[02:50:02] * jkridner|work (n=a0321898@nat/ti/x-865732188419e216) Quit ("Leaving.")
[05:22:46] * BThompson (n=BThompso@cpe-76-183-86-15.tx.res.rr.com) Quit ("Trillian (http://www.ceruleanstudios.com")
[07:50:27] * RogerMonk (n=a0740758@nat/ti/x-1726fd27dc77c756) Quit (Remote closed the connection)
[09:08:24] * RogerMonk (n=a0740758@nat/ti/x-b689a4da9c34ae68) has joined #beagle
[10:50:41] * RogerMonk (n=a0740758@nat/ti/x-b689a4da9c34ae68) Quit (Remote closed the connection)
[11:03:23] * RogerMonk (n=a0740758@nat/ti/x-7ae9b37a100c437f) has joined #beagle
[12:02:03] * jkridne1 (n=a0321898@nat/ti/x-a76f2c5981cab8d1) has joined #beagle
[12:15:21] * khasim (n=a0393720@192.163.20.231) has joined #beagle
[12:16:09] * khasim (n=a0393720@192.163.20.231) has left #beagle
[12:16:14] * khasim (n=a0393720@192.163.20.231) has joined #beagle
[12:16:16] <jkridne1> hi khasim!
[12:16:20] <khasim> hello
[12:16:33] <jkridne1> sorry you were not able to see everyone from the pibb.
[12:16:43] * jkridne1 is now known as jkridner|work
[12:16:48] <khasim> yeah, with Pidgin I am able to
[12:17:13] <khasim> Hi Koen
[12:21:05] <Crofton> probably still asleep .....
[12:21:30] <koen> no, in bed with a cold
[12:21:32] <koen> same difference :)
[12:21:46] <koen> khasim: I sent the patches to linux-omap@vger
[12:22:31] <koen> Crofton: http://dominion.thruhere.net/koen/cms/beagleboard-in-the-wild
[12:23:15] <khasim> thanks koen
[12:23:45] <koen> khasim: any idea how I can get omapfb to work on linux-omap git?
[12:26:13] <jkridner|work> khasim: I just found http://tiexpressdsp.com yesterday. Heard the URL might change, but it looks nice for covering more than http://www.davincidsp.com
[12:28:38] <khasim> Koen: I dont see your patches at linux omap vger
[12:29:19] <koen> khasim: I did send them, maybe the mailserver here is being slow again
[12:35:25] <khasim> Koen : OK.
[12:35:45] <khasim> Koen: The FB driver is bit different from what we have in our release on code.google.com
[12:35:54] <koen> I didn't get a reply for the majordomo at vger either
[12:36:04] <khasim> we have to work on migrating the GIT one ...
[12:36:09] <khasim> it will take some time
[12:37:53] <koen> I couldn't figure out how other boards 'register' the omapfb driver
[12:38:08] <koen> sakomans code registers the LCD, which according to him works
[12:38:27] <koen> his code for the EVM, btw
[12:39:19] <DJWillis> koen: it works, using Steve's code here (EVM)
[12:39:28] <koen> DJWillis: I don't doubt that :)
[12:39:49] <koen> but the beagle doesn't have an lcd, so I can't copy-paste the code like I did for usb and mmc
[12:40:03] <DJWillis> No, I realise that.
[12:40:13] <Crofton> doh, I forgot the EVM was LCD
[12:40:37] <koen> Crofton: I sent sakoman a link to a new kdrive (1.5rc) to test on the evm
[12:40:48] <DJWillis> Well the EVM has a DVI port also but I have yet to figure out how to get life out of it ;-)
[12:42:27] <koen> I wonder if the dvi and svideo ports can be driven independantly on the beagle
[12:42:32] <DJWillis> koen: I have a patch I was sent (can't remember where from, guess jkridner) to enable the DVI on the EVM, assuming that it is the same for the Beagle (and I can't see why not) I can mail you that.
[12:42:55] <koen> k.kooi@student.utwente.nl :)
[12:49:18] <koen> "Confirmation for subscribe linux-omap" there we go
[12:52:20] <DJWillis> koen: Mail found and sent on :-)
[12:52:28] <koen> thanks
[12:52:42] <koen> http://marc.info/?l=linux-omap&m=120843662011934&w=2 http://marc.info/?l=linux-omap&m=120843662111939&w=2 http://marc.info/?l=linux-omap&m=120843662211945&w=2
[12:58:01] * koen hates email
[12:59:03] <DJWillis> yep
[13:08:38] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
[13:09:55] * DarrenEtheridge (n=a0867391@nat/ti/x-7d54f0430931f895) has joined #beagle
[13:12:10] <jkridner|work> DJWillis, I believe it is somewhat different. Certainly the pin configs are different. Also, Beagle has DDI(?) that allows reading of the EDID, but the EVM does not. :(
[13:13:28] <jkridner|work> koen: yes, DVI and S-Video can be driven independently. you are limited to the 1 graphics window and 2 video windows, but you can choose where each one is displayed.
[13:13:57] <Crofton> btw, how much power does the beagle use?
[13:14:22] <jkridner|work> the u-boot up on code.google.com should emit a TV color-bar pattern on the S-Video and a red screen on the DVI-D.
[13:14:44] <jkridner|work> I used the patches posted to the mailing list against 1.3.2 and the DVI-D will display the BeagleBoard.org logo.
[13:15:10] <jkridner|work> Crofton: we are still optimizing the power consumption of the board and could use some help in that area.
[13:15:18] <jkridner|work> the DVI-D chip can be a pig.
[13:15:27] <jkridner|work> let me dig up some numbers.
[13:16:14] <koen> 5V x 500mA max :)
[13:16:35] <Crofton> ah, that is the limit that usb can supply
[13:16:57] <koen> jkridner|work: at some point I'll rebuild uboot to have ext2load, but atm I'm satisfied with 1.1.4
[13:19:09] <jkridner|work> koen, right.
[13:19:25] <jkridner|work> actually, I think it is something like 4.7Vx500mA is all you can guarantee.
[13:19:49] * BThompson (n=BThompso@nat/ti/x-99b592c0533d25f2) has joined #beagle
[13:22:28] <koen> I wonder how much video performance the beagle can deliver without using the iva2.2
[13:24:46] <prpplague> hmm, sounds like koen wants to play lots of pr0n videos on his beagle board
[13:24:58] <prpplague> koen: did you get your beagle delivered?
[13:25:28] <koen> yes: http://www.flickr.com/photos/koenkooi/2420765852/ :)
[13:26:15] <BThompson> http://www.bdti.com/bdtimark/cortex_a8.htm gives some basic performance figures on the CortexA8 itself (no IVA)
[13:27:05] <koen> BThompson: I don't know how optimized libavcodec is for the cortex
[13:27:18] <prpplague> wow its smaller that i imagined
[13:27:40] <koen> prpplague: about 3x3 inches
[13:28:21] <prpplague> the board seems more geared to software developers though
[13:29:24] * Crofton wonders if prpplague is ever happy :)
[13:29:39] <Crofton> koen, dailtone really ran?
[13:29:49] <prpplague> Crofton: hehe, never
[13:30:09] <prpplague> Crofton: at best, i'm just "not unhappy"
[13:30:25] <Crofton> here we have TI seriously trying to help the opensource community buy providing useful cheap hw
[13:30:27] <koen> Crofton: with 2.6.22, yes
[13:30:30] <Crofton> ok
[13:30:38] <Crofton> I'll forward to gnu radio list
[13:30:40] <BThompson> koen: I agree I am not sure how helpful those figures are from a usage stand point, but they are the only benchmarks I have come across, I figure they could be useful if you compared them to other processors running the bdti tests
[13:30:45] <Crofton> was the compiler neon aware?
[13:31:32] <prpplague> Crofton: wasn't really complaining about the board per se, it's a wonderful step in the right direction for a good OMAP board
[13:31:38] <Crofton> yeah
[13:31:43] <Crofton> I'm just poking at you
[13:40:15] <prpplague> Crofton: ahh ok
[14:26:48] <koen> ah, 2.6.25 is out
[14:28:12] <prpplague> koen: yep, just dl'd and started a fresh compile
[14:28:36] <prpplague> koen: hehe, someone actually commited my enc28j60 driver into 2.6.25
[14:33:00] <koen> would be nice to have beagle and evm into 2.6.26
[14:33:19] <jkridner|work> 2.6.25 is already on bug fixes, isn't it?
[14:34:08] <prpplague> koen: well, based on the info i've got from rmk and my arch maintainer(s3c24xx), we've got about a 2 week window right now for new device patches
[14:34:17] <prpplague> jkridner|work: no, just released this morning
[14:34:52] <koen> prpplague: I dunno how tony handles merge windows
[14:35:56] <prpplague> koen: usually for arm if its not submitted by rc6 or 7 , it won't get in for the next release
[14:36:33] <koen> and RMK will bitch anyway
[14:36:50] <koen> I unsubscribed from lak a while ago
[14:37:05] <koen> it was turning into the RMK real life soap too much
[14:37:19] <Crofton> just feed to stuff to linux-omap
[14:37:37] <Crofton> they are pretty good about staying close to upstream anyway
[14:37:53] <Crofton> it is nice to have a buffer from lak
[14:37:56] <prpplague> koen: indeed
[14:37:57] <prpplague> http://www.arm.linux.org.uk/developer/patches/search.php?summary=omap
[14:38:10] <prpplague> looks like tony has a whole round of omap patches schedule for 2.6.26
[15:20:03] <koen> koen@lieve:/OE$ avahi-browse -at | grep beag
[15:20:03] <koen> + eth0 IPv4 beagleboard [00:10:60:84:4f:26] Workstation local
[15:20:06] <koen> + eth0 IPv4 beagleboard SSH Remote Terminal local
[15:20:10] <koen> + eth0 IPv4 SFTP File Transfer on beagleboard SFTP File Transfer local
[15:20:13] * koen hugs avahi
[15:38:50] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
[15:39:14] <ds2> they ought to make ezOMAP3 (a la ez430)
[16:08:39] * horizon (i=horizon@141.30.223.123) has joined #beagle
[16:08:48] <horizon> hi.
[16:13:21] <Crofton> gm
[16:38:02] * DarrenEtheridge (n=a0867391@nat/ti/x-7d54f0430931f895) Quit (Remote closed the connection)
[16:51:18] * RogerMonk (n=a0740758@nat/ti/x-7ae9b37a100c437f) Quit (Remote closed the connection)
[17:01:30] * dirk2 (n=dirk@F3205.f.strato-dslnet.de) has joined #beagle
[17:10:24] * DarrenEtheridge (n=a0867391@nat/ti/x-fb455794396a9c36) has joined #beagle
[17:16:43] <koen> jkridner|work: do you have that optimized mplayer build online somewhere?
[17:17:08] <koen> the armv6 mplayer build doesnt perform to well :(
[17:17:51] <koen> it's 30% slower than run on a 300MHz omap2
[17:22:14] * koen tries an armv7 build
[17:22:19] * koen crosses fingers
[17:25:52] <prpplague> dirk2: ping
[17:26:09] * khasim (n=a0393720@192.163.20.231) has joined #beagle
[17:27:37] <dirk2> hi
[17:28:03] <prpplague> dirk2: hey, we are getting ready to switch over to the new server
[17:28:16] <dirk2> GO! I have backups ;)
[17:28:25] <prpplague> dirk2: your current pages will still be available at http://movial.elinux.org
[17:28:47] <dirk2> Thanks. It will only take minutes to copy & paste stuff :)
[17:28:51] <prpplague> dirk2: i'll be going thru to make sure all the pages that have been updated on the old server are moved over
[17:29:07] <prpplague> dirk2: but i just didn't want you to freak out when they weren't there initially
[17:29:56] <dirk2> Okay, I then will wait until tomorrow before touching them again.
[17:30:43] <prpplague> dirk2: should have a green light by then
[17:31:03] <prpplague> dirk2: thanks for your good attitude about this all, it has been a real mess
[17:31:47] <dirk2> Would be interesting to know the reason. Do you know what happened?
[17:33:03] <dirk2> All: How do I create a "bootable partition on MMC/SD Card" under Linux without this "HP USB Disk Storage Format Tool 2.0.6" for Windows?
[17:33:13] <dirk2> http://code.google.com/p/beagleboard/wiki/BootingBeagleBoard
[17:33:13] <prpplague> dirk2: sorry no, godaddy was less than helpful, we've made extra arrangments to have daily backups and a backup server ready
[17:33:49] <NishanthMenon> dirk2: formatting with the app will do - format it in FAT32 mode.
[17:33:49] <prpplague> dirk2: should be able to use fsdisk/fdisk
[17:33:59] <NishanthMenon> linux -> fdisk is good enough
[17:35:01] * prpplague is a sfdisk fan
[17:35:27] <dirk2> ;) man fdisk: sfdisk is for hackers only
[17:36:50] * prpplague laughs
[17:37:02] <NishanthMenon> kind of wierd that windows format utility on xp does not make the partition bootable OR have an option to create bootable partition..
[17:37:59] <dirk2> I don't want to reboot into Windows. Put precompiled MLO, u-boot.bin and self compiled uImage on SDcard. But nothing happens :(
[17:38:38] <dirk2> In serial console still the .... 40T .... from UART bootloader
[17:38:49] <NishanthMenon> 40T is the ASIC ID
[17:39:19] <dirk2> Yes. But I want xloader output from SDcard ;)
[17:39:27] <NishanthMenon> the boot sequence failed for MMC.. hmm... aaah.. SD..
[17:39:58] <NishanthMenon> khasim tried with these cards: Kingston 1GB SD Card , PQI Hi-Speed 150 1GB SD card, etc..
[17:40:21] <NishanthMenon> what card are u using? some 4GB High capacity card?
[17:40:41] <koen> a sandisk 1gb SD card works
[17:40:42] <dirk2> No. An old one. TOSHIBA 256MB SD Memory Card
[17:40:56] <koen> but I had to format it as fat32, not fat16
[17:41:10] <dirk2> Under Win or Linux?
[17:41:22] <NishanthMenon> dirk2: koen: yeah. FAT32 did work for me too.. i did it on windows though.. :(
[17:41:50] * koen used a vm with xp in for formatting
[17:42:01] <koen> and parted to resize the fat partition
[17:42:14] <NishanthMenon> here is a snippet of a similar discussion "Put the card in a reader and connect to a Linux PC. Run fdisk and check if the partition is active. If not, mark it (I think the ?a? command) then the ?w? command. Then try to boot the card. "
[17:43:14] <NishanthMenon> later in the thread: "After marking boot partition as ?active? (0x80) almost all cards worked."
[17:43:25] <NishanthMenon> then more back and froth ...
[17:44:06] <NishanthMenon> another recommendation i heard from someone:" One thing you could try is to delete the first sector (set all bytes to zero) and reformat "
[17:46:25] <prpplague> what i've found in the past also works is to remove all partitions, then format the entire device as a single filesystem(fat or ext2 doesn't matter) then repartition and set the partition active
[17:46:36] <prpplague> its round about, but seems to work
[17:51:23] * NishanthMenon thinks getting nand boot working and a flashing solution is useful instead of hunting for a sd/mmc card..
[17:52:01] <dirk2> Yes, serial download (and then flashing) would be next step
[17:52:03] <koen> aha
[17:52:15] <koen> a big cpu load kills usb on omap-git
[17:52:26] <prpplague> oh nand booting isn't working on the beagle yet?
[17:52:36] * prpplague is kind of surprised
[17:52:49] <NishanthMenon> prpplague: nand boot works. we need images, and some port of x-loader i think
[17:52:50] <koen> prpplague: uboot and xloader support it
[17:53:00] <koen> but the kernel doesn't
[17:53:09] <prpplague> jtag and primary flash support are the two things i get working first on a new board
[17:53:11] <koen> no mtd drivers enabled and stuff
[17:53:27] <koen> a 5 minute patch job, though :)
[17:53:47] <ds2> is it nnand or onenand?
[17:53:59] <NishanthMenon> the new boards have nand - 16bit
[17:54:17] <NishanthMenon> at least koen has a new board i think look at the flickr snap ;)
[17:54:48] <koen> 256Mbit NAND if I believe uboot
[17:54:50] <prpplague> i'm just not a fan of uboot
[17:54:59] * NishanthMenon big fan of uboot
[17:55:12] <koen> I like uboot over redboot
[17:55:12] <ds2> Mb???
[17:55:34] <ds2> :(
[17:55:57] <prpplague> if/when i get a beagle board, i'd like to get apex working on it
[17:57:29] <NishanthMenon> koen i seem to have it as 256Mbyte
[17:57:42] <NishanthMenon> rather 256MByte..
[17:58:12] <koen> 32mb flash is plenty of space
[17:58:30] <koen> you can fit X11 in there quite easily
[17:58:39] <koen> and still have 16MB free :)
[17:58:43] <NishanthMenon> untill u start adding hoggy stuff and store mp3 and video files :D
[17:58:50] <prpplague> actually i've gotten x running on much less
[17:59:23] <koen> prpplague: you can run X in 2-3MB flash
[17:59:36] <prpplague> http://www.elinux.org/Pixter_Multimedia
[17:59:38] <koen> but that's with cramfs, uclibc and other things
[18:00:00] <NishanthMenon> actually for those interested in nand: beagle new versions like what koen has seems to share same as zoom MDK.. openomap.org/pub has code for it
[18:00:11] <prpplague> koen: talk about "tight"
[18:00:30] <koen> I've seem people putting X into their PC bios flash chips :)
[18:00:36] <dirk2> How do I detect "new" versions?
[18:00:56] <prpplague> koen: nice
[18:01:04] <NishanthMenon> dirk2: i think the new ones have micron nand
[18:01:15] <prpplague> koen: the pixter-mm was a nice little device, you can still get them for around $20
[18:01:18] <prpplague> USD
[18:01:33] <koen> OMAP3430-GP rev 2, CPU-OPP2 L3-133MHz
[18:01:34] <koen> TI 3430Beagle 2.0 Version + mDDR (Boot ONND)
[18:01:34] <koen> DRAM: 128 MB
[18:01:34] <koen> Flash: 0 kB
[18:01:34] <koen> NAND:256 MiB
[18:01:34] <ds2> hmmmmm zoom really exists?
[18:01:39] <koen> MiB indeed
[18:02:12] <NishanthMenon> what is MiB?
[18:02:20] <koen> MebiByte
[18:02:26] <koen> 1024 KibiBytes
[18:02:38] <koen> instead of the marketing 1000 kilobytes
[18:03:03] <NishanthMenon> koen: thanks :)
[18:03:14] <prpplague> ds2: yea well, i have less than favorable things to say about logicpd
[18:03:17] * DarrenEtheridg1 (n=a0867391@nat/ti/x-6a24c791b8fabdcc) has joined #beagle
[18:03:36] <koen> prpplague: you should have heard Lennert about logicpd :)
[18:03:43] <NishanthMenon> ds2: http://www.logicpd.com/products/devkit/ti/zoom_mobile_development_kit
[18:03:44] <ds2> :(
[18:03:52] <prpplague> koen: hehe, good or bad?
[18:04:13] <dirk2> Texas Instruments X-Loader 1.41
[18:04:15] <koen> depends on how you interpret "&%$*#%$&*# CPLD"
[18:04:15] <dirk2> Starting on with MMC
[18:04:22] <prpplague> koen: hehehe
[18:04:22] <dirk2> Reading boot sector
[18:04:29] <dirk2> Error: reading boot sector
[18:04:33] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
[18:04:52] <ds2> NishanthMenon - is it orrderable?
[18:05:02] <NishanthMenon> dirk2: ouch! x-loader mmc support... so rom code did get x-loader started in sram :)
[18:05:34] <NishanthMenon> ds2: I dont know that answer..
[18:05:57] <NishanthMenon> ds2: they seem to have a link for pricing i think..
[18:05:57] <koen> you can register for a beta unit
[18:06:01] * DarrenEtheridg2 (n=a0867391@nat/ti/x-6f30f79de89b1253) has joined #beagle
[18:06:38] <dirk2> NishanthMenon: Something wrong with boot sector on SDcard?
[18:06:46] <ds2> that i my complaint....4m+ wait and counting
[18:07:07] <ds2> koen, where?
[18:07:20] <NishanthMenon> dirk2: i think something going wrong with x-loader reading uImage
[18:07:31] <NishanthMenon> dirk2: sorry.. u-boot.bin
[18:08:09] <dirk2> okay, will have a look for it
[18:08:13] <koen> http://www.openomap.org/beta.html
[18:08:15] <NishanthMenon> dirk2: ./fs/fat/fat.c
[18:08:23] * DarrenEtheridge (n=a0867391@nat/ti/x-fb455794396a9c36) Quit (Remote closed the connection)
[18:09:44] * DarrenEtheridg1 (n=a0867391@nat/ti/x-6a24c791b8fabdcc) Quit (Remote closed the connection)
[18:10:29] <DJWillis> The amount of low cost (and not so low cost) boards with OMAP3's is a very promising sign. Things like Beagle and others really start to get the hardware into hobby hands and I really like that.
[18:10:53] <ds2> thanks
[18:11:48] <koen> DJWillis: hopefully that will mean that we end up with better (read: Free) dsp tools
[18:11:58] <Crofton> amen
[18:12:03] <NishanthMenon> amen
[18:12:17] <koen> although I have access to CCS on the uni :)
[18:12:18] <Crofton> everyone, place your hand on the computer!
[18:12:31] <Crofton> ~praise free DSP tools
[18:12:38] <DJWillis> koen: I really hope so, there are so many people clamering to use the DSP aspect.
[18:12:56] <ds2> i wold like tto see a ezOMAP stick
[18:13:04] <prpplague> DJWillis: indeed, now if OEM's can actually purchase the SoC's in smaller qty so they can do prototype runs
[18:13:08] <prpplague> DJWillis: that would be promising
[18:13:13] * NishanthMenon wondering what singular OS can use heterogenous multiprocessing well..
[18:13:30] <prpplague> DJWillis: thats the #1 reasons OMAP has lost out on our designs for the last 4 years
[18:13:51] <Crofton> http://mailman.tu-chemnitz.de/pipermail/gcc-c6x/2007-March/000031.html
[18:14:44] <prpplague> ds2: omap version of the nail board?
[18:14:53] <DJWillis> prpplague: I can only speak for our project but TI have really supprised us (we wrote OMAP off at 1st). TI seem to want to try and take a step change with OMAP3, more help to the little guys ;-), I guess they get a better platform out of it with lots of free hackers.
[18:15:30] <prpplague> DJWillis: yea, thats what i'm hearing
[18:16:20] <prpplague> DJWillis: i'm planning to request a couple beagle boards to eval, if they look nice, we might try to do a small prototype run for testing
[18:16:29] <ds2> prpplague, sure or a smller one. like the ez430 from TI
[18:18:14] <prpplague> ds2: imho the ez430 is too small
[18:18:21] <prpplague> ds2: but thats just me
[18:18:29] <DJWillis> prpplague: as soon as our boards are further along I would like to get flyswatter going, are you happy to sell them to people in quantities of 1 ;-). Historicly we have just put Wigglers for jTAG use with the old ARM9 SoC's
[18:18:36] <prpplague> ds2: it would be nice to have a pin for pin compatible omap-hammer
[18:18:53] <prpplague> DJWillis: http://www.tincantools.com
[18:19:00] <prpplague> DJWillis: thats what we are in business for, hehe
[18:19:03] <ds2> the price is righht nd i can use the ccarrier board as is
[18:20:57] <DJWillis> prpplague: true, I guess, homebrewhardware seems a fair description ;-)
[18:23:01] <Crofton> dirk2,
[18:23:04] <prpplague> DJWillis: TinCanTools is a sister company to American Microsystems Ltd. http://www.amltd.com
[18:23:08] <Crofton> root@davinci-sffsdr:~# flash_eraseall /dev/mtd2
[18:23:08] <Crofton> Erasing 128 Kibyte @ 7d40000 -- 99 % complete.
[18:23:15] <dirk2> prpplague: New server is up
[18:23:16] <Crofton> I supect it nuked the kernel again
[18:23:16] <prpplague> DJWillis: we do industrial handheld designs
[18:23:21] <prpplague> dirk2: yep
[18:23:23] <Crofton> based on the 128M number
[18:23:38] <prpplague> DJWillis: we usually do a dev board when bringing up a new product
[18:23:43] <Crofton> yep
[18:24:16] <prpplague> DJWillis: and afterwards it usually goes by the wayside(i.e. read trash), so we've started tincantools to resale some of our unused dev designs
[18:24:43] <dirk2> prpplague: Or shall I move my pages *now*? Will take only 5min and you don't need to do it :)
[18:25:33] <prpplague> dirk2: any help would be great
[18:25:45] <DJWillis> prpplague: nice idea, I know AML's stuff well but we tend to us others (Kontron, Psion et. all) at work :(
[18:26:07] <dirk2> Do I have to care about the 'Please do not make any updates until this notice is removed' on the main page?
[18:26:51] <prpplague> DJWillis: yea, 95% of our stuff isn't retail, its strictly B2B
[18:27:10] <prpplague> dirk2: nope
[18:27:20] <prpplague> dirk2: that on the main page?
[18:27:29] * DarrenEtheridge (n=a0867391@nat/ti/x-8f0fcac0a05a465f) has joined #beagle
[18:28:46] <DJWillis> prpplague: my day job is Technical Architect for a big Utility - We use a lot of rugged kit in all maner of ways ;-)
[18:29:07] <prpplague> DJWillis: ahh
[18:29:56] <DJWillis> prpplague: this hackery is just what I call 'fun' (sometimes) and the OMAP3 project I am tied to is also something I am doing for fun.
[18:30:23] <prpplague> DJWillis: ahh ok, similiar to my tincantools stuff, hehe
[18:30:55] <DJWillis> Less hardware, more bad code ;-), and no income stream for me, but the gist is right ;-)
[18:31:26] * koen enables armv6 SIMD in the mplayer build
[18:31:52] <koen> with armv7 pipeline correction it's only 20% slower than my nokia 770
[18:32:20] <koen> so something is still very wrong
[18:32:42] <koen> a 600MHz omap3 is supposed to be faster than a 250MHz omap1 :)
[18:33:24] <DJWillis> One would hope ;-)
[18:33:46] <dirk2> prpplague: Pages moved, image upload fails
[18:33:57] <prpplague> dirk2: yea, just found that
[18:34:01] <prpplague> dirk2: looking now
[18:34:16] <DJWillis> koen: we have a quick build of mplayer on the EVM, not sure what quick port consists of but performance is very good.
[18:35:50] * DarrenEtheridg2 (n=a0867391@nat/ti/x-6f30f79de89b1253) Quit (Remote closed the connection)
[18:36:58] <Crofton> DJWillis, what toolchain do you use?
[18:38:13] <NishanthMenon> koen: omap3 has Neon +ARM-Cortex A8 -> thumb2 etc.. other than the clocking changes -> kinda fast
[18:38:59] <DJWillis> Crofton: currently (well that) was CodeSourcery 2007 Q4, so ARM6.
[18:39:35] <DJWillis> Crofton: my current is a handbuild 4.3 with ARMV7 patches, about to try CodeSourcery 2008 Q1 as soon as I get a chance.
[18:45:26] <DJWillis> Crofton: of course ARM6 = ARMV6 ;-)
[18:46:25] <Crofton> koen, should be using gcc-4.3 I think
[18:47:14] <Crofton> is there any way to look for pipeline stalls and cache misses easily?
[18:47:44] <koen> DJWillis: http://rafb.net/p/HAG6Lv51.txt
[18:48:19] <koen> I get ~180 seconds on the beagle without armv6 simd, but with gcc 4.3.0 using -march=cortex and stuff
[18:48:42] <koen> omap1710 can get ~139 seconds
[18:48:42] <DJWillis> koen: if the EVM was here I would do that :(, once I am home I will see that I can do :D
[18:50:15] <DJWillis> That 139 sounds about right for my N770
[18:53:58] <koen> maybe the kernel is locking up somewhere
[18:54:21] <koen> preempt is turned off atm
[18:56:16] <DJWillis> Well for ease i'll try and test on the 2.6.22 kernel and ROOTFS from TI, I think I have my .25 kernel based off Steve's on there at the moment.
[18:56:56] * dirk2 (n=dirk@F3205.f.strato-dslnet.de) has left #beagle
[19:19:43] * dirk2 (n=dirk@F3205.f.strato-dslnet.de) has joined #beagle
[19:21:57] <dirk2> back from booting to win and formatting SD card. Now I get a U-Boot prompt :)
[19:22:20] <koen> :D
[19:22:57] <dirk2> Have to find out the correct fdisk/mkfs.vfat to get the same result like with the win tool...
[19:23:39] <dirk2> ... fdisk/mkfs.vfat *options* to ...
[19:25:12] <dirk2> How do I start uImage now from u-boot prompt?
[19:26:30] <BThompson> boot?
[19:26:51] <koen> dirk2: tried fatload?
[19:27:07] <dirk2> which options?
[19:27:08] <koen> dirk2: mmcinit ; fatload mmc 0:1 0x80000000 uimage ; bootm
[19:27:20] <dirk2> :) thanks
[19:27:39] <koen> to make that happen every boot: set bootcmd 'mmcinit ; fatload mmc 0:1 0x80000000 uimage ; bootm' ; saveenv
[19:29:32] <dirk2> uname -a: Linux 2430SDP-2430 2.6.25-rc9-omap1-04453-g9d0cbf9-dirty #2 Thu Apr 17 17:48:12 CEST 2008 armv7l unknown
[19:29:35] <dirk2> :)))
[19:30:55] <koen> ah, with armv7 SIMD I'm down to 166.215s
[19:31:06] <koen> ehm armv6 SIMD
[19:31:54] <dirk2> Thanks for all your help! Will document all this in wiki. Hopefully I'm the last one asking this stuff then ;)
[19:38:06] * dirk2 (n=dirk@F3205.f.strato-dslnet.de) has left #beagle
[19:52:47] * DarrenEtheridg1 (n=a0867391@nat/ti/x-aa29f226169d7709) has joined #beagle
[19:55:04] <koen> drat, the mplayer VFP stuff isn't upstream
[19:59:12] * DarrenEtheridge (n=a0867391@nat/ti/x-8f0fcac0a05a465f) Quit (Remote closed the connection)
[20:14:29] * DarrenEtheridg1 (n=a0867391@nat/ti/x-aa29f226169d7709) has left #beagle
[20:15:59] * koen_ (n=koen@s55917625.adsl.wanadoo.nl) has joined #beagle
[20:16:14] * koen (n=koen@s55917625.adsl.wanadoo.nl) Quit (Nick collision from services.)
[20:16:20] * koen_ is now known as koen
[20:46:21] * jkridner|work (n=a0321898@nat/ti/x-a76f2c5981cab8d1) Quit (Remote closed the connection)
[21:01:03] * jkridne1 (n=a0321898@nat/ti/x-fa80518784d55136) has joined #beagle
[21:11:29] <horizon> one question, what pins are available on the extension connector? gpio only or also some sort of higher-speed interface?
[21:15:51] * PibbRelay (n=supybot@nat/janrain/x-60ab231533f87839) has joined #beagle
[21:26:06] * BThompson (n=BThompso@nat/ti/x-99b592c0533d25f2) Quit ("Trillian (http://www.ceruleanstudios.com")
[21:27:01] * NishanthMenon (n=gnat@nat/ti/x-b2dd36971ff155e7) Quit ("Make it idiot proof and someone will make a better idiot.")
[21:34:31] * Crofton wonders if they have McBSP on the extension connector ...
[21:35:53] * Paulius (n=Paulius@69-196-135-97.dsl.teksavvy.com) has joined #beagle
[21:37:46] <Paulius> The beagle board looks so awesome.
[21:37:57] <Paulius> When will it be available and how much is it going to be?
[21:38:54] * Paulius (n=Paulius@unaffiliated/paulius) Quit (Remote closed the connection)
[21:39:17] * Paulius (n=Paulius@unaffiliated/paulius) has joined #beagle
[21:41:03] <ali_as> I think the developers are concentrating on making it work first with things like that left till later.
[21:41:13] <horizon> Crofton: looks like i2c, i2s, spi and mmc/sd only
[21:41:17] <Paulius> awww
[21:41:29] <Crofton> what do you want to hook up?
[21:43:51] <horizon> i thought of sata
[21:44:15] <Crofton> ah
[21:44:23] <Paulius> sata on an embedded device?
[21:46:53] <Crofton> beagle/omap3 may create interesting solutions to some problems :)
[21:49:45] <ali_as> It will read and write to hard disks using butterflies.
[22:12:20] * jkridne1 (n=a0321898@nat/ti/x-fa80518784d55136) Quit (Remote closed the connection)
[22:22:24] * Nishanth_Home (n=Nishanth@cpe-24-175-68-203.tx.res.rr.com) has joined #beagle
[22:27:30] * Nishanth_Home (n=Nishanth@cpe-24-175-68-203.tx.res.rr.com) Quit (Remote closed the connection)
[22:27:47] * Nishanth_Home (n=Nishanth@cpe-24-175-68-203.tx.res.rr.com) has joined #beagle
[22:37:04] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
[22:51:05] * Nishanth_Home (n=Nishanth@cpe-24-175-68-203.tx.res.rr.com) Quit (Read error: 110 (Connection timed out))
[22:51:08] * Nishanth_Home (n=Nishanth@cpe-24-175-68-203.tx.res.rr.com) has joined #beagle
[23:58:16] * Nishanth_Home (n=Nishanth@cpe-24-175-68-203.tx.res.rr.com) Quit (Read error: 110 (Connection timed out))