This primarily impacts the Windows MSVC and Windows itanium
environments. Windows MSVC does not use `__cxa_atexit` and Itanium
follows suit. Simplify the logic for the default value calculation and
blanket the Windows environments to default to off for use of
`__cxa_atexit`.
llvm-svn: 312941
Summary:
While `goog.setTestOnly` usually appears in the imports section of a file, it is
not actually an import, and also usually doesn't take long parameters (nor
namespaces as a parameter, it's a description/message that should be wrapped).
This fixes a regression where a `goog.setTestOnly` call nested in a function was
not wrapped.
Reviewers: djasper
Subscribers: klimek
Differential Revision: https://reviews.llvm.org/D37685
llvm-svn: 312918
When using a virtual file-system (VFS) and a preamble file (PCH) is generated,
it is generated on-disk in the real file-system instead of in the VFS (which
makes sense, since the VFS is read-only). However, when subsequently reading
the generated PCH, the frontend passes through the VFS it has been given --
resulting in an error and a failed parse (since the VFS doesn't contain the
PCH; the real filesystem does).
This patch fixes that by detecting when a VFS is being used for a parse that
needs to work with a PCH file, and creating an overlay VFS that includes the
PCH file from the real file-system.
This allows tests to be written which make use of both PCH files and a VFS.
Differential Revision: https://reviews.llvm.org/D37474
llvm-svn: 312917
Summary:
This causes template arguments to be traversed before the templated
declaration, which is useful for clients that expect the nodes in
the same order as they are in the source code. Additionally, there
seems to be no good reason not to do so.
This was moved here from LexicallyOrderedRecursiveASTVisitor. The tests
still reside in LexicallyOrderedRecursiveASTVisitorTest.cpp under
VisitTemplateDecls.
Reviewers: arphaman, rsmith, klimek
Subscribers: cfe-commits, klimek
Differential Revision: https://reviews.llvm.org/D37662
llvm-svn: 312911
Summary:
**Short overview:**
Fixed bug: https://bugs.llvm.org/show_bug.cgi?id=34001
Clang-format bug resulting in a strange behavior of control statements short blocks. Different flags combinations do not guarantee expected result. Turned on option AllowShortBlocksOnASingleLine does not work as intended.
**Description of the problem:**
Cpp source file UnwrappedLineFormatter does not handle AllowShortBlocksOnASingleLine flag as it should. Putting a single-line control statement without any braces, clang-format works as expected (depending on AllowShortIfStatementOnASingleLine or AllowShortLoopsOnASingleLine value). Putting a single-line control statement in braces, we can observe strange and incorrect behavior.
Our short block is intercepted by tryFitMultipleLinesInOne function. The function returns a number of lines to be merged. Unfortunately, our control statement block is not covered properly. There are several if-return statements, but none of them handles our block. A block is identified by the line first token and by left and right braces. A function block works as expected, there is such an if-return statement doing proper job. A control statement block, from the other hand, falls into strange conditional construct, which depends on BraceWrapping.AfterFunction flag (with condition that the line’s last token is left brace, what is possible in our case) or goes even further. That should definitely not happen.
**Description of the patch:**
By adding three different if statements, we guarantee that our short control statement block, however it looks like (different brace wrapping flags may be turned on), is handled properly and does not fall into wrong conditional construct. Depending on appropriate options we return either 0 (when something disturbs our merging attempt) or let another function (tryMergeSimpleBlock) take the responsibility of returned result (number of merged lines). Nevertheless, one more correction is required in mentioned tryMergeSimpleBlock function. The function, previously, returned either 0 or 2. The problem was that this did not handle the case when our block had the left brace in a separate line, not the header one. After change, after adding condition, we return the result compatible with block’s structure. In case of left brace in the header’s line we do everything as before the patch. In case of left brace in a separate line we do the job similar to the one we do in case of a “non-header left brace” function short block. To be precise, we try to merge the block ignoring the header line. Then, if success, we increment our returned result.
**After fix:**
**CONFIG:**
```
AllowShortBlocksOnASingleLine: true
AllowShortIfStatementsOnASingleLine: true
BreakBeforeBraces: Custom
BraceWrapping: {
AfterClass: true, AfterControlStatement: true, AfterEnum: true, AfterFunction: true, AfterNamespace: false, AfterStruct: true, AfterUnion: true, BeforeCatch: true, BeforeElse: true
}
```
**BEFORE:**
```
if (statement) doSomething();
if (statement) { doSomething(); }
if (statement) {
doSomething();
}
if (statement)
{
doSomething();
}
if (statement)
doSomething();
if (statement) {
doSomething1();
doSomething2();
}
```
**AFTER:**
```
if (statement) doSomething();
if (statement) { doSomething(); }
if (statement) { doSomething(); }
if (statement) { doSomething(); }
if (statement) doSomething();
if (statement)
{
doSomething1();
doSomething2();
}
```
Contributed by @PriMee!
Reviewers: krasimir, djasper
Reviewed By: krasimir
Subscribers: cfe-commits, klimek
Differential Revision: https://reviews.llvm.org/D37140
llvm-svn: 312904
Summary:
The find_first_existing_file and find_first_existing_vc_file macros in
lib/Basic/CMakeLists.txt are removed. The macros are also defined in
{LLVM}/cmake/modules/AddLLVM.cmake for the same purpose. This change
serves the following 2 objectives:
1. To remove the redundant code in clang to use the same
macros in llvm,
2. The macros in AddLLVM.cmake can also handle repo for
displaying correct version information.
Reviewers: jordan_rose, cfe-commits, modocache, hintonda
Reviewed By: hintonda
Subscribers: mgorny
Differential Revision: https://reviews.llvm.org/D35533
llvm-svn: 312865
This fixes a possible crash on certain kinds of corrupted AST file, but
checking in an AST file corrupted in just the right way will be a maintenance
nightmare because the format changes frequently.
llvm-svn: 312851
Treat typedef enum as named enums instead of anonymous enums. Anonymous enums
are ignored by the warning, so previously, typedef enums were ignored as well.
llvm-svn: 312842
When performing a NSFastEnumeration, the compiler synthesizes a call to
`countByEnumeratingWithState:objects:count:` where the `count` parameter
is of type `NSUInteger` and the return type is a `NSUInteger`. We would
previously always use a `UnsignedLongTy` for the `NSUInteger` type. On
32-bit targets, `long` is 32-bits which is the same as `unsigned int`.
Most 64-bit targets are LP64, where `long` is 64-bits. However, on
LLP64 targets, such as Windows, `long` is 32-bits. Introduce new
`getNSUIntegerType` and `getNSIntegerType` helpers to allow us to
determine the correct type for the `NSUInteger` type. Wire those
through into the generation of the message dispatch to the selector.
llvm-svn: 312835
This is to fix PR34347. EmitAtomicExpr now only uses alignment information from
Type, instead of Decl, so when the declaration of an atomic variable is marked
to have the alignment equal as its size, EmitAtomicExpr doesn't know about it and
will generate libcall instead of atomic op. The patch uses EmitPointerWithAlignment
to get the precise alignment information.
Differential Revision: https://reviews.llvm.org/D37310
llvm-svn: 312830
The current coverage implementation doesn't handle region termination
very precisely. Take for example an `if' statement with a `return':
void f() {
if (true) {
return; // The `if' body's region is terminated here.
}
// This line gets the same coverage as the `if' condition.
}
If the function `f' is called, the line containing the comment will be
marked as having executed once, which is not correct.
The solution here is to create a deferred region after terminating a
region. The deferred region is completed once the start location of the
next statement is known, and is then pushed onto the region stack.
In the cases where it's not possible to complete a deferred region, it
can safely be dropped.
Testing: lit test updates, a stage2 coverage-enabled build of clang
This is a reapplication but there are no changes from the original commit.
With D36813, the segment builder in llvm will be able to handle deferred
regions correctly.
llvm-svn: 312818
This is to fix PR34347. EmitAtomicExpr now only uses alignment information from
Type, instead of Decl, so when the declaration of an atomic variable is marked
to have the alignment equal as its size, EmitAtomicExpr doesn't know about it and
will generate libcall instead of atomic op. The patch uses EmitPointerWithAlignment
to get the precise alignment information.
Differential Revision: https://reviews.llvm.org/D37310
llvm-svn: 312801
This is a recommit of r312781; in some build configurations
variable names are omitted, so changed the new regression
test accordingly.
llvm-svn: 312794
Summary:
1.Updated annotations for include/clang/StaticAnalyzer/Core/PathSensitive/Store.h, which belong to the old version of clang.
2.Delete annotations for CodeGenFunction::getEvaluationKind() in clang/lib/CodeGen/CodeGenFunction.h, which belong to the old version of clang.
Reviewers: bkramer, krasimir, klimek
Reviewed By: bkramer
Subscribers: MTC
Differential Revision: https://reviews.llvm.org/D36330
Contributed by @MTC!
llvm-svn: 312790
Summary: The crash occured when FunctionDecl was parsed with an initializer.
Reviewers: bkramer, klimek, francisco.lopes
Reviewed By: bkramer
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D37382
llvm-svn: 312788
In case of copy constructor is implicitly deleted it's still shown.
PR34402 describes a way to reproduce that.
Patch by Ivan Donchevskii!
Differential Revision: https://reviews.llvm.org/D37435
llvm-svn: 312785
This adds _Float16 as a source language type, which is a 16-bit floating point
type defined in C11 extension ISO/IEC TS 18661-3.
In follow up patches documentation and more tests will be added.
Differential Revision: https://reviews.llvm.org/D33719
llvm-svn: 312781
When we have enabled cache for global completions we did not have
diagnostics for Bar and could not complete Ba as in provided code
example.
template <typename T>
struct Foo { T member; };
template<typename T> using Bar = Foo<T>;
int main() {
Ba
}
(This is the fixed version of r 311442, which was reverted in r311445.)
Patch by Ivan Donchevskii!
Differential Revision: https://reviews.llvm.org/D35355
llvm-svn: 312780
Summary:
This is a first half(?) of a fix for the following bug:
https://bugs.llvm.org/show_bug.cgi?id=34147 (gcc -Wtype-limits)
GCC's -Wtype-limits does warn on comparison of unsigned value
with signed zero (as in, with 0), but clang only warns if the
zero is unsigned (i.e. 0U).
Also, be careful not to double-warn, or falsely warn on
comparison of signed/fp variable and signed 0.
Yes, all these testcases are needed.
Testing: $ ninja check-clang-sema check-clang-semacxx
Also, no new warnings for clang stage-2 build.
Reviewers: rjmccall, rsmith, aaron.ballman
Reviewed By: rjmccall
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D37565
llvm-svn: 312750
The ToolChain class validates the -mthread-model flag in the constructor which
doesn't work correctly since the thread model methods are virtual methods. The
check is moved into Clang::ConstructJob() when constructing the internal
command line.
https://reviews.llvm.org/D37496
Patch by: Ian Tessier!
llvm-svn: 312748
Summary:
That is, instead of "1 error generated", we now say "1 error generated
when compiling for sm_35".
This (partially) solves a usability foogtun wherein e.g. users call a
function that's only defined on sm_60 when compiling for sm_35, and they
get an unhelpful error message.
Reviewers: tra
Subscribers: sanjoy, cfe-commits
Differential Revision: https://reviews.llvm.org/D37548
llvm-svn: 312736
For now CUDA-9 is not included in the list of CUDA versions clang
searches for, so the path to CUDA-9 must be explicitly passed
via --cuda-path=.
On LLVM side NVPTX added sm_70 GPU type which bumps required
PTX version to 6.0, but otherwise is equivalent to sm_62 at the moment.
Differential Revision: https://reviews.llvm.org/D37576
llvm-svn: 312734
Block captures can have different physical locations
in memory segments depending on the use case (as a function
call or as a kernel enqueue) and in different vendor
implementations.
Therefore it's unclear how to add address space to capture
addresses uniformly. Currently it has been decided to disallow
taking addresses of captured variables until further
clarifications in the spec.
Differential Revision: https://reviews.llvm.org/D36410
llvm-svn: 312728
The current description of AllowAllParametersOfDeclarationOnNextLine in
the Clang-Format Style Options guide suggests that it is possible to
format function declaration, which fits in a single line (what is not
supported in current clang-format version). Also the example was not
reproducible and mades no sense.
Patch by Lucja Mazur, thank you!
llvm-svn: 312721
Summary:
Looks like we are out of sync between the doc and the code.
Reviewers: djasper
Reviewed By: djasper
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D37558
llvm-svn: 312716
Summary:
Tests have to live in the test-suite, and so will come in a separate
patch.
Fixes PR34360.
Reviewers: tra
Subscribers: llvm-commits, sanjoy
Differential Revision: https://reviews.llvm.org/D37539
llvm-svn: 312681
The implementation is in AnalysisDeclContext.cpp and the class is called
AnalysisDeclContext.
Making those match up has numerous benefits, including:
- Easier jump from header to/from implementation.
- Easily identify filename from class.
Differential Revision: https://reviews.llvm.org/D37500
llvm-svn: 312671
The new bare metal support only supports the single thread model. This causes
the builtin atomic functions (e.g.: __atomic_fetch_add) to not generate
thread-safe assembly for these operations, which breaks our firmware. We target
bare metal, and need to atomically modify variables in our interrupt routines,
and task threads.
Internally, the -mthread-model flag determines whether to lower or expand
atomic operations (see D4984).
This change removes the overridden thread model methods, and instead relies on
the base ToolChain class to validate the thread model (which already includes
logic to validate single thread model support). If the single thread model is
required, the -mthread-model flag will have to be provided.
As a workaround "-mthread-model posix" could be provided, but it only works due
to a bug in the validation of the -mthread-model flag (separate patch coming to
fix this).
https://reviews.llvm.org/D37493
Patch by: Ian Tessier!
llvm-svn: 312651
__kmpc_for_static_fini().
Added special flags for calls of __kmpc_for_static_fini(), like previous
ly for __kmpc_for_static_init(). Added flag OMP_IDENT_WORK_DISTRIBUTE
for distribute cnstruct, OMP_IDENT_WORK_SECTIONS for sections-based
constructs and OMP_IDENT_WORK_LOOP for loop-based constructs in
location flags.
llvm-svn: 312642
move constructor.
Previously user-defined reduction initializer was considered as an
assignment expression, not as initializer. Fixed this by treating the
initializer expression as an initializer.
llvm-svn: 312638
Summary:
This adds an option "-gen-clang-data-collectors" to the Clang TableGen
that is used to generate StmtDataCollectors.inc.
Reviewers: arphaman, teemperor!
Subscribers: mgorny, cfe-commits
Differential Revision: https://reviews.llvm.org/D37383
llvm-svn: 312634
Summary:
This affects overloaded operators, which are represented by a
CXXOperatorCallExpr whose first child is always a DeclRefExpr referring to the
operator. For infix, postfix and call operators we want the first argument
to be traversed before the operator.
Reviewers: arphaman
Subscribers: klimek, cfe-commits
Differential Revision: https://reviews.llvm.org/D37200
llvm-svn: 312633
Summary:
We need to specialize this because RecursiveASTVisitor visits template
template parameters after the templated declaration, unlike the order in
which they appear in the source code.
Reviewers: arphaman
Reviewed By: arphaman
Subscribers: klimek, cfe-commits
Differential Revision: https://reviews.llvm.org/D36998
llvm-svn: 312631
Summary:
As the attributed statements are considered simple statements no
stoppoint was generated before emitting attributed do/while/for/range-
statement. This lead to faulty debug locations.
Reviewers: echristo, aaron.ballman, dblaikie
Reviewed By: dblaikie
Subscribers: bjope, aprantl, cfe-commits
Differential Revision: https://reviews.llvm.org/D37428
llvm-svn: 312623
This code has been `#if 0`'ed out for a very long time. After speaking
with Duncan, opt to remove it even if it is something which should be
fixed. If the underlying issue is still valid, this can be restored
with proper explanation and tests.
llvm-svn: 312618
This patch changes the default behavior of `diagtool tree` to only
display warning flags and not the internal warnings flags. The latter is
an implementation detail of the compiler and usually not what the users
wants.
Furthermore, flags that are enabled by default are now also printed in
green. Originally, this was only the case for the diagnostic names.
Differential revision: https://reviews.llvm.org/D37390
llvm-svn: 312546
Correct implementation: Apparently I managed in r311683 to submit the wrong
version of the patch for this, so I'm correcting it now.
Differential Revision: https://reviews.llvm.org/D37079
llvm-svn: 312542
Summary:
Currently clang segfaults when invoked with `clang --autocomplete=`.
This patch adds the necessary boundary checks and some tests for corner cases like this.
Reviewers: yamaguchi
Reviewed By: yamaguchi
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D37465
llvm-svn: 312533
In current OpenCL implementation some options are set in OpenCL RT/Driver, which causes discrepancy between online and offline paths.
Implement infrastructure to move options from OpenCL RT/Driver to AMDGPUToolChain using overloaded TranslateArgs() method.
Create map for default options values, as Options.td doesn't support default values (in contrast with OPTIONS.def).
Add two driver options: -On and -mNN (like -O3, -m64).
Some minor formatting changes to follow the clang-format style.
Differential Revision: https://reviews.llvm.org/D37386
llvm-svn: 312524
Based off the Intel Intrinsics guide, we should expect a void const* argument.
Prevents 'passing 'const void *' to parameter of type 'void *' discards qualifiers' warnings.
Differential Revision: https://reviews.llvm.org/D37449
llvm-svn: 312523
By exposing the constant initializer, the optimizer can fold many
of these constructs.
This is a recommit of r311857 that was reverted in r311898 because
an assert was hit when building Chromium.
We have to take into account that the GlobalVariable may be first
created with a different type than the initializer. This can
happen for example when the variable is a struct with tail padding
while the initializer does not have padding. In such case, the
variable needs to be destroyed an replaced with a new one with the
type of the initializer.
Differential Revision: https://reviews.llvm.org/D34992
llvm-svn: 312512
Sometimes we create the ASTContext and thus the TranslationUnitDecl before we know the LangOptions. This should fix the asan buildbot failures after r312467.
llvm-svn: 312506
Summary:
So far we used a value of 10 which was useful for testing but produces many false-positives in real programs. The usual suspicious clones we find seem to be at around a complexity value of 70 and for normal clone-reporting everything above 50 seems to be a valid normal clone for users, so let's just go with 50 for now and set this as the new default value.
This patch also explicitly sets the complexity value for the regression tests as they serve more of a regression testing/debugging purpose and shouldn't really be reported by default in real programs. I'll add more tests that reflect actual found bugs that then need to pass with the default setting in the future.
Reviewers: NoQ
Subscribers: cfe-commits, javed.absar, xazax.hun, v.g.vassilev
Differential Revision: https://reviews.llvm.org/D34178
llvm-svn: 312468
This follows the scheme agreed with Nathan Sidwell, which can be found here:
https://gcc.gnu.org/wiki/cxx-modules?action=AttachFile
This will be proposed to the itanium-cxx-abi list once we have some experience
with how well it works; the ABI for this TS should be considered unstable until
it is part of the Itanium C++ ABI.
llvm-svn: 312467
Because it is common to treat vector types as an array of their elements, or
even some other type that's not the element type, and thus index into them, we
can't use struct-path TBAA for these accesses. Even though we already treat all
vector types as equivalent to 'char', we were using field-offset information
for them with TBAA, and this renders undefined the intra-value indexing we
intend to allow. Note that, although 'char' is universally aliasing, with path
TBAA, we can still differentiate between access to s.a and s.b in
struct { char a, b; } s;. We can't use this capability as-is for vector types.
Fixes PR33967.
llvm-svn: 312447
Not all targets support vararg (e.g. amdgpu). Instead of using vararg in the emitted functions for enqueue_kernel,
this patch creates a temporary array of size_t, stores the size arguments in the temporary array
and passes it to the emitted functions for enqueue_kernel.
Differential Revision: https://reviews.llvm.org/D36678
llvm-svn: 312441
Summary:
This patch contains performance improvements for the `MinComplexityConstraint`. It reduces the constraint time when running on the SQLite codebase by around 43% (from 0.085s down to 0.049s).
The patch is essentially doing two things:
* It introduces a possibility for the complexity value to early exit when reaching the limit we were checking for. This means that once we noticed that the current clone is larger than the limit the user has set, we instantly exit and no longer traverse the tree or do further expensive lookups in the macro stack.
* It also removes half of the macro stack lookups we do so far. Previously we always checked the start and the end location of a Stmt for macros, which was only a middle way between checking all locations of the Stmt and just checking one location. In practice I rarely found cases where it really matters if we check start/end or just the start of a statement as code with lots of macros that somehow just produce half a statement are very rare.
Reviewers: NoQ
Subscribers: cfe-commits, xazax.hun, v.g.vassilev
Differential Revision: https://reviews.llvm.org/D34361
llvm-svn: 312440
Extract the target specific option application. This is a huge switch
which was inlined into the `ConstructJob` option which adds a large
amount of code to the already large function. Extract it to simply
reduce the line count. NFC
llvm-svn: 312436
Out-of-line the logic for selecting the debug information handling.
This is still split across the new function and partially inline in the
job construction. This is needed since the split portion attempts to
record the "-cc1" arguments. This needs to be the very last item to
ensure that all the flags are recorded. NFC.
llvm-svn: 312435
Extract the handling of the `-fbuiltin` family of flags to the driver.
This centralises the handling of those options, keeping the long
standing `#if 0`'ed block of code. This requires some additional code
archaeology to determine if we need to enable this functionality.
llvm-svn: 312392
Extract the logic for the floating point handling into its own function.
None of this information is needed for calculating the remainder of the
arguments to the frontend. NFC
llvm-svn: 312385
"target" implementation
A small set of refactors that'll make it easier for me to implement 'target'
support.
First, extract the CPUSupports functionality into its own function.
THis has the advantage of not wasting time in this builtin to deal with
arguments.
Second, pulls both CPUSupports and CPUIs implementation into a member-function,
so that it can be called from the resolver generation that I'm working on.
Third, creates an overload that takes simply the feature/cpu name (rather than
extracting it from a callexpr), since that info isn't available later.
Note that despite how the 'diff' looks, the EmitX86CPUSupports function simply
takes the implementation out of the 'switch'.
llvm-svn: 312355
Extract a function to render the diagnostics options to the clang
frontend. This continues the simplification of the clang cc1 command
line invocation generation. NFC
llvm-svn: 312351
Extract the ObjC option rendering for the frontend. This localises the
option translation. It augments the existing `AddRuntimeObjCOptions`
which handles the runtime/ABI versioning flags only. This new function
handles the non-runtime selecting flags. This logic was previously
inlined into the `ConstructJob` function.
Minor change to the flag ordering to group the blocks related flags
together.
llvm-svn: 312344
Extract a function to render the options related to modules. This
reduces the cyclomatic complexity of the `ConstructJob` function. NFC.
llvm-svn: 312330
rule classes
This commit changes the way that the refactoring results are produced. Instead
of using different `RefactoringActionRule` subclasses for each result type,
Clang now use a single `RefactoringResultConsumer`. This was suggested by
Manuel in https://reviews.llvm.org/D36075.
Differential Revision: https://reviews.llvm.org/D37291
llvm-svn: 312316
This fixes cases where dynamic classes produced RTTI data with
external linkage, producing linker errors about duplicate symbols.
This touches code close to what was changed in SVN r244266, but
this change doesn't break the tests added in that revision.
The previous version had missed to update CodeGenCXX/virt-dtor-key.cpp,
which had a behaviour change only when running the testsuite on windows.
Differential revision: https://reviews.llvm.org/D37327
llvm-svn: 312306
step>1.
If the loop is a loot with random access iterators and the iteration
construct is represented it += n, then the compiler crashed because of
reusing of the same MaterializedTemporaryExpr around N. Patch fixes it
by using the expression as written, without any special kind of
wrappings.
llvm-svn: 312292
Summary:
This allows `linkageSpecDecl` to be used from `clang-query`.
See also D31869 which similary adds `isStaticStorageClass`.
Reviewers: aaron.ballman
Reviewed By: aaron.ballman
Subscribers: cfe-commits, klimek
Differential Revision: https://reviews.llvm.org/D37346
llvm-svn: 312283
Summary:
Currently object format is taken from the default target triple. For toolchains with a non-COFF default target this may result in an object format inappropriate for pc-windows and lead to compilation issues.
For example, the default triple `aarch64-linux-elf` may produce something like `aarch64-pc-windows-msvc19.0.24215-elf` in CL mode. Clang creates `MicrosoftARM64TargetInfo` for such triple with data layout `e-m:w-p:64:64-i32:32-i64:64-i128:128-n32:64-S128`. On the other hand, the AArch64 backend in `computeDataLayout` detects a non-COFF target and selects `e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128` as data layout for little endian. Different layouts used by clang and the backend cause an error:
```
error: backend data layout 'e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128'
does not match expected target description 'e-m:w-p:64:64-i32:32-i64:64-i128:128-n32:64-S128'
```
This can be observed on the clang's Driver/cl-pch.c test with AArch64 as a default target.
This patch enforces COFF in CL mode.
Reviewers: hans
Reviewed By: hans
Subscribers: cfe-commits, aemerson, asl, kristof.beyls
Differential Revision: https://reviews.llvm.org/D37336
llvm-svn: 312275
EXPENSIVE_CHECKS enables libstdc++'s library consistency checks, which
includes checking the container passed to std::priority_queue for its
well-formedness. This makes the clang-format memoization too expensive,
so disable it.
(it's a necessary feature of libstdc++'s consistency checks that it
ruins the required scalability of C++ standard library features - so
these workarounds are to be expected if a test ever tries to test
scalability in some way, like this test does)
llvm-svn: 312268
Extract the ARC migration tool flag handling into its own function.
This simplifies the flow of the clang frontend command line construction
function. NFC.
llvm-svn: 312244
We are starting to use data flow tracing in kernel.
The interface is not subject to change anymore.
Reviewed in https://reviews.llvm.org/D37303
llvm-svn: 312232
This fixes cases where dynamic classes produced RTTI data with
external linkage, producing linker errors about duplicate symbols.
This touches code close to what was changed in SVN r244266, but
this change doesn't break the tests added in that revision.
Differential revision: https://reviews.llvm.org/D37206
llvm-svn: 312224
Summary:
This patch aims at optimizing the CloneChecker for larger programs. Before this
patch we took around 102 seconds to analyze sqlite3 with a complexity value of
50. After this patch we now take 2.1 seconds to analyze sqlite3.
The biggest performance optimization is that we now put the constraint for group
size before the constraint for the complexity. The group size constraint is much
faster in comparison to the complexity constraint as it only does a simple
integer comparison. The complexity constraint on the other hand actually
traverses each Stmt and even checks the macro stack, so it is obviously not able
to handle larger amounts of incoming clones. The new order filters out all the
single-clone groups that the type II constraint generates in a faster way before
passing the fewer remaining clones to the complexity constraint. This reduced
runtime by around 95%.
The other change is that we also delay the verification part of the type II
clones back in the chain of constraints. This required to split up the
constraint into two parts - a verification and a hash constraint (which is also
making it more similar to the original design of the clone detection algorithm).
The reasoning for this is the same as before: The verification constraint has to
traverse many statements and shouldn't be at the start of the constraint chain.
However, as the type II hashing has to be the first step in our algorithm, we
have no other choice but split this constrain into two different ones. Now our
group size and complexity constrains filter out a chunk of the clones before
they reach the slow verification step, which reduces the runtime by around 8%.
I also kept the full type II constraint around - that now just calls it's two
sub-constraints - in case someone doesn't care about the performance benefits
of doing this.
Reviewers: NoQ
Reviewed By: NoQ
Subscribers: klimek, v.g.vassilev, xazax.hun, cfe-commits
Differential Revision: https://reviews.llvm.org/D34182
llvm-svn: 312222
Extend the -fmodule-file option to support the [<name>=]<file> value format.
If the name is omitted, then the old semantics is preserved (the module file
is loaded whether needed or not). If the name is specified, then the mapping
is treated as just another prebuilt module search mechanism, similar to
-fprebuilt-module-path, and the module file is only loaded if actually used
(e.g., via import). With one exception: this mapping also overrides module
file references embedded in other modules (which can be useful if module files
are moved/renamed as often happens during remote compilation).
This override semantics requires some extra work: we now store the module name
in addition to the file name in the serialized AST representation.
Reviewed By: rsmith
Differential Revision: https://reviews.llvm.org/D35020
llvm-svn: 312220
r312167 made it so that we emit Wdelete-non-virtual-dtor from delete statements
that are in system headers (e.g. std::unique_ptr). That works great on Linux
and macOS, but on Windows there are non-final classes that are defined in
system headers that have virtual methods but non-virtual destructors and yet
get deleted through a base class pointer (e.g. ATL::CAccessToken::CRevert). So
paddle back a bit and don't emit the warning if it's about a class defined in a
system header.
https://reviews.llvm.org/D37324
llvm-svn: 312216
Tests fail on ARM targets due to ABI name between define and void. Added reg ex to skip.
Patch by Glenn Howe (and expanded on by Douglas Yung)!
Differential Revision: https://reviews.llvm.org/D33410
llvm-svn: 312181
Makes the warning useful again in a std::unique_ptr world, PR28460.
Also make the warning not fire in unevaluated contexts, since system libraries
(e.g. libc++) do do that. This would've been a good change before we started
emitting this warning in system headers too, but "normal" code seems to be less
template-heavy, so we didn't notice until now.
https://reviews.llvm.org/D37235
llvm-svn: 312167
This patch implements the broadcastf32x2/broadcasti32x2 intrinsics using __builtin_shufflevector.
Differential Revision: https://reviews.llvm.org/D37287
llvm-svn: 312135
form of PseudoObjectExpr
The new commit adjusts unittest test code compilation options so that the
Objective-C code in the unittest can be parsed on non-macOS platforms.
Original message:
The AST selection finder now constructs a selection tree that contains only the
syntactic form of PseudoObjectExpr. This form of selection tree is more
meaningful when doing downstream analysis as we're interested in the syntactic
features of the AST and the correct lexical parent relation.
llvm-svn: 312132
of PseudoObjectExpr
The AST selection finder now constructs a selection tree that contains only the
syntactic form of PseudoObjectExpr. This form of selection tree is more
meaningful when doing downstream analysis as we're interested in the syntactic
features of the AST and the correct lexical parent relation.
llvm-svn: 312127
Summary:
This is an implementation for [bug 17362](https://bugs.llvm.org/attachment.cgi?bugid=17362) which adds support for indenting preprocessor statements inside if/ifdef/endif. This takes previous work from fmauch (https://github.com/fmauch/clang/tree/preprocessor_indent) and makes it into a full feature.
The context of this patch is that I'm a VMware intern, and I implemented this because VMware needs the feature. As such, some decisions were made based on what VMware wants, and I would appreciate suggestions on expanding this if necessary to use-cases other people may want.
This adds a new enum config option, `IndentPPDirectives`. Values are:
* `PPDIS_None` (in config: `None`):
```
#if FOO
#if BAR
#include <foo>
#endif
#endif
```
* `PPDIS_AfterHash` (in config: `AfterHash`):
```
#if FOO
# if BAR
# include <foo>
# endif
#endif
```
This is meant to work whether spaces or tabs are used for indentation. Preprocessor indentation is independent of indentation for non-preprocessor lines.
Preprocessor indentation also attempts to ignore include guards with the checks:
1. Include guards cover the entire file
2. Include guards don't have `#else`
3. Include guards begin with
```
#ifndef <var>
#define <var>
```
This patch allows `UnwrappedLineParser::PPBranchLevel` to be decremented to -1 (the initial value is -1) so the variable can be used for indent tracking.
Defects:
* This patch does not handle the case where there's code between the `#ifndef` and `#define` but all other conditions hold. This is because when the #define line is parsed, `UnwrappedLineParser::Lines` doesn't hold the previous code line yet, so we can't detect it. This is out of the scope of this patch.
* This patch does not handle cases where legitimate lines may be outside an include guard. Examples are `#pragma once` and `#pragma GCC diagnostic`, or anything else that does not change the meaning of the file if it's included multiple times.
* This does not detect when there is a single non-preprocessor line in front of an include-guard-like structure where other conditions hold because `ScopedLineState` hides the line.
* Preprocessor indentation throws off `TokenAnnotator::setCommentLineLevels` so the indentation of comments immediately before indented preprocessor lines is toggled on each run. Fixing this issue appears to be a major change and too much complexity for this patch.
Contributed by @euhlmann!
Reviewers: djasper, klimek, krasimir
Reviewed By: djasper, krasimir
Subscribers: krasimir, mzeren-vmw, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D35955
llvm-svn: 312125
Extract the analyzer flag handling into its own function to reduce the
overall complexity of the construction of the clang compiler arguments.
NFC.
llvm-svn: 312124
Extend the -fmodule-file option to support the [<name>=]<file> value format.
If the name is omitted, then the old semantics is preserved (the module file
is loaded whether needed or not). If the name is specified, then the mapping
is treated as just another prebuilt module search mechanism, similar to
-fprebuilt-module-path, and the module file is only loaded if actually used
(e.g., via import). With one exception: this mapping also overrides module
file references embedded in other modules (which can be useful if module files
are moved/renamed as often happens during remote compilation).
This override semantics requires some extra work: we now store the module name
in addition to the file name in the serialized AST representation.
Reviewed By: rsmith
Differential Revision: https://reviews.llvm.org/D35020
llvm-svn: 312105
Out-of-line the SSP argument handling for the sake of readability. Pass
along some state information to avoid re-computing the command line
flags.
llvm-svn: 312084
Extract the argument forwarding for OpenCL arguments. Make this more
data driven as we are just repeating the argument name and spelling.
This costs a slight bit more memory due to the string duplication, but
makes it easier to follow. It should be possible to forward the
internal string representation from the TableGen data to avoid this.
But, this makes the code simpler to follow for now.
llvm-svn: 312083
Summary:
Recent changes canonicalized clang_rt library names to refer to
"i386" on all x86 targets. Android historically uses i686.
This change adds a special case to keep i686 in all clang_rt
libraries when targeting Android.
Reviewers: hans, mgorny, beanz
Subscribers: srhines, cfe-commits, llvm-commits
Differential Revision: https://reviews.llvm.org/D37278
llvm-svn: 312048
Summary:
An implementation of ubsan runtime library suitable for use in production.
Minimal attack surface.
* No stack traces.
* Definitely no C++ demangling.
* No UBSAN_OPTIONS=log_file=/path (very suid-unfriendly). And no UBSAN_OPTIONS in general.
* as simple as possible
Minimal CPU and RAM overhead.
* Source locations unnecessary in the presence of (split) debug info.
* Values and types (as in A+B overflows T) can be reconstructed from register/stack dumps, once you know what type of error you are looking at.
* above two items save 3% binary size.
When UBSan is used with -ftrap-function=abort, sometimes it is hard to reason about failures. This library replaces abort with a slightly more informative message without much extra overhead. Since ubsan interface in not stable, this code must reside in compiler-rt.
Reviewers: pcc, kcc
Subscribers: srhines, mgorny, aprantl, krytarowski, llvm-commits
Differential Revision: https://reviews.llvm.org/D36810
llvm-svn: 312029
Capturing of the global variables occurs only in target regions. Patch
fixes it and allows capturing of globals in all target executable
directives.
llvm-svn: 312024
Summary:
We wrote many codes in HandleImediateArgs, so I've refactored it into
handleAutocompletions.
Reviewers: v.g.vassilev, teemperor
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D37249
llvm-svn: 312018
Overriding a method from a virtual base with a covariant return type
consumes a slot from the vftable in the virtual base. This can make it
impossible to implement certain diamond inheritance hierarchies, but we
have to follow along for compatibility in the simple cases.
This patch only affects our vtable dumper and member pointer function
mangling, since all other callers of getMethodVFTableLocation seem to
recompute VBTableIndex instead of using the one in the method location.
Patch by David Majnemer
llvm-svn: 312017
If a TS module name has more than one component (e.g., foo.bar) then we
erroneously activated the submodule semantics when encountering a module
declaration in the module implementation unit (e.g., 'module foo.bar;').
Reviewed By: rsmith
Differential Revision: https://reviews.llvm.org/D35678
llvm-svn: 312007
Summary:
This patch detects the leading '<' in likely xml files and stops formatting in
that case. A recent use of a Qt xml file with a .ts extension triggered this:
http://doc.qt.io/qt-4.8/linguist-ts-file-format.html
Reviewers: djasper
Reviewed By: djasper
Subscribers: sammccall, cfe-commits, klimek
Differential Revision: https://reviews.llvm.org/D37136
llvm-svn: 311999
Summary:
Bug: https://bugs.llvm.org/show_bug.cgi?id=34016 - **Typedef enum part**
**Problem:**
Clang format does not allow the flag **BraceWrapping.AfterEnum** control the case when our **enum** is preceded by **typedef** keyword (what is common in C language).
**Patch description:**
Added case to the **"AfterEnum"** flag when our enum does not start a line - is preceded by **typedef** keyword.
**After fix:**
**CONFIG:**
```
BreakBeforeBraces: Custom
BraceWrapping: {
AfterClass: true, AfterControlStatement: true, AfterEnum: true, AfterFunction: true, AfterNamespace: false, AfterStruct: true, AfterUnion: true, BeforeCatch: true, BeforeElse: true
}
```
**BEFORE:**
```
typedef enum
{
a,
b,
c
} SomeEnum;
```
**AFTER:**
```
typedef enum
{
a,
b,
c
} SomeEnum;
```
Contributed by @PriMee!
Reviewers: krasimir, djasper
Reviewed By: djasper
Subscribers: cfe-commits, klimek
Differential Revision: https://reviews.llvm.org/D37143
llvm-svn: 311998
Summary:
This adds a test that checks if the using declaration in classes still works as intended with modules.
The motivation for this is that we tried to add a shortcut to `removeDecl` that would skip the removal of declarations from the lookup table if they are hidden. This optimization passed the clang test suite but actually broke the using declaration in combination with -fmodules-local-submodule-visibility. In this mode we hide all decls from other modules such as by chance the parent method, in which case don't remove the parent method from the lookup table and get ambiguous lookup errors. After this patch we now correctly see if this behavior is broken by a patch like this in the test suite.
Reviewers: v.g.vassilev
Reviewed By: v.g.vassilev
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D37180
llvm-svn: 311991
Looks like this one was forgotten for clang_parseTranslationUnit*, as
LIBCLANG_NOTHREADS is checked for/in:
clang_saveTranslationUnit()
clang_reparseTranslationUnit()
clang_codeCompleteAt()
clang_indexTranslationUnit()
clang_indexSourceFile()
Patch by Nikolai Kosjar!
Differential Revision: https://reviews.llvm.org/D36821
llvm-svn: 311990
Summary:
Previously, clang-format would try to wrap template string substitutions
by indenting relative to the openening `${`. This helped with
indenting structured strings, such as strings containing HTML, as the
substitutions would be aligned according to the structure of the string.
However it turns out that the overwhelming majority of template string +
substitution usages are for substitutions into non-structured strings,
e.g. URLs or just plain messages. For these situations, clang-format
would often produce very ugly indents, in particular for strings
containing no line breaks:
return `<a href='http://google3/${file}?l=${row}'>${file}</a>(${
row
},${
col
}): `;
This change makes clang-format indent template string substitutions as
if they were string concatenation operations. It wraps +4 on overlong
lines and keeps all operands on the same line:
return `<a href='http://google3/${file}?l=${row}'>${file}</a>(${
row},${col}): `;
While this breaks some lexical continuity between the `${` and `row}`
here, the overall effects are still a huge improvement, and users can
still manually break the string using `+` if desired.
Reviewers: djasper
Subscribers: klimek, cfe-commits
Differential Revision: https://reviews.llvm.org/D37142
llvm-svn: 311988
Information about clang executable name components, such as target and
driver mode, was passes in std::pair. With this change it is passed in
a special structure. It improves readability and makes access to this
information more convenient.
NFC.
Differential Revision: https://reviews.llvm.org/D36057
llvm-svn: 311981
Summary:
Add support for autocompleting values of -std= by including
LangStandards.def. This patch relies on D36782, and is using two-stage
code generation.
Reviewers: v.g.vassilev, teemperor, ruiu
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D36820
llvm-svn: 311971
The standard is not clear on how these are supposed to be handled, so we
conservatively treat as non-constant any cases whose value is unknown or whose
evaluation might result in undefined behavior.
llvm-svn: 311970
This reverts commit 7c46b80c022e18d43c1fdafb117b0c409c5a6d1e.
r311552 broke lld buildbot because I've changed OptionInfos type from
ArrayRef to vector. However the bug is fixed, so I'll commit this again.
llvm-svn: 311958
Replace "long" with __UINTPTR_TYPE__
to make the test added in rL311935 Windows-friendly.
Caught by the buildbot llvm-clang-x86_64-expensive-checks-win.
llvm-svn: 311947
This diff fixes modeling of arithmetic
expressions where pointers are treated as integers
(i.e. via C-style / reinterpret casts).
For now we return UnknownVal unless the operation is a comparison.
Test plan: make check-all
Differential revision: https://reviews.llvm.org/D37120
llvm-svn: 311935
Use llvm::Triple::getArchTypeName() when looking for compiler-rt
libraries, rather than the exact arch string from the triple. This is
more correct as it matches the values used when building compiler-rt
(builtin-config-ix.cmake) which are the subset of the values allowed
in triples.
For example, this fixes an issue when the compiler set for
i686-pc-linux-gnu triple would not find an i386 compiler-rt library,
while this is the exact arch that is detected by compiler-rt. The same
applies to any other i?86 variant allowed by LLVM.
This also makes the special case for MSVC unnecessary, since now i386
will be used reliably for all 32-bit x86 variants.
Differential Revision: https://reviews.llvm.org/D26796
llvm-svn: 311923
Diagnostic Categories are fairly annoying, and are only enforced
by a runtime-debug-only assert. This puts in a touch more work
to get this all done at compile-time with static asserts
Differential Revision: https://reviews.llvm.org/D37122
llvm-svn: 311905
It caused PR759744.
> Emit static constexpr member as available_externally definition
>
> By exposing the constant initializer, the optimizer can fold many
> of these constructs.
>
> Differential Revision: https://reviews.llvm.org/D34992
llvm-svn: 311898
This patch implements the initial support for refactoring action rules. The
first rule that's supported is a "source change" rule that returns a set of
atomic changes. This patch is based on the ideas presented in my RFC:
http://lists.llvm.org/pipermail/cfe-dev/2017-July/054831.html
The following pieces from the RFC are added by this patch:
- `createRefactoringRule` (known as `apply` in the RFC)
- `requiredSelection` refactoring action rule requirement.
- `selection::SourceSelectionRange` selection constraint.
Differential Revision: https://reviews.llvm.org/D36075
llvm-svn: 311884
This way the unrolling can be restricted for loops which will take at most a
given number of steps. It is defined as 128 in this patch and it seems to have
a good number for that purpose.
Differential Revision: https://reviews.llvm.org/D37181
llvm-svn: 311883
Added check if the execution of the last step of the given unrolled loop has
generated more branches. If yes, than treat it as a normal (non-unrolled) loop
in the remaining part of the analysis.
Differential Revision: https://reviews.llvm.org/D36962
llvm-svn: 311881
1. The LoopUnrolling feature needs the LoopExit included in the CFG so added this
dependency via the config options
2. The LoopExit element can be encountered even if we haven't encountered the
block of the corresponding LoopStmt. So the asserts were not right.
3. If we are caching out the Node then we get a nullptr from generateNode which
case was not handled.
Differential Revision: https://reviews.llvm.org/D37103
llvm-svn: 311880
This adds builtin_cpu_init which will emit a call to cpu_indicator_init in libgcc or compiler-rt.
This is needed to support builtin_cpu_supports/builtin_cpu_is in an ifunc resolver.
Differential Revision: https://reviews.llvm.org/D36336
llvm-svn: 311874
By exposing the constant initializer, the optimizer can fold many
of these constructs.
Differential Revision: https://reviews.llvm.org/D34992
llvm-svn: 311857
Prior to this patch clang would not error here:
template <class T> struct B;
template <class T> struct A {
void foo();
void foo2();
void test1() {
B<T>::foo(); // OK, foo is declared in A<int> - matches type of 'this'.
B<T>::foo2(); // This should be an error!
// foo2 is found in B<int>, 'base unrelated' to 'this'.
}
};
template <class T> struct B : A<T> {
using A<T>::foo2;
};
llvm-svn: 311851
This allows multi-module / incremental compilation environments to have unique
initializer symbols.
Patch by Axel Naumann with minor modifications by me!
llvm-svn: 311844
When isIncrementalProcessingEnabled is on we might want to produce multiple
llvm::Modules. This patch allows the clients to start a new llvm::Module,
allowing CodeGen to continue working after a HandleEndOfTranslationUnit call.
This should give the necessary facilities to write a unittest for D34059.
As discussed in the review this is meant to give us a way to proceed forward
in our efforts to upstream our interpreter-related patches. The design of this
will likely change soon.
llvm-svn: 311843
Prior to this patch, clang would do the wrong thing here (see inline comments for pre-patch behavior):
struct A {
void bar(int) { }
static void bar(double) { }
void g(int*);
static void g(char *);
};
struct B {
void f() {
A::bar(3); // selects (double) ??!!
A::g((int*)0); // Instead of no object argument, states conversion error?!!
}
};
The fix is as follows: When we detect that what appears to be an implicit member function call (A::bar) is actually a call to a member of a class (A) unrelated to the type (B) that contains the member function (B::f) from which the call is being made, don't treat it (A::bar) as an Implicit Member Call Expression.
P.S. I wonder if there is an existing bug report related to this? (Surprisingly, a cursory search did not find one).
llvm-svn: 311839
Use llvm::Triple::getArchTypeName() when looking for compiler-rt
libraries, rather than the exact arch string from the triple. This is
more correct as it matches the values used when building compiler-rt
(builtin-config-ix.cmake) which are the subset of the values allowed
in triples.
For example, this fixes an issue when the compiler set for
i686-pc-linux-gnu triple would not find an i386 compiler-rt library,
while this is the exact arch that is detected by compiler-rt. The same
applies to any other i?86 variant allowed by LLVM.
This also makes the special case for MSVC unnecessary, since now i386
will be used reliably for all 32-bit x86 variants.
Differential Revision: https://reviews.llvm.org/D26796
llvm-svn: 311836
This patch adds a flag -fclang-abi-compat that can be used to request that
Clang attempts to be ABI-compatible with some older version of itself.
This is provided on a best-effort basis; right now, this can be used to undo
the ABI change in r310401, reverting Clang to its prior C++ ABI for pass/return
by value of class types affected by that change, and to undo the ABI change in
r262688, reverting Clang to using integer registers rather than SSE registers
for passing <1 x long long> vectors. The intent is that we will maintain this
backwards compatibility path as we make ABI-breaking fixes in future.
The reversion to the old behavior for r310401 is also applied to the PS4 target
since that change is not part of its platform ABI (which is essentially to do
whatever Clang 3.2 did).
llvm-svn: 311823
alignments
Indent should be compared before nesting level to determine if a token
is on the same scope as the one we align with. Because it was inverted,
clang-format sometimes tried to align tokens with tokens from outer
scopes, causing the assert(Shift >= 0) to fire.
This fixes bug #33507. Patch by Beren Minor, thank you!
llvm-svn: 311792
In ASTUnit::LoadFromASTFile, the context object is set up using
default-constructed LangOptions (which only later get populated). As the
language options are used in the constructor of PrintingPolicy, this
needs to be updated explicitly after the language options are available.
Patch by Johann Klähn!
Differential Revision: https://reviews.llvm.org/D35271
llvm-svn: 311787
-Wobjc-messaging-id is a new, non-default warning that warns about
message sends to unqualified id in Objective-C. This warning is useful
for projects that would like to avoid any potential future compiler
errors/warnings, as the system frameworks might add a method with the same
selector which could make the message send to id ambiguous.
rdar://33303354
llvm-svn: 311779
Clang's DiagnosticError is an llvm::Error payload that stores a partial
diagnostic and its location. I'll be using it in the refactoring engine.
Differential Revision: https://reviews.llvm.org/D36969
llvm-svn: 311778
expressions
C++ allows us to reference static variables through member expressions. Prior to
this commit, non-integer static variables that were referenced using a member
expression were always emitted using lvalue loads. The old behaviour introduced
an inconsistency between regular uses of static variables and member expressions
uses. For example, the following program compiled and linked successfully:
struct Foo {
constexpr static const char *name = "foo";
};
int main() {
return Foo::name[0] == 'f';
}
but this program failed to link because "Foo::name" wasn't found:
struct Foo {
constexpr static const char *name = "foo";
};
int main() {
Foo f;
return f.name[0] == 'f';
}
This commit ensures that constant static variables referenced through member
expressions are emitted in the same way as ordinary static variable references.
rdar://33942261
Differential Revision: https://reviews.llvm.org/D36876
llvm-svn: 311772
Summary:
If await_suspend returns a coroutine_handle, as in the example below:
```
coroutine_handle<> await_suspend(coroutine_handle<> h) {
coro.promise().waiter = h;
return coro;
}
```
suspensionExpression processing will resume the coroutine pointed at by that handle.
Related LLVM change rL311751 makes resume calls of this kind `musttail` at any optimization level.
This enables unlimited symmetric control transfer from coroutine to coroutine without blowing up the stack.
Reviewers: GorNishanov
Reviewed By: GorNishanov
Subscribers: rsmith, EricWF, cfe-commits
Differential Revision: https://reviews.llvm.org/D37131
llvm-svn: 311762
Summary: With accurate sample profile, we can do more aggressive size optimization. For some size-critical application, this can reduce the text size by 20%
Reviewers: davidxl, rsmith
Reviewed By: davidxl, rsmith
Subscribers: mehdi_amini, eraman, sanjoy, cfe-commits
Differential Revision: https://reviews.llvm.org/D37091
llvm-svn: 311707
Do not sanitize the 'this' pointer of a member call operator for a lambda with
no capture-default, since that call operator can legitimately be called with a
null this pointer from the static invoker function. Any actual call with a null
this pointer should still be caught in the caller (if it is being sanitized).
This reinstates r311589 (reverted in r311680) with the above fix.
llvm-svn: 311695
Discovered due to a goofy git setup, the test system-headerline-directive.c
(and a few others) failed because the token-consumption will consume only the
'\r' in CRLF, making the preprocessor's printed value give the wrong line number
when returning from an include. For example:
(line 1):#include <noline.h>\r\n
The "file exit" code causes the printer to try to print the 'returned to the
main file' line. It looks up what the current line number is. However, since the
current 'token' is the '\n' (since only the \r was consumed), it will give the
line number as '1", not '2'. This results in a few failed tests, but more
importantly, results in error messages being incorrect when compiling a
previously preprocessed file.
Differential Revision: https://reviews.llvm.org/D37079
llvm-svn: 311683
Summary:
This recommits https://reviews.llvm.org/D36956 with an update to the added test
case to not use raw string literals, since this makes gcc unhappy.
Reviewers: djasper
Reviewed By: djasper
Subscribers: cfe-commits, klimek
Differential Revision: https://reviews.llvm.org/D37109
llvm-svn: 311672
In patch r205628 using abs.[ds] instruction is forced, as they should behave
in accordance with flags Has2008 and ABS2008. Unfortunately for revisions
prior mips32r6 and mips64r6, abs.[ds] is not generating correct result when
working with NaNs. To generate a sequence which always produce a correct
result but also to allow user more control on how his code is compiled,
option -mabs is added where user can choose legacy or 2008.
By default legacy mode is used on revisions prior R6. Mips32r6 and mips64r6
use abs2008 mode by default.
Patch by Aleksandar Beserminji
Differential Revision: https://reviews.llvm.org/D35982
llvm-svn: 311669
This commit adds the base AST source selection component to the refactoring
library. AST selection is represented using a tree of SelectedASTNode values.
Each selected node gets its own selection kind, which can actually be None even
in the middle of tree (e.g. statement in a macro whose child is in a macro
argument). The initial version constructs a "raw" selection tree, without
applying filters and canonicalisation operations to the nodes.
Differential Revision: https://reviews.llvm.org/D35012
llvm-svn: 311655
This patch is intended to enable the use of basic double letter constraints used in GCC extended inline asm {Yi Y2 Yz Y0 Ym Yt}.
Supersedes D35205
llvm counterpart: D36369
Differential Revision: https://reviews.llvm.org/D36371
llvm-svn: 311643
This reverts commit r311457. It reveals some dormant bugs in comment
reflowing, like breaking a single line jsdoc type annotation before a
parameter into multiple lines.
llvm-svn: 311641
`id` needs to be handled specially since it is a `TypedefType` which is
sugar for an `ObjCObjectPointerType` whose pointee is an
`ObjCObjectType` with base `BuiltinType::ObjCIdType` and no protocols
and the first level of pointer gets it own type implementation. `Class`
is similar with the `ObjCClassType` as the base instead.
The qualifiers on the base type of the `ObjCObjectType` need to be
dropped because the innermost `mangleType` will handle the qualifiers
itself.
`id` is desugared to `struct objc_object *` which should be encoded as
`PAUobjc_object@@`. `Class` is desugared to `struct objc_class *` which
should be encoded as `PAUobjc_class@@`.
We were previously applying an extra modifier `A` which will be handled
during the recursive call.
This now properly decorates interface types as well as `Class` and `id`.
This corrects the interactions between C++ and ObjC++ for the type
specifier decoration.
llvm-svn: 311617
Summary:
Most DIExpressions are empty or very simple. When they are complex, they
tend to be unique, so checking them inline is reasonable.
This also avoids the need for CodeGen passes to append to the
llvm.dbg.mir named md node.
See also PR22780, for making DIExpression not be an MDNode.
Reviewers: aprantl, dexonsmith, dblaikie
Subscribers: qcolombet, javed.absar, eraman, hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D37075
llvm-svn: 311594
GCC will interpret `__attribute__((__aligned__))` as 8-byte alignment on
ARM, but clang will not. Explicitly specify the alignment. This
mirrors the declaration in libunwind.
llvm-svn: 311576
Summary:
This adds shortcuts j and k to jump between changes.
It is especially useful in diffs with few changes.
Reviewers: arphaman
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D36685
llvm-svn: 311570
Summary:
This moves the data collection macro calls for Stmt nodes
to lib/AST/StmtDataCollectors.inc
Users can subclass ConstStmtVisitor and include StmtDataCollectors.inc
to define visitor methods for each Stmt subclass. This makes it also
possible to customize the visit methods as exemplified in
lib/Analysis/CloneDetection.cpp.
Move helper methods for data collection to a new module,
AST/DataCollection.
Add data collection for DeclRefExpr, MemberExpr and some literals.
Reviewers: arphaman, teemperor!
Subscribers: mgorny, xazax.hun, cfe-commits
Differential Revision: https://reviews.llvm.org/D36664
llvm-svn: 311569
The C++ ABI requires that the exception object (which under AEABI is the
`_Unwind_Control_Block`) is double-word aligned. The attribute was
applied to the `_Unwind_Exception` type, but not the
`_Unwind_Control_Block`. This should fix the libunwind test for the
alignment of the exception type.
llvm-svn: 311563
This makes -Wunreachable-code work for programs containing SEH (except for
__finally, which is still missing for now).
__try is modeled like try (but simpler since it can only have a single __except
or __finally), __except is fairly similar to catch (but simpler, since it can't
contain declarations). __leave is implemented similarly to break / continue.
Use the existing addTryDispatchBlock infrastructure (which
FindUnreachableCode() in ReachableCode.cpp uses via cfg->try_blocks_begin()) to
mark things in the __except blocks as reachable.
Re-use TryTerminatedBlock. This means we add EH edges from calls to the __try
block, but not from all other statements. While this is incomplete, it matches
LLVM's SEH codegen support. Also, in practice, BuildOpts.AddEHEdges is always
false in practice from what I can tell, so we never even insert the call EH
edges either.
https://reviews.llvm.org/D36914
llvm-svn: 311561
Summary:
This patch makes the splits emitted for the beginning of comment lines during
reformatting absolute. Previously, they were relative to the start of the
non-whitespace content of the line, which messes up further TailOffset
calculations in breakProtrudingToken. This fixes an assertion failure reported
in bug 34236: https://bugs.llvm.org/show_bug.cgi?id=34236.
Reviewers: djasper
Reviewed By: djasper
Subscribers: klimek, cfe-commits
Differential Revision: https://reviews.llvm.org/D36956
llvm-svn: 311559
Summary:
This is a patch for clang autocomplete feature.
It will collect values which -analyzer-checker takes, which is defined in
clang/StaticAnalyzer/Checkers/Checkers.inc, dynamically.
First, from ValuesCode class in Options.td, TableGen will generate C++
code in Options.inc. Options.inc will be included in DriverOptions.cpp, and
calls OptTable's addValues function. addValues function will add second
argument to Option's Values class. Values contains string like "foo,bar,.."
which is handed to Values class
in OptTable.
Reviewers: v.g.vassilev, teemperor, ruiu
Subscribers: hiraditya, cfe-commits
Differential Revision: https://reviews.llvm.org/D36782
llvm-svn: 311552
Summary:
ColumnLimit = 0 means no limit, so comment should always be aligned if requested. This was broken with
https://llvm.org/svn/llvm-project/cfe/trunk@304687
introduced via
https://reviews.llvm.org/D33830
and is included in 5.0.0-rc2. This commit fixes it and adds a unittest for this property.
Should go into clang-5.0 IMHO.
Contributed by @pboettch!
Reviewers: djasper, krasimir
Reviewed By: djasper, krasimir
Subscribers: hans, klimek
Differential Revision: https://reviews.llvm.org/D36967
llvm-svn: 311532
As of now, libFuzzer is located in compiler-rt and is bundled with
Clang's toolchain by default.
Differential Revision: https://reviews.llvm.org/D37037
llvm-svn: 311514
As before, not every platform supports LTO. Make sure the platform
we're targeting is one that supports it (regardless of the *host*
platform).
llvm-svn: 311488
This is the same assertion as in https://reviews.llvm.org/D25206 that is
triggered when RecordLayoutBuilder tries to compute the size of a field
(for capture "typo_boo" in the test case) whose type hasn't been
deduced.
The fix is to add CorrectDelayedTyposInExpr call to the cases when we
aren't disambiguating between an Obj-C message send and a lambda
expression.
rdar://problem/31760839
Reviewers: rsmith, ahatanak
Reviewed By: arphaman
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D36853
llvm-svn: 311480
of class fails to map class static variable.
If the global variable is captured and it has several redeclarations,
sometimes it may lead to a compiler crash. Patch fixes this by working
only with canonical declarations.
llvm-svn: 311479
Summary:
Rename stop-after to stop-diff-after. When building LLVM with
-DLLVM_BUILD_LLVM_DYLIB=ON, stop-after collides with the stop-after
already present in LLVM.
Reviewers: johannes, arphaman
Subscribers: klimek, aheejin, cfe-commits
Differential Revision: https://reviews.llvm.org/D36989
llvm-svn: 311476
Summary:
This patch is an alternative to https://reviews.llvm.org/D36614, by resolving a
non-idempotency issue by breaking non-trailing comments:
Consider formatting the following code with column limit at `V`:
```
V
const /* comment comment */ A = B;
```
The comment is not a trailing comment, breaking before it doesn't bring it under
the column limit. The formatter breaks after it, resulting in:
```
V
const /* comment comment */
A = B;
```
For a next reformat, the formatter considers the comment as a trailing comment,
so it is free to break it further, resulting in:
```
V
const /* comment
comment */
A = B;
```
This patch improves the situation by directly producing the third case.
Reviewers: djasper
Reviewed By: djasper
Subscribers: cfe-commits, klimek
Differential Revision: https://reviews.llvm.org/D37007
llvm-svn: 311457
Summary:
This patch fixes a regression after https://reviews.llvm.org/rL305665,
which updates the structure of the `lines` variable.
Reviewers: djasper
Reviewed By: djasper
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D37011
llvm-svn: 311456
Rename the function getSupportedNanEncoding() to getIEEE754Standard(), since
this function will be used for non-nan related features.
Patch by Aleksandar Beserminji.
Differential Revision: https://reviews.llvm.org/D36824
llvm-svn: 311454
This commit fixes a bug introduced in r307903. The attribute ambiguity checker
that was introduced in r307903 checked all property attributes, which caused
errors for source-compatible properties, like:
@property (nonatomic, readonly) NSObject *prop;
@property (nonatomic, readwrite) NSObject *prop;
because the readwrite property would get implicit 'strong' attribute. The
ambiguity checker should be concerned about explicitly specified attributes
only.
rdar://33748089
llvm-svn: 311443
When we have enabled cache for global completions we did not have
diagnostics for Bar and could not complete Ba as in provided code
example.
template <typename T>
struct Foo { T member; };
template<typename T> using Bar = Foo<T>;
int main() {
Ba
}
Patch by Ivan Donchevskii!
Differential Revision: https://reviews.llvm.org/D35355
llvm-svn: 311442
Summary:
If a node referring to a name is within a class or namespace, do not use
the full qualified name, but strip the namespace prefix.
Reviewers: arphaman, bkramer
Subscribers: klimek
Differential Revision: https://reviews.llvm.org/D36681
llvm-svn: 311433
supplied.
With this change, -fno-exceptions disables unwind tables unless
-funwind-tables is supplied too or the target is x86-64 (x86-64 requires
emitting unwind tables).
rdar://problem/33934446
llvm-svn: 311397
This is a reasonably non-intrusive change, which I've verified
works for both x86 and x64 DevDiv-internal builds.
The idea is to change `bool IsVS2017OrNewer` into a 3-state
`ToolsetLayout VSLayout`. Either a build is DevDiv-internal,
released VS 2017 or newer, or released VS 2015 or older. When looking at
the directory structure, if instead of `"VC"` we see `"x86ret"`, `"x86chk"`,
`"amd64ret"`, or `"amd64chk"`, we recognize this as a DevDiv-internal build.
After we get past the directory structure validation, we use this knowledge
to regenerate paths appropriately. `llvmArchToDevDivInternalArch()` knows how
we use `"i386"` subdirectories, and `MSVCToolChain::getSubDirectoryPath()`
uses that. It also knows that DevDiv-internal builds have an `"inc"`
subdirectory instead of `"include"`.
This may still not be the "right" fix in any sense, but I believe that it's
non-intrusive in the sense that if the special directory names aren't found,
no codepaths are affected. (`ToolsetLayout::OlderVS` and
`ToolsetLayout::VS2017OrNewer` correspond to `IsVS2017OrNewer` being `false`
or `true`, respectively.) I searched for all references to `IsVS2017OrNewer`,
which are places where Clang cares about VS's directory structure, and the
only one that isn't being patched is some logic to deal with
cross-compilation. I'm fine with that not working for DevDiv-internal builds
for the moment (we typically test the native compilers), so I added a comment.
Fixes D36860.
llvm-svn: 311391
The LoopExit CFG information provides the opportunity to not mark the loops but
having a stack which tracks if a loop is unrolled or not. So in case of
simulating a loop we just add it and the information if it meets the
requirements to be unrolled to the top of the stack.
Differential Revision: https://reviews.llvm.org/D35684
llvm-svn: 311346
This patch adds handling of the LoopExit CFGElements to the StaticAnalyzer.
This is reached by introducing a new ProgramPoint.
Tests will be added in a following commit.
Differential Revision: https://reviews.llvm.org/D35670
llvm-svn: 311344
Summary:
The crash occurs when the first token after a preamble is a macro
expansion.
Fixed by moving replayPreambleConditionalStack from Parser into
Preprocessor. It is now called right after the predefines file is
processed.
Reviewers: erikjv, bkramer, klimek, yvvan
Reviewed By: bkramer
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D36872
llvm-svn: 311330
The comment markers accepted by the assembler vary between different targets,
but '//' is always accepted, so we should use that for consistency.
Differential revision: https://reviews.llvm.org/D36666
llvm-svn: 311325
Summary:
Add separate tests for the top-down and the bottom-up phase, as well as
one for the optimal matching.
Reviewers: arphaman
Subscribers: klimek
Differential Revision: https://reviews.llvm.org/D36185
llvm-svn: 311284
Summary:
Ignore macros and implicit AST nodes, as well as anything outside of the
main source file.
Reviewers: arphaman
Subscribers: klimek
Differential Revision: https://reviews.llvm.org/D36184
llvm-svn: 311280
clang/test/Tooling/clang-diff-ast.cpp:45:12: error: expected string not found in input
// CHECK: CXXConstructorDecl: X(void (char, int))
^
<stdin>:43:25: note: scanning from here
AccessSpecDecl: public(42)
^
<stdin>:44:2: note: possible intended match here
CXXConstructorDecl: X(void (char, int) __attribute__((thiscall)))(43)
^
llvm-svn: 311270
Summary:
Until we find a decent heuristic on how to choose between multiple
identical trees, there is no point in supporting multiple mappings.
This also enables matching of nodes with parents of different types,
because there are many instances where this is appropriate. For
example for and foreach statements; functions in the global or
other namespaces.
Reviewers: arphaman
Subscribers: klimek
Differential Revision: https://reviews.llvm.org/D36183
llvm-svn: 311251
This patch introduces a new CFG element CFGLoopExit that indicate when a loop
ends. It does not deal with returnStmts yet (left it as a TODO).
It hidden behind a new analyzer-config flag called cfg-loopexit (false by
default).
Test cases added.
The main purpose of this patch right know is to make loop unrolling and loop
widening easier and more efficient. However, this information can be useful for
future improvements in the StaticAnalyzer core too.
Differential Revision: https://reviews.llvm.org/D35668
llvm-svn: 311235
Adding escape check for the counter variable of the loop.
It is achieved by jumping back on the ExplodedGraph to its declStmt.
Differential Revision: https://reviews.llvm.org/D35657
llvm-svn: 311234
Summary:
This is done with -ast-dump; the JSON variant has been renamed to
-ast-dump-json.
Reviewers: arphaman
Differential Revision: https://reviews.llvm.org/D36180
llvm-svn: 311232
This patch, by hamzasood, implements P0409R2, and allows [=, this] pre-C++2a as an extension (with appropriate warnings) for consistency.
https://reviews.llvm.org/D36572
Thanks Hamza!
llvm-svn: 311224
Summary:
This also changes the output order of the changes. Now the matches are
printed in pre-order, intertwined with insertions, updates, and moves.
Deletions are printed afterwards.
Reviewers: arphaman
Subscribers: klimek
Differential Revision: https://reviews.llvm.org/D36179
llvm-svn: 311200
This made it awkward to switch over an enum where some entries
are partial and is unlikley to catch any bugs.
Differential revision: https://reviews.llvm.org/D36777
llvm-svn: 311191
Summary:
Augment SanitizerCoverage to insert maximum stack depth tracing for
use by libFuzzer. The new instrumentation is enabled by the flag
-fsanitize-coverage=stack-depth and is compatible with the existing
trace-pc-guard coverage. The user must also declare the following
global variable in their code:
thread_local uintptr_t __sancov_lowest_stack
https://bugs.llvm.org/show_bug.cgi?id=33857
Reviewers: vitalybuka, kcc
Reviewed By: vitalybuka
Subscribers: kubamracek, hiraditya, cfe-commits, llvm-commits
Differential Revision: https://reviews.llvm.org/D36839
llvm-svn: 311186
Summary:
Arguments can be specified after -ignore_remaining_args=1 to modify
the compiler invocation. For example, the following command-line
will fuzz LLVM with a custom optimization level and target triple:
clang-proto-fuzzer CORPUS/ -ignore_remaining_args -O3 \
-triple arm64-apple-ios9
Reviewers: vitalybuka, kcc
Reviewed By: vitalybuka
Subscribers: aemerson, cfe-commits, kristof.beyls
Differential Revision: https://reviews.llvm.org/D36882
llvm-svn: 311185
This diff fixes analyzer's crash (triggered assert) on the newly added test case.
The assert being discussed is assert(!B.lookup(R, BindingKey::Direct))
in lib/StaticAnalyzer/Core/RegionStore.cpp, however the root cause is different.
For classes with empty bases the offsets might be tricky.
For example, let's assume we have
struct S: NonEmptyBase, EmptyBase {
...
};
In this case Clang applies empty base class optimization and
the offset of EmptyBase will be 0, it can be verified via
clang -cc1 -x c++ -v -fdump-record-layouts main.cpp -emit-llvm -o /dev/null.
When the analyzer tries to perform zero initialization of EmptyBase
it will hit the assert because that region
has already been "written" by the constructor of NonEmptyBase.
Test plan:
make check-all
Differential revision: https://reviews.llvm.org/D36851
llvm-svn: 311182
Summary:
Support command line options for build path and extra arguments
This emulates the options accepted by clang tools that use CommonOptionsParser.
Add a flag for controlling the maximum size parameter for bottom up matching.
Reviewers: arphaman
Subscribers: klimek
Differential Revision: https://reviews.llvm.org/D36177
llvm-svn: 311173
Fix to the computation of the rightmost descendant.
Prevents root nodes from being mapped if they are already mapped. This
only makes a difference when we compare AST Nodes other than entire
translation units, a feature which still has to be tested.
Reviewers: arphaman
Subscribers: klimek
Differential Revision: https://reviews.llvm.org/D36176
llvm-svn: 311172
If a struct would end up half in GPRs and half on SP the ABI says it should
actually go entirely on the stack. We were getting this wrong in GlobalISel
before, causing compatibility issues.
llvm-svn: 311137
Base::TraverseStmt when visiting the then/else branches of if statements
This ensures that the statement stack is correctly tracked and correct
multi-statement fixit is generated inside of an if (@available)
llvm-svn: 311088
This commit adds the functionality of performing reference counting on the
callee side for Integer Set Library (ISL) to Clang Static Analyzer's
RetainCountChecker.
Reference counting on the callee side can be extensively used to perform
debugging within a function (For example: Finding leaks on error paths).
Patch by Malhar Thakkar!
Differential Revision: https://reviews.llvm.org/D36441
llvm-svn: 311063
In dependent contexts we end up referencing these, so make sure they
have USRs, and have their declarations indexed. For the most part they
behave like typedefs, but we also need to worry about having multiple
using declarations with the same "name".
rdar://problem/33883650
llvm-svn: 311053
-no-integrated-as is not supported on some targets (e.g.,
x86_64-pc-windows-msvc). Testing using -save-temps is good enough to cover the
relevant logic, and that should work everywhere.
llvm-svn: 311043
Using Output.getFilename() to construct the file name used for optimization
recording in Clang::ConstructJob, when -c is provided, does not work correctly
if we're not using the integrated assembler. With -no-integrated-as (or
-save-temps) Output.getFilename() gives the name of the temporary assembly
file, not the final output file. Instead, use the final output (as provided by
-o). If this is not available, then fall back to using a name based on the
input file.
Fixes PR31532.
llvm-svn: 311041
Summary:
Even in the case of the input file is a preprocessed source, clang uses the file name of the preprocesses source for debug info (DW_AT_name attribute for DW_TAG_compile_unit). However, gcc uses the file name specified in the first linemarker instead. This makes more sense because the one specified in the linemarker represents the "actual" source file name.
Clang already uses the file name specified in the first linemarker for Module name (https://github.com/llvm-mirror/clang/blob/master/lib/Frontend/FrontendAction.cpp#L779) if the input is preprocessed. This patch makes clang to use the same value for debug info as well.
Reviewers: compnerd, rnk, dblaikie, rsmith
Reviewed By: rnk
Subscribers: aprantl, cfe-commits
Differential Revision: https://reviews.llvm.org/D36474
llvm-svn: 311037
If worksharing construct has at least one linear item, an implicit
synchronization point must be emitted to avoid possible conflict with
the loading/storing values to the original variables. Added implicit
barrier if the linear item is found before actual start of the
worksharing construct.
llvm-svn: 311013
If exceptions are enabled, there may be a problem with the codegen of
the finalization functions from OpenMP runtime. It happens because of
the problem with the getting of thread identifier value. Patch tries to
fix it by using the result of the call of function
__kmpc_global_thread_num() rather than loading of value of outlined
function parameter.
llvm-svn: 311007
When r310905 moved the pointer and bool out of a PointerIntPair, it made
them end up uninitialized and caused UBSan failures when copying the
uninitialized boolean. However, making the pointer be null should avoid
the reference to the boolean entirely.
llvm-svn: 310994
constructors when deciding whether classes should be passed indirectly.
This fixes ABI differences between Clang and GCC:
* Previously, Clang ignored the move constructor when making this
determination. It now takes the move constructor into account, per
https://github.com/itanium-cxx-abi/cxx-abi/pull/17 (this change may
seem recent, but the ABI change was agreed on the Itanium C++ ABI
list a long time ago).
* Previously, Clang's behavior when the copy constructor was deleted
was unstable -- depending on whether the lazy declaration of the
copy constructor had been triggered, you might get different behavior.
We now eagerly declare the copy constructor whenever its deletedness
is unclear, and ignore deleted copy/move constructors when looking for
a trivial such constructor.
This also fixes an ABI difference between Clang and MSVC:
* If the copy constructor would be implicitly deleted (but has not been
lazily declared yet), for instance because the class has an rvalue
reference member, we would pass it directly. We now pass such a class
indirectly, matching MSVC.
Based on a patch by Vassil Vassilev, which was based on a patch by Bernd
Schmidt, which was based on a patch by Reid Kleckner!
This is a re-commit of r310401, which was reverted in r310464 due to ARM
failures (which should now be fixed).
llvm-svn: 310983
Saleem Abdulrasool