LGTM
On Wed, Sep 28, 2016 at 10:09 PM, Eric Fiselier wrote:
> EricWF created this revision.
> EricWF added reviewers: rsmith, GorNishanov.
> EricWF added a subscriber: cfe-commits.
> Herald added a subscriber: mehdi_amini.
>
> If there is an error finding the `coroutine_traits` template when bui
Author: rsmith
Date: Thu Sep 29 16:30:12 2016
New Revision: 282777
URL: http://llvm.org/viewvc/llvm-project?rev=282777&view=rev
Log:
Switch to a different workaround for unimplementability of P0145R3 in MS ABIs.
Instead of ignoring the evaluation order rule, ignore the "destroy parameters
in rever
Author: rsmith
Date: Thu Sep 29 17:49:46 2016
New Revision: 282800
URL: http://llvm.org/viewvc/llvm-project?rev=282800&view=rev
Log:
P0035R4: add std::align_val_t overloads of operator new/delete in C++17 mode.
Added:
cfe/trunk/test/PCH/cxx1z-aligned-alloc.cpp
Modified:
cfe/trunk/include/
+def fcoroutines : Flag <["-"], "fcoroutines-ts">, Group,
+ Flags<[DriverOption, CC1Option]>,
+ HelpText<"Enable support for the C++ Coroutines TS">;
+def fno_coroutines : Flag <["-"], "fno-coroutines-ts">, Group,
These should be named fcoroutines_ts, fno_coroutines_ts (see comment at the
top of
rsmith added a comment.
LGTM, go ahead when Gor is happy.
https://reviews.llvm.org/D25078
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
rsmith accepted this revision.
rsmith added a comment.
This revision is now accepted and ready to land.
Minor comment, otherwise LGTM.
> DeclPrinter.cpp:1350-1351
> +
> + // PR 28885: Use the correct record name when the using declaration is used
> + // for inheriting constructors.
> + for (
Author: rsmith
Date: Fri Sep 30 17:41:36 2016
New Revision: 282974
URL: http://llvm.org/viewvc/llvm-project?rev=282974&view=rev
Log:
P0035R4: add predefined __STDCPP_DEFAULT_NEW_ALIGNMENT__ macro. By default, we
assume that ::operator new provides no more alignment than is necessary for any
primit
Author: rsmith
Date: Fri Sep 30 18:16:08 2016
New Revision: 282981
URL: http://llvm.org/viewvc/llvm-project?rev=282981&view=rev
Log:
Fix bogus "inline namespace cannot be reopened as non-inline" diagnostic to
just warn that the second declaration is missing the 'inline' keyword. This is
valid, and
rsmith added inline comments.
> SemaDeclCXX.cpp:8157-8160
> + CurContext->getRedeclContext()->isStdNamespace()) {
> + PrevNS = getStdExperimentalNamespace();
> + IsStdExperimental = true;
> + AddToKnown = !IsInline;
If you're going to track the namespace on creation,
Author: rsmith
Date: Fri Sep 30 18:47:58 2016
New Revision: 282984
URL: http://llvm.org/viewvc/llvm-project?rev=282984&view=rev
Log:
Fix tests to not assume they know exactly what will be in Clang's predefines
buffer.
Modified:
clang-tools-extra/trunk/test/pp-trace/pp-trace-conditional.cpp
rsmith accepted this revision.
rsmith added a comment.
This revision is now accepted and ready to land.
One question, but otherwise this looks fine.
> coroutines.m:1
> +// RUN: %clang -### %s 2>&1 | FileCheck -check-prefix=CHECK-NO-CORO %s
> +// RUN: %clang -fcoroutines-ts -### %s 2>&1 | FileCh
rsmith accepted this revision.
rsmith added inline comments.
> Sema.h:720-723
> + /// \brief The C++ "std::experimental" namespace, where the experimental
> parts
> + /// of the standard library resides.
> + NamespaceDecl *StdExperimentalNamespace;
> +
This is in the middle of a block of imp
rsmith added inline comments.
> LanguageExtensions.rst:1886
> + bool __builtin_coro_done(void *addr);
> + void __builtin_coro_promise(void *addr, int alignment, bool from_promise)
> +
Return type here is `void*`, not `void`.
> LanguageExtensions.rst:1916
> +
> +Other coroutine builtins are ei
rsmith added a comment.
I'm concerned about the complexity of this approach; it's hard for me to be
confident that `BuildScopeInformation` is correct and will remain correct in
the presence of future changes to the AST, and if it's not, we'll potentially
silently miscompile.
A simpler but less
rsmith added a comment.
OK, so it seems like all the other approaches we discussed have problems too.
Let's move forward with this for now.
Is there some reasonable base set of functionality between this and
JumpDiagnostics that could be factored out and shared?
> VarBypassDetector.cpp:45
>
rsmith accepted this revision.
rsmith added a comment.
This revision is now accepted and ready to land.
A couple of comment nits, then this LGTM. Thanks!
> SemaDecl.cpp:8659-8664
> + // void func();
> + // template class C1 { friend void func() { } };
> + // template class
rsmith added inline comments.
> SemaOverload.cpp:11750
> ExprResult
> Sema::CreateOverloadedBinOp(SourceLocation OpLoc,
> BinaryOperatorKind Opc,
We should never be calling this function outside of C++ mode.
It looks like the bug is in `Sema::BuildBinOp` -- in C,
rsmith added inline comments.
> SemaChecking.cpp:11053-11056
> + if ((RequiredAlignment > AlignRecord) ||
> + ((Context.toCharUnitsFromBits(
> +Context.getFieldOffset(cast(MD))) %
> +RequiredAlignment) != 0)) {
This doesn't seem to correctly handle the c
rsmith added inline comments.
> PrintPreprocessedOutput.cpp:320-321
> + * if we're emitting a '#'-style directive (escape both CR and LF). Paths
> are quoted in some
> + * instances, so escape quotes. Escaping is with a backslash (and
> backslashes themselves have
> + * to be escaped); for CR
rsmith added a comment.
There are a bunch of cases here that do this:
if (auto t = getThing())
ID.addThing(t);
if (auto t = getOtherThing())
ID.addThing(t);
That will result in hash collisions between objects with thing and objects with
otherthing (for instance, `struct A { int n :
rsmith added inline comments.
> SemaTemplateInstantiateDecl.cpp:4849
> + DeclarationName Name = D->getDeclName();
> + if (auto *DD = dyn_cast(D))
> +Name =
Do we need to do this for conversion function names too? (Eg, `operator C1*`)
https://reviews.llvm.org/D24969
rsmith added inline comments.
> PrintPreprocessedOutput.cpp:328
> + case '"': // paths are enclosed in quotes; escape them
> + case '*': // don't allow a "*/" sequence to accidentally open the
> comment
> + case '\\': // escape the escape character itself.
Putting a \ before a
On Mon, Oct 3, 2016 at 2:59 PM, Peter Collingbourne wrote:
> pcc added a reviewer: rsmith.
> pcc added a comment.
>
> It seems to me that this sanitizer would break the semantics of otherwise
> well-defined programs. For example:
>
> int *x = nullptr;
> delete x;
> if (x != nullptr) {
>
rsmith added a comment.
This looks like it's going in the right direction.
> Decl.cpp:2269-2272
> + // If we have hit a point where the user provided a specialization of
> + // this template, we're done looking.
> + if (VarTemplate->isMemberSpecialization())
> +break;
I
Author: rsmith
Date: Wed Oct 5 17:41:02 2016
New Revision: 283406
URL: http://llvm.org/viewvc/llvm-project?rev=283406&view=rev
Log:
PR22924, PR22845, some of CWG1464: When checking the initializer for an array
new expression, distinguish between the case of a constant and non-constant
initializer
LGTM
On 6 Oct 2016 8:58 am, "David Sheinkman via cfe-commits" <
cfe-commits@lists.llvm.org> wrote:
> Author: davidsh
> Date: Thu Oct 6 10:47:36 2016
> New Revision: 283460
>
> URL: http://llvm.org/viewvc/llvm-project?rev=283460&view=rev
> Log:
> [OpenMP] Check if the template specialization is m
Author: rsmith
Date: Thu Oct 6 15:30:51 2016
New Revision: 283489
URL: http://llvm.org/viewvc/llvm-project?rev=283489&view=rev
Log:
[modules] Be sure to emit local specializations of imported templates, even if
the resulting specialization is not referenced by the rest of the AST. This
both avoid
rsmith added inline comments.
> SemaDecl.cpp:3692-3693
> + // Demote the newly parsed definition to a fake declaration.
> + // FIXME: Sema::AddInitializationToDecl still allows two definitions,
> + // which make the AST variants inconsistent.
> + assert (Def != New && "There i
Author: rsmith
Date: Thu Oct 6 16:31:04 2016
New Revision: 283498
URL: http://llvm.org/viewvc/llvm-project?rev=283498&view=rev
Log:
Add another .def file to module map to fix modules buildbot's displeasure.
Modified:
cfe/trunk/include/clang/module.modulemap
Modified: cfe/trunk/include/clang
Author: rsmith
Date: Thu Oct 6 18:12:58 2016
New Revision: 283508
URL: http://llvm.org/viewvc/llvm-project?rev=283508&view=rev
Log:
PR25890: Fix incoherent error handling in PerformImplicitConversion and
CheckSingleAssignmentConstraints. These no longer produce ExprError() when they
have not emit
rsmith added a comment.
I'm concerned that name lookup will find an inappropriate declaration after
this merging. If we have two visible declarations of a function in the same
scope, and one has a prototype but the other does not, the user should be able
to rely on the prototype being used. `is
rsmith accepted this revision.
rsmith added a comment.
This revision is now accepted and ready to land.
Please add a test to test/PCH for the serialization code. Otherwise, LGTM.
Comment at: clang/include/clang/Basic/DiagnosticParseKinds.td:1032
+ "force_cuda_host_device end
rsmith added a comment.
I only skipped this one because it seemed pointless to me. I have no objection
to providing it, and this change makes perfect sense if libc++ wants to
guarantee that `<__config>` is included whenever any standard C++ header is
included.
https://reviews.llvm.org/D25361
Author: rsmith
Date: Mon Oct 10 01:55:42 2016
New Revision: 283724
URL: http://llvm.org/viewvc/llvm-project?rev=283724&view=rev
Log:
Mark aligned allocation as done.
Modified:
cfe/trunk/lib/Frontend/InitPreprocessor.cpp
cfe/trunk/test/Lexer/cxx-features.cpp
cfe/trunk/www/cxx_status.ht
Author: rsmith
Date: Mon Oct 10 13:54:32 2016
New Revision: 283789
URL: http://llvm.org/viewvc/llvm-project?rev=283789&view=rev
Log:
Re-commit r283722, reverted in r283750, with a fix for a CUDA-specific use of
past-the-end iterator.
Original commit message:
P0035R4: Semantic analysis and code g
On Mon, Oct 10, 2016 at 10:45 AM, Hal Finkel via cfe-commits <
cfe-commits@lists.llvm.org> wrote:
> hfinkel added a comment.
>
> In https://reviews.llvm.org/D25308#566176, @arphaman wrote:
>
> > The updated patch now makes clang warn every time it encounters this
> attribute in C++ mode. Would tha
rsmith accepted this revision.
rsmith added inline comments.
This revision is now accepted and ready to land.
Comment at: include/clang/AST/Decl.h:1222
+ void demoteThisDefinitionToDeclaration() {
+assert (!isThisDeclarationADemotedDefinition() && "Aleady demoted!");
+as
rsmith added a comment.
As discussed on IRC, I have a mild concern about using
`-fsave-optimization-record` (with no argument) to enable the feature, and
`-fsave-optimization-record=X` to enable the feature and specify a filename; in
most (but not all) cases, `-option arg` and `-option=arg` mea
Author: rsmith
Date: Mon Oct 10 19:21:10 2016
New Revision: 283830
URL: http://llvm.org/viewvc/llvm-project?rev=283830&view=rev
Log:
Aligned allocation versus CUDA: make deallocation function preference order
match other CUDA preference orders, per discussion with jlebar. We now model
this in an a
rsmith added a comment.
It sounds like this will hinder our ability to typo-correct to builtins. I
think we only want to suppress implicitly declaring *library* builtins here
(those that are expected to be provided by a header), not all builtins.
https://reviews.llvm.org/D25458
rsmith added inline comments.
Comment at: include/clang/AST/Decl.h:1213
+ bool isThisDeclarationADemotedDefinition() const {
+return NonParmVarDeclBits.IsThisDeclarationADemotedDefinition;
+ }
This is the bug. You can't read this bit here without first chec
rsmith added inline comments.
Comment at: lib/Sema/SemaTemplateInstantiateDecl.cpp:4849
+ DeclarationName Name = D->getDeclName();
+ if (auto *DD = dyn_cast(D))
+Name =
ahatanak wrote:
> rsmith wrote:
> > Do we need to do this for conversion fun
rsmith added inline comments.
Comment at: lib/Sema/SemaChecking.cpp:11356-11357
+ // Compute the EffectiveAlignment as the alignment of the whole chain.
+ CharUnits EffectiveAlignment = Context.getTypeAlignInChars(
+ ReverseMemberChain.back()->getParent()->getTypeForDecl()
Author: rsmith
Date: Wed Oct 12 18:29:02 2016
New Revision: 284081
URL: http://llvm.org/viewvc/llvm-project?rev=284081&view=rev
Log:
Revert r284008. This is us to fail to instantiate static data members in some
cases. I'm working on reducing a testcase.
Removed:
cfe/trunk/test/Modules/Inputs/
Do we need the same change for if-statements too?
On 12 Oct 2016 6:26 pm, "Vedant Kumar via cfe-commits" <
cfe-commits@lists.llvm.org> wrote:
> vsk created this revision.
> vsk added reviewers: arphaman, ikudrin.
> vsk added a subscriber: cfe-commits.
>
> Generate coverage mappings for in switch
t; > On Jul 20, 2016, at 12:10 PM, Richard Smith via cfe-commits <
> cfe-commits@lists.llvm.org> wrote:
> >
> > Author: rsmith
> > Date: Wed Jul 20 14:10:16 2016
> > New Revision: 276159
> >
> > URL: http://llvm.org/viewvc/llvm-project?rev=276159&view=re
Author: rsmith
Date: Thu Oct 13 18:04:14 2016
New Revision: 284176
URL: http://llvm.org/viewvc/llvm-project?rev=284176&view=rev
Log:
Reinstate r281429, reverted in r281452, with a fix for its mishandling of
compiles without -fmodules-local-submodule-visibility. Original commit message:
[modules]
rsmith added inline comments.
Comment at: lib/CodeGen/CGBuiltin.cpp:1037-1039
+llvm::APSInt AlignInBits;
+if (!E->getArg(1)->EvaluateAsInt(AlignInBits, CGM.getContext()))
+ break;
This takes the alignment in **bits**? That's so ridiculously dumb that
Author: rsmith
Date: Thu Oct 13 20:12:20 2016
New Revision: 284184
URL: http://llvm.org/viewvc/llvm-project?rev=284184&view=rev
Log:
Use SubstInitializer instead of SubstExpr when instantiating a default
argument, in order to correctly instantiate the initializer.
Modified:
cfe/trunk/lib/Sema
Author: rsmith
Date: Thu Oct 13 21:35:11 2016
New Revision: 284187
URL: http://llvm.org/viewvc/llvm-project?rev=284187&view=rev
Log:
Fix bogus assert breaking modules self-host.
Modified:
cfe/trunk/lib/AST/ASTContext.cpp
Modified: cfe/trunk/lib/AST/ASTContext.cpp
URL:
http://llvm.org/viewvc
On Fri, Oct 14, 2016 at 12:49 AM, Eric Fiselier via cfe-commits <
cfe-commits@lists.llvm.org> wrote:
> Author: ericwf
> Date: Fri Oct 14 02:49:15 2016
> New Revision: 284210
>
> URL: http://llvm.org/viewvc/llvm-project?rev=284210&view=rev
> Log:
> XFAIL aligned allocation test failures with UBSAN
Author: rsmith
Date: Fri Oct 14 14:51:36 2016
New Revision: 284271
URL: http://llvm.org/viewvc/llvm-project?rev=284271&view=rev
Log:
[linux] When pre-reserving stack pages to work around broken address space
layout for PIE binaries, ask the OS how much stack space is already in use to
avoid stack
Author: rsmith
Date: Fri Oct 14 16:41:24 2016
New Revision: 284284
URL: http://llvm.org/viewvc/llvm-project?rev=284284&view=rev
Log:
Reinstate r284008 reverted in r284081, with two fixes:
1) Merge and demote variable definitions when we find a redefinition in
MergeVarDecls, not only when we find
rsmith added a comment.
I reverted this in r284081, and relanded with fixes described here as r284284.
Comment at: lib/Sema/SemaDecl.cpp:9712
+
+ // Demote the newly parsed definition to a fake declaration.
+ if (!VDecl->isThisDeclarationADemotedDefinition())
On Fri, Oct 14, 2016 at 11:44 AM, Bruno Cardoso Lopes <
bruno.card...@gmail.com> wrote:
> Hi Richard,
>
> I have a patch on top of your suggested patch from a year ago, that
> break the cyclic dependency we're seeing, with this (and a few changes
> to the SDK) we can bootstrap clang with submodule
Oh, I have another idea: could it be that you're also turning some
optimization on when UBSan is enabled? Note that the operator new/operator
delete pair is elidable in each of these tests, and Clang will remove the
calls when compiling with optimizations enabled.
On Fri, Oct 14, 2016 at 2:38 PM,
On Fri, Oct 14, 2016 at 3:34 PM, Eric Fiselier via cfe-commits <
cfe-commits@lists.llvm.org> wrote:
> Oh, I have another idea: could it be that you're also turning some
> optimization on when UBSan is enabled? Note that the operator new/operator
> delete pair is elidable in each of these tests, an
Author: rsmith
Date: Fri Oct 14 20:59:52 2016
New Revision: 284300
URL: http://llvm.org/viewvc/llvm-project?rev=284300&view=rev
Log:
Disable a silly GCC diagnostic for combining a scanf length specifier with the
'*' specifier. Apparently the GNU folks want to discourage self-documenting
code.
Mod
rsmith accepted this revision.
rsmith added a comment.
This revision is now accepted and ready to land.
Thanks for your patience, this looks great to me. Do you need someone to commit
this for you?
https://reviews.llvm.org/D25153
___
cfe-commits ma
Author: rsmith
Date: Sun Oct 16 01:23:29 2016
New Revision: 284331
URL: http://llvm.org/viewvc/llvm-project?rev=284331&view=rev
Log:
Extend this test and make it a bit clearer which cases Clang is getting wrong.
Modified:
cfe/trunk/test/CXX/temp/temp.fct.spec/temp.deduct/temp.deduct.conv/p4.c
rsmith added a comment.
Please also add a test to test/CXX/drs/dr15xx.cpp for core issue 1518, which
this paper was resolving.
Comment at: lib/AST/DeclCXX.cpp:564
+// C++1z [dcl.init.aggr]p1:
+// - no user-provided, explicit, or inherited constructors,
+if (getAST
Author: rsmith
Date: Sun Oct 16 12:54:23 2016
New Revision: 284337
URL: http://llvm.org/viewvc/llvm-project?rev=284337&view=rev
Log:
P0012R1: Make exception specifications be part of the type system. This
implements the bulk of the change (modifying the type system to include
exception specificati
Author: rsmith
Date: Thu Nov 30 00:18:21 2017
New Revision: 319408
URL: http://llvm.org/viewvc/llvm-project?rev=319408&view=rev
Log:
Preserve the "last diagnostic was suppressed" flag across SFINAE checks.
Sometimes we check the validity of some construct between producing a
diagnostic and produc
Author: rsmith
Date: Thu Nov 30 15:07:29 2017
New Revision: 319495
URL: http://llvm.org/viewvc/llvm-project?rev=319495&view=rev
Log:
Update website to mention that you still need -frelaxed-template-template-args
to enable the corresponding C++17 feature in Clang 5.
Modified:
cfe/trunk/www/cx
Author: rsmith
Date: Thu Nov 30 17:07:10 2017
New Revision: 319509
URL: http://llvm.org/viewvc/llvm-project?rev=319509&view=rev
Log:
[c++2a] P0515R3: lexer support for new <=> token.
Added:
cfe/trunk/test/Lexer/cxx2a-spaceship.cpp
Modified:
cfe/trunk/include/clang/Basic/DiagnosticLexKinds
Author: rsmith
Date: Thu Nov 30 18:13:10 2017
New Revision: 319513
URL: http://llvm.org/viewvc/llvm-project?rev=319513&view=rev
Log:
[c++2a] P0515R3: Support for overloaded operator<=>.
No CodeGen support for MSABI yet, we don't know how to mangle this there.
Added:
cfe/trunk/test/CodeGenCXX
Author: rsmith
Date: Fri Dec 1 13:24:36 2017
New Revision: 319584
URL: http://llvm.org/viewvc/llvm-project?rev=319584&view=rev
Log:
[c++17] When deducing the type of a non-type template parameter from the type
of its argument, perform function-to-pointer and array-to-pointer decay on the
paramete
Author: rsmith
Date: Fri Dec 1 16:55:48 2017
New Revision: 319601
URL: http://llvm.org/viewvc/llvm-project?rev=319601&view=rev
Log:
Move comment back to being next to the code it's a comment for.
Modified:
cfe/trunk/lib/AST/Decl.cpp
Modified: cfe/trunk/lib/AST/Decl.cpp
URL:
http://llvm.org
Author: rsmith
Date: Fri Dec 1 18:48:42 2017
New Revision: 319605
URL: http://llvm.org/viewvc/llvm-project?rev=319605&view=rev
Log:
PR35456: Track definedness of variable template specializations separately from
whether they have an initializer.
We cannot distinguish between a declaration of a v
Author: rsmith
Date: Mon Dec 4 17:31:47 2017
New Revision: 319727
URL: http://llvm.org/viewvc/llvm-project?rev=319727&view=rev
Log:
Generalize "static data member instantiated" notification to cover variable
templates too.
While here, split the "point of instantiation changed" notification out
On 5 December 2017 at 09:08, Artem Dergachev via Phabricator via
cfe-commits wrote:
> NoQ created this revision.
> Herald added subscribers: cfe-commits, rnkovacs.
>
> Yet another situation where we cannot easily guess, by looking at the CFG,
> the target region for C++ constructor call.
>
> Cons
Author: rsmith
Date: Tue Dec 5 11:39:37 2017
New Revision: 319817
URL: http://llvm.org/viewvc/llvm-project?rev=319817&view=rev
Log:
Give this test a triple to avoid failures on MS bots.
Modified:
cfe/trunk/test/Modules/var-templates.cpp
Modified: cfe/trunk/test/Modules/var-templates.cpp
URL
e-checks-win/builds/6568
>
> Failing Tests (3):
> . . .
> Clang :: Modules/var-templates.cpp
> . . .
>
> Please have a look?
>
> Thanks
>
> Galina
>
> On Mon, Dec 4, 2017 at 5:31 PM, Richard Smith via cfe-commits <
> cfe-commits@lists.llvm.org> wrot
Author: rsmith
Date: Tue Dec 5 13:29:36 2017
New Revision: 319835
URL: http://llvm.org/viewvc/llvm-project?rev=319835&view=rev
Log:
Use an even more precise triple to avoid errors on Darwin, where we don't use
comdats for inline entities.
Modified:
cfe/trunk/test/Modules/var-templates.cpp
Should be fixed in r319835. I'd forgotten that Darwin / MachO uses a
different mechanism from COMDATs here.
On 5 December 2017 at 13:27, Ahmed Bougacha via cfe-commits <
cfe-commits@lists.llvm.org> wrote:
> On Tue, Dec 5, 2017 at 11:40 AM, Richard Smith via cfe-commits
> wrote
Author: rsmith
Date: Tue Dec 5 15:54:25 2017
New Revision: 319858
URL: http://llvm.org/viewvc/llvm-project?rev=319858&view=rev
Log:
P0722R2: The first parameter in an implicit call to a destroying operator
delete should be a cv-unqualified pointer to the deleted object.
Modified:
cfe/trunk/l
Author: rsmith
Date: Tue Dec 5 19:00:51 2017
New Revision: 319875
URL: http://llvm.org/viewvc/llvm-project?rev=319875&view=rev
Log:
Fix a bunch of wrong "tautological unsigned enum compare" diagnostics in C++.
An enumeration with a fixed underlying type can have any value in its
underlying type,
Author: rsmith
Date: Wed Dec 6 11:23:19 2017
New Revision: 319942
URL: http://llvm.org/viewvc/llvm-project?rev=319942&view=rev
Log:
Delete special-case "out-of-range" handling for bools, and just use the normal
codepath plus the new "minimum / maximum value of type" diagnostic to get the
same eff
Author: rsmith
Date: Thu Dec 7 16:45:25 2017
New Revision: 320122
URL: http://llvm.org/viewvc/llvm-project?rev=320122&view=rev
Log:
Unify implementation of our two different flavours of -Wtautological-compare.
In so doing, fix a handful of remaining bugs where we would report false
positives or
Author: rsmith
Date: Thu Dec 7 17:00:27 2017
New Revision: 320124
URL: http://llvm.org/viewvc/llvm-project?rev=320124&view=rev
Log:
Fold together the in-range and out-of-range portions of -Wtautological-compare.
Modified:
cfe/trunk/lib/Sema/SemaChecking.cpp
Modified: cfe/trunk/lib/Sema/Sema
On 8 Dec 2017 08:54, "Hans Wennborg via cfe-commits" <
cfe-commits@lists.llvm.org> wrote:
Author: hans
Date: Fri Dec 8 08:54:08 2017
New Revision: 320162
URL: http://llvm.org/viewvc/llvm-project?rev=320162&view=rev
Log:
Revert "Unify implementation of our two different flavours of
-Wtautological
I'm going to reland without the changes to bit-field handling. If we want
those changes (which I think we do, based on the bugs it found in
selfhost), that'll need to be done more carefully, and I don't want to get
the refactoring and bugfixes here tangled up with that.
On 8 December 2017 at 13:27
for the self-host.
>
> On Fri, Dec 8, 2017 at 1:33 PM, Richard Smith via cfe-commits
> wrote:
> > I'm going to reland without the changes to bit-field handling. If we want
> > those changes (which I think we do, based on the bugs it found in
> selfhost),
> > tha
Author: rsmith
Date: Fri Dec 8 14:57:11 2017
New Revision: 320211
URL: http://llvm.org/viewvc/llvm-project?rev=320211&view=rev
Log:
Unify implementation of our two different flavours of -Wtautological-compare,
and fold together into a single function.
In so doing, fix a handful of remaining bugs
Author: rsmith
Date: Fri Dec 8 15:29:59 2017
New Revision: 320212
URL: http://llvm.org/viewvc/llvm-project?rev=320212&view=rev
Log:
Remove creation of out-of-bounds value of enumeration type (resulting in UB).
Also remove unnecessary initialization of out-parameters with this value, so
that MSan
ning is easy to fix. The
>> "-1 vs enum" comparisons are more plentiful and less straightforward.
>> I'm guessing it's the same for the self-host.
>>
>> On Fri, Dec 8, 2017 at 1:33 PM, Richard Smith via cfe-commits
>> wrote:
>> > I'm g
Thank you and everyone else who has helped to make this happen!
On 9 Dec 2017 04:10, "Tim Northover via cfe-commits" <
cfe-commits@lists.llvm.org> wrote:
> Author: tnorthover
> Date: Sat Dec 9 04:09:54 2017
> New Revision: 320250
>
> URL: http://llvm.org/viewvc/llvm-project?rev=320250&view=rev
>
On 12 Dec 2017 16:02, "Erich Keane via cfe-commits" <
cfe-commits@lists.llvm.org> wrote:
Author: erichkeane
Date: Tue Dec 12 08:02:06 2017
New Revision: 320489
URL: http://llvm.org/viewvc/llvm-project?rev=320489&view=rev
Log:
Fix ICE when __has_unqiue_object_representations called with invalid de
Author: Richard Smith
Date: 2022-08-29T21:42:11-07:00
New Revision: 740a75c9d6adf8a26ebd13fc1d3460c728550624
URL:
https://github.com/llvm/llvm-project/commit/740a75c9d6adf8a26ebd13fc1d3460c728550624
DIFF:
https://github.com/llvm/llvm-project/commit/740a75c9d6adf8a26ebd13fc1d3460c728550624.diff
Author: Richard Smith
Date: 2022-08-30T18:31:53-07:00
New Revision: 146ae4138a081a9a10e4901bbec61b629331da8a
URL:
https://github.com/llvm/llvm-project/commit/146ae4138a081a9a10e4901bbec61b629331da8a
DIFF:
https://github.com/llvm/llvm-project/commit/146ae4138a081a9a10e4901bbec61b629331da8a.diff
Author: Richard Smith
Date: 2022-09-02T16:08:30-07:00
New Revision: b484256f59850e702df4d4532c5f31f478879bb9
URL:
https://github.com/llvm/llvm-project/commit/b484256f59850e702df4d4532c5f31f478879bb9
DIFF:
https://github.com/llvm/llvm-project/commit/b484256f59850e702df4d4532c5f31f478879bb9.diff
Author: Richard Smith
Date: 2022-09-02T18:43:27-07:00
New Revision: 220c2cb0a88f0f1cc97ea657018abbffb8697a73
URL:
https://github.com/llvm/llvm-project/commit/220c2cb0a88f0f1cc97ea657018abbffb8697a73
DIFF:
https://github.com/llvm/llvm-project/commit/220c2cb0a88f0f1cc97ea657018abbffb8697a73.diff
Author: Richard Smith
Date: 2022-09-02T19:09:15-07:00
New Revision: 632e1e9fd81d899ffa01300eb47fd417b76bbae5
URL:
https://github.com/llvm/llvm-project/commit/632e1e9fd81d899ffa01300eb47fd417b76bbae5
DIFF:
https://github.com/llvm/llvm-project/commit/632e1e9fd81d899ffa01300eb47fd417b76bbae5.diff
Author: Richard Smith
Date: 2022-09-06T17:12:57-07:00
New Revision: a002063de37c00133b93d2ce1f0ea78fc4b807ff
URL:
https://github.com/llvm/llvm-project/commit/a002063de37c00133b93d2ce1f0ea78fc4b807ff
DIFF:
https://github.com/llvm/llvm-project/commit/a002063de37c00133b93d2ce1f0ea78fc4b807ff.diff
@@ -802,9 +842,19 @@ InitListChecker::FillInEmptyInitializations(const
InitializedEntity &Entity,
}
}
} else {
+ InitListExpr *SForm =
+ ILE->isSyntacticForm() ? ILE : ILE->getSyntacticForm();
// The fields beyond ILE->getNumInits() are d
@@ -465,7 +460,8 @@ class InitListChecker {
void FillInEmptyInitForField(unsigned Init, FieldDecl *Field,
const InitializedEntity &ParentEntity,
InitListExpr *ILE, bool &RequiresSecondPass,
-
@@ -727,6 +729,44 @@ void InitListChecker::FillInEmptyInitForField(unsigned
Init, FieldDecl *Field,
if (hadError || VerifyOnly) {
// Do nothing
} else if (Init < NumInits) {
zygoloid wrote:
Is it OK that we don't warn on the `else` case here? Do
@@ -727,6 +729,44 @@ void InitListChecker::FillInEmptyInitForField(unsigned
Init, FieldDecl *Field,
if (hadError || VerifyOnly) {
// Do nothing
} else if (Init < NumInits) {
+ if (MaybeEmitMFIWarning) {
+auto CheckAnonMember = [&](const FieldDecl *FD
@@ -727,6 +729,44 @@ void InitListChecker::FillInEmptyInitForField(unsigned
Init, FieldDecl *Field,
if (hadError || VerifyOnly) {
// Do nothing
} else if (Init < NumInits) {
zygoloid wrote:
Hmm, how does that work? If this `if` condition doesn't
@@ -727,6 +734,38 @@ void InitListChecker::FillInEmptyInitForField(unsigned
Init, FieldDecl *Field,
if (hadError || VerifyOnly) {
// Do nothing
} else if (Init < NumInits) {
+ if (WarnIfMissing) {
+auto CheckAnonMember = [&](const FieldDecl *FD,
+
901 - 1000 of 4192 matches
Mail list logo
