Re: [PATCH 1/1] nvme-fcloop: verify wwnn and wwpn format

[Hannes Reinecke]

On 6/4/20 8:54 AM, Chaitanya Kulkarni wrote:
> On 6/3/20 11:46 PM, Dongli Zhang wrote:
> > May I get feedback for this?
> >
> > For the first time I use fcloop, I set:
> >
> > # echo "wwnn=0x3,wwpn=0x1" > /sys/class/fcloop/ctl/add_target_port
> >
> > However, I would not be able to move forward if I use "0x3" or "0x1" for nvme-fc
> > target or host further. Instead, the address and port should be
> > 0x0000000000000003 and 0x0000000000000001.
> >
> > This patch would sync the requirements of input format for nvme-fc and
> > nvme-fcloop, unless this would break existing test suite (e.g., blktest).
> If I remember correctly I don't think we have fc-loop testcases (correct
> me if I'm wrong).
Well, I sent some testcases a while back (cf 'fcloop and ANA fixes').
Should I resend them?

> Not an fc expert, but having uniform format for the input make sense to
> me (unless there is an explicit reason). I'll let James have a final say.

I would stick to use the full 64bit number for both wwpn and wwnn; one 
gets into too many arguments otherwise (big-endian? little-endian?).
And one could argue that '0x0000000000000001' is invalid anyway as per 
FC-FS3 a '0' in word 0 byte 0 means 'Name not present' :-)

Cheers,

Hannes
--
Dr. Hannes Reinecke            Teamlead Storage & Networking
hare@xxxxxxx                               +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 NÃrnberg
HRB 36809 (AG NÃrnberg), GeschÃftsfÃhrer: Felix ImendÃrffer