IRC log for #nslu2-linux on 20070429

00:24.13*** join/#nslu2-linux jbot (i=ibot@pdpc/supporter/active/TimRiker/bot/apt)
00:24.13*** topic/#nslu2-linux is nslu2-linux developer channel - use #nslu2-general for end-user questions
00:52.55*** join/#nslu2-linux pierrelux (n=pierre-l@144-125.sh.cgocable.ca)
02:44.59*** join/#nslu2-linux drath_ (i=vmaster@p5B07E7F1.dip.t-dialin.net)
03:12.03*** join/#nslu2-linux caplink811_log (n=caplink8@dslb-088-073-155-080.pools.arcor-ip.net)
03:26.00*** part/#nslu2-linux pierrelux (n=pierre-l@144-125.sh.cgocable.ca)
03:30.33CIA-3003osas * r5992 10optware/trunk/make/asterisk14-core-sounds-en-ulaw.mk: asterisk14-core-sounds-en-ulaw: 1.4.5 -> 1.4.6
03:31.39CIA-3003osas * r5993 10optware/trunk/make/asterisk14-extra-sounds-en-ulaw.mk: asterisk14-extra-sounds-en-ulaw: 1.4.4 -> 1.4.5
03:32.07CIA-3003osas * r5994 10optware/trunk/make/asterisk14-extra-sounds-en-gsm.mk: asterisk14-extra-sounds-en-gsm: 1.4.4 -> 1.4.5
03:33.59CIA-3003osas * r5995 10optware/trunk/make/asterisk14-core-sounds-en-gsm.mk: asterisk14-core-sounds-en-gsm: new package
03:34.41*** join/#nslu2-linux yvasilev (n=yvasilev@189.140.140.159)
04:09.11CIA-3003osas * r5996 10optware/trunk/make/asterisk14.mk: asterisk14: reducing the size of the package (we have a separate package for core-sounds-en-gsm)
04:35.25*** join/#nslu2-linux UkG (n=Sinclair@p5b203569.dip0.t-ipconnect.de)
06:13.59*** join/#nslu2-linux eiffel (n=eiffel@p5B052A5D.dip0.t-ipconnect.de)
06:15.52*** join/#nslu2-linux DrZimmerman (n=theo@84-74-134-243.dclient.hispeed.ch)
07:01.02*** join/#nslu2-linux pleemans (n=peter@d51A5E76A.access.telenet.be)
08:43.50*** join/#nslu2-linux Tobbe (n=Tobbe@120.19.235.85.dk-no.res.dyn.perspektivbredband.net)
08:46.21*** join/#nslu2-linux caplink811_log (n=caplink8@dslb-088-073-004-163.pools.arcor-ip.net)
09:09.06*** join/#nslu2-linux bullet (n=bullet@8.204.203.62.cust.bluewin.ch)
10:40.53*** join/#nslu2-linux Pete_ (n=pete@nock.demon.co.uk)
10:55.06Pete_anyone here using asterisk on their slug?
11:46.18*** join/#nslu2-linux yvasilev (n=yvasilev@189.140.140.159)
12:08.35CIA-3003osas * r5997 10optware/trunk/Makefile: asterisk14-core-sounds-en-gsm: promoting asterisk14-core-sounds-en-gsm package
12:10.47*** join/#nslu2-linux osas (n=nnnnnnnn@CABLE-72-53-75-252.cia.com)
12:56.28*** join/#nslu2-linux GPSFan (n=kenm@12.10.255.248)
14:40.19*** join/#nslu2-linux pleemans (n=peter@d51A5E76A.access.telenet.be)
15:43.07*** join/#nslu2-linux AdamBaker (n=aab@userch028.dsl.pipex.com)
16:08.39*** join/#nslu2-linux H4rvy (n=norm@80-193-172-141.cable.ubr05.pres.blueyonder.co.uk)
16:14.20*** join/#nslu2-linux caplink811_log (n=caplink8@dslb-088-073-004-163.pools.arcor-ip.net)
16:49.01*** join/#nslu2-linux ByronT (n=byron-po@nslu2-linux/ByronT)
17:13.55*** join/#nslu2-linux H4rvy (n=norm@80-193-172-141.cable.ubr05.pres.blueyonder.co.uk)
18:53.09*** join/#nslu2-linux caplink811-log (n=caplink8@dslb-088-073-026-062.pools.arcor-ip.net)
18:53.45*** join/#nslu2-linux eno (n=eno@nslu2-linux/eno)
19:02.50*** join/#nslu2-linux pleemans (n=peter@d51A5E76A.access.telenet.be)
19:18.12*** join/#nslu2-linux Bitmaster (n=Bitmaste@c-ed60e253.325-1-64736c12.cust.bredbandsbolaget.se)
19:44.15*** join/#nslu2-linux pleemans (n=peter@d51A5E76A.access.telenet.be)
19:52.12*** join/#nslu2-linux caplink811_log (n=caplink8@dslb-088-073-026-062.pools.arcor-ip.net)
22:21.43VoodooZa plus tard...
22:21.49VoodooZoops sorry
22:22.23VoodooZanybody built HEAD since friday? mine failed on npe..
22:25.51NAiLrwhitby has upgraded the microcode afaik
22:25.55NAiLhaven't tried building for ixp
22:26.43caplink811_logit's never to late....
22:27.41NAiLhehe
22:27.54caplink811_log;)
22:27.55NAiLGot a wide range of powerpc devices I'm working on atm.
22:36.32VoodooZI just know that make slugosbe-image failed.
22:37.22VoodooZactually....
22:37.51VoodooZI just noticed that the npe package needs to be downloaded manually...
22:38.13VoodooZOdd. I thought were weren't using intel's stuff anymore....
22:42.04rwhitbyVoodooZ: we can't get away from the microcode
22:42.11rwhitbywe don't use the access library any more.
22:47.45VoodooZI see.
22:47.45VoodooZI've downloaded the latest and building.. thanks.
22:47.45VoodooZI haven't touched the robot/slug stuff at all for a while. I have to start soon though...
22:47.48VoodooZlater
22:51.07NAiLrwhitby: btw, since we're not using apex, why aren't we using the upx compression that john reiser worked on?
22:51.47NAiLI'm discussing upx compression for a minimal kernel for powerpc right now, and that reduced the kernel from 1.4mb to 1mb (!)
22:53.58NAiL(Still in development, so it's not actually working yet)
22:54.28NAiLthe memory usage is largely unaffected, since it uses in-place decompression
22:55.48jacquessounds good to me - what's involved in supporting it? kernel mods?
22:56.38NAiLThe kernel needs a new decompressor, so I guess it's a patch.
22:56.46NAiLit also needs some tools available when building
22:56.52NAiLie, the compressor tools.
22:57.14jacquesdoes assembler code need to be written?
22:57.27NAiLafaik, it works for ARM already.
22:57.42NAiLso I don't think any new code needs to be written
22:58.32jacquesare the patches available?
22:58.45NAiLI compressed the userspace binaries in my em-mode rootfs. The binaries generally compressed to about 50% their original size.
22:59.22jacqueswow, it would work on initramfs too?
22:59.34jacquesI could definitely use this on ppc
22:59.39NAiLcan't see any reason for it not to work.
23:00.17jacquesI assume there's a longer decompression time
23:00.39jacquesbut even a couple of seconds would be worth it
23:00.40NAiLI think he mentioned that. I don't think the decompression time was that much longer.
23:00.44NAiLindeed
23:01.45NAiLI wonder if it'd be possible to get openembedded to automatically compress binaries when building.
23:02.18NAiLbusybox compressed to 1/3rd of the original size on my fs.
23:04.09NAiL(Given, it's a tiny busybox to begin with)
23:04.11jacquesthe board I have been working on has 4MB flash, now I'm trying to put linux on a similar board which only has 2MB flash
23:04.18NAiLa semi-default busybox for powerpc:
23:04.19NAiL<PROTECTED>
23:04.30NAiL<PROTECTED>
23:04.49jacquesmy current kernel + initramfs image is 3MB Including some stuff I can remove, like bash ;-)
23:04.50NAiLthat's got most of the utilities enabled.
23:04.58NAiLbash is ... huge! :)
23:05.03jacquesI know
23:05.09jacques600kB+
23:05.12NAiLand it requires ncurses, doesn't it?
23:05.20jacquesyep, the way I built it it does
23:07.35NAiLis that a 2.6-kernel btw?
23:08.14jacquesyep, currently 2.6.19 because it works and has proven stable
23:08.26NAiLhow big is the kernel?
23:08.40NAiLI'm really struggling to get a 2.6-kernel down to an acceptable size
23:08.58NAiLA bit of the problem is that I have to have support for a couple of filesystems.
23:09.08NAiLbut even without that, it's a bit large.
23:09.15jacquesI think it's right around 1MB but I'm not sure
23:09.54jacquesdo you have to have ext3 builtin? that's big
23:10.09NAiLyes :-\
23:10.12NAiLThe worst bit is xfs
23:10.18jacqueswith initramfs, I don't need *any* disk based filesystems
23:10.28jacquesone of the benefits over initrd
23:10.35NAiLah
23:10.47NAiLI'm making a new em-mode for the linkstations ;)
23:10.53jacquesneat
23:11.16jacquesI need to flash my HG with u-boot and put it back together
23:11.19NAiLso I need to fit a 2.6-kernel with xfs, ext[23] and basic utilities to support installing/fixing stuff
23:11.33jacquesxfs? ouch
23:11.41jacquesxfs *and* ext3 == huuuuge
23:11.53NAiLxfs is the killer though... Without xfs-support in the rootfs, I'm at 950kb. With xfs-support, I'm at 1.44mb.
23:12.08NAiL(This is with upx max compression, ext2 rootfs.gz)
23:12.16*** join/#nslu2-linux H4rvy (n=norm@80-193-172-141.cable.ubr05.pres.blueyonder.co.uk)
23:12.28jacquesxfs is a filesystem right? what does it need in the rootfs?
23:12.55NAiLUtilities to create, grow, shrink, fix, diagnose etc, etc, etc.
23:13.02jacquesah
23:17.03NAiLsome of the utilities compress a lot though :)
23:17.03NAiL<PROTECTED>
23:17.24NAiLon average, the xfs support utils:
23:17.25NAiL<PROTECTED>
23:17.30jacquesheh that's pretty good
23:20.44NAiLI've got a freshly generated rootfs
23:20.51NAiL4580    root/
23:23.00NAiL3380    root/
23:23.15NAiL1.2mb savings
23:23.59jacqueshow does it compare to gzip?
23:24.34NAiLI dunno. I haven't tried compressing the binaries with gzip.
23:24.44NAiLThat size is the whole rootfs before compression
23:25.45*** join/#nslu2-linux caplink811_log (n=caplink8@dslb-088-073-026-062.pools.arcor-ip.net)
23:26.06NAiLremoving an unused library, adding a missing one, then making the whole thing an ext2 fs and compressing it gives me:
23:26.10NAiL1508 -rw-r--r--   1 repvik repvik 1536391 2007-04-30 01:26 rootfs.powerpc.ext2.gz
23:26.31NAiLThat's a fully featured rootfs with ext2,ext3,xfs and dmraid support.
23:26.40jacqueswow, there's something I don;t understand then
23:27.00jacquesupx compresses the executables and then gzip does further compression?
23:27.36NAiLI don't think gzip can compress the upx-compressed binaries much
23:27.47jacquesyeah I was gonna say
23:27.48NAiLbut gzip compresses the whole rootfs including the libraries that upx can't compress
23:27.52jacquesah OK
23:28.15jacquesso the upx compressed binaries, they self-decompress when called?
23:28.23NAiLyes
23:28.38jacquesok, now I begin to understand
23:28.41NAiLObviously, this is a speed hit
23:28.47jacquesnod
23:29.08NAiLyou save a bit of time reading them from flash, but you have to wait a bit for the decompression again
23:29.29jacquesthe main issue is usually *fitting* everything into flash
23:29.41jacquesmost ppl would gladly give up a bit of speed to fit more in
23:29.43jacquesI know I would
23:29.57NAiLIn this case, speed really isn't an issue.
23:30.11NAiLIt's not designed to be used except in case of emergencies.
23:30.23jacquesyep
23:30.52jacquespretty much anything faster than mailing it off to get unbricked would be acceptable ;-)
23:31.16jacqueswell technically not bricked
23:31.29NAiLhehe
23:31.30jacquesyou'd have to remove the drive, attach to a PC and fix it
23:31.34NAiLyeah
23:31.43NAiLwhich is a pain, especially if you're doing development work
23:32.13jacquesyeah
23:32.27NAiLbut now, with the all-singing, all-dancing u-boot and this fancy all-singing, all-dancing em-mode rootfs, this should almost be a joy ;)
23:32.38jacquesheh :-)
23:33.13NAiLbtw, I'm compressing using a very old version of the upx compressor. From what I see, it's far more efficient in its later incarnations.
23:33.56jacqueswell heck, get 2.93 then!
23:34.34NAiLhehe, yeah
23:35.05NAiLah
23:35.07NAiLThe main news since UPX 2 are:
23:35.07NAiL* new format: added support for linux elf/arm
23:35.07NAiL* better compression using the LZMA algorithm (option '--lzma')
23:35.45jacquesneat
23:35.53jacqueshmm, what about elf/armeb  ;-)
23:38.23rwhitbyNAiL: the upx thingo couldn't get our kernel down by a full block size, so no point.
23:40.30jacquesI forget the block size on the nslu2 128kB, 256kB?
23:41.09NAiL2976    root/
23:41.24NAiLdown from 4850kb, that's not bad ;)
23:41.46jacqueslol not half bad
23:42.41NAiL1352 -rw-r--r--   1 repvik repvik 1379629 2007-04-30 01:42 rootfs.powerpc.ext2.gz
23:42.50NAiLThat's the complete compressed rootfs
23:43.27NAiLconsidering the amount of support in that rootfs, that's quite impressive.
23:44.02NAiLrwhitby: how much could it compress the default kernel? I'm seeing huge gains with the new (--lzma) compression.
23:45.14NAiLcompressing it from 1.4mb down to 1mb is in my eyes nothing short of boggling.
23:45.34jacquesyah
23:46.28rwhitbyNAiL: it's worth looking at again.
23:46.39rwhitbybut only if it gets it down to 768KiB-16
23:46.52NAiLwhy that small?
23:46.58rwhitbyoh, make the 1024KiB-128KiB-16
23:47.04rwhitby892 I guess
23:47.25NAiLredboot reads 1mb no matter what, doesn't it?

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.