summaryrefslogtreecommitdiffstats
path: root/llvm/lib
Commit message (Collapse)AuthorAgeFilesLines
...
* Add an explicit cast to pacify implicit boolean conversion warnings.Benjamin Kramer2014-08-271-1/+1
| | | | llvm-svn: 216539
* [x86] Fix a regression introduced with r213897 for 32-bit targets whereChandler Carruth2014-08-271-4/+2
| | | | | | | | | | | | we stopped efficiently lowering sextload using the SSE41 instructions for that operation. This is a consequence of a bad predicate I used thinking of the memory access needs. The code actually handles the cases where the predicate doesn't apply, and handles them much better. =] Simple fix and a test case added. Fixes PR20767. llvm-svn: 216538
* [SDAG] Re-instate r215611 with a fix to a pesky X86 DAG combine.Chandler Carruth2014-08-272-12/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This combine is essentially combining target-specific nodes back into target independent nodes that it "knows" will be combined yet again by a target independent DAG combine into a different set of target-independent nodes that are legal (not custom though!) and thus "ok". This seems... deeply flawed. The crux of the problem is that we don't combine un-legalized shuffles that are introduced by legalizing other operations, and thus we don't see a very profitable combine opportunity. So the backend just forces the input to that combine to re-appear. However, for this to work, the conditions detected to re-form the unlegalized nodes must be *exactly* right. Previously, failing this would have caused poor code (if you're lucky) or a crasher when we failed to select instructions. After r215611 we would fall back into the legalizer. In some cases, this just "fixed" the crasher by produces bad code. But in the test case added it caused the legalizer and the dag combiner to iterate forever. The fix is to make the alignment checking in the x86 side of things match the alignment checking in the generic DAG combine exactly. This isn't really a satisfying or principled fix, but it at least make the code work as intended. It also highlights that it would be nice to detect the availability of under aligned loads for a given type rather than bailing on this optimization. I've left a FIXME to document this. Original commit message for r215611 which covers the rest of the chang: [SDAG] Fix a case where we would iteratively legalize a node during combining by replacing it with something else but not re-process the node afterward to remove it. In a truly remarkable stroke of bad luck, this would (in the test case attached) end up getting some other node combined into it without ever getting re-processed. By adding it back on to the worklist, in addition to deleting the dead nodes more quickly we also ensure that if it *stops* being dead for any reason it makes it back through the legalizer. Without this, the test case will end up failing during instruction selection due to an and node with a type we don't have an instruction pattern for. It took many million runs of the shuffle fuzz tester to find this. llvm-svn: 216537
* Clang-format over X86AsmInstrumentation.*.Evgeniy Stepanov2014-08-272-183/+216
| | | | llvm-svn: 216536
* [SKX] Added new versions of cmp instructions in avx512_icmp_cc multiclass, ↵Robert Khasanov2014-08-271-34/+148
| | | | | | | | added VL multiclass. Added encoding tests llvm-svn: 216532
* AVX-512: Added intrinsic for VMOVSS store form with mask.Elena Demikhovsky2014-08-271-0/+10
| | | | llvm-svn: 216530
* Simplify creation of a bunch of ArrayRefs by using None, makeArrayRef or ↵Craig Topper2014-08-2728-88/+62
| | | | | | just letting them be implicitly created. llvm-svn: 216525
* Fix some cases were ArrayRefs were being passed by reference. Also remove ↵Craig Topper2014-08-273-6/+6
| | | | | | 'const' from some other ArrayRef uses since its implicitly const already. llvm-svn: 216524
* InstCombine: Optimize GEP's involving ptrtoint betterDavid Majnemer2014-08-271-11/+29
| | | | | | | | | | | | | | We supported transforming: (gep i8* X, -(ptrtoint Y)) to: (inttoptr (sub (ptrtoint X), (ptrtoint Y))) However, this only fired if 'X' had type i8*. Generalize this to support various types of different sizes. This results in much better CodeGen, especially for pointers to packed structs. llvm-svn: 216523
* Remove type unit skeletons. GDB no longer needs them & this saves a heap of ↵David Blaikie2014-08-271-22/+1
| | | | | | space. llvm-svn: 216521
* [FastISel][AArch64] Fix address simplification.Juergen Ributzka2014-08-271-8/+103
| | | | | | | | | | | | When a shift with extension or an add with shift and extension cannot be folded into the memory operation, then the address calculation has to be materialized separately. While doing so the code forgot to consider a possible sign-/zero- extension. This fix folds now also the sign-/zero-extension into the add or shift instruction which is used to materialize the address. This fixes rdar://problem/18141718. llvm-svn: 216511
* [FastISel][AArch64] Fold Sign-/Zero-Extend into the shift immediate instruction.Juergen Ributzka2014-08-271-55/+249
| | | | llvm-svn: 216510
* Pass a std::unique_ptr<MemoryBuffer>& to getLazyBitcodeModule.Rafael Espindola2014-08-264-15/+15
| | | | | | | By taking a reference we can do the ownership transfer in one place instead of expecting every caller to do it. llvm-svn: 216492
* Pass a MemoryBufferRef when we can avoid taking ownership.Rafael Espindola2014-08-269-49/+40
| | | | | | | | | | | | | The attached patch simplifies a few interfaces that don't need to take ownership of a buffer. For example, both parseAssembly and parseBitcodeFile will parse the entire buffer before returning. There is no need to take ownership. Using a MemoryBufferRef makes it obvious in the type signature that there is no ownership transfer. llvm-svn: 216488
* Give ExecutionEngine of top level buffers.Rafael Espindola2014-08-263-3/+12
| | | | | | | Long term the idea if for the engine to not own the buffers, but for now this is consistent with the rest of the API. llvm-svn: 216484
* MC: Split the x86 asm matcher implementations by dialectReid Kleckner2014-08-262-33/+198
| | | | | | | | | | | | | | | | | | | The existing matcher has lots of AT&T assembly dialect assumptions baked into it. In particular, the hack for resolving the size of a memory operand by appending the four most common suffixes doesn't work at all. The Intel assembly dialect mnemonic table has ambiguous entries, so we need to try matching multiple times with different operand sizes, since that's the only way to choose different instruction variants. This makes us more compatible with gas's implementation of Intel assembly syntax. MSVC assumes you want byte-sized operations for the instructions that we reject as ambiguous. Reviewed By: grosbach Differential Revision: http://reviews.llvm.org/D4747 llvm-svn: 216481
* Revert r210342 and r210343, add test case for the crasher.Joerg Sonnenberger2014-08-261-91/+0
| | | | | | PR 20642. llvm-svn: 216475
* Convert MC command line option for fatal assembler warnings into aJoerg Sonnenberger2014-08-262-8/+4
| | | | | | proper flag. llvm-svn: 216471
* Return a std::unique_ptr from the IRReader.h functions. NFC.Rafael Espindola2014-08-261-16/+15
| | | | llvm-svn: 216466
* Simplify LTOModule::makeLTOModule a bit. NFC.Rafael Espindola2014-08-261-3/+1
| | | | | | | Just call parseBitcodeFile instead of getLazyBitcodeModule followed by materializeAllPermanently. llvm-svn: 216461
* Merge TempDir and system_temp_directory.Rafael Espindola2014-08-263-90/+106
| | | | | | | | | We had two functions for finding the temp or cache directory. Each had a different set of smarts about OS specific APIs. With this patch system_temp_directory becomes the only way to do it. llvm-svn: 216460
* Silence unused function warning in Release builds.Benjamin Kramer2014-08-261-0/+2
| | | | llvm-svn: 216458
* Change the return value of "getEnd()" from a MachineInstr* to a ↵James Molloy2014-08-261-1/+1
| | | | | | | | MachineBasicBlock::iterator. It seems on Darwin the illegal round-trip ::iterator -> MachineInstr* -> ::iterator breaks execution horribly when the iterator is not a real MachineInstr, like ::end(). llvm-svn: 216455
* ARM: Add patterns for dbgYi Kong2014-08-262-2/+3
| | | | llvm-svn: 216451
* This patch enables SimplifyUsingDistributiveLaws() to handle following pattens.Dinesh Dwivedi2014-08-262-54/+45
| | | | | | | | | | | | (X >> Z) & (Y >> Z) -> (X&Y) >> Z for all shifts. (X >> Z) | (Y >> Z) -> (X|Y) >> Z for all shifts. (X >> Z) ^ (Y >> Z) -> (X^Y) >> Z for all shifts. These patterns were previously handled separately in visitAnd()/visitOr()/visitXor(). Differential Revision: http://reviews.llvm.org/D4951 llvm-svn: 216443
* InstSimplify: Fold gep X, (sub 0, ptrtoint(X)) to nullDavid Majnemer2014-08-261-21/+32
| | | | | | | Save InstCombine some work if we can perform this fold during InstSimplify. llvm-svn: 216441
* InstSimplify: Simplify trivial pointer expressions like b + (e - b)David Majnemer2014-08-261-5/+36
| | | | | | | | | | | | | | | | | | | | | | | | consider: long long *f(long long *b, long long *e) { return b + (e - b); } we would lower this to something like: define i64* @f(i64* %b, i64* %e) { %1 = ptrtoint i64* %e to i64 %2 = ptrtoint i64* %b to i64 %3 = sub i64 %1, %2 %4 = ashr exact i64 %3, 3 %5 = getelementptr inbounds i64* %b, i64 %4 ret i64* %5 } This should fold away to just 'e'. N.B. This adds m_SpecificInt as a convenient way to match against a particular 64-bit integer when using LLVM's match interface. llvm-svn: 216439
* AArch64: use std::fill instead of memsetDylan Noblesmith2014-08-261-1/+1
| | | | | | Followup based on review. llvm-svn: 216436
* Revert "AArch64: use std::vector for temp array"Dylan Noblesmith2014-08-261-4/+5
| | | | | | This reverts commit r216365. llvm-svn: 216433
* Analysis: cleanupDylan Noblesmith2014-08-261-3/+2
| | | | | | Address review comments. llvm-svn: 216432
* Revert "Analysis: unique_ptr-ify DependenceAnalysis::collectCoeffInfo"Dylan Noblesmith2014-08-261-8/+8
| | | | | | This reverts commit r216358. llvm-svn: 216431
* Revert "NVPTX: remove another raw delete call"Dylan Noblesmith2014-08-261-1/+3
| | | | | | This reverts commit r216364. llvm-svn: 216430
* Revert "Support/APFloat: unique_ptr-ify temp arrays"Dylan Noblesmith2014-08-261-8/+11
| | | | | | This reverts commit rr216359. llvm-svn: 216429
* Revert "Support/Path: remove raw delete"Dylan Noblesmith2014-08-261-2/+2
| | | | | | This reverts commit r216360. llvm-svn: 216428
* ExecutionEngine: address review commentsDylan Noblesmith2014-08-261-7/+6
| | | | llvm-svn: 216427
* CodeGen/LiveVariables: use vector::assign()Dylan Noblesmith2014-08-261-8/+4
| | | | | | Address review comments. llvm-svn: 216426
* musttail: Don't eliminate varargs packs if there is a forwarding callReid Kleckner2014-08-261-2/+7
| | | | | | Also clean up and beef up this grep test for the feature. llvm-svn: 216425
* fix typos in commentsSanjay Patel2014-08-261-4/+4
| | | | llvm-svn: 216424
* Declare that musttail calls in variadic functions forward the ellipsisReid Kleckner2014-08-263-3/+30
| | | | | | | | | | | | | | | | | Summary: There is no functionality change here except in the way we assemble and dump musttail calls in variadic functions. There's really no need to separate out the bits for musttail and "is forwarding varargs" on call instructions. A musttail call by definition has to forward the ellipsis or it would fail verification. Reviewers: chandlerc, nlewycky Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D4892 llvm-svn: 216423
* ArgPromotion: Don't touch variadic functionsReid Kleckner2014-08-251-0/+7
| | | | | | | | | | | | | | | | Adding, removing, or changing non-pack parameters can change the ABI classification of pack parameters. Clang and other frontends encode the classification in the IR of the call site, but the callee side determines it dynamically based on the number of registers consumed so far. Changing the prototype affects the number of registers consumed would break such code. Dead argument elimination performs a similar task and already has a similar check to avoid this problem. Patch by Thomas Jablin! llvm-svn: 216421
* [MCJIT][SystemZ] Use a simpler expression for indirect relocation offsets.Lang Hames2014-08-251-1/+1
| | | | | | | | The expressions 'Reloc.Addend - Addend' and 'Reloc.Offset' should always be equal in this context. The latter is prefered - we want to remove the RelocationValueRef::Addend field in the future. llvm-svn: 216418
* Fix bug in llvm::sys::argumentsFitWithinSystemLimits().Rafael Espindola2014-08-251-2/+2
| | | | | | | | | | | | | This patch fixes a subtle bug in the UNIX implementation of llvm::sys::argumentsFitWithinSystemLimits() regarding the misuse of a static variable. This bug causes our cached number that stores the system command line maximum length to be halved after each call to the function. With a sufficient number of calls to this function, it will eventually report any given command line string to be over system limits. Patch by Rafael Auler. llvm-svn: 216415
* [MCJIT] Dump section memory both before and after relocations are applied.Lang Hames2014-08-251-5/+8
| | | | | | Also switch section memory dump format from 8 to 16 columns. llvm-svn: 216413
* Refactor argument serialization logic when executing process. NFC.Rafael Espindola2014-08-251-13/+17
| | | | | | | | | | | | This patch refactors the argument serialization logic used in the Execute function, used to launch new Windows processes. There is a critical step that joins char** arguments into a single string, building the command line used to launch the new process, and the readability of this code is improved if this part is refactored in its own helper function. Patch by Rafael Auler! llvm-svn: 216411
* [FastISel][AArch64] Refactor float zero materialization. NFCI.Juergen Ributzka2014-08-251-20/+29
| | | | llvm-svn: 216403
* [MCJIT] Make RuntimeDyld dump section contents in -debug mode.Lang Hames2014-08-252-4/+34
| | | | llvm-svn: 216400
* Modernize raw_fd_ostream's constructor a bit.Rafael Espindola2014-08-2515-79/+72
| | | | | | | | | | Take a StringRef instead of a "const char *". Take a "std::error_code &" instead of a "std::string &" for error. A create static method would be even better, but this patch is already a bit too big. llvm-svn: 216393
* [x86] Fix a bug in r216319 where I was missing a 'break'.Chandler Carruth2014-08-251-0/+2
| | | | | | | | | | | This actually was caught by existing tests but those tests were disabled with an XFAIL because of PR20736. While working on fixing that, I noticed the test failure, and tracked it down to this. We even have a really nice Clang warning that would have caught this but it isn't enabled in LLVM! =[ I may look at enabling it. llvm-svn: 216391
* Remove dangling initializers in GlobalDCEBruno Cardoso Lopes2014-08-252-1/+10
| | | | | | | | | | | | | | GlobalDCE deletes global vars and updates their initializers to nullptr while leaving underlying constants to be cleaned up later by its uses. The clean up may never happen, fix this by forcing it every time it's safe to destroy constants. Final patch by Rafael Espindola http://reviews.llvm.org/D4931 <rdar://problem/17523868> llvm-svn: 216390
* [AArch32] Add patterns for VCVT{A,N,P,M}.Chad Rosier2014-08-251-9/+13
| | | | | | | Patterns for lowering libm calls to VCVT{A,N,P,M} are also included. Phabricator Revision: http://reviews.llvm.org/D5033 llvm-svn: 216388
OpenPOWER on IntegriCloud