**** BEGIN LOGGING AT Sat Sep 19 02:59:59 2009 Sep 19 03:28:53 rektide: hi Sep 19 03:52:22 tinker-f595: having fun? Sep 19 03:52:30 hi Sep 19 03:53:16 what's a good httpd to use with php, now where you made me look over here? :) Sep 19 03:53:23 morfic: just wondering why Dhavel did not post the OpenRD Client kernel patches up stream Sep 19 03:53:53 whicc distro? Sep 19 03:54:00 ubuntu Sep 19 03:54:17 hmmm apt-get install apache2 Sep 19 03:54:38 or lighttpd with php on fastcgi Sep 19 03:54:41 unfortunately ubuntu renamed stuff Sep 19 03:54:55 something more lightweight? i know sheeva is beefy compared to slug, but apache2 was unexpected Sep 19 03:55:10 morfic: lighttpd + fastcgi Sep 19 03:55:22 I looked at lighttpd and it was too restricted for what I wanted Sep 19 03:55:36 tinker-f595: what did it miss? Sep 19 03:55:39 apache2 uses hardly any resources Sep 19 03:55:59 once you spend a hell lot of time configuring it Sep 19 03:56:10 i just want to run a little php script a friend made that thumbnails all images in a dir, as simple gallery Sep 19 03:56:31 morfic: lighttpd will work Sep 19 03:56:42 arachnist: how's your openrd btw? Sep 19 03:57:33 morfic: well, currently it's doing "emerge -uDNv world" from a stage Sep 19 03:57:45 so you need all the weight of php on top Sep 19 03:57:57 having wifi problems this end Sep 19 03:58:14 (would have probably finished by now, but i forgot about setting up /etc/locale.gen...) Sep 19 03:58:34 emerge world Sep 19 03:58:52 i prefer -uDNv world Sep 19 03:59:05 no i was just repeating and wondering Sep 19 03:59:28 with those switches portage at least pretends to calculate dependencies correctly Sep 19 04:00:18 arachnist: have you tried eSata ? Sep 19 04:00:31 tinker-f595: not yet, but i intend to Sep 19 04:00:50 uBoot seems to have problem with eSata Sep 19 04:01:12 I am going to write an email about what I have been seeing Sep 19 04:01:16 does uboot support initrd? Sep 19 04:01:31 yes Sep 19 04:01:41 jsut another thing you call Sep 19 04:01:43 i don't mind having to load initrd first Sep 19 04:02:50 but I am finding uBoot "ide start" is flaky for eSata and ext2ls messes up as well Sep 19 04:02:59 i nicked the Gentoo 'plug initrd to bring up my DB-78100 board: http://dev.gentoo.org/~armin76/arm/sheevaplug/install/sheevaplug.initramfs Sep 19 04:03:22 Wish I could find the correct spacers so I could plug in my drive directly to the header on the board Sep 19 04:03:53 I found some initrd instructions on the OpenRD cd Sep 19 04:04:09 well Sep 19 04:04:13 but more instructions the merrier!!! Sep 19 04:04:37 http://dev.gentoo.org/~armin76/arm/sheevaplug/install.xml <--- there are the gentoo sheevaplug instructions. substitute your kernel and the correct arcNumber for it Sep 19 04:04:44 any idea why Dhavel did not post the OpenRD Client patches on the 17th? Sep 19 04:04:53 robbat2|na: cool Sep 19 04:05:18 robbat2|na: thanks Sep 19 04:05:23 btw 3.5" drive won't fit in the aluminum case openrd-client is shipped with, right? Sep 19 04:06:04 it is designed for 2.5 inch drives Sep 19 04:06:49 so i'll have to buy an extra drive, probably Sep 19 04:06:56 the header is for 2.5 inch SATA drives. One of the docs has the spacer specs. But being metric htey are virtually impossible to find in the US Sep 19 04:07:20 (or just stick with connecting my 1TB 3.5" over esata) Sep 19 04:07:49 tinker-f595: why did you ctcp me? Sep 19 04:07:55 well let me know how that works with eSATA Sep 19 04:08:10 clicked on the wrong thing !!! Sep 19 04:11:52 page 9 of OPENRD_16-00050-02.pdf had the dimension of the spacers needed to mount a 2.5 inch SATA drive Sep 19 04:12:44 it is in ../openrd_hardware/schematics_bom/.. on the CD Sep 19 04:21:40 morfic: having interesting adventures? Sep 19 04:23:07 not really, you? Sep 19 04:24:40 yes banging my head trying to get eSata to work with the OpenRD Client. Given up for the moment. Will write dhavel an email over the weekend describing what I am seeing. Sep 19 04:25:39 Also will complain about the unavailability of spacers for mounting drive inside Sep 19 04:25:57 they should of supplied the spacers with the board Sep 19 04:26:45 morfic so what have you done with your device? Sep 19 04:30:36 tinker-f595: stop ctcping me Sep 19 04:30:39 tinker-f595: just put nfs back up, setup znc, used thttpd but moving it to lighttpd now Sep 19 04:32:29 morfic so nfs runs well? Sep 19 04:34:08 not piinging from here Sep 19 04:34:37 tinker-f595: well, not much better than slug, but better Sep 19 04:34:56 morfic your using arch? Sep 19 04:34:57 062200 [fnode] tinker-f595 [n=tinker-f@m208-148.dsl.rawbw.com] requested CTCP VERSION from arachnist: Sep 19 04:35:00 5MB/s write vs 4MB/s on slug, but 10MB/s read vs 4MB/s read on slug Sep 19 04:35:01 062203 [fnode] tinker-f595 [n=tinker-f@m208-148.dsl.rawbw.com] requested CTCP TIME from arachnist: Sep 19 04:35:04 :( Sep 19 04:35:07 tinker-f595: no, still ubuntu Sep 19 04:35:38 arachnist must be some weird setting in this client Sep 19 04:35:46 or bug Sep 19 04:36:17 morfic not sure how good the nfs stack is in ubuntu...if they have done anything weird to it. Sep 19 04:36:43 Did a lot of NFS testing back in the early 1990s Sep 19 04:37:01 is sshfs an option? or is the overhead too much? Sep 19 04:37:49 also i am on ext3 on sheeva, looks like xfs might gain me some, buddy who had spare space and uses lvm did a ext3 vs xfs real quick when he said xfs might be better if i have slowishness Sep 19 04:37:53 I have not tried it with the sheevaplug but I know people that use it with far lesser hardware and it works fine Sep 19 04:37:57 was a little fasterm nothing crazy though Sep 19 04:38:24 like i said slug did ok, but now i get 125% write and 250% read speeds Sep 19 04:38:52 googled up a ton of nfs//etx3 tuning stuff until they all repeated what other links already said Sep 19 04:38:52 oh ! going faster than light ;) Sep 19 04:39:26 and like i told you, my sleeping usb disks are no issue on a booted linux system :) just u-boot hating them Sep 19 04:39:51 04:39:26 up 6 days, 3:40, 2 users, load average: 0.17, 0.11, 0.09 Sep 19 04:40:04 luckily booting is something i am not too worried about right now Sep 19 04:40:05 usb disks can have issues Sep 19 04:40:35 i might try a active usb hub for u-boot alone, not worried about linux itself, i want them to sleep and wake up on access Sep 19 04:40:38 my uptime got ruined the other day. Electric went out Sep 19 04:40:55 i am dumb too without a UPS, so don't feel bad Sep 19 04:40:59 but i need one soon Sep 19 04:41:26 I have not seen a USB with acceptable battery tech Sep 19 04:41:58 If I could find one that used LiFePO4 tech I would be happy Sep 19 04:43:06 lead acid gel is cheap though Sep 19 04:43:57 a USB? not a UPS? Sep 19 04:46:03 morfic: I meant UPS Sep 19 04:46:17 lead acid is nasty even as gel Sep 19 04:46:27 * robbat2|na wants LiS tech to mature Sep 19 04:46:47 http://www.batteryspace.com/lifepo426650battery128v102ah13056wh21aratewithpcbreplacenimhpackwith4timeslongerlife.aspx Sep 19 04:46:51 kinda costly :) Sep 19 04:47:32 LiFePO4 is relatively inexpensive, no electrolyte, and does not burn hot Sep 19 04:49:18 There's all sorts of nanophosphate and nanotube stuff that's "better" than LiFePO4, but it's expensive Sep 19 04:49:34 ShadowJK: that is expensive Sep 19 04:50:09 I have some 3.1 AH that are about $20 each Sep 19 04:50:19 at 12V? Sep 19 04:50:40 oh Sep 19 04:50:44 ok Sep 19 04:50:54 so that works out to about same price Sep 19 04:51:04 plus it has the required pcb Sep 19 04:51:18 hopefully they have the correct logic in the pcb Sep 19 04:51:25 hah Sep 19 04:51:48 the PCBs tend to stop working if you put them into regular use Sep 19 04:52:49 for some real runtime on your UPS you want to build it out of these anyway: http://www.batteryspace.com/lifepo4prismaticmodule32v200ah3crate640whceunapproved.aspx Sep 19 04:53:09 hmmmm that does not sound good Sep 19 04:53:09 the PCB should last for thousands of cycles Sep 19 04:54:06 hmm wonder who makes that? BYD ? Sep 19 04:54:43 they're really designed for the "Oh you fucked up your device and charger"-scenario, as a last line of protection, not as first and only line of protection Sep 19 04:56:49 well the chargers need to be properly designed to match the cells Sep 19 04:57:03 yikes, installing imagemagick needs quite a few deps Sep 19 04:57:10 ha Sep 19 04:57:36 well I hope you are installing it on the drive Sep 19 04:58:52 of course Sep 19 04:59:25 but i am limited on space to 1GB either way and dir that grows would end up on md0 Sep 19 04:59:34 hmmmm Sep 19 04:59:37 1GB was never an issue on slug, i doubt it is now Sep 19 05:00:23 oh noes Sep 19 05:00:31 hmmm thought you had more storage than that Sep 19 05:00:34 this blows, there goes another 131MB Sep 19 05:00:50 tinker-f595: / is 1GB, rest is more Sep 19 05:01:02 well right now it's just /home Sep 19 05:01:22 hmmmm Sep 19 05:16:04 morfic: For sshfs just get fuse, add yourself to the fuse group if needed, then "sshfs -P xxx foo@xx.xx.xx.xx:/home/foo /home/foo/server Sep 19 06:02:28 this is awesome, script seems to be right, php seems to run, but somehow it can not display any of the thumbs/images Sep 19 06:05:44 ? Sep 19 06:06:09 oh the ImageMagick Sep 19 06:09:06 is it an arm version? Sep 19 06:10:39 it seems to convert them, but something is not working on the script after they are converted, i can see them with a filesize > 0 in the cache dir Sep 19 06:11:13 SuperRoach: well, i am trusting ubuntu here, but fact convert runs suggests it is arm :) Sep 19 06:11:16 morfic: do you have logging on? Sep 19 06:11:55 morfic: the apt config points to the correct repository Sep 19 06:12:44 the one it has on their filesystem, just moved to disk Sep 19 06:16:27 tinker-f595: i get access and error logs, not telling me much though Sep 19 06:16:46 dmesg? Sep 19 06:19:32 nope, nfs has some lockdv1 issues it seems, not flooding messages, so going to check into that later Sep 19 06:21:07 acts like it can't see the thumbs are there and runs convert to make them Sep 19 06:21:19 dir is readable for sure Sep 19 06:21:36 what time is it htere? Sep 19 06:23:06 1:23 am Sep 19 06:23:10 files are 666 Sep 19 06:25:52 hmm Sep 19 06:27:14 script runs so php works, convert runs andi get non zero .jpg that are 666, urls come up the way the script writer expects, things look the way they used to, hm, no clue, runs great on apache, it's where i used to use the script Sep 19 06:27:25 maybe i just found out the script is very apache centric Sep 19 06:27:49 it looks i might be able to have gentoo installed on internal flash, with /usr on the sdhc card and portage tree, kernel sources, distfiles and tmpdir mounted over nfs over gigabit ethernet Sep 19 06:28:03 or lighttpd is not very good for this purpose Sep 19 06:31:50 tinker-f595: or that Sep 19 06:32:05 ? Sep 19 06:32:10 script has only been run on apache, might have to go and install apache after all Sep 19 06:32:22 tinker-f595: re: lighttpd not being very good at that Sep 19 06:32:57 not sure how well lighttpd is at spawning processes Sep 19 06:36:22 morfic: I just hit my apache for you with top running http://pastebin.com/d79a69396 Sep 19 06:37:17 convincing Sep 19 06:38:03 convincing of what? Sep 19 06:38:39 thats a good question - lighthttpd or apache2? Sep 19 06:44:02 to try apache2 Sep 19 06:46:44 ah Sep 19 06:48:00 well I am going to sleep Sep 19 06:48:03 night Sep 19 06:49:06 /dev/sda1 966M 886M 32M 97% / Sep 19 06:49:09 weee Sep 19 06:49:16 * morfic must tidy things up here Sep 19 06:49:18 ? Sep 19 06:49:25 almost full Sep 19 06:50:31 oh well Sep 19 06:54:14 no biggie, i have ways to fix that, better brush up on my "i installed apache so long ago, i get to google up configurationa again" Sep 19 06:54:16 'night Sep 19 06:54:44 morfic I think I have a link for apache on ubuntu tutorial Sep 19 06:55:03 ubuntu did some weird stuff with config files and renaming Sep 19 06:56:10 remembering what goes where with all the "enabled dirs" and such is what i forget by time i reinstall apache2 somewhere else ;) Sep 19 06:56:35 i can google apache + ubuntu, trying not to keep you up, you already said you are going to sleep Sep 19 06:58:04 http://www.control-escape.com/web/configuring-apache2-debian.html Sep 19 06:58:27 bye Sep 19 07:02:07 what did you have on it morfic? Sep 19 07:02:55 SuperRoach: what i had it on? i think my desktop in the shed had it Sep 19 07:03:12 pretty sure actually, just been mothballed for so long now Sep 19 07:04:15 ah ok Sep 19 07:04:21 i thought you were talking about your sheevaplug Sep 19 07:04:43 with the rev 1.0 firmware, as anyone tried it to format a mmc to boot from? Sep 19 07:37:22 rere morfic Sep 19 07:37:49 working fine on apache2, the script that is Sep 19 07:38:04 bfs locked up X on here after something over 5 hours Sep 19 07:38:20 odd Sep 19 07:38:38 not really Sep 19 07:42:22 i havnt saw x lockup before Sep 19 07:42:51 has anyone used the 1.0 firmware installer's mmc option? Sep 19 07:43:20 SuperRoach: with bfs as scheduler? Sep 19 07:43:44 yeah Sep 19 07:48:38 SuperRoach: which version/what gfx Sep 19 07:48:40 no Sep 19 07:48:45 hold that thought Sep 19 07:48:50 bootin230 Sep 19 07:52:08 lol Sep 19 08:04:36 re Sep 19 08:12:00 thanks Sep 19 08:12:02 'night Sep 19 08:12:20 having a kernel compile looping, buh bye Sep 19 08:26:09 hi. the installer worked fine with the provided kernel Sep 19 08:26:27 ? Sep 19 20:38:22 on and off a lot morfic Sep 19 20:39:00 bfs, but this will change, moving morfic to znc on the sheeva Sep 19 20:39:14 so all you will see is when znc switches to "away" Sep 19 20:39:41 SuperRoach: bfs, as in bfs' fault for not being stable yet Sep 19 20:40:16 now i thought bfs was a filesystem Sep 19 20:40:41 didn't know irc could do state switching though, thats nice Sep 19 20:43:24 well, 6:43 am here, I better try to sleep for now. Sep 19 20:45:52 SuperRoach: bfs, the scheduler Sep 20 00:26:18 tinker-f595: howdy Sep 20 00:27:07 tinker-f595: i too am suffering from the lack of upstream patched. i need the rt2x00 tree for my wifi, and the marvell tree for my openrd. or to merge the two, which i do not want to do Sep 20 00:46:14 rektide: hi Sep 20 01:05:27 heya Sep 20 01:05:31 tinker-f595: Sep 20 01:26:02 rektide: I am back Sep 20 01:28:31 hey how goes the adventures Sep 20 01:28:55 have not messed with the OpenRD Client for a couple of days Sep 20 01:29:26 Seems uBoot needs some work Sep 20 01:31:38 oi Sep 20 01:31:46 thats "un-good" Sep 20 01:32:07 the stock one isnt good enough for you? Sep 20 01:32:27 well it seems to have problems with eSata. Sep 20 01:32:50 nothing was working for me (usb, sd...) so I gave up and just started using NAND / ubifs; nothing fancy. minimal demands from u-boot. Sep 20 01:32:53 ide reset with eSata is not very reliable Sep 20 01:32:56 ahhh Sep 20 01:33:22 you're not comfortable putting the kernel in mtd ? Sep 20 01:33:27 I would use sata but the spacers for mounting the drive on the board are impossible to find Sep 20 01:34:37 rektide: I want to be able to put a drive with the OpenRD Client, change a few u-boot parameters and have it working. Sep 20 01:36:37 I also noticed that if you have esata drive connected and an usb keyboard and mouse attached that you have to reset after 1st boot to get them to work. I see same problem with drive connected via USB Sep 20 01:37:14 Seems to be a power bus management problem which I suspect is uboot related Sep 20 01:37:21 holy moly Sep 20 01:37:49 I might play with it tomorrow so I can send a good write up to dhavel etc Sep 20 01:38:26 I suspect it is uboot related and that they have not actually tried using external drives via usb / esata Sep 20 01:38:45 i was able to boot via usb just fine & have an operational system Sep 20 01:38:50 Prpbably uboot drivers Sep 20 01:39:06 rektide was that using a USB HDD ? Sep 20 01:39:29 USB flash sticks, i tried two. they should present a comperable usb mass storage interface. Sep 20 01:39:42 the caveat was that Sep 20 01:39:45 rektide they are not. Sep 20 01:39:57 rektide: they use less power Sep 20 01:40:04 ahh yes, that is a difference. Sep 20 01:40:40 i have a sound card on here, so everything is on a powered hub (#1 lesson of computer audio: use separate power buses wherever possible) Sep 20 01:41:24 but the caveat i found was Sep 20 01:41:31 adding anything more than the most trivial usb setup Sep 20 01:41:33 rektide: so uboot needs to make sure the usb power management bus is set up and also wait a a few milliseconds for the initial spin up surge and then requery the usb bus. At least that is my understanding of how those sort of things are supposed to work Sep 20 01:41:38 caused the usb init to fail... badly Sep 20 01:42:09 rektide: I remember you mentioning that Sep 20 01:42:19 http://plugcomputer.org/plugforum/index.php?topic=653.0 Sep 20 01:42:31 rektide: any idea how to get uboot to spit out diagnostic info? Sep 20 01:42:39 i did some archaeology, and apparently the problem was ubiquotous to sheevapluga s well. Sep 20 01:42:51 no, thats a great question though. Sep 20 01:44:31 it seems like there have been plenty of openrd changes in the marvell git Sep 20 01:44:40 rektide: I am hoping it is a software fix for uboot that is required Sep 20 01:44:54 but that kind of seems end of the line Sep 20 01:45:43 tinker-f595: seems very very probable; all the periperhials work in linux, so its got to just be uboots elemntary drivers Sep 20 01:45:51 rektide: the only changes I saw were for the OpenRD base. dhavel said he was going to post the OpenRD Client changes upstream on the 17th...Date went by and nothing Sep 20 01:46:35 rektide: but without a good boot system headaches Sep 20 01:46:39 brb Sep 20 01:48:12 rektide which do you have? Sep 20 01:48:33 base Sep 20 01:48:52 so you have all the software so far Sep 20 01:49:17 fact of the matter is, until openrd getsmainline support, we're hanging in the wind Sep 20 01:49:41 hey maybe you could ask dhavel what happened to the openrd client updates to the git tree. I don't want to bug him again http://groups.google.com/group/openrd/browse_thread/thread/b4981ecb40628a00?hl=en# Sep 20 01:49:47 i tried building and using uboot from uboot mainline, but i never really figured out what i was doing and my first couple attempts didnt seem to boot Sep 20 01:50:00 rektide: haha Sep 20 01:50:24 i just pointed the openocd scripts at the image i'd built and crossed my fingers Sep 20 01:50:42 rektide: as I said my plan is to document the weird uboot stuff I have been seeing and report it to dhavel Sep 20 01:51:26 they really to have a solid boot loader and a decent linux distro and kernel for no other reason for sales demonstrations Sep 20 01:51:30 good plan, but ultimately the goal has to be mainlining. Sep 20 01:52:25 rektide: also I think I will see if I can fedora 11 booting Sep 20 01:52:37 real support in OS requires being in the real software archives Sep 20 01:52:45 i'm using debian now, & you ? Sep 20 01:52:51 rektide: true Sep 20 01:53:15 oh on the OpendRD Client it still has the Fed 8 that it came with Sep 20 01:54:06 back to your earlier inquiry-- to get esata up-- Sep 20 01:54:20 writing a new kernel to mtd is easy Sep 20 01:55:03 then just boot your mtd kernel with root=/dev/sdXY Sep 20 01:55:03 rektide: I know Sep 20 01:55:52 i'm not sure what parallels Fedora has for the convenient debootstrap with which to build a root FS Sep 20 01:55:53 rektide: I might do that as a temporary measure Sep 20 01:56:26 rektide I already have an RFS already to go Sep 20 01:56:36 and built a dirty kernel Sep 20 01:57:33 its easy and it works well, until u-boot gets into shape Sep 20 01:58:40 ug i'm not so sure that convincing Dharval to open his own trees was the right idea Sep 20 01:59:02 please please please may some of these trees be turned into patches & commited Sep 20 02:02:15 dhavel well I am not quite sure why he did that Sep 20 02:03:49 rektide: came across this the other day http://www.linux-arm.org/Main/WebHome Sep 20 02:04:28 rektide: looks like Marvell is playing with getting into the server space http://www.linux-arm.org/Main/LinuxArmOrg Sep 20 02:05:05 i now have 3 uboot trees. hard-core Sep 20 02:05:46 rektide: I how do they differ? Sep 20 02:05:50 hey & debian represent Sep 20 02:06:01 ???? Sep 20 02:06:53 dhavel's, marvell's & mainline. i just got dhavels. and marvel's had simon's old openrd patches on it for a whille. and uboot is miles ahead. Sep 20 02:07:04 the blade systems run debian Sep 20 02:08:30 yes they do run debian Sep 20 02:08:44 no fans as well Sep 20 02:09:34 64 bit memory controller Sep 20 02:09:37 dhavell forgot the hardware crypto stuff. It is not in his tree Sep 20 02:09:38 on the cpu Sep 20 02:09:53 dual issue FPU Sep 20 02:09:59 rektide: there is a dual core version of that Sep 20 02:10:07 yeah Sep 20 02:10:23 but i suspect its probably extremely similar in terms of actual architecture Sep 20 02:10:42 http://www.marvell.com/products/embedded_processors/discovery_innovation/index.jsp Sep 20 02:10:47 same cpu, just not hooked up to a drinking straw Sep 20 02:10:55 haha Sep 20 02:11:00 instead, big pipes and a big cpu Sep 20 02:11:16 err big pipes and big peripeherials Sep 20 02:11:17 and 2 cores in the higher end modle Sep 20 02:11:24 yes Sep 20 02:11:29 with 4 ether ports Sep 20 02:11:46 way bigger crossbar with way more peripherials on it, yes Sep 20 02:12:11 I have been wondering when Marvell will have someone sell a board based on that Sep 20 02:12:16 i'm terrified to find out what volume pricing is Sep 20 02:12:40 it drives me nuts that theres only a 16bit interface on this kirkwoodp rocessor Sep 20 02:12:55 it takes 2GB of ram to get 512 usable MB Sep 20 02:13:25 rektide: I wonder what the pricing is on the discovery series Sep 20 02:14:20 take kirkwood price, add 0 Sep 20 02:14:42 i dont really know, but thats typically how these games are played Sep 20 02:15:06 add what? Sep 20 02:15:42 a digit Sep 20 02:18:56 decimal shift left Sep 20 02:19:05 yes Sep 20 02:19:23 looks like Simon actually got his uboot patches posted upstream Sep 20 02:19:27 hopefully those really make it in Sep 20 02:20:03 the hw crypto is in the latest 2.6.31 vanilla kernel Sep 20 02:21:10 robbat2|na: but not in dhavels tree Sep 20 02:21:31 robbat2|na: I told dhavel about that Sep 20 02:21:51 rektide: which uboot patches? Sep 20 02:22:02 does his have support for the watchdog on the mv78100 boards at all? Sep 20 02:22:35 tinker-f595: gawd you are right, the -client patches still dont really exist Sep 20 02:22:43 except in dhavel's tree, afaik Sep 20 02:23:00 rektide: that is what I have been saying Sep 20 02:23:17 rektide: and his tree is missing other stuff Sep 20 02:24:36 robbat2|na: so you have a mv78100 board? Sep 20 02:24:45 yup Sep 20 02:24:55 it's generating the new ARM builds for Gentoo now Sep 20 02:24:56 robbat2|na: how do you like it? Sep 20 02:25:18 just sitting at home now, i've been trying to find a suitable small chassis for it so I can put it in a datacentre, with more bandwidth Sep 20 02:25:31 reasonable, just wish i'd know about the 3GiB RAM limit earlier on Sep 20 02:25:37 i cnat find simon kagstrom's openrd patches atm. Sep 20 02:26:09 and I want to poke the PCI code still, i suspect something is odd or just not pushed to vanilla trees yet Sep 20 02:26:21 robbat2|na: see the chassis here Sep 20 02:26:23 http://www.linux-arm.org/Main/LinuxArmOrg Sep 20 02:26:46 i don't have that much rackspace available ;-) Sep 20 02:26:57 i'm trying to find a 1 or 2U full-length BTX rackmount Sep 20 02:27:08 i've found microBTX rackmounts, but not full-length Sep 20 02:27:19 robbat2|na: we were conjecturing about how expensive the mv78100 chips must be Sep 20 02:27:39 yes Sep 20 02:28:14 the mv78200 looks interesting Sep 20 02:29:12 avnet lists the 78100 chips for $86 each Sep 20 02:30:50 http://tinyurl.com/n5c4jw Sep 20 02:32:52 they list the dual core mv78200 for $108 each Sep 20 02:38:19 higher clock version is 129 Sep 20 02:40:20 wifi cut out Sep 20 02:42:48 not bad. cant wait for aCortex A9 compairson Sep 20 02:42:59 hopefully Cortex A9 starts dropping SoC prices everywhere Sep 20 02:43:19 I'd love a quad core P4040 Freescale for sub $180. Sep 20 02:47:49 hmmmm Sep 20 02:51:16 avnet requires a quote for that part Sep 20 02:54:03 brand spanking new-- not sure if they even know the price. i figued it was as quite a bit below the old old 8 core p4080 at $495 and the old old 2 core p2020 at... uh... what was that price? $160 or something? Sep 20 02:55:06 i'm guessing its probably circa $300 now. Sep 20 02:57:00 sorry, $105 for the p2020 Sep 20 02:57:05 rektide: avnet has listed dev boards for the mv78100 and mv78200 but special order with 9 week lead time Sep 20 02:57:19 yikes; thats globalscale speeds Sep 20 02:58:04 someone saidthe other day they got there openrd client from globalscale in 2 weeks Sep 20 02:58:16 how much is the dev kit ? Sep 20 02:58:29 the mv78200, its $ ? Sep 20 02:58:31 rektide there was no price... Sep 20 02:58:43 fooey; that would've been interesting to see Sep 20 02:59:01 108 or 129 for the higher clock speed **** ENDING LOGGING AT Sun Sep 20 02:59:57 2009