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 2005-11-12

00:00:16ender`is it any different from the battery in H120?
00:00:33linuxstb_I've no idea. But I would expect them to be the same
00:01:15amiconnBattery runtime might be *slightly* higher for h120 than h140
00:02:34ender`due to disk?
00:03:03amiconnTwo reasons. (1) The disk probably needs less energy to spin up (1 platter instead of 2)
00:03:25amiconn(2) The RAM refresh needs less bandwidth
00:03:41amiconnBah, (2) is for h100
00:03:58linuxstb_The h100 has 16MB RAM?
00:04:36ender`i was looking at my PS2 the other day and realized that it, too has 32MB ram...
00:05:02linuxstb_Do we have any stats on how many downloads there have been for each target?
00:05:21ender`btw, what's the unicode status?
00:05:41Bagderlinuxstb: not really
00:07:52 Join len0x [0] (n=len0x@
00:09:54 Quit len0x (Client Quit)
00:10:54 Join matsl [0] (
00:27:13 Join DJDD [0] (
00:28:25 Quit DJDD_ (Read error: 110 (Connection timed out))
00:50:58 Join ashridah [0] (
00:57:27***Saving seen data "./dancer.seen"
01:02:53amiconnMy asm monster finally works as intended
01:03:06linuxstb_Congratulations. Is an ARM version next? :)
01:07:35XShocKhmmm... i look at trickle chrger in ihp120.. and see that it has usb pin... so, does it mean that player can be chrged from usb?
01:09:02XShocKfrom what i see there will be nothing hard at all to make a player charge and work from usb..
01:09:13XShocKit is just a matter of one wire.
01:21:07 Quit DJDD ("Trillian (")
01:27:31ashridahideally, any kind of usb-drawing charger will actually have a usb device built in in order to control current flow.
01:28:00ashridahbut i've heard of people hacking usb 1.1 cables in order to do usb-drawn charging
01:28:08ashridahjust make sure you do it on a port you don't mind frying ;)
01:40:09 Quit dpassen1 ()
01:44:35 Quit ender` (Read error: 104 (Connection reset by peer))
01:49:54 Join _aLF [0] (
02:12:39 Quit actionshrimp ("a bird in the bush is worth two in your house")
02:14:05XShocKashridah, usb-calble-hacking doesn't look very good. one wire inside of a player looks much nicer. :)
02:15:06XShocKanyway, i will not do it just because i don't really need it. btw, i could fit my rtc schema inside of my player
02:15:39XShocKit barely fit, but it is inside now.
02:16:24XShocKi didn't connect any wires yet, i don't have solder in my room, i will solder it tomorrow. :)
02:16:48XShocKand will start writing a driver for it.
02:17:29XShocKpretty easy, since it is connected using i2c to the same bus tuner is using.
02:24:29 Quit matsl (Remote closed the connection)
02:35:20XavierGrxshock: where exatcly to solder the cable for charge and sybc?
02:35:33XavierGralso it would need a switch
02:39:13XShocKpin number 2 on TI trickle charger goes directly to usb cable.. adding a cap is also not a bad idea.
02:39:25XShocKthat is for charging.
02:39:40XShocKground i think is already connected.
02:39:59XavierGrno need for switch then?
02:40:12XShocKno switches..
02:40:46XavierGrso already roclbox has the option to charge from other targets.
02:41:14XShocKhmm. i dont' understand what you mean.
02:41:29XavierGrIyeah I said it very wrong
02:42:56XavierGrI mean that rockbox option "tickle charge" will controll the charging when this wire is added, right?
02:43:51XavierGror I completely misundestood what you said.
02:44:25XShocKthat addition of the wire will make your battery charge whenever you connect your player to computer using usb.
02:44:38XShocKit has nothing to do with software and rockbox in particular
02:44:56XavierGrexaplain this: If you add the wire that you say, which part of hardware (or software) will control if the user wants to charge?
02:45:05XavierGrso there will be no option.
02:45:19XavierGrwhen you insert USB you charge without an option.
02:45:22XShocKyes. no option.. it will charge automatically..
02:45:49XavierGrbut I think that this is not acceptable.
02:45:56XavierGrobviously you don't want to charge all the time.
02:46:31XShocKhmm. why not?
02:47:07XavierGrbad for the battery?
02:47:25XavierGrimagine that you charged fully and then you remembered to put some new files
02:48:00XavierGrI guess that li-ion is not spoiled by continous charging but....
02:49:04XShocKi guess that TI trickle charger will terminate the charge..
02:49:29XavierGrI just had my worst nightmare with those modded USB cables that charge and sync
02:49:32XShocKi didn't read the documentation very deep.
02:49:51XavierGrI made one of those when I first got the player and it worked quite fine.
02:50:05XavierGrbut now something went wrong.
02:50:20XShocKhow do you control chargin in those ipods that have the feature of usb charging?
02:50:35XavierGrwhen I coupled it on my my PC, Windoze crashed
02:50:57XavierGrthen my player crashed in a way that no led was lit
02:51:17XavierGrSo I didn't realised that iriver had crashed too.
02:51:28XavierGrthen I pushed play only to see that nothing happens
02:51:30XShocKi gotta go.
02:51:41XShocKhave a meeting at 9pm.
02:51:42XavierGrok good night
02:51:51XShocKu2. cya
02:52:37XavierGranyway I then realised that the HD was spinning and the reset did the work. For one sec I really thought that my unit died...!
02:57:28***Saving seen data "./dancer.seen"
03:02:54amiconnlinuxstb: Your table.c commit is incomplete => red builds
03:07:31linuxstb_Sorry - forgot SOURCES. It's committed now.
03:09:39linuxstb_Why didn't the sim builds turn red as well?
03:11:04amiconnGood question...
03:11:52ashridahdo the platforms all rebuild whenever there's a commit, or daily?
03:12:08linuxstb_All platforms after every commit
03:12:42ashridahwhoa. how long does the build process take for all of them?
03:13:10amiconnAbout half an hour, depending on the server load and how much was changed
03:13:20ashridahaah, does it does incrementals if it can?
03:13:32amiconnIt uses some compiler cache
03:14:58linuxstb_And it's going to get worse with the ipod targets. We could quite quickly have 4 or 5 new targets
03:15:19linuxstb_All with codecs to build
03:15:28amiconnYes, and some sims as well...
03:15:41linuxstb_And bootloaders for each ipod.
03:16:26amiconnEven today not all possible combinations are built
03:16:54linuxstb_The ipods are very very similar. I'm sure you've noticed the minor differences between the two ipods
03:18:15linuxstb_I'm not sure the ipod sims would need building. They would probably be very close to the H3x0
03:18:55amiconnAt least the lcd resolution is different
03:19:02amiconn...and the buttons
03:19:19amiconnThen we might decide to run the H300 lcd in 18bit
03:19:22linuxstb_Yes, I forgot the buttons. I thought the resolution was the same (for the color ipod)
03:19:35linuxstb_My ipod is 220x176
03:19:56linuxstb_So I guess we will need some sims.
03:20:02 Quit _aLF ("^^")
03:20:18amiconnThe nano is less
03:20:26amiconnThen there are the greyscale ipods
03:20:46linuxstb_Yes - the nano is less. But I think the greyscales are the same as the H1x0
03:20:54amiconnTwo resolutions
03:21:06amiconnStandard ipod is same as H1x0
03:21:10amiconnMini is again less
03:22:18linuxstb_So we're back to needing all the sims then?
03:22:24amiconnI wonder whether the pixel-flicking greyscale technique would be possible on these ipods...
03:22:50linuxstb_I can tell your tempted to buy one :)
03:23:31amiconnHmpf. I already have 5 rockboxes, more than enough for my limited time :/
03:25:11amiconnThen I don't really like the ipod design (well, mostly the available colours)
03:25:52amiconnI heard that the nano is available in black though
03:26:26linuxstb_Yes - the Nano comes in black and white
03:28:12amiconnHmm, afaik ipodlinux doesn't have the pixel-flicking greyscale...
03:28:39linuxstb_The lack of buttons is my biggest criticism. Also, the non-standard USB connector, but charging via USB is very useful.
03:29:00amiconnHmm, no standard connector?
03:29:09amiconnHow many buttons?
03:30:12linuxstb_There are four real buttons and the scroll-wheel which is touch-sensitive giving (i think) 96 values depending on where you press it.
03:30:31linuxstb_So you could argue there are 100 buttons...
03:31:01linuxstb_So you basically have a wheel that turns in two directions and 4 buttons.
03:31:20linuxstb_Plus a hold switch
03:31:58amiconnThat's not much...
03:32:41amiconn...but a tiny bit better than the Ondio, which has 6 buttons, one of which is the power button
03:32:54amiconnI guess the ipod has no separate power button?
03:33:12linuxstb_No - a long press on PLAY sends it to sleep.
03:33:31linuxstb_And there is the hardware reset key combination as well.
03:33:35amiconnDo you know whether button combos are restricted somehow?
03:33:49linuxstb_I don't know.
03:34:39linuxstb_I haven't really looked at the button driver yet.
03:35:04linuxstb_I've been working tonight on getting Rockbox itself to compile. I've now reached the plugins.
03:35:42amiconnI'm curious how the wheel is handled in the firmware
03:36:00linuxstb_Which firmware?
03:36:08 Quit ]RowaN[ (Read error: 104 (Connection reset by peer))
03:36:34amiconnI mean, how is it hooked up, what signals are received when the wheel is operated?
03:37:53linuxstb_I think the hardware provides an integer between 0 and 95, with all the hard work done in the hardware.
03:38:44linuxstb_There was a thread on the IPL forums about it, but the IPL website is down at the moment so I can't find it.
03:40:55amiconnI'm undecided whether I should commit the memcpy() monster. 958 bytes...
03:44:36 Quit ghode|afk (Read error: 110 (Connection timed out))
03:45:38linuxstb_I forgot the middle (select) button on the ipod. It's 5 buttons, plus the wheel, plus hold.
03:46:25linuxstb_Are you planning on putting the memcpy in iram?
03:46:51amiconnIt always was in iram
03:51:08linuxstb_I think you should - it's very well-used function.
03:51:37amiconnAfaik, some codecs use an own function instead
03:51:58amiconnThese should then use the new memcpy()
03:54:07linuxstb_Yes, that shouldn't be a problem to fix.
03:54:53linuxstb_Apart from the plugins, Rockbox now builds for the ipod. I now need to write some startup code.
03:57:24linuxstb_I think it's time for sleep. Goodnight.
04:01:24 Quit ashridah ("out")
04:08:45 Part linuxstb_
04:30:59 Join solexx_ [0] (
04:41:56 Quit solexx (Read error: 110 (Connection timed out))
04:50:17 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
04:57:31***Saving seen data "./dancer.seen"
05:10:58 Join andy [0] (
05:11:32andyiriver recording committed, waiting for build results :)
05:21:46 Join DJDD [0] (
05:48:50 Join webguest64 [0] (
05:49:51 Quit webguest64 (Client Quit)
05:57:24 Quit andy ()
06:06:08 Quit RotAtoR ()
06:57:34***Saving seen data "./dancer.seen"
07:03:54 Quit BirdFish ("reboot")
07:08:00 Join psycho_maniac [0] (
07:08:21 Join BirdFish [0] (n=bradbox8@
07:26:13 Quit psycho_maniac ("CGI:IRC")
07:56:43 Quit BirdFish ("reboot (back in a minute)")
08:06:46 Join BirdFish [0] (n=bradbox8@
08:28:49 Quit DJDD (Read error: 104 (Connection reset by peer))
08:31:26 Join DJDD [0] (
08:57:35***Saving seen data "./dancer.seen"
09:32:09 Join ashridah [0] (
09:34:46 Join LinusN [0] (
09:34:57LinusNXShocK: u there?
09:39:22LinusNlast time i investigated usb charging on the h100, the AC and USB inputs were connected, so it's not a matter of connecting a wire, you have to lift the (tiny) pin from the charging chip
09:39:51LinusNpretty hard to do without breaking the pin
09:40:42LinusNhi :-)
09:40:43XShocKhmm. i didn't think about it..
09:41:09 Join amiconn_ [0] (
09:41:24 Join StrathAFK [0] (
09:41:33 Quit amiconn (Nick collision from services.)
09:41:33 Nick amiconn_ is now known as amiconn (
09:41:39LinusNXShocK: still, it's not impossible
09:42:32LinusNhi amiconn
09:42:46amiconnGood morning
09:43:02XShocKgood morning
09:45:15amiconn10 codecs now supported in iriver rockbox. A true multi-codec jukebox :-)
09:53:14amiconnRed build alert
09:54:59LinusNamiconn: saw my mail in the dev list?
09:55:36*amiconn checks
09:56:28amiconnYeah, totally wrong
09:56:47amiconnI wonder what the bool monitor is for anyway
09:57:04 Quit Strath (Read error: 110 (Connection timed out))
09:57:54amiconnIt seems this selects whether the signal is put on the DAC, but imho this should always be done
09:58:04LinusNi think so too
09:58:20amiconn...and the volume should be changeable from the recording screen
09:58:45LinusNi'll fix that bug
09:59:16amiconnThere's a (imho sensible) suggestion in the Ondio forum:
10:00:02LinusNsaw that too, a great idea
10:00:34amiconnI think the proper fix would be to completely remove the 'bool monitor'
10:01:05LinusNi think so too
10:01:24amiconnOf course the renaming of functions in mpeg.c should stay
10:01:29LinusNi believe you can control the monitor volume in the iriver recording screen, can't you?
10:01:36amiconn(mpeg_* -> audio_* )
10:01:57amiconnIf it's the same recording screen as on archos, you can't (yet)
10:03:05amiconnit is
10:04:14LinusNyes, and you can't toggle monitor mode in the screen, only in the settings
10:04:50amiconnHmm, so volume in the recording screens would be really useful, and the monitoring should be always enabled and the option dropped
10:05:11MaximeLinusN: so it is 'possible' to do USB Charging on H1xx ? :x
10:05:33amiconn...and I have an idea how to make the gain settings look more 'professional' and take one line less
10:06:31amiconnInstead of having separate 'Gain', 'Left' and 'Right' for line recording, we could have 'Left' and 'Right' only, with the option to couple them
10:06:55amiconnThe coupling would be clearly indicated by inverting both lines
10:08:29amiconnHmm, and two pointers when using pointer mode?
10:08:40LinusNMaxime: yes, but you need to do some pretty tricky hardware modifications
10:09:02MaximeLinusN: hum, how? some soldering and else?
10:09:02LinusNamiconn: good idea
10:09:41LinusNMaxime: unsolder and lift a pin on the charging chip, and connect it to the USB 5V with a wire
10:09:58Maximeah yes I see..
10:10:04LinusNunfortunately not as easy as it sounds
10:10:37Maximeyeah :x
10:10:39LinusNthe pin is horrendously small
10:11:11XShocKLinusN, that was a QFN package right? trickle charger
10:11:18XShocKi think it was
10:11:38XShocKah. it actually isnt
10:12:45XShocKsorry. it is late here
10:13:47XShocKi hope tomorrow i find someone to open electronic laboratory for me. doh.. they only give access to the building only yo graduate students with electrical engineering major.
10:20:31LinusNXShocK: how come you call it trickle charger?
10:21:01LinusNit can charge at full rate too
10:38:08XShocKmm.. i just looked at documentationv very quickly...
10:38:46XShocKsorry for confusion
10:39:20XShocKi gotta go sleep.
10:39:29XShocKgood night all
10:50:43 Join arkascha [0] (
10:52:06LinusNamiconn: imho, it might be simpler to have one gain gauge when l&r are coupled
10:52:35amiconnHmm, but then it's not obvious that left & right can be set separately
10:52:56LinusNno, not in the rec screen itself
10:54:24amiconnHmm, I have 2 variants of superfast coldfire memcpy(), one is 718 bytes, the other is 958 bytes
10:54:46amiconnThe larger one is of course faster
10:55:16amiconnI'm undecided which one to commit...
10:57:17LinusNi don't think 200 extra bytes is a problem
10:57:37***Saving seen data "./dancer.seen"
10:57:57LinusNif that is a problem, we'll just reduce the stack with 200 bytes
10:58:13amiconnHmm, memcpy() is in iram, and my next step (memmove) will double the size again
10:58:35amiconnFor SH1 memmove() is already working
10:58:36LinusNstill a few hundred bytes isn't much imho
10:58:47LinusNsh1 is worse, since the iram is so small
10:59:30amiconnOn sh1 the full memmove() is 464 bytes
10:59:55amiconnIt still fits in iram, even with debug builds
11:00:31amiconnThe problem why the coldfire memcpy is so big is that we have to keep line alignment in order to profit from burst mode
11:01:22amiconnSo memcpy has to deal with 256 possible alignment combinations (start address at line_start + 0..15 each for source and destination)
11:01:51amiconnThe source alignment is handled in head loops, after which the source is always line aligned
11:02:24amiconn-> 16 possible alignments for the main loops
11:03:04amiconnAll main loops read full lines with burst mode
11:04:05amiconnThe big memcpy() has a total of 10 main loops. For all longword and word aligned destinations it also writes full lines in burst mode
11:04:38amiconnFor byte alignment, writing is only done as longwords (still quite fast)
11:05:38amiconnThe smaller memcpy() version ditches burst writing for the word aligned destinations, and only does longword writes there as well
11:05:47amiconn-> Number of main loops reduced to 7
11:07:20LinusNhow different do you think the performance would be?
11:11:20 Join amiconn_ [0] (
11:13:53 Join DangerousDan [0] (
11:19:14 Quit amiconn (Nick collision from services.)
11:19:15 Nick amiconn_ is now known as amiconn (
11:21:30amiconnLinusN: 50% faster for word aligned destinations
11:22:46amiconnNote that I did all my measurements in sdram so far, I still have to test with iram destination
11:23:04LinusNand how would the real life performance gain be do you think?
11:23:26LinusNi mean, how many word aligned memcpy() are there?
11:23:35LinusNhard to say i guess
11:23:50amiconnThat's hard to say... the codecs do quite some copying, but I don't know about the alignment
11:24:11LinusNi'd say go for the maximum optimization
11:24:27LinusNno need to work that hard for an incomplete optimization
11:25:12amiconnI can leave the alternative in. It's a simple #ifdef
11:26:49amiconnCopy speed in sdram (full version) is 10MB/s for longword aligned dest, 9MB/s for word aligned, and 6 MB/s for byte aligned
11:27:38amiconn...versus the C version: 4 MB/s for longword alignment (only when aligned right at the start), 1 MB/s otherwise
11:28:12LinusNamiconn: half-hearted test of the recording gui (only for archos):
11:28:23LinusNtry it in the sim
11:28:53LinusNamiconn: commit! commit! :-)
11:33:58amiconn(all speeds measured at 45 MHz)
11:40:49amiconnBtw, I just tried it: it is very easy to trigger folder skip with short-long right if you intend to skip to next track and seek there
11:41:30LinusNok, then i am old and slow :-)
11:42:04LinusNfor me, it mostly takes about a second to hear which track it is and realize that i want to seek at all
11:42:50amiconnOf course, if you want to hear first what track it is then it doesn't happen
11:43:10amiconnThe problem arises when you *know* that you want to seek in the next track
11:44:57amiconnHmm, your recording screen patch is a slightly different idea than mine.
11:45:36amiconnI meant that the channel coupling could be selected within the recording screen, just with up/down like before
11:45:53LinusNaha, how?
11:46:27amiconni.e. the screen would have 'Left' and 'Right' and start with the channels coupled (both lines inverted/with cursor)
11:46:48LinusNok, and how do you decouple?
11:47:00amiconnGoing Up would select Left only, going down Right only
11:47:28LinusNhow do you change the volume then?
11:47:31amiconnLeft <=> coupled <=> Right, starting with 'coupled'
11:47:43amiconnThe volume would be an additional line
11:48:09LinusNah, so you essentially want the "gain" to be between the left and right?
11:48:21amiconnYes, as a 'virtual' line
11:49:01LinusNthat leaves us with a screen estate problem, doesn't it?
11:49:26amiconnCurrently there is one spare line, so we can put volume in even without removing 'Gain'
11:51:13LinusNah, i thought the empty line was used for prerecording or trigger info
11:51:19amiconnOny the recorder there's currently no button bar in the recording screen. If we want one (would make sense), we can only put volume in with removing 'Gain'
11:53:56amiconnNo, prerecording is displayed instead of 'size' in line 2, and trigger info isn't displayed in the recording screen
11:53:59LinusNyou're right, it seems the empty line isn't used at all, silly me
11:54:39amiconn...but the bottom line is used for showing quality & sample rate, which is covered by the gain when recording from line in
11:57:11LinusNhmm, so the bottom line could be there if we coupled the gain like in my suggestion
11:58:02LinusNgotta go, bbl
11:58:08 Part LinusN
11:58:26 Join einhirn [0] (
12:03:47 Join ender` [0] (i=ychat@
12:30:55 Quit ender` (Read error: 113 (No route to host))
12:57:38***Saving seen data "./dancer.seen"
12:57:52 Join linuxstb_ [0] (
13:08:13 Join ender` [0] (i=ychat@
13:13:02 Quit linuxstb (Read error: 110 (Connection timed out))
13:17:50Slasherihmmh, i think i somehow now managed to brick my player. I still don't understand how, because i haven't touched the original boot loader area
13:17:56Slasheribut it broke after i pressed reset
13:18:16Slasheristill it reacts to hold somehow, but it displays nothing on the screen
13:18:22ashridahwon't turn back on? hold play while you press reset?
13:18:35Slasheristill nothing on the screen
13:18:46ashridahdarn :/
13:19:02Slasheriand it will not turn off. If i press play with hold in, the disk spill spin down but unit still remains on
13:19:05Slasherithat's weird
13:19:40Slasheriand i verified with a hexdump that i haven't replaced the original boot loader and reset vectors was ok.. but somehow the flash got corrupted
13:21:12Slasheribut i will try to get the unit powered off and then try again. Maybe the reset button doesn't reset everything
13:23:23Slasheriand yes, it did boot at the first time but then when i pressed reset, it no longer boots at all.. that is interesting :)
13:25:32ashridahyay, a challenge! :)
13:25:46Slasherianyway, i probably need a bdm now
13:26:32SlasheriI found that i wrote the rombox into wrong position (it still didn't overwrite anything). But maybe it caused cpu to excecute some wrong commands that erased the flash
13:26:56amiconnYou now see why flashing routines in the core can be bad?
13:27:14Slasheriyes, but those routines were not in core (yet) :/
13:27:16Slasherionly in plugin
13:28:01Slasherihmm, but maybe the plugin remained in ram.. at least i have noticed that short power cycle does not erase the dram
13:29:08amiconnYeah, the RAM is guarantted to hold the data for 64 milliseconds, but often it holds the data much longer
13:29:13 Quit arkascha (Read error: 110 (Connection timed out))
13:29:16amiconnNot all bits are reliable this way
13:29:36amiconn(guaranteed to hold the data without refresh, that is)
13:30:04Slasheriamiconn: i found out that it holds the audio buffer even over 5s-10s
13:31:31Slasherii think it might be a good idea to erase the whole dram on bootloader before executing anything from the ram..
13:31:43Slasherior erasing on shutdown
13:31:52 Quit DJDD ("Trillian (")
13:32:15 Nick StrathAFK is now known as Strath (
13:32:22Slasherito prevent that kind of things happening :)
13:32:22amiconnThe startup code clears the bss anyway
13:32:51amiconnWell, maybe the plugin ram should be cleared after running a flash plugin
13:41:13Slasheriamiconn: btw, what is the meaning of %vbr register?
13:41:43amiconnIt's the vector base register, pointing to the exception vector table
13:41:47 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
13:41:57Slasheriah, ok
13:41:59amiconnIt can only point at 1MB-aligned destinations
13:43:03 Nick paugh is now known as AliasCoffee (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
13:43:22linuxstb_Does anyone know anything about databox? Building it for ARM generates a couple of serious compiler warnings, and I'm not sure the best way to fix it.
13:44:00linuxstb_The problem is that the TOKEN_ defines are in the range -1 to 131, but they are stored in a "char".
13:44:27linuxstb_This gives me: "databox.c:133: warning: comparison is always true due to limited range of data type"
13:44:44linuxstb_(and in editparser.c:55)
13:45:35amiconn-1 to 131 is clearly out of range for chars
13:46:52linuxstb_Could it be that a char on ARM is always unsigned?
13:47:11amiconnIt's out of range either way
13:47:45linuxstb_I get a similar warning in libfaad for the line "if (x < 0)" where x is defined as a char - Comparison always false
13:48:01amiconnBut of course it can be that char defaults to unsigned on arm
13:48:36amiconnchar is the only type whose signedness depends on the platform
13:48:52amiconnYou can explicitly change that to 'signed char' as a workaround
13:49:13linuxstb_Yes - I've just done that (in inttypes.h) and that fixes it.
13:50:06linuxstb_I mean it fixes libfaad. But that's still a problem in databox.
13:50:39*amiconn sees something he didn't expect at all
13:51:07 Join yngwi [0] (
13:51:42*amiconn is silly
13:56:08yngwihi, anyone online, i've a question concerning compiling the uisimulator with cygwin
13:58:52 Quit yngwi ("Chatzilla 0.9.68a [Firefox 1.0.7/20050919]")
14:02:07linuxstb_Anyone know if there is a reason why some functions have the explicit "__attribute__ ((section(".icode")));" instead of using the ICODE_ATTR define? Or is it just historical and should be changed?
14:02:16Slasheriamiconn: hmm, is it possible that reset button doesn't reset everything? Something that might cause the kernel to crash
14:02:50amiconnI guess it hard-resets the cpu (only)
14:02:59amiconnlinuxstb_: which ones?
14:03:06linuxstb_sleep() for example.
14:03:23Slasheriyep.. the bootloader certainly starts because it reacts to the hold button but cannot go so far that it could show anything on the display
14:03:28linuxstb_amiconn: Plus I think the ata functions. I don't have a clean source tree at the moment - but grep will find them for you.
14:04:11 Join Sandking [0] (
14:04:12amiconnHmm. I guess this is just historical
14:04:15linuxstb_I'm wondering if that's used to put some functions in iram for all platforms, and some functions only for iriver.
14:04:48amiconnICODE_ATTR is defined for all targets
14:04:57amiconn...for the core
14:05:09linuxstb_Apart from iPod - I have a problem with that at the moment.
14:05:14amiconnThe only difference is that ICODE_ATTR is not defined on archos for plugins
14:05:50amiconn(too little iram to share between core and plugins)
14:07:34linuxstb_So it's fine for me to change them then?
14:10:48amiconnI think so
14:11:18amiconnBtw, in order to put code in iram, you need a proper crt0.S implementation. Maybe that is the problem on ipod?
14:11:38linuxstb_It's a compile-time problem.
14:12:15linuxstb_Give me a second, and I'll quote the error message from the linker.
14:12:25linuxstb_bookmark.c:(.text+0x13e8): relocation truncated to fit: R_ARM_PC24 against symbol `strlen' defined in .icode section in /home/dave/devel/rockbox-cvs/build-ipod/librockbox.a(strlen.o)
14:13:01linuxstb_DRAM is located at 0x1000000 and IRAM is at 0x4000000 - I think this large jump is causing problems. But I haven't investigated it properly yet.
14:13:52linuxstb_I have changed for ipod to be more or less identical to the iriver.
14:14:24amiconnIt seems gcc generates pc-relative jumps, with a 24-bit displacement
14:14:51linuxstb_The option -mlong-jumps sounds like it should fix it - but it doesn't.
14:27:55amiconnHmm, -mlong-calls should indeed fix the problem, and in fact it shouldn't even arise:
14:28:17amiconn"The exception to this rule is that weak function definitions, functions with the `long-call' attribute or the `section' attribute, and functions that are within the scope of a `#pragma long_calls' directive, will always be turned into long calls."
14:30:05Slasheriamiconn: huh, i think the might be not bricked. It has just a problem with the reset cookie. Original bootloader initially booted my secondary bootloader, but that crashed. When i pressed reset, the reset cookie was set and now it tries to load immediately my secondary bootloader and that crashes.. So i have to wait the battery to run out (i haven't T5 screw driver at home so i could disconnect the power)
14:35:23 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
14:36:33 Join linuxstb__ [0] (
14:38:37linuxstb__amiconn: Yes, I read that it in the man page (about it being enabled for functions with the section attribute).
14:40:42linuxstb__I've just tried adding -mlong-calls again, and that actually causes it to fail with the same error, but in a different function - lcd-16bit.c instead of bookmark.c
14:41:59 Join frederic [0] (
14:45:31 Join Superman [0] (
14:47:16 Quit linuxstb__ (Read error: 104 (Connection reset by peer))
14:47:48 Join linuxstb__ [0] (
14:47:50Supermanhow can I see when progress is made for the h300 port?
14:49:34fredericcheck the wiki
14:49:50 Join linuxstb [0] (
14:50:08Supermanah thanks :)
14:51:12Supermanwhats a I2C driver?
14:52:12 Quit linuxstb_ (Read error: 110 (Connection timed out))
14:57:39***Saving seen data "./dancer.seen"
14:58:32 Quit frederic ("Chatzilla [Firefox 1.5/undefined]")
15:04:58 Quit ashridah ("Leaving")
15:05:45Slasheriamiconn: the problem was that the vbr was not 1M aligned.. So the secondary bootloader crashes on kernel system calls
15:05:58SlasheriIt should be fine when the player runs out of battery
15:06:26 Quit linuxstb__ (Read error: 110 (Connection timed out))
15:07:00 Join amiconn_ [0] (
15:11:05 Join ep0ch| [0] (n=ep0ch|@
15:11:44ep0ch|Slasheri: if you have some very small flat head screwdrivers you can get away with using that instead of a T5
15:12:36Slasheriep0ch|: probably i have somewhere but it could be still hard.. I think i will just wait the battery to go empty. Fortunately the disk is continuously spinning
15:13:05ep0ch|i hope you aint bricked it :)
15:13:12Slasherihehe, i hope too :)
15:16:16 Join linuxstb_ [0] (
15:23:32 Join muesli_- [0] (
15:23:53amiconn_linuxstb: What function was that? (in lcd-16bit.c)
15:24:32 Join andy [0] (
15:25:02amiconn_Slasheri: I hope you're right and your iriver isn't bricked. Too bad that the crash prevention cookie turned into a crash-trigger cookie, since you're using the bootloader 'the other way round'
15:25:26 Quit amiconn (Read error: 110 (Connection timed out))
15:25:26 Nick amiconn_ is now known as amiconn (
15:25:29 Join preglowII [0] (
15:26:12andyamiconn: I'm removing the monitor option from audio_set_recording_options..
15:26:41amiconnThe monitor function within the mas is something completely different
15:26:49andyaha, sorry =)
15:27:18amiconnIt doesn't switch the output signal on and off, but selects whether the mas is ready-to-record or actually encoding
15:28:23amiconnThe ready-to-record status is called monitoring, and it saves energy to use this status as long as the recording did not actually start
15:28:34amiconn(energy consumption similar to decoding mode)
15:28:48preglowIISlasheri: careful...
15:30:12amiconnhi preglowII
15:33:04 Quit linuxstb (Read error: 110 (Connection timed out))
15:33:12linuxstb_Hello preglowII. You'll be pleased to know that the ipod bootloader now compiles cleanly (in CVS), and rockbox itself is now very close to a clean compile.
15:33:43preglowIIahh, excellent
15:33:55preglowIIi'll start hacking again when i'm back tomorrow
15:34:43preglowIIprinting some arm manuals as we speak, hehe
15:34:46amiconnOkay, iram destination with sophisticated bitshifting isn't significantly slower than simple unaligned copy, so no need to handle it as special case
15:35:02amiconnSounds like I'll commit memset tonight
15:35:11preglowIIarm is a dream come true for fixed point maths
15:35:22preglowIIamiconn: memcpy, you mean?
15:35:30amiconnYes of course
15:35:40preglowIIlooking forward to it
15:35:52amiconnRead the numbers in today's log?
15:36:06amiconnspeeeed :-)
15:36:21preglowIIshould be most excellent with regard to the codecs and voice ui
15:36:26linuxstb_andy: Do you have plans to try to incorporate the wavpack encoder into your recording code?
15:36:57amiconnaway again
15:36:58preglowIIbtw, i take it we wont support simultaneous playback and recording?
15:37:18andylinuxstb_: is the encoder working?
15:37:26linuxstb_Recording from radio whilst playing back files could be a nice feature.
15:37:34linuxstb_andy: Yes - see the wav2wv plugin
15:37:44preglowIIi'm just wondering about iram
15:37:45linuxstb_It's much faster than realtime, so should work well.
15:37:52preglowIIwe'll need plugins for encoders as well
15:37:58preglowIIand there's not too much iram left for them...
15:38:31andylinuxstb_: nice.. i'm personally more interested in speex but we need a plugin api anyhow.. hm
15:38:59linuxstb_andy: Yes, the main work is creating a recording codec api
15:39:21preglowIIshouldn't be too much work, to be honest
15:39:38preglowIIwe already have the codec plugin system
15:40:03preglowIIso the ground work is done
15:40:15preglowIIandy: the speex encoder might be a tough nut, but i'm interested in speex too
15:41:01 Join Paul_The_Nerd [0] (
15:41:10andypreglowII: maybe someone on the speex-team is interested? :)
15:43:15preglowIIproblem is that libspeex has quite a lot of floating point code left
15:43:26preglowIIperhaps they are interested, i haven't asked
15:44:10andypreglowII: ok..
15:45:26 Quit Superman ("CGI:IRC (EOF)")
15:45:48preglowIIbut i gotta go
15:45:53 Quit preglowII ("CGI:IRC")
15:52:00linuxstb_andy: Does your "fix for broken simulators" commit also need to be applied to the other targets - h100, h300, ipod* ?
15:53:49Slasheriandy: Hmm, i think we can use the playback codec api with little modifications for recording too
15:59:25 Quit Paul_The_Nerd ("Chatzilla 0.9.68a [Firefox 1.0.7/20050915]")
16:06:42 Join len0x [0] (
16:09:45 Part len0x
16:17:03linuxstb_Slasheri: Sure. There are some differences though - such as a codec-specific configuration screen, byte/time-splitting the output at an appropriate place for each codec, making sure that all recorded data is safely written to a file.
16:30:41 Quit ender` (Read error: 104 (Connection reset by peer))
16:31:10 Join ender` [0] (i=ychat@
16:35:54 Quit ehntoo (Remote closed the connection)
16:38:18 Join linuxstb [0] (
16:40:32 Join ehntoo [0] (
16:41:00 Quit linuxstb (Read error: 104 (Connection reset by peer))
16:41:34 Join linuxstb [0] (
16:43:22 Join mashalla [0] (
16:44:15 Quit ehntoo (Remote closed the connection)
16:44:42 Join ehntoo [0] (
16:46:59 Quit ehntoo (Remote closed the connection)
16:47:47andylinuxstb_: no, that fix (config-h120.h) was because I enabled HAVE_RECORDING on h120 och that broke the h120 simulator
16:50:58linuxstbandy: OK, thanks for clarifying.
16:51:14linuxstbMaybe you should enable it for the h100 as well though.
16:51:41linuxstbBut I don't know anything about the h100.
16:53:03 Quit linuxstb_ (Read error: 110 (Connection timed out))
16:53:16 Join arkascha [0] (
16:55:06 Join ehntoo [0] (
16:57:43***Saving seen data "./dancer.seen"
16:58:25 Quit andy ()
17:16:41 Quit ehntoo ("Leaving")
17:16:58ep0ch|looks like enabling replaygain adds about 2.5% to the boost ratio
17:18:41 Join ehntoo [0] (
17:18:49 Join RotAtoR [0] (
17:21:22 Join _FireFly_ [0] (
17:23:18 Join muesli- [0] (
17:24:16 Quit muesli_- (Read error: 110 (Connection timed out))
17:25:14 Quit korpse (Remote closed the connection)
17:25:41 Join korpse [0] (
17:26:19 Join linuxstb_ [0] (
17:26:52 Quit linuxstb_ (Client Quit)
17:26:59 Join linuxstb_ [0] (
17:27:44linuxstb_amiconn: Does the tools/ipod_fw.c utility compile OK for you in cygwin?
17:28:04linuxstb_Someone has mentioned a problem on the forums - I'm guessing cygwin is involved.
17:30:33 Join Mxm`Pas`Bien [0] (
17:30:46 Quit korpse (Remote closed the connection)
17:31:09 Join _arkascha [0] (
17:36:16 Join korpse [0] (
17:36:23 Quit korpse (Client Quit)
17:38:32 Quit Maxime (Read error: 110 (Connection timed out))
17:43:55 Quit arkascha (Read error: 110 (Connection timed out))
18:03:53 Quit ep0ch| (Read error: 104 (Connection reset by peer))
18:12:10 Quit _arkascha (Read error: 104 (Connection reset by peer))
18:18:26 Quit muesli- (Read error: 110 (Connection timed out))
18:18:29amiconnlinuxstb: I just tried (no need to rebuild the tools everytime)
18:18:43amiconnipod_fw compiles cleanly on cygwin
18:19:38linuxstb_amcionn: Thanks for checking. It's fine on Linux as well. I'll ask the poster for more details.
18:20:20amiconnI have a fairly recent cygwin installation, I'm updating it regularly
18:20:27amiconnPerhaps it's a gcc problem
18:27:09 Quit XavierGr (Read error: 110 (Connection timed out))
18:31:43 Quit _FireFly_ (Read error: 113 (No route to host))
18:31:45 Join _FireFly_ [0] (
18:32:18 Join webguest17 [0] (
18:32:26 Quit webguest17 (Client Quit)
18:33:29amiconnlinuxstb: The idea behind the LCD_BLACK and LCD_WHITE macros was that you don't need to #ifdef between colour and b&w lcds
18:34:28amiconnJust #define LCD_WHITE ((struct rgb){LCD_MAX_RED-1, LCD_MAX_GREEN-1, LCD_MAX_BLUE-1}
18:34:54amiconnand #define LCD_BLACK ((struct rgb){0, 0, 0}
18:40:51linuxstb_amiconn: OK. I just copied that code from elsewhere in the same file. But I'll look at fixing it.
18:42:49 Join Philip_0729 [0] (
18:55:03Slasheriargh, the player is still running.. Almost 6 hours now with hd continuosly spinning.. :)
18:57:08 Join dpassen1 [0] (
18:57:46***Saving seen data "./dancer.seen"
19:01:55Slasherii can't boot it again before the battery goes empty :P
19:03:08_FireFly_?? why
19:03:28 Join ravelo [0] (
19:04:08Slasheribecause i flashed a secondary boot loader which crashes, and now the reset cookie prevents booting before it's reset.. And only power off can do that
19:04:36_FireFly_moep ;)
19:07:34_FireFly_argh connecting to is currently very slow
19:10:27_FireFly_i will only update my wps-widget on the patch-tracker
19:13:27 Join muesli_- [0] (
19:22:19_FireFly_ah updated :)
19:32:07_FireFly_i didn't get it
19:32:15_FireFly_what smee means
19:32:23muesli_-is me
19:32:24 Join arkascha [0] (
19:32:27muesli_-its me
19:33:07_FireFly_ah ok but my "ah updated" doesn't related to you ;)
19:33:19muesli_-ah ok ;)
19:36:33 Join Maxime [0] (
19:48:39 Quit Mxm`Pas`Bien (Read error: 110 (Connection timed out))
20:02:03 Join muesli- [0] (
20:06:27 Quit ravelo (Connection timed out)
20:09:16 Join XavierGr [0] (
20:15:05 Quit AliasCoffee ("bbl")
20:16:00thegeek_it works!
20:16:25thegeek_I just soldered in the new audio-connector LinusN sent me
20:16:37thegeek_and the connection was still bad
20:16:42thegeek_so I poked around a little
20:16:51thegeek_and found that the ground channel was bad
20:17:15thegeek_so I reheated/resoldered the tiny smd resistors that were the first inline with it
20:17:19thegeek_and .. it works:)
20:17:24thegeek_I feel like a pro
20:21:15XavierGrwhat was the problem?
20:22:13thegeek_it seems it was not the connector itself
20:22:18thegeek_but the small smd resistors
20:22:26thegeek_they sit on the "screen" side of the pcb
20:22:57XavierGrI mean why you resoldered again? What was you problem in the first place?
20:22:59thegeek_if you look at the pcb with the screen towards you, they are the two small components at the very top left
20:23:09thegeek_the sound was breaking up
20:23:15thegeek_I thought it was the connector itself
20:23:50XavierGrbreaking up? Like one time you have sound and then you don't ?
20:24:07 Quit muesli_- (Read error: 110 (Connection timed out))
20:24:09thegeek_it was only the ground channel
20:24:15thegeek_so the sound was "hollow" sometimes
20:24:17XavierGrhmm interesting...
20:24:19thegeek_I could bend the connector
20:24:23thegeek_and it would sound fine
20:24:42XavierGrI have something like this whenver I poke the plug with my finger
20:24:46thegeek_turns out the reason the sound was fine when I was bending, was because I was bending the pcb itself
20:25:08thegeek_I got like this when I once dropped the iriver, with my headphones plugged in, and it landed on the plug itself
20:25:40XavierGralso I hear clicks when a fadein/out is commenced or I change the volume (very faint clicks)
20:25:57XavierGrI have never dropped mine
20:26:42_FireFly_the faint clicks have i also when changing the volume sometimes when i navigating in the tree
20:26:50XavierGrbut maybe I will need your advice about that, maybe my problem with clicks is related to yours (well at least I have the same with you when I poke the plug so)
20:27:00thegeek_I dont have clicks
20:27:11XavierGr_FireFly_: not with the remote
20:27:16XavierGrmain unit
20:27:30XavierGrand these clicks appear on fade out it on the iriver fw too.
20:27:35XavierGrso it is harware related and not a rockbox bug.
20:27:59XavierGr_FireFly_: But please play with fade in/out settings on iriver fw and teel me your results
20:28:03 Quit Philip_0729 ("Miranda IM! Smaller, Faster, Easier.")
20:30:39_FireFly_i can find only fade in on the iriver fw
20:30:47_FireFly_as option
20:31:25XavierGrwhen I say clicks on fade in/out: It is like very small gaps with no sound (very very small but audible)
20:31:31XavierGryes I must be wrong
20:31:39XavierGronly fade in for the iiver
20:32:09XavierGrdoes it maks those little faint gaps when it fades in after a pause or stop?
20:32:45_FireFly_no or i can't hear it
20:33:09XavierGrthen my plauer is faulty as I thought
20:34:04XavierGralso sometimes I can hear a very faint click on track changes (on iriver firmware) but I am not sure about it. (I think that in the past I hadn't such problem.)
20:34:16XavierGrmaybe I will look at the audio connector again
20:35:08thegeek_sounds very wierd that the connector would do anything like that
20:35:47XavierGrthat might not a be a connector problem
20:35:56 Quit Kohlriba ("Leaving")
20:36:01XavierGrand as I think more of it, it can't be the connector to blame
20:36:10XavierGrbut what can it be?
20:37:34thegeek_I don't know;)
20:39:24thegeek_for the record, the resistors that caused my problems were L5 and L6
20:40:45 Join Kohlrabi [0] (
20:40:53 Quit Kohlrabi (Client Quit)
20:42:08 Join Kohlrabi [0] (
20:43:39 Quit Kohlrabi (Client Quit)
20:44:59XavierGrok thanks
20:45:48 Join Kohlrabi [0] (
20:47:50 Join POP|OPP|PPO [0] (
20:47:58POP|OPP|PPOdo Archos' use 2.5 ide?
20:48:04POP|OPP|PPOfor hard drives
20:48:42 Quit Kohlrabi (Client Quit)
20:48:50 Join Kohlrabi [0] (
20:48:52 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
20:49:50 Join Kohlrabi [0] (
20:57:49***Saving seen data "./dancer.seen"
20:59:06 Quit Kohlrabi ("Leaving")
21:01:29 Join Kohlrabi [0] (
21:01:37 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
21:02:06 Join Kohlrabi [0] (
21:24:22 Quit muesli- (Read error: 110 (Connection timed out))
21:56:27 Join TiMiD [0] (
22:00:52 Quit Kohlrabi (Nick collision from services.)
22:00:56 Join Kohlriba [0] (
22:09:34 Quit mashalla (Read error: 110 (Connection timed out))
22:11:07 Join muesli_- [0] (
22:12:51 Join muesli- [0] (
22:20:33muesli-TiMiD ?
22:22:18TiMiDyes ?
22:22:23TiMiDhi all !
22:22:35muesli-we will french fried tonite ;)
22:23:20muesli-->soccer ;)
22:24:57TiMiDsoccer ?
22:25:03 Quit _FireFly_ (Read error: 113 (No route to host))
22:25:31TiMiDyou go to play football in the night oO
22:25:54muesli-naah...its france vs germany tonight
22:26:07muesli-but 0:0 so far
22:26:09 Join Moos [0] (
22:26:21MoosHello all!
22:26:28TiMiDhi Moos
22:26:32muesli-another fried french ;)
22:26:40*TiMiD doesn't like soccer ...
22:26:41Moosyeah :)
22:26:56Moos0-0 but Germany play fine
22:27:47Moosmuesli: french team is'nt like few years old
22:28:14Moosnow it's "easy" to win it
22:29:02muesli-but they can still feed from their credits
22:29:16muesli-henry is a god
22:29:30MoosI don't like it
22:29:41muesli-doesnt matter how bad the rest it he can win the game alone
22:30:48Moosmuesli: Zidane is a God XD
22:31:32muesli-true too :D
22:32:24muesli-but henry got 3 times best scorer in premiere league..thats awesome either
22:33:03Moos1st league it's easier to make goal
22:33:14Moosnot like in Italia or Espana ;)
22:33:42muesli-but i guess france will win today anyway
22:34:14MoosOlympique Lyonais is a good team, maybe it will do a good Champions League
22:34:20muesli-they are doing next to nothing at the moment but once they will have a chance france will use it
22:34:57 Quit muesli_- (Read error: 110 (Connection timed out))
22:35:10muesli-yeah..moderator told they won the french champiosnship 4x in a row
22:35:45Moos2 years ago AS Monaco 2nd best team in Champions League
22:35:54 Quit Kohlriba ("Leaving")
22:36:03POP|OPP|PPOanyone upgraded the archos HD?
22:36:25 Join Kohlrabi [0] (
22:36:49Moosmuesli: OL is a good team cause his president is bilionaire
22:36:59muesli-who is it?
22:37:10muesli-money rulez the world ;)
22:37:21MoosJean Michel Aulas, one of the french who have got the more of euros :)
22:37:39muesli-POP|OPP|PPO i guess you want to know if archos' are using 2.5 drives
22:38:42Moosmuesli: Aulos is a bit like Chelsa Russian presidant ;)
22:39:08POP|OPP|PPOno i know that
22:39:16POP|OPP|PPOI want to know if anyone as upgraded
22:39:26POP|OPP|PPOI read the manual on the website
22:39:32POP|OPP|PPOit makes a little sense..
22:39:34Moosthe firmware?
22:39:40POP|OPP|PPOHD upgrade
22:40:14Moosyou probably can change your HD yes with the same size of original one
22:40:18muesli-did you tried it here
22:40:42POP|OPP|PPOdo any retailers still sell the old archos'
22:40:50POP|OPP|PPOi saw one at compusa but no one helped me..
22:42:29 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
22:45:49 Join ashridah [0] (
22:46:27Bagderstand by for my BMP change commit
22:47:02 Join _FireFly_ [0] (
22:51:08muesli-what does it do?
22:52:06 Quit ehntoo (Remote closed the connection)
22:52:12 Quit POP|OPP|PPO ()
22:52:58 Join ehntoo [0] (
22:53:19muesli-so we get a bmp-dir for gfx=
22:53:37Bagderand a wps dir
22:53:52Bagderthe bmp dir should be named like the wps itself
22:54:03muesli-very kewl :D
22:54:24muesli-so abc.wps ->abc directory?
22:54:43muesli-do we have to change our wps?
22:54:54Bagderhm, no
22:55:02Bagdernot if you didn't use any path before
22:55:15Bagderjust create a dir and move the bmps to that
22:55:15muesli-no, i did not
22:56:16muesli-very good idea
22:56:29BagderI think it is a step forward
22:57:01muesli-definitively..those x bmps in .rockbox are quite messy
22:57:50***Saving seen data "./dancer.seen"
22:59:09TiMiDThat's one small step for Bagder ; one giant leap for rockbox
22:59:34Bagderand the patch for it was surprisingly small
23:00:17muesli-btw didnt know that Bagder is an animal before i'Ve seen it in a dictionary ;)
23:00:29Bagderbadger is an aminal
23:00:32Bagderbagder is not
23:00:46muesli-tricky ;)
23:01:31linuxstb_Bagder: Do you have any ideas about my ARM and ICODE problem? (see today's IRC logs)
23:05:03Bagderno idea
23:05:57BagderI noticed your commit about it and was a bit curious ;-)
23:09:36 Quit ender` (Read error: 113 (No route to host))
23:14:06 Join ender` [0] (i=ychat@
23:21:21amiconnBagder: Yellow build alert
23:22:26 Quit _FireFly_ (Read error: 113 (No route to host))
23:25:15 Quit muesli- (Read error: 110 (Connection timed out))
23:34:02XShocKi soldered RTC in my iriver ihp120
23:34:13XShocKnow need to write a driver for it
23:35:07 Join _arkascha [0] (
23:41:28 Quit arkascha (Read error: 110 (Connection timed out))
23:45:44thegeek_nice XShocK
23:47:03 Join lokki [0] (
23:49:01lokkiso does this basically mean you will have all the cool features like clock calendar wakeup in the morning etc?
23:53:43XShocKi didn't add wakeup.
23:53:53XShocKwill add it later
23:55:09XShocKhave a floating pin of an interrupt. i first want it to work just ac a calendar and clock, when everything is bug free will add wakeup
23:56:11lokkithat would/will be amazing
23:57:45lokkican you set up a wiki page about it?
23:58:08Moospic could be fun :)
23:58:49XShocKok. i will

Previous day | Next day