Ok, although that results in client code having to deal with multiple logical block sizes. The transition from 512 to 4k has been sufficiently involved and lengthy that introduction of yet another logical block size in the foreseeable future seems rather
unlikely. I'm ok with either approach.
Thanks, --David ++Sent from Android smartphone ...
-------- Original message --------
From: Christoph Hellwig <hch@xxxxxx>
Date: 7/19/16 10:37 AM (GMT+01:00)
To: "Black, David" <david.black@xxxxxxx>
Cc: Christoph Hellwig <hch@xxxxxx>, ietf@xxxxxxxx, draft-ietf-nfsv4-scsi-layout@xxxxxxxx, spencerdawkins.ietf@xxxxxxxxx, nfsv4@xxxxxxxx, nfsv4-chairs@xxxxxxxx
Subject: Re: [nfsv4] Last Call: <draft-ietf-nfsv4-scsi-layout-06.txt> (Parallel NFS (pNFS) SCSI Layout) to Proposed Standard
On Tue, Jul 19, 2016 at 08:26:38AM +0000, Black, David wrote:
> - Yes, making 4k alignment a MUST would be a fine thing to do now.
Can we converge on a MUST for the device logical block size, be that
512, 4k or any other value that might happen in the future?
> - Yes, making 4k alignment a MUST would be a fine thing to do now.
Can we converge on a MUST for the device logical block size, be that
512, 4k or any other value that might happen in the future?