• [00:02:41] * ldesnogu_ (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) Quit ()
  • [00:14:21] * methril|lunch (n=methril@135.72.220.87.dynamic.jazztel.es) Quit (Read error: 104 (Connection reset by peer))
  • [00:15:05] * methril|lunch (n=methril@135.72.220.87.dynamic.jazztel.es) has joined #beagle
  • [00:15:14] * methril|lunch (n=methril@135.72.220.87.dynamic.jazztel.es) Quit (Read error: 104 (Connection reset by peer))
  • [00:27:36] * emeb (n=ericb@ip72-223-95-46.ph.ph.cox.net) has joined #beagle
  • [00:38:42] * jeh (n=jeh@vpn-pub2.hard.ware.fi) has joined #beagle
  • [00:47:13] * Crofton (n=balister@65.208.187.35) has joined #beagle
  • [01:06:16] * fulgas (n=fn@a83-132-158-61.cpe.netcabo.pt) Quit ("changing servers")
  • [01:06:55] * FuL|OUT (n=fn@a83-132-158-61.cpe.netcabo.pt) has joined #beagle
  • [01:07:57] * N_MQZ (n=chatzill@201.174.193.110) has joined #beagle
  • [01:12:50] <N_MQZ> Hello all
  • [01:14:19] <N_MQZ> quick question, what linux distro is everyone using? Angstrom? I would like to use the best one! :P
  • [01:23:02] <Crofton> Angstrom is the best :)
  • [01:23:05] * Crofton is biased
  • [01:24:16] <N_MQZ> lemme guess, you worked on the port?
  • [01:25:50] * gregoiregentil (n=zonbu@adsl-71-135-100-69.dsl.pltn13.pacbell.net) has left #beagle
  • [01:43:50] <vlad_> there isn't a "best one"
  • [01:43:54] <vlad_> depends what you want/need
  • [01:50:58] <N_MQZ> USB OTG for ethernet, ssh, drivers for I2C, SPI, Ethernet, no x-org or fancy HDMI output needed
  • [01:55:16] * TAKI (n=Administ@dslb-088-072-233-142.pools.arcor-ip.net) has joined #beagle
  • [01:55:58] * TAK2004 (n=Administ@dslb-088-072-233-142.pools.arcor-ip.net) Quit (Read error: 104 (Connection reset by peer))
  • [02:00:41] * hli_ (i=chaton@vig91-2-82-232-97-149.fbx.proxad.net) has joined #beagle
  • [02:12:34] <vlad_> the majority of what you mentioned is all kernel-provided, and that's largely up to you and is independent of distro
  • [02:12:41] <vlad_> and ssh is pretty generic
  • [02:18:04] <ds2> still got to decide between openssh, dropbear, putty, etc ;)
  • [02:18:56] * hli (i=chaton@vig91-2-82-232-97-149.fbx.proxad.net) Quit (Read error: 110 (Connection timed out))
  • [02:20:38] * TAKI (n=Administ@dslb-088-072-233-142.pools.arcor-ip.net) Quit (Read error: 54 (Connection reset by peer))
  • [02:27:50] * Beagle5 (n=Beagle5@c-69-142-105-149.hsd1.nj.comcast.net) has joined #beagle
  • [02:28:29] <Beagle5> ? about the omap POP chip - is this a standard part one can buy (w/ sdram + nand) or is that special?
  • [02:29:26] <Beagle5> Example: I see the 2 layer PCB layout app note from TI ...
  • [02:29:55] <Beagle5> A 2layer board is simple ... but creating my own POP chip is non-trivial
  • [02:30:45] <Beagle5> and NO TI documentation I can find speaks of on-chip memory like beagle has.
  • [02:31:31] <geist> it's not a single chip
  • [02:31:35] <geist> it's stacked
  • [02:31:44] <geist> so the memory is still external to the omap
  • [02:32:31] <Beagle5> Oh I understand it is a POP (stack chip) but who did the stacking?
  • [02:32:37] <geist> you
  • [02:32:46] <geist> whoever makes the board
  • [02:33:09] <Beagle5> That suddenly makes it nearly impossible for an experimentor to do stuff :-(
  • [02:33:37] <geist> i dunno, i dont see how much harder than it is putting the chip on your board in the first place
  • [02:34:55] <ds2> if you can do bga's...
  • [02:35:14] <geist> right, i would assume you can stack chips as well
  • [02:35:19] <geist> and of course you dont have to get the pop version
  • [02:35:26] <geist> though it might be simpler if you did
  • [02:35:42] <ds2> i think the POP version still supports routing memory through the norml balls
  • [02:35:48] <Beagle5> Yes, and where would one buy the nand & sdram pop components
  • [02:36:03] <geist> probably the same place you buy the regular ones from
  • [02:36:20] <geist> if you want to put together a omap3 board, you have a lot of stuff to buy and get assembled
  • [02:36:25] <ds2> beagle5: contact micron and ask for the P/N from the beagle BOM
  • [02:36:29] <geist> cpu and memory are probably the easiest parts to find
  • [02:36:34] <ds2> it is currently not listed but available if you ask for the specific part
  • [02:36:41] <geist> you'll also need a twl4030, power supplies, connectors, board layout, etc
  • [02:36:43] <ds2> lookat Gerald's responses on the list
  • [02:37:27] <Beagle5> I want an ARM(linux) solution w/ an fpga on it - (ie: something the free web pack works with)
  • [02:38:33] <Beagle5> the beagle looked "close", ie: "just add fpga" but that does not seem to be the case.
  • [02:38:48] <ds2> you can add an FPGA
  • [02:39:40] <Beagle5> yes, but building it becomes complicated in qty 3 to 5 runs
  • [02:40:39] <Beagle5> No IO expansion connector on the beagle...
  • [02:40:55] <ds2> there is an IO expansion connector on the beagle
  • [02:41:09] <Beagle5> what, the MMC and I2C connector ...
  • [02:41:31] <geist> and the 28 pin header on the side
  • [02:41:40] <ds2> yes
  • [02:41:50] <ds2> you can select between GPIO, MMC, I2C, etc
  • [02:43:00] <Beagle5> I saw no bus type interface... or camera type interface
  • [02:43:33] <geist> guess the camera isn't brought out to that
  • [02:43:46] <geist> what do you mean by 'bus type interface'?
  • [02:43:57] <Beagle5> address, data, rd/wr, etc.
  • [02:44:06] <geist> ah, sram style bus
  • [02:44:14] <geist> nope, it's not brought out, that's true
  • [02:44:19] <ds2> for the most part, ARMs do not have an exposed bus
  • [02:44:25] <ds2> so good luck finding it
  • [02:44:32] <geist> well, sram buses they usually do
  • [02:44:39] <geist> but it's not used on this omap design
  • [02:45:05] <geist> and in general they're becoming less used as stuff moves from NOR to nand flash
  • [02:45:08] <ds2> but they are dedicated and not GP... NOR flash support uses something similar
  • [02:45:29] <geist> well, you can put other stuff on it
  • [02:45:34] <ds2> so a SPI interface is probally more appropriate
  • [02:45:41] <Beagle5> no - a spi is slow.
  • [02:45:45] <ds2> there is limited number of chipselects
  • [02:45:52] <ds2> uh... SPI slow? how much data are you move?
  • [02:45:55] <geist> sure, but it's better than nothing
  • [02:46:16] <geist> you could put a device or two, or one uberdevice with your own address selection logic
  • [02:46:27] <geist> you'd use it to drop an ethernet chip or something
  • [02:46:29] <Beagle5> yes, uberdevice=fpga with internal logic
  • [02:46:34] <Beagle5> or yes, even an ethernet
  • [02:46:35] <ds2> using a 40MHz clock, it can do 5Mbyte/sec
  • [02:46:49] <geist> but anyway, you're right. beagle does not bring it out to any headers, so you're kind of hosed
  • [02:47:10] <Beagle5> the 40mhz clock is dubious - there is turn around - and other "protocol like" delays
  • [02:47:14] <geist> you'd have to interface with spi
  • [02:47:33] <ds2> uh...
  • [02:47:35] <geist> spi is pretty minimal, there's not much to it
  • [02:47:35] <Beagle5> Actuall IIS would be better
  • [02:47:40] <ds2> SPI is a shift register
  • [02:48:06] <Beagle5> spi is a shift register - read something - you make decisions you write something
  • [02:48:11] <geist> though you'd probably end up building your own protocol on top of it, and the turnaround time could get out of hand, if you want to do a bunch of little transactions
  • [02:48:11] <ds2> there is no other delays other then software overahead and I believe there is DMA support on the McBSP
  • [02:48:27] <ds2> the overheads would be the same as a parallel bus
  • [02:48:42] <geist> not really ds2. if it was a memory bus the cpu handles the transactions for you
  • [02:48:49] <geist> you just have to access it, and it's handled in n cycles
  • [02:49:01] <geist> versus having to fire up an entire dma transaction, run a couple million lines of code, etc
  • [02:49:25] <ds2> for small stuff like that, I think you can just write to a register
  • [02:49:28] <geist> especially with something like linux, which is not terribly great for stuff like that
  • [02:49:47] <geist> same thing though, you have the entire host software stack to dive through to get the transaction out
  • [02:49:59] <geist> not that it's the end of the world, mind you, it just can add up
  • [02:50:56] <ds2> for decent performance on a parallel bus, you still have issues with waiting
  • [02:51:05] <Beagle5> that 5mb/sec goes away pretty fast - yes, spi works... but - ugh.
  • [02:51:10] <ds2> the ARM core is spinning at 500+ MHz and the bus is a lot slower
  • [02:51:17] <ds2> so...
  • [02:51:25] <geist> yeah, but when it happens it happens then
  • [02:51:42] <geist> not when the next interrupt comes in and you decide the dma is finished and context switch back to the thread, etc etc
  • [02:52:36] <geist> i've seen this sort of thing on the omap3s i2c bus. you have to go poll something 100 times a second or so and pretty substantial latencies start showing up
  • [02:52:41] <geist> lots of thread churn
  • [02:52:45] <geist> it's pretty annoying
  • [02:52:49] <Beagle5> For example - if I wanted to create an arbitrary wave form generator...
  • [02:53:06] <Beagle5> or something like an fpga based logic analizer (slow one, not fast)
  • [02:53:08] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit ()
  • [02:53:44] <Beagle5> Example: the miniLA one on source forge
  • [02:53:54] * _altered (n=Beagle0@208-100-140-231.bendbroadband.com) Quit ()
  • [02:54:06] <ds2> for a LA, I'd just use the GPIO's as is ;)
  • [02:54:26] <Beagle5> gpios are not fast enough, nor are they synced
  • [02:54:52] <ds2> anyways
  • [02:55:08] <ds2> feel free to build a custom board :)
  • [02:55:19] <ds2> or look into the logicPD modules
  • [02:55:22] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [02:55:30] <geist> yeah, i'd do something like that first
  • [02:55:35] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) Quit (Read error: 104 (Connection reset by peer))
  • [02:55:37] <geist> building a custom board is a pretty big task
  • [02:55:59] <Beagle5> I'm familiar with LogicPD - but the weird 500pin connectors to the carrier board - ugh.
  • [02:56:25] <geist> can't imagine it's as big of an ugh as building your own
  • [02:56:49] <Beagle5> You are right, building your own is probably the same
  • [02:56:55] <ds2> FPGAs are nice and all but using one next to be beagle is like putting a 1500W space heater in the same box as a refridgerator
  • [02:58:52] <Beagle5> Thanks for the information gotta go
  • [02:59:20] * Beagle5 (n=Beagle5@c-69-142-105-149.hsd1.nj.comcast.net) Quit ()
  • [03:00:44] * Beagle5 (n=Beagle5@208-100-140-231.bendbroadband.com) has joined #beagle
  • [03:00:44] * Beagle5 is now known as _altered
  • [03:12:52] * emeb (n=ericb@ip72-223-95-46.ph.ph.cox.net) has left #beagle
  • [03:15:03] * tulpe (n=ask@p54AF7495.dip.t-dialin.net) Quit (Read error: 60 (Operation timed out))
  • [03:15:21] * tulpe (n=ask@p54AF7143.dip.t-dialin.net) has joined #beagle
  • [03:15:30] * jeh (n=jeh@vpn-pub2.hard.ware.fi) Quit (Read error: 104 (Connection reset by peer))
  • [03:15:43] * jeh (n=jeh@vpn-pub2.hard.ware.fi) has joined #beagle
  • [03:28:34] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit ()
  • [03:42:44] * emeb_mac (n=ericb@ip72-223-95-46.ph.ph.cox.net) has joined #beagle
  • [04:00:10] * shyam_k (n=user@117.196.143.52) Quit (Remote closed the connection)
  • [04:06:26] * hli_ (i=chaton@vig91-2-82-232-97-149.fbx.proxad.net) Quit (Read error: 104 (Connection reset by peer))
  • [04:14:21] * codepope (n=codepope@ip-217.146.111.161.merula.net) has joined #beagle
  • [04:43:52] * gduncan (n=spec@pool-96-239-44-239.nycmny.east.verizon.net) has joined #beagle
  • [04:49:18] * codepope (n=codepope@ip-217.146.111.161.merula.net) Quit ()
  • [05:45:15] * jsync (n=root@59.160.172.220) has joined #beagle
  • [06:09:46] <calculus> there are things like this with arm+fpga, http://www.embeddedarm.com/products/board-detail.php?product=TS-7800
  • [06:14:45] * rsalveti (n=salveti@189.70.53.187) has joined #beagle
  • [06:19:04] * tobb (i=phikz@user152.77-105-212.netatonce.net) Quit (Read error: 110 (Connection timed out))
  • [06:36:14] * emeb_mac (n=ericb@ip72-223-95-46.ph.ph.cox.net) Quit ()
  • [06:48:16] * codepope (n=codepope@ip-217.146.111.161.merula.net) has joined #beagle
  • [06:50:28] * tobb (i=phikz@user152.77-105-212.netatonce.net) has joined #beagle
  • [06:50:45] * jsync (n=root@59.160.172.220) has left #beagle
  • [07:20:59] * jonnor (n=jon@ti0016a380-2455.bb.online.no) has joined #beagle
  • [07:22:32] * emeb_mac (n=ericb@ip72-223-95-46.ph.ph.cox.net) has joined #beagle
  • [07:39:04] * emeb_mac (n=ericb@ip72-223-95-46.ph.ph.cox.net) Quit ()
  • [07:41:13] * N_MQZ (n=chatzill@201.174.193.110) Quit ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
  • [07:44:34] * nathanm (n=nathan@adsl-69-106-242-112.dsl.pltn13.pacbell.net) has joined #beagle
  • [07:44:54] <nathanm> Anyone know when BeagleBoards will be back in stock?
  • [07:48:46] * methril|gone is now known as methril
  • [08:04:33] * Curtis (i=3edc0622@gateway/web/ajax/mibbit.com/x-75c67bf0747a6850) has joined #beagle
  • [08:06:18] * xzcvczx (n=simon@gentoo/user/xzcvczx) has joined #beagle
  • [08:06:37] <xzcvczx> does anyone have a copy of spruf98a.pdf?
  • [08:08:10] <Curtis> http://www-s.ti.com/sc/techlit/spruf98 ?
  • [08:10:38] <xzcvczx> ah, thanks i was following the link on muru.com which is dead
  • [08:13:21] <Curtis> no problem
  • [08:15:39] * JerryLi (n=Jerry@211.147.253.143) has joined #beagle
  • [08:16:24] <JerryLi> Hi
  • [08:19:47] <adj_> argh, all components for my beagle daughterboard have arrived from farnell EXCEPT level translators which i won't get until around mid-january :(
  • [08:20:48] <zalassi_> D:
  • [08:20:51] * zalassi_ is now known as zalassi
  • [08:20:56] <xzcvczx> adj_: thats farnell for you :P
  • [08:21:36] * codepope is still having little success getting his beagle to talk on serial.... I get what appears to be noise, in time with where I should be seeing bootup, but only a errored character or two...
  • [08:23:27] <adj_> ah, well, if i can't get those ic's by money i'll just have to order them as a free sample from maxim :)
  • [08:25:02] <jeh> codepope: wrong serial cable?
  • [08:25:37] <codepope> Nope, second cable, checked for correct crossover
  • [08:25:59] <kulve> codepope: did you check the FAQ through=?
  • [08:26:02] <codepope> (and three different serial ports on different devices tried)
  • [08:26:09] <codepope> Yup....
  • [08:26:24] <codepope> I'm suspecting it's a bad dog....
  • [08:31:26] * Psychiatras (i=1000@212.122.84.232) has joined #beagle
  • [08:34:48] * abitos (n=nixgibts@dslb-084-057-136-160.pools.arcor-ip.net) has joined #beagle
  • [08:42:36] * Beagle2 (n=Beagle2@s-inf-pc117.oulu.fi) has joined #beagle
  • [08:43:51] * Beagle2 (n=Beagle2@s-inf-pc117.oulu.fi) Quit (Remote closed the connection)
  • [08:44:14] * codepope notes that it's now crashing minicom.... gets rolled up newspaper ...
  • [08:53:33] <JerryLi> when rev c will come out?
  • [08:55:35] <kulve> JerryLi: I think I heard something about february, by the earliest
  • [08:56:18] <codepope> What's changing in rev c?
  • [08:56:40] <kulve> es3.0 of the cpu and host usb at least
  • [08:59:58] <JerryLi> expecting for lcd support
  • [09:00:52] <Curtis> mid or end Q1
  • [09:01:50] <Curtis> changes: http://elinux.org/BeagleBoard#Revision_C
  • [09:03:54] * Curtis (i=3edc0622@gateway/web/ajax/mibbit.com/x-75c67bf0747a6850) Quit (Remote closed the connection)
  • [09:09:10] * Curtis (i=3edc0622@gateway/web/ajax/mibbit.com/x-ba62de29c2fb182b) has joined #beagle
  • [09:40:30] <JerryLi> exit
  • [09:40:39] * JerryLi (n=Jerry@211.147.253.143) Quit ("Leaving")
  • [09:50:25] * rsalveti (n=salveti@189.70.53.187) Quit (Read error: 113 (No route to host))
  • [10:09:05] * jsync (n=jess@59.160.172.220) has joined #beagle
  • [10:15:36] * geckosen1tor (n=sean@c-24-8-193-190.hsd1.co.comcast.net) has joined #beagle
  • [10:20:45] * jsync (n=jess@59.160.172.220) Quit (Read error: 60 (Operation timed out))
  • [10:35:01] * geckosenator (n=sean@c-24-8-193-190.hsd1.co.comcast.net) Quit (Read error: 110 (Connection timed out))
  • [10:37:26] * jsync (n=jess@59.160.172.220) has joined #beagle
  • [10:43:57] * JoaoPauloB (n=root@galileo.intelbras.com.br) has joined #beagle
  • [10:44:22] <JoaoPauloB> Hello!
  • [10:45:35] <JoaoPauloB> Does any one has examples of how to setup timers? I need set up a periodic timer but i am not familiarized... :/
  • [10:56:52] * geckosenator (n=sean@c-24-8-193-190.hsd1.co.comcast.net) has joined #beagle
  • [10:56:53] * geckosen1tor (n=sean@c-24-8-193-190.hsd1.co.comcast.net) Quit (Read error: 54 (Connection reset by peer))
  • [11:00:29] * ldesnogu_ (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) has joined #beagle
  • [11:23:09] * shriram (i=7aa60de8@gateway/web/ajax/mibbit.com/x-55b181b34ef7d1a5) has joined #beagle
  • [11:30:50] * wiking (n=wiking@huwico/staff/wiking) has joined #beagle
  • [11:43:28] * DJWillis (i=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) has joined #beagle
  • [11:47:52] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [11:56:47] * rsalveti (n=salveti@200.184.118.132) has joined #beagle
  • [11:58:47] <jkridner> good morning all
  • [11:58:56] * dcramer (n=davec@dcdsl.ebox.com) Quit (Read error: 60 (Operation timed out))
  • [11:59:51] <ldesnogu_> gm Jason
  • [12:01:20] * dcramer (n=davec@dcdsl.ebox.com) has joined #beagle
  • [12:01:23] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit ()
  • [12:04:17] * florian_ (n=fuchs@217.146.132.69) has joined #beagle
  • [12:07:08] * florian_ (n=fuchs@217.146.132.69) Quit (Remote closed the connection)
  • [12:10:59] <jeh> JoaoPauloB: look at QTimer
  • [12:11:16] <jeh> uh, no, this was not #qt, sorry :)
  • [12:19:11] * florian_ (n=fuchs@217.146.132.69) has joined #beagle
  • [12:20:05] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [12:20:55] * florian_ is now known as florian
  • [12:34:22] <kulve> koen: noticed the latest EVM patch on the linux-omap? It fixes the red tint on my board (with the omap_rev())
  • [12:38:23] * cbrake_away is now known as cbrake
  • [12:47:33] * and-ri (n=x0103690@nat/ti/x-6284790e7229d140) has joined #beagle
  • [12:47:39] <and-ri> hello there
  • [12:47:47] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit ()
  • [12:48:32] * jkridner|work (n=a0321898@nat/ti/x-842443dfb245e9c8) Quit ("Leaving.")
  • [12:51:12] <and-ri> one question: do anybody know if there already exist a product which use the beagleboard as a part of themself? I mean: is there a commercial product which includes the beagleboard?
  • [12:51:47] <kulve> I don't think such exists yet
  • [12:52:00] <kulve> the beagleboard itself is still unfinished
  • [12:54:20] <and-ri> is there any company who official want to use the beagleboard?
  • [13:01:24] * Beagle1 (n=Beagle1@s-inf-pc117.oulu.fi) has joined #beagle
  • [13:02:01] * Beagle1 is now known as ewerw
  • [13:04:24] <ewerw> hi I would like to connect to the internet with the beagleboard with a help of an USB Ethernet adapter, the network is an open network which uses DHCP so it gives IP address automatically
  • [13:04:54] <ewerw> is there some special settings which I should use?
  • [13:06:45] <ewerw> the /etc/network/interfaces say auto eth0 iface eth0 inet dhcp so it should work
  • [13:07:52] <ewerw> but I do not get an IP address
  • [13:08:17] * FuL|OUT (n=fn@a83-132-158-61.cpe.netcabo.pt) Quit (Read error: 104 (Connection reset by peer))
  • [13:08:43] * JoaoPauloB (n=root@galileo.intelbras.com.br) has left #beagle
  • [13:08:59] * FuL|OUT (n=fn@a83-132-158-61.cpe.netcabo.pt) has joined #beagle
  • [13:09:01] <and-ri> which OS du you use? Angstroem?
  • [13:09:31] <and-ri> which OS do you use? Angstroem?
  • [13:10:28] * John_PB (n=root@galileo.intelbras.com.br) has joined #beagle
  • [13:10:46] <John_PB> Hii
  • [13:10:48] <ewerw> yes
  • [13:12:04] <John_PB> i downloaded the GIT (git://source.mvista.com/git/linux-omap-2.6.git) and after type "make uImage" i get the error: Assembler messages:
  • [13:12:04] <John_PB> Error: unknown architecture `armv7-a'
  • [13:12:04] <John_PB> Error: unrecognized option -march=armv7-a
  • [13:12:04] <John_PB> make[2]: *** [scripts/mod/empty.o] Error 1
  • [13:13:31] <and-ri> ewerw: usually he should get an IP during the bootprocess, so you schould read the bootmessages over the serial Port to find out the reason. But i had the same problem and the reason was a bad USB-HUB, but it could be also the Ethernet-adapter.
  • [13:14:11] <John_PB> ahm, before i type "make omap3_beagle_defconfig"
  • [13:15:07] * jrmuizel (n=jrmuizel@mozca02.ca.mozilla.com) has joined #beagle
  • [13:15:29] <adj_> John_PB: which compiler are you using?
  • [13:16:29] <adj_> you should try e.g. CROSS_COMPILE=arm-none-linux-gnueabi- make uImage
  • [13:17:01] <John_PB> adj_: codesourcery tool chain
  • [13:17:12] * robclark (n=robclark@nat/ti/x-ba43273d1a701ad7) has joined #beagle
  • [13:17:54] <ewerw> and-ri: during the boot process it tries to connect, but it stucks and can not boot up, it is waiting for an IP address I guess so it does not go further
  • [13:19:41] <and-ri> so, if this happend, it think you have some trouble with the USB-HUB or the ETHERNET-Adapter, if you can use the internet with other PC's
  • [13:20:03] <John_PB> adj_: worked, thanks ;P
  • [13:20:24] * methril is now known as methril|lunch
  • [13:20:36] <and-ri> do you use the USB-HUB?
  • [13:20:53] <ewerw> yes I have a hub
  • [13:20:58] <John_PB> adj_: BTW, what is the difference between the GIT and the kernel from
  • [13:21:02] <John_PB> http://www.beagleboard.org/uploads/2.6_kernel_revb-v2.tar.gz
  • [13:21:07] <ewerw> which works fine with the keyboard and the mouse and even with a webcam
  • [13:22:17] <ewerw> and-ri: the disconnect part, happens when I disconnect the ehternet adapter and then the Angstrom finnish the booting fine
  • [13:22:21] <and-ri> ewerw: try to use a direct connection without the hub and look at the boot-process to find out if the probelm is caused by the hub
  • [13:22:58] <ewerw> and-ri: ok I try it
  • [13:23:05] * abitos (n=nixgibts@dslb-084-057-136-160.pools.arcor-ip.net) Quit (Nick collision from services.)
  • [13:23:22] * abitos (n=nixgibts@dslb-084-057-195-204.pools.arcor-ip.net) has joined #beagle
  • [13:23:58] <adj_> John_PB: the kernel from mvista is newer than the one at beagleboard.org (actually GIT is not a kernel but rather a program to fetch current source code)
  • [13:25:21] <adj_> John_PB: also be aware that the kernel from mvista changes constantly and is broken very often (e.g. does not boot, console missing, etc.)
  • [13:26:43] <and-ri> ewerw: whats the status?
  • [13:27:48] <ewerw> and-ri: it booted up but there is no internet
  • [13:28:27] <ewerw> with ifconfig -a there is no eth0
  • [13:29:47] <and-ri> ewerw: another thing, what you can do. connect the Ethernet-adapter to a PC and try to connect to the internet by using only the adapter
  • [13:30:23] <John_PB> adj_: hmm ok.
  • [13:31:23] <and-ri> if it is working with a Desktop pc then there must be another problem
  • [13:31:49] * fulgas81 (n=fn@a83-132-158-61.cpe.netcabo.pt) has joined #beagle
  • [13:32:03] * fulgas81 is now known as fulgas
  • [13:32:04] <ewerw> and-ri: ok I will try that too, at the moment there is no power in the adapter, so none of the leds are blinking or on
  • [13:32:37] * FuL|OUT (n=fn@a83-132-158-61.cpe.netcabo.pt) Quit (Read error: 104 (Connection reset by peer))
  • [13:34:16] <ewerw> and-ri: i think it does not recognized it at all
  • [13:35:20] <and-ri> ewerw: if there is no led on, it could possible that the adpater has no power. do you have a powersupply connectet with the hub?
  • [13:35:41] * Gordon (i=d2d4fc47@gateway/web/ajax/mibbit.com/x-8e6a2ef2dffba487) has joined #beagle
  • [13:36:11] * Gordon is now known as Guest24866
  • [13:37:03] <Guest24866> please anybody tell me briefly how the content under the url http://elinux.org/BeagleBoard/DSP_Howto help me get started on dsp in beagle....
  • [13:38:25] * ewerw (n=Beagle1@s-inf-pc117.oulu.fi) Quit (Read error: 60 (Operation timed out))
  • [13:40:54] * magnet (n=magnet@AMontpellier-259-1-7-173.w90-37.abo.wanadoo.fr) Quit ("Lost terminal")
  • [13:41:20] <and-ri> ewerw: if you use only the usb connection to the hub, maybe the devices like the ethernet adpater do not get enought power because there existing a lot of different USB spezifikation. Some USB-port deliver less than 5V
  • [13:42:02] <and-ri> do you use the powersupply for the HUB?
  • [13:44:31] * John_PB (n=root@galileo.intelbras.com.br) has left #beagle
  • [13:46:30] * JuanG (n=Juan@nat/ti/x-077ad2d920b1eae7) has joined #beagle
  • [13:47:56] * Beagle3 (n=Beagle3@s-inf-pc117.oulu.fi) has joined #beagle
  • [13:48:33] * Beagle3 is now known as ewerw
  • [13:48:38] * jkridner|work (n=a0321898@nat/ti/x-729e2b9343e48207) has joined #beagle
  • [13:49:38] * JuanG (n=Juan@nat/ti/x-077ad2d920b1eae7) has left #beagle
  • [13:52:07] * magnet (n=magnet@AMontpellier-259-1-7-173.w90-37.abo.wanadoo.fr) has joined #beagle
  • [13:52:14] <ewerw> and-ri: I use an external power supply as well, during the boot it finds in /proc/bus/usb it is there but probably it needs too much power so that is why it does not work here it says max 450mA and that USB is capable only 100mA
  • [13:52:43] <Guest24866> i'm getting this error ./ping.out: cannot execute binary file when i put the command ./ping.out to try a ping test....wat can i do?
  • [13:53:57] * dcramer (n=davec@dcdsl.ebox.com) Quit ()
  • [13:54:11] <ewerw> and-ri: in the /proc/bus/usb/devices :Manufacturer=ASIX Elec. Corp. S: Product=AX88178 S: SerialNumber=000003 C: #Ifs= 1 Cfg#= 1 Atr=a0 MxPwr=450mA
  • [13:54:35] <Guest24866> kulve: i'm getting this error ./ping.out: cannot execute binary file when i put the command ./ping.out to try a ping test....wat can i do?
  • [13:58:37] <and-ri> do you recieved the beagle kit demo from TI with the adapter or do you use your own adapter? because i worked with the beagle-Kit adapter and it was working
  • [14:00:47] <and-ri> received
  • [14:02:05] <Crofton> koen, ping
  • [14:02:23] <ewerw> and-ri: I use my own: A-LINK NA1GU USB 2.0 Network adapter
  • [14:03:04] * emeb (n=ericb@ip72-223-95-46.ph.ph.cox.net) has joined #beagle
  • [14:03:33] * FuL|OUT (n=fn@a83-132-158-61.cpe.netcabo.pt) has joined #beagle
  • [14:04:44] <magnet> ewerw: usb can deliver up to 500ma, your device need to register as 'high speed'
  • [14:04:48] <and-ri> ewerw: to be honest, i have no more ideas if you have try it on a desktop pc an it doesn't work. but if you didn't try it till now, maybe it is really an energy supply problem
  • [14:05:02] <magnet> default is low speed, with 100 Ma.
  • [14:06:14] <magnet> if I remenber right, to be recognized as high speed device, there is a resistor to put on one of the data cable, I think wikipedia as more informations.
  • [14:06:44] <ewerw> and-ri: I have tryed it on a PC and it worked there
  • [14:07:05] <Guest24866> crofton: i'm getting this error ./ping.out: cannot execute binary file when i put the command ./ping.out to try a ping test....wat can i do?
  • [14:07:16] <pbrook> magnet: IIRC you have to negotiate addidtional power.
  • [14:07:33] * Beagle3 (n=Beagle3@bzq-219-125-195.static.bezeqint.net) has joined #beagle
  • [14:07:37] * Beagle3 is now known as BeagleDeveloper
  • [14:07:44] <BeagleDeveloper> Hi There,
  • [14:07:52] <BeagleDeveloper> Need some help
  • [14:08:29] <BeagleDeveloper> I'm trying to compaile an hello world application
  • [14:08:53] <BeagleDeveloper> I'm trying to do this using Sourcery G++
  • [14:08:56] <and-ri> so, then it is indeed a powersupply problem. my ony suggestion is, that you search another adpater with lesser energy consumption
  • [14:09:23] <BeagleDeveloper> I'm doing the follow : user@ubuntu810desktop:~/BeagleBoard$ arm-none-linux-gnueabi-g++ -o hello hello.c
  • [14:09:33] <BeagleDeveloper> and i'm getting the following results:
  • [14:10:02] <ewerw> and-ri: yes it seems so
  • [14:10:09] <BeagleDeveloper> home/user/CodeSourcery/Sourcery_G++_Lite/bin/../lib/gcc/arm-none-linux-gnueabi/4.2.1/../../../../arm-none-linux-gnueabi/bin/ld: cannot find -lgcc_s
  • [14:10:11] <Crofton> ping.out was built with what tools?
  • [14:10:16] <AV500> BeagleDeveloper: http://pastebin.com/ is your friend
  • [14:10:41] * JoeyBorn (n=jborn@h-64-105-84-39.chcgilgm.dynamic.covad.net) Quit (Read error: 110 (Connection timed out))
  • [14:11:49] <Guest24866> Crofton: actually the directory is full of .dll64P files...
  • [14:13:39] <ewerw> magnet: when I plug it in it recognizes it as high speed device. usb 1-1.4: new high speed USB device using musb_hdrc and address 7 usb 1-1.4: configuration #1 chosen from 1 choice eth0: register 'asix' at usb-musb_hdrc-1.4, ASIX AX88178 USB 2.0 Ethernet, 00:16
  • [14:13:50] <Guest24866> crofton: i tried 'file ping.out' ...thats working but next './ping.out' is not working...
  • [14:14:23] <Crofton> are you sure you aren't trying to run dsp apps on the arm?
  • [14:14:54] * fulgas (n=fn@a83-132-158-61.cpe.netcabo.pt) Quit (Read error: 110 (Connection timed out))
  • [14:15:42] <magnet> ewerw: oh so it sound good :\
  • [14:15:50] <robclark> BeagleDeveloper: go to wherever the compiler is installed, and 'find . -name libgcc_s.\*`
  • [14:16:00] <robclark> maybe something is wrong with your compiler install?
  • [14:16:11] <Guest24866> Crofton :no ..i'm trying only the DSP apps now....got the dspbridge and binaries...
  • [14:16:37] <BeagleDeveloper> robclark: one sec, checking
  • [14:16:45] <Guest24866> crofton: i'm following the 'DSP How to' in elinux wiki
  • [14:16:46] <and-ri> ewerw: i used the linksys adapter and it works
  • [14:16:47] <and-ri> http://www.linksys.com/servlet/Satellite?c=L_Product_C2&childpagename=US%2FLayout&cid=1115416834080&pagename=Linksys%2FCommon%2FVisitorWrapper&lid=3408033077B01
  • [14:17:20] * NishanthMenon (n=nMenon@nat/ti/x-0268630e548d96c3) has joined #beagle
  • [14:17:22] * NishanthMenon (n=nMenon@nat/ti/x-0268630e548d96c3) Quit (Client Quit)
  • [14:18:27] <robclark> BeagleDeveloper: fyi, this is the layout I have.. you should have the libgcc_s.so in same path relative to your gcc/ld/etc: http://pastebin.com/d2d475a3
  • [14:19:00] <Crofton> I am not familiar with that ..
  • [14:19:11] <ewerw> and-ri: thanks, I might buy one of those
  • [14:20:49] <robclark> BeagleDeveloper: I'm not sure that it should be required, but I noticed that I was using a '-march=armv7a' in my makefiles, and sometimes a -mthumb2... if you have the libgcc_s.so in the right place, you might try adding -march=armv7a
  • [14:22:45] <BeagleDeveloper> robclark: see http://pastebin.com/m437920d0 for my results, I think it's ok
  • [14:23:10] * ClaudeQC (n=claude@bas1-quebec03-1279635609.dsl.bell.ca) has joined #beagle
  • [14:24:05] <robclark> BeagleDeveloper: hmm, seems a bit odd to only see the .so.1's and not the .so's (which should be symlinks to the .so.1's... any chance you are using some filesystem that doesn't support symlinks?
  • [14:24:16] <Guest24866> crofton: will it run in qemu emulator?
  • [14:25:01] <robclark> (but anyways, if the libgcc_s.so is missing, and you only have libgcc_s.so.1, I'm pretty sure that is the problem)
  • [14:28:06] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
  • [14:29:28] <Guest24866> robclark: will QEMU support beagle?
  • [14:29:31] * RogerMonk (n=a0740758@nat/ti/x-25e395e9ceac1ef0) has joined #beagle
  • [14:29:56] <robclark> Guest24866: no idea about QEMU, I've never used it
  • [14:30:09] <Curtis> Guest24866: qemu support arm support beagle
  • [14:30:50] <Guest24866> Curtis: with DSP binaries and bridge drivers, will it support for DsP in beagle?
  • [14:30:57] <Curtis> for example see http://elinux.org/BeagleBoardDebian - using qemu for installing debian
  • [14:30:59] <BeagleDeveloper> robclark: how do you find if the file system supports symlinks?
  • [14:31:03] <Curtis> of course no
  • [14:31:42] <Guest24866> Curtis: why not? please be clear...
  • [14:31:53] <robclark> BeagleDeveloper: well, I think a normal linux type filesystem (ext2/ext3, etc) should support symlinks... windows type filesystems won't. I guess you could try:
  • [14:32:00] <robclark> touch foo; ln -s foo bar
  • [14:32:09] <Curtis> Guest24866: why not what?
  • [14:33:49] <Guest24866> Curtis: tell me why the dsp binaries wont work in qemu with needed drivers....i'm following 'http://elinux.org/BeagleBoard/DSP_Howto'
  • [14:35:04] <Curtis> Guest24866: couse qemu simulates the arm not the dsp
  • [14:35:42] <Guest24866> Curtis: will it work on the actual board?
  • [14:36:04] <Crofton> qemu emultes the arm not the dsp processor
  • [14:36:34] <Crofton> if you do not understand the difference between the arm processor and the dsp processor, you need to do some background reading ......
  • [14:37:11] <Guest24866> Curtis: oh, now i'm not asking about the qemu...i'm askin whether the binaries and drivers work for the actual board?
  • [14:38:07] <Curtis> Guest24866: never tried that
  • [14:38:11] * methril|lunch is now known as methril
  • [14:43:01] <BeagleDeveloper> robclark: it seams to work, I'll try to reinstall
  • [14:43:51] <BeagleDeveloper> robclark: one important thing, is that I manage to compile the kernel with the corrent status
  • [14:43:53] <BeagleDeveloper> strange
  • [14:44:38] <robclark> hmm, I guess the kernel probably doesn't depend on any shared lib's, which might explain why it compiled ok
  • [14:44:49] * shriram (i=7aa60de8@gateway/web/ajax/mibbit.com/x-55b181b34ef7d1a5) Quit ("http://www.mibbit.com ajax IRC Client")
  • [14:45:35] <robclark> if the re-install doesn't help, you could always make a small script to create the missing symlinks... any place where you have a libfoo.so.1 you should have libfoo.so -> libfoo.so.1
  • [14:45:51] * Guest24866 (i=d2d4fc47@gateway/web/ajax/mibbit.com/x-8e6a2ef2dffba487) Quit ("http://www.mibbit.com ajax IRC Client")
  • [14:47:14] * cringthis (n=cringthi@nat/ti/x-08048e3d44e09a14) has joined #beagle
  • [14:49:42] * cringthis (n=cringthi@nat/ti/x-08048e3d44e09a14) has left #beagle
  • [14:56:27] <BeagleDeveloper> robclark\; how do i write this script?
  • [15:00:50] <robclark> BeagleDeveloper: I guess something roughly like: http://pastebin.com/d689b7c47 (run from directory where compiler is installed)
  • [15:01:29] <robclark> I've not tried running it, so it might not be completely bug free, but I guess it is a starting point
  • [15:12:21] * BThompson (n=BThompso@nat/ti/x-ec46b59714d6e8bb) has joined #beagle
  • [15:13:58] * bazbell (n=a0192809@nat/ti/x-be6994318200a68f) has joined #beagle
  • [15:16:45] * florian (n=fuchs@217.146.132.69) Quit (Read error: 104 (Connection reset by peer))
  • [15:18:00] * florian (n=fuchs@217.146.132.69) has joined #beagle
  • [15:18:22] * cbrake (n=cbrake@oh-69-34-21-229.sta.embarqhsd.net) Quit (Remote closed the connection)
  • [15:18:33] * cbrake (n=cbrake@oh-69-34-21-229.sta.embarqhsd.net) has joined #beagle
  • [15:26:44] * ewerw (n=Beagle3@s-inf-pc117.oulu.fi) Quit (Remote closed the connection)
  • [15:41:06] <ldesnogu_> Curtis, qemu doesn't support OMAP3
  • [15:41:21] <ldesnogu_> and qemu doesn't support the C64x+
  • [15:42:12] * methril is now known as methril|gone
  • [15:50:09] <sakoman> jkridner, jkridner|work: ping
  • [15:50:44] * peter|work (i=9d161582@gateway/web/ajax/mibbit.com/x-56b3a74f8f7d8c03) has joined #beagle
  • [15:51:33] <peter|work> anyone running powertop on thier targets?
  • [15:52:18] <peter|work> running powertop produces the error "Error opening terminal: vt102"
  • [15:57:26] * cringthis (n=cringthi@nat/ti/x-d1c062587719362a) has joined #beagle
  • [15:57:42] <pbrook> Sounds like your curses install is busted.
  • [16:04:38] <BeagleDeveloper> robclark: thanks it's working
  • [16:05:00] <robclark> cool
  • [16:05:25] * kozak (n=subbu@117.192.15.222) has joined #beagle
  • [16:05:27] * cringthis (n=cringthi@nat/ti/x-d1c062587719362a) has left #beagle
  • [16:05:47] <BeagleDeveloper> i reinstalled the sorceery.. now i have a simple hello world runs on the beagle board.
  • [16:06:03] <BeagleDeveloper> next step: working with eclipse..
  • [16:07:23] <Crofton|1rssi> sakoman: is ALSA working on overo in recent git?
  • [16:07:35] <Crofton|1rssi> it no longer finds the hw on beagle :(
  • [16:08:13] <sakoman> Crofton|1rssi: I just tried top of tree this morning and noticed that sound got broken
  • [16:08:20] <sakoman> haven't researched it yet
  • [16:08:25] <Crofton|1rssi> great :)
  • [16:08:37] <Crofton|1rssi> shared problems are always better
  • [16:08:51] <sakoman> yeah, something always seems to get broken each rc
  • [16:09:13] <Crofton|1rssi> there have been loads of changes in the twl thing
  • [16:09:24] <Crofton|1rssi> I assume that contributes to the problems
  • [16:10:59] <Crofton|1rssi> I wish I had time to hunt for where things went bad :(
  • [16:11:47] <BeagleDeveloper> question: what should i do in order to use sourcery arm toolchain with eclipse? when i create new C project i have only linuxGCC in the toolchains list
  • [16:13:03] <BeagleDeveloper> i installed sourcery and its compiled simple application using the terminal
  • [16:14:35] * cringthis (n=cringthi@nat/ti/x-4752b805b7f56ca0) has joined #beagle
  • [16:18:05] * cringthis (n=cringthi@nat/ti/x-4752b805b7f56ca0) has left #beagle
  • [16:23:42] <koen> Crofton: pong
  • [16:25:24] * Curtis (i=3edc0622@gateway/web/ajax/mibbit.com/x-ba62de29c2fb182b) Quit ("http://www.mibbit.com ajax IRC Client")
  • [16:30:48] * peter|work (i=9d161582@gateway/web/ajax/mibbit.com/x-56b3a74f8f7d8c03) Quit ("http://www.mibbit.com ajax IRC Client")
  • [16:38:19] * garren (n=garren@dsl-244-213-247.telkomadsl.co.za) has joined #beagle
  • [16:38:22] * denix (n=denys@nat/ti/x-de99858759a7c6fd) has joined #beagle
  • [16:48:07] * Fl1pFl0p (n=flipflop@146.244.179.206) has joined #beagle
  • [16:52:14] * koen spots a seemingly unneeded ifdef in the omap3 cpufreq code
  • [16:53:02] <Crofton|1rssi> anyone have the url for the OMAP3 PM work handy?
  • [16:54:30] <Crofton|1rssi> koen: greetings
  • [16:54:40] <Crofton|1rssi> ASoC on beagle is busted again :(
  • [16:54:50] <Crofton|1rssi> sakoman: has the same issue on overo I think
  • [16:55:08] <Crofton|1rssi> this having a day job is getting old
  • [16:55:38] * koen agrees with the day job comment
  • [16:55:50] <koen> but that's because traffic was gridlocked today
  • [16:56:04] <Crofton|1rssi> heh
  • [16:56:17] <Crofton|1rssi> I thought you all rode bicycles ...
  • [16:56:35] <koen> not if the job is 100km away :)
  • [16:56:44] <Crofton|1rssi> heh
  • [16:56:53] <Crofton|1rssi> you fitting armor again?
  • [16:56:58] <koen> yes
  • [16:57:04] <koen> refitting to be exact
  • [16:57:15] <koen> people decided to loose weight
  • [16:57:32] <Crofton|1rssi> funny how exercise wil do that
  • [16:57:39] * Crofton|1rssi needs more exercise
  • [16:57:41] <koen> 1008 subscribers on the beagle list
  • [16:57:45] * koen -> food
  • [16:57:51] * and-ri (n=x0103690@nat/ti/x-6284790e7229d140) Quit ()
  • [17:05:01] <garren> hi all
  • [17:05:03] * nathanm (n=nathan@adsl-69-106-242-112.dsl.pltn13.pacbell.net) Quit (Read error: 113 (No route to host))
  • [17:05:54] * FuL|OUT (n=fn@a83-132-158-61.cpe.netcabo.pt) Quit (Read error: 104 (Connection reset by peer))
  • [17:06:36] * FuL|OUT (n=fn@a83-132-158-61.cpe.netcabo.pt) has joined #beagle
  • [17:14:08] * _altered (n=Beagle5@208-100-140-231.bendbroadband.com) Quit ()
  • [17:25:51] * jkridner|work (n=a0321898@nat/ti/x-729e2b9343e48207) Quit (Read error: 110 (Connection timed out))
  • [17:25:52] * robclark (n=robclark@nat/ti/x-ba43273d1a701ad7) Quit (Read error: 110 (Connection timed out))
  • [17:28:30] * Geicus (n=chatzill@64.132.1.226) Quit ("ChatZilla 0.9.83-rdmsoft [XULRunner 1.8.0.9/2006120508]")
  • [17:31:16] * gregoiregentil (n=zonbu@adsl-71-135-100-69.dsl.pltn13.pacbell.net) has joined #beagle
  • [17:33:29] * hli (i=chaton@vig91-2-82-232-97-149.fbx.proxad.net) has joined #beagle
  • [17:33:29] * Nikita (i=d2d4fc47@gateway/web/ajax/mibbit.com/x-9cb2634f3b7541cd) has joined #beagle
  • [17:37:32] <kulve> koen: did you notice that there's a patch (and a comment) for the EVM red tint? Worksforme.
  • [17:48:21] <adj_> sakoman: thanks for digging up that mtd patch, it also solved my issue! :)
  • [17:49:22] <koen> kulve: yes, I already applied it to the kernel in OE :)
  • [17:49:37] <koen> kulve: I was meaning to mail you about it :)
  • [17:50:11] <koen> kulve: I was planning to try making the evm work with tomba's dss, but it seems TI is going to do that work
  • [17:51:25] * koen hopes the new DSS will soon be in l-o
  • [17:51:27] <kulve> I tested the tomba's dss on beagle but zuh's xf86-video-omapfb doesn't work with it :/
  • [17:51:35] <kulve> probably something wrong in the zuh's driver
  • [17:51:45] <kulve> it doesn't set mode correctly or something
  • [17:52:02] <koen> it works for me after fbset -depth 16
  • [17:52:08] <kulve> oh
  • [17:52:11] <koen> haven't tried XV, but I do get a picture
  • [17:52:13] <kulve> I have a greep display
  • [17:52:18] <kulve> green
  • [17:52:31] <koen> I did need the proper fbset
  • [17:52:53] <kulve> and if I set the something->bits_per_pixel to 16 before FBIOPUT, I don't need use the fbset..
  • [17:53:02] <kulve> I used the proper one too
  • [17:53:15] * Beagle1 (n=Beagle1@sceapdsd43-72.989studios.com) has joined #beagle
  • [17:53:35] <koen> I start Xorg, fbset -depth 16, killall Xorg, start Xorg
  • [17:53:40] * Beagle1 is now known as _altered
  • [17:53:44] <kulve> koen: do you still have the 16x16 resolution with the EVM touchscreen?
  • [17:53:56] <kulve> I tried to same (I think)
  • [17:53:59] <koen> kulve: no, I appled gravisomethings patch
  • [17:54:16] <kulve> from the linux-omap list?
  • [17:54:46] <koen> yes
  • [17:54:48] <koen> kulve: http://gitweb.openembedded.net/?p=openembedded.git;a=commit;h=2bfbd7904f474550844a5c223d2bdf081d269a50
  • [17:55:08] <koen> kulve: I do have scaling issues with xorg and tslib
  • [17:55:38] <koen> Xorg seems to multiply the x coord with 4/3 and y with 3
  • [17:56:07] <kulve> if I get the ts working I can check if I get the same
  • [17:56:11] * jkridner|work (n=a0321898@nat/ti/x-939c2eb0d0b4baf2) has joined #beagle
  • [17:56:18] <kulve> what about eth, do you have that working on evm?
  • [17:56:22] <koen> but that seems to be a general xorg <> tslib interaction, the same problem exists on the openmoko phones
  • [17:56:32] <koen> yes, eth works with the old 1.1.4 uboot
  • [17:56:44] <kulve> ok
  • [17:57:02] <kulve> Maybe I could get a working evm kernel on linux.onarm.com finally :)
  • [17:57:05] <koen> I'd rather uses sakomans tree, but that didn't have eth working last I checked
  • [17:57:06] * valhalla (n=valhalla@81-174-21-31.dynamic.ngi.it) has joined #beagle
  • [17:57:59] <koen> drat
  • [17:58:08] <koen> only 16 minutes of battery left in this laptop
  • [17:58:25] * koen pokes people to test the dsp_pll uImage mentioned on the beagle list
  • [17:58:38] <Ikarus> time to boot up another laptop ? :P
  • [17:58:54] <kulve> yeah, sakoman's tree would be nicer, but I'm happy with the old 1.1.4 u-boot too
  • [17:58:54] <koen> or locate the psu
  • [17:59:12] <koen> I'm hitting the i2c bug in 1.1.4 pretty often
  • [17:59:34] <koen> it decides to decrease a clock divider with a value of '1'
  • [17:59:43] * bazbell (n=a0192809@nat/ti/x-be6994318200a68f) Quit (Remote closed the connection)
  • [17:59:59] <koen> and a divider of '0' doesn't work to well
  • [18:00:00] <kulve> what's the i2c bug? Or it's symptom?
  • [18:00:21] <koen> I2C_STAT something something
  • [18:00:53] <koen> on the beagle it's fatal, on the evm you just get funny runtime behaviour
  • [18:02:04] <kulve> ok
  • [18:02:05] * koen afk now
  • [18:03:05] * Nikita (i=d2d4fc47@gateway/web/ajax/mibbit.com/x-9cb2634f3b7541cd) Quit ("http://www.mibbit.com ajax IRC Client")
  • [18:07:56] * bazbell (n=a0192809@nat/ti/x-e490ed7f271fa9c6) has joined #beagle
  • [18:08:29] * Fl1pFl0p (n=flipflop@146.244.179.206) Quit ()
  • [18:08:42] * bazbell (n=a0192809@nat/ti/x-e490ed7f271fa9c6) Quit (Client Quit)
  • [18:10:52] * Geicus (n=chatzill@64.132.1.226) has joined #beagle
  • [18:13:37] <garren> how do I add programs like gdb-server to be built with OE with my console-image
  • [18:15:17] * florian (n=fuchs@217.146.132.69) Quit (Remote closed the connection)
  • [18:20:40] <koen> add them to IMAGE_INSTALL in the image recipe
  • [18:22:00] * gregoiregentil (n=zonbu@adsl-71-135-100-69.dsl.pltn13.pacbell.net) Quit (Read error: 110 (Connection timed out))
  • [18:22:25] * fulgas59 (n=fn@a83-132-158-61.cpe.netcabo.pt) has joined #beagle
  • [18:22:34] <garren> cool thanks... and how do I set the path for a build of my local kernel instead of downloading... ie so I can add or edit my kernel and deploy
  • [18:22:37] * fulgas59 is now known as fulgas
  • [18:22:48] * FuL|OUT (n=fn@a83-132-158-61.cpe.netcabo.pt) Quit (Read error: 104 (Connection reset by peer))
  • [18:23:36] <koen> why not add patches to the existing kernel?
  • [18:24:12] <garren> well because I'm a n00b and I need to learn all of this first before contributing
  • [18:32:51] * Geicus (n=chatzill@64.132.1.226) Quit (Read error: 104 (Connection reset by peer))
  • [18:34:47] * rsalveti (n=salveti@200.184.118.132) Quit (Read error: 113 (No route to host))
  • [18:55:16] * geckosen1tor (n=sean@c-24-8-193-190.hsd1.co.comcast.net) has joined #beagle
  • [18:56:40] * RogerMonk (n=a0740758@nat/ti/x-25e395e9ceac1ef0) has left #beagle
  • [18:56:41] * zedstar (n=john@fsf/member/zedstar) has joined #beagle
  • [18:59:19] * geckosenator (n=sean@c-24-8-193-190.hsd1.co.comcast.net) Quit (Read error: 60 (Operation timed out))
  • [19:04:32] * none1 (n=Odesus@200.77.100.197) has joined #beagle
  • [19:05:31] * bazbell (n=a0192809@nat/ti/x-2060072e3c664237) has joined #beagle
  • [19:06:58] * JoeyBorn (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net) has joined #beagle
  • [19:14:14] * none1 (n=Odesus@200.77.100.197) has left #beagle
  • [19:19:51] <garren> does everyone build the kernel via OE each time they make changes to it and are testing or how does everyone do that?
  • [19:21:43] <geist> kernel is pretty easy to build standalone
  • [19:22:38] <garren> and then just use it with the rootfs creat by OE?
  • [19:23:49] <geist> yeah
  • [19:24:02] <geist> only issue would be if you have external modules that you'd need to recompile and link
  • [19:24:13] * gduncan (n=spec@pool-96-239-44-239.nycmny.east.verizon.net) Quit ()
  • [19:24:59] <garren> is that the standard way of working on the kernel with the Beagleboard... how do you then get any new patches submitted to OE?
  • [19:28:58] * guillaum1 (n=Guillaum@AMontsouris-153-1-47-150.w90-2.abo.wanadoo.fr) has joined #beagle
  • [19:29:41] * DaQatz (n=db@c-66-30-48-54.hsd1.nh.comcast.net) Quit (Remote closed the connection)
  • [19:31:48] <kulve> garren: I guess the patches should be posted to linux-omap if they are good enough for the kernel
  • [19:34:36] <kulve> garren: I build the kernel "in a normal way", i.e. outside OE
  • [19:34:46] <kulve> and just scp it to the mmc
  • [19:35:36] <kulve> I git clone the kernel, so it's easy to add the patches as separate git commits
  • [19:36:15] <garren> ok cool thats what I wanted to know thanks...
  • [19:36:27] <kulve> I guess you can compile the modules in the normal way too and then just copy the modified module to the beagle
  • [19:36:34] <garren> and then to add new programs like gdbserver I can do that via OE?
  • [19:36:41] <kulve> yeah
  • [19:37:00] <kulve> (I don't use OE, I use "ALIP" http://linux.onarm.com)
  • [19:37:08] <garren> great thanks for the help...
  • [19:37:46] <garren> I thought most people used OE.. how is the advantage to ALIP without trying to start a flame war :-)
  • [19:37:54] <garren> *what is the advantage
  • [19:38:34] <kulve> probably 99% of the people on this channel uses OE
  • [19:38:42] <kulve> it has applications, ALIP doesn't
  • [19:38:53] <kulve> OE cross compiles, ALIP uses Scratchbox for it
  • [19:39:04] * _altered (n=Beagle1@sceapdsd43-72.989studios.com) Quit ()
  • [19:39:25] <garren> ok I see
  • [19:40:44] <kulve> I want to pretend that I'm compiling natively and that's why I use SB
  • [19:41:09] <kulve> SB hides most of the cross compilation issues from me (and introduces few other problems..)
  • [19:41:23] <garren> ok cool. once I understand OE a little more I'll have to look into SB
  • [19:44:12] <garren> how does bitbake know to go into this directory /oe/openembedded/packages/linux/linux-omap2-git/beagleboard to build the kernel and does it applie all patches in that directory or can you specify?
  • [19:44:48] <kulve> it does what the repice tell to do (some .bb file)
  • [19:44:55] <kulve> the .bb file lists the patches too
  • [19:46:52] <garren> yes of course... thanks
  • [19:47:02] * flo_lap (n=fuchs@g227198235.adsl.alicedsl.de) has joined #beagle
  • [19:56:12] * FuL|OUT (n=fn@a83-132-158-61.cpe.netcabo.pt) has joined #beagle
  • [20:07:54] * fulgas (n=fn@a83-132-158-61.cpe.netcabo.pt) Quit (Read error: 110 (Connection timed out))
  • [20:08:25] * valhalla (n=valhalla@81-174-21-31.dynamic.ngi.it) Quit ("Leaving")
  • [20:16:23] * dirk2 (n=dirk@F32ad.f.strato-dslnet.de) has joined #beagle
  • [20:27:18] * felipec (n=felipec@a91-153-251-222.elisa-laajakaista.fi) has joined #beagle
  • [20:29:07] * robclark (n=robclark@nat/ti/x-83607d454aedb72b) has joined #beagle
  • [20:31:01] <garren> cheers all
  • [20:31:03] * garren (n=garren@dsl-244-213-247.telkomadsl.co.za) Quit ("Ex-Chat")
  • [20:32:04] * FuL|OUT is now known as fulgas
  • [20:57:31] * cringthis (n=cringthi@nat/ti/x-e2f0b15f58e51a03) has joined #beagle
  • [20:57:40] * Fl1pFl0p (n=flipflop@146.244.179.138) has joined #beagle
  • [20:58:08] * cringthis (n=cringthi@nat/ti/x-e2f0b15f58e51a03) has left #beagle
  • [21:00:52] * dirk2 (n=dirk@F32ad.f.strato-dslnet.de) has left #beagle
  • [21:08:47] * RyoKimball (n=Ryo_Kimb@75-121-108-124.dyn.centurytel.net) has joined #beagle
  • [21:11:20] <RyoKimball> Hey!
  • [21:11:20] <RyoKimball> Q: How hard is it for a long-time PC (Linux experienced...) user (not so much a programmer, though) to port programs and/or system apps to ARM-based stuff, like... well, the Beagle Board.
  • [21:14:55] <suihkulokki> if you have compiled programs before, it shouldn't be big challenge
  • [21:18:57] <RyoKimball> So it's just a matter of compiling with an ARM-compiler? Where might I find one? (Must I use an ARM emulator or can I just compile from GNU compiler to ARM?)
  • [21:19:17] <RyoKimball> (Sorry if that's too much at once...)
  • [21:22:53] <adj_> RyoKimball: I consider myself as a beginner when it comes to C/Linux programming and beagle is my first cross-compile target
  • [21:23:44] <adj_> still i have not encountered any major issues and things are mostly working in a way i want them to
  • [21:24:14] <geist> RyoKimball: i'd recommend the free one from codesourcery
  • [21:24:23] <mru> version 2007q3
  • [21:24:25] <geist> its a good gcc compiler + linux libraries
  • [21:24:31] <geist> yeah, that one. later versions have some issues
  • [21:24:51] <mru> the latest non-lite version seems to work
  • [21:24:59] <mru> but 2007q3 gives better code
  • [21:25:27] <geist> yeah, we've been using 2007q3 at work for a year now, as far as i know i've only seen one bad codegen problem
  • [21:25:30] <geist> and it was thumb2
  • [21:25:48] <geist> actually that reminds me, i need to narrow that one down and send in a bug report to codesourcery
  • [21:26:48] <mru> ffmpeg built with the latest release is 5% slower than with 2007q3
  • [21:27:10] <mru> and that's with heavy use of hand-written assembler
  • [21:27:29] <mru> so the C code is probably closer to 10% slower
  • [21:27:50] <RyoKimball> Oh, okay. (sorry it took me a moment to notice messages)
  • [21:27:50] <RyoKimball> I am wanting to do some things I know have occurred in Linux but I have no idea how to go about doing it on the Beagle Board and with AnstromOS.
  • [21:27:50] <geist> yeah, though to be fair sometimes with tight code like that all it takes is one or two out of place instructions to foul it all up
  • [21:27:59] <geist> seen some stuff like that with pulseaudio
  • [21:28:00] * RogerMonk (n=a0740758@nat/ti/x-53b494235a20455d) has joined #beagle
  • [21:28:07] <geist> move one line, 4x faster
  • [21:28:34] * JoeyBorn (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net) Quit (Read error: 60 (Operation timed out))
  • [21:28:44] <geist> not that it really gets it off the hook, it just is that sometimes it makes a dramatic difference
  • [21:29:07] <RyoKimball> I'll look into this as soon as I find time (it's already bookmarked). Thanks you all.
  • [21:29:31] <mru> I have a test case where ARM's compiler creates 15x faster code than gcc
  • [21:29:46] <mru> and my hand-written assembler is 3x faster again
  • [21:30:00] <Ikarus> mru: egads
  • [21:30:05] <Ikarus> mru: what kind of code is that
  • [21:30:13] <geist> not surprising
  • [21:30:19] <mru> multiplying two float arrays element by element
  • [21:30:27] <geist> yep
  • [21:30:39] <geist> that's almost precisely what i've seen in pulseaudio
  • [21:30:43] <Ikarus> mru: so matrix multiplications ?
  • [21:30:47] <Ikarus> yup
  • [21:30:49] <geist> since it was doing a volume adjustment, two float arrays
  • [21:31:01] <mru> a[i] *= b[i]
  • [21:31:05] * Ikarus makes a note to watch out for stuff like that
  • [21:31:06] <calculus> dot product
  • [21:31:09] <geist> the compiler just isn't smart enough to do more effective load/store
  • [21:31:16] <mru> dot product would be similar
  • [21:31:29] <geist> even rewriting it in C trying to unroll the loop a bit helps a bunch
  • [21:31:30] <mru> the ARM compiler actually vectorised the loop with neon
  • [21:31:54] <Ikarus> geist: hrm, though the unrolled loop might ruin cache
  • [21:32:01] <geist> but i expect bad code like that, especially since on x86 it isn't that big of a problem
  • [21:32:34] <geist> Ikarus: yep, though on an omap3 it's much nicer
  • [21:32:55] <geist> omap3 kind of blows a lot of the earlier arm optimization tricks out of the water in a lot of cases, due to the L2
  • [21:33:55] <mru> I unrolled it 16 times, simd-processing 4 elements at a time
  • [21:34:26] <geist> meh, i dont expect gcc to be even remotely intelligent with that kind of stuff
  • [21:34:39] <geist> i used to hack on cell processors, and it was basically the same thing
  • [21:34:47] <mru> the trick with floating point is to pipeline the code properly so the long latencies don't bite you
  • [21:34:49] <geist> if you wanna do fast vector code, you gotta write it in intrinsics
  • [21:34:59] <mru> s/intrinsics/assembler/
  • [21:35:02] <geist> same thing
  • [21:35:12] <mru> no, the compiler can screw up intrinsics
  • [21:35:20] <geist> did a good enough job with cell
  • [21:35:23] <mru> and what gcc can screw up, gcc does screw up
  • [21:35:35] <geist> but since there were 128 registers, it didn't have to be that smart with register allocation
  • [21:35:56] <mru> register allocation is a sore spot for gcc
  • [21:36:12] <geist> i'd have to think it's largely tuned for x86
  • [21:36:19] <geist> but i'm not sure what it's failure case is
  • [21:36:26] <mru> yet that's where it falls down the hardest
  • [21:36:44] <geist> anyway, back to work
  • [21:40:22] * _fulgas (n=fn@a83-132-158-61.cpe.netcabo.pt) has joined #beagle
  • [21:40:25] * fulgas (n=fn@a83-132-158-61.cpe.netcabo.pt) Quit (Read error: 104 (Connection reset by peer))
  • [21:42:30] <zalassi> gcc register allocation for sse was pretty bad <_<
  • [21:52:24] * bazbell (n=a0192809@nat/ti/x-2060072e3c664237) Quit ("Leaving.")
  • [21:53:22] * fulgas (n=fn@a83-132-158-61.cpe.netcabo.pt) has joined #beagle
  • [21:55:57] * RyoKimball (n=Ryo_Kimb@75-121-108-124.dyn.centurytel.net) has left #beagle
  • [21:59:20] * BThompson (n=BThompso@nat/ti/x-ec46b59714d6e8bb) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [22:04:44] * _fulgas (n=fn@a83-132-158-61.cpe.netcabo.pt) Quit (Read error: 110 (Connection timed out))
  • [22:04:55] * AV500 (n=av500@p50996ded.dip0.t-ipconnect.de) Quit (Read error: 145 (Connection timed out))
  • [22:07:32] * fulgas is now known as FuL|OUT
  • [22:08:11] * nemequ (n=nemequ@ip68-111-215-155.sd.sd.cox.net) has joined #beagle
  • [22:16:13] * RogerMonk (n=a0740758@nat/ti/x-53b494235a20455d) Quit (Remote closed the connection)
  • [22:20:12] * nathanm (n=nathan@adsl-69-106-242-112.dsl.pltn13.pacbell.net) has joined #beagle
  • [22:21:03] * Fl1pFl0p (n=flipflop@146.244.179.138) Quit (Read error: 145 (Connection timed out))
  • [22:21:05] * RogerMonk (n=a0740758@nat/ti/x-a16f422cd3bbbe39) has joined #beagle
  • [22:23:02] * AV500 (n=av500@p50996ded.dip0.t-ipconnect.de) has joined #beagle
  • [22:36:59] * dcramer (n=davec@static-72-71-192-9.cncdnh.fios.verizon.net) has joined #beagle
  • [22:43:55] * Beagle0 (n=Beagle0@sceapdsd43-72.989studios.com) has joined #beagle
  • [22:43:59] * Beagle0 is now known as _altered
  • [22:57:09] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) has joined #beagle
  • [23:03:10] * abitos (n=nixgibts@dslb-084-057-195-204.pools.arcor-ip.net) Quit (No route to host)
  • [23:08:41] * DaQatz (n=db@c-66-30-48-54.hsd1.nh.comcast.net) has joined #beagle
  • [23:08:53] * DaQatz (n=db@c-66-30-48-54.hsd1.nh.comcast.net) Quit (Read error: 104 (Connection reset by peer))
  • [23:09:57] * DaQatz (n=db@c-66-30-48-54.hsd1.nh.comcast.net) has joined #beagle
  • [23:10:06] * DaQatz (n=db@c-66-30-48-54.hsd1.nh.comcast.net) Quit (Remote closed the connection)
  • [23:11:21] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
  • [23:13:39] * codepope (n=codepope@ip-217.146.111.161.merula.net) Quit ()
  • [23:22:36] * JuanG (n=Juan@nat/ti/x-826ea3de672c2f0f) has joined #beagle
  • [23:22:48] * JuanG (n=Juan@nat/ti/x-826ea3de672c2f0f) has left #beagle
  • [23:25:40] * guillaum1 (n=Guillaum@AMontsouris-153-1-47-150.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
  • [23:25:52] * guillaum1 (n=gl@AMontsouris-153-1-47-150.w90-2.abo.wanadoo.fr) has joined #beagle
  • [23:29:19] * jrmuizel (n=jrmuizel@mozca02.ca.mozilla.com) Quit ()
  • [23:35:28] * JuanG_ (n=Juan@nat/ti/x-9869130187b67169) has joined #beagle
  • [23:37:13] * gduncan (n=spec@static-71-249-182-7.nycmny.east.verizon.net) has joined #beagle
  • [23:49:56] * JoeyBorn (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net) has joined #beagle
  • [23:50:13] * Fl1pFl0p (n=flipflop@146.244.179.18) has joined #beagle
  • [23:53:10] * robclark (n=robclark@nat/ti/x-83607d454aedb72b) Quit ()