apr-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Gavin Sherry <...@alcove.com.au>
Subject Re: Segfault in testlockperf test on AIX
Date Thu, 18 Jun 2009 00:45:47 GMT
2009/6/18 Bojan Smojver <bojan@rexursive.com>

> On Thu, 2009-06-18 at 02:18 +0200, Gavin Sherry wrote:
> > $2 = 24
> > ptype proc_mutex_unlock_it
> > type = struct flock {
> >     short l_type;
> >     short l_whence;
> >     off_t l_start;
> >     off_t l_len;
> >     unsigned int l_sysid;
> >     pid_t l_pid;
> >     int l_vfs;
> > }
>
> I'm guessing here: short 2 bytes, off_t 8 bytes, int 4 bytes, pid_t 4
> bytes. So, we have 2 + 2 + 8 + 8 + 4 + 4 + 4 = 32. Is that right?
>

No, it comes to 24 bytes. short = 2 + 2, off_t = 4 + 4, + 4 + 4 + 4 = 24.
off_t = 4 for this build because I've forced it to 32 bit (gcc -m32) based
on some other requirements.

Thanks,
Gavin

Mime
View raw message