[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH v2 5/5] hw/block/nvme: report non-mdts command size limit for
From: |
Keith Busch |
Subject: |
Re: [PATCH v2 5/5] hw/block/nvme: report non-mdts command size limit for dsm |
Date: |
Tue, 2 Mar 2021 00:37:30 +0900 |
User-agent: |
Mutt/1.12.1 (2019-06-15) |
On Mon, Mar 01, 2021 at 12:15:26PM +0100, Klaus Jensen wrote:
> On Feb 22 22:12, Klaus Jensen wrote:
> > On Feb 23 05:55, Keith Busch wrote:
> > > On Mon, Feb 22, 2021 at 07:47:59PM +0100, Klaus Jensen wrote:
> > > > +typedef struct NvmeIdCtrlNvm {
> > > > + uint8_t vsl;
> > > > + uint8_t wzsl;
> > > > + uint8_t wusl;
> > > > + uint8_t dmrl;
> > > > + uint32_t dmrsl;
> > > > + uint64_t dmsl;
> > > > + uint8_t rsvd16[4080];
> > > > +} NvmeIdCtrlNvm;
> > >
> > > TP 4040a still displays these fields with preceding '...' indicating
> > > something comes before this. Is that just left-over from the integration
> > > for TBD offsets, or is there something that still hasn't been accounted
> > > for?
> >
> > Good question.
> >
> > But since the TBDs have been assigned I believe it is just a left-over.
> > I must admit that I have not cross checked this with all other TPs, but
> > AFAIK this is the only ratified TP that adds something to the
> > NVM-specific identify controller data structure.
>
> Are you otherwise OK with this?
Yes, I compared other TP's and it appears to be set for good once an
actual numeric value is assigned, so okay to go here.
Reviewed-by: Keith Busch <kbusch@kernel.org>