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 2016-09-23

00:06:55lebelliumpamaury: I confirm it doesn't work for me. I just tried again. I don't know if that's due to different hardware revision or if you have several bootloader releases wrongly named "v2"
00:07:57 Join ruhans [0] (uid76353@gateway/web/
00:07:58pamauryas far as I remember, I only made two releases so I doubt there would be several v2
00:08:32lebelliumI mean, maybe you made a test bootloader and named it v2. And you think you use the same bootloader as the released one
00:08:41pamauryah good point
00:08:47 Join ulmutul [0] (
00:08:51pamauryLet me try the official v2 on mine
00:14:07 Quit ZincAlloy (Quit: Leaving.)
00:15:51pamaurydamn, the OF doesn't want to flash it
00:15:55 Part b0hoon ("GTG... Bye.")
00:16:22lebelliumI guess the OF only flash it if you use the USB from OF and not Rockbox
00:16:49pamauryquite possible actually
00:16:59pamaurythere must be a flag set by the usb code
00:18:59 Join lebellium_z3c [0] (
00:19:05 Quit lebellium (Quit: ChatZilla 0.9.92 [Firefox 49.0/20160916101415])
00:20:22 Quit ender` (Quit: Perform mass upper blepharectomy and nobody bats an eye. Perform a mass encephalectomy and everyone loses their mind!)
00:21:06pamaurylebellium_z3c: it works
00:21:56lebellium_z3cI don't understand how that can be
00:23:01lebellium_z3cIs it possible that the file generated by RButil is not exactly the same?
00:25:29lebellium_z3cI tried again
00:25:35lebellium_z3cAnd now it worked
00:25:42lebellium_z3cThis makes me crazy
00:28:35 Quit edhelas (Ping timeout: 250 seconds)
00:31:17lebellium_z3cI'm sure it didn't work although I got the confirmation message on screen
00:31:52lebellium_z3cMaybe a specific sequence makes it not working
00:32:10lebellium_z3cI'll investigate further tomorrow
00:32:29lebellium_z3cThank you and good night
00:33:05 Quit lebellium_z3c (Quit: Bye)
00:37:32[Saint]vifino: sorry, I had to run off for a bit.
00:37:32[Saint]vifino: ideally, the majority of the additions will be entirely transparent to the user, but there are a few forward facing additions you might notice directly or indirectly.
00:37:56[Saint]One very obvious one is that it is now almost impossible to deep discharge the device and get into a boot loop cycle.
00:39:21[Saint]The bootloader will manage detection of a deeply discharged device and will withhold the boot process (and hence disk spinup - which will cause a brownout, and start a vicious bootloop cycle) until the device has charged enough that we know with absolute certainty that it can complete a boot.
00:40:02 Join idonob [0] (~Owner@
00:40:17[Saint]Another forward facing aspect, but one that is largely transparent to the user, is the correct setup of the hardware so that the hardware is set up correctly at a very low level from boot.
00:40:47[Saint]Most users don't seem to use any accessories or advanced feature sets of these devices, though.
00:41:20[Saint]Another is USB boot failure recovery being actually reliable and consistent now.
00:42:35[Saint]If the Rockbox binary is damaged or missing, you can mount the storage volume from the bootloader directly and repair it.
00:43:16[Saint]Previously, this "worked", but was wildly unreliable on Windows systems, and even for some *nix systems for some users (but I myself never had any issues).
00:44:19[Saint]In that case, prior to the dualboot bootloader, if a user could not get the bootloader to mount the storage volume it was basically impossible to recover the device without a full iTunes restore.
00:44:52[Saint](because the user couldn't boot the original firmware or disk mode)
00:45:49[Saint]Now the user has the option of delegating mounting damaged storage to the original firmware or disk mode, or from the bootloader directly, without having to rely in any way on the Rockbox binary itself.
00:47:26[Saint]The changes users will actually notice are:
00:47:26[Saint] -the ability to boot the original firmware, disk mode, diagmode (original firmware image for diagnostics), and
00:47:26[Saint] -deep discharge mitigation
00:48:48[Saint]Functionally, in terms of actually booting Rockbox, there is no difference.
00:55:38ulmutulb0hoon: I'm not fully convinced of declaring HDD6330 to stable, because on some targets (including mine) there are problems with disk access.
00:55:48 Join alexweissman [0] (
00:56:43ulmutulHere is a discussion about it:
00:57:05vifino[Saint]: Sounds like a very nice upgrade, thank you very much for both guiding me through the upgrade and explaining what changed!
00:57:05ulmutuland here:
00:57:50[Saint]vifino: Not a problem my man.
00:58:10[Saint]vifino: if you have any more questions I would be happy to help, assuming I can answer them.
00:58:43vifinoI've noticed the boot loop when I dug up my iPod, solved it by putting the iPod on hold, which prevented it from booting, waiting for it to charge a little.
00:59:07[Saint]vifino: One fairly big difference in the new bootloader is that by extension of the fact that it is significantly more modern and has had many longstanding emCORE bugs addressed and absorbed in to it, it is more widely compatible with aftermarket storage additions.
00:59:25vifinoThat is very good!
00:59:36[Saint]I very highly recommend the 'iFlash' microSD adapter RAID cards from Tarkan.
00:59:52vifinoI'll put that to the test in a few weeks once my kit arrives :P
01:00:36[Saint]You can put in ~256GB (or more) of solid state storage in this device for around $100 USD
01:00:48vifinoI bought a thick cover, "3000mah" battery replacement and a 200gb microsd card, didn't order the iFlash Quad yet.
01:00:59[Saint]Have you already ordered the iFlash, or are you using some other media?
01:01:03[Saint]Ah. I see.
01:01:30vifino[Saint]: Currently, a SanDisk 200gb MicroSD card is on sale on Amazon for 60€, which I absolutely had to buy.
01:01:41[Saint]Yeah, I use an iFlash Quad with four 64GB SanDisk Extreme Pro Backflip Super L337+++ cards.
01:01:52vifinoSeeing as 256gb is 150€, it really made no sense to buy a bigger one.
01:02:19[Saint]I just had a bunch of the 64GB Extreme Pro cards from SanDisk on hand at the time.
01:02:23[Saint]So I used what I had.
01:02:51vifinoI may add another 200gb card, but for now, I think upgrading the 80gb hdd with just a single card is gonna be a very good thing.
01:03:23vifinoWhat model/generation do you have, if I may ask, [Saint]?
01:04:35[Saint]I have at least two of each.
01:04:46[Saint]Most of them in various states of modification.
01:05:22[Saint]Your 3000mAh battery is almost assuredly completely full of shit, just so you know.
01:05:49[Saint]It's pretty much impossible to get more than 2Ah of energy density in there.
01:05:50vifinoI'm happy if it will be half the capacity.
01:06:08[Saint]And even 2Ah is a massive stretch of the imagination.
01:06:39[Saint]I'd be very surprised if it wasn't just the .85Ah battery from the CEATA 160GB model.
01:06:51[Saint]Very, very surprised.
01:06:59vifinoNope, they are much bigger.
01:07:18[Saint]Hmmm, interesting.
01:07:34vifinoThe reason I need a thick case back is because the battery takes the space :)
01:07:53 Quit girafe (Read error: Connection reset by peer)
01:07:58vifinoIt also only works for the thin models, obviously with the thick backs.
01:08:10[Saint]I have quite a few modified iPods of varying generation and model.
01:09:12[Saint]My favorite is a solid state converted iPod Colour that has a 512GB CF card in it, and the CF adapter plate I used also offered a microUSb connection.
01:09:34[Saint]So I broke out the microUSb port to the exterior of the case and made it so that it can also handle charging the device.
01:10:00[Saint]So I can use either the ridax port (the usual iPod connector) or a microUSb for addressing storage and charging.
01:10:05vifino[Saint]: This is the one I ordered, as you can see in the picture, the battery is much bigger.
01:10:44[Saint]I have a few solid state converted iPod Videos, as well.
01:11:13vifinoThat reminds me, I have a question. Which one sounds the best, of all the iPods capable of running rockbox?
01:11:42[Saint]One 512GB SSD, one 256GB CF, and one 800GB iFlash/microSD.
01:12:05[Saint]"best" is entirely subjective.
01:12:39 Join munch [0] (pls@gateway/shell/elitebnc/x-qahnhzeiuircqrsi)
01:12:51[Saint]I can say that if you ignore "best" and take it as "which iPod has the most accurate signal reproduction under load" the anser is most certainly the iPod Classic.
01:13:16vifinoI just read quite a bit about the 5.5g sounding better than the 6g and up, allegedly because of the switch from a wolfson DAC to a cirrus one.
01:13:34vifinoAwesome. Thank you once again.
01:14:06[Saint]Bah. People make all sorts of claims about Wolfson being some form of infallible golden God of DAC silicon.
01:14:10[Saint]They're really not.
01:14:15 Quit alexweissman (Remote host closed the connection)
01:14:42 Join alexweissman [0] (
01:14:47[Saint]The Cirrus Logic DAC in the Classic generation iPods is way more capable than I would expect it to be given the costing and the device class.
01:15:47[Saint]Under load the output is pretty flat, with very low distortion. Only relatively small and typical rolloff in frequencies that are going to be pretty much inaudible to a wide margin of the population.
01:16:19[Saint]Basically any male over 30 isn't going to notice the rolloff in the high shelf, that's certain.
01:16:44[Saint]Unless that male happened to not be homo sapiens sapiens, and was instead...a bat.
01:18:09[Saint]So, if you happen to be a bat, you might well notice the output curve isn't entirely flat.
01:18:24[Saint]Or a dolphin.
01:18:44[Saint]If you're a blue whale you might notice the rolloff in the low shelf, too.
01:21:22***Saving seen data "./dancer.seen"
01:26:41 Quit alexweissman (Remote host closed the connection)
01:28:18 Join paulk-collins [0] (
01:28:53vifinoAnother question, if I were to sync music to the iPod via iTunes, booted in to the official firmware, of course, would I be able to play it using RockBox, [Saint]?
01:29:17 Join alexweissman [0] (
01:29:56 Quit pamaury (Ping timeout: 244 seconds)
01:30:21 Quit alexweissman (Remote host closed the connection)
01:31:05 Join alexweissman [0] (
01:31:51[Saint]That is pretty much the only useful thing the 'Database' function can do.
01:32:18[Saint]When iPods sync, they do some file level obfuscation of the media library.
01:32:56[Saint]But the Database just scans over the entire disk area and makes a library of the available media based on metadata, not file location.
01:34:02[Saint]So - basically - if your media is correctly tagged with nice metadata (including non-progressive jpg or bitmap album art) it will be indistinguishable from any other content on the device that has been put there by any other means.
01:34:31vifinoNice to know.
01:34:55[Saint]I very highly advise you look over our fine manual.
01:35:26vifinoGood idea.
01:35:31[Saint]The iPod Clasisc doesn't have a manual publicly available, due to the entirely unofficial and unsupported third party installation route.
01:36:03[Saint]But you may use the iPod Video manual while for most intents and purposes would be identical to an iPod Classic manual, if one were to exist.
01:36:33[Saint]I guarantee you that you will discover at least one major feature or funtion you had absolutely no idea existed.
01:36:43[Saint]More than likely you will discover several.
01:37:09vifinoAh, that must be the reason I never even looked at the manual - Because I couldn't find one.
01:37:21vifinoI'll do so and I very much hope I do.
01:37:34[Saint]Right. For the iPods, there is very little difference for the majority of the ports.
01:37:52[Saint]The iPod Video manual would be applicable to almost every other iPod in almost every other aspect.
01:38:14[Saint]It will just include some information about features that may or may not be present in the older generation iPods.
01:38:24[Saint]But the information it does have that applies will be correct.
01:38:58[Saint]The iPod Classic is basically a spiritual clusterfuck between the iPod Nano 2G and the iPod Video.
01:39:29[Saint]Internally it is very similar to the iPod Nano 2G. Externally and functionally, it is very similar to the iPod Video.
01:52:20 Quit ulmutul (Quit: ChatZilla 0.9.92 [Firefox 48.0.2/20160823121617])
02:02:38 Quit alexweissman (Remote host closed the connection)
02:05:56 Part robertd1
02:34:21 Join alexweissman [0] (
02:50:27 Join soadkombucha [0] (dennis@gateway/shell/
02:50:27 Quit soadkombucha (Changing host)
02:50:27 Join soadkombucha [0] (dennis@unaffiliated/soadkombucha)
02:50:27 Quit soadkombucha (Changing host)
02:50:27 Join soadkombucha [0] (dennis@gateway/shell/
03:00:24 Quit krabador (Quit: Leaving)
03:04:50 Quit alexweissman (Remote host closed the connection)
03:14:02 Join JdGordon_ [0] (~jonno@rockbox/developer/JdGordon)
03:15:29 Quit __builtin (Quit: - Chat comfortably. Anywhere.)
03:15:57 Join alexweissman [0] (
03:15:59 Join __builtin [0] (~xray@unaffiliated/franklin)
03:16:40 Quit JdGordon (Ping timeout: 264 seconds)
03:21:25***Saving seen data "./dancer.seen"
04:16:57 Join CrashBash-Kun [0] (~CrashBash@unaffiliated/crashbash-kun)
04:38:52 Quit Smx (Quit: EliteBNC free bnc service - - be a part of the Elite!)
04:43:04 Join Smx [0] (Elite8556@gateway/shell/elitebnc/x-wmqchjmkmslgyonw)
05:21:28***Saving seen data "./dancer.seen"
06:21:29 Quit [7] (Ping timeout: 248 seconds)
06:21:59 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
06:28:41 Quit hoshi (Ping timeout: 264 seconds)
06:30:40 Join hoshi [0] (
07:01:36 Join ZincAlloy [0] (~Adium@2a02:8108:8b80:1700:8cb2:547c:a71:c12e)
07:21:31***Saving seen data "./dancer.seen"
07:36:59 Join Strife89|Quassel [0] (
07:39:56 Quit Strife89 (Ping timeout: 265 seconds)
07:47:44 Join edhelas [0] (~edhelas@
07:49:42 Quit amiconn (Quit: - Chat comfortably. Anywhere.)
07:49:42 Quit pixelma (Quit: .)
07:49:53 Join pixelma [0] (~pixelma@rockbox/staff/pixelma)
07:49:54 Join amiconn [0] (~amiconn@rockbox/developer/amiconn)
07:57:50 Quit ZincAlloy (Quit: Leaving.)
08:12:49 Quit edhelas (Ping timeout: 260 seconds)
08:24:20 Join einhirn [0] (
08:26:30 Join smoke_fumus [0] (~smoke_fum@
08:32:12 Join ender` [0] (
08:42:00 Quit nlogex (Ping timeout: 260 seconds)
08:42:02 Join edhelas [0] (~edhelas@
08:48:23 Join Strife89 [0] (
08:51:56 Quit Strife89|Quassel (Ping timeout: 265 seconds)
09:09:09 Quit edhelas (Ping timeout: 255 seconds)
09:21:35***Saving seen data "./dancer.seen"
09:37:49 Join wodz [0] (
09:44:50 Join elensil [0] (~edhelas@2001:1c02:1903:d800:a06b:b22e:8640:9910)
10:03:56 Join einhirn_ [0] (
10:05:22 Quit einhirn (Ping timeout: 250 seconds)
10:13:14 Join MrZeus [0] (~MrZeus@
10:21:31 Join foolsh [0] (
10:21:44 Quit CrashBash-Kun (Ping timeout: 260 seconds)
10:22:20 Quit foolsh (Client Quit)
10:23:34 Quit elensil (Ping timeout: 250 seconds)
10:25:19 Join elensil [0] (~edhelas@2001:1c02:1903:d800:a06b:b22e:8640:9910)
10:30:50 Join foolsh [0] (
10:40:09 Quit foolsh (Ping timeout: 276 seconds)
10:42:01 Join foolsh [0] (
11:06:29 Quit The_Prospector (Quit: when in doubt, kernel panic)
11:08:58 Join The_Prospector [0] (~The_Prosp@unaffiliated/cornman)
11:21:36***Saving seen data "./dancer.seen"
12:14:20 Join robertd [0] (~root@
12:23:21 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
12:29:17 Join b0hoon [0] (~quassel@
12:30:44b0hoonulmutul: i know about these threads. I'm present in one of them.
12:31:51b0hoonulmutul: The problem is that last time i tried, i couldn't reproduce it in any way.
12:35:18 Join petur [0] (~petur@rockbox/developer/petur)
12:35:57b0hoonulmultul: Are you sure that it's not the dying hdd's problem? Like you mentioned in one of the threads?
12:36:27 Quit robertd (Quit: Leaving.)
12:37:09 Join robertd1 [0] (~root@
12:37:47b0hoonulmutul: I have some time this weekend, so i will do extensive tests and i'll try to reproduce it.
12:41:43b0hoonulmutul: I don't get it. You say that you power on the DAP, wait for disk spin down, then run some file (plugin, audio) and message error appears? :-O
12:42:20b0hoonulmutul: What kind of message exactly? Is it panic error or something else?
12:43:40b0hoonulmutul: Could you give me info about your disk from Debug menu, please?
12:45:11b0hoonulmultul: And could you try to recreate it with the current version, just to make sure.
12:46:05 Quit pamaury (Ping timeout: 264 seconds)
13:13:26 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
13:20:33 Join CrashBash-Kun [0] (~CrashBash@unaffiliated/crashbash-kun)
13:21:39 Join ZincAlloy [0] (~Adium@2a02:8108:8b80:1700:413c:a846:d25c:f393)
13:21:42***Saving seen data "./dancer.seen"
13:25:29 Quit CrashBash-Kun (Ping timeout: 260 seconds)
13:27:34 Join CrashBash-Kun [0] (~CrashBash@unaffiliated/crashbash-kun)
13:30:18 Quit CrashBash-Kun (Client Quit)
13:30:30 Quit paulk-collins (Quit: Leaving)
13:52:26 Quit thum (Ping timeout: 250 seconds)
13:53:21 Join thum [0] (
14:19:17 Join ulmutul [0] (
14:32:43 Quit Rondom (Remote host closed the connection)
14:33:20 Join Rondom [0] (
14:35:15gevaertsscorche|sh: I think I want new user approval on the forums back. I'm not *that* fond of removing spam...
14:35:18gevaertssaratoga: ^^
14:35:32gevaerts[Saint]: ^^
14:36:01gevaertsother regular forum cleanup people: ^^
14:45:27 Part robertd1
14:46:51ulmutulb0hoon: testing now. I don't think it's a dying hard disk; currently I'm trying usb mode (in windows). The Philips bootloader mass storage conection works flawless, in rockbox usb mode I get occasional "Disk not formatted" or "Can't open file. Be sure that drive contains storage media" (I don't know the exact english messages)
14:49:03ulmutulThe errors only appear if the drive was spun down. As long as the drive is spinning, everything works.
14:51:18ulmutulAn in case I get an error, I just need a second try and it works.
14:53:31ulmutulI don't know if it's related, but sometimes windows needs a very long time to display the files. You hear the disk spin up, and then you'll have to wait for 10sec or so.
14:54:10b0hoonulmutul: By rockbox usb mode you mean usb in rockbox, not in rockbox bootloader?
14:54:35ulmutulYes, usb in rockbox
14:55:59ulmutulHm, panic after disconecting from usb. "mount:0 pc:00051210 sp:000D2648"
14:56:19b0hoonulmutul: wow, i must check it on mine once again, and it's on the current build?
14:58:11ulmutulYes, it's the current build. I guess it's depending on different drives.
14:59:45ulmutulDisk Info: TOSHIBA MK3006GAL, Firmware BY102A, 28615MB
14:59:46b0hoonulmutul: may be, could you give your disk details from debug menu, please?
15:00:08ulmutulDo you need more details?
15:00:09b0hoonulmutul: spin time?
15:00:44b0hoonulmutul: yeah give me everything if it's possible
15:02:09b0hoonulmutul: and there is no problem on hdd1630?
15:02:33ulmutulSpinup Time: 0ms, Power mgmt: enabled, Noise mgmt: unsupported, Read ahead: enabled, PIO modes:0 1 2 3 4, Cycle times: 120ns/120ns, Phys.sector size: 512, IORDY supported:yes, IORDY disable:yes
15:02:39b0hoonor other pp target you own?
15:04:18 Part thum
15:04:45ulmutulI havn't used the 1630 much, but I can't remember any problems.
15:07:18ulmutulI also have disk problems in my Vibe500, but they are different and I'm not sure if it's a dying hard disk problem. I've had data corruption literally every time rockbox writes something to disk, but I need to check, if this also happens under OF.
15:09:25b0hoonulmutul: ok, thanks i'll try to do something with it later, then we will decide about degrading it to unstable again, ok?
15:10:27ulmutulI suspect that rockbox fails to wake up the HDD6330s drive fully before access, and so drive commands get lost or fail.
15:11:19ulmutulI already tried to randomly enlarge timing values in the ATA driver code, but it did't help :)
15:13:59b0hoonulmutul: this target is specific because it doesn't have ata power on/off function implemented so rockbox does not control it.
15:15:11ulmutulWhere do I find ata power management in the source?
15:16:34ulmutulSeems to be hidden somewhere in ata.c
15:21:08b0hoonulmutul: portalplayer specific ata code is in firmware/target/arm/pp/ata-p5020.c, ata poweroff/poweron is usually implemented in target's powermgmt-xxxxx.c
15:21:43***Saving seen data "./dancer.seen"
15:21:59ulmutulThanks. Can the power status be read someway?
15:36:00b0hoonulmutul: when you have implemented it, ide_powered() function gives the result, otherwise i don't know, and it's not in powermgmt-xxx.c but power-xxx.c, sorry
15:40:50ulmutulThanks, I'll experiment a bit.
15:57:53 Part b0hoon ("GTG... Bye.")
15:58:58 Quit fs-bluebot (Read error: Connection reset by peer)
15:59:14 Join fs-bluebot [0] (
16:00:54 Quit bluebrother^ (Ping timeout: 255 seconds)
16:03:08 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
16:18:05 Join nlogex [0] (
16:26:14 Join robertd1 [0] (~root@
16:27:44 Join xorly [0] (
16:49:47 Join JanC_ [0] (~janc@lugwv/member/JanC)
16:50:58 Nick JanC is now known as Guest73904 (~janc@lugwv/member/JanC)
16:50:58 Nick JanC_ is now known as JanC (~janc@lugwv/member/JanC)
16:51:28 Quit Guest73904 (Ping timeout: 240 seconds)
16:57:44 Join girafe [0] (
17:19:04 Quit wodz (Ping timeout: 248 seconds)
17:21:44***Saving seen data "./dancer.seen"
17:53:27 Join paulk-collins [0] (
17:54:04 Quit elensil (Quit: Leaving.)
18:03:03ulmutulb0hoon: g#1397 fixes my issues! Has to be worked out a bit more :)
18:03:04fs-bluebotGerrit review #1397 at : [HACK] fix Philips HDD6330 disk issues by Sebastian Leonhardt
18:03:14ulmutulgtg now, bye!
18:03:19 Quit alexweissman (Remote host closed the connection)
18:03:24 Quit ulmutul (Quit: ChatZilla 0.9.92 [Firefox 48.0.2/20160823121617])
18:09:57 Join gbl08ma [0] (
18:22:00 Quit michaelni (Quit: Leaving)
18:33:20 Quit MrZeus (Ping timeout: 276 seconds)
18:33:22 Join michaelni [0] (
19:06:29 Quit einhirn_ (Ping timeout: 276 seconds)
19:13:48 Join alexweissman [0] (~alexweiss@
19:18:25 Quit alexweissman (Ping timeout: 244 seconds)
19:21:48***Saving seen data "./dancer.seen"
19:28:47 Join lebellium [0] (
19:37:23 Join cereal_killer [0] (2e7df905@gateway/web/freenode/ip.
19:38:45 Quit pamaury (Remote host closed the connection)
19:39:30cereal_killerthanks to gevaerts for helping me having a current build on my D2. It is usable and except for creating bookmarks very stable.
19:40:01cereal_killerbuilds before the filesystem rework just wouldt let me create a bookmark, the current build panics.
19:40:34cereal_killerwhere in the code can I disable the bookmark feature and not using it accidently?
19:52:21 Part robertd1
19:56:34 Join alexweissman [0] (~alexweiss@
19:59:58 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
20:20:44 Quit alexweissman (Remote host closed the connection)
20:21:20 Quit paulk-collins (Quit: Leaving)
20:27:44 Quit xorly (Ping timeout: 276 seconds)
20:52:06 Quit Bray90820 (Ping timeout: 244 seconds)
21:02:22 Join alexweissman [0] (~alexweiss@
21:05:04 Quit krnlyng (Ping timeout: 240 seconds)
21:05:05 Quit alexweissman (Remote host closed the connection)
21:06:37 Quit pamaury (Ping timeout: 250 seconds)
21:06:40 Join pamaury_ [0] (~pamaury@rockbox/developer/pamaury)
21:07:31 Quit cereal_killer (Quit: Page closed)
21:17:42 Quit pamaury_ (Ping timeout: 265 seconds)
21:17:46 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
21:18:08 Join krnlyng [0] (
21:21:51***Saving seen data "./dancer.seen"
21:23:31 Join pamaury_ [0] (~pamaury@rockbox/developer/pamaury)
21:23:32 Quit pamaury (Ping timeout: 255 seconds)
21:33:09 Join edhelas [0] (~edhelas@
21:35:50 Quit pamaury_ (Ping timeout: 244 seconds)
21:36:08 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
21:38:52 Join alexweissman [0] (
21:41:17 Quit edhelas (Ping timeout: 250 seconds)
21:51:51 Join pamaury_ [0] (~pamaury@rockbox/developer/pamaury)
21:52:01 Quit pamaury (Ping timeout: 265 seconds)
21:56:51 Quit Guest25806 (Read error: error:1408F10B:SSL routines:SSL3_GET_RECORD:wrong version number)
22:00:05pamaury_gevaerts: do you remember the command to make a bootloader display a specific version ?
22:00:19pamaury_I think it's something make VERSION="V2" but I'm not sure
22:00:41 Join alexbobp [0] (
22:01:02 Nick alexbobp is now known as Guest36199 (
22:01:06gevaertsI think so...
22:07:35pamaury_gevaerts: I'm not sure if it's a bug but apparently "make clean" does not remove rbversion.H
22:07:54pamaury_actually I am not sure which tool create rbversion.h
22:08:07gevaertsSome makefile magic IIRC
22:08:20CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
22:08:20*pamaury_ tries to find out and fix this makefile bug
22:08:36*gevaerts recommends care
22:08:48gevaertsI seem to remember it being a bit fiddly
22:09:33 Join robertd1 [0] (~root@
22:10:32 Join edhelas [0] (~edhelas@
22:10:47pamaury_it's firmware/firmware.make
22:10:52 Nick pamaury_ is now known as pamaury (~pamaury@rockbox/developer/pamaury)
22:11:46pamaurygevaerts: this is firmware.make:
22:11:47pamauryCLEANOBJS += $(BUILDDIR)/sysfont.* $(BUILDDIR)/version.*
22:11:57pamaurylooks like a typo to me
22:12:13 Join krabador [0] (~krabador@unaffiliated/krabador)
22:12:29gevaertsCould well be
22:13:29pamauryhum, we do have firmware/common/version.c
22:13:49pamaurythat's the file that stores the version
22:13:59pamaury#include "rbversion.h"
22:13:59pamauryconst char rbversion[] = RBVERSION;
22:14:18pamauryso if you remove version.o, it's recompiled but it will be the same because rbversion.h hasn't changed
22:14:19 Quit girafe (Read error: Connection reset by peer)
22:15:37pamauryoh my, there is some funny business in the makefile
22:17:38 Quit edhelas (Ping timeout: 265 seconds)
22:20:18*pamaury is confused, this is the code:
22:20:18pamaurySVNVERSION:=$(shell $(TOOLSDIR)/ $(ROOTDIR))
22:20:18pamauryOLDSVNVERSION:=$(shell grep 'RBVERSION' $(BUILDDIR)/rbversion.h 2>/dev/null|cut -d '"' -f 2 || echo "NOREVISION")
22:20:18DBUGEnqueued KICK pamaury
22:20:18pamauryifneq ($(SVNVERSION),$(OLDSVNVERSION))
22:20:18pamaury.PHONY: $(BUILDDIR)/rbversion.h
22:20:19***Alert Mode level 1
22:21:07pamauryobversion: if I rm rbversion.h and make VERSION="X", then rbversion.H contains X. If I then make then rbversion.h is regenerated from git version. But if I then make VERSION="X", nothing happens
22:23:21 Join edhelas [0] (~edhelas@
22:25:00 Quit petur (Quit: Leaving)
22:28:37gevaertsIt's worse than that
22:28:53gevaertsIf I change a file somewhere (to get the M), it keeps rebuilding
22:29:10pamaurythe code in tools/ is....interesting
22:30:20***Alert Mode OFF
22:30:47gevaertsWell yes
22:30:51pamaurygevaerts: isn't the code in completely useless? It seems to me like it always put the new version in rbversion.h
22:30:58gevaertsNo, it doesn't
22:31:01pamaurywhy this complicated detour with _rbversion.h
22:31:37gevaertsBecause if the version is the same the file must not be changed
22:32:03pamauryah yeah, that's a makefile thing
22:32:04pamaurymake sense
22:32:20pamauryprobably worth a comment
22:33:51pamaurystill, I don't yet understand how my make VERSION="X" eventually endsup in rbversion.h
22:34:32gevaertsThat's tools/
22:35:07pamauryI am not so sure. I added code to firmware.make to print OLDSVNVERSION and SVNVERSION
22:35:25pamauryand if I make VERSION="X", then it does not affect the output
22:36:13pamauryah wait, that's crazy
22:36:24pamauryfirst the makefile runs to see if it was changed
22:36:40pamaurybut then it gives the string "" to that re-runs it
22:36:48pamauryand somehow this second run leads to a different output
22:39:22pamauryapparently this line:
22:39:22pamaury$(shell $(TOOLSDIR)/ $(ROOTDIR))
22:39:22pamaurydoes not transfer the VERSION environement variable to the subshell, but this one does:
22:39:22pamaury$(call PRINTS,GEN $(@F))$(TOOLSDIR)/ $(BUILDDIR) $(TOOLSDIR)/ $(ROOTDIR)
22:42:14 Quit amayer (Quit: Leaving)
22:44:28pamaury is unclear on $(shell ...) behaves w.r.t. env variables
22:45:22 Join Bray90820 [0] (
22:46:34pamaurythis thread contains an explanation:
22:46:34pamauryapparently shells invoked with $(shell ) do not contain exported variable, whereas recipe commands do...
22:47:05pamauryand since is invoked in the recipe command, it get VERSION and when it runs, it has the proper version
22:47:26pamaurygevaerts: any reason why re-runs ?
22:47:48pamauryone workaround is to explicitely pass variable on the shell
22:48:06pamaury$(shell VERSION='$(VERSION)' ...)
22:50:19gevaertsYes, that does seem to make things better
22:50:34pamaurymaybe is invoked somewhere else?
22:51:40CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
22:51:40*gevaerts finds that his earlier observation about a modified git version was wrong
22:52:12pamaurygevaerts: what do you think about g#1398 ?
22:52:13fs-bluebotGerrit review #1398 at : Fix makefile not rebuilding rbversion.h in some cases by Amaury Pouly
22:52:46pamauryoops I left some debugging
22:56:20 Quit hoshi (Read error: Connection reset by peer)
22:56:37pamauryI don't see the point of rerunning the command so I suggest it takes the version directly
22:57:33gevaertspamaury: I think that looks sensible
22:58:15gevaertsBut the history of rbversion.h makes me cautious :)
23:00:31 Join hoshi [0] (
23:00:49pamaurywell let's look at the fact: looks at the VERSION variable but it is not passed to the shell. There is clearly a mismatch between the intent and the implementation. Maybe it's only a GNU make bug
23:01:15gevaertsYes, it clearly wasn't working well for hand-specified versions
23:01:44gevaertsAnd apart from that, I don't see anything functinally different now, so I'm fairly confident it's ok
23:02:22gevaertsWell, passing the value directly is different I guess, but it's still the same flow :)
23:02:52*pamaury commits this
23:03:25fs-bluebotBuild Server message: New build round started. Revision 50c1de7, 255 builds, 14 clients.
23:03:37 Join girafe [0] (
23:14:59fs-bluebotBuild Server message: Build round completed after 694 seconds.
23:15:00fs-bluebotBuild Server message: Revision 50c1de7 result: All green
23:17:29 Join paulk-veyron-min [0] (
23:20:40 Quit ruskie (Excess Flood)
23:21:53***Saving seen data "./dancer.seen"
23:30:10 Quit girafe (Read error: Connection reset by peer)
23:33:21 Quit lebellium (Quit: ChatZilla 0.9.92 [Firefox 49.0/20160916101415])
23:43:53 Join ruskie [0] (ruskie@sourcemage/mage/ruskie)
23:54:00 Quit paulk-veyron-min (Read error: Connection reset by peer)

Previous day | Next day