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 2020-11-24

00:58:24_bilgus__tomf, you should be able to make the drive(s) mark bad sectors
01:03:15tomfooh, what would I search for that, _bilgus__?
01:04:20tomfwould this be a simple chkdsk?
01:04:31_bilgus__in windows no clue
01:04:46tomfI can use any OS −− ideally not windows :)
01:04:57_bilgus__ah good
01:06:34_bilgus__badblocks should allow you to export a list of bad blocks, just the act of a write error is supposed to be handled by the flash controller but if you are out of re-provisioning blocks then I suspect the drive could be on its way out
01:07:20_bilgus__now I wonder if there is a fat32 way to mark bad blocks
01:07:52tomfthe microSDs are fairly new −− I wouldn't be surprised if its a goofy JBOD issue with the iFlash
01:09:02_bilgus__well bad blocks will tell you that
01:10:42_bilgus__hmm I swear fsck had an option to do this but I'm not seeing it on my pc or the manpage
01:11:28_bilgus__well first I'd try some non-destructive tests on the whole drive together then on each card individually
01:11:28tomfyeah, I'll dig into it. first I'll try the ol 'copy and rename the directory' bit to see if that magic still works.
01:12:28_bilgus__yeah I suppose if its always the same file then just make a big dummy file named DONTDELETEME or something
01:13:00tomfyeah, I'll give it a swing. Thanks _bilgus__!
01:13:24tomfits been a week of minor, goofy things like this
01:13:41_bilgus__that sounds a lot like a bad card
01:13:54_bilgus__especially if you are using the latest dev versions
01:14:12_bilgus__but whooo knows
01:14:20tomfooh. I should move up to a newer version, actually. I'm not sure which one I'm on, but I'll update anyways
01:14:28_bilgus__most def
01:14:40_bilgus__always test bugs of fthe latest versions
01:14:52tomfI love rockbox so much :)(
01:15:12_bilgus__1 to see if its been fixed and 2 its hard to fix stuff thats been chnaged already
01:16:14tomfwhere can I find a list of what the error codes line up to?
01:16:30_bilgus__have to look in the source
01:16:39tomfyeah, are they all in errno.h?
01:17:12_bilgus__dc_writeback is probably in the fat driver
01:17:22_bilgus__easiest is to search for the error message
01:17:47tomfok, I'm on it! thanks again!
01:17:56_bilgus__you error quite literally said I tried multiple times to write this sector and failed
01:18:40tomfrb doesn't write a lot outside of bookmarks, does it?
01:18:50_bilgus__we really try not to
01:19:06_bilgus__settings changes cause writes on shutdown
01:20:09tomfok cool. thanks!
01:20:29_bilgus__I suppose all the repeatedly written files should probably employ the leave the old file write a new file delete old file rename new file
01:55:47***No seen item changed, no save performed.
03:14:23 Join petur [0] (~petur@
03:14:23 Quit petur (Changing host)
03:14:23 Join petur [0] (~petur@rockbox/developer/petur)
03:55:48***Saving seen data "./dancer.seen"
04:39:24 Join pamaury [0] (
04:39:24 Quit pamaury (Changing host)
04:39:24 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
05:33:08 Join dfkt [0] (
05:52:40 Join dfkt [0] (
05:55:49***Saving seen data "./dancer.seen"
06:30:59 Quit ThaneVim (Remote host closed the connection)
06:34:59 Quit Stanley00 (Remote host closed the connection)
06:36:37 Join prof_wolfff [0] (
07:55:40 Quit pamaury (Remote host closed the connection)
07:55:54***Saving seen data "./dancer.seen"
07:56:40 Join pamaury [0] (
07:56:40 Quit pamaury (Changing host)
07:56:40 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
08:03:39 Quit Rower ()
08:17:24 Join Rower [0] (
08:23:33 Quit Rower ()
08:25:11 Join Rower [0] (
08:58:51 Quit Rower ()
09:08:12 Join massiveH [0] (
09:09:14 Join Rower [0] (
09:44:03 Quit mutax (Read error: Connection reset by peer)
09:46:32 Join mutax [0] (
09:55:58***Saving seen data "./dancer.seen"
10:00:51 Quit mutax (Read error: Connection reset by peer)
10:06:38 Join mutax [0] (
10:33:51 Quit massiveH (Quit: Leaving)
10:35:57 Join Stanley00 [0] (~stanley00@unaffiliated/stanley00)
10:40:29 Quit Stanley00 (Ping timeout: 260 seconds)
11:20:03 Quit Moarc (Read error: Connection reset by peer)
11:20:32 Join Moarc [0] (
11:48:28 Join thanevim [0] (
11:56:02***Saving seen data "./dancer.seen"
12:10:08 Quit pamaury (Quit: Konversation terminated!)
12:29:05 Quit ufdm (Quit: Leaving)
12:50:09 Join lebellium [0] (
13:27:23 Join amiconn_ [0] (jens@rockbox/developer/amiconn)
13:27:23 Nick amiconn is now known as Guest16470 (jens@rockbox/developer/amiconn)
13:27:23 Quit Guest16470 (Killed ( (Nickname regained by services)))
13:27:23 Nick amiconn_ is now known as amiconn (jens@rockbox/developer/amiconn)
13:27:48 Join pixelma_ [0] (marianne@rockbox/staff/pixelma)
13:27:49 Nick pixelma is now known as Guest39682 (marianne@rockbox/staff/pixelma)
13:27:49 Quit Guest39682 (Killed ( (Nickname regained by services)))
13:27:49 Nick pixelma_ is now known as pixelma (marianne@rockbox/staff/pixelma)
13:43:16 Quit petur (Quit: Connection reset by beer)
13:46:24 Join ufdm [0] (
13:56:07***Saving seen data "./dancer.seen"
14:35:28 Quit jdarnley (Ping timeout: 256 seconds)
14:38:46 Join J_Darnley [0] (
15:40:49 Join livvy [0] (~livvy@gateway/tor-sasl/livvy)
15:44:40 Quit thanevim (Remote host closed the connection)
15:56:09***Saving seen data "./dancer.seen"
16:02:37braewoodsspeachy: we're probably going to want some way to manage the allocation of usb endpoints... from the user's end. like being able to control which drivers will be activated when USB is connected.
16:03:07braewoodsit might also help if we could know how many endpoints each driver uses so we know if a specific combination is even possible
16:03:12speachythat's already implemented, though not in a clean generic manner.
16:03:23braewoodsi see
16:03:32braewoodswell, with the desire for more drivers this becomes even more important.
16:03:56braewoodsusb audio appears to use a ton of endpoints... at least 3 if this right
16:04:20 Join petur [0] (~petur@rockbox/developer/petur)
16:04:32speachywe're not going to expose everything simultaneously. Though I suppose UMS + Audio would at least make sense.
16:04:36braewoodsi was thinking usb audio could be implemented as a bridge to rockbox's audio outputs
16:04:55braewoodsand inputs
16:05:30braewoodsthough i don't entirely see the point of using usb for this
16:05:47braewoods"do usb audio cables make a difference?"
16:05:57braewoodsi doubt it? usb is a digital transfer.
16:06:07braewoodsat most it might effect the error rate
16:13:23speachyacting as an ADC/DAC is a common enough feature of DAPs these days.
16:14:18speachyUSB audio requires isocronous transfers though, which will definitely require work on a per-usb-core-driver basis.
16:15:07speachylast night I started trying to gether all the software bits I'd need to build an image for the PineCube or CherryPi-V3S boards.
16:15:32speachythe former is in mainline u-boot and linux now, the latter shouldn't require more than minor devicetree tweaks.
16:16:09braewoodsspeachy: i heard not all of them support iso but there's a good chance this'll work on anything that HID does
16:16:29braewoodsreading the usb spec i realize it's basically based on TCP/IP principles
16:16:40braewoodsUDP vs TCP
16:16:42braewoodsin many cases
16:17:04speachyIIRC audio is about the only acutual user of isoc transfers.
16:17:18braewoodsthe HID driver also uses one
16:17:25braewoodsa single iso endpoint to send input events
16:18:00braewoodsthought it was iso
16:18:02speachyyeah, HID used interrupts
16:18:13speachy(again, about the only actual user of that..)
16:18:25braewoodsat least as far as rockbox is concerned
16:19:06speachyisoc matters because it has relatively tight timing requirments, which generally also means DMA.
16:19:54braewoodsmeaning there's not many devices where it'll work properly?
16:21:03braewoodsaccording to the source
16:21:05speachymeaning it'll likely only be "newer" devices.
16:21:14braewoodsonly 2 USBOTG drivers support ISO
16:21:32braewoodsis mainly PP devices
16:21:36speachyDWG or ARC cores will work.
16:21:51speachythe ingenic USB core will too.
16:22:07braewoodsISP1362 will too if we ever get it working
16:22:14braewoodsit's limited to 1.1 but it claims to support ISO
16:24:30braewoodsspeachy: is this right? ARC is limited to 3 endpoints?
16:25:29speachydunno, I've never looked.
16:25:50braewoodsthat's what pp5020.h defines
16:26:25braewoodswhich is enough for the current stack of HID + UMS
16:26:26speachythere might be multiple variations of the ARC core too, with different features/capabilities.
16:26:56speachywe already have the ability to enable/disable HID on devices that can support it
16:27:03speachyditto on UMS.
16:27:46speachyI'd always envisioned my cheesy "mount y/n" prompt on USB insertion to become a full menu once more capabilities are added.
16:27:57speachy(that's the easy part!)
16:28:14braewoodsif we want MTP, it requires 3 endpoints
16:29:31braewoodsoddly this isp1362 has one thing going for it in spite of being slow
16:29:36braewoodsit has a shitton of endpoints
16:29:50 Quit Rower ()
16:31:39braewoodsi wouldn't be surprised if 3 is the limit for PP5020 since they never envisioned needing to support more than 3 endpoints or USB devices over the same link
16:32:18braewoodsso for some devices they may end up being a one trick pony for what they do at any one time over USB
16:42:50 Join ac_laptop [0] (~ac_laptop@
16:52:01 Join Rower [0] (
16:55:34 Quit livvy (Remote host closed the connection)
17:08:36 Join livvy [0] (~livvy@gateway/tor-sasl/livvy)
17:12:07 Quit livvy (Remote host closed the connection)
17:14:30 Join borkitall [0] (
17:15:03 Quit borkitall (Client Quit)
17:15:10 Join livvy [0] (~livvy@gateway/tor-sasl/livvy)
17:18:52 Quit igitoor (Ping timeout: 260 seconds)
17:19:25 Join igitoor [0] (igitur@2a00:d880:3:1::c1ca:a648)
17:23:45 Join vmx [0] (~vmx@
17:27:00 Join borkitall [0] (
17:30:21 Quit igitoor (Changing host)
17:30:21 Join igitoor [0] (igitur@unaffiliated/contempt)
17:31:44 Quit Rower ()
17:36:32 Join Rower [0] (
17:38:58 Quit lebellium (Quit: Leaving)
17:44:03 Quit livvy (Remote host closed the connection)
17:46:32 Join livvy [0] (~livvy@gateway/tor-sasl/livvy)
17:48:20 Quit Rower ()
17:54:43 Join cockroach [0] (~blattodea@pdpc/supporter/active/cockroach)
17:55:56 Join Rower [0] (
17:56:10***Saving seen data "./dancer.seen"
18:07:46 Join Rower- [0] (
18:09:32 Quit Rower- (Client Quit)
18:10:36 Quit Rower (Ping timeout: 240 seconds)
18:13:25 Quit vmx (Quit: Leaving)
19:08:09 Quit bluebrother (Disconnected by services)
19:08:14 Join bluebrother^ [0] (~dom@rockbox/developer/bluebrother)
19:08:26 Join fs-bluebot [0] (
19:10:56 Quit fs-bluebot_ (Ping timeout: 256 seconds)
19:26:11 Quit petur (Quit: Leaving)
19:56:12***Saving seen data "./dancer.seen"
20:41:11braewoodsspeachy: the usb audio spec mentions a need for synchronization. what method could we use in rockbox i wonder?
20:41:53braewoodssection 3.3
21:07:31speachyprobably astnchronous. I don't think we will realistically be able to slave things to the incoming data frames.
21:08:14speachy(that would require being able to slave our DAC to a recovered clock, which.. might be possible on some targets, but isn't something I would count on.
21:09:55speachyis SOF generated by the source or the sink?
21:12:12speachyOTOH, the DAC and USB core are often on a shared clock source
21:12:44speachyso jitter notwithstanding, they're not going to drift with respect to each other.
21:13:12braewoodsi just don't see myself doing this alone.
21:13:15speachybut I don't know the spec well enough to know how the sink tells the source to slow down
21:13:38braewoodsit may only need 1.1 but it's a specialized area
21:14:28braewoodsi also posted a topic on the forums asking if people thought the h300 could use OTG support
21:17:34speachyMTP would be simpler. :)
21:18:38braewoodsit would also help if we had something to base it on
21:18:46braewoodsmost stuff is focused on host usb audio not the other side
21:19:08speachythough to be honest at this point it may make more sense to focus on making hosted rockbox behave better rather than low-level hackery like that.
21:19:28braewoodsyou mean the H300?
21:21:41braewoodsanyway in these older ports the main innovations i see is in the OTG area
21:21:51braewoodsthey're otherwise mostly maxed out already
21:40:36 Quit cockroach (Quit: leaving)
21:41:24 Join Stanley00 [0] (~stanley00@unaffiliated/stanley00)
21:56:14***Saving seen data "./dancer.seen"
22:33:23 Quit livvy (Ping timeout: 240 seconds)
22:44:06borkitallso just confirming, the search function on the website is broken? im trying to add lyrics
22:45:20borkitallif it is, its possibly something i could help out with, if noone else has the time to fix it
22:46:11borkitall2014 is far too long for it to be broken
22:49:38speachy Huh. Guess I should look into that.
22:51:52borkitallin the meantime, would someone link me the lyrics page?
22:53:16borkitallim far too low on sleep atm to bother lol
22:54:30borkitallbut ill try find it
22:55:16 Quit michaelni (Ping timeout: 240 seconds)
22:56:31borkitalla rockbox bang command for duckduckgo would be a good idea for something for me to do
23:02:19borkitallone of the requirements for a bang command is it must be useful to more than 500 people, i think that requirement is satisfied
23:04:26borkitallokay ill have a look into it further later, once i get more sleep maybe
23:04:49borkitallor no
23:04:59borkitallim so tired
23:08:00 Join michaelni [0] (
23:11:33borkitallspeachy: thanks
23:13:53borkitalllooks like im gonna have to actually do some dev work :)
23:16:08borkitallhave to restart computer to clone the git repo
23:16:27borkitalloh wait no i dont lol
23:16:33borkitallso tired
23:17:47borkitallugghhh my ssl config for konversation is fucking with my ability to clone git repos
23:18:13speachygit-over-http is faily inefficient
23:18:32borkitalloh yeah, i could git another way hey
23:18:33CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
23:18:33*speachy sighs and yawns.
23:18:35borkitalli should know this
23:18:46speachybeen at this sice about 0800. it's now 23:18
23:19:01borkitalli feel ya
23:19:27speachynot going to stay up to watch it finish though.
23:19:35speachyor more likely, go all explode-y.
23:19:35borkitallall good man
23:19:45speachy(it'll be at least another 2 hours)
23:19:50borkitall hope you have a good sleep :)
23:21:23borkitallim in the process of switching from firefox to konqueror as well so that doesn't help..
23:22:16borkitalljust have to do my git config
23:23:11speachyja, you too
23:24:55borkitallokay we have progress
23:26:47borkitalloh no wonder im confused, i already have a private window open :P
23:28:13borkitallokay that wasn't necessary..
23:29:28borkitallreally need to get something better than i3...
23:31:20borkitallneed to write a reason why it would be helpful for duckduckgo users..
23:31:54borkitallif anyone feels like doing this for me, feel free lol
23:32:03borkitallotherwise i might just do it a bit later
23:32:57borkitallthe bang works btw
23:33:51borkitallit was a good excuse to finally get my git config set up tho
23:56:17***Saving seen data "./dancer.seen"

Previous day | Next day