| *** tpb has joined #timvideos | 00:00 | |
| shenki | mithro: pong | 00:59 |
|---|---|---|
| mithro | See pull requests | 01:03 |
| mithro | shenki: we should also get working on that SFC application ASAP so we don't miss the next round of acceptances | 01:22 |
| mithro | I believe we have till the 10th of June | 01:22 |
| shenki | mithro: okay | 01:23 |
| mithro | xfxf: you could probably help with that too if you have time | 01:23 |
| *** mithro has quit IRC | 01:36 | |
| *** cfelton has quit IRC | 01:36 | |
| *** kierank has quit IRC | 01:36 | |
| *** CARAM__ has quit IRC | 01:36 | |
| *** kierank has joined #timvideos | 01:40 | |
| *** mithro has joined #timvideos | 01:40 | |
| *** ChanServ sets mode: +o mithro | 01:40 | |
| *** cfelton has joined #timvideos | 01:40 | |
| *** CARAM__ has joined #timvideos | 01:44 | |
| *** puck has joined #timvideos | 02:11 | |
| puck | Hey, I see that there now prebuilt HDMI2USB images from the end of March, but a few bug fixes are in the main HDMI2USB git repo, any plans to built another set of prebuilt images anytime soon? | 02:15 |
| mithro | puck: I don't see any major bug fixes since the last build? | 02:42 |
| puck | mithro, Just looks like forcing the input to be 1024x768 | 02:46 |
| mithro | puck: yes, but I think there are more bugs related to that which need to be fixed | 03:00 |
| mithro | ~seen tija | 03:00 |
| tpb | mithro: tija was last seen in #timvideos 6 days, 9 hours, 31 minutes, and 23 seconds ago: <tija> gitk | 03:00 |
| puck | Ah, fair enough. | 03:01 |
| puck | As it is, I'm failng to get the hdmi2usb binary loaded onto my board. Stillt trying to work that out. | 03:01 |
| mithro | puck: How are you loading it? | 03:01 |
| puck | I'm trying fx2loader from makestuff | 03:02 |
| puck | No output is appearing, but running it with strace I get: | 03:02 |
| puck | recvmsg(3, 0x7fffc1a34bb0, 0) = -1 EAGAIN (Resource temporarily unavailable) | 03:03 |
| puck | write(2, "libFPGAlink/makestuff/apps/fx2lo"..., 94libFPGAlink/makestuff/apps/fx2loader/lin.x64/rel/fx2loader: usbOpenDevice(): device not found | 03:03 |
| mithro | Have you got a more complete log of what you are running? | 03:04 |
| puck | yes | 03:05 |
| mithro | https://github.com/timvideos/HDMI2USB/blob/master/tools/flash.sh | 03:05 |
| tpb | Title: HDMI2USB/flash.sh at master · timvideos/HDMI2USB · GitHub (at github.com) | 03:05 |
| puck | That's effectively the process I went through, but I followed: https://github.com/timvideos/HDMI2USB/wiki/libFPGALink | 03:06 |
| tpb | Title: libFPGALink · timvideos/HDMI2USB Wiki · GitHub (at github.com) | 03:06 |
| mithro | Yes, they look correct too | 03:08 |
| mithro | We really need to fix up some of the documentation | 03:08 |
| CarlFK | puck: sec.. I think I know whats wrong... | 03:12 |
| puck | Cool | 03:13 |
| CarlFK | puck: never mind. I wrote that script and at first was missing a cd, but it is fixed, so that isn't it. | 03:22 |
| puck | I didn't actually use that script, but I used the commands from the wiki | 03:27 |
| CarlFK | https://github.com/timvideos/HDMI2USB/wiki/libFPGALink thats what I wrote | 03:31 |
| tpb | Title: libFPGALink · timvideos/HDMI2USB Wiki · GitHub (at github.com) | 03:31 |
| puck | Yeah, that's the process I was following - for "First Time" | 03:31 |
| puck | Sorry, on the phone now | 03:32 |
| *** tija has joined #timvideos | 04:44 | |
| *** hyades has joined #timvideos | 04:44 | |
| mithro | hey tija | 04:55 |
| tija | mithro: Hi | 04:55 |
| tija | FYI the resolution is not changing using USB command. | 04:55 |
| tija | Today I am planning to pull the Jahanzeb pre built repo and see did I break anything. Or it never worked. | 04:56 |
| mithro | tija: can you explain how you are testing at the moment? | 04:58 |
| mithro | tija: then I can try and replicate the issue here too | 04:58 |
| tija | mithro: put sw(1) = 0; open gtkterm and connect to cdcport of atlys; send command 48 30. | 05:08 |
| mithro | tija: and then.... | 05:09 |
| tija | mithro: then send 48 53 to get the HDMI status | 05:14 |
| tija | or use the fpga_debug tool to see the resolution. I hope it works correctly | 05:18 |
| mithro | tija: but you are not checking what the computer connected to board is seeing? | 05:19 |
| tija | mithro: I did not understand ^ | 05:20 |
| mithro | tija: ? - you mean you don't understand what I'm saying? | 05:20 |
| tija | yes | 05:20 |
| mithro | So, the ultimate behavioural change we are trying to get working is that a computer connected to the Atlys's HDMI port sees either a HDMI 720p EDID or a 1024x768 EDID | 05:22 |
| tija | okay | 05:23 |
| tija | Well if the computer connected to the board sees a 720p EDID it will change it's resolution and hence output 720p eventually? | 05:25 |
| mithro | tija: yes | 05:25 |
| tija | hence checking the resolution should be sufficient to check whether things are working. | 05:27 |
| mithro | tija: On the Atlys boards we can't force a computer to renegotiate the EDID values because we don't control the HPD lines | 06:19 |
| mithro | tija: so after a change you might need to unplug and replug in the computer before the effect "sticks" | 06:19 |
| tija | mithro: hmm this is something I didn't do. Damn! You are too smart! | 06:20 |
| mithro | tija: or I have banged my head against the problem for hours before and remeber the pain well :P | 06:24 |
| tija | mithro: what you told was right. The switch is working when I plug and unplug the HDMI cable. | 07:04 |
| tija | but 720p is not streaming. Errors! | 07:05 |
| mithro | tija: what if you unplug the usb and replug it? | 07:29 |
| mithro | tija: the new firmware that Joel and rohit worked on might not be correctly resetting after the descriptor information after the video resolution change | 07:30 |
| tija | mithro: I haven't tried it. But when I switch it back to 1280x768 then it works | 07:31 |
| tija | I am using the old firmware | 07:31 |
| mithro | tija: do you know about v4l-info? | 07:32 |
| tija | mithro: No | 07:33 |
| mithro | tija: try installing it and see what it says about the HDMI2USB device | 07:34 |
| tija | mithro: okay | 07:34 |
| mithro | tija: I would expect the output to change saying the video is now streaming 720p rather than 1024x768 resolution | 07:35 |
| mithro | The `v4l2-ctl --all` command might be another option | 07:35 |
| tija | mithro: OKay I will try it tonight. | 07:37 |
| mithro | tija: great! | 07:40 |
| *** _florent_ has joined #timvideos | 07:55 | |
| tija | mithro: I got some free time so I did it right now. | 08:01 |
| tija | mithro: http://paste.ubuntu.com/11515544/ | 08:05 |
| tpb | Title: Ubuntu Pastebin (at paste.ubuntu.com) | 08:05 |
| tija | things don't change when I switch to 720p | 08:07 |
| tija | There is no 720p test pattern it is always 1280x768 | 08:14 |
| tija | So the test pattern works | 08:14 |
| tija | but when I switch to HDMI0, video streaming stops. | 08:15 |
| tija | http://paste.ubuntu.com/11515692/ | 08:16 |
| tpb | Title: Ubuntu Pastebin (at paste.ubuntu.com) | 08:16 |
| *** _florent_ has quit IRC | 08:47 | |
| *** _florent_ has joined #timvideos | 09:14 | |
| *** xfxf has quit IRC | 10:53 | |
| mithro | tija: well the test pattern should switch to generating at 720p | 12:29 |
| tija | mithro: that's not how it is written in HDL | 12:29 |
| tija | mithro: it will output 1280x768 irrespective of the what we. | 12:31 |
| tija | https://github.com/timvideos/HDMI2USB/blob/master/hdl/misc/pattern.vhd | 12:31 |
| tpb | Title: HDMI2USB/pattern.vhd at master · timvideos/HDMI2USB · GitHub (at github.com) | 12:31 |
| *** Niharika has joined #timvideos | 12:42 | |
| mithro | Which is why I said should :-) | 13:32 |
| tija | mithro: I thought you were claiming it should switch. | 13:38 |
| mithro | It should switch, but it does look like the vhd doesn't at the moment | 13:39 |
| mithro | making software is hard :P | 13:41 |
| tija | mithro: So we have another bug. HDMI2USB does not work at 720p. | 13:43 |
| mithro | It appears so | 13:43 |
| tija | I will pull Jahabzeb's prebuilt and test it. Actually we should have a checklist of this to check on every commit, since Jahanzeb's era (if I can say so) | 13:45 |
| tija | What things worked earlier and now are broken. What new feature got added which should be checked in the future commits etc. | 13:46 |
| mithro | tija: sure, but non-automated testing is always a mixed bag | 13:47 |
| tija | mithro: It can be automated to some level by sending commands and receiving status via the cdc port | 13:49 |
| mithro | tija: if you can figure out a way to automate the testing, that is very useful | 13:49 |
| mithro | tija: but I think that is a big project that requires a lot more time then you have? | 13:49 |
| *** _florent_ has quit IRC | 13:49 | |
| *** _florent_ has joined #timvideos | 13:50 | |
| tija | mithro: Shenki in his LCA presentation: If you have a big project, give it to an intern. | 13:50 |
| mithro | Ha, your no longer an intern, right? :P | 13:51 |
| mithro | tija: think the best bet is to try and get 720p working again | 13:51 |
| tija | mithro: I am not but few more will come next GSoC :) | 13:52 |
| mithro | tija: yeah - it's in the issue tracker as a GSoC project :) | 13:52 |
| CarlFK | can we roll back to when 720p worked? | 13:52 |
| tija | mithro: How about shifting our focus to zynq? Try to minimize HDLs. | 13:54 |
| mithro | CarlFK: we don't know when 720p worked | 13:54 |
| CarlFK | mithro: k. never mind that. | 13:54 |
| mithro | CarlFK: otherwise it would be really simple | 13:55 |
| mithro | CarlFK: I was pretty sure it did work at some point, but it was long ago and we haven't been testing with it frequently since | 13:56 |
| mithro | actually, it might have been even before we had a test pattern | 13:59 |
| mithro | tija: How much time do you have? | 14:01 |
| CarlFK | would it help to remove support for 4:3 ? | 14:01 |
| CarlFK | I can see the value, but I can live without it it it gets 16:9 functioning sooner | 14:02 |
| tija | mithro: Honestly not much. I have started working. Currently I am working from home. Weekends will be free. | 14:02 |
| mithro | tija: yeah, so lets work on small things which you have a chance of finishing? | 14:04 |
| tija | mithro: like?? | 14:04 |
| mithro | tija: getting 720p working on the current system | 14:04 |
| tija | mithro: Yeah we could do that. | 14:05 |
| mithro | tija: small incremental fixes to the current system seem good | 14:05 |
| mithro | tija: I think there are a bunch of low hanging fruit on the issue tracker | 14:06 |
| tija | mithro: But if you ask my I would prefer porting HDMI capture to ZYBO. Even though it does not appear reasonable. | 14:09 |
| tija | mithro: If you ask me why? It is because making something new seems more exciting. | 14:10 |
| tija | but yeah that system too will eventually reach this stage. | 14:11 |
| mithro | tija: If you want to try newer more interesting stuff, it might be worth seeing if you can continue my work on the misoc/migen stuff | 14:12 |
| mithro | tija: I've been trying to get misoc/migen working on the Atlys board to test out their implementation of a DVI receiver / source | 14:13 |
| mithro | tija: because it should be portable to many more architectures then just the spartan 6 | 14:13 |
| mithro | tija: It looks like they have some support related to the Zybo already too | 14:14 |
| mithro | tija: they also have a really nice ethernet core that _florent_ wrote | 14:15 |
| tija | mithro: hmm this seems interesting. | 14:16 |
| mithro | tija: I currently think that any rewrite of HDMI2USB firmware would probably end up using that | 14:16 |
| tija | mithro: So basically a JPEG encoder and we have another firmware. | 14:17 |
| mithro | tija: happy to chat another night about it | 14:17 |
| mithro | tija: yeah | 14:17 |
| mithro | tija: which is *fully* open source | 14:17 |
| mithro | tija: not using "generated cores" like our current one does | 14:17 |
| mithro | but anyway, I should go to bed now | 14:18 |
| tija | Good night! | 14:18 |
| mithro | tija: I'm happy to help you get misoc/migen set up tomorrow or on the weekend if you think its something your excited about | 14:18 |
| * cfelton this is unfortunate | 14:19 | |
| mithro | cfelton: why? | 14:19 |
| mithro | cfelton: you think we should do myhdl? :P | 14:19 |
| cfelton | not necessarily | 14:21 |
| cfelton | but I know what you should not do it in | 14:21 |
| CarlFK | cfelton: what does this refer to? | 14:21 |
| mithro | cfelton: vhdl? :P | 14:22 |
| cfelton | mithro: ABEL | 14:25 |
| mithro | LGTM -> http://en.wikipedia.org/wiki/Advanced_Boolean_Expression_Language#/media/File:ABEL_HDL_example_SN74162.png | 14:27 |
| tpb | Title: Advanced Boolean Expression Language - Wikipedia, the free encyclopedia (at en.wikipedia.org) | 14:27 |
| mithro | anyway, really going now | 14:27 |
| *** cfelton has left #timvideos | 14:28 | |
| tija | mithro: Jahanzeb's prebuilt does not have working 720p | 14:44 |
| *** Niharika has quit IRC | 15:37 | |
| *** Niharika has joined #timvideos | 16:09 | |
| *** tija has quit IRC | 16:54 | |
| *** CarlFK has quit IRC | 19:39 | |
| *** Niharika has quit IRC | 19:58 | |
| *** CarlFK has joined #timvideos | 20:38 | |
| *** ChanServ sets mode: +v CarlFK | 20:38 | |
| *** mithro has quit IRC | 22:09 | |
| *** kierank has quit IRC | 22:27 | |
| *** _florent_ has quit IRC | 22:43 | |
| *** mithro has joined #timvideos | 22:45 | |
| *** ChanServ sets mode: +o mithro | 22:45 | |
| *** hyades has quit IRC | 22:48 | |
| *** kierank has joined #timvideos | 23:07 | |
| *** CarlFK has quit IRC | 23:23 | |
| *** CarlFK has joined #timvideos | 23:50 | |
| *** ChanServ sets mode: +v CarlFK | 23:50 | |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!
