Hi Alistair
> -----Original Message-----
> From: Alistair Francis <alistai...@gmail.com>
> Sent: Saturday, February 13, 2021 3:41 AM
> To: Sai Pavan Boddu <saip...@xilinx.com>
> Cc: Markus Armbruster <arm...@redhat.com>; Kevin Wolf
> <kw...@redhat.com>; Max Reitz <mre...@redhat.com>; Vladimir Sementsov-
> Ogievskiy <vsement...@virtuozzo.com>; Eric Blake <ebl...@redhat.com>;
> Joel Stanley <j...@jms.id.au>; Cédric Le Goater <c...@kaod.org>; Vincent
> Palatin <vpala...@chromium.org>; Dr. David Alan Gilbert
> <dgilb...@redhat.com>; Thomas Huth <th...@redhat.com>; Stefan Hajnoczi
> <stefa...@redhat.com>; Peter Maydell <peter.mayd...@linaro.org>; Alistair
> Francis <alistair.fran...@wdc.com>; Edgar Iglesias <edg...@xilinx.com>; Luc
> Michel <luc.mic...@greensocs.com>; Paolo Bonzini <pbonz...@redhat.com>;
> Sai Pavan Boddu <saip...@xilinx.com>; qemu-devel@nongnu.org Developers
> <qemu-devel@nongnu.org>; Qemu-block <qemu-bl...@nongnu.org>
> Subject: Re: [RFC PATCH 10/15] sd: emmc: Update CID structure for eMMC
>
> On Thu, Feb 11, 2021 at 12:30 AM Sai Pavan Boddu
> <sai.pavan.bo...@xilinx.com> wrote:
> >
> > CID structure is little different for eMMC, w.r.t to product name and
> > manufacturing date.
> >
> > Signed-off-by: Sai Pavan Boddu <sai.pavan.bo...@xilinx.com>
> > Signed-off-by: Edgar E. Iglesias <edgar.igles...@xilinx.com>
> > ---
> > hw/sd/sd.c | 52 +++++++++++++++++++++++++++++++++++-----------------
> > 1 file changed, 35 insertions(+), 17 deletions(-)
> >
> > diff --git a/hw/sd/sd.c b/hw/sd/sd.c
> > index 7aab647..45311fa 100644
> > --- a/hw/sd/sd.c
> > +++ b/hw/sd/sd.c
> > @@ -345,23 +345,41 @@ static void sd_set_scr(SDState *sd)
> >
> > static void sd_set_cid(SDState *sd)
> > {
> > - sd->cid[0] = MID; /* Fake card manufacturer ID (MID) */
> > - sd->cid[1] = OID[0]; /* OEM/Application ID (OID) */
> > - sd->cid[2] = OID[1];
> > - sd->cid[3] = PNM[0]; /* Fake product name (PNM) */
> > - sd->cid[4] = PNM[1];
> > - sd->cid[5] = PNM[2];
> > - sd->cid[6] = PNM[3];
> > - sd->cid[7] = PNM[4];
> > - sd->cid[8] = PRV; /* Fake product revision (PRV) */
> > - sd->cid[9] = 0xde; /* Fake serial number (PSN) */
> > - sd->cid[10] = 0xad;
> > - sd->cid[11] = 0xbe;
> > - sd->cid[12] = 0xef;
> > - sd->cid[13] = 0x00 | /* Manufacture date (MDT) */
> > - ((MDT_YR - 2000) / 10);
> > - sd->cid[14] = ((MDT_YR % 10) << 4) | MDT_MON;
> > - sd->cid[15] = (sd_crc7(sd->cid, 15) << 1) | 1;
> > + if (sd->emmc) {
> > + sd->cid[0] = MID;
> > + sd->cid[1] = 0x1; /* CBX */
> > + sd->cid[2] = OID[0]; /* OEM/Application ID (OID) */
> > + sd->cid[3] = PNM[0]; /* Fake product name (PNM) 48bit */
> > + sd->cid[4] = PNM[1];
> > + sd->cid[5] = PNM[2];
> > + sd->cid[6] = PNM[3];
> > + sd->cid[7] = PNM[4];
>
> Aren't the majority of these the same between the two cases? It's probably
> cleaner to split them out then.
[Sai Pavan Boddu] Yes, I would try to re-order. If I see only the PNM fields
are same, rest all fields kind of moved a byte below.
Regards,
Sai Pavan
>
> Alistair
>
> > + sd->cid[8] = 0x0;
> > + sd->cid[9] = PRV; /* Fake product revision (PRV) */
> > + sd->cid[10] = 0xde; /* Fake serial number (PSN) */
> > + sd->cid[11] = 0xad;
> > + sd->cid[12] = 0xbe;
> > + sd->cid[13] = 0xef;
> > + sd->cid[14] = ((MDT_YR - 1997) % 0x10); /* MDT */
> > + } else {
> > + sd->cid[0] = MID; /* Fake card manufacturer ID (MID) */
> > + sd->cid[1] = OID[0]; /* OEM/Application ID (OID) */
> > + sd->cid[2] = OID[1];
> > + sd->cid[3] = PNM[0]; /* Fake product name (PNM) 40bit */
> > + sd->cid[4] = PNM[1];
> > + sd->cid[5] = PNM[2];
> > + sd->cid[6] = PNM[3];
> > + sd->cid[7] = PNM[4];
> > + sd->cid[8] = PRV; /* Fake product revision (PRV) */
> > + sd->cid[9] = 0xde; /* Fake serial number (PSN) */
> > + sd->cid[10] = 0xad;
> > + sd->cid[11] = 0xbe;
> > + sd->cid[12] = 0xef;
> > + sd->cid[13] = 0x00 | /* Manufacture date (MDT) */
> > + ((MDT_YR - 2000) / 10);
> > + sd->cid[14] = ((MDT_YR % 10) << 4) | MDT_MON;
> > + }
> > + sd->cid[15] = (sd_crc7(sd->cid, 15) << 1) | 1;
> > }
> >
> > #define HWBLOCK_SHIFT 9 /* 512 bytes */
> > --
> > 2.7.4
> >
> >
