Tuesday, 2016-07-26

*** tpb has joined #timvideos00:00
mithrocr1901_modern: You should set up remote access (either serial or network) to your board.00:02
mithrocr1901_modern: That way you can look at the kernel output even when X is borked00:02
mithrocr1901_modern: also, turn on the extra debugging stuff I linked you to00:03
cr1901_modernmithro: Already have a serial connection. I need to figure out why X is segfaulting. Will do re: debugging stuff00:13
cr1901_modernmithro: What I know so far. It's segfaulting here: https://github.com/robclark/xserver/blob/master/hw/xfree86/common/xf86Cursor.c#L108 b/c pScr->modes is in fact a NULL pointer. Whether that's due to my board not liking HDMI2USB or some other interaction-of-components, I don't know yet.00:17
tpbTitle: xserver/xf86Cursor.c at master · robclark/xserver · GitHub (at github.com)00:17
cr1901_modernI'm (perhaps naively) assuming that if there's not a NULL check here, the code's assuming that pScr->modes can't ever be NULL, sans h/w problems or radiation :).00:18
mithrocr1901_modern: I'm guessing it is having problems reading or parsing the EDID data00:31
cr1901_modernmithro: Interestingly enough, Xorg.0.log claims to be able to parse the data... at the very least, I can read the string "HDMI2USB" and the display mode 800x600 is detected.00:34
cr1901_modernmithro: EDIDs don't match, so that's prob a hint00:41
tpbTitle: edid_fail.txt · GitHub (at gist.github.com)00:43
*** sb0 has joined #timvideos00:48
mithrocr1901_modern: check dmesg00:49
cr1901_modernmithro: Need to find that debug link to make messages more verbose00:49
cr1901_moderndmesg hasn't put out anything since 23 seconds after boot00:50
*** paradisaeidae_ has joined #timvideos00:54
cr1901_modernmithro: Okay... let's see what happens...00:54
cr1901_modern[email protected]:~# cat /sys/module/drm/parameters/debug00:54
cr1901_modernmithro: Aha... here we go. Linux doesn't like it :) https://gist.github.com/cr1901/5dcb886856b2b93fc5b371ec70e73de1#file-dmesg-log00:56
tpbTitle: Xorg.0.log · GitHub (at gist.github.com)00:56
*** xiaohongaiweno has joined #timvideos01:03
*** xiaohongaiweno has left #timvideos01:04
mithroNot using 800x600 mode: BAD02:25
mithrocr1901_modern: I wonder if our modeline is borked in some way02:26
mithrocr1901_modern: go look up what a valid 800x600 modeline should be and use that? :P02:26
cr1901_modernmithro: I can do that :P. But the received EDID != what was intended to be sent02:32
mithrocr1901_modern: Oh?02:32
cr1901_modernmithro: I formatted what was sent vs recv'd here https://gist.github.com/cr1901/5dcb886856b2b93fc5b371ec70e73de1#file-edid_fail-txt02:33
tpbTitle: Xorg.0.log · GitHub (at gist.github.com)02:33
cr1901_modernFirst 21 bytes are identical, but looks like synchronization goes bad or something?02:33
mithrocr1901_modern: when you set the mode in the firmware, it's not going to be the default value, right?02:33
mithrocr1901_modern: How did you figure out what is suppose to be sent?02:34
cr1901_modernmithro: I thought the default_edid WAS what was sent02:34
cr1901_modernmithro: You're saying that the Memory initialized at line 24 of edid.py is altered programmatically during firmware init?02:36
mithrocr1901_modern: Yes02:36
mithrocr1901_modern: see the edid.c file02:36
cr1901_modernmithro: Then ignore my comment for now :P02:36
*** paradisaeidae_ has quit IRC03:03
mithroxfxf: http://www.jaycar.com.au/p/XC4618?utm_source=XC4618&utm_medium=Email&utm_campaign=HCAUAugust1603:31
tpbTitle: Pan and Tilt Action Camera Bracket Mount for 9G Servos | Jaycar Electronics (at www.jaycar.com.au)03:31
cr1901_modernmithro: This is the function that decides the EDID is bad. http://lxr.free-electrons.com/source/drivers/gpu/drm/omapdrm/omap_connector.c#L205 I'll play around more tomorrow. Lines 238-244 are of interest. Not sure how they work yet04:11
tpbTitle: Linux/drivers/gpu/drm/omapdrm/omap_connector.c - Linux Cross Reference - Free Electrons (at lxr.free-electrons.com)04:11
*** rohitksingh_work has joined #timvideos04:43
*** Bertl_zZ is now known as Bertl05:06
*** Bertl is now known as Bertl_oO05:39
*** 7YUABOYZ3 has joined #timvideos06:59
*** 7YUABOYZ3 has quit IRC07:42
*** rohitksingh_work has quit IRC07:42
*** rohitksingh_work has joined #timvideos08:03
*** rohitksingh_work has quit IRC08:07
*** MaZderMind has quit IRC08:13
*** MaZderMind has joined #timvideos08:14
*** rohitksingh_work has joined #timvideos08:15
*** rohitksingh_work has quit IRC09:53
*** rohitksingh_work has joined #timvideos09:57
*** rohitksingh_work has quit IRC12:36
*** rohitksingh_work has joined #timvideos12:40
*** springermac_ has joined #timvideos12:48
*** springermac is now known as Guest9216512:48
*** springermac_ is now known as springermac12:48
*** Guest92165 has quit IRC12:50
*** rohitksingh_work has quit IRC13:05
*** rohitksingh has joined #timvideos13:44
*** rohitksingh1 has joined #timvideos14:06
*** rohitksingh has quit IRC14:06
*** rohitksingh has joined #timvideos14:13
*** rohitksingh1 has quit IRC14:13
*** danielki has joined #timvideos15:13
*** springermac_ has joined #timvideos16:44
*** springermac is now known as Guest7437216:44
*** springermac_ is now known as springermac16:45
*** Guest74372 has quit IRC16:46
*** rohitksingh has quit IRC20:01
*** danielki has quit IRC21:31
*** ssk1328 has joined #timvideos21:46
*** danielki has joined #timvideos21:48
ssk1328mithro: I have added description about a problem I am facing in a neat doc. This also contains ideas of how I am going to proceed further and some block desciptions on connections. Can you please review this and help me out. Link to code in the doc itself.21:48
tpbTitle: Mixer Block Design - Google Docs (at docs.google.com)21:48
*** danielki has quit IRC21:49

Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!