• [00:00:42] * RogerMonk (n=a0740758@nat/ti/x-e47deaa3999554de) Quit (Remote closed the connection)
  • [00:06:57] * robclark (n=robclark@nat/ti/x-88cb553c606a42dd) Quit (Remote closed the connection)
  • [00:10:43] * JoeBorn_ is now known as JoeBOrn
  • [00:10:48] * JoeBOrn is now known as JoeBorn
  • [00:50:48] * JoeBorn (n=jborn@dsl017-022-252.chi1.dsl.speakeasy.net) Quit ("home home home")
  • [01:23:29] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [01:42:46] * mru (n=mru@thrashbarg.mansr.com) Quit (Read error: 110 (Connection timed out))
  • [01:45:27] * jkridner_ is now known as jkridner
  • [02:05:22] * mru (n=mru@thrashbarg.mansr.com) has joined #beagle
  • [02:22:28] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [02:45:21] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [03:30:45] * NishanthMenon (n=nmenon@nat/ti/x-6c2d67c88292b287) Quit ("Ex-Chat")
  • [03:43:25] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [04:05:22] * nomis (i=simon@vmax.unix-ag.uni-siegen.de) Quit (Remote closed the connection)
  • [04:05:30] * nomis (i=simon@vmax.unix-ag.uni-siegen.de) has joined #beagle
  • [04:12:24] * pdg37 (n=paul@wsip-24-234-78-58.lv.lv.cox.net) Quit (Read error: 110 (Connection timed out))
  • [04:25:19] * Bari\ (n=ly@c-24-12-190-159.hsd1.il.comcast.net) has joined #beagle
  • [04:35:04] <Bari\> anyone know what firmware is used to init the SOC and what bootloader is used for Linux? u-boot? something custum?
  • [04:36:17] <kulve> U-Boot is the bootloader
  • [04:38:35] <kulve> Bari\: I think beagle is booted very similarly to the Zoom boards: http://www.omapzoom.org/wiki/tiki-index.php?page=Booting+and+Flashing
  • [04:39:19] <Bari\> kulve: thanks
  • [04:42:45] <jkridner> Bari\, check out http://elinux.org/BeagleBoard. there has been work on u-boot-v2. I also expect WinCE support in the future.
  • [04:43:44] * n0mis (i=simon@vmax.unix-ag.uni-siegen.de) has joined #beagle
  • [04:44:06] * nomis (i=simon@vmax.unix-ag.uni-siegen.de) Quit (Read error: 104 (Connection reset by peer))
  • [04:44:36] <jkridner> kulve: thanks for sharing the link. I like all the drawings.
  • [04:47:58] <Bari\> I was just looking at what it would take to port coreboot to the beagle, it may only make sense if the beagle design was expanded to support PCI/pci-e
  • [04:49:31] <jkridner> what is 'coreboot'?
  • [04:50:27] <Bari\> http://www.coreboot.org/
  • [04:50:33] <jkridner> more to the point, why would you need PCI?
  • [04:50:50] <jkridner> yeah, I found the website pretty fast with Google.
  • [04:50:56] <jkridner> what makes it better than other bootloaders?
  • [04:51:02] <Bari\> open source firmware, mainly used for x86 but not limited to
  • [04:51:30] <jkridner> I'm not partial to u-boot, but that is also open source.
  • [04:52:13] <jkridner> the ROM on the OMAP3530 enables boot from 4 sources on the Beagle board: the PoP NAND, SD, USB OTG, and serial.
  • [04:52:52] <jkridner> generally they are just utilized to load a next stage bootloader, such as u-boot, but I wouldn't see any reason not to have coreboot.
  • [04:53:03] <jkridner> I just don't know any reasons TO have it yet either.
  • [04:53:22] <jkridner> I guess it is really good at handling PCI/PCI-E devices?
  • [04:53:30] <Bari\> yeah, I was just kicking the idea around
  • [04:53:44] <jkridner> EHCI Debug Port looks interesting.
  • [04:53:56] <Bari\> I've been using u-boot as well for several years
  • [04:54:07] <jkridner> The next rev of Beagle will have an EHCI port, in addition to the OTG port that is there on Rev. B.
  • [04:57:24] <Bari\> coreboot may just make it simpler to support GRUB2 or etherboot
  • [04:58:30] <jkridner> if it had good USB support, that would be interesting.
  • [04:59:16] <jkridner> right now, many u-boot patches are still required to make the bootloader respond to keyboard via USB, as the more typical use case of u-boot is over network or serial ports.
  • [04:59:51] <Bari\> coreboot just init's the devices, the USB support would come from Linux, but coreboot can use Linux as the bootloader
  • [05:00:40] <Bari\> for example I have coreboot + kernel + busybox in 900KB of ROM on x86
  • [05:02:33] <Bari\> kexec is used to jump from the kernel in ROM to another kernel located on HD, flash, network etc
  • [05:07:46] <Bari\> coreboot + kernel +busybox +initrd has been stuffed into 256KB with older 2.2 kernels
  • [05:19:00] <Bari\> I haven't looked too deeply into the MPU datasheet, is the L2 cache always on and can a portion of it be protected after being written to?
  • [05:22:28] <bazbell> Bari\: No, the L2 cache is not enabled by default. Yes, a portion can be locked.
  • [05:26:43] <Bari\> bazbell: thanks, coreboot uses cache as ram similar to u-boot
  • [05:29:55] <Bari\> so coreboot would init the L2 cache and run from there until RAM is set up and then run from RAM, after RAM is up a kernel could loaded and run for USB or any other support
  • [05:46:49] * NeuroTech2 (n=NeuroTec@adsl-75-51-176-138.dsl.lsan03.sbcglobal.net) has joined #Beagle
  • [06:23:16] * macneib (n=macneib@d221-71-252.commercial.cgocable.net) Quit (Read error: 110 (Connection timed out))
  • [06:24:47] * glp (n=glp@2303ds1-fm.0.fullrate.dk) has joined #beagle
  • [06:25:50] <glp> good morning
  • [06:29:47] <NeuroTech2> morning
  • [06:52:27] <glp> I thought I would like to inquire about what systems presently are running on the beagle board?
  • [06:54:02] <kulve> systems?
  • [06:54:57] <glp> maybe a unclear formulation, with systems
  • [06:57:04] <glp> is it a custom-build linux system, something based on openembedded or ...?
  • [06:58:05] <kulve> ?ngstrom distro built with OE I think
  • [06:59:57] <glp> has the board been ported to other distro's?
  • [07:00:35] <kulve> there is no board, yet.
  • [07:00:40] <kulve> just some engineering samples
  • [07:01:10] <kulve> but I guess most of the distros running on arm will run on beagle just fine (debian, ubuntu, etc)
  • [07:04:07] * NeuroTech2 (n=NeuroTec@adsl-75-51-176-138.dsl.lsan03.sbcglobal.net) Quit (Read error: 110 (Connection timed out))
  • [07:07:15] * rsalveti (n=salveti@189.70.96.19) Quit (Remote closed the connection)
  • [07:07:31] * rsalveti (n=salveti@189.70.96.19) has joined #beagle
  • [07:33:30] * n0mis is now known as nomis
  • [08:05:15] * kevlo (n=kevlo@kevlo.org) has joined #beagle
  • [08:06:29] * kevlo (n=kevlo@kevlo.org) has left #beagle
  • [08:25:18] * like2wise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [08:48:58] * RogerMonk (n=a0740758@nat/ti/x-2e25acf330a0bada) has joined #beagle
  • [09:19:42] * hagisbasheruk (n=hagisbas@78.148.135.75) has joined #beagle
  • [10:05:44] * like2wise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  • [10:09:22] * jjg (n=jjg@12.40.200.74) Quit (Read error: 104 (Connection reset by peer))
  • [10:09:52] * trickie|work (n=trickie@basesoft.demon.nl) has joined #beagle
  • [10:15:38] * Genesis (n=Ronan@AMontsouris-153-1-33-226.w90-2.abo.wanadoo.fr) has joined #beagle
  • [10:15:54] <Genesis> yop =)
  • [10:18:13] * like2wise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [10:20:44] <hagisbasheruk> Crofton Crofton|work , got a couple of links for you , http://www.nitehawk.com/w3sz/dttspw3sz.htm ,http://groups.yahoo.com/group/dttsp-linux , http://dttsp.org/wiki/index.php?title=Main_Page
  • [10:21:35] <Crofton|work> so many links, so little time
  • [10:23:45] <hagisbasheruk> :P
  • [10:25:46] <hagisbasheruk> dttsp is what the SDR-Shell program uses so i thought i would pass on some links regarding it
  • [10:30:46] <Crofton|work> there are a lot of potentially very interesting stuff for the beagle
  • [10:30:59] <Crofton|work> what we need is a mobile display :)
  • [10:31:36] <hagisbasheruk> yup , that would be nice
  • [10:32:51] <hagisbasheruk> a cheap photoframe LCD with beagle on the back flashes in my mind right now
  • [10:35:08] * like2wise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  • [11:06:44] * NeuroTech2 (n=NeuroTec@adsl-75-51-176-138.dsl.lsan03.sbcglobal.net) has joined #Beagle
  • [11:07:05] * NeuroTech2 is now known as NeuroTech
  • [11:08:41] * rsalveti (n=salveti@189.70.96.19) Quit (Remote closed the connection)
  • [11:08:57] * rsalveti (n=salveti@189.70.96.19) has joined #beagle
  • [11:10:00] <DJWillis> Crofton|work: A Beagle with a mobile display? Pandora board (well it's fairly close ;-))
  • [11:10:09] <Crofton|work> yeah
  • [11:10:30] <Crofton|work> how does the softrock interface?
  • [11:10:41] <Crofton|work> interesting
  • [11:10:47] <DJWillis> ahh
  • [11:11:13] <NeuroTech> I think Pandora are at a hold point, because of USB port related issues
  • [11:11:20] <NeuroTech> could be wrong though
  • [11:12:16] <Crofton|work> http://www.amqrp.org/kits/softrock40/
  • [11:12:21] <Crofton|work> usb interface
  • [11:12:56] <DJWillis> Crofton|work: I was going to ask, Softrock is that music UART interface, somewhat like the hardware side of MIDIShare?
  • [11:13:19] <Crofton|work> no, interface for the 40m amateur band
  • [11:13:37] <Crofton|work> we could create a portable sdr for amateurs :)
  • [11:14:10] <DJWillis> NeuroTech: We have a similar issue to the Beagle with the USB EHCI PHY that is being worked on.
  • [11:14:17] <koen> softrock used USB and analog audio jacks
  • [11:14:26] <DJWillis> Crofton|work: I should have guessed.
  • [11:15:07] <DJWillis> koen: in that case we have analog audio out and in and the OTG/ECHI (well as far as it goes), crofton, maybe there ius something that could be done.
  • [11:16:16] <Crofton|work> does neuros have audio in?
  • [11:16:26] <Crofton|work> ah, yes
  • [11:16:31] <Genesis> has someone robotic project with such card ?
  • [11:16:39] <Crofton|work> so the softrock is RX ony :(
  • [11:16:59] <Crofton|work> still, running this on Pandora would still be cool
  • [11:17:02] <hagisbasheruk> back later , going to view new house
  • [11:17:16] * DJWillis goes off to read up on this so he does not feel quite so stupid.
  • [11:17:18] <Crofton|work> I wonder if it tunes under 7Mhz so you could listen to SW broadcasters
  • [11:17:35] <jkridner> Genesis: do you mean Beagle? I don't think anyone has done a robotics project with it yet.
  • [11:17:38] <Crofton|work> also, if you can tune a DRM (digital radio mondial) that is a "harder" demod
  • [11:17:55] <Genesis> jkridner : yeap i should look better at spec
  • [11:18:03] <koen> the softrock40 could be enhanched with an usb-audio chip
  • [11:18:04] <Genesis> extension port is poor in gpio
  • [11:18:05] <jkridner> there has been lots of interest in using Beagle for robotics.
  • [11:18:14] <Genesis> which ?
  • [11:18:15] <koen> (or complete device, they are like $0.50 on ebay)
  • [11:18:25] <jkridner> well, each of the serial ports can be converted into GPIO.
  • [11:18:38] <NeuroTech> DJWillis: I get that feeling alot - it passes :)
  • [11:18:46] <jkridner> just about every pin on the device can become GPIO.
  • [11:18:47] <koen> just put an gpio expander on the i2c bus
  • [11:18:57] <Genesis> i should have a look ; this boards interess me because i can do some video stuff on it
  • [11:19:06] <jkridner> you do need to watch the voltages as many are 1.8V
  • [11:19:23] <jkridner> but, it is fairly straight to add level shifters.
  • [11:19:28] <NeuroTech> koen: which USB audio chip are you suggesting?
  • [11:20:04] <koen> NeuroTech: haven't looked at specific chips, but they are dirt cheap
  • [11:20:24] <NeuroTech> k
  • [11:20:59] <Crofton|work> ok, back to the real world for a bit :)
  • [11:21:31] <NeuroTech> mind if I ask a silly question - but would the USB host port issues be solved by putting a USB hub chip directly on it
  • [11:21:51] <NeuroTech> of course it would take extra board space...
  • [11:21:54] <koen> they wouldn't
  • [11:21:55] <jkridner> NeuroTech: no.
  • [11:22:17] <NeuroTech> k.. just making sure
  • [11:22:37] <jkridner> NeuroTech: however, you can make use of the OTG port as a host, if you don't mind using another source of power.
  • [11:24:01] <NeuroTech> jkrider: sounds logical
  • [11:24:40] <NeuroTech> for one of my projects, i'm still looking for a good way to isolate (optically) the USB port from the rest of the circuitry
  • [11:30:23] <NeuroTech> jkrider: Do you ever work with FPGAs? I was thinking about how to make HMDI more compatible with some of the controllers that dont have full compatibility for things like audio and multiple screens
  • [11:34:09] <Crofton> Neuros may need to increase the bounty amount since the dollar continues to weaken :)
  • [11:53:59] * NeuroTech is now known as NeuroTech2
  • [11:54:01] * darkerror05_ (n=darkerro@cpe-74-69-2-202.rochester.res.rr.com) has joined #beagle
  • [11:57:04] <jkridner> NeuroTech2: I have, but it has been several years. Improving the DVI-D connection would be quite welcome.
  • [11:57:22] <jkridner> To keep Beagle costs low, we won't add any additional hardware to the board.
  • [11:57:55] <jkridner> Also, I expect some good single-board computer companies to build upon Beagle and make better platforms for hardware hobbyists.
  • [11:58:11] <NeuroTech2> jkrider: understood..
  • [11:59:10] <Genesis> there is already too hw on board for me
  • [11:59:10] <Genesis> :o)
  • [12:00:37] <NeuroTech2> I thought that the the OMAP apparently can support dual displays - not sure it supports high enough resolution to actually need DVI-D as opposed to DVI-I
  • [12:01:31] <NeuroTech2> Genesis: short of making one of those boards with headers and no connectors - I thought it was basically the minimum config
  • [12:06:30] <Genesis> lot video output stuff, usb only for camera
  • [12:07:02] <hagisbasheruk> <Crofton|work> so the softrock is RX ony :( , there is a RXTX version now ,<Crofton|work> I wonder if it tunes under 7Mhz so you could listen to SW broadcasters , yes via DSS or Xtal board
  • [12:07:22] * JonHunter (n=a0741266@nat/ti/x-9b4b1e0d673c4dd6) has joined #beagle
  • [12:08:28] <NeuroTech2> I think as development progresses, it'll be possible to create a working board with the features you need
  • [12:08:53] <NeuroTech2> as long as the bootloader and BSP work, creating a new board is fairly straight forward
  • [12:10:41] * darkerror05 (n=darkerro@cpe-74-69-7-154.rochester.res.rr.com) Quit (Read error: 110 (Connection timed out))
  • [12:17:41] <hagisbasheruk> Crofton: some softrock mods here http://www.ewjt.com/kd5tfd/sdr1k-notebook/sr40/index.html but latest ones are on the yahoo group for softrock40
  • [12:25:28] * bazbell (n=a0192809@nat/ti/x-2dccdf26a205e1f1) Quit (Remote closed the connection)
  • [12:28:11] * NishanthMenon (n=nmenon@nat/ti/x-2b83a321112df854) has joined #beagle
  • [12:28:12] <hagisbasheruk> a nice DSS kit for $30 is avalible here http://www.amqrp.org/kits/dds60/ , it does 0 to 60Mhz :)
  • [12:32:59] * jkridner|work (n=a0321898@nat/ti/x-f8e24ba3429c2c5e) has joined #beagle
  • [12:33:05] <koen> Crofton: a pure armv7a image compiled by gcc 4.3.1 boots on omap3
  • [12:33:17] <Crofton> awesome
  • [12:33:21] <Crofton> is that in .dev?
  • [12:33:27] <koen> yes
  • [12:33:32] <Crofton> I supposed I could look at what built last night :)
  • [12:35:28] <koen> Crofton: http://rafb.net/p/0PiHvc78.txt
  • [12:36:10] <Crofton> ah, so .dev still uses arm-1136
  • [12:36:30] <Genesis> hey koen :)
  • [12:36:35] <koen> hey Genesis
  • [12:36:38] <Crofton> does 4.3.1 do neon and vfp?
  • [12:36:47] <koen> it does vfp
  • [12:36:52] <koen> haven't tried neon yet
  • [12:37:03] <koen> 2008q1 doesn't do neon either
  • [12:40:04] * BThompson (n=BThompso@nat/ti/x-eab6c48cbc60ced9) has joined #beagle
  • [12:42:43] <ldesnogu> 2008q1 doesn't do neon unless you vectorize (and in this case it's likely the compiler will end with ICE :))
  • [12:53:25] * rsalveti (n=salveti@189.70.96.19) Quit (Read error: 110 (Connection timed out))
  • [12:55:10] * rsalveti (n=salveti@200.184.118.132) has joined #beagle
  • [13:03:11] * koen_ (n=koen@s55917625.adsl.wanadoo.nl) has joined #beagle
  • [13:04:01] * koen (n=koen@s55917625.adsl.wanadoo.nl) Quit (Nick collision from services.)
  • [13:04:08] * koen_ is now known as koen
  • [13:09:28] * bazbell (n=a0192809@nat/ti/x-87755339cdb47483) has joined #beagle
  • [13:18:35] * Lopoi (n=db@c-66-30-48-54.hsd1.nh.comcast.net) has joined #beagle
  • [13:19:21] * DaQatz (n=db@c-66-30-48-54.hsd1.nh.comcast.net) Quit (Read error: 104 (Connection reset by peer))
  • [13:25:33] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
  • [13:26:12] <Genesis> yo prpplague :)
  • [13:26:28] <prpplague> Genesis: greetings
  • [13:27:40] <Genesis> i wonder i'll recycle my juicebox in a vaova vacu R40 bot
  • [14:05:38] * NeuroTech2 (n=NeuroTec@adsl-75-51-176-138.dsl.lsan03.sbcglobal.net) Quit (Read error: 110 (Connection timed out))
  • [14:33:40] * glp (n=glp@2303ds1-fm.0.fullrate.dk) Quit ("leaving")
  • [14:55:56] <khasim> I dont know if some one from this list is on FBDEV list
  • [14:56:24] <khasim> I have posted a question to handle multiple output panels wiith FBDEV
  • [14:56:30] <hagisbasheruk> I just sent an email to my m8 sweetlilmre who ported a Gp2x gui to the gizmondo http://gmenu2x.sourceforge.net/images/c/ca/Gp2x-gmenu2x.png <-GP2x , Gizmondo Versions-> http://www.gizmondoforums.com/forums/index.php?showtopic=12440&st=160 Hopefully he comes onboard , i think he will as he loves hardware hacking also
  • [14:56:57] <khasim> basically many of our boards have Dual output paths like LCD/DVI/SVIDEO
  • [14:57:16] <khasim> I am looking for a standard way to switch between these dynamically
  • [14:58:14] <khasim> at present we use sysfs entries, which I think is the right way to go forward
  • [15:00:18] <koen> khasim: http://thread.gmane.org/gmane.linux.ports.arm.omap/8709
  • [15:01:16] <koen> the current omapfb driver in git is lacking a lot of features
  • [15:05:28] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [15:09:58] * Olipro (n=Olipro@apn-89-223-209-43.vodafone.hu) has joined #beagle
  • [15:19:02] * trickie|work (n=trickie@basesoft.demon.nl) Quit ("Leaving")
  • [15:26:44] <sakoman> NishanthMenon: alsa SoC output works quite well now that I cleaned up the POPN_SET stuff
  • [15:27:02] <sakoman> Thanks for the clue that led me to look there!
  • [15:27:43] <hagisbasheruk> Nice progress sakoman :)
  • [15:28:36] <sakoman> Only works on EVM right now. I'll do a beagle version as soon as I am happy with EVM
  • [15:30:09] <hagisbasheruk> ignore my ignorance ,whats EVM ,only found out aabout beagle a few days ago
  • [15:31:25] <ldesnogu> hagisbasheruk: http://focus.ti.com/docs/toolsw/folders/print/tmdxevm3503.html
  • [15:31:34] <sakoman> It is the TI 35XX evaluation module
  • [15:31:47] * hagisbasheruk is making wiggler for mobilepro 900c right now :)
  • [15:31:55] <sakoman> Cost is about 10x the beagle price
  • [15:32:15] <hagisbasheruk> ahh .... :)
  • [15:36:25] <Crofton|work> but it has an LCD panel!
  • [15:37:13] <hagisbasheruk> the EMV ?
  • [15:37:23] <Crofton|work> the EVM, yes
  • [15:40:53] <sakoman> all kinds of push buttons, test points, & expansion connectors too :-)
  • [15:42:18] <hagisbasheruk> you found any cheap LCD for the beagle ?
  • [15:44:50] <prpplague> i didn't think the lcd data/control lines were brought out on the beagle
  • [15:45:26] <hagisbasheruk> awe :(
  • [15:47:23] <Crofton|work> we'll need some form of usb or spi/i2c led panel
  • [15:48:33] <Crofton> or just use a monitor
  • [15:49:59] <hagisbasheruk> that mabey a good plan , just embed it all onto the back of my monitor
  • [15:50:26] <hagisbasheruk> then mabey add a touchscreen at some point
  • [15:52:41] <NishanthMenon> sakoman, congrats :) glad to see things moving for you
  • [15:53:09] <sakoman> Yeah, I hate hitting my head on brick walls!
  • [15:53:34] <sakoman> That was such a strange symptom. It just wasn't making sense!
  • [15:53:44] <NishanthMenon> hmm.. i know the feeling..
  • [15:54:08] <sakoman> But taking the pop reduction into account it makes complete sense
  • [15:54:23] <NishanthMenon> one more thing to be careful of is AGC
  • [15:54:34] <sakoman> say more words!
  • [15:54:51] <NishanthMenon> just dumping things as they come to mind.. :( too long since i looked at t2 :(
  • [15:55:45] <NishanthMenon> i dont know how you plan to push this to Tony...
  • [15:56:08] <sakoman> OK, well I'll just watch out for AGC problems :-)
  • [15:56:56] <sakoman> I figured I would start with a RFC on the patch set with both linux-omap and alsa-soc
  • [15:57:39] <NishanthMenon> good luck on the patches.. cya later (i2cdev on u-bootv2 is killing me :( )
  • [15:57:50] <sakoman> thanks again!
  • [15:57:59] <NishanthMenon> you are welcome :)
  • [15:58:03] <Olipro> so, when do we get to fitting an RF GSM chip to this?
  • [15:59:27] <hagisbasheruk> hay Olipro , fancy seeing you here
  • [16:01:42] <hagisbasheruk> won't there be some usb/sdio gsm dongle already out there Olipro ?
  • [16:01:56] <Olipro> no
  • [16:01:59] <Olipro> I doubt it
  • [16:02:20] <Olipro> GSM chips have NDAs up the ying yang
  • [16:03:15] <hagisbasheruk> So whats you plans with the beagleboard anyway ?
  • [16:05:15] * JonHunter (n=a0741266@nat/ti/x-9b4b1e0d673c4dd6) has left #beagle
  • [16:06:29] * robclark (n=robclark@nat/ti/x-35225ed9e09df6d7) has joined #beagle
  • [16:09:42] <hagisbasheruk> there are usb gsm dongles Olipro http://www.rfsolutions.co.uk/acatalog/USB_GSM_Modem.html
  • [16:10:51] <Olipro> oh that's just a modem
  • [16:10:53] <Olipro> it won't make calls
  • [16:12:59] <hagisbasheruk> i am sure there is something out there
  • [16:17:30] <hagisbasheruk> or use skype for phonecalls via gsm modem dongle :P
  • [16:19:10] <sakoman> woo-hoo! audio input works now too
  • [16:19:16] <koen> sakoman: yay!
  • [16:19:27] <hagisbasheruk> :)
  • [16:19:58] <sakoman> just have a mono mic, but I get voice on the left channel
  • [16:20:25] <sakoman> If I could find the right cable I'd connect my ipod to try a stereo signal
  • [16:21:02] <sakoman> grumble ... I dread having to sift though my box of tangled cables :-(
  • [16:25:37] <koen> heh
  • [16:25:55] <koen> I have an abundance of jack<>jack cables
  • [16:26:11] <koen> the blackfin board had 6 of them each
  • [16:32:22] * hagisbasheruk (n=hagisbas@78.148.135.75) Quit (Remote closed the connection)
  • [16:37:34] <sakoman> koen: so is it safe to use gcc 4.3.1 for my builds now?
  • [16:42:14] * ldesnogu_ (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) has joined #beagle
  • [16:44:40] * DJWillis (i=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) Quit (Read error: 104 (Connection reset by peer))
  • [16:45:06] * DJWillis (i=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) has joined #beagle
  • [17:09:03] <Genesis> ++
  • [17:09:09] * Genesis (n=Ronan@AMontsouris-153-1-33-226.w90-2.abo.wanadoo.fr) Quit (Read error: 104 (Connection reset by peer))
  • [17:16:52] * sob0l (n=sobol@devel4.net) Quit (Remote closed the connection)
  • [17:17:04] * sob0l (n=sobol@devel4.net) has joined #beagle
  • [17:30:18] * hagisbasheruk (n=hagisbas@78.148.135.75) has joined #beagle
  • [17:30:25] * RogerMonk (n=a0740758@nat/ti/x-2e25acf330a0bada) Quit (Remote closed the connection)
  • [17:41:48] * gkarabi1 (n=gkarabin@64.212.175.249) has joined #beagle
  • [17:59:51] * gkarabin (n=gkarabin@64.212.175.249) Quit (Read error: 110 (Connection timed out))
  • [18:00:28] * robclark (n=robclark@nat/ti/x-35225ed9e09df6d7) Quit ("Leaving")
  • [18:01:25] <koen> sakoman: RP tested minimal-image on his zoom, and that worked
  • [18:01:43] <koen> sakoman: I haven't tried to boot a kernel yet, so I'm keeping it as local diff atm
  • [18:02:22] * darkerror05_ (n=darkerro@cpe-74-69-2-202.rochester.res.rr.com) Quit (kubrick.freenode.net irc.freenode.net)
  • [18:02:22] * ferret_0567 (n=ferret_0@unaffiliated/ferret0567/x-681468) Quit (kubrick.freenode.net irc.freenode.net)
  • [18:02:22] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) Quit (kubrick.freenode.net irc.freenode.net)
  • [18:02:22] * suihkulokki (i=rvoipio@nat/movial/x-3240b41bdb909ddc) Quit (kubrick.freenode.net irc.freenode.net)
  • [18:02:22] * keesj (n=keesj@ip49-193-210-87.adsl2.versatel.nl) Quit (kubrick.freenode.net irc.freenode.net)
  • [18:02:22] * kulve (i=98860559@emperor.pingu.fi) Quit (kubrick.freenode.net irc.freenode.net)
  • [18:03:33] * darkerror05_ (n=darkerro@cpe-74-69-2-202.rochester.res.rr.com) has joined #beagle
  • [18:03:33] * ferret_0567 (n=ferret_0@unaffiliated/ferret0567/x-681468) has joined #beagle
  • [18:03:33] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
  • [18:03:33] * keesj (n=keesj@ip49-193-210-87.adsl2.versatel.nl) has joined #beagle
  • [18:03:33] * kulve (i=98860559@emperor.pingu.fi) has joined #beagle
  • [18:03:33] * suihkulokki (i=rvoipio@nat/movial/x-3240b41bdb909ddc) has joined #beagle
  • [18:05:06] * keesj (n=keesj@ip49-193-210-87.adsl2.versatel.nl) Quit (Read error: 104 (Connection reset by peer))
  • [18:05:06] * keesj_ (n=keesj@ip49-193-210-87.adsl2.versatel.nl) has joined #beagle
  • [18:12:46] * hagisbasheruk_ (n=hagisbas@78.148.135.75) has joined #beagle
  • [18:16:57] * like2wise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [18:21:11] * keesj_ is now known as keesj
  • [18:21:31] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) Quit (kubrick.freenode.net irc.freenode.net)
  • [18:21:31] * ferret_0567 (n=ferret_0@unaffiliated/ferret0567/x-681468) Quit (kubrick.freenode.net irc.freenode.net)
  • [18:21:31] * suihkulokki (i=rvoipio@nat/movial/x-3240b41bdb909ddc) Quit (kubrick.freenode.net irc.freenode.net)
  • [18:21:31] * kulve (i=98860559@emperor.pingu.fi) Quit (kubrick.freenode.net irc.freenode.net)
  • [18:21:31] * darkerror05_ (n=darkerro@cpe-74-69-2-202.rochester.res.rr.com) Quit (kubrick.freenode.net irc.freenode.net)
  • [18:22:55] * hagisbasheruk (n=hagisbas@78.148.135.75) Quit (Read error: 110 (Connection timed out))
  • [18:24:28] * ferret_0567 (n=ferret_0@unaffiliated/ferret0567/x-681468) has joined #beagle
  • [18:24:58] * darkerror05_ (n=darkerro@cpe-74-69-2-202.rochester.res.rr.com) has joined #beagle
  • [18:24:58] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
  • [18:24:58] * kulve (i=98860559@emperor.pingu.fi) has joined #beagle
  • [18:24:58] * suihkulokki (i=rvoipio@nat/movial/x-3240b41bdb909ddc) has joined #beagle
  • [18:25:01] * suihkulokki (i=rvoipio@nat/movial/x-3240b41bdb909ddc) Quit (Remote closed the connection)
  • [18:25:04] * suihkulokki (i=rvoipio@nat/movial/x-cb7d3851f087b3fe) has joined #beagle
  • [18:29:46] <Crofton|work> sakoman, is this still current for flashing beagle NAND?
  • [18:29:52] <Crofton|work> http://www.sakoman.net/omap3/flash%20procedure.txt
  • [18:31:52] <Crofton|work> bother, I have u-boot 1.3.2, where is 1.3.3 for mmc?
  • [18:33:04] <koen> Crofton|work: http://www.sakoman.net/omap3/beagle/u-boot-beagleboard-git-r3.bin
  • [18:33:12] <sakoman> Croftton: that's what I use
  • [18:33:15] <Crofton> ok
  • [18:33:31] <Crofton> I just need to get 1.3.3 on my mmc card first
  • [18:34:10] * traeak (n=bolsen@208.42.251.146) has joined #beagle
  • [18:34:11] <koen> I need a faster buildmachine
  • [18:34:22] <traeak> woah busy here
  • [18:34:27] <Crofton|work> you need more things to do while you are building :)
  • [18:34:35] <mru> still no neon support in gcc 4.3.1 :-(
  • [18:34:49] <koen> mru: there is neon support in 4.3.1
  • [18:34:58] <mru> theoretically, yes
  • [18:35:02] <koen> it will probably ICE the compiler, though
  • [18:35:04] <mru> the compiler crashes if it's enabled
  • [18:35:21] <koen> mru: http://gcc.gnu.org/bugzilla/
  • [18:35:29] <mru> I don't call ICE support
  • [18:35:46] <mru> I've less than good experience with reporting gcc bugs
  • [18:36:02] <mru> ffmpeg tends to trigger quite a few
  • [18:36:33] <Crofton|work> we need to start reporting stuff
  • [18:36:48] * darkerror05__ (n=darkerro@cpe-74-69-2-202.rochester.res.rr.com) has joined #beagle
  • [18:36:57] <traeak> you guys working on hardware then?
  • [18:37:09] <traeak> or just "getting ready"?
  • [18:37:21] <Crofton|work> some of us have Beagle Boards
  • [18:37:28] <Crofton|work> sakoman, even has an EVM
  • [18:37:35] <traeak> i just saw this today
  • [18:37:40] <traeak> been watching the "pandora" guys
  • [18:38:20] * mru is currently trying to find a combination of compiler version and flags that reliably produces working code
  • [18:38:29] <kulve> traeak: here's both the sw people and hw people if I have understood correctly
  • [18:38:29] <kulve> u-boot, kernel and distro guys from the sw side
  • [18:39:03] * Olipro (n=Olipro@apn-89-223-209-43.vodafone.hu) Quit (Read error: 110 (Connection timed out))
  • [18:39:04] * darkerror05_ (n=darkerro@cpe-74-69-2-202.rochester.res.rr.com) Quit (Read error: 110 (Connection timed out))
  • [18:39:47] <Crofton|work> xloader.bin.ift is what goes to NAND?
  • [18:40:27] <traeak> btw, what x86 processor setup does this board best compare with?
  • [18:40:29] <traeak> just curious
  • [18:41:00] <mru> I've never heard of an x86+dsp setup
  • [18:41:10] <mru> besides, arm is fun, x86 is not
  • [18:41:12] <Crofton|work> you would need to speak to ARM/TI marketing
  • [18:41:15] <traeak> exactly
  • [18:41:22] <Crofton|work> I'm sure they can make something up :)
  • [18:41:27] <mru> ;-)
  • [18:41:27] <traeak> i am a linux guy first and foremost
  • [18:41:38] <traeak> :-p
  • [18:44:32] <koen> Crofton|work: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36500
  • [18:45:01] <koen> mru: try csl 2007q3
  • [18:45:10] <ldesnogu_> mru: how do you want to use NEON with gcc? I think the best way at the moment is to use intrinsics (or even assembler)
  • [18:45:43] <mru> koen: yes, I finally managed to build the csl compiler the way I wanted it
  • [18:45:52] <mru> ldesnogu_: I'd be using assembler
  • [18:45:57] <Crofton> that should be easy for them to reproduce :)
  • [18:46:02] <mru> I don't trust gcc intrinsics
  • [18:46:07] <ldesnogu_> :)
  • [18:46:22] <ldesnogu_> I think it's the safest way
  • [18:46:24] <koen> Crofton: Funny how helloworld contains floats ;)
  • [18:46:29] <mru> but I still need to pass -mfpu=neon or the assembler will choke on neon instructions
  • [18:46:46] <mru> and -mfpu=neon crashes stock gcc
  • [18:46:47] <ds2> don't the new SSE3 stuff have a MAC? and doesn't that sort of qualify it as DSP capable?
  • [18:46:51] * ghaz`` (n=kvirc@e178066172.adsl.alicedsl.de) has joined #beagle
  • [18:46:58] <ldesnogu_> can't you just pass -mcpu=cortex-a8 to the assembler only?
  • [18:47:03] <mru> ds2: not really
  • [18:47:45] <mru> ldesnogu_: gcc inserts some assembler directives based on command line flags
  • [18:48:09] <mru> notably it places ".fpu vfp" or ".fpu neon" at the start of the file
  • [18:48:17] <ldesnogu_> using directly gas is no option?
  • [18:48:24] <mru> sometimes
  • [18:48:53] <mru> for large functions like idct that's how I usually do it
  • [18:49:16] <mru> but ffmpeg has lots of little macros with assembler optimisations
  • [18:50:00] <koen> sakoman: gcc 4.3.1 still miscompiles the kernel :(
  • [18:50:41] * hagisbasheruk (n=_2M0UFO_@78.148.135.75) has joined #beagle
  • [18:51:33] <mru> koen: too bad, but thanks for trying
  • [18:52:57] <Crofton> can we read files from mmc with u-bbot?
  • [18:53:07] <Crofton> loading via kermit is a bit slow ....
  • [18:53:17] <koen> Crofton: yes we can
  • [18:53:28] <koen> Crofton: how do you think uboot boots the kernel?
  • [18:53:38] <Crofton> true
  • [18:54:08] <traeak> ugh no eth
  • [18:54:08] <Crofton> what does mw.b do?
  • [18:54:13] <Crofton> yeah
  • [18:54:19] <Crofton> for understandable reasons
  • [18:55:10] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) Quit (kubrick.freenode.net irc.freenode.net)
  • [18:55:10] * kulve (i=98860559@emperor.pingu.fi) Quit (kubrick.freenode.net irc.freenode.net)
  • [18:56:30] <Crofton> must be filling with ff
  • [18:56:32] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
  • [18:56:32] * kulve (i=98860559@emperor.pingu.fi) has joined #beagle
  • [18:56:36] <sakoman> yes.
  • [18:56:54] <sakoman> you might want to use linux for flashing the rootfs
  • [18:57:37] <Crofton> I am going to do it the painful way once
  • [18:57:43] * kulve (i=98860559@emperor.pingu.fi) Quit (Read error: 104 (Connection reset by peer))
  • [18:57:47] * kulve (i=35c567d5@emperor.pingu.fi) has joined #beagle
  • [18:57:56] <sakoman> did you figure out how to read files from mmc?
  • [18:58:12] <koen> fatload mmc 0 0x80200000 u-boot-beagleboard-git-r1.bin
  • [18:58:21] <koen> nand unlock
  • [18:58:21] <koen> nand ecc sw
  • [18:58:21] <koen> nand erase 80000 160000
  • [18:58:21] <koen> nand write 0x80200000 80000 160000
  • [18:58:23] <sakoman> there you go!
  • [18:58:38] <sakoman> don't forget mmcinit
  • [18:58:41] <koen> (the nand ecc will probably fail)
  • [18:58:43] <ldesnogu_> mru: I could get inline assembler to recognize NEON instructions this way: ./arm-2008q1/bin/arm-none-linux-gnueabi-gcc -mcpu=cortex-a8 -mfloat-abi=softfp -mfpu=neon -Wall -c test.c
  • [18:58:57] <mru> I know
  • [18:59:09] <mru> but that only works with gcc-csl 2007q3
  • [18:59:11] <ldesnogu_> however I am not sure it will mix with float and double...
  • [18:59:14] <mru> all other versions crash
  • [18:59:19] <sakoman> you only need the nand ecc with the 1.1.4 uboot
  • [18:59:21] <ldesnogu_> it works with 2008q1 too
  • [18:59:36] <koen> 2008.1 crashes with -ftree-vectorize
  • [18:59:57] <ldesnogu_> koen, that's alas a given, no vectorization with 2008q1...
  • [19:00:08] <sakoman> koen: what breaks in the kernel with 4.3.1?
  • [19:00:20] <koen> and 2007q3 generates a segfaulting rootfs
  • [19:00:21] <suihkulokki> I think -ftree-vectorize is not really stable on x86 either
  • [19:00:27] <koen> sakoman: the null pointer in rcu-update
  • [19:00:47] <sakoman> :-( I had high hopes
  • [19:00:49] <koen> suihkulokki: 4.3.1 listed a ton of vectorizer patches, so it's clearly WIP
  • [19:01:03] <traeak> umm....we've been building our high performance stuff with ftree-vectorize on amd64 for a while
  • [19:01:09] <ldesnogu_> koen, anyway if mru wants to use assembler, he has almost no need for vectorization ;)
  • [19:01:13] <traeak> not sure free-vectorize actually *does* anything though
  • [19:01:28] <koen> traeak: it does do something
  • [19:01:33] <ldesnogu_> traeak, wrong it does some things, but the code has to be dead simple
  • [19:01:38] <traeak> nothing compared with the intel compiler
  • [19:01:52] <koen> since it makes gcc crash if it tries to use NEON instructions for vectorizing
  • [19:01:59] <ldesnogu_> ^_^
  • [19:02:27] <traeak> our system has an element of image processing
  • [19:02:30] <koen> it vectorizes: glibc timezone code, freetype and libxcursor
  • [19:02:38] <traeak> seems to be stable running days on 8cpus
  • [19:02:57] <traeak> but yeah..mostly simple stl stuff I guess
  • [19:03:00] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) Quit (kubrick.freenode.net irc.freenode.net)
  • [19:05:06] <mru> stl as in c++ stl? simple?
  • [19:05:20] <traeak> heh
  • [19:05:24] <koen> funny: http://kerneltrap.org/Quote/x86_Is_So_Totally_Dominant+
  • [19:05:43] <koen> I'm willing to bet that there are more ARM machines running linux than x86 systems
  • [19:05:44] <traeak> <iterator> in stl is really the most useful
  • [19:05:51] <traeak> i'd bet so too
  • [19:06:06] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
  • [19:06:46] <traeak> i'm not a fan of x86 either
  • [19:06:57] <traeak> if someone gave me a commercial arm setup with 32 or more cores
  • [19:07:01] <traeak> i'd be waaaay happy
  • [19:07:09] <traeak> of course i'd need double precision
  • [19:07:22] <koen> power6 :)
  • [19:07:36] <traeak> but i can build an 8 core high power amd64 system for $1000
  • [19:07:41] <traeak> find me a power6 like that ?
  • [19:07:54] <traeak> i'm not going to buy one for development :-p
  • [19:08:00] <koen> you can build a 1/8th core power6 system for $1000
  • [19:08:16] <koen> in europe you can even get a burger at BK for $1000
  • [19:08:23] <traeak> heh
  • [19:08:30] <koen> can't*
  • [19:10:04] <traeak> ugh this hardware ref manual is 133 pages
  • [19:11:29] <suihkulokki> traeak: which chapter did you download? <g>
  • [19:11:54] <traeak> hehe...bastard...
  • [19:13:45] * hagisbasheruk (n=_2M0UFO_@78.148.135.75) Quit ("Leaving")
  • [19:18:24] <Crofton> bugger all, no boot from NAND
  • [19:18:27] <Crofton> only 40T
  • [19:19:15] <koen> wrong xloader?
  • [19:19:15] <sakoman> no xload message?
  • [19:19:25] <Crofton> no xload message
  • [19:19:36] <sakoman> sound familiar koen?
  • [19:19:42] <Crofton> x-load.bin.ift
  • [19:20:01] <koen> sakoman: very :)
  • [19:20:27] <sakoman> md5 = 301ece8b7e87bdef210ebb6f592b1f80 ?
  • [19:20:58] <Crofton> that is good
  • [19:21:05] <Crofton> I guess I try reloading it
  • [19:21:10] <Crofton> then booting :)
  • [19:21:16] <mru> that's what I'm using
  • [19:21:20] <Crofton> I don't need to push any buttons?
  • [19:21:31] <sakoman> nope, should just boot
  • [19:21:43] <sakoman> you will need to push buttons for mmc boot now
  • [19:21:47] <Crofton> putting the mmc back in works
  • [19:22:07] <sakoman> that's because your nand xload is bad
  • [19:22:23] <koen> I needed uboot 1.1.4 to write xloader
  • [19:22:24] <traeak> hmm...seems this beagleboard will run as a desktop machine
  • [19:22:24] <sakoman> once it is good you will need to push the user button
  • [19:22:52] <Crofton> Skipping bad block at 0x00060000
  • [19:23:06] <mru> I quite frequently get crc errors when uboot is loading the kernel
  • [19:23:15] <mru> any ideas what might be the cause?
  • [19:23:25] <mru> yay, finally my rootfs is booting again
  • [19:23:29] <mru> broken bash...
  • [19:23:51] <sakoman> koen: so kernel will build with 4.3.1 but die upon boot from a null pointer?
  • [19:23:58] <koen> sakoman: correct
  • [19:24:10] <Crofton> crap, no more 40T
  • [19:24:10] <sakoman> bummer
  • [19:24:44] <traeak> NEON has double precision capability ?
  • [19:25:28] <suihkulokki> traeak: if, you are conserned about float/double crunchin, arm is not really for you
  • [19:25:42] <Crofton> crap
  • [19:25:54] <Crofton> I'm going to take a break and hope things get better
  • [19:25:55] <suihkulokki> arm is about low power consumption, small die size, easy embedding
  • [19:26:08] <Crofton> I need to run over and help a guy with some stuff
  • [19:26:18] <koen> suihkulokki: and being smug about architectures that need power cords :)
  • [19:26:29] <traeak> i know :-p
  • [19:26:33] <suihkulokki> koen: or heat sinks :)
  • [19:26:41] <koen> :)
  • [19:26:48] <traeak> cortex A8 is a movement towards better FP support though
  • [19:27:13] <traeak> always exploring my options guys :-p
  • [19:28:41] <traeak> and i work with a guy who does arm based power plant control systems anyways
  • [19:36:07] <Crofton> I'm worried I had a bad block in the xloader area
  • [19:36:46] <koen> Crofton: first 16k are garanteed good
  • [19:38:18] <Crofton|work> something funky is going on
  • [19:38:23] <Crofton|work> this is annoying
  • [19:38:55] <Crofton|work> feel free to substitute various obscenities for the word "annoying"
  • [19:40:25] <Crofton> ok, I need to run
  • [19:40:42] * koen decides it's rum o'clock
  • [19:40:45] * koen nudges prpplague
  • [19:40:45] <Crofton> hopefully I have done something done like pull out the serial cable
  • [19:40:53] <Crofton> um rum
  • [19:41:00] <prpplague> koen: sounds good to me
  • [19:41:14] <prpplague> koen: after dealing with poky all morning, i'm ready to drink
  • [19:42:05] <Crofton> see, you should have used .dev :)
  • [19:48:32] * ghaz`` (n=kvirc@e178066172.adsl.alicedsl.de) Quit (Read error: 110 (Connection timed out))
  • [20:04:22] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) Quit (kubrick.freenode.net irc.freenode.net)
  • [20:04:22] * koen (n=koen@s55917625.adsl.wanadoo.nl) Quit (kubrick.freenode.net irc.freenode.net)
  • [20:04:22] * nomis (i=simon@vmax.unix-ag.uni-siegen.de) Quit (kubrick.freenode.net irc.freenode.net)
  • [20:04:22] * jkridner (n=jason@c-76-31-18-64.hsd1.tx.comcast.net) Quit (kubrick.freenode.net irc.freenode.net)
  • [20:04:22] * ldesnogu (n=ldesnogu@fw-tnat.cambridge.arm.com) Quit (kubrick.freenode.net irc.freenode.net)
  • [20:06:30] * koen (n=koen@s55917625.adsl.wanadoo.nl) has joined #beagle
  • [20:06:30] * nomis (i=simon@vmax.unix-ag.uni-siegen.de) has joined #beagle
  • [20:06:30] * jkridner (n=jason@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
  • [20:06:30] * ldesnogu (n=ldesnogu@fw-tnat.cambridge.arm.com) has joined #beagle
  • [20:10:13] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
  • [20:16:20] * traeak (n=bolsen@208.42.251.146) has left #beagle
  • [20:20:48] <koen> sweet: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36500
  • [20:22:14] <ldesnogu_> now we have to wait for 4.3.2 :)
  • [20:25:55] <ldesnogu_> it looks like the fix only touches gcc/config/arm/arm.c
  • [20:49:04] <koen> it probably won't fix the kernel build
  • [20:55:14] <mru> I'm getting inexplicable random crashes
  • [20:56:03] <mru> like you'd get from bad memory and such
  • [20:58:07] <mru> any suggestions?
  • [20:58:43] <koen> do a memtest in uboot?
  • [20:58:50] <sakoman> koen: is there a good gui mp3 player for x11-image?
  • [20:58:50] <koen> (takes a few hours)
  • [20:59:02] <koen> sakoman: xmms :)
  • [20:59:18] <sakoman> opkg install xmms?
  • [20:59:25] <koen> that should do the trick
  • [20:59:41] <sakoman> I want to put a nice demo image together for the EVM
  • [21:00:27] <sakoman> midori and xmms ought to be nice
  • [21:00:36] <sakoman> any other suggestions?
  • [21:00:42] <mru> koen: mtest?
  • [21:00:54] <koen> mru: I think that was the one
  • [21:01:03] <koen> sakoman: you could try epiphany instead of midori
  • [21:01:07] <mru> it seems to be doing something
  • [21:01:21] <mru> is there a way to lower the clock speed to see that makes any difference?
  • [21:01:25] <sakoman> koen: I'll try it again. Last time it didn't work
  • [21:02:27] <koen> sakoman: it might need things like icon themes that get dragged in by other stuff I install
  • [21:03:06] <sakoman> could well be. I was impatient and just moved on to something else :-)
  • [21:05:48] <ldesnogu_> gn all
  • [21:05:57] <koen> NEON works now with gcc 4.3.1
  • [21:06:02] <koen> haven't tried a kernel build
  • [21:06:19] <mru> koen: huh?
  • [21:06:27] <mru> it crashed here
  • [21:07:02] <koen> in OE's gcc 4.3.1
  • [21:07:08] <mru> patched?
  • [21:07:11] <koen> yes
  • [21:08:02] <koen> mru: http://gcc.gnu.org/viewcvs?diff_format=h&view=rev&revision=136661
  • [21:08:35] <sakoman> fingers crossed on the kernel build!
  • [21:09:04] <mru> I won't trust gcc 4.3 for yet some time
  • [21:09:20] * ldesnogu_ (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) Quit ("Leaving")
  • [21:09:21] <mru> by 4.3.3 maybe
  • [21:09:38] <mru> it usually takes them until .3 to stabilise things
  • [21:12:17] * BThompson (n=BThompso@nat/ti/x-eab6c48cbc60ced9) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [21:13:03] <koen> nah
  • [21:13:06] <koen> 4.3.4 :)
  • [21:13:48] <mru> yes, 4.3 has so far been much worse than any other series
  • [21:15:24] <koen> that is very true
  • [21:15:40] <koen> well, recent CSL gccs are pretty bad as well
  • [21:16:24] <mru> third-party hacked up gcc tends to be the worst of the lot
  • [21:26:37] * Olipro (n=Olipro@apn-89-223-208-196.vodafone.hu) has joined #beagle
  • [21:27:10] <hagisbasheruk_> hi Olipro
  • [21:43:38] <koen> sakoman: kernel crash is still there :(
  • [21:44:05] <sakoman> :-(
  • [21:44:59] <koen> Backtrace:
  • [21:44:59] <koen> [<c007ec30>] (__rcu_process_callbacks+0x0/0x298) from [<c007eee0>] (rcu_process_callbacks+0x18/0x38)
  • [21:45:23] <koen> CPU: 0 Not tainted (2.6.26-rc5-omap1 #1)
  • [21:45:23] <koen> PC is at __rcu_process_callbacks+0x200/0x298
  • [21:45:23] <koen> LR is at __slab_free+0x88/0x330
  • [21:49:33] <Crofton> can anyone give me any suggestions what to do with a beagle that does not print 40T anymore?
  • [21:51:09] <Crofton> thank you diety
  • [21:51:24] <Crofton> combination button pushing got me a boot from mmc
  • [21:51:28] <mru> koen: that looks all too familiar
  • [21:51:42] <mru> Crofton: what combination?
  • [21:52:20] <mru> mine has x-load and u-boot in nand, and now it refuses to load from mmc even when I push buttons
  • [21:53:20] <Crofton> not sure
  • [21:53:28] <Crofton> I held both down
  • [21:53:33] <Crofton> then released one
  • [21:53:41] <Crofton> then I tried the other combination
  • [21:53:55] <Crofton> this is what I thinnk I did :)
  • [21:54:26] <Crofton> ok, push both down
  • [21:54:49] <Crofton> then release the one on the "inside" of the board
  • [21:55:37] <Crofton> ~praise TI for including a way to recover from a bad NAND flash!
  • [22:00:03] * bazbell (n=a0192809@nat/ti/x-87755339cdb47483) Quit (Remote closed the connection)
  • [22:13:57] <Crofton> Does xloader print anything to the serial port?
  • [22:14:08] <Crofton> it must be getting control, then dying ...
  • [22:14:39] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) has joined #beagle
  • [22:16:23] <Crofton> when I write xloader it says it writes about .5M
  • [22:16:29] <Crofton> that seems a bit large?
  • [22:17:30] <Crofton> when I boot from mmc I see this message
  • [22:17:32] <Crofton> *** Warning - bad CRC or NAND, using default environment
  • [22:18:07] <Crofton> x-load.bin.ift is 9812 bytes
  • [22:20:35] * rsalveti (n=salveti@200.184.118.132) Quit (Read error: 113 (No route to host))
  • [22:24:00] <Crofton> ahhh, working throught he boot by hand "nand read ...." then bootm gets me to
  • [22:24:07] <Crofton> usb usb2: configuration #1 chosen from 1 choice
  • [22:24:07] <Crofton> hub 2-0:1.0: USB hub found
  • [22:24:07] <Crofton> hub 2-0:1.0: 1 port detected
  • [22:24:11] <Crofton> and it stops
  • [22:30:32] * Raven (n=R4v3N@81-208-36-90.ip.fastwebnet.it) has joined #beagle
  • [22:34:29] <sakoman> Crofton: the bad crc error is normal -- you don't have a valid env in the u-boot env partition
  • [22:34:47] <sakoman> after your first saveenv command it will go away
  • [22:35:38] <sakoman> If the nand were programmed during the board mfg process an end user wouldn't see this
  • [22:35:38] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
  • [22:36:06] * Raven (n=R4v3N@81-208-36-90.ip.fastwebnet.it) Quit ("Sto andando via")
  • [22:36:13] <sakoman> You'll need to set up an env and do a saveenv command
  • [22:36:24] <sakoman> The default env is something silly IIRC
  • [22:37:00] <sakoman> My nand flash instructions have a minimal env setup for nand boot at the bottom of the page
  • [22:37:29] * gkarabi1 (n=gkarabin@64.212.175.249) Quit ("Leaving.")
  • [22:37:31] <sakoman> I should edit the u-boot default env and set it to something reasonable in the next build
  • [22:41:35] <Crofton> yeah, I did that by hand and got further
  • [22:41:42] <Crofton> guess I'll try to saveenv
  • [22:42:21] <Crofton> it looks like the kernel is in NAND
  • [22:42:31] <Crofton> but xloader is not loading u-boot ....
  • [22:50:35] * bazbell (n=a0192809@nat/ti/x-ad96c680cfd14300) has joined #beagle
  • [22:55:06] <mru> hmm... using an old u-boot my crashes are gone
  • [22:55:21] <mru> it must be configuring something differently
  • [23:11:06] * darkerror05__ (n=darkerro@cpe-74-69-2-202.rochester.res.rr.com) Quit ("Leaving")
  • [23:11:16] * like2wise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  • [23:23:35] * Olipro_ (n=Olipro@apn-89-223-209-28.vodafone.hu) has joined #beagle
  • [23:24:16] * Olipro (n=Olipro@apn-89-223-208-196.vodafone.hu) Quit (Nick collision from services.)
  • [23:24:50] * Olipro_ is now known as Olipro
  • [23:26:37] * rsalveti (n=salveti@189.70.96.19) has joined #beagle
  • [23:28:12] * hag900c (n=henchman@78.148.135.75) has joined #beagle
  • [23:32:14] * hagisbasheruk_ (n=hagisbas@78.148.135.75) Quit (Remote closed the connection)
  • [23:37:34] <mru> is there a git repo u-boot for beagle somewhere?
  • [23:37:48] <mru> or do I have to hunt for patches?
  • [23:46:04] <jkridner|work> sakoman has set one up.
  • [23:46:25] <mru> any idea where?
  • [23:47:22] <jkridner|work> git://www.sakoman.net/git/u-boot-omap3.git
  • [23:47:43] <jkridner|work> I attempted to make a mirror, but I don't have it setup properly.
  • [23:48:17] <mru> thanks
  • [23:49:22] * hag900c is away: bed time
  • [23:50:17] * rsalveti (n=salveti@189.70.96.19) Quit (Remote closed the connection)
  • [23:50:55] * rsalveti (n=salveti@189.70.96.19) has joined #beagle
  • [23:52:33] <jkridner|work> There are a lot of new nicks in the room tonight...
  • [23:53:18] <jkridner|work> for those of you who don't know me, I'm Jason Kridner and I'm one of the guys that is responsible for Beagle.
  • [23:53:24] <mru> are there even that many boards?
  • [23:53:33] <jkridner|work> If you have questions that don't get answers, let me know.
  • [23:53:46] <jkridner|work> no, there are more people here than there are boards in the public.