Re: [PATCH v2] aspeed: Deprecate the tacoma-bmc machine

[Guenter Roeck <linux@roeck-us.net>]

On 8/28/24 05:35, Cédric Le Goater wrote:
> On 8/26/24 20:50, Guenter Roeck wrote:
>> Hi,
>>
>> On 8/26/24 02:58, Cédric Le Goater wrote:
>>> Hello Guenter,
>>>
>>> On 8/9/24 00:05, Guenter Roeck wrote:
>>>> Hi,
>>>>
>>>> On Tue, Jun 25, 2024 at 09:08:30AM +0200, Cédric Le Goater wrote:
>>>>> The tacoma-bmc machine was a board including an AST2600 SoC based BMC
>>>>> and a witherspoon like OpenPOWER system. It was used for bring up of
>>>>> the AST2600 SoC in labs. It can be easily replaced by the rainier-bmc
>>>>> machine which is part of a real product offering.
>>>>>
>>>>> Signed-off-by: Cédric Le Goater <clg@redhat.com>
>>>>
>>>> I have been using tacoma-bmc to test tcg,tpm-tis-i2c functionality
>>>> on arm. rainier-bmc doesn't support that, and other IBM BMCs which
>>>> do support it (bonnell, everest, system1) are not supported in qemu.
>>>>
>>>> Do you have a suggested alternative ?
>>>
>>> Could you use the ast2600-evb machine instead ? as done in
>>> machine_aspeed.py, see routine test_arm_ast2600_evb_buildroot_tpm.
>>>
>>
>> Unfortunately, that does not work for me because that requires instantiating
>> the tpm chip from the CLI by writing into the new_device sysfs attribute,
>> and I can not do that in my test environment.
> 
> Ah. too bad.
> 
>>> We can't add a "tpm-tis-i2c" device to the tacoma-bmc machine init
>>> routine because a TPM backend is required.
>>>
>>
>> Not sure I understand; tacoma-bmc instantiates the TPM chip through its
>> devicetree file which is what I was looking for.
> 
> I meant at the "HW" board level in QEMU.
> 
> We can not instantiate the TPM I2C chip device model in the tacoma-bmc
> machine init routine and attach it to the I2C bus because of the required
> TPM backend. This means that the device is necessarily defined on the QEMU
> command line and this makes the ast2600-evb and tacoma-bmc machine very
> similar in terms of HW definitions.
> 

Yes, I found that as well.

>> I solved the problem by adding support for IBM Bonnell (which instantiates
>> the TPM chip through its devicetree file, similar to tacoma-bmc) to my local
>> copy of qemu. 
> 
> Hmm, did you copy the rainier-bmc machine definition ?
> 
For aspeed_machine_bonnell_class_init(), pretty much yes, since I don't know
the actual hardware. For I2C initialization I used the devicetree file.
You can find the patch in the master-local or v9.1.0-local branches
of my qemu clone at https://github.com/groeck/qemu if you are interested.

>> It isn't perfect since I don't know the correct HW pin strapping
>> and reused the strapping from Rainier, but it works for me.
> 
> Keeping the tacoma-bmc machine is fine if there is a use for it. Testing
> the TPM I2C device driver is certainly a good use but we should reflect
> that in QEMU also (so that we don't forget). Could we change the test in
> machine_aspeed.py to use the tacoma-bmc machine instead ? and revert the
> deprecation patch of course.
> 

No need from my perspective. This isn't the only patch I carry on top of
upstream qemu, after all.

Thanks,
Guenter