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 2021-06-13

00:00:29speachyanyway, the new-new code now tracks the prior header's offset, so when we have two-in-a-row matches it can return the first header and seek back to the correct position before ejecting from the loop.
00:06:04Bilguslooks like less churn
00:14:49speachyyeah, instead of reading bytewise forward, then seeknig forward and back, then resuming the bytewise search
00:17:22 Join skipwich [0] (~skipwich@user/skipwich)
00:36:26 Quit ufdm_ (Quit: Leaving)
00:36:40 Join ufdm [0] (
00:42:56 Quit Bilgus (Ping timeout: 252 seconds)
00:43:34 Join Bilgus [0] (~WW@
01:26:52desowinbraewoods: the chip is marked TNETV105PAP, I believe in digikey it is listed as XTNETV105PAP - atleast the description pretty much matches
01:28:08desowinspeachy: yes, the wiki is updated
01:28:54desowinI am thinking about "fast boot" exploit. Basically now, the bootloader payload is limited to 2095088 bytes, in the "fast" option it would be limited to 901104 bytes
01:29:20desowinthe change would be where in flash it goes (and thus the binary will be incompatible)
01:29:41desowinthe current exploit basically calculates SHA1 over 16 + bootloader length bytes
01:30:13***Saving seen data "./dancer.seen"
01:30:25desowinthe fast boot would result in calculating SHA1 over 24 bytes (16 bytes mandatory header, 4 bytes ldr instruction, 4 bytes value for ldr instruction)
01:36:54desowinhmm, 20 bytes should be possible as I think I could stuff the value somewhere in the signature itself
02:09:14braewoodsdesowin: that should be more than enough for a bootloader.
02:09:25braewoodsa lot of older targets have far less space than that
02:20:33braewoodsdesowin: sheesh. it's some mystery chip.
02:20:47braewoodsthere's no datasheet that i can tell.
02:21:16braewoodsi wonder what it does that would make them do that
02:24:09braewoodsdesowin: i see you expanded the serial driver?
02:24:26braewoodsdesowin: i was going to look into adding some kind of binary protocol to it at a later time
03:06:32desowinthe expansion is just to make it appear as COM port in Windows and ttyACM0 in Linux, the data travelling is unaffected
03:07:29desowinI would prefer a text protocol with simple human readable commands, adding some target specific commands would make it easy to just try some things (like toggling gpio) without loading new binary
03:08:36desowinand well, it shouldn't be too hard to multiplex logf there... as long as commands are human readable, the user will be able to distinguish between logf and command if these happen to be merged
03:17:03desowinhmm, maybe even 16 bytes in fast boot should be possible, as the immutable signature bytes seem to form valid and harmless instructions
03:18:15desowinas "moveq r0, #137" followed by "andeq r0, r0, r5" does not really make any difference
03:19:23desowinif needed you can stuff quite some payload into the signature itself (e.g. copying the flash to ram at some address), but for Rockbox just a jump is enough
03:30:14***No seen item changed, no save performed.
04:54:58desowinI need help with the linker script. For fast boot exploit I would have to build Rockbox Sansa Connect bootloader in a way that it would occupy flash starting at 0x00424010, with limit up to 0x000DBFF0 bytes
04:55:36desowinSo I opened firmware/target/arm/tms320dm320/ and modified Sansa Connect FLASHMEMORIG to 0x00424010 and FLASHMEMSIZE to 0x000DBFF0
04:56:06desowinbootloader binary is just 114 KiB so there's plenty room left
04:56:20desowinhowever, the linking fails with "Not enough room for program headers, try linking with -N"
04:56:46desowinif I add -N option to linker, then the resulting elf does not contain the startup code that's supposed to go at 0x00424010
05:30:17***No seen item changed, no save performed.
07:06:39desowinchanging FLASHMEMORIG to 0x00430010 resulted in working binary
07:07:07desowinhowever, the "fast" is not as fast as I anticipated, it is in tens of milliseconds range
07:07:17 Join F3l1x_10m [0] (~Al3x_10m@user/f3l1x-10m/x-3393542)
07:21:09desowinconsidering the boot takes around 5 seconds before the execution passes to Rockbox bootloader, the less than 100 ms save seems negligible
07:22:00desowinI wonder what takes so long, my guess would be the bignum operations on garbled data
07:30:18***No seen item changed, no save performed.
09:01:46 Join lebellium [0] (~lebellium@2a01:cb10:2e:2000:6920:551f:cdfb:44c1)
09:30:22***No seen item changed, no save performed.
09:41:40 Join dconrad [0] (~dconrad@
11:04:11 Join amachronic [0] (~amachroni@user/amachronic)
11:16:21 Quit amachronic (Quit: amachronic)
11:30:26***Saving seen data "./dancer.seen"
11:39:50 Quit dconrad (Remote host closed the connection)
11:41:59 Join dconrad [0] (~dconrad@
11:46:45 Quit dconrad (Ping timeout: 272 seconds)
11:48:56 Join dconrad [0] (~dconrad@
12:19:56rb-bluebotBuild Server message: New build round started. Revision 756c0d2ac8, 297 builds, 9 clients.
12:27:20jschwartwhat is the largest flash drive that works with an iPod mini?
12:28:34rb-bluebotBuild Server message: Build round completed after 519 seconds.
12:28:38speachyin theoy anything up to 2TB.
12:28:38rb-bluebotBuild Server message: Revision 756c0d2ac8 result: All green
12:28:54speachyat least from rockbox's perspective. no idea what sort of upper limit the Apple firmware has
12:29:17jschwartspeachy: interesting
12:29:39jschwartI have a CF to SD adapter now, that someone claimed worked for their ipod
12:30:53speachymake sure you're using current nightly/dev builds rather than the 3.15 release.
12:31:33jschwartah why is that?
12:32:01speachymasty bugs due to the iflash adapters not being quite compliant with the ATA spec.
12:32:44speachy(ie data corruption)
12:33:18jschwartah I see, I was already struggling lately to put stuff on it, but another cable solved that
12:36:56jschwartas for the brand here, I really have no clue, Patriot, Kingston, Team Group, ADATA, Transcend, Goodram, Samsung, Lexar, Silicon Power, Kioxia, Intenso
12:37:03jschwartthey all seem in the same price range
12:37:58jschwartfew euro more it's Toshiba, Sandisk
12:38:09jschwartEmtec, Verbatim
12:38:16jschwartcrazy market it seems
12:39:59jschwartsimply buying a compact flash card is also way more expensive than an sd card plus adapter
13:17:21speachyyeah, classid CF cards are mostly gone now, save for industrial use
13:20:47jschwartI'll just get the Samsung one at 256GB as the brand is known and it is the cheapest
13:21:16jschwartif it doesn't work I think I can probably swap it with a Sandisk that's 128GB that I use for my Tinker Board
13:30:27***Saving seen data "./dancer.seen"
13:59:55 Join TheLemonMan [0] (~lemonboy@irssi/staff/TheLemonMan)
14:33:31 Quit dconrad (Remote host closed the connection)
14:42:39 Quit j-r (Ping timeout: 244 seconds)
14:43:17 Join j-r [0] (
15:08:47 Quit TheLemonMan (Quit: "It's now safe to turn off your computer.")
15:30:31***Saving seen data "./dancer.seen"
15:31:33 Join parity [0] (~ash@user/parity)
15:33:48 Quit emacsoma1 (Quit: WeeChat 3.1)
15:34:51 Join emacsomancer [0] (
15:47:45 Quit rasher (Changing host)
15:47:45 Join rasher [0] (~rasher@user/rasher)
16:47:22 Join dconrad [0] (~dconrad@
16:53:50 Quit lebellium (Quit: Leaving)
17:02:24 Join ufdm_ [0] (
17:04:16 Join skipwich_ [0] (~skipwich@user/skipwich)
17:04:39 Quit kadoban (Read error: Connection reset by peer)
17:04:41 Quit blbro[m] (Read error: Connection reset by peer)
17:05:33 Join kadoban [0] (~kadoban@user/kadoban)
17:05:34 Join ufdm__ [0] (
17:05:38 Join speachy1 [0] (~speachy@
17:05:57 Quit skipwich (Ping timeout: 245 seconds)
17:05:57 Quit speachy (Ping timeout: 245 seconds)
17:05:57 Quit ufdm (Ping timeout: 245 seconds)
17:05:57 Join asabas [0] (~asabas@
17:06:41 Quit asaba (Ping timeout: 252 seconds)
17:06:41 Quit pablocastellanos (Ping timeout: 252 seconds)
17:07:09 Join pablocastellanos [0] (~pidgin@user/pablocastellanos)
17:08:25 Join scorche [0] (
17:08:47 Nick scorche is now known as Guest9984 (
17:09:25 Join ufdm [0] (
17:09:52 Join blbro[m] [0] (~blbrostra@2001:470:69fc:105::8f7)
17:11:21 Join skipwich [0] (~skipwich@user/skipwich)
17:11:34 Join ParkerR_ [0] (
17:11:35 Join parity_ [0] (
17:11:38 Join wolfshappen_ [0] (
17:11:50 Join speachy [0] (~speachy@
17:11:50Mode"#rockbox +v speachy" by ChanServ (
17:12:05 Join gevaerts_ [0] (~fg@user/gevaerts)
17:12:09 Join rudi_s_ [0] (~simon@user/rudi-s/x-7673890)
17:12:49 Quit ufdm_ (Ping timeout: 272 seconds)
17:12:49 Quit Bilgus (Ping timeout: 272 seconds)
17:14:46 Quit ParkerR (Ping timeout: 264 seconds)
17:14:46 Quit Galois (Ping timeout: 264 seconds)
17:14:47 Quit wolfshappen (Read error: Connection reset by peer)
17:14:47 Quit speachy1 (Ping timeout: 264 seconds)
17:14:47 Quit parity (Ping timeout: 264 seconds)
17:14:47 Quit tchan (Ping timeout: 264 seconds)
17:14:48 Quit Guest2644 (Ping timeout: 264 seconds)
17:14:48 Quit ufdm__ (Ping timeout: 264 seconds)
17:14:48 Quit gevaerts (Ping timeout: 264 seconds)
17:14:48 Quit michaelni (Ping timeout: 264 seconds)
17:14:48 Quit rudi_s (Ping timeout: 264 seconds)
17:14:48 Quit skipwich_ (Ping timeout: 264 seconds)
17:14:49 Quit Romster (Ping timeout: 264 seconds)
17:15:20 Join michaelni [0] (
17:15:24 Join ufdm_ [0] (
17:19:47 Quit gsora (Ping timeout: 265 seconds)
17:20:14 Quit XDjackieXD (Ping timeout: 265 seconds)
17:22:59 Quit Natch (Ping timeout: 265 seconds)
17:23:24 Quit CasBot (Ping timeout: 265 seconds)
17:25:28 Join Romster [0] (~romster@user/romster)
17:25:29 Join asaba [0] (~asabas@
17:25:29 Join gsora_ [0] (~gsora@
17:25:29 Join Guest2644 [0] (
17:25:30 Join jackie [0] (
17:25:30 Join CasBot1 [0] (
17:25:31 Quit ufdm (*.net *.split)
17:25:31 Quit Guest9984 (*.net *.split)
17:25:31 Quit pablocastellanos (*.net *.split)
17:25:31 Quit asabas (*.net *.split)
17:27:12 Join pablocastellanos [0] (~pidgin@user/pablocastellanos)
17:28:15 Join Natch [0] (
17:30:35***Saving seen data "./dancer.seen"
17:38:23 Join tchan [0] (
17:43:23 Nick rudi_s_ is now known as rudi_s (~simon@user/rudi-s/x-7673890)
18:53:19 Join ddevault [0] (znc@sourcehut/staff/ddevault)
19:06:23 Join Galois [0] (
19:30:37***No seen item changed, no save performed.
20:24:19 Quit dconrad ()
21:30:40***Saving seen data "./dancer.seen"
21:31:00 Join Bilgus [0] (~WW@
23:30:43***No seen item changed, no save performed.

Previous day | Next day