All the mail mirrored from lore.kernel.org
 help / color / mirror / Atom feed
From: "Limaye, Aniket" <a-limaye@ti.com>
To: Ryan Eatmon <reatmon@ti.com>, <meta-ti@lists.yoctoproject.org>
Cc: <b-padhi@ti.com>
Subject: Re: [meta-ti][master/scarthgap][PATCH] conf: machine: add AM69-SK machine configuration
Date: Mon, 29 Apr 2024 23:26:22 +0530	[thread overview]
Message-ID: <9a083386-48a0-4604-87d5-721cd4d8dce5@ti.com> (raw)
In-Reply-To: <308e8874-351e-4c61-94c5-9de218c038f8@ti.com>


On 4/29/2024 8:30 PM, Ryan Eatmon wrote:
>
>
> On 4/29/2024 8:51 AM, Limaye, Aniket wrote:
>>
>> On 4/26/2024 1:06 AM, Ryan Eatmon wrote:
>>>
>>>
>>> On 4/24/2024 10:59 PM, Limaye, Aniket wrote:
>>>> On 4/24/2024 6:53 PM, Ryan Eatmon wrote:
>>>>>
>>>>> We have consistently named the machines with -evm and not -sk.  Is 
>>>>> there a good reason to not name this am69-evm?
>>>>>
>>>> Well this is just following the naming convention that got accepted 
>>>> in u-boot:
>>>>
>>>> https://git.ti.com/cgit/ti-u-boot/ti-u-boot/tree/configs?h=ti-u-boot-2024.04 
>>>>
>>>>
>>>> Yeah other parts have been like "am62x_evm_*_defconfig" and such, 
>>>> but for am69 we have gone with "am69_sk_*_defconfig"
>>>>
>>>> Even the dts files for these, everywhere, have -sk suffixes.
>>>>
>>>> Hence the change here as well.
>>>
>>> Just because the configs in uboot have sk does not imply that the 
>>> machine name in meta-ti needs to have sk.  From my understanding, 
>>> generally the sk comes first and latest a full up evm.  But the 
>>> Yocto machine usually supports all of the variations and so they 
>>> were named with -evm to be the broadest possible name.
>>>
>> The yocto machine did used to support all of the variations up until 
>> now, but starting the next LTS stream, the j784s4-evm build will not 
>> work for the am69-sk. U-boot will only be packing one dtb in the 
>> u-boot img, which means we require a source code / build config 
>> change between j784s4-evm and am68-sk (and all future boards with 
>> different dtbs), to be able to boot the board.
>>
>> The above only explains the need for a new yocto machine for the 
>> SKs... As for the naming, everywhere on ti.com we have always used 
>> am69-*sk* as the product, including documentation. For example, [1] 
>> vs [2].
>> [1]: https://www.ti.com/tool/J784S4XEVM
>> [2]: https://www.ti.com/tool/SK-AM69
>>
>> I don't think we should deviate from ti.com naming, u-boot board 
>> names, and the current public documentation, just to have consistency 
>> in meta-ti?
>>
>> For future devices, we can keep using -evm as the default since those 
>> boards come first (not the corresponding SKs, not sure if it was a 
>> typo in your msg). And then add the -sk boards for the SOCs as 
>> applicable.
>>
>>> My personal preference would be to eliminate the -evm for all of the 
>>> machines and just go with the soc name.  But if we are going to have 
>>> it then I think they should be consistent.
>>
>> am69-sk and j784s4-evm will be including the same soc include 
>> (j784s4.inc), but they do need different dtbs, u-boot configs, 
>> GP/HS-FS defaults... That is to say we are configuring board specific 
>> settings in our builds for our boards. So I don't think we should 
>> move to just using the soc name.
>
> I'm still not happy about introducing the inconsistency, but I'll 
> accept this patch for now and work on a better longer term solution 
> for the machine naming that can be both high level (multi-board 
> support) and low level (single board support).
>
> I have some ideas on that, I just need time to put together an RFC so 
> that we can have a larger discussion.

Oh, that sounds good...
Will check it out when ready!

Thanks,
Aniket

>
>
>>>
>>>> Regards,
>>>> Aniket
>>>>
>>>>>
>>>>>
>>>>> On 4/24/2024 3:35 AM, Aniket Limaye wrote:
>>>>>> From: Beleswar Padhi <b-padhi@ti.com>
>>>>>>
>>>>>> As of commits [1] and [2] in ti-u-boot-2024.04 and upstream 
>>>>>> respectively
>>>>>> there is a separate defconfig to build u-boot for j784s4-evm and 
>>>>>> am69-sk
>>>>>>
>>>>>> Hence, introduce new yocto machine configs for am69-sk. Also 
>>>>>> remove am69
>>>>>> devicetree builds from the j784s4-evm conf
>>>>>>
>>>>>> [1]: 
>>>>>> https://git.ti.com/cgit/ti-u-boot/ti-u-boot/commit/?h=ti-u-boot-2024.04&id=8a46af7a39e6218e4de5c6195697af36201c778c
>>>>>> [2]: 
>>>>>> https://source.denx.de/u-boot/u-boot/-/commit/740759d558e632e1540d620baadbcc46fdf24db2
>>>>>>
>>>>>> Signed-off-by: Beleswar Padhi <b-padhi@ti.com>
>>>>>> Signed-off-by: Aniket Limaye <a-limaye@ti.com>
>>>>>> ---
>>>>>>   meta-ti-bsp/conf/machine/am69-sk-k3r5.conf | 11 +++++++++++
>>>>>>   meta-ti-bsp/conf/machine/am69-sk.conf      | 20 
>>>>>> ++++++++++++++++++++
>>>>>>   meta-ti-bsp/conf/machine/j784s4-evm.conf   |  1 -
>>>>>>   3 files changed, 31 insertions(+), 1 deletion(-)
>>>>>>   create mode 100644 meta-ti-bsp/conf/machine/am69-sk-k3r5.conf
>>>>>>   create mode 100644 meta-ti-bsp/conf/machine/am69-sk.conf
>>>>>>
>>>>>> diff --git a/meta-ti-bsp/conf/machine/am69-sk-k3r5.conf 
>>>>>> b/meta-ti-bsp/conf/machine/am69-sk-k3r5.conf
>>>>>> new file mode 100644
>>>>>> index 00000000..9d2b4cd6
>>>>>> --- /dev/null
>>>>>> +++ b/meta-ti-bsp/conf/machine/am69-sk-k3r5.conf
>>>>>> @@ -0,0 +1,11 @@
>>>>>> +#@TYPE: Machine
>>>>>> +#@NAME: AM69 SK (R5F)
>>>>>> +#@DESCRIPTION: Machine configuration for the TI AM69 SK (R5F core)
>>>>>> +
>>>>>> +require conf/machine/include/k3r5.inc
>>>>>> +
>>>>>> +SYSFW_SOC = "j784s4"
>>>>>> +SYSFW_CONFIG = "evm"
>>>>>> +SYSFW_SUFFIX = "hs-fs"
>>>>>> +
>>>>>> +UBOOT_MACHINE = "am69_sk_r5_defconfig"
>>>>>> diff --git a/meta-ti-bsp/conf/machine/am69-sk.conf 
>>>>>> b/meta-ti-bsp/conf/machine/am69-sk.conf
>>>>>> new file mode 100644
>>>>>> index 00000000..233a4618
>>>>>> --- /dev/null
>>>>>> +++ b/meta-ti-bsp/conf/machine/am69-sk.conf
>>>>>> @@ -0,0 +1,20 @@
>>>>>> +#@TYPE: Machine
>>>>>> +#@NAME: AM69 SK
>>>>>> +#@DESCRIPTION: Machine configuration for the TI AM69 SK
>>>>>> +
>>>>>> +require conf/machine/include/j784s4.inc
>>>>>> +
>>>>>> +TFA_K3_USART = "0x8"
>>>>>> +OPTEE_K3_USART = "0x8"
>>>>>> +
>>>>>> +KERNEL_DEVICETREE_PREFIX = " \
>>>>>> +    ti/k3-am69-sk \
>>>>>> +    ti/k3-fpdlink \
>>>>>> +    ti/k3-v3link \
>>>>>> +"
>>>>>> +
>>>>>> +KERNEL_DEVICETREE = " \
>>>>>> +    ti/k3-am69-sk.dtb \
>>>>>> +"
>>>>>> +
>>>>>> +UBOOT_MACHINE = "am69_sk_a72_defconfig"
>>>>>> diff --git a/meta-ti-bsp/conf/machine/j784s4-evm.conf 
>>>>>> b/meta-ti-bsp/conf/machine/j784s4-evm.conf
>>>>>> index e451bf47..6573e5c5 100644
>>>>>> --- a/meta-ti-bsp/conf/machine/j784s4-evm.conf
>>>>>> +++ b/meta-ti-bsp/conf/machine/j784s4-evm.conf
>>>>>> @@ -11,7 +11,6 @@ KERNEL_DEVICETREE_PREFIX = " \
>>>>>>       ti/k3-j784s4 \
>>>>>>       ti/k3-fpdlink \
>>>>>>       ti/k3-j7200-evm-mcspi-loopback \
>>>>>> -    ti/k3-am69-sk \
>>>>>>       ti/k3-v3link \
>>>>>>   "
>>>>>
>>>
>


      reply	other threads:[~2024-04-29 17:56 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-24  8:35 [meta-ti][master/scarthgap][PATCH] conf: machine: add AM69-SK machine configuration Aniket Limaye
2024-04-24 13:23 ` Ryan Eatmon
2024-04-25  3:59   ` Limaye, Aniket
2024-04-25 19:36     ` Ryan Eatmon
2024-04-26 19:01       ` Denys Dmytriyenko
2024-04-26 21:08         ` Ryan Eatmon
2024-04-29 13:51       ` Limaye, Aniket
2024-04-29 15:00         ` Ryan Eatmon
2024-04-29 17:56           ` Limaye, Aniket [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=9a083386-48a0-4604-87d5-721cd4d8dce5@ti.com \
    --to=a-limaye@ti.com \
    --cc=b-padhi@ti.com \
    --cc=meta-ti@lists.yoctoproject.org \
    --cc=reatmon@ti.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.