Re: [scsi 5/7 RESEND] scsi_debug: schedule_resp fix input variable check

[James Bottomley <James.Bottomley@HansenPartnership.com>]

On Tue, 2015-08-25 at 21:03 +0000, Winkler, Tomas wrote:
> 
> > > +	/* this should never happen */
> > > +	if (WARN_ON(!cmnd))
> > > +		return SCSI_MLQUEUE_HOST_BUSY;
> > >
> > > -	if (NULL == cmnd || NULL == devip) {
> > > -		pr_warn("called with NULL cmnd or devip pointer\n");
> > > +	if (NULL == devip) {
> > > +		pr_warn("called devip == NULL\n");
> > >  		/* no particularly good error to report back */
> > >  		return SCSI_MLQUEUE_HOST_BUSY;
> > >  	}
> > 
> > Please refer to the patch I just posted, we can't return _HOST_BUSY here
> > if devip == NULL.  I posted a fix against the current "misc" branch as
> > I don't see this patch applied, let me know if I need to update it.
> 
> 
> I'm just not sure why the patches are not merged or even rejected.

Because ideally I want a Maintainer ack.  That's Doug Gilbert.

> I'm submitting patches to the Linux kernel for more than 10 years to
> various trees and  I can agree that these are not some  urgent fixes,
> but  this is the first time my effort is ignored for such long time by
> the maintainer. 

Well, OK, I trust martin, I'll override the lack of Maintainer ack if
you fix as Ewan suggests.

James


> Thanks
> Tomas
> 
> NrybXǧv^)޺{.n+{"{ay\x1dʇڙ,j\afhz\x1ew\fj:+vwjm\azZ+ݢj"!