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 2016-11-10

00:02:19 Quit petur (Quit: Leaving)
00:04:44 Join The_Prospector|2 [0] (
00:07:45 Quit The_Prospector (Ping timeout: 258 seconds)
00:18:09 Quit TheLemonMan (Quit: "It's now safe to turn off your computer.")
00:19:13 Join thrillho [0] (~Luke@unaffiliated/rockandorroll)
00:24:04 Quit thrillho (Ping timeout: 265 seconds)
00:46:55 Quit ender` (Quit: In politics, absurdity is not a handicap. — Napoleon Bonaparte)
00:49:43***Saving seen data "./dancer.seen"
00:59:14 Join thrillho [0] (~Luke@unaffiliated/rockandorroll)
01:03:37 Quit thrillho (Ping timeout: 244 seconds)
01:10:46 Quit dfkt (Disconnected by services)
01:10:56 Join dfkt_ [0] (~dfkt@unaffiliated/dfkt)
01:15:15 Quit paulk-collins (Quit: Leaving)
01:33:19 Join thrillho [0] (~Luke@unaffiliated/rockandorroll)
01:37:43 Quit thrillho (Ping timeout: 244 seconds)
01:50:51 Join JanC_ [0] (~janc@lugwv/member/JanC)
01:51:17 Quit JanC (Ping timeout: 245 seconds)
01:52:07 Nick JanC_ is now known as JanC (~janc@lugwv/member/JanC)
01:52:18 Quit TheEaterOfSouls (Quit: ChatZilla 0.9.92 [Firefox 49.0.2/20161019084923])
02:13:29 Quit anormal (Ping timeout: 260 seconds)
02:13:54 Join thrillho [0] (~Luke@unaffiliated/rockandorroll)
02:13:56 Join Bilgus [0] (ae6611d9@gateway/web/freenode/ip.
02:14:27Bilguscouple questions.. where doew the sim save its .cfg?
02:15:40Bilgusand I want to put my menu into settings is there a way to do this without using the standard defined ways? Ie can I just have the menu call my function to write global_settings.bl_selective_backlight_val?
02:16:17Bilgushere is the function i'm interested in adding to the menu
02:17:00BilgusI was able to call it in a callback but the menu overwrites what I have set in there with True/false etc.
02:17:04 Quit ZincAlloy (Quit: Leaving.)
02:18:24 Quit thrillho (Ping timeout: 252 seconds)
02:24:34 Quit [Saint] (Ping timeout: 260 seconds)
02:24:59 Quit krnlyng (Ping timeout: 260 seconds)
02:33:06 Join thrillho [0] (~Luke@unaffiliated/rockandorroll)
02:37:38 Join krnlyng [0] (
02:37:39 Quit thrillho (Ping timeout: 248 seconds)
02:49:17__builtinok, I can't seem to figure this out
02:49:45__builtindoes rockbox on the ipod 7g support those 4-conductor inline remotes?
02:49:46***Saving seen data "./dancer.seen"
02:56:03__builtinand the mics on them
03:00:59 Join thrillho [0] (~Luke@unaffiliated/rockandorroll)
03:02:25 Quit soap (Read error: Connection reset by peer)
03:05:47 Quit thrillho (Ping timeout: 256 seconds)
03:12:38 Join [Saint] [0] (~sinner@rockbox/staff/saint)
03:25:05 Join bzed_ [0] (
03:26:30 Quit bzed (Ping timeout: 260 seconds)
03:26:30 Quit maraz (Ping timeout: 260 seconds)
03:26:33 Nick bzed_ is now known as bzed (
03:26:47 Join maraz [0] (
03:26:58 Quit cryham (Ping timeout: 252 seconds)
03:26:58 Quit ranmacha1 (Ping timeout: 260 seconds)
03:28:13 Join thrillho [0] (~Luke@unaffiliated/rockandorroll)
03:28:21 Join ranmachan [0] (
03:32:57 Quit thrillho (Ping timeout: 250 seconds)
03:50:43 Quit smoke_fumus (Ping timeout: 248 seconds)
04:08:30 Join thrillho [0] (~Luke@unaffiliated/rockandorroll)
04:13:11 Quit thrillho (Ping timeout: 252 seconds)
04:49:48***Saving seen data "./dancer.seen"
04:50:58 Join thrillho [0] (~Luke@unaffiliated/rockandorroll)
04:52:37 Quit alexweissman (Remote host closed the connection)
04:55:40 Quit thrillho (Ping timeout: 258 seconds)
05:08:38 Join alexweissman [0] (
05:09:56 Quit alexweissman (Remote host closed the connection)
05:30:41 Join thrillho [0] (~Luke@unaffiliated/rockandorroll)
05:35:25 Quit thrillho (Ping timeout: 268 seconds)
05:43:54 Join alexweissman [0] (
05:47:32 Quit [Saint] (Remote host closed the connection)
05:49:17 Join [Saint] [0] (~sinner@rockbox/staff/saint)
05:59:14 Join thrillho [0] (~Luke@unaffiliated/rockandorroll)
06:03:54 Quit thrillho (Ping timeout: 258 seconds)
06:14:36 Quit TheSeven (Ping timeout: 260 seconds)
06:14:57 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
06:22:29 Join thrillho [0] (~Luke@unaffiliated/rockandorroll)
06:27:04 Quit thrillho (Ping timeout: 244 seconds)
06:49:47 Quit alexweissman (Remote host closed the connection)
06:49:50***Saving seen data "./dancer.seen"
06:52:54 Join dfkt [0] (~dfkt@unaffiliated/dfkt)
07:00:17 Quit [Saint] (Quit: Quit.)
07:01:42 Join [Saint] [0] (~sinner@rockbox/staff/saint)
07:06:19Bilgusthink i have it figured out
07:14:16 Join alexweissman [0] (
08:27:18 Join ender` [0] (
08:43:36 Join wodz [0] (
08:46:25 Join lebellium [0] (
08:49:51***Saving seen data "./dancer.seen"
08:50:42 Join thrillho [0] (~Luke@unaffiliated/rockandorroll)
08:54:19 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
08:59:36 Quit thrillho (Ping timeout: 240 seconds)
09:11:57 Join thrillho [0] (~Luke@unaffiliated/rockandorroll)
09:16:39wodzpamaury: Today I found this:
09:17:05 Quit pamaury (Ping timeout: 256 seconds)
09:18:09wodzpamaury: This are sources from action for otg which seems to be very close to the one used in atj213x. I think DMA part is different as atj213x seems to miss separate LL DMA engine for usb
09:18:55 Join anormal [0] (5dbc8f01@gateway/web/freenode/ip.
09:21:05 Quit thrillho (Quit: WeeChat 1.5)
09:50:45anormalhi everyone, regarding my question yesterday about opening Fiio X1 II, i wrote them an email (as suggested in reddit)
09:50:51anormal>>We are making the disassemble guidance in English now , and it will be ready in next week. could you kindly wait for it . Do you have any experience for disassembling any unit before ? Since the X1 2nd gen is not easy to disassemble, the unit may be broken. To be honest , we do not suggest you to open your X1 2nd gen , but all is up to you.
09:51:02anormalSo maybe next week they release the video
09:54:16 Join elensil [0] (~edhelas@2001:1c02:1903:d800:dcd1:6f28:6bdc:719f)
10:06:31duo8it's built like a phone apparently
10:06:35duo8expect glue
10:16:56 Join pamaury [0] (
10:16:56 Quit pamaury (Changing host)
10:16:56 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:18:27pamauryanormal: don't worry if you can't open it, we can probably done without
10:33:14wodzpamaury: Have you seen my statement earlier today?
10:33:38pamauryabout linux sources? yes, had a quick look but the driver is huge
10:36:52wodzpamaury: yes it is very convoluted
10:37:09pamaurybut maybe it can be useful for register documentation
10:49:30wodzpamaury: The strange thing is I can't find anywhere special ZLP handling in this sources
10:49:53***Saving seen data "./dancer.seen"
10:51:09pamaurywodz: in aotg_udc.c:write_ep0_fifo
10:53:22pamaurywhen in state EP0_IN_DATA_PHASE and there is nothing more to transfer, it will stay one more time in transfer state to send a ZLP (you can trace the logic to write_ep0_fifo:is_last)
10:53:37wodzpamaury: ok, thats for IN, and where it is for OUT?
10:54:05pamaurywell for out there is just the final ack, it's not really a zlp, let me see
10:56:48pamauryin handle_ep0_out in state EP0_OUT_DATA_PHASE, if read_ep0_fifo returns 1 (meaning it received all the data), then it udc_handle_status() which clears NAK
10:57:02pamaurythat's the status stage
11:16:40 Join pamaury_ [0] (~pamaury@rockbox/developer/pamaury)
11:27:16 Quit pamaury_ (Ping timeout: 260 seconds)
11:41:48 Join pamaury_ [0] (~pamaury@rockbox/developer/pamaury)
11:45:01 Join robertd1 [0] (~as@
11:51:32 Quit pamaury_ (Ping timeout: 244 seconds)
12:15:52 Join TheLemonMan [0] (~root@unaffiliated/thelemonman)
12:41:23 Join petur [0] (
12:41:25 Quit petur (Changing host)
12:41:25 Join petur [0] (~petur@rockbox/developer/petur)
12:44:06 Join ZincAlloy [0] (~Adium@2a02:8108:8b80:1700:6828:40c7:4ad2:5dd8)
12:49:55***Saving seen data "./dancer.seen"
12:55:57 Quit Bilgus (Quit: Page closed)
13:09:52wodzpamaury: in OUT handler if read_ep0() returns less then max pkt size that means end of transfer, right?
13:12:17pamauryhum, at least that's what the spec says but let me check the code
13:12:24wodzsooo, what happens if read_ep0() is short and we are still below expected len?
13:12:57pamauryread_ep0_fifo() does not seem to care about short packet and continues until it receives the whole packet
13:13:19pamaurythe host will never send a short packet in the middle of a OUT transfer
13:13:22pamauryit's forbidden
13:13:34wodzpamaury: you mean in actions code or hwstub?
13:14:16pamauryever, when the host sends you data of a control transfer, only the last packet will be short
13:14:50 Join Bray90820_ [0] (
13:14:51wodzpamaury: yes I understand that but what if host sends less then stated in setup?
13:14:52pamaurytechnically if you receive a short packet and you haven't received everything, that's a protocol error
13:14:58pamauryit's forbidden
13:15:08pamaurythe host always sends what he promised
13:15:18wodzlots of things are forbidden and still occures :-)
13:16:11pamauryIN and OUT transfers are not symmetric: in out transfers, the host knows how much it is sending and must do so. for in transfers on the other hands, the host tells you how much it expects at the maximum but you can send less
13:16:46wodzwhat should happen on protocol error as you call it
13:16:56 Quit Bray90820 (Ping timeout: 265 seconds)
13:17:22pamauryI don't know, the spec probably says what should be done, most likely set stall
13:19:29 Join paulk-collins [0] (
13:21:05wodzpamaury: ok we stall in higher layer on req->wLength and actual transfer size mismatch
13:22:01 Join pamaury_ [0] (~pamaury@rockbox/developer/pamaury)
13:25:50pamauryGiven how our code works, the higher layer should be the one stalling: after a setup for control OUT, the driver is supposed to do size = usb_drv_recv(EP0, bufferm, setup.wLength); and stall if size != setup.wLength
13:26:47pamauryI am not sure if we do it, but in rockbox firmware, control OUT transfers are fundamentally broken anyway
13:27:23pamaurybut if you look at handle_read in main.c of stub, I check the return value
13:27:35wodzpamaury: At least in hwstub it is not prudently checked. Often size of transfer is checked against expected hwstub protocol header size instead of setup.wLength
13:28:03pamaurydo you have a concrete example?
13:28:40pamaury(there is only one type of control OUT in hwstub and it's for writes and as I said above we do check)
13:29:22wodzpamaury: in handle_read() int size = usb_drv_recv(EP_CONTROL, usb_buffer, req->wLength); if(size != sizeof(struct hwstub_read_req_t))
13:29:57wodzpamaury: I think it should be checked both against wLength and size of expected struct
13:30:00pamaurythat's ok, it's safe
13:30:32wodzit is ok as long and host and stub have the same size of this struct
13:30:48pamauryif they disagree the transfers will be stalled
13:31:23wodzyou are right
13:31:31 Quit pamaury_ (Ping timeout: 256 seconds)
13:31:42pamaurythe only bad thing that could happen is if the host says "I will send you 16 bytes" but actually sends you 10 and the size of the struct is 10. Then yeah technically that's a protocol error and we still accept it
13:32:19pamauryI won't object if you add it, it's just incredibly unlikely, linux won't ever let you do that
13:32:20wodzThis. Although possibility is tiny
13:32:31pamaurywindows either
13:33:15wodzpamaury: why OUT transfers are broken in rockbox?
13:33:25pamaurybecause usb_drv_recv is not blocking
13:33:36pamaury(it is blocking in hwstub)
13:34:07wodzand what stops us from make it blocking?
13:34:34pamauryfixing all the drivers
13:35:40pamauryI mean at the moment it's fine, we don't use control out transfers
13:35:54pamaurybut for example for usb audio I had to fix in some drivers
13:37:25pamauryalso it's not that simple, you want some recv not to be blocking (like non-control transfers and ZLP), so you need two variants (blocking and non-blocking)
13:38:00wodzI really hate usb :/
13:38:09 Join soap [0] (~soap@rockbox/staff/soap)
13:38:40pamaurysee g#1009 if you are curious
13:38:42fs-bluebot_Gerrit review #1009 at : Add USB Audio 1.0 support (EXPERIMENTAL) by Amaury Pouly
13:57:49fs-bluebot_Build Server message: New build round started. Revision ba9f405, 255 builds, 14 clients.
14:03:52 Quit lebellium (Quit: ChatZilla 0.9.92 [Firefox 49.0.2/20161019084923])
14:06:37fs-bluebot_Build Server message: Build round completed after 527 seconds.
14:06:38fs-bluebot_Build Server message: Revision ba9f405 result: All green
14:20:44 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
14:21:10 Join fs-bluebot [0] (
14:21:35 Join einhirn [0] (
14:23:35 Quit fs-bluebot_ (Ping timeout: 260 seconds)
14:23:57 Quit bluebrother^ (Ping timeout: 244 seconds)
14:26:00 Quit einhirn (Ping timeout: 250 seconds)
14:26:25 Join einhirn [0] (
14:33:16 Quit anormal (Quit: Page closed)
14:49:58***Saving seen data "./dancer.seen"
14:50:40 Join pamaury_ [0] (~pamaury@rockbox/developer/pamaury)
14:55:14 Quit pamaury_ (Ping timeout: 246 seconds)
14:57:39 Join fyre^OS [0] (~nnscript@
15:01:24 Quit fyre^OS (Client Quit)
15:24:19 Quit wodz (Ping timeout: 268 seconds)
15:31:11 Quit amayer (Quit: Leaving)
15:31:33 Join amayer [0] (
15:33:18 Quit amayer (Client Quit)
15:33:39 Join amayer [0] (
15:56:50 Join JanC_ [0] (~janc@lugwv/member/JanC)
15:58:04 Nick JanC is now known as Guest87424 (~janc@lugwv/member/JanC)
15:58:04 Quit Guest87424 (Killed ( (Nickname regained by services)))
15:58:04 Nick JanC_ is now known as JanC (~janc@lugwv/member/JanC)
16:05:32duo8wow, software usb audio
16:05:40duo8does it work well?
16:17:01dongsnothing too special about that, just shows up as USB Audio class, either for input or output (probly just output
16:19:53dongslooks like it does setup for both input and output.
16:50:02***Saving seen data "./dancer.seen"
16:51:38 Quit petur (Quit: Connection reset by beer)
17:10:56 Quit pamaury (Remote host closed the connection)
18:03:00 Join krabador [0] (~krabador@unaffiliated/krabador)
18:05:04 Quit elensil (Quit: Leaving.)
18:22:06 Quit toli (Ping timeout: 256 seconds)
18:22:46 Quit alexweissman (Remote host closed the connection)
18:28:37 Join toli [0] (
18:32:38 Join girafe [0] (
18:50:05***Saving seen data "./dancer.seen"
19:27:20 Join alexweissman [0] (
19:29:38 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
19:44:48pamaurydongs: duo8: the patch is not working properly, my implementation is incomplete
19:45:12pamaurybut fundamentally nothing prevents from implementing usb audio in software
20:27:46 Quit alexweissman (Remote host closed the connection)
20:31:59 Quit vincentflyson (Quit: ZNC 1.6.3 -
20:32:12 Join vflyson [0] (
20:33:10 Join alexweissman [0] (
20:50:06***Saving seen data "./dancer.seen"
21:25:03 Quit rela (Ping timeout: 260 seconds)
22:01:41 Join Bilgus [0] (ae6611d9@gateway/web/freenode/ip.
22:02:27BilgusIs there anyway to write a setting to the .cfg file without interfacing it through the menu system?
22:02:39Bilguscustom_setting perhaps?
22:03:34pamauryBilgus: why would you do that?
22:04:07pamauryI guess that if you just put the setting in setting.c but don't create an entry for it in the menus it will achieve what you want
22:05:09BilgusI have an option to enable selective actions when no it returns false and writes 0 to the mask when yes it calls action_select and I want to write the mask to a var which will be saved to cfg
22:06:42 Quit zoktar (Quit: -)
22:06:51Bilgusi tried that and it doesn't persist after power cycle is there a way to force sace?
22:09:23pamauryI am not sure I understand, if you have no menu to set, you have to write in the config file directly
22:10:07BilgusNo i take that back i put it in the user_data array in settings.h do I need to add it to a function in settings.c as well?
22:15:30BilgusI could do menu_function but I still can't figure out htf to save the mask
22:17:54pamauryyes you need the code in settings.c
22:18:10pamaurysettings.c is about loading/saving settings
22:18:21pamaurymenu_setting is about changing the setting with the user interface
22:18:46pamaurynot sure what you are trying to achieve, I still don't see what you want to do with a setting that you cannot change...
22:19:11 Part robertd1
22:20:12 Quit ruhans (Quit: Connection closed for inactivity)
22:22:19 Quit alexweissman (Remote host closed the connection)
22:26:01Bilgusok so I have a setting off/on enable selective backlight and enable selective softlock NO return false and set the mask to 0
22:26:36Bilgusselect YES call action_select and write the user selected mask to the cfg
22:27:14Bilgusdo you have a better idea?
22:29:09pamauryis the setting on/off or is it a mask?
22:29:41 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
22:29:51 Join alexweissman [0] (
22:31:42Bilguswell its a mask returned from action_select( 0 = disabled 2=vol+- 4 = play 8 = stop 16=seek 32 = skip
22:50:09***Saving seen data "./dancer.seen"
23:16:08 Quit alexweissman (Remote host closed the connection)
23:28:18Bilguspamaury: I had the mask defined as a bool :/ For future reference though:any setting type in settinglist.c will be saved to cfg you just need to define the range and inc and the variable in struct user_settings (settings.h) .. and in this case there is no matching menu in 'display_menu.c'
23:29:36BilgusDo you think instead I should display the settings chosen ie. none|Vol+-,Play,Skip
23:33:22 Quit pamaury (Ping timeout: 245 seconds)
23:42:52 Join alexweissman [0] (
23:50:24 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)

Previous day | Next day