Thursday, January 21st, 2021, 00:05 UTC | ||
[00:05:18] | MythBuild: | Build [#2](https://code.mythtv.org/buildbot/#builders/211/builds/2) of `http-debian-buster-64bit` 3completed successfully. |
[00:08:25] | MythBuild: | Build [#2](https://code.mythtv.org/buildbot/#builders/208/builds/2) of `http-freebsd12–64bit` 3completed successfully. |
[00:21:55] | pvr4me_ (pvr4me_!~craigtrel@d24-150-215-113.home.cgocable.net) has quit (Quit: pvr4me_) | |
[00:39:26] | MythBuild: | Build [#2](https://code.mythtv.org/buildbot/#builders/221/builds/2) of `http-centos8–64bit` 3completed successfully. |
[00:46:29] | MythBuild: | Build [#4](https://code.mythtv.org/buildbot/#builders/220/builds/4) of `http-ubuntu-18_04–64bit` 3completed successfully. |
[00:47:59] | MythBuild: | Build [#4](https://code.mythtv.org/buildbot/#builders/207/builds/4) of `http-ubuntu-lts-64bit` 3completed successfully. |
[00:54:06] | MythBuild: | Build [#4](https://code.mythtv.org/buildbot/#builders/218/builds/4) of `http-fedora-rawhide-64bit` 3completed successfully. |
[01:58:07] | MythBuild: | Build [#5](https://code.mythtv.org/buildbot/#builders/217/builds/5) of `http-osx-64bit` 3completed successfully. |
[04:53:20] | ooshlablu (ooshlablu!~ooshlablu@98.217.124.239) has quit (Ping timeout: 256 seconds) | |
[04:53:44] | ooshlablu (ooshlablu!~ooshlablu@98.217.124.239) has joined #mythtv | |
[05:29:32] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has quit (Ping timeout: 246 seconds) | |
[05:31:19] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
[05:31:19] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth | |
[06:13:30] | Warped_ (Warped_!~Warped@108.85.160.119) has joined #mythtv | |
[06:13:31] | Warped (Warped!~Warped@unaffiliated/warped) has quit (Ping timeout: 272 seconds) | |
[06:30:00] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has quit (Quit: WeeChat 2.3) | |
[06:32:01] | peterbennett (peterbennett!~pi@2601:183:100:5f1:bcae:8e96:15ff:a1af) has joined #mythtv | |
[06:32:02] | peterbennett (peterbennett!~pi@2601:183:100:5f1:bcae:8e96:15ff:a1af) has quit (Changing host) | |
[06:32:02] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has joined #mythtv | |
[06:32:02] | Mode for #mythtv by ChanServ!ChanServ@services. : +v peterbennett | |
[07:00:38] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has quit (Ping timeout: 264 seconds) | |
[07:02:21] | peterbennett (peterbennett!~pi@2601:183:100:e:a0f4:3c6:66fb:3fd5) has joined #mythtv | |
[07:02:21] | peterbennett (peterbennett!~pi@2601:183:100:e:a0f4:3c6:66fb:3fd5) has quit (Changing host) | |
[07:02:21] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has joined #mythtv | |
[07:02:21] | Mode for #mythtv by ChanServ!ChanServ@services. : +v peterbennett | |
[07:06:19] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has quit (Read error: Connection reset by peer) | |
[07:06:28] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
[07:06:28] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth | |
[07:10:41] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has quit (Ping timeout: 246 seconds) | |
[07:11:52] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
[07:11:53] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth | |
[07:41:22] | SteveGoodey (SteveGoodey!~steve@host86-150-62-91.range86-150.btcentralplus.com) has joined #mythtv | |
[07:41:22] | Mode for #mythtv by ChanServ!ChanServ@services. : +v SteveGoodey | |
[07:54:06] | Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7d83:6501:d1ff:6177:c50e:b4be) has joined #mythtv | |
[07:54:06] | Mode for #mythtv by ChanServ!ChanServ@services. : +v Steve-Goodey | |
[08:30:09] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has quit (Ping timeout: 246 seconds) | |
[08:31:18] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
[08:31:18] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth | |
[09:15:23] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has quit (Ping timeout: 260 seconds) | |
[09:16:00] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has joined #mythtv | |
[09:16:00] | Mode for #mythtv by ChanServ!ChanServ@services. : +v peterbennett | |
[09:37:15] | MythBuild (MythBuild!~MythBuild@mizar.mythtv.org) has quit (Ping timeout: 256 seconds) | |
[09:37:25] | MythBuild (MythBuild!~MythBuild@mizar.mythtv.org) has joined #mythtv | |
[09:57:18] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has quit (Ping timeout: 246 seconds) | |
[09:58:25] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
[09:58:25] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth | |
[10:22:10] | warpme_: | guys: quick Q: is there way to totally disable upnp (by cmdline) in frontend? |
[10:22:47] | stuarta: | --disable-upnp (which incidentally shuts down http status pages) |
[10:23:04] | stuarta: | which is wrong, but we are working on that |
[10:27:03] | warpme_: | stuarta: is this switch working also frontend? |
[10:28:07] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has quit (Ping timeout: 272 seconds) | |
[10:28:24] | peterbennett (peterbennett!~pi@2601:183:100:e:a0f4:3c6:66fb:3fd5) has joined #mythtv | |
[10:28:24] | peterbennett (peterbennett!~pi@2601:183:100:e:a0f4:3c6:66fb:3fd5) has quit (Changing host) | |
[10:28:24] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has joined #mythtv | |
[10:28:25] | Mode for #mythtv by ChanServ!ChanServ@services. : +v peterbennett | |
[10:28:44] | warpme_: | https://www.irccloud.com/pastebin/KfVgn3vQ/ |
[10:35:58] | MythBuild: | Build [#6](https://code.mythtv.org/buildbot/#builders/213/builds/6) of `http-fedora-armv7hl` 3completed successfully. |
[10:37:50] | markk_ (markk_!~mark@host86-150-127-175.range86-150.btcentralplus.com) has joined #mythtv | |
[10:38:05] | markk_: | warpme_: --noupnp |
[10:38:16] | markk_: | is what you need |
[10:39:06] | stuarta: | markk_: ah yes, that sounds better than my nonsense |
[10:40:25] | SteveGoodey: | !trout stuarta |
[10:40:25] | ** MythLogBot slaps stuarta with a trout on behalf of SteveGoodey... ** | |
[10:40:37] | SteveGoodey: | Is that acceptable? |
[10:40:38] | stuarta: | !zap SteveGoodey |
[10:40:45] | stuarta: | dang, i never implemented that |
[10:40:52] | SteveGoodey: | Haha, |
[10:40:57] | stuarta: | SteveGoodey: totally :) |
[10:41:10] | stuarta: | it is after all, what it's for |
[10:45:38] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has quit (Ping timeout: 264 seconds) | |
[10:45:54] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has joined #mythtv | |
[10:45:54] | Mode for #mythtv by ChanServ!ChanServ@services. : +v peterbennett | |
[10:47:22] | stuarta: | markk_: welcome back |
[10:53:13] | warpme_: | guys: wonder your opinion: i have single box which is segfaulting frontend at start – but only in EGLFS. RAM is too small to launch debug build (with symbols) of frontend under gdb. Debug level log is like this: http://warped.inet2.org/mythfrontend.20210121104728.4962.log I wonder how to move forward with this? |
[10:59:15] | stuarta: | warpme_: profile build, contains sufficient symbol info to be able to debug it |
[10:59:32] | stuarta: | be warned profile build turns on optimizations so it will take 4x longer to build |
[11:01:27] | warpme_: | or maybe: 2 builds (debug & release) and use launch mix of both with only few libs/bins from debug (those where issue is expected)? |
[11:02:11] | stuarta: | that is worse. use profile build, it is the half way house you are looking for |
[11:03:03] | markk_: | warpme_: your log is behind a paywall:) If eglfs segfaults – check you have a display connected |
[11:03:11] | markk_: | stuarta: thanks:) |
[11:03:38] | warpme_: | markk_: for sure i have :-) |
[11:04:03] | markk_: | warpme_: so you are not getting any tearing/picture breakup with DRM video? what about other alpha blending problems etc? |
[11:04:16] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has quit (Ping timeout: 260 seconds) | |
[11:05:01] | peterbennett (peterbennett!~pi@2601:183:100:e:a0f4:3c6:66fb:3fd5) has joined #mythtv | |
[11:05:01] | peterbennett (peterbennett!~pi@2601:183:100:e:a0f4:3c6:66fb:3fd5) has quit (Changing host) | |
[11:05:01] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has joined #mythtv | |
[11:05:01] | Mode for #mythtv by ChanServ!ChanServ@services. : +v peterbennett | |
[11:07:09] | warpme_: | well – short tests on h6 & rk3328 (slowest crap i have) – playback is really good. no smoothness like with 2x framerate DI but experience is "good". Looking on scrolling i have impression of regular A/V sync "dancing" (a bit: slower-faster-slower-faster with cycle approx 200–300ms) |
[11:08:41] | warpme_: | need to verify isn't this "A/V sync" results of frame rate missmatch (content was 50i; monitor iirc was 60Hz) |
[11:10:23] | warpme_: | markk_: btw: we have HW DI working on H6 (as ffmpeg filter). Probably it will be worth to add this in future. I have patches for: kernel + ffmpeg and also KODI code snippets to enable ffmpeg DI filter |
[11:10:34] | markk_: | warpme_: I can't get decent video rendering on any device I've tested. tried the tanix box at 4k on a 4k display yesterday – terrible... |
[11:11:02] | markk_: | warpme_: can you point to the code? |
[11:11:31] | warpme_: | markk_: let me check this on 4k monitor. currently my test were with hd monitor.... |
[11:12:02] | warpme_: | re: DI: sure. give me sec |
[11:14:56] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has quit (Ping timeout: 240 seconds) | |
[11:16:00] | peterbennett (peterbennett!~pi@2601:183:100:e:a0f4:3c6:66fb:3fd5) has joined #mythtv | |
[11:16:00] | peterbennett (peterbennett!~pi@2601:183:100:e:a0f4:3c6:66fb:3fd5) has quit (Changing host) | |
[11:16:00] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has joined #mythtv | |
[11:16:01] | Mode for #mythtv by ChanServ!ChanServ@services. : +v peterbennett | |
[11:19:43] | warpme_: | markk_: Here it is: kernel driver: https://github.com/warpme/minimyth2/blob/mast . . . ce-WiP.patch and kernel DT enablement: https://github.com/warpme/minimyth2/blob/mast . . . e-node.patch ffmpeg: |
[11:19:43] | warpme_: | https://github.com/warpme/minimyth2/blob/mast . . . kernel.patch KODI enablement code: https://github.com/LibreELEC/LibreELEC.tv/blo . . . ilters.patch and |
[11:19:43] | warpme_: | https://github.com/LibreELEC/LibreELEC.tv/blo . . . filter.patch |
[11:25:30] | warpme_: | markk_: "check you have a display connected" – i have physically connected monitor – but maybe kernel/drm reports to Qt/myth wrong/false info about monitor? Is there way to verify this in logs? |
[11:28:49] | warpme_: | DRM says following: https://pastebin.com/EXcXyHpP |
[11:31:15] | markk_: | warpme_: drm info looks good. was just a suggestion:) the one time I had the code crashing hard at startup with no logs was because no display was connected. |
[11:35:20] | warpme_: | btw: this EGLFS issue i have on ALL amlogic boxes I have (s905/mali450; s912/t720 and sm1/g31) so it seems to be independent on GL hardware. Exactly the same sw. stack works well on rk/aw/rpi/intel/amd. Qt X11/GL(glamour) works OK on AML – only EGLFS traps... |
[11:36:56] | stuarta: | specific to the amlogic box then |
[11:43:55] | warpme_: | stuarta: my initial hypothesis was: EGLFS uses mesa GLES call(s) which are exposing issue in aml drm driver. It was developed on assumption: difference in observed behaviour (aml segfaults; aw/rk/rpi/intel/amd not segfaults) comes from difference in sw/hw. Player/qt/mesa are the same binary for all cases. Difference is: drm & hw. As mali/t720/g31 segfaults the same way – drm seems highest correlating to |
[11:43:55] | warpme_: | sefault:nosegfaul cases. But.....strace on trapping frontend shows no touches to drm (/dev/dri/card0). It looks like without gdb trace i can't move forward with this. Let me try rebuild with --compile-type="profile" |
[12:02:19] | markspieth2 (markspieth2!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
[12:02:20] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth2 | |
[12:05:02] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has quit (Ping timeout: 246 seconds) | |
[12:05:26] | warpme_: | https://www.irccloud.com/pastebin/7YfXrOv2/ |
[12:06:53] | warpme_: | markk_: indeed – playing 4 content on 4k monitor gives me jumpy playback. I think issue is about v4l2 buffers tweaking as fe log says to me: |
[12:06:58] | warpme_: | 2021-01–21 13:03:38.455046 I GLVid: New frame format: None:None 3840x2160 (Tex: 2D) -> DRM-PRIME:RGBA32 3840x2160 (Tex: OES) |
[12:06:58] | warpme_: | 2021-01–21 13:03:38.652602 N Player(0): Waited 103ms for video buffers PAAAAA |
[12:06:58] | warpme_: | 2021-01–21 13:03:39.796736 E AFD: video avcodec_send_packet error: Cannot allocate memory (-12) gotpicture:0 |
[12:06:58] | warpme_: | 2021-01–21 13:03:39.802581 E AFD: video avcodec_send_packet error: Cannot allocate memory (-12) gotpicture:0 |
[12:10:14] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has quit (Ping timeout: 264 seconds) | |
[12:10:39] | peterbennett (peterbennett!~pi@2601:183:100:e:a0f4:3c6:66fb:3fd5) has joined #mythtv | |
[12:10:40] | peterbennett (peterbennett!~pi@mythtv/developer/peterbennett) has joined #mythtv | |
[12:10:40] | peterbennett (peterbennett!~pi@2601:183:100:e:a0f4:3c6:66fb:3fd5) has quit (Changing host) | |
[12:10:40] | Mode for #mythtv by ChanServ!ChanServ@services. : +v peterbennett | |
[12:12:46] | markk_ (markk_!~mark@host86-150-127-175.range86-150.btcentralplus.com) has quit (Ping timeout: 264 seconds) | |
[13:03:05] | markk_ (markk_!~mark@host86-150-127-175.range86-150.btcentralplus.com) has joined #mythtv | |
[13:06:56] | warpme_: | markk_ stuarta: i managed to get started myth under gdb to get trace of trap on amlogic eglfs (profile builds also exceeds memory at start – so i selectivity replace libs build with symbols). Here is trace: https://pastebin.com/8pYxBs2K Is this useful? |
[13:07:11] | markk_: | warpme_: you shouldn't be hitting the log line where it says 'GLVid: New frame format: None:None 3840x2160 (Tex: 2D) -> DRM-PRIME:RGBA32 3840x2160 (Tex: OES)'. Is there any error earlier than that? |
[13:09:54] | warpme_: | markk_: let me give you full log from playback :) |
[13:15:11] | warpme_: | here it is: https://pastebin.com/yMgdRhyC |
[13:15:43] | markk_: | warpme_: re crash – I'm fixing that crash – but it may not help. for some reason Qt has no screens at that point – which will be a problem elsewhere. |
[13:16:01] | warpme_: | perfect ! |
[13:16:05] | markk_: | warpme_: QT_QPA_EGLFS_KMS user overrides detected <<<<<<<<< no drm video with that log |
[13:17:21] | markk_: | actually – that's a bit of a bug – but regardless – no drm video |
[13:25:17] | warpme_: | markk_: re no drm video with "QT_QPA_EGLFS_KMS user overrides detected". It somehow works as crappy rk3328 wasn't unwatchable at all with EGL drm_prime but now works really well........ maybe it works now well because we have EGL drm_prime on EGLFS (not on X11)? |
[13:33:43] | markk_: | warpme_: definitely better performance without x. what environment variables have you set? |
[13:35:05] | warpme_: | markk_: probably this env set by me is issue: QT_QPA_EGLFS_KMS_CONFIG=/var/run/qt/eglfs_kms_config.json |
[13:35:56] | warpme_: | done this after reading https://doc.qt.io/qt-5/embedded-linux.html and your remark about broken install.... |
[13:38:48] | warpme_: | but without this env i'm getting "Failed to open 'eglfs_kms_config.json' for writing. DRM setup incomplete" |
[13:39:28] | warpme_: | this is from g63bd210347 (so before some fixes from you!) |
[13:42:11] | markk_: | warpme: crash fix is in. I'm guessing QT_QPA_EGLFS_KMS_CONFIG=/var/run/qt/eglfs_kms_config.json is also causing the no screen issue. any reason you can't give mythfrontend some write permissions. without allowing it to write that config file for Qt, you will have to hand code a config for each setup. |
[13:46:09] | warpme_: | "any reason you can't give mythfrontend some write permissions" – absolutely no reason. I wasn't able to deduce to where mythfrontend process needs to have white permissions. pls advice |
[13:47:17] | pvr4me_ (pvr4me_!~craigtrel@d24-150-215-113.home.cgocable.net) has joined #mythtv | |
[13:49:12] | markk_: | warpme_: when that code is processed, we do not have any settings or database and no directories are setup (i.e. we do not know about ~/.mythtv etc). So I just use the working directory (i.e. the directory mythfrontend was run from). I could change that to the current users home directory? ie ~/eglfs_kms_config.json |
[13:50:24] | stuarta: | markk_: if the ENV MYTHCONFDIR exists, create it in there, else $HOME/.mythtv (the default value of MYTHCONFDIR) |
[13:51:58] | markk_: | stuarta: thanks – hadn't spotted MYTHCONFDIR – sounds like a plan |
[13:52:07] | warpme_: | indeed. done test with changing dir to mythfrontend user home and launching frontend. now i have: |
[13:52:13] | warpme_: | https://www.irccloud.com/pastebin/lSW8tD7M/ |
[13:52:20] | warpme_: | good |
[13:52:32] | warpme_: | let see now how it works :-) |
[13:52:45] | stuarta: | markk_: i know about it because it allows me to have both fixes/31 and master on the same machine :) |
[13:53:26] | warpme_: | yeah. now i see tearing :-p |
[13:56:06] | markk_: | warpme_: :( |
[13:56:20] | stuarta: | see i thought that was good because it was expected? |
[14:16:50] | warpme_: | markk_: im scratching my head: how we can have tearing when we have atomic flips? what comes to mind is v_blanks events propagated to player loop are not in pair/synced (or some to them are lost) with crtc hw vblank. it is possible to get logging timestamps where commits are triggered by player loop? Then we can compare them with drm logged commits... |
[14:22:57] | MythBuild: | Build [#6](https://code.mythtv.org/buildbot/#builders/212/builds/6) of `http-debian-buster-rpi2-armv7l` 3completed successfully. |
[14:24:16] | stuarta: | yay, only took it 5hrs or so |
[14:39:32] | warpme_: | markk_: i looked on playback in slowmotion. what is see is more judder than tearing. i have 50i content on 60Hz tv so...juder is expected. What is good wan to set refresh rate to 50Hz in EGLFS? Should i create custom eglfs_kms_config.json? |
[14:39:52] | markk_: | warpme_: yeah – it makes no sense to me either. the only idea I have is that qt is allowing a modeset on each commit: https://github.com/qt/qtbase/blob/5.15.2/src/ . . . ice.cpp#L961 |
[14:40:30] | markk_: | complete guess. really need to get some drm debugging setup and compare it to some other player |
[14:42:02] | markk_: | warpme_: there is definitely significant tearing and out of order presentation. consistent across different devices – so not driver related. |
[14:42:40] | stuarta: | consistency++ |
[14:42:51] | warpme_: | kodi guys tells me drm should be set to allow DRM_MODE_ATOMIC_ALLOW_MODESET. but in my case any analysis when i have fps not equal to refresh rate is useless... |
[14:47:46] | markk_: | warpme_: for best testing I would suggest downloading some progressive 60fps material. |
[14:48:03] | warpme_: | i was told: starting point to debug drm playback is logging commit data+timestamps. I agree with this statement as having efficient logging (i mean logging not impacting/skewing logged code execution) will be very useful. we can see i.e any artefacts in time dimension (like lost vblanks, coalescent vblanks, too short to code loop exec time vblanks, etc). just brain storming here :-) |
[15:10:42] | warpme_: | isn't that by https://github.com/qt/qtbase/blob/40143c189b7 . . . ice.cpp#L219 i can declare refrestrate by declaring "mode": "1920x1080@50" in custom eglfs_kms_config.json? |
[15:11:03] | warpme_: | refrestrate->refreshrate |
[15:18:20] | warpme_: | markk_: setting refresh by custom .json works. now my monitor says 50Hz. is this log saying drm_prime is in drm planes mode? https://pastebin.com/rB7xtPjB |
[15:20:57] | warpme_: | nope. seems egl mode :-( |
[15:22:31] | markk_ (markk_!~mark@host86-150-127-175.range86-150.btcentralplus.com) has quit (Ping timeout: 246 seconds) | |
[15:28:33] | markk_ (markk_!~mark@host86-150-127-175.range86-150.btcentralplus.com) has joined #mythtv | |
[15:29:35] | Guest6892 (Guest6892!maliutamat@gateway/shell/matrix.org/x-sivxqmqbpqbvvfhp) has joined #mythtv | |
[15:34:21] | markspieth2 (markspieth2!~yaaic@mythtv/developer/markspieth) has quit (Ping timeout: 246 seconds) | |
[15:51:09] | Warped_ (Warped_!~Warped@108.85.160.119) has quit (Quit: ChatZilla 0.9.92-rdmsoft [XULRunner 35.0.1/20150122214805]) | |
[16:04:42] | markk_: | warpme_: so I *think* I *might* know what is going on. We are mapping drm frames to a framebuffer, commiting that framebuffer and then releasing the video frame before the page flip. the decoder then starts to overwrite that frame before or while it is actually rendered. |
[16:10:47] | markk_: | nailed it:) |
[16:26:26] | warpme_: | markk_: this may explain sporadic jumps by just frame at playback i see. issue with testing i have is refresh rate vs. content fps. it looks like custom .json (where i can declare refresh) turns-off drm planes rendering. turn-off drm planes by custom .json is design decision or WiP? btw: i got working amlogic. issue is wrong display name in .json file. Generated by Qt(or myth; i don't know yet who generates it) is: |
[16:26:26] | warpme_: | HDMI1 while modeset tool says: HDMI-A-1. Changing this is .json file makes frontend working. Q for me is: it this Qt bug or amlogic drm driver reports wrongly HDMI1 (so Qt wrongly writes this to .json for further usage and fails due wrong name).... |
[16:27:50] | warpme_: | qll you find & nailed it!!! |
[16:54:14] | warpme_: | markk_: just fyi: looking on mythdrmdevice.cpp i see name is taken from drmGetDeviceNameFromFd2. So it looks like libdrm call to drmGetDeviceNameFromFd2 returns wrong name? (is HDMI1 and should be HDMI-A-1 as this name works...)? |
[16:59:35] | markk_ (markk_!~mark@host86-150-127-175.range86-150.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
[17:19:07] | markspieth (markspieth!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
[17:19:07] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth | |
[17:21:43] | Guest6892 (Guest6892!maliutamat@gateway/shell/matrix.org/x-sivxqmqbpqbvvfhp) has quit (Changing host) | |
[17:21:43] | Guest6892 (Guest6892!maliutamat@unaffiliated/maliuta) has joined #mythtv | |
[17:21:43] | Guest6892 (Guest6892!maliutamat@gateway/shell/matrix.org/x-sivxqmqbpqbvvfhp) has joined #mythtv | |
[17:21:44] | Guest6892 (Guest6892!maliutamat@unaffiliated/maliuta) has quit (Changing host) | |
[18:51:16] | Warped (Warped!~Warped@unaffiliated/warped) has joined #mythtv | |
[19:27:59] | ijc (ijc!~ijc@benson.default.arb33.uk0.bigv.io) has quit (*.net *.split) | |
[19:28:07] | ijc (ijc!~ijc@benson.default.arb33.uk0.bigv.io) has joined #mythtv | |
[21:45:09] | pvr4me_ (pvr4me_!~craigtrel@d24-150-215-113.home.cgocable.net) has quit (Quit: pvr4me_) | |
[22:32:08] | SteveGoodey (SteveGoodey!~steve@host86-150-62-91.range86-150.btcentralplus.com) has quit (Quit: Konversation terminated!) | |
[22:32:36] | Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7d83:6501:d1ff:6177:c50e:b4be) has quit (Quit: Konversation terminated!) |
IRC Logs collected by
BeirdoBot.
Please use the above link to report any bugs.