| 00:26:32 | *** | Saving seen data "./dancer.seen" |
| 02:00 |
| 02:21:10 | | Join livvy [0] (~livvy@gateway/tor-sasl/livvy) |
| 02:26:36 | *** | No seen item changed, no save performed. |
| 03:00 |
| 03:03:18 | | Join dys [0] (~dys@116.203.150.34) |
| 03:39:50 | | Quit __Bilgus_ (Remote host closed the connection) |
| 04:00 |
| 04:26:38 | *** | Saving seen data "./dancer.seen" |
| 04:40:46 | | Join pamaury [0] (~pamaury@rockbox/developer/pamaury) |
| 06:00 |
| 06:06:28 | | Join kugel_ [0] (~kugel@ip5b40dc1f.dynamic.kabel-deutschland.de) |
| 06:06:36 | | Quit kugel_ (Changing host) |
| 06:06:37 | | Join kugel_ [0] (~kugel@rockbox/developer/kugel) |
| 06:06:59 | | Quit kugel (Ping timeout: 240 seconds) |
| 06:26:40 | *** | Saving seen data "./dancer.seen" |
| 06:32:25 | | Join petur [0] (~petur@77.77.179.66) |
| 06:32:26 | | Quit petur (Changing host) |
| 06:32:26 | | Join petur [0] (~petur@rockbox/developer/petur) |
| 06:47:18 | | Quit livvy (Remote host closed the connection) |
| 06:47:29 | | Join livvy [0] (~livvy@gateway/tor-sasl/livvy) |
| 08:00 |
| 08:01:38 | | Join ac_laptop [0] (~ac_laptop@186.2.247.129) |
| 08:26:44 | *** | Saving seen data "./dancer.seen" |
| 08:28:59 | | Quit Natch (Ping timeout: 240 seconds) |
| 08:30:28 | | Join Natch [0] (~Natch@c-b471e255.014-297-73746f25.bbcust.telenor.se) |
| 08:53:42 | | Quit ac_laptop (Ping timeout: 244 seconds) |
| 09:00 |
| 09:16:38 | Strife89 | https://photos.app.goo.gl/Yp2yEthTSii86SVy9 A look at some of the bugginess in the m200v4 port. |
| 09:21:47 | | Quit Acou_Bass (Ping timeout: 244 seconds) |
| 09:23:46 | speachy | so the display driver is buggy as hell, looks like a good starting point. :) |
| 09:26:04 | | Join Acou_Bass [0] (~eddie@cpc97736-bolt17-2-0-cust152.10-3.cable.virginm.net) |
| 09:28:49 | fs-bluebot | Build Server message: New build round started. Revision 5d1d431, 280 builds, 9 clients. |
| 09:29:31 | | Quit petur (Quit: Connection reset by beer) |
| 09:43:57 | fs-bluebot | Build Server message: Build round completed after 908 seconds. |
| 09:44:01 | fs-bluebot | Build Server message: Revision 5d1d431 result: All green |
| 10:00 |
| 10:07:03 | | Quit livvy (Ping timeout: 240 seconds) |
| 10:07:39 | | Join livvy [0] (~livvy@gateway/tor-sasl/livvy) |
| 10:18:42 | | Join __Bilgus_ [0] (41ba23be@65.186.35.190) |
| 10:19:40 | __Bilgus_ | the snake bug is probably just incorrect keymap |
| 10:21:11 | __Bilgus_ | display bugs OTOH might just be plain wrong or it could also be something to do with the lcd_write commands |
| 10:26:45 | *** | Saving seen data "./dancer.seen" |
| 10:28:03 | | Join sakax [0] (~r0b0t@unaffiliated/r0b0t) |
| 10:29:37 | | Join ac_laptop [0] (~ac_laptop@186.2.247.129) |
| 10:53:53 | __Bilgus_ | I wonder if the m200 and the m200v4 actually have the same lcd () |
| 10:54:14 | __Bilgus_ | Strife89 in the debug menu does it give the dbop clock? |
| 10:58:50 | Strife89 | __Bilgus_: Yep, 62 MHz SET, 31 MHz ACTUAL. |
| 10:59:28 | __Bilgus_ | are you able to compile? |
| 10:59:34 | Strife89 | YEs |
| 11:00 |
| 11:02:03 | Strife89 | Here's a photo of that whole page from the Debug screen https://usercontent.irccloud-cdn.com/file/74FRVPD7/IMG_20200806_105743.jpg |
| 11:05:30 | __Bilgus_ | in firmware/export/config/sansam200v4.h comment out HAVE_ADJUSTABLE_CPU_FREQ |
| 11:09:44 | __Bilgus_ | after that revisit your bug testing if its still there we'll look at the dbop clock, if its too fast or too slow weird stuff happens |
| 11:10:23 | __Bilgus_ | more so too fast but too slow makes it act really funny too |
| 11:17:46 | __Bilgus_ | after that if it still glitches go to firmware/target/arm/as3525/clock-target.h and mess around with the clocks |
| 11:17:57 | __Bilgus_ | AS3525_DBOP_FREQ try dividing by 2 |
| 11:18:57 | Strife89 | Promising start, the WPS isn't glitching yet |
| 11:20:43 | __Bilgus_ | it could just be the voltage is set too low when it adjust the cpu freq |
| 11:22:02 | __Bilgus_ | are you by chance using a rechargable battery? I'd think the voltage regulator would take care of it but who knows.. |
| 11:22:20 | Strife89 | Nah, I've got a regular AAA in there |
| 11:24:10 | Strife89 | Which just died :/ |
| 11:25:12 | Strife89 | Popped in a new one, Rockbox is behaving right as rain |
| 11:26:55 | __Bilgus_ | hmm so we should just be able to up the cpu voltage |
| 11:27:08 | __Bilgus_ | give me a minute let me find that |
| 11:27:53 | speachy | one of these days I'll try to take another crack at the jz47xx reclocking |
| 11:28:06 | Strife89 | And the LCD is - so far - consistently correct when Rockbox starts. |
| 11:29:24 | Strife89 | Rockbox does still shut down if I raise the volume high enough, though |
| 11:29:46 | Strife89 | Said volume doesn't sound very loud, though, even at +-0 |
| 11:30:46 | | Quit heredoc (Ping timeout: 244 seconds) |
| 11:31:33 | Strife89 | I shouldn't say that Rockbox shuts down; the player itself suddenly shuts off |
| 11:32:36 | Strife89 | I know my unit can do better in the volume department (I listened to it's radio for a bit in the OF, before I installed Rockbox) |
| 11:33:43 | Strife89 | __Bilgus_: So far I've only tested with #HAVE_ADJUSTABLE_CPU_FREQ commented out; haven't messed with the clocks |
| 11:37:29 | __Bilgus_ | its the cpu voltage i'm pretty sure |
| 11:38:07 | __Bilgus_ | ill paste bin you a change to try once I figure out the define for the target |
| 11:42:00 | Strife89 | Battery life is a bit concerning, although I don't have a baseline to compare to. |
| 11:44:03 | Strife89 | ~ -0.028 per minute with music playing (MP3 160k) |
| 11:45:45 | Strife89 | Oh, that was at peak anyway. Guess that was when buffering. |
| 11:46:06 | Strife89 | I'll just do a battery bench once I switch to another fresh batt. |
| 11:50:39 | __Bilgus_ | Strife89 https://pastebin.com/pak4exTu |
| 11:50:56 | __Bilgus_ | also put back the have adjustable cpu voltage |
| 11:50:58 | Strife89 | As a reference, -10dB on the m200v4 sounds roughly equivalent to -50dB on an iPod Color |
| 11:51:39 | __Bilgus_ | that could easily be a misconfigured ground |
| 11:53:07 | __Bilgus_ | or even bad impedance matching.. |
| 11:57:44 | | Join salty-horse [0] (~salty-hor@unaffiliated/salty-horse) |
| 11:58:26 | salty-horse | with xduoo x3ii, when starting playback or seeking, I get a short high pitch distortion sound. known issue? |
| 11:58:54 | __Bilgus_ | Strife89 https://github.com/Rockbox/rockbox/blob/master/firmware/drivers/audio/as3514.c#L157 |
| 11:59:06 | __Bilgus_ | try adding M200V4 here as well |
| 12:00 |
| 12:00:37 | __Bilgus_ | other things probably need to change here as well but see if it increases your volume first' |
| 12:04:47 | __Bilgus_ | I'll need to look at the data sheet for the AS3514/AS3543 before I can tell you what else to try but I gtg till this eve |
| 12:06:50 | salty-horse | should the build process for the xduoo have generated the cabbie theme files? I have a cabbiev2.cfg but no bitmaps. Been years since I built for a new target so I'm probably missing a step |
| 12:10:59 | __Bilgus_ | bah I don't have the 3514 datasheet on this PC |
| 12:11:36 | __Bilgus_ | https://www.y-ic.it/datasheet/04/AS3514.pdf |
| 12:16:57 | __Bilgus_ | oh it also has a settable gain, which might be the issue but low battery life pushes me towards something being misconfigured |
| 12:18:26 | Strife89 | I can't actually confirm that it has "low battery life" |
| 12:18:40 | Strife89 | But it did go from 100% to 82% in 20 minutes |
| 12:18:58 | __Bilgus_ | a single AAA you say? |
| 12:19:02 | Strife89 | Yes |
| 12:19:23 | Strife89 | A good Duracell, not a cheapo |
| 12:21:02 | speachy | I think NiMH rechargeables have more usable power than an Alkakine. more instantaneous current too |
| 12:21:42 | __Bilgus_ | pretty sure aaa have around 1000mah' |
| 12:22:13 | __Bilgus_ | but thats also going through a boost convertor |
| 12:22:35 | __Bilgus_ | at best 80% efficiency so already at 800 mah probably |
| 12:22:36 | speachy | I'm sure the m200 uses a switching buck/boost supply so the lower max voltage won't matter |
| 12:23:41 | Strife89 | The "set batt capacity" option has 1500mAh as it's *lowest* option for some reason |
| 12:23:47 | Strife89 | Goes up to 3000 |
| 12:24:12 | Strife89 | I know that's mainly for runtime calculation, but hey |
| 12:25:18 | __Bilgus_ | its been a long while since I went through the AS3514 stuff but I remember there are a bunch settings for the power regulators and weird things that happen with the audio driver when settings are mis-configured like higher power draw |
| 12:26:49 | *** | Saving seen data "./dancer.seen" |
| 12:26:59 | Strife89 | I had to step AFK earlier, finally made the changes. Compiling now. |
| 12:28:19 | __Bilgus_ | let me know how it turns out once I go through the datasheet we can try a few things but if it works get a battery bench pref. the same pack of batteries and we can tweak a few things and see how it fares |
| 12:30:03 | Strife89 | I'm working through at 16-pack of the same batteries, we should be good there. :) |
| 12:30:14 | Strife89 | s/at/a/ |
| 12:30:48 | Strife89 | At work ATM but there's 12 of 'em at home |
| 12:32:25 | Strife89 | I should buy some more soon, though, I also use them for my older Palms |
| 12:32:36 | Strife89 | and a Game Boy Pocket |
| 12:33:56 | __Bilgus_ | its amazing how many batteries were needed from 1980 - 2004 |
| 12:35:40 | Strife89 | Welp, I disconnected the player, had RoLo reload, and the LCD immediately went dark (same as it did in the video, with text barely readable) |
| 12:36:22 | salty-horse | speachy, may I bother you with some xduoo questions? after building rockbox, I'm missing the cabbie v2 theme. Is there a build step I forgot? I'm also getting some distortion (as noted above) |
| 12:36:32 | Strife89 | Second boot the same, third boot correct contrast |
| 12:37:00 | Strife89 | WPS immediately made the screen blank |
| 12:37:07 | Strife89 | Going AFK for a bit |
| 12:37:12 | __Bilgus_ | same |
| 12:37:13 | speachy | salty-horse: I don't actually have an x3ii. and afaik you're the first person to report any such issue |
| 12:37:51 | | Join MrZeus [0] (~MrZeus@4e6942be.skybroadband.com) |
| 12:38:03 | speachy | but the default wps theme is built in, yeah. |
| 12:38:16 | salty-horse | speachy, ah. you edited the wiki page :D. should I open a bug about the distortion? |
| 12:38:43 | speachy | sure, I just don't have any way to investigate, much less fix, the problem. |
| 12:38:48 | salty-horse | speachy, built-in means I didn't miss a step...? by default I got the old "theme" without graphics, since the bitmaps were missing |
| 12:39:15 | salty-horse | speachy, are there people involved with the port I can talk to? |
| 12:41:52 | speachy | did you install this thing from rockbox.org files, or did you start with the original xvortex port? |
| 12:46:30 | salty-horse | only the rockbox repository |
| 12:46:48 | salty-horse | I also have a keymapping suggestion |
| 12:59:06 | | Join heredoc [0] (heredoc@2a01:7e01::f03c:91ff:fec1:de1d) |
| 13:00 |
| 13:14:01 | Strife89 | Any iRiver fans in here? I just snagged an H320. https://www.ebay.com/itm/124234803070 |
| 13:14:53 | Strife89 | The seller wrote, "One thing I did notice is that when it is turned on it makes a ticking noise that is noticeable," which concerns me a little. Wondering if it's a known problem among H320 users in here. |
| 13:15:09 | speachy | heh, I saw that listing |
| 13:15:15 | Strife89 | (Sorry, I know this isn't a Rockbox question.) |
| 13:15:29 | Strife89 | speachy: They offered me half off :D |
| 13:15:48 | Strife89 | Well, s/me/all watchers/ |
| 13:16:10 | Strife89 | But obviously I was the first to jump |
| 13:17:04 | speachy | salty-horse: don't know what to say; beyond file bugs with as much detail as possible. and confirm it's recreatable with the builds you can download as opposed to those you make yourself.. |
| 13:18:11 | salty-horse | speachy, what can go wrong with the builds I make myself if I make no code changes? |
| 13:19:06 | speachy | salty-horse: it's possible there's something wrong with your environment or some lingering config/files on the player itself. |
| 13:19:44 | speachy | pristine install from a supposedly-good source will at least confirm or eliminate tha possibility |
| 13:20:16 | speachy | and thus affect the decision tree |
| 13:21:00 | salty-horse | ok, will test with official build before reporting. for now I'm making a patch for improved button mapping |
| 13:21:41 | speachy | the keymaps were largely copied from the original X3, as the xvortex port sources we started with didn't include any. |
| 13:21:58 | speachy | so patches will be (very) welcome |
| 13:24:33 | salty-horse | now I'm copying button mapping for the power key from e200 :D Sad to have a merged play/select on the xduoo but ¯\_(ツ)_/¯ |
| 13:25:36 | speachy | pamaury: I'm thinking of trying to dust off your not-touched-in-a-decade USB MTP patches. Any thoughts, beyond perhaps that I should do something more useful with my life? |
| 13:27:48 | salty-horse | I see that the official xduoo daily build is also missing the cabbiev2 bitmaps. the daily build for sansa e200 has it - is that a problem I should report? |
| 13:37:07 | speachy | hmm. give me a moment |
| 13:45:00 | fs-bluebot | Build Server message: New build round started. Revision d51dfbf, 280 builds, 10 clients. |
| 13:45:15 | speachy | salty-horse: this build should have the missing bitmaps/etc |
| 13:45:31 | speachy | do a git pull/build yourseif if you don't want to wait |
| 13:50:45 | bluebrother | Strife89: I remember a ticking noise being a problem on the h100 / h300 players when a remote is attached. Can't remember too much details about it though, it's been really long and I never noticed that on my h120. |
| 13:51:50 | salty-horse | speachy, thanks |
| 13:53:53 | Strife89 | bluebrother: Interesting to know at least, thanks. |
| 13:54:46 | bluebrother | I guess the logs should have something on that. Like ... 10 years ago :) |
| 13:55:26 | Strife89 | I thought I would go ahead an do an informal Battery Bench since I'm going to be busy ... but the m250 didn't like that. https://usercontent.irccloud-cdn.com/file/OMaefcCL/IMG_20200806_135245.jpg |
| 13:57:30 | Strife89 | I think it was about to play the next track when the Panic occurred. |
| 13:58:19 | fs-bluebot | Build Server message: Build round completed after 799 seconds. |
| 13:58:22 | fs-bluebot | Build Server message: Revision d51dfbf result: All green |
| 13:59:31 | Strife89 | salty-horse: ^ |
| 13:59:46 | salty-horse | thanks |
| 14:00 |
| 14:00:57 | salty-horse | question about button mapping. when a button code says BUTTON_POWER|BUTTON_REL I understand it's when the power button is short-pressed. what does this mean? BUTTON_POWER|BUTTON_RIGHT |
| 14:01:59 | Strife89 | I think that might simply be a two-button code? |
| 14:02:12 | Strife89 | Not a dev at all here. |
| 14:02:51 | Strife89 | But I know that some targets have features where two buttons must be pressed simultaneously to trigger. |
| 14:19:44 | | Quit Natch (Remote host closed the connection) |
| 14:26:50 | *** | Saving seen data "./dancer.seen" |
| 14:29:44 | bluebrother | speachy: why did you add bzip2 to Rockbox Utility instead of simply linking the system one? |
| 14:30:37 | speachy | my understanding was that rbutil was to be a depdendency-free as possible, including external libraries |
| 14:31:31 | bluebrother | ok, fair point :) |
| 14:31:37 | speachy | / |
| 14:32:02 | bluebrother | it breaks when building for Windows, since Qt seems to export those symbols too. A similar issue exists with zlib. |
| 14:32:19 | bluebrother | I'm going to not link bzip2 on Windows for now, so at least that will build again. |
| 14:32:43 | speachy | and sorting this stuff out is why you're paid the big bucks! |
| 14:33:03 | bluebrother | I'm getting paid? o.O |
| 14:33:12 | * | bluebrother hasn't seen any money :D |
| 14:33:17 | speachy | with exposure! |
| 14:33:39 | bluebrother | oh noes! |
| 14:33:41 | * | bluebrother hides |
| 14:34:53 | bluebrother | ok, I'm going to drop Qt4 support soonish. Plus some changes that will kill warnings with current Qt (but that won't work on Qt4 anymore) |
| 14:38:16 | | Join Natch [0] (~Natch@c-b471e255.014-297-73746f25.bbcust.telenor.se) |
| 14:38:54 | Strife89 | bluebrother: All the money you're paid gets sucked right back into the server costs, black hole style :P |
| 14:39:24 | bluebrother | strange. I'm not paying for those servers. |
| 14:39:40 | * | bluebrother has the feeling something is off here. And he's getting ripped off. :P |
| 14:42:34 | speachy | drat, you caught us. you're the only one not getting showered in gold |
| 14:43:46 | fs-bluebot | Build Server message: New build round started. Revision 7c5b9a4, 280 builds, 10 clients. |
| 14:46:10 | | Join petur [0] (~petur@rockbox/developer/petur) |
| 14:56:02 | bluebrother | ok, so I uploaded a Windows development binary of Rockbox Utility for all those people whining on the forums about AGPtek. |
| 14:58:04 | fs-bluebot | Build Server message: Build round completed after 859 seconds. |
| 14:58:07 | fs-bluebot | Build Server message: Revision 7c5b9a4 result: All green |
| 15:00 |
| 15:01:18 | speachy | bluebrother, I'd been meaning to ask about FS #12947 |
| 15:01:19 | fs-bluebot | http://www.rockbox.org/tracker/task/12947 Rockbox Utility manual version and installed program version can fail to accord (bugs, unconfirmed) |
| 15:02:05 | speachy | rbutil always grabs the latest nightly manual build, even when one installs a release package |
| 15:07:10 | bluebrother | needs fixing I guess :) |
| 15:07:53 | bluebrother | though I don't consider that too important −− those build issues actually are a major problem to me. |
| 15:09:34 | speachy | yeah; we generally want folks to use newer manuals, but with the archos retirements, there are no longer daily manuals for them. |
| 15:23:45 | | Quit J_Darnley (Ping timeout: 240 seconds) |
| 15:26:37 | | Join J_Darnley [0] (~J_Darnley@d51A44418.access.telenet.be) |
| 16:00 |
| 16:00:31 | | Join Rower [0] (~Rower@78-73-72-39-no2340.tbcn.telia.com) |
| 16:04:40 | | Quit Rower (Ping timeout: 246 seconds) |
| 16:16:11 | pamaury | speachy: re-MTP: possibly you should do something better with your life because as I recall it was annoying to understand what needs to be done to make Windows happy. But some people might find it useful and it's better than rotting on gerrit |
| 16:22:25 | | Quit salty-horse (Quit: Leaving) |
| 16:26:52 | *** | Saving seen data "./dancer.seen" |
| 16:38:22 | | Quit ac_laptop (Ping timeout: 265 seconds) |
| 16:54:19 | | Quit livvy (Quit: Network issue or system upgrade probably) |
| 17:00 |
| 17:01:10 | | Quit J_Darnley (Ping timeout: 260 seconds) |
| 17:04:29 | | Join J_Darnley [0] (~J_Darnley@d51A44418.access.telenet.be) |
| 17:05:15 | fs-bluebot | Build Server message: New build round started. Revision 04d289f, 280 builds, 10 clients. |
| 17:17:01 | fs-bluebot | Build Server message: Build round completed after 706 seconds. |
| 17:17:03 | fs-bluebot | Build Server message: Revision 04d289f result: All green |
| 17:18:52 | Strife89 | Whoops, I started to render video during that build round |
| 17:21:09 | Strife89 | I heard the fans making noise but I managed to convince myself that it was just a strange echo of the air conditioner sound |
| 17:21:10 | Lonoxmont | haha computer go brrr |
| 17:44:17 | speachy | the jz4760 pll is ... annoying complicated. |
| 17:49:51 | | Join ac_laptop [0] (~ac_laptop@186.2.247.129) |
| 17:53:13 | pamaury | speachy: the doc is not particularly clear on that subject for sure |
| 17:54:12 | pamaury | speachy: also beware of the difference between JZ4760 and JZ4760B, I think something was changed wrt to clocks |
| 17:54:30 | speachy | the rule about how the outclock * outdivisor has to be between 500 and 1500MHz is particularly fun |
| 17:55:25 | speachy | discovered that after I'd recalculated the codec clocks as low as I could make it. |
| 17:56:04 | speachy | but I'm able to clock the PLL at 127MHz for 44KHz audio, which is a quarter of its old speed. |
| 17:56:44 | pamaury | that's pretty good, should improve battery usage a lot |
| 18:00 |
| 18:00:57 | speachy | main CPU is still blazing along at 492MHz |
| 18:02:04 | mendel_munkis | pamaury: I just noticed the imx233 rev 4 data sheet you put on the wiki Is missing some fields in the rev 1 generally available online. |
| 18:15:15 | pamaury | mendel_munkis: yeah it's weird |
| 18:15:43 | pamaury | I think I've used the rev 1 as a reference mostly |
| 18:16:28 | mendel_munkis | I know I have. I just noticed the rev 4 yesterday |
| 18:20:29 | pamaury | I don't remember spotted big differences though, and sometimes the stmp3700 datasheets contains interesting information too |
| 18:20:54 | pamaury | it's clear some functionality from the stmp3700 was removed from the manual but not from the hardware |
| 18:26:56 | *** | Saving seen data "./dancer.seen" |
| 18:29:54 | | Quit pamaury (Ping timeout: 256 seconds) |
| 18:35:00 | | Quit pixelma (Quit: .) |
| 18:35:00 | | Quit amiconn (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.) |
| 18:37:53 | | Join pixelma [0] (marianne@rockbox/staff/pixelma) |
| 18:37:53 | | Join amiconn [0] (jens@rockbox/developer/amiconn) |
| 18:44:18 | | Quit sakax (Remote host closed the connection) |
| 19:00 |
| 19:07:49 | | Quit petur (Remote host closed the connection) |
| 19:33:17 | | Quit MrZeus (Read error: Connection reset by peer) |
| 20:00 |
| 20:22:12 | speachy | probably IP owned by some 3rd party that didn't transfer to Freescale |
| 20:27:00 | *** | Saving seen data "./dancer.seen" |
| 20:38:53 | | Join t0mato8 [0] (~t0mato@193.32.127.154) |
| 20:39:45 | | Quit t0mato (Ping timeout: 240 seconds) |
| 20:39:45 | | Nick t0mato8 is now known as t0mato (~t0mato@193.32.127.154) |
| 21:00 |
| 21:11:25 | | Join reductum [0] (~weechat@cpe-104-175-169-123.socal.res.rr.com) |
| 21:16:25 | | Join advcomp2019 [0] (~advcomp20@65-131-173-185.sxct.qwest.net) |
| 21:16:25 | | Quit advcomp2019 (Changing host) |
| 21:16:25 | | Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019) |
| 21:16:31 | __Bilgus_ | Strife89 I just changed the battery bench code to hopefully stop stack overflows is that built off the current head? |
| 21:16:53 | __Bilgus_ | just did it as in last few days |
| 21:18:28 | Strife89 | Built off 5d1d431b97 |
| 21:19:13 | Strife89 | So not current head but pretty close |
| 21:19:25 | | Quit advcomp2019_ (Ping timeout: 264 seconds) |
| 21:21:53 | | Join saratoga [0] (620acd42@cpe-98-10-205-66.rochester.res.rr.com) |
| 21:22:03 | saratoga | speachy: sorry just saw your question about merging my gerrit accounts, if that is possible it would be great |
| 21:29:48 | __Bilgus_ | Strife89 that looks to be after my patch unfortunately |
| 21:45:22 | | Quit reductum (Quit: WeeChat 2.9) |
| 21:45:36 | | Quit ac_laptop (Ping timeout: 244 seconds) |
| 22:00 |
| 22:02:48 | | Quit heredoc (Quit: ZNC - http://znc.in) |
| 22:06:19 | | Quit saratoga (Ping timeout: 245 seconds) |
| 22:16:17 | __Bilgus_ | Strife89 batterybench.c try changing #define BUF_SIZE 16000 to #define BUF_SIZE 8000 |
| 22:16:43 | __Bilgus_ | its higly doubtful its going to get 8 hrs let along 16 |
| 22:16:47 | __Bilgus_ | alone* |
| 22:27:01 | *** | Saving seen data "./dancer.seen" |
| 22:27:08 | Strife89 | __Bilgus_: Alright, I'll give it a try. |
| 22:27:58 | Strife89 | Building off the same revision, and the only other local change will be commenting out HAVE_ADJUSTABLE_CPU_FREQ in sansam200v4.h |
| 22:30:46 | Strife89 | Then I'll test a clean build, same revision. |
| 22:37:27 | Strife89 | Damn, it didn't even make it to the end of the first track. Undefined instruction |
| 22:41:47 | Strife89 | I'll just try a clean build to make sure |
| 22:46:48 | Strife89 | __Bilgus_: https://photos.app.goo.gl/sBd39CnfU4UnZWFo7 |
| 22:53:26 | Strife89 | If I was smart, I would have built and copied a voice file for the clean build |
| 22:54:02 | Strife89 | lol this is new, the LCD output is upside-down *and* mirrored |
| 22:55:58 | Strife89 | Nope, still crashes in the middle of the first track |
| 22:56:10 | Strife89 | With the added fun of not being able to see anything on the screen |
| 22:57:53 | | Join ac_laptop [0] (~ac_laptop@186.2.247.129) |
| 22:59:43 | Strife89 | It looks like Battery Bench did manage to write something. https://www.irccloud.com/pastebin/k8oygo7n/battery_bench.txt |
| 23:00 |
| 23:01:43 | Strife89 | Upside-mirrored screen: https://photos.app.goo.gl/8ny5qkeT1iRAXCwn8 |
| 23:02:13 | | Join S|h|a|w|n [0] (~shawn156@unaffiliated/shawn156) |
| 23:03:24 | __Bilgus_ | strife that is interesting |
| 23:04:29 | __Bilgus_ | hmm wonder what the undefined instruction is could you pastebin the map file for the build and for battery bench or the relevant portions thereof? |
| 23:04:40 | Strife89 | Built off head, no changes, and another build with one change (BUF_SIZE 8000). Both crash before the end of a short track |
| 23:05:05 | __Bilgus_ | which build is that undefined ins error from? |
| 23:05:33 | __Bilgus_ | or is it the same pc and sp\? |
| 23:06:03 | Strife89 | That was from 5d1d431b97, with change to BUF_SIZE and commenting out HAVE_ADJUSTABLE_CPU_FREQ |
| 23:06:11 | Strife89 | Though I imagine I could reproduce |
| 23:06:40 | __Bilgus_ | well the pc is going to be different between builds most likely |
| 23:06:42 | Strife89 | Can't read the crash errors on this build because the LCD stays blank when it crashes :D |
| 23:06:53 | __Bilgus_ | lol |
| 23:07:18 | Strife89 | So you still want the map file? |
| 23:07:33 | __Bilgus_ | not sure where the not having a screen comes in at though tbh |
| 23:07:37 | Strife89 | Or should I roll back and try to reproduce the exact same error |
| 23:08:18 | __Bilgus_ | if the map matches the build that made the undefined ins error screen yes |
| 23:08:21 | Strife89 | The screen frequently goes blank on me when entering the WPS |
| 23:09:28 | __Bilgus_ | that could be the lcd controller locking up did you try halving the dbop clock? |
| 23:09:42 | Strife89 | Haven't tried that |
| 23:09:52 | Strife89 | Where is it |
| 23:10:51 | __Bilgus_ | firmware/target/arm/as3525/clock-target.h and mess around with the clocks AS3525_DBOP_FREQ try dividing by 2 |
| 23:11:01 | Strife89 | Just made a new build, using head, with CPUFREQ commented out |
| 23:11:29 | Strife89 | I will try to get it to reproduce the crash and then try halving the clock |
| 23:12:02 | __Bilgus_ | cpufreq? |
| 23:12:17 | Strife89 | HAVE_ADJUSTABLE_CPU_FREQ |
| 23:13:27 | __Bilgus_ | ah ok |
| 23:13:48 | Strife89 | Damn, I thought it was going to make it |
| 23:13:59 | Strife89 | Track 1 is 53 seconds long, it crashed on 52 |
| 23:14:52 | Strife89 | __Bilgus_: Panic message: https://photos.app.goo.gl/9hpdnFksPEBTPYYo7 |
| 23:15:43 | __Bilgus_ | ah you'll need to do the other thing too :/ |
| 23:15:57 | Strife89 | Which other thing |
| 23:16:15 | __Bilgus_ | cutting down the buffer in battery bench |
| 23:16:42 | Strife89 | Oka |
| 23:17:20 | __Bilgus_ | I wonder if the undefined instruction long jumping could cause the stack ovf |
| 23:18:37 | Strife89 | I'm not sure which part to even change in clock-target.h so I'm going to leave that alone for now |
| 23:18:47 | Strife89 | It's not part of the battery bench testing besides. |
| 23:19:11 | speachy | okay, here's a strange question −− does rockbox automagically resample everything to 44.1KHz? |
| 23:21:43 | speachy | because it looks like, even for targets that support 48KHz, the mixer rate gets initialized at startup to HW_SAMPR_DEFAULT and is never actually changed by anything |
| 23:23:37 | | Join Rower [0] (~Rower@78-73-72-39-no2340.tbcn.telia.com) |
| 23:23:44 | speachy | unless HAVE_PLAY_FREQ is defined, and that's only done by the iBasso DX50/DX90 |
| 23:26:34 | Strife89 | If that's the case ... whoops. |
| 23:28:09 | Strife89 | __Bilgus_, here's your crash. https://photos.app.goo.gl/oBvNP6YMzFnsm7vo9 |
| 23:28:39 | Strife89 | I played a longer song this time, and it still crashed around 52 seconds |
| 23:30:01 | Strife89 | (I changed songs because I tried playing the 53-second song with no batt bench, as a sanity check, and this happened immediately after clicking on it in the file browser. https://photos.app.goo.gl/HcDVgqpKqYQj7XJr9) |
| 23:30:55 | __Bilgus_ | now just need the map so we can see whats at 0x302082B8 |
| 23:32:19 | __Bilgus_ | oh and maybe the battery bench one too |
| 23:33:10 | __Bilgus_ | 0x30212758 |
| 23:34:29 | Strife89 | https://usercontent.irccloud-cdn.com/file/NTgsfJSS/battery_bench.map |
| 23:35:30 | Strife89 | https://usercontent.irccloud-cdn.com/file/lSRhbw0U/rockbox.map |
| 23:36:29 | __Bilgus_ | thank you sir bbiab |
| 23:38:14 | __Bilgus_ | huh 302082B8 points to main() |
| 23:38:44 | __Bilgus_ | sounds like an unhandled stack overflow perhaps |
| 23:40:37 | __Bilgus_ | and 30212758 isn't even in the maps |
| 23:41:25 | * | Strife89 raises an eyebrow |
| 23:42:29 | __Bilgus_ | theres something funky going on maybe the chip is glitching? |
| 23:42:42 | __Bilgus_ | I assume it works fine with the OF |
| 23:43:08 | Strife89 | As far as I know it does, but I didn't test it much |
| 23:43:39 | __Bilgus_ | before we go chasing rabbits you might wanna check that out a bit |
| 23:43:44 | fs-bluebot | Build Server message: New build round started. Revision f554c78, 280 builds, 11 clients. |
| 23:43:53 | Strife89 | Will do |
| 23:44:26 | | Quit TheSeven (Disconnected by services) |
| 23:44:36 | | Join [7] [0] (~quassel@rockbox/developer/TheSeven) |
| 23:46:37 | Strife89 | For now I'm going to bed. Night, folks. |
icon identifies that the person is a core developer (has commit access).