| Message ID | 1477008706-13387-1-git-send-email-andre.przywara@arm.com |
|---|---|
| State | Accepted |
| Commit | 6301e80ccfdf8b8daf921c67ffb5c5a17960f895 |
| Delegated to: | Hans de Goede |
| Headers | show |
On Fri, Oct 21, 2016 at 5:41 AM, Andre Przywara <andre.przywara@arm.com> wrote: > The sun8i-emac driver works fine with the A64 Ethernet IP, but we are > missing an alias entry to trigger the driver instantiation by U-Boot. > Add the line to point U-Boot to the Ethernet DT node. > This enables TFTP boot on the Pine64. > > Signed-off-by: Andre Przywara <andre.przywara@arm.com> > --- > arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi b/arch/arm/dts/sun50i-a64-pine64-common.dtsi > index d5a7249..c0fde44 100644 > --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi > +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi > @@ -46,6 +46,7 @@ > > aliases { > serial0 = &uart0; > + ethernet0 = &emac; I think alias doesn’t require for probing emac, it will straight away probed like > dm tree .... eth [ + ] ethernet@01c30000 Did you find any issue while detecting eth? thanks!
Hi, On 21/10/16 10:31, Jagan Teki wrote: > On Fri, Oct 21, 2016 at 5:41 AM, Andre Przywara <andre.przywara@arm.com> wrote: >> The sun8i-emac driver works fine with the A64 Ethernet IP, but we are >> missing an alias entry to trigger the driver instantiation by U-Boot. >> Add the line to point U-Boot to the Ethernet DT node. >> This enables TFTP boot on the Pine64. >> >> Signed-off-by: Andre Przywara <andre.przywara@arm.com> >> --- >> arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + >> 1 file changed, 1 insertion(+) >> >> diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >> index d5a7249..c0fde44 100644 >> --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >> +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >> @@ -46,6 +46,7 @@ >> >> aliases { >> serial0 = &uart0; >> + ethernet0 = &emac; > > I think alias doesn’t require for probing emac, it will straight away > probed like > >> dm tree > .... > eth [ + ] ethernet@01c30000 > > Did you find any issue while detecting eth? Yes, it just didn't work ;-) I don't have a board here, but can post the error message later tonight. In fact I was wondering about that already, maybe it's worth investigating this further. But aside from that I think the MAC address calculation based on the SID serial number does not get triggered without an alias, so we need this line anyway. Cheers, Andre.
Hi, On 21-10-16 12:06, Andre Przywara wrote: > Hi, > > On 21/10/16 10:31, Jagan Teki wrote: >> On Fri, Oct 21, 2016 at 5:41 AM, Andre Przywara <andre.przywara@arm.com> wrote: >>> The sun8i-emac driver works fine with the A64 Ethernet IP, but we are >>> missing an alias entry to trigger the driver instantiation by U-Boot. >>> Add the line to point U-Boot to the Ethernet DT node. >>> This enables TFTP boot on the Pine64. >>> >>> Signed-off-by: Andre Przywara <andre.przywara@arm.com> >>> --- >>> arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + >>> 1 file changed, 1 insertion(+) >>> >>> diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>> index d5a7249..c0fde44 100644 >>> --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>> +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>> @@ -46,6 +46,7 @@ >>> >>> aliases { >>> serial0 = &uart0; >>> + ethernet0 = &emac; >> >> I think alias doesn’t require for probing emac, it will straight away >> probed like >> >>> dm tree >> .... >> eth [ + ] ethernet@01c30000 >> >> Did you find any issue while detecting eth? > > Yes, it just didn't work ;-) I don't have a board here, but can post the > error message later tonight. > In fact I was wondering about that already, maybe it's worth > investigating this further. > > But aside from that I think the MAC address calculation based on the SID > serial number does not get triggered without an alias, so we need this > line anyway. Correct, the MAC address code relies on the alias. Regards, Hans
On 21/10/16 11:28, Hans de Goede wrote: > Hi, > > On 21-10-16 12:06, Andre Przywara wrote: >> Hi, >> >> On 21/10/16 10:31, Jagan Teki wrote: >>> On Fri, Oct 21, 2016 at 5:41 AM, Andre Przywara >>> <andre.przywara@arm.com> wrote: >>>> The sun8i-emac driver works fine with the A64 Ethernet IP, but we are >>>> missing an alias entry to trigger the driver instantiation by U-Boot. >>>> Add the line to point U-Boot to the Ethernet DT node. >>>> This enables TFTP boot on the Pine64. >>>> >>>> Signed-off-by: Andre Przywara <andre.przywara@arm.com> >>>> --- >>>> arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + >>>> 1 file changed, 1 insertion(+) >>>> >>>> diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>> b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>> index d5a7249..c0fde44 100644 >>>> --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>> +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>> @@ -46,6 +46,7 @@ >>>> >>>> aliases { >>>> serial0 = &uart0; >>>> + ethernet0 = &emac; >>> >>> I think alias doesn’t require for probing emac, it will straight away >>> probed like >>> >>>> dm tree >>> .... >>> eth [ + ] ethernet@01c30000 >>> >>> Did you find any issue while detecting eth? >> >> Yes, it just didn't work ;-) I don't have a board here, but can post the >> error message later tonight. .... Net: phy interface7 Error: ethernet@01c30000 address not set. No ethernet found. .... >> In fact I was wondering about that already, maybe it's worth >> investigating this further. So the reason is that CONFIG_NET_RANDOM_ETHADDR isn't defined, so without the DT alias triggering the SID MAC generation there will be _no_ MAC address at all, which makes the driver give up. AFAIK on the A64 the MAC generation from the SID serial number works just fine, so we should in any case add the alias. I can resend the patch with an amended commit message. Cheers, Andre. >> >> But aside from that I think the MAC address calculation based on the SID >> serial number does not get triggered without an alias, so we need this >> line anyway. > > Correct, the MAC address code relies on the alias. > > Regards, > > Hans >
On Tue, Oct 25, 2016 at 3:59 AM, André Przywara <andre.przywara@arm.com> wrote: > On 21/10/16 11:28, Hans de Goede wrote: >> Hi, >> >> On 21-10-16 12:06, Andre Przywara wrote: >>> Hi, >>> >>> On 21/10/16 10:31, Jagan Teki wrote: >>>> On Fri, Oct 21, 2016 at 5:41 AM, Andre Przywara >>>> <andre.przywara@arm.com> wrote: >>>>> The sun8i-emac driver works fine with the A64 Ethernet IP, but we are >>>>> missing an alias entry to trigger the driver instantiation by U-Boot. >>>>> Add the line to point U-Boot to the Ethernet DT node. >>>>> This enables TFTP boot on the Pine64. >>>>> >>>>> Signed-off-by: Andre Przywara <andre.przywara@arm.com> >>>>> --- >>>>> arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + >>>>> 1 file changed, 1 insertion(+) >>>>> >>>>> diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>> b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>> index d5a7249..c0fde44 100644 >>>>> --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>> +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>> @@ -46,6 +46,7 @@ >>>>> >>>>> aliases { >>>>> serial0 = &uart0; >>>>> + ethernet0 = &emac; >>>> >>>> I think alias doesn’t require for probing emac, it will straight away >>>> probed like >>>> >>>>> dm tree >>>> .... >>>> eth [ + ] ethernet@01c30000 >>>> >>>> Did you find any issue while detecting eth? >>> >>> Yes, it just didn't work ;-) I don't have a board here, but can post the >>> error message later tonight. > > .... > Net: phy interface7 > > Error: ethernet@01c30000 address not set. > No ethernet found. > .... > >>> In fact I was wondering about that already, maybe it's worth >>> investigating this further. > > So the reason is that CONFIG_NET_RANDOM_ETHADDR isn't defined, so > without the DT alias triggering the SID MAC generation there will be > _no_ MAC address at all, which makes the driver give up. > AFAIK on the A64 the MAC generation from the SID serial number works > just fine, so we should in any case add the alias. I wondered how the alias related to MAC generation, tried fec_mxc without ethernet0 alias but couldn't find the change/generation on MAC at-all. thanks!
Hi, On 26-10-16 09:00, Jagan Teki wrote: > On Tue, Oct 25, 2016 at 3:59 AM, André Przywara <andre.przywara@arm.com> wrote: >> On 21/10/16 11:28, Hans de Goede wrote: >>> Hi, >>> >>> On 21-10-16 12:06, Andre Przywara wrote: >>>> Hi, >>>> >>>> On 21/10/16 10:31, Jagan Teki wrote: >>>>> On Fri, Oct 21, 2016 at 5:41 AM, Andre Przywara >>>>> <andre.przywara@arm.com> wrote: >>>>>> The sun8i-emac driver works fine with the A64 Ethernet IP, but we are >>>>>> missing an alias entry to trigger the driver instantiation by U-Boot. >>>>>> Add the line to point U-Boot to the Ethernet DT node. >>>>>> This enables TFTP boot on the Pine64. >>>>>> >>>>>> Signed-off-by: Andre Przywara <andre.przywara@arm.com> >>>>>> --- >>>>>> arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + >>>>>> 1 file changed, 1 insertion(+) >>>>>> >>>>>> diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>>> b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>>> index d5a7249..c0fde44 100644 >>>>>> --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>>> +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>>> @@ -46,6 +46,7 @@ >>>>>> >>>>>> aliases { >>>>>> serial0 = &uart0; >>>>>> + ethernet0 = &emac; >>>>> >>>>> I think alias doesn’t require for probing emac, it will straight away >>>>> probed like >>>>> >>>>>> dm tree >>>>> .... >>>>> eth [ + ] ethernet@01c30000 >>>>> >>>>> Did you find any issue while detecting eth? >>>> >>>> Yes, it just didn't work ;-) I don't have a board here, but can post the >>>> error message later tonight. >> >> .... >> Net: phy interface7 >> >> Error: ethernet@01c30000 address not set. >> No ethernet found. >> .... >> >>>> In fact I was wondering about that already, maybe it's worth >>>> investigating this further. >> >> So the reason is that CONFIG_NET_RANDOM_ETHADDR isn't defined, so >> without the DT alias triggering the SID MAC generation there will be >> _no_ MAC address at all, which makes the driver give up. >> AFAIK on the A64 the MAC generation from the SID serial number works >> just fine, so we should in any case add the alias. > > I wondered how the alias related to MAC generation, See the setup_environment() function in board/sunxi/board.c, we use the alias-es to determine for which interface we need to generate a random MAC and set ethaddr / eth1addr in the environment. Regards, Hans
On Wed, Oct 26, 2016 at 1:14 PM, Hans de Goede <hdegoede@redhat.com> wrote: > Hi, > > > On 26-10-16 09:00, Jagan Teki wrote: >> >> On Tue, Oct 25, 2016 at 3:59 AM, André Przywara <andre.przywara@arm.com> >> wrote: >>> >>> On 21/10/16 11:28, Hans de Goede wrote: >>>> >>>> Hi, >>>> >>>> On 21-10-16 12:06, Andre Przywara wrote: >>>>> >>>>> Hi, >>>>> >>>>> On 21/10/16 10:31, Jagan Teki wrote: >>>>>> >>>>>> On Fri, Oct 21, 2016 at 5:41 AM, Andre Przywara >>>>>> <andre.przywara@arm.com> wrote: >>>>>>> >>>>>>> The sun8i-emac driver works fine with the A64 Ethernet IP, but we are >>>>>>> missing an alias entry to trigger the driver instantiation by U-Boot. >>>>>>> Add the line to point U-Boot to the Ethernet DT node. >>>>>>> This enables TFTP boot on the Pine64. >>>>>>> >>>>>>> Signed-off-by: Andre Przywara <andre.przywara@arm.com> >>>>>>> --- >>>>>>> arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + >>>>>>> 1 file changed, 1 insertion(+) >>>>>>> >>>>>>> diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>>>> b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>>>> index d5a7249..c0fde44 100644 >>>>>>> --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>>>> +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>>>> @@ -46,6 +46,7 @@ >>>>>>> >>>>>>> aliases { >>>>>>> serial0 = &uart0; >>>>>>> + ethernet0 = &emac; >>>>>> >>>>>> >>>>>> I think alias doesn’t require for probing emac, it will straight away >>>>>> probed like >>>>>> >>>>>>> dm tree >>>>>> >>>>>> .... >>>>>> eth [ + ] ethernet@01c30000 >>>>>> >>>>>> Did you find any issue while detecting eth? >>>>> >>>>> >>>>> Yes, it just didn't work ;-) I don't have a board here, but can post >>>>> the >>>>> error message later tonight. >>> >>> >>> .... >>> Net: phy interface7 >>> >>> Error: ethernet@01c30000 address not set. >>> No ethernet found. >>> .... >>> >>>>> In fact I was wondering about that already, maybe it's worth >>>>> investigating this further. >>> >>> >>> So the reason is that CONFIG_NET_RANDOM_ETHADDR isn't defined, so >>> without the DT alias triggering the SID MAC generation there will be >>> _no_ MAC address at all, which makes the driver give up. >>> AFAIK on the A64 the MAC generation from the SID serial number works >>> just fine, so we should in any case add the alias. >> >> >> I wondered how the alias related to MAC generation, > > > See the setup_environment() function in board/sunxi/board.c, we use > the alias-es to determine for which interface we need to generate > a random MAC and set ethaddr / eth1addr in the environment. Ohh,OK. But not a good way of setting MAC on the board files. I think if we add .read_rom_hwaddr on the driver by placing SIM code on arch. thanks!
On Fri, Oct 21, 2016 at 5:41 AM, Andre Przywara <andre.przywara@arm.com> wrote: > The sun8i-emac driver works fine with the A64 Ethernet IP, but we are > missing an alias entry to trigger the driver instantiation by U-Boot. > Add the line to point U-Boot to the Ethernet DT node. > This enables TFTP boot on the Pine64. > > Signed-off-by: Andre Przywara <andre.przywara@arm.com> > --- > arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi b/arch/arm/dts/sun50i-a64-pine64-common.dtsi > index d5a7249..c0fde44 100644 > --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi > +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi > @@ -46,6 +46,7 @@ > > aliases { > serial0 = &uart0; > + ethernet0 = &emac; Better to have this alias on sun50i-a64.dtsi since the node is defined there? thanks!
On 26/10/16 19:51, Jagan Teki wrote: Hi, > On Fri, Oct 21, 2016 at 5:41 AM, Andre Przywara <andre.przywara@arm.com> wrote: >> The sun8i-emac driver works fine with the A64 Ethernet IP, but we are >> missing an alias entry to trigger the driver instantiation by U-Boot. >> Add the line to point U-Boot to the Ethernet DT node. >> This enables TFTP boot on the Pine64. >> >> Signed-off-by: Andre Przywara <andre.przywara@arm.com> >> --- >> arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + >> 1 file changed, 1 insertion(+) >> >> diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >> index d5a7249..c0fde44 100644 >> --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >> +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >> @@ -46,6 +46,7 @@ >> >> aliases { >> serial0 = &uart0; >> + ethernet0 = &emac; > > Better to have this alias on sun50i-a64.dtsi since the node is defined there? Mmh, I find examples for both ways (.dtsi vs. .dts) in the kernel. I need to learn what's the best practice here. But the U-Boot DTs need to get replaced anyway soon-ish, since the Linux DTs (which are WIP) have already diverged. So what about we keep this alias here next to the existing one for the 2016.11 release and possibly fix this later once we replace the DTs anyway with what gets merged into the kernel eventually? In the end of the day it's an easy and obvious fix which makes many people happy (since it allows TFTP boot of the Pine64). Cheers, Andre.
Hi André, Am 26.10.2016 um 22:50 schrieb André Przywara: > On 26/10/16 19:51, Jagan Teki wrote: >> On Fri, Oct 21, 2016 at 5:41 AM, Andre Przywara <andre.przywara@arm.com> wrote: >>> The sun8i-emac driver works fine with the A64 Ethernet IP, but we are >>> missing an alias entry to trigger the driver instantiation by U-Boot. >>> Add the line to point U-Boot to the Ethernet DT node. >>> This enables TFTP boot on the Pine64. >>> >>> Signed-off-by: Andre Przywara <andre.przywara@arm.com> >>> --- >>> arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + >>> 1 file changed, 1 insertion(+) >>> >>> diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>> index d5a7249..c0fde44 100644 >>> --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>> +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>> @@ -46,6 +46,7 @@ >>> >>> aliases { >>> serial0 = &uart0; >>> + ethernet0 = &emac; >> >> Better to have this alias on sun50i-a64.dtsi since the node is defined there? > > Mmh, I find examples for both ways (.dtsi vs. .dts) in the kernel. I > need to learn what's the best practice here. Last I was told by kernel maintainers (in an Amlogic context) aliases should be in the board's .dts[i] so that only the enabled nodes get numbered. If they're enabled at .dtsi level then I guess nothing is wrong with having the alias there. So I'm guessing the patch is good. Reviewed-by: Andreas Färber <afaerber@suse.de> Cheers, Andreas
On Thu, Oct 27, 2016 at 2:20 AM, André Przywara <andre.przywara@arm.com> wrote: > On 26/10/16 19:51, Jagan Teki wrote: > Hi, > >> On Fri, Oct 21, 2016 at 5:41 AM, Andre Przywara <andre.przywara@arm.com> wrote: >>> The sun8i-emac driver works fine with the A64 Ethernet IP, but we are >>> missing an alias entry to trigger the driver instantiation by U-Boot. >>> Add the line to point U-Boot to the Ethernet DT node. >>> This enables TFTP boot on the Pine64. >>> >>> Signed-off-by: Andre Przywara <andre.przywara@arm.com> >>> --- >>> arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + >>> 1 file changed, 1 insertion(+) >>> >>> diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>> index d5a7249..c0fde44 100644 >>> --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>> +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>> @@ -46,6 +46,7 @@ >>> >>> aliases { >>> serial0 = &uart0; >>> + ethernet0 = &emac; >> >> Better to have this alias on sun50i-a64.dtsi since the node is defined there? > > Mmh, I find examples for both ways (.dtsi vs. .dts) in the kernel. I > need to learn what's the best practice here. > > But the U-Boot DTs need to get replaced anyway soon-ish, since the Linux > DTs (which are WIP) have already diverged. > So what about we keep this alias here next to the existing one for the > 2016.11 release and possibly fix this later once we replace the DTs > anyway with what gets merged into the kernel eventually? OK, but please stick with the current discussion and still I see a good point to see alias on dtsi because the dts files which are using this dtsi shouldn't separately define the alias. And also status on dtsi is showing "disabled" are you enabled it some other patches? or missed? thanks!
Hi, On 21-10-16 02:11, Andre Przywara wrote: > The sun8i-emac driver works fine with the A64 Ethernet IP, but we are > missing an alias entry to trigger the driver instantiation by U-Boot. > Add the line to point U-Boot to the Ethernet DT node. > This enables TFTP boot on the Pine64. > > Signed-off-by: Andre Przywara <andre.przywara@arm.com> Thank you I've applied this to my tree and will include it in my next pull-req. Regards, Hans > --- > arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi b/arch/arm/dts/sun50i-a64-pine64-common.dtsi > index d5a7249..c0fde44 100644 > --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi > +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi > @@ -46,6 +46,7 @@ > > aliases { > serial0 = &uart0; > + ethernet0 = &emac; > }; > > soc { >
On Fri, Oct 28, 2016 at 11:21 PM, Jagan Teki <jagan@openedev.com> wrote: > On Thu, Oct 27, 2016 at 2:20 AM, André Przywara <andre.przywara@arm.com> wrote: >> On 26/10/16 19:51, Jagan Teki wrote: >> Hi, >> >>> On Fri, Oct 21, 2016 at 5:41 AM, Andre Przywara <andre.przywara@arm.com> wrote: >>>> The sun8i-emac driver works fine with the A64 Ethernet IP, but we are >>>> missing an alias entry to trigger the driver instantiation by U-Boot. >>>> Add the line to point U-Boot to the Ethernet DT node. >>>> This enables TFTP boot on the Pine64. >>>> >>>> Signed-off-by: Andre Przywara <andre.przywara@arm.com> >>>> --- >>>> arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + >>>> 1 file changed, 1 insertion(+) >>>> >>>> diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>> index d5a7249..c0fde44 100644 >>>> --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>> +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>> @@ -46,6 +46,7 @@ >>>> >>>> aliases { >>>> serial0 = &uart0; >>>> + ethernet0 = &emac; >>> >>> Better to have this alias on sun50i-a64.dtsi since the node is defined there? >> >> Mmh, I find examples for both ways (.dtsi vs. .dts) in the kernel. I >> need to learn what's the best practice here. >> >> But the U-Boot DTs need to get replaced anyway soon-ish, since the Linux >> DTs (which are WIP) have already diverged. >> So what about we keep this alias here next to the existing one for the >> 2016.11 release and possibly fix this later once we replace the DTs >> anyway with what gets merged into the kernel eventually? > > OK, but please stick with the current discussion and still I see a > good point to see alias on dtsi because the dts files which are using > this dtsi shouldn't separately define the alias. > > And also status on dtsi is showing "disabled" are you enabled it some > other patches? or missed? Any response for this? thanks!
On 30/10/16 08:48, Jagan Teki wrote: > On Fri, Oct 28, 2016 at 11:21 PM, Jagan Teki <jagan@openedev.com> wrote: >> On Thu, Oct 27, 2016 at 2:20 AM, André Przywara <andre.przywara@arm.com> wrote: >>> On 26/10/16 19:51, Jagan Teki wrote: >>> Hi, >>> >>>> On Fri, Oct 21, 2016 at 5:41 AM, Andre Przywara <andre.przywara@arm.com> wrote: >>>>> The sun8i-emac driver works fine with the A64 Ethernet IP, but we are >>>>> missing an alias entry to trigger the driver instantiation by U-Boot. >>>>> Add the line to point U-Boot to the Ethernet DT node. >>>>> This enables TFTP boot on the Pine64. >>>>> >>>>> Signed-off-by: Andre Przywara <andre.przywara@arm.com> >>>>> --- >>>>> arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + >>>>> 1 file changed, 1 insertion(+) >>>>> >>>>> diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>> index d5a7249..c0fde44 100644 >>>>> --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>> +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>> @@ -46,6 +46,7 @@ >>>>> >>>>> aliases { >>>>> serial0 = &uart0; >>>>> + ethernet0 = &emac; >>>> >>>> Better to have this alias on sun50i-a64.dtsi since the node is defined there? >>> >>> Mmh, I find examples for both ways (.dtsi vs. .dts) in the kernel. I >>> need to learn what's the best practice here. >>> >>> But the U-Boot DTs need to get replaced anyway soon-ish, since the Linux >>> DTs (which are WIP) have already diverged. >>> So what about we keep this alias here next to the existing one for the >>> 2016.11 release and possibly fix this later once we replace the DTs >>> anyway with what gets merged into the kernel eventually? >> >> OK, but please stick with the current discussion and still I see a >> good point to see alias on dtsi because the dts files which are using >> this dtsi shouldn't separately define the alias. >> >> And also status on dtsi is showing "disabled" are you enabled it some >> other patches? or missed? We have the status = "okay" in sun50i-a64-pine64-plus.dts, which is the only dts that is used these days. The non-plus board version has a different PHY (100 MBit only) and a quick test a while ago showed that this didn't work out of the box, so we just don't enable it there. It seems like that not many people have this version of the board and they would need to hack U-Boot to use it anyway (because defconfig select -plus.dts). This will be revisited when the SPL FIT support introduces board detection and we actually choose the proper DT for each board. Cheers, Andre.
On Sun, Oct 30, 2016 at 3:16 PM, André Przywara <andre.przywara@arm.com> wrote: > On 30/10/16 08:48, Jagan Teki wrote: >> On Fri, Oct 28, 2016 at 11:21 PM, Jagan Teki <jagan@openedev.com> wrote: >>> On Thu, Oct 27, 2016 at 2:20 AM, André Przywara <andre.przywara@arm.com> wrote: >>>> On 26/10/16 19:51, Jagan Teki wrote: >>>> Hi, >>>> >>>>> On Fri, Oct 21, 2016 at 5:41 AM, Andre Przywara <andre.przywara@arm.com> wrote: >>>>>> The sun8i-emac driver works fine with the A64 Ethernet IP, but we are >>>>>> missing an alias entry to trigger the driver instantiation by U-Boot. >>>>>> Add the line to point U-Boot to the Ethernet DT node. >>>>>> This enables TFTP boot on the Pine64. >>>>>> >>>>>> Signed-off-by: Andre Przywara <andre.przywara@arm.com> >>>>>> --- >>>>>> arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + >>>>>> 1 file changed, 1 insertion(+) >>>>>> >>>>>> diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>>> index d5a7249..c0fde44 100644 >>>>>> --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>>> +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi >>>>>> @@ -46,6 +46,7 @@ >>>>>> >>>>>> aliases { >>>>>> serial0 = &uart0; >>>>>> + ethernet0 = &emac; >>>>> >>>>> Better to have this alias on sun50i-a64.dtsi since the node is defined there? >>>> >>>> Mmh, I find examples for both ways (.dtsi vs. .dts) in the kernel. I >>>> need to learn what's the best practice here. >>>> >>>> But the U-Boot DTs need to get replaced anyway soon-ish, since the Linux >>>> DTs (which are WIP) have already diverged. >>>> So what about we keep this alias here next to the existing one for the >>>> 2016.11 release and possibly fix this later once we replace the DTs >>>> anyway with what gets merged into the kernel eventually? >>> >>> OK, but please stick with the current discussion and still I see a >>> good point to see alias on dtsi because the dts files which are using >>> this dtsi shouldn't separately define the alias. >>> >>> And also status on dtsi is showing "disabled" are you enabled it some >>> other patches? or missed? > > We have the status = "okay" in sun50i-a64-pine64-plus.dts, which is the > only dts that is used these days. > The non-plus board version has a different PHY (100 MBit only) and a > quick test a while ago showed that this didn't work out of the box, so > we just don't enable it there. It seems like that not many people have > this version of the board and they would need to hack U-Boot to use it > anyway (because defconfig select -plus.dts). > This will be revisited when the SPL FIT support introduces board > detection and we actually choose the proper DT for each board. OK. Reviewed-by: Jagan Teki <jagan@openedev.com> thanks!
diff --git a/arch/arm/dts/sun50i-a64-pine64-common.dtsi b/arch/arm/dts/sun50i-a64-pine64-common.dtsi index d5a7249..c0fde44 100644 --- a/arch/arm/dts/sun50i-a64-pine64-common.dtsi +++ b/arch/arm/dts/sun50i-a64-pine64-common.dtsi @@ -46,6 +46,7 @@ aliases { serial0 = &uart0; + ethernet0 = &emac; }; soc {
The sun8i-emac driver works fine with the A64 Ethernet IP, but we are missing an alias entry to trigger the driver instantiation by U-Boot. Add the line to point U-Boot to the Ethernet DT node. This enables TFTP boot on the Pine64. Signed-off-by: Andre Przywara <andre.przywara@arm.com> --- arch/arm/dts/sun50i-a64-pine64-common.dtsi | 1 + 1 file changed, 1 insertion(+)