Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2019-08-12

00:24:01 Quit mendelmunkis_ (Ping timeout: 248 seconds)
00:48:26***Saving seen data "./dancer.seen"
01:00
01:00:14 Join Bilgus [0] (~Bilgus@unaffiliated/bilgus)
01:04:56 Quit _Bilgus (Ping timeout: 276 seconds)
01:23:37 Quit speachy (Ping timeout: 260 seconds)
02:00
02:20:05 Quit ZincAlloy (Quit: Leaving.)
02:48:30***Saving seen data "./dancer.seen"
03:00
03:49:03 Join mendelmunkis [0] (~mendelmun@cpe-67-245-191-31.nyc.res.rr.com)
04:00
04:06:58 Quit mendelmunkis (Ping timeout: 276 seconds)
04:48:34***Saving seen data "./dancer.seen"
05:00
05:26:27 Quit TheSeven (Ping timeout: 264 seconds)
05:26:40 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
06:00
06:48:36***Saving seen data "./dancer.seen"
07:00
07:00:48 Quit bzed (Read error: Connection reset by peer)
07:02:59 Join bzed [0] (~bzed@shell.bzed.at)
07:28:26fs-bluebotBuild Server message: New build round started. Revision 3d06d35, 280 builds, 11 clients.
07:46:17fs-bluebotBuild Server message: Build round completed after 1071 seconds.
07:46:18fs-bluebotBuild Server message: Revision 3d06d35 result: All green
08:00
08:10:13 Quit pixelma (Quit: .)
08:10:13 Quit amiconn (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)
08:11:15 Join pixelma [0] (marianne@rockbox/staff/pixelma)
08:11:16 Join amiconn [0] (jens@rockbox/developer/amiconn)
08:28:15 Quit [7] (Ping timeout: 264 seconds)
08:47:47 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
08:48:39***Saving seen data "./dancer.seen"
08:50:50 Join petur [0] (~petur@rockbox/developer/petur)
09:00
09:03:33 Join quadjfet [0] (quadjfet@gateway/vpn/privateinternetaccess/quadjfet)
09:05:33 Quit advcomp2019__ (Read error: Connection reset by peer)
09:06:10 Join advcomp2019__ [0] (~advcomp20@unaffiliated/advcomp2019)
09:11:03 Quit quadjfet (Quit: Leaving)
09:43:40 Join dys [0] (~dys@2003:5b:203b:100:a64c:c8ff:fef4:13a6)
09:46:17 Join deevious [0] (~Thunderbi@193.226.142.214)
10:00
10:42:41 Quit advcomp2019__ (*.net *.split)
10:42:46 Quit plum (*.net *.split)
10:42:46 Quit St3ak (*.net *.split)
10:42:50 Quit ufdm (*.net *.split)
10:42:52 Quit Guest33065 (*.net *.split)
10:42:55 Quit bremner (*.net *.split)
10:42:56 Quit TheSphinX^ (*.net *.split)
10:42:57 Quit ender| (*.net *.split)
10:43:02 Join plum [0] (~plum@unaffiliated/plum)
10:43:44 Join Guest33065 [0] (0x56@unaffiliated/x56)
10:47:56 Join bremner [0] (~bremner@fethera.tethera.net)
10:47:56 Join advcomp2019__ [0] (~advcomp20@unaffiliated/advcomp2019)
10:47:56 Join St3ak [0] (~st3ak@st3ak3000.powered.by.lunarbnc.net)
10:47:56 Join TheSphinX^ [0] (~briehl@srv001.nosupamu.de)
10:47:56 Join ender| [0] (~ender1@2a01:260:4094:1:225:90ff:fe2a:83dc)
10:48:42***Saving seen data "./dancer.seen"
12:00
12:09:36 Join vmx [0] (~vmx@ip5f5ac62a.dynamic.kabel-deutschland.de)
12:48:43***No seen item changed, no save performed.
13:00
13:32:10 Quit bremner (Changing host)
13:32:10 Join bremner [0] (~bremner@debian/developer/bremner)
13:45:16 Join krabador [0] (~krabador@unaffiliated/krabador)
13:45:19 Quit krabador (Read error: Connection reset by peer)
13:57:36 Join massiveH [0] (~massiveH@ool-18e4ed57.dyn.optonline.net)
14:00
14:09:27 Join krabador [0] (~krabador@unaffiliated/krabador)
14:33:01 Join Rower [0] (~husvagn@m83-185-94-13.cust.tele2.se)
14:48:44***Saving seen data "./dancer.seen"
15:00
15:15:09 Quit massiveH (Quit: Leaving)
15:34:53 Join speachy [0] (40eebded@64.238.189.237)
15:36:17 Quit krabador (Remote host closed the connection)
15:44:58 Quit igitoor (Remote host closed the connection)
16:00
16:01:48 Join igitoor [0] (igitur@2a00:d880:3:1::c1ca:a648)
16:15:59 Quit igitoor (Changing host)
16:15:59 Join igitoor [0] (igitur@unaffiliated/contempt)
16:48:47***Saving seen data "./dancer.seen"
17:00
17:06:14 Join ufdm [0] (~ufdm@c-75-72-5-49.hsd1.mn.comcast.net)
17:08:06 Quit vmx (Remote host closed the connection)
17:29:59 Quit petur (Quit: Connection reset by beer)
18:00
18:38:27 Quit dys (Ping timeout: 264 seconds)
18:44:26 Join dys [0] (~dys@tmo-104-211.customers.d1-online.com)
18:48:50***Saving seen data "./dancer.seen"
19:00
19:10:32 Join lebellium [0] (~lebellium@89-92-69-110.hfc.dyn.abo.bbox.fr)
19:32:33Bilgusspeachy, in opus I remember you messing with the codec to rectify codec thread overflow didn't you try to relocate some of the structures to the heap to decrease stack usage or am I mis-remembering?
19:33:39BilgusI saw whrer you increased the stack but the problem still occurs so I'm trying to figure out the best way to fix it or atleast fail gracefully
19:47:54 Join ZincAlloy [0] (~Adium@2a02:8108:9440:dfc:d108:3622:4566:e73b)
19:48:17 Quit ZincAlloy (Client Quit)
19:48:26 Join ZincAlloy [0] (~Adium@2a02:8108:9440:dfc:d108:3622:4566:e73b)
20:00
20:02:08speachyBilgus, I moved a few structs off the stack onto the heap. I know that one of them definitely wasn't part of a hot path, but I didn't look at the other two.
20:02:33speachysaved 640 bytes of stack (on MIPS, IIRC).
20:03:09 Join IncognitoMan__ [0] (~Incognito@unaffiliated/incognitoman)
20:03:10speachyI didn't have any access to IRAM-less targets that could have been adversely affected by the change, so couldn't benchmark things to see if it hurt performance there.
20:04:18speachyIIRC Opus is quite stack-heavy.. I think we had to bump the codec stack by 2K to keep it from crashing during even light seeking.
20:06:17 Quit IncognitoMan_ (Ping timeout: 245 seconds)
20:17:36speachy g#1932
20:17:37fs-bluebotGerrit review #1932 at http://gerrit.rockbox.org/r/1932 : opus: shrink stack usage by 640 bytes by Solomon Peachy
20:22:19speachythe patch is bitrotten, need to fix it up.
20:42:06speachyokay, it's updated. savings is nearly 700 bytes now, and I can confirm none of the structs are referenced in the hot decoding path.
20:48:51***Saving seen data "./dancer.seen"
21:00
21:00:57 Join vmx [0] (~vmx@ip5f5ac62a.dynamic.kabel-deutschland.de)
21:10:26 Quit dys (Ping timeout: 248 seconds)
21:18:15Bilgusspeachy I was in the midst of doing the same
21:18:53Bilgusbut as far as I can tell we can free the header as soon as we are done processing the header packet
21:20:23Bilgushttps://pastebin.com/TtmmzF2g
21:20:29Bilguslike so ^
21:24:34*speachy nods.
21:24:54speachyI did that original patch rather mechanically.
21:26:28speachythough it's not clear if freeing the header will do much good.
21:27:18Bilgusit has like another 600 bytes in it IIRC
21:27:43Bilgusno like 300
21:28:15Bilgusso maybe some but not much but still no point in keeping it around in the heap
21:28:26Bilgusor the stack..
21:35:35speachythat header is about 300, the stream state about 350, and the sync state about 40.
21:37:08speachybut I was thinking more about the way the ogg codec works −− unless the freed chunk is at the "end" of the heap, it can't fall back into rockbox buffers.
21:56:48 Join dys [0] (~dys@tmo-099-124.customers.d1-online.com)
22:00
22:26:26 Quit speachy (Ping timeout: 260 seconds)
22:48:53***Saving seen data "./dancer.seen"
23:00
23:14:09 Join quadjfet [0] (quadjfet@gateway/vpn/privateinternetaccess/quadjfet)
23:15:45 Quit ZincAlloy (Quit: Leaving.)
23:16:06 Quit lebellium (Quit: Leaving)
23:28:03 Quit kugel (Quit: Lost terminal)
23:29:11 Join kugel [0] (~kugel@ip5b414158.dynamic.kabel-deutschland.de)
23:29:11 Quit kugel (Changing host)
23:29:11 Join kugel [0] (~kugel@rockbox/developer/kugel)

Previous day | Next day