rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
Comment at: tools/driver/driver.cpp:464-466
@@ +463,5 @@
+ PrependedOpts);
+ // Insert right after the program name to prepend to
rnk added inline comments.
Comment at: tools/driver/driver.cpp:451
@@ +450,3 @@
+ if (Tokenizer == &llvm::cl::TokenizeWindowsCommandLine) {
+// The first instance of '#' should be replaced with '='.
+auto AdjustOpt = [] (const char *Opt) {
I'd also rather
rnk created this revision.
rnk added reviewers: majnemer, hans.
rnk added a subscriber: cfe-commits.
The vtable takes its DLL storage class from the class, not the key
function. When they disagree, the vtable won't be exported by the DLL
that defines the key function. The easiest way to ensure tha
This revision was automatically updated to reflect the committed changes.
Closed by commit rL244488: [dllimport] A non-imported class with an imported
key can't have a key (authored by rnk).
Changed prior to commit:
http://reviews.llvm.org/D11913?vs=31697&id=31710#toc
Repository:
rL LLVM
ht
Author: rnk
Date: Mon Aug 10 14:39:01 2015
New Revision: 244488
URL: http://llvm.org/viewvc/llvm-project?rev=244488&view=rev
Log:
[dllimport] A non-imported class with an imported key can't have a key
Summary:
The vtable takes its DLL storage class from the class, not the key
function. When they
Author: rnk
Date: Mon Jul 18 12:49:40 2016
New Revision: 275826
URL: http://llvm.org/viewvc/llvm-project?rev=275826&view=rev
Log:
[clang-cl] Reinstate -gline-tables-only as a CoreOption while we sort out the
fate of /Zd
Modified:
cfe/trunk/include/clang/Driver/Options.td
Modified: cfe/trunk
rnk added a comment.
In https://reviews.llvm.org/D22426#486129, @DaveBartolomeo wrote:
> One reasonable solution would be to choose the toolset that is hosted on the
> same architecture as the host of clang.exe (e.g. x64-hosted Clang looks for
> x64-hosted MSVC). If this sounds good, I can make
rnk added a comment.
The approach makes sense to me, but the tests suggest it isn't doing what I'd
expect.
Comment at: test/CodeGenCXX/dllimport-rtti.cpp:7
@@ -6,3 +6,1 @@
} s;
-// MSVC: [[VF_S:.*]] = private unnamed_addr constant [2 x i8*]
-// MSVC-DAG: @"\01??_SS@@6B@" = unn
rnk added inline comments.
Comment at: test/CodeGenCXX/dllimport-rtti.cpp:7
@@ -6,3 +6,1 @@
} s;
-// MSVC: [[VF_S:.*]] = private unnamed_addr constant [2 x i8*]
-// MSVC-DAG: @"\01??_SS@@6B@" = unnamed_addr alias i8*, getelementptr inbounds
([2 x i8*], [2 x i8*]* [[VF_S]], i32 0
Author: rnk
Date: Wed Jul 20 09:40:25 2016
New Revision: 276120
URL: http://llvm.org/viewvc/llvm-project?rev=276120&view=rev
Log:
[MS] Improve VPtrInfo field names and doc comments
'ReusingBase' was a terrible name. It might actually refer to the most
derived class, which is not a base. 'BaseWith
rnk added inline comments.
Comment at: lib/CodeGen/CGDebugInfo.cpp:1096
@@ +1095,3 @@
+ QualType Ty = CGM.getContext().getTypeDeclType(RD);
+ llvm::DIType *nestedType = getOrCreateType(Ty, getOrCreateMainFile());
+ elements.push_back(nestedType);
Is getOrCreate
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
https://reviews.llvm.org/D22577
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Author: rnk
Date: Thu Jul 21 16:07:47 2016
New Revision: 276335
URL: http://llvm.org/viewvc/llvm-project?rev=276335&view=rev
Log:
Disable a flaky test on Windows that uses "echo >>"
Modified:
cfe/trunk/test/Modules/embed-files.cpp
Modified: cfe/trunk/test/Modules/embed-files.cpp
URL:
http:/
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
Repository:
rL LLVM
https://reviews.llvm.org/D21385
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailm
rnk added inline comments.
Comment at: test/CodeGenCUDA/kernel-args-alignment.cu:1-2
@@ +1,3 @@
+// REQUIRES: x86-registered-target
+// REQUIRES: nvptx-registered-target
+
Typically clang doesn't need a registered backend for a target to generate IR
for that targ
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
https://reviews.llvm.org/D22879
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
rnk created this revision.
rnk added reviewers: amccarth, aprantl, dblaikie, dexonsmith.
rnk added a subscriber: cfe-commits.
rnk added a dependency: D22884: [codeview] Emit vftable records.
MSVC emits codeview records describing all the methods contained in a
vftable. This is observable in windbg
rnk added a subscriber: rnk.
rnk added a comment.
Needs tests, though. The MSDN sample code is probably a good starting point.
Comment at: lib/Sema/SemaDeclAttr.cpp:2659
@@ -2658,1 +2658,3 @@
+ // code_seg only ever applies to functions.
+ if (Attr.getName()->getName() == "c
Author: rnk
Date: Fri Jul 29 14:43:28 2016
New Revision: 277205
URL: http://llvm.org/viewvc/llvm-project?rev=277205&view=rev
Log:
Fix naked.cpp test on 32-bit Windows
Modified:
cfe/trunk/test/CodeGenCXX/naked.cpp
Modified: cfe/trunk/test/CodeGenCXX/naked.cpp
URL:
http://llvm.org/viewvc/llvm
rnk added a comment.
I think the correctness problem is in DSE. It should be safe for clang to put
'tail' here. The tail call does not capture the address of any local variables,
but it does load from them via 'byval'.
https://reviews.llvm.org/D22900
rnk added a comment.
I think Nick was responsible for adding this DSE optimization *years* ago.
https://reviews.llvm.org/D22900
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Author: rnk
Date: Mon Aug 1 13:56:13 2016
New Revision: 277367
URL: http://llvm.org/viewvc/llvm-project?rev=277367&view=rev
Log:
[codeview] Skip injected class names in nested record emission
We were already trying to do this, but our check wasn't quite right.
Fixes PR28790
Added:
cfe/trun
rnk added a comment.
So, if clang were to use a temporary alloca for the byval parameter, then yes,
I agree marking it as a tail call would be incorrect. However, clang doesn't
use an alloca, it forwards the byval pointer parameter directly to the callee:
define i32 @_ZThn4_N1C4SeekE6_LARGE(%
rnk added a comment.
Thanks! lgtm
Repository:
rL LLVM
https://reviews.llvm.org/D23105
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Author: rnk
Date: Tue Aug 9 12:23:56 2016
New Revision: 278139
URL: http://llvm.org/viewvc/llvm-project?rev=278139&view=rev
Log:
[clang-cl] Make -gline-tables-only imply -gcodeview
It's surprising that you have to pass /Z7 in addition to -gcodeview to
get debug info. The sanitizer runtime, for e
rnk added inline comments.
Comment at: lib/Driver/Driver.cpp:93
@@ +92,3 @@
+ ArrayRef Args) {
+ auto Default = ToolChain::getTargetAndModeFromProgramName(ProgramName);
+ StringRef DefaultMode(Default.second);
Why not change ToolInvoc
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
https://reviews.llvm.org/D23454
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
rnk added a subscriber: rnk.
rnk added a comment.
We should convince cmake to emit "arguments" instead of "command" so that we
don't have this ambiguity.
https://reviews.llvm.org/D23455
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http:/
rnk added a comment.
Sorry for the delay, was trying to stay focused on debugging
Comment at: clang/include/clang/AST/Decl.h:1647
@@ +1646,3 @@
+ /// clears this list) needs to be a logically-const operation.
+ mutable std::unique_ptr> DeferredDiags;
+
Decl me
rnk added a comment.
+cmake people for the feasibility of emitting 'arguments' instead of 'command'
into the JSON compilation database.
https://reviews.llvm.org/D23455
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cg
rnk added a comment.
In https://reviews.llvm.org/D23455#515486, @brad.king wrote:
> > the feasibility of emitting 'arguments' instead of 'command' into the JSON
> > compilation database.
>
>
> CMake constructs the command lines internally using string replacement on
> templates. We never actua
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
https://reviews.llvm.org/D23241
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
rnk added inline comments.
Comment at: include/clang/AST/ASTContext.h:1771-1782
@@ -1770,1 +1770,14 @@
+ /// Return true is the given types are compatible in C from MSVC's point of
+ /// view.
+ //
+ // Conditions:
+ // 1. Both types are equally-qualified, sized and align
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
Should I land this for you?
http://reviews.llvm.org/D16779
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/
Author: rnk
Date: Mon Feb 1 12:58:24 2016
New Revision: 259383
URL: http://llvm.org/viewvc/llvm-project?rev=259383&view=rev
Log:
Fix attribute((mode([word|unwind_word]))) for x32
Patch by H.J. Lu
```
typedef unsigned int gcc_word __attribute__((mode(word)));
```
and
```
typedef unsigned int gc
This revision was automatically updated to reflect the committed changes.
Closed by commit rL259383: Fix attribute((mode([word|unwind_word]))) for x32
(authored by rnk).
Changed prior to commit:
http://reviews.llvm.org/D16779?vs=46558&id=46559#toc
Repository:
rL LLVM
http://reviews.llvm.org
rnk added a comment.
If the ABI is still open to small changes and clarifications, can we make sure
that C and C++ empty structs are passed the same way? Unfortunately, on normal
x86_64 SysV GCC doesn't pass them the same way and Clang has to follow suit.
http://reviews.llvm.org/D16808
rnk added inline comments.
Comment at: include/clang/AST/ASTContext.h:1772-1783
@@ -1771,1 +1771,14 @@
+ /// Return true is the given typedef types are compatible in C from MSVC's
+ /// point of view.
+ //
+ // Conditions:
+ // 1. Both typedef types are either integer, en
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
http://reviews.llvm.org/D16626
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
http://reviews.llvm.org/D16761
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Author: rnk
Date: Mon Feb 8 20:51:17 2016
New Revision: 260194
URL: http://llvm.org/viewvc/llvm-project?rev=260194&view=rev
Log:
Avoid forcing emission of delayed dllexported classes on template instantiation
Fixes PR26490
Modified:
cfe/trunk/lib/Sema/SemaTemplateInstantiate.cpp
cfe/tru
Author: rnk
Date: Tue Feb 9 11:48:27 2016
New Revision: 260241
URL: http://llvm.org/viewvc/llvm-project?rev=260241&view=rev
Log:
Revert "Avoid forcing emission of delayed dllexported classes on template
instantiation"
This reverts commit r260194.
It caused PR26549. There's probably a better wa
Author: rnk
Date: Tue Feb 9 13:53:30 2016
New Revision: 260267
URL: http://llvm.org/viewvc/llvm-project?rev=260267&view=rev
Log:
Pass /bigobj when building lib/ASTMatchers/Dynamic/Registry.cpp
This is the third time it has crossed the 2^16 section limit. We've
already spent time optimizing this
On Tue, Feb 9, 2016 at 12:32 PM, Aaron Ballman
wrote:
> On Tue, Feb 9, 2016 at 2:53 PM, Reid Kleckner via cfe-commits
> wrote:
> > Author: rnk
> > Date: Tue Feb 9 13:53:30 2016
> > New Revision: 260267
> >
> > URL: http://llvm.org/viewvc/llvm-project?r
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
Thanks, lgtm!
http://reviews.llvm.org/D15373
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
http://reviews.llvm.org/D13622
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm if you clarify that bit
Comment at: lib/CodeGen/MicrosoftCXXABI.cpp:1676
@@ -1675,2 +1675,3 @@
- llvm::GlobalValue::LinkageTypes VFTableLinkage = CGM.getVTableLinkage(RD);
+
Author: rnk
Date: Wed Feb 10 13:09:15 2016
New Revision: 260411
URL: http://llvm.org/viewvc/llvm-project?rev=260411&view=rev
Log:
Silence some MSVC false positive warnings about integer zexts and falling off
the end of a covered switch
Modified:
cfe/trunk/lib/CodeGen/MicrosoftCXXABI.cpp
Author: rnk
Date: Wed Feb 10 15:28:38 2016
New Revision: 260435
URL: http://llvm.org/viewvc/llvm-project?rev=260435&view=rev
Log:
[clang-cl] /Z7 now generates normal debug info, not just line info
Previously LLVM could not process any debug info we produced, so it
didn't make sense to spend time
rnk added a comment.
Ultimately, Sema should be responsible for rejecting this, correct? In the
meantime we can have CodeGen reject this and emit a null value to avoid
crashing.
Comment at: lib/CodeGen/CGCUDABuiltin.cpp:105
@@ -99,3 +104,3 @@
llvm::Value *P = Builder.Cr
rnk added inline comments.
Comment at: lib/CodeGen/CGCUDABuiltin.cpp:105
@@ -99,3 +104,3 @@
llvm::Value *P = Builder.CreateStructGEP(AllocaTy, Alloca, I - 1);
llvm::Value *Arg = Args[I].RV.getScalarVal();
Builder.CreateAlignedStore(Arg, P,
DL.getPrefTypeAlign
Author: rnk
Date: Wed Feb 10 18:14:04 2016
New Revision: 260472
URL: http://llvm.org/viewvc/llvm-project?rev=260472&view=rev
Log:
Fix two tests relying on LLVM -O1 behavior
Something changed the inference of nonnull.
Modified:
cfe/trunk/test/CodeGenCXX/init-invariant.cpp
cfe/trunk/test/C
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
http://reviews.llvm.org/D16808
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
Comment at: lib/CodeGen/CGCUDABuiltin.cpp:90
@@ +89,3 @@
+CGM.ErrorUnsupported(E, "non-scalar arg to printf");
+return RValue::getIgnored();
+ }
Doesn
rnk added a subscriber: rnk.
rnk accepted this revision.
rnk added a reviewer: rnk.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
This code seems like it's intended to avoid operating on dependent types, so it
shouldn't need the template keyword. It's hard to have a
Author: rnk
Date: Tue Feb 16 13:16:20 2016
New Revision: 260993
URL: http://llvm.org/viewvc/llvm-project?rev=260993&view=rev
Log:
[typo-correction] Apply name specifier corrections when forming a NNS
Previously we would leave behind the old name specifier prefix, which
creates an invalid AST. Ot
Author: rnk
Date: Tue Feb 16 13:16:28 2016
New Revision: 260994
URL: http://llvm.org/viewvc/llvm-project?rev=260994&view=rev
Log:
Exclude PCH/missing-file.cpp on Windows, it does not pass reliably
Tag the test with "REQUIRES: can-remove-opened-file", which is what we
use for the similar test Modu
Author: rnk
Date: Tue Feb 16 14:34:27 2016
New Revision: 261005
URL: http://llvm.org/viewvc/llvm-project?rev=261005&view=rev
Log:
Stop using "template" when printing qualtype names
Summary:
The keyword "template" isn't necessary when
printing a fully-qualified qualtype name, and, in fact,
results
This revision was automatically updated to reflect the committed changes.
Closed by commit rL261005: Stop using "template" when printing qualtype names
(authored by rnk).
Changed prior to commit:
http://reviews.llvm.org/D17214?vs=47854&id=48095#toc
Repository:
rL LLVM
http://reviews.llvm.or
rnk added a comment.
Thanks, but I don't think we should do this. There's already a lot of code out
there (ffmpeg) that is already using clang its current form and it does stuff
like this:
#if defined(_MSC_VER) && !defined(__clang__)
# define inline __inline
// More MSVC compatibility hac
rnk created this revision.
rnk added reviewers: thakis, rtrieu.
rnk added a subscriber: cfe-commits.
This only warns on casts of the address of a function defined in the
current TU. In this case, the fix is likely to be local and the warning
useful.
Here are some things we could experiment with i
This revision was automatically updated to reflect the committed changes.
Closed by commit rL261290: Add call to find_package to load LLVM dependencies
(authored by rnk).
Changed prior to commit:
http://reviews.llvm.org/D13622?vs=37031&id=48431#toc
Repository:
rL LLVM
http://reviews.llvm.or
Author: rnk
Date: Thu Feb 18 18:56:56 2016
New Revision: 261290
URL: http://llvm.org/viewvc/llvm-project?rev=261290&view=rev
Log:
Add call to find_package to load LLVM dependencies
ClangConfig requires LLVMConfig, so add find_package call in
ClangConfig so find_package(clang REQUIRED CONFIG) will
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
Looks good, thanks. I like the new approach of avoiding the creation of a
LateParsedMethodDeclaration if we couldn't get a NamedDecl back. I'll land it
soon.
http://reviews.llvm.org/D17072
This revision was automatically updated to reflect the committed changes.
Closed by commit rL261292: [Sema] PR25181 Fix crash when method declaration
with throw spec fails to… (authored by rnk).
Changed prior to commit:
http://reviews.llvm.org/D17072?vs=47923&id=48435#toc
Repository:
rL LLVM
Author: rnk
Date: Thu Feb 18 19:15:08 2016
New Revision: 261292
URL: http://llvm.org/viewvc/llvm-project?rev=261292&view=rev
Log:
[Sema] PR25181 Fix crash when method declaration with throw spec fails to parse
correctly
Fixes crash referenced in PR25181 where dyn_cast is called on a null
instanc
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
http://reviews.llvm.org/D17313
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
rnk added a comment.
I think we should do this because MSVC doesn't make _Static_assert available to
C code. David says that, according to the C standard, assert.h is supposed to
`#define static_assert _Static_assert`. MSVC doesn't do that because they
provide static_assert directly as a keywor
An alternative fix would be to interpose MSVC's assert.h in clang's builtin
headers with #include_next, and do the define there. I don't like
interposing headers if we can avoid it though.
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lis
rnk updated this revision to Diff 55044.
rnk added a comment.
- Implement suggestions to avoid warning twice
http://reviews.llvm.org/D18271
Files:
include/clang/Basic/Diagnostic.h
include/clang/Basic/DiagnosticGroups.td
include/clang/Basic/DiagnosticSemaKinds.td
include/clang/Sema/Sema.
rnk added a subscriber: rnk.
rnk added a comment.
+1 for canCallMismatchedFunctionType.
I guess there's a more general issue here which is that LLVM appears to be more
permissive about prototype mismatch than wasm. Specifically, I'm thinking about
how swifterror relies on being able to pass ext
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm, thanks this is a nice cleanup!
Comment at: include/clang/Sema/Sema.h:338-342
@@ +337,7 @@
+PSK_Reset = 0x0,// #pragma ()
+PSK_Set = 0x1,
rnk updated this revision to Diff 55359.
rnk added a comment.
- address comments
http://reviews.llvm.org/D18271
Files:
include/clang/Basic/Diagnostic.h
include/clang/Basic/DiagnosticGroups.td
include/clang/Basic/DiagnosticSemaKinds.td
include/clang/Sema/Sema.h
lib/Sema/AnalysisBasedWa
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
Comment at: lib/CodeGen/CodeGenModule.cpp:492-498
@@ +491,9 @@
+ if (uint32_t PLevel = Context.getLangOpts().PIELevel) {
+llvm::PIELevel::Level PL = llvm::PIELevel::Defaul
rnk added a comment.
still lgtm
Comment at: lib/CodeGen/CodeGenModule.cpp:480
@@ -479,8 +479,3 @@
if (uint32_t PLevel = Context.getLangOpts().PICLevel) {
-llvm::PICLevel::Level PL = llvm::PICLevel::Default;
-switch (PLevel) {
-case 0: break;
-case 1: PL = llvm
This revision was automatically updated to reflect the committed changes.
Closed by commit rL267957: Avoid -Wshadow warnings about constructor parameters
named after fields (authored by rnk).
Changed prior to commit:
http://reviews.llvm.org/D18271?vs=55359&id=55515#toc
Repository:
rL LLVM
h
Author: rnk
Date: Thu Apr 28 19:37:43 2016
New Revision: 267957
URL: http://llvm.org/viewvc/llvm-project?rev=267957&view=rev
Log:
Avoid -Wshadow warnings about constructor parameters named after fields
Usually these parameters are used solely to initialize the field in the
initializer list, and t
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
Comment at: test/SemaCXX/pr27047-default-init-expr-name-conflict.cpp:6
@@ +5,3 @@
+template
+struct B {
+
More reduced:
template struct A { int A = 0; };
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
Comment at: include/clang/Sema/Sema.h:354-357
@@ -356,6 +353,6 @@
};
void Act(SourceLocation PragmaLocation,
PragmaMsStackAction Action,
l
Author: rnk
Date: Fri Apr 29 13:06:53 2016
New Revision: 268082
URL: http://llvm.org/viewvc/llvm-project?rev=268082&view=rev
Log:
Fix crash in BuildCXXDefaultInitExpr.
Fix crash in BuildCXXDefaultInitExpr when member of template class has
same name as the class itself.
Based on patch by Raphael
This revision was automatically updated to reflect the committed changes.
Closed by commit rL268082: Fix crash in BuildCXXDefaultInitExpr. (authored by
rnk).
Changed prior to commit:
http://reviews.llvm.org/D19721?vs=55580&id=55631#toc
Repository:
rL LLVM
http://reviews.llvm.org/D19721
Fil
rnk accepted this revision.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
http://reviews.llvm.org/D19749
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
rnk created this revision.
rnk added reviewers: majnemer, hans.
rnk added a subscriber: cfe-commits.
Before this change, we would pass all non-HFA record arguments on
Windows with byval. Byval often blocks optimizations and results in bad
code generation. Windows now uses the existing workaround t
rnk marked an inline comment as done.
rnk added a comment.
In http://reviews.llvm.org/D19756#417718, @hans wrote:
> This is awesome! lgtm
Great!
> Want to reference PR27522 in the patch description?
>
> Also in the description:
>
> > I also expanded the workaround handle C++ records with
Author: rnk
Date: Mon May 2 12:41:07 2016
New Revision: 268261
URL: http://llvm.org/viewvc/llvm-project?rev=268261&view=rev
Log:
Expand aggregate arguments more often on 32-bit Windows
Before this change, we would pass all non-HFA record arguments on
Windows with byval. Byval often blocks optimi
This revision was automatically updated to reflect the committed changes.
Closed by commit rL268261: Expand aggregate arguments more often on 32-bit
Windows (authored by rnk).
Changed prior to commit:
http://reviews.llvm.org/D19756?vs=55695&id=55845#toc
Repository:
rL LLVM
http://reviews.ll
rnk added a subscriber: rnk.
rnk added a comment.
I think threading this through PP options is reasonable.
Comment at: test/PCH/pragma-once.h:1
@@ +1,2 @@
+#pragma once
+
This should be in test/PCH/Inputs
http://reviews.llvm.org/D19815
_
Author: rnk
Date: Mon May 2 17:42:34 2016
New Revision: 268321
URL: http://llvm.org/viewvc/llvm-project?rev=268321&view=rev
Log:
Fix argument expansion of reference fields of structs
r268261 made Clang "expand" more struct arguments on Windows. It removed
the check for 'RD->isCLike()', which was
rnk added a comment.
I think Richard has a counterexample that shows that the "NullOut" approach to
computing abi_tag sets isn't the right way to go. I wasn't able to craft it
myself, but I figured I should send along the feedback that maybe a separate,
up-front pass over the return type with a
Author: rnk
Date: Tue May 3 13:44:29 2016
New Revision: 268418
URL: http://llvm.org/viewvc/llvm-project?rev=268418&view=rev
Log:
[MS] Pass CalleeDecl to adjustThisArgumentForVirtualFunctionCall
If we are devirtualizing, then we want to compute the 'this' adjustment
of the devirtualized target, n
Author: rnk
Date: Tue May 3 13:48:50 2016
New Revision: 268419
URL: http://llvm.org/viewvc/llvm-project?rev=268419&view=rev
Log:
Fix use of LLVM IR names in lit test
Modified:
cfe/trunk/test/CodeGenCXX/microsoft-abi-virtual-inheritance.cpp
Modified: cfe/trunk/test/CodeGenCXX/microsoft-abi-v
Author: rnk
Date: Tue May 3 21:58:24 2016
New Revision: 268473
URL: http://llvm.org/viewvc/llvm-project?rev=268473&view=rev
Log:
[X86] Add -malign-double support
The -malign-double flag causes i64 and f64 types to have alignment 8
instead of 4. On x86-64, the behavior of -malign-double is enable
rnk added a subscriber: rnk.
rnk accepted this revision.
rnk added a reviewer: rnk.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
To be clear, loop unrolling just lowers its size threshold when -Os is on:
// Apply size attributes
if (L->getHeader()->getParent()->
rnk added inline comments.
Comment at: lib/AST/MicrosoftMangle.cpp:1583-1584
@@ -1579,2 +1582,4 @@
case QMM_Result:
+// Presence of __unaligned qualifier shouldn't affect mangling here.
+Quals.removeUnaligned();
if ((!IsPointer && Quals) || isa(T)) {
--
FYI, this change broke bionic, which has this exact pattern:
https://android.googlesource.com/platform/bionic/+/master/tools/relocation_packer/src/debug.h#84
struct A {
enum E {
X = 3
};
};
typedef A::E T;
using T::X;
I expect users are going to have lots of issues with this, and it proba
On Thu, May 5, 2016 at 11:15 AM, Richard Smith
wrote:
> Given that GCC rejects this code (and has done for as long as it's
> accepted the Enum::Member syntax), I don't expect the problems to be too
> widespread. If they are, we could trivially accept this as an extension; it
> seems like a pointl
rnk accepted this revision.
rnk added a reviewer: rnk.
rnk added a comment.
lgtm
http://reviews.llvm.org/D19275
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
rnk updated this revision to Diff 56802.
rnk marked 3 inline comments as done.
rnk added a comment.
- address comments
http://reviews.llvm.org/D17348
Files:
include/clang/Basic/DiagnosticSemaKinds.td
lib/Sema/SemaCast.cpp
test/Sema/callingconv-cast.c
Index: test/Sema/callingconv-cast.c
=
rnk added inline comments.
Comment at: include/clang/Basic/DiagnosticGroups.td:294
@@ -293,2 +293,3 @@
def SelTypeCast : DiagGroup<"cast-of-sel-type">;
+def CastCallingConvention : DiagGroup<"cast-calling-convention">;
def FunctionDefInObjCContainer : DiagGroup<"function-def-in-
Author: rnk
Date: Tue May 10 16:00:03 2016
New Revision: 269116
URL: http://llvm.org/viewvc/llvm-project?rev=269116&view=rev
Log:
Add -Wcast-calling-convention to warn when casting away calling conventions
Summary:
This only warns on casts of the address of a function defined in the
current TU. I
901 - 1000 of 1562 matches
Mail list logo
