aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen/InlineSpiller.cpp
AgeCommit message (Collapse)Author
2011-02-09Set an allocation hint when rematting before a COPY.Jakob Stoklund Olesen
This almost guarantees that the COPY will be coalesced. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125140 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-08Avoid folding a load instruction into an instruction that redefines the ↵Jakob Stoklund Olesen
register. The target hook doesn't know how to do that. (Neither do I). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125108 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-09Teach TargetRegisterInfo how to cram stack slot indexes in with the virtual andJakob Stoklund Olesen
physical register numbers. This makes the hack used in LiveInterval official, and lets LiveInterval be oblivious of stack slots. The isPhysicalRegister() and isVirtualRegister() predicates don't know about this, so when a variable may contain a stack slot, isStackSlot() should always be tested first. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123128 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-18Apparently, operandices is not a word.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122135 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-18Teach the inline spiller to attempt folding a load instruction into its singleJakob Stoklund Olesen
use before rematerializing the load. This allows us to produce: addps LCPI0_1(%rip), %xmm2 Instead of: movaps LCPI0_1(%rip), %xmm3 addps %xmm3, %xmm2 Saving a register and an instruction. The standard spiller already knows how to do this. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122133 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-18Pass a Banner argument to the machine code verifier both fromJakob Stoklund Olesen
createMachineVerifierPass and MachineFunction::verify. The banner is printed before the machine code dump, just like the printer pass. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122113 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-10Rip out live range splitting support from the inline spiller.Jakob Stoklund Olesen
The spiller should only spill. The register allocator will drive live range splitting, it has the needed information about register pressure and interferences. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@121590 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-15When spilling a register defined by an early clobber, make sure that the newJakob Stoklund Olesen
live ranges for the spill register are also defined at the use slot instead of the normal def slot. This fixes PR8612 for the inline spiller. A use was being allocated to the same register as a spilled early clobber def. This problem exists in all the spillers. A fix for the standard spiller is forthcoming. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119182 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-10Hook up AliasAnalysis in InlineSpiller. This is used for rematerializingJakob Stoklund Olesen
constant loads. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118741 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-10RABasic is nearly functionally complete. There are a few remainingAndrew Trick
benchmarks hitting an assertion. Adds LiveIntervalUnion::collectInterferingVRegs. Fixes "late spilling" by checking for any unspillable live vregs among all physReg aliases. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118701 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-10Simplify the LiveRangeEdit::canRematerializeAt() interface a bit.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118661 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-04Disable fancy splitting during spilling unless -extra-spiller-splits is given.Jakob Stoklund Olesen
This way, InlineSpiller does the same amount of splitting as the standard spiller. Splitting should really be guided by the register allocator, and doesn't belong in the spiller at all. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118216 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-03Tag debug output as regallocJakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118193 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-01Don't assign new registers created during a split to the same stack slot, butJakob Stoklund Olesen
give them individual stack slots once the are actually spilled. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117945 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-30Print out register class of spilled register.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117761 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-29Run a verification pass before any splitting to better distribute blame.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117629 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-28Make MachineDominators available for SplitEditor. We are going to need it forJakob Stoklund Olesen
proper SSA updating. This doesn't cause MachineDominators to be recomputed since we are already requiring MachineLoopInfo which uses dominators as well. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117598 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-28Add a temporary command line option to verify machine code after each spill orJakob Stoklund Olesen
split. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117597 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-26InlineSpiller can also update LiveStacks.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117338 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-20Move some of the InlineSpiller rematerialization code into LiveRangeEdit.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116951 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-15Move stack slot assignments into LiveRangeEdit.Jakob Stoklund Olesen
All registers created during splitting or spilling are assigned to the same stack slot as the parent register. When splitting or rematting, we may not spill at all. In that case the stack slot is still assigned, but it will be dead. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116546 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-14Create a new LiveRangeEdit class to keep track of the new registers created whenJakob Stoklund Olesen
splitting or spillling, and to help with rematerialization. Use LiveRangeEdit in InlineSpiller and SplitKit. This will eventually make it possible to share remat code between InlineSpiller and SplitKit. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116543 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-05Update SplitEditor API to reflect the fact that the original live interval isJakob Stoklund Olesen
never kept after splitting. Keeping the original interval made sense when the split region doesn't modify the register, and the original is spilled. We can get the same effect by detecting reloaded values when spilling around copies. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@115695 91177308-0d34-0410-b5e6-96231b3b80d8
2010-09-26Fixed some tests to avoid LiveIntervals::getInstructionFromIndex(..) ↵Lang Hames
overhead where possible. Thanks to Jakob for the suggestions. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@114798 91177308-0d34-0410-b5e6-96231b3b80d8
2010-09-25Removed VNInfo::isDefAccurate(). Def "accuracy" can be checked by testing ↵Lang Hames
whether LiveIntervals::getInstructionFromIndex(def) returns NULL. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@114791 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-13Clean up the Spiller.h interface.Jakob Stoklund Olesen
The earliestStart argument is entirely specific to linear scan allocation, and can be easily calculated by RegAllocLinearScan. Replace std::vector with SmallVector. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@111055 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-13Implement splitting inside a single block.Jakob Stoklund Olesen
When a live range is contained a single block, we can split it around instruction clusters. The current approach is very primitive, splitting before and after the largest gap between uses. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@111043 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-12Implement single block splitting.Jakob Stoklund Olesen
Before spilling a live range, we split it into a separate range for each basic block where it is used. That way we only get one reload per basic block if the new smaller ranges can allocate to a register. This type of splitting is already present in the standard spiller. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@110934 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-10Avoid editing the current live interval during remat.Jakob Stoklund Olesen
The live interval may be used for a spill slot as well, and that spill slot could be shared by split registers. We cannot shrink it, even if we know the current register won't need the spill slot in that range. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@110721 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-06Lazily defer duplicating the live interval we are splitting until we know it isJakob Stoklund Olesen
necessary. Sometimes, live range splitting doesn't shrink the current interval, but simply changes some instructions to use a new interval. That makes the original more suitable for spilling. In this case, we don't need to duplicate the original. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@110481 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-06Keep the MachiuneFunctionPass pointer around. It is useful for verification.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@110464 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-04Coalesce stack slot accesses that arise when spilling both sides of a COPY.Jakob Stoklund Olesen
This helps avoid silly code: %R0<def = LOAD <fi#5> STORE <fi#5>, %R0<kill> git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@110266 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-04Checkpoint SplitKit progress.Jakob Stoklund Olesen
We are now at a point where we can split around simple single-entry, single-exit loops, although still with some bugs. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@110257 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-26Add SplitEditor to SplitKit. This class will be used to edit live intervals andJakob Stoklund Olesen
rewrite instructions for live range splitting. Still work in progress. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@109469 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-20Change the createSpiller interface to take a MachineFunctionPass argument.Jakob Stoklund Olesen
The spillers can pluck the analyses they need from the pass reference. Switch some never-null pointers to references. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@108969 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-20Beginning SplitKit - utility classes for live range splitting.Jakob Stoklund Olesen
This is a work in progress. So far we have some basic loop analysis to help determine where it is useful to split a live range around a loop. The actual loop splitting code from Splitter.cpp is also going to move in here. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@108842 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-19Spillers may alter MachineLoopInfo when breaking critical edges, so make itJakob Stoklund Olesen
non-const. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@108734 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-09Change TII::foldMemoryOperand API to require the machine instruction to beJakob Stoklund Olesen
inserted in a MBB, and return an already inserted MI. This target API change is necessary to allow foldMemoryOperand to call storeToStackSlot and loadFromStackSlot when folding a COPY to a stack slot reference in a target independent way. The foldMemoryOperandImpl hook is going to change in the same way, but I'll wait until COPY folding is actually implemented. Most targets only fold copies and won't need to specialize this hook at all. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@107991 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-02Remove invalid assertJakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@107505 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-02Properly handle debug values during inline spilling.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@107503 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-02Rematerialize as much as possible before inserting spills and reloads.Jakob Stoklund Olesen
This allows us to recognize the common case where all uses could be rematerialized, and no stack slot allocation is necessary. If some values could be fully rematerialized, remove them from the live range before allocating a stack slot for the rest. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@107492 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-01Add memory operand folding support to InlineSpiller.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@107355 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-30Add support for rematerialization to InlineSpiller.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@107351 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-30Remember to track spill slot uses in VirtRegMap when inserting loads and stores.Jakob Stoklund Olesen
LocalRewriter::runOnMachineFunction uses this information to mark dead spill slots. This means that InlineSpiller now also works for functions that spill. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@107302 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-29Begin implementation of an inline spiller.Jakob Stoklund Olesen
InlineSpiller inserts loads and spills immediately instead of deferring to VirtRegMap. This is possible now because SlotIndexes allows instructions to be inserted and renumbered. This is work in progress, and is mostly a copy of TrivialSpiller so far. It works very well for functions that don't require spilling. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@107227 91177308-0d34-0410-b5e6-96231b3b80d8