On 04-08-2026 09:08 pm, Krzysztof Kozlowski wrote:
On 04/08/2026 17:18, Vishnu Santhosh wrote:
On 30-07-2026 12:53 pm, Krzysztof Kozlowski wrote:
On Wed, Jul 29, 2026 at 02:21:01PM +0530, Vishnu Santhosh wrote:
The BAM-DMUX node lives as a direct child of the modem PAS remoteproc
No, it does not. Look:
$ git describe
next-20260729
$ git grep bam-dmux
No child node shown
The commit message overstated this. The dt changes are yet to be merged.
I'll correct it in v3.
Why isn't this part of Shikra remoteproc commit/patch? How any code
could pass validation otherwise?
By the time the BAM-DMUX support was brought up and validated, the
remoteproc changes was already in a stable state. Therefore, the
No, it was not in stable state if it was not merged.
bam-dmux change is being submitted separately.
No, bindings should be complete.
Understood on both. I'll correct the commit message to describe this
as enabling the bam-dmux subnode in the binding, rather than
asserting the node already exists in a DT or that this fixes a
failure that doesn't exist today.
node, but qcom,shikra-pas.yaml has unevaluatedProperties: false and
does not list bam-dmux, so any DT adding this subnode fails schema
validation.
Fails how? How can I reproduce the error?
There is no bam-dma in this binding example.
Best regards,
Krzysztof
Did you mean bam-dmux (the property this patch adds)? If so, I'll add
an example with a bam-dmux subnode in v3.
You claim something fails. Prove it.
Best regards,
Krzysztof
I will fix the commit message, removing the false claim, and will fold
the bam-dmux example node into the yaml in the next revision so
dt_binding_check can be used to verify this patch.
Thanks,
Vishnu
