Re: cvs commit: apache-2.0/src/modules/mpm/mpmt_pthread acceptlock.c scoreboard.c

[Jim Jagielski <ji...@devsys.jaguNET.com>]

rbb@hyperreal.org wrote:
> 
> rbb         99/09/11 08:58:44
> 
>   Modified:    src/modules/mpm/dexter acceptlock.c
>                src/modules/mpm/mpmt_pthread acceptlock.c scoreboard.c
>   Log:
>   This change should allow Apache to build on all platforms with the popenf
>   changes.  I have not been able to test this, but it should be evident what I
>   am doing if it doesn't work.
>   
>   Revision  Changes    Path
>   1.7       +6 -3      apache-2.0/src/modules/mpm/dexter/acceptlock.c
>   
>   Index: acceptlock.c
>   ===================================================================
>   RCS file: /home/cvs/apache-2.0/src/modules/mpm/dexter/acceptlock.c,v
>   retrieving revision 1.6
>   retrieving revision 1.7
>   diff -u -r1.6 -r1.7
>   --- acceptlock.c	1999/09/08 18:58:46	1.6
>   +++ acceptlock.c	1999/09/11 15:58:35	1.7
>   @@ -574,12 +574,14 @@
>     */ 
>    void accept_mutex_child_init(ap_context_t *p)
>    {
>   +    ap_file_t *tempfile = NULL;
>        int i;
>      
>        for (i = 0; i < lock_count; i++) {
>            char *lock_fname = expand_lock_fname(p, i);
>        
>   -        lock_fd[i] = ap_popenf(p, lock_fname, O_WRONLY, 0600);
>   +        ap_open(p, lock_fname, APR_WRITE, APR_UREAD | APR_UWRITE, &tempfile);
>   +        ap_get_os_file(tempfile, *lock_fd[i]);

'*lock_fd[i]'

Is that right?

-- 
===========================================================================
   Jim Jagielski   [|]   jim@jaguNET.com   [|]   http://www.jaguNET.com/
                "Are you suggesting coconuts migrate??"

Re: cvs commit: apache-2.0/src/modules/mpm/mpmt_pthread acceptlock.c scoreboard.c

[Ryan Bloom <rb...@raleigh.ibm.com>]

> >   -        lock_fd[i] = ap_popenf(p, lock_fname, O_WRONLY, 0600);
> >   +        ap_open(p, lock_fname, APR_WRITE, APR_UREAD | APR_UWRITE, &tempfile);
> >   +        ap_get_os_file(tempfile, *lock_fd[i]);
> 
> '*lock_fd[i]'
> 
> Is that right?

Nope.  That should be &lock_fd[i].  Sorry, ussually stupid typos are
caught early with compilers and testing.  Couldn't do either this time,
but I wanted to get the changes in.  I'll fix it this morning.

Ryan

_______________________________________________________________________
Ryan Bloom		rbb@raleigh.ibm.com
4205 S Miami Blvd	
RTP, NC 27709		It's a beautiful sight to see good dancers 
			doing simple steps.  It's a painful sight to
			see beginners doing complicated patterns.