2010-08-21 00:00 and these tools allow you to make physical RAM pretend that it's swap? 2010-08-21 00:00 is there any way to compile the hello world C application for nanonote and execute it on the platform without actually recompiling the whole kernel 2010-08-21 00:00 welcomes qwebirc 2010-08-21 00:02 we need to publish the toolchain or some kind of SDK... 2010-08-21 00:03 haha 2010-08-21 00:04 not just pretend to be swap 2010-08-21 00:04 pretend to be swap, and compress and decompress on the fly in and out 2010-08-21 00:04 nebajoth: just making sure I understand what you're doing. =D 2010-08-21 00:04 yes !!!! :) 2010-08-21 00:04 most distros seem to be using this already 2010-08-21 00:04 at least for their installers 2010-08-21 00:04 turtlee: yes 2010-08-21 00:04 anybody having a solution for my problem 2010-08-21 00:04 wolfspraul: keep is simple. toolchain and selected headers/libraries. like in openmoko. that worked great. 2010-08-21 00:05 s/is/it/ 2010-08-21 00:05 why would you have to recompile the kernel? 2010-08-21 00:05 nebajoth: probably as part of building all of openwrt ... 2010-08-21 00:05 yes exactly 2010-08-21 00:06 oh 2010-08-21 00:06 I dunno anything about that 2010-08-21 00:06 I run openwrt on my routers :P 2010-08-21 00:06 not my gtd nn 2010-08-21 00:06 cant we have any toolchain only to compile the C file and port it on nanonote using usbnet and run it 2010-08-21 00:07 qwebirc91025: i just bit the bullet, ran the whole build, and then symlinked all the executables into /usr/local/bin 2010-08-21 00:08 qwebirc91025: yeah that should work 2010-08-21 00:08 hell I'm running openwrt compiled sound modules in my debian I think 2010-08-21 00:09 but how ? 2010-08-21 00:09 can you plz tell me the procedure 2010-08-21 00:09 I'm not sure I know it 2010-08-21 00:09 I have to compile with a mips toolchain myself 2010-08-21 00:09 qwebirc91025: here it is: http://en.qi-hardware.com/wiki/Building_Software_Image 2010-08-21 00:09 right now I just need to develop a C aplication to print some thing on console 2010-08-21 00:09 I was either going to do it with openwrt 2010-08-21 00:09 or right on the nn 2010-08-21 00:10 you can compile on the NN with debian :P 2010-08-21 00:10 once done,  ln -s staging_dir/toolchain-mipsel_gcc-4.3.3+cs_uClibc-0.9.32/usr/bin/mipsel-openwrt-linux-* /usr/local/bin/ 2010-08-21 00:11 wpwrak : if i go with this procedure then i need to port the kernel image ? 2010-08-21 00:11 am I correct  ? 2010-08-21 00:11 does the basic pyneo debian install slow the NN noticably more than the stock OpenWRT? 2010-08-21 00:12 i think it builds a kernel in the process. you will be asked a few questions. just hit enter to accept the defaults. 2010-08-21 00:12 turtlee: I didn't keep openwrt on there long enough to know really 2010-08-21 00:12 but yeah probabl 2010-08-21 00:12 y 2010-08-21 00:12 I still find it totally usable though 2010-08-21 00:12 bootup takes a little longer 2010-08-21 00:12 it could probably be optimized though 2010-08-21 00:12 but ultimately we need to port the whole kernel image ? isnt 2010-08-21 00:13 qwebirc91025: why worry about the kernel ? there's plenty more junk in openwrt ;-) 2010-08-21 00:14 qwebirc91025: the build process takes care of it all. you don't have to pay attention to all the things it builds. 2010-08-21 00:14 qwebirc91025: for the configuration, instead of "menuconfig" use "oldconfig". i.e., accept things as they are. 2010-08-21 00:15 rofl: "When code from the staging tree is loaded in the kernel, a warning message will be printed to the kernel log, and the kernel will be tainted with the TAINT_CRAP flag." 2010-08-21 00:15 epic 2010-08-21 00:16 sounds like Greg :) 2010-08-21 00:16 would snort his beverage out his nose if he had a beverage. 2010-08-21 00:18 wpwrak: yep 2010-08-21 00:19 I am worrried about that because in software development for a minor in my application I will have to port the whole kernel again and again 2010-08-21 00:20 install debian, compile on the Nanonote itself :D 2010-08-21 00:21 http://www.aurel32.net/info/debian_mips_qemu.php 2010-08-21 00:21 qwebirc91025: the kernel is already ported, openwrt merely compiles it one more time :) 2010-08-21 00:21 http://code.google.com/p/jz-hacking/wiki/qemujz 2010-08-21 00:21 qwebirc91025: and no, you don't need to go through that ritual again once you have built your toolchain 2010-08-21 00:22 qwebirc91025: you'll have a mipsel-openwrt-linux-gcc, mipsel-openwrt-linux-ld, etc., which you then use for cross-compiling 2010-08-21 00:26 I am little bit confused now 2010-08-21 00:27 let me tell you what all I have already done 2010-08-21 00:27 I have downloaded source code and toolchian from Ben nanonote website 2010-08-21 00:28 qwebirc91025: the "toolchain" or the "tools" ? 2010-08-21 00:29 and configured it as per http://en.qi-hardware.com/wiki/Building_Software_Image 2010-08-21 00:29 does this thing have an internal clock? 2010-08-21 00:30 I am having one folder namely toolchain having binutils , gcc etc varous folders 2010-08-21 00:31 and tools folder is also present but not in toolchain directory 2010-08-21 00:31 qwebirc91025: sounds good so far 2010-08-21 00:32 nebajoth: what do you mean? 2010-08-21 00:32 rtc? 2010-08-21 00:33 I mean 2010-08-21 00:33 that it loses the date 2010-08-21 00:33 and I have to rdate it back 2010-08-21 00:33 I haven't looked into it at all 2010-08-21 00:33 maybe this has already been addressed on the wiki or mailing list 2010-08-21 00:34 maybe its debian specific 2010-08-21 00:34 do you have that issue with jLime? 2010-08-21 00:35 nebajoth: no.. rtc works okey 2010-08-21 00:35 ok 2010-08-21 00:36 so its a debian thing 2010-08-21 00:36 nebajoth: but it did not do before.. I had to add /dev/rtc* device files.. 2010-08-21 00:36 so hwclock -w .. etc works 2010-08-21 00:36 aha 2010-08-21 00:36 just add the device files and it works? 2010-08-21 00:37 nebajoth: well.. it has several parts.. you need a kernel with rtc driver, etc.. 2010-08-21 00:37 awesome 2010-08-21 00:37 then you need the proper files under /dev/* 2010-08-21 00:37 and then to test hwclock/date commands ;) 2010-08-21 00:37 yeah, I'm about to compile 2.6.36 anyway 2010-08-21 00:37 I'll check for rtc driver now 2010-08-21 00:37 you have it compiled into the kernel or as a module? 2010-08-21 00:37 now if I go to openwrt-xburst directory and execute make menuconfig FORCE=1 cmd then kernel configuration screen pops up 2010-08-21 00:37 inside the kernel 2010-08-21 00:38 now tell me what to do next ? 2010-08-21 00:38 I'm sure its turned on actually 2010-08-21 00:38 I grabbed your config file 2010-08-21 00:38 to build on 2010-08-21 00:38 qwebirc91025: i would run  make oldconfig, not menuconfig. just hit ENTER for all the questions. 2010-08-21 00:38 ok 2010-08-21 00:39 or yes "" | make oldconfig 2010-08-21 00:39 qwebirc91025: also, FORCE doesn't work reliably. you'll have to edit out the silly test in include/prereq-build.mk 2010-08-21 00:39 nebajoth: you can check you rtc driver in nn kernel with : 2010-08-21 00:39 Jlime$ dmesg | grep -i rtc 2010-08-21 00:39 jz4740-rtc jz4740-rtc: rtc core: registered jz4740-rtc as rtc0 2010-08-21 00:39 jz4740-rtc jz4740-rtc: setting system clock to 2010-06-20 20:30:31 UTC (1277065831) 2010-08-21 00:39 "yes "" | make oldconfig" 2010-08-21 00:39 (for example) 2010-08-21 00:39 damned double quotes 2010-08-21 00:39 but you get my point 2010-08-21 00:40 this is my favourite irc channel 2010-08-21 00:40 yes 2010-08-21 00:40 I'm using the older qi-hw kernel 2010-08-21 00:40 2.6.32.10 2010-08-21 00:41 and yes, rtc is on 2010-08-21 00:41 but 2010-08-21 00:41 [    9.660000] jz4740-rtc jz4740-rtc: setting system clock to 1970-01-01 00:00:00 UTC (0) 2010-08-21 00:41 1970! 2010-08-21 00:42 /dev/rtc0 exists 2010-08-21 00:42 did you change the time at some moment? :) 2010-08-21 00:43 date,...hwclock ? 2010-08-21 00:43 [commit] Werner Almesberger: Added register definitons and simplified register naming. http://qi-hw.com/p/ben-wpan/1d777aa 2010-08-21 00:43 [commit] Werner Almesberger: fw/atspi/atspi.c (reset_rf): added reset timing measurement http://qi-hw.com/p/ben-wpan/44f0c38 2010-08-21 00:43 [commit] Werner Almesberger: Enter TRX_OFF to enable DVDD regulator. Plus minor cleanup. http://qi-hw.com/p/ben-wpan/7a534d2 2010-08-21 00:43 what has to be done to remove FORCE=1 option ? 2010-08-21 00:43 I've set it with rdate 2010-08-21 00:43 multiple times 2010-08-21 00:43 rdate -s time.nrc.ca 2010-08-21 00:44 aha 2010-08-21 00:44 but 2010-08-21 00:44 qwebirc91025: where it says "define Require/non-root", change the next line to "true" 2010-08-21 00:44 after running rdate 2010-08-21 00:44 the command "date" shows the new correct time 2010-08-21 00:44 but hwclock 2010-08-21 00:44 still shows 1970 2010-08-21 00:45 ok 2010-08-21 00:46 ran hwclock -w 2010-08-21 00:46 lets see how it goes now :P 2010-08-21 00:46 thx rafa 2010-08-21 00:47 can you plz tell me the syntac 2010-08-21 00:48 in next line "[ "$$(shell whoami)" != "root" ] is written 2010-08-21 00:48 nebajoth: you are welcome 2010-08-21 00:51 I have made it "== root" 2010-08-21 00:51 so dont need to use FORCE=1 2010-08-21 00:51 now what ? 2010-08-21 00:53 is the LED connected via GPIO? 2010-08-21 00:59 qwebirc91025: yes "" | make oldconfig 2010-08-21 00:59 qwebirc91025: make -j 2010-08-21 01:00 e.g., make -j5 2010-08-21 01:00 lean back and relax while openwrt builds all the stuff 2010-08-21 01:01 your toolchain will be in staging_dir/toolchain-mipsel_gcc-4.3.3+cs_uClibc-0.9.32/usr/bin/ 2010-08-21 01:02 hmm. i better get some sleep. barbecue in 12 hours. terrace oblige :) 2010-08-21 01:03 is not barbecue some meat made with fire only? :) 2010-08-21 01:04 barbecue, nice - 'night 2010-08-21 01:04 how to chek that hwo many cores are there 2010-08-21 01:06 haha terrace oblige 2010-08-21 01:06 how many cores are where? 2010-08-21 01:06 in your computer? 2010-08-21 01:07 cat /proc/cpuinfo |grep processor|wc -l 2010-08-21 01:07 are you talking about processor or ? 2010-08-21 01:09 OK gentlemen 2010-08-21 01:09 I believe he was 2010-08-21 01:09 what other kinds of core are there? 2010-08-21 01:09 OK gentlemen, I'm undertaking a kernel compile FOR the nanonote, ON the nanonote 2010-08-21 01:09 I am timing it :P 2010-08-21 01:10 I'll let you know what the results are 2010-08-21 01:10 ... in a few days 2010-08-21 01:10 ... when its complete 2010-08-21 01:11 fine 2010-08-21 01:11 I got it 2010-08-21 01:18 nebajoth: thanks for the advice. wish me luck on my debian install. 2010-08-21 01:18 ah yes 2010-08-21 01:18 good luck :D 2010-08-21 01:18 lmk how it goes 2010-08-21 01:18 I'll be here for a bit still 2010-08-21 01:19 Heh. I won't. time for sleeping. will try the flash over the next day or two. 2010-08-21 01:19 ok 2010-08-21 01:19 good luck :) 2010-08-21 01:20 I eventually intend to release a new rootfs 2010-08-21 01:20 but I want to get this new kernel working first 2010-08-21 01:20 NN is gonna be busy for the next couple days 2010-08-21 01:20 compiling :P 2010-08-21 01:20 Very cool. 2010-08-21 01:20 Oh! 2010-08-21 01:20 one more thing. 2010-08-21 01:21 any recommendation on how to go about formatting and setting up a swap part on the nand? 2010-08-21 01:21 are you going to use the whole thing or partition it? 2010-08-21 01:21 actually 2010-08-21 01:21 not sure it would be very easy to partition it 2010-08-21 01:22 ummm, you can either use the whole mtd partition 2010-08-21 01:22 or you can make a swap file 2010-08-21 01:22 up to you 2010-08-21 01:22 guess you don't need that big a swap file 2010-08-21 01:22 er that big a swap 2010-08-21 01:23 so I'd make a swap file on the nand and use that 2010-08-21 01:23 that tickles my memory of something on the mailing list involving xiang-fu 2010-08-21 01:23 ohh, 2010-08-21 01:23 but I didn't see it while cruising through the last 6 months yesterday 2010-08-21 01:23 swap file is super-obvious. sorry. 2010-08-21 01:24 alright. time for sleeping. 2010-08-21 01:24 see you later. 2010-08-21 01:27 he's a nice fellow 2010-08-21 01:47 nebajoth: yes we are lucky with the people that show up. it drives me nuts that the NanoNote is not more useful from a regular user standpoint today. argh. but the direction is right... 2010-08-21 02:02 we'll get there 2010-08-21 04:55 wolfspraul: one program I lack in the nanonote is something that would allow me to store and navigate maps 2010-08-21 04:55 I have to try that nanomap still. 2010-08-21 04:59 viric: yes, nanomap :-) but I haven't tried it myself yet... 2010-08-21 04:59 definitely I want that too though 2010-08-21 04:59 bbl 2010-08-21 08:18 [commit] Andres Calderon: ddr-vref improved placement http://qi-hw.com/p/xue/1ba7525 2010-08-21 09:05 Hello 2010-08-21 09:05 anyone using mpg321 in the nanonote? 2010-08-21 09:08 nope that i'm aware off 2010-08-21 09:08 did you tried gmu? 2010-08-21 09:18 there is also a command line program to play music called: oggplay 2010-08-21 09:20 I'll try 2010-08-21 09:20 I simply saw that mpg321 seems to require more cpu power than the ben has 2010-08-21 09:21 while madplay seems to have plenty of cpu 2010-08-21 09:35 mpg321? well gmu plays mp3 fine but uses mpg123 i think 2010-08-21 09:36 cpu power shouldnt be any problem at all for playing music 2010-08-21 09:38 hm 2010-08-21 09:39 mpg123 should be using floating point, while mpg321 should not 2010-08-21 09:39 (for what I read about them) 2010-08-21 09:39 so mpg321 should work *faster* in a cpu without fpu 2010-08-21 09:39 and at the end, mpg321 uses libmad, that of madplay too. 2010-08-21 09:39 wier.d 2010-08-21 09:40 well mpg123 works fine so just use it 2010-08-21 09:41 I'll try. 2010-08-21 09:42 if you need binarys, use the ones from the dingoo version of gmu 2010-08-21 09:42 no no, I build myself 2010-08-21 09:42 thank you 2010-08-21 09:44 mpg321 is very old and generally should not be used 2010-08-21 09:45 also, mpg123 has both floating point and fixed point decoders included 2010-08-21 09:45 ahh 2010-08-21 09:46 There was a time when mpg321 was newer than mpg123 :) 2010-08-21 09:46 that is long ago 2010-08-21 09:46 mpg123 is actively maintained 2010-08-21 09:46 ahh 2010-08-21 09:47 last mpg123 release is from july 11th 2010 2010-08-21 09:47 great 2010-08-21 09:48 yeah, works really good 2010-08-21 09:56 do you use softvol? 2010-08-21 09:56 I'd like to get from someone the alsa config for that. 2010-08-21 09:56 I could not manage to get it working 2010-08-21 10:03 [commit] Andres Calderon: some eth-phy to s6 nets has been routed http://qi-hw.com/p/xue/4193109 2010-08-21 10:14 [commit] Andres Calderon: some eth-phy to s6 nets has been routed http://qi-hw.com/p/xue/195bfb9 2010-08-21 12:08 hello everybody 2010-08-21 12:16 I'm trying to compile the toolchain for the new image, but I get this error: 2010-08-21 12:16 In file included from ./include/bits/syscalls.h:11, 2010-08-21 12:16                  from ./include/sys/syscall.h:34, 2010-08-21 12:16                  from ./libc/sysdeps/linux/common/sysdep.h:20, 2010-08-21 12:16                  from ./libc/sysdeps/linux/mips/sysdep.h:25, 2010-08-21 12:16                  from libpthread/nptl/sysdeps/mips/tcb-offsets.c:1: 2010-08-21 12:16 ./include/errno.h:69: error: thread-local storage not supported for this target 2010-08-21 12:16 ./include/errno.h:69: warning: `tls_model' attribute ignored 2010-08-21 12:16 make[4]: *** [libpthread/nptl/sysdeps/mips/tcb-offsets.s] Error 1 2010-08-21 12:16 wejp: I just built mpg123 (generic-cpu optimizations), and it looks like not having enough cpu either 2010-08-21 12:17 99% of the CPU spent on 'system' 2010-08-21 12:17 using the 'alsa backend' 2010-08-21 12:17 clearly something goes wrong 2010-08-21 12:17 viric, you nee to build it with generic_no-fpu 2010-08-21 12:17 the generic decoder uses floating point math, you really need to tell configure to use the no_fpu decoder 2010-08-21 12:18 ah ok 2010-08-21 12:18 ah, hence the 'system' load! I guess it's using the kernel fpu emulation 2010-08-21 12:18 with that decoder gmu is able to decode mp3 files with any bitrate with less than 25 % cpu load on the ben 2010-08-21 12:18 I've seen 22% on madplay 2010-08-21 12:18 hehe, yeah kernel fpu emulation is especially slow ;) 2010-08-21 12:19 yeah, madplay is a little slower than mpg123 but still okay 2010-08-21 12:19 nevertheless I don't know what code my gcc generated, so that the kernel fpu operations are used 2010-08-21 12:19 for ARM cpus mpg123 has an even more optimized decoder, but that doesn't help you on mips cpus ;) 2010-08-21 12:19 I did not expect my gcc to produce code for fpu 2010-08-21 12:19 I have a sheevaplug, but no sound card on it. 2010-08-21 12:20 Maybe for the gp2x I could do something :) 2010-08-21 12:20 you could attach an usb soundcard :D 2010-08-21 12:20 :) yes 2010-08-21 12:20 yeah, gmu on the gp2x uses mpg123 with the arm optimized version :) 2010-08-21 12:23 How can I know if gcc generated fpu instructions? 2010-08-21 12:24 there may be an elf flag for it 2010-08-21 12:24 you can tell gcc to explicitly use its own fpu emulation layer, i think it is -mfloat-abi=softfp 2010-08-21 12:24 but you don't want that 2010-08-21 12:24 I know how to set that explicit 2010-08-21 12:24 it is much faster than kernel emulation, but still damn slow 2010-08-21 12:25 Wa! Here I go! 2010-08-21 12:25 13% user cpu, 2% system1 :) 2010-08-21 12:25 :) 2010-08-21 12:25 192kbps stereo 2010-08-21 12:25 yep :) 2010-08-21 12:25 far below 25% :) 2010-08-21 12:25 yeah, just decoding mp3 is pretty fast 2010-08-21 12:25 well, it plays. 2010-08-21 12:25 so it goes to the sound card 2010-08-21 12:26 gmu does some graphics output too, which needs some cpu, but it is still pretty fast even with that 2010-08-21 12:26 ok 2010-08-21 12:26 nothing I need, that gmu :) 2010-08-21 12:26 thank you very much 2010-08-21 12:26 you're welcome :) 2010-08-21 12:26 Aren't you using softvol, btw? 2010-08-21 12:26 yeah 2010-08-21 12:26 can you paste your softvol configuration so I could take a look? 2010-08-21 12:27 I could not manage to configure it 2010-08-21 12:27 what exactly do you mean? the source code? 2010-08-21 12:27 the asoundrd file 2010-08-21 12:27 asoundrc 2010-08-21 12:27 oh, sorry, no i do not use alsa for software volume control 2010-08-21 12:28 it should have the description of the softvol mixer setting 2010-08-21 12:28 Then what volume do you use? 2010-08-21 12:28 the hardware mixer? :) 2010-08-21 12:28 i scale the audio signal itself before giving it to alsa (or whatever audio driver is being used) 2010-08-21 12:28 ahm 2010-08-21 12:28 mpg123 can do that? 2010-08-21 12:28 no 2010-08-21 12:29 oh :) 2010-08-21 12:29 i use mpg123 just for decoding mp3 data, the volume control and audio output takes place at another point 2010-08-21 12:29 ah.. 2010-08-21 12:29 maybe I could pipe it to sox 2010-08-21 12:30 that should be possible 2010-08-21 12:30 i think the mpg123 executable can output the data to stdout 2010-08-21 12:30 but doing that on the *decoded* signal looks like a heavy job. 2010-08-21 12:30 many samples a second 2010-08-21 12:31 it is not that bad and actually you could not even do it easily on the encoded signal 2010-08-21 12:31 and even if you could, you would not want to do that in a music player that is capable of decoding lots of different file formats 2010-08-21 12:31 I mean that if mpg123 did that, it would not that be that a heavy task I guess 2010-08-21 12:31 for the cpu 2010-08-21 12:32 anyway, I have mp3 playing. Great! :) 2010-08-21 12:32 I wonder what I need more... 2010-08-21 12:32 it could be even worse, the decoding process is quite complex ;) 2010-08-21 12:32 I should try to start X some day. 2010-08-21 12:32 :D 2010-08-21 12:32 i think X is a bit overkill on such a device 2010-08-21 12:33 there is the tiny keith X 2010-08-21 12:33 it should run fine there 2010-08-21 12:33 I could cross-build it, but I did not go into configuring. :) 2010-08-21 12:33 i've used X on the zipit z2 (which has similar specs) and it is rather slow and the screen resolution does not help either ;) 2010-08-21 12:33 hm 2010-08-21 12:33 but it is possible 2010-08-21 12:34 should be possible on the Ben too 2010-08-21 12:34 What other options are, to run GTK or qt programs? 2010-08-21 12:34 you can use fbdev for video output 2010-08-21 12:34 for mplayer I know 2010-08-21 12:34 maybe people use qtopia on the fbdev? 2010-08-21 12:34 you can use fbdev for X and there is a gtk port which outputs to fbdev directly 2010-08-21 12:34 there were some problems with that on the ben, though 2010-08-21 12:35 ah 2010-08-21 12:35 yeah, qtopia runs on fbdev too 2010-08-21 12:35 there are plenty of things to learn :) 2010-08-21 12:35 maybe one could run a tiny x server instead of xorg 2010-08-21 12:36 The keith server I said 2010-08-21 12:36 It comes with xorg. 2010-08-21 12:36 build the server with "--enable-keith" and you are there 2010-08-21 12:36 yeah, you could try that one 2010-08-21 12:36 it runs on fbdev 2010-08-21 12:37 --enable-kdrive, sorry 2010-08-21 12:37 I built it, but I did not try to start it 2010-08-21 12:37 yep right, kdrive was the name :) 2010-08-21 12:37 kdrive is made by Keith. I always mix that :) 2010-08-21 12:38 I'm busy with another mips... kdrive for another day. 2010-08-21 12:38 guys, sorry to interrupt, but I'm trying to compile the mpg123 lib for gmu 2010-08-21 12:38 but I get errors when compiling the tool-chain :( 2010-08-21 12:39 what kind of errors? could you show us some error log? 2010-08-21 12:40 In file included from ./include/bits/syscalls.h:11, 2010-08-21 12:40                  from ./include/sys/syscall.h:34, 2010-08-21 12:40                  from ./libc/sysdeps/linux/common/sysdep.h:20, 2010-08-21 12:40                  from ./libc/sysdeps/linux/mips/sysdep.h:25, 2010-08-21 12:40                  from libpthread/nptl/sysdeps/mips/tcb-offsets.c:1: 2010-08-21 12:40 ./include/errno.h:69: error: thread-local storage not supported for this target 2010-08-21 12:40 ./include/errno.h:69: warning: `tls_model' attribute ignored 2010-08-21 12:40 make[4]: *** [libpthread/nptl/sysdeps/mips/tcb-offsets.s] Error 1 2010-08-21 12:40 this is where it stops... I'm on Ubuntu 10.04 and I have all the dependencies needed 2010-08-21 12:41 hm 2010-08-21 12:41 I just added qt and libSDL to the config file 2010-08-21 12:51 woop 2010-08-21 12:51 373 minutes, 20 seconds 2010-08-21 12:52 to compile 2.6.36 kernel 2010-08-21 12:52 ON the nanonote 2010-08-21 12:52 actually that's not as bad as I thought it would be 2010-08-21 12:58 oO 2010-08-21 12:59 sup 2010-08-21 13:06 nebajoth: with gcc? 2010-08-21 13:06 nebajoth: I would not expect gcc to have enough ram 2010-08-21 13:07 it doesn't 2010-08-21 13:07 that's why I gave it 1G of swap :P 2010-08-21 13:07 I run debian 2010-08-21 13:07 I had to assign swap anyway, just to run apt-get 2010-08-21 13:07 uh :) 2010-08-21 13:08 swap on SD? 2010-08-21 13:08 or a network block device? 2010-08-21 13:08 (just not to wear down Flash memory :) 2010-08-21 13:08 microsd 2010-08-21 13:08 You don't love that microsd very much :) 2010-08-21 13:09 #1 no 2010-08-21 13:09 I don't 2010-08-21 13:09 I have 4 others identical to it 2010-08-21 13:09 #2 wearing out takes forever 2010-08-21 13:09 much longer than I'll likely have a practical use for it 2010-08-21 13:09 people get really upset about wearing of flash memory 2010-08-21 13:09 and I don't really understand why 2010-08-21 13:09 when they generally outlive their mechanical cousins 2010-08-21 13:10 ok 2010-08-21 13:10 I never got a flash memory to wear down too 2010-08-21 13:11 right 2010-08-21 13:11 I'm sure it does happen 2010-08-21 13:11 but last time I looked at the numbers 2010-08-21 13:11 it was like 2010-08-21 13:11 years 2010-08-21 13:11 I have some friends that worked on embedded systems, and they are used to flash wearing down 2010-08-21 13:11 But maybe they write the flash a lot. 2010-08-21 13:12 argh, there is 2.6.35.3 already. 2010-08-21 13:16 you can swap to zram too, by the way 2010-08-21 13:16 a compressed block device stored in RAM 2010-08-21 13:17 gives you some extra memory before swapping to SD 2010-08-21 13:21 how much you set for zram? 2010-08-21 13:22 Ayla set it to 10 MB on Dingux, afaik 2010-08-21 13:23 dingux is a device or a gnu distribution? 2010-08-21 13:23 Dingux is Linux for the Dingoo A320/A330 2010-08-21 13:23 does that have also 32MB of RAM? 2010-08-21 13:23 yes 2010-08-21 13:23 ah 2010-08-21 13:23 zram is new in 2.6.35 2010-08-21 13:24 great 2010-08-21 13:24 I backported it to 2.6.34 though 2010-08-21 13:24 I run 2.6.35 2010-08-21 13:24 I could try. 2010-08-21 13:24 (with 2.6.34, mmc does not work for me) 2010-08-21 13:25 there was a bug fixed in the MMC insert detection, I don't know if that's related? 2010-08-21 13:42 mth: yes 2010-08-21 13:42 I have it compiled into the kernel that took 6h to compile :P 2010-08-21 13:43 zram, that is 2010-08-21 13:43 in fact, that's the entire reason I undertook to build the kernel 2010-08-21 13:43 6h? you compile it on the NanoNote itself? 2010-08-21 13:43 yup 2010-08-21 13:43 I was interested in how long it would take 2010-08-21 13:43 6h was a lot shorter than I expected, tbh 2010-08-21 13:44 I thought it'd still be going when I woke up toda 2010-08-21 13:44 y 2010-08-21 13:44 mth: no, the insert detection worked fine. It simply did not understand my sd card 2010-08-21 14:05 nebajoth, 6hours isnt that bad :) good to know that it works 2010-08-21 14:40 does anybody know if there's anything special that must be done to compiled kernel before it can be flashed to the NN? 2010-08-21 15:33 nothing special, you just need the compiled uImage 2010-08-21 17:39 [commit] Andres Calderon: s6 to eth-phy connections has been completed http://qi-hw.com/p/xue/30b1282 2010-08-21 19:08 [commit] Andres Calderon: nand routing just started http://qi-hw.com/p/xue/de965fa 2010-08-21 19:16 nebajoth: you really need to learn how to use a toolchain man 2010-08-21 19:16 nebajoth: building on host is nice and fun.. but it is not to suffer 2010-08-21 19:17 nebajoth: you can do the same using a toolchain.. and you will have a kernel built in just a few seconds.. or minutes 2010-08-21 22:29 rafa: oh totally, I was just curious and didn't feel like setting up a toolchain yet 2010-08-21 22:29 I'll definitely be doing that this week