aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen
AgeCommit message (Collapse)Author
2012-04-16Add a somewhat hacky heuristic to do something different from whole-loopChandler Carruth
rotation. When there is a loop backedge which is an unconditional branch, we will end up with a branch somewhere no matter what. Try placing this backedge in a fallthrough position above the loop header as that will definitely remove at least one branch from the loop iteration, where whole loop rotation may not. I haven't seen any benchmarks where this is important but loop-blocks.ll tests for it, and so this will be covered when I flip the default. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154812 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-16Tweak the loop rotation logic to check whether the loop is naturallyChandler Carruth
laid out in a form with a fallthrough into the header and a fallthrough out of the bottom. In that case, leave the loop alone because any rotation will introduce unnecessary branches. If either side looks like it will require an explicit branch, then the rotation won't add any, do it to ensure the branch occurs outside of the loop (if possible) and maximize the benefit of the fallthrough in the bottom. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154806 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-16Remove dead SD nodes after the combining pass. Fixes PR12201.Hal Finkel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154786 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-16Rewrite how machine block placement handles loop rotation.Chandler Carruth
This is a complex change that resulted from a great deal of experimentation with several different benchmarks. The one which proved the most useful is included as a test case, but I don't know that it captures all of the relevant changes, as I didn't have specific regression tests for each, they were more the result of reasoning about what the old algorithm would possibly do wrong. I'm also failing at the moment to craft more targeted regression tests for these changes, if anyone has ideas, it would be welcome. The first big thing broken with the old algorithm is the idea that we can take a basic block which has a loop-exiting successor and a looping successor and use the looping successor as the layout top in order to get that particular block to be the bottom of the loop after layout. This happens to work in many cases, but not in all. The second big thing broken was that we didn't try to select the exit which fell into the nearest enclosing loop (to which we exit at all). As a consequence, even if the rotation worked perfectly, it would result in one of two bad layouts. Either the bottom of the loop would get fallthrough, skipping across a nearer enclosing loop and thereby making it discontiguous, or it would be forced to take an explicit jump over the nearest enclosing loop to earch its successor. The point of the rotation is to get fallthrough, so we need it to fallthrough to the nearest loop it can. The fix to the first issue is to actually layout the loop from the loop header, and then rotate the loop such that the correct exiting edge can be a fallthrough edge. This is actually much easier than I anticipated because we can handle all the hard parts of finding a viable rotation before we do the layout. We just store that, and then rotate after layout is finished. No inner loops get split across the post-rotation backedge because we check for them when selecting the rotation. That fix exposed a latent problem with our exitting block selection -- we should allow the backedge to point into the middle of some inner-loop chain as there is no real penalty to it, the whole point is that it *won't* be a fallthrough edge. This may have blocked the rotation at all in some cases, I have no idea and no test case as I've never seen it in practice, it was just noticed by inspection. Finally, all of these fixes, and studying the loops they produce, highlighted another problem: in rotating loops like this, we sometimes fail to align the destination of these backwards jumping edges. Fix this by actually walking the backwards edges rather than relying on loopinfo. This fixes regressions on heapsort if block placement is enabled as well as lots of other cases where the previous logic would introduce an abundance of unnecessary branches into the execution. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154783 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-15When emulating vselect using OR/AND/XOR make sure to bitcast the result back ↵Nadav Rotem
to the original type. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154764 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-13misched: Added CanHandleTerminators.Andrew Trick
This is a special flag for targets that really want their block terminators in the DAG. The default scheduler cannot handle this correctly, so it becomes the specialized scheduler's responsibility to schedule terminators. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154712 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-13Reduce malloc traffic in DwarfAccelTableBenjamin Kramer
- Don't copy offsets into HashData, the underlying vector won't change once the table is finalized. - Allocate HashData and HashDataContents in a BumpPtrAllocator. - Allocate string map entries in the same allocator. - Random cleanups. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154694 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-12HexagonPacketizer patch.Sirish Pande
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154616 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-11Reapply 154397. Original message:Nadav Rotem
Fix a dagcombine optimization which assumes that the vsetcc result type is always of the same size as the compared values. This is ture for SSE/AVX/NEON but not for all targets. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154490 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-11Fix an overly indented line. Remove an 'else' after an 'if' that returns.Craig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154479 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-11Inline implVisitAluOverflow by introducing a nested switch to convert the ↵Craig Topper
intrinsic to an nodetype. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154478 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-11Optimize code a bit by calling push_back only once in some loops. Reduces ↵Craig Topper
compiled code size a bit. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154473 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-11Tweak MachineLICM heuristics for cheap instructions.Jakob Stoklund Olesen
Allow cheap instructions to be hoisted if they are register pressure neutral or better. This happens if the instruction is the last loop use of another virtual register. Only expensive instructions are allowed to increase loop register pressure. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154455 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-11Only check for PHI uses inside the current loop.Jakob Stoklund Olesen
Hoisting a value that is used by a PHI in the loop will introduce a copy because the live range is extended to cross the PHI. The same applies to PHIs in exit blocks. Also use this opportunity to make HasLoopPHIUse() non-recursive. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154454 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-10Move the constant-folding support for FP_ROUND in SelectionDAG from the ↵Owen Anderson
one-operand version of getNode() to the two-operand version, since it became a two-operand node at sound point. Zap a testcase that this allows us to completely fold away. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154447 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-10Add a comment noting that the fdiv -> fmul conversion won't generateDuncan Sands
multiplication by a denormal, and some tests checking that. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154431 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-10To ensure that we have more accurate line information for a blockEric Christopher
don't elide the branch instruction if it's the only one in the block, otherwise it's ok. PR9796 and rdar://11215207 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154417 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-10Revert r154397, which was causing make check failures on the buildbots.Owen Anderson
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154414 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-10Fix a dagcombine optimization which assumes that the vsetcc result type is ↵Nadav Rotem
always of the same size as the compared values. This is ture for SSE/AVX/NEON but not for all targets. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154397 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-10Make a somewhat subtle change in the logic of block placement. SometimesChandler Carruth
the loop header has a non-loop predecessor which has been pre-fused into its chain due to unanalyzable branches. In this case, rotating the header into the body of the loop in order to place a loop exit at the bottom of the loop is a Very Bad Idea as it makes the loop non-contiguous. I'm working on a good test case for this, but it's a bit annoynig to craft. I should get one shortly, but I'm submitting this now so I can begin the (lengthy) performance analysis process. An initial run of LNT looks really, really good, but there is too much noise there for me to trust it much. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154395 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-10Transform div to mul with reciprocal only when fp imm is legal.Anton Korobeynikov
This fixes PR12516 and uncovers one weird problem in legalize (workarounded) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154394 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-10Make the code slightly more palatable.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154378 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-10Fix a long standing tail call optimization bug. When a libcall is emittedEvan Cheng
legalizer always use the DAG entry node. This is wrong when the libcall is emitted as a tail call since it effectively folds the return node. If the return node's input chain is not the entry (i.e. call, load, or store) use that as the tail call input chain. PR12419 rdar://9770785 rdar://11195178 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154370 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-10Don't try to zExt just to check if an integer constant is zero, it mightRafael Espindola
not fit in a i64. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154364 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-09Have TargetLowering::getPICJumpTableRelocBase return a node that points to theAkira Hatanaka
GOT if jump table uses 64-bit gp-relative relocation. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154341 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-09Patch r153892 for PR11861 apparently broke an external project (see PR12493).Lang Hames
This patch restores TwoAddressInstructionPass's pre-r153892 behaviour when rescheduling instructions in TryInstructionTransform. Hopefully this will fix PR12493. To refix PR11861, lowering of INSERT_SUBREGS is deferred until after the copy that unties the operands is emitted (this seems to be a more appropriate fix for that issue anyway). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154338 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-09Pattern match a setcc of boolean value with 0 as a truncate.Rafael Espindola
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154322 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-09Remove unnecessary type check when combining and/or/xor of swizzles. Move ↵Craig Topper
some checks to allow better early out. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154309 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-09Remove unnecessary 'else' on an 'if' that always returnsCraig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154308 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-09Optimize code slightly. No functionality change.Craig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154307 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-09Replace some explicit checks with asserts for conditions that should never ↵Craig Topper
happen. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154305 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-08Optimize code a bit. No functional change intended.Craig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154299 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-08Silence sign-compare warning.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154297 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-08Only have codegen turn fdiv by a constant into fmul by the reciprocalDuncan Sands
when -ffast-math, i.e. don't just always do it if the reciprocal can be formed exactly. There is already an IR level transform that does that, and it does it more carefully. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154296 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-08Simplify code that tries to do vector extracts for shuffles when the mask ↵Craig Topper
width and the input vector widths don't match. No need to check the min and max are in range before calculating the start index. The range check after having the start index is sufficient. Also no need to check for an extract from the beginning differently. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154295 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-08Move the TLSModel information into the TargetMachine rather than hidingChandler Carruth
in TargetLowering. There was already a FIXME about this location being odd. The interface is simplified as a consequence. This will also make it easier to change TLS models when compiling with PIE. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154292 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-08Remove an over zealous assert. The assert was trying to catch placesChandler Carruth
where a chain outside of the loop block-set ended up in the worklist for scheduling as part of the contiguous loop. However, asserting the first block in the chain is in the loop-set isn't a valid check -- we may be forced to drag a chain into the worklist due to one block in the chain being part of the loop even though the first block is *not* in the loop. This occurs when we have been forced to form a chain early due to un-analyzable branches. No test case here as I have no idea how to even begin reducing one, and it will be hopelessly fragile. We have to somehow end up with a loop header of an inner loop which is a successor of a basic block with an unanalyzable pair of branch instructions. Ow. Self-host triggers it so it is unlikely it will regress. This at least gets block placement back to passing selfhost and the test suite. There are still a lot of slowdown that I don't like coming out of block placement, although there are now also a lot of speedups. =[ I'm seeing swings in both directions up to 10%. I'm going to try to find time to dig into this and see if we can turn this on for 3.1 as it does a really good job of cleaning up after some loops that degraded with the inliner changes. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154287 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-08Add a debug-only 'dump' method to the BlockChain structure to easeChandler Carruth
debugging. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154286 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-07Turn avx2 vinserti128 intrinsic calls into INSERT_SUBVECTOR DAG nodes and ↵Craig Topper
remove patterns for selecting the intrinsic. Similar was already done for avx1. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154272 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-07Remove 'else' after 'if' that ends in return.Craig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154267 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-071. Remove the part of r153848 which optimizes shuffle-of-shuffle into a newNadav Rotem
shuffle node because it could introduce new shuffle nodes that were not supported efficiently by the target. 2. Add a more restrictive shuffle-of-shuffle optimization for cases where the second shuffle reverses the transformation of the first shuffle. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154266 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-07Convert floating point division by a constant into multiplication by theDuncan Sands
reciprocal if converting to the reciprocal is exact. Do it even if inexact if -ffast-math. This substantially speeds up ac.f90 from the polyhedron benchmarks. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154265 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-05Patch to set is_stmt a little better for prologue lines in a function.Eric Christopher
This enables debuggers to see what are interesting lines for a breakpoint rather than any line that starts a function. rdar://9852092 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154120 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-05Don't break the IV update in TLI::SimplifySetCC().Jakob Stoklund Olesen
LSR always tries to make the ICmp in the loop latch use the incremented induction variable. This allows the induction variable to be kept in a single register. When the induction variable limit is equal to the stride, SimplifySetCC() would break LSR's hard work by transforming: (icmp (add iv, stride), stride) --> (cmp iv, 0) This forced us to use lea for the IC update, preventing the simpler incl+cmp. <rdar://problem/7643606> <rdar://problem/11184260> git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154119 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-05Treat f16 the same as f80/f128 for the purposes of generating constants ↵Owen Anderson
during instruction selection. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154113 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-04REG_SEQUENCE expansion to COPY instructions wasn't taking account of sub ↵Pete Cooper
register indices on the source registers. No simple test case git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154051 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-04f16 FREM can now be legalized by promoting to f32Pete Cooper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154039 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-04Remove spurious debug output.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154032 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-04Always compute all the bits in ComputeMaskedBits.Rafael Espindola
This allows us to keep passing reduced masks to SimplifyDemandedBits, but know about all the bits if SimplifyDemandedBits fails. This allows instcombine to simplify cases like the one in the included testcase. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154011 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-04Remove default case from switch that was already covering all cases.Craig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@153996 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-09-21 09:29:32 +0000