Petar Jovanovic [Tue, 9 May 2017 16:24:03 +0000 (16:24 +0000)]
[mips] Impose a threshold for coercion of aggregates
Modified MipsABIInfo::classifyArgumentType so that it now coerces aggregate
structures only if the size of said aggregate is less than 16/64 bytes,
depending on the ABI.
Faisal Vali [Tue, 9 May 2017 04:17:15 +0000 (04:17 +0000)]
Fix PR32638 : Make sure we switch Sema's CurContext to the substituted FunctionDecl when instantiating the exception specification.
This fixes the bug: https://bugs.llvm.org/show_bug.cgi?id=32638
int main()
{
[](auto x) noexcept(noexcept(x)) { } (0);
}
In the above code, prior to this patch, when substituting into the noexcept expression, i.e. transforming the DeclRefExpr that represents 'x' - clang attempts to capture 'x' because Sema's CurContext is still pointing to the pattern FunctionDecl (i.e. the templated-decl set in FinishTemplateArgumentDeduction) which does not match the substituted 'x's DeclContext, which leads to an attempt to capture and an assertion failure.
We fix this by adjusting Sema's CurContext to point to the substituted FunctionDecl under which the noexcept specifier's argument should be transformed, and so the ParmVarDecl that 'x' refers to has the same declcontext and no capture is attempted.
I briefly investigated whether the SwitchContext should occur right after VisitMethodDecl creates the new substituted FunctionDecl, instead of only during instantiating the exception specification - but seeing no other code that seemed to rely on that, I decided to leave it just for the duration of the exception specification instantiation.
[Sema] Make typeof(OverloadedFunctionName) not a pointer.
We were sometimes doing a function->pointer conversion in
Sema::CheckPlaceholderExpr, which isn't the job of CheckPlaceholderExpr.
So, when we saw typeof(OverloadedFunctionName), where
OverloadedFunctionName referenced a name with only one function that
could have its address taken, we'd give back a function pointer type
instead of a function type. This is incorrect.
I kept the logic for doing the function pointer conversion in
resolveAndFixAddressOfOnlyViableOverloadCandidate because it was more
consistent with existing ResolveAndFix* methods.
Richard Trieu [Tue, 9 May 2017 03:24:34 +0000 (03:24 +0000)]
[ODRHash] Loosen checks on typedefs.
When a type in a class is from a typedef, only check the canonical type. Skip
checking the intermediate underlying types. This is in response to PR 32965
Akira Hatanaka [Tue, 9 May 2017 01:54:51 +0000 (01:54 +0000)]
[Sema][ObjC] Clean up possible null dereference.
It appears that the code is actually dead since unbridged-cast
placeholder types are created by calling CastOperation::complete and
ImplicitCastExprs are never passed to it.
Akira Hatanaka [Tue, 9 May 2017 01:20:05 +0000 (01:20 +0000)]
[CodeGen][ObjC] Emit @objc_retain at -O0 for variables captured by
blocks.
r302270 made changes to avoid emitting clang.arc.use at -O0 and instead
emit @objc_release. We also have to emit @objc_retain for the captured
variable at -O0 to match the @objc_release instead of just storing the
pointer to the capture field.
[XRay] Add __xray_customeevent(...) as a clang-supported builtin
Summary:
We define the `__xray_customeevent` builtin that gets translated to
IR calls to the correct intrinsic. The default implementation of this is
a no-op function. The codegen side of this follows the following logic:
- When `-fxray-instrument` is not provided in the driver, we elide all
calls to `__xray_customevent`.
- When `-fxray-instrument` is enabled and a function is marked as "never
instrumented", we elide all calls to `__xray_customevent` in that
function; if either marked as "always instrumented" or subject to
threshold-based instrumentation, we emit a call to the
`llvm.xray.customevent` intrinsic from LLVM for each
`__xray_customevent` occurrence in the function.
This change depends on D27503 (to land in LLVM first).
[Modules] Allow umbrella frameworks to define private submodules for subframeworks
In r298391 we fixed the umbrella framework model to work when submodules
named "Private" are used. This complements the work by allowing the
umbrella framework model to work in general.
Vedant Kumar [Mon, 8 May 2017 21:11:55 +0000 (21:11 +0000)]
[Driver] Don't enable -fsanitize-use-after-scope when ASan is disabled
When enabling any sanitizer, -fsanitize-use-after-scope is enabled by
default. This doesn't actually turn ASan on, because we've been getting
lucky and there are extra checks in BackendUtil that stop this from
happening.
However, this has been causing a behavior change: extra lifetime markers
are emitted in some cases where they aren't needed or expected.
[Sema] Fix typos handling in an overloadable call.
In C typos in arguments in a call of an overloadable function lead
to a failure of construction of CallExpr and following recovery does
not handle created delayed typos. This causes an assertion fail in
Sema::~Sema since Sema::DelayedTypos remains not empty.
The patch fixes that behavior by handling a call with arguments
having dependant types in the way that C++ does.
Daniel Jasper [Mon, 8 May 2017 15:08:00 +0000 (15:08 +0000)]
[clang-format] Convert AlignEscapedNewlinesLeft to an enum, adding
DontAlign
This converts the clang-format option AlignEscapedNewlinesLeft from a
boolean to an enum, named AlignEscapedNewlines, with options Left (prev.
true), Right (prev. false), and a new option DontAlign.
When set to DontAlign, the backslashes are placed just after the last token in each line:
#define EXAMPLE \
do { \
int x = aaaaa; \
int b; \
int dddddddddd; \
} while (0)
Both of these have sizeof == 0, even in C++ mode, for GCC compatibility. The
first one also doesn't occupy a register when passed by value in GNU C++ mode,
unlike everything else.
On Darwin, we want to ignore the lot (and especially don't want to try to use
an i0 as we were).
Richard Smith [Fri, 5 May 2017 22:18:51 +0000 (22:18 +0000)]
Add support for building modules from preprocessed source.
To support this, an optional marker "#pragma clang module contents" is
recognized in module map files, and the rest of the module map file from that
point onwards is treated as the source of the module. Preprocessing a module
map produces the input module followed by the marker and then the preprocessed
contents of the module.
Ignoring line markers, a preprocessed module might look like this:
module A {
header "a.h"
}
#pragma clang module contents
#pragma clang module begin A
// ... a.h ...
#pragma clang module end
The preprocessed output generates line markers, which are not accepted by the
module map parser, so -x c++-module-map-cpp-output should be used to compile
such outputs.
A couple of major parts do not work yet:
1) The files that are listed in the module map must exist on disk, in order to
build the on-disk header -> module lookup table in the PCM file. To fix
this, we need the preprocessed output to track the file size and other stat
information we might use to build the lookup table.
2) Declaration ownership semantics don't work properly yet, since mapping from
a source location to a module relies on mapping from FileIDs to modules,
which we can't do if module transitions can occur in the middle of a file.
Jakub Kuderski [Fri, 5 May 2017 21:01:12 +0000 (21:01 +0000)]
Add cxxStdInitializerListExpr AST matcher
Summary:
This adds a new ASTMatcher for CXXStdInitializerListExprs that matches C++ initializer list expressions.
The primary motivation is to use it to fix [[ https://bugs.llvm.org/show_bug.cgi?id=32896 | PR32896 ]] (review here [[ https://reviews.llvm.org/D32767 | D32767 ]]).
Richard Trieu [Fri, 5 May 2017 20:47:50 +0000 (20:47 +0000)]
[ODRHash] Fix typo, NFC
NestedNameSpecifer to NestedNameSpecifier. This was not a problem before since
one of the included headers transitively brought in the definition of the class
and only manifested as a problem when using the typoed NestedNameSpecifer and
getting an incomplete type error instead of a typo correction.
CodeGen: avoid use of @clang.arc.use intrinsic at O0
The clang.arc.use intrinsic is removed via the ARC Contract Pass. This
pass is only executed in optimized builds (>= opt level 1). Prevent the
optimization implemented in SVN r301667 from triggering at optimization
level 0 like every other ARC use intrinsic usage.
Alex Lorenz [Fri, 5 May 2017 16:42:44 +0000 (16:42 +0000)]
Add a fix-it for -Wunguarded-availability
This patch adds a fix-it for the -Wunguarded-availability warning. This fix-it
is similar to the Swift one: it suggests that you wrap the statement in an
`if (@available)` check. The produced fixits are indented (just like the Swift
ones) to make them look nice in Xcode's fix-it preview.
Alex Lorenz [Fri, 5 May 2017 14:52:48 +0000 (14:52 +0000)]
[Driver] Add a "-mmacos_version_min" option that's an alias for
"-mmacosx_version_min"
The option -mmacosx_version_min will still be the canonical option for now, but
in the future we will switch over to -mmacos_version_min and make
-mmacosx_version_min an alias instead.
James Y Knight [Thu, 4 May 2017 21:31:17 +0000 (21:31 +0000)]
Fix whitespace before token-paste of an argument.
The whitespace should come from the argument name in the macro
expansion, rather than from the token passed to the macro (same as it
does when not pasting).
Added a new test case for the change in behavior to stringize_space.c.
FileCheck'ized macro_paste_commaext.c, tweaked the test case, and
added a comment; no behavioral change to this test.
Nico Weber [Thu, 4 May 2017 19:54:50 +0000 (19:54 +0000)]
Turn on -Wmicrosoft-enum-forward-reference by default.
clang-cl already errs or warns on everything that cl
warns on in /permissive- mode, except for enum foward
declarations (and ATL attributes).
So warn on enum forward declarations by default.
Reid Kleckner [Thu, 4 May 2017 19:51:05 +0000 (19:51 +0000)]
Fix bugs checking va_start in lambdas and erroneous contexts
Summary:
First, getCurFunction looks through blocks and lambdas, which is wrong.
Inside a lambda, va_start should refer to the lambda call operator
prototype. This fixes PR32737.
Second, we shouldn't use any of the getCur* methods, because they look
through contexts that we don't want to look through (EnumDecl,
CapturedStmtDecl). We can use CurContext directly as the calling
context.
Finally, this code assumed that CallExprs would never appear outside of
code contexts (block, function, obj-c method), which is wrong. Struct
member initializers are an easy way to create and parse exprs in a
non-code context.
Nico Weber [Thu, 4 May 2017 19:36:26 +0000 (19:36 +0000)]
Turn -Wmicrosoft-enum-value off by default.
The warning is currently way too noisy to be useful. The plan is
to make it warn when an MS enum that's negative is compared to
something, but until that's done the warning shouldn't default
to on.
Driver: only inject libstdc++ paths when needed on CrossWindows
When building with libc++ don't bother with injecting the libstdc++
search paths into the linker search path. This will make it easier to
switch between ld and lld.
Reid Kleckner [Thu, 4 May 2017 18:19:52 +0000 (18:19 +0000)]
[ms-inline-asm] Use the frontend size only for ambiguous instructions
This avoids problems on code like this:
char buf[16];
__asm {
movups xmm0, [buf]
mov [buf], eax
}
The frontend size in this case (1) is wrong, and the register makes the
instruction matching unambiguous. There are also enough bytes available
that we shouldn't complain to the user that they are potentially using
an incorrectly sized instruction to access the variable.
Martin Probst [Thu, 4 May 2017 15:04:04 +0000 (15:04 +0000)]
clang-format: [JS] exponentiation operator
Summary: While its precedence should be higher than multiplicative, LLVM does not have a level for that, so for the time being just treat it as multiplicative.
Alex Lorenz [Thu, 4 May 2017 13:56:51 +0000 (13:56 +0000)]
DiagnosticsEngine should clear DelayedDiagID before reporting the
delayed diagnostic
This fix avoids an infinite recursion that was uncovered in one of our internal
tests by r301992. The testcase is the most reduced version of that
auto-generated test.
This is an improved version of the reverted commit r302037. The previous fix
actually managed to expose another subtle bug whereby `fatal_too_many_errors`
error was reported twice, with the second report setting the
`FatalErrorOccurred` flag. That prevented the notes that followed the diagnostic
the caused `fatal_too_many_errors` to be emitted. This commit ensures that notes
that follow `fatal_too_many_errors` but that belong to the diagnostic that
caused `fatal_too_many_errors` won't be emitted by setting the
`FatalErrorOccurred` when emitting `fatal_too_many_errors`.
Alex Lorenz [Thu, 4 May 2017 10:42:06 +0000 (10:42 +0000)]
[Index] The relation between the declarations in template
specializations that 'override' declarations in the base template should use
the 'specializationOf' relation instead of 'specializationOf | overrideOf'.
The indexer relations are meant to be orthogonal, so 'specializationOf' is
better than the combined relation.
Xiuli Pan [Thu, 4 May 2017 07:31:20 +0000 (07:31 +0000)]
[OpenCL] Add intel_reqd_sub_group_size attribute support
Summary:
Add intel_reqd_sub_group_size attribute support as intel extension cl_intel_required_subgroup_size from
https://www.khronos.org/registry/OpenCL/extensions/intel/cl_intel_required_subgroup_size.txt
IR: Use pointers instead of GUIDs to represent edges in the module summary. NFCI.
When profiling a no-op incremental link of Chromium I found that the functions
computeImportForFunction and computeDeadSymbols were consuming roughly 10% of
the profile. The goal of this change is to improve the performance of those
functions by changing the map lookups that they were previously doing into
pointer dereferences.
This is achieved by changing the ValueInfo data structure to be a pointer to
an element of the global value map owned by ModuleSummaryIndex, and changing
reference lists in the GlobalValueSummary to hold ValueInfos instead of GUIDs.
This means that a ValueInfo will take a client directly to the summary list
for a given GUID.
Richard Smith [Thu, 4 May 2017 00:29:54 +0000 (00:29 +0000)]
Add #pragma clang module begin/end pragmas and generate them when preprocessing a module.
These pragmas are intended to simulate the effect of entering or leaving a file
with an associated module. This is not completely implemented yet: declarations
between the pragmas will not be attributed to the correct module, but macro
visibility is already functional.
Modules named by #pragma clang module begin must already be known to clang (in
some module map that's either loaded or on the search path).
Tim Northover [Wed, 3 May 2017 19:20:45 +0000 (19:20 +0000)]
Tests: strengthen CHECK line to avoid picking up stray path.
A bot had "-LTO" in its working directory, which matched the regex used in this
test. Since the arg is quoted, we can exploit that instead. Still broken if
there's a path with a quote in, but I think that's pretty niche.
Reid Kleckner [Wed, 3 May 2017 18:17:31 +0000 (18:17 +0000)]
[IR] Abstract away ArgNo+1 attribute indexing as much as possible
Summary:
Do three things to help with that:
- Add AttributeList::FirstArgIndex, which is an enumerator currently set
to 1. It allows us to change the indexing scheme with fewer changes.
- Add addParamAttr/removeParamAttr. This just shortens addAttribute call
sites that would otherwise need to spell out FirstArgIndex.
- Remove some attribute-specific getters and setters from Function that
take attribute list indices. Most of these were only used from
BuildLibCalls, and doesNotAlias was only used to test or set if the
return value is malloc-like.
I'm happy to split the patch, but I think they are probably easier to
review when taken together.
This patch should be NFC, but it sets the stage to change the indexing
scheme to this, which is more convenient when indexing into an array:
0: func attrs
1: retattrs
2...: arg attrs
Alex Lorenz [Wed, 3 May 2017 15:41:16 +0000 (15:41 +0000)]
DiagnosticsEngine should clear DelayedDiagID before reporting the
delayed diagnostic
This avoids an infinite loop that was uncovered in one of our internal tests
by r301992. The testcase is the most reduced version of that auto-generated
test.
Carlo Bertolli [Wed, 3 May 2017 15:28:48 +0000 (15:28 +0000)]
[OpenMP] Extended parse for 'always' map modifier
https://reviews.llvm.org/D32807
This patch allows the map modifier 'always' to be separated by the map type (to, from, tofrom) only by a whitespace, rather than strictly by a comma as in current trunk.
Oren Ben Simhon [Wed, 3 May 2017 14:05:00 +0000 (14:05 +0000)]
Reusing an existing attribute diagnostic
In a previous patch, a new generic error diagnostic for inconsistent attributes was added.
In this commit I reuse this diagnostic for ns_returns_retained attribute check.
It was written as "Memory Error" in most places and as "Memory error" in a few
other places, however it is the latter that is more consistent with
other categories (such as "Logic error").
Daniel Jasper [Wed, 3 May 2017 11:27:34 +0000 (11:27 +0000)]
Undo turning ExtBehavior into a bitfield.
This produces warnings that I can't explain in a GCC build:
In file included from ../tools/clang/include/clang/Lex/LexDiagnostic.h:13:0,
from /usr/local/google/home/djasper/llvm/tools/clang/lib/Lex/PTHLexer.cpp:19:
../tools/clang/include/clang/Basic/Diagnostic.h:219:34: warning: ‘clang::DiagnosticsEngine::DiagState::ExtBehavior’ is too small to hold all values of ‘enum class clang::diag::Severity’ [enabled by default]
diag::Severity ExtBehavior : 4; // Map extensions to warnings or errors?
^
While I don't think this warning makes sense, I'd like to get this back to being
warning-free. This only seems to trigger for "enum class".
Daniel Jasper [Wed, 3 May 2017 07:48:27 +0000 (07:48 +0000)]
Silences gcc's -Wnarrowing.
I think this is a false positive in GCC's warning, but nonetheless, we
should try to be warning-free. Smaller reproducer (reproduces with GCC
6.3):
https://godbolt.org/g/cJuO2z
Yuka Takahashi [Wed, 3 May 2017 04:58:39 +0000 (04:58 +0000)]
Fix a bug that -isysroot is completely ignored on Unix
-isysroot is the flag which set the system root directory.
This bug report https://bugs.llvm.org//show_bug.cgi?id=11503
shows that -isysroot is not handled at all on Unix, so fixed this bug.
After this diff, I could get this result https://pastebin.com/TeCmn9mj .
Richard Smith [Wed, 3 May 2017 00:28:49 +0000 (00:28 +0000)]
[modules] Round-trip -Werror flag through explicit module build.
The intent for an explicit module build is that the diagnostics produced within
the module are those that were configured when the module was built, not those
that are enabled within a user of the module. This includes diagnostics that
don't actually show up until the module is used (for instance, diagnostics
produced during template instantiation and weird cases like -Wpadded).
We serialized and restored the diagnostic state for individual warning groups,
but previously did not track the state for flags like -Werror and -Weverything,
which are implemented as separate bits rather than as part of the diagnostics
mapping information.
Vedant Kumar [Tue, 2 May 2017 23:46:56 +0000 (23:46 +0000)]
[ubsan] Skip overflow checks on safe arithmetic (fixes PR32874)
Currently, ubsan emits overflow checks for arithmetic that is known to
be safe at compile-time, e.g:
1 + 1 => CheckedAdd(1, 1)
This leads to breakage when using the __builtin_prefetch intrinsic. LLVM
expects the arguments to @llvm.prefetch to be constant integers, and
when ubsan inserts unnecessary checks on the operands to the intrinsic,
this contract is broken, leading to verifier failures (see PR32874).
Instead of special-casing __builtin_prefetch for ubsan, this patch fixes
the underlying problem, i.e that clang currently emits unnecessary
overflow checks.
Testing: I ran the check-clang and check-ubsan targets with a stage2,
ubsan-enabled build of clang. I added a regression test for PR32874, and
some extra checking to make sure we don't regress runtime checking for
unsafe arithmetic. The existing ubsan-promoted-arithmetic.cpp test also
provides coverage for this change.
Faisal Vali [Tue, 2 May 2017 21:02:46 +0000 (21:02 +0000)]
[NFC] Add original test that triggered crash post r301735
- this is added just for completeness sake (though the general case should be represented by the test added in the revision to that patch: https://reviews.llvm.org/rL301972 )
When computing the appropriate cv-qualifiers for the 'this' capture, we have to examine each enclosing lambda - but when using the FunctionScopeInfo stack we have to ensure that the lambda below (outer) is the decl-context of the closure-class of the current lambda.
https://bugs.llvm.org/show_bug.cgi?id=32831
This patch was initially committed here: https://reviews.llvm.org/rL301735
Then reverted here: https://reviews.llvm.org/rL301916
The issue with the original patch was a failure to check that the closure type has been created within the LambdaScopeInfo before querying its DeclContext - instead of just assuming it has (silly!). A reduced example such as this highlights the problem:
struct X {
int data;
auto foo() { return [] { return [] -> decltype(data) { return 0; }; }; }
};
When 'data' within decltype(data) tries to determine the type of 'this', none of the LambdaScopeInfo's have their closure types created at that point.
Reid Kleckner [Tue, 2 May 2017 20:10:03 +0000 (20:10 +0000)]
Simplify some va_start checking logic
Combine the logic doing the ms_abi/sysv_abi checks into one function so
that each check and its logical opposite are near each other. Now we
don't need two Sema entry points for MS va_start and regular va_start.
Refactor the code that checks if the va_start caller is a function,
block, or obj-c method. We do this in three places, and they are all
buggy for variadic lambdas (PR32737). After this change, I have one
place to apply the functional fix.
When discovering the initializer for A::a, the bounds of A::b aren't known yet.
It is unclear whether warning about errors should be deferred until the end of
the translation unit, possibly resolving errors that can be resolved. In
practice, the compiler can know the bounds of all arrays in this example.
Credits for reproducers and explanation go to Richard Smith. Richard, please
add more info in case my explanation is wrong.
Nick Lewycky [Tue, 2 May 2017 01:06:16 +0000 (01:06 +0000)]
Revert r301785 (and r301787) because they caused PR32864.
The fix is that ExprEvaluatorBase::VisitInitListExpr should handle transparent exprs instead of exprs with one element. Fixing that uncovers one testcase failure because the AST for "constexpr _Complex float test2 = {1};" is wrong (the _Complex prvalue should not be const-qualified), and a number of test failures in test/OpenMP where the captured stmt contains an InitListExpr that is in syntactic form.
Richard Smith [Mon, 1 May 2017 22:10:47 +0000 (22:10 +0000)]
Fix initial diagnostic state setup for an explicit module with no diagnostic pragmas.
If a file has no diagnostic pragmas, we build its diagnostic state lazily, but
in this case we never set up the root state to be the diagnostic state in which
the module was originally built, so the diagnostic flags for files in the
module with no diagnostic pragmas were incorrectly based on the user of the
module rather than the diagnostic state when the module was built.
projects / clang / log