RE: [PATCH 05/10] ravb: Initialize GbEthernet DMAC

[Biju Das <biju.das.jz@bp.renesas.com>]

> -----Original Message-----
> From: Sergei Shtylyov <sergei.shtylyov@gmail.com>
> Sent: 04 October 2021 16:51
> To: Biju Das <biju.das.jz@bp.renesas.com>; Sergey Shtylyov
> <s.shtylyov@omp.ru>; David S. Miller <davem@davemloft.net>; Jakub Kicinski
> <kuba@kernel.org>
> Cc: Geert Uytterhoeven <geert+renesas@glider.be>; Sergey Shtylyov
> <s.shtylyov@omprussia.ru>; Adam Ford <aford173@gmail.com>; Andrew Lunn
> <andrew@lunn.ch>; Yuusuke Ashizuka <ashiduka@fujitsu.com>; Yoshihiro
> Shimoda <yoshihiro.shimoda.uh@renesas.com>; netdev@vger.kernel.org; linux-
> renesas-soc@vger.kernel.org; Chris Paterson <Chris.Paterson2@renesas.com>;
> Biju Das <biju.das@bp.renesas.com>; Prabhakar Mahadev Lad
> <prabhakar.mahadev-lad.rj@bp.renesas.com>
> Subject: Re: [PATCH 05/10] ravb: Initialize GbEthernet DMAC
> 
> On 10/4/21 4:12 PM, Biju Das wrote:
> 
> >> Subject: Re: [PATCH 05/10] ravb: Initialize GbEthernet DMAC
> >>
> >> Hello!
> >>
> >> On 10/1/21 6:06 PM, Biju Das wrote:
> >>
> >>> Initialize GbEthernet DMAC found on RZ/G2L SoC.
> >>>
> >>> Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
> >>> Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
> >>> ---
> >>> RFC->v1:
> >>>  * Removed RIC3 initialization from DMAC init, as it is
> >>>    same as reset value.
> >>
> >>    I'm not sure we do a reset everytime...
> >>
> >>>  * moved stubs function to earlier patches.
> >>>  * renamed "rgeth" with "gbeth"
> >>> ---
> >>>  drivers/net/ethernet/renesas/ravb.h      |  3 ++-
> >>>  drivers/net/ethernet/renesas/ravb_main.c | 30
> >>> +++++++++++++++++++++++-
> >>>  2 files changed, 31 insertions(+), 2 deletions(-)
> >>>
> >> [...]
> >>> diff --git a/drivers/net/ethernet/renesas/ravb_main.c
> >>> b/drivers/net/ethernet/renesas/ravb_main.c
> >>> index dc817b4d95a1..5790a9332e7b 100644
> >>> --- a/drivers/net/ethernet/renesas/ravb_main.c
> >>> +++ b/drivers/net/ethernet/renesas/ravb_main.c
> >>> @@ -489,7 +489,35 @@ static void ravb_emac_init(struct net_device
> >>> *ndev)
> >>>
> >>>  static int ravb_dmac_init_gbeth(struct net_device *ndev)  {
> >>> -	/* Place holder */
> >>> +	int error;
> >>> +
> >>> +	error = ravb_ring_init(ndev, RAVB_BE);
> >>> +	if (error)
> >>> +		return error;
> >>> +
> >>> +	/* Descriptor format */
> >>> +	ravb_ring_format(ndev, RAVB_BE);
> >>> +
> >>> +	/* Set AVB RX */
> >>
> >>    AVB? We don't have it, do we?
> >
> > Good catch. I Will update the comment in next RFC patch.
> 
>    That's trifles, not worth a patch on its own...
> 
> >>
> >>> +	ravb_write(ndev, 0x60000000, RCR);
> >>
> >>    Not even RCR.EFFS? And what do bits 29..30 mean?
> >
> > RZ/G2L Bit 31 is reserved.
> > Bit 16:30 Reception fifo critical level.
> > Bit 15:1 reserved
> > Bit 0 : EFFS
> >
> > I am not sure, where do you get 29..30? can you please clarify.
> 
>    0x60000000 has bits 29..30 set and gen3 manual has these bits reserved.
OK.

> 
> >> [...]
> >>> +	/* Set FIFO size */
> >>> +	ravb_write(ndev, 0x00222200, TGC);
> >>
> >>    Do TBD<n> (other than TBD0) fields even exist?
> >
> > Only TBD (Bit 8..9) is available to write,
> 
>    Thought so! :-)
> 
> > rest all are reserved with remaining values as in "0x00222200"
> 
>   Oh, so the defaluts are the sme on RZ/G2L, despite only 1 TX queue?

Yep.

> 
> > Regds,
> > Biju
> 
> MBR, Sergey