cvs commit: src/sys/kern vfs_subr.c

[Jeff Roberson]

On Fri, 3 Dec 2004, Nate Lawson wrote:

> Poul-Henning Kamp wrote:
> > phk         2004-12-03 12:09:34 UTC
> >
> >   FreeBSD src repository
> >
> >   Modified files:
> >     sys/kern             vfs_subr.c
> >   Log:
> >   Improve vprint() a little bit:  break long lines, reduce indent and tell
> >   if the VI_LOCK() is held.
> >
> >   Revision  Changes    Path
> >   1.555     +7 -4      src/sys/kern/vfs_subr.c
> >
> >
> > --- src/sys/kern/vfs_subr.c:1.554	Wed Dec  1 23:16:36 2004
> > +++ src/sys/kern/vfs_subr.c	Fri Dec  3 12:09:34 2004
> > @@ -2651,9 +2651,9 @@
> >  		printf("%s: %p: ", label, (void *)vp);
> >  	else
> >  		printf("%p: ", (void *)vp);
> > -	printf("tag %s, type %s, usecount %d, writecount %d, refcount %d,",
> > -	    vp->v_tag, typename[vp->v_type], vp->v_usecount,
> > -	    vp->v_writecount, vp->v_holdcnt);
> > +	printf("tag %s, type %s\n    ", vp->v_tag, typename[vp->v_type]);
> > +	printf("usecount %d, writecount %d, refcount %d mountedhere %p\n",
> > +	    vp->v_usecount, vp->v_writecount, vp->v_holdcnt, vp->v_mountedhere);
> >  	buf[0] = '\0';
> >  	if (vp->v_vflag & VV_ROOT)
> >  		strcat(buf, "|VV_ROOT");
> > @@ -2672,7 +2672,10 @@
> >  	if (vp->v_vflag & VV_OBJBUF)
> >  		strcat(buf, "|VV_OBJBUF");
> >  	if (buf[0] != '\0')
> > -		printf(" flags (%s),", &buf[1]);
> > +		printf("    flags (%s)", &buf[1]);
> > +	if (mtx_owned(VI_MTX(vp)))
> > +		printf(" VI_LOCKed");
> > +	printf("\n    ");
> >  	lockmgr_printinfo(vp->v_vnlock);
> >  	printf("\n");
> >  	if (vp->v_data != NULL)
>
> I think that breaking vprint() into multiple lines is not a good idea.
> Bruce pointed this out to me when I was doing the v_tag changes a while ago.

What's the reason?

>
> -Nate
>