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 2017-10-09

00:23:08 Quit Halamix2 (Quit: Leaving)
00:30:37 Quit lebellium (Quit: ChatZilla 0.9.93 [Firefox 56.0/20170926190823])
00:30:41 Join Strife1989 [0] (
00:34:14 Quit Strife89 (Ping timeout: 255 seconds)
00:45:37 Quit ender` (Quit: The sum of intelligence on the planet is a constant; the population is growing.)
01:06:49 Join krabador_ [0] (~krabador@unaffiliated/krabador)
01:07:16 Quit krabador_ (Remote host closed the connection)
01:09:35 Join krabador [0] (~krabador@unaffiliated/krabador)
01:23:15 Quit Bilgus_PH (Quit: Page closed)
01:46:23***Saving seen data "./dancer.seen"
01:47:26 Quit krabador (Quit: Leaving)
03:00:10 Quit man_in_shack (Read error: Connection reset by peer)
03:01:10 Join man_in_shack [0] (~chat@unaffiliated/man-in-shack/x-4279753)
03:03:59 Quit Moarc (Ping timeout: 248 seconds)
03:06:30 Quit _meg (Ping timeout: 260 seconds)
03:07:54 Join _meg [0] (~notsure@
03:08:47 Join Moarc [0] (
03:46:25***Saving seen data "./dancer.seen"
05:30:12 Quit Ruhan (Quit: Connection closed for inactivity)
05:46:28***Saving seen data "./dancer.seen"
06:05:58 Quit Strife1989 (Ping timeout: 240 seconds)
06:06:08 Quit TheSeven (Ping timeout: 258 seconds)
06:06:57 Join Strife89 [0] (
06:09:02 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
06:29:31 Quit TheSeven (Ping timeout: 258 seconds)
06:30:02 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
06:56:46 Quit Jinx (Ping timeout: 246 seconds)
07:09:09 Join wodz [0] (
07:12:15 Quit alexweissman (Remote host closed the connection)
07:15:36 Join alexweissman [0] (
07:46:32***Saving seen data "./dancer.seen"
07:49:15 Join deevious [0] (~Thunderbi@
08:01:02 Quit _meg (Ping timeout: 248 seconds)
08:04:45 Join _meg [0] (~notsure@
08:08:02 Quit alexweissman (Remote host closed the connection)
08:20:36 Join ender` [0] (
08:40:54 Join PimpiN8 [0] (~textual@2a02:a454:38ea:1:cdca:b:ffe0:ac4e)
08:45:16 Quit olspookishmagus (Quit: All for nothing)
08:49:15 Join The_Prospector|2 [0] (
08:51:34 Quit The_Prospector (Ping timeout: 246 seconds)
08:55:09 Join olspookishmagus [0] (
08:58:42 Join petur [0] (~petur@rockbox/developer/petur)
09:00:37 Join dys [0] (
09:03:44 Join p3tur [0] (~petur@
09:03:44 Quit p3tur (Changing host)
09:03:44 Join p3tur [0] (~petur@rockbox/developer/petur)
09:03:44 Quit petur (Read error: Connection reset by peer)
09:30:51 Nick p3tur is now known as petur (~petur@rockbox/developer/petur)
09:46:34***Saving seen data "./dancer.seen"
09:51:14 Quit PimpiN8 (Quit: My MacBook has gone to sleep. ZZZzzz…)
10:04:33 Join almog1006 [0] (4d8bf186@gateway/web/freenode/ip.
10:07:03 Join almog1006_ [0] (4d8bf186@gateway/web/freenode/ip.
10:07:07 Quit almog1006_ (Client Quit)
10:12:14 Join prof_wolfff [0] (
10:30:06 Join almog [0] (4d8bf186@gateway/web/freenode/ip.
10:30:51almog1006I extracted FW file by "atjboottool" tool and the output file is afi/sql file. how can i unpack/repack this sql file?
10:35:00 Quit dys (Ping timeout: 255 seconds)
10:35:43wodzalmog1006: afi is *not* sql file
10:37:14almog1006I can just extract it to afi file or sql file
10:37:48wodzmind sharing the fwu? I might look at it
10:37:56wodzwhat player it is for?
10:38:05almog1006ruizu X26
10:38:36almog1006look at this topic that i opened:,52004.0.html
10:39:13almog1006link to the sql file:
10:40:33almog1006link to the fw before unpack:
10:42:23wodzAh, ok so it is sql actually. Weird. Search for sqlight managers to retrieve resources.
10:43:34wodzalmog1006: This might be useful as well
10:44:54almog1006And how i can repack it back to FW file?
10:45:17 Join PimpiN8 [0] (~textual@
10:45:44wodzalmog1006: You can't. Our tool is currently one way only. We may support repacking in the future but not now.
10:47:15almog1006Sorry but what do you mean in the future?
10:50:14wodzalmog1006: According to dictionary "the time or the events that will come after the present"
10:51:36almog1006My question is whether it's a matter of days or a few weeks
10:52:08wodzalmog1006: Can't answer that. Pamaury is the author of the tool.
10:57:30 Quit prof_wolfff (Ping timeout: 255 seconds)
11:27:27 Join TheLemonMan [0] (~lemonboy@irssi/staff/TheLemonMan)
11:36:14 Join p3tur [0] (~petur@rockbox/developer/petur)
11:36:15 Quit petur (Read error: Connection reset by peer)
11:46:38***Saving seen data "./dancer.seen"
11:49:53 Join petur [0] (~petur@rockbox/developer/petur)
11:53:13 Quit p3tur (Ping timeout: 246 seconds)
12:26:15 Part robertd1
12:39:43 Join robertd1 [0] (~root@
13:09:08 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
13:16:33pamauryalmog1006: for repack for fwu firmware, I don't know, I'm working on it but ot
13:16:54pamauryit's not so easy because the encryption is complicated, I need to find out how to decrypt
13:17:23pamauryfor sql dumping it's fairly trivial, I'll add code to my github online, at least for unpacking
13:34:42wodzpamaury: I can't get mmu to work. When I test in hwstub basic operations (nuking tlb entries, adding static mapping for dram) it seems to work. But when I do the same in crt0.S and link hwstub in vma it fails randomly (i.e it mostly works but read32_cop doesn't). Rockbox even doesn't boot linked into vma
13:35:33wodzpamaury: I must be missing something very basic
13:36:29pamaurywodz: are you sure the linking script is correct? Could it be a cache problem?
13:37:24wodzpamaury: I nuke cache just after initialization of mmu and before jumping into vma so I think this should be ok.
13:38:29wodzpamaury: and there is no really much to change in hwstub linker script.
13:38:58pamauryare you sure the stack address is right?
13:39:21wodzwill check
13:46:39***Saving seen data "./dancer.seen"
13:47:18wodzpamaury: so stack address looks ok
13:55:45wodzpamaury: This is with crt0.S and target-config.h changed to link at 0xc1000000
13:58:36pamaurywodz: I'm busy right now but I'll have a look. I might have a try at making hwstub with mmu on the fiio tonight
14:14:43wodzpamaury: hmm, I reverted all my changes to hwstub and cop operations still fail. I did test it this morning and it was all working with vanila hwstub :/
14:19:28 Join prof_wolfff [0] (
14:27:39 Quit jhMikeS (Ping timeout: 240 seconds)
14:30:47 Join jhMikeS [0] (
14:33:03 Quit Strife89 (Ping timeout: 248 seconds)
14:34:57 Quit jhMikeS (Ping timeout: 240 seconds)
14:40:06 Quit Huntereb (Ping timeout: 240 seconds)
14:40:38 Join Huntereb [0] (
14:50:51pamauryalmog1006: you can use pamaury/pa-tools/tree/master/qsql/dump"> to dump the sql firmware. Build qsql_view (using qmake and make, you'll Qt and sqllite probably), then run:
14:50:51pamauryqsql_dump X26.afi X26/
14:50:51pamauryThis will extract all files to X26/ directory. Note that the SQL firmware contains much more information than files, it's quite complicated actually. You can use
14:50:51DBUGEnqueued KICK pamaury
14:50:51pamauryto view the rest of the file (not sure that's necessary though):
14:50:53pamauryqsql_view X26.afi
14:55:16 Join amayer [0] (
14:58:22wodzpamaury: Ok, figured out why cop read was failing. It will not work when CONFIG_FLUSH_CACHES is not defined.
15:01:31pamaurywodz: why is that?
15:01:41pamauryah yes of course
15:01:45wodzpamaury: :-)
15:02:00pamaurybecause the only way to read the cop is to generated the code on the fly
15:02:38pamaurymaybe add a comment about that in the source? Or a warning like if CONFIG_FLUSH_CACHES is not defined print a #warning that needs to be explicitely disable per target?
15:02:49pamaurybecause that's sneaky
15:02:58wodzpamaury: you could use uncached alias as well
15:04:01pamaurymmmh, the thing is I wanted to make as few assumptions as possible, for example on the JZ there is this weird situation with the TCSM
15:04:14pamaurybut since I ended up working about it, you are right that this might be an option
15:07:03pamauryactually I'm using the TCSM on JZ so it's a problem
15:08:24pamauryI suspect this could be fixed easily with some linker magic, or some macro to compute the uncached address
15:08:55pamaurydon't know what is best between flushing caches and some uncached address
15:09:29 Quit amayer (Quit: Leaving)
15:09:46 Join amayer [0] (
15:10:17 Quit amayer (Client Quit)
15:10:40 Join amayer [0] (
15:12:25wodzpamaury: I'd say what is easier. This is debugging tool, speed is not important. So adding preprocessor warning and disabling cop code when CONFIG_FLUSH_CACHES is not defined.
15:21:25pamaurywodz: ok sounds good, do you want me to add or you do it?
15:21:55wodzpamaury: do the honor, its your code
15:25:10 Join Strife89 [0] (
15:27:30 Quit wodz (Ping timeout: 255 seconds)
15:40:44almog1006pamaury: Should I unpack the fw to sql or to afi file?
15:41:10almog1006pamaury: And how can I repack the files back to the afi/sql file?
15:43:08pamauryalmog1006: it doesn't matter the extension of the file. Use the qsql_dump tool on the file produced by atjboottool (it has .afi extension by default, but it's not necessarily an afi file, in your case it's an sql fil).
15:43:11almog1006pamaury: I mean after dumping them as you explained
15:43:12pamauryYou can't repack for now
15:43:16 Join alexweissman [0] (
15:45:17pamaurybut for now you don't need to repack because you don't even know how to change the translation, you'll have to figure it out by yourself, find the right file and how to change it.
15:46:42***Saving seen data "./dancer.seen"
15:48:03 Quit alexweissman (Ping timeout: 246 seconds)
15:49:52almog1006pamaury: In my humble opinion, this does not seem to me to be phase I am supposed to worry about
15:49:57almog1006pamaury: I am concerned about the re-packaging phase
15:49:59 Join alexweissman [0] (
15:51:31almog1006pamaury: (Both re-packaging back to the sql file and also from sql file to fw)
16:11:35pamauryalmog1006: you still can start working on finding the right file. As I said repacking is not entirely trivial, for the sql file I think it's not hard but for the encryption firmware, there are couple things in the format that are unclear, I can't solve it in one hour
16:21:27 Join Ruhan [0] (uid76353@gateway/web/
16:32:42 Join krabador [0] (~krabador@unaffiliated/krabador)
16:36:56 Quit krabador (Remote host closed the connection)
16:39:58 Join krabador [0] (~krabador@unaffiliated/krabador)
16:43:58 Quit PimpiN8 (Quit: My MacBook has gone to sleep. ZZZzzz…)
16:43:59 Quit krabador (Client Quit)
16:45:10 Join PimpiN8 [0] (~textual@
17:07:17 Quit PimpiN8 (Quit: My MacBook has gone to sleep. ZZZzzz…)
17:32:50 Quit petur (Quit: Connection reset by beer)
17:36:23 Join krabador [0] (~krabador@unaffiliated/krabador)
17:46:44***Saving seen data "./dancer.seen"
17:53:12 Join dys [0] (
18:31:03 Quit Ruhan (Quit: Connection closed for inactivity)
18:35:46 Join Halamix2 [0] (~Halamix2@unaffiliated/halamix2)
18:37:03 Quit TheLemonMan (Quit: "It's now safe to turn off your computer.")
19:04:55 Quit pamaury (Ping timeout: 260 seconds)
19:19:48 Join petur [0] (~petur@rockbox/developer/petur)
19:43:46 Quit Halamix2 (Quit: Leaving)
19:44:11 Join Ruhan [0] (uid76353@gateway/web/
19:44:42 Join thomasjfox [0] (~thomasjfo@rockbox/developer/thomasjfox)
19:46:21 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
19:46:45***Saving seen data "./dancer.seen"
19:50:15 Quit _meg (Ping timeout: 246 seconds)
19:52:24 Join _meg [0] (~notsure@
19:56:00 Join lebellium [0] (
20:11:27 Quit usvi (Quit: Lost terminal)
20:16:53 Quit almog (Quit: Page closed)
20:25:19 Join jhMikeS [0] (
20:44:06 Join b0hoon [0] (1f007f70@gateway/web/freenode/ip.
20:45:22b0hoonpamaury: ping
20:46:08pamauryb0hoon: pong
20:47:08b0hoonpamaury: hi, my machines don't want to do any job for sony, i have arm-rb-gcc494 in my archlist in
20:48:01b0hoonit's impossible, i don't know why
20:49:17b0hoonany idea?
20:49:34pamauryb0hoon: does your machine do other jobs?
20:49:49 Join johnb5 [0] (
20:50:01b0hoonb0hoon: well, let me check...
20:50:09b0hoonpamaury: ^
20:50:35 Quit johnb5 (Client Quit)
20:50:48pamauryb0hoon: is it vm1-b0hoon and vm2-b0hoon?
20:50:49b0hoonpamaury: yes, sdl, arm-eabi
20:52:04pamauryand you are sure there is no typo?
20:52:16pamaurymaybe the build server don't like your client :-p
20:52:43pamaurymy clients have -archlist=arm-rb-gcc494,arm-eabi-gcc444
20:54:19b0hoonit's like this => -archlist=m68k-gcc452,sh,mipsel,sdl,arm-eabi-gcc444,arm-rb-gcc494
20:55:00b0hoonpamaury: maybe it doesn't like me, yes
20:55:16pamaurydid you stop and rerun the client since you changed the list?
20:55:43b0hoonpamaury: yes, sure
20:56:16b0hooni'll check processes list again
20:57:34b0hoonperl is running the with this line
20:58:15b0hooni just don't know, maybe it's to long or something
20:59:11b0hooni will try to make a shorter list like yours
21:01:47 Join JanC_ [0] (~janc@lugwv/member/JanC)
21:03:04 Quit JanC (Killed ( (Nickname regained by services)))
21:03:04 Nick JanC_ is now known as JanC (~janc@lugwv/member/JanC)
21:20:55 Join maffe [0] (
21:20:58 Quit maffe (Client Quit)
21:34:39b0hoonha! gerrit created for me something like a second profile, but i don't have a developer rights, although my ssh keys are still present :/
21:35:06gevaertsYes, it did that for most people
21:41:06pamauryb0hoon: you can still do most gerrit stuff through the gerrit ssh command line
21:41:10pamauryit's a bit obscure though
21:41:41b0hoongevaerts: is there anything i can do about it?
21:42:28gevaertsHmmm, if I can use that to assign users to groups, we could climb out of this...
21:42:35b0hoonpamaury: ok, i will try it, thanks
21:43:55CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
21:43:55*gevaerts experiments a bit
21:44:49pamaurygevaerts: you can assign users to group using command line, althoug it's extremely clumsy, but I don't think it solves the problem :-/
21:46:04gevaertsYes and no
21:46:48***Saving seen data "./dancer.seen"
21:47:19gevaertsI mean, if I'm an admin again, that we have an admin again, and we can consider migrating to the "new" set of users, or at least give them some rights
21:47:38gevaertsThat's definitely far from ideal, but at least it's an option
21:48:08gevaertsI'd prefer finding a way to fix the issue, but that will also require admin anyway
21:48:38pamaurygevaerts: no I mean I don't think gerrit really has two sets of users
21:49:12pamauryat least not two sets that you can distinguish from each other
21:51:58*pamaury tries
21:52:45pamaurygevaerts: you might be right actually:
21:52:56pamaury> git gerrit-cli ls-members 'Rockbox\ Committers' | grep -i amaury
21:53:03pamaury1000040 pamaury Amaury Pouly
21:53:16pamaurybut on the gerrit website my ID is 1000536
21:53:28*pamaury adds the second himself to committers
21:54:44gevaertsWhat I *don't* know is if gerrit tracks changes by numeric id or by email address
21:55:24gevaertsActually I think I do know that. "My changes" doesn't list anything for the 'web' user
21:55:37pamaury> git gerrit-cli set-members 'Rockbox\ Committers' −−add 1000536
21:55:37pamauryfatal: Cannot add member: Amaury Pouly
21:55:41pamaurynot sure why thouh
21:55:58pamaurygevaerts: I think by ID
21:56:12gevaertsLet me try that command
21:56:20gevaertsMembers added to group Rockbox Committers:
21:56:25gevaertsAdmin stuff?
21:56:54b0hoonpamaury: heh, now you are twice on the list
21:57:09b0hoonwith the same email
21:57:24pamauryweird, so it worked but returned an error ?!
21:57:43pamaurygevaerts: oh maybe I don't have enough rights
21:57:45 Quit man_in_shack (Read error: Connection reset by peer)
21:57:54gevaertsYes, I suspect that's the issue
21:58:01*pamaury feels superpowerful again
21:58:25 Join man_in_shack [0] (~chat@unaffiliated/man-in-shack/x-4279753)
21:58:29pamauryit would be better if we could merge those identities though...
21:59:03gevaertsYes, I want to see if there's enough to achieve that
21:59:18gevaertsI suspect not without database access
21:59:56pamaurygevaerts: the command line seems to provide some sql access but not sure how powerful
22:00:03pamauryI can't try it, it's reserved for admin
22:00:04b0hoonyeah with database access it's probably possible
22:00:19gevaertspamaury: maybe I should make you an admin too
22:01:03pamauryalthough I suspect we want to make a database backup before even trying
22:01:35pamauryit only requires the ssh admin access as you can see
22:01:38 Join diox [0] (
22:02:20pamauryit seems simple enough though, very low risk even
22:03:35gevaertsOh, good. gerrit gsql gives me a nice interactive H2 prompt
22:03:41gevaertsNow I get to learn H2 stuff :)
22:04:22pamaurygevaerts: maybe try the script above then, try to merge someone with himself
22:04:32pamaurybut not you because your are our link to the database ;)
22:04:58gevaertsYes, I know. I'd like a backup admin, really :)
22:05:07b0hoonyou can try on me
22:05:46gevaertsRight now I'm checking if all tables used in that script actually exist
22:06:37gevaertsHmmm, things differ apparently
22:08:01pamauryI can't imagine why gerrit doesn't have such functionality built-in really, this is like THE feature everyone complains about
22:08:58gevaertsWhat I'm *pretty* sure I can do (but haven't tested yet) is resetting remote auth stuff so the old account gets the new google (or whatever) link
22:09:56 Quit diox (Ping timeout: 264 seconds)
22:10:57 Quit krabador (Quit: Leaving)
22:11:01gevaerts are the tables I see
22:12:29*gevaerts is trying to find the ssh keys
22:12:30pamaurygevaerts: the patch suggest it's sufficient to look in all tables and change the IDs
22:14:29gevaertsWait, there's a link to an updated version
22:15:15gevaertsThat looks more like what I'm seeing
22:16:32*gevaerts makes a backup
22:17:11pamaurygevaerts: lol, it's like the first line but I missed it
22:17:29pamaurysounds like what we need
22:19:09 Join diox [0] (
22:23:12gevaertsOK, I can try with b0hoon. Am I right that we want to $KEEP the old account and $MERGE the new one?
22:27:12gevaertspamaury: what do you think?
22:28:16b0hoonUPDATE something SET account_id=$KEEP WHERE account_id=$MERGE; yes i guess...
22:29:27gevaertsb0hoon: you'll end up with the "old" email (non-gmail) address as the main one (unrelated to auth I think). Is that OK?
22:29:42b0hoongevaerts: yes
22:30:09b0hoonif it's unrelated to the auth like you say
22:32:11gevaertsb0hoon: ok, might be done
22:32:14gevaertsCan you check?
22:35:25b0hoonyes give me a second
22:38:24b0hoonwhen i try to login through the google account i get a Forbidden page :/
22:39:03*gevaerts has a look
22:40:03gevaertsI'm going to clear out all but the "good" ACCOUNT_EXTERNAL_IDS data for your account
22:42:11gevaertsb0hoon: can you try again?
22:44:10b0hoonthe same thing
22:46:30gevaertsDoes ssh access work?
22:47:27gevaertsI suspect it's a matter of email addresses not matching
22:48:09pamaurygevaerts: yes we want to merge the new one
22:48:22gevaertsBased on the web UI, you *can* have multiple email addresses, but I suspect there might be an issue with a "linked account" having an email address that isn't one of your known addresses
22:48:38 Join diox_ [0] (
22:49:53b0hoongevaerts: yes, cherry picking is working for me
22:49:56gevaertsb0hoon: can you try again?
22:50:44 Quit diox (Ping timeout: 264 seconds)
22:50:52b0hoonForbidden again
22:52:59b0hoonmaybe my name is the reason i have a Huawei Zte name in the google profile :P
22:53:10b0hooninstead of Szymon Dziok
22:53:51gevaertsHave you tried clearing cookies for Maybe it's something there?
22:54:14b0hooni didn't to be honest, wait i will do that
22:55:06gevaertsYour account data looks very similar to my "web" account data (apart from groups and things like that), so I'd expect things to work
22:57:59b0hoonForbidden again, cleared firefox
22:58:36__builtinjhMikeS: (re: g#1664) I misunderstood what you meant when saying to change the enums
22:58:39fs-bluebotGerrit review #1664 at : Wolfenstein 3D / Duke Nukem 3D / Quake 3-in-One Mega-Patch by Franklin Wei
23:00:39__builtinrather than change the enum typedefs, I thought you meant to change each reference to them in the code
23:00:49__builtinthe former is much more doable
23:00:56*gevaerts isn't sure what to try next
23:02:28b0hoonchanging the name in the google profile din't help
23:03:38b0hoonoh, after refreshing the page somethimes there is a "server error" instead "forbidden" :O
23:03:50gevaertsOn the gerrit server?
23:04:36gevaertsI'm going to merge my accounts. If ssh access survives, I'll be fine
23:06:55gevaertsOK, same symptoms
23:07:38pamauryso you can access with ssh but not web?
23:09:26*pamaury doesn't know much about oauth and stuff
23:09:47pamauryis there a support channel for gerrit?
23:16:10 Quit petur (Remote host closed the connection)
23:17:06 Quit amayer (Quit: Leaving)
23:19:55gevaertsAlthough that doesn't make me understand the issue of course
23:20:09pamauryso maybe we need to update?
23:20:43gevaerts suspect so
23:21:12pamaurywe need the swedes for this
23:21:15*pamaury goes to bed
23:21:33 Quit utrack (Ping timeout: 248 seconds)
23:22:00 Quit lebellium (Quit: ChatZilla 0.9.93 [Firefox 56.0/20170926190823])
23:24:55*gevaerts agrees
23:24:59gevaertsThis is for tomorrow...
23:25:26 Quit pamaury (Ping timeout: 240 seconds)
23:25:29gevaertsb0hoon: if you need access today, I can clear out the google link, which will allow you to create another new account
23:26:18b0hoongevaerts: no, no rush, i have to go to, thanks for everything
23:27:29 Part b0hoon
23:30:05 Quit thomasjfox (Quit: Konversation terminated!)
23:40:31 Join utrack [0] (~utrack@unaffiliated/utrack)
23:42:34__builtinpamaury (logs): #gerrit
23:46:52***Saving seen data "./dancer.seen"
23:55:18 Join krabador [0] (~krabador@unaffiliated/krabador)

Previous day | Next day