On Thu, Feb 11, 2021 at 11:44:44AM -0800, Sagi Grimberg wrote:

So you suggest that we keep the retry w.a. but continue to send the
wrong value in 1.3 or lower?

I personally think that this would be extremely confusing to users, if
they do trip on this, but I agree that they are less likely to hit
it because existing controllers obviously accept wrong values...

Yes.  While I'd like to move to standards compliance as much as possible,
introducing a long delay for every existing working setup is 
very counter productive.

As for option to override. I really hate it.. What is the least exposure
we can get for this but still allow the user to control this?

We can of course still override it using an option.

Wish we had a good and reliable quirk mechanism for fabrics...

We do, using the serial number.  But that does not help for an issue
during connection establishment.

_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme