llvm-6502/lib/CodeGen
Jakub Staszak 23f34f1c22 Allow getBlockFreq to return 0.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135742 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-22 02:24:57 +00:00
..
AsmPrinter Refactor. 2011-07-20 23:00:27 +00:00
SelectionDAG PR10421: Fix a straightforward bug in the widening logic for CONCAT_VECTORS. 2011-07-20 18:14:33 +00:00
AggressiveAntiDepBreaker.cpp Rename TargetSubtarget to TargetSubtargetInfo for consistency. 2011-07-01 21:01:15 +00:00
AggressiveAntiDepBreaker.h Rename TargetSubtarget to TargetSubtargetInfo for consistency. 2011-07-01 21:01:15 +00:00
AllocationOrder.cpp Rename TRI::getAllocationOrder() to getRawAllocationOrder(). 2011-06-16 23:31:16 +00:00
AllocationOrder.h Get allocation orders from RegisterClassInfo when possible. 2011-06-06 21:02:04 +00:00
Analysis.cpp land David Blaikie's patch to de-constify Type, with a few tweaks. 2011-07-18 04:54:35 +00:00
AntiDepBreaker.h Update DBG_VALUEs while breaking anti dependencies. 2011-06-02 21:26:52 +00:00
BranchFolding.cpp When tail-merging multiple blocks, make sure to correctly update the live-in list on the merged block to correctly account for the live-outs of all the predecessors. They might not be the same in all cases (the testcase I have involves a PHI node where one of the operands is an IMPLICIT_DEF). 2011-07-06 23:41:48 +00:00
BranchFolding.h When tail-merging multiple blocks, make sure to correctly update the live-in list on the merged block to correctly account for the live-outs of all the predecessors. They might not be the same in all cases (the testcase I have involves a PHI node where one of the operands is an IMPLICIT_DEF). 2011-07-06 23:41:48 +00:00
CalcSpillWeights.cpp Don't inflate register classes used by inline asm. 2011-07-01 01:24:25 +00:00
CallingConvLower.cpp Rename the ParmContext enum values to make a bit more sense and add a small 2011-06-10 20:37:36 +00:00
CMakeLists.txt Add MachineBlockFrequency analysis. 2011-07-16 20:23:20 +00:00
CodeGen.cpp Remove the experimental (and unused) pre-ra splitting pass. Greedy regalloc can split live ranges. 2011-06-27 23:40:45 +00:00
CodePlacementOpt.cpp
CriticalAntiDepBreaker.cpp More refactoring. Move getRegClass from TargetOperandInfo to TargetInstrInfo. 2011-06-27 21:26:13 +00:00
CriticalAntiDepBreaker.h Teach antidependency breakers to use RegisterClassInfo. 2011-06-16 21:56:21 +00:00
DeadMachineInstructionElim.cpp Track live-out physical registers in MachineDCE. 2011-06-27 15:00:36 +00:00
DwarfEHPrepare.cpp Convert CallInst and InvokeInst APIs to use ArrayRef. 2011-07-15 08:37:34 +00:00
EdgeBundles.cpp Function::getNumBlockIDs() should be used instead of Function::size() to set the upper limit on the block IDs since basic blocks might get removed (simplified away) after being initially numbered. Plus the test case, in which SelectionDAGBuilder::visitBr() calls llvm::MachineFunction::removeFromMBBNumbering(), which introduces the hole in numbering leading to an assert in llc (prior to the fix). 2011-06-16 00:03:21 +00:00
ELF.h
ELFCodeEmitter.cpp
ELFCodeEmitter.h
ELFWriter.cpp Goodbye TargetAsmInfo. This eliminate last bit of CodeGen and Target in llvm-mc. 2011-07-20 19:50:42 +00:00
ELFWriter.h Fix a FIXME by making GlobalVariable::getInitializer() return a 2011-06-19 18:37:11 +00:00
ExpandISelPseudos.cpp - Rename TargetInstrDesc, TargetOperandInfo to MCInstrDesc and MCOperandInfo and 2011-06-28 19:10:37 +00:00
GCMetadata.cpp
GCMetadataPrinter.cpp
GCStrategy.cpp
IfConversion.cpp Revert patch which broke some IfConversion tests. 2011-07-22 00:55:15 +00:00
InlineSpiller.cpp Fix PR10387. 2011-07-18 05:31:59 +00:00
InterferenceCache.cpp Reapply r135121 with a fixed copy constructor. 2011-07-14 05:35:11 +00:00
InterferenceCache.h Reapply r135121 with a fixed copy constructor. 2011-07-14 05:35:11 +00:00
IntrinsicLowering.cpp land David Blaikie's patch to de-constify Type, with a few tweaks. 2011-07-18 04:54:35 +00:00
LatencyPriorityQueue.cpp
LiveDebugVariables.cpp Typo. 2011-07-06 23:09:51 +00:00
LiveDebugVariables.h Update LiveDebugVariables after live range splitting. 2011-05-06 18:00:02 +00:00
LiveInterval.cpp
LiveIntervalAnalysis.cpp When a physreg is live-in and live through a basic block, make sure its live 2011-04-30 19:12:33 +00:00
LiveIntervalUnion.cpp Be more aggressive about following hints. 2011-07-08 20:46:18 +00:00
LiveIntervalUnion.h Be more aggressive about following hints. 2011-07-08 20:46:18 +00:00
LiveRangeEdit.cpp Fix PR10277. 2011-07-05 15:38:41 +00:00
LiveRangeEdit.h Migrate LLVM and Clang to use the new makeArrayRef(...) functions where previously explicit non-default constructors were used. 2011-07-18 12:00:32 +00:00
LiveStackAnalysis.cpp
LiveVariables.cpp
LLVMTargetMachine.cpp Goodbye TargetAsmInfo. This eliminate last bit of CodeGen and Target in llvm-mc. 2011-07-20 19:50:42 +00:00
LocalStackSlotAllocation.cpp
LowerSubregs.cpp
MachineBasicBlock.cpp - Rename TargetInstrDesc, TargetOperandInfo to MCInstrDesc and MCOperandInfo and 2011-06-28 19:10:37 +00:00
MachineBlockFrequency.cpp Allow getBlockFreq to return 0. 2011-07-22 02:24:57 +00:00
MachineBranchProbabilityInfo.cpp Introduce MachineBranchProbabilityInfo class, which has similar API to 2011-06-16 20:22:37 +00:00
MachineCSE.cpp - Rename TargetInstrDesc, TargetOperandInfo to MCInstrDesc and MCOperandInfo and 2011-06-28 19:10:37 +00:00
MachineDominators.cpp
MachineFunction.cpp land David Blaikie's patch to de-constify Type, with a few tweaks. 2011-07-18 04:54:35 +00:00
MachineFunctionAnalysis.cpp
MachineFunctionPass.cpp
MachineFunctionPrinterPass.cpp
MachineInstr.cpp If known DebugLocs do not match then two DBG_VALUE machine instructions are not identical. For example, 2011-07-07 17:45:33 +00:00
MachineLICM.cpp Sink SubtargetFeature and TargetInstrItineraries (renamed MCInstrItineraries) into MC. 2011-06-29 01:14:12 +00:00
MachineLoopInfo.cpp
MachineLoopRanges.cpp
MachineModuleInfo.cpp Goodbye TargetAsmInfo. This eliminate last bit of CodeGen and Target in llvm-mc. 2011-07-20 19:50:42 +00:00
MachineModuleInfoImpls.cpp
MachinePassRegistry.cpp
MachineRegisterInfo.cpp Remove RegClass2VRegMap from MachineRegisterInfo. 2011-06-27 23:54:40 +00:00
MachineSink.cpp
MachineSSAUpdater.cpp
MachineVerifier.cpp - Rename TargetInstrDesc, TargetOperandInfo to MCInstrDesc and MCOperandInfo and 2011-06-28 19:10:37 +00:00
Makefile
ObjectCodeEmitter.cpp
OcamlGC.cpp
OptimizePHIs.cpp
Passes.cpp Update comment. 2011-04-30 03:13:08 +00:00
PeepholeOptimizer.cpp - Rename TargetInstrDesc, TargetOperandInfo to MCInstrDesc and MCOperandInfo and 2011-06-28 19:10:37 +00:00
PHIElimination.cpp Remove unused STL header includes. 2011-04-23 19:53:52 +00:00
PHIEliminationUtils.cpp
PHIEliminationUtils.h
PostRASchedulerList.cpp Rename TargetSubtarget to TargetSubtargetInfo for consistency. 2011-07-01 21:01:15 +00:00
ProcessImplicitDefs.cpp
PrologEpilogInserter.cpp Move CallFrameSetupOpcode and CallFrameDestroyOpcode to TargetInstrInfo. 2011-06-28 21:14:33 +00:00
PrologEpilogInserter.h
PseudoSourceValue.cpp
README.txt
RegAllocBase.h Switch AllocationOrder to using RegisterClassInfo instead of a BitVector 2011-06-03 20:34:53 +00:00
RegAllocBasic.cpp Better diagnostics when inline asm fails to allocate. 2011-07-02 07:17:37 +00:00
RegAllocFast.cpp Better diagnostics when inline asm fails to allocate. 2011-07-02 07:17:37 +00:00
RegAllocGreedy.cpp Migrate LLVM and Clang to use the new makeArrayRef(...) functions where previously explicit non-default constructors were used. 2011-07-18 12:00:32 +00:00
RegAllocLinearScan.cpp Remove the experimental (and unused) pre-ra splitting pass. Greedy regalloc can split live ranges. 2011-06-27 23:40:45 +00:00
RegAllocPBQP.cpp There is only one register coalescer. Merge it into the base class and 2011-06-26 22:34:10 +00:00
RegisterClassInfo.cpp Add TargetRegisterInfo::getRawAllocationOrder(). 2011-06-16 17:42:25 +00:00
RegisterClassInfo.h Migrate LLVM and Clang to use the new makeArrayRef(...) functions where previously explicit non-default constructors were used. 2011-07-18 12:00:32 +00:00
RegisterCoalescer.cpp Fix an easy fixme. 2011-07-03 05:26:42 +00:00
RegisterCoalescer.h Remove dead code. 2011-06-30 13:17:24 +00:00
RegisterScavenging.cpp Handle <def,undef> in the second loop as well. 2011-05-02 20:36:53 +00:00
RenderMachineFunction.cpp Add TargetRegisterInfo::getRawAllocationOrder(). 2011-06-16 17:42:25 +00:00
RenderMachineFunction.h
ScheduleDAG.cpp - Rename TargetInstrDesc, TargetOperandInfo to MCInstrDesc and MCOperandInfo and 2011-06-28 19:10:37 +00:00
ScheduleDAGEmit.cpp createMCInstPrinter doesn't need TargetMachine anymore. 2011-07-06 19:45:42 +00:00
ScheduleDAGInstrs.cpp Rename TargetSubtarget to TargetSubtargetInfo for consistency. 2011-07-01 21:01:15 +00:00
ScheduleDAGInstrs.h Update DBG_VALUEs while breaking anti dependencies. 2011-06-02 21:26:52 +00:00
ScheduleDAGPrinter.cpp
ScoreboardHazardRecognizer.cpp Sink SubtargetFeature and TargetInstrItineraries (renamed MCInstrItineraries) into MC. 2011-06-29 01:14:12 +00:00
ShadowStackGC.cpp land David Blaikie's patch to de-constify Type, with a few tweaks. 2011-07-18 04:54:35 +00:00
ShrinkWrapping.cpp
SjLjEHPrepare.cpp land David Blaikie's patch to de-constify Type, with a few tweaks. 2011-07-18 04:54:35 +00:00
SlotIndexes.cpp
Spiller.cpp Remove unused STL header includes. 2011-04-23 19:53:52 +00:00
Spiller.h
SpillPlacement.cpp
SpillPlacement.h
SplitKit.cpp Fix a crash when building 177.mesa for armv6. 2011-07-18 18:47:13 +00:00
SplitKit.h Extract parts of RAGreedy::splitAroundRegion as SplitKit methods. 2011-07-15 21:47:57 +00:00
Splitter.cpp merge SimpleRegisterCoalescing.h into RegisterCoalescer.h. 2011-06-26 21:54:28 +00:00
Splitter.h
StackProtector.cpp land David Blaikie's patch to de-constify Type, with a few tweaks. 2011-07-18 04:54:35 +00:00
StackSlotColoring.cpp - Rename TargetInstrDesc, TargetOperandInfo to MCInstrDesc and MCOperandInfo and 2011-06-28 19:10:37 +00:00
StrongPHIElimination.cpp
TailDuplication.cpp Move most of the pre BB code to TailDuplicateAndUpdate. Change the 2011-07-04 01:21:42 +00:00
TargetInstrInfoImpl.cpp - Rename TargetInstrDesc, TargetOperandInfo to MCInstrDesc and MCOperandInfo and 2011-06-28 19:10:37 +00:00
TargetLoweringObjectFileImpl.cpp Goodbye TargetAsmInfo. This eliminate last bit of CodeGen and Target in llvm-mc. 2011-07-20 19:50:42 +00:00
TwoAddressInstructionPass.cpp - Rename TargetInstrDesc, TargetOperandInfo to MCInstrDesc and MCOperandInfo and 2011-06-28 19:10:37 +00:00
UnreachableBlockElim.cpp Fix PR10029 - VerifyCoalescing failure on patterns_dfa.c of 445.gobmk. 2011-05-27 05:04:51 +00:00
VirtRegMap.cpp Also count identity copies. 2011-05-06 17:59:57 +00:00
VirtRegMap.h Be more aggressive about following hints. 2011-07-08 20:46:18 +00:00
VirtRegRewriter.cpp - Rename TargetInstrDesc, TargetOperandInfo to MCInstrDesc and MCOperandInfo and 2011-06-28 19:10:37 +00:00
VirtRegRewriter.h

//===---------------------------------------------------------------------===//

Common register allocation / spilling problem:

        mul lr, r4, lr
        str lr, [sp, #+52]
        ldr lr, [r1, #+32]
        sxth r3, r3
        ldr r4, [sp, #+52]
        mla r4, r3, lr, r4

can be:

        mul lr, r4, lr
        mov r4, lr
        str lr, [sp, #+52]
        ldr lr, [r1, #+32]
        sxth r3, r3
        mla r4, r3, lr, r4

and then "merge" mul and mov:

        mul r4, r4, lr
        str lr, [sp, #+52]
        ldr lr, [r1, #+32]
        sxth r3, r3
        mla r4, r3, lr, r4

It also increase the likelihood the store may become dead.

//===---------------------------------------------------------------------===//

bb27 ...
        ...
        %reg1037 = ADDri %reg1039, 1
        %reg1038 = ADDrs %reg1032, %reg1039, %NOREG, 10
    Successors according to CFG: 0x8b03bf0 (#5)

bb76 (0x8b03bf0, LLVM BB @0x8b032d0, ID#5):
    Predecessors according to CFG: 0x8b0c5f0 (#3) 0x8b0a7c0 (#4)
        %reg1039 = PHI %reg1070, mbb<bb76.outer,0x8b0c5f0>, %reg1037, mbb<bb27,0x8b0a7c0>

Note ADDri is not a two-address instruction. However, its result %reg1037 is an
operand of the PHI node in bb76 and its operand %reg1039 is the result of the
PHI node. We should treat it as a two-address code and make sure the ADDri is
scheduled after any node that reads %reg1039.

//===---------------------------------------------------------------------===//

Use local info (i.e. register scavenger) to assign it a free register to allow
reuse:
        ldr r3, [sp, #+4]
        add r3, r3, #3
        ldr r2, [sp, #+8]
        add r2, r2, #2
        ldr r1, [sp, #+4]  <==
        add r1, r1, #1
        ldr r0, [sp, #+4]
        add r0, r0, #2

//===---------------------------------------------------------------------===//

LLVM aggressively lift CSE out of loop. Sometimes this can be negative side-
effects:

R1 = X + 4
R2 = X + 7
R3 = X + 15

loop:
load [i + R1]
...
load [i + R2]
...
load [i + R3]

Suppose there is high register pressure, R1, R2, R3, can be spilled. We need
to implement proper re-materialization to handle this:

R1 = X + 4
R2 = X + 7
R3 = X + 15

loop:
R1 = X + 4  @ re-materialized
load [i + R1]
...
R2 = X + 7 @ re-materialized
load [i + R2]
...
R3 = X + 15 @ re-materialized
load [i + R3]

Furthermore, with re-association, we can enable sharing:

R1 = X + 4
R2 = X + 7
R3 = X + 15

loop:
T = i + X
load [T + 4]
...
load [T + 7]
...
load [T + 15]
//===---------------------------------------------------------------------===//

It's not always a good idea to choose rematerialization over spilling. If all
the load / store instructions would be folded then spilling is cheaper because
it won't require new live intervals / registers. See 2003-05-31-LongShifts for
an example.

//===---------------------------------------------------------------------===//

With a copying garbage collector, derived pointers must not be retained across
collector safe points; the collector could move the objects and invalidate the
derived pointer. This is bad enough in the first place, but safe points can
crop up unpredictably. Consider:

        %array = load { i32, [0 x %obj] }** %array_addr
        %nth_el = getelementptr { i32, [0 x %obj] }* %array, i32 0, i32 %n
        %old = load %obj** %nth_el
        %z = div i64 %x, %y
        store %obj* %new, %obj** %nth_el

If the i64 division is lowered to a libcall, then a safe point will (must)
appear for the call site. If a collection occurs, %array and %nth_el no longer
point into the correct object.

The fix for this is to copy address calculations so that dependent pointers
are never live across safe point boundaries. But the loads cannot be copied
like this if there was an intervening store, so may be hard to get right.

Only a concurrent mutator can trigger a collection at the libcall safe point.
So single-threaded programs do not have this requirement, even with a copying
collector. Still, LLVM optimizations would probably undo a front-end's careful
work.

//===---------------------------------------------------------------------===//

The ocaml frametable structure supports liveness information. It would be good
to support it.

//===---------------------------------------------------------------------===//

The FIXME in ComputeCommonTailLength in BranchFolding.cpp needs to be
revisited. The check is there to work around a misuse of directives in inline
assembly.

//===---------------------------------------------------------------------===//

It would be good to detect collector/target compatibility instead of silently
doing the wrong thing.

//===---------------------------------------------------------------------===//

It would be really nice to be able to write patterns in .td files for copies,
which would eliminate a bunch of explicit predicates on them (e.g. no side 
effects).  Once this is in place, it would be even better to have tblgen 
synthesize the various copy insertion/inspection methods in TargetInstrInfo.

//===---------------------------------------------------------------------===//

Stack coloring improvements:

1. Do proper LiveStackAnalysis on all stack objects including those which are
   not spill slots.
2. Reorder objects to fill in gaps between objects.
   e.g. 4, 1, <gap>, 4, 1, 1, 1, <gap>, 4 => 4, 1, 1, 1, 1, 4, 4

//===---------------------------------------------------------------------===//

The scheduler should be able to sort nearby instructions by their address. For
example, in an expanded memset sequence it's not uncommon to see code like this:

  movl $0, 4(%rdi)
  movl $0, 8(%rdi)
  movl $0, 12(%rdi)
  movl $0, 0(%rdi)

Each of the stores is independent, and the scheduler is currently making an
arbitrary decision about the order.

//===---------------------------------------------------------------------===//

Another opportunitiy in this code is that the $0 could be moved to a register:

  movl $0, 4(%rdi)
  movl $0, 8(%rdi)
  movl $0, 12(%rdi)
  movl $0, 0(%rdi)

This would save substantial code size, especially for longer sequences like
this. It would be easy to have a rule telling isel to avoid matching MOV32mi
if the immediate has more than some fixed number of uses. It's more involved
to teach the register allocator how to do late folding to recover from
excessive register pressure.