On 8/19/25 13:10, Marek Vasut wrote:
> On 8/13/25 10:53 AM, Patrice CHOTARD wrote:
>>
>>
>> On 8/8/25 17:24, Marek Vasut wrote:
>>> On 8/8/25 5:11 PM, Patrice Chotard wrote:
>>>> This series introduces 2 new drivers:
>>>> _ STM32 Octo Memory Manager driver
>>>> _ STM32 OSPI driver
>>>>
>>>> It adds resource check helpers which are used by OMM driver.
>>>> It adds and enables OMM and OSPI driver support for stm32mp257f-ev1 board
>>> Can I test this with mainline TFA and mainline OpTee-OS ?
>>
>> Hi Marek
>>
>> Currently, it misses regulator support in OPTEE.
>> But you can enable "manually" VDDIO3 and VDDIO4 in U-Boot console
> Actually, I need to enable VDDIO3/4 even earlier, otherwise I get no UART and
> no I2C, so that's not a problem. VDDIO3/4 handling seems to be broken in the
> ST downstream TFA with almost 1000 patches on top of old version of TFA, and
> upstream TFA for STM32MP25xx is entirely bare bones that it barely works.
>
> I also didn't manage to start upstream U-Boot on STM32MP25xx at all yet, I
> still have to upport about 400 patches (reduced from some 800 from ST
> downstream old U-Boot fork), and the result won't work unless I use the
> obsolete ST downstream OpTee-OS 4.0 fork with almost 1700 patches on top.
>
> Maybe the upstream OpTee-OS is missing something, like the entire SCMI
> implementation ?
Hi Marek
If you refer to my previous e-mail with the stm32mp257f-ev1 boot log,
with the following upstream components:
TF-A: v2.13.0-304-g000fe221b8
OP-TEE: 4.7.0-26-g847ee2932
U-Boot: v2025.07
I am able to boot a STM32MP257f-EV1 board.
Patrice
