Am 17. April 2025 21:40:44 UTC schrieb BALATON Zoltan <bala...@eik.bme.hu>:
>On Thu, 17 Apr 2025, Thomas Huth wrote:
>> On 01/11/2022 23.29, Philippe Mathieu-Daudé wrote:
>>> Adds missing functionality to e500plat machine which increases the
>>> chance of given "real" firmware images to access SD cards.
>>>
>>> Signed-off-by: Bernhard Beschow <shen...@gmail.com>
>>> Message-Id: <20221018210146.193159-8-shen...@gmail.com>
>>> [PMD: Simplify using create_unimplemented_device("esdhc")]
>>> Signed-off-by: Philippe Mathieu-Daudé <phi...@linaro.org>
>>> ---
>>
>> Hi!
>>
>> I recently noticed that the QEMU advent calendar 2018 day 19 image
>> (https://www.qemu-advent-calendar.org/2018/download/day19.tar.xz) is
>> flooding the console with "mmc0: Internal clock never stabilised" messages
>> when it's started like this:
>>
>> /qemu-system-ppc64 -vga none -nographic -monitor none -M ppce500 -cpu e5500
>> -kernel ../day19/uImage
>>
>> This was not the case when I assembled the image in 2018. I bisected the
>> problem and apparently this patch here is causing the problem. Do you know
>> whether there is a way to fix this again?
>
>That patch added the MMC/SD controller so no wonder it did not appear before.
>The error seems to be in Linux esdhci driver where it checks for:
>
>sdhci_readl(host, ESDHC_PRSSTAT) & ESDHC_CLOCK_STABLE
>
>So I think that bit should be set somewhere. Berhard had some patches for e500
>so maybe there's already a fix for this somewhere.
This logic is already implemented in the USDHC device model. Given that the
e500 and the i.mx machines need the same modifications to make u-boot work, and
given that the imx25 machine also uses the USDHC model even though the original
has an ESDHC, the USDHC might be a better fit here. AFAICS only the interrupt
bits specific to the ESDHC need special treatment here -- which could be
achieved e.g. by subclassing.
I could send a patch for that. Volunteers ahead for a timely solution.
Best regards,
Bernhard
>
>Regards,
>BALATON Zoltan
