Re: [PATCH] arm64: dts: qcom: Add device tree for TUXEDO Elite 14 Gen1

[Georg Gottleuber <g.gottleuber@tuxedocomputers.com>]

Am 21.03.25 um 16:51 schrieb Dmitry Baryshkov:
> On Fri, 21 Mar 2025 at 17:49, Ettore Chimenti
> <ettore.chimenti@linaro.org> wrote:
>>
>> Hi Dmitry,
>>
>> On Tue, Mar 18, 2025 at 11:36:32PM +0200, Dmitry Baryshkov wrote:
>>> On Tue, Mar 18, 2025 at 04:24:27PM +0100, Georg Gottleuber wrote:
>>>> Am 07.03.25 um 07:45 schrieb Dmitry Baryshkov:
>>>> [...]
>>>>>>> diff --git a/arch/arm64/boot/dts/qcom/x1e80100-tuxedo-elite-14-gen1.dts b/arch/arm64/boot/dts/qcom/x1e80100-tuxedo-elite-14-gen1.dts
>>>>>>> new file mode 100644
>>>>>>> index 000000000000..86bdec4a2dd8
>>>>>>> --- /dev/null
>>>>>>> +++ b/arch/arm64/boot/dts/qcom/x1e80100-tuxedo-elite-14-gen1.dts
>>>>>>
>>>>>>> +&gpu {
>>>>>>> +       status = "okay";
>>>>>>> +
>>>>>>> +       zap-shader {
>>>>>>> +               firmware-name = "qcom/a740_zap.mbn";
>>>>>>
>>>>>> Are the laptop's OEM key/security fuses not blown?
>>>>>
>>>>> Can this laptop use "qcom/x1e80100/gen70500_zap.mbn" which is already a
>>>>> part of linux-firmware?
>>>>
>>>> It seems so.
>>>>
>>>> Because there were no logs about loading zap.mbn, I activated dyndbg
>>>> (dyndbg="file drivers/base/firmware_loader/main.c +fmp"). See attachment
>>>> for dmesg output. But GUI freezes after sddm login.
>>>
>>> Does it happen only with this ZAP or does it happen with the ZAP from
>>> WIndows too? Can you run some simple GPU workload, like kmscube from the
>>> console?
>>>
>>
>> It seems to work fine changing the `firmware-name` property to
>> "qcom/x1e80100/gen70500_zap.mbn" and updating to latest mesa on Debian
>> Sid (25.0.1).
>>
>> Also tried with linux-firmware binary blob.
> 
> Then please post an updated patch, using firmware & ZAP from linux-firmware.

I'm sorry, all our devices are engineering examples (including Ettores).
The mass production devices will be fused.