summaryrefslogtreecommitdiffstats
path: root/llvm/lib
Commit message (Collapse)AuthorAgeFilesLines
* [CGP] Fix the detection of trivial case for addressing modeSerguei Katkov2017-10-311-10/+9
| | | | | | | The address can be presented as a bitcast of baseReg. In this case it is still trivial but OriginalValue != baseReg. llvm-svn: 316980
* [IRCE][NFC] Rename fields of InductiveRangeCheckMax Kazantsev2017-10-311-23/+23
| | | | | | | | | | Rename `Offset`, `Scale`, `Length` into `Begin`, `Step`, `End` respectively to make naming of similar entities for Ranges and Range Checks more consistent. Differential Revision: https://reviews.llvm.org/D39414 llvm-svn: 316979
* [X86] Make AVX512_512_SET0 XMM16-31 lower to 128-bit XOR when AVX512VL is ↵Craig Topper2017-10-311-13/+2
| | | | | | | | | | enabled. Use 128-bit VLX instruction when VLX is enabled. Unfortunately, this weakens our ability to do domain fixing when AVX512DQ is not enabled, but it is consistent with our 256-bit behavior. Maybe we should add custom handling to domain fixing to allow EVEX integer XOR/AND/OR/ANDN to switch to VEX encoded fp instructions if the high registers aren't being used? llvm-svn: 316978
* [NFC] Get rid of variables used in assert onlyMax Kazantsev2017-10-311-6/+6
| | | | llvm-svn: 316977
* [IndVarSimplify] Simplify code using preheader assumptionPhilip Reames2017-10-311-44/+6
| | | | | | | | As noted in the nice block comment, the previous code didn't actually handle multi-entry loops correctly, it just assumed SCEV didn't analyze such loops. Given SCEV has comments to the contrary, that seems a bit suspect. More importantly, the pass actually requires loopsimplify form which ensures a loop-preheader is available. Remove the excessive generaility and shorten the code greatly. Note that we do successfully analyze many multi-entry loops, but we do so by converting them to single entry loops. See the added test case. llvm-svn: 316976
* Reapply "[GVN] Prevent LoadPRE from hoisting across instructions that don't ↵Max Kazantsev2017-10-311-0/+97
| | | | | | | | | | | | | | | | | | pass control flow to successors" This patch fixes the miscompile that happens when PRE hoists loads across guards and other instructions that don't always pass control flow to their successors. PRE is now prohibited to hoist across such instructions because there is no guarantee that the load standing after such instruction is still valid before such instruction. For example, a load from under a guard may be invalid before the guard in the following case: int array[LEN]; ... guard(0 <= index && index < LEN); use(array[index]); Differential Revision: https://reviews.llvm.org/D37460 llvm-svn: 316975
* [SimplifyIndVar] Extract out invariant expression handlingPhilip Reames2017-10-311-82/+107
| | | | | | | | Previously, the code returned early from the *function* when it couldn't find a free expansion, it should be returning from the *transform*. I don't have a test case, noticed this via inspection. As a follow up, I'm going to revisit the logic in the extract function. I think that essentially the whole helper routine can be replaced with SCEVExpander, but I wanted to do that in a series of separate commits. llvm-svn: 316974
* [X86] Clang-format some code. NFCCraig Topper2017-10-311-2/+8
| | | | llvm-svn: 316973
* Undo accidental commitPhilip Reames2017-10-312-245/+82
| | | | | | These files shouldn't have been submitted in 316967 llvm-svn: 316968
* [CGP] Fix crash on i96 bit multiplyPhilip Reames2017-10-303-83/+246
| | | | | | | | Issue found by llvm-isel-fuzzer on OSS fuzz, https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=3725 If anyone actually cares about > 64 bit arithmetic, there's a lot more to do in this area. There's a bunch of obviously wrong code in the same function. I don't have the time to fix all of them and am just using this to understand what the workflow for fixing fuzzer cases might look like. llvm-svn: 316967
* Fix unused variable warnings. NFCI.Simon Pilgrim2017-10-301-3/+0
| | | | llvm-svn: 316964
* [SelectionDAG] Tidyup computeKnownBits extension/truncation cases. NFCI.Simon Pilgrim2017-10-301-17/+4
| | | | | | We don't need to extend/truncate the Known structure before calling computeKnownBits - it will reset at the start of the function. llvm-svn: 316962
* [AArch64]: range loopify frame-loweringJaved Absar2017-10-301-2/+2
| | | | llvm-svn: 316960
* InferAddressSpaces: Fix bug about replacing addrspacecastYaxun Liu2017-10-301-0/+5
| | | | | | | | | | | | | | InferAddressSpaces assumes the pointee type of addrspacecast is the same as the operand, which is not always true and causes invalid IR. This bug cause build failure in HCC. This patch fixes that. Differential Revision: https://reviews.llvm.org/D39432 llvm-svn: 316957
* [X86] Add AVX512 support to fast isel's X86ChooseCmpOpcode.Craig Topper2017-10-301-2/+3
| | | | llvm-svn: 316955
* [NewGVN] Stop assuming PHI args ordering when looking at phi-of-ops.Davide Italiano2017-10-301-1/+1
| | | | | | | | It's not guaranteed. There's a bug open to sort them in predecessor order, but it won't happen anytime soon. In the meanwhile, passes will have to do an O(#preds) scan. Such is life. llvm-svn: 316953
* Revert "[PowerPC] Try to simplify a Swap if it feeds a Splat"Stefan Pintilie2017-10-301-47/+0
| | | | | | | | | | Revert r316478. A test case has failed. Will recommit this change once we find and fix the failure. This reverts commit 7c330fabaedaba3d02c58bc3cc1198896c895f34. llvm-svn: 316952
* Create instruction classes for identifying any atomicity of memory ↵Daniel Neilson2017-10-304-14/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | intrinsic. (NFC) Summary: For reference, see: http://lists.llvm.org/pipermail/llvm-dev/2017-August/116589.html This patch fleshes out the instruction class hierarchy with respect to atomic and non-atomic memory intrinsics. With this change, the relevant part of the class hierarchy becomes: IntrinsicInst -> MemIntrinsicBase (methods-only class) -> MemIntrinsic (non-atomic intrinsics) -> MemSetInst -> MemTransferInst -> MemCpyInst -> MemMoveInst -> AtomicMemIntrinsic (atomic intrinsics) -> AtomicMemSetInst -> AtomicMemTransferInst -> AtomicMemCpyInst -> AtomicMemMoveInst -> AnyMemIntrinsic (both atomicities) -> AnyMemSetInst -> AnyMemTransferInst -> AnyMemCpyInst -> AnyMemMoveInst This involves some class renaming: ElementUnorderedAtomicMemCpyInst -> AtomicMemCpyInst ElementUnorderedAtomicMemMoveInst -> AtomicMemMoveInst ElementUnorderedAtomicMemSetInst -> AtomicMemSetInst A script for doing this renaming in downstream trees is included below. An example of where the Any* classes should be used in LLVM is when reasoning about the effects of an instruction (ex: aliasing). --- Script for renaming AtomicMem* classes: PREFIXES="[<,([:space:]]" CLASSES="MemIntrinsic|MemTransferInst|MemSetInst|MemMoveInst|MemCpyInst" SUFFIXES="[;)>,[:space:]]" REGEX="(${PREFIXES})ElementUnorderedAtomic(${CLASSES})(${SUFFIXES})" REGEX2="visitElementUnorderedAtomic(${CLASSES})" FILES=$( grep -E "(${REGEX}|${REGEX2})" -r . | tr ':' ' ' | awk '{print $1}' | sort | uniq ) SED_SCRIPT="s~${REGEX}~\1Atomic\2\3~g" SED_SCRIPT2="s~${REGEX2}~visitAtomic\1~g" for f in $FILES; do echo "Processing: $f" sed -i ".bak" -E "${SED_SCRIPT};${SED_SCRIPT2};${EA_SED_SCRIPT};${EA_SED_SCRIPT2}" $f done Reviewers: sanjoy, deadalnix, apilipenko, anna, skatkov, mkazantsev Reviewed By: sanjoy Subscribers: hfinkel, jholewinski, arsenm, sdardis, nhaehnle, JDevlieghere, javed.absar, llvm-commits Differential Revision: https://reviews.llvm.org/D38419 llvm-svn: 316950
* [GVNHoist] Fix non-deterministic sort order of PHIs for identical instructionsMandeep Singh Grang2017-10-301-1/+1
| | | | | | | | | | | | | | Summary: This fixes failure in Transforms/GVNHoist/hoist.ll uncovered by D39245. Reviewers: hiraditya, spop, dberlin Reviewed By: dberlin Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D39410 llvm-svn: 316949
* [SelectionDAG] Add VSELECT demanded elts support to computeKnownBits Simon Pilgrim2017-10-301-4/+4
| | | | llvm-svn: 316947
* [SelectionDAG] Add VSELECT support to computeKnownBits Simon Pilgrim2017-10-301-0/+1
| | | | llvm-svn: 316944
* [SelectionDAG] Add SELECT demanded elts support to ComputeNumSignBitsSimon Pilgrim2017-10-301-4/+5
| | | | llvm-svn: 316933
* [MC] Split out register def/use idx calls to make debugging simpler. NFCI.Simon Pilgrim2017-10-301-3/+4
| | | | llvm-svn: 316927
* [X86][AVX512] Adding a pattern for broadcastm intrinsic.Jina Nahias2017-10-301-0/+64
| | | | | | | Differential Revision: https://reviews.llvm.org/D38312 Change-Id: I71c8605a8e4c98013ef25289694afc5cfd46bb0b llvm-svn: 316921
* Move isDSOLocal check and add a comment.Rafael Espindola2017-10-301-2/+12
| | | | llvm-svn: 316920
* [PPC CodeGen] Fix the bitreverse.i64 intrinsic.Fangrui Song2017-10-301-71/+34
| | | | | | | | | | | | Summary: The two 32-bit words were swapped. Update a test omitted in reverted r316270. Reviewers: jtony, aaron.ballman Subscribers: nemanjai, kbarton Differential Revision: https://reviews.llvm.org/D39163 llvm-svn: 316916
* [X86] Make sure we don't create locked inc/dec instructions when the carry ↵Craig Topper2017-10-304-19/+84
| | | | | | | | | | | | | | | | | | | | | | | flag is being used. Summary: INC/DEC don't update the carry flag so we need to make sure we don't try to use it. This patch introduces new X86ISD opcodes for locked INC/DEC. Teaches lowerAtomicArithWithLOCK to emit these nodes if INC/DEC is not slow or the function is being optimized for size. An additional flag is added that allows the INC/DEC to be disabled if the caller determines that the carry flag is being requested. The test_sub_1_cmp_1_setcc_ugt test is currently showing this bug. The other test case changes are recovering cases that were regressed in r316860. This should fully fix PR35068 finishing the fix started in r316860. Reviewers: RKSimon, zvi, spatel Reviewed By: zvi Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D39411 llvm-svn: 316913
* [X86] Remove AVX512 early out from X86FastISel::X86SelectCmp.Craig Topper2017-10-301-3/+0
| | | | | | This shouldn't be needed anymore since i1 isn't a legal type. llvm-svn: 316912
* [AMDGPU] Emit metadata for hidden arguments for kernel enqueueYaxun Liu2017-10-302-4/+51
| | | | | | | | | | | | | Identifies kernels which performs device side kernel enqueues and emit metadata for the associated hidden kernel arguments. Such kernels are marked with calls-enqueue-kernel function attribute by AMDGPUOpenCLEnqueueKernelLowering pass and later on hidden kernel arguments metadata HiddenDefaultQueue and HiddenCompletionAction are emitted for them. Differential Revision: https://reviews.llvm.org/D39255 llvm-svn: 316907
* [CodeGen][ExpandMemcmp] Allow memcmp to expand to vector loads (2).Clement Courbet2017-10-307-38/+74
| | | | | | | | | | | | - Targets that want to support memcmp expansions now return the list of supported load sizes. - Expansion codegen does not assume that all power-of-two load sizes smaller than the max load size are valid. For examples, this is not the case for x86(32bit)+sse2. Fixes PR34887. llvm-svn: 316905
* [Hexagon] Allow the RDF optimizations to be run in .mir testcasesKrzysztof Parzyszek2017-10-302-5/+7
| | | | llvm-svn: 316904
* [GlobalISel|ARM] : Allow legalizing G_FSUBJaved Absar2017-10-303-8/+13
| | | | | | | | Adding support for VSUB. Reviewed by: @rovka Differential Revision: https://reviews.llvm.org/D39261 llvm-svn: 316902
* Invalid used of 'w' suffix on push and pop using 64-bit register.Andrew V. Tischenko2017-10-301-1/+3
| | | | | | Differential Revision: https://reviews.llvm.org/D38626 llvm-svn: 316898
* Revert "[X86][AVX512] Adding a pattern for broadcastm intrinsic."Jina Nahias2017-10-301-64/+0
| | | | | | | This reverts commit r316890. Change-Id: I683cceee9848ef309b452293086b1f26a941950d llvm-svn: 316894
* Recommit r315288: [SCCP] Propagate integer range info for parameters in IPSCCP.Florian Hahn2017-10-301-7/+95
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This version of the patch includes a fix addressing a stage2 LTO buildbot failure and addressed some additional nits. Original commit message: This updates the SCCP solver to use of the ValueElement lattice for parameters, which provides integer range information. The range information is used to remove unneeded icmp instructions. For the following function, f() can be optimized to ret i32 2 with this change source_filename = "sccp.c" target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128" target triple = "x86_64-unknown-linux-gnu" ; Function Attrs: norecurse nounwind readnone uwtable define i32 @main() local_unnamed_addr #0 { entry: %call = tail call fastcc i32 @f(i32 1) %call1 = tail call fastcc i32 @f(i32 47) %add3 = add nsw i32 %call, %call1 ret i32 %add3 } ; Function Attrs: noinline norecurse nounwind readnone uwtable define internal fastcc i32 @f(i32 %x) unnamed_addr #1 { entry: %c1 = icmp sle i32 %x, 100 %cmp = icmp sgt i32 %x, 300 %. = select i1 %cmp, i32 1, i32 2 ret i32 %. } attributes #1 = { noinline } Reviewers: davide, sanjoy, efriedma, dberlin Reviewed By: davide, dberlin Subscribers: mcrosier, gberry, mssimpso, dberlin, llvm-commits Differential Revision: https://reviews.llvm.org/D36656 llvm-svn: 316891
* [X86][AVX512] Adding a pattern for broadcastm intrinsic.Jina Nahias2017-10-301-0/+64
| | | | | | | Differential Revision: https://reviews.llvm.org/D38312 Change-Id: I6551fb13879e098aed74de410e29815cf37d9ab5 llvm-svn: 316890
* [IRCE][NFC] Store Length as SCEV in RangeCheck instead of ValueMax Kazantsev2017-10-301-6/+6
| | | | llvm-svn: 316889
* Revert r316887 to fix buildbot failures.Florian Hahn2017-10-301-93/+7
| | | | llvm-svn: 316888
* Recommit r315288: [SCCP] Propagate integer range info for parameters in IPSCCP.Florian Hahn2017-10-301-7/+93
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This version of the patch includes a fix addressing a stage2 LTO buildbot failure and addressed some additional nits. Original commit message: This updates the SCCP solver to use of the ValueElement lattice for parameters, which provides integer range information. The range information is used to remove unneeded icmp instructions. For the following function, f() can be optimized to ret i32 2 with this change source_filename = "sccp.c" target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128" target triple = "x86_64-unknown-linux-gnu" ; Function Attrs: norecurse nounwind readnone uwtable define i32 @main() local_unnamed_addr #0 { entry: %call = tail call fastcc i32 @f(i32 1) %call1 = tail call fastcc i32 @f(i32 47) %add3 = add nsw i32 %call, %call1 ret i32 %add3 } ; Function Attrs: noinline norecurse nounwind readnone uwtable define internal fastcc i32 @f(i32 %x) unnamed_addr #1 { entry: %c1 = icmp sle i32 %x, 100 %cmp = icmp sgt i32 %x, 300 %. = select i1 %cmp, i32 1, i32 2 ret i32 %. } attributes #1 = { noinline } Reviewers: davide, sanjoy, efriedma, dberlin Reviewed By: davide, dberlin Subscribers: mcrosier, gberry, mssimpso, dberlin, llvm-commits Differential Revision: https://reviews.llvm.org/D36656 llvm-svn: 316887
* [GVN][NFC] Mark instruction for deletion instead of immediate erasing in LoadPREMax Kazantsev2017-10-301-2/+1
| | | | | | | | It is done to uniformly handle instructions removal. Differential Revision: https://reviews.llvm.org/D39369 llvm-svn: 316884
* [X86] Rearrange code in X86InstrInfo.cpp to put all the ↵Craig Topper2017-10-301-270/+270
| | | | | | | | foldMemoryOperandImpl methods together without partial/undef register handling in the middle. NFC I have a future patch that wants to make use of the one of the partial functions in one of the earlier memory folding methods and the current ordering prevents that. llvm-svn: 316883
* [X86] Simplify code by removing an unnecessary temporary variable. NFCCraig Topper2017-10-301-2/+1
| | | | llvm-svn: 316882
* [X86] Move some EVEX->VEX code to a helper function to prepare for a future ↵Craig Topper2017-10-301-24/+33
| | | | | | patch. NFC llvm-svn: 316881
* [SelectionDAG] Add SEXT/AND/XOR/Or demanded elts support to ComputeNumSignBitsSimon Pilgrim2017-10-291-7/+11
| | | | llvm-svn: 316875
* [(new) Pass Manager] instantiate SimplifyCFG with the same options as the old PMSanjay Patel2017-10-291-2/+7
| | | | | | | | | | | | The old PM sets the options of what used to be known as "latesimplifycfg" on the instantiation after the vectorizers have run, so that's what we'redoing here. FWIW, there's a later SimplifyCFGPass instantiation in both PMs where we do not set the "late" options. I'm not sure if that's intentional or not. Differential Revision: https://reviews.llvm.org/D39407 llvm-svn: 316869
* [SelectionDAG] Add SRA/SHL demanded elts support to ComputeNumSignBitsSimon Pilgrim2017-10-291-3/+29
| | | | | | Introduce a isConstOrDemandedConstSplat helper function that can recognise a constant splat build vector for at least the demanded elts we care about. llvm-svn: 316866
* [X86] Remove combine that turns X86ISD::LSUB into X86ISD::LADD. Update ↵Craig Topper2017-10-292-29/+8
| | | | | | | | | | | | patterns that depended on this. If the carry flag is being used, this transformation isn't safe. This does prevent some test cases from using DEC now, but I'll try to look into that separately. Fixes PR35068. llvm-svn: 316860
* [X86] Fix typo in comment. NFCCraig Topper2017-10-291-1/+1
| | | | llvm-svn: 316859
* [X86] Use the extended vector register classes in fast isel with AVX512F/VL.Craig Topper2017-10-291-10/+10
| | | | llvm-svn: 316857
* [X86] Add AVX512 support to X86FastISel::X86SelectFPExt and ↵Craig Topper2017-10-291-4/+12
| | | | | | X86FastISel::X86SelectFPTrunc. llvm-svn: 316856
OpenPOWER on IntegriCloud