| Wednesday, September 12th, 2018, 00:29 UTC | ||
| [00:29:13] | jpharvey_ (jpharvey_!~jpharvey@95.149.163.5) has quit (Ping timeout: 244 seconds) | |
| [00:33:28] | hampton (hampton!~hampton@mythtv/developer/hampton) has quit (Ping timeout: 250 seconds) | |
| [00:48:50] | gigem: | peterbennett: The current fire tv stick uses a mediatek 8127d cpu, mali 450 gpu and has 1 gb or ram. If mythtv runs on it, I wouldn't expect it to run well. |
| [01:50:24] | amessina (amessina!~amessina@unaffiliated/amessina) has joined #mythtv | |
| [02:33:39] | jpabq (jpabq!~quassel@mythtv/developer/jpabq) has quit (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.) | |
| [02:33:43] | amessina (amessina!~amessina@unaffiliated/amessina) has quit (Quit: Konversation terminated!) | |
| [03:00:53] | hampton (hampton!~hampton@mythtv/developer/hampton) has joined #mythtv | |
| [03:00:53] | Mode for #mythtv by ChanServ!ChanServ@services. : +v hampton | |
| [03:02:08] | DarkWhite_ (DarkWhite_!~sailfish@dyn-207-165-dsl.vsp.fi) has quit (Read error: Connection reset by peer) | |
| [07:45:43] | willcooke (willcooke!~willcooke@ubuntu/member/willcooke) has joined #mythtv | |
| [08:00:04] | markspieth2 (markspieth2!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
| [08:00:04] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth2 | |
| [08:15:54] | ** stuarta upgrades shield ** | |
| [08:16:41] | stuarta: | peterbennett: i'm hoping the newer raspbians move towards more standard interfaces, so we can reduce the rpi specific bits |
| [08:22:31] | markspieth2 (markspieth2!~yaaic@mythtv/developer/markspieth) has quit (Read error: Connection reset by peer) | |
| [08:22:37] | markspieth22 (markspieth22!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
| [08:22:37] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth22 | |
| [09:01:36] | Merlin83b (Merlin83b!~Daniel@2a00:1ee0:3:1337:1116:e9c7:1bcf:d646) has joined #mythtv | |
| [11:33:25] | Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7da3:4501:4a51:b7ff:fe21:2979) has joined #mythtv | |
| [11:33:26] | Mode for #mythtv by ChanServ!ChanServ@services. : +v Steve-Goodey | |
| [12:20:48] | Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7da3:4501:4a51:b7ff:fe21:2979) has quit (Remote host closed the connection) | |
| [12:41:17] | peterbennett: | stuarta: Unfortunately as raspberry pi moves towards more standard software, it runs slower and can no longer play video smoothly, or so it seems. |
| [12:45:43] | stuarta: | i am hopeful the vc4 work being done in mesa will help. i suspect it's slow progress tho |
| [12:46:40] | stuarta: | reasonable opengles speed is the aim |
| [13:41:44] | ooshlablu (ooshlablu!~ooshlablu@2601:18d:4600:5f31:fc64:7b43:de46:2015) has joined #mythtv | |
| [13:44:54] | amessina (amessina!~amessina@unaffiliated/amessina) has joined #mythtv | |
| [13:53:25] | sphery (sphery!~mdean@mythtv/developer/sphery) has quit (Ping timeout: 246 seconds) | |
| [13:53:42] | sphery (sphery!~mdean@mythtv/developer/sphery) has joined #mythtv | |
| [13:53:45] | Mode for #mythtv by ChanServ!ChanServ@services. : +v sphery | |
| [13:57:57] | knowledgejunkie (knowledgejunkie!~knowledge@cpc77018-warw18-2-0-cust959.3-2.cable.virginm.net) has joined #mythtv | |
| [14:46:18] | gigem: | stuarta: I updated one of my shields last night. In very brief testing, it worked all right. I'll use it more tonight before updating the other shield. |
| [15:10:36] | gregl (gregl!~greg@cpe-24-194-253-7.nycap.res.rr.com) has quit (Remote host closed the connection) | |
| [15:12:33] | gigem: | So close and yet so far! In the slow resume case for the watch recording screen, I've narrowed the problem down to two threads which use 100% cpu for several seconds to several minutes. In top, they are labeld "tv.mythfrontend" (I assume that's really "org.mythtv.mythfrontend" truncated to fit) and "QtMainThread" (of which there are several with that name). Unfortunately, top doesn't report the unique, |
| [15:12:34] | gigem: | thread ID for each one. Instead, it only reports the common process ID for all of the threads. Does anyone know how to get the unique, thread IDs? |
| [15:24:08] | gigem: | Found it. I need to specify a custom, fields list which includes TID. |
| [15:26:06] | gregl (gregl!~greg@cpe-24-194-253-7.nycap.res.rr.com) has joined #mythtv | |
| [15:35:08] | SteveGoodey (SteveGoodey!~steve@2a00:23c5:7da3:4501:7a24:afff:fe9d:c233) has joined #mythtv | |
| [15:35:08] | Mode for #mythtv by ChanServ!ChanServ@services. : +v SteveGoodey | |
| [15:41:54] | amessina (amessina!~amessina@unaffiliated/amessina) has quit (Remote host closed the connection) | |
| [15:42:21] | amessina (amessina!~amessina@unaffiliated/amessina) has joined #mythtv | |
| [15:56:31] | gigem: | This time there was only 1 thread using 100% cpu. It was thread 1 which I think is the Qt event loop and is what I suspected. The stack trace is useless, however. The relevant part is at https://pastebin.com/i8QJxhQY . If anyone knows how to get a better stack trace, please let me know. Time to figure out how to debug the event loop. |
| [16:20:18] | Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7da3:4501:4a51:b7ff:fe21:2979) has joined #mythtv | |
| [16:20:19] | Mode for #mythtv by ChanServ!ChanServ@services. : +v Steve-Goodey | |
| [16:40:30] | ooshlablu (ooshlablu!~ooshlablu@2601:18d:4600:5f31:fc64:7b43:de46:2015) has quit (Remote host closed the connection) | |
| [16:41:32] | ooshlablu (ooshlablu!~ooshlablu@2601:18d:4600:5f31:a41d:40b0:725b:6115) has joined #mythtv | |
| [16:46:57] | peterbennett: | gigem: To get a better stack trace you may have to package non-stripped executables. It seems that the qt packaging tool strips the executables, but there may be an option to prevent that. |
| [17:03:05] | Merlin83b (Merlin83b!~Daniel@2a00:1ee0:3:1337:1116:e9c7:1bcf:d646) has quit (Quit: Leaving) | |
| [17:03:15] | willcooke (willcooke!~willcooke@ubuntu/member/willcooke) has quit (Quit: Do your hobbies) | |
| [17:16:11] | natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv | |
| [17:16:11] | Mode for #mythtv by ChanServ!ChanServ@services. : +v natanojl | |
| [17:20:59] | jpabq (jpabq!~quassel@mythtv/developer/jpabq) has joined #mythtv | |
| [17:21:00] | Mode for #mythtv by ChanServ!ChanServ@services. : +v jpabq | |
| [17:21:17] | jpabq: | gigem: is the Fire Cube any better? |
| [17:39:30] | Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7da3:4501:4a51:b7ff:fe21:2979) has quit (Ping timeout: 264 seconds) | |
| [18:39:06] | peterbennett: | stuarta: Trac is performing well these days. But the commits take a long time to update the tickets (many hours). Is that expected? |
| [18:44:27] | brfransen (brfransen!~brfransen@47.42.27.251) has quit (Ping timeout: 240 seconds) | |
| [18:45:39] | brfransen (brfransen!~brfransen@47.42.27.251) has joined #mythtv | |
| [18:52:26] | stuarta: | peterbennett: it'll either take 30 or 60s or not at all until i poke it |
| [18:53:38] | stuarta: | right, last 3 commits didn't work properly |
| [18:55:16] | stuarta: | fixed |
| [18:55:34] | stuarta: | that github plugin for trac isn't great, but sadly it's better than nothing |
| [19:00:09] | peterbennett: | stuarta: I see, so when it does not work after a couple of minutes I should contact you |
| [19:01:10] | peterbennett: | stuarta: I hvae committed the vaapi2 stuff, so if that fixes the wayland and other issues we can close those tickets also |
| [19:01:58] | peterbennett: | #13298, #13104, #13144 |
| [19:01:58] | ** MythLogBot https://code.mythtv.org/trac/ticket/13298 ** | |
| [19:01:58] | ** MythLogBot https://code.mythtv.org/trac/ticket/13104 ** | |
| [19:01:58] | ** MythLogBot https://code.mythtv.org/trac/ticket/13144 ** | |
| [19:05:32] | natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 272 seconds) | |
| [19:05:41] | stuarta: | peterbennett: yes it worked in my earlier testing |
| [19:06:01] | stuarta: | haven't tested yet with the committed version |
| [19:06:45] | peterbennett: | stuarta: There is now a setting in video setup for the vaapi device |
| [19:07:23] | stuarta: | peterbennett: you can close #13104, but we haven't actually added wayland support #13144 |
| [19:07:23] | ** MythLogBot https://code.mythtv.org/trac/ticket/13104 ** | |
| [19:07:23] | ** MythLogBot https://code.mythtv.org/trac/ticket/13144 ** | |
| [19:08:01] | peterbennett: | stuarta: Also, the file that "wedged" – see if it at least can fail over to software decoding – if not get me a sample and I will see if I can make it work. |
| [19:08:29] | stuarta: | i suspect hardware related, so we may have to blacklist high profile on older hardware |
| [19:09:33] | peterbennett: | I am trying to make it so that it will silently fail over if it does not work, rather than trying to figure out in advance whether it will work. |
| [19:10:00] | stuarta: | that assumes that the api returns a failure rather than never returning |
| [19:11:04] | peterbennett: | One can hope... |
| [19:11:15] | stuarta: | yeah, i'm hoping it's the former too |
| [19:11:25] | stuarta: | although i have a suspicion it's the latter |
| [19:12:24] | peterbennett: | So #13144 says add support to fix #13104. I am not sure about what that means. |
| [19:12:24] | ** MythLogBot https://code.mythtv.org/trac/ticket/13144 ** | |
| [19:12:24] | ** MythLogBot https://code.mythtv.org/trac/ticket/13104 ** | |
| [19:13:39] | stuarta: | my original plan to fix it, was to rework MythDisplay class, so it kept track of which it was using, by first trying wayland, and then falling back to x connections |
| [19:14:11] | stuarta: | the if it's wayland, you do the vaInit with a wayland display rather than an x display |
| [19:14:50] | stuarta: | see https://github.com/stuarta/opengl-qt5/blob/ma . . . main.cpp#L22 to line 31 |
| [19:15:08] | peterbennett: | I think ffmpeg now sorts that out for you . We do not do a vaInit or any va call with vaapi2 |
| [19:15:26] | stuarta: | no, that's a nicety of the ffmpeg api |
| [19:15:49] | stuarta: | i guess the proper status of that bug, is closed : wontfix |
| [19:16:01] | stuarta: | since it's no longer needed to explicitly support it |
| [19:16:40] | peterbennett: | If we can leave as much as possible to ffmpeg, qt and other frameworks, it will make our life easier |
| [19:19:32] | stuarta: | agreed, that should be the preferred direction |
| [19:26:01] | natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv | |
| [19:26:01] | Mode for #mythtv by ChanServ!ChanServ@services. : +v natanojl | |
| [19:50:21] | hampton: | amen :-) |
| [20:55:49] | stuarta: | peterbennett: so a quick test on the committed vaapi2 code. works fine on everything i have except the one recording that locks up the decoding |
| [21:19:27] | natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 240 seconds) | |
| [21:28:34] | gigem: | jpabq: The Cube is better than the Stick. It's essentially a 3rd generation Fire TV with an always listening Alexa and some Harmony Hub capabilities (can switch TV inputs, turn devices on/off, etc.) Both use Amlogic 905 cpus. Neither is as powerful cpu-wise as the 2nd generation Fire TV. |
| [21:29:24] | gigem: | peterbennett: I'll have to look into the stripped libraries issue later. All of a sudden today I'm juggling too many balls and feel like pulling out what little hair I have left. |
| [21:31:18] | peterbennett: | gigem: So each generation fire TV is less powerful than the last ? |
| [21:48:32] | gigem: | peterbennett: No, second was better than first. Third is worse than second but better than first. I think third was mainly a cost savings iteration. It's only us power hungry, HTPC enthusiasts who think the third generation was a step back. |
| [21:49:37] | SteveGoodey (SteveGoodey!~steve@2a00:23c5:7da3:4501:7a24:afff:fe9d:c233) has quit (Quit: Konversation terminated!) | |
| [22:39:22] | markspieth22 (markspieth22!~yaaic@mythtv/developer/markspieth) has quit (Ping timeout: 256 seconds) | |
| [22:40:21] | amessina (amessina!~amessina@unaffiliated/amessina) has quit (Quit: Konversation terminated!) | |
| [22:42:33] | markspieth2 (markspieth2!~yaaic@mythtv/developer/markspieth) has joined #mythtv | |
| [22:42:33] | Mode for #mythtv by ChanServ!ChanServ@services. : +v markspieth2 | |
IRC Logs collected by
BeirdoBot.
Please use the above link to report any bugs.