Author: rsmith
Date: Tue Oct 31 21:52:12 2017
New Revision: 317076
URL: http://llvm.org/viewvc/llvm-project?rev=317076&view=rev
Log:
Fix -Wunused-private-field to fire regardless of which implicit special members
have been implicitly declared.
Modified:
cfe/trunk/lib/Sema/Sema.cpp
cfe/tr
Author: chapuni
Date: Tue Oct 31 21:43:22 2017
New Revision: 317075
URL: http://llvm.org/viewvc/llvm-project?rev=317075&view=rev
Log:
clang/lib/Format/Format.cpp: Fix warnings introduced in rL316903. [-Wpedantic]
Modified:
cfe/trunk/lib/Format/Format.cpp
Modified: cfe/trunk/lib/Format/Format
dcoughlin added a reviewer: george.karpenkov.
dcoughlin added a comment.
I think this is a great idea, and I expect it to find some nasty bugs!
However, I'm worried about false positives in the following not-too-uncommon
idiom:
dispatch_semaphore_t semaphore = dispatch_semaphore_create(0);
Author: marshall
Date: Tue Oct 31 21:03:35 2017
New Revision: 317074
URL: http://llvm.org/viewvc/llvm-project?rev=317074&view=rev
Log:
More statuses, a new patch, too
Modified:
libcxx/trunk/www/upcoming_meeting.html
Modified: libcxx/trunk/www/upcoming_meeting.html
URL:
http://llvm.org/viewv
Author: rtrieu
Date: Tue Oct 31 20:57:27 2017
New Revision: 317073
URL: http://llvm.org/viewvc/llvm-project?rev=317073&view=rev
Log:
Change assertion to quick exit from checking function.
Remove the assertion that could be triggered by invalid code. Replace it with
an early exit from the checkin
alexshap updated this revision to Diff 121100.
alexshap added a comment.
cleanup
Repository:
rL LLVM
https://reviews.llvm.org/D39438
Files:
lib/StaticAnalyzer/Checkers/StackAddrEscapeChecker.cpp
test/Analysis/stack-async-leak.m
Index: test/Analysis/stack-async-leak.m
===
alexshap updated this revision to Diff 121099.
alexshap added a comment.
more tests
Repository:
rL LLVM
https://reviews.llvm.org/D39438
Files:
lib/StaticAnalyzer/Checkers/StackAddrEscapeChecker.cpp
test/Analysis/stack-async-leak.m
Index: test/Analysis/stack-async-leak.m
Author: george.karpenkov
Date: Tue Oct 31 19:29:04 2017
New Revision: 317070
URL: http://llvm.org/viewvc/llvm-project?rev=317070&view=rev
Log:
[analyzer] Removing unused stored field.
Modified:
cfe/trunk/include/clang/Analysis/AnalysisDeclContext.h
cfe/trunk/lib/Analysis/AnalysisDeclConte
Author: ctopper
Date: Tue Oct 31 19:18:49 2017
New Revision: 317069
URL: http://llvm.org/viewvc/llvm-project?rev=317069&view=rev
Log:
[X86] Define i586 and pentium preprocessor defines for -march=lakemont to match
GCC
Modified:
cfe/trunk/lib/Basic/Targets/X86.cpp
cfe/trunk/test/Preproces
Author: rsmith
Date: Tue Oct 31 18:37:11 2017
New Revision: 317066
URL: http://llvm.org/viewvc/llvm-project?rev=317066&view=rev
Log:
[c++17] Refine resolution of constructor / conversion function disambiguation.
Given a choice between a constructor call and a conversion function in C++17,
we pref
This revision was automatically updated to reflect the committed changes.
Closed by commit rL317065: [Analyzer] Use value storage for BodyFarm (authored
by george.karpenkov).
Changed prior to commit:
https://reviews.llvm.org/D39428?vs=120861&id=121098#toc
Repository:
rL LLVM
https://reviews
Author: george.karpenkov
Date: Tue Oct 31 18:36:01 2017
New Revision: 317065
URL: http://llvm.org/viewvc/llvm-project?rev=317065&view=rev
Log:
[Analyzer] Use value storage for BodyFarm
Differential Revision: https://reviews.llvm.org/D39428
Modified:
cfe/trunk/include/clang/Analysis/AnalysisD
Author: marshall
Date: Tue Oct 31 18:27:25 2017
New Revision: 317063
URL: http://llvm.org/viewvc/llvm-project?rev=317063&view=rev
Log:
Update the synopsis to match the P/R of 2945. No code changes.
Modified:
libcxx/trunk/include/optional
Modified: libcxx/trunk/include/optional
URL:
http://
Author: arphaman
Date: Tue Oct 31 18:12:56 2017
New Revision: 317062
URL: http://llvm.org/viewvc/llvm-project?rev=317062&view=rev
Log:
[refactor][extract] code extracted from inline method should be placed
in a function defined before the outer class
Added:
cfe/trunk/test/Refactor/Extract/Fro
dcoughlin accepted this revision.
dcoughlin added a comment.
Looks good to me with the comments by Gabor addressed.
In https://reviews.llvm.org/D39428#912126, @george.karpenkov wrote:
> On top of that, reference is part of the type, not part of the variable name,
The reference is parsed as par
Author: arphaman
Date: Tue Oct 31 17:20:55 2017
New Revision: 317056
URL: http://llvm.org/viewvc/llvm-project?rev=317056&view=rev
Log:
[refactor][extract] prohibit extraction of ObjC property setters
Added:
cfe/trunk/test/Refactor/Extract/ObjCProperty.m
Modified:
cfe/trunk/include/clang/B
Author: arphaman
Date: Tue Oct 31 17:07:12 2017
New Revision: 317054
URL: http://llvm.org/viewvc/llvm-project?rev=317054&view=rev
Log:
[refactor][selection] code ranges can be selected in objc methods
Modified:
cfe/trunk/lib/Tooling/Refactoring/ASTSelection.cpp
cfe/trunk/unittests/Tooling
Author: wolfgangp
Date: Tue Oct 31 17:01:20 2017
New Revision: 317053
URL: http://llvm.org/viewvc/llvm-project?rev=317053&view=rev
Log:
Making a couple of tests a bit more flexible wrt thunk mangling. Fixes checkin
for r317047.
Modified:
cfe/trunk/test/CodeGenCXX/tmp-md-nodes1.cpp
cfe/tr
nathawes planned changes to this revision.
nathawes added a comment.
Thanks @arphaman! I'll work through your comments and update.
Comment at: include/clang/Index/IndexDataStoreSymbolUtils.h:13
+
+#include "indexstore/indexstore.h"
+#include "clang/Index/IndexSymbol.h"
efriedma added a comment.
> The internal canonical types are compared, so all this typedef sugar will be
> silently ignored.
Yes, and that's precisely the problem. On many 32-bit platforms, both "size_t"
and "uint32_t" are typedefs for "unsigned int"; on some 32-bit platforms,
"size_t" is an
Author: d0k
Date: Tue Oct 31 16:23:36 2017
New Revision: 317052
URL: http://llvm.org/viewvc/llvm-project?rev=317052&view=rev
Log:
Make helper function static. NFC.
Modified:
cfe/trunk/lib/AST/Type.cpp
Modified: cfe/trunk/lib/AST/Type.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/li
efriedma added a comment.
Please make these consistent with the non-__builtin versions.
Granted, I'm not sure all of those are right... I'm pretty sure, for example,
cbrt can't set errno, and carg can. But I'd prefer to deal with that in a
separate patch.
https://reviews.llvm.org/D39481
_
arphaman added inline comments.
Comment at: include/clang/Index/IndexDataStoreSymbolUtils.h:13
+
+#include "indexstore/indexstore.h"
+#include "clang/Index/IndexSymbol.h"
It looks to me like this header, `"indexstore/indexstore.h"`, and
`IndexDataStoreUtils.cpp`
This revision was automatically updated to reflect the committed changes.
Closed by commit rL317047: Fix for PR33930. Short-circuit metadata mapping when
cloning a varargs thunk. (authored by wolfgangp).
Changed prior to commit:
https://reviews.llvm.org/D39396?vs=120876&id=121078#toc
Repositor
Author: wolfgangp
Date: Tue Oct 31 15:49:48 2017
New Revision: 317047
URL: http://llvm.org/viewvc/llvm-project?rev=317047&view=rev
Log:
Fix for PR33930. Short-circuit metadata mapping when cloning a varargs thunk.
The cloning happens before all metadata nodes are resolved. Prevent the value
mapper
Author: vlad.tsyrklevich
Date: Tue Oct 31 15:39:44 2017
New Revision: 317044
URL: http://llvm.org/viewvc/llvm-project?rev=317044&view=rev
Log:
[CFI] Add CFI-icall pointer type generalization
Summary:
This change allows generalizing pointers in type signatures used for
cfi-icall by enabling the -f
spatel created this revision.
Herald added a subscriber: mcrosier.
As the description in Builtins.def says: "e = const, but only when
-fmath-errno=0".
This is step 2 of 3 to fix builtins and math calls as discussed in
https://reviews.llvm.org/D39204.
I went through the lists at:
http://en.cppr
sugak added inline comments.
Comment at: lib/Format/UsingDeclarationsSorter.cpp:50
std::string computeUsingDeclarationLabel(const FormatToken *UsingTok) {
assert(UsingTok && UsingTok->is(tok::kw_using) && "Expecting a using token");
std::string Label;
and
sugak created this revision.
This fixes clang-format internal assertion for the following code:
/* override */ using std::string;
Added add a unit test.
https://reviews.llvm.org/D39478
Files:
lib/Format/UsingDeclarationsSorter.cpp
unittests/Format/UsingDeclarationsSorterTest.cpp
Index
alexshap added inline comments.
Comment at: test/Analysis/stack-async-leak.m:60
+ void (^b)(void) = ^void(void) {
+*p = 1;
+ };
NoQ wrote:
> You may enjoy adding an extra note piece (as in D24278) to the offending
> statement within the block. It might be
smeenai added a comment.
This solves the issue in https://reviews.llvm.org/D39149, at least.
Repository:
rL LLVM
https://reviews.llvm.org/D39462
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinf
smeenai abandoned this revision.
smeenai added a comment.
I confirmed that these warnings go away with https://reviews.llvm.org/D39462
applied, and they reappear if I manually specify
`-Wmaybe-tautological-constant-compare`. Thank you!
https://reviews.llvm.org/D39149
___
alexshap updated this revision to Diff 121055.
alexshap added a comment.
Address the comments, add more tests
https://reviews.llvm.org/D39438
Files:
lib/StaticAnalyzer/Checkers/StackAddrEscapeChecker.cpp
test/Analysis/stack-async-leak.m
Index: test/Analysis/stack-async-leak.m
=
Author: vitalybuka
Date: Tue Oct 31 13:49:57 2017
New Revision: 317035
URL: http://llvm.org/viewvc/llvm-project?rev=317035&view=rev
Log:
[clang-fuzzer] Fix incremental builds of the fuzzer
Summary:
Don't use BUILD_IN_SOURCE keep git checkout clean
Don't forward CMAKE_GENERATOR as ExternalProject_
This revision was automatically updated to reflect the committed changes.
Closed by commit rL317035: [clang-fuzzer] Fix incremental builds of the fuzzer
(authored by vitalybuka).
Changed prior to commit:
https://reviews.llvm.org/D39445?vs=120926&id=121045#toc
Repository:
rL LLVM
https://rev
lebedev.ri updated this revision to Diff 121044.
lebedev.ri added a comment.
In https://reviews.llvm.org/D39462#912011, @rjmccall wrote:
> The actual choice of integer type is not stable across targets any more than
> the size is. In practice, people often don't use int and long, they use
> st
zaks.anna added a comment.
> Also if you check the code snippets in the coding standard:
> https://llvm.org/docs/CodingStandards.html#beware-unnecessary-copies-with-auto
> you can see that where we officially put the references.
Correct! The reference should not go with the type name.
George, p
rsmith closed this revision.
rsmith added a comment.
Committed as r317032, thanks!
https://reviews.llvm.org/D36101
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Author: rsmith
Date: Tue Oct 31 13:29:22 2017
New Revision: 317032
URL: http://llvm.org/viewvc/llvm-project?rev=317032&view=rev
Log:
Fix usage of right shift operator in fold expressions
The right shift operator was not seen as a valid operator in a fold expression,
which is PR32563.
Patch by N
xazax.hun added a comment.
In https://reviews.llvm.org/D39428#912126, @george.karpenkov wrote:
> @xazax.hun I'm really not convinced:
Unfortunately, the LLVM codebase right now does not strictly follow the style
guide. But clang-format puts the references next to the variable (or function
in
Author: spatel
Date: Tue Oct 31 13:19:39 2017
New Revision: 317031
URL: http://llvm.org/viewvc/llvm-project?rev=317031&view=rev
Log:
[CodeGen] map sqrt libcalls to llvm.sqrt when errno is not set
The LLVM sqrt intrinsic definition changed with:
D28797
...so we don't have to use any relaxed FP set
This revision was automatically updated to reflect the committed changes.
Closed by commit rL317031: [CodeGen] map sqrt libcalls to llvm.sqrt when errno
is not set (authored by spatel).
Changed prior to commit:
https://reviews.llvm.org/D39204?vs=120902&id=121042#toc
Repository:
rL LLVM
http
spatel added a comment.
In https://reviews.llvm.org/D39204#912084, @efriedma wrote:
> Let's do this one step at a time; first this patch, then fix the __builtin_*
> functions to use "e", then add all the missing cases
> CodeGenFunction::EmitBuiltinExpr.
Sounds good.
> This patch LGTM, assumi
george.karpenkov added a comment.
@xazax.hun I'm really not convinced:
george@/Volumes/Transcend/code/llvm (master)≻ rg "\w+\&"
tools/clang/include/clang/StaticAnalyzer
tools/clang/include/clang/StaticAnalyzer/Core/Checker.h
31: static void _checkDecl(void *checker, const Decl *D, Analysi
NoQ added a comment.
In https://reviews.llvm.org/D35109#837723, @NoQ wrote:
> It's something similar to assuming that the string length is within range [0,
> INT_MAX/4] in CStringChecker: we can easily assume that no overflow is
> happening in computations involving string lengths or iterator p
efriedma accepted this revision.
efriedma added a comment.
This revision is now accepted and ready to land.
Let's do this one step at a time; first this patch, then fix the __builtin_*
functions to use "e", then add all the missing cases
CodeGenFunction::EmitBuiltinExpr.
This patch LGTM, assumi
rnk added a comment.
I'd also like to highlight that LLVM has many users with different needs.
Sometimes we need to "disagree and commit". Flags are one of the primary ways
that we have to do that.
https://reviews.llvm.org/D39079
___
cfe-commits m
tmsriram added a comment.
Ping. How do we take this forward?
https://reviews.llvm.org/D39079
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
rjmccall added a comment.
The actual choice of integer type is not stable across targets any more than
the size is. In practice, people often don't use int and long, they use
standard typedefs like size_t and uint64_t, but the actual type chosen for
size_t is arbitrary when there are multiple
lebedev.ri added a comment.
In https://reviews.llvm.org/D39149#910825, @mclow.lists wrote:
> I dislike this change fairly strongly.
> I would much rather pursue a clang-based solution (since clang is being
> unhelpful here)
> Don't know if we can get one, though.
@smeenai
https://reviews.l
ABataev updated this revision to Diff 121018.
ABataev added a comment.
Added info about supported devices.
https://reviews.llvm.org/D39457
Files:
docs/OpenMPSupport.rst
docs/index.rst
Index: docs/index.rst
===
--- docs/index.
ABataev added a comment.
In https://reviews.llvm.org/D39457#911950, @Hahnfeld wrote:
> Do we want to add documentation about which target devices Clang can generate
> code for?
Ok
Comment at: docs/OpenMPSupport.rst:46
+
+* #pragma omp distribute parallel for [simd]: :partia
And i have again forgot to add the appropriate -commits list from the start.
Is there any reason at all why phab does not/should not do that automatically?
-- Forwarded message --
From: Roman Lebedev via Phabricator
Date: Tue, Oct 31, 2017 at 8:44 PM
Subject: [PATCH] D39462: [Sema
Hahnfeld added a comment.
Do we want to add documentation about which target devices Clang can generate
code for?
Comment at: docs/OpenMPSupport.rst:46
+
+* #pragma omp distribute parallel for [simd]: :partial:`Partial`. No full
codegen support.
+
I think tha
JonasToth added a comment.
@aaron.ballman could you take a (i think) finishing look on the check? Most
issues should be resolved and i think its ready for the finishing line :)
https://reviews.llvm.org/D37808
___
cfe-commits mailing list
cfe-commit
JonasToth updated this revision to Diff 121011.
JonasToth added a comment.
- fix rebasing issues in docs
https://reviews.llvm.org/D37808
Files:
clang-tidy/hicpp/CMakeLists.txt
clang-tidy/hicpp/HICPPTidyModule.cpp
clang-tidy/hicpp/MultiwayPathsCoveredCheck.cpp
clang-tidy/hicpp/MultiwayPa
JonasToth updated this revision to Diff 121009.
JonasToth added a comment.
- improved docs and comments
- remove some newlines
- use ternary operator instead of smallvec
https://reviews.llvm.org/D37808
Files:
clang-tidy/hicpp/CMakeLists.txt
clang-tidy/hicpp/HICPPTidyModule.cpp
clang-tidy/
SjoerdMeijer updated this revision to Diff 121003.
SjoerdMeijer added a comment.
Addressed the comments about portability, and added a note that Float16 is
available in both C and C++ mode.
https://reviews.llvm.org/D35295
Files:
docs/LanguageExtensions.rst
Index: docs/LanguageExtensions.rs
ABataev created this revision.
Some info about supported features of OpenMP 4.5-5.0.
https://reviews.llvm.org/D39457
Files:
docs/OpenMPSupport.rst
docs/index.rst
Index: docs/index.rst
===
--- docs/index.rst
+++ docs/index.rst
vsk planned changes to this revision.
vsk added a comment.
Apart from Alex's point, this patch is missing a compatibility test for v5 of
the profile format, and is missing coverage for the new entries in `HashType`.
Comment at: lib/CodeGen/CodeGenPGO.cpp:186
+if (Hash.getH
bob.wilson added a comment.
I'm excited to see some progress on this, but since there is overhead to adding
a new hashing scheme, I think we should do more before introducing a new
scheme. One of the problems with the previous scheme is that is did not take
into account nesting. Distinguishing
spatel added a comment.
In https://reviews.llvm.org/D39204#911316, @efriedma wrote:
> > I was going to include the builtins too, but that exposes another bug
> > (marked here with FIXME) - the builtins are all defined 'const'.
>
> Probably just need to change c->e in Builtins.def?
Yes - at lea
arphaman added inline comments.
Comment at: lib/CodeGen/CodeGenPGO.cpp:186
+if (Hash.getHashVersion() != PGO_HASH_V1)
+ Type = getHashType(Hash.getHashVersion(), S);
+
Am I missing something or will this call always return the same type as before
since
morehouse accepted this revision.
morehouse added a comment.
This revision is now accepted and ready to land.
LGTM.
https://reviews.llvm.org/D39445
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinf
alexshap added inline comments.
Comment at: lib/StaticAnalyzer/Checkers/StackAddrEscapeChecker.cpp:144
+return;
+ if (dyn_cast_or_null(Region->getMemorySpace())) {
+ExplodedNode *N = C.generateErrorNode();
NoQ wrote:
> `getMemorySpace()` is n
aaron.ballman added a comment.
In https://reviews.llvm.org/D39121#911745, @aaron.ballman wrote:
> Hmm, this is a good point -- I was thinking of the generic +N case with the
> original example, but with an explicit +1, you can't run into that situation
> with Win32 APIs. I will think on this a
kosarev created this revision.
kosarev added a project: clang.
The basic idea behind this patch is that since in strict aliasing mode all
accesses to union members require their outermost enclosing union objects to be
specified explicitly, then for a couple given accesses to union members of the
NoQ added a comment.
What i was trying to say is - Hey this check looks useful!~ Great idea.
Repository:
rL LLVM
https://reviews.llvm.org/D39438
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinf
NoQ added inline comments.
Comment at: lib/StaticAnalyzer/Checkers/StackAddrEscapeChecker.cpp:128
+ !Call.isGlobalCFunction("dispatch_async") &&
+ !Call.isGlobalCFunction("dispatch_once"))
+return;
`dispatch_once` isn't really asynchronous; i think
aaron.ballman added a comment.
In https://reviews.llvm.org/D39121#911744, @baloghadamsoftware wrote:
> In https://reviews.llvm.org/D39121#911741, @aaron.ballman wrote:
>
> > As I pointed out earlier in the thread, it is common to have
> > double-null-terminated strings in Win32 APIs. This is a c
baloghadamsoftware added a comment.
In https://reviews.llvm.org/D39121#911741, @aaron.ballman wrote:
> As I pointed out earlier in the thread, it is common to have
> double-null-terminated strings in Win32 APIs. This is a case where strlen(s +
> N) is valid. Since 1-byte strings would also be a
aaron.ballman added a comment.
In https://reviews.llvm.org/D39121#911736, @baloghadamsoftware wrote:
> Can you give me please an example where `malloc(strlen(s+1))` is intentional.
> It allocates `2` byte less than needed to store `s`. If it is really the goal
> (e.g. `s` has a `2` character pr
v.g.vassilev accepted this revision.
v.g.vassilev added a comment.
This revision is now accepted and ready to land.
LGTM!
https://reviews.llvm.org/D39416
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/l
baloghadamsoftware added a comment.
Can you give me please an example where `malloc(strlen(s+1))` is intentional.
It allocates `2` byte less than needed to store `s`. If it is really the goal
(e.g. `s` has a `2` character prefix which we do not want to copy) then the
correct solution is to use
NoQ added a comment.
Hey, i just recalled that we have documentation for `ExprInspection` functions
in `docs/analyzer/DebugChecks.rst`, you may want to add your function there as
well :)
Repository:
rL LLVM
https://reviews.llvm.org/D38844
___
c
aaron.ballman added a comment.
In https://reviews.llvm.org/D39121#910802, @baloghadamsoftware wrote:
> I thought on something like this, but I still do not like my phrasing:
>
> "Addition operator is applied to the argument of strlen(). instead of its
> result; move the '+ 1' outside of the call
xazax.hun added a comment.
In https://reviews.llvm.org/D37437#909537, @zaks.anna wrote:
> Just to be clear, since this leads to regression to the checker API, I am
> asking to look into other ways of solving this problem. For example, is there
> a way to ensure that the checker names are set at
Rakete updated this revision to Diff 120977.
Rakete added a comment.
Added some more tests.
If this patch is still ok can you please commit it @rsmith ? Thanks!
https://reviews.llvm.org/D36101
Files:
include/clang/Parse/Parser.h
lib/Parse/ParseExpr.cpp
test/Parser/cxx1z-fold-expre
CarlosAlbertoEnciso added a comment.
Hi Tamas,
Thanks very much for your message.
In https://reviews.llvm.org/D39239#910797, @tberghammer wrote:
> - Can you do a diff of the debug_info dump before and after your change?
> Understanding what have changed should give us a pretty good clue about
Rakete updated this revision to Diff 120966.
Rakete added a comment.
Used Parser's `GreaterThanIsOperator` just to be sure. :)
https://reviews.llvm.org/D36101
Files:
include/clang/Parse/Parser.h
lib/Parse/ParseExpr.cpp
test/Parser/cxx1z-fold-expressions.cpp
Index: test/Parser/cx
mstorsjo added a comment.
After trying to build a bit more code with clang and a static libcxx, I agree
with this change, so +1 from me. @compnerd ?
Repository:
rL LLVM
https://reviews.llvm.org/D33620
___
cfe-commits mailing list
cfe-commits@lis
Author: rengolin
Date: Tue Oct 31 05:21:32 2017
New Revision: 316991
URL: http://llvm.org/viewvc/llvm-project?rev=316991&view=rev
Log:
Merge r311574: ARM: explicitly specify the 8-byte alignment
Fixing the last libunwind failure on ARM.
Modified:
libunwind/branches/release_50/include/unwin
This revision was automatically updated to reflect the committed changes.
Closed by commit rL316988: [CodeGen] Propagate may-alias'ness of lvalues with
TBAA info (authored by kosarev).
Changed prior to commit:
https://reviews.llvm.org/D39008?vs=120828&id=120958#toc
Repository:
rL LLVM
https
Author: kosarev
Date: Tue Oct 31 04:05:34 2017
New Revision: 316988
URL: http://llvm.org/viewvc/llvm-project?rev=316988&view=rev
Log:
[CodeGen] Propagate may-alias'ness of lvalues with TBAA info
This patch fixes various places in clang to propagate may-alias
TBAA access descriptors during constru
schroedersi added a comment.
Ping
https://reviews.llvm.org/D30946
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
xazax.hun created this revision.
In order to reduce the load on the other code owners, I would like to step up
as an owner of the new library.
https://reviews.llvm.org/D39452
Files:
CODE_OWNERS.TXT
Index: CODE_OWNERS.TXT
===
-
xazax.hun added inline comments.
Comment at: lib/StaticAnalyzer/Checkers/StackAddrEscapeChecker.cpp:145
+ if (dyn_cast_or_null(Region->getMemorySpace())) {
+ExplodedNode *N = C.generateErrorNode();
+if (!N)
This will stop the analysis on this
baloghadamsoftware updated this revision to Diff 120954.
baloghadamsoftware added a comment.
Herald added subscribers: szepet, whisperity.
Tests updated.
https://reviews.llvm.org/D22374
Files:
lib/StaticAnalyzer/Core/ExprEngineCXX.cpp
test/Analysis/ctor.mm
Index: test/Analysis/ctor.mm
xazax.hun added inline comments.
Comment at: test/Analysis/string-with-signedness.c:1
+// RUN: %clang_analyze_cc1 -Wno-incompatible-library-redeclaration
-analyzer-checker=core,unix.cstring,alpha.unix.cstring -verify %s
+
NoQ wrote:
> We do have a warning for th
xazax.hun added inline comments.
Comment at: lib/StaticAnalyzer/Checkers/ExprInspectionChecker.cpp:68
// (globals should not be invalidated, etc), hence the use of evalCall.
- FnCheck Handler = llvm::StringSwitch(C.getCalleeName(CE))
-.Case("clang_analyzer_eval", &ExprIns
xazax.hun accepted this revision.
xazax.hun added a comment.
Some minor nits otherwise LGTM!
Comment at: include/clang/Analysis/AnalysisDeclContext.h:479
+ /// Get a reference to {@code BodyFarm} instance.
+ BodyFarm& getBodyFarm();
The reference is on the
baloghadamsoftware added a comment.
Any progress reviewing this? Iterator checkers are pending for more than half a
year because of this.
An alternative solution is to always extend the type and change
`ProgramState::assumeInbound()` so it does not move everything to the bottom of
the range, b
baloghadamsoftware added inline comments.
Comment at: lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp:595
+
+ if (origWidth < 128) {
+auto newWidth = std::max(2 * origWidth, (uint32_t) 8);
danielmarjamaki wrote:
> I would like that "128" is rew
baloghadamsoftware added a comment.
It seems that review on https://reviews.llvm.org/D35109 is stuck forever. So
maybe we should forget about this simplification and return to the local
solution I tried to use here originally. It is Part2, and we need to go through
all parts as soon as possible
lichray created this revision.
This is a side-effect brought in by p0620r0, which allows other placeholder
types (derived from `auto` and `decltype(auto)`) to be usable in a `new`
expression with a single-clause //braced-init-list// as its initializer (8.3.4
[expr.new]/2). N3922 defined its se
95 matches
Mail list logo
