Re: HEADS UP: destroy_dev_sched() KPI in the tree

On Sun, Jul 08, 2007 at 09:47:41AM +0800, Tai-hwa Liang wrote:
> On Sat, 7 Jul 2007, Kostik Belousov wrote:
> >On Sat, Jul 07, 2007 at 08:36:23PM +0800, Jiawei Ye wrote:
> >>On 7/5/07, Kostik Belousov <kostikbel_at_gmail.com> wrote:
> >>>
> >>>Today, I reverted the part of commit that tried to transform 
> >>>destroy_dev()
> >>>from d_close() into destroy_dev_sched(). See kern_conf.c, rev. 1.208.
> >>>You shall call destroy_dev_sched() explicitely.
> >>>
> >>>
> >>Does this mean that mount_smbfs has to explicitly call it now? It is
> >>currently stuck in the devdrn state with the latest kernel :(
> >>
> >>Jiawei Ye
> >Exactly. Patch by Tai-hwa Liang is pending. I also remember that sg(4)
> 
>   I'll ask for re_at_'s approval tomorrow.  Hopefully the new KPI would have
> more exposure before my patch is committed.
> 
> >had the same problem, but I do not know it current status.
> 
>   Though it was reviewed before destroy_dev_sched() KPI enters to the tree,
> I'd be appreciate it if you can reviewed the attached patch again.
So, this is still the problem for scsi_targ ?

It probably make sense to postpone free of softc until all threads
finished using it. You may use destroy_dev_sched_cb() to run the
function after the device is actually destroyed. It would just call
free().
> 
> -- 
> Cheers,
> 
> Tai-hwa Liang

> --- cam/scsi/scsi_target.c.orig	Wed May  2 11:42:46 2007
> +++ cam/scsi/scsi_target.c	Wed May  2 11:43:11 2007
> _at__at_ -210,7 +210,7 _at__at_ targclose(struct cdev *dev, int flag, in
>  	softc = (struct targ_softc *)dev->si_drv1;
>  	if ((softc->periph == NULL) ||
>  	    (softc->state & TARG_STATE_LUN_ENABLED) == 0) {
> -		destroy_dev(dev);
> +		destroy_dev_sched(dev);
>  		FREE(softc, M_TARG);
>  		return (0);
>  	}
> _at__at_ -229,7 +229,7 _at__at_ targclose(struct cdev *dev, int flag, in
>  			cam_periph_invalidate(softc->periph);
>  			softc->periph = NULL;
>  		}
> -		destroy_dev(dev);
> +		destroy_dev_sched(dev);
>  		FREE(softc, M_TARG);
>  	}
>  	cam_periph_unlock(periph);