Differential Revision: https://reviews.llvm.org/D70268
This is a recommit of f978ea4983 with a fix for the PowerPC failure.
The issue was that:
* `CompilerInstance::ExecuteAction` calls
`getTarget().adjust(getLangOpts());`.
* `PPCTargetInfo::adjust` changes `LangOptions::HasAltivec`.
* This happens after the first few calls to `getModuleHash`.
There’s even a FIXME saying:
```
// FIXME: We shouldn't need to do this, the target should be immutable once
// created. This complexity should be lifted elsewhere.
```
This only showed up on PowerPC because it's one of the few targets that
almost always changes a hashed langopt.
I looked into addressing the fixme, but that would be a much larger
change, and it's not the only thing that happens in `ExecuteAction` that
can change the module context hash. Instead I changed the code to not
call `getModuleHash` until after it has been modified in `ExecuteAction`.
MSVC 2013 would refuse to pass highly aligned things (typically vectors
and aggregates) by value. Users would receive this error:
t.cpp(11) : error C2719: 'w': formal parameter with __declspec(align('32')) won't be aligned
t.cpp(11) : error C2719: 'q': formal parameter with __declspec(align('32')) won't be aligned
However, in MSVC 2015, this behavior was changed, and highly aligned
things are now passed indirectly. To avoid breaking backwards
incompatibility, objects that do not have a *required* high alignment
(i.e. double) are still passed directly, even though they are not
naturally aligned. This change implements the new behavior of passing
things indirectly.
The new behavior is:
- up to three vector parameters can be passed in [XYZ]MM0-2
- remaining arguments with required alignment greater than 4 bytes are
passed indirectly
Previously, MSVC never passed things truly indirectly, meaning clang
would always apply the byval attribute to indirect arguments. We had to
go to the trouble of adding inalloca so that non-trivially copyable C++
types could be passed in place without copying the object
representation. When inalloca was added, we asserted that all arguments
passed indirectly must use byval. With this change, that assert no
longer holds, and I had to update inalloca to handle that case. The
implicit sret pointer parameter was already handled this way, and this
change generalizes some of that logic to arguments.
There are two cases that this change leaves unfixed:
1. objects that are non-trivially copyable *and* overaligned
2. vectorcall + inalloca + vectors
For case 1, I need to touch C++ ABI code in MicrosoftCXXABI.cpp, so I
want to do it in a follow-up.
For case 2, my fix is one line, but it will require updating IR tests to
use lots of inreg, so I wanted to separate it out.
Related to D71915 and D72110
Fixes most of PR44395
Reviewed By: rjmccall, craig.topper, erichkeane
Differential Revision: https://reviews.llvm.org/D72114
Now with concepts support merged and mostly complete, we do not need -fconcepts-ts
(which was also misleading as we were not implementing the TS) and can enable
concepts features under C++2a. A warning will be generated if users still attempt
to use -fconcepts-ts.
Summary:
I initially encountered those assertions when trying to create
this IR `alignment` attribute from clang's `__attribute__((assume_aligned(imm)))`,
because until D72994 there is no sanity checking for the value of `imm`.
But even then, we have `llvm::Value::MaximumAlignment` constant (which is `536870912`),
which is enforced for clang attributes, and then there are some other magical constant
(`0x40000000` i.e. `1073741824` i.e. `2 * 536870912`) in
`Attribute::getWithAlignment()`/`AttrBuilder::addAlignmentAttr()`.
I strongly suspect that `0x40000000` is incorrect,
and that also should be `llvm::Value::MaximumAlignment`.
Reviewers: erichkeane, hfinkel, jdoerfert, gchatelet, courbet
Reviewed By: erichkeane
Subscribers: hiraditya, cfe-commits, llvm-commits
Tags: #llvm, #clang
Differential Revision: https://reviews.llvm.org/D72998
This patch implements P1141R2 "Yet another approach for constrained declarations".
General strategy for this patch was:
- Expand AutoType to include optional type-constraint, reflecting the wording and easing the integration of constraints.
- Replace autos in parameter type specifiers with invented parameters in GetTypeSpecTypeForDeclarator, using the same logic
previously used for generic lambdas, now unified with abbreviated templates, by:
- Tracking the template parameter lists in the Declarator object
- Tracking the template parameter depth before parsing function declarators (at which point we can match template
parameters against scope specifiers to know if we have an explicit template parameter list to append invented parameters
to or not).
- When encountering an AutoType in a parameter context we check a stack of InventedTemplateParameterInfo structures that
contain the info required to create and accumulate invented template parameters (fields that were already present in
LambdaScopeInfo, which now inherits from this class and is looked up when an auto is encountered in a lambda context).
Resubmit after fixing MSAN failures caused by incomplete initialization of AutoTypeLocs in TypeSpecLocFiller.
Differential Revision: https://reviews.llvm.org/D65042
Summary:
Immediate vmvnq is code-generated as a simple vector constant in IR,
and left to the backend to recognize that it can be created with an
MVE VMVN instruction. The predicated version is represented as a
select between the input and the same constant, and I've added a
Tablegen isel rule to turn that into a predicated VMVN. (That should
be better than the previous VMVN + VPSEL: it's the same number of
instructions but now it can fold into an adjacent VPT block.)
The unpredicated forms of VBIC and VORR are done by enabling the same
isel lowering as for NEON, recognizing appropriate immediates and
rewriting them as ARMISD::VBICIMM / ARMISD::VORRIMM SDNodes, which I
then instruction-select into the right MVE instructions (now that I've
also reworked those instructions to use the same MC operand encoding).
In order to do that, I had to promote the Tablegen SDNode instance
`NEONvorrImm` to a general `ARMvorrImm` available in MVE as well, and
similarly for `NEONvbicImm`.
The predicated forms of VBIC and VORR are represented as a vector
select between the original input vector and the output of the
unpredicated operation. The main convenience of this is that it still
lets me use the existing isel lowering for VBICIMM/VORRIMM, and not
have to write another copy of the operand encoding translation code.
This intrinsic family is the first to use the `imm_simd` system I put
into the MveEmitter tablegen backend. So, naturally, it showed up a
bug or two (emitting bogus range checks and the like). Fixed those,
and added a full set of tests for the permissible immediates in the
existing Sema test.
Also adjusted the isel pattern for `vmovlb.u8`, which stopped matching
because lowering started turning its input into a VBICIMM. Now it
recognizes the VBICIMM instead.
Reviewers: dmgreen, MarkMurrayARM, miyuki, ostannard
Reviewed By: dmgreen
Subscribers: kristof.beyls, hiraditya, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D72934
After rGb4a99a061f517e60985667e39519f60186cbb469, passing a response file such as -Wp,@a.rsp wasn't working anymore because .rsp expansion happens inside clang's main() function.
This patch adds response file expansion in the -cc1 tool.
Differential Revision: https://reviews.llvm.org/D73120
We previously had to guard against older MSVC and GCC versions which had rvalue
references but not support for marking functions with ref qualifiers. However,
having bumped our minimum required version to MSVC 2017 and GCC 5.1 mean we can
unconditionally enable this feature. Rather than keeping the macro around, this
replaces use of the macro with the actual ref qualifier.
This patch implements P1141R2 "Yet another approach for constrained declarations".
General strategy for this patch was:
- Expand AutoType to include optional type-constraint, reflecting the wording and easing the integration of constraints.
- Replace autos in parameter type specifiers with invented parameters in GetTypeSpecTypeForDeclarator, using the same logic
previously used for generic lambdas, now unified with abbreviated templates, by:
- Tracking the template parameter lists in the Declarator object
- Tracking the template parameter depth before parsing function declarators (at which point we can match template
parameters against scope specifiers to know if we have an explicit template parameter list to append invented parameters
to or not).
- When encountering an AutoType in a parameter context we check a stack of InventedTemplateParameterInfo structures that
contain the info required to create and accumulate invented template parameters (fields that were already present in
LambdaScopeInfo, which now inherits from this class and is looked up when an auto is encountered in a lambda context).
Resubmit after incorrect check in NonTypeTemplateParmDecl broke lldb.
Differential Revision: https://reviews.llvm.org/D65042
Add a simple cache for constraint satisfaction results. Whether or not this simple caching
would be permitted in final C++2a is currently being discussed but it is required for
acceptable performance so we use it in the meantime, with the possibility of adding some
cache invalidation mechanisms later.
Differential Revision: https://reviews.llvm.org/D72552
This patch implements P1141R2 "Yet another approach for constrained declarations".
General strategy for this patch was:
- Expand AutoType to include optional type-constraint, reflecting the wording and easing the integration of constraints.
- Replace autos in parameter type specifiers with invented parameters in GetTypeSpecTypeForDeclarator, using the same logic
previously used for generic lambdas, now unified with abbreviated templates, by:
- Tracking the template parameter lists in the Declarator object
- Tracking the template parameter depth before parsing function declarators (at which point we can match template
parameters against scope specifiers to know if we have an explicit template parameter list to append invented parameters
to or not).
- When encountering an AutoType in a parameter context we check a stack of InventedTemplateParameterInfo structures that
contain the info required to create and accumulate invented template parameters (fields that were already present in
LambdaScopeInfo, which now inherits from this class and is looked up when an auto is encountered in a lambda context).
Differential Revision: https://reviews.llvm.org/D65042
This patch broke the Sanitizer buildbots. Please see the commit's
differential revision for more information
(https://reviews.llvm.org/D67678).
This reverts commit b72a8c65e4.
Target regions have implicit outer region which may erroneously capture
some globals when it should not. It may lead to a compiler crash at the
compile time.
Summary:
The documentation for IndentCaseLabels claimed that the "Switch
statement body is always indented one level more than case labels". This
is technically false for the code block immediately following the label.
Its closing bracket aligns with the start of the label.
If the case label are not indented, it leads to a style where the
closing bracket of the block aligns with the closing bracket of the
switch statement, which can be hard to parse.
This change introduces a new option, IndentCaseBlocks, which when true
treats the block as a scope block (which it technically is).
(Note: regenerated ClangFormatStyleOptions.rst using tools/dump_style.py)
Reviewed By: MyDeveloperDay
Patch By: capn
Tags: #clang-format, #clang
Differential Revision: https://reviews.llvm.org/D72276
Implement support for C++2a requires-expressions.
Re-commit after compilation failure on some platforms due to alignment issues with PointerIntPair.
Differential Revision: https://reviews.llvm.org/D50360
Currently there are 4 different mechanisms for controlling denormal
flushing behavior, and about as many equivalent frontend controls.
- AMDGPU uses the fp32-denormals and fp64-f16-denormals subtarget features
- NVPTX uses the nvptx-f32ftz attribute
- ARM directly uses the denormal-fp-math attribute
- Other targets indirectly use denormal-fp-math in one DAGCombine
- cl-denorms-are-zero has a corresponding denorms-are-zero attribute
AMDGPU wants a distinct control for f32 flushing from f16/f64, and as
far as I can tell the same is true for NVPTX (based on the attribute
name).
Work on consolidating these into the denormal-fp-math attribute, and a
new type specific denormal-fp-math-f32 variant. Only ARM seems to
support the two different flush modes, so this is overkill for the
other use cases. Ideally we would error on the unsupported
positive-zero mode on other targets from somewhere.
Move the logic for selecting the flush mode into the compiler driver,
instead of handling it in cc1. denormal-fp-math/denormal-fp-math-f32
are now both cc1 flags, but denormal-fp-math-f32 is not yet exposed as
a user flag.
-cl-denorms-are-zero, -fcuda-flush-denormals-to-zero and
-fno-cuda-flush-denormals-to-zero will be mapped to
-fp-denormal-math-f32=ieee or preserve-sign rather than the old
attributes.
Stop emitting the denorms-are-zero attribute for the OpenCL flag. It
has no in-tree users. The meaning would also be target dependent, such
as the AMDGPU choice to treat this as only meaning allow flushing of
f32 and not f16 or f64. The naming is also potentially confusing,
since DAZ in other contexts refers to instructions implicitly treating
input denormals as zero, not necessarily flushing output denormals to
zero.
This also does not attempt to change the behavior for the current
attribute. The LangRef now states that the default is ieee behavior,
but this is inaccurate for the current implementation. The clang
handling is slightly hacky to avoid touching the existing
denormal-fp-math uses. Fixing this will be left for a future patch.
AMDGPU is still using the subtarget feature to control the denormal
mode, but the new attribute are now emitted. A future change will
switch this and remove the subtarget features.
A TemplateIdAnnotation represents only a template-id, not a
nested-name-specifier plus a template-id. Don't make a redundant copy of
the CXXScopeSpec and store it on the template-id annotation.
This slightly improves error recovery by more properly handling the case
where we would form an invalid CXXScopeSpec while parsing a typename
specifier, instead of accidentally putting the token stream into a
broken "annot_template_id with a scope specifier, but with no preceding
annot_cxxscope token" state.
Extend -fxray-instrumentation-bundle to split function-entry and
function-exit into two separate options, so that it is possible to
instrument only function entry or only function exit. For use cases
that only care about one or the other this will save significant overhead
and code size.
Differential Revision: https://reviews.llvm.org/D72890
XRay allows tuning by minimum function size, but also always instruments
functions with loops in them. If the minimum function size is set to a
large value the loop instrumention ends up causing most functions to be
instrumented anyway. This adds a new flag, -fxray-ignore-loops, to disable
the loop detection logic.
Differential Revision: https://reviews.llvm.org/D72873
Summary: Currently, an attempt to rewrite source code inside a macro expansion succeeds, but results in empty text, rather than failing with an error. This patch restructures to the code to explicitly validate ranges before attempting to edit them.
Reviewers: gribozavr
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D72274
Right now every dataflow algorithm uses its own worklist implementation.
This is a first step to reduce this duplication. Some upcoming
algorithms such as the lifetime analysis is going to use the factored
out implementations.
Differential Revision: https://reviews.llvm.org/D72380
Partially reverts 0a2be46cfd as it turned
out to cause redundant module rebuilds in multi-process incremental builds.
When a module was getting out of date, all compilation processes started at the
same time were marking it as `ToBuild`. So each process was building the same
module instead of checking if it was built by someone else and using that
result. In addition to the work duplication, contention on the same .pcm file
wasn't making builds faster.
Note that for a single-process build this change would cause redundant module
reads and validations. But reading a module is faster than building it and
multi-process builds are more common than single-process. So I'm willing to
make such a trade-off.
rdar://problem/54395127
Reviewed By: dexonsmith
Differential Revision: https://reviews.llvm.org/D72860
This is (more?) usable by GDB pretty printers and seems nicer to write.
There's one tricky caveat that in C++14 (LLVM's codebase today) the
static constexpr member declaration is not a definition - so odr use of
this constant requires an out of line definition, which won't be
provided (that'd make all these trait classes more annoyidng/expensive
to maintain). But the use of this constant in the library implementation
is/should always be in a non-odr context - only two unit tests needed to
be touched to cope with this/avoid odr using these constants.
Based on/expanded from D72590 by Christian Sigg.
This needs somewhat careful disambiguation, as C++2a explicit(bool) is a
breaking change. We only enable it in cases where the source construct
could not possibly be anything else.
Flags are clang's default UI is flags.
We can have an env var in addition to that, but in D69825 nobody has yet
mentioned why this needs an env var, so omit it for now. If someone
needs to set the flag via env var, the existing CCC_OVERRIDE_OPTIONS
mechanism works for it (set CCC_OVERRIDE_OPTIONS=+-fno-integrated-cc1
for example).
Also mention the cc1-in-process change in the release notes.
Also spruce up the test a bit so it actually tests something :)
Differential Revision: https://reviews.llvm.org/D72769
This is applied to the vector types defined in <arm_mve.h> for use
with the intrinsics for the ARM MVE vector architecture.
Its purpose is to inhibit lax vector conversions, but only in the
context of overload resolution of the MVE polymorphic intrinsic
functions. This solves an ambiguity problem with polymorphic MVE
intrinsics that take a vector and a scalar argument: the scalar
argument can often have the wrong integer type due to default integer
promotions or unsuffixed literals, and therefore, the type of the
vector argument should be considered trustworthy when resolving MVE
polymorphism.
As part of the same change, I've added the new attribute to the
declarations generated by the MveEmitter Tablegen backend (and
corrected a namespace issue with the other attribute while I was
there).
Reviewers: aaron.ballman, dmgreen
Reviewed By: aaron.ballman
Subscribers: kristof.beyls, JDevlieghere, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D72518
These driver options perform some checking and delegate to MC options -x86-align-branch* and -x86-branches-within-32B-boundaries.
Reviewed By: skan
Differential Revision: https://reviews.llvm.org/D72463
Add support for type-constraints in template type parameters.
Also add support for template type parameters as pack expansions (where the type constraint can now contain an unexpanded parameter pack).
Differential Revision: https://reviews.llvm.org/D44352
Summary:
Before this change, X86_32ABIInfo::classifyArgument would be called
twice on vector arguments to vectorcall functions. This function has
side effects to track GPR register usage, and this would lead to
incorrect GPR usage in some cases. The specific case I noticed is from
running out of XMM registers with mixed FP and vector arguments and no
aggregates of any kind. Consider this prototype:
void __vectorcall vectorcall_indirect_vec(
double xmm0, double xmm1, double xmm2, double xmm3, double xmm4,
__m128 xmm5,
__m128 ecx,
int edx,
__m128 mem);
classifyArgument has no effects when called on a plain FP type, but when
called on a vector type, it modifies FreeRegs to model GPR consumption.
However, this should not happen during the vector call first pass.
I refactored the code to unify vectorcall HVA logic with regcall HVA
logic. The conventions pass HVAs in registers differently (expanded vs.
not expanded), but if they do not fit in registers, they both pass them
indirectly by address.
Reviewers: erichkeane, craig.topper
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D72110
Add checks for some structural invariants when building and mutating
the syntax trees.
Fix a bug failing the invariants after mutations: the parent of nodes
added into the tree was null.
Summary:
This cleans up a lot of ugly `foreach` bodges that I've been using to
work around the lack of those two language features. Now they both
exist, I can make then all into something more legible!
In particular, in the common pattern in `ARMInstrMVE.td` where a
multiclass defines an `Instruction` instance plus one or more `Pat` that
select it, I've used a `defvar` to wrap `!cast<Instruction>(NAME)` so
that the patterns themselves become a little more legible.
Replacing a `foreach` with a `defvar` removes a level of block
structure, so several pieces of code have their indentation changed by
this patch. Best viewed with whitespace ignored.
NFC: the output of `llvm-tblgen -print-records` on the two affected
Tablegen sources is exactly identical before and after this change, so
there should be no effect at all on any of the other generated files.
Reviewers: MarkMurrayARM, miyuki
Reviewed By: MarkMurrayARM
Subscribers: kristof.beyls, hiraditya, dmgreen, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D72690
The option will limit debug info by only emitting complete class
type information when its constructor is emitted.
This patch changes comparisons with LimitedDebugInfo to use the new
level instead.
Differential Revision: https://reviews.llvm.org/D72427
GCC supports the conditional operator on VectorTypes that acts as a
'select' in C++ mode. This patch implements the support. Types are
converted as closely to GCC's behavior as possible, though in a few
places consistency with our existing vector type support was preferred.
Note that this implementation is different from the OpenCL version in a
number of ways, so it unfortunately required a different implementation.
First, the SEMA rules and promotion rules are significantly different.
Secondly, GCC implements COND[i] != 0 ? LHS[i] : RHS[i] (where i is in
the range 0- VectorSize, for each element). In OpenCL, the condition is
COND[i] < 0 ? LHS[i]: RHS[i].
In the process of implementing this, it was also required to make the
expression COND ? LHS : RHS type dependent if COND is type dependent,
since the type is now dependent on the condition. For example:
T ? 1 : 2;
Is not typically type dependent, since the result can be deduced from
the operands. HOWEVER, if T is a VectorType now, it could change this
to a 'select' (basically a swizzle with a non-constant mask) with the 1
and 2 being promoted to vectors themselves.
While this is a change, it is NOT a standards incompatible change. Based
on my (and D. Gregor's, at the time of writing the code) reading of the
standard, the expression is supposed to be type dependent if ANY
sub-expression is type dependent.
Differential Revision: https://reviews.llvm.org/D71463
With this patch, the clang tool will now call the -cc1 invocation directly inside the same process. Previously, the -cc1 invocation was creating, and waiting for, a new process.
This patch therefore reduces the number of created processes during a build, thus it reduces build times on platforms where process creation can be costly (Windows) and/or impacted by a antivirus.
It also makes debugging a bit easier, as there's no need to attach to the secondary -cc1 process anymore, breakpoints will be hit inside the same process.
Crashes or signaling inside the -cc1 invocation will have the same side-effect as before, and will be reported through the same means.
This behavior can be controlled at compile-time through the CLANG_SPAWN_CC1 cmake flag, which defaults to OFF. Setting it to ON will revert to the previous behavior, where any -cc1 invocation will create/fork a secondary process.
At run-time, it is also possible to tweak the CLANG_SPAWN_CC1 environment variable. Setting it and will override the compile-time setting. A value of 0 calls -cc1 inside the calling process; a value of 1 will create a secondary process, as before.
Differential Revision: https://reviews.llvm.org/D69825
which is the default TLS model for non-PIC objects. This allows large/
many thread local variables or a compact/fast code in an executable.
Specification is same as that of GCC. For example, the code model
option precedes the TLS size option.
TLS access models other than local-exec are not changed. It means
supoort of the large code model is only in the local exec TLS model.
Patch By KAWASHIMA Takahiro (kawashima-fj <t-kawashima@fujitsu.com>)
Reviewers: dmgreen, mstorsjo, t.p.northover, peter.smith, ostannard
Reviewd By: peter.smith
Committed by: peter.smith
Differential Revision: https://reviews.llvm.org/D71688
Summary:
The analysis for const-ness of local variables required a view generally useful
matchers that are extracted into its own patch.
They are `decompositionDecl` and `forEachArgumentWithParamType`, that works
for calls through function pointers as well.
Reviewers: aaron.ballman
Reviewed By: aaron.ballman
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D72505
All 130+ f_Group flags that take an argument allow it after a '=',
except for fdebug-complation-dir. Add a Joined<> alias so that
it behaves consistently with all the other f_Group flags.
(Keep the old Separate flag for backwards compat.)
In the backend, this feature is implemented with the function attribute
"patchable-function-entry". Both the attribute and XRay use
TargetOpcode::PATCHABLE_FUNCTION_ENTER, so the two features are
incompatible.
Reviewed By: ostannard, MaskRay
Differential Revision: https://reviews.llvm.org/D72222
This feature is generic. Make it applicable for AArch64 and X86 because
the backend has only implemented NOP insertion for AArch64 and X86.
Reviewed By: nickdesaulniers, aaron.ballman
Differential Revision: https://reviews.llvm.org/D72221
Summary:
This checker verifies if default placement new is provided with pointers
to sufficient storage capacity.
Noncompliant Code Example:
#include <new>
void f() {
short s;
long *lp = ::new (&s) long;
}
Based on SEI CERT rule MEM54-CPP
https://wiki.sei.cmu.edu/confluence/display/cplusplus/MEM54-CPP.+Provide+placement+new+with+properly+aligned+pointe
This patch does not implement checking of the alignment.
Reviewers: NoQ, xazax.hun
Subscribers: mgorny, whisperity, xazax.hun, baloghadamsoftware, szepet,
rnkovacs, a.sidorin, mikhail.ramalho, donat
Tags: #clang
Differential Revision: https://reviews.llvm.org/D71612
Summary:
Avoid using the `nocf_check` attribute with Control Flow Guard. Instead, use a
new `"guard_nocf"` function attribute to indicate that checks should not be
added on indirect calls within that function. Add support for
`__declspec(guard(nocf))` following the same syntax as MSVC.
Reviewers: rnk, dmajor, pcc, hans, aaron.ballman
Reviewed By: aaron.ballman
Subscribers: aaron.ballman, tomrittervg, hiraditya, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D72167
Summary:
A copy-paste error in `arm_mve.td` meant that the MVE `vqrshrun`
intrinsic family was generating the `vqshrun` machine instruction,
because in the IR intrinsic call, the rounding flag argument was set
to 0 rather than 1.
Reviewers: dmgreen, MarkMurrayARM, miyuki, ostannard
Reviewed By: dmgreen
Subscribers: kristof.beyls, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D72496
If a system header provides an (inline) implementation of some of their
function, clang still matches on the function name and generate the appropriate
llvm builtin, e.g. memcpy. This behavior is in line with glibc recommendation «
users may not provide their own version of symbols » but doesn't account for the
fact that glibc itself can provide inline version of some functions.
It is the case for the memcpy function when -D_FORTIFY_SOURCE=1 is on. In that
case an inline version of memcpy calls __memcpy_chk, a function that performs
extra runtime checks. Clang currently ignores the inline version and thus
provides no runtime check.
This code fixes the issue by detecting functions whose name is a builtin name
but also have an inline implementation.
Differential Revision: https://reviews.llvm.org/D71082
The language wording change forgot to update overload resolution to rank
implicit conversion sequences based on qualification conversions in
reference bindings. The anticipated resolution for that oversight is
implemented here -- we order candidates based on qualification
conversion, not only on top-level cv-qualifiers, including ranking
reference bindings against non-reference bindings if they differ in
non-top-level qualification conversions.
For OpenCL/C++, this allows reference binding between pointers with
differing (nested) address spaces. This makes the behavior of reference
binding consistent with that of implicit pointer conversions, as is the
purpose of this change, but that pre-existing behavior for pointer
conversions is itself probably not correct. In any case, it's now
consistently the same behavior and implemented in only one place.
This reinstates commit de21704ba9,
reverted in commit d8018233d1, with
workarounds for some overload resolution ordering problems introduced by
CWG2352.
explicit functions that are not candidates.
It's not always obvious that the reason a conversion was not possible is
because the function you wanted to call is 'explicit', so explicitly say
if that's the case.
It would be nice to rank the explicit candidates higher in the
diagnostic if an implicit conversion sequence exists for their
arguments, but unfortunately we can't determine that without potentially
triggering non-immediate-context errors that we're not permitted to
produce.
This change introduces three new builtins (which work on both pointers
and integers) that can be used instead of common bitwise arithmetic:
__builtin_align_up(x, alignment), __builtin_align_down(x, alignment) and
__builtin_is_aligned(x, alignment).
I originally added these builtins to the CHERI fork of LLVM a few years ago
to handle the slightly different C semantics that we use for CHERI [1].
Until recently these builtins (or sequences of other builtins) were
required to generate correct code. I have since made changes to the default
C semantics so that they are no longer strictly necessary (but using them
does generate slightly more efficient code). However, based on our experience
using them in various projects over the past few years, I believe that adding
these builtins to clang would be useful.
These builtins have the following benefit over bit-manipulation and casts
via uintptr_t:
- The named builtins clearly convey the semantics of the operation. While
checking alignment using __builtin_is_aligned(x, 16) versus
((x & 15) == 0) is probably not a huge win in readably, I personally find
__builtin_align_up(x, N) a lot easier to read than (x+(N-1))&~(N-1).
- They preserve the type of the argument (including const qualifiers). When
using casts via uintptr_t, it is easy to cast to the wrong type or strip
qualifiers such as const.
- If the alignment argument is a constant value, clang can check that it is
a power-of-two and within the range of the type. Since the semantics of
these builtins is well defined compared to arbitrary bit-manipulation,
it is possible to add a UBSAN checker that the run-time value is a valid
power-of-two. I intend to add this as a follow-up to this change.
- The builtins avoids int-to-pointer casts both in C and LLVM IR.
In the future (i.e. once most optimizations handle it), we could use the new
llvm.ptrmask intrinsic to avoid the ptrtoint instruction that would normally
be generated.
- They can be used to round up/down to the next aligned value for both
integers and pointers without requiring two separate macros.
- In many projects the alignment operations are already wrapped in macros (e.g.
roundup2 and rounddown2 in FreeBSD), so by replacing the macro implementation
with a builtin call, we get improved diagnostics for many call-sites while
only having to change a few lines.
- Finally, the builtins also emit assume_aligned metadata when used on pointers.
This can improve code generation compared to the uintptr_t casts.
[1] In our CHERI compiler we have compilation mode where all pointers are
implemented as capabilities (essentially unforgeable 128-bit fat pointers).
In our original model, casts from uintptr_t (which is a 128-bit capability)
to an integer value returned the "offset" of the capability (i.e. the
difference between the virtual address and the base of the allocation).
This causes problems for cases such as checking the alignment: for example, the
expression `if ((uintptr_t)ptr & 63) == 0` is generally used to check if the
pointer is aligned to a multiple of 64 bytes. The problem with offsets is that
any pointer to the beginning of an allocation will have an offset of zero, so
this check always succeeds in that case (even if the address is not correctly
aligned). The same issues also exist when aligning up or down. Using the
alignment builtins ensures that the address is used instead of the offset. While
I have since changed the default C semantics to return the address instead of
the offset when casting, this offset compilation mode can still be used by
passing a command-line flag.
Reviewers: rsmith, aaron.ballman, theraven, fhahn, lebedev.ri, nlopes, aqjune
Reviewed By: aaron.ballman, lebedev.ri
Differential Revision: https://reviews.llvm.org/D71499
In common with most MVE immediate shift instructions, the left shift
takes an immediate in the range [0,n-1], while the right shift takes
one in the range [1,n]. I had absent-mindedly made them both the
latter.
While I'm here, I've added a set of regression tests checking both
ends of the immediate range for a representative sample of the
immediate shifts.
Function trailing requires clauses now parsed, supported in overload resolution and when calling, referencing and taking the address of functions or function templates.
Differential Revision: https://reviews.llvm.org/D43357
This matcher matches any node and at the same time executes all its
inner matchers to produce any possbile result bindings.
This is useful when a user wants certain supplementary information
that's not always present along with the main match result.
This batch of intrinsics fills in all the shift instructions that take
a variable shift distance in a register, instead of an immediate. Some
of these instructions take a single shift distance in a scalar
register and apply it to all lanes; others take a vector of per-lane
distances.
These instructions are all basically one family, varying in whether
they saturate out-of-range values, and whether they round when bits
are shifted off the bottom. I've implemented them at the IR level by a
much smaller family of IR intrinsics, which take flag parameters to
indicate saturating and/or rounding (along with the usual one to
specify signed/unsigned integers).
An oddity is that all of them are //left// shift instructions – but if
you pass a negative shift count, they'll shift right. So the vector
shift distances are always vectors of //signed// integers, regardless
of whether you're considering the other input vector to be of signed
or unsigned. Also, even the simplest `vshlq` instruction in this
family (neither saturating nor rounding) has to be implemented as an
IR intrinsic, because the ordinary LLVM IR `shl` operation would
consider an out-of-range shift count to be undefined behavior.
Reviewers: dmgreen, MarkMurrayARM, miyuki, ostannard
Reviewed By: dmgreen
Subscribers: kristof.beyls, hiraditya, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D72329
This batch of intrinsics covers two sets of immediate shift
instructions, which have in common that they only overwrite part of
their output register and so they need an extra input giving its
previous value.
The VSLI and VSRI instructions shift each lane of the input vector
left or right just as if they were normal immediate VSHL/VSHR, but
then they only overwrite the output bits that correspond to actual
shifted bits of the input. So VSLI will leave the low n bits of each
output lane unchanged, and VSRI the same with the top n bits.
The V[Q][R]SHR[U]N family are all narrowing shifts: they take an input
vector of 2n-bit integers, shift each lane right by a constant, and
then narrowing the shifted result to only n bits. So they only
overwrite half of the n-bit lanes in the output register, and the B/T
suffix indicates whether it's the bottom or top half of each 2n-bit
lane.
I've implemented the whole of the latter family using a single IR
intrinsic `vshrn`, which takes a lot of i32 parameters indicating
which instruction it expands to (by specifying signedness of the input
and output types, whether it saturates and/or rounds, etc).
Reviewers: dmgreen, MarkMurrayARM, miyuki, ostannard
Reviewed By: dmgreen
Subscribers: kristof.beyls, hiraditya, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D72328
An empty string literal in an asm label does not make a whole lot of sense. GCC
does not diagnose such a construct, but it also generates code that cannot be
assembled by gas should two symbols have an empty asm label within the same TU.
This does not affect an asm statement with an empty string literal, which is
still a useful construct.
Summary:
It now handles `typedef`s that include comma-separated multiple types, and handles embedded struct definitions, which previously could not be automatically converted.
For example, with this patch `modernize-use-using` now can convert:
typedef struct { int a; } R_t, *R_p;
to:
using R_t = struct { int a; };
using R_p = R_t*;
`-ast-dump` showed that the `CXXRecordDecl` definitions and multiple `TypedefDecl`s come consecutively in the tree, so `check()` stores information between calls to determine when it is receiving a second or additional `TypedefDecl` within a single `typedef`, or when the current `TypedefDecl` refers to an embedded `CXXRecordDecl` like a `struct`.
Reviewers: alexfh, aaron.ballman
Patch by: poelmanc
Subscribers: riccibruno, sammccall, cfe-commits, aaron.ballman
Tags: clang-tools-extra, clang
Differential Revision: https://reviews.llvm.org/D70270
If standalone OpenMP declaration pragma, like declare mapper or declare
reduction, is declared in the class context, it may reference a member
(data or function) in its internal expressions/statements. So, the
parsing of such pragmas must be dalayed just like the parsing of the
member initializers/definitions before the completion of the class
declaration.
It turns out it is useful to be able to define the deref type as void.
In case we have a type erased owner, we want to express that the pointee
can be basically any type. It should not be unnatural to have a void
deref type as we already familiar with "pointers to void".
Differential Revision: https://reviews.llvm.org/D72097
pack expansion.
Previously, if all parameter / argument pairs for a pack expansion
deduction were non-deduced contexts, we would not deduce the arity of
the pack, and could end up deducing a different arity (leading to
failures during substitution) or defaulting to an arity of 0 (leading to
bad diagnostics about passing the wrong number of arguments to a
variadic function). Instead, we now always deduce the arity for all
involved packs any time we deduce a pack expansion.
This will result in less substitution happening in some cases, which
could avoid non-SFINAEable errors, and should generally improve the
quality of diagnostics when passing initializer lists to variadic
functions.
This makes the range loop warnings part of -Wall.
Fixes PR32823: Warn about accidental coping of data in range based for
Differential Revision: https://reviews.llvm.org/D68912
Recomitted after fixing the warnings it created.
Summary:
Running an end-to-end test last week I noticed that a lot of the ACLE
intrinsics that operate differently on vectors of signed and unsigned
integers were ending up generating the signed version of the
instruction unconditionally. This is because the IR intrinsics had no
way to distinguish signed from unsigned: the LLVM type system just
calls them both `v8i16` (or whatever), so you need either separate
intrinsics for signed and unsigned, or a flag parameter that tells
ISel which one to choose.
This patch fixes all the problems of that kind that I've noticed, by
adding an i32 flag parameter to many of the IR intrinsics which is set
to 1 for unsigned (matching the existing practice in cases where we
got it right), and conditioning all the isel patterns on that flag. So
the fundamental change is in `IntrinsicsARM.td`, changing the
low-level IR intrinsics API; there are knock-on changes in
`arm_mve.td` (adjusting code gen for the ACLE intrinsics to use the
modified API) and in `ARMInstrMVE.td` (adjusting isel to expect the
new unsigned flags). The rest of this patch is boringly updating tests.
Reviewers: dmgreen, miyuki, MarkMurrayARM
Reviewed By: dmgreen
Subscribers: kristof.beyls, hiraditya, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D72270
Summary:
The ACLE intrinsics with `gather_base` or `scatter_base` in the name
are wrappers on the MVE load/store instructions that take a vector of
base addresses and an immediate offset. The immediate offset can be up
to 127 times the alignment unit, and it can be positive or negative.
At the MC layer, we got that right. But in the Sema error checking for
the wrapping intrinsics, the offset was erroneously constrained to be
positive.
To fix this I've adjusted the `imm_mem7bit` class in the Tablegen that
defines the intrinsics. But that causes integer literals like
`0xfffffffffffffe04` to appear in the autogenerated calls to
`SemaBuiltinConstantArgRange`, which provokes a compiler warning
because that's out of the non-overflowing range of an `int64_t`. So
I've also tweaked `MveEmitter` to emit that as `-0x1fc` instead.
Updated the tests of the Sema checks themselves, and also adjusted a
random sample of the CodeGen tests to actually use negative offsets
and prove they get all the way through code generation without causing
a crash.
Reviewers: dmgreen, miyuki, MarkMurrayARM
Reviewed By: dmgreen
Subscribers: kristof.beyls, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D72268
-mpacked-stack is currently not supported with -mbackchain, so this should
result in a compilation error message instead of being silently ignored.
Review: Ulrich Weigand
The OpenMP specification disallows having zero-length array
sections in the depend clause (OpenMP 5.0 2.17.11).
Differential Revision: https://reviews.llvm.org/D71969
Adds a new ASTMatcher condition called 'hasInitStatement()' that matches if,
switch and range-for statements with an initializer. Reworked clang-tidy
readability-else-after-return to handle variables in the if condition or init
statements in c++17 ifs. Also checks if removing the else would affect object
lifetimes in the else branch.
Fixes PR44364.
This makes the range loop warnings part of -Wall.
Fixes PR32823: Warn about accidental coping of data in range based for
Differential Revision: https://reviews.llvm.org/D68912
Summary:
These overloads make it possible to wrap unless(), anyOf(), has() etc
with the traverse matcher.
Reviewers: aaron.ballman
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D71977
This reverts commit de21704ba9.
Regressed/causes this to error due to ambiguity:
void f(const int * const &);
void f(int *);
int main() {
int * x;
f(x);
}
(in case it's important - the original case where this turned up was a
member function overload in a class template with, essentially:
f(const T1&)
f(T2*)
(where T1 == X const *, T2 == X))
It's not super clear to me if this ^ is expected behavior, in which case
I'm sorry about the revert & happy to look into ways to fix the original
code.
Revert "Fix -Wunused-lambda-capture warnings."
This reverts commit 2369560f4a.
This reverts commit 522ee29a4f.
clang/lib/ASTMatchers/Dynamic/Parser.cpp:610:13: warning: implicit conversion turns string literal into bool: 'const char [35]' to 'bool' [-Wstring-conversion]
assert(!"Newline should never be found here");
This removes the OpenMPProcBindClauseKind enum in favor of
llvm::omp::ProcBindKind which lives in OpenMPConstants.h and was
introduced in D70109.
No change in behavior is expected.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D70289
The function and its called static helpers don't modify the received
CXXRecordDecl arguments at all as the method's result is put into an
output parameter. Thus they can be const which allows for neatly
grabbing the conversion methods in a context where we only have a const
ASTUnit at hand.
Differential Revision: https://reviews.llvm.org/D71805
The validateOutputSize and validateInputSize need to check whether
AVX or AVX512 are enabled. But this can be affected by the
target attribute so we need to factor that in.
This patch moves some of the code from CodeGen to create an
appropriate feature map that we can pass to the function.
Differential Revision: https://reviews.llvm.org/D68627
Summary:
Basic codegen for the declarations marked as nontemporal. Also, if the
base declaration in the member expression is marked as nontemporal,
lvalue for member decl access inherits nonteporal flag from the base
lvalue.
Reviewers: rjmccall, hfinkel, jdoerfert
Subscribers: guansong, arphaman, caomhin, kkwli0, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D71708
Added support for constraint satisfaction checking and partial ordering of constraints in constrained partial specialization and function template overloads.
Re-commit after fixing another crash (added regression test).
Differential Revision: https://reviews.llvm.org/D41910
Commit d77ae1552f
("[DebugInfo] Support to emit debugInfo for extern variables")
added deebugInfo for extern variables for BPF target.
The commit is reverted by 891e25b02d
as the committed tests using %clang instead of %clang_cc1 causing
test failed in certain scenarios as reported by Reid Kleckner.
This patch fixed the tests by using %clang_cc1.
Differential Revision: https://reviews.llvm.org/D71818
NFCs factored out of the following patches:
- Change all of the `Expr *` to `const Expr *` in SequenceChecker for
const-correctness. SequenceChecker should not modify AST nodes.
- Add some comments.
- clang-format
Differential Revision: https://reviews.llvm.org/D57659
Reviewed By: xbolva00
Added support for constraint satisfaction checking and partial ordering of constraints in constrained partial specialization and function template overloads.
Re-commit after fixing some crashes and warnings.
Differential Revision: https://reviews.llvm.org/D41910
getLastArgIntValue is a useful utility function to get command line argument as an integer.
Currently it is in Frontend so that it can only be used by clang -cc1. Move it to basic so
that it can also be used by clang driver.
Differential Revision: https://reviews.llvm.org/D71080
These annotations will be used in an upcomming static analyzer check
that finds handle leaks, use after releases, and double releases.
Differential Revision: https://reviews.llvm.org/D70469
Because the name of a direct method must be agreed upon by the caller
and the implementation, certain bad practices that one can get away with
when using dynamism are fatal with direct methods.
To avoid really weird and unscruttable linker error, tighten the
front-end error reporting.
Rule 1:
Direct methods can only have at most one declaration in an @interface
container. Any redeclaration is strictly forbidden.
Today some amount of redeclaration is tolerated between the main
interface and categories for dynamic methods, but we can't have that.
Rule 2:
Direct method implementations can only be declared in a matching
@interface container: when implemented in the primary @implementation
then the declaration must be in the primary @interface or an
extension, and when implemented in a category, the declaration must be
in the @interface for the same category.
Also fix another issue with ObjCMethod::getCanonicalDecl(): when an
implementation lives in the primary @interface, then its canonical
declaration can be in any extension, even when it's not an accessor.
Add Sema tests to cover the new errors, and CG tests to beef up testing
around function names for categories and extensions.
Radar-Id: <rdar://problem/58054563>
Differential Revision: https://reviews.llvm.org/D71694
In the current implementation of clang the canonicalization of paths in
diagnostic messages (when using -fdiagnostics-absolute-paths) only works
if the symbolic link is in the directory part of the filename, not if
the file itself is a symbolic link to another file.
This patch adds support to canonicalize the complete path including the
file.
Reviewers: rsmith, hans, rnk, ikudrin
Reviewed By: rnk
Differential Revision: https://reviews.llvm.org/D70527
The language wording change forgot to update overload resolution to rank
implicit conversion sequences based on qualification conversions in
reference bindings. The anticipated resolution for that oversight is
implemented here -- we order candidates based on qualification
conversion, not only on top-level cv-qualifiers.
For OpenCL/C++, this allows reference binding between pointers with
differing (nested) address spaces. This makes the behavior of reference
binding consistent with that of implicit pointer conversions, as is the
purpose of this change, but that pre-existing behavior for pointer
conversions is itself probably not correct. In any case, it's now
consistently the same behavior and implemented in only one place.
Method '-[NSCoder decodeValueOfObjCType:at:]' is not only deprecated
but also a security hazard, hence a loud check.
Differential Revision: https://reviews.llvm.org/D71728
Enabling `-Wcast-qual` identified many casts in various system headers
that were dropping the `const` qualifier. Fixing those missing
qualifiers pointed out that a few of the definitions of the builtins
did not properly identify their arguments as `const` pointers. This
commit fixes those builtin definitions, and the system header files
so that they no longer drop the qualifier.
Differential Revision: https://reviews.llvm.org/D71718
Recognize -mrecord-mcount from the command line and add a function attribute
"mrecord-mcount" when passed.
Only valid on SystemZ (when used with -mfentry).
Review: Ulrich Weigand
https://reviews.llvm.org/D71627
GCC 7 and earlier, when targeting MinGW, seems to have a bug in
layout/size of bitfield structs if they contain a nested enum,
making the size of the struct 8 bytes, while we have a static assert
requiring it to be 4 bytes or less.
While this clearly is a GCC bug, the workaround (moving the enum out
of the bitfield) also is very nonintrusive and matches other existing
enums there.
Differential Revision: https://reviews.llvm.org/D71650
Summary: Per D62731, the behavior of clang with `-frounding-math` is no worse than when the rounding flag was completely ignored, so remove this unnecessary warning.
Reviewers: mibintc, chandlerc, echristo, rjmccall, kpn, erichkeane, rsmith, andrew.w.kaylor
Reviewed By: mibintc
Subscribers: merge_guards_bot, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D71671
Summary:
The vector pattern `(a + b + 1) / 2` was previously selected to an
avgr_u instruction regardless of nuw flags, but this is incorrect in
the case where either addition may have an unsigned wrap. This CL
changes the existing pattern to require both adds to have nuw flags
and adds builtin functions and intrinsics for the avgr_u instructions
because the corrected pattern is not representable in C.
Reviewers: aheejin
Subscribers: dschuff, sbc100, jgravelle-google, hiraditya, sunfish, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D71648
implementing the resolution of CWG2352.
No functionality change, except that we now convert the referent of a
reference binding to the underlying type of the reference in more cases;
we used to happen to preserve the type sugar from the referent if the
only type change was in the cv-qualifiers.
This exposed a bug in how we generate code for trivial assignment
operators: if the type sugar (particularly the may_alias attribute)
got lost during reference binding, we'd use the "wrong" TBAA information
for the load during the assignment.
TypeSourceInfo is a thin wrapper around TypeLocs. Notionally, the best
place for it to live would be TypeLoc.h, but Decl.h requires it to be
complete, so it needs to be lower in the dependency graph. Type.h seems
like the next best place.
By itself, this change has no impact on build time, because it doesn't
remove a single dependency edge from a .cpp file to a .h file, but it is
an incremental step towards making the AST headers less interdependent.
Reviewers: rsmith
Differential Revision: https://reviews.llvm.org/D71427
Michael Spencer