
Triple::MinGW64 is deprecated and removed. We can use Triple::MinGW32 generally.
2011-02-17 12:24:17 +00:00
2009-06-04-VirtualLiveIn.ll
Remove support for parsing the "type i32" syntax for defining a numbered
2011-06-19 00:03:46 +00:00
2009-06-05-ScalarToVectorByteMMX.ll
Attempt to fix -mtriple=i686-{cygwin|mingw|win32} regressions. Nakamura,
2011-09-20 00:08:12 +00:00
2009-06-05-sitofpCrash.ll
…
2009-06-05-VariableIndexInsert.ll
…
2009-06-05-VZextByteShort.ll
Enable element promotion type legalization by deafault.
2011-10-16 20:31:33 +00:00
2009-06-06-ConcatVectors.ll
…
2009-06-07-ExpandMMXBitcast.ll
Stabilize the test by specifying an exact cpu target
2011-10-30 08:07:50 +00:00
2009-06-12-x86_64-tail-call-conv-out-of-sync-bug.ll
…
2009-06-15-not-a-tail-call.ll
…
2009-06-18-movlp-shuffle-register.ll
- Move all MOVSS and MOVSD patterns close to their definitions
2011-08-31 03:04:20 +00:00
2009-07-06-TwoAddrAssert.ll
…
2009-07-07-SplitICmp.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
2009-07-09-ExtractBoolFromVector.ll
…
2009-07-15-CoalescerBug.ll
…
2009-07-16-CoalescerBug.ll
…
2009-07-19-AsmExtraOperands.ll
…
2009-07-20-CoalescerBug.ll
…
2009-07-20-DAGCombineBug.ll
…
2009-08-02-mmx-scalar-to-vector.ll
…
2009-08-06-branchfolder-crash.ll
make the asmparser reject function and type redefinitions. 'Merging' hasn't been
2011-06-17 07:06:44 +00:00
2009-08-06-inlineasm.ll
Experiment with changing the default 32-bit linux stack alignment to
2011-01-13 06:47:10 +00:00
2009-08-08-CastError.ll
…
2009-08-12-badswitch.ll
…
2009-08-14-Win64MemoryIndirectArg.ll
Remove support for parsing the "type i32" syntax for defining a numbered
2011-06-19 00:03:46 +00:00
2009-08-19-LoadNarrowingMiscompile.ll
…
2009-08-23-linkerprivate.ll
…
2009-08-23-SubRegReuseUndo.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
2009-09-10-LoadFoldingBug.ll
Update more tests to the new EH scheme.
2011-08-31 21:39:05 +00:00
2009-09-10-SpillComments.ll
Use pushq / popq instead of subq $8, %rsp / addq $8, %rsp to adjust stack in
2011-01-03 22:53:22 +00:00
2009-09-16-CoalescerBug.ll
…
2009-09-19-earlyclobber.ll
Remove support for parsing the "type i32" syntax for defining a numbered
2011-06-19 00:03:46 +00:00
2009-09-21-NoSpillLoopCount.ll
…
2009-09-22-CoalescerBug.ll
…
2009-09-23-LiveVariablesBug.ll
…
2009-10-14-LiveVariablesBug.ll
…
2009-10-16-Scope.ll
Add a new wrapper node for a DILexicalBlock that encapsulates it and a
2011-10-11 22:59:11 +00:00
2009-10-19-atomic-cmp-eflags.ll
Convert a bunch more tests over to the new atomic instructions.
2011-09-26 23:15:09 +00:00
2009-10-19-EmergencySpill.ll
…
2009-10-25-RewriterBug.ll
…
2009-11-04-SubregCoalescingBug.ll
…
2009-11-13-VirtRegRewriterBug.ll
…
2009-11-16-MachineLICM.ll
…
2009-11-16-UnfoldMemOpBug.ll
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is
2011-06-18 06:05:24 +00:00
2009-11-17-UpdateTerminator.ll
…
2009-11-18-TwoAddrKill.ll
…
2009-11-25-ImpDefBug.ll
Update more tests to the new EH scheme.
2011-08-31 21:39:05 +00:00
2009-12-01-EarlyClobberBug.ll
Be more aggressive about following hints.
2011-07-08 20:46:18 +00:00
2009-12-11-TLSNoRedZone.ll
…
2010-01-05-ZExt-Shl.ll
…
2010-01-07-ISelBug.ll
…
2010-01-07-UAMemFeature.ll
…
2010-01-08-Atomic64Bug.ll
Convert a bunch more tests over to the new atomic instructions.
2011-09-26 23:15:09 +00:00
2010-01-11-ExtraPHIArg.ll
…
2010-01-13-OptExtBug.ll
…
2010-01-15-SelectionDAGCycle.ll
…
2010-01-18-DbgValue.ll
…
2010-01-19-OptExtBug.ll
…
2010-02-01-DbgValueCrash.ll
…
2010-02-01-TaillCallCrash.ll
…
2010-02-03-DualUndef.ll
…
2010-02-04-SchedulerBug.ll
…
2010-02-11-NonTemporal.ll
…
2010-02-12-CoalescerBug-Impdef.ll
Land the long talked about "type system rewrite" patch. This
2011-07-09 17:41:24 +00:00
2010-02-15-ImplicitDefBug.ll
…
2010-02-19-TailCallRetAddrBug.ll
Fix one more batch of X86 tests to be register allocation dependent.
2011-04-05 20:20:30 +00:00
2010-02-23-DAGCombineBug.ll
…
2010-02-23-DIV8rDefinesAX.ll
…
2010-02-23-RematImplicitSubreg.ll
…
2010-02-23-SingleDefPhiJoin.ll
…
2010-03-04-Mul8Bug.ll
…
2010-03-05-ConstantFoldCFG.ll
…
2010-03-05-EFLAGS-Redef.ll
…
2010-03-17-ISelBug.ll
…
2010-04-06-SSEDomainFixCrash.ll
Update more tests to the new EH scheme.
2011-08-31 21:39:05 +00:00
2010-04-07-DbgValueOtherTargets.ll
…
2010-04-08-CoalescerBug.ll
Add Mode64Bit feature and sink it down to MC layer.
2011-07-07 21:06:52 +00:00
2010-04-13-AnalyzeBranchCrash.ll
…
2010-04-21-CoalescerBug.ll
…
2010-04-23-mmx-movdq2q.ll
Enable element promotion type legalization by deafault.
2011-10-16 20:31:33 +00:00
2010-04-29-CoalescerCrash.ll
…
2010-04-30-LocalAlloc-LandingPad.ll
Convert to the new EH model.
2011-11-08 00:17:28 +00:00
2010-05-03-CoalescerSubRegClobber.ll
Fix one more batch of X86 tests to be register allocation dependent.
2011-04-05 20:20:30 +00:00
2010-05-05-LocalAllocEarlyClobber.ll
…
2010-05-06-LocalInlineAsmClobber.ll
…
2010-05-07-ldconvert.ll
…
2010-05-10-DAGCombinerBug.ll
…
2010-05-12-FastAllocKills.ll
…
2010-05-16-nosseconversion.ll
…
2010-05-25-DotDebugLoc.ll
Force a triple to make this test pass on Darwin.
2011-05-27 23:12:48 +00:00
2010-05-26-DotDebugLoc.ll
Make size computation less brittle.
2011-05-27 22:05:41 +00:00
2010-05-26-FP_TO_INT-crash.ll
…
2010-05-28-Crash.ll
Run LiveDebugVariables in RegAllocBasic and RegAllocGreedy.
2011-04-05 21:40:37 +00:00
2010-06-01-DeadArg-DbgInfo.ll
Run LiveDebugVariables in RegAllocBasic and RegAllocGreedy.
2011-04-05 21:40:37 +00:00
2010-06-09-FastAllocRegisters.ll
…
2010-06-14-fast-isel-fs-load.ll
…
2010-06-15-FastAllocEarlyCLobber.ll
…
2010-06-24-g-constraint-crash.ll
…
2010-06-25-asm-RA-crash.ll
…
2010-06-25-CoalescerSubRegDefDead.ll
Add Mode64Bit feature and sink it down to MC layer.
2011-07-07 21:06:52 +00:00
2010-06-28-DbgEntryPC.ll
Add an option to pad an uleb128 to MCObjectWriter and remove the uleb128 encoding from the DWARF asm printer.
2011-11-05 11:52:44 +00:00
2010-06-28-FastAllocTiedOperand.ll
…
2010-06-28-matched-g-constraint.ll
…
2010-07-02-asm-alignstack.ll
Use pushq / popq instead of subq $8, %rsp / addq $8, %rsp to adjust stack in
2011-01-03 22:53:22 +00:00
2010-07-02-UnfoldBug.ll
…
2010-07-06-asm-RIP.ll
…
2010-07-06-DbgCrash.ll
…
2010-07-11-FPStackLoneUse.ll
Fix broken x86_64 tests which specify non-64-bit cpu's.
2011-07-08 22:29:33 +00:00
2010-07-13-indirectXconstraint.ll
…
2010-07-15-Crash.ll
…
2010-07-29-SetccSimplify.ll
…
2010-08-04-MaskedSignedCompare.ll
…
2010-08-04-MingWCrash.ll
Convert to the new EH model.
2011-11-08 00:17:28 +00:00
2010-08-04-StackVariable.ll
Teach dwarf writer to handle complex address expression for .debug_loc entries.
2011-04-28 02:22:40 +00:00
2010-08-10-DbgConstant.ll
test/CodeGen/X86/2010-08-10-DbgConstant.ll: Add explicit -mtriple=i686-linux. It must be for elf!
2011-10-28 10:50:52 +00:00
2010-09-01-RemoveCopyByCommutingDef.ll
…
2010-09-16-asmcrash.ll
…
2010-09-16-EmptyFilename.ll
…
2010-09-17-SideEffectsInChain.ll
Add Mode64Bit feature and sink it down to MC layer.
2011-07-07 21:06:52 +00:00
2010-09-30-CMOV-JumpTable-PHI.ll
Eliminate the unused CodeGenPrepare option to split critical edges.
2011-03-02 03:31:46 +00:00
2010-10-08-cmpxchg8b.ll
Convert a bunch more tests over to the new atomic instructions.
2011-09-26 23:15:09 +00:00
2010-11-02-DbgParameter.ll
…
2010-11-09-MOVLPS.ll
Land the long talked about "type system rewrite" patch. This
2011-07-09 17:41:24 +00:00
2010-11-18-SelectOfExtload.ll
…
2010-12-02-MC-Set.ll
…
2011-01-07-LegalizeTypesCrash.ll
Fix the other problem reported in PR8582. Testcase and patch by
2011-01-06 23:45:22 +00:00
2011-01-10-DagCombineHang.ll
Fix PR 8916 (qv for analysis), at least the immediate problem.
2011-01-10 21:53:07 +00:00
2011-01-24-DbgValue-Before-Use.ll
Always use the string pool, even when it makes the .o larger. This may help
2011-10-28 05:29:47 +00:00
2011-02-04-FastRegallocNoFP.ll
Mark that the return is using EAX so that we don't use it for some other
2011-02-04 22:44:08 +00:00
2011-02-12-shuffle.ll
Move test for appropriate directory.
2011-05-17 19:06:43 +00:00
2011-02-21-VirtRegRewriter-KillSubReg.ll
VirtRegRewriter assertion fix.
2011-02-22 06:52:56 +00:00
2011-02-23-UnfoldBug.ll
Fix bug in X86 folding / unfolding table. Int_CMPSDrm and Int_CMPSSrm memory
2011-02-24 02:36:52 +00:00
2011-02-27-Fpextend.ll
Windows codegen also dies on this, so restrict to the platform it was
2011-02-28 14:22:08 +00:00
2011-03-02-DAGCombiner.ll
Test case for r126864. Radar 9056407.
2011-03-02 23:41:40 +00:00
2011-03-08-Sched-crash.ll
Strip cruft.
2011-03-08 20:19:10 +00:00
2011-03-09-Physreg-Coalescing.ll
Make physreg coalescing independent on the number of uses of the virtual register.
2011-03-09 19:27:06 +00:00
2011-03-30-CreateFixedObjCrash.ll
Don't try to create zero-sized stack objects.
2011-03-30 23:44:13 +00:00
2011-04-13-SchedCmpJmp.ll
Fix broken x86_64 tests which specify non-64-bit cpu's.
2011-07-08 22:29:33 +00:00
2011-05-09-loaduse.ll
Fixes a bug in the DAGCombiner. LoadSDNodes have two values (data, chain).
2011-05-11 14:40:50 +00:00
2011-05-26-UnreachableBlockElim.ll
Fix PR10029 - VerifyCoalescing failure on patterns_dfa.c of 445.gobmk.
2011-05-27 05:04:51 +00:00
2011-05-27-CrossClassCoalescing.ll
Add a GR32_NOREX_NOSP register class and fix a bug where getMatchingSuperRegClass()
2011-05-27 22:26:04 +00:00
2011-06-01-fildll.ll
Andy pointed out a dumb omission in this test case. Thanks Andy!
2011-06-02 19:26:49 +00:00
2011-06-03-x87chain.ll
Tweak this test for ARM-hosted 'bot.
2011-06-07 15:23:11 +00:00
2011-06-06-fgetsign80bit.ll
Test case for PR10085.
2011-06-06 20:03:22 +00:00
2011-06-12-FastAllocSpill.ll
Be less aggressive about hinting in RAFast.
2011-06-13 03:26:46 +00:00
2011-06-14-mmx-inlineasm.ll
Test case for x86 MMX inline asm. rdar://problem/8886707
2011-06-14 21:51:38 +00:00
2011-06-14-PreschedRegalias.ll
lit support for REQUIRES: asserts.
2011-06-22 23:23:19 +00:00
2011-06-19-QuicksortCoalescerBug.ll
Fix PR9960 by teaching SimpleRegisterCoalescing::AdjustCopiesBackFrom() to preserve
2011-05-20 03:54:04 +00:00
2011-07-13-BadFrameIndexDisplacement.ll
Make sure we don't combine a large displacement and a frame index in the same addressing mode on x86-64. It can overflow, leading to a crash/miscompile.
2011-07-13 21:29:53 +00:00
2011-08-23-PerformSubCombine128.ll
PerformSubCombine to work on integers larger than i128. Fixes a crasher.
2011-08-23 19:01:24 +00:00
2011-08-23-Trampoline.ll
Another forgotten trampoline testcase.
2011-09-07 10:05:14 +00:00
2011-08-29-BlockConstant.ll
Compress Repeated Byte Output
2011-08-31 17:30:56 +00:00
2011-08-29-InitOrder.ll
Darwin wants ctors/dtors to be ordered the other way round to linux.
2011-09-02 18:07:19 +00:00
2011-09-14-valcoalesce.ll
Test case trial and error. Not sure the proper way to check MBB names.
2011-09-16 03:57:19 +00:00
2011-09-18-sse2cmp.ll
setOperationAction should be done on the return value of the type, not the operands.
2011-09-18 14:57:03 +00:00
2011-09-21-setcc-bug.ll
add another testcase for pr10902
2011-09-21 17:13:40 +00:00
2011-10-11-SpillDead.ll
Add value numbers when spilling dead defs.
2011-10-14 00:34:31 +00:00
2011-10-11-srl.ll
Add support for legalization of vector SHL/SRA/SRL instructions
2011-10-11 14:36:35 +00:00
2011-10-12-MachineCSE.ll
We need to verify that the machine instruction we're using as a replacement for
2011-10-12 23:03:40 +00:00
2011-10-18-FastISel-VectorParams.ll
Force SSE for this test.
2011-10-29 19:43:44 +00:00
2011-10-19-LegelizeLoad.ll
Fix a bug in the legalization of vector anyext-load and trunc-store. Mem Index starts with zero.
2011-10-18 22:32:43 +00:00
2011-10-19-widen_vselect.ll
Add support for the vector-widening of vselect and vector-setcc
2011-10-19 09:45:11 +00:00
2011-10-21-widen-cmp.ll
Fix pr11193.
2011-10-22 12:39:25 +00:00
2011-10-27-tstore.ll
Add a new DAGCombine optimization for BUILD_VECTOR.
2011-10-29 21:23:04 +00:00
2011-10-30-padd.ll
Fix pr11266.
2011-10-30 13:24:22 +00:00
2011-11-07-LegalizeBuildVector.ll
Add a bunch of calls to RemoveDeadNode in LegalizeDAG, so legalization doesn't get confused by CSE later on. Fixes PR11318.
2011-11-08 01:25:24 +00:00
2011-11-09-FoldImpDefs.ll
Strip old implicit operands after foldMemoryOperand.
2011-11-10 00:17:03 +00:00
2011-11-22-AVX2-Domains.ll
Fix PR11422.
2011-11-23 04:03:08 +00:00
2011-11-30-or.ll
Add test arch to make it pass on non x86 targets
2011-11-30 17:34:28 +00:00
2011-12-8-bitcastintprom.ll
Fix a bug in the integer-promotion of bitcast operations on vector types.
2011-12-08 13:10:01 +00:00
2011-12-06-AVXVectorExtractCombine.ll
Fix an optimization involving EXTRACT_SUBVECTOR in DAGCombine so it behaves correctly. PR11494.
2011-12-07 00:11:56 +00:00
2011-12-06-BitcastVectorGlobal.ll
Support vector bitcasts in the AsmPrinter. PR11495.
2011-12-07 00:50:54 +00:00
2011-12-08-AVXISelBugs.ll
This is the second fix related to VZEXT_MOVL node.
2011-12-20 13:34:28 +00:00
2011-12-15-vec_shift.ll
Make sure that the lower bits on the VSELECT condition are properly set.
2011-12-17 01:08:46 +00:00
2011-12-26-extractelement-duplicate-load.ll
Make sure DAGCombiner doesn't introduce multiple loads from the same memory location. PR10747, part 2.
2011-12-26 22:49:32 +00:00
2011-12-28-vselecti8.ll
Revert 147426 because it caused pr11696.
2012-01-03 22:19:42 +00:00
2011-20-21-zext-ui2fp.ll
Fix pr11194. When promoting and splitting integers we need to use
2011-10-21 17:35:19 +00:00
2012-1-10-buildvector.ll
Fix a bug in the legalization of shuffle vectors. When we emulate shuffles using BUILD_VECTORS we may be using a BV of different type. Make sure to cast it back.
2012-01-10 14:28:46 +00:00
2012-01-10-UndefExceptionEdge.ll
Disable test that seems to expose an unrelated Linux issue.
2012-01-11 03:42:27 +00:00
9601.ll
Oddly people want to use the 'r' constraint for fp constants on x86.
2011-05-19 21:33:47 +00:00
20090313-signext.ll
…
abi-isel.ll
Teach LiveInterval::isZeroLength about null SlotIndexes.
2011-05-16 23:50:05 +00:00
add-of-carry.ll
X86: Add a bunch of peeps for add and sub of SETB.
2011-05-08 18:36:07 +00:00
add.ll
Prepare remaining tests for -join-physreg going away.
2011-05-04 23:54:59 +00:00
adde-carry.ll
Revert a part of r126557 which could create unschedulable DAGs.
2011-06-29 13:47:25 +00:00
addr-label-difference.ll
Update tests.
2011-01-16 18:02:57 +00:00
aliases.ll
Add 129518 back with a fix for when we are producing eh just because of debug info.
2011-04-15 15:11:06 +00:00
aligned-comm.ll
…
alignment-2.ll
Filecheck-ize.
2011-09-30 23:40:29 +00:00
alignment.ll
Add new line at end of file.
2011-09-30 23:21:11 +00:00
all-ones-vector.ll
…
alldiv-divdi3.ll
…
alloca-align-rounding-32.ll
…
alloca-align-rounding.ll
…
allrem-moddi3.ll
Add _allrem, _aullrem and _allmul to the runtime for MSVC.
2011-07-08 21:40:25 +00:00
and-or-fold.ll
…
and-su.ll
…
andimm8.ll
Prepare remaining tests for -join-physreg going away.
2011-05-04 23:54:59 +00:00
anyext.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
apm.ll
Fix sfence, lfence, mfence, and clflush to be able to be selected when AVX is enabled. Fix monitor and mwait to require SSE3 or AVX, previously they worked even if SSE3 was disabled. Make prefetch instructions not set the execution domain since they don't use XMM registers.
2012-01-01 19:40:22 +00:00
arg-cast.ll
…
asm-block-labels.ll
…
asm-global-imm.ll
FileCheck-ize and simplify RUNs.
2011-07-02 20:42:48 +00:00
asm-indirect-mem.ll
…
asm-label2.ll
Update more tests to the new EH scheme.
2011-08-31 21:39:05 +00:00
asm-label.ll
Really fix the fall-through logic.
2011-06-12 05:57:01 +00:00
asm-modifier-P.ll
…
asm-modifier.ll
…
atomic_add.ll
Convert a bunch more tests over to the new atomic instructions.
2011-09-26 23:15:09 +00:00
atomic_op.ll
Last batch of test conversions to new atomic instructions.
2011-09-27 00:17:29 +00:00
atomic-load-store-wide.ll
Error out on CodeGen of unaligned load/store. Fix test so it isn't accidentally testing that case.
2011-09-13 20:50:54 +00:00
atomic-load-store.ll
Fix atomic load and store on x86 to pass -verify-machineinstrs (and possibly fix some subtle bugs involving passes which check mayStore()).
2011-09-07 18:48:32 +00:00
atomic-or.ll
Convert a bunch more tests over to the new atomic instructions.
2011-09-26 23:15:09 +00:00
Atomics-64.ll
Last batch of test conversions to new atomic instructions.
2011-09-27 00:17:29 +00:00
attribute-sections.ll
Be nice to Xcore and the XMOS assembler and avoid quoting section names
2011-03-04 20:03:14 +00:00
avoid-lea-scale2.ll
test/CodeGen/X86: FileCheck-ize and add actions for x86_64-linux and x86_64-win32.
2011-03-16 13:53:07 +00:00
avoid-loop-align-2.ll
…
avoid-loop-align.ll
…
avx2-arith.ll
Add AVX2 instruction lowering for add, sub, and mul.
2011-11-09 07:28:55 +00:00
avx2-cmp.ll
Add instruction selection for AVX2 integer comparisons.
2011-11-09 08:06:13 +00:00
avx2-intrinsics-x86.ll
Fix VINSERTF128/VEXTRACTF128 to be marked as FP instructions. Allow execution dependency fix pass to convert them to their integer equivalents when AVX2 is enabled.
2011-11-29 05:37:58 +00:00
avx2-logic.ll
Use 256-bit vcmpeqd for creating an all ones vector when AVX2 is enabled.
2011-11-19 22:34:59 +00:00
avx2-nontemporal.ll
Clean up patterns for MOVNT*. Not sure why there were floating point types on MOVNTPS and MOVNTDQ. And v4i64 was completely missing.
2012-01-09 06:52:46 +00:00
avx2-phaddsub.ll
Add instruction selection support for horizontal add/sub of 256-bit floating point vectors. Also add the test case for 256-bit integer vectors.
2011-12-02 07:16:01 +00:00
avx2-shift.ll
Make LowerSIGN_EXTEND_INREG split 256-bit vectors when AVX1 is enabled and use AVX2 shifts when AVX2 is enabled.
2011-11-21 01:12:36 +00:00
avx2-unpack.ll
Don't try to match 'unpackl/h v, v' for 32xi8 and 16xi16 when only AVX1 is supported. Fix 'unpackh v, v' for 256-bit types to understand 128-bit lanes.
2011-12-16 08:06:31 +00:00
avx2-vbroadcast.ll
Fix a crash in AVX2 when trying to broadcast a double into a 128-bit vector. There is no vbroadcastsd xmm, but we do need to support 64-bit integers broadcasted into xmm. Also factor the AVX check into the isVectorBroadcast function. This makes more sense since the AVX2 check was already inside.
2012-01-10 08:23:59 +00:00
avx2-vperm2i128.ll
Merge isSHUFPMask and isCommutedSHUFPMask into single function that can do both. Do the same for the 256-bit version. Use loops to reduce size of isVSHUFPYMask. Fix test cases that were incorrectly passing due to isCommutedSHUFPMask not checking for the vector being 128-bit. This caused some 256-bit shuffles to be incorrectly commuted.
2011-12-06 04:59:07 +00:00
avx-arith.ll
Fixed vsqrt.ss intrinsic usage - order of input operands was wrong.
2011-11-29 15:00:45 +00:00
avx-basic.ll
Fix assert in LowerBUILD_VECTOR for v16i16 type on AVX.
2011-12-15 21:34:44 +00:00
avx-bitcast.ll
Add a AVX version of a simple i64 -> f64 bitcast. This could be
2011-09-08 21:52:33 +00:00
avx-blend.ll
Apply Duncan's test fix from r139986 to the avx version of that test too.
2011-09-18 00:41:38 +00:00
avx-cast.ll
Fix VINSERTF128/VEXTRACTF128 to be marked as FP instructions. Allow execution dependency fix pass to convert them to their integer equivalents when AVX2 is enabled.
2011-11-29 05:37:58 +00:00
avx-cmp.ll
Match X86ISD::FSETCCsd and X86ISD::FSETCCss while in AVX mode. This fix
2011-09-19 21:29:24 +00:00
avx-cvt.ll
Reapply a more appropriate solution than in r137114. AVX supports
2011-08-09 17:39:13 +00:00
avx-intrinsics-x86_64.ll
…
avx-intrinsics-x86.ll
Allow CRC32 instructions to be selected when AVX is enabled.
2012-01-01 19:51:58 +00:00
avx-load-store.ll
Remove X86-dependent stuff from SSEDomainFix.
2011-09-27 23:50:46 +00:00
avx-logic.ll
Properly qualify AVX2 specific parts of execution dependency table. Also enable converting between 256-bit PS/PD operations when AVX1 is enabled. Fixes PR11370.
2011-11-15 05:55:35 +00:00
avx-minmax.ll
test/CodeGen/X86/avx-minmax.ll: Unbreak Win32.
2011-09-20 14:11:35 +00:00
avx-movdup.ll
Rename and tidy up tests
2011-08-09 03:04:23 +00:00
avx-select.ll
Add 256-bit support for v8i32, v4i64 and v4f64 ISD::SELECT. Fix PR10556
2011-08-09 23:27:13 +00:00
avx-shift.ll
Make LowerSIGN_EXTEND_INREG split 256-bit vectors when AVX1 is enabled and use AVX2 shifts when AVX2 is enabled.
2011-11-21 01:12:36 +00:00
avx-shuffle-x86_32.ll
FileCheck hygiene.
2012-01-05 00:43:34 +00:00
avx-shuffle.ll
Enhance DAGCombine for transforming 128->256 casts into a vmovaps, rather
2012-01-03 21:05:52 +00:00
avx-splat.ll
Fix VINSERTF128/VEXTRACTF128 to be marked as FP instructions. Allow execution dependency fix pass to convert them to their integer equivalents when AVX2 is enabled.
2011-11-29 05:37:58 +00:00
avx-unpack.ll
Don't try to match 'unpackl/h v, v' for 32xi8 and 16xi16 when only AVX1 is supported. Fix 'unpackh v, v' for 256-bit types to understand 128-bit lanes.
2011-12-16 08:06:31 +00:00
avx-varargs-x86_64.ll
Pass AVX vectors which are arguments to varargs functions on the stack. <rdar://problem/10463281>.
2011-12-01 04:49:21 +00:00
avx-vbroadcast.ll
Fix a crash in AVX2 when trying to broadcast a double into a 128-bit vector. There is no vbroadcastsd xmm, but we do need to support 64-bit integers broadcasted into xmm. Also factor the AVX check into the isVectorBroadcast function. This makes more sense since the AVX2 check was already inside.
2012-01-10 08:23:59 +00:00
avx-vextractf128.ll
Add a dag combine to xform 256-bit shuffles into simple vector
2011-08-11 21:50:44 +00:00
avx-vinsertf128.ll
Add a DAGCombine for subvector extracts to remove useless chains of
2011-09-20 23:19:33 +00:00
avx-vmovddup.ll
Add support for AVX 256-bit version of MOVDDUP!
2011-08-25 21:40:37 +00:00
avx-vperm2f128.ll
Merge isSHUFPMask and isCommutedSHUFPMask into single function that can do both. Do the same for the 256-bit version. Use loops to reduce size of isVSHUFPYMask. Fix test cases that were incorrectly passing due to isCommutedSHUFPMask not checking for the vector being 128-bit. This caused some 256-bit shuffles to be incorrectly commuted.
2011-12-06 04:59:07 +00:00
avx-vpermil.ll
Fix issues in shuffle decoding around VPERM* instructions. Fix shuffle decoding for VSHUFPS/D for 256-bit types. Add pattern matching for memory forms of VPERMILPS/VPERMILPD.
2011-11-29 07:49:05 +00:00
avx-vshufp.ll
Add patterns for integer forms of SHUFPD/VSHUFPD with a memory load.
2011-12-31 23:24:49 +00:00
avx-vzeroupper.ll
Enhanced vzeroupper insertion pass that avoids inserting vzeroupper where it is unnecessary through local analysis. Patch from Bruno Cardoso Lopes, with some additional changes.
2011-11-04 23:46:11 +00:00
barrier-sse.ll
Last batch of test conversions to new atomic instructions.
2011-09-27 00:17:29 +00:00
barrier.ll
Last batch of test conversions to new atomic instructions.
2011-09-27 00:17:29 +00:00
basic-promote-integers.ll
Move the legalizer tests to the X86 directory because the test uses the x86
2011-06-07 05:23:58 +00:00
bc-extract.ll
Make X86::FsFLD0SS / FsFLD0SD real pseudo-instructions.
2011-11-29 22:27:25 +00:00
bigstructret2.ll
…
bigstructret.ll
…
bit-test-shift.ll
…
bitcast2.ll
Revert r133452: "Emit movq for 64-bit register to XMM register moves..."
2011-06-21 17:35:13 +00:00
bitcast-int-to-vector.ll
…
bitcast.ll
…
block-placement.ll
Take two on rotating the block ordering of loops. My previous attempt
2011-11-27 13:34:33 +00:00
bmi.ll
Use standard promotion for i8 CTTZ nodes and i8 CTLZ nodes when the
2011-12-24 12:12:34 +00:00
bool-zext.ll
Fixed regression due to commit 131709, which disables vararg tail call optimizations on Win64
2011-05-20 17:49:39 +00:00
br-fold.ll
…
brcond.ll
Reverted commit #147601 upon Evan's request.
2012-01-08 17:20:33 +00:00
break-anti-dependencies.ll
Turn on list-ilp scheduling by default on x86 and x86-64, fix up
2011-03-08 02:42:25 +00:00
break-sse-dep.ll
Relax expressions and add explicit triplets -linux and -win32.
2011-02-22 07:19:20 +00:00
bss_pagealigned.ll
Be nice to Xcore and the XMOS assembler and avoid quoting section names
2011-03-04 20:03:14 +00:00
bswap-inline-asm.ll
Recognize inline asm 'rev /bin/bash, ' as a bswap intrinsic call.
2011-01-08 01:24:27 +00:00
bswap.ll
Force CPU type on test so it doesn't accidentally emit movbe instead of bswap on Intel Atom CPUs.
2011-10-13 14:27:54 +00:00
bt.ll
…
btq.ll
X86: Use btq for bit tests if the immediate can't be encoded in 32 bits.
2011-11-23 13:54:17 +00:00
byval2.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
byval3.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
byval4.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
byval5.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
byval6.ll
…
byval7.ll
Prepare remaining tests for -join-physreg going away.
2011-05-04 23:54:59 +00:00
byval-align.ll
Recommit r132764 since it didn't cause the windows buildbot failures.
2011-06-09 15:39:01 +00:00
byval.ll
test/CodeGen/X86/byval*.ll: Win64 has not supported byval yet.
2011-03-16 13:52:20 +00:00
call-imm.ll
…
call-push.ll
look for the verboten argument slot access in any order, thanks to Frits
2011-04-09 17:00:34 +00:00
change-compare-stride-0.ll
LSR: rewrite inner loops only.
2011-09-29 01:33:38 +00:00
change-compare-stride-1.ll
Use a bigger hammer to fix PR11314 by disabling the "forcing two-address
2011-11-10 07:43:16 +00:00
change-compare-stride-trickiness-0.ll
…
change-compare-stride-trickiness-1.ll
FileCheck-ize and modernize the RUN line.
2011-07-02 20:42:44 +00:00
change-compare-stride-trickiness-2.ll
…
clz.ll
Use standard promotion for i8 CTTZ nodes and i8 CTLZ nodes when the
2011-12-24 12:12:34 +00:00
cmov.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
cmp.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
cmpxchg16b.ll
FileCheck hygiene.
2012-01-05 00:43:34 +00:00
coalesce-esp.ll
Turn on list-ilp scheduling by default on x86 and x86-64, fix up
2011-03-08 02:42:25 +00:00
coalescer-commute1.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
coalescer-commute2.ll
Prepare remaining tests for -join-physreg going away.
2011-05-04 23:54:59 +00:00
coalescer-commute3.ll
…
coalescer-commute4.ll
…
coalescer-commute5.ll
…
coalescer-cross.ll
Remove support for parsing the "type i32" syntax for defining a numbered
2011-06-19 00:03:46 +00:00
coalescer-dce.ll
Add test case for r138018.
2011-08-19 04:30:24 +00:00
coalescer-remat.ll
Convert a bunch more tests over to the new atomic instructions.
2011-09-26 23:15:09 +00:00
code_placement_eh.ll
Update to new EH scheme.
2011-08-27 04:53:41 +00:00
code_placement.ll
…
codegen-prepare-cast.ll
…
codegen-prepare-extload.ll
Relax expressions and add explicit triplets -linux and -win32.
2011-02-22 07:19:28 +00:00
codemodel.ll
…
combine-lds.ll
…
combiner-aa-0.ll
…
combiner-aa-1.ll
…
commute-intrinsic.ll
…
commute-two-addr.ll
Turn on list-ilp scheduling by default on x86 and x86-64, fix up
2011-03-08 02:42:25 +00:00
compare_folding.ll
…
compare-add.ll
…
compare-inf.ll
…
compiler_used.ll
…
complex-asm.ll
…
complex-fca.ll
…
conditional-indecrement.ll
Add some x86 specific dagcombines for conditional increments.
2010-12-21 21:41:44 +00:00
constant-pool-remat-0.ll
When dead code elimination removes all but one use, try to fold the single def into the remaining use.
2011-04-05 20:20:26 +00:00
constant-pool-sharing.ll
Relax expressions and add explicit triplets -linux and -win32.
2011-02-22 07:19:37 +00:00
constpool.ll
…
convert-2-addr-3-addr-inc64.ll
test/CodeGen/X86: FileCheck-ize and add actions for x86_64-linux and x86_64-win32.
2011-03-16 13:53:07 +00:00
copysign-zero.ll
…
crash-nosse.ll
Eliminate copies of undefined values during coalescing.
2011-07-26 23:00:24 +00:00
crash-O0.ll
…
crash.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
critical-edge-split-2.ll
fix PR8642: if a critical edge has a PHI value that can trap,
2010-12-19 04:58:57 +00:00
cstring.ll
…
ctpop-combine.ll
Relax expressions and add explicit triplets -linux and -win32.
2011-02-22 07:19:46 +00:00
dag-rauw-cse.ll
FileCheck-ize
2011-07-02 20:42:36 +00:00
dagcombine-buildvector.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
dagcombine-cse.ll
…
darwin-bzero.ll
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is
2011-06-18 06:05:24 +00:00
darwin-no-dead-strip.ll
…
darwin-quote.ll
…
darwin-stub.ll
…
dbg-at-specficiation.ll
Do not unnecessarily use AT_specification DIE because it does not add any value.
2011-09-21 23:41:11 +00:00
dbg-byval-parameter.ll
…
dbg-const-int.ll
Select DW_AT_const_value size based on variable size.
2011-05-27 16:45:18 +00:00
dbg-const.ll
Do not drop constant values when a variable's content is described using .debug_loc entries.
2011-06-01 22:03:25 +00:00
dbg-declare-arg.ll
A dbg.declare may not be in entry block, even if it is referring to an incoming argument. However, It is appropriate to emit DBG_VALUE referring to this incoming argument in entry block in MachineFunction.
2011-04-25 16:33:52 +00:00
dbg-file-name.ll
Don't emit a directory entry for the value in DW_AT_comp_dir, that is always
2011-11-02 20:55:33 +00:00
dbg-i128-const.ll
Dwarf: [PR11022] Fix emitting DW_AT_const_value(>i64), to be host-endian-neutral.
2011-10-28 14:12:22 +00:00
dbg-large-unsigned-const.ll
While emitting constant value, look through derived type and use underlying basic type to determine size and signness of the constant value.
2011-07-20 21:57:04 +00:00
dbg-merge-loc-entry.ll
Don't print an unused label before .cfi_endproc.
2012-01-09 00:17:29 +00:00
dbg-prolog-end.ll
Identify end of prologue (and beginning of function body) using DW_LNS_set_prologue_end line table opcode.
2011-05-11 19:22:19 +00:00
dbg-subrange.ll
DISubrange supports unsigned lower/upper array bounds, so let's not fake it in the end while emitting DWARF. If a FE needs to encode signed lower/upper array bounds then we need to extend DISubrange or ad DISignedSubrange.
2011-11-17 23:43:15 +00:00
dbg-value-dag-combine.ll
While replacing all uses of a SDValue with another value, do not forget to transfer SDDbgValue.
2011-05-23 17:35:08 +00:00
dbg-value-inlined-parameter.ll
Always use the string pool, even when it makes the .o larger. This may help
2011-10-28 05:29:47 +00:00
dbg-value-isel.ll
While extending definition range of a debug variable, consult lexical scopes also. There is no point extending debug variable out side its lexical block. This provides 6x compile time speedup in some cases.
2011-08-10 21:25:34 +00:00
dbg-value-location.ll
Always use the string pool, even when it makes the .o larger. This may help
2011-10-28 05:29:47 +00:00
dbg-value-range.ll
Make size computation less brittle.
2011-05-27 22:05:41 +00:00
dec-eflags-lower.ll
test/CodeGen/X86/dec-eflags-lower.ll: Relax expression for win32 x64.
2011-11-15 22:30:37 +00:00
dg.exp
…
discontiguous-loops.ll
…
div8.ll
Re-commit 131641 with fixes; de-pseudoize MOVSX16rr8 and friends.
2011-05-20 19:04:40 +00:00
divide-by-constant.ll
Be more aggressive about following hints.
2011-07-08 20:46:18 +00:00
divrem.ll
…
dll-linkage.ll
Target/X86/X86FastISel: [PR6275] Fix Win32's dllimport function with fastisel.
2011-02-21 04:50:06 +00:00
dllexport.ll
…
dollar-name.ll
…
dwarf-comp-dir.ll
Don't emit a directory entry for the value in DW_AT_comp_dir, that is always
2011-11-02 20:55:33 +00:00
dyn-stackalloc.ll
test/CodeGen/X86/dyn-stackalloc.ll: FileCheck-ize.
2011-03-16 13:52:08 +00:00
eh_frame.ll
Fix a silly error I introduce in r131951.
2011-06-07 23:26:45 +00:00
empty-functions.ll
Don't print an unused label before .cfi_endproc.
2012-01-09 00:17:29 +00:00
empty-struct-return-type.ll
…
epilogue.ll
…
extend.ll
…
extern_weak.ll
…
extmul64.ll
…
extmul128.ll
…
extract-combine.ll
…
extract-extract.ll
…
extractelement-from-arg.ll
…
extractelement-load.ll
Make sure this DAGCombine actually returns an UNDEF of the correct type; PR10476.
2011-07-25 22:25:42 +00:00
extractelement-shuffle.ll
…
extractps.ll
…
f16c-intrinsics.ll
Add intrinsics for X86 vcvtps2ph and vcvtph2ps instructions
2011-11-04 06:59:21 +00:00
fabs.ll
…
fast-cc-callee-pops.ll
…
fast-cc-merge-stack-adj.ll
…
fast-cc-pass-in-regs.ll
…
fast-isel-agg-constant.ll
PR10077: fix fast-isel of extractvalue of aggregate constants.
2011-06-06 05:46:34 +00:00
fast-isel-atomic.ll
Convert a bunch more tests over to the new atomic instructions.
2011-09-26 23:15:09 +00:00
fast-isel-avoid-unnecessary-pic-base.ll
…
fast-isel-bail.ll
Remove support for parsing the "type i32" syntax for defining a numbered
2011-06-19 00:03:46 +00:00
fast-isel-bc.ll
Eliminate more linear scan tests.
2011-11-12 20:35:26 +00:00
fast-isel-call.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
fast-isel-cmp-branch.ll
Update more tests to the new EH scheme.
2011-08-31 21:39:05 +00:00
fast-isel-constpool.ll
…
fast-isel-extract.ll
Add x86 fast-isel for calls returning first-class aggregates. rdar://9435872.
2011-05-17 18:29:03 +00:00
fast-isel-fneg.ll
Force a triple on a couple of tests; we don't support fast-isel of ret on Win64.
2011-05-18 17:16:37 +00:00
fast-isel-gep.ll
Disable expensive two-address optimizations at -O0. rdar://10453055
2011-11-16 18:44:48 +00:00
fast-isel-gv.ll
…
fast-isel-i1.ll
Fix test on Windows.
2011-05-26 18:00:32 +00:00
fast-isel-mem.ll
…
fast-isel-ret-ext.ll
Add fast-isel support for zeroext and signext ret instructions on x86.
2011-05-19 22:16:13 +00:00
fast-isel-tailcall.ll
…
fast-isel-tls.ll
PR10991: make fast-isel correctly check whether accessing a global through an alias involves thread-local storage. (I'm not entirely sure how this is supposed to work, but this patch makes fast-isel consistent with the normal isel path.)
2011-09-22 23:41:28 +00:00
fast-isel-x86-64.ll
Make X86::FsFLD0SS / FsFLD0SD real pseudo-instructions.
2011-11-29 22:27:25 +00:00
fast-isel-x86.ll
fast-isel sret calls, try 2. We actually do need to do something on x86-32. rdar://problem/9303592 .
2011-04-28 20:19:12 +00:00
fast-isel.ll
Force a triple on a couple of tests; we don't support fast-isel of ret on Win64.
2011-05-18 17:16:37 +00:00
fastcall-correct-mangling.ll
…
fastcc3struct.ll
…
fastcc-2.ll
…
fastcc-byval.ll
…
fastcc-sret.ll
…
fastcc.ll
…
field-extract-use-trunc.ll
…
fildll.ll
…
fltused.ll
…
fma4-intrinsics-x86_64.ll
Change FMA4 memory forms to use memopv* instead of alignedloadv*. No need to force alignment on these instructions. Add a couple testcases for memory forms.
2011-12-30 02:18:36 +00:00
fma.ll
Add an intrinsic and codegen support for fused multiply-accumulate. The intent
2011-07-08 21:39:21 +00:00
fmul-zero.ll
…
fold-add.ll
FileCheck-ize
2011-07-02 20:42:33 +00:00
fold-and-shift.ll
Teach the X86 instruction selection to do some heroic transforms to
2012-01-11 08:41:08 +00:00
fold-call-2.ll
…
fold-call-3.ll
…
fold-call.ll
…
fold-imm.ll
…
fold-load.ll
…
fold-mul-lohi.ll
test/CodeGen/X86: FileCheck-ize and add explicit -mtriple=x86_64-linux. They are useless to Win64 target.
2011-03-16 13:52:38 +00:00
fold-pcmpeqd-0.ll
Use a bigger hammer to fix PR11314 by disabling the "forcing two-address
2011-11-10 07:43:16 +00:00
fold-pcmpeqd-1.ll
…
fold-pcmpeqd-2.ll
Eliminate more linear scan tests.
2011-11-12 20:35:26 +00:00
fold-sext-trunc.ll
Remove support for parsing the "type i32" syntax for defining a numbered
2011-06-19 00:03:46 +00:00
fold-xmm-zero.ll
Add -mattr=+sse2 to make the buildbots happy.
2011-06-10 08:26:26 +00:00
fold-zext-trunc.ll
PR9055: extend the fix to PR4050 (r70179) to apply to zext and anyext.
2011-04-16 23:25:34 +00:00
force-align-stack.ll
…
fp2sint.ll
…
fp_constant_op.ll
…
fp_load_cast_fold.ll
…
fp_load_fold.ll
…
fp-elim.ll
…
fp-immediate-shorten.ll
…
fp-in-intregs.ll
test/CodeGen/X86/fp-in-intregs.ll: FileCheck-ize.
2010-12-29 03:58:36 +00:00
fp-stack-2results.ll
Land the long talked about "type system rewrite" patch. This
2011-07-09 17:41:24 +00:00
fp-stack-compare.ll
Have the X86 back-end emit the alias instead of what's being aliased. In most
2011-04-14 01:11:51 +00:00
fp-stack-direct-ret.ll
…
fp-stack-O0-crash.ll
Handle IMPLICIT_DEF instructions in X86FloatingPoint.
2011-08-03 16:33:19 +00:00
fp-stack-O0.ll
Fix APFloat::convert so that it handles narrowing conversions correctly; it
2011-11-26 03:38:02 +00:00
fp-stack-ret-conv.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
fp-stack-ret-store.ll
…
fp-stack-ret.ll
FileCheckize a couple of tests.
2011-06-28 06:25:03 +00:00
fp-stack-retcopy.ll
…
fp-stack-set-st1.ll
…
fp-stack.ll
…
fp-trunc.ll
Make tests more useful.
2011-04-25 10:12:01 +00:00
fsgsbase.ll
Test case for X86 FS/GS Base intrinsics
2011-10-31 02:15:47 +00:00
fsxor-alignment.ll
…
full-lsr.ll
…
ga-offset.ll
…
gather-addresses.ll
Relax expressions and add explicit triplets -linux and -win32.
2011-02-22 07:20:10 +00:00
ghc-cc64.ll
…
ghc-cc.ll
…
global-sections-tls.ll
Be nice to Xcore and the XMOS assembler and avoid quoting section names
2011-03-04 20:03:14 +00:00
global-sections.ll
Fix an obvious typo that's preventing x86 (32-bit) from using .literal16.
2011-07-19 23:14:32 +00:00
h-register-addressing-32.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
h-register-addressing-64.ll
…
h-register-store.ll
test/CodeGen/X86: Add a pattern for Win64.
2011-03-16 13:52:51 +00:00
h-registers-0.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
h-registers-1.ll
test/CodeGen/X86/h-registers-1.ll: Add explicit -mtriple=x86_64-linux. It does not need to be checked on x86_64-win32 (aka Win64).
2011-03-17 04:24:40 +00:00
h-registers-2.ll
FileCheck-ize a test, avoiding a temporary file.
2011-07-02 20:42:31 +00:00
h-registers-3.ll
…
haddsub.ll
Add instruction selection support for horizontal add/sub of 256-bit floating point vectors. Also add the test case for 256-bit integer vectors.
2011-12-02 07:16:01 +00:00
hidden-vis-2.ll
…
hidden-vis-3.ll
…
hidden-vis-4.ll
…
hidden-vis-pic.ll
Replace the -unwind-tables option with a per function flag. This is more
2011-05-25 03:44:17 +00:00
hidden-vis.ll
PR10989: Don't print .hidden on Windows.
2011-09-23 00:13:02 +00:00
hoist-common.ll
Add target triple so test doesn't fail on Windows machines.
2011-05-17 00:15:58 +00:00
hoist-invariant-load.ll
Adding test for machine-licm operating on invariant load instructions
2011-11-08 19:06:53 +00:00
i2k.ll
…
i64-mem-copy.ll
test/CodeGen/X86: FileCheck-ize and add actions for x86_64-linux and x86_64-win32.
2011-03-16 13:53:07 +00:00
i128-and-beyond.ll
…
i128-immediate.ll
…
i128-mul.ll
…
i128-ret.ll
Relax expressions and add explicit triplets -linux and -win32.
2011-02-22 07:20:18 +00:00
i128-sdiv.ll
Don't crash on 128-bit sdiv by constant. Found by inspection.
2011-10-27 02:06:39 +00:00
i256-add.ll
…
iabs.ll
Catch more cases where 2-address pass should 3-addressify instructions. rdar://9002648.
2011-03-02 01:08:17 +00:00
illegal-insert.ll
…
illegal-vector-args-return.ll
…
imul-lea-2.ll
…
imul-lea.ll
…
inline-asm-2addr.ll
…
inline-asm-error.ll
Better diagnostics when inline asm fails to allocate.
2011-07-02 07:17:37 +00:00
inline-asm-flag-clobber.ll
…
inline-asm-fpstack.ll
Don't clobber pending ST regs when FP regs are killed.
2011-08-08 17:15:43 +00:00
inline-asm-h.ll
Fix darwin bots.
2011-01-01 21:58:41 +00:00
inline-asm-modifier-n.ll
…
inline-asm-mrv.ll
remove asmparser support for the old getresult instruction, which has been subsumed by extractvalue.
2011-06-17 06:57:15 +00:00
inline-asm-out-regs.ll
…
inline-asm-pic.ll
…
inline-asm-ptr-cast.ll
…
inline-asm-q-regs.ll
Make the 'x' constraint work for AVX registers as well.
2012-01-07 01:02:09 +00:00
inline-asm-R-constraint.ll
…
inline-asm-tied.ll
Switch a couple -O0 tests to RABasic.
2011-11-12 20:11:04 +00:00
inline-asm-x-scalar.ll
…
inline-asm.ll
Check register class matching instead of width of type matching
2011-07-14 20:13:52 +00:00
ins_subreg_coalesce-1.ll
…
ins_subreg_coalesce-2.ll
…
ins_subreg_coalesce-3.ll
…
insert-positions.ll
…
insertelement-copytoregs.ll
…
insertelement-legalize.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
int-intrinsic.ll
…
invalid-shift-immediate.ll
…
isel-sink2.ll
…
isel-sink3.ll
Fix register-dependent X86 tests.
2011-04-05 00:32:44 +00:00
isel-sink.ll
FileCheck-ize and simplify this test.
2011-07-02 20:42:28 +00:00
isint.ll
Revert r144611 and r144613.
2011-11-15 07:13:03 +00:00
isnan2.ll
…
isnan.ll
…
ispositive.ll
…
iv-users-in-other-loops.ll
Teach two-address pass to re-schedule two-address instructions (or the kill
2011-11-14 19:48:55 +00:00
jump_sign.ll
test/CodeGen/X86/jump_sign.ll: Add -mcpu=pentiumpro for non-x86 hosts. It uses "cmov".
2012-01-04 03:52:23 +00:00
label-redefinition.ll
…
large-gep-scale.ll
…
ldzero.ll
…
lea-2.ll
…
lea-3.ll
Remove working entry from README.
2011-04-17 02:36:27 +00:00
lea-4.ll
…
lea-recursion.ll
…
lea.ll
Relax expressions and add explicit triplets -linux and -win32.
2011-02-22 07:20:26 +00:00
leaf-fp-elim.ll
…
legalize-fmp-oeq-vector-select.ll
…
legalize-libcalls.ll
Reapply r143206, with fixes. Disallow physical register lifetimes
2011-11-03 21:49:52 +00:00
legalize-sub-zero-2.ll
The change for PR9190 wasn't quite right. We need to avoid making the
2011-02-16 01:10:03 +00:00
legalize-sub-zero.ll
Add testcase for PR9190.
2011-02-16 01:08:31 +00:00
legalizedag_vec.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
lfence.ll
Last batch of test conversions to new atomic instructions.
2011-09-27 00:17:29 +00:00
licm-dominance.ll
Add dominance check for the instruction being hoisted.
2011-10-11 18:09:58 +00:00
licm-nested.ll
Make this test more specific. There are 3 stats that matched "machine-licm".
2011-10-11 23:34:31 +00:00
licm-symbol.ll
Use pushq / popq instead of subq $8, %rsp / addq $8, %rsp to adjust stack in
2011-01-03 22:53:22 +00:00
limited-prec.ll
…
live-out-reg-info.ll
…
liveness-local-regalloc.ll
…
lock-inst-encoding.ll
Convert a bunch more tests over to the new atomic instructions.
2011-09-26 23:15:09 +00:00
long-setcc.ll
Revert r128175.
2011-03-23 23:11:02 +00:00
longlong-deadload.ll
FileCheck-ize a test that had the strangest TCL quote I've seen yet: an
2011-06-28 02:03:10 +00:00
loop-blocks.ll
Re-apply r124518 with fix. Watch out for invalidated iterator.
2011-01-29 04:46:23 +00:00
loop-hoist.ll
…
loop-strength-reduce2.ll
FileCheck-ize
2011-07-02 20:42:25 +00:00
loop-strength-reduce3.ll
LSR: rewrite inner loops only.
2011-09-29 01:33:38 +00:00
loop-strength-reduce4.ll
Fix register-dependent X86 tests.
2011-04-05 00:32:44 +00:00
loop-strength-reduce5.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
loop-strength-reduce6.ll
…
loop-strength-reduce7.ll
…
loop-strength-reduce8.ll
…
loop-strength-reduce-2.ll
…
loop-strength-reduce-3.ll
…
loop-strength-reduce.ll
…
lsr-delayed-fold.ll
…
lsr-i386.ll
…
lsr-interesting-step.ll
Make tests more useful.
2011-04-25 10:12:01 +00:00
lsr-loop-exit-cond.ll
test/CodeGen/X86/lsr-loop-exit-cond.ll: Try to appease linux and freebsd bots to specify explicit -mtriple=x86_64-darwin.
2011-11-10 14:18:59 +00:00
lsr-negative-stride.ll
…
lsr-nonaffine.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
lsr-normalization.ll
…
lsr-overflow.ll
When checking for signed multiplication overflow, watch out for INT_MIN and -1.
2011-05-23 21:07:39 +00:00
lsr-quadratic-expand.ll
Don't re-use existing addrec expansions if they contain casts.
2011-03-02 01:34:10 +00:00
lsr-redundant-addressing.ll
FileCheck-ize another codegen test.
2011-07-02 20:42:22 +00:00
lsr-reuse-trunc.ll
Teach two-address pass to re-schedule two-address instructions (or the kill
2011-11-14 19:48:55 +00:00
lsr-reuse.ll
Turn on list-ilp scheduling by default on x86 and x86-64, fix up
2011-03-08 02:42:25 +00:00
lsr-sort.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
lsr-static-addr.ll
…
lsr-wrap.ll
…
lzcnt.ll
Use standard promotion for i8 CTTZ nodes and i8 CTLZ nodes when the
2011-12-24 12:12:34 +00:00
machine-cp.ll
Added a late machine instruction copy propagation pass. This catches
2012-01-07 03:02:36 +00:00
machine-cse.ll
Allow machine-cse to look across MBB boundary when cse'ing instructions that
2012-01-10 02:02:58 +00:00
MachineSink-CritEdge.ll
…
MachineSink-DbgValue.ll
While sinking machine instructions, sink matching DBG_VALUEs also otherwise live debug variable pass will drop DBG_VALUEs on the floor.
2011-09-07 00:07:58 +00:00
MachineSink-eflags.ll
Update live-in lists when splitting critical edges.
2011-10-14 17:25:46 +00:00
MachineSink-PHIUse.ll
Roll r127459 back in:
2011-03-11 21:52:04 +00:00
masked-iv-safe.ll
Use a bigger hammer to fix PR11314 by disabling the "forcing two-address
2011-11-10 07:43:16 +00:00
masked-iv-unsafe.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
maskmovdqu.ll
…
mcinst-avx-lowering.ll
Add test case for r147881.
2012-01-10 23:09:53 +00:00
mcinst-lowering.ll
…
mem-promote-integers.ll
Reduce the runtime of the test. Keep only the interesting cases.
2011-06-19 08:12:43 +00:00
membarrier.ll
Last batch of test conversions to new atomic instructions.
2011-09-27 00:17:29 +00:00
memcmp.ll
Relax expressions and add explicit triplets -linux and -win32.
2011-02-22 07:20:52 +00:00
memcpy-2.ll
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is
2011-06-18 06:05:24 +00:00
memcpy.ll
Add Mode64Bit feature and sink it down to MC layer.
2011-07-07 21:06:52 +00:00
memset64-on-x86-32.ll
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is
2011-06-18 06:05:24 +00:00
memset-2.ll
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is
2011-06-18 06:05:24 +00:00
memset-3.ll
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is
2011-06-18 06:05:24 +00:00
memset.ll
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is
2011-06-18 06:05:24 +00:00
mfence.ll
Last batch of test conversions to new atomic instructions.
2011-09-27 00:17:29 +00:00
mingw-alloca.ll
…
misaligned-memset.ll
…
mmx-arg-passing2.ll
…
mmx-arg-passing.ll
…
mmx-arith.ll
…
mmx-bitcast-to-i64.ll
…
mmx-builtins.ll
Don't disable MMX support when AVX is enabled. Fix predicates for MMX instructions that were added along with SSE instructions to check for AVX in addition to SSE level.
2012-01-09 00:11:29 +00:00
mmx-copy-gprs.ll
Fix register-dependent X86 tests.
2011-04-05 00:32:44 +00:00
mmx-emms.ll
…
mmx-insert-element.ll
…
mmx-pinsrw.ll
stabalize tests by specifying the exact sse level
2011-10-17 19:45:38 +00:00
mmx-punpckhdq.ll
…
mmx-s2v.ll
…
mmx-shift.ll
…
mmx-shuffle.ll
Remove support for using "foo" as symbols instead of %"foo". This is ancient
2011-06-17 06:36:20 +00:00
movbe.ll
test/CodeGen/X86/movbe.ll: Give explicit -mtriple=x86_64-linux, to unbreak win32 hosts.
2011-10-11 03:41:03 +00:00
movfs.ll
…
movgs.ll
Attempt to fix -mtriple=i686-{cygwin|mingw|win32} regressions. Nakamura,
2011-09-20 00:08:12 +00:00
movmsk.ll
Don't try to form FGETSIGN after legalization; it is possible in some cases, but the existing code can't do it correctly. PR11570.
2011-12-15 02:07:20 +00:00
movntdq-no-avx.ll
Don't produce a vmovntdq if we don't have AVX support.
2011-05-14 00:30:01 +00:00
mul64.ll
…
mul128.ll
…
mul-legalize.ll
…
mul-remat.ll
…
mul-shift-reassoc.ll
…
muloti.ll
Fix UMULO support for 2x register width to allow the full
2011-06-18 00:09:57 +00:00
mult-alt-generic-i686.ll
…
mult-alt-generic-x86_64.ll
…
mult-alt-x86.ll
…
multiple-loop-post-inc.ll
Use a bigger hammer to fix PR11314 by disabling the "forcing two-address
2011-11-10 07:43:16 +00:00
multiple-return-values-cross-block.ll
remove asmparser support for the old getresult instruction, which has been subsumed by extractvalue.
2011-06-17 06:57:15 +00:00
nancvt.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
narrow_op-1.ll
…
narrow-shl-cst.ll
X86: Try to use a smaller encoding by transforming (X << C1) & C2 into (X & (C2 >> C1)) & C1. (Part of PR5039)
2011-04-22 15:30:40 +00:00
narrow-shl-load.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
neg_fp.ll
…
neg-shl-add.ll
…
negate-add-zero.ll
…
negative_zero.ll
…
negative-sin.ll
test/CodeGen/X86/negative-sin.ll: FileCheck-ize.
2010-12-29 03:58:47 +00:00
negative-stride-fptosi-user.ll
…
negative-subscript.ll
…
no-cfi.ll
Convert tests to the new EH model.
2011-11-08 00:09:27 +00:00
nobt.ll
…
non-lazy-bind.ll
Add a new function attribute, nonlazybind, which inhibits lazy-loading
2011-06-15 20:36:13 +00:00
nontemporal.ll
Kill and collapse outstanding DomainValues.
2011-11-07 23:08:21 +00:00
norex-subreg.ll
Add TEST8ri_NOREX pseudo to constrain sub_8bit_hi copies.
2011-10-08 18:28:28 +00:00
nosse-error1.ll
…
nosse-error2.ll
…
nosse-varargs.ll
…
object-size.ll
Switch a couple -O0 tests to RABasic.
2011-11-12 20:11:04 +00:00
opt-ext-uses.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
opt-shuff-tstore.ll
test/CodeGen/X86/opt-shuff-tstore.ll: Add explicit -mtriple=x86_64-linux.
2011-08-10 22:52:48 +00:00
optimize-max-0.ll
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is
2011-06-18 06:05:24 +00:00
optimize-max-1.ll
…
optimize-max-2.ll
…
optimize-max-3.ll
Prepare remaining tests for -join-physreg going away.
2011-05-04 23:54:59 +00:00
or-address.ll
Pseudo CMOV instructions don't clobber EFLAGS.
2011-09-02 23:52:55 +00:00
or-branch.ll
…
overlap-shift.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
packed_struct.ll
…
palignr-2.ll
…
palignr.ll
Really un-XFAIL the testcase, like I said I would in r139458.
2011-09-10 02:02:27 +00:00
peep-setb.ll
X86: Add a bunch of peeps for add and sub of SETB.
2011-05-08 18:36:07 +00:00
peep-test-0.ll
…
peep-test-1.ll
…
peep-test-2.ll
…
peep-test-3.ll
Be more aggressive about following hints.
2011-07-08 20:46:18 +00:00
peep-vector-extract-concat.ll
test/CodeGen/X86: Add a pattern for Win64.
2011-03-16 13:52:51 +00:00
peep-vector-extract-insert.ll
Make X86::FsFLD0SS / FsFLD0SD real pseudo-instructions.
2011-11-29 22:27:25 +00:00
personality.ll
Remove old declare statements.
2011-08-31 21:41:20 +00:00
phaddsub.ll
Test cases for SSSE3/AVX integer horizontal add/sub.
2011-11-19 09:03:33 +00:00
phi-bit-propagation.ll
Fix PR9324 / <rdar://problem/9052489> by handling the case where a PHI has no uses.
2011-02-27 08:06:01 +00:00
phi-immediate-factoring.ll
…
phys_subreg_coalesce-2.ll
Prepare remaining tests for -join-physreg going away.
2011-05-04 23:54:59 +00:00
phys_subreg_coalesce-3.ll
Prepare remaining tests for -join-physreg going away.
2011-05-04 23:54:59 +00:00
phys_subreg_coalesce.ll
…
phys-reg-local-regalloc.ll
…
pic_jumptable.ll
Partially FileCheck-ize a test to remove a weird quoting situation.
2011-07-02 20:42:20 +00:00
pic-load-remat.ll
…
pic.ll
Be careful about scheduling nodes above previous calls. It increase usages of
2011-04-26 21:31:35 +00:00
pmul.ll
Prepare remaining tests for -join-physreg going away.
2011-05-04 23:54:59 +00:00
pmulld.ll
test/CodeGen/X86: Add a pattern for Win64.
2011-03-16 13:52:51 +00:00
pointer-vector.ll
test/CodeGen/X86/pointer-vector.ll: Add explicit -mtriple=i686-linux.
2011-12-05 07:54:57 +00:00
popcnt.ll
…
postalloc-coalescing.ll
…
postra-licm.ll
Fix register-dependent X86 tests.
2011-04-05 00:32:44 +00:00
powi.ll
…
pr1462.ll
…
pr1489.ll
…
pr1505.ll
…
pr1505b.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
pr2177.ll
…
pr2182.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
pr2326.ll
…
pr2656.ll
…
pr2659.ll
Prepare remaining tests for -join-physreg going away.
2011-05-04 23:54:59 +00:00
pr2849.ll
…
pr2924.ll
…
pr2982.ll
…
pr3154.ll
…
pr3216.ll
FileCheck-ize another test, and upgrade its syntax a bit.
2011-07-02 20:42:17 +00:00
pr3241.ll
…
pr3243.ll
…
pr3244.ll
…
pr3250.ll
…
pr3317.ll
Land the long talked about "type system rewrite" patch. This
2011-07-09 17:41:24 +00:00
pr3366.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
pr3457.ll
…
pr3495.ll
Eliminate more linear scan tests.
2011-11-12 20:35:26 +00:00
pr3522.ll
Update more tests to the new EH scheme.
2011-08-31 21:39:05 +00:00
pr7882.ll
…
pr9127.ll
Reapply 132424 with fixes. This fixes PR10068.
2011-06-03 23:53:54 +00:00
pr9743.ll
Use the dwarf->llvm mapping to print register names in the cfi
2011-05-30 20:20:15 +00:00
pr10068.ll
Add test for PR10068.
2011-06-02 20:02:48 +00:00
pr10420.ll
Assume .cfi_startproc is the first thing in a function. If the function is
2011-08-02 20:24:22 +00:00
pr11202.ll
Don't print a label before .cfi_startproc when we don't need to. This makes
2012-01-07 22:42:19 +00:00
pr11415.ll
Add triple to the test.
2011-11-22 06:36:25 +00:00
prefetch.ll
Add test for r146163.
2011-12-08 19:21:39 +00:00
private-2.ll
…
private.ll
make the asmparser reject function and type redefinitions. 'Merging' hasn't been
2011-06-17 07:06:44 +00:00
promote-assert-zext.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
promote-i16.ll
Commit the right set of tests for r133124. Sorry 'bout that!
2011-06-16 01:35:45 +00:00
promote-trunc.ll
Fix PromoteIntRes_TRUNCATE: Add support for cases where the
2011-06-20 07:15:58 +00:00
promote.ll
Fix incorrect widening of the bitcast sdnode in case the incoming operand is integer-promoted.
2012-01-03 22:12:28 +00:00
ptr-rotate.ll
New pointer rotate test.
2011-07-20 03:09:11 +00:00
ptrtoint-constexpr.ll
…
rdtsc.ll
…
red-zone2.ll
…
red-zone.ll
test/CodeGen/X86/red-zone.ll: Add explicit -mtriple=x86_64-linux.
2011-02-22 07:18:55 +00:00
reghinting.ll
Hoist spills within a basic block.
2011-07-09 00:25:03 +00:00
regpressure.ll
…
rem-2.ll
…
rem.ll
…
remat-constant.ll
…
remat-mov-0.ll
Relax expressions and add explicit triplets -linux and -win32.
2011-02-22 07:21:25 +00:00
remat-scalar-zero.ll
Turn on list-ilp scheduling by default on x86 and x86-64, fix up
2011-03-08 02:42:25 +00:00
ret-addr.ll
…
ret-i64-0.ll
…
ret-mmx.ll
Remove dead code. Fix associated test to use FileCheck.
2011-05-16 21:28:22 +00:00
rip-rel-address.ll
…
rodata-relocs.ll
Update tests.
2011-01-16 18:02:57 +00:00
rot16.ll
…
rot32.ll
…
rot64.ll
…
rotate2.ll
…
rotate.ll
…
rounding-ops.ll
X86: Add patterns for the various rounding ops for SSE4.1 and AVX.
2011-12-09 15:44:03 +00:00
scalar_sse_minmax.ll
…
scalar_widen_div.ll
fix a typo in the test
2011-10-16 20:43:41 +00:00
scalar-extract.ll
…
scalar-min-max-fill-operand.ll
test/CodeGen/X86: FileCheck-ize and add explicit -mtriple=x86_64-linux. They are useless to Win64 target.
2011-03-16 13:52:38 +00:00
scalarize-bitcast.ll
…
scev-interchange.ll
Update more tests to the new EH scheme.
2011-08-31 21:39:05 +00:00
sdiv-exact.ll
Emit a more efficient magic number multiplication for exact sdivs.
2011-07-08 10:31:30 +00:00
segmented-stacks.ll
Check for stack space more intelligently.
2011-12-03 09:32:07 +00:00
select.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
setcc.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
setoeq.ll
Reapply 132424 with fixes. This fixes PR10068.
2011-06-03 23:53:54 +00:00
setuge.ll
…
sext-i1.ll
Revert r128175.
2011-03-23 23:11:02 +00:00
sext-load.ll
…
sext-ret-val.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
sext-subreg.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
sext-trunc.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
sfence.ll
Last batch of test conversions to new atomic instructions.
2011-09-27 00:17:29 +00:00
shift-and.ll
…
shift-coalesce.ll
…
shift-codegen.ll
FileCheck-ize another X86 test, making it more precisely verify the
2011-07-02 20:43:16 +00:00
shift-combine.ll
…
shift-double.ll
…
shift-folding.ll
Prevent a DAGCombine from firing where there are two uses of
2012-01-05 11:05:55 +00:00
shift-i128.ll
…
shift-i256.ll
…
shift-one.ll
…
shift-pair.ll
Change this DAGCombine to build AND of SHR instead of SHR of AND; this matches the ordering we prefer in instcombine. Part of rdar://9562809.
2011-06-09 22:14:44 +00:00
shift-parts.ll
…
shl_elim.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
shl_undef.ll
TargetLoweringOpt is a struct used by DAGCombine, not a pass.
2011-06-15 16:48:02 +00:00
shl-anyext.ll
…
shrink-compare.ll
test/CodeGen/X86/shrink-compare.ll: Relax expressions for Win64.
2011-04-23 00:15:45 +00:00
shrink-fp-const1.ll
…
shrink-fp-const2.ll
…
sibcall-2.ll
…
sibcall-3.ll
…
sibcall-4.ll
…
sibcall-5.ll
…
sibcall-byval.ll
Test case for r133858 (tail call optimize in the presence of byval).
2011-06-25 02:44:56 +00:00
sibcall.ll
make the asmparser reject function and type redefinitions. 'Merging' hasn't been
2011-06-17 07:06:44 +00:00
sincos.ll
…
sink-hoist.ll
Add dominance check for the instruction being hoisted.
2011-10-11 18:09:58 +00:00
small-byval-memcpy.ll
Revert r122955. It seems using movups to lower memcpy can cause massive regression (even on Nehalem) in edge cases. I also didn't see any real performance benefit.
2011-01-07 19:35:30 +00:00
smul-with-overflow.ll
Merge and FileCheckize test cases.
2011-05-21 18:31:48 +00:00
soft-fp.ll
…
splat-scalar-load.ll
Enable stack protectors for all arrays, not just char arrays. rdar://5875909
2011-11-23 07:13:56 +00:00
split-eh-lpad-edges.ll
Update more tests to the new EH scheme.
2011-08-31 21:39:05 +00:00
split-vector-bitcast.ll
Prevent x86-specific DAGCombine from creating nodes with illegal type (which could not be selected). Fixes a minor isel issue that was breaking the testcase from r136130.
2011-07-26 21:02:58 +00:00
split-vector-rem.ll
…
sret.ll
…
sse1.ll
Default 64-bit target features and SSE2 on when a triple specifies x86-64. Clean up all the other hacks which are now unnecessary.
2011-07-08 22:16:47 +00:00
sse2-blend.ll
Revert 147426 because it caused pr11696.
2012-01-03 22:19:42 +00:00
sse2.ll
Use a bigger hammer to fix PR11314 by disabling the "forcing two-address
2011-11-10 07:43:16 +00:00
sse3.ll
Teach two-address pass to re-schedule two-address instructions (or the kill
2011-11-14 19:48:55 +00:00
sse41-blend.ll
Revert 147426 because it caused pr11696.
2012-01-03 22:19:42 +00:00
sse41.ll
improve the setcc -> setcc_carry optimization to happen more
2010-12-19 22:08:31 +00:00
sse42_64.ll
Renamed llvm.x86.sse42.crc32 intrinsics; crc64 doesn't exist.
2011-05-26 23:13:19 +00:00
sse42.ll
Renamed llvm.x86.sse42.crc32 intrinsics; crc64 doesn't exist.
2011-05-26 23:13:19 +00:00
sse_reload_fold.ll
Simplify local live range splitting's safeguard to fix PR10070.
2011-06-06 23:55:20 +00:00
sse-align-0.ll
test/CodeGen/X86: FileCheck-ize and add explicit -mtriple=x86_64-linux. They are useless to Win64 target.
2011-03-16 13:52:38 +00:00
sse-align-1.ll
…
sse-align-2.ll
…
sse-align-3.ll
Kill and collapse outstanding DomainValues.
2011-11-07 23:08:21 +00:00
sse-align-4.ll
…
sse-align-5.ll
…
sse-align-6.ll
…
sse-align-7.ll
test/CodeGen/X86: FileCheck-ize and add explicit -mtriple=x86_64-linux. They are useless to Win64 target.
2011-03-16 13:52:38 +00:00
sse-align-8.ll
…
sse-align-9.ll
…
sse-align-10.ll
…
sse-align-11.ll
Experiment with changing the default 32-bit linux stack alignment to
2011-01-13 06:47:10 +00:00
sse-align-12.ll
…
sse-commute.ll
test/CodeGen/X86: FileCheck-ize and add explicit -mtriple=x86_64-linux. They are useless to Win64 target.
2011-03-16 13:52:38 +00:00
sse-domains.ll
Make X86::FsFLD0SS / FsFLD0SD real pseudo-instructions.
2011-11-29 22:27:25 +00:00
sse-fcopysign.ll
…
sse-load-ret.ll
…
sse-minmax.ll
Make X86::FsFLD0SS / FsFLD0SD real pseudo-instructions.
2011-11-29 22:27:25 +00:00
sse-varargs.ll
…
stack-align2.ll
Default stack alignment for 32bit x86 should be 4 Bytes, not 8 Bytes.
2012-01-10 22:43:53 +00:00
stack-align.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
stack-protector-linux.ll
…
stdarg.ll
test/CodeGen/X86: FileCheck-ize and add explicit -mtriple=x86_64-linux. They are useless to Win64 target.
2011-03-16 13:52:38 +00:00
stdcall-notailcall.ll
…
stdcall.ll
…
store_op_load_fold2.ll
…
store_op_load_fold.ll
…
store-empty-member.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
store-fp-constant.ll
…
store-global-address.ll
…
store-narrow.ll
…
storetrunc-fp.ll
…
stride-nine-with-base-reg.ll
test/CodeGen/X86: FileCheck-ize and add explicit -mtriple=x86_64-linux. They are useless to Win64 target.
2011-03-16 13:52:38 +00:00
stride-reuse.ll
test/CodeGen/X86: FileCheck-ize and add explicit -mtriple=x86_64-linux. They are useless to Win64 target.
2011-03-16 13:52:38 +00:00
sub-with-overflow.ll
llvm/test/CodeGen/X86/sub-with-overflow.ll: Add explicit -mtriple=i686-linux.
2012-01-11 07:34:14 +00:00
sub.ll
Update test.
2011-07-26 22:45:39 +00:00
subreg-to-reg-0.ll
…
subreg-to-reg-1.ll
…
subreg-to-reg-2.ll
…
subreg-to-reg-3.ll
…
subreg-to-reg-4.ll
…
subreg-to-reg-6.ll
…
switch-bt.ll
Don't emit a bit test if there is only one case the test can yield false. A simple SETNE is sufficient.
2011-07-14 01:38:42 +00:00
switch-crit-edge-constant.ll
…
switch-or.ll
…
switch-zextload.ll
…
SwitchLowering.ll
…
swizzle.ll
…
tail-call-got.ll
This test requires sse, otherwise x87 ops will block tailcall optimization
2011-08-31 16:49:05 +00:00
tail-dup-addr.ll
Don't print a label before .cfi_startproc when we don't need to. This makes
2012-01-07 22:42:19 +00:00
tail-opts.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
tail-threshold.ll
Add triple.
2011-06-14 23:47:36 +00:00
tailcall1.ll
…
tailcall-fastisel.ll
…
tailcall-i1.ll
…
tailcall-largecode.ll
Use pushq / popq instead of subq $8, %rsp / addq $8, %rsp to adjust stack in
2011-01-03 22:53:22 +00:00
tailcall-returndup-void.ll
Do early taildup of ret in CodeGenPrepare for potential tail calls that have a
2011-03-24 04:52:10 +00:00
tailcall-ri64.ll
Target/X86: Tweak win64's tailcall.
2011-01-26 02:04:09 +00:00
tailcall-stackalign.ll
…
tailcall-structret.ll
…
tailcall-void.ll
…
tailcallbyval64.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
tailcallbyval.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
tailcallfp2.ll
…
tailcallfp.ll
…
tailcallpic1.ll
…
tailcallpic2.ll
…
tailcallstack64.ll
Fix some tests that depend on register allocation.
2011-06-03 22:45:21 +00:00
test-nofold.ll
In the pre-RA scheduler, maintain cmp+br proximity.
2011-04-14 05:15:06 +00:00
test-shrink-bug.ll
…
test-shrink.ll
Relax expressions and add explicit triplets -linux and -win32.
2011-02-22 07:21:33 +00:00
testl-commute.ll
FileCheck-ize another test, making it much more precise for testing the
2011-07-02 20:42:11 +00:00
tls1.ll
…
tls2.ll
…
tls3.ll
…
tls4.ll
…
tls5.ll
…
tls6.ll
…
tls7.ll
…
tls8.ll
…
tls9.ll
…
tls10.ll
…
tls11.ll
…
tls12.ll
…
tls13.ll
…
tls14.ll
…
tls15.ll
…
tls-pic.ll
…
tlv-1.ll
Fix TLS lowering bug. The CopyFromReg must be glued to the TLSCALL. rdar://10291355
2011-10-19 22:22:54 +00:00
tlv-2.ll
Use pushq / popq instead of subq $8, %rsp / addq $8, %rsp to adjust stack in
2011-01-03 22:53:22 +00:00
trap.ll
…
trunc-ext-ld-st.ll
When promoting integer vectors we often create ext-loads. This patch adds a
2011-09-18 10:39:32 +00:00
trunc-to-bool.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
twoaddr-coalesce-2.ll
…
twoaddr-coalesce.ll
…
twoaddr-lea.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
twoaddr-pass-sink.ll
…
twoaddr-sink-terminator.ll
PR10998: It is not legal to sink an instruction past the terminator of a block; make sure we don't do that.
2011-09-23 22:41:57 +00:00
uint64-to-float.ll
Revert r144611 and r144613.
2011-11-15 07:13:03 +00:00
uint_to_fp-2.ll
Explicitly zero out parts of a vector which are required to be zero by the algorithm in LowerUINT_TO_FP_i32. This only has a substantial effect on the generated code when the input is extracted from a vector register; other ways of loading an i32 do the appropriate zeroing implicitly. Fixes PR10802.
2011-08-29 21:15:46 +00:00
uint_to_fp.ll
…
umul-with-carry.ll
…
umul-with-overflow.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
unaligned-load.ll
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is
2011-06-18 06:05:24 +00:00
undef-label.ll
Add a test for the recent regression.
2011-06-14 20:38:50 +00:00
unknown-location.ll
Emit less labels for debug info and stop emitting .loc directives for DBG_VALUEs.
2011-03-25 17:20:59 +00:00
unreachable-loop-sinking.ll
…
unreachable-stack-protector.ll
Fix the GCC test suite issue exposed by r127477, which was caused by stack
2011-03-11 21:51:56 +00:00
urem-i8-constant.ll
…
use-add-flags.ll
Fix a batch of x86 tests to be coalescer independent.
2011-05-04 23:54:51 +00:00
v2f32.ll
Attempt to fix -mtriple=i686-{cygwin|mingw|win32} regressions. Nakamura,
2011-09-20 00:08:12 +00:00
v4f32-immediate.ll
…
v-binop-widen2.ll
…
v-binop-widen.ll
Change the x86 32-bit scheduler to register pressure and fix up the
2011-03-11 01:05:58 +00:00
vararg_tailcall.ll
Don't attempt to tail call optimize for Win64.
2011-05-20 00:59:28 +00:00
variable-sized-darwin-bzero.ll
rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is
2011-06-18 06:05:24 +00:00
variadic-node-pic.ll
…
vec_add.ll
…
vec_align.ll
…
vec_anyext.ll
Enable support for vector sext and trunc:
2011-02-24 21:01:34 +00:00
vec_call.ll
…
vec_cast.ll
test/CodeGen/X86/vec_cast.ll: [PR8311] Add explicit -mtriple=x86_64-linux and -mtriple=x86_64-win32. Thanks to Nadav, it might be fixed in r126424.
2011-03-05 02:38:02 +00:00
vec_clear.ll
…
vec_compare-2.ll
test/CodeGen/X86/vec_compare-2.ll: Add explicit -mtriple=i686-linux.
2011-12-08 15:24:09 +00:00
vec_compare-sse4.ll
PR11033: Make sure we don't generate PCMPGTQ and PCMPEQQ if the target CPU does not support them.
2011-09-28 21:00:25 +00:00
vec_compare.ll
…
vec_ctbits.ll
Manually upgrade the test suite to specify the flag to cttz and ctlz.
2011-12-12 11:59:10 +00:00
vec_ext_inreg.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
vec_extract-sse4.ll
Make sure to replace the chain properly when DAGCombining a LOAD+EXTRACT_VECTOR_ELT into a single LOAD. Fixes PR10747/PR11393.
2011-11-16 23:50:22 +00:00
vec_extract.ll
Fixes a bug in the DAGCombiner. LoadSDNodes have two values (data, chain).
2011-05-11 14:40:50 +00:00
vec_fneg.ll
…
vec_fpext.ll
Fix type-checking for load transformation which is not legal on floating-point types. PR11674.
2011-12-28 21:24:44 +00:00
vec_i64.ll
…
vec_ins_extract-1.ll
…
vec_ins_extract.ll
…
vec_insert-2.ll
FileCheck-ize another test. This one is more clear and runs fewer
2011-07-02 20:42:08 +00:00
vec_insert-3.ll
…
vec_insert-4.ll
…
vec_insert-5.ll
…
vec_insert-6.ll
…
vec_insert-7.ll
…
vec_insert-8.ll
…
vec_insert-9.ll
…
vec_insert.ll
…
vec_loadsingles.ll
…
vec_logical.ll
…
vec_return.ll
Fix test for Windows as well.
2011-11-07 23:10:43 +00:00
vec_set-2.ll
…
vec_set-3.ll
…
vec_set-4.ll
…
vec_set-5.ll
…
vec_set-6.ll
…
vec_set-7.ll
…
vec_set-8.ll
Revert r133452: "Emit movq for 64-bit register to XMM register moves..."
2011-06-21 17:35:13 +00:00
vec_set-9.ll
Revert r133452: "Emit movq for 64-bit register to XMM register moves..."
2011-06-21 17:35:13 +00:00
vec_set-A.ll
FileCheck-ize a test, no functionality changed.
2011-07-02 20:42:06 +00:00
vec_set-B.ll
…
vec_set-C.ll
Attempt to fix -mtriple=i686-{cygwin|mingw|win32} regressions. Nakamura,
2011-09-20 00:08:12 +00:00
vec_set-D.ll
…
vec_set-E.ll
…
vec_set-F.ll
…
vec_set-G.ll
…
vec_set-H.ll
…
vec_set-I.ll
…
vec_set-J.ll
…
vec_set.ll
…
vec_sext.ll
Enable support for vector sext and trunc:
2011-02-24 21:01:34 +00:00
vec_shift2.ll
…
vec_shift3.ll
…
vec_shift4.ll
…
vec_shift.ll
…
vec_shuffle-11.ll
…
vec_shuffle-14.ll
Revert r133452: "Emit movq for 64-bit register to XMM register moves..."
2011-06-21 17:35:13 +00:00
vec_shuffle-15.ll
…
vec_shuffle-16.ll
Fix a batch of x86 tests to be coalescer independent.
2011-05-04 23:54:51 +00:00
vec_shuffle-17.ll
Revert r133452: "Emit movq for 64-bit register to XMM register moves..."
2011-06-21 17:35:13 +00:00
vec_shuffle-18.ll
…
vec_shuffle-19.ll
…
vec_shuffle-20.ll
…
vec_shuffle-22.ll
…
vec_shuffle-23.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
vec_shuffle-24.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
vec_shuffle-25.ll
…
vec_shuffle-26.ll
…
vec_shuffle-27.ll
Add comment for testcase for 122206
2010-12-20 00:54:26 +00:00
vec_shuffle-28.ll
…
vec_shuffle-30.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
vec_shuffle-31.ll
…
vec_shuffle-34.ll
…
vec_shuffle-35.ll
…
vec_shuffle-36.ll
Move test to Transforms/InstCombine.
2011-05-19 05:53:22 +00:00
vec_shuffle-37.ll
Add a new DAGCombine optimization for BUILD_VECTOR.
2011-10-29 21:23:04 +00:00
vec_shuffle-38.ll
Add another missing pattern. llvm-gcc likes f64 but clang likes i64 so it was generating poor code for some SSE builtins.
2011-11-29 22:48:34 +00:00
vec_shuffle-39.ll
Another missing X86ISD::MOVLPD pattern. rdar://10450317
2011-11-16 22:24:44 +00:00
vec_shuffle.ll
test/CodeGen/X86/vec_shuffle.ll: Add explicit -mtriple=i686-linux. We may see some suboptimal frame (%ebp) emission on certain hosts. Possible [PR11031]
2011-11-08 03:46:25 +00:00
vec_splat-2.ll
…
vec_splat-3.ll
…
vec_splat-4.ll
…
vec_splat.ll
…
vec_ss_load_fold.ll
…
vec_udiv_to_shift.ll
Added missing newline
2011-11-08 00:03:24 +00:00
vec_uint_to_fp.ll
Rename the "sandybridge" subtarget to "corei7-avx", for GCC compatibility.
2011-05-20 15:11:26 +00:00
vec_zero_cse.ll
Expand V_SET0 to xorps by default.
2011-11-07 19:15:58 +00:00
vec_zero-2.ll
…
vec_zero.ll
Expand V_SET0 to xorps by default.
2011-11-07 19:15:58 +00:00
vec_zext.ll
Fix 9267; Add vector zext support.
2011-02-20 12:37:50 +00:00
vec-sign.ll
Just rename the functions, relying on matching a instruction that has the same name as a symbol is way too fragile.
2010-12-18 14:23:57 +00:00
vec-trunc-store.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
vector-gep.ll
Add support for vectors of pointers.
2011-12-05 06:29:09 +00:00
vector-intrinsics.ll
…
vector-rem.ll
…
vector-variable-idx2.ll
Changed SSE4/AVX <2 x i64> extract and insert ops to be Custom lowered
2011-11-14 19:38:42 +00:00
vector-variable-idx.ll
…
vector.ll
Revert "Don't check liveness of unallocatable registers."
2011-07-30 00:57:25 +00:00
vfcmp.ll
…
visibility2.ll
Adding a test case for revision 132825.
2011-06-10 02:44:19 +00:00
visibility.ll
Print visibility info for external variables.
2011-04-05 15:51:32 +00:00
volatile.ll
Upgrade syntax of tests using volatile instructions to use 'load volatile' instead of 'volatile load', which is archaic.
2011-11-27 06:54:59 +00:00
vortex-bug.ll
…
vshift_scalar.ll
…
vshift_split2.ll
…
vshift_split.ll
…
vshift-1.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
vshift-2.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
vshift-3.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
vshift-4.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
vshift-5.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
vsplit-and.ll
Clean the triple, add check lines.
2011-10-17 07:07:51 +00:00
weak.ll
…
wide-integer-fold.ll
…
widen_arith-1.ll
Enable element promotion type legalization by deafault.
2011-10-16 20:31:33 +00:00
widen_arith-2.ll
Enable element promotion type legalization by deafault.
2011-10-16 20:31:33 +00:00
widen_arith-3.ll
Fix a bug in the integer-promotion of bitcast operations on vector types.
2011-12-08 13:10:01 +00:00
widen_arith-4.ll
Revert r141569 and r141576.
2011-10-10 23:18:02 +00:00
widen_arith-5.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
widen_arith-6.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
widen_cast-1.ll
Enable element promotion type legalization by deafault.
2011-10-16 20:31:33 +00:00
widen_cast-2.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
widen_cast-3.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
widen_cast-4.ll
Enable element promotion type legalization by deafault.
2011-10-16 20:31:33 +00:00
widen_cast-5.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
widen_cast-6.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
widen_conv-1.ll
Enable element promotion type legalization by deafault.
2011-10-16 20:31:33 +00:00
widen_conv-2.ll
As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
2011-04-14 01:46:37 +00:00
widen_conv-3.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
widen_conv-4.ll
Enable element promotion type legalization by deafault.
2011-10-16 20:31:33 +00:00
widen_extract-1.ll
now that generic vector types aren't selected onto MMX registers, these
2010-12-19 20:12:58 +00:00
widen_load-0.ll
stabalize tests by specifying the exact sse level
2011-10-17 19:45:38 +00:00
widen_load-1.ll
Revert r145273 and fix in SelectionDAG::InferPtrAlignment() instead.
2011-11-28 22:37:34 +00:00
widen_load-2.ll
Changed test to check for correct load size instead of shift as the shift might change if optimised
2011-10-27 18:15:58 +00:00
widen_shuffle-1.ll
Fix testcase.
2011-11-16 03:03:52 +00:00
win64_alloca_dynalloca.ll
Fix some tests that depend on register allocation.
2011-06-03 22:45:21 +00:00
win64_params.ll
Target/X86: Tweak allocating shadow area (aka home) on Win64. It must be enough for caller to allocate one.
2011-02-05 15:11:32 +00:00
win64_vararg.ll
Target/X86: Tweak va_arg for Win64 not to miss taking va_start when number of fixed args > 4.
2011-03-09 11:33:15 +00:00
win_chkstk.ll
Target/X86: [PR8777][PR8778] Tweak alloca/chkstk for Windows targets.
2011-03-24 07:07:00 +00:00
x86_64-mul-by-const.ll
Cosmetic changes.
2010-12-20 20:10:50 +00:00
x86-64-and-mask.ll
Fix a batch of x86 tests to be coalescer independent.
2011-05-04 23:54:51 +00:00
x86-64-arg.ll
…
x86-64-asm.ll
…
x86-64-dead-stack-adjust.ll
…
x86-64-disp.ll
…
x86-64-extend-shift.ll
Don't depend on the physreg coalescing order.
2011-05-04 01:01:47 +00:00
x86-64-frameaddr.ll
…
x86-64-gv-offset.ll
…
x86-64-jumps.ll
…
x86-64-mem.ll
…
x86-64-pic-1.ll
…
x86-64-pic-2.ll
…
x86-64-pic-3.ll
…
x86-64-pic-4.ll
…
x86-64-pic-5.ll
…
x86-64-pic-6.ll
…
x86-64-pic-7.ll
…
x86-64-pic-8.ll
…
x86-64-pic-9.ll
…
x86-64-pic-10.ll
…
x86-64-pic-11.ll
…
x86-64-ret0.ll
…
x86-64-shortint.ll
manually upgrade a bunch of tests to modern syntax, and remove some that
2011-06-17 03:14:27 +00:00
x86-64-sret-return.ll
…
x86-64-tls-1.ll
…
x86-64-varargs.ll
…
x86-frameaddr2.ll
…
x86-frameaddr.ll
…
x86-shifts.ll
Teach the x86 backend a couple tricks for dealing with v16i8 sra by a constant splat value. Fixes PR11289.
2011-11-01 21:18:39 +00:00
x86-store-gv-addr.ll
…
xmm-r64.ll
…
xor-icmp.ll
…
xor.ll
Expand V_SET0 to xorps by default.
2011-11-07 19:15:58 +00:00
zero-remat.ll
Make X86::FsFLD0SS / FsFLD0SD real pseudo-instructions.
2011-11-29 22:27:25 +00:00
zext-extract_subreg.ll
Roll r127459 back in:
2011-03-11 21:52:04 +00:00
zext-fold.ll
llvm/test/CodeGen/X86/zext-fold.ll: Relax an expression in stack offset.
2012-01-11 07:34:22 +00:00
zext-inreg-0.ll
…
zext-inreg-1.ll
…
zext-sext.ll
Turn on list-ilp scheduling by default on x86 and x86-64, fix up
2011-03-08 02:42:25 +00:00
zext-shl.ll
…
zext-trunc.ll
…