r40 can support tx_delay, so we add it.
Signed-off-by: fuyao <[email protected]>
---
drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
index 137741b94122..fd07573afc9b 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
@@ -127,6 +127,7 @@ static const struct emac_variant emac_variant_r40 = {
.support_mii = true,
.support_rgmii = true,
.rx_delay_max = 7,
+ .tx_delay_max = 7,
};
static const struct emac_variant emac_variant_a64 = {
--
2.39.2
--
CYG Technology.
Dne sreda, 20. december 2023 ob 10:08:25 CET je fuyao napisal(a):
> r40 can support tx_delay, so we add it.
I don't see how. User manual, 3.3.4.64. GMAC Clock Register talks only
about RX delay.
Best regards,
Jernej
>
> Signed-off-by: fuyao <[email protected]>
> ---
> drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
> index 137741b94122..fd07573afc9b 100644
> --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
> +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
> @@ -127,6 +127,7 @@ static const struct emac_variant emac_variant_r40 = {
> .support_mii = true,
> .support_rgmii = true,
> .rx_delay_max = 7,
> + .tx_delay_max = 7,
> };
>
> static const struct emac_variant emac_variant_a64 = {
>
On Wed, Dec 20, 2023 at 06:04:28PM +0100, Jernej Škrabec wrote:
> Dne sreda, 20. december 2023 ob 10:08:25 CET je fuyao napisal(a):
> > r40 can support tx_delay, so we add it.
>
> I don't see how. User manual, 3.3.4.64. GMAC Clock Register talks only
> about RX delay.
>
I don't see it in the user manual also.
it works fine in 100M, but it doesn't work good in 1G without tx_delay
in my board.
Read the sdk from allwinner I found in:
https://sourcegraph.com/github.com/tinalinux/linux-3.10/-/blob/arch/arm/boot/dts/sun8iw11p1.dtsi?L1826
https://sourcegraph.com/github.com/tinalinux/[email protected]/-/blob/drivers/net/ethernet/allwinner/sunxi-gmac.c?L877
And I add it to my board, it works fine.
So, I think it may be exist register but not documented.
Best regards,
fuyao
> Best regards,
> Jernej
>
> >
> > Signed-off-by: fuyao <[email protected]>
> > ---
> > drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
> > index 137741b94122..fd07573afc9b 100644
> > --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
> > +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c
> > @@ -127,6 +127,7 @@ static const struct emac_variant emac_variant_r40 = {
> > .support_mii = true,
> > .support_rgmii = true,
> > .rx_delay_max = 7,
> > + .tx_delay_max = 7,
> > };
> >
> > static const struct emac_variant emac_variant_a64 = {
> >
>
>
>
>
On Thu, 21 Dec 2023 15:16:54 +0800 fuyao wrote:
> I don't see it in the user manual also.
>
> it works fine in 100M, but it doesn't work good in 1G without tx_delay
> in my board.
>
> Read the sdk from allwinner I found in:
>
> https://sourcegraph.com/github.com/tinalinux/linux-3.10/-/blob/arch/arm/boot/dts/sun8iw11p1.dtsi?L1826
> https://sourcegraph.com/github.com/tinalinux/[email protected]/-/blob/drivers/net/ethernet/allwinner/sunxi-gmac.c?L877
>
> And I add it to my board, it works fine.
>
> So, I think it may be exist register but not documented.
Unless Allwinner can chime in we should at least document the situation
clearly in the commit message (incl. what HW+config was tested, exactly)
and add a comment in the code?
--
pw-bot: cr
Dne torek, 02. januar 2024 ob 16:58:07 CET je Jakub Kicinski napisal(a):
> On Thu, 21 Dec 2023 15:16:54 +0800 fuyao wrote:
> > I don't see it in the user manual also.
> >
> > it works fine in 100M, but it doesn't work good in 1G without tx_delay
> > in my board.
> >
> > Read the sdk from allwinner I found in:
> >
> > https://sourcegraph.com/github.com/tinalinux/linux-3.10/-/blob/arch/arm/boot/dts/sun8iw11p1.dtsi?L1826
> > https://sourcegraph.com/github.com/tinalinux/[email protected]/-/blob/drivers/net/ethernet/allwinner/sunxi-gmac.c?L877
> >
> > And I add it to my board, it works fine.
> >
> > So, I think it may be exist register but not documented.
>
> Unless Allwinner can chime in we should at least document the situation
> clearly in the commit message (incl. what HW+config was tested, exactly)
> and add a comment in the code?
>
I agree. Fuyao, please do that.
Best regards,
Jernej