Re: [PATCH 1/1] dpt_i2o: delete unnecessary null test on array

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

On Fri, 2014-08-08 at 19:03 +0200, Julia Lawall wrote:
> > > diff --git a/drivers/scsi/dpt_i2o.c b/drivers/scsi/dpt_i2o.c
> > > index 67283ef..62e276b 100644
> > > --- a/drivers/scsi/dpt_i2o.c
> > > +++ b/drivers/scsi/dpt_i2o.c
> > > @@ -1169,11 +1169,6 @@ static struct adpt_device* adpt_find_device(adpt_hba* pHba, u32 chan, u32 id, u6
> > >  	if(chan < 0 || chan >= MAX_CHANNEL)
> > >  		return NULL;
> > >
> > > -	if( pHba->channel[chan].device == NULL){
> > > -		printk(KERN_DEBUG"Adaptec I2O RAID: Trying to find device before they are allocated\n");
> > > -		return NULL;
> > > -	}
> >
> > dpt_i2o is always weirdly fun, but I think, based on the message, this
> > check is supposed to be
> >
> > if( pHba->channel[chan].device[id] == NULL){
> >
> > Since device is an array of device pointers which are allocated by
> > parsing data.
> 
> That seems to be already checked immediately below:
> 
>         d = pHba->channel[chan].device[id];
> 	if(!d || d->tid == 0) {
>                 return NULL;

Yes, I know, but no message is emitted.  The message seems to be for a
violation of the state machine which device[id] = NULL implies.

James