[analyzer] added cache for SMT queries in the SMTConstraintManager
Summary:
This patch implements a new cache for the result of SMT queries; with this patch the regression tests are 25% faster.
It's implemented as a `llvm::DenseMap` where the key is the hash of the set of the constraints in a state.
There is still one method that does not use the cache, `getSymVal`, because it needs to get a symbol interpretation from the SMT, which is not cached yet.
[analyzer] Moved all CSA code from the SMT API to a new header, `SMTConv.h`. NFC.
Summary:
With this patch, the SMT backend is almost completely detached from the CSA.
Unfortunate consequence is that we missed the `ConditionTruthVal` from the CSA and had to use `Optional<bool>`.
The Z3 solver implementation is still in the same file as the `Z3ConstraintManager`, in `lib/StaticAnalyzer/Core/Z3ConstraintManager.cpp` though, but except for that, the SMT API can be moved to anywhere in the codebase.
[analyzer] Templatefy SMTConstraintManager so more generic code can be moved from solver specific implementations. NFC.
Summary:
By making SMTConstraintManager a template and passing the SMT constraint type and expr, we can further move code from the Z3ConstraintManager class to the generic SMT constraint Manager.
Now, each SMT specific constraint manager only needs to implement the method `bool canReasonAbout(SVal X) const`.
Update avr attributes test for output change in r340519
After this commit there is an addrspace(1) before the attribute #. Since
these tests are only checking the value of the attribute add a {{.*}} to
make the test resilient to future output changes.
Ivan Donchevskii [Thu, 23 Aug 2018 09:48:11 +0000 (09:48 +0000)]
[libclang] Fix cursors for arguments of Subscript and Call operators
The DeclRefExpr of CXXOperatorCallExpr refering to the custom operator
is visited before the arguments to the operator call. For the Call and
Subscript operator the range of this DeclRefExpr includes the whole call
expression, so that all tokens in that range were mapped to the operator
function, even the tokens of the arguments.
Fix this by ensuring that this particular DeclRefExpr is visited last.
Chandler Carruth [Thu, 23 Aug 2018 06:06:38 +0000 (06:06 +0000)]
[x86/retpoline] Split the LLVM concept of retpolines into separate
subtarget features for indirect calls and indirect branches.
This is in preparation for enabling *only* the call retpolines when
using speculative load hardening.
I've continued to use subtarget features for now as they continue to
seem the best fit given the lack of other retpoline like constructs so
far.
The LLVM side is pretty simple. I'd like to eventually get rid of the
old feature, but not sure what backwards compatibility issues that will
cause.
This does remove the "implies" from requesting an external thunk. This
always seemed somewhat questionable and is now clearly not desirable --
you specify a thunk the same way no matter which set of things are
getting retpolines.
I really want to keep this nicely isolated from end users and just an
LLVM implementation detail, so I've moved the `-mretpoline` flag in
Clang to no longer rely on a specific subtarget feature by that name and
instead to be directly handled. In some ways this is simpler, but in
order to preserve existing behavior I've had to add some fallback code
so that users who relied on merely passing -mretpoline-external-thunk
continue to get the same behavior. We should eventually remove this
I suspect (we have never tested that it works!) but I've not done that
in this patch.
Petr Hosek [Wed, 22 Aug 2018 22:56:46 +0000 (22:56 +0000)]
[Driver] Check normalized triples for multiarch runtime path
Previously we only used target triple as provided which matches the
GCC behavior, but it also means that all clients have to be consistent
in their spelling of target triples since e.g. x86_64-linux-gnu and
x86_64-unknown-linux-gnu will result in Clang driver looking at two
different paths when searching for runtime libraries.
Unfortunatelly, as it turned out many clients aren't consistent in
their spelling of target triples, e.g. many Linux distributions use
the shorter spelling but config.guess and rustc insist on using the
normalized variant which is causing issues. To avoid having to ship
multiple copies of runtimes for different triple spelling or rely on
symlinks which are not portable, we should also check the normalized
triple when constructing paths for multiarch runtimes.
Raphael Isemann [Wed, 22 Aug 2018 22:50:45 +0000 (22:50 +0000)]
[ASTImporter] Actually test ArrayInitLoopExpr in the array-init-loop-expr test.
Summary:
The `array-init-loop-expr` test is currently not testing the importing of ArrayInitLoopExprs.
This is because we import the `S` struct into the `test.cpp` context
and only do a copy-assignment in `test.cpp`, so the actual ArrayInitLoopExpr we wanted to
import is generated by clang directly in the target context. This means we actually
never test the importing of ArrayInitLoopExpr with this test, which becomes obvious
when looking at the missing test coverage for the respective VisitArrayInitLoopExpr method.
This patch moves the copy-assignment of our struct to the `S.cpp` context, which means
that `test.cpp` now actually has to import the ArrayInitLoopExpr.
Raphael Isemann [Wed, 22 Aug 2018 22:49:32 +0000 (22:49 +0000)]
[ASTImporter] Remove duplicated and dead CXXNamedCastExpr handling code.
Summary:
`CXXNamedCastExpr` importing is already handled in the respective `VisitCXXNamedCastExpr` method.
So this code here can never be reached under normal circumstances and we might as well remove it.
This patch shouldn't change any observable behavior of the ASTImporter.
Currently clang does not emit unused static constants. GCC emits these
constants by default when there is no optimization.
GCC's option -fno-keep-static-consts can be used to not emit
unused static constants.
In Clang, since default behavior does not keep unused static constants,
-fkeep-static-consts can be used to emit these if required. This could be
useful for producing identification strings like SVN identifiers
inside the object file even though the string isn't used by the program.
Chih-Hung Hsieh [Wed, 22 Aug 2018 17:13:40 +0000 (17:13 +0000)]
[Tooling] Allow -flto flags and filter out -Wa, flags
This change fixes the problem in https://bugs.llvm.org/show_bug.cgi?id=38332
by allowing driver::Action::BackendJobClass to run with the analyzer.
Otherwise, such jobs will look up the non-existing compilation database
and then run without flags.
Also filter out the -Wa,* flags that could be passed to and ignored
by the clang compiler. Clang-tidy gives warnings about unused -Wa,* flags.
Akira Hatanaka [Wed, 22 Aug 2018 13:41:19 +0000 (13:41 +0000)]
[CodeGen] Look at the type of a block capture field rather than the type
of the captured variable when determining whether the capture needs
special handing when the block is copied or disposed.
This fixes bugs in the handling of variables captured by a block that is
nested inside a lambda that captures the variables by reference.
Henry Wong [Wed, 22 Aug 2018 13:30:46 +0000 (13:30 +0000)]
[analyzer] Improve `CallDescription` to handle c++ method.
Summary:
`CallDecription` can only handle function for the time being. If we want to match c++ method, we can only use method name to match and can't improve the matching accuracy through the qualifiers.
This patch add the support for `QualifiedName` matching to improve the matching accuracy.
Gabor Marton [Wed, 22 Aug 2018 11:52:14 +0000 (11:52 +0000)]
Fix import of class templates partial specialization
Summary:
Currently there are several issues with the import of class template
specializations. (1) Different TUs may have class template specializations
with the same template arguments, but with different set of instantiated
MethodDecls and FieldDecls. In this patch we provide a fix to merge these
methods and fields. (2) Currently, we search the partial template
specializations in the set of simple specializations and we add partial
specializations as simple specializations. This is bad, this patch fixes it.
Set __mips_fpr to 0 if o32 ABI is used with either -mfpxx
or none of -mfp32, -mfpxx, -mfp64 being specified.
Introduce additional checks:
-mfpxx is only to be used in conjunction with the o32 ABI.
report an error when incompatible options are provided.
Formerly no errors were raised when combining n32/n64 ABIs
with -mfp32 and -mfpxx.
There are other cases when __mips_fpr should be set to 0
that are not covered, ex. using o32 on a mips64 cpu
which is valid but not supported in the backend as of yet.
Simon Tatham [Wed, 22 Aug 2018 09:20:39 +0000 (09:20 +0000)]
[clang-tblgen] Add -print-records and -dump-json modes.
Currently, if clang-tblgen is run without a mode option, it defaults
to the first mode in its 'enum Action', which happens to be
-gen-clang-attr-classes. I think it makes more sense for it to behave
the same way as llvm-tblgen, i.e. print a diagnostic dump if it's not
given any more specific instructions.
I've also added the same -dump-json that llvm-tblgen supports. This
means any tblgen command line (whether llvm- or clang-) can be
mechanically turned into one that processes the same input into JSON.
Hiroshi Inoue [Wed, 22 Aug 2018 05:43:27 +0000 (05:43 +0000)]
[AST] correct the behavior of -fvisibility-inlines-hidden option (don't make static local variables hidden)
The command line option -fvisibility-inlines-hidden makes inlined method hidden, but it is expected not to affect the visibility of static local variables in the function.
However, Clang makes the static local variables in the function also hidden as reported in PR37595. This problem causes LLVM bootstarp failure on Fedora 28 if configured with -DBUILD_SHARED_LIBS=ON.
This patch makes the behavior of -fvisibility-inlines-hidden option to be consistent with that of gcc; the option does not change the visibility of the static local variables if the containing function does not associated with explicit visibility attribute and becomes hidden due to this option.
Martin Storsjo [Tue, 21 Aug 2018 20:41:17 +0000 (20:41 +0000)]
[CodeGen] Implicitly set stackrealign on the main function, if custom stack alignment is used
If using a custom stack alignment, one is expected to make sure
that all callers provide such alignment, or realign the stack in
all entry points (and callbacks).
Despite this, the compiler can assume that the main function will
need realignment in these cases, since the startup routines calling
the main function most probably won't provide the custom alignment.
This matches what GCC does in similar cases; if compiling with
-mincoming-stack-boundary=X -mpreferred-stack-boundary=X, GCC normally
assumes such alignment on entry to a function, but specifically for
the main function still does realignment.
Erik Pilkington [Tue, 21 Aug 2018 17:24:06 +0000 (17:24 +0000)]
Add a new flag and attributes to control static destructor registration
This commit adds the flag -fno-c++-static-destructors and the attributes
[[clang::no_destroy]] and [[clang::always_destroy]]. no_destroy specifies that a
specific static or thread duration variable shouldn't have it's destructor
registered, and is the default in -fno-c++-static-destructors mode.
always_destroy is the opposite, and is the default in -fc++-static-destructors
mode.
A variable whose destructor is disabled (either because of
-fno-c++-static-destructors or [[clang::no_destroy]]) doesn't count as a use of
the destructor, so we don't do any access checking or mark it referenced. We
also don't emit -Wexit-time-destructors for these variables.
Louis Dionne [Tue, 21 Aug 2018 15:54:24 +0000 (15:54 +0000)]
[clang][NFC] Fix typo in the name of a note
Summary:
r306722 introduced a new note called note_silence_unligned_allocation_unavailable
where I believe what was meant is note_silence_aligned_allocation_unavailable.
Kristof Umann [Tue, 21 Aug 2018 12:16:59 +0000 (12:16 +0000)]
[analyzer][UninitializedObjectChecker] Explicit namespace resolution for inherited data members
For the following example:
struct Base {
int x;
};
// In a different translation unit
struct Derived : public Base {
Derived() {}
};
For a call to Derived::Derived(), we'll receive a note that
this->x is uninitialized. Since x is not a direct field of Derived,
it could be a little confusing. This patch aims to fix this, as well
as the case when the derived object has a field that has the name as
an inherited uninitialized data member:
struct Base {
int x; // note: uninitialized field 'this->Base::x'
};
struct Derived : public Base {
int x = 5;
Derived() {}
};
libclang: add install/distribution targets for python
Add installation support for the python bindings for libclang. Add an
additional CMake configuration variable to enumerate the python versions for
which the bindings should be installed. This allows for a LLVM/clang
distribution to distribute the python bindings for libclang as part of the
image. Because the python versions need to be explicitly stated by the user,
the default image remains unchanged.
Richard Smith [Mon, 20 Aug 2018 21:47:29 +0000 (21:47 +0000)]
Model type attributes as regular Attrs.
Specifically, AttributedType now tracks a regular attr::Kind rather than
having its own parallel Kind enumeration, and AttributedTypeLoc now
holds an Attr* instead of holding an ad-hoc collection of Attr fields.
This reinstates r339623, reverted in r339638, with a fix to not fail
template instantiation if we instantiate a QualType with no associated
type source information and we encounter an AttributedType.
David Blaikie [Mon, 20 Aug 2018 20:14:08 +0000 (20:14 +0000)]
DebugInfo: Add the ability to disable DWARF name tables entirely
This changes the current default behavior (from emitting pubnames by
default, to not emitting them by default) & moves to matching GCC's
behavior* with one significant difference: -gno(-gnu)-pubnames disables
pubnames even in the presence of -gsplit-dwarf (though -gsplit-dwarf
still by default enables -ggnu-pubnames). This allows users to disable
pubnames (& the new DWARF5 accelerated access tables) when they might
not be worth the size overhead.
* GCC's behavior is that -ggnu-pubnames and -gpubnames override each
other, and that -gno-gnu-pubnames and -gno-pubnames act as synonyms and
disable either kind of pubnames if they come last. (eg: -gpubnames
-gno-gnu-pubnames causes no pubnames (neither gnu or standard) to be
emitted)
Alexey Bataev [Mon, 20 Aug 2018 18:03:40 +0000 (18:03 +0000)]
[OPENMP] Fix crash on the emission of the weak function declaration.
If the function is actually a weak reference, it should not be marked as
deferred definition as this is only a declaration. Patch adds checks for
the definitions if they must be emitted. Otherwise, only declaration is
emitted.
Adrian Prantl [Mon, 20 Aug 2018 17:10:27 +0000 (17:10 +0000)]
Close FileEntries of cached files in ModuleManager::addModule().
While investigating why LLDB (which can build hundreds of clang
modules during one debug session) was getting "too many open files"
errors, I found that most of them are .pcm files that are kept open by
ModuleManager. Pretty much all of the open file dscriptors are
FileEntries that are refering to `.pcm` files for which a buffer
already exists in a CompilerInstance's PCMCache.
Before PCMCache was added it was necessary to hold on to open file
descriptors to ensure that all ModuleManagers using the same
FileManager read the a consistent version of a given `.pcm` file on
disk, even when a concurrent clang process overwrites the file halfway
through. The PCMCache makes this practice unnecessary, since it caches
the entire contents of a `.pcm` file, while the FileManager caches all
the stat() information.
This patch adds a call to FileEntry::closeFile() to the path where a
Buffer has already been created. This is necessary because even for a
freshly written `.pcm` file the file is stat()ed once immediately
after writing to generate a FileEntry in the FileManager. Because a
freshly-generated file's contents is stored in the PCMCache, it is
fine to close the file immediately thereafter. The second change this
patch makes is to set the `ShouldClose` flag to true when reading a
`.pcm` file into the PCMCache for the first time.
[For reference, in 1 Clang instance there is
- 1 FileManager and
- n ModuleManagers with
- n PCMCaches.]
Raphael Isemann [Mon, 20 Aug 2018 16:20:01 +0000 (16:20 +0000)]
[ASTImporter] Add test for C++ casts and fix broken const_cast importing.
Summary:
The ASTImporter does currently not handle const_casts. This patch adds the
missing const_cast importer code and the test case that discovered this.
Alexey Bataev [Mon, 20 Aug 2018 16:00:22 +0000 (16:00 +0000)]
[OPENMP][BLOCKS]Fix PR38923: reference to a global variable is captured
by a block.
Added checks for capturing of the variable in the block when trying to
emit correct address for the variable with the reference type. This
extra check allows correctly identify the variables that are not
captured in the block context.
Sanjay Patel [Sun, 19 Aug 2018 16:50:30 +0000 (16:50 +0000)]
[CodeGen] add rotate builtins that map to LLVM funnel shift
This is a partial retry of rL340137 (reverted at rL340138 because of gcc host compiler crashing)
with 1 change:
Remove the changes to make microsoft builtins also use the LLVM intrinsics.
This exposes the LLVM funnel shift intrinsics as more familiar bit rotation functions in clang
(when both halves of a funnel shift are the same value, it's a rotate).
We're free to name these as we want because we're not copying gcc, but if there's some other
existing art (eg, the microsoft ops) that we want to replicate, we can change the names.
The funnel shift intrinsics were added here:
https://reviews.llvm.org/D49242
With improved codegen in:
https://reviews.llvm.org/rL337966
https://reviews.llvm.org/rL339359
And basic IR optimization added in:
https://reviews.llvm.org/rL338218
https://reviews.llvm.org/rL340022
...so these are expected to produce asm output that's equal or better to the multi-instruction
alternatives using primitive C/IR ops.
In the motivating loop example from PR37387:
https://bugs.llvm.org/show_bug.cgi?id=37387#c7
...we get the expected 'rolq' x86 instructions if we substitute the rotate builtin into the source.
Ivan A. Kosarev [Sun, 19 Aug 2018 16:30:57 +0000 (16:30 +0000)]
[NEON] Define fp16 vld and vst intrinsics conditionally
This patch fixes definitions of vld and vst NEON intrinsics so
that we only define them if half-precision arithmetic is
supported on the target platform, as prescribed in ACLE 2.0.
Sanjay Patel [Sun, 19 Aug 2018 14:44:47 +0000 (14:44 +0000)]
[CodeGen] add/fix rotate builtins that map to LLVM funnel shift (retry)
This is a retry of rL340135 (reverted at rL340136 because of gcc host compiler crashing)
with 2 changes:
1. Move the code into a helper to reduce code duplication (and hopefully work-around the crash).
2. The original commit had a formatting bug in the docs (missing an underscore).
Original commit message:
This exposes the LLVM funnel shift intrinsics as more familiar bit rotation functions in clang
(when both halves of a funnel shift are the same value, it's a rotate).
We're free to name these as we want because we're not copying gcc, but if there's some other
existing art (eg, the microsoft ops that are modified in this patch) that we want to replicate,
we can change the names.
The funnel shift intrinsics were added here:
https://reviews.llvm.org/D49242
With improved codegen in:
https://reviews.llvm.org/rL337966
https://reviews.llvm.org/rL339359
And basic IR optimization added in:
https://reviews.llvm.org/rL338218
https://reviews.llvm.org/rL340022
...so these are expected to produce asm output that's equal or better to the multi-instruction
alternatives using primitive C/IR ops.
In the motivating loop example from PR37387:
https://bugs.llvm.org/show_bug.cgi?id=37387#c7
...we get the expected 'rolq' x86 instructions if we substitute the rotate builtin into the source.
Sanjay Patel [Sun, 19 Aug 2018 13:48:06 +0000 (13:48 +0000)]
revert r340135: [CodeGen] add rotate builtins
At least a couple of bots (PPC only?) are showing the compiler dying while trying to compile:
http://lab.llvm.org:8011/builders/clang-ppc64be-linux-multistage/builds/11065/steps/build%20stage%201/logs/stdio
http://lab.llvm.org:8011/builders/clang-ppc64be-linux-lnt/builds/18267/steps/build%20stage%201/logs/stdio
Sanjay Patel [Sun, 19 Aug 2018 13:12:40 +0000 (13:12 +0000)]
[CodeGen] add rotate builtins
This exposes the LLVM funnel shift intrinsics as more familiar bit rotation functions in clang
(when both halves of a funnel shift are the same value, it's a rotate).
We're free to name these as we want because we're not copying gcc, but if there's some other
existing art (eg, the microsoft ops that are modified in this patch) that we want to replicate,
we can change the names.
The funnel shift intrinsics were added here:
D49242
With improved codegen in:
rL337966
rL339359
And basic IR optimization added in:
rL338218
rL340022
...so these are expected to produce asm output that's equal or better to the multi-instruction
alternatives using primitive C/IR ops.
In the motivating loop example from PR37387:
https://bugs.llvm.org/show_bug.cgi?id=37387#c7
...we get the expected 'rolq' x86 instructions if we substitute the rotate builtin into the source.
George Karpenkov [Sat, 18 Aug 2018 01:45:50 +0000 (01:45 +0000)]
[analyzer] [NFC] Split up RetainSummaryManager from RetainCountChecker
ARCMigrator is using code from RetainCountChecker, which is a layering
violation (and it also does it badly, by using a different header, and
then relying on implementation being present in a header file).
This change splits up RetainSummaryManager into a separate library in
lib/Analysis, which can be used independently of a checker.
Alex Lorenz [Fri, 17 Aug 2018 22:18:08 +0000 (22:18 +0000)]
[ObjC] Error out when using forward-declared protocol in a @protocol
expression
Clang emits invalid protocol metadata when a @protocol expression is used with a
forward-declared protocol. The protocol metadata is missing protocol conformance
list of the protocol since we don't have access to the definition of it in the
compiled translation unit. The linker then might end up picking the invalid
metadata when linking which will lead to incorrect runtime protocol conformance
checks.
This commit makes sure that Clang fails to compile code that uses a @protocol
expression with a forward-declared protocol. This ensures that Clang does not
emit invalid protocol metadata. I added an extra assert in CodeGen to ensure
that this kind of issue won't happen in other places.
Reid Kleckner [Fri, 17 Aug 2018 22:11:31 +0000 (22:11 +0000)]
Don't warn on returning the address of a label from a statement expression
Summary:
There isn't anything inherently wrong with returning a label from a
statement expression. In practice, the Linux kernel uses this pattern to
materialize PCs.
George Karpenkov [Fri, 17 Aug 2018 21:40:38 +0000 (21:40 +0000)]
[analyzer] Drop support for GC mode in RetainCountChecker
A lot of code in RetainCountChecker deals with GC mode.
Given that GC mode is deprecated, Apple does not ship runtime for it,
and modern compiler toolchain does not support it, it makes sense to
remove the code dealing with it in order to aid understanding of
RetainCountChecker.
Reid Kleckner [Fri, 17 Aug 2018 20:59:27 +0000 (20:59 +0000)]
[MS] Mangle a hash of the main file path into anonymous namespaces
Summary:
This is needed to avoid conflicts in mangled names for codeview types in
anonymous namespaces. In CodeView, types refer to each other typically
through forward declarations, which contain mangled names. These names
have to be unique, otherwise the debugger will look up the mangled name
and find the wrong definition.
Furthermore, ThinLTO will deduplicate the types, and debug info
verification can fail when the types have the wrong sizes. This is
PR38608.
Richard Smith [Fri, 17 Aug 2018 19:43:40 +0000 (19:43 +0000)]
Improve diagnostic for missing comma in template parameter list.
Given 'typename T typename U', we would correctly diagnose the missing
comma, but incorrectly disambiguate the first parameter as being a
non-type parameter and complain that the 'T' is not a qualified-id.
projects / clang / log