Author: sepavloff
Date: Thu May 11 03:25:22 2017
New Revision: 302777
URL: http://llvm.org/viewvc/llvm-project?rev=302777&view=rev
Log:
Reverted r302775
Added:
cfe/trunk/test/Driver/arm-darwin-builtin.c
- copied unchanged from r302774,
cfe/trunk/test/Driver/arm-darwin-builtin.c
Modifie
Sorry.
I have to investigate a fail in clang-tidy.
Thanks,
--Serge
2017-05-11 15:43 GMT+07:00 Richard Smith :
> When you revert a patch (even one of your own), please say why you
> reverted it in the commit message.
>
>
> On 11 May 2017 1:38 am, "Serge Pavlov via cfe-comm
Ping.
Thanks,
--Serge
2016-10-13 11:51 GMT+07:00 Serge Pavlov :
> sepavloff updated the summary for this revision.
>
> https://reviews.llvm.org/D21508
>
>
>
>
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman
sepavloff created this revision.
sepavloff added a subscriber: cfe-commits.
If initializer contains parentheses around braced list where it is not allowed,
as in
construct `int({0})`, clang issued message like `functional-style cast from
'void' to
'int' is not allowed`, which does not much help.
sepavloff updated this revision to Diff 75404.
sepavloff added a comment.
Addressed reviewr's notes.
https://reviews.llvm.org/D25816
Files:
include/clang/Basic/DiagnosticSemaKinds.td
include/clang/Sema/Sema.h
lib/Sema/SemaDecl.cpp
lib/Sema/SemaExprCXX.cpp
test/SemaCXX/cxx0x-initialize
sepavloff marked 5 inline comments as done.
sepavloff added inline comments.
Comment at: include/clang/Basic/DiagnosticSemaKinds.td:1762
def err_init_incomplete_type : Error<"initialization of incomplete type %0">;
+def err_list_init_in_parens : Error<"list-initializer for non-c
sepavloff added inline comments.
Comment at: lib/Sema/SemaExprCXX.cpp:4231
+const CXXConstructorDecl *Constructor = nullptr;
+if (const ConstructorUsingShadowDecl *CSD =
+dyn_cast(ND)) {
Use `auto` here. Type of `CSD` is clear from `dy
Ping.
Thanks,
--Serge
2016-10-18 0:09 GMT+07:00 Serge Pavlov :
> Ping.
>
> Thanks,
> --Serge
>
> 2016-10-13 11:51 GMT+07:00 Serge Pavlov :
>
>> sepavloff updated the summary for this revision.
>>
>> https://reviews.llvm.org/D21508
>>
>>
>>
>>
>
___
cfe
sepavloff created this revision.
sepavloff added a subscriber: cfe-commits.
Clang makes check for function redefinition after it merged the new
declaration with the existing one. As a result, it produces poor
diagnostics in the case of a friend function defined inline, as in
the code:
void func
Author: sepavloff
Date: Mon Oct 31 00:11:12 2016
New Revision: 285543
URL: http://llvm.org/viewvc/llvm-project?rev=285543&view=rev
Log:
Make output of ast-print closer to C++ code
Put semicolon after non-defining method declaration and a class
specialization body.
Modified:
cfe/trunk/lib/AST
sepavloff created this revision.
sepavloff added a subscriber: cfe-commits.
This change adds to the method isThisDeclarationADefinition additional
conditions under which a function declaration becomes a definition.
These includes the case of a declaration that does not have a function
body because
sepavloff updated this revision to Diff 76520.
sepavloff added a comment.
Do not use getFriendObjectKind(), friend kind sometimes is set too late.
https://reviews.llvm.org/D26065
Files:
lib/Sema/SemaDeclCXX.cpp
test/CXX/dcl.dcl/dcl.spec/dcl.fct.spec/p4.cpp
Index: test/CXX/dcl.dcl/dcl.spec
sepavloff added a comment.
Summary of proposal and discussion in mail list.
**Problems**
The two main problems this feature addresses are:
1. Ability to modify default compiler settings.
As an example, the warning `-Wundefined-var-template` can be helpful for people
doing module-enabled build
Author: sepavloff
Date: Thu Nov 3 02:32:02 2016
New Revision: 285882
URL: http://llvm.org/viewvc/llvm-project?rev=285882&view=rev
Log:
Enhancement to test for -ast-print
Present tests for the functionality provided by command lime option
`-ast-print` check only absence of crash. This change trie
Author: sepavloff
Date: Thu Nov 3 03:58:44 2016
New Revision: 285889
URL: http://llvm.org/viewvc/llvm-project?rev=285889&view=rev
Log:
Reverted r285882 (Enhancement to test for -ast-print)
It broke buildbot on Windows.
Modified:
cfe/trunk/test/Coverage/ast-printing.c
Modified: cfe/trunk/te
Author: sepavloff
Date: Fri Nov 4 01:03:34 2016
New Revision: 285979
URL: http://llvm.org/viewvc/llvm-project?rev=285979&view=rev
Log:
Do not print enum underlying type if language is not C++11
Output generated by option '-ast-print' must not contains enum
base type specifications if source lang
Author: sepavloff
Date: Fri Nov 4 01:09:23 2016
New Revision: 285981
URL: http://llvm.org/viewvc/llvm-project?rev=285981&view=rev
Log:
Enhancement to test for -ast-print
Present tests for the functionality provided by command lime option
`-ast-print` check only absence of crash. This change trie
This following commit r285981 (Enhancement to test for -ast-print tests)
tests this change. Without it the test 'Coverage/ast-printing.c' fails on
Windows.
Thanks,
--Serge
2016-11-04 19:09 GMT+07:00 Aaron Ballman :
> On Fri, Nov 4, 2016 at 2:03 AM, Serge Pavlov via cfe-com
sepavloff added a comment.
> For Chromium, our build system provides a specific Clang version close to
> ToT, and obviously what flags to use when invoking it. (Developers can of
> course override the flags when configuring if they want.) Now maybe a
> developer has a ~/clang.cfg because they w
sepavloff created this revision.
sepavloff added reviewers: ABataev, gribozavr, rtrieu.
sepavloff added a subscriber: cfe-commits.
Output generated by option -ast-print looks like C/C++ code, and it
really is for plain C. For C++ the produced output was not valid C++
code, but the differences were
m, "Serge Pavlov via cfe-commits" <
> cfe-commits@lists.llvm.org> wrote:
>
>> Author: sepavloff
>> Date: Mon Oct 31 00:11:12 2016
>> New Revision: 285543
>>
>> URL: http://llvm.org/viewvc/llvm-project?rev=285543&view=rev
>> Log:
>&
This revision was automatically updated to reflect the committed changes.
Closed by commit rL286439: Make output of -ast-print a valid C++ code.
(authored by sepavloff).
Changed prior to commit:
https://reviews.llvm.org/D26452?vs=77354&id=77452#toc
Repository:
rL LLVM
https://reviews.llvm.o
I don't see any reason why we should adhere particular solution. If you can
solve the problem with another approach, please go forward.
Thanks,
--Serge
2016-11-10 10:12 GMT+07:00 Akira Hatanaka :
> ahatanak added a comment.
>
> Do you think we can transform the Exprs for the uninstantiated defau
sepavloff added a comment.
>> Default config file is searched for only in the directory where clang
>> executable resides. It allows SDK suppliers to customize compiler, for
>> instance by turning off unneeded warnings, by providing their own clang.cfg.
>> On the other hand developers cannot by
sepavloff added a comment.
>> Config file set by environment variable is similar to default config file as
>> user does not specify anything in command line. But the user must explicitly
>> specify the file, it is not made automatically.
>
> Sorry, I don't understand this part.
If a user speci
Is it OK to commit this patch?
Thanks,
--Serge
2016-10-21 18:21 GMT+07:00 Alex Lorenz :
> arphaman added a subscriber: rsmith.
> arphaman added a comment.
>
> LGTM, I added Richard in case he has something to add.
>
> > I chose to retain current clang behavior and reject questionable code.
> GCC
Author: sepavloff
Date: Sat Nov 12 09:38:55 2016
New Revision: 286721
URL: http://llvm.org/viewvc/llvm-project?rev=286721&view=rev
Log:
Use descriptive message if list initializer is incorrectly parenthesized.
If initializer contains parentheses around braced list where it is not allowed,
as in c
This revision was automatically updated to reflect the committed changes.
Closed by commit rL286721: Use descriptive message if list initializer is
incorrectly parenthesized. (authored by sepavloff).
Changed prior to commit:
https://reviews.llvm.org/D25816?vs=75404&id=77725#toc
Repository:
r
sepavloff updated this revision to Diff 34673.
sepavloff added a comment.
Updated diagnostics according to Sean's notes ('textual module' -> 'textual
header')
http://reviews.llvm.org/D11844
Files:
include/clang/Basic/DiagnosticParseKinds.td
include/clang/Basic/DiagnosticSemaKinds.td
incl
sepavloff updated this revision to Diff 35115.
sepavloff added a comment.
Updated patch.
http://reviews.llvm.org/D11844
Files:
include/clang/Basic/DiagnosticParseKinds.td
include/clang/Basic/DiagnosticSemaKinds.td
include/clang/Parse/Parser.h
include/clang/Sema/Sema.h
lib/Parse/ParseD
sepavloff added a comment.
Thanks for helpful notes!
There are similar cases in Objective C. Should they be implemented in this
patch or separate patch is OK?
Comment at: include/clang/Basic/DiagnosticParseKinds.td:1016
@@ -1015,2 +1015,3 @@
"expected ';' after module name"
This revision was automatically updated to reflect the committed changes.
Closed by commit rL248085: [Modules] More descriptive diagnostics for misplaced
import directive (authored by sepavloff).
Changed prior to commit:
http://reviews.llvm.org/D11844?vs=35115&id=35161#toc
Repository:
rL LLV
Author: sepavloff
Date: Sat Sep 19 00:32:57 2015
New Revision: 248085
URL: http://llvm.org/viewvc/llvm-project?rev=248085&view=rev
Log:
[Modules] More descriptive diagnostics for misplaced import directive
If an import directive was put into wrong context, the error message was
obscure,
complain
sepavloff updated this revision to Diff 31831.
sepavloff added a comment.
Fixed module boundary treatment.
This version must fix problems in the treatment of annot_modulbegin and
annot_module_end. Error recover for the token annot_module_include found
in wrong context can be made as previously,
sepavloff updated this revision to Diff 32085.
sepavloff added a comment.
Updated patch
Thanks to all for fruitful discussion!
The new version tries to addresses review notes. It differs from the previous
version in:
- Method tryParseMisplacedModuleImport now depends only on current token and
sepavloff updated this revision to Diff 32310.
sepavloff added a comment.
Updated method isLexicallyWithinFunctionOrMethod according to Richard's notes.
http://reviews.llvm.org/D11194
Files:
include/clang/AST/DeclBase.h
lib/AST/DeclBase.cpp
lib/Sema/SemaTemplateInstantiate.cpp
lib/Sema/
sepavloff added inline comments.
Comment at: lib/AST/DeclBase.cpp:273
@@ +272,3 @@
+return true;
+ if (const CXXRecordDecl *ClassD = dyn_cast(LDC))
+return ClassD->isLocalClass() && !ClassD->isLambda();;
rsmith wrote:
> It's not necessary for this change,
sepavloff updated this revision to Diff 32597.
sepavloff added a comment.
Lambda functions are now treated as local classes.
http://reviews.llvm.org/D11194
Files:
include/clang/AST/DeclBase.h
lib/AST/DeclBase.cpp
lib/Sema/SemaTemplateInstantiate.cpp
lib/Sema/SemaTemplateInstantiateDecl.
Author: sepavloff
Date: Sun Aug 23 05:22:28 2015
New Revision: 245810
URL: http://llvm.org/viewvc/llvm-project?rev=245810&view=rev
Log:
Instantiate function declarations in instantiated functions.
If a function declaration is found inside a template function as in:
template void f() {
This revision was automatically updated to reflect the committed changes.
Closed by commit rL245810: Instantiate function declarations in instantiated
functions. (authored by sepavloff).
Changed prior to commit:
http://reviews.llvm.org/D11194?vs=32597&id=32923#toc
Repository:
rL LLVM
http:/
Author: sepavloff
Date: Sun Aug 23 06:09:40 2015
New Revision: 245811
URL: http://llvm.org/viewvc/llvm-project?rev=245811&view=rev
Log:
Changes missed from r245810
Modified:
cfe/trunk/lib/AST/DeclBase.cpp
Modified: cfe/trunk/lib/AST/DeclBase.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/
sepavloff created this revision.
sepavloff added a subscriber: cfe-commits.
Instantiation of static class members can be not obvious in some cases. Using
modules can cause problems even more difficult to diagnose. PR24425 describes
one of such cases. As a way to assist a user, compiler could issue
sepavloff added a subscriber: sepavloff.
sepavloff added a comment.
Ping.
Thanks,
--Serge
http://reviews.llvm.org/D11844
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Ping.
Thanks,
--Serge
2015-08-22 3:15 GMT+06:00 Sean Silva :
> silvas added inline comments.
>
>
> Comment at: test/Modules/auto-module-import.m:89
> @@ +88,3 @@
> + expected-error {{import of module
> 'NoUmbrella.A' appears within function 'includeN
Any feedback?
Thanks,
--Serge
2015-08-26 0:26 GMT+06:00 Serge Pavlov :
> sepavloff created this revision.
> sepavloff added a subscriber: cfe-commits.
>
> Instantiation of static class members can be not obvious in some cases.
> Using
> modules can cause problems even more difficult to diagnose.
sepavloff created this revision.
sepavloff added a subscriber: cfe-commits.
This change fixes performance degradation reported in PR24667. The reason
of the problem is leaving module after each header, even if the next header
is of the same module. With this fix module state record counts the numb
sepavloff created this revision.
sepavloff added a subscriber: cfe-commits.
http://reviews.llvm.org/D14224
Files:
lib/AST/ASTImporter.cpp
test/ASTMerge/Inputs/class3.cpp
test/ASTMerge/Inputs/exprs3.cpp
test/ASTMerge/class2.cpp
test/ASTMerge/exprs.cpp
Index: test/ASTMerge/exprs.cpp
sepavloff added a comment.
The fix must contain tests.
Comment at: lib/AST/ASTImporter.cpp:35
@@ +34,3 @@
+void ImportMultipleItems(IIter Ibegin, IIter Iend, OIter Obegin) {
+ ASTImporter &_Importer = Importer;
+ std::transform(Ibegin, Iend, Obegin,
---
2015-11-06 20:28 GMT+06:00 Aleksei Sidorin :
> a.sidorin marked 7 inline comments as done.
> a.sidorin added a comment.
>
> Thank you for your comments. I leaved some replies and will update
> revision.
> Something about lacking tests.
>
> 1. We cannot check if expression import is correct until w
sepavloff added a comment.
The patch requires tests.
For instance, import of CXXBaseSpecifier can be tested along with static_cast:
if a class indeed is a base, static_cast will be processed silent, otherwise a
message must appear. ArraySubscript could be tested using constexpr function.
Materi
sepavloff created this revision.
sepavloff added a subscriber: cfe-commits.
If an import directive was put into wrong context, the error message was
obscure,
complaining on misbalanced braces. To get more descriptive messages, annotation
tokens related to modules are processed where they must not
Unit test ImportVAArgExpr fails on Windows. The source file:
void declToImport(__builtin_va_list list, ...) {
(void)__builtin_va_arg(list, int);
}
When compiled on Windows it produces AST:
TranslationUnitDecl 0x638f150 <>
`-FunctionDecl 0x638f780 line:1:6 declToImport
'void (__builtin
sepavloff added a comment.
With this patch unit tests pass on Windows as well.
I do not have formal authority to approve patches, but for me the patch is good
enough to be accepted.
Comment at: lib/AST/ASTImporter.cpp:3422
@@ +3421,3 @@
+ D->getTrailingObjects();
+ for (i
sepavloff updated this revision to Diff 65376.
sepavloff added a comment.
Added one more case in shouldLinkDependentDeclWithPrevious
https://reviews.llvm.org/D16989
Files:
include/clang/Sema/Sema.h
lib/Sema/SemaDecl.cpp
test/SemaCXX/PR25848.cpp
test/SemaCXX/friend2.cpp
Index: test/Sema
sepavloff created this revision.
sepavloff added a subscriber: cfe-commits.
Function specializations used in friend declarations in class templates, like:
```
template class C1 {
friend void func<>(int);
```
previously were processed incorrectly: class instantiation made them ordinary
sepavloff added inline comments.
Comment at: test/SemaTemplate/default-expr-arguments-3.cpp:20
@@ +19,2 @@
+ }
+}
Please add the following test to the patch:
```
template void f1() {
enum class foo { a, b };
struct S {
int g1(foo n = foo::a);
};
}
temp
Author: sepavloff
Date: Sun Aug 7 23:02:15 2016
New Revision: 277976
URL: http://llvm.org/viewvc/llvm-project?rev=277976&view=rev
Log:
Pass information in a record instead of stack. NFC
Functions of Sema that work with building of nested name specifiers have too
many parameters (BuildCXXNestedNa
Ping.
Thanks,
--Serge
2016-08-01 21:47 GMT+07:00 Serge Pavlov :
> sepavloff created this revision.
> sepavloff added a subscriber: cfe-commits.
>
> Function specializations used in friend declarations in class templates,
> like:
> ```
> template class C1 {
> friend void func<>(int);
Whether enable this warning or not should be determined by user feedback.
The warning was implemented to help users in complicated cases that arise
in module-enabled builds. It seems however that it makes troubles for other
users. Based on feedback on this list I feel that it is better to make this
sepavloff updated this revision to Diff 46575.
sepavloff added a comment.
Added regression test.
http://reviews.llvm.org/D15450
Files:
lib/CodeGen/CodeGenAction.cpp
unittests/Frontend/CMakeLists.txt
unittests/Frontend/CodeGenActionTest.cpp
Index: unittests/Frontend/CodeGenActionTest.cpp
Any feedback?
Thanks,
--Serge
2016-02-02 3:31 GMT+06:00 Serge Pavlov :
> sepavloff updated this revision to Diff 46575.
> sepavloff added a comment.
>
> Added regression test.
>
>
> http://reviews.llvm.org/D15450
>
> Files:
> lib/CodeGen/CodeGenAction.cpp
> unittests/Frontend/CMakeLists.txt
sepavloff added a comment.
Any feedback?
Thanks,
--Serge
http://reviews.llvm.org/D15450
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
sepavloff created this revision.
sepavloff added a reviewer: rsmith.
sepavloff added a subscriber: cfe-commits.
Previously if a file-level function was defined inside befriending
template class, it was treated as defined in the code like:
```
void func(int);
template class C1 {
friend void
CodeGen action defined in the unit test does not call HandleTranslationUnit,
so the hack implemented in `BackendConsumer::HandleTranslationUnit` does
not run and clang crashes. With the fix no crash occurs.
Thanks,
--Serge
2016-02-05 1:17 GMT+06:00 Serge Pavlov :
> sepavloff added a comment.
>
>
Any feedback?
Thanks,
--Serge
2016-01-26 20:55 GMT+06:00 Serge Pavlov :
> sepavloff created this revision.
> sepavloff added a subscriber: cfe-commits.
>
> Declaration of friend function may depend on template parameters,
> however it does not become a template function:
>
> template class C
sepavloff added a comment.
In http://reviews.llvm.org/D15450#354606, @yaron.keren wrote:
> It certainly makes sense to redirect the module request to its owner instead
> of duplicating it in a local copy.
It is not even a duplication, it is simultaneous sharing by two unique_ptr's.
> There ma
Author: sepavloff
Date: Thu Feb 18 10:42:09 2016
New Revision: 261222
URL: http://llvm.org/viewvc/llvm-project?rev=261222&view=rev
Log:
Avoid double deletion in Clang driver.
Llvm module object is shared between CodeGenerator and BackendConsumer,
in both classes it is stored as std::unique_ptr, w
This revision was automatically updated to reflect the committed changes.
Closed by commit rL261222: Avoid double deletion in Clang driver. (authored by
sepavloff).
Changed prior to commit:
http://reviews.llvm.org/D15450?vs=46575&id=48332#toc
Repository:
rL LLVM
http://reviews.llvm.org/D154
Author: sepavloff
Date: Thu Feb 18 10:43:24 2016
New Revision: 261223
URL: http://llvm.org/viewvc/llvm-project?rev=261223&view=rev
Log:
File missed from r261222
Added:
cfe/trunk/unittests/Frontend/CodeGenActionTest.cpp
Added: cfe/trunk/unittests/Frontend/CodeGenActionTest.cpp
URL:
http://ll
Author: sepavloff
Date: Fri Feb 19 06:06:23 2016
New Revision: 261323
URL: http://llvm.org/viewvc/llvm-project?rev=261323&view=rev
Log:
Removed unused local variable
Modified:
cfe/trunk/lib/Sema/SemaTemplate.cpp
Modified: cfe/trunk/lib/Sema/SemaTemplate.cpp
URL:
http://llvm.org/viewvc/llvm-
Could someone provide a feedback?
Thanks,
--Serge
2016-01-26 20:55 GMT+06:00 Serge Pavlov :
> sepavloff created this revision.
> sepavloff added a subscriber: cfe-commits.
>
> Declaration of friend function may depend on template parameters,
> however it does not become a template function:
>
>
sepavloff updated this revision to Diff 56015.
sepavloff added a comment.
Updated patch.
http://reviews.llvm.org/D16989
Files:
include/clang/Sema/Sema.h
lib/Sema/SemaDecl.cpp
test/SemaCXX/PR25848.cpp
test/SemaCXX/friend2.cpp
Index: test/SemaCXX/friend2.cpp
=
2016-04-26 0:55 GMT+06:00 Richard Smith :
> rsmith added inline comments.
>
>
> Comment at: lib/Sema/SemaDecl.cpp:8611-8612
> @@ -8609,3 +8610,4 @@
> } else {
> - // This needs to happen first so that 'inline' propagates.
> - NewFD->setPreviousDeclaration(cast(OldDe
sepavloff updated this revision to Diff 56607.
sepavloff added a comment.
Changed implementation of friend function set.
Friend functions declared in template classes are not added to redeclaration
chains,
but access to them is needed to make some checks. This change allows using the
set of frie
Any feedback?
Thanks,
--Serge
2016-05-03 22:40 GMT+06:00 Serge Pavlov :
> 2016-04-26 0:55 GMT+06:00 Richard Smith :
>
>> rsmith added inline comments.
>>
>>
>> Comment at: lib/Sema/SemaDecl.cpp:8611-8612
>> @@ -8609,3 +8610,4 @@
>> } else {
>> - // This needs to happen
In this case moving implementation of `Singleton::getInstance()` into a
.cpp file would prevent compiler from instantiation of the method body when
it sees `Singleton::getInstance()`. In this case
`Singleton::getInstance()` must be instantiated in some source file
either explicitly or implicitly. I
sepavloff updated this revision to Diff 52527.
sepavloff added a comment.
Updated patch
Added a check for presence of template declaration corresponding to the friend
function. Presence of similar functions but with specializations as argument
types is also checked now.
Made an atempt to make me
sepavloff updated this revision to Diff 53547.
sepavloff added a comment.
Updated patch according to review notes.
The patch adds diagnostics for undefined function templates, in a similar way
as for template variables. These two warnings are classified into different
groups.
Warnings on undefin
2016-04-14 2:44 GMT+06:00 Richard Smith :
> rsmith added a comment.
>
> I would prefer to avoid adding the `%qt` format if we can. But if we do
> provide it, the template parameter names we use should match what was
> written in the declaration of the template itself.
>
> New format was not a good
sepavloff updated this revision to Diff 53853.
sepavloff added a comment.
Herald added a subscriber: klimek.
Updated patch
Get rid of '%qt' format, instead print template parameters always. It requires
about 10 tests to be fixed. However printing parameters still can be suppressed,
it is required
sepavloff updated this revision to Diff 54085.
sepavloff added a comment.
Updated patch
Changed text of messages as proposed by reviewer. As a result, printing
template parameters is not needed anymore, related code is removed.
http://reviews.llvm.org/D16396
Files:
include/clang/AST/DeclBase
Author: sepavloff
Date: Tue Apr 19 01:19:52 2016
New Revision: 266719
URL: http://llvm.org/viewvc/llvm-project?rev=266719&view=rev
Log:
Warn if function or variable cannot be implicitly instantiated
With this patch compiler emits warning if it tries to make implicit
instantiation
of a template b
This revision was automatically updated to reflect the committed changes.
Closed by commit rL266719: Warn if function or variable cannot be implicitly
instantiated (authored by sepavloff).
Changed prior to commit:
http://reviews.llvm.org/D16396?vs=54085&id=54163#toc
Repository:
rL LLVM
http
> static LSubKindOperand* Create(int index, Zone* zone) {
>> if (index < kNumCachedOperands) return &cache[index];
>> return new(zone) LSubKindOperand(index);
>> }
>> private:
>> static LSubKindOperand* cache;
>> explicit LSubKindOperand(int index
2016-03-29 1:08 GMT+06:00 Serge Pavlov :
> 2016-03-18 20:50 GMT+06:00 Richard Smith :
>
>> rsmith added a comment.
>>
>> Can we instead not add the function to the redeclaration chain until it's
>> instantiated (like we do if it's dependent)?
>>
>>
> I prepared implementation that uses this approa
sepavloff created this revision.
sepavloff added a reviewer: rsmith.
sepavloff added a subscriber: cfe-commits.
When friend template function is parsed, it obtains wrong type. For
instance, in the code:
template void func(T1 *x);
template struct C1 {
template void func(T1 *x);
};
the friend
Let me demonstrate the problem using excerpt from v8 sources:
-- lithium.h
template
struct LSubKindOperand {
static int* Create(int index) { return &cache[index]; }
static void SetUpCache();
static int* cache;
};
struct LOperand {
2016-04-21 2:57 GMT+06:00 Richard Smith :
> rsmith added a comment.
>
> Sorry, I don't think this approach can work. Consider:
>
> template struct X {
> template friend void f(T);
> template friend void f(U);
> };
>
> These two friend declarations declare different friend function temp
sepavloff updated this revision to Diff 58983.
sepavloff added a comment.
Updated patch
http://reviews.llvm.org/D16989
Files:
include/clang/Sema/Sema.h
lib/Sema/SemaDecl.cpp
test/SemaCXX/PR25848.cpp
test/SemaCXX/friend2.cpp
Index: test/SemaCXX/friend2.cpp
==
sepavloff marked 2 inline comments as done.
sepavloff added a comment.
In http://reviews.llvm.org/D16989#432282, @rsmith wrote:
> Please also add some testcases for the corresponding case for a friend
> function template:
>
> template void f();
> template struct A {
> template void f() {
sepavloff added inline comments.
Comment at: lib/AST/ASTImporter.cpp:2373
@@ +2372,3 @@
+ Error = true;
+ToInfo = TemplateArgumentLocInfo(TSI);
+ } else {
Maybe `else` before this statement so that in the case of error `ToInfo`
remained default initiali
sepavloff updated this revision to Diff 59532.
sepavloff added a comment.
Fixed test for friend functions.
http://reviews.llvm.org/D16989
Files:
include/clang/Sema/Sema.h
lib/Sema/SemaDecl.cpp
test/SemaCXX/PR25848.cpp
test/SemaCXX/friend2.cpp
Index: test/SemaCXX/friend2.cpp
===
Author: sepavloff
Date: Thu Jun 9 23:39:07 2016
New Revision: 272366
URL: http://llvm.org/viewvc/llvm-project?rev=272366&view=rev
Log:
Fix recognition of shadowed template parameter
Crash reported in PR28023 is caused by the fact that non-type template
parameters are found by tag name lookup. I
sepavloff created this revision.
sepavloff added a subscriber: cfe-commits.
If definition of default function argument uses itself, clang crashed,
because corresponding function parameter is not associated with the default
argument yet. With this fix clang emits appropriate error message.
This ch
Author: sepavloff
Date: Mon Jun 13 21:55:56 2016
New Revision: 272623
URL: http://llvm.org/viewvc/llvm-project?rev=272623&view=rev
Log:
Detect recursive default argument definition
If definition of default function argument uses itself, clang crashed,
because corresponding function parameter is n
This revision was automatically updated to reflect the committed changes.
Closed by commit rL272623: Detect recursive default argument definition
(authored by sepavloff).
Changed prior to commit:
http://reviews.llvm.org/D21301?vs=60564&id=60644#toc
Repository:
rL LLVM
http://reviews.llvm.or
sepavloff updated this revision to Diff 61220.
sepavloff added a comment.
Updated patch
Added handling of FunctionTemplateDecl to shouldLinkDependentDeclWithPrevious.
It is not
used now but is required for subsequent patch. Also position comment correctly.
http://reviews.llvm.org/D16989
Files
sepavloff created this revision.
sepavloff added a reviewer: rsmith.
sepavloff added a subscriber: cfe-commits.
Similar to friend functions, if a friend function template is defined in a
class template, its definition is available only if the class is instantiated.
http://reviews.llvm.org/D21508
sepavloff updated this revision to Diff 61989.
sepavloff added a comment.
Make more correct template function definition lookup
Lookup is made recursive to cover more cases.
http://reviews.llvm.org/D21508
Files:
include/clang/AST/DeclTemplate.h
lib/AST/DeclTemplate.cpp
lib/Sema/SemaDecl.
sepavloff created this revision.
sepavloff added a reviewer: rsmith.
sepavloff added a subscriber: cfe-commits.
If a function template was defined in a friend declaration in a
template class, it was not instantiated because template definition
was not found.
http://reviews.llvm.org/D21767
Files:
201 - 300 of 434 matches
Mail list logo
