> On Feb 13, 2016, at 11:34 PM, NAKAMURA Takumi wrote:
>
> I guess it would break layering violation.
Not sure what you mean, what is the layering violation ?
I don’t see an issue with a tool binary linking both against the libclang C API
and a C++ library. I could add yet another tool binary
I guess it would break layering violation.
libclang exports just C API and it involves clangIndex, but coremain
requires clangIndex and other dependent modules.
I think you should do;
- Export C++ API in libclang. (But I don't expect it would work for win32)
- Link libclang as static. Then c-
Author: akirtzidis
Date: Sun Feb 14 01:08:31 2016
New Revision: 260844
URL: http://llvm.org/viewvc/llvm-project?rev=260844&view=rev
Log:
[c-index-test] Fix a gcc build error.
Modified:
cfe/trunk/tools/c-index-test/core_main.cpp
Modified: cfe/trunk/tools/c-index-test/core_main.cpp
URL:
http:
Author: akirtzidis
Date: Sun Feb 14 00:53:20 2016
New Revision: 260843
URL: http://llvm.org/viewvc/llvm-project?rev=260843&view=rev
Log:
[index] Fix gcc builds.
Modified:
cfe/trunk/include/clang/Index/IndexingAction.h
cfe/trunk/tools/c-index-test/core_main.cpp
Modified: cfe/trunk/include
Author: akirtzidis
Date: Sun Feb 14 00:39:03 2016
New Revision: 260841
URL: http://llvm.org/viewvc/llvm-project?rev=260841&view=rev
Log:
[index] Allow calling createIndexingAction() without passing another action to
wrap over.
Modified:
cfe/trunk/include/clang/Index/IndexingAction.h
cfe/
Author: akirtzidis
Date: Sun Feb 14 00:39:11 2016
New Revision: 260842
URL: http://llvm.org/viewvc/llvm-project?rev=260842&view=rev
Log:
[index] Enhance c-index-test tool and have it link and test the clangIndex
library directly.
Added:
cfe/trunk/test/Index/Core/
cfe/trunk/test/Index/Cor
LegalizeAdulthood updated this revision to Diff 47929.
LegalizeAdulthood marked an inline comment as done.
LegalizeAdulthood added a comment.
Add FIXME for macro argument test case.
http://reviews.llvm.org/D16529
Files:
clang-tidy/modernize/CMakeLists.txt
clang-tidy/modernize/ModernizeTidyM
LegalizeAdulthood updated this revision to Diff 47928.
LegalizeAdulthood added a comment.
Add FIXME for non-ASCII string literals.
Allow delimiter stem to be configured.
Avoid some string concatenation.
http://reviews.llvm.org/D16529
Files:
clang-tidy/modernize/CMakeLists.txt
clang-tidy/mod
LegalizeAdulthood updated this revision to Diff 47927.
LegalizeAdulthood marked 4 inline comments as done.
LegalizeAdulthood added a comment.
Update from review comments.
Avoid some string concatenation when delimiter is empty.
http://reviews.llvm.org/D16529
Files:
clang-tidy/modernize/CMakeL
ryee88 updated this revision to Diff 47926.
ryee88 added a comment.
Messed up the diff. Uploading it again.
http://reviews.llvm.org/D16965
Files:
include/clang/Basic/DiagnosticSemaKinds.td
lib/Sema/SemaDecl.cpp
test/Parser/forward-declaration.cpp
Index: test/Parser/forward-declaration.cp
ryee88 updated this revision to Diff 47925.
ryee88 added a comment.
Sorry-- re-uploading the diff. Didn't notice I dropped the code changes.
http://reviews.llvm.org/D16949
Files:
include/clang/Basic/DiagnosticSemaKinds.td
lib/Sema/SemaOverload.cpp
test/Parser/overloaded-pointer-vs-referen
ryee88 updated this revision to Diff 47924.
ryee88 added a comment.
Added unit test as per review comments.
http://reviews.llvm.org/D16949
Files:
test/Parser/overloaded-pointer-vs-reference-hint.cpp
Index: test/Parser/overloaded-pointer-vs-reference-hint.cpp
=
Hubert,
Ok on keeping the tests simple.
I removed the parentheses and transformed the fp checks into _Static_assert.
JT
On Sat, Feb 13, 2016 at 8:11 PM, Hubert Tong
wrote:
> On Sat, Feb 13, 2016 at 7:10 PM, Hubert Tong
> wrote:
>>
>> Hi Jorge,
>>
>> I do not intend to "overcomplicate" the te
hintonda updated this revision to Diff 47923.
hintonda added a comment.
- Only call HandleMemberFunctionDeclDelays() for valid NamedDecl pointers.
HandleMemberFunctionDeclDelays() adds LateParsedMethodDeclaration elements to
LateParsedDeclarations if the method has default args or
exception-s
joker.eph added a subscriber: joker.eph.
joker.eph added a comment.
What the build time of clang before and after? (Your goal is to improve build
time right?)
http://reviews.llvm.org/D17130
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
ht
ryee88 updated this revision to Diff 47922.
ryee88 added a comment.
Added unit tests.
http://reviews.llvm.org/D16965
Files:
test/Parser/forward-declaration.cpp
Index: test/Parser/forward-declaration.cpp
===
--- /dev/null
+++ tes
hubert.reinterpretcast added inline comments.
Comment at: lib/Frontend/InitPreprocessor.cpp:718
@@ -717,1 +717,3 @@
+ if (TI.hasFloat128Type())
+DefineFloatMacros(Builder, "FLT128", &TI.getFloat128Format(), "Q");
I am concerned that setting these macros are
On Sat, Feb 13, 2016 at 7:10 PM, Hubert Tong <
hubert.reinterpretc...@gmail.com> wrote:
> Hi Jorge,
>
> I do not intend to "overcomplicate" the testing as you put it, so there
> will be cases which are not caught by these tests.
> For *_MAX, if overflow occurs to the maximum positive finite value
Hi Jorge,
I do not intend to "overcomplicate" the testing as you put it, so there
will be cases which are not caught by these tests.
For *_MAX, if overflow occurs to the maximum positive finite value
(assuming no infinities), then it is possible for that value to be less
than 10^37 and the test wi
carlo.bertolli updated this revision to Diff 47921.
carlo.bertolli added a comment.
Reflect changes in http://reviews.llvm.org/D17148 and show full context.
Repository:
rL LLVM
http://reviews.llvm.org/D17170
Files:
include/clang/AST/StmtOpenMP.h
lib/CodeGen/CGOpenMPRuntime.cpp
lib/Code
Author: sfantao
Date: Sat Feb 13 17:35:10 2016
New Revision: 260837
URL: http://llvm.org/viewvc/llvm-project?rev=260837&view=rev
Log:
[OpenMP] Rename the offload entry points.
Summary:
Unlike other outlined regions in OpenMP, offloading entry points have to have
be visible (external linkage) for
Thanks Hubert.
I'm curious to see how you will handle corner cases without library
support, such as unsigned/signed zero and 0+x cases, with abs(x) <
EPS.
How does the parser/preprocessor interpret fp literals that are too
"precise" for that machine but are fine for other targets (cross
compile)?
Hi Jorge,
Looks fine to me. I'll work on committing this (with minor changes) over
the weekend.
Basically, I intend to remove some extraneous parentheses and adjust the *
_EPSILON, *_MIN and *_TRUE_MIN checks to reject values equal to 0.
-- HT
On Sat, Feb 13, 2016 at 2:33 PM, Jorge Teixeira
wr
Author: akirtzidis
Date: Sat Feb 13 15:46:50 2016
New Revision: 260833
URL: http://llvm.org/viewvc/llvm-project?rev=260833&view=rev
Log:
[AST] Add a print() method in DeclarationName that accepts a PrintingPolicy.
Modified:
cfe/trunk/include/clang/AST/DeclarationName.h
cfe/trunk/lib/AST/D
Hi,
this is my error message:
In file included from z.cc:1:
/code/build-llvm/bin/../include/c++/v1/ext/hash_map:213:5: warning:
Use of the header is deprecated. Migrate to
[-W#warnings]
# warning Use of the header is deprecated. Migrate
to
^
/code/build-llvm/bin/../include/c++/v1/ext/h
I guess refreshing the build directory fixed the bots now.
You may want to look into ccache as possibly the issue here.
> On Feb 13, 2016, at 12:08 PM, Argyrios Kyrtzidis wrote:
>
> clangIndex library is not part of that command so I don’t understand how my
> changes affect linking the clang bi
omtcyf0 created this revision.
omtcyf0 added reviewers: alexfh, LegalizeAdulthood, Eugene.Zelenko, djasper.
omtcyf0 added a subscriber: cfe-commits.
This patch introduces a proof-of-concept check for clang-tidy aiming to suggest
using ternary operator in order to improve readability.
It's curren
rjmccall added a comment.
Yes, thank you, that looks great. LGTM.
http://reviews.llvm.org/D16846
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
clangIndex library is not part of that command so I don’t understand how my
changes affect linking the clang binary, clangIndex is only used for libclang.
> On Feb 13, 2016, at 11:55 AM, Argyrios Kyrtzidis wrote:
>
> Sorry, I looked at it but it wasn’t clear to me what the problem is:
>
> FAIL
Sorry, I looked at it but it wasn’t clear to me what the problem is:
FAILED: :
/home/linaro/buildbot/clang-cmake-armv7-a15/stage1/tools/clang/clang.order:
file not recognized: File truncated
I thought it was some build SNAFU, do you have some hint on what I need to do
to fix this ?
> On Feb 1
Hubert,
You're right about the *_MIN relationships, and I fixed them on the
attached patch.
As for the enums, since there we're not even testing if the literals
are integers or fp numbers, and the Std. already reserves ranges for
implementation-specific values for some macros, it felt more natura
On Sat, Feb 13, 2016 at 1:11 PM, Benjamin Kramer via cfe-commits
wrote:
> Author: d0k
> Date: Sat Feb 13 12:11:49 2016
> New Revision: 260831
>
> URL: http://llvm.org/viewvc/llvm-project?rev=260831&view=rev
> Log:
> Fix a leak in the generated code for attributes with strings.
>
> Storing std::str
Author: d0k
Date: Sat Feb 13 12:11:49 2016
New Revision: 260831
URL: http://llvm.org/viewvc/llvm-project?rev=260831&view=rev
Log:
Fix a leak in the generated code for attributes with strings.
Storing std::strings in attributes simply doesn't work, we never call
the destructor. Use an array of Str
hintonda updated this revision to Diff 47914.
hintonda added a comment.
- Simplify diff
http://reviews.llvm.org/D17143
Files:
lib/Sema/SemaExprCXX.cpp
test/SemaCXX/pr25156-crash-on-invalid.cpp
Index: test/SemaCXX/pr25156-crash-on-invalid.cpp
dim updated this revision to Diff 47913.
dim added a comment.
Add checks for -lm and -lm_p, when profiling is on.
http://reviews.llvm.org/D16264
Files:
lib/Driver/ToolChains.cpp
lib/Driver/ToolChains.h
test/Driver/freebsd.cpp
Index: test/Driver/freebsd.cpp
===
+#if ((FLT_MIN < DBL_MIN) || (DBL_MIN < LDBL_MIN))
+#error "Mandatory macros {FLT,DBL,LDBL}_MIN are invalid."
This value again depends on the minimum exponent, and so the relationship
being tested here is not required to hold.
+#endif
For the enumeration-like cases, perhaps it would be better
Author: d0k
Date: Sat Feb 13 10:00:13 2016
New Revision: 260823
URL: http://llvm.org/viewvc/llvm-project?rev=260823&view=rev
Log:
RValue refs do not work that way.
Modified:
cfe/trunk/lib/CodeGen/CGCall.cpp
Modified: cfe/trunk/lib/CodeGen/CGCall.cpp
URL:
http://llvm.org/viewvc/llvm-project/
Author: d0k
Date: Sat Feb 13 09:49:17 2016
New Revision: 260822
URL: http://llvm.org/viewvc/llvm-project?rev=260822&view=rev
Log:
Don't copy a DenseMap just to do lookup in it.
Also remove the now unused isPodLike specialization. DenseMap only uses
it for copies.
Modified:
cfe/trunk/lib/Sema
On 12 February 2016 at 23:11, Argyrios Kyrtzidis via cfe-commits
wrote:
> Author: akirtzidis
> Date: Fri Feb 12 17:10:59 2016
> New Revision: 260760
>
> URL: http://llvm.org/viewvc/llvm-project?rev=260760&view=rev
> Log:
> [libclang] Separate the underlying indexing functionality of libclang and
Author: d0k
Date: Sat Feb 13 07:42:41 2016
New Revision: 260814
URL: http://llvm.org/viewvc/llvm-project?rev=260814&view=rev
Log:
Fix use after free.
Found by asan.
Modified:
cfe/trunk/lib/Lex/Pragma.cpp
Modified: cfe/trunk/lib/Lex/Pragma.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/tr
Author: d0k
Date: Sat Feb 13 07:42:54 2016
New Revision: 260815
URL: http://llvm.org/viewvc/llvm-project?rev=260815&view=rev
Log:
Reduce the number of implicit StringRef->std::string conversions by threading
StringRef through more APIs.
No functionality change intended.
Modified:
cfe/trunk/
Hi Tim,
as I did the change: Sorry that it broke for you on Darwin! I could not test
myself and there doesn't seem to be a build-bot...
To the change: I think the explicit passing of -stdlib=libc++ will override
CLANG_DEFAULT_CXX_STDLIB on Darwin.
Instead, Darwin::AddClangCXXStdlibIncludeArgs s
> The "hash_map" part of the name means they belong to "",
and the actual type is defined in that header.
Woops, Sorry I guess the name is used in both "std::__1::" in unordered_map
and "__gnu_cxx::" in . My intention in the changes was to
forward declare "std::__1::__hash_map_iterator".
On Sat,
> hash_map still looks broken to me.
Uh oh, Sorry :-S. The original breakage was caused by the removal of a
typedef "hash_map" used, but I'm sure I fixed that.
> should not they be prefixed with __gnu_cxx:: ?
No, they are in the correct namespace. The "hash_map" part of the name
means they belon
44 matches
Mail list logo
