Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <6349abb3-9fb5-8bd1-7d71-e8a1435b0cef@suse.de>
Date:   Tue, 8 Nov 2022 07:57:09 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.2.2
Subject: Re: [PATCH] scsi: core: Add BLIST_NO_ASK_VPD_SIZE for some VDASD
Content-Language: en-US
To:     "Martin K. Petersen" <martin.petersen@oracle.com>
Cc:     Bart Van Assche <bvanassche@acm.org>,
        Lee Duncan <leeman.duncan@gmail.com>,
        "James E.J. Bottomley" <jejb@linux.ibm.com>,
        linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org,
        Lee Duncan <lduncan@suse.com>, Martin Wilck <mwilck@suse.com>
References: <20220928181350.9948-1-leeman.duncan@gmail.com>
 <11a582f0-723c-95e1-0e44-0a19e1a8a9a8@acm.org>
 <4a1da181-8a54-d2f8-6d19-d9c1982ab044@suse.de>
 <yq1k046kwyv.fsf@ca-mkp.ca.oracle.com>
From:   Hannes Reinecke <hare@suse.de>
In-Reply-To: <yq1k046kwyv.fsf@ca-mkp.ca.oracle.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk

On 11/8/22 03:50, Martin K. Petersen wrote:
> 
> Hannes,
> 
> I have been contemplating this for a bit.
> 
>>> Has it been considered instead of introducing a blacklist flag to not
>>> use the reported VPD page size if the device reports that the VPD
>>> page size is zero? I am not aware of any VPD pages for which zero is
>>> a valid size.
> 
> That would also be my preferred approach, I think. I haven't received
> any bug reports about devices returning short VPD pages since this
> change was introduced. So I think I'd prefer falling back to a
> (hopefully small) default if a device returns a 0 page length.
> 
> Now, my question is which VPD pages are actually supported by this
> device and how large are they?
> 
>> But pre-SPC drives will ignore the VPD bit in the inquiry size. And
>> these devices do not set an additional length in the inquiry data
> 
> Can you elaborate a bit on your experience with older devices? I checked
> SCSI-2 (1991) and don't see any indication this would be valid behavior
> even back then.
> 
This is primarily crappy USB devices, which implement only the absolute 
minimum to get SCSI rolling.
In particular, if devices do _not_ check the VPD bit in the inquiry 
command they will continue to return the standard inquiry data.
And if the additional length is zero we have exactly the scenario above.

However, we _could_ turn things around, and use the BLIST_NO_VPD flag 
for these cases; so I'd be fine with having a default length for the VPD 
page and delegate any fallout from the to use the BLIST_NO_VPD flags.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke                Kernel Storage Architect
hare@suse.de                              +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew
Myers, Andrew McDonald, Martje Boudien Moerman