Re: [PATCH v3 7/7] scsi: scsi_devinfo: extend BLIST_NO_LUN_1F to MATSHITA and NEC PD-1 variants

[Hannes Reinecke <hare@suse.de>]

On 5/6/26 02:52, Phil Pemberton wrote:
> On 27/04/2026 12:56, Hannes Reinecke wrote:
>> On 4/26/26 21:09, Phil Pemberton wrote:
>>> The Panasonic LF-1095/LF-1195 PD/CD combo drive was sold under three
>>> OEM identities: COMPAQ "PD-1", MATSHITA "PD-1", and NEC "PD-1 ODX654P".
>>> All three are the same drive mechanism with the same firmware family,
>>> so they should share the BLIST_NO_LUN_1F quirk that was applied to the
>>> COMPAQ variant: PDT 0x1f / PQ 0 INQUIRY responses on non-existent LUNs
>>> are treated as "LUN not present" rather than as a phantom sdev.
>>>
>>> This patch is offered for completeness.  It has not been tested on the
>>> MATSHITA or NEC variants -- the author only has access to the COMPAQ
>>> unit -- but the drives are functionally identical and the flag is a
>>> no-op on devices that do not exhibit the PDT 0x1f response.  Drop or
>>> hold this patch if confirmation on real hardware is preferred before
>>> extending the quirk.
>>>
>>> Signed-off-by: Phil Pemberton <philpem@philpem.me.uk>
>>> ---
>>>   drivers/scsi/scsi_devinfo.c | 6 ++++--
>>>   1 file changed, 4 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/scsi/scsi_devinfo.c b/drivers/scsi/scsi_devinfo.c
>>> index bfc2cbd43897..ab1ffa9433b7 100644
>>> --- a/drivers/scsi/scsi_devinfo.c
>>> +++ b/drivers/scsi/scsi_devinfo.c
>>> @@ -201,7 +201,8 @@ static struct {
>>>       {"LASOUND", "CDX7405", "3.10", BLIST_MAX5LUN | BLIST_SINGLELUN},
>>>       {"Marvell", "Console", NULL, BLIST_SKIP_VPD_PAGES},
>>>       {"Marvell", "91xx Config", "1.01", BLIST_SKIP_VPD_PAGES},
>>> -    {"MATSHITA", "PD-1", NULL, BLIST_FORCELUN | BLIST_SINGLELUN},
>>> +    {"MATSHITA", "PD-1", NULL, BLIST_FORCELUN | BLIST_SINGLELUN |
>>> +                   BLIST_NO_LUN_1F},
>>>       {"MATSHITA", "DMC-LC5", NULL, BLIST_NOT_LOCKABLE | 
>>> BLIST_INQUIRY_36},
>>>       {"MATSHITA", "DMC-LC40", NULL, BLIST_NOT_LOCKABLE | 
>>> BLIST_INQUIRY_36},
>>>       {"Medion", "Flash XL  MMC/SD", "2.6D", BLIST_FORCELUN},
>>> @@ -212,7 +213,8 @@ static struct {
>>>       {"nCipher", "Fastness Crypto", NULL, BLIST_FORCELUN},
>>>       {"NAKAMICH", "MJ-4.8S", NULL, BLIST_FORCELUN | BLIST_SINGLELUN},
>>>       {"NAKAMICH", "MJ-5.16S", NULL, BLIST_FORCELUN | BLIST_SINGLELUN},
>>> -    {"NEC", "PD-1 ODX654P", NULL, BLIST_FORCELUN | BLIST_SINGLELUN},
>>> +    {"NEC", "PD-1 ODX654P", NULL, BLIST_FORCELUN | BLIST_SINGLELUN |
>>> +                      BLIST_NO_LUN_1F},
>>>       {"NEC", "iStorage", NULL, BLIST_REPORTLUN2},
>>>       {"NRC", "MBR-7", NULL, BLIST_FORCELUN | BLIST_SINGLELUN},
>>>       {"NRC", "MBR-7.4", NULL, BLIST_FORCELUN | BLIST_SINGLELUN},
>>
>> Any specific reason why this patch is not merged with the previous one?
>> Otherwise:
>>
>> Reviewed-by: Hannes Reinecke <hare@suse.de>
> 
> Only that these are optional to the feature work and untested as I don't 
> have hardware. I believe these drives use the same PD-1 mechanism and 
> firmware so should behave the same, but I can't prove it.
> 
> The intent was to allow the 1-6 set to be merged (as these are tested) 
> without 7/7 (which is not) to minimise the risk of regressions.
> 
So drop it, then.
We can always add it later once someone shows up who actually has the 
hardware.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke                  Kernel Storage Architect
hare@suse.de                                +49 911 74053 688
SUSE Software Solutions GmbH, Frankenstr. 146, 90461 Nürnberg
HRB 36809 (AG Nürnberg), GF: I. Totev, A. McDonald, W. Knoblich