| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |
| Who | What | When |
|---|---|---|
| sailus | kbingham: Pung. :-) | [07:42] |
| ......................................................................... (idle for 6h2mn) | ||
| kbingham | sailus: pang :) | [13:44] |
| narmstrong | hverkuil: I'm not sure the Amlogic CEC support snooping. Do you mean if I do not set any logical address it will receive all messages ? | [13:48] |
| hverkuil | narmstrong: normally a CEC adapter will receive broadcast messages and directed messages, i.e. messages where the destination matches the logical address(es) of the adapter. | [13:50] |
| narmstrong | hverkuil: ok, I don't know then | [13:51] |
| hverkuil | In monitoring mode it will also receive messages directed to other CEC adapters. So yes, you'll see all traffic.
It's usually called monitoring or snooping. | [13:51] |
| narmstrong | ok | [13:54] |
| hverkuil | narmstrong: or 'listen', I've seen that as well.
Easiest to test is by connecting the amlogic to a TV and a DVD/Bluray/other amlogic to the same TV on another HDMI input. Put one in monitoring mode and you should see the traffic between TV and the other device. Or connect two amlogics together (yes, HDMI-out to HDMI-out works fine for CEC!). Also a good test to see if you can use CEC without a hotplug signal. Let me know if you want to do that because testing this is a little bit tricky. | [13:54] |
| sailus | kbingham: How do you do? | [13:59] |
| kbingham | sailus: I'm good - I hope your headache is better now! | [14:00] |
| narmstrong | hverkuil: ok
hverkuil: for dw-hdmi, looking at the databook, it does not support monitoring, but you can enable all addresses and explicitly NACK all incoming messages, could it work ? | [14:01] |
| kbingham | sailus: Anything I can help you with (from your earliest ping)?
sailus: I have a 12 hour daytime flight back to London on Sunday - so I will probably be doing some offline work / reviews. | [14:02] |
| sailus | About the fwnode patches --- did you have a chance to test them?
This set in particular: http://www.spinics.net/lists/devicetree/msg179120.html | [14:06] |
| hverkuil | narmstrong: whether a message is acked or not is entirely dependent on the logical addresses. I.e. the hw determines what to do based on the logical addresses. So enabling all addresses will cause the HW to always ack all addresses, but that's not what monitoring is. | [14:13] |
| narmstrong | hverkuil: yes but there is a setup where all messages are NACKed
it's their "standby" feature "1: CEC controller responds NACK to all messages and generates wakeup status for opcode. It only responds NACK when the EOM is received. It means only the last block of a frame would be responded NACK. The follower acknowledges the message when there is only one head block pointed to the follower, if the follower is in the standby mode" | [14:13] |
| sailus | kbingham: Ping? | [14:17] |
| kbingham | sailus: In essence yes - my current work is based on those patches - but I haven't tested the v5 or v6 explicitly.
sailus: Unfortunately here in Japan I've had hardware that hasn't been working | [14:17] |
| sailus | Ok. Let me know when you're able to test them. The last two patches are essentially untested. | [14:19] |
| kbingham | sailus: Next week I will be back in the UK on earlier hardware revision which should put me back to a known working state - and then I can rebase on to your v6, and submit a tested-by: | [14:19] |
| sailus | That'd be wonderful! | [14:19] |
| hverkuil | narmstrong: non-standard behavior, but I've seen such things before. I don't quite understand what they mean with that last sentence ("The follower..."). | [14:19] |
| sailus | I asked Rob if I could have an immutable branch for the OF patches these depend on in order to get them in during the same merge window. | [14:20] |
| kbingham | sailus: Ah excellent - that will help my patch progression as well! Are there functional differences between v5/6 and the patches I'm on top of ? (sailus/acpi-graph-cleaned from 2017-05-23?)
Essentially the endpoint matching is working - the issues I am having are unrelated. sailus: On monday when I'm back to known good hardware I can do some explicit tests on your patches though to make sure they are in active code paths. sailus: Is "v4l: async: Match parent devices" in any of your series currently? | [14:21] |
| sailus | Between v5 and v6 there's just line wrapping and Mika's reviewed-by to the set. | [14:24] |
| kbingham | looks like v4 was the last posted version. | [14:24] |
| sailus | v6 is. | [14:24] |
| ......... (idle for 41mn) | ||
| *** | kbingham has left "Ex-Chat" | [15:05] |
| ............ (idle for 58mn) | ||
| benjiG has left | [16:03] | |
| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |