summaryrefslogtreecommitdiffstats
path: root/llvm
Commit message (Collapse)AuthorAgeFilesLines
...
* [x86] replace div/rem with shift/mask for better shuffle combining perfSanjay Patel2017-06-141-13/+31
| | | | | | | | | | | | We know that shuffle masks are power-of-2 sizes, but there's no way (?) for LLVM to know that, so hack combineX86ShufflesRecursively() to be much faster by replacing div/rem with shift/mask. This makes the motivating compile-time bug in PR32037 ( https://bugs.llvm.org/show_bug.cgi?id=32037 ) about 9% faster overall. Differential Revision: https://reviews.llvm.org/D34174 llvm-svn: 305398
* Add an include of gtest-printers.h to appease the buildbots.Zachary Turner2017-06-141-1/+2
| | | | llvm-svn: 305396
* [gtest] Create a shared include directory for gtest utilities.Zachary Turner2017-06-1422-373/+441
| | | | | | | | | | | | | | | | | | | | | | | Many times unit tests for different libraries would like to use the same helper functions for checking common types of errors. This patch adds a common library with helpers for testing things in Support, and introduces helpers in here for integrating the llvm::Error and llvm::Expected<T> classes with gtest and gmock. Normally, we would just be able to write: EXPECT_THAT(someFunction(), succeeded()); but due to some quirks in llvm::Error's move semantics, gmock doesn't make this easy, so two macros EXPECT_THAT_ERROR() and EXPECT_THAT_EXPECTED() are introduced to gloss over the difficulties. Consider this an exception, and possibly only temporary as we look for ways to improve this. Differential Revision: https://reviews.llvm.org/D33059 llvm-svn: 305395
* Resubmit "[codeview] Make obj2yaml/yaml2obj support .debug$S..."Zachary Turner2017-06-1428-321/+572
| | | | | | | | | This was originally reverted because of some non-deterministic failures on certain buildbots. Luckily ASAN eventually caught this as a stack-use-after-scope, so the fix is included in this patch. llvm-svn: 305393
* Revert "[ARM] Support constant pools in data when generating execute-only code."Alexandros Lamprineas2017-06-144-93/+15
| | | | | | | | | | | This reverts commit 3a204faa093c681a1e96c5e0622f50649b761ee0. I've upset a buildbot which runs the address sanitizer: ERROR: AddressSanitizer: stack-use-after-scope lib/Target/ARM/ARMISelLowering.cpp:2690 That Twine variable is used illegally. llvm-svn: 305390
* [mips] Fix multiprecision arithmetic.Simon Dardis2017-06-1411-463/+648
| | | | | | | | | | | | | | | | | | | | | | | | | | For multiprecision arithmetic on MIPS, rather than using ISD::ADDE / ISD::ADDC, get SelectionDAG to break down the operation into ISD::ADDs and ISD::SETCCs. For MIPS, only the DSP ASE has a carry flag, so in the general case it is not useful to directly support ISD::{ADDE, ADDC, SUBE, SUBC} nodes. Also improve the generation code in such cases for targets with TargetLoweringBase::ZeroOrOneBooleanContent by directly using the result of the comparison node rather than using it in selects. Similarly for ISD::SUBE / ISD::SUBC. Address optimization breakage by moving the generation of MIPS specific integer multiply-accumulate nodes to before legalization. This revolves PR32713 and PR33424. Thanks to Simonas Kazlauskas and Pirama Arumuga Nainar for reporting the issue! Reviewers: slthakur Differential Revision: https://reviews.llvm.org/D33494 llvm-svn: 305389
* [MathExtras] fix documentation comments; NFCSanjay Patel2017-06-141-61/+46
| | | | llvm-svn: 305388
* [ARM] Support constant pools in data when generating execute-only code.Alexandros Lamprineas2017-06-144-15/+93
| | | | | | | | | | | | | | The ARM backend asserts against constant pool lowering when it generates execute-only code in order to prevent the generation of constant pools in the text section. It appears that target independent optimizations might generate DAG nodes that represent constant pools. By lowering such nodes as global addresses we don't violate the semantics of execute-only code and also it is guaranteed that execute-only behaves correct with the position-independent addressing modes that support execute-only code. Differential Revision: https://reviews.llvm.org/D33773 llvm-svn: 305387
* Align definition of DW_OP_plus with DWARF spec [3/3]Florian Hahn2017-06-1433-114/+200
| | | | | | | | | | | | | | | | | | | | | | | | Summary: This patch is part of 3 patches that together form a single patch, but must be introduced in stages in order not to break things. The way that LLVM interprets DW_OP_plus in DIExpression nodes is basically that of the DW_OP_plus_uconst operator since LLVM expects an unsigned constant operand. This unnecessarily restricts the DW_OP_plus operator, preventing it from being used to describe the evaluation of runtime values on the expression stack. These patches try to align the semantics of DW_OP_plus and DW_OP_minus with that of the DWARF definition, which pops two elements off the expression stack, performs the operation and pushes the result back on the stack. This is done in three stages: • The first patch (LLVM) adds support for DW_OP_plus_uconst. • The second patch (Clang) contains changes all its uses from DW_OP_plus to DW_OP_plus_uconst. • The third patch (LLVM) changes the semantics of DW_OP_plus and DW_OP_minus to be in line with its DWARF meaning. This patch includes the bitcode upgrade from legacy DIExpressions. Patch by Sander de Smalen. Reviewers: echristo, pcc, aprantl Reviewed By: aprantl Subscribers: fhahn, javed.absar, aprantl, llvm-commits Differential Revision: https://reviews.llvm.org/D33894 llvm-svn: 305386
* [mips] Fix machine verifier errors in the long branch passSimon Dardis2017-06-142-26/+26
| | | | | | | | | | | | | | | This patch fixes two systemic machine verifier errors in the long branch pass. The first is the incorrect basic block successors and the second was the incorrect construction of several jump instructions. This partially resolves PR27458 and the associated PR32146. Reviewers: slthakur Differential Revision: https://reviews.llvm.org/D33378 llvm-svn: 305382
* Revert r304907 as it is causing some failures that I cannot reproduce.Nemanja Ivanovic2017-06-146-593/+6
| | | | | | Reverting this until a test case can be provided to aid the investigation. llvm-svn: 305372
* Revert "[codeview] Make obj2yaml/yaml2obj support .debug$S..."Zachary Turner2017-06-1424-567/+308
| | | | | | | | This is causing failures on linux bots with an invalid stream read. It doesn't repro in any configuration on Windows, so reverting until I have a chance to investigate on Linux. llvm-svn: 305371
* Use make_shared instead of make_unique.Zachary Turner2017-06-141-2/+2
| | | | llvm-svn: 305369
* Fix some more errors.Zachary Turner2017-06-143-38/+11
| | | | llvm-svn: 305368
* Add some #includes to appease the build.Zachary Turner2017-06-142-0/+2
| | | | llvm-svn: 305367
* [codeview] Make obj2yaml/yaml2obj support .debug$S/T sections.Zachary Turner2017-06-1424-301/+585
| | | | | | | | | | | | | | | This allows us to use yaml2obj and obj2yaml to round-trip CodeView symbol and type information without having to manually specify the bytes of the section. This makes for much easier to maintain tests. See the tests under lld/COFF in this patch for example. Before they just said SectionData: <blob> whereas now we can use meaningful record descriptions. Note that it still supports the SectionData yaml field, which could be useful for initializing a section to invalid bytes for testing, for example. Differential Revision: https://reviews.llvm.org/D34127 llvm-svn: 305366
* Support: Remove MSVC 2013 workarounds in ThreadPool class.Peter Collingbourne2017-06-142-41/+6
| | | | | | | | I have confirmed that these are no longer needed with MSVC 2015. Differential Revision: https://reviews.llvm.org/D34187 llvm-svn: 305347
* [libFuzzer] really restrict the new test to Linux (fails on Mac/Windows ↵Kostya Serebryany2017-06-141-1/+3
| | | | | | currently) llvm-svn: 305346
* Re-enable tests on power pc since the bug has been fixed.Eric Beckmann2017-06-142-8/+0
| | | | | | | | | | Summary: just flip them on. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D34186 llvm-svn: 305345
* Added partial verification for .apple_names accelerator table in ↵Spyridoula Gravani2017-06-146-1/+254
| | | | | | | | | | | llvm-dwarfdump output. This patch adds code which verifies that each bucket in the .apple_names accelerator table is either empty or has a valid hash index. Differential Revision: https://reviews.llvm.org/D34177 llvm-svn: 305344
* Reverted r305339 as MSVC is not happy with noreturn in lambda.Galina Kistanova2017-06-131-1/+2
| | | | llvm-svn: 305343
* [globalisel][legalizer] G_LOAD/G_STORE NarrowScalar should not emit G_GEP x, 0.Daniel Sanders2017-06-134-20/+57
| | | | | | | | | | | | | | | | | | | Summary: When legalizing G_LOAD/G_STORE using NarrowScalar, we should avoid emitting %0 = G_CONSTANT ty 0 %1 = G_GEP %x, %0 since it's cheaper to not emit the redundant instructions than it is to fold them away later. Reviewers: qcolombet, t.p.northover, ab, rovka, aditya_nandakumar, kristof.beyls Reviewed By: qcolombet Subscribers: javed.absar, llvm-commits, igorb Differential Revision: https://reviews.llvm.org/D32746 llvm-svn: 305340
* Specified LLVM_ATTRIBUTE_NORETURN for ReportError.Galina Kistanova2017-06-131-2/+1
| | | | llvm-svn: 305339
* [InstCombine] Add test cases demonstrating failure to handle (select (icmp ↵Craig Topper2017-06-131-0/+31
| | | | | | | | eq (and X, C1), 0), Y, (or Y, C2)) when the icmp portion gets turned into a truncate and a signed compare with 0. InstCombine has an optimization that recognizes an and with the sign bit of legal type size and turns it into a truncate and compare that checks the sign bit. But the select handling code doesn't recognize this idiom. llvm-svn: 305338
* [libFuzzer] restrict the new test to Linux (fails on Mac currently)Kostya Serebryany2017-06-131-0/+1
| | | | llvm-svn: 305335
* [ADT] Revert r305326 changes in BitVector.h to fix broken builds.Eugene Zelenko2017-06-131-28/+22
| | | | llvm-svn: 305332
* [libFuzzer] initial support of -fsanitize-coverage=inline-8bit-counters in ↵Kostya Serebryany2017-06-138-9/+83
| | | | | | libFuzzer. This is not fully functional yet, but simple tests work llvm-svn: 305331
* [AMDGPU] Remove now dead defaultOffsetS13(). NFCI.Davide Italiano2017-06-131-5/+0
| | | | | | Fixes the GCC7 build with -Werror. llvm-svn: 305329
* [InstrProf] Don't take the address of alwaysinline available_externally ↵Vedant Kumar2017-06-132-1/+39
| | | | | | | | | | | | | | | | | | | | | | | | functions Doing so breaks compilation of the following C program (under -fprofile-instr-generate): __attribute__((always_inline)) inline int foo() { return 0; } int main() { return foo(); } At link time, we fail because taking the address of an available_externally function creates an undefined external reference, which the TU cannot provide. Emitting the function definition into the object file at all appears to be a violation of the langref: "Globals with 'available_externally' linkage are never emitted into the object file corresponding to the LLVM module." Differential Revision: https://reviews.llvm.org/D34134 llvm-svn: 305327
* [ADT] Fix some Clang-tidy modernize-use-using and Include What You Use ↵Eugene Zelenko2017-06-1319-238/+368
| | | | | | warnings; other minor fixes (NFC). llvm-svn: 305326
* Use reference to iterate through string table instead of copying.Eric Beckmann2017-06-131-1/+1
| | | | | | | | | | Summary: just a quick patch Subscribers: ruiu, llvm-commits, hiraditya Differential Revision: https://reviews.llvm.org/D34171 llvm-svn: 305324
* Fix a bug introduced in r305092 on big-endian systems.Eric Beckmann2017-06-131-3/+1
| | | | | | | | | | | | | | Summary: We were writing the length of the string based on system-endianness, and not universally little-endian. This fixes that. Reviewers: zturner Subscribers: hiraditya, llvm-commits Differential Revision: https://reviews.llvm.org/D34159 llvm-svn: 305322
* [PGO] Update VP metadata after memory intrinsic optimizationTeresa Johnson2017-06-132-2/+15
| | | | | | | | | | | | | | | Summary: Leave an updated VP metadata on the fallback memcpy intrinsic after specialization. This can be used for later possible expansion based on the average of the remaining values. Reviewers: davidxl Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D34164 llvm-svn: 305321
* Fix alignment complaint.Eric Beckmann2017-06-131-3/+2
| | | | | | | | | | Summary: Apparently we need to write using a void* pointer on some architectures, or else alignment error is caused. Subscribers: hiraditya, llvm-commits Differential Revision: https://reviews.llvm.org/D34166 llvm-svn: 305320
* [ADT] PointerUnion::getAddrOfPtr1(): fix/silence -Wcast-qual warning.Roman Lebedev2017-06-131-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: Previously, when D33102 landed, this broke -Werror buildbots. http://lab.llvm.org:8011/builders/clang-with-lto-ubuntu/builds/3249 ``` FAILED: /home/buildbot/Buildbot/Slave1a/clang-with-lto-ubuntu/install/stage1/bin/clang++ -DGTEST_HAS_RTTI=0 -DLLVM_BUILD_GLOBAL_ISEL -D_DEBUG -D_GNU_SOURCE -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS -Ilib/CodeGen/AsmPrinter -I/home/buildbot/Buildbot/Slave1a/clang-with-lto-ubuntu/llvm.src/lib/CodeGen/AsmPrinter -Iinclude -I/home/buildbot/Buildbot/Slave1a/clang-with-lto-ubuntu/llvm.src/include -fPIC -fvisibility-inlines-hidden -Werror -Werror=date-time -std=c++11 -Wall -W -Wno-unused-parameter -Wwrite-strings -Wcast-qual -Wmissing-field-initializers -pedantic -Wno-long-long -Wcovered-switch-default -Wnon-virtual-dtor -Wdelete-non-virtual-dtor -Wstring-conversion -fcolor-diagnostics -ffunction-sections -fdata-sections -O3 -UNDEBUG -fno-exceptions -fno-rtti -MMD -MT lib/CodeGen/AsmPrinter/CMakeFiles/LLVMAsmPrinter.dir/CodeViewDebug.cpp.o -MF lib/CodeGen/AsmPrinter/CMakeFiles/LLVMAsmPrinter.dir/CodeViewDebug.cpp.o.d -o lib/CodeGen/AsmPrinter/CMakeFiles/LLVMAsmPrinter.dir/CodeViewDebug.cpp.o -c /home/buildbot/Buildbot/Slave1a/clang-with-lto-ubuntu/llvm.src/lib/CodeGen/AsmPrinter/CodeViewDebug.cpp In file included from /home/buildbot/Buildbot/Slave1a/clang-with-lto-ubuntu/llvm.src/lib/CodeGen/AsmPrinter/CodeViewDebug.cpp:14: In file included from /home/buildbot/Buildbot/Slave1a/clang-with-lto-ubuntu/llvm.src/lib/CodeGen/AsmPrinter/CodeViewDebug.h:17: In file included from /home/buildbot/Buildbot/Slave1a/clang-with-lto-ubuntu/llvm.src/lib/CodeGen/AsmPrinter/DbgValueHistoryCalculator.h:15: In file included from /home/buildbot/Buildbot/Slave1a/clang-with-lto-ubuntu/llvm.src/include/llvm/IR/DebugInfoMetadata.h:26: In file included from /home/buildbot/Buildbot/Slave1a/clang-with-lto-ubuntu/llvm.src/include/llvm/IR/Metadata.h:23: /home/buildbot/Buildbot/Slave1a/clang-with-lto-ubuntu/llvm.src/include/llvm/ADT/PointerUnion.h:161:19: error: cast from 'void **' to 'const llvm::DISubprogram **' must have all intermediate pointers const qualified to be safe [-Werror,-Wcast-qual] return (PT1 *)Val.getAddrOfPointer(); ^ /home/buildbot/Buildbot/Slave1a/clang-with-lto-ubuntu/llvm.src/include/llvm/ADT/TinyPtrVector.h:177:18: note: in instantiation of member function 'llvm::PointerUnion<const llvm::DISubprogram *, llvm::SmallVector<const llvm::DISubprogram *, 4> *>::getAddrOfPtr1' requested here return Val.getAddrOfPtr1(); ^ /home/buildbot/Buildbot/Slave1a/clang-with-lto-ubuntu/llvm.src/lib/CodeGen/AsmPrinter/CodeViewDebug.cpp:1885:33: note: in instantiation of member function 'llvm::TinyPtrVector<const llvm::DISubprogram *>::begin' requested here for (const DISubprogram *SP : MethodItr.second) { ^ 1 error generated. ``` Reviewers: dblaikie, akyrtzi Reviewed By: dblaikie Subscribers: joerg, mehdi_amini, llvm-commits Differential Revision: https://reviews.llvm.org/D34153 llvm-svn: 305319
* Revert r305313 & r305303, self-hosting build-bot isn’t liking it.Frederich Munch2017-06-135-51/+13
| | | | llvm-svn: 305318
* [powerpc] deactivate tests combined.test and object.test on powerpc64 beBill Seurer2017-06-132-0/+8
| | | | | | | | | | | These tests fail on powerpc64 BE (only, not LE) and are thus intefering with the running of 3 of the powerpc buildbots. The author of the tests has been notified and is working on fixing them but in the meantime I am disabling them to get the bots working again. See https://bugs.llvm.org//show_bug.cgi?id=33429 llvm-svn: 305317
* [WebAssembly] Cleanup WebAssemblyWasmObjectWriterSam Clegg2017-06-135-33/+26
| | | | | | Differential Revision: https://reviews.llvm.org/D34131 llvm-svn: 305316
* Improve error messages in order to help with fixing a big-endian bug.Eric Beckmann2017-06-135-8/+20
| | | | | | | | | | | | Summary: Added output to stderr so that we can actually see what is happening when the test fails on big endian. Reviewers: zturner Subscribers: llvm-commits, hiraditya Differential Revision: https://reviews.llvm.org/D34155 llvm-svn: 305314
* Fix self hosting build-bot failure from r305303 by adjusting ↵Frederich Munch2017-06-131-0/+1
| | | | | | DynamicLibraryTests compile flags. llvm-svn: 305313
* [IR] Remove a couple explicitly deleted default constructors on classes that ↵Craig Topper2017-06-132-3/+0
| | | | | | | | have a non-default constructor. NFC The non-default constructor will implicitly delete the default constructor. llvm-svn: 305311
* [AArch64][Falkor] Fix sched details for FDIV, FSQRT, SDIV, UDIVGeoff Berry2017-06-131-11/+50
| | | | llvm-svn: 305310
* Test commit - NFC.Kit Barton2017-06-131-1/+1
| | | | | | Modified a comment to confirm commit access functionality. llvm-svn: 305309
* Fix m_[Ord|Unord][FMin|FMax] matchers to correctly match ordering.Craig Topper2017-06-132-34/+79
| | | | | | | | | | | Previously, the matching was done incorrectly for the case where operands for FCmpInst and SelectInst were in opposite order. Patch by Andrei Elovikov. Differential Revision: https://reviews.llvm.org/D33185 llvm-svn: 305308
* [Hexagon] Generate store-immediate instructions for stack objectsKrzysztof Parzyszek2017-06-133-4/+111
| | | | | | | | | Store-immediate instructions have a non-extendable offset. Since the actual offset for a stack object is not known until much later, only generate these stores when the stack size (at the time of instruction selection) is small. llvm-svn: 305305
* Align definition of DW_OP_plus with DWARF spec [1/3]Florian Hahn2017-06-135-3/+19
| | | | | | | | | | | | | | | | | | | | | | | | Summary: This patch is part of 3 patches that together form a single patch, but must be introduced in stages in order not to break things. The way that LLVM interprets DW_OP_plus in DIExpression nodes is basically that of the DW_OP_plus_uconst operator since LLVM expects an unsigned constant operand. This unnecessarily restricts the DW_OP_plus operator, preventing it from being used to describe the evaluation of runtime values on the expression stack. These patches try to align the semantics of DW_OP_plus and DW_OP_minus with that of the DWARF definition, which pops two elements off the expression stack, performs the operation and pushes the result back on the stack. This is done in three stages: • The first patch (LLVM) adds support for DW_OP_plus_uconst. • The second patch (Clang) contains changes all its uses from DW_OP_plus to DW_OP_plus_uconst. • The third patch (LLVM) changes the semantics of DW_OP_plus and DW_OP_minus to be in line with its DWARF meaning. This patch includes the bitcode upgrade from legacy DIExpressions. Patch by Sander de Smalen. Reviewers: pcc, echristo, aprantl Reviewed By: aprantl Subscribers: fhahn, aprantl, javed.absar, llvm-commits Differential Revision: https://reviews.llvm.org/D33892 llvm-svn: 305304
* Force RegisterStandardPasses to construct std::function in the IPO library.Frederich Munch2017-06-135-13/+50
| | | | | | | | | | | | | | Summary: Fixes an issue using RegisterStandardPasses from a statically linked object before PassManagerBuilder::addGlobalExtension is called from a dynamic library. Reviewers: efriedma, theraven Reviewed By: efriedma Subscribers: mehdi_amini, mgorny, llvm-commits Differential Revision: https://reviews.llvm.org/D33515 llvm-svn: 305303
* [Hexagon] Generate multiply-high instruction in iselKrzysztof Parzyszek2017-06-132-0/+32
| | | | llvm-svn: 305302
* bpf: clang-format on BPFAsmPrinter.cppYonghong Song2017-06-131-2/+3
| | | | | Signed-off-by: Yonghong Song <yhs@fb.com> llvm-svn: 305301
* [Hexagon] Don't kill live registers when creating mux out of tfrKrzysztof Parzyszek2017-06-133-8/+37
| | | | | | | | | | | When a mux instruction is created from a pair of complementary conditional transfers, it can be placed at the location of either the earlier or the later of the transfers. Since it will use the operands of the original transfers, putting it in the earlier location may hoist a kill of a source register that was originally further down. Make sure the kill flag is removed if the register is still used afterwards. llvm-svn: 305300
OpenPOWER on IntegriCloud