summaryrefslogtreecommitdiff
path: root/include
Commit message (Collapse)AuthorAgeFilesLines
* [analyzer] Track class member initializer constructors path-sensitively.Artem Dergachev2018-06-141-7/+10
| | | | | | | | | | | | | | | | The reasoning behind this change is similar to the previous commit, r334681. Because members are already in scope when construction occurs, we are not suffering from liveness problems, but we still want to figure out if the object was constructed with construction context, because in this case we'll be able to avoid trivial copy, which we don't always model perfectly. It'd also have more importance when copy elision is implemented. This also gets rid of the old CFG look-behind mechanism. Differential Revision: https://reviews.llvm.org/D47350 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334682 91177308-0d34-0410-b5e6-96231b3b80d8
* [analyzer] NFC: Merge code for finding and tracking construction target.Artem Dergachev2018-06-141-15/+8
| | | | | | | | | | | | | | | When analyzing C++ code, a common operation in the analyzer is to discover target region for object construction by looking at CFG metadata ("construction contexts"), and then track the region path-sensitively until object construction is resolved, where the amount of information, again, depends on construction context. Scan construction context only once for both purposes. Differential Revision: https://reviews.llvm.org/D47304 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334678 91177308-0d34-0410-b5e6-96231b3b80d8
* [COFF] Add ARM64 intrinsics: __yield, __wfe, __wfi, __sev, __sevlMandeep Singh Grang2018-06-131-0/+6
| | | | | | | | | | | | | | Summary: These intrinsics result in hint instructions. They are provided here for MSVC ARM64 compatibility. Reviewers: mstorsjo, compnerd, javed.absar Reviewed By: mstorsjo Subscribers: kristof.beyls, chrib, cfe-commits Differential Revision: https://reviews.llvm.org/D48132 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334639 91177308-0d34-0410-b5e6-96231b3b80d8
* [CodeGen] make nan builtins pure rather than const (PR37778)Sanjay Patel2018-06-131-8/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | https://bugs.llvm.org/show_bug.cgi?id=37778 ...shows a miscompile resulting from marking nan builtins as 'const'. The nan libcalls/builtins take a pointer argument: http://www.cplusplus.com/reference/cmath/nan-function/ ...and the chars dereferenced by that arg are used to fill in the NaN constant payload bits. "const" means that the pointer argument isn't dereferenced. That's translated to "readnone" in LLVM. "pure" means that the pointer argument may be dereferenced. That's translated to "readonly" in LLVM. This change prevents the IR optimizer from killing the lead-up to the nan call here: double a() { char buf[4]; buf[0] = buf[1] = buf[2] = '9'; buf[3] = '\0'; return __builtin_nan(buf); } ...the optimizer isn't currently able to simplify this to a constant as we might hope, but this patch should solve the miscompile. Differential Revision: https://reviews.llvm.org/D48134 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334628 91177308-0d34-0410-b5e6-96231b3b80d8
* Add -fforce-emit-vtablesPiotr Padlewski2018-06-133-0/+11
| | | | | | | | | | | | | | | | | | | | Summary: In many cases we can't devirtualize because definition of vtable is not present. Most of the time it is caused by inline virtual function not beeing emitted. Forcing emitting of vtable adds a reference of these inline virtual functions. Note that GCC was always doing it. Reviewers: rjmccall, rsmith, amharc, kuhar Subscribers: llvm-commits, cfe-commits Differential Revision: https://reviews.llvm.org/D47108 Co-authored-by: Krzysztof Pszeniczny <krzysztof.pszeniczny@gmail.com> git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334600 91177308-0d34-0410-b5e6-96231b3b80d8
* [libclang] Optionally add code completion results for arrow instead of dotIvan Donchevskii2018-06-132-20/+107
| | | | | | | | Follow up for D41537 - libclang part. Differential Revision: https://reviews.llvm.org/D46862 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334593 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Remove masking from avx512vbmi2 concat and shift by immediate ↵Craig Topper2018-06-131-18/+18
| | | | | | builtins. Use select builtins instead. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334577 91177308-0d34-0410-b5e6-96231b3b80d8
* [CUDA][HIP] Allow CUDA __global__ functions to have amdgpu kernel attributesYaxun Liu2018-06-121-1/+1
| | | | | | | | | | | | | There are HIP applications e.g. Tensorflow 1.3 using amdgpu kernel attributes, however currently they are only allowed on OpenCL kernel functions. This patch will allow amdgpu kernel attributes to be applied to CUDA/HIP __global__ functions. Differential Revision: https://reviews.llvm.org/D47958 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334561 91177308-0d34-0410-b5e6-96231b3b80d8
* [analyzer] [NFC] Remove "removeInvalidation" from visitor APIGeorge Karpenkov2018-06-121-7/+0
| | | | | | | | | | | removeInvalidation is a very problematic API, as it makes suppression order-dependent. Moreover, it was used only once, and could be rewritten in a much cleaner way. Differential Revision: https://reviews.llvm.org/D48045 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334542 91177308-0d34-0410-b5e6-96231b3b80d8
* [analyzer] [NFC] Move ::dump methods from BugReporter.cpp to PathDiagnostics.cppGeorge Karpenkov2018-06-121-0/+6
| | | | | | | | | | BugReporter.cpp is already severely overloaded, and those dump methods are on PathDiagnostics and should belong in the corresponding implementation file. Differential Revision: https://reviews.llvm.org/D48035 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334541 91177308-0d34-0410-b5e6-96231b3b80d8
* [analyzer] [NFC] Remove most usages of getEndPathGeorge Karpenkov2018-06-121-6/+10
| | | | | | | | | | | | | | | getEndPath is a problematic API, because it's not clear when it's called (hint: not always at the end of the path), it crashes at runtime with more than one non-nullptr returning implementation, and diagnostics internal depend on it being called at some exact place. However, most visitors don't actually need that: all they want is a function consistently called after all nodes are traversed, to perform finalization and to decide whether invalidation is needed. Differential Revision: https://reviews.llvm.org/D48042 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334540 91177308-0d34-0410-b5e6-96231b3b80d8
* [AArch64] Support reserving x20 registerPetr Hosek2018-06-121-0/+2
| | | | | | | | | | | | Register x20 is a callee-saved register which may be used for other purposes in certain contexts, for example to hold special variables within the kernel. This change adds support for reserving this register both to frontend and backend to make this register usable for these purposes. Differential Revision: https://reviews.llvm.org/D46552 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334531 91177308-0d34-0410-b5e6-96231b3b80d8
* [analyzer] [NFC] Remove unused Extensive diagnostic setting,George Karpenkov2018-06-121-1/+10
| | | | | | | | | | | | | | Rename AlternateExtensive to Extensive. In 2013, five years ago, we have switched to AlternateExtensive diagnostics by default, and Extensive was available under unused, undocumented flag. This change remove the flag, renames the Alternate diagnostic to Extensive (as it's no longer Alternate), and ports the test. Differential Revision: https://reviews.llvm.org/D47670 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334524 91177308-0d34-0410-b5e6-96231b3b80d8
* [AArch64] Corrected FP16 Intrinsic range checks in Clang + added Sema testsLuke Geeson2018-06-121-9/+9
| | | | | | | | | | | | | | | | | | | Summary: This fixes the ranges for the vcvth family of FP16 intrinsics in the clang front end. Previously it was accepting incorrect ranges -Changed builtin range checking in SemaChecking -added tests SemaCheck changes - included in their own file since no similar one exists -modified existing tests to reflect new ranges Reviewers: SjoerdMeijer, javed.absar Reviewed By: SjoerdMeijer Subscribers: kristof.beyls, cfe-commits Differential Revision: https://reviews.llvm.org/D47592 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334489 91177308-0d34-0410-b5e6-96231b3b80d8
* Fix that AlignedAllocation.h doesn't compile because of VersionTupleRaphael Isemann2018-06-121-4/+4
| | | | | | | | | | | | Summary: rL334399 put VersionTuple in the llvm namespace, but this header still assumes it's in the clang namespace. This leads to compilation failures with enabled modules when building Clang. Subscribers: cfe-commits Differential Revision: https://reviews.llvm.org/D48062 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334471 91177308-0d34-0410-b5e6-96231b3b80d8
* [Driver] Add aliases for -Qn/-QyMikhail Maltsev2018-06-111-1/+2
| | | | | | | | | | This patch adds aliases for -Qn (-fno-ident) and -Qy (-fident) which look less cryptic than -Qn/-Qy. The aliases are compatible with GCC. Differential Revision: https://reviews.llvm.org/D48021 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334414 91177308-0d34-0410-b5e6-96231b3b80d8
* clang-format: Introduce BreakInheritanceList optionFrancois Ferrand2018-06-111-12/+31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: This option replaces the BreakBeforeInheritanceComma option with an enum, thus introducing a mode where the colon stays on the same line as constructor declaration: // When it fits on line: class A : public B, public C { ... }; // When it does not fit: class A : public B, public C { ... }; This matches the behavior of the `BreakConstructorInitializers` option, introduced in https://reviews.llvm.org/D32479. Reviewers: djasper, klimek Reviewed By: djasper Subscribers: mzeren-vmw, cfe-commits Differential Revision: https://reviews.llvm.org/D43015 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334408 91177308-0d34-0410-b5e6-96231b3b80d8
* Move VersionTuple from clang/Basic to llvm/SupportPavel Labath2018-06-1114-170/+17
| | | | | | | | | | | | | | | | | | | | | | | Summary: This kind of functionality is useful to other project apart from clang. LLDB works with version numbers a lot, but it does not have a convenient abstraction for this. Moving this class to a lower level library allows it to be freely used within LLDB. Since this class is used in a lot of places in clang, and it used to be in the clang namespace, it seemed appropriate to add it to the list of adopted classes in LLVM.h to avoid prefixing all uses with "llvm::". Also, I didn't find any tests specific for this class, so I wrote a couple of quick ones for the more interesting bits of functionality. Reviewers: zturner, erik.pilkington Subscribers: mgorny, cfe-commits, llvm-commits Differential Revision: https://reviews.llvm.org/D47887 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334399 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Remove masking from dbpsadbw builtins, use select builtin instead.Craig Topper2018-06-111-3/+3
| | | | git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334385 91177308-0d34-0410-b5e6-96231b3b80d8
* [NEON] Support VST1xN intrinsics in AArch32 mode (Clang part)Ivan A. Kosarev2018-06-101-6/+9
| | | | | | | | | | We currently support them only in AArch64. The NEON Reference, however, says they are 'ARMv7, ARMv8' intrinsics. Differential Revision: https://reviews.llvm.org/D47446 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334362 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Remove masking from the 512-bit packed floating point add/sub/mul/div ↵Craig Topper2018-06-101-8/+8
| | | | | | builtins. Use select in IR instead. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334359 91177308-0d34-0410-b5e6-96231b3b80d8
* [analyzer] Add dangling internal buffer check.Reka Kovacs2018-06-091-0/+5
| | | | | | | | | | | | This check will mark raw pointers to C++ standard library container internal buffers 'released' when the objects themselves are destroyed. Such information can be used by MallocChecker to warn about use-after-free problems. In this first version, 'std::basic_string's are supported. Differential Revision: https://reviews.llvm.org/D47135 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334348 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Add avx512 feature flags to __builtin_ia32_select*.Craig Topper2018-06-081-18/+18
| | | | | | | | There are many masked intrinsics that just wrap a select around a legacy intrinsic from a pre-avx512 instruciton set. If that intrinsic is implemented as a macro, nothing prevents it from being used when only the older feature was enabled. This likely generates very poor code since we don't have a good way to convert from the scalar masked type used by the intrinsic into a vector control for a legacy blend instruction. If we even have a blend instruction to use. By adding a feature to the select builtins we can prevent and diagnose misuse of these intrinsics. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334334 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Add back some masked vector truncate builtins. Custom IRgen a a few ↵Craig Topper2018-06-081-0/+2
| | | | | | | | | | others. I'd like to make the select builtins require an avx512f, avx512bw, or avx512vl fature to match what is normally required to get masking. Truncate is special in that there are instructions with a 128/256-bit masked result even without avx512vl. By using special buitlins we can emit a select without using the 128/256-bit select builtins. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334331 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Fold masking into subvector extract builtins.Craig Topper2018-06-081-12/+12
| | | | | | | | I'm looking into making the select builtins require avx512f, avx512bw, or avx512vl since masking operations generally require those features. The extract builtins are funny because the 512-bit versions return a 128 or 256 bit vector with masking even when avx512vl is not supported. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334330 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Add builtins for vpermq/vpermpd instructions to enable target feature ↵Craig Topper2018-06-081-0/+4
| | | | | | checking. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334311 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Change immediate type for some builtins from char to int.Craig Topper2018-06-081-12/+12
| | | | | | These builtins are all handled by CGBuiltin.cpp so it doesn't much matter what the immediate type is, but int matches the intrinsic spec. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334310 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Add builtins for shufps and shufpd to enable target feature and ↵Craig Topper2018-06-081-0/+6
| | | | | | immediate range checking. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334266 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Add builtins for pshufd, pshuflw, and pshufhw to enable target feature ↵Craig Topper2018-06-081-0/+9
| | | | | | and immediate range checking. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334265 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Add subvector insert and extract builtins to enable target feature ↵Craig Topper2018-06-081-0/+32
| | | | | | | | checking and immediate range checking. Test changes are due to differences in how we generate undef elements now. We also changed the types used for extractf128_si256/insertf128_si256 to match the signature of the builtin that previously existed which this patch resurrects. This also matches gcc. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334261 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Add builtins for vpermilps/pd instructions to enable target feature ↵Craig Topper2018-06-081-0/+6
| | | | | | checking. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334256 91177308-0d34-0410-b5e6-96231b3b80d8
* Reapply "[Parse] Use CapturedStmt for @finally on MSVC"Shoaib Meenai2018-06-083-1/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reapplies r334224 and adds explicit triples to some tests to fix them on Windows (where otherwise they would have run with the default windows-msvc triple, which I'm changing the behavior for). Original commit message: The body of a `@finally` needs to be executed on both exceptional and non-exceptional paths. On landingpad platforms, this is straightforward: the `@finally` body is emitted as a normal (non-exceptional) cleanup, and then a catch-all is emitted which branches to that cleanup (the cleanup has code to conditionally re-throw based on a flag which is set by the catch-all). Unfortunately, we can't use the same approach for MSVC exceptions, where the catch-all will be emitted as a catchpad. We can't just branch to the cleanup from within the catchpad, since we can only exit it via a catchret, at which point the exception is destroyed and we can't rethrow. We could potentially emit the finally body inside the catchpad and have the normal cleanup path somehow branch into it, but that would require some new IR construct that could branch into a catchpad. Instead, after discussing it with Reid Kleckner, we decided that frontend outlining was the best approach, similar to how SEH `__finally` works today. We decided to use CapturedStmt (which was also suggested by Reid) rather than CaptureFinder (which is what `__finally` uses) since the latter doesn't handle a lot of cases we care about, e.g. self accesses, property accesses, block captures, etc. Extending CaptureFinder to handle those additional cases proved unwieldy, whereas CapturedStmt already took care of all of those. In theory `__finally` could also be moved over to CapturedStmt, which would remove some existing limitations (e.g. the inability to capture this), although CaptureFinder would still be needed for SEH filters. The one case supported by `@finally` but not CapturedStmt (or CaptureFinder for that matter) is arbitrary control flow out of the `@finally`, e.g. having a return statement inside a `@finally`. We can add that support as a follow-up, but in practice we've found it to be used very rarely anyway. Differential Revision: https://reviews.llvm.org/D47564 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334251 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Add builtins for blend with immediate control to enforce target ↵Craig Topper2018-06-081-0/+8
| | | | | | feature requirements and check immediate range. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334249 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Add builtins for shuff32x4/shuff64x2/shufi32x4/shuff64x2 to enable ↵Craig Topper2018-06-071-0/+8
| | | | | | target feature checking and immediate range checking. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334244 91177308-0d34-0410-b5e6-96231b3b80d8
* [Frontend] Disallow non-MSVC exception models for windows-msvc targetsShoaib Meenai2018-06-071-0/+2
| | | | | | | | | | | | | | The windows-msvc target is used for MSVC ABI compatibility, including the exceptions model. It doesn't make sense to pair a windows-msvc target with a non-MSVC exception model. This would previously cause an assertion failure; explicitly error out for it in the frontend instead. This also allows us to reduce the matrix of target/exception models a bit (see the modified tests), and we can possibly simplify some of the personality code in a follow-up. Differential Revision: https://reviews.llvm.org/D47853 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334243 91177308-0d34-0410-b5e6-96231b3b80d8
* Revert "[Parse] Use CapturedStmt for @finally on MSVC"Shoaib Meenai2018-06-073-4/+1
| | | | | | | | | | This reverts commit r334224. This is causing buildbot failures on Windows, presumably because some tests don't specify a triple. I'll test this on Windows locally and recommit with the tests fixed. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334240 91177308-0d34-0410-b5e6-96231b3b80d8
* [MS] Re-add support for the ARM interlocked bittest intrinscsReid Kleckner2018-06-072-4/+12
| | | | | | | | | | | | | | | Adds support for these intrinsics, which are ARM and ARM64 only: _interlockedbittestandreset_acq _interlockedbittestandreset_rel _interlockedbittestandreset_nf _interlockedbittestandset_acq _interlockedbittestandset_rel _interlockedbittestandset_nf Refactor the bittest intrinsic handling to decompose each intrinsic into its action, its width, and its atomicity. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334239 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Add builtins for VALIGNQ/VALIGND to enable proper target feature checking.Craig Topper2018-06-071-0/+6
| | | | | | | | We still emit shufflevector instructions we just do it from CGBuiltin.cpp now. This ensures the intrinsics that use this are only available on CPUs that support the feature. I also added range checking to the immediate, but only checked it is 8 bits or smaller. We should maybe be stricter since we never use all 8 bits, but gcc doesn't seem to do that. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334237 91177308-0d34-0410-b5e6-96231b3b80d8
* [Parse] Use CapturedStmt for @finally on MSVCShoaib Meenai2018-06-073-1/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The body of a `@finally` needs to be executed on both exceptional and non-exceptional paths. On landingpad platforms, this is straightforward: the `@finally` body is emitted as a normal (non-exceptional) cleanup, and then a catch-all is emitted which branches to that cleanup (the cleanup has code to conditionally re-throw based on a flag which is set by the catch-all). Unfortunately, we can't use the same approach for MSVC exceptions, where the catch-all will be emitted as a catchpad. We can't just branch to the cleanup from within the catchpad, since we can only exit it via a catchret, at which point the exception is destroyed and we can't rethrow. We could potentially emit the finally body inside the catchpad and have the normal cleanup path somehow branch into it, but that would require some new IR construct that could branch into a catchpad. Instead, after discussing it with Reid Kleckner, we decided that frontend outlining was the best approach, similar to how SEH `__finally` works today. We decided to use CapturedStmt (which was also suggested by Reid) rather than CaptureFinder (which is what `__finally` uses) since the latter doesn't handle a lot of cases we care about, e.g. self accesses, property accesses, block captures, etc. Extending CaptureFinder to handle those additional cases proved unwieldy, whereas CapturedStmt already took care of all of those. In theory `__finally` could also be moved over to CapturedStmt, which would remove some existing limitations (e.g. the inability to capture this), although CaptureFinder would still be needed for SEH filters. The one case supported by `@finally` but not CapturedStmt (or CaptureFinder for that matter) is arbitrary control flow out of the `@finally`, e.g. having a return statement inside a `@finally`. We can add that support as a follow-up, but in practice we've found it to be used very rarely anyway. Differential Revision: https://reviews.llvm.org/D47564 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334224 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Add back builtins for _mm_slli_si128/_mm_srli_si128 and similar ↵Craig Topper2018-06-071-0/+6
| | | | | | | | | | intrinsics. We still lower them to native shuffle IR, but we do it in CGBuiltin.cpp now. This allows us to check the target feature and ensure the immediate fits in 8 bits. This also improves our -O0 codegen slightly because we're able to see the zeroinitializer in the shuffle. It looks like it got lost behind a store+load previously. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334208 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Add back _mask, _maskz, and _mask3 builtins for some 512-bit ↵Craig Topper2018-06-071-4/+16
| | | | | | | | | | | | | | | | | | | | | | | fmadd/fmsub/fmaddsub/fmsubadd builtins. Summary: We recently switch to using a selects in the intrinsics header files for FMA instructions. But the 512-bit versions support flavors with rounding mode which must be an Integer Constant Expression. This has forced those intrinsics to be implemented as macros. As it stands now the mask and mask3 intrinsics evaluate one of their macro arguments twice. If that argument itself is another intrinsic macro, we can end up over expanding macros. Or if its something we can CSE later it would show up multiple times when it shouldn't. I tried adding __extension__ around the macro and making it an expression statement and declaring a local variable. But whatever name you choose for the local variable can never be used as the name of an input to the macro in user code. If that happens you would end up with the same name on the LHS and RHS of an assignment after expansion. We might be safe if we use __ in front of the variable names because those names are reserved and user code shouldn't use that, but I wasn't sure I wanted to make that claim. The other option which I've chosen here, is to add back _mask, _maskz, and _mask3 flavors of the builtin which we will expand in CGBuiltin.cpp to replicate the argument as needed and insert any fneg needed on the third operand to make a subtract. The _maskz isn't truly necessary if we have an unmasked version or if we use the masked version with a -1 mask and wrap a select around it. But I've chosen to make things more uniform. I separated out the scalar builtin handling to avoid too many things going on in EmitX86FMAExpr. It was different enough due to the extract and insert that the minor duplication of the CreateCall was probably worth it. Reviewers: tkrupa, RKSimon, spatel, GBuella Reviewed By: tkrupa Subscribers: cfe-commits Differential Revision: https://reviews.llvm.org/D47724 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334159 91177308-0d34-0410-b5e6-96231b3b80d8
* ClangTidy fix - 'clang::Sema::checkAllowedCUDAInitializer' has a definition ↵Han Shen2018-06-071-1/+1
| | | | | | with different parameter names. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334155 91177308-0d34-0410-b5e6-96231b3b80d8
* Change the wording of RTTI errors to make them more generic.Sunil Srivastava2018-06-071-2/+2
| | | | | | | | | | | | | | | | | An attempt to use dynamic_cast while rtti is disabled, used to emit the error: cannot use dynamic_cast with -fno-rtti and a similar one for typeid. This patch changes that to: use of dynamic_cast requires -frtti Differential Revision: https://reviews.llvm.org/D47291 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334153 91177308-0d34-0410-b5e6-96231b3b80d8
* [CUDA] Check initializers of instantiated template variables.Artem Belevich2018-06-061-0/+10
| | | | | | | | | We were already performing checks on non-template variables, but the checks on templated ones were missing. Differential Revision: https://reviews.llvm.org/D45231 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334143 91177308-0d34-0410-b5e6-96231b3b80d8
* [Frontend] Honor UserFilesAreVolatile flag getting file buffer in ASTUnitIvan Donchevskii2018-06-061-1/+1
| | | | | | | | | Do not memory map the main file if the flag UserFilesAreVolatile is set to true in ASTUnit when calling FileSystem::getBufferForFile. Differential Revision: https://reviews.llvm.org/D47460 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334070 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Move the vec_set/vec_ext builtins for 64-bit elements to ↵Craig Topper2018-06-062-4/+4
| | | | | | | | BuiltinsX86_64.def. The instructions these correspond to and the intrinsics that use them are only available in 64-bit mode. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334061 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Add builtins for vector element insert and extract for different 128 ↵Craig Topper2018-06-061-0/+17
| | | | | | | | | | | | | | and 256 bit vector types. Use them to implement the extract and insert intrinsics. Previously we were just using extended vector operations in the header file. This unfortunately allowed non-constant indices to be used with the intrinsics. This is incompatible with gcc, icc, and MSVC. It also introduces a different performance characteristic because non-constant index gets lowered to a vector store and an element sized load. By adding the builtins we can check for the index to be a constant and ensure its in range of the vector element count. User code still has the option to use extended vector operations themselves if they need non-constant indexing. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334057 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Make __builtin_ia32_vec_ext_v2si require ICE for its index argument. ↵Craig Topper2018-06-051-1/+1
| | | | | | | | Add warnings for out of range indices for __builtin_ia32_vec_ext_v2si, __builtin_ia32_vec_ext_v4hi, and __builtin_ia32_vec_set_v4hi. These should take a constant value for an index and that constant should be a valid element number. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@334051 91177308-0d34-0410-b5e6-96231b3b80d8
* [X86] Mark all the builtins and intrinsics that require MMX and an SSE ↵Craig Topper2018-06-051-37/+37
| | | | | | | | feature as requiring both mmx and the sse feature. Previously we only checked the sse feature, but this means that if you passed -mno-mmx, the builtins/intrinsics wouldn't be disabled in the frontend and would instead fail backend isel. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@333980 91177308-0d34-0410-b5e6-96231b3b80d8
* Reimplement the bittest intrinsic family as builtins with inline asmReid Kleckner2018-06-051-1/+12
| | | | | | | | | | | We need to implement _interlockedbittestandset as a builtin for windows.h, so we might as well do the whole family. It reduces code duplication anyway. Fixes PR33188, a long standing bug in our bittest implementation encountered by Chakra. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@333978 91177308-0d34-0410-b5e6-96231b3b80d8
View Lorry Controller Status