aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen
AgeCommit message (Collapse)Author
2011-09-15Some legalization fixes for atomic load and store.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139851 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-15Add an option to disable spill hoisting.Jakob Stoklund Olesen
When -split-spill-mode is enabled, spill hoisting is performed by SplitKit instead of by InlineSpiller. This hidden command line option is for testing the splitter spill mode. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139845 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-15VirtRegMap is counting spill slots, not register spills.Jakob Stoklund Olesen
Fix the stats counters to reflect that. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139819 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-15Count correctly when a COPY turns into a spill or reload.Jakob Stoklund Olesen
The number of spills could go negative since a folded COPY is just a spill, and it may be eliminated. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139815 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-15Count inserted spills and reloads more accurately.Jakob Stoklund Olesen
Adjust counters when removing spill and reload instructions. We still don't account for reloads being removed by eliminateDeadDefs(). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139806 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-15Trace through sibling PHIs in bulk.Jakob Stoklund Olesen
When traceSiblingValue() encounters a PHI-def value created by live range splitting, don't look at all the predecessor blocks. That can be very expensive in a complicated CFG. Instead, consider that all the non-PHI defs jointly dominate all the PHI-defs. Tracing directly to all the non-PHI defs is much faster that zipping around in the CFG when there are many PHIs with many predecessors. This significantly improves compile time for indirectbr interpreters. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139797 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-15Speed up LiveIntervals::shrinkToUse with some caching.Jakob Stoklund Olesen
Blocks with multiple PHI successors only need to go on the worklist once. Use a SmallPtrSet to track the live-out blocks that have already been handled. This is a lot faster than the two live range check we would otherwise do. Also stop recomputing hasPHIKill flags. Like RenumberValues(), it is conservatively correct to leave them in, and they are not used for anything important. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139792 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-15Revert r139782, "RemoveCopyByCommutingDef doesn't need hasPHIKill()."Jakob Stoklund Olesen
It does, after all. RemoveCopyByCommutingDef rewrites the uses of one particular value number in A. It doesn't know how to rewrite phi uses, so there can't be any. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139787 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-15Stop verifying hasPHIKill() flags.Jakob Stoklund Olesen
There is only one legitimate use remaining, in addIntervalsForSpills(). All other calls to hasPHIKill() are only used to update PHIKill flags. The addIntervalsForSpills() function is part of the old spilling framework, only used by linearscan. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139783 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-15RemoveCopyByCommutingDef doesn't need hasPHIKill().Jakob Stoklund Olesen
Instead, let HasOtherReachingDefs() test for defs in B that overlap any phi-defs in A as well. This test is slightly different, but almost identical. A perfectly precise test would only check those phi-defs in A that are reachable from AValNo. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139782 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-15It is safe to remat a value killed by phis.Jakob Stoklund Olesen
The source live range is recomputed using shrinkToUses() which does handle phis correctly. The hasPHIKill() condition was relevant in the old days when ReMaterializeTrivialDef() tried to recompute the live range itself. The shrinkToUses() function will mark the original def as dead when no more uses and phi kills remain. It is then removed by runOnMachineFunction(). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139781 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-15Leave hasPHIKill flags alone in LiveInterval::RenumberValues.Jakob Stoklund Olesen
It is conservatively correct to keep the hasPHIKill flags, even after deleting PHI-defs. The calculation can be very expensive after taildup has created a quadratic number of indirectbr edges in the CFG, and the hasPHIKill flag isn't used for anything after RenumberValues(). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139780 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-15[regcoalescing] bug fix for RegistersDefinedFromSameValue.Andrew Trick
An improper SlotIndex->VNInfo lookup was leading to unsafe copy removal. Fixes PR10920 401.bzip2 miscompile with no IV rewrite. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139765 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-14Add support to emit debug info for C++0x nullptr type.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139751 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-14Ignore the cloning of unknown registers.Jakob Stoklund Olesen
THe LRE_DidCloneVirtReg callback may be called with vitual registers that RAGreedy doesn't even know about yet. In that case, there are no data structures to update. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139702 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-14Hoist back-copies to the least busy dominator.Jakob Stoklund Olesen
When a back-copy is hoisted to the nearest common dominator, keep looking up the dominator tree for a less loopy dominator, and place the back-copy there instead. Don't do this when a single existing back-copy dominates all the others. Assume the client knows what he is doing, and keep the dominating back-copy. This prevents us from hoisting back-copies into loops in most cases. If a value is defined in a loop with multiple exits, we may still hoist back-copies into that loop. That is the speed/size tradeoff. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139698 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-14Add integer promotion support for vselectNadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139692 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-13Distinguish complex mapped values from forced recomputation.Jakob Stoklund Olesen
When a ParentVNI maps to multiple defs in a new interval, its live range may still be derived directly from RegAssign by transferValues(). On the other hand, when instructions have been rematerialized or hoisted, it may be necessary to completely recompute live ranges using LiveRangeCalc::extend() to all uses. Use a bit in the value map to indicate that a live range must be recomputed. Rename markComplexMapped() to forceRecompute(). This fixes some live range verification errors when -split-spill-mode=size hoists back-copies by recomputing source ranges when RegAssign kills can't be moved. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139660 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-13Implement -split-spill-mode=size.Jakob Stoklund Olesen
Whenever the complement interval is defined by multiple copies of the same value, hoist those back-copies to the nearest common dominator. This ensures that at most one copy is inserted per value in the complement inteval, and no phi-defs are needed. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139651 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-13Fix check for unaligned load/store so it doesn't catch over-aligned load/store.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139649 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-13Error out on CodeGen of unaligned load/store. Fix test so it isn't ↵Eli Friedman
accidentally testing that case. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139641 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-13Fix the assertion which checks the size of the input operand.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139633 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-13Add vselect target support for targets that do not support blend but do supportNadav Rotem
xor/and/or (For example SSE2). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139623 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-13Use a cache to maintain list of machine basic blocks for a given UserValue.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139616 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-13Add SplitEditor::markOverlappedComplement().Jakob Stoklund Olesen
This function is used to flag values where the complement interval may overlap other intervals. Call it from overlapIntv, and use the flag to fully recompute those live ranges in transferValues(). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139612 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-13Eliminate the extendRange() wrapper.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139608 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-13Switch extendInBlock() to take a kill slot instead of the last use slot.Jakob Stoklund Olesen
Three out of four clients prefer this interface which is consistent with extendIntervalEndTo() and LiveRangeCalc::extend(). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139604 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-13Use a separate LiveRangeCalc for the complement in spill modes.Jakob Stoklund Olesen
The complement interval may overlap the other intervals created, so use a separate LiveRangeCalc instance to compute its live range. A LiveRangeCalc instance can only be shared among non-overlapping intervals. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139603 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-13Unbreak msvc.NAKAMURA Takumi
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139581 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-13Extract live range calculations from SplitKit.Jakob Stoklund Olesen
SplitKit will soon need two copies of these data structures, and the algorithms will also be useful when LiveIntervalAnalysis becomes independent of LiveVariables. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139572 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-12Introduce a bit of a hack.Bill Wendling
Splitting a landing pad takes considerable care because of PHIs and other nasties. The problem is that the jump table needs to jump to the landing pad block. However, the landing pad block can be jumped to only by an invoke instruction. So we clone the landingpad instruction into its own basic block, have the invoke jump to there. The landingpad instruction's basic block's successor is now the target for the jump table. But because of PHI nodes, we need to create another basic block for the jump table to jump to. This is definitely a hack, because the values for the PHI nodes may not be defined on the edge from the jump table. But that's okay, because the jump table is simply a construct to mimic what is happening in the CFG. So the values are mysteriously there, even though there is no value for the PHI from the jump table's edge (hence calling this a hack). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139545 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-12Remove the -compact-regions flag.Jakob Stoklund Olesen
It has been enabled by default for a while, it was only there to allow performance comparisons. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139501 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-12Add an interface for SplitKit complement spill modes.Jakob Stoklund Olesen
SplitKit always computes a complement live range to cover the places where the original live range was live, but no explicit region has been allocated. Currently, the complement live range is created to be as small as possible - it never overlaps any of the regions. This minimizes register pressure, but if the complement is going to be spilled anyway, that is not very important. The spiller will eliminate redundant spills, and hoist others by making the spill slot live range overlap some of the regions created by splitting. Stack slots are cheap. This patch adds the interface to enable spill modes in SplitKit. In spill mode, SplitKit will assume that the complement is going to spill, so it will allow it to overlap regions in order to avoid back-copies. By doing some of the spiller's work early, the complement live range becomes simpler. In some cases, it can become much simpler because no extra PHI-defs are required. This will speed up both splitting and spilling. This is only the interface to enable spill modes, no implementation yet. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139500 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-12Update comments to reflect some (not so) recent changes.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139498 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-10Fix asserts in CodeGen from:Richard Trieu
assert("error"); to: assert(0 && "error"); git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139449 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-09tidy up a bitChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139419 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-09Make the SelectionDAG verify that all the operands of BUILD_VECTOR have the ↵Eli Friedman
same type. Teach DAGCombiner::visitINSERT_VECTOR_ELT not to make invalid BUILD_VECTORs. Fixes PR10897. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139407 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-09Reapply r139247: Cache intermediate results during traceSiblingValue.Jakob Stoklund Olesen
In some cases such as interpreters using indirectbr, the CFG can be very complicated, and live range splitting may be forced to insert a large number of phi-defs. When that happens, traceSiblingValue can spend a lot of time zipping around in the CFG looking for defs and reloads. This patch causes more information to be cached in SibValues, and the cached values are used to terminate searches early. This speeds up spilling by 20x in one interpreter test case. For more typical code, this is just a 10% speedup of spilling. The previous version had bugs that caused miscompilations. They have been fixed. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139378 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08Directly point debug info to the stack slot of the arugment, instead of ↵Devang Patel
trying to keep track of vreg in which it the arugment is copied. The LiveDebugVariable can keep track of variable's ranges. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139330 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-07Revert r139247 "Cache intermediate results during traceSiblingValue."Jakob Stoklund Olesen
It broke the self host and clang-x86_64-darwin10-RA. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139259 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-07Cache intermediate results during traceSiblingValue.Jakob Stoklund Olesen
In some cases such as interpreters using indirectbr, the CFG can be very complicated, and live range splitting may be forced to insert a large number of phi-defs. When that happens, traceSiblingValue can spend a lot of time zipping around in the CFG looking for defs and reloads. This patch causes more information to be cached in SibValues, and the cached values are used to terminate searches early. This speeds up spilling by 20x in one interpreter test case. For more typical code, this is just a 10% speedup of spilling. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139247 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-07Refactor instprinter and mcdisassembler to take a SubtargetInfo. Add -mattr= ↵James Molloy
handling to llvm-mc. Reviewed by Owen Anderson. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139237 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-07Relax the MemOperands on atomics a bit. Fixes -verify-machineinstrs ↵Eli Friedman
failures for atomic laod/store on ARM. (The fix for the related failures on x86 is going to be nastier because we actually need Acquire memoperands attached to the atomic load instrs, etc.) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139221 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-07While sinking machine instructions, sink matching DBG_VALUEs also otherwise ↵Devang Patel
live debug variable pass will drop DBG_VALUEs on the floor. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139208 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-06Add codegen support for vector select (in the IR this means a selectDuncan Sands
with a vector condition); such selects become VSELECT codegen nodes. This patch also removes VSETCC codegen nodes, unifying them with SETCC nodes (codegen was actually often using SETCC for vector SETCC already). This ensures that various DAG combiner optimizations kick in for vector comparisons. Passes dragonegg bootstrap with no testsuite regressions (nightly testsuite as well as "make check-all"). Patch mostly by Nadav Rotem. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139159 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-06Split the init.trampoline intrinsic, which currently combines GCC'sDuncan Sands
init.trampoline and adjust.trampoline intrinsics, into two intrinsics like in GCC. While having one combined intrinsic is tempting, it is not natural because typically the trampoline initialization needs to be done in one function, and the result of adjust trampoline is needed in a different (nested) function. To get around this llvm-gcc hacks the nested function lowering code to insert an additional parent variable holding the adjust.trampoline result that can be accessed from the child function. Dragonegg doesn't have the luxury of tweaking GCC code, so it stored the result of adjust.trampoline in the memory GCC set aside for the trampoline itself (this is always available in the child function), and set up some new memory (using an alloca) to hold the trampoline. Unfortunately this breaks Go which allocates trampoline memory on the heap and wants to use it even after the parent has exited (!). Rather than doing even more hacks to get Go working, it seemed best to just use two intrinsics like in GCC. Patch mostly by Sanjoy Das. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139140 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-03Fix a truly heinous bug in DAGCombine related to AssertZext.Owen Anderson
If we have a chain of zext -> assert_zext -> zext -> use, the first zext would get simplified away because of the later zext, and then the later zext would get simplified away because of the assert. The solution is to teach SimplifyDemandedBits that assert_zext demands all of the high bits of its input, rather than only those demanded by its users. No testcase because the only example I have manifests as llvm-gcc miscompiling LLVM, and I haven't found a smaller case that reproduces this problem. Fixes <rdar://problem/10063365>. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139059 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-02Simplify by using isFullCopy().Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139019 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-02Darwin wants ctors/dtors to be ordered the other way round to linux.Duncan Sands
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139015 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-01Revert r131152, r129796, r129761. This code is currently consideredDan Gohman
to be unreliable on platforms which require memcpy calls, and it is complicating broader legalize cleanups. It is hoped that these cleanups will make memcpy byval easier to implement in the future. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@138977 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2025-07-21 05:35:05 +0000