On 06/02/2014 01:14 AM, Paul Eggert wrote:
> Pádraig Brady wrote:
>> Well the ; is needed in C++ but optional in C.
>> I was worried about users leaving out the ; by mistake
>> and not noticing in the normal case of compiling in C.
>
> But then the other case should be "# define _GL_UNUSED_LABEL ;
Another argument for omitting the semicolon is, suppose we add another
label-related attribute later? _GL_HOT, say? Then, 'lab: _GL_HOT
_GL_UNUSED_LABEL' would work, but 'lab: _GL_UNUSED_LABEL _GL_HOT' would
not, and users would have to remember to always put _GL_UNUSED_LABEL
last among all t
Pádraig Brady wrote:
Well the ; is needed in C++ but optional in C.
I was worried about users leaving out the ; by mistake
and not noticing in the normal case of compiling in C.
But then the other case should be "# define _GL_UNUSED_LABEL ;", no?
Otherwise a semicolon would be appended sometim
On 06/02/2014 12:53 AM, Paul Eggert wrote:
> Paul Eggert wrote:
>> Pádraig Brady wrote:
>>
>>> +# define _GL_UNUSED_LABEL _GL_UNUSED;
>>
>> Why is there a semicolon at the end of that macro definition?
>
> I removed it just now (since I was syncing to Emacs and I couldn't stand
> seeing the typo
Paul Eggert wrote:
Pádraig Brady wrote:
+# define _GL_UNUSED_LABEL _GL_UNUSED;
Why is there a semicolon at the end of that macro definition?
I removed it just now (since I was syncing to Emacs and I couldn't stand
seeing the typo there...).
Pádraig Brady wrote:
+# define _GL_UNUSED_LABEL _GL_UNUSED;
Why is there a semicolon at the end of that macro definition? It might
cause problems, no? _GL_UNUSED_LABEL is supposed to not affect the
semantics of the program, but with a semicolon there it could change things.
On 06/01/2014 11:39 PM, Ben Walton wrote:
> * lib/acl-internal.h (acl_nontrivial, acl_ace_nontrivial): Mark
>
> Signed-off-by: Ben Walton
> ---
> lib/acl-internal.h | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/lib/acl-internal.h b/lib/acl-internal.h
> index f
Ben Walton wrote:
WIth your suggested followup
(prerequisite), I think that it's a better solution.
Better yet, how about putting the goto-containing code into a subsdiary
static function that can return rather than goto? That'd be cleaner anyway.
On Sun, Jun 1, 2014 at 9:59 PM, Pádraig Brady wrote:
> On 06/01/2014 09:34 AM, Ben Walton wrote:
>> * Avoid possible compiler warnings/errors by defining the out label
>> only when it may be accessed.
>>
>> Signed-off-by: Ben Walton
>> ---
>>
>> Hi All,
>>
>> When building coreutils 8.22 on
* lib/acl-internal.h (acl_nontrivial, acl_ace_nontrivial): Mark
Signed-off-by: Ben Walton
---
lib/acl-internal.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/acl-internal.h b/lib/acl-internal.h
index fdffe64..b238006 100644
--- a/lib/acl-internal.h
+++ b/lib/ac
On 06/01/2014 09:59 PM, Pádraig Brady wrote:
> On 06/01/2014 09:34 AM, Ben Walton wrote:
>> * Avoid possible compiler warnings/errors by defining the out label
>> only when it may be accessed.
>>
>> Signed-off-by: Ben Walton
>> ---
>>
>> Hi All,
>>
>> When building coreutils 8.22 on Solaris
On 06/01/2014 09:34 AM, Ben Walton wrote:
> * Avoid possible compiler warnings/errors by defining the out label
> only when it may be accessed.
>
> Signed-off-by: Ben Walton
> ---
>
> Hi All,
>
> When building coreutils 8.22 on Solaris with -Werror=unused-label, the build
> fails with:
>
On 1 Jun 2014 21:16, "Pádraig Brady" wrote:
>
> On 06/01/2014 09:42 AM, Ben Walton wrote:
> > * lib/acl-internal.h (acl_ace_nontrivial): Apply pure attribute
> > * lib/file-has-acl.c: Disable pure attribute error.
> > - The AIX version of acl_nontrivial isn't pure while other
> >
On 06/01/2014 09:42 AM, Ben Walton wrote:
> * lib/acl-internal.h (acl_ace_nontrivial): Apply pure attribute
> * lib/file-has-acl.c: Disable pure attribute error.
> - The AIX version of acl_nontrivial isn't pure while other
> versions are. We cannot apply the attribute globally
I'm not sure if the post should be sent to the autoconf mailing list
instead. I see the following warnings from gl_EARLY when running
autoconf (version 2.69):
configure.ac:42: warning: AC_RUN_IFELSE was called before
AC_USE_SYSTEM_EXTENSIONS
../../lib/autoconf/specific.m4:314: AC_GNU_SOURCE is exp
* lib/acl-internal.h (acl_ace_nontrivial): Apply pure attribute
* lib/file-has-acl.c: Disable pure attribute error.
- The AIX version of acl_nontrivial isn't pure while other
versions are. We cannot apply the attribute globally.
Signed-off-by: Ben Walton
---
When building co
Hi,
I noticed that lib/parse-duration.c does:
#define TIME_MAX 0x7FFF
This is naturally entirely unacceptable as the 2038 bug ((time_t)
INT32_MAX) is coming up soon and most operating systems are 64-bit
time_t capable by now. This should be rewritten as something involving
sizeof(time_t) in
* Avoid possible compiler warnings/errors by defining the out label
only when it may be accessed.
Signed-off-by: Ben Walton
---
Hi All,
When building coreutils 8.22 on Solaris with -Werror=unused-label, the build
fails with:
lib/rename.c: In function 'rpl_rename':
lib/rename.c:465:2: err
I am (would be) _very_ happy - ecstatic even - about my present situation. My
concern is one regarding certain memory leaks I found I could not avoid with
regard to parent processes. I had hoped to limit the page tables accessed by
these parent processes; although it appears I may not have fully
19 matches
Mail list logo
