diff mbox series

[meta-ti,dunfell,3/4] conf: machine: j721s2: Add OP-TEE flavour

Message ID 20221010134649.20977-4-j-choudhary@ti.com
State Superseded
Delegated to: Ryan Eatmon
Headers show
Series TRNG support for J721S2 and J784S4 | expand

Commit Message

Jayesh Choudhary Oct. 10, 2022, 1:46 p.m. UTC
Add the new OP-TEE flavour for J721S2 instead of default k3-j721e.
The TI-SCI ID for MCU domain sa2ul/sa3ul are not the same for
J721E and J721S2. So a new flavour is added in upstream OP-TEE.

Signed-off-by: Jayesh Choudhary <j-choudhary@ti.com>
---
 conf/machine/j721s2-evm.conf | 1 +
 1 file changed, 1 insertion(+)

Comments

Andrew Davis Oct. 10, 2022, 2:26 p.m. UTC | #1
On 10/10/22 8:46 AM, Jayesh Choudhary wrote:
> Add the new OP-TEE flavour for J721S2 instead of default k3-j721e.
> The TI-SCI ID for MCU domain sa2ul/sa3ul are not the same for
> J721E and J721S2. So a new flavour is added in upstream OP-TEE.
> 

You are not really adding a new flavor for J721S2 but rather
re-using J784s4 as they both use SA3UL. Would be better to
add a different commit message here to make that clear.

Otherwise the series LGTM,

Acked-by: Andrew Davis <afd@ti.com>

> Signed-off-by: Jayesh Choudhary <j-choudhary@ti.com>
> ---
>   conf/machine/j721s2-evm.conf | 1 +
>   1 file changed, 1 insertion(+)
> 
> diff --git a/conf/machine/j721s2-evm.conf b/conf/machine/j721s2-evm.conf
> index ba5e8def..ec9b7454 100644
> --- a/conf/machine/j721s2-evm.conf
> +++ b/conf/machine/j721s2-evm.conf
> @@ -9,6 +9,7 @@ MACHINE_FEATURES += "gpu"
>   SERIAL_CONSOLES = "115200;ttyS10"
>   SERIAL_CONSOLES_CHECK = "${SERIAL_CONSOLES}"
>   TFA_K3_USART = "0x8"
> +OPTEEMACHINE = "k3-j784s4"
>   OPTEE_K3_USART = "0x8"
>   
>   KERNEL_DEVICETREE = " \
Ryan Eatmon Oct. 10, 2022, 3:07 p.m. UTC | #2
On 10/10/2022 9:26, Andrew Davis wrote:
> On 10/10/22 8:46 AM, Jayesh Choudhary wrote:
>> Add the new OP-TEE flavour for J721S2 instead of default k3-j721e.
>> The TI-SCI ID for MCU domain sa2ul/sa3ul are not the same for
>> J721E and J721S2. So a new flavour is added in upstream OP-TEE.
>>
> 
> You are not really adding a new flavor for J721S2 but rather
> re-using J784s4 as they both use SA3UL. Would be better to
> add a different commit message here to make that clear.


On a similar note, it was bad enough that we used j721s2 for the j784s2 
since some things were not ready, but this patch goes the other way and 
uses j784s4 in in the j721s2 config...  Why are we doing that?  That 
seems really confusing.

Are doing this because we are not upstreaming something to optee for 
j721s2 and are just reusing something from j784s4?


> Otherwise the series LGTM,
> 
> Acked-by: Andrew Davis <afd@ti.com>
> 
>> Signed-off-by: Jayesh Choudhary <j-choudhary@ti.com>
>> ---
>>   conf/machine/j721s2-evm.conf | 1 +
>>   1 file changed, 1 insertion(+)
>>
>> diff --git a/conf/machine/j721s2-evm.conf b/conf/machine/j721s2-evm.conf
>> index ba5e8def..ec9b7454 100644
>> --- a/conf/machine/j721s2-evm.conf
>> +++ b/conf/machine/j721s2-evm.conf
>> @@ -9,6 +9,7 @@ MACHINE_FEATURES += "gpu"
>>   SERIAL_CONSOLES = "115200;ttyS10"
>>   SERIAL_CONSOLES_CHECK = "${SERIAL_CONSOLES}"
>>   TFA_K3_USART = "0x8"
>> +OPTEEMACHINE = "k3-j784s4"
>>   OPTEE_K3_USART = "0x8"
>>   KERNEL_DEVICETREE = " \
Jayesh Choudhary Oct. 10, 2022, 3:11 p.m. UTC | #3
On 10/10/22 20:37, Ryan Eatmon wrote:
> 
> 
> On 10/10/2022 9:26, Andrew Davis wrote:
>> On 10/10/22 8:46 AM, Jayesh Choudhary wrote:
>>> Add the new OP-TEE flavour for J721S2 instead of default k3-j721e.
>>> The TI-SCI ID for MCU domain sa2ul/sa3ul are not the same for
>>> J721E and J721S2. So a new flavour is added in upstream OP-TEE.
>>>
>>
>> You are not really adding a new flavor for J721S2 but rather
>> re-using J784s4 as they both use SA3UL. Would be better to
>> add a different commit message here to make that clear.
> 
> 
> On a similar note, it was bad enough that we used j721s2 for the j784s2 
> since some things were not ready, but this patch goes the other way and 
> uses j784s4 in in the j721s2 config...  Why are we doing that?  That 
> seems really confusing.
>
> Are doing this because we are not upstreaming something to optee for 
> j721s2 and are just reusing something from j784s4?
> 

Yes Ryan,
sa3ul support has been added to upstream optee for j784s4 already.
For j721s2, no additional changes are required.
So we are reusing the same optee flavour.

-Jayesh


> 
>> Otherwise the series LGTM,
>>
>> Acked-by: Andrew Davis <afd@ti.com>
>>
>>> Signed-off-by: Jayesh Choudhary <j-choudhary@ti.com>
>>> ---
>>>   conf/machine/j721s2-evm.conf | 1 +
>>>   1 file changed, 1 insertion(+)
>>>
>>> diff --git a/conf/machine/j721s2-evm.conf b/conf/machine/j721s2-evm.conf
>>> index ba5e8def..ec9b7454 100644
>>> --- a/conf/machine/j721s2-evm.conf
>>> +++ b/conf/machine/j721s2-evm.conf
>>> @@ -9,6 +9,7 @@ MACHINE_FEATURES += "gpu"
>>>   SERIAL_CONSOLES = "115200;ttyS10"
>>>   SERIAL_CONSOLES_CHECK = "${SERIAL_CONSOLES}"
>>>   TFA_K3_USART = "0x8"
>>> +OPTEEMACHINE = "k3-j784s4"
>>>   OPTEE_K3_USART = "0x8"
>>>   KERNEL_DEVICETREE = " \
>
diff mbox series

Patch

diff --git a/conf/machine/j721s2-evm.conf b/conf/machine/j721s2-evm.conf
index ba5e8def..ec9b7454 100644
--- a/conf/machine/j721s2-evm.conf
+++ b/conf/machine/j721s2-evm.conf
@@ -9,6 +9,7 @@  MACHINE_FEATURES += "gpu"
 SERIAL_CONSOLES = "115200;ttyS10"
 SERIAL_CONSOLES_CHECK = "${SERIAL_CONSOLES}"
 TFA_K3_USART = "0x8"
+OPTEEMACHINE = "k3-j784s4"
 OPTEE_K3_USART = "0x8"
 
 KERNEL_DEVICETREE = " \