Date: Sun, 07 Feb 2021 22:12:21 -0600 From: "Mike Karels" <mike@karels.net> To: "Matthew Grooms" <mgrooms@shrew.net> Cc: "Klaus =?utf-8?q?K=C3=BCchemann?=" <maciphone2@googlemail.com>, "Torfinn Ingolfsen" <torfinn.ingolfsen@getmail.no>, freebsd-arm@freebsd.org Subject: Re: rpi4 network boot Message-ID: <31C01D43-A85B-4D91-9F0B-671CEADA75A4@karels.net> In-Reply-To: <cfd37d1a-3178-3ddf-b06b-ef13cdf15b86@shrew.net> References: <d385f2b3-87e0-1bd5-d0f8-76ba1eb99816@shrew.net> <3a1acba5-8662-55cb-c5c4-98ec7873f78f@shrew.net> <e324f942-43ac-d2ab-3e7f-47ca91e40bee@shrew.net> <082BC0B9-2538-43D5-9D5E-C18B4DEE9ECB@googlemail.com> <de86ca7c-dd8a-511a-6063-ada6eb451488@shrew.net> <C6845ECB-23BE-48DB-BD78-A56814FCB39C@googlemail.com> <d02ac38c-ac8e-fc84-ba65-1c49e9b73bbf@shrew.net> <2D8B4921-CCEB-4203-B616-C0C86612C1E5@googlemail.com> <24127be9-ff90-ad97-ada5-2c2175f34fd4@shrew.net> <20210207201956.f289206d5eb560c034fea7a4@getmail.no> <7DB4B9E5-2CA4-4AA3-B3EB-61A4807CFC9E@googlemail.com> <cfd37d1a-3178-3ddf-b06b-ef13cdf15b86@shrew.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On 7 Feb 2021, at 19:24, Matthew Grooms wrote:
> On 2/7/2021 2:00 PM, Klaus K=C3=BCchemann wrote:
>> Am 07.02.2021 um 20:19 schrieb Torfinn Ingolfsen=20
>> <torfinn.ingolfsen@getmail.no>:
>>> I believe he is asking you to verify that the fstab file for the=20
>>> _client_ (aka the diskless machine your'e trying to boot) is=20
>>> correct. That file exists as
>>> ${NFSROOTDIR}/etc/fstab on the boot server.
>>>
>>> HTH
>>> --=20
>>> Torfinn Ingolfsen <torfinn.ingolfsen@getmail.no>
>>> _______________________________________________
>>> freebsd-arm@freebsd.org mailing list
>>> https://lists.freebsd.org/mailman/listinfo/freebsd-arm
>>> To unsubscribe, send any mail to=20
>>> "freebsd-arm-unsubscribe@freebsd.org=E2=80=9C
>> Yes, the fstab-setup is important=E2=80=A6... but then I realized Matt=
hew`s=20
>> boot environment :
>> He has u-boot sitting on the pxe-Server, whereas the eeprom was=20
>> flashed for completely diskless netboot
>> as a requirement .
>> With that setup in fact we will end up in n infinite loop of an=20
>> 'genet0: gen_intr2=E2=80=98 -console-output,
>> regardless of fstab or whatever configuration elsewhere I tried.
> Thanks for testing and re-producing that. Corroborating test results=20
> were also posted by others in the FreeBSD forum thread I referenced=20
> earlier.
>> I have now refreshed the eeprom to it=E2=80=99s default(where diskless=
=20
>> pxe-boot is disabled )
>> And again successfully net-booted (14_current in this case) :
>>
>> https://dmesgd.nycbug.org/index.cgi?do=3Dview&id=3D5917
>>
>> Since netboot from u-boot on the uSD requiresdonly 1 command by hand=20
>> (load net0:/boot/kernel/kernel
>> )(+ command :=E2=80=99boot=E2=80=99) and there=E2=80=99s no problem wi=
th the=20
>> genet-driver I guess that
>> It=E2=80=99s an issue of the pxe-flashed eeprom where we don=E2=80=99t=
yet know=20
>> exactly what happens.
>
> I have a slightly different take: The network device appears to work=20
> well initially when used by the eeprom and after the hand-off to=20
> u-boot. The FreeBSD genet driver generally works well unless the=20
> device is first used by eeprom/u-boot. I suspect that u-boot is=20
> performing some task that leaves the device in a state that is=20
> incompatible what FreeBSD expects and/or can work with. Perhaps there=20
> is a way to force some a full device reset of genet either before=20
> u-boot hands things off to FreeBSD or right after FreeBSD loads the=20
> genet driver? If so, maybe that would be worth trying, even as a hacky=20
> test, to see if it's the network device state that's wonky.
That=E2=80=99s a reasonable theory, but I don=E2=80=99t actually know how=
to reset=20
the chip
(I=E2=80=99ve never seen documentation for it). I=E2=80=99m also not con=
fident that=20
the
driver would fully initialize the chip from a reset. However, there is
probably more to the problem. Klaus told me that the irq values are
different when using PXE, which means the OS is getting different config
information as well. The irqs print in the main auto configuration line
for genet0. But it=E2=80=99s possible that something as simple as callin=
g the
same interrupt routine from both interrupts might work.
Mike
>> So for now I guess netboot from uSD is a painless and well working=20
>> setup,
>> while to be honest:
>> I don=E2=80=99t remember exactly at the moment how I figured to get it=
to=20
>> work after trials&errors :-) Ha Ha
>> But I=E2=80=99ll leave the uSD untouched or copy the folder to have a =
point=20
>> for a tutorial or so if someone is interested in.
>
> Can that be done without human intervention? Perhaps with some sort of=20
> u-boot or EFI script that gets executed? If this method requires the=20
> user to enter a command to boot from the network, it's not an=20
> alternative to PXE.
>
> Thanks,
>
> -Matthew
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?31C01D43-A85B-4D91-9F0B-671CEADA75A4>