From: Sam Ravnborg <sam@ravnborg.org>
To: Wolfram Sang <wsa@kernel.org>, Sascha Hauer <sha@pengutronix.de>,
barebox@lists.infradead.org
Subject: Re: [PATCH 0/3] mtd: nand: atmel: legacy: fix boot on USB-A9G20
Date: Tue, 21 Mar 2023 20:02:34 +0100 [thread overview]
Message-ID: <ZBn/ShAwKik7TdvZ@ravnborg.org> (raw)
In-Reply-To: <ZBnnbU1SUkNm4ryz@sai.fritz.box>
Hi Wolfram.
On Tue, Mar 21, 2023 at 06:20:45PM +0100, Wolfram Sang wrote:
> Hi Sascha!
>
> > Nice to hear from you here ;)
>
> Yeah, it has been only 10 years... :)
>
> > I have no idea how the SDRAM setup is done on the USB-A9G20. There seems
> > to be SDRAM setup code for the USB-A9263, but not for the USB-A9G20. Is
> > there some AT91Bootstrap required?
>
> Yes. There is a bootstrap required: https://github.com/linux4sam/at91bootstrap
> However, an old version is needed. I am still fighting the details.
>
> I accidently erased it, so my journey for unbricking the device began.
> at91bootstrap, openocd, barebox - they all once supported A9G20, but
> nowadays all of them were broken. Well, I fixed openocd and barebox
> mostly so far, the bootstrap is still the missing piece.
You may be hit by the newer gcc optimize the delays in
the sdram init code in at91bootstrap.
This code:
/* Step#4 The minimum pause of 200 us is provided to precede any single toggle */
for (i = 0; i < 1000; i++) ;
Are optimized away with newer gcc's.
You can find it repeated several times in the init code.
I think this was the only part what broke at91bootstrap for me.
It was Ahmad, IIRC, that told me about it - so not my finding.
>
> Super interesting about the A9263 setup code. I missed it so far. Does
> that mean that barebox can replace the at91bootstrap?
Yes. The patch series I pointed at generates a bare variant that I
successfully used to boot my board direct from ROMBOOT.
It is a direct replacement for at91bootstrap and loads barebox as the
next step bootloader.
I really like the way it is all integrated and how the two barebox
variants are built in one go.
Sam
>
> > Anyway, what's really missing is DT support. I scribbled a patch to get
> > you started in case you are motivated. Basically it's: Compile in the
>
> Motivated? No way ;) Obliged? More so. But first, I want to unbrick my
> device. First make it work, then make it beautiful.
>
> That being said, the two patches from this series are okay? I'd think
> they are still needed for other legacy systems even if I convert A9G20
> to DT in some ages.
>
> Thanks for your help!
>
> Wolfram
next prev parent reply other threads:[~2023-03-21 19:04 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-19 15:49 [PATCH 0/3] mtd: nand: atmel: legacy: fix boot on USB-A9G20 Wolfram Sang
2023-03-19 15:49 ` [PATCH 1/3] mtd: nand: atmel: legacy: add 'algo' to use Wolfram Sang
2023-03-19 15:49 ` [PATCH 2/3] mtd: nand: atmel: legacy: use proper ecc_shift Wolfram Sang
2023-03-19 15:49 ` [PATCH 3/3] mtd: nand: atmel: legacy: remove superfluous code Wolfram Sang
2023-03-19 18:51 ` [PATCH 0/3] mtd: nand: atmel: legacy: fix boot on USB-A9G20 Sam Ravnborg
2023-03-19 21:28 ` Wolfram Sang
2023-03-21 8:58 ` Sascha Hauer
2023-03-21 9:44 ` Alexander Dahl
2023-03-21 16:48 ` Sam Ravnborg
2023-03-21 17:20 ` Wolfram Sang
2023-03-21 19:02 ` Sam Ravnborg [this message]
2023-03-21 20:53 ` Sam Ravnborg
2023-03-21 21:42 ` Wolfram Sang
2023-03-22 7:37 ` Sascha Hauer
2023-04-27 17:25 ` Wolfram Sang
2023-04-27 17:26 ` Wolfram Sang
2023-03-22 7:41 ` Sascha Hauer
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ZBn/ShAwKik7TdvZ@ravnborg.org \
--to=sam@ravnborg.org \
--cc=barebox@lists.infradead.org \
--cc=sha@pengutronix.de \
--cc=wsa@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.