svn commit: trunk/uClibc/include/sys

Bernhard Fischer rep.nop at aon.at
Sat May 6 14:46:33 UTC 2006


On Sat, May 06, 2006 at 03:43:19PM +0200, Peter S. Mazinger wrote:
>On Fri, 5 May 2006 vapier at uclibc.org wrote:
>
>> Author: vapier
>> Date: 2006-05-05 23:34:45 -0700 (Fri, 05 May 2006)
>> New Revision: 15020
>> 
>> Log:
>> sync with glibc
>> 
>> Modified:
>>    trunk/uClibc/include/sys/queue.h
>>    trunk/uClibc/include/sys/syslog.h
>> 
>> 
>> Changeset:
>> Modified: trunk/uClibc/include/sys/queue.h
>> ===================================================================
>> --- trunk/uClibc/include/sys/queue.h	2006-05-06 05:30:46 UTC (rev 15019)
>> +++ trunk/uClibc/include/sys/queue.h	2006-05-06 06:34:45 UTC (rev 15020)
>>  
>> -#define LIST_INSERT_AFTER(listelm, elm, field) {			\
>> +#define	LIST_INSERT_AFTER(listelm, elm, field) do {			\
>> +	QUEUEDEBUG_LIST_OP((listelm), field)				\
>>  	if (((elm)->field.le_next = (listelm)->field.le_next) != NULL)	\
>>  		(listelm)->field.le_next->field.le_prev =		\
>>  		    &(elm)->field.le_next;				\
>>  	(listelm)->field.le_next = (elm);				\
>>  	(elm)->field.le_prev = &(listelm)->field.le_next;		\
>> -}
>> +} while (/*CONSTCOND*/0)
>
>I have looked also into this change and don't see how it will ever work 
>for uClibc, QUEUEDEBUG_* does not exist. Haven't found how/where glibc 
>defines these.

I only see it defined in netbsd..
http://cvsweb.netbsd.org/bsdweb.cgi/src/sys/sys/queue.h
which suggests that it's something that should expand to '' unless heavy
debugging is turned on..
>
>On another note, I have locally all the non-arch specific headers synced 
>w/ glibc and almost all are cleaned up (disabling all prototypes for 
>functions not provided by uClibc). I am working now on the arch specific 
>ones. Parts of it are already in the patch you have access on your ppc 
>box.
>
>Peter



More information about the uClibc mailing list