[cfe-commits] r166039 - in /cfe/trunk: include/clang/Basic/DiagnosticGroups.td include/clang/Basic/DiagnosticSemaKinds.td lib/Sema/SemaCast.cpp test/Analysis/CFContainers.mm test/Analysis/idempotent-operations.c test/Analysis/misc-ps-region-store

[Matt Beaumont-Gay]

On Wed, Oct 17, 2012 at 11:25 AM, David Blaikie <dblaikie at gmail.com> wrote:
> On Wed, Oct 17, 2012 at 11:16 AM, Matt Beaumont-Gay
> <matthewbg at google.com> wrote:
>> On Tue, Oct 16, 2012 at 11:53 AM, David Blaikie <dblaikie at gmail.com> wrote:
>>> Author: dblaikie
>>> Date: Tue Oct 16 13:53:14 2012
>>> New Revision: 166039
>>>
>>> URL: http://llvm.org/viewvc/llvm-project?rev=166039&view=rev
>>> Log:
>>> Implement GCC's -Wint-to-pointer-cast.
>>>
>>> This implementation doesn't warn on anything that GCC doesn't warn on with the
>>> exception of templates specializations (GCC doesn't warn, Clang does). The
>>> specific skipped cases (boolean, constant expressions, enums) are open for
>>> debate/adjustment if anyone wants to demonstrate that GCC is being overly
>>> conservative here. The only really obvious false positive I found was in the
>>> Clang regression suite's MPI test - apparently MPI uses specific flag values in
>>> pointer constants. (eg: #define FOO (void*)~0)
>>>
>>> Modified:
>>>     cfe/trunk/include/clang/Basic/DiagnosticGroups.td
>>>     cfe/trunk/include/clang/Basic/DiagnosticSemaKinds.td
>>>     cfe/trunk/lib/Sema/SemaCast.cpp
>>>     cfe/trunk/test/Analysis/CFContainers.mm
>>>     cfe/trunk/test/Analysis/idempotent-operations.c
>>>     cfe/trunk/test/Analysis/misc-ps-region-store.m
>>>     cfe/trunk/test/Analysis/taint-tester.c
>>>     cfe/trunk/test/CXX/dcl.dcl/dcl.spec/dcl.constexpr/p5.cpp
>>>     cfe/trunk/test/Sema/block-return.c
>>>     cfe/trunk/test/Sema/cast.c
>>>     cfe/trunk/test/Sema/i-c-e.c
>>>     cfe/trunk/test/SemaCXX/cast-conversion.cpp
>>>     cfe/trunk/test/SemaCXX/decl-expr-ambiguity.cpp
>>>
>>> Modified: cfe/trunk/include/clang/Basic/DiagnosticGroups.td
>>> URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/include/clang/Basic/DiagnosticGroups.td?rev=166039&r1=166038&r2=166039&view=diff
>>> ==============================================================================
>>> --- cfe/trunk/include/clang/Basic/DiagnosticGroups.td (original)
>>> +++ cfe/trunk/include/clang/Basic/DiagnosticGroups.td Tue Oct 16 13:53:14 2012
>>> @@ -365,6 +365,8 @@
>>>
>>>  def TypeSafety : DiagGroup<"type-safety">;
>>>
>>> +def IntToPointerCast : DiagGroup<"int-to-pointer-cast">;
>>> +
>>>  def Extra : DiagGroup<"extra", [
>>>      MissingFieldInitializers,
>>>      IgnoredQualifiers,
>>> @@ -412,7 +414,7 @@
>>>  // Note that putting warnings in -Wall will not disable them by default. If a
>>>  // warning should be active _only_ when -Wall is passed in, mark it as
>>>  // DefaultIgnore in addition to putting it here.
>>> -def : DiagGroup<"all", [Most, Parentheses, Switch]>;
>>> +def : DiagGroup<"all", [Most, Parentheses, Switch, IntToPointerCast]>;
>>
>> Any reason not to put the new flag in -Wmost?
>
> It's on by default. The only reason it's in -Wall is because that's
> where it is in GCC. I suppose that doesn't rule out putting it in
> -Wmost instead, though - not sure what the tradeoffs/benefits are
> here.

-Wmost seems to be where we have the big list of random DiagGroups.
I'd just as soon only have one of those, rather than having specific
warnings arbitrarily placed in -Wmost or -Wall.