| 00:16.05 | *** join/#elinux eggers (n=eggers@cpe-24-28-73-7.austin.res.rr.com) |
| 00:30.07 | *** join/#elinux joe_bleau (n=joe_blea@adsl-65-67-66-115.dsl.fyvlar.swbell.net) |
| 00:31.04 | TimRiker | andersee: it was just a backup. I'll re-mirror it once I re-install it. |
| 01:09.10 | *** join/#elinux weo1 (n=weo@p5499E696.dip.t-dialin.net) |
| 01:10.44 | *** join/#elinux noclouds (n=mhfan@60.166.39.7) |
| 01:49.39 | *** join/#elinux ljp (n=lpotter@203.94.178.46) |
| 02:15.26 | *** join/#elinux Crofton (n=balister@66-207-66-26.black.dmt.ntelos.net) |
| 02:37.36 | *** join/#elinux thraxisp (n=thraxisp@ottawa-hs-64-26-167-172.d-ip.magma.ca) |
| 02:57.48 | *** join/#elinux we0 (n=weo@p5499FE97.dip.t-dialin.net) |
| 04:07.45 | *** join/#elinux MonMotha (n=monmotha@martinbv-3.student.rose-hulman.edu) |
| 04:59.22 | *** join/#elinux joe_bleau (n=joe_blea@adsl-69-152-208-148.dsl.fyvlar.swbell.net) |
| 05:55.04 | *** part/#elinux joe_bleau (n=joe_blea@adsl-69-152-208-148.dsl.fyvlar.swbell.net) |
| 08:22.46 | *** join/#elinux vrm (n=vrm@183.55.101-84.rev.gaoland.net) |
| 09:04.26 | *** join/#elinux TorbaX2 (n=TorbaX@host-84-220-141-244.cust-adsl.tiscali.it) |
| 09:26.50 | *** join/#elinux ade|desk (n=adavey@194.200.143.249) |
| 09:58.42 | *** join/#elinux |LinuxCa| (n=kvirc@212.106.244.50.adsl.jazztel.es) |
| 09:59.17 | *** join/#elinux CosmicPenguin (n=nobody@aus-ext-proxy01.amd.com) |
| 09:59.17 | *** mode/#elinux [+v CosmicPenguin] by ChanServ |
| 10:42.22 | *** join/#elinux bkankur (n=ankur@59.95.195.8) |
| 10:43.00 | bkankur | which webserver i should use for embedded device which will support MONO ? Also it should be taking less footprint. |
| 11:41.57 | *** join/#elinux T`2_ (n=total@tor/session/x-cc12848bac90b0e1) |
| 12:15.27 | *** join/#elinux Crofton (n=balister@66-207-66-26.black.dmt.ntelos.net) |
| 14:12.24 | *** join/#elinux prpplague (n=billybob@72.22.146.125) |
| 14:20.54 | *** join/#elinux GPSFan (n=Ken@65.121.49.208) |
| 14:49.45 | jamie | hi folks... |
| 14:49.53 | jamie | another ext3/2.4 uclinux/ARM question... |
| 14:50.21 | jamie | I'm doing "mke2fs" on a 40G hard disk, and it takes more than an hour... It seems to run quickly until memory fills, then goes very slowly, but doesn't crash. |
| 14:50.45 | jamie | Is that something anyone else has seen? Is there another forum I should be asking about this sort of thing? Cheers, |
| 15:05.04 | *** join/#elinux cbrake (n=cbrake2@cpe-24-166-121-47.neo.res.rr.com) |
| 15:14.48 | CosmicPenguin | jamie: no - thats very broken |
| 15:18.26 | *** join/#elinux thraxisp (n=thraxisp@ottgate.precidia.com) |
| 15:27.03 | jamie | CosmicPenguin: Thanks. Someone mentioned you've worked on a Sigma Designs device, is that right? |
| 15:27.10 | jamie | It's what I have here, with these ext3 problems. |
| 15:27.47 | jamie | I also find that after writing to a file (even a 1 char edit of a text file), then sync, then power cycle, often the edit is lost, or filesystem structures are corrupted in odd ways. |
| 15:28.20 | jamie | I was wondering if it's caused by uclinux oddness, if anyone ever used uclinux/ext3 with large disks before, or if it's clearly a bug with this device. |
| 15:30.46 | CosmicPenguin | Its been a long time since I've worked with a Sigma Designs device |
| 15:31.01 | CosmicPenguin | heck, when I worked with them, they were just pure video decode chips - I know the newer ones have working cores on them |
| 15:32.01 | jamie | Yes, the current ones have an ARM core, and the video decoder glued together. The ARM runs uclinux-2.4.26 (with Sigma mods, of course), and communicates with the video decoder much like the old video decoder over PCI. |
| 15:32.25 | jamie | I guess my problems are with the ARM and IDE interfaces (the IDE driver is a Sigma special...) |
| 15:32.51 | jamie | (Well, my _other_ problems are with the video decoder :/ but that's another story. When it works it's quite good). |
| 15:33.31 | jamie | Do you happen to know what governs "maximum number of dirty pages before writes are synchronous"? |
| 15:34.09 | jamie | I see significant problems (including kernel panics) when writing "too fast", which leads me to suspect memory is filling with dirty pages until the system is wedged or nearly so. |
| 15:35.09 | jamie | Standard Linux will pause the writing process - making writes synchronous - when the amount of dirty data exceeds a threshold. That's why I'm wondering if (a) uclinux breaks that (affecting the memory subsystem as much as it does), and (b) if it can be fixed by tweaking something in /proc. |
| 15:35.52 | CosmicPenguin | I'm not exactly sure how uclinux works in that regard |
| 15:35.57 | jamie | Me neither. |
| 15:36.53 | jamie | I suspect not a lot of people are using uclinux with 40G hard disks, and those that are are using 2.6, which has much more integrated uclinux and a different memory subsystem. |
| 15:37.22 | psypete | tried using vfat? or do you need the ext2-ness |
| 15:37.32 | jamie | I need the journalling. |
| 15:37.38 | psypete | ah |
| 15:37.39 | jamie | I might try reiserfs. |
| 15:37.50 | jamie | It's a big disk, so fsck on boot is too slow. |
| 15:38.01 | jamie | And power can be withdrawn without notice. |
| 15:38.15 | psypete | journal is no guarantee of safety |
| 15:38.23 | jamie | Indeed, as I'm seeing :/ |
| 15:38.36 | jamie | It should be though :) |
| 15:38.44 | psypete | for my car pc, since i only need to play back files, i mount all disks read-only |
| 15:39.06 | jamie | What do you do when writing new files to the disk? |
| 15:39.12 | jamie | And do you log anything on it? |
| 15:39.15 | psypete | if i need to write something (mostly only when rsyncing media files) i'll switch to read-write and 'sync' every 5 seconds |
| 15:39.37 | jamie | I am doing 'sync' after writing, before power off in tests. I still get the corruption. 'sync' is not working. |
| 15:39.59 | psypete | i don't log anything yet, but when i was doing wardriving i would write to fat16 thumb drives.... i tried writing to the drive about 30 times and unplugged the drive each time during writes, it never corrupted |
| 15:40.03 | jamie | I don't have problems (that I've noticed) when not writing anything, so the read-only mount is moot. |
| 15:40.40 | psypete | so to me the flash disk thingies seem safer than 'normal' media |
| 15:41.08 | jamie | That might be true. I have made a point of disabling the hard-disk's write cache - I know a few disks don't honour that request, but most do. |
| 15:41.49 | psypete | only other option you have is a small battery or big capacitor and provide power for a second or two to sync and umount |
| 15:42.14 | jamie | How would that help? I _am_ doing 'sync', and still seeing corrupt data after that. |
| 15:42.30 | psypete | umount after sync |
| 15:42.38 | jamie | Is that necessary? |
| 15:42.51 | psypete | well that way you can make sure this corruption isn't something else |
| 15:43.06 | psypete | it seems strange to me a sync wouldn't grab everything |
| 15:43.13 | jamie | I thought sync was roughly equivalent to umount followed by mount. |
| 15:43.19 | jamie | It seems strange to me, too. |
| 15:43.19 | psypete | no |
| 15:43.27 | psypete | sync just flushes buffers, iirc |
| 15:43.45 | jamie | And umount + mount does something else? |
| 15:43.48 | psypete | a umount/mount changes things like the i-was-umounted-safely flag |
| 15:43.55 | jamie | Yes, I'm aware of that. |
| 15:44.16 | psypete | i'm not intimately familiar but i do know theres one or two differences. sync is even supposedly unsafe in certain conditions |
| 15:44.24 | psypete | or at least, not guaranteed |
| 15:44.34 | jamie | I did notice, with suspicion, that e2fsck said "ext3 recovery flag is clear but journal has data", which shouldn't ever happen. |
| 15:44.56 | jamie | I believe that if you're not writing anything, then 'sync' is supposed to be safe. |
| 15:45.48 | jamie | I'll try umounting, to see if the writes become more reliable. It's possible the "recovery flag clear" would be causing journal replay to not happen on the next mount - and that would be enough to corrupt the fs. |
| 15:46.30 | jamie | But not enough to explain why mke2fs takes more than an hour, and why occasionally the contents of a file are spontaneously replaced by completely different contents. |
| 15:46.50 | jamie | (I suppose a corrupted free bitmap would explain the latter, but enough speculation). |
| 15:50.11 | psypete | fyi: new reiser is supposed to be better in some ways than ext3 |
| 15:50.16 | psypete | and ext3 is kind of a hack anyway |
| 15:50.58 | jamie | sure, I read a lot of conflicting opinions about ext3 vs. reiser. Some say one is more reliable than the other, and vice versa. No useful information. |
| 15:51.19 | jamie | ext3 is kind of a hack, but it is a _simple_ kind of a hack that ought to be trivially correct as it were. |
| 15:52.36 | *** join/#elinux noclouds (n=mhfan@60.166.46.119) |
| 16:01.03 | *** join/#elinux mallum (n=mallum@host86-133-116-67.range86-133.btcentralplus.com) |
| 16:01.04 | *** mode/#elinux [+v mallum] by ChanServ |
| 16:14.04 | *** join/#elinux weo (n=weo@p5499FE97.dip.t-dialin.net) |
| 17:00.46 | psypete | slap an LD_PRELOAD wrapper around em |
| 17:01.06 | psypete | i made one for tar that uses pipes but LD_PRELOAD would work too |
| 17:05.27 | *** join/#elinux mbuf (n=mbuf@59.92.34.6) |
| 17:07.20 | *** join/#elinux TimRiker (n=timr@pdpc/supporter/bronze/TimRiker) |
| 17:07.20 | *** mode/#elinux [+o TimRiker] by ChanServ |
| 17:10.17 | *** join/#elinux Genesis (n=genesis@dev.aimao.org) |
| 17:11.40 | Genesis | bonsoir |
| 17:21.06 | psypete | que tal |
| 17:30.05 | LinuxCart | good spanish psypete! ;) |
| 17:33.09 | *** join/#elinux Crofton_|laptop (n=balister@hc6521c42.dhcp.vt.edu) |
| 17:34.49 | psypete | ty |
| 17:36.13 | LinuxCart | ty¿? |
| 17:37.43 | *** join/#elinux zinosat (n=Davide@host37-89.pool8256.interbusiness.it) |
| 17:38.04 | psypete | ty == thank you |
| 17:38.16 | LinuxCart | ahh, ok, sorry |
| 17:39.00 | LinuxCart | maybe I can't teach you much about micros, but if you need some spanish words... ;) |
| 17:40.21 | psypete | thanks :) |
| 17:41.38 | LinuxCart | should we go on with our last conversation¿? |
| 17:42.49 | LinuxCart | it was about booting and flash memories.. |
| 17:43.14 | LinuxCart | but if you are busy it doesn't matter |
| 17:49.31 | psypete | if you have questions, feel free to ask |
| 17:54.18 | LinuxCart | yes it was the problems that could arise to boot |
| 17:54.32 | LinuxCart | I mean, you need a bootrom |
| 17:54.55 | LinuxCart | the bootrom should be in some storage memory |
| 17:55.11 | LinuxCart | but it is not advisable to be nand flash |
| 17:55.34 | LinuxCart | where is the problem, in the SW or the HW¿? |
| 17:57.37 | *** join/#elinux hufnus (n=slonsiki@DSL135-071.labridge.com) |
| 18:01.51 | psypete | my guess, the hardware |
| 18:01.57 | psypete | but i am by no means an ARM expert |
| 18:06.39 | LinuxCart | the micro I've selected includes 128KB internal ROM, I think it is possible to include the bootrom there... |
| 18:07.12 | LinuxCart | ...but if not is it problematic to download it to external flash through jtag¿? |
| 18:07.21 | psypete | someone mentioned a bootloader <32kb |
| 18:07.41 | psypete | might have been CosmicPenguin |
| 18:10.02 | LinuxCart | i'll ask him when he's available |
| 18:10.06 | LinuxCart | ty :) |
| 18:10.28 | LinuxCart | could I ask what are you developing now¿? |
| 18:14.24 | psypete | i am working on a computer for my car, to control mp3s, videos, navigation, etc |
| 18:15.21 | psypete | right now it is like buildroot except it doesn't use uClibc and it doesn't bootstrap |
| 18:15.24 | LinuxCart | oh true.. now I remember |
| 18:15.54 | LinuxCart | what distro do you use¿? (if any) |
| 18:16.05 | psypete | to develop it, i use Slackware 10.1 |
| 18:16.17 | psypete | but the end result is to make its own distro |
| 18:17.24 | CosmicPenguin | Twasn't me |
| 18:21.38 | LinuxCart | CosmicPenguin: yes, it is about the 32kb bootloader.. |
| 18:21.46 | LinuxCart | which one is it¿? |
| 18:22.16 | LinuxCart | psypete: so you started from scratch in the target... |
| 18:22.56 | psypete | well actually i don't, i start with busybox and add applications, put them in a root directory, then copy libraries needed and some /etc/ files |
| 18:23.11 | psypete | a few more little tricks and i basically construct a little chroot-type system |
| 18:24.42 | LinuxCart | how much resources do you have, cpu, ram, flash...?¿ |
| 18:27.40 | CosmicPenguin | I don't know anything about a 32kb bootloader |
| 18:27.41 | CosmicPenguin | sorry |
| 18:29.17 | LinuxCart | it's ok CosmicPenguin |
| 18:38.56 | LinuxCart | i'm leaving cu |
| 19:08.49 | T0mW | jamie, I wonder if the corruption problems you are seeing are due to a hardware problem of the Sigma ARM processor? |
| 19:09.12 | jamie | Maybe. How do I tell? |
| 19:10.20 | T0mW | jamie: the LPC2000 ARM I'm working with has a MAM to accelerate the memory access. If it is not setup correctly, it can corrupt a byte if you do a write followed by in immediate read |
| 19:11.11 | jamie | Um, is that memory access for the IDE controller, or memory access in general? |
| 19:11.19 | T0mW | eratta sheets are a start, is the uclinux a supported version by Sigma? |
| 19:11.52 | T0mW | MAM == Memory Access Manager, it is on the Philips parts, but not sure about others |
| 19:12.22 | T0mW | the LPC2000 is too small to run uClinux, at this time. |
| 19:12.36 | T0mW | ARM7TDMI |
| 19:13.01 | T0mW | jamie: but reading the scrollback I kept thinking that this sounds like a hardware issue. |
| 19:13.53 | jamie | I don't know if it's supported by Sigma. I have no direct relationship with Sigma, only a lot of code that is from them, supplied by a third-party who are the board manufacturer. |
| 19:14.09 | T0mW | jamie: one thing, uclinux migrated very quickly from 2.4 kernels to 2.6, there were a lot of memory fragmentation issues with 2.4 |
| 19:14.14 | jamie | Of course I have no errata sheets. In fact, I have no access to any documentation from Sigma. |
| 19:14.36 | jamie | I have seen at least on panic that looked like it was due to memory fragmentation. |
| 19:14.45 | T0mW | yeah |
| 19:14.59 | T0mW | jamie: you considered going to 2.6? |
| 19:15.05 | jamie | I'd love to use 2.6, but I suspect it would be a huge amount of work. |
| 19:15.23 | jamie | Not least because I'd need to port all the Sigma-specific drivers/patches. |
| 19:15.32 | T0mW | that could be |
| 19:15.59 | T0mW | just something to keep in mind though, IMO 2.4 uClinux was not that stable |
| 19:16.09 | jamie | And if I did, and it sorta worked, but it crashed occasionally, I couldn't ask anyone upstream to debug it. |
| 19:16.29 | jamie | That's why I keep asking. (Alas, still waiting for "approval" to join the uclinux-dev list). |
| 19:16.30 | T0mW | IIRC, we only were on 2.4 for about a year before we went to 2.6 in late 2000? |
| 19:16.41 | jamie | Asking whether 2.4 uclinux has known instabilities that look like what I'm seeing. |
| 19:16.49 | jamie | Not many people here seem to recognise them. |
| 19:16.49 | T0mW | jamie: uClinux is so disorganized, turf wars |
| 19:17.00 | jamie | yeah, I can tell from the web site that it's fragmented. |
| 19:17.27 | jamie | 2.6 ought to be more stable because it's integrated (or largely so) into the standard 2.6 tree, isn't it? |
| 19:17.50 | jamie | Alas, that whole problem with not having any kind of support from my suppliers if I start trying 2.6 on it. |
| 19:18.06 | T0mW | it got accepted yes, but what is important to note is that more time has been spent on the 2.6 port |
| 19:18.54 | jamie | I expect I'd end up with a lovely, stable 2.6 generic uclinux, and dodgy drivers. |
| 19:19.20 | T0mW | we rapidly moved into 2.6 as serious changes needed to be done to attack the memory fragmentation issue and it was felt that such an effort needed to be spent on a kernel that was going to have ongoing development |
| 19:19.52 | jamie | Well, I like 2.6 tee, having contributed to the futex, clone and epoll features ;-) |
| 19:20.16 | jamie | Still. When I get to specify a kernel version one day... then I'll insist on it. |
| 19:21.18 | T0mW | jamie: I'm out of the loop on uclinux for a few years now, I was looking for realtime stuff and the horsepower needed to crank linux over fast enough to be responsive needed faster chips. But, when you goto faster chips, you usually get an MMU and a host of issues disappear. |
| 19:21.41 | jamie | Yes, and MMU would be nice. Considering it's a 266MHz ARM or something I'm surprised there isn't one. |
| 19:21.56 | jamie | Probably wouldv'e cost extra licensing from ARM or something daft like that. |
| 19:22.03 | T0mW | jamie: you see any recent messages on uClinux list from "gerg" or "David M" ? |
| 19:22.36 | jamie | Am still waiting to be allowed to join the uclinux list... first time I've seen an open source dev list that moderates who can subscribe... |
| 19:23.13 | psypete | you mean, besides the kernel vulnerability disclosure list |
| 19:23.19 | jamie | maybe. not been there. |
| 19:23.21 | T0mW | maybe from the archives? If either of those two are still active, then you should be able to contact them directly. |
| 19:23.33 | jamie | they are familiar with 2.4's problems? |
| 19:23.38 | T0mW | yes |
| 19:23.54 | T0mW | gerg is very proficient with uClinux |
| 19:24.04 | T0mW | so is David M |
| 19:24.37 | T0mW | I'm trying to recall gerg's company in Australia, they made uClinux routers |
| 19:24.55 | T0mW | company has been bought out |
| 19:25.04 | T0mW | SnapGear! |
| 19:25.09 | jamie | Ah, to work with someone who's not stuck in 2.4 land :) |
| 19:26.02 | T0mW | <PROTECTED> |
| 19:26.15 | T0mW | <PROTECTED> |
| 19:26.34 | T0mW | those two guys have been with the project since the early daze |
| 19:26.44 | T0mW | www.uclinux.org |
| 19:27.47 | T0mW | D. Jeff Dionne is problematic (ego) |
| 19:28.10 | T0mW | Mike Durrant follows Jeff |
| 19:28.23 | T0mW | two somewhat opposing factions |
| 19:28.25 | T0mW | heh |
| 19:29.01 | jamie | I couldn't care less about who's following who etc. Just want to know if my problems are kernel or driver or hardware problems. And whether we might need to redesign the board. |
| 19:29.21 | T0mW | jamie: understood |
| 19:30.00 | jamie | There have been other hardware problems, and reliability problems, so I'm not ruling out h/w. And that's urgent: if it is h/w, I need to get the right diagnostic info to the h/w design people before they make another batch in December. |
| 19:30.03 | T0mW | jamie: anyway, get ahold of either of those two guys and they should get you on the list |
| 19:30.06 | jamie | cheers |
| 19:32.39 | T0mW | jamie: final thought, Arcturus Networks is the old founders of uClinux. I've often got the sense that the uClinux.org side of things were something of a competition for their commercial enterprise. A "poor relation". |
| 19:32.53 | T0mW | jamie: cheers |
| 19:35.16 | *** join/#elinux darkschneider (n=gab@213-140-6-96.ip.fastwebnet.it) |
| 19:46.48 | jamie | bleargh, I just remembered another reason why jumping to 2.6 would be messy. Binary drivers. Poo. |
| 19:46.49 | *** join/#elinux GPSFan_ (n=Ken@65.121.49.208) |
| 19:48.43 | T0mW | jamie: sux |
| 19:48.48 | *** join/#elinux eggers (n=eggers@pixpat.austin.ibm.com) |
| 19:50.47 | T0mW | jamie: did you try to strip down kernel of all but necessary modules / configs needed to run the IDE, then see if the corruption still exists? |
| 21:22.45 | *** join/#elinux darkschneider (n=gab@213-140-6-96.ip.fastwebnet.it) |
| 21:40.45 | jamie | T0mW: That's a good idea. Not an easy one to try, but a good one. |
| 21:41.37 | T0mW | jamie: may help to narrow down the considerations. Prove what is "right" |
| 21:44.01 | jamie | Hmm. I have a big kernel patch from Sigma here which includes the IDE drivers. |
| 21:44.18 | jamie | Do you reckon they're implicitly GPL'd and I should submit them to the main kernel tree? :-) |
| 21:46.42 | T0mW | jamie: why did they need to patch IDE? That is a pretty generic interface. |
| 21:47.09 | T0mW | unless they are memory mapped IDE registers |
| 21:47.24 | T0mW | of oddball addresssing |
| 21:48.13 | CosmicPenguin | If I remember correctly Sigma wasn't ever very excited about the GPL |
| 21:48.29 | CosmicPenguin | T0mW: most controllers will have some sort of glue with them for DMA and such |
| 21:48.35 | CosmicPenguin | refrence drivers/ide/pci/* |
| 21:51.19 | T0mW | CosmicPenguin: just not used to dealing with DMA in a controller |
| 21:51.20 | T0mW | heh |
| 21:51.39 | CosmicPenguin | heh - I'll bet |
| 21:51.53 | T0mW | that is like the Lexas of peripherals |
| 21:52.29 | CosmicPenguin | we're putting bus masters in pretty much everything these days |
| 21:52.39 | CosmicPenguin | I still love the Alchemy DDMA engine |
| 21:54.26 | *** join/#elinux andersee (n=andersee@codepoet.org) |
| 21:54.26 | *** mode/#elinux [+o andersee] by ChanServ |
| 22:06.43 | Genesis | bonne nuit |
| 22:35.35 | jamie | well, without IDE DMA, the thing can't read the disk fast enough to play video files so it's quite important |
| 22:37.22 | *** join/#elinux file[laptop] (n=jcolp@mctnnbsa24w-142167049176.nb.aliant.net) |
| 22:46.55 | MonMotha | TimRiker: you there? |
| 23:02.29 | *** join/#elinux GPSFan (n=Ken@65.121.49.208) |
| 23:21.16 | *** join/#elinux Sgt-Donan (n=Loutre@mto30-1-82-242-53-8.fbx.proxad.net) |
| 23:24.42 | TimRiker | sorta |
| 23:24.54 | MonMotha | TimRiker: I' |
| 23:24.55 | MonMotha | bleh |
| 23:25.07 | MonMotha | I've been told you're fairly knowledgable about the GPL |
| 23:25.08 | MonMotha | ? |
| 23:26.17 | MonMotha | mind if I /msg you? |
| 23:26.43 | TimRiker | sure. |
| 23:26.52 | TimRiker | er, no. I don't mind. =) |
| 23:28.45 | CosmicPenguin | grr... |
| 23:29.05 | jamie | CosmicPenguin: Hmm, on further study, this kernel pkg from Sigma says: |
| 23:29.07 | jamie | The ARM utilities package (to the exception of the 'peripherals' folder) is |
| 23:29.07 | jamie | released under the terms of the GNU General Public License (the GPL); see the |
| 23:29.08 | jamie | file COPYING for more information about the GPL. |
| 23:29.36 | CosmicPenguin | Oh, nice |
| 23:29.39 | CosmicPenguin | thats different then I remember |
| 23:29.41 | CosmicPenguin | good for them |
| 23:29.48 | jamie | Looks like I could legitimately send the patch kernel's way, then. |
| 23:30.36 | jamie | Well, Sigma were threatened by the Xine folks at one point, for using bits of Xine's code. I gather they replaced that, and maybe learned a few things. |
| 23:33.22 | *** join/#elinux aldo (n=aldo@dsl-200-78-99-58.prod-infinitum.com.mx) |
| 23:33.30 | aldo | hi, good day |
| 23:33.39 | aldo | does anyone have compiled the MySQLdb module for python with uclibc??? |
| 23:55.54 | *** join/#elinux joe_bleau (n=joe_blea@adsl-69-152-223-244.dsl.fyvlar.swbell.net) |