All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Etheridge, Darren" <detheridge@ti.com>
To: Denys Dmytriyenko <denis@denix.org>
Cc: <nm@ti.com>, <reatmon@ti.com>, Yogesh Siraswar <yogeshs@ti.com>,
	Praneeth Bajjuri <praneeth@ti.com>,
	Denys Dmytriyenko <denys@konsulko.com>,
	<meta-arago@lists.yoctoproject.org>
Subject: Re: [meta-arago][dunfell][PATCH 2/3] packagegroup-*-graphics: add virtual provider for AM62x GPU driver v2
Date: Tue, 22 Mar 2022 17:44:39 -0500	[thread overview]
Message-ID: <1a1e58fe-70a3-bc92-9b33-75d02c92a283@ti.com> (raw)
In-Reply-To: <20220322222702.GV23554@denix.org>



On 3/22/2022 5:27 PM, Denys Dmytriyenko wrote:
> On Tue, Mar 22, 2022 at 05:16:20PM -0500, Etheridge, Darren wrote:
>>
>>
>> On 3/22/2022 5:13 PM, Denys Dmytriyenko wrote:
>>> On Tue, Mar 22, 2022 at 03:58:04PM -0500, Etheridge, Darren via lists.yoctoproject.org wrote:
>>>>
>>>>
>>>> On 3/22/2022 3:49 PM, Nishanth Menon via lists.yoctoproject.org wrote:
>>>>> On 14:35-20220322, Ryan Eatmon via lists.yoctoproject.org wrote:
>>>>>> From: Krunal Bhargav <k-bhargav@ti.com>
>>>>>>
>>>>>> Signed-off-by: Krunal Bhargav <k-bhargav@ti.com>
>>>>>> Signed-off-by: Ryan Eatmon <reatmon@ti.com>
>>>>>> ---
>>>>>>   .../conf/distro/include/arago-prefs.inc              | 12 ++++++------
>>>>>>   1 file changed, 6 insertions(+), 6 deletions(-)
>>>>>>
>>>>>> diff --git a/meta-arago-distro/conf/distro/include/arago-prefs.inc b/meta-arago-distro/conf/distro/include/arago-prefs.inc
>>>>>> index a356c76c..7bc2ba52 100644
>>>>>> --- a/meta-arago-distro/conf/distro/include/arago-prefs.inc
>>>>>> +++ b/meta-arago-distro/conf/distro/include/arago-prefs.inc
>>>>>> @@ -14,12 +14,12 @@ PREFERRED_PROVIDER_virtual/libgles2 = "ti-sgx-ddk-um"
>>>>>>   PREFERRED_PROVIDER_virtual/libgbm = "ti-sgx-ddk-um"
>>>>>>   PREFERRED_PROVIDER_virtual/gpudriver = "ti-sgx-ddk-km"
>>>>>> -# J7 has a Rogue GPU core
>>>>>> -PREFERRED_PROVIDER_virtual/egl_j7 = "ti-img-rogue-umlibs"
>>>>>> -PREFERRED_PROVIDER_virtual/libgles1_j7 = "ti-img-rogue-umlibs"
>>>>>> -PREFERRED_PROVIDER_virtual/libgles2_j7 = "ti-img-rogue-umlibs"
>>>>>> -PREFERRED_PROVIDER_virtual/libgbm_j7 = "ti-img-rogue-umlibs"
>>>>>> -PREFERRED_PROVIDER_virtual/gpudriver_j7 = "ti-img-rogue-driver"
>>>>>> +# K3 TI Platforms has a Rogue GPU core
>>>>>> +PREFERRED_PROVIDER_virtual/egl = "ti-img-rogue-umlibs"
>>>>>> +PREFERRED_PROVIDER_virtual/libgles1 = "ti-img-rogue-umlibs"
>>>>>> +PREFERRED_PROVIDER_virtual/libgles2 = "ti-img-rogue-umlibs"
>>>>>> +PREFERRED_PROVIDER_virtual/libgbm = "ti-img-rogue-umlibs"
>>>>>> +PREFERRED_PROVIDER_virtual/gpudriver = "ti-img-rogue-driver"
>>>>>
>>>>> How does this apply to k3 and leave non-k3 alone? Also, am65x for example is
>>>>> sgx.
>>>>>
>>>>
>>>> That is a very good point.  This will mess up SGX platforms.   I
>>>> wish we had k3_ as the prefix for the name here for all the k3
>>>> platforms.
>>>
>>> There is "k3" defined and you can do _k3 override to address all K3 platforms,
>>> but again, some are SGX and some are RGX...
>>>
>>>
>>>> So I guess the only solution is leave all the _j7 as they were
>>>>
>>>> and add:
>>>> +PREFERRED_PROVIDER_virtual/egl_am62 = "ti-img-rogue-umlibs"
>>>> +PREFERRED_PROVIDER_virtual/libgles1_am62 = "ti-img-rogue-umlibs"
>>>> +PREFERRED_PROVIDER_virtual/libgles2_am62 = "ti-img-rogue-umlibs"
>>>> +PREFERRED_PROVIDER_virtual/libgbm_am62 = "ti-img-rogue-umlibs"
>>>> +PREFERRED_PROVIDER_virtual/gpudriver_am62 = "ti-img-rogue-driver"
>>>
>>> And there's no "am62" defined, but "am62xx" :)
>>>
>>
>> Isn't it a partial pattern match?   If it is not the _j7 won't work
>> for j721s2 either.
> 
> No, it's not a partial pattern match. Overrides use exact matches only, so
> those have to be defined. Specifically these overrides come from SOC_FAMILY
> variable and there are definitions for "k3", "j7", but there's no "j721s2"
> SOC_FAMILY defined, so you cannot use _j721s2, nor _j7 will match it.
> https://git.yoctoproject.org/meta-ti/tree/conf/machine/include?h=dunfell
> 

Thanks for the explanation Denys,  I was making totally wrong assumption 
on how that was being derived.

So we could actually use _k3 and get am62xx-evm, j721s2-evm and j7-evm 
with a single override.  All k3's use the IMG Rogue driver if they have 
a GPU.



  reply	other threads:[~2022-03-22 22:44 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-03-22 19:35 [meta-arago][dunfell][PATCH 0/3] Adding support for am62xx Ryan Eatmon
2022-03-22 19:35 ` [meta-arago][dunfell][PATCH 1/3] ti-tisdk-makefile: adding uboot r5 config variable " Ryan Eatmon
2022-03-22 20:48   ` Nishanth Menon
2022-03-22 19:35 ` [meta-arago][dunfell][PATCH 2/3] packagegroup-*-graphics: add virtual provider for AM62x GPU driver v2 Ryan Eatmon
2022-03-22 20:49   ` Nishanth Menon
2022-03-22 20:58     ` Etheridge, Darren
2022-03-22 22:13       ` Denys Dmytriyenko
2022-03-22 22:16         ` Etheridge, Darren
2022-03-22 22:27           ` Denys Dmytriyenko
2022-03-22 22:44             ` Etheridge, Darren [this message]
2022-03-22 23:06               ` Denys Dmytriyenko
2022-03-22 19:35 ` [meta-arago][dunfell][PATCH 3/3] busybox: Enable rtcwake Ryan Eatmon
2022-03-22 20:51   ` Nishanth Menon
2022-03-23 19:10     ` Ryan Eatmon
2022-03-24 13:14       ` Nishanth Menon

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=1a1e58fe-70a3-bc92-9b33-75d02c92a283@ti.com \
    --to=detheridge@ti.com \
    --cc=denis@denix.org \
    --cc=denys@konsulko.com \
    --cc=meta-arago@lists.yoctoproject.org \
    --cc=nm@ti.com \
    --cc=praneeth@ti.com \
    --cc=reatmon@ti.com \
    --cc=yogeshs@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.