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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2022-09-04

00:37:10 Quit m01 (Quit: Konversation terminated.)
00:37:35***Saving seen data "./dancer.seen"
00:39:08 Join m01 [0] (~quassel@vps-b172b88b.vps.ovh.net)
00:45:29 Quit massiveH (Quit: Leaving)
01:00
01:02:56 Quit tomato (*.net *.split)
01:02:56 Quit foolsh (*.net *.split)
01:02:56 Quit Bobathan (*.net *.split)
01:02:56 Quit rb-bluebot (*.net *.split)
01:02:56 Quit danwellby (*.net *.split)
01:02:56 Quit Galois (*.net *.split)
01:02:57 Quit ddevault (*.net *.split)
01:02:57 Quit TorC (*.net *.split)
01:02:57 Quit reductum (*.net *.split)
01:02:57 Quit Malinux- (*.net *.split)
01:02:57 Quit _builtin (*.net *.split)
01:02:57 Quit Pokey (*.net *.split)
01:02:57 Quit JanC (*.net *.split)
01:02:57 Quit SammysHP (*.net *.split)
01:02:57 Quit cnx (*.net *.split)
01:02:58 Quit kugel (*.net *.split)
01:02:58 Quit skipwich (*.net *.split)
01:02:58 Quit prg (*.net *.split)
01:02:58 Quit Riviera (*.net *.split)
01:02:58 Quit funman (*.net *.split)
01:02:58 Quit Ckat (*.net *.split)
01:03:25 Join tomato [0] (~tomato@user/tomato)
01:03:25 Join foolsh [0] (~foolsh@17.sub-174-210-230.myvzw.com)
01:03:25 Join Bobathan [0] (~admin@cpe-65-29-248-157.wi.res.rr.com)
01:03:25 Join rb-bluebot [0] (~rb-bluebo@rockbox/bot/utility)
01:03:25 Join danwellby [0] (~danwellby@cpc1-cdif16-2-0-cust352.5-1.cable.virginm.net)
01:03:25 Join Galois [0] (djao@efnet.math.uwaterloo.ca)
01:03:25 Join ddevault [0] (e7b4bb6755@sourcehut/staff/ddevault)
01:03:25 Join Malinux- [0] (~malin@2001:4641:4dfa::12c:c4a7)
01:03:25 Join TorC [0] (~Tor@fsf/member/TorC)
01:03:25 Join reductum [0] (~reductum@2603-8000-b400-8764-dea6-32ff-fe16-a622.res6.spectrum.com)
01:03:25 Join _builtin [0] (~quassel@rockbox/developer/builtin)
01:03:25 Join Pokey [0] (~pokey@spikeyCactus/hoosky)
01:03:25 Join JanC [0] (~janc@user/janc)
01:03:25 Join SammysHP [0] (~SammysHP@faol.sammyshp.de)
01:03:25 Join cnx [0] (c364d5be87@2604:bf00:561:2000::3f)
01:03:25 Join kugel [0] (~kugel@ip4d164466.dynamic.kabel-deutschland.de)
01:03:25 Join skipwich [0] (~skipwich@user/skipwich)
01:03:25 Join prg [0] (~prg@2001:19f0:5:9c1:1d2f:7937:5cb0:da6f)
01:03:25 Join Riviera [0] (Riviera@user/riviera)
01:03:25 Join funman [0] (~fun@chui-pas.net)
01:03:25 Join Ckat [0] (~Ckat@xn--z7x.xn--6frz82g)
01:04:57 Quit rogeliodh (Quit: Ping timeout (120 seconds))
01:05:20 Join rogeliodh [0] (~rogeliodh@rogeliodh.dev)
01:06:33 Quit Xeha (*.net *.split)
01:06:33 Quit munkis (*.net *.split)
01:06:34 Quit bleb (*.net *.split)
01:06:34 Quit gevaerts (*.net *.split)
01:06:34 Quit hactar|ant (*.net *.split)
01:06:34 Quit advcomp2019_ (*.net *.split)
01:06:34 Quit asaba (*.net *.split)
01:07:09 Join Xeha [0] (~Xeha@user/Xeha)
01:07:09 Join munkis [0] (~mendel_mu@ool-ae2cb229.dyn.optonline.net)
01:07:09 Join bleb [0] (~cm@user/bleb)
01:07:09 Join gevaerts [0] (~fg@user/gevaerts)
01:07:09 Join hactar|ant [0] (~zem@c-24-21-103-100.hsd1.or.comcast.net)
01:07:09 Join advcomp2019_ [0] (~advcomp20@user/advcomp2019)
01:07:09 Join asaba [0] (~asabas@103.113.159.250)
01:07:16 Quit MayeulC (*.net *.split)
01:07:17 Quit bonswouar[m] (*.net *.split)
01:07:18 Quit benjaoming (*.net *.split)
01:08:08 Quit m01 (*.net *.split)
01:08:09 Quit tinybronca[m] (*.net *.split)
01:08:09 Quit popcorn9499[m] (*.net *.split)
01:08:10 Quit ats (*.net *.split)
01:08:10 Quit melmothX (*.net *.split)
01:08:10 Quit rasher (*.net *.split)
01:08:32 Quit larbob (*.net *.split)
01:08:32 Quit paulcarroty (*.net *.split)
01:08:33 Quit blbro[m] (*.net *.split)
01:08:33 Quit berber_l (*.net *.split)
01:08:34 Quit jackie (*.net *.split)
01:09:10 Quit LjL (*.net *.split)
01:09:10 Quit bluebrother (*.net *.split)
01:09:10 Quit hexadecagram (*.net *.split)
01:09:10 Quit koniu (*.net *.split)
01:09:11 Quit Nezumi-sama (*.net *.split)
01:09:11 Quit spork (*.net *.split)
01:09:11 Quit toruvinn (*.net *.split)
01:09:11 Quit michaelni (*.net *.split)
01:09:11 Quit kirvesAxe (*.net *.split)
01:09:11 Quit dbohdan3 (*.net *.split)
01:09:12 Quit lizzie (*.net *.split)
01:09:12 Quit GeekShadow (*.net *.split)
01:09:12 Quit yosafbridge (*.net *.split)
01:10:09 Quit ParkerR (*.net *.split)
01:10:09 Quit mrkrisprolls (*.net *.split)
01:10:10 Quit Retr0id (*.net *.split)
01:10:22 Quit pixelma (*.net *.split)
01:10:22 Quit amiconn (*.net *.split)
01:10:23 Quit paulk (*.net *.split)
01:10:23 Quit rudi_s (*.net *.split)
01:10:23 Quit edward (*.net *.split)
01:10:23 Quit Longclaw (*.net *.split)
01:10:24 Quit hook54321 (*.net *.split)
01:10:24 Quit Arsen (*.net *.split)
01:10:24 Quit launchd (*.net *.split)
01:10:31 Join ParkerR [0] (ParkerR@znc.withg.org)
01:10:31 Join mrkrisprolls [0] (mrkrisprol@mrkrisprolls.orange.ircnow.org)
01:10:31 Join Retr0id [0] (~Retr0id@user/retr0id)
01:10:43 Join amiconn [0] (jens@p200300ea87273400305e95fffec66ff3.dip0.t-ipconnect.de)
01:10:43 Join pixelma [0] (marianne@p200300ea87273400305e95fffec66ff3.dip0.t-ipconnect.de)
01:10:43 Join paulk [0] (~paulk@vpn-0-22.aquilenet.fr)
01:10:43 Join rudi_s [0] (~simon@user/rudi-s/x-7673890)
01:10:43 Join edward [0] (~edward@user/edward)
01:10:43 Join Longclaw [0] (bouncer@xn--n8jr8c7341a.xn--q9jyb4c)
01:10:43 Join hook54321 [0] (sid149355@user/hook54321)
01:10:43 Join Arsen [0] (arsen@managarm/dev/Arsen)
01:10:43 Join launchd [0] (launchd@bitbot/launchd)
01:10:51 Quit pixelma (*.net *.split)
01:10:51 Quit amiconn (*.net *.split)
01:10:51 Quit paulk (*.net *.split)
01:10:52 Quit rudi_s (*.net *.split)
01:10:52 Quit edward (*.net *.split)
01:10:52 Quit Longclaw (*.net *.split)
01:10:52 Quit hook54321 (*.net *.split)
01:10:52 Quit Arsen (*.net *.split)
01:10:52 Quit launchd (*.net *.split)
01:10:52 Quit ParkerR (*.net *.split)
01:10:52 Quit mrkrisprolls (*.net *.split)
01:10:52 Quit Retr0id (*.net *.split)
01:10:53 Quit Xeha (*.net *.split)
01:10:53 Quit munkis (*.net *.split)
01:10:53 Quit bleb (*.net *.split)
01:10:53 Quit gevaerts (*.net *.split)
01:10:53 Quit hactar|ant (*.net *.split)
01:10:53 Quit advcomp2019_ (*.net *.split)
01:10:53 Quit asaba (*.net *.split)
01:10:53 Quit tomato (*.net *.split)
01:10:54 Quit foolsh (*.net *.split)
01:10:54 Quit Bobathan (*.net *.split)
01:10:54 Quit rb-bluebot (*.net *.split)
01:10:54 Quit danwellby (*.net *.split)
01:10:54 Quit Galois (*.net *.split)
01:10:54 Quit ddevault (*.net *.split)
01:10:54 Quit TorC (*.net *.split)
01:10:54 Quit reductum (*.net *.split)
01:10:54 Quit Malinux- (*.net *.split)
01:10:54 Quit _builtin (*.net *.split)
01:10:54 Quit Pokey (*.net *.split)
01:10:54 Quit JanC (*.net *.split)
01:10:54 Quit SammysHP (*.net *.split)
01:10:54 Quit cnx (*.net *.split)
01:10:55 Quit kugel (*.net *.split)
01:10:55 Quit skipwich (*.net *.split)
01:10:55 Quit prg (*.net *.split)
01:10:55 Quit Riviera (*.net *.split)
01:10:55 Quit funman (*.net *.split)
01:10:55 Quit Ckat (*.net *.split)
01:10:55 Quit rogeliodh (*.net *.split)
01:10:56 Quit Piece_Maker (*.net *.split)
01:10:56 Quit woffs (*.net *.split)
01:10:56 Quit wolfshappen (*.net *.split)
01:10:56 Quit yang (*.net *.split)
01:10:56 Quit tchan (*.net *.split)
01:10:56 Quit akaWolf (*.net *.split)
01:10:56 Quit Maxdamantus (*.net *.split)
01:10:56 Quit Natch (*.net *.split)
01:10:56 Quit _bilgus (*.net *.split)
01:10:56 Quit dys (*.net *.split)
01:10:56 Quit cb (*.net *.split)
01:11:07 Join launchd [0] (launchd@bitbot/launchd)
01:11:07 Join Arsen [0] (arsen@managarm/dev/Arsen)
01:11:07 Join hook54321 [0] (sid149355@user/hook54321)
01:11:07 Join Longclaw [0] (bouncer@xn--n8jr8c7341a.xn--q9jyb4c)
01:11:07 Join edward [0] (~edward@user/edward)
01:11:07 Join rudi_s [0] (~simon@user/rudi-s/x-7673890)
01:11:07 Join paulk [0] (~paulk@vpn-0-22.aquilenet.fr)
01:11:07 Join pixelma [0] (marianne@p200300ea87273400305e95fffec66ff3.dip0.t-ipconnect.de)
01:11:07 Join amiconn [0] (jens@p200300ea87273400305e95fffec66ff3.dip0.t-ipconnect.de)
01:11:07 Join Retr0id [0] (~Retr0id@user/retr0id)
01:11:07 Join mrkrisprolls [0] (mrkrisprol@mrkrisprolls.orange.ircnow.org)
01:11:07 Join ParkerR [0] (ParkerR@znc.withg.org)
01:11:07 Join asaba [0] (~asabas@103.113.159.250)
01:11:07 Join advcomp2019_ [0] (~advcomp20@user/advcomp2019)
01:11:07 Join hactar|ant [0] (~zem@c-24-21-103-100.hsd1.or.comcast.net)
01:11:07 Join gevaerts [0] (~fg@user/gevaerts)
01:11:07 Join bleb [0] (~cm@user/bleb)
01:11:07 Join munkis [0] (~mendel_mu@ool-ae2cb229.dyn.optonline.net)
01:11:07 Join Xeha [0] (~Xeha@user/Xeha)
01:11:07 Join rogeliodh [0] (~rogeliodh@rogeliodh.dev)
01:11:07 Join Ckat [0] (~Ckat@xn--z7x.xn--6frz82g)
01:11:07 Join funman [0] (~fun@chui-pas.net)
01:11:07 Join Riviera [0] (Riviera@user/riviera)
01:11:07 Join prg [0] (~prg@2001:19f0:5:9c1:1d2f:7937:5cb0:da6f)
01:11:07 Join skipwich [0] (~skipwich@user/skipwich)
01:11:07 Join kugel [0] (~kugel@ip4d164466.dynamic.kabel-deutschland.de)
01:11:07 Join cnx [0] (c364d5be87@2604:bf00:561:2000::3f)
01:11:07 Join SammysHP [0] (~SammysHP@faol.sammyshp.de)
01:11:07 Join JanC [0] (~janc@user/janc)
01:11:07 Join Pokey [0] (~pokey@spikeyCactus/hoosky)
01:11:07 Join _builtin [0] (~quassel@rockbox/developer/builtin)
01:11:07 Join reductum [0] (~reductum@2603-8000-b400-8764-dea6-32ff-fe16-a622.res6.spectrum.com)
01:11:07 Join TorC [0] (~Tor@fsf/member/TorC)
01:11:07 Join Malinux- [0] (~malin@2001:4641:4dfa::12c:c4a7)
01:11:07 Join ddevault [0] (e7b4bb6755@sourcehut/staff/ddevault)
01:11:07 Join Galois [0] (djao@efnet.math.uwaterloo.ca)
01:11:07 Join danwellby [0] (~danwellby@cpc1-cdif16-2-0-cust352.5-1.cable.virginm.net)
01:11:07 Join rb-bluebot [0] (~rb-bluebo@rockbox/bot/utility)
01:11:07 Join Bobathan [0] (~admin@cpe-65-29-248-157.wi.res.rr.com)
01:11:07 Join foolsh [0] (~foolsh@17.sub-174-210-230.myvzw.com)
01:11:07 Join tomato [0] (~tomato@user/tomato)
01:11:07 Join cb [0] (~calvin@fctnnbsc38w-47-55-87-222.dhcp-dynamic.fibreop.nb.bellaliant.net)
01:11:07 Join dys [0] (~dys@user/dys)
01:11:07 Join _bilgus [0] (~bilgus@162.154.213.134)
01:11:07 Join Natch [0] (~natch@c-9e07225c.038-60-73746f7.bbcust.telenor.se)
01:11:07 Join Maxdamantus [0] (~Maxdamant@user/maxdamantus)
01:11:07 Join akaWolf [0] (~akaWolf@akawolf.org)
01:11:07 Join tchan [0] (~tchan@c-98-206-141-238.hsd1.il.comcast.net)
01:11:07 Join yang [0] (sid23779@fsf/member/yang)
01:11:07 Join wolfshappen [0] (~waff@irc.furworks.de)
01:11:07 Join woffs [0] (3cd46299b2@woffs.de)
01:11:07 Join Piece_Maker [0] (~eddie@cpc95736-bolt17-2-0-cust330.10-3.cable.virginm.net)
01:11:59 Join m01 [0] (~quassel@vps-b172b88b.vps.ovh.net)
01:11:59 Join ats [0] (~ats@cartman.offog.org)
01:11:59 Join melmothX [0] (~marco@amusewiki/marco)
01:11:59 Join rasher [0] (~rasher@user/rasher)
01:12:16 Join benjaoming [0] (~benjaomin@37.139.19.237)
01:12:54 Join larbob [0] (~larbob@159.65.42.191)
01:12:54 Join blbro[m] [0] (~blbrostra@2001:470:69fc:105::8f7)
01:12:54 Join berber_l [0] (~berber@v2202101107577140883.nicesrv.de)
01:12:54 Join jackie [0] (~jackie@banana-new.kilobyte22.de)
01:13:20 Join LjL [0] (~ljl@user/ljl)
01:13:20 Join bluebrother [0] (~dom@user/bluebrother)
01:13:20 Join koniu [0] (~koniu@cpc107003-dals23-2-0-cust91.20-2.cable.virginm.net)
01:13:20 Join Nezumi-sama [0] (~narf@rrcs-67-53-148-69.west.biz.rr.com)
01:13:20 Join spork [0] (topic@i121118.upc-i.chello.nl)
01:13:20 Join toruvinn [0] (~toruvinn@87-205-81-180.adsl.inetia.pl)
01:13:20 Join michaelni [0] (~michael@213-47-68-29.cable.dynamic.surfer.at)
01:13:20 Join kirvesAxe [0] (kirvesaxe@user/kirvesaxe)
01:13:20 Join dbohdan3 [0] (~dbohdan@user/dbohdan)
01:13:20 Join lizzie [0] (~capitalth@meowface.org)
01:13:20 Join GeekShadow [0] (~antoine@82-64-164-139.subs.proxad.net)
01:13:20 Join yosafbridge [0] (~yosafbrid@static.38.6.217.95.clients.your-server.de)
01:13:21 Quit rogeliodh (*.net *.split)
01:13:21 Quit Piece_Maker (*.net *.split)
01:13:21 Quit woffs (*.net *.split)
01:13:22 Quit wolfshappen (*.net *.split)
01:13:22 Quit yang (*.net *.split)
01:13:22 Quit tchan (*.net *.split)
01:13:22 Quit akaWolf (*.net *.split)
01:13:22 Quit Maxdamantus (*.net *.split)
01:13:22 Quit Natch (*.net *.split)
01:13:22 Quit _bilgus (*.net *.split)
01:13:22 Quit dys (*.net *.split)
01:13:22 Quit cb (*.net *.split)
01:13:24 Join LjL^ [0] (~ljl@user/ljl)
01:13:30 Quit LjL (Max SendQ exceeded)
01:13:54 Quit hook54321 (Ping timeout: 264 seconds)
01:14:05 Join rogeliodh [0] (~rogeliodh@rogeliodh.dev)
01:14:05 Join Piece_Maker [0] (~eddie@cpc95736-bolt17-2-0-cust330.10-3.cable.virginm.net)
01:14:05 Join woffs [0] (3cd46299b2@woffs.de)
01:14:05 Join wolfshappen [0] (~waff@irc.furworks.de)
01:14:05 Join yang [0] (sid23779@fsf/member/yang)
01:14:05 Join tchan [0] (~tchan@c-98-206-141-238.hsd1.il.comcast.net)
01:14:05 Join akaWolf [0] (~akaWolf@akawolf.org)
01:14:05 Join Maxdamantus [0] (~Maxdamant@user/maxdamantus)
01:14:05 Join Natch [0] (~natch@c-9e07225c.038-60-73746f7.bbcust.telenor.se)
01:14:05 Join _bilgus [0] (~bilgus@162.154.213.134)
01:14:05 Join dys [0] (~dys@user/dys)
01:14:05 Join cb [0] (~calvin@fctnnbsc38w-47-55-87-222.dhcp-dynamic.fibreop.nb.bellaliant.net)
01:14:13 Quit blbro[m] (Ping timeout: 268 seconds)
01:17:00 Nick LjL^ is now known as LjL (~ljl@user/ljl)
01:17:04 Join hook54321 [0] (sid149355@user/hook54321)
01:19:04 Quit e-user[m] (Ping timeout: 268 seconds)
01:19:51 Join hexadecagram [0] (~acc@user/hexadecagram)
01:28:33 Join bonswouar[m] [0] (~bonswouar@2001:470:69fc:105::2:6bc9)
01:35:54 Join blbro[m] [0] (~blbrostra@2001:470:69fc:105::8f7)
01:56:35 Join popcorn9499[m] [0] (~popcorn94@2001:470:69fc:105::2:11cc)
01:56:48 Join paulcarroty [0] (~paulcarro@2001:470:69fc:105::216)
02:00
02:19:00 Join MayeulC [0] (~mayeulc@2001:470:69fc:105::35e)
02:32:45 Join lebellium [0] (~lebellium@2a01cb040109a60040cd08c0f11c1538.ipv6.abo.wanadoo.fr)
02:37:38***Saving seen data "./dancer.seen"
03:00
03:21:51 Join e-user[m] [0] (~e-usermat@2001:470:69fc:105::1:2763)
04:00
04:00:55 Join tinybronca[m] [0] (~tinybronc@2001:470:69fc:105::2:1af6)
04:37:40***No seen item changed, no save performed.
05:00
05:39:14 Quit Galois (Ping timeout: 244 seconds)
06:00
06:37:43***Saving seen data "./dancer.seen"
07:00
07:12:22 Join megamaced [0] (~megamaced@static-90-255-226-34.vodafonexdsl.co.uk)
07:35:23 Join Galois [0] (djao@efnet.math.uwaterloo.ca)
08:00
08:10:25 Quit Galois (Ping timeout: 252 seconds)
08:37:44***Saving seen data "./dancer.seen"
10:00
10:07:21 Join Galois [0] (djao@efnet.math.uwaterloo.ca)
10:37:48***No seen item changed, no save performed.
11:00
11:20:45 Quit foolsh (Ping timeout: 244 seconds)
11:33:02 Join foolsh [0] (~foolsh@17.sub-174-210-230.myvzw.com)
12:00
12:20:59 Quit megamaced (Quit: megamaced)
12:28:21 Join smithjd [0] (~enderw@node-1w7jra28y73i5sm7fzgdtjvwy.ipv6.telus.net)
12:34:50 Join SonoSooS [0] (~SonoSooS@91-82-3-129.pool.digikabel.hu)
12:37:40 Join dconrad [0] (~dconrad@152.117.104.235)
12:37:49***Saving seen data "./dancer.seen"
12:38:22SonoSooSgood (UGT) morning
12:40:11SonoSooSI think I have reached a point where I should probably ask some technical stuff, as I might be digging something someone has already solved, and I have ran out of ways to try to figure out the problems I'm facing myself (the IRC logs have been help to fix some of my issues, but most of my complaints I can't solve myself)
12:42:16SonoSooSI have already tried basically everything: enabling caches (I tested them confirmed working), rearranging memory into logical blocks, mapping basically the entirety of the RAM chip to Rockbox, arranging memory mapping to optimize for RAS/CAS latency minimization, and I have fixed my timer code, and I'm still getting terrible performance, and
12:42:17SonoSooSplayback takes on average half a minute to start, and I have ran out of ideas what else to try
12:45:22SonoSooS(oops, forgot to state the context: I'm porting Rockbox to a new device, and it works(tm), but performance is rather sub-optimal for the hardware it's running on, to put it mildly)
12:47:40dconradwhat's the player and chip? Clock speed?
12:48:51SonoSooSit's custom hardware (Nintendo 3DS), ARM11 MPCore (2 or 4 cores, although I did not implement Rockbox's coprocessor support yet), 268111856Hz, overclockable to 804335568Hz on newer modells
12:49:33dconradoh, is this running as an application on the 3DS's OS?
12:50:26SonoSooSnope, that's too difficult, and the threading modell is most likely incompatible with Rockbox's hosting mode (it's bad even on purpose-made native applications, so I have no hopes of getting Rockbox to work in host mode)
12:50:34SonoSooSI'm doing a native port
12:50:52dconraddang
12:51:33dconradthat's impressive, even if it's not there yet
12:51:59SonoSooSI mean, it does work, with themes and stuff, it's just... unbearably slow
12:52:51SonoSooSI even implemented things which I'll disable in the final build (namely the remote display is one thing I'll disable, as it gives me a lot of compiler hurdles, and too much runtime overhead)
12:52:56dconradthere should be more than enough processor power, considering rb runs very well on what, 30 mhz ipods? And I imagine the 3DS has more than enough ram as well, >2-4 mb?
12:53:22SonoSooS128M on old modells, 256Megs on overclockable modells
12:53:36SonoSooSI have mapped the entirety of the 128Megs to Rockbox in some way
12:54:39SonoSooS(the MMU had to be enabled for cache support and due to hardware bugs, so I just mapped the memory at that point)
12:55:23dconradI can't claim to be anything close to an expert, but my two gut reactions are not effectively using any on-chip "fast ram" type of thing, or maybe some sort of compiler de-optimization issues?
12:55:26 Quit smithjd (Ping timeout: 260 seconds)
12:55:31dconradwild guesses
12:55:36hactar|antoh hey SonoSooS
12:55:45hactar|antglad to see you're still at it
12:55:47SonoSooSthere is also 512k of fast RAM (AXIWRAM), but I can't fit anything into it (it's definitely be not being that much familiar with Rockbox, I'm sure other players with less IRAM managed to fit stuff there just fine)
12:55:49SonoSooShi
12:56:08SonoSooSyeah, I just took a break to try to reflect on why it could be broken, but could not come up with anything
12:57:12SonoSooSwell, I know that FCRAM is awfully slow, but I'm using caches, and using very accurate(tm) performance-measuring techniques of plotting CPU usage on the LCD and eyeballing it, there seems to be barely any CPU usage
12:57:34SonoSooS(memory stall also counts as CPU usage)
12:58:32SonoSooSI also did disassembly a lot on the binary due to me not being able to code properly for some reason (typos which compile successfully are my worst enemy!), and I did not see any signs of bad compiler optimization flags
12:58:44dconradis non-memory based stuff responsive, like scrolling through a menu or something? I don't know if we have any good ways of benchmarking the cpu without involving ram
12:59:10SonoSooSwell DOOM also runs at bad FPS, but that could be something else
12:59:29SonoSooSmenu navigation performance seems to change if I make big edits, but in the current build it's pretty good
12:59:55SonoSooSI think it would be even faster if I have killed the PCM driver
13:00
13:00:41SonoSooSit also seems to be more than fast enough to decode audio in real time, as after the ~30second wait period, it plays audio with absolutely no issues
13:01:23SonoSooSCPU could be benchmarked from cached instruction memory
13:02:05SonoSooS(ARM11 has ways to make sure that the code you're running is 100% in cache, though it's quite tedious to do)
13:02:45dconradis it possible that disk access is just crazy slow for some reason?
13:03:34SonoSooSI mean, the SD controller is total garbage, but even then it's 100% async (handled by ARM9, the "security coprocessor"), and can reach at least 200-500kb/s
13:04:56SonoSooSI mean, it's async, but the driver-side is not (now that I mention it, I did not really look into if there is even support for async disk IO, time to put that on my TODO list)
13:06:10SonoSooSit doesn't seem to be a disk issue either, as skipping between files takes like half a second, approximately
13:06:47dconradthat's definitely strange
13:07:50dconradlike the audio buffer takes forever to allocate, but once it's allocated it performs well?
13:07:59SonoSooSyup
13:08:27SonoSooSonce playback starts, the buffer stays basically >95% all the time
13:08:27dconradwild
13:09:17SonoSooSpcmbufdesc doesn't go below 61/64
13:10:05SonoSooSand pcm doesn't go below around 49xxxx/524288
13:10:58SonoSooSat least with my friend's driver it finally no longer panics
13:11:14SonoSooS(the PCM hardware doesn't have interrupts, so I schedule it via HSync)
13:12:07SonoSooSwhile it consumes a lot of CPU as a result, it still doesn't seem to put a dent into the total CPU usage
13:12:49SonoSooSbad phrasing, it consumes most of the CPU, but even then that's not a lot of CPU time
13:13:28dconradI presume all the allocation stuff is handled by buflib.c and core_alloc.c? I don't know how much of that memory management stuff is target-specific
13:14:04SonoSooSit's all Rockbox, I did nothing custom
13:14:45SonoSooSI do no dynamic allocation, as the function stated in the wiki to do dynamic allocation, I can't even find its existance of that function
13:15:07SonoSooSbut I did without dynamic allocation anyways, so it's all in .bss
13:15:11dconraddoesn't surprise me that the wiki would be out of date
13:15:43SonoSooSI would say I would update it, but it's a weakness of mine
13:15:51SonoSooScurse of the "Real Programmer"
13:16:10dconradhaha the code IS the documentation!
13:16:24SonoSooSyeah, you just gotta find it
13:16:45SonoSooSI did learn that most of the times it's in the header files, sometimes even with nice text graphics
13:19:24SonoSooShuh, this is very odd
13:20:57SonoSooSI started spamming L and R buttons during file loading, and that somehow made the initial starting faster
13:21:33dconradis it repeatable, or just wild coincidence?
13:22:15SonoSooSyes and no
13:22:26SonoSooSit's somewhat faster
13:22:29SonoSooSnot by a lot
13:24:44SonoSooSit's probably placebo
13:25:23SonoSooSyeah, sorry, it was placebo
13:25:44dconradaww, I was starting to think maybe something to do with the cpu erroneously going to sleep or something
13:25:54SonoSooSyeah, I don't even have HID IRQ enabled
13:27:15SonoSooSit should panic if I do
13:27:46SonoSooSoh, I forgot I implemented the backlight driver
13:28:05SonoSooSI'm not used to backlight fade on this hardware at all
13:28:17SonoSooSit looks really good
13:30:12SonoSooShmm, I just had a think... how much RAM does the average Rockbox-compatible player have? I just realized that giving GTA4 too many CPU cores will make it run bad, so perhaps Rockbox is not intended for super amazing specs hardware?
13:31:35SonoSooSoh wait, I can answer the average RAM, it should be in the linker script
13:31:51dconradI think the most ram-heavy targets have 64 MB, but the bare minimum is ~4 MB, 2 MB if you start cutting features to make it fit, I think
13:46:32hactar|antyeah the later generation ipods have 64mb i believe
13:46:50hactar|antstarting with the 60/80gb 5th gens, bumped from 32 to better accomodate the larger potential library size
13:59:24SonoSooSyeah, I think I'll fiddle around with RAM sizes, and also fix the linker script, now that I understand its syntax more
14:00
14:04:34SonoSooSoh, and as a result of fixing the linker script, I'll try to make IRAM work, just in case it gives any performance increase over cached RAM
14:08:23 Join lebellium_ [0] (~lebellium@2a01cb040109a600d1935bb671971f56.ipv6.abo.wanadoo.fr)
14:09:20 Join hombrelaser [0] (~my@user/hombrelaser)
14:10:38 Quit lebellium (Ping timeout: 240 seconds)
14:37:51***Saving seen data "./dancer.seen"
14:39:06 Quit SonoSooS (Ping timeout: 264 seconds)
14:42:15dconradSonoSooS: one last thought (and maybe you've already thought this through), if the audio buffer is getting excessively large, it may be buffering the entire audio file before starting playback - maybe you want to look at the code that starts playback? I think start_audio_playback() might be a good place to start, and go up and down the call tree
14:44:24dconradcould just be taking a long time to read that much from disk into ram
15:00
15:36:04 Join smithjd [0] (~enderw@node-1w7jra28y73i5sm7fzgdtjvwy.ipv6.telus.net)
16:00
16:07:21 Quit smithjd (Ping timeout: 260 seconds)
16:30:34 Join smithjd [0] (~enderw@node-1w7jra28y73i5sm7fzgdtjvwy.ipv6.telus.net)
16:37:53***Saving seen data "./dancer.seen"
17:00
17:03:21 Quit smithjd (Ping timeout: 260 seconds)
17:18:27 Quit lebellium_ (Quit: Leaving)
17:33:16 Join smithjd [0] (~enderw@node-1w7jra28y73i5sm7fzgdtjvwy.ipv6.telus.net)
17:48:51 Quit smithjd (Ping timeout: 260 seconds)
18:00
18:22:43 Quit dconrad ()
18:37:56***Saving seen data "./dancer.seen"
18:57:42 Join massiveH [0] (~massiveH@2001:470:1f07:76a:d83f:325:52fc:d33c)
20:00
20:02:48 Join smithjd [0] (~enderw@d142-179-244-211.abhsia.telus.net)
20:38:00***No seen item changed, no save performed.
21:00
21:01:00 Quit tomato (Quit: The Lounge - https://thelounge.chat)
21:25:23 Quit smithjd (Ping timeout: 248 seconds)
21:32:53 Join tomato [0] (~tomato@user/tomato)
22:00
22:38:01***Saving seen data "./dancer.seen"
22:40:25 Join dconrad [0] (~dconrad@152.117.104.235)
23:00
23:21:04 Quit massiveH (Quit: Leaving)
23:54:40 Join smithjd [0] (~enderw@node-1w7jra28y73i5sm7fzgdtjvwy.ipv6.telus.net)

Previous day | Next day