| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |
| Who | What | When |
|---|---|---|
| sailus | neg: Hejssan! | [09:09] |
| neg | sailus: hej | [09:10] |
| sailus | Hur mår du? | [09:10] |
| neg | bara bra, själv? | [09:12] |
| sailus | Jag mår bra också, tack!
Om lappserie "[PATCH 00/32] rcar-vin: Add Gen3 with media controller support" --- för vad behöver du de två första lappar? Jag trodde man borde ta hänsyn till de rutterna också när man promenerar grafen. neg: Kan du förstå mig? :-) | [09:13] |
| neg | yes, but maybe we should stick to english I'm guessing not everyone speeks swedish here :-)
Are you talking about the two patches from Laurent which adds the entity route functions? | [09:21] |
| sailus | Oh, yes, that's entirely possible.
Yes, I am. | [09:22] |
| neg | I need them to propperly interact with the ADV7482 driver where I need to know which sink is connected to what source internaly in that driver | [09:24] |
| sailus | Right.
Do you also intend to allow changing the routes? Or is it static for now? | [09:24] |
| neg | it's static for now in since there are other issues with the ADV7482 which needs to be sourted out, but in theory it could be made configuarable today | [09:25] |
| sailus | Do you use the V4L2 power count helpers in the bridge driver? | [09:26] |
| neg | the prototype driver have a static route from one of the AIN to the 1-lane CSI2 transmitter for example but that could be made configurable | [09:26] |
| sailus | They depend on walking the graph, and the walk starts from sub-devices. So with that, it has to start from a specific pad.
I luckily have patches for that. :-) | [09:26] |
| neg | I use the v4l2_pipeline_pm_use() functions if that was what you where asking? | [09:28] |
| sailus | Yes, I was asking about that.
So if the routing configuration changes, that has to be taken into account. If you don't change it, that's all right I guess. | [09:29] |
| neg | ahh yes, I have not considerd that since I only have static routes inside the ADV7482
but why would that change? Dose v4l2_pipeline_pm_use() not operate on a entity level, so what ever rotuing there is inside the ADV7482 would not matter for the power counts? | [09:31] |
| sailus | It operates on graph level.
E.g. if you have an open file handle to a video device with links set up towards the adv7482, the adv7482 will be powered up as well. | [09:33] |
| neg | yes and if the internal routing inside the adv7482 changes from say AIN1 -> 1 Lane CSI2 to AIN4 -> 1 Lane CSI2 no change in the power is needed, right? | [09:37] |
| sailus | If you have links set up from sub-devices to AIN1 or AIN4, then yes.
Because those sub-devices will be powered up based on the routing configuration then. | [09:39] |
| neg | ahh ofc | [09:39] |
| sailus | But if the routing configuration is static, then this isn't strictly required yet. | [09:41] |
| larsc | lappserie fraan lappland? | [09:41] |
| sailus | :-D | [09:41] |
| neg | but for that case I have limit set by the hardware where no links can be changed if any other link is active, and I have solved that limitation by not allowing new links if any entity->use_count > 0 in the whole MC graph, so there can't be any opend file handles at that time so no device are powerd when link changes happen | [09:41] |
| sailus | Lapland. Lappland is different. :-)
And there's no S_ROUTING yet either. I think it'd be good to add a comment that the configuration must be static. | [09:42] |
| neg | Yes, I will add a comment before I send out the next version, thanks for your feedback | [09:48] |
| ..... (idle for 23mn) | ||
| sailus | neg: I sent you another patch you might like. | [10:11] |
| ........................ (idle for 1h57mn) | ||
| neg: Ping? | [12:08] | |
| .... (idle for 19mn) | ||
| larsc | sailus: I hope you have manual heating control: http://metropolitan.fi/entry/ddos-attack-halts-heating-in-finland-amidst-winter | [12:27] |
| neg | sailus: pong
Thanks for the patch, it looks interesting. I will see if it can make things easier for me before I send out the next version | [12:31] |
| ................................................................................ (idle for 6h37mn) | ||
| *** | awalls has left | [19:09] |
| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |