| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |
| Who | What | When |
|---|---|---|
| *** | GrayShade has left | [11:49] |
| .............. (idle for 1h6mn) | ||
| hverkuil | pinchartl: ping | [12:55] |
| ...................... (idle for 1h45mn) | ||
| pinchartl | hverkuil: pong | [14:40] |
| hverkuil | I've resumed work on cec for omap4 and omap5. I believe I mentioned it before, but with the current mainline kernel if you boot without a display connected, and then connect the display after it has booted, then it won't enable the HDMI output.
Is any work being done to fix that? Tested with pandaboard and an omap5 board. | [14:42] |
| pinchartl | I can add it to my todo list, but won't have time to look very soon
tomba: have you tested that recently ? | [14:42] |
| hverkuil | pinchartl: I plan on posting my cec patches for omap4 and omap5 next week, most likely. The only thing I was unable to test is what happens if a display is connect after the device finished booting, but that's due to this bug.
If a display was connected at boot time, then after that you can safely disconnect and reconnect and it will work fine. | [14:50] |
| pinchartl | I'll try to give it a try
sorry for the delay I'm fighting with different SoCs right now :-) | [14:52] |
| tomba | pinchartl: hmm no I haven't tested that. | [14:53] |
| hverkuil | No problem, I've been sitting on this series myself for waaaay too long :-) | [14:53] |
| tomba | hverkuil: what do you mean with "won't enable HDMI output". That the fbdev doesn't appear on the screen? or that the HDMI is not usable at all, even if you start a KMS application?
hverkuil: did you find omap5 hdmi spec somewhere? or did you reverse engineer? | [14:54] |
| hverkuil | fbdev doesn't appear on the screen, but I'm fairly certain the HDMI is completely unusable since hdmi_power_on_full() is never called.
tomba: you mean HDMI CEC? Tomi helped me with that a few years back. I don't believe I ever had the spec for omap5. | [14:57] |
| tomba | hverkuil: yes, it's Tomi here =). but I couldn't give you omap5 hdmi documentation, so I was wondering if you somehow managed to get it working. | [14:58] |
| hverkuil | 'hverkuil': that's a proper nick. At least you *know* who you are talking to :-)
I had CEC working for the most part for a long time now, but I had problems with an interrupt crash. And then I had to wait for pinchartl's 'bridge' work to be merged, and then I was too busy with other things, etc. etc. | [14:59] |
| tomba | HDMI won't be turned on unless something turns it on. It could be a KMS app, or it could be fbdev. I'm not quite sure when fbdev decides to turn on a display, might not happen on hotplug.
Oh, right, I'm via Matrix, so you can't even see my real name in the details. But I'm not aware of any change there. So if it used to work before, it should work now too. | [15:00] |
| hverkuil | It does enter fbdev but it fails on something. I'll try to do more testing today or next week. | [15:02] |
| tomba | Ok. I'm no longer working for TI, so I haven't been actively looking at the omapdrm for a while now. | [15:03] |
| hverkuil | Other SoCs have no problem with this. It might be something that appeared with the bridge work. I am fairly certain it used to work on the pandaboard before that series was merged. | [15:03] |
| tomba | actually I can test this out right now. I'm working on capture side, but on a TI board.
well, I don't see any error, but as you noticed, the HDMI output is not enabled automatically. Running a KMS app works fine, though, and after that the fbdev is also visible. | [15:04] |
| hverkuil | which kms app do you use? | [15:07] |
| tomba | my kmstest from kms++ library
I think this is how it's "always" been. But if other platforms work differently, then probably omapdrm should be changed. Hmm, no... Or did it work like this... Heh. Memory fails. I don't usually ever use fbdev, so I wouldn't notice this | [15:07] |
| Looks like just opening the card0 and querying info will turn on the fbdev. I don't see any reason why the hotplug itself shouldn't turn the display on. | [15:15] | |
| .................................... (idle for 2h56mn) | ||
| sailus | There's some serious loss of messages that traverse vger.kernel.org again. :-( | [18:11] |
| ................... (idle for 1h34mn) | ||
| jm_h | sailus, yes, does it mean we need to resend later or they will appear one day :-p | [19:45] |
| ......... (idle for 43mn) | ||
| sailus | It's been around for some time. Also Patchwork is getting the same e-mails I'm getting through the list, so it seems to be lost early.
Sometimes the rest of the e-mails have appeared later, at other times not. | [20:28] |
| Actually, now that I think about it, the e-mail that has been lost was sent through the company's mail servers.
The rest was just delayed. Hmm. So perhaps it may not be a vger.kernel.org issue after all. But without logs on either side it's hard to tell for sure. | [20:34] | |
| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |