summaryrefslogtreecommitdiffstats
path: root/clang/lib/CodeGen
Commit message (Collapse)AuthorAgeFilesLines
...
* [AArch64] Enable fp16 data type for the Builtin for AArch64 only.Abderrazek Zaafrani2017-12-212-24/+37
| | | | | | Differential Revision: https:://reviews.llvm.org/D41360 llvm-svn: 321301
* Revert "Fix for PR32990"Reid Kleckner2017-12-211-33/+9
| | | | | | | | | | | | This reverts commit r321239. It broke the Chromium DLL build: [8834/50217] LINK(DLL) icui18n.dll icui18n.dll.lib icui18n.dll.pdb FAILED: icui18n.dll icui18n.dll.lib icui18n.dll.pdb zrule.obj : error LNK2001: unresolved external symbol "__declspec(dllimport) public: void __cdecl icu_60::UnicodeString::`vbase destructor'(void)" (__imp_??_DUnicodeString@icu_60@@QEAAXXZ) llvm-svn: 321298
* [CodeGen] Fix crash when a function taking transparent union is redeclared.Volodymyr Sapsai2017-12-211-5/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When a function taking transparent union is declared as taking one of union members earlier in the translation unit, clang would hit an "Invalid cast" assertion during EmitFunctionProlog. This case corresponds to function f1 in test/CodeGen/transparent-union-redecl.c. We decided to cast i32 to union because after merging function declarations function parameter type becomes int, CGFunctionInfo::ArgInfo type matches with ABIArgInfo type, so we decide it is a trivial case. But these types should also be castable to parameter declaration type which is not the case here. The fix is in checking for the trivial case if ABIArgInfo type matches with parameter declaration type. It exposed inconsistency that we check hasScalarEvaluationKind for different types in EmitParmDecl and EmitFunctionProlog, and comment says they should match. Additional tests in Sema/transparent-union.c capture current behavior and make sure there are no regressions. rdar://problem/34949329 Reviewers: rjmccall, rafael Reviewed By: rjmccall Subscribers: aemerson, cfe-commits, kristof.beyls Differential Revision: https://reviews.llvm.org/D41311 llvm-svn: 321296
* [AARch64] Add ARMv8.2-A FP16 vector intrinsicsAbderrazek Zaafrani2017-12-213-7/+179
| | | | | | | | Putting back the code that was reverted few weeks ago. Differential Revision: https://reviews.llvm.org/D34161 llvm-svn: 321294
* [CodeGen] Fix access sizes in new-format TBAA tagsIvan A. Kosarev2017-12-211-1/+3
| | | | | | | | | | | | The new format requires to specify both the type of the access and its size. This patch fixes setting access sizes for TBAA tags that denote accesses to structure members. This fix affects all future TBAA metadata tests for the new format, so I guess we don't need any special tests for this fix. Differential Revision: https://reviews.llvm.org/D41452 llvm-svn: 321250
* Fix for PR32990Erich Keane2017-12-211-9/+33
| | | | | | | | | This fixes the bug in https://bugs.llvm.org/show_bug.cgi?id=32990. Patch By: zahiraam Differential Revision: https://reviews.llvm.org/D39063 llvm-svn: 321239
* [ubsan] Diagnose noreturn functions which returnVedant Kumar2017-12-215-17/+45
| | | | | | | | | | | | | | | | | | Diagnose 'unreachable' UB when a noreturn function returns. 1. Insert a check at the end of functions marked noreturn. 2. A decl may be marked noreturn in the caller TU, but not marked in the TU where it's defined. To diagnose this scenario, strip away the noreturn attribute on the callee and insert check after calls to it. Testing: check-clang, check-ubsan, check-ubsan-minimal, D40700 rdar://33660464 Differential Revision: https://reviews.llvm.org/D40698 llvm-svn: 321231
* [hwasan] Implement -fsanitize-recover=hwaddress.Evgeniy Stepanov2017-12-201-1/+5
| | | | | | | | | | | | Summary: Very similar to AddressSanitizer, with the exception of the error type encoding. Reviewers: kcc, alekseyshl Subscribers: cfe-commits, kubamracek, llvm-commits, hiraditya Differential Revision: https://reviews.llvm.org/D41417 llvm-svn: 321203
* [Complex] Don't use __div?c3 when building with fast-math.Florian Hahn2017-12-201-8/+29
| | | | | | | | | | | | Summary: Plant an inline version of "((ac+bd)/(cc+dd)) + i((bc-ad)/(cc+dd))" instead. Patch by Paul Walker. Reviewed By: hfinkel Differential Revision: https://reviews.llvm.org/D40299 llvm-svn: 321183
* Silence a bunch of implicit fallthrough warningsAdrian Prantl2017-12-196-26/+26
| | | | llvm-svn: 321115
* [Coverage] Fix use-after free in coverage emissionEli Friedman2017-12-191-1/+5
| | | | | | | | | | | Fixes regression from r320533. This fixes the undefined behavior, but I'm not sure it's really right... I think we end up with missing coverage for code in modules. Differential Revision: https://reviews.llvm.org/D41374 llvm-svn: 321052
* Fix the reference to the now renamed member of TBAAStructFieldIvan A. Kosarev2017-12-181-1/+1
| | | | | | See https://reviews.llvm.org/D39956 for details. llvm-svn: 320994
* Revert r320978 "No -fsanitize=function warning when calling noexcept ↵Stephan Bergmann2017-12-181-6/+2
| | | | | | | | | | | | | function through non-noexcept pointer in C++17" At least <http://lab.llvm.org:8011/builders/sanitizer-x86_64-linux-android/ builds/6013/steps/annotate/logs/stdio> complains about __ubsan::__ubsan_handle_function_type_mismatch_abort (compiler-rt lib/ubsan/ubsan_handlers.cc) returning now despite being declared 'noreturn', so looks like a different approach is needed for the function_type_mismatch check to be called also in cases that may ultimately succeed. llvm-svn: 320982
* No -fsanitize=function warning when calling noexcept function through ↵Stephan Bergmann2017-12-181-2/+6
| | | | | | | | | | | | | | | | | | | | | | | | | non-noexcept pointer in C++17 As discussed in the mail thread <https://groups.google.com/a/isocpp.org/forum/ #!topic/std-discussion/T64_dW3WKUk> "Calling noexcept function throug non- noexcept pointer is undefined behavior?", such a call should not be UB. However, Clang currently warns about it. There is no cheap check whether two function type_infos only differ in noexcept, so pass those two type_infos as additional data to the function_type_mismatch handler (with the optimization of passing a null "static callee type" info when that is already noexcept, so the additional check can be avoided anyway). For the Itanium ABI (which appears to be the only one that happens to be used on platforms that support -fsanitize=function, and which appears to only record noexcept information for pointer-to-function type_infos, not for function type_infos themselves), we then need to check the mangled names for occurrence of "Do" representing "noexcept". This is the cfe part of a patch covering both cfe and compiler-rt. Differential Revision: https://reviews.llvm.org/D40720 llvm-svn: 320978
* Refactor overridden methods iteration to avoid double lookups.Benjamin Kramer2017-12-171-1/+1
| | | | | | Convert most uses to range-for loops. No functionality change intended. llvm-svn: 320954
* [Driver, CodeGen] pass through and apply -fassociative-mathSanjay Patel2017-12-161-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are 2 parts to getting the -fassociative-math command-line flag translated to LLVM FMF: 1. In the driver/frontend, we accept the flag and its 'no' inverse and deal with the interactions with other flags like -ffast-math -fno-signed-zeros -fno-trapping-math. This was mostly already done - we just need to translate the flag as a codegen option. The test file is complicated because there are many potential combinations of flags here. Note that we are matching gcc's behavior that requires 'nsz' and no-trapping-math. 2. In codegen, we map the codegen option to FMF in the IR builder. This is simple code and corresponding test. For the motivating example from PR27372: float foo(float a, float x) { return ((a + x) - x); } $ ./clang -O2 27372.c -S -o - -ffast-math -fno-associative-math -emit-llvm | egrep 'fadd|fsub' %add = fadd nnan ninf nsz arcp contract float %0, %1 %sub = fsub nnan ninf nsz arcp contract float %add, %2 So 'reassoc' is off as expected (and so is the new 'afn' but that's a different patch). This case now works as expected end-to-end although the underlying logic is still wrong: $ ./clang -O2 27372.c -S -o - -ffast-math -fno-associative-math | grep xmm addss %xmm1, %xmm0 subss %xmm1, %xmm0 We're not done because the case where 'reassoc' is set is ignored by optimizer passes. Example: $ ./clang -O2 27372.c -S -o - -fassociative-math -fno-signed-zeros -fno-trapping-math -emit-llvm | grep fadd %add = fadd reassoc float %0, %1 $ ./clang -O2 27372.c -S -o - -fassociative-math -fno-signed-zeros -fno-trapping-math | grep xmm addss %xmm1, %xmm0 subss %xmm1, %xmm0 Differential Revision: https://reviews.llvm.org/D39812 llvm-svn: 320920
* [X86] Implement kand/kandn/kor/kxor/kxnor/knot intrinsics using native IR.Craig Topper2017-12-161-0/+29
| | | | llvm-svn: 320919
* [X86] Add builtins and tests for 128 and 256 bit vpopcntdq.Craig Topper2017-12-161-0/+4
| | | | llvm-svn: 320915
* [CodeGen] Specialize mixed-sign mul-with-overflow (fix PR34920)Vedant Kumar2017-12-161-0/+95
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch introduces a specialized way to lower overflow-checked multiplications with mixed-sign operands. This fixes link failures and ICEs on code like this: void mul(int64_t a, uint64_t b) { int64_t res; __builtin_mul_overflow(a, b, &res); } The generic checked-binop irgen would use a 65-bit multiplication intrinsic here, which requires runtime support for _muloti4 (128-bit multiplication), and therefore fails to link on i386. To get an ICE on x86_64, change the example to use __int128_t / __uint128_t. Adding runtime and backend support for 65-bit or 129-bit checked multiplication on all of our supported targets is infeasible. This patch solves the problem by using simpler, specialized irgen for the mixed-sign case. llvm.org/PR34920, rdar://34963321 Testing: Apart from check-clang, I compared the output from this fairly comprehensive test driver using unpatched & patched clangs: https://gist.github.com/vedantk/3eb9c88f82e5c32f2e590555b4af5081 Differential Revision: https://reviews.llvm.org/D41149 llvm-svn: 320902
* [OPENMP] Codegen `declare simd` for function declarations.Alexey Bataev2017-12-152-4/+3
| | | | | | | Previously the attributes were emitted only for function definitions. Patch adds emission of the attributes for function declarations. llvm-svn: 320826
* Fix many -Wsign-compare and -Wtautological-constant-compare warnings.Zachary Turner2017-12-141-1/+1
| | | | | | | | | | | | Most of the -Wsign-compare warnings are due to the fact that enums are signed by default in the MS ABI, while the tautological comparison warnings trigger on x86 builds where sizeof(size_t) is 4 bytes, so N > numeric_limits<unsigned>::max() is always false. Differential Revision: https://reviews.llvm.org/D41256 llvm-svn: 320750
* [CodeGen][X86] Implement _InterlockedCompareExchange128 intrinsicReid Kleckner2017-12-141-0/+39
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary: InterlockedCompareExchange128 is a bit more complicated than the other InterlockedCompareExchange functions, so it requires a bit more work. It doesn't directly refer to 128bit ints, instead it takes pointers to 64bit ints for Destination and ComparandResult, and exchange is taken as two 64bit ints (high & low). The previous value is written to ComparandResult, and success is returned. This implementation does the following in order to produce a cmpxchg instruction: 1. Cast everything to 128bit ints or int pointers, and glues together the Exchange values 2. Reads from CompareandResult to get the comparand 3. Calls cmpxchg volatile (on X86 this will produce a lock cmpxchg16b instruction) 1. Result 0 (previous value) is written back to ComparandResult 2. Result 1 (success bool) is zext'ed to a uchar and returned Resolves bug https://llvm.org/PR35251 Patch by Colden Cullen! Reviewers: rnk, agutowski Reviewed By: rnk Subscribers: majnemer, cfe-commits Differential Revision: https://reviews.llvm.org/D41032 llvm-svn: 320730
* In an ARC lambda-to-block conversion thunk, reclaim the return value ofJohn McCall2017-12-141-2/+5
| | | | | | | | the lambda so that we don't over-release it. Patch by Dan Zimmerman! llvm-svn: 320721
* [OPENMP] Add codegen for target data constructs with `nowait` clause.Alexey Bataev2017-12-141-3/+64
| | | | | | Added codegen for the `nowait` clause in target data constructs. llvm-svn: 320717
* [c++20] P0515R3: Parsing support and basic AST construction for operator <=>.Richard Smith2017-12-142-0/+2
| | | | | | | | | | | | | | | Adding the new enumerator forced a bunch more changes into this patch than I would have liked. The -Wtautological-compare warning was extended to properly check the new comparison operator, clang-format needed updating because it uses precedence levels as weights for determining where to break lines (and several operators increased their precedence levels with this change), thread-safety analysis needed changes to build its own IL properly for the new operator. All "real" semantic checking for this operator has been deferred to a future patch. For now, we use the relational comparison rules and arbitrarily give the builtin form of the operator a return type of 'void'. llvm-svn: 320707
* IRGen: When performing CFI checks, load vtable pointer from vbase when ↵Peter Collingbourne2017-12-135-12/+47
| | | | | | | | | | | | | | necessary. Under the Microsoft ABI, it is possible for an object not to have a virtual table pointer of its own if all of its virtual functions were introduced by virtual bases. In that case, we need to load the vtable pointer from one of the virtual bases and perform the type check using its type. Differential Revision: https://reviews.llvm.org/D41036 llvm-svn: 320638
* [OPENMP] Add codegen for `nowait` clause in target directives.Alexey Bataev2017-12-131-3/+50
| | | | | | Added basic codegen for `nowait` clauses in target-based directives. llvm-svn: 320613
* [Hexagon] Intrinsic support for V62 and V65Krzysztof Parzyszek2017-12-132-0/+58
| | | | llvm-svn: 320609
* [OPENMP] Initial codegen for `target teams distribute simd` directive.Alexey Bataev2017-12-133-11/+53
| | | | | | | Host + generic device codegen for `target teams distribute simd` directive. llvm-svn: 320608
* [OPENMP] Support `reduction` clause on target-based directives.Alexey Bataev2017-12-131-9/+17
| | | | | | | OpenMP 5.0 added support for `reduction` clause in target-based directives. Patch adds this support to clang. llvm-svn: 320596
* [Coverage] Always emit unused coverage mappings in the same order.Eli Friedman2017-12-132-14/+4
| | | | | | | | Non-determinism is confusing at best. Differential Revision: https://reviews.llvm.org/D41140 llvm-svn: 320533
* [OpenMP] Add function attribute for triggering data sharing.Gheorghe-Teodor Bercea2017-12-121-0/+2
| | | | | | | | | | | | | | | | Summary: The backend should only emit data sharing code for the cases where it is needed. A new function attribute is used by Clang to enable data sharing only for the cases where OpenMP semantics require it and there are variables that need to be shared. Reviewers: hfinkel, Hahnfeld, ABataev, carlo.bertolli, caomhin Reviewed By: ABataev Subscribers: cfe-commits, jholewinski Differential Revision: https://reviews.llvm.org/D41123 llvm-svn: 320527
* [Driver][CodeGen] Add -mprefer-vector-width driver option and attribute ↵Craig Topper2017-12-111-0/+5
| | | | | | | | | | | | | | | | during CodeGen. This adds a new command line option -mprefer-vector-width to specify a preferred vector width for the vectorizers. Valid values are 'none' and unsigned integers. The driver will check that it meets those constraints. Specific supported integers will be managed by the targets in the backend. Clang will take the value and add it as a new function attribute during CodeGen. This represents the alternate direction proposed by Sanjay in this RFC: http://lists.llvm.org/pipermail/llvm-dev/2017-November/118734.html The syntax here matches gcc, though gcc treats it as an x86 specific command line argument. gcc only allows values of 128, 256, and 512. I'm not having clang check any values. Differential Revision: https://reviews.llvm.org/D40230 llvm-svn: 320419
* Hardware-assisted AddressSanitizer (clang part).Evgeniy Stepanov2017-12-095-4/+25
| | | | | | | | | | | | | | Summary: Driver, frontend and LLVM codegen for HWASan. A clone of ASan, basically. Reviewers: kcc, pcc, alekseyshl Subscribers: srhines, javed.absar, cfe-commits Differential Revision: https://reviews.llvm.org/D40936 llvm-svn: 320232
* [CodeGen][X86] Fix handling of __fp16 vectors.Akira Hatanaka2017-12-093-11/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This commit fixes a bug in IRGen where it generates completely broken code for __fp16 vectors on X86. For example when the following code is compiled: half4 hv0, hv1, hv2; // these are vectors of __fp16. void foo221() { hv0 = hv1 + hv2; } clang generates the following IR, in which two i16 vectors are added: @hv1 = common global <4 x i16> zeroinitializer, align 8 @hv2 = common global <4 x i16> zeroinitializer, align 8 @hv0 = common global <4 x i16> zeroinitializer, align 8 define void @foo221() { %0 = load <4 x i16>, <4 x i16>* @hv1, align 8 %1 = load <4 x i16>, <4 x i16>* @hv2, align 8 %add = add <4 x i16> %0, %1 store <4 x i16> %add, <4 x i16>* @hv0, align 8 ret void } To fix the bug, this commit uses the code committed in r314056, which modified clang to promote and truncate __fp16 vectors to and from float vectors in the AST. It also fixes another IRGen bug where a short value is assigned to an __fp16 variable without any integer-to-floating-point conversion, as shown in the following example: __fp16 a; short b; void foo1() { a = b; } @b = common global i16 0, align 2 @a = common global i16 0, align 2 define void @foo1() #0 { %0 = load i16, i16* @b, align 2 store i16 %0, i16* @a, align 2 ret void } rdar://problem/20625184 Differential Revision: https://reviews.llvm.org/D40112 llvm-svn: 320215
* [OPENMP] Simplify codegen for loop iteration variables in loop preamble.Alexey Bataev2017-12-081-1/+6
| | | | | | | Initial patch could cause trouble in the optimized code because of the incorrectly generated lifetime intrinsics. llvm-svn: 320191
* [ubsan] array-bounds: Ignore params with constant sizeVedant Kumar2017-12-081-8/+0
| | | | | | | | | | This is a follow-up to r320128. Eli pointed out that there is some gray area in the language standard about whether the constant size is exact, or a lower bound. https://reviews.llvm.org/D40940 llvm-svn: 320185
* [OPENMP] Initial codegen for `target teams distribute` directive.Alexey Bataev2017-12-083-14/+58
| | | | | | Host + default devices codegen for `target teams distribute` directive. llvm-svn: 320149
* [Blocks] Inherit sanitizer options from parent declVedant Kumar2017-12-081-1/+3
| | | | | | | | | | | | There is no way to apply sanitizer suppressions to ObjC blocks. A reasonable default is to have blocks inherit their parent's sanitizer options. rdar://32769634 Differential Revision: https://reviews.llvm.org/D40668 llvm-svn: 320132
* [ubsan] Use pass_object_size info in bounds checksVedant Kumar2017-12-082-0/+59
| | | | | | | | | Teach UBSan's bounds check to opportunistically use pass_object_size information to check array accesses. rdar://33272922 llvm-svn: 320128
* [driver] Set the 'simulator' environment for Darwin when compiling forAlex Lorenz2017-12-071-4/+1
| | | | | | | | | | iOS/tvOS/watchOS simulator rdar://35135215 Differential Revision: https://reviews.llvm.org/D40682 llvm-svn: 320073
* CodeGen: Fix invalid bitcasts for memcpyYaxun Liu2017-12-071-4/+4
| | | | | | | | | | | | CreateCoercedLoad/CreateCoercedStore assumes pointer argument of memcpy is in addr space 0, which is not correct and causes invalid bitcasts for triple amdgcn---amdgiz. It is fixed by using alloca addr space instead. Differential Revision: https://reviews.llvm.org/D40806 llvm-svn: 320000
* Fix PR35542: Correct adjusting of private reduction variableJonas Hahnfeld2017-12-061-3/+6
| | | | | | | | | | | | The adjustment is calculated with CreatePtrDiff() which returns the difference in (base) elements. This is passed to CreateGEP() so make sure that the GEP base has the correct pointer type: It needs to be a pointer to the base type, not a pointer to a constant sized array. Differential Revision: https://reviews.llvm.org/D40911 llvm-svn: 319931
* [OPENMP] Initial codegen for `teams distribute simd` directive.Alexey Bataev2017-12-061-13/+23
| | | | | | Host + default devices codegen for `teams distribute simd` directive. llvm-svn: 319896
* [OpenCL] Fix layering violation by getOpenCLTypeAddrSpaceSven van Haastregt2017-12-061-3/+3
| | | | | | | | | | | | Commit 7ac28eb0a5 / r310911 ("[OpenCL] Allow targets to select address space per type", 2017-08-15) made Basic depend on AST, introducing a circular dependency. Break this dependency by adding the OpenCLTypeKind enum in Basic and map from AST types to this enum in ASTContext. Differential Revision: https://reviews.llvm.org/D40838 llvm-svn: 319883
* [OPENMP] Fix PR35486: crash when collapsing loops with dependent iteration ↵Alexey Bataev2017-12-041-0/+3
| | | | | | | | | | spaces. Though it is incorrect from point of view of OpenMP standard to have dependent iteration space in OpenMP loops, compiler should not crash. Patch fixes this problem. llvm-svn: 319700
* [OpenMP] Initial implementation of code generation for pragma 'teams ↵Carlo Bertolli2017-12-041-12/+22
| | | | | | | | | | distribute parallel for simd' on host https://reviews.llvm.org/D40795 This includes regression tests for all associated clauses. llvm-svn: 319696
* [OPENMP] Codegen for `distribute simd` directive.Alexey Bataev2017-12-041-21/+54
| | | | | | Initial codegen support for `distribute simd` directive. llvm-svn: 319661
* Revert "[CodeGen] Add initial support for union members in TBAA"Hal Finkel2017-12-034-67/+32
| | | | | | | | This reverts commit r319413. See PR35503. We can't use "union member" as the access type here like this. llvm-svn: 319629
* [CodeGen] fix mapping from fmod calls to frem instructionSanjay Patel2017-12-021-14/+18
| | | | | | Similar to D40044 and discussed in D40594. llvm-svn: 319619
OpenPOWER on IntegriCloud