aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen
AgeCommit message (Collapse)Author
2011-11-22Fix an obvious omission in the SelectionDAGBuilder where we wereChandler Carruth
dropping weights on the floor for invokes. This was impeding my writing further test cases for invoke when interacting with probabilities and block placement. No test case as there doesn't appear to be a way to test this stuff. =/ Suggestions for a test case of course welcome. I hope to be able to add test cases that indirectly cover this eventually by adding probabilities to the exceptional edge and reordering blocks as a result. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@145060 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-22If a register is both an early clobber and part of a tied use, handle the useRafael Espindola
before the clobber so that we copy the value if needed. Fixes pr11415. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@145056 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-20The logic for breaking the CFG in the presence of hot successors didn'tChandler Carruth
properly account for the *global* probability of the edge being taken. This manifested as a very large number of unconditional branches to blocks being merged against the CFG even though they weren't particularly hot within the CFG. The fix is to check whether the edge being merged is both locally hot relative to other successors for the source block, and globally hot compared to other (unmerged) predecessors of the destination block. This introduces a new crasher on GCC single-source, but it's currently behind a flag, and Ben has offered to work on the reduction. =] git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@145010 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-19Move the handling of unanalyzable branches out of the loop-driven chainChandler Carruth
formation phase and into the initial walk of the basic blocks. We essentially pre-merge all blocks where unanalyzable fallthrough exists, as we won't be able to update the terminators effectively after any reorderings. This is quite a bit more principled as there may be CFGs where the second half of the unanalyzable pair has some analyzable predecessor that gets placed first. Then it may get placed next, implicitly breaking the unanalyzable branch even though we never even looked at the part that isn't analyzable. I've included a test case that triggers this (thanks Benjamin yet again!), and I'm hoping to synthesize some more general ones as I dig into related issues. Also, to make this new scheme work we have to be able to handle branches into the middle of a chain, so add this check. We always fallback on the incoming ordering. Finally, this starts to really underscore a known limitation of the current implementation -- we don't consider broken predecessors when merging successors. This can caused major missed opportunities, and is something I'm planning on looking at next (modulo more bug reports). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144994 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-17DISubrange supports unsigned lower/upper array bounds, so let's not fake it ↵Devang Patel
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. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144937 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-17When fast iseling a GEP, accumulate the offset rather than emitting a series ofChad Rosier
ADDs. MaxOffs is used as a threshold to limit the size of the offset. Tradeoffs being: (1) If we can't materialize the large constant then we'll cause fast-isel to bail. (2) Too large of an offset can't be directly encoded in the ADD resulting in a MOV+ADD. Generally not a bad thing because otherwise we would have had ADD+ADD, but on Thumb this turns into a MOVS+MOVT+ADD. Working on a fix for that. (3) Conversely, too low of a threshold we'll miss opportunities to coalesce ADDs. rdar://10412592 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144886 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16Make sure to replace the chain properly when DAGCombining a ↵Eli Friedman
LOAD+EXTRACT_VECTOR_ELT into a single LOAD. Fixes PR10747/PR11393. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144863 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16Add fast-isel stats to determine who's doing all the work, the Chad Rosier
target-independent selector or the target-specific selector. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144833 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16Fix the stats collection for fast-isel. The failed count was only accountingChad Rosier
for a single miss and not all predecessor instructions that get selected by the selection DAG instruction selector. This is still not exact (e.g., over states misses when folded/dead instructions are present), but it is a step in the right direction. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144832 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16Disable expensive two-address optimizations at -O0. rdar://10453055Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144806 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16Disable the assertion again. Looks like fastisel is still generating bad ↵Evan Cheng
kill markers. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144804 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16Sink codegen optimization level into MCCodeGenInfo along side relocation modelEvan Cheng
and code model. This eliminates the need to pass OptLevel flag all over the place and makes it possible for any codegen pass to use this information. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144788 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16Record landing pads with a SmallSetVector to avoid multiple entries.Bob Wilson
There may be many invokes that share one landing pad, and the previous code would record the landing pad once for each invoke. Besides the wasted effort, a pair of volatile loads gets inserted every time the landing pad is processed. The rest of the code can get optimized away when a landing pad is processed repeatedly, but the volatile loads remain, resulting in code like: LBB35_18: Ltmp483: ldr r2, [r7, #-72] ldr r2, [r7, #-68] ldr r2, [r7, #-72] ldr r2, [r7, #-68] ldr r2, [r7, #-72] ldr r2, [r7, #-68] ldr r2, [r7, #-72] ldr r2, [r7, #-68] ldr r2, [r7, #-72] ldr r2, [r7, #-68] ldr r2, [r7, #-72] ldr r2, [r7, #-68] ldr r2, [r7, #-72] ldr r2, [r7, #-68] ldr r2, [r7, #-72] ldr r2, [r7, #-68] ldr r4, [r7, #-72] ldr r2, [r7, #-68] git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144787 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16Update the SP in the SjLj jmpbuf whenever it changes. <rdar://problem/10444602>Bob Wilson
This same basic code was in the older version of the SjLj exception handling, but it was removed in the recent revisions to that code. It needs to be there. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144782 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16Revert r144568 now that r144730 has fixed the fast-isel kill marker bug.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144776 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16If the 2addr instruction has other kills, don't move it below any other uses ↵Evan Cheng
since we don't want to extend other live ranges. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144772 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16RescheduleKillAboveMI() must backtrack to before the rescheduled DBG_VALUE ↵Evan Cheng
instructions. rdar://10451185 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144771 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16Process all uses first before defs to accurately capture register liveness. ↵Evan Cheng
rdar://10449480 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144770 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16CONCAT_VECTORS can have more than two operands. PR11389.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144768 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16Add a couple asserts so it will be easier to debug if we accidentally pass ↵Eli Friedman
indexed loads/stores to the legalizer. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144767 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-16Rename MVT::untyped to MVT::Untyped to match similar nomenclature.Owen Anderson
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144747 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-15Stabilize the output of the dwarf accelerator tables. Fixes a comparisonEric Christopher
failure during bootstrap with it turned on. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144731 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-15GEPs with all zero indices are trivially coalesced by fast-isel. For example,Chad Rosier
%arrayidx135 = getelementptr inbounds [4 x [4 x [4 x [4 x i32]]]]* %M0, i32 0, i64 0 %arrayidx136 = getelementptr inbounds [4 x [4 x [4 x i32]]]* %arrayidx135, i32 0, i64 %idxprom134 Prior to this commit, the GEP instruction that defines %arrayidx136 thought that %arrayidx135 was a trivial kill. The GEP that defines %arrayidx135 doesn't generate any code and thus %M0 gets folded into the second GEP. Thus, we need to look through GEPs with all zero indices. rdar://10443319 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144730 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-15Added custom lowering for load->dec->store sequence in x86 when the EFLAGS ↵Pete Cooper
registers is used by later instructions. Only done for DEC64m right now. Fixes <rdar://problem/6172640> git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144705 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-15Insert modified DBG_VALUE into LiveDbgValueMap. Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144696 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-15We currently use a callback to handle an IL pass deleting a BB that stillRafael Espindola
has a reference to it. Unfortunately, that doesn't work for codegen passes since we don't get notified of MBB's being deleted (the original BB stays). Use that fact to our advantage and after printing a function, check if any of the IL BBs corresponds to a symbol that was not printed. This fixes pr11202. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144674 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-15Remove all remaining uses of Value::getNameStr().Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144648 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-15Twinify GraphWriter a little bit.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144647 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-15Check all overlaps when looking for used registers.Jakob Stoklund Olesen
A function using any RC alias is enough to enable the ExeDepsFix pass. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144636 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-15Make use of MachinePointerInfo::getFixedStack.Jay Foad
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144635 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-15Remove some unnecessary includes of PseudoSourceValue.h.Jay Foad
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144634 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-15Set SeenStore to true to prevent loads from being moved; also eliminates a ↵Evan Cheng
non-deterministic behavior. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144628 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-15Rather than trying to use the loop block sequence *or* the functionChandler Carruth
block sequence when recovering from unanalyzable control flow constructs, *always* use the function sequence. I'm not sure why I ever went down the path of trying to use the loop sequence, it is fundamentally not the correct sequence to use. We're trying to preserve the incoming layout in the cases of unreasonable control flow, and that is only encoded at the function level. We already have a filter to select *exactly* the sub-set of blocks within the function that we're trying to form into a chain. The resulting code layout is also significantly better because of this. In several places we were ending up with completely unreasonable control flow constructs due to the ordering chosen by the loop structure for its internal storage. This change removes a completely wasteful vector of basic blocks, saving memory allocation in the common case even though it costs us CPU in the fairly rare case of unnatural loops. Finally, it fixes the latest crasher reduced out of GCC's single source. Thanks again to Benjamin Kramer for the reduction, my bugpoint skills failed at it. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144627 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-15Break false dependencies before partial register updates.Jakob Stoklund Olesen
Two new TargetInstrInfo hooks lets the target tell ExecutionDepsFix about instructions with partial register updates causing false unwanted dependencies. The ExecutionDepsFix pass will break the false dependencies if the updated register was written in the previoius N instructions. The small loop added to sse-domains.ll runs twice as fast with dependency-breaking instructions inserted. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144602 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-15Track register ages more accurately.Jakob Stoklund Olesen
Keep track of the last instruction to define each register individually instead of per DomainValue. This lets us track more accurately when a register was last written. Also track register ages across basic blocks. When entering a new basic block, use the least stale predecessor def as a worst case estimate for register age. The register age is used to arbitrate between conflicting domains. The most recently defined register wins. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144601 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-14Avoid dereferencing off the beginning of lists.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144569 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-14At -O0, multiple uses of a virtual registers in the same BB are being markedEvan Cheng
"kill". This looks like a bug upstream. Since that's going to take some time to understand, loosen the assertion and disable the optimization when multiple kills are seen. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144568 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-14Teach two-address pass to re-schedule two-address instructions (or the killEvan Cheng
instructions of the two-address operands) in order to avoid inserting copies. This fixes the few regressions introduced when the two-address hack was disabled (without regressing the improvements). rdar://10422688 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144559 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-14Fix early-clobber handling in shrinkToUses.Jakob Stoklund Olesen
I broke this in r144515, it affected most ARM testers. <rdar://problem/10441389> git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144547 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-14It helps to deallocate memory as well as allocate it. =] This actuallyChandler Carruth
cleans up all the chains allocated during the processing of each function so that for very large inputs we don't just grow memory usage without bound. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144533 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-14Remove an over-eager assert that was firing on one of the ARM regressionChandler Carruth
tests when I forcibly enabled block placement. It is apparantly possible for an unanalyzable block to fallthrough to a non-loop block. I don't actually beleive this is correct, I believe that 'canFallThrough' is returning true needlessly for the code construct, and I've left a bit of a FIXME on the verification code to try to track down why this is coming up. Anyways, removing the assert doesn't degrade the correctness of the algorithm. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144532 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-14Begin chipping away at one of the biggest quadratic-ish behaviors inChandler Carruth
this pass. We're leaving already merged blocks on the worklist, and scanning them again and again only to determine each time through that indeed they aren't viable. We can instead remove them once we're going to have to scan the worklist. This is the easy way to implement removing them. If this remains on the profile (as I somewhat suspect it will), we can get a lot more clever here, as the worklist's order is essentially irrelevant. We can use swapping and fold the two loops to reduce overhead even when there are many blocks on the worklist but only a few of them are removed. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144531 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-14Under the hood, MBPI is doing a linear scan of every successor everyChandler Carruth
time it is queried to compute the probability of a single successor. This makes computing the probability of every successor of a block in sequence... really really slow. ;] This switches to a linear walk of the successors rather than a quadratic one. One of several quadratic behaviors slowing this pass down. I'm not really thrilled with moving the sum code into the public interface of MBPI, but I don't (at the moment) have ideas for a better interface. My direction I'm thinking in for a better interface is to have MBPI actually retain much more state and make *all* of these queries cheap. That's a lot of work, and would require invasive changes. Until then, this seems like the least bad (ie, least quadratic) solution. Suggestions welcome. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144530 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-14Reuse the logic in getEdgeProbability within getHotSucc in order toChandler Carruth
correctly handle blocks whose successor weights sum to more than UINT32_MAX. This is slightly less efficient, but the entire thing is already linear on the number of successors. Calling it within any hot routine is a mistake, and indeed no one is calling it. It also simplifies the code. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144527 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-14Fix an overflow bug in MachineBranchProbabilityInfo. This pass relied onChandler Carruth
the sum of the edge weights not overflowing uint32, and crashed when they did. This is generally safe as BranchProbabilityInfo tries to provide this guarantee. However, the CFG can get modified during codegen in a way that grows the *sum* of the edge weights. This doesn't seem unreasonable (imagine just adding more blocks all with the default weight of 16), but it is hard to come up with a case that actually triggers 32-bit overflow. Fortuately, the single-source GCC build is good at this. The solution isn't very pretty, but its no worse than the previous code. We're already summing all of the edge weights on each query, we can sum them, check for an overflow, compute a scale, and sum them again. I've included a *greatly* reduced test case out of the GCC source that triggers it. It's a pretty lame test, as it clearly is just barely triggering the overflow. I'd like to have something that is much more definitive, but I don't understand the fundamental pattern that triggers an explosion in the edge weight sums. The buggy code is duplicated within this file. I'll colapse them into a single implementation in a subsequent commit. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144526 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-14Use getVNInfoBefore() when it makes sense.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144517 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-14Teach machine block placement to cope with unnatural loops. These don'tChandler Carruth
get loop info structures associated with them, and so we need some way to make forward progress selecting and placing basic blocks. The technique used here is pretty brutal -- it just scans the list of blocks looking for the first unplaced candidate. It keeps placing blocks like this until the CFG becomes tractable. The cost is somewhat unfortunate, it requires allocating a vector of all basic block pointers eagerly. I have some ideas about how to simplify and optimize this, but I'm trying to get the logic correct first. Thanks to Benjamin Kramer for the reduced test case out of GCC. Sadly there are other bugs that GCC is tickling that I'm reducing and working on now. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144516 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-13Use kill slots instead of the previous slot in shrinkToUses.Jakob Stoklund Olesen
It's more natural to use the actual end points. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144515 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-13Cleanup some 80-columns violations and poor formatting. These snuck byChandler Carruth
when I was reading through the code for style. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144513 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-13Terminate all dead defs at the dead slot instead of the 'next' slot.Jakob Stoklund Olesen
This makes no difference for normal defs, but early clobber dead defs now look like: [Slot_EarlyClobber; Slot_Dead) instead of: [Slot_EarlyClobber; Slot_Register). Live ranges for normal dead defs look like: [Slot_Register; Slot_Dead) as before. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144512 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-09-23 16:44:07 +0000