llvm-project/llvm/lib/CodeGen
Evan Cheng ede2ce71aa Fix up support for OptionalDefOperand when it defaults to an actual register def. I need this to get ready for major Thumb1 surgery.
llvm-svn: 75328
2009-07-11 01:06:50 +00:00
..
AsmPrinter Accidently dropped this while removing dead code in previous commit. 2009-07-07 21:55:14 +00:00
SelectionDAG Fix up support for OptionalDefOperand when it defaults to an actual register def. I need this to get ready for major Thumb1 surgery. 2009-07-11 01:06:50 +00:00
BranchFolding.cpp Fix PR4188. TailMerging can't tolerate inexact 2009-05-11 21:54:13 +00:00
CMakeLists.txt Fix the cmake build - patch by Xerxes Rånby. 2009-07-06 14:28:32 +00:00
CodePlacementOpt.cpp Fix CodePlacementOpt::OptimizeIntraLoopEdges so that its return value 2009-05-18 21:02:18 +00:00
DeadMachineInstructionElim.cpp Rename AliasSet to SubRegs, to reflect changes in the surrounding code. 2008-10-16 01:06:18 +00:00
DwarfEHPrepare.cpp More LLVMContext-ification. 2009-07-05 22:41:43 +00:00
ELF.h shrinking down #includes 2009-07-02 18:29:24 +00:00
ELFCodeEmitter.cpp Changed ELFCodeEmitter to inherit from ObjectCodeEmitter 2009-07-06 09:26:48 +00:00
ELFCodeEmitter.h Changed ELFCodeEmitter to inherit from ObjectCodeEmitter 2009-07-06 09:26:48 +00:00
ELFWriter.cpp Changed ELFCodeEmitter to inherit from ObjectCodeEmitter 2009-07-06 09:26:48 +00:00
ELFWriter.h Changed ELFCodeEmitter to inherit from ObjectCodeEmitter 2009-07-06 09:26:48 +00:00
GCMetadata.cpp Eliminate several more unnecessary intptr_t casts. 2009-02-18 05:09:16 +00:00
GCMetadataPrinter.cpp Use raw_ostream throughout the AsmPrinter. 2008-08-21 00:14:44 +00:00
GCStrategy.cpp Eliminate several more unnecessary intptr_t casts. 2009-02-18 05:09:16 +00:00
IfConversion.cpp Fewer static variables, part 3 of many. 2009-06-24 23:41:44 +00:00
IntrinsicLowering.cpp Push LLVMContext _back_ through IRBuilder. 2009-07-08 20:50:47 +00:00
LLVMTargetMachine.cpp Add the Object Code Emitter class. Original patch by Aaron Gray, I did some 2009-07-06 05:09:34 +00:00
LatencyPriorityQueue.cpp Add initial support for back-scheduling address computations, 2008-12-16 03:35:01 +00:00
LazyLiveness.cpp Owen Anderson 2009-06-15: Use a SmallPtrSet here, for speed and to match df_iterator. 2009-06-15 22:54:48 +00:00
LiveInterval.cpp Improved tracking of value number kills. VN kills are now represented 2009-07-09 03:57:02 +00:00
LiveIntervalAnalysis.cpp Avoid compiler warnings if assertions turned off. 2009-07-10 20:07:07 +00:00
LiveStackAnalysis.cpp Update to in-place spilling framework. Includes live interval scaling and trivial rewriter. 2009-06-02 16:53:25 +00:00
LiveVariables.cpp Avoid adding a duplicate def. This fixes PR4478. 2009-07-06 21:34:05 +00:00
LowerSubregs.cpp Do not fold away subreg_to_reg if the source register has a sub-register index. That means the source register is taking a sub-register of a larger register. e.g. On x86 2009-03-23 07:19:58 +00:00
MachO.h Cleanup MachO writer and code emitter. Fix 80 cols problems, remove extra spaces, shrink down includes and move some methods out-of-line 2009-07-06 06:40:51 +00:00
MachOCodeEmitter.cpp Cleanup MachO writer and code emitter. Fix 80 cols problems, remove extra spaces, shrink down includes and move some methods out-of-line 2009-07-06 06:40:51 +00:00
MachOCodeEmitter.h Cleanup MachO writer and code emitter. Fix 80 cols problems, remove extra spaces, shrink down includes and move some methods out-of-line 2009-07-06 06:40:51 +00:00
MachOWriter.cpp Cleanup MachO writer and code emitter. Fix 80 cols problems, remove extra spaces, shrink down includes and move some methods out-of-line 2009-07-06 06:40:51 +00:00
MachOWriter.h Cleanup MachO writer and code emitter. Fix 80 cols problems, remove extra spaces, shrink down includes and move some methods out-of-line 2009-07-06 06:40:51 +00:00
MachineBasicBlock.cpp If a MachineBasicBlock has multiple ways of reaching another block, 2009-05-05 21:10:19 +00:00
MachineDominators.cpp Eliminate several more unnecessary intptr_t casts. 2009-02-18 05:09:16 +00:00
MachineFunction.cpp Add an "alignment" field to the MachineFunction object. It makes more sense to 2009-06-30 22:38:32 +00:00
MachineInstr.cpp Add a bit IsUndef to MachineOperand. This indicates the def / use register operand is defined by an implicit_def. That means it can def / use any register and passes (e.g. register scavenger) can feel free to ignore them. 2009-06-30 08:49:04 +00:00
MachineLICM.cpp MachineLICM CSE should match destination register classes; avoid hoisting implicit_def's. 2009-02-27 00:02:22 +00:00
MachineLoopInfo.cpp Eliminate a layer of indirection in LoopInfo and MachineLoopInfo. 2009-06-27 21:22:48 +00:00
MachineModuleInfo.cpp DebugLabelFolder ruthlessly deletes redundant labels. However, sometimes the redundant labels is referenced by debug info somewhere else. This patch provies a way so that dwarf writer can mark labels as used. 2009-04-10 18:58:59 +00:00
MachinePassRegistry.cpp Remove attribution from file headers, per discussion on llvmdev. 2007-12-29 20:36:04 +00:00
MachineRegisterInfo.cpp Part 1. 2009-06-15 08:28:29 +00:00
MachineSink.cpp fix two problems with machine sinking: 2009-04-10 16:38:36 +00:00
MachineVerifier.cpp Rename MachineVerifier pass to avoid command line collision. 2009-05-17 19:37:14 +00:00
Makefile Removed trailing whitespace from Makefiles. 2009-01-09 16:44:42 +00:00
ObjectCodeEmitter.cpp Just forgot to include the two new files 2009-07-06 05:16:40 +00:00
OcamlGC.cpp Registry.h should not depend on CommandLine.h. 2009-01-16 07:02:28 +00:00
PBQP.cpp reorder #include order, patch by Kenneth Boyd! 2008-10-06 03:54:25 +00:00
PBQP.h Fix typos pointed out by Duncan. Also untabify these files. 2008-10-03 17:11:58 +00:00
PHIElimination.cpp Revert 74898. It broke several tests. 2009-07-07 17:50:43 +00:00
Passes.cpp Clean up the use of static and anonymous namespaces. This turned up 2008-05-13 00:00:25 +00:00
PostRASchedulerList.cpp Move getInstrOperandRegClass from the scheduler to TargetInstrInfo. 2009-05-05 00:30:09 +00:00
PreAllocSplitting.cpp Improved tracking of value number kills. VN kills are now represented 2009-07-09 03:57:02 +00:00
PrologEpilogInserter.cpp Targets sometimes assign fixed stack object to spill certain callee-saved 2009-07-09 06:53:48 +00:00
PrologEpilogInserter.h PEI: rename PEI.h to PrologEpilogInserter.h to adhere to file naming standard 2009-05-13 17:52:11 +00:00
PseudoSourceValue.cpp Now that errs() is properly non-buffered, there's no need to 2009-03-23 15:57:19 +00:00
README.txt Remove tabs. 2008-08-22 00:04:26 +00:00
RegAllocBigBlock.cpp Adjust the sizes for a few SmallVectors to reflect their usage. 2009-02-12 17:29:01 +00:00
RegAllocLinearScan.cpp Handle IMPLICIT_DEF with isUndef operand marker, part 2. This patch moves the code to annotate machineoperands to LiveIntervalAnalysis. It also add markers for implicit_def that define physical registers. The rest, is just a lot of details. 2009-07-01 01:59:31 +00:00
RegAllocLocal.cpp Fix pr4100. Do not remove no-op copies when they are dead. The register 2009-05-07 23:47:03 +00:00
RegAllocPBQP.cpp VNInfo cleanup. 2009-06-17 21:01:20 +00:00
RegAllocSimple.cpp Fix pr3954. The register scavenger asserts for inline assembly with 2009-04-09 17:16:43 +00:00
RegisterCoalescer.cpp Clean up the use of static and anonymous namespaces. This turned up 2008-05-13 00:00:25 +00:00
RegisterScavenging.cpp Remove special handling of implicit_def. Fix a couple more bugs in liveintervalanalysis and coalescer handling of implicit_def. 2009-07-01 08:19:36 +00:00
ScheduleDAG.cpp When scheduling a block in parts, keep track of the overall 2009-02-11 04:27:20 +00:00
ScheduleDAGEmit.cpp When scheduling a block in parts, keep track of the overall 2009-02-11 04:27:20 +00:00
ScheduleDAGInstrs.cpp When scheduling a block in parts, keep track of the overall 2009-02-11 04:27:20 +00:00
ScheduleDAGInstrs.h When scheduling a block in parts, keep track of the overall 2009-02-11 04:27:20 +00:00
ScheduleDAGPrinter.cpp Get rid of the global CFGOnly flag by threading a ShortNames parameters through the GraphViz rendering code. 2009-06-24 17:37:09 +00:00
ShadowStackGC.cpp Push LLVMContext _back_ through IRBuilder. 2009-07-08 20:50:47 +00:00
ShrinkWrapping.cpp PEI: rename PEI.h to PrologEpilogInserter.h to adhere to file naming standard 2009-05-13 17:52:11 +00:00
SimpleRegisterCoalescing.cpp Remove TargetInstrInfo::CommuteChangesDestination and added findCommutedOpIndices which returns the operand indices which are swapped (when applicable). This allows for some code clean up and future enhancements. 2009-07-10 19:15:51 +00:00
SimpleRegisterCoalescing.h Rename RemoveCopiesFromValNo to TurnCopiesFromValNoToImpDefs. 2009-06-16 07:15:05 +00:00
Spiller.cpp Improved tracking of value number kills. VN kills are now represented 2009-07-09 03:57:02 +00:00
Spiller.h More VNInfo tweaking, plus a little progress on intra-block splitting. 2009-06-19 02:17:53 +00:00
StackProtector.cpp This started as a small change, I swear. Unfortunately, lots of things call the [I|F]CmpInst constructors. Who knew!? 2009-07-09 23:48:35 +00:00
StackSlotColoring.cpp Fixed a stack slot coloring with reg bug: do not update implicit use / def when doing forward / backward propagation. 2009-05-12 18:31:57 +00:00
StrongPHIElimination.cpp Improved tracking of value number kills. VN kills are now represented 2009-07-09 03:57:02 +00:00
TargetInstrInfoImpl.cpp Use findCommutedOpIndices to find the operands to commute. 2009-07-10 23:26:12 +00:00
TwoAddressInstructionPass.cpp Two-address pass should use findCommutedOpIndices to determine what registers are commuted. 2009-07-11 00:04:23 +00:00
UnreachableBlockElim.cpp Rename getAnalysisToUpdate to getAnalysisIfAvailable. 2009-01-28 13:14:17 +00:00
VirtRegMap.cpp Part 1. 2009-06-15 08:28:29 +00:00
VirtRegMap.h Move register allocation preference (or hint) from LiveInterval to MachineRegisterInfo. This allows more passes to set them. 2009-06-14 20:22:55 +00:00
VirtRegRewriter.cpp Remove TargetInstrInfo::CommuteChangesDestination and added findCommutedOpIndices which returns the operand indices which are swapped (when applicable). This allows for some code clean up and future enhancements. 2009-07-10 19:15:51 +00:00
VirtRegRewriter.h Fixed a bug in LiveInterval scaling (failure to scale VNI defs correctly), removed old TODO comments. 2009-06-24 02:17:32 +00:00

README.txt

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

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 likelyhood the store may become dead.

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

I think we should have a "hasSideEffects" flag (which is automatically set for
stuff that "isLoad" "isCall" etc), and the remat pass should eventually be able
to remat any instruction that has no side effects, if it can handle it and if
profitable.

For now, I'd suggest having the remat stuff work like this:

1. I need to spill/reload this thing.
2. Check to see if it has side effects.
3. Check to see if it is simple enough: e.g. it only has one register
destination and no register input.
4. If so, clone the instruction, do the xform, etc.

Advantages of this are:

1. the .td file describes the behavior of the instructions, not the way the
   algorithm should work.
2. as remat gets smarter in the future, we shouldn't have to be changing the .td
   files.
3. it is easier to explain what the flag means in the .td file, because you
   don't have to pull in the explanation of how the current remat algo works.

Some potential added complexities:

1. Some instructions have to be glued to it's predecessor or successor. All of
   the PC relative instructions and condition code setting instruction. We could
   mark them as hasSideEffects, but that's not quite right. PC relative loads
   from constantpools can be remat'ed, for example. But it requires more than
   just cloning the instruction. Some instructions can be remat'ed but it
   expands to more than one instruction. But allocator will have to make a
   decision.

4. As stated in 3, not as simple as cloning in some cases. The target will have
   to decide how to remat it. For example, an ARM 2-piece constant generation
   instruction is remat'ed as a load from constantpool.

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

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 improvments:

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