#beagle IRC Log on BeagleBoard.org

Join the chat at beagleboard.org/chat

IRC Log for 2008-12-18

Timestamps are in UTC.

[00:00:28] <atin_> its the hsw value I don't understand. how is that calculated from the modeline?
[00:01:05] <atin_> Modeline "1024x768@60" 64.56 1024 1056 1296 1328 768 783 791 807 gives me:
[00:01:07] <torus> oh - I found about the non-aligned reads in the spec. to bad it does not work for ldmia...
[00:01:26] <mru> the modeline tells you active, sync start, sync end, total
[00:01:30] <atin_> hfp 32, hsw 240, hbp 32
[00:01:32] <torus> what a downer..
[00:01:44] <atin_> unless I am not getting what hsw is
[00:01:56] * maelcum (n=quassel@e178168192.adsl.alicedsl.de) has joined #beagle
[00:01:59] <mru> hsw is the sync end - sync start
[00:02:02] <mru> sync width
[00:02:12] * abitos (n=nixgibts@dslb-084-057-157-130.pools.arcor-ip.net) Quit (Read error: 113 (No route to host))
[00:02:13] <atin_> ok, so I get 240
[00:02:23] <mru> correct
[00:02:39] <atin_> but all the numbers in the file are like at best 32, 64 - mine seems way over
[00:03:03] <atin_> let me see something..
[00:03:44] * guillaum1 (n=Guillaum@AMontsouris-153-1-20-180.w86-212.abo.wanadoo.fr) Quit ("Leaving.")
[00:04:07] * guillaum1 (n=gl@AMontsouris-153-1-20-180.w86-212.abo.wanadoo.fr) has joined #beagle
[00:05:32] <atin_> what the heck, let me try it
[00:07:58] * jrmuizel (n=jrmuizel@mozca02.ca.mozilla.com) Quit ()
[00:22:10] <atin_> hey gecko, it cloned successfully
[00:22:13] <atin_> took a very long time
[00:23:42] * Sept (n=bakljg@c-98-240-226-129.hsd1.mn.comcast.net) Quit (Read error: 110 (Connection timed out))
[00:24:18] <atin_> geckosenatore: it looks like he's worked in this area for sure.
[00:24:25] <atin_> geckosenator
[00:27:14] * emeb (n=ericb@ip72-223-90-212.ph.ph.cox.net) has left #beagle
[00:29:24] <geckosenator> yes
[00:29:28] <geckosenator> in omap2 directory
[00:30:04] <geckosenator> atin_: we need the timings to be done with ioctls so you can use fbset
[00:30:18] <geckosenator> atin_: are you able to set the timings at all through sysfs?
[00:30:30] <atin_> I haven't actually built this.
[00:30:38] <atin_> I just downloaded it :)
[00:30:41] <geckosenator> just reading the readme?
[00:30:46] <geckosenator> I'm going to try to clone again
[00:30:57] <atin_> ok. it just took a very long time
[00:31:00] <atin_> but no error.
[00:34:31] <atin_> anyone know which .bb file in packages/linux is the one that is used to build virtual/kernel?
[00:38:47] * torus (n=nils@c130105.adsl.hansenet.de) Quit ()
[00:42:20] * Openfree (n=df@222.65.143.118) Quit (Read error: 60 (Operation timed out))
[00:45:18] <geckosenator> atin_: still cloning
[00:49:17] * jkridner|work (n=a0321898@nat/ti/x-5034db1444ebffe8) Quit (Remote closed the connection)
[00:51:05] <geckosenator> error: Unable to get pack file http://www.bat.org/~tomba/git/linux-omap-dss.git/objects/pack/pack-cb2e352988bfaf793400d41ce05ba0bd92abcec1.pack
[00:51:45] <geckosenator> oh wait
[00:51:50] <geckosenator> my problem is I have no disk space again
[00:52:13] <ds2> using a 8M SD card again? :D
[00:53:47] <geckosenator> well it didn't work on the beagle for a different reason
[00:58:19] * n6pfk (n=mike@c-76-104-40-104.hsd1.va.comcast.net) has joined #beagle
[00:58:56] <n6pfk> Hi, Are the source listings of the boot rom available?
[01:00:55] <ds2> no
[01:01:55] <n6pfk> Is there a list of what the boot rom configures?
[01:02:12] * fulgas is now known as FuL|OUT
[01:02:19] * Beagle6 (n=Beagle6@c-69-142-105-149.hsd1.nj.comcast.net) has joined #beagle
[01:03:07] <ds2> I don't think it configures much
[01:03:20] <ds2> all it does is read X-Loader into SRAM and jump to it
[01:03:45] <n6pfk> There are a few things mentioned in the harware manual.
[01:04:04] <ds2> which one?
[01:04:51] <n6pfk> The beagleboard reference manual.
[01:05:05] <ds2> besides, it is safer to have U-boot do it then to expect much from the bootrom
[01:05:32] <Beagle6> ? Just got a beagle board today, user doc PDF BBSRM_latest.pdf - section 12.2 says - I can use USB (hyperterm) to talk to the board, does not seem to work. Lights light, but nothing else. If i RESET+USR WinXP says found OMAP3439 ... and wants a driver. Is that what i need?
[01:05:57] <n6pfk> Well, my board is supposed to arrive on Monday and I am trying to understand it as much as possible.
[01:06:24] <ds2> are we both talking about the same thing (Boot Rom)?
[01:07:00] <n6pfk> Yes.
[01:07:33] <ds2> okay... unless you are writing a bootloader.... the Bootrom shouldn't matter too much; all it does is act on SYSBOOT's config
[01:08:27] <n6pfk> I like to know all of the details.
[01:08:45] <ds2> that's probally a TI question...
[01:08:58] <n6pfk> True.
[01:09:16] <ds2> Getting USB boot and Serial boot to work reliably with other tools has been close to practicing voodoo
[01:11:23] <n6pfk> I like black arts.
[01:11:50] <ds2> N6... are you still on the west coast?
[01:11:56] <Beagle6> ? Hello - any suggestion about usb+hyperterm? or is this java client not working? -Duane.
[01:12:12] <n6pfk> NO, I am near Richmond, VA.
[01:12:40] <n6pfk> Hi tech black hle.
[01:13:12] <ds2> were you ever on the west coast??????
[01:13:23] <geist> talking about the ham id, i guess
[01:13:28] <ds2> yeah
[01:13:32] <ds2> 6 is west coast
[01:13:32] <n6pfk> Yes, with an n6, where else.
[01:13:41] <n6pfk> Silicon Valley.
[01:14:00] <ds2> wasn't sure if the FCC gave up on the regions since they no longer reassign for moves
[01:14:19] <n6pfk> AE ticket.
[01:14:28] <n6pfk> The hard way.
[01:14:37] <ds2> you know code I take it?
[01:14:56] <n6pfk> Yes, a lot of fun but psk31 is too.
[01:15:39] <n6pfk> Mostly use QRP.
[01:16:01] <ds2> like the 1transistor stuff?
[01:16:22] <n6pfk> What's ur call and qth?
[01:16:36] <ds2> kf6pyk, SanJose, CA
[01:16:54] <n6pfk> No, digital radios.
[01:17:10] <n6pfk> I lived in Milpitas, many years.
[01:17:12] * RogerMonk (n=a0740758@nat/ti/x-71e3f36fcd9ad5e1) has left #beagle
[01:18:51] <n6pfk> Last job there was with INTEL.
[01:19:05] * Beagle6 (n=Beagle6@c-69-142-105-149.hsd1.nj.comcast.net) Quit ()
[01:19:05] <ds2> oh THOSE power guzzling folks ;)
[01:19:39] <n6pfk> I worked on dialogic telephony boards.
[01:22:13] <n6pfk> Before that I was the total test department for Metricom, the Ricochet RF modem, etc.
[01:24:41] <n6pfk> I am retired now and need a new toy to play with.
[01:26:55] <geckosenator> how is intel?
[01:27:09] <n6pfk> Don't know or care.
[01:27:36] <geckosenator> n6pfk: can you make the beagle get online over ham radio?
[01:28:23] <geckosenator> I would like that for worldwide coverage
[01:28:30] <n6pfk> Should be able to do that. There is a software radio project.
[01:29:04] <n6pfk> And the dsp should easily handle digital modes.
[01:29:13] <geckosenator> do you know how much power it would consume
[01:29:29] <geckosenator> to do say 9600 baud to the opposite side of the world from the base station
[01:31:46] <n6pfk> Too much, psk31 is much better for bandwidth and power.
[01:32:43] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
[01:32:58] <geckosenator> oh
[01:33:26] <ds2> mmmmm ricochet modems
[01:34:29] <n6pfk> Freqwuncy hopping at 900 MHz. Early mesh network.
[01:35:03] <ds2> isn't there a ALSApacket modem already?
[01:35:46] <n6pfk> Dont hnow.
[01:35:51] <n6pfk> know
[01:37:23] <n6pfk> Anyway, gotta run. ttul
[01:37:26] <n6pfk> bye
[01:37:47] * n6pfk (n=mike@c-76-104-40-104.hsd1.va.comcast.net) Quit ("Leaving")
[01:45:02] <geckosenator> ds2: for sound or data?
[01:45:38] <geckosenator> I wonder if I can surf the web with psk
[01:45:51] <geckosenator> probably too slow even for text
[01:46:14] * khasim (n=a0393720@192.163.20.231) has joined #beagle
[01:47:10] <geckosenator> but it could do irc :-P
[01:47:25] <Crofton|work> psk31 ir REALLY slow
[01:47:36] <Crofton|work> k4ep btw
[01:49:25] <geckosenator> well it could communicate to a base station
[01:49:42] <geckosenator> and that base station would connect to irc with a bot
[01:50:22] <geckosenator> could do telnet too
[01:54:29] <ds2> geckosenator: data
[01:54:40] <geckosenator> it would be bad for data
[01:54:49] <ds2> not really
[01:54:54] <ds2> no worse then a 9600 link
[01:55:01] <geckosenator> oh, I thought it was 31 baud
[01:55:04] <ds2> even 300bps is usable
[01:55:12] <geckosenator> but 31 is not
[01:55:25] <ds2> details....
[01:56:14] <geckosenator> heh
[02:00:02] * florian (n=fuchs@f048224011.adsl.alicedsl.de) Quit (Read error: 113 (No route to host))
[02:03:08] <mike18> can someone point me to the libc6-dev package online? I don't have networking on my beagle yet and this is timing out: http://www.angstrom-distribution.org/repo/?pkgname=libc6-dev :(
[02:04:21] <geckosenator> why don't you have networking?
[02:04:52] <geckosenator> that should be unpossible
[02:07:38] <mike18> cause I don't have the wireless usb dongle yet
[02:07:41] <mike18> or a wired one ;)
[02:07:44] <geckosenator> then use usb
[02:07:45] <mike18> moving stuff on flash drive
[02:08:03] <geckosenator> that's what I do
[02:08:03] <mike18> how use USB for networking
[02:08:16] <geckosenator> plug usb to your already networked computer
[02:08:25] <geckosenator> and now the beagle is on the network too
[02:08:33] <mike18> that easy eh
[02:08:47] <geckosenator> just make sure you forward packets
[02:09:03] <geckosenator> it's pretty easy, it works quite well
[02:09:44] <geckosenator> iptables -t nat -A POSTROUTING -s 192.168.1.0/24 -o wifi0 -j MASQUERADE
[02:09:49] <geckosenator> echo 1 > /proc/sys/net/ipv4/ip_forward
[02:09:55] <geckosenator> run that on your host computer
[02:10:03] <geckosenator> replace wifi0 with your netowkr device
[02:10:10] <geckosenator> setup ips on both computers so they can ping eachother
[02:10:18] <geckosenator> make the host the beagle's gateway
[02:10:19] <mike18> my host is winxp
[02:10:25] <geckosenator> oh crap
[02:10:37] <geckosenator> that's bad
[02:10:41] <mike18> :(
[02:11:11] <mike18> i can virtual machine linux but wont be true host
[02:11:26] <geckosenator> it might work
[02:11:37] <geckosenator> if you can get usb in the virtal host
[02:11:44] <mike18> ill run a ubuntu vm in a sec
[02:11:56] <geckosenator> the beagle might even show up as an ethernet device
[02:12:05] <geckosenator> in which case it should work
[02:12:29] <mike18> that my friend would be rad
[02:12:37] <geckosenator> it would be horrible
[02:13:03] <geckosenator> it would be like mounting a bike in a car and pedaling it to make it go
[02:13:23] <geckosenator> but feel free to try
[02:20:44] <mike18> flintstones style
[02:21:44] <mike18> maybe I should force myself to use linux for 1 month straight so I can learn it
[02:22:51] <geckosenator> are you using linux on the beagle?
[02:23:12] <mike18> yeah I've got Angstrom running
[02:23:20] <mike18> I know enough in linux to be dangerous that's about it
[02:23:34] <mike18> stupidly dangerous
[02:23:38] <geckosenator> windows can actually damage your brain though
[02:24:18] <mike18> unfortunately I probably couldn't do my job without windows
[02:24:31] <geckosenator> just be careful
[02:24:36] <mike18> hahaha
[02:25:03] <mike18> i do kick ass @ windows tho
[02:25:14] <mike18> sometimes it amazes me how dumb my friends can be
[02:25:37] <atin_> hey geckosenator, did you get the dss kernel built?
[02:25:44] <mike18> "my computer doesn't work" "can you fix it?" ... my favorite lately is http://www.letmegooglethisforyou.com
[02:27:24] <atin_> my change to the omapfb_main.c took, but my monitor still complains about "Not optimum mode, recommended mode 1600x1200 60Hz" stupid thing - I don't want 1600x1200! I want 1024x768!
[02:27:50] <geckosenator> adoes it work?
[02:27:57] <geckosenator> can you see stuff on the monitor?
[02:28:20] * rsalveti (n=salveti@200.184.118.130) Quit (Remote closed the connection)
[02:28:24] <atin_> I didn't build the dss kernel - I built the 2.6.27-r4 right now with my change to omapfb_main.c
[02:28:28] <geckosenator> try running 800x600, it shouldn't be as blurry if the monitor's native res is 1600x1200
[02:28:29] <atin_> and that doesn't go.
[02:28:44] <geckosenator> doesn't go?
[02:28:45] <atin_> it doesnt show anything - just complains about "not optimum mode"
[02:28:49] <geckosenator> oh ok
[02:29:01] <geckosenator> well maybe the problem is the beagle can't drive it
[02:29:05] <atin_> which usually means a config problem with clocks or something
[02:29:11] <geckosenator> since the monitor wants 1600x1200 faster than the beagle can do
[02:29:24] <atin_> I'm trying to run it at 1024x768 right now
[02:29:27] <geckosenator> can you hook the monitor to a normal linux box with framebuffer?
[02:29:32] <atin_> and fbset says its ok
[02:29:42] <geckosenator> but the monitor doesn't work right?
[02:29:50] <geckosenator> even though fbset says 1024x768?
[02:30:01] <atin_> http://pastebin.com/d3881d4ca
[02:30:02] <atin_> yeah
[02:30:11] <geckosenator> ok
[02:30:13] <atin_> but monitor still complains
[02:30:27] <geckosenator> can you plug the monitor into a computer with linux framebuffer?
[02:30:46] <geckosenator> then type "fbset -g x y x y bpp -t t1 t2 t3 t4 t5
[02:30:53] <atin_> I can plug it into a ubuntu box
[02:30:55] <geckosenator> but replace with different values until it works
[02:31:00] <geckosenator> it works on ubuntu?
[02:31:06] <atin_> yeah - there its happy
[02:31:11] <geckosenator> ok
[02:31:14] <mike18> geckosenator: i don't see anything in dmesg | tail about my beagleboard in ubuntu VM :(
[02:31:16] <atin_> on a pc running ubuntu that is.
[02:31:17] <geckosenator> so run fbset on ubuntu
[02:31:23] <geckosenator> and get the timings
[02:31:30] <geckosenator> and I bet you they are different from on the beagle
[02:31:50] <geckosenator> mike18: do you get anything in lsusb when the beagle is plugged in usb?
[02:31:50] <atin_> yes, what I did was use http://www.tkk.fi/Misc/Electronics/faq/vga2rgb/calc.html
[02:31:59] <atin_> to calculate the values to put into omapfb_main.c
[02:32:12] <atin_> I used the modeline: Modeline "1024x768" 64.56 1024 1056 1296 1328 768 783 791 807
[02:32:16] <geckosenator> uh
[02:32:19] <geckosenator> that is xorg right?
[02:32:26] <atin_> which is what my ubuntu machine thinks should be in xconf
[02:32:35] <geckosenator> you need the fbdev timings
[02:32:35] <atin_> yes
[02:32:39] <geckosenator> I don't think they are compatible
[02:32:44] <atin_> let me see.
[02:32:59] <geckosenator> look for a file "fb.modes"
[02:33:00] <atin_> I converted like someone told me above (mru?)
[02:33:05] <mike18> geckosenator: Bus 001 Devie --1: ID ld6b:001 Linux Foundation 1.1 root hub
[02:33:07] <geckosenator> oh
[02:33:13] <geckosenator> yeah you can convert
[02:33:18] <mike18> is the only thing
[02:33:24] <geckosenator> mike18: what about ifconfig?
[02:33:35] <geckosenator> mike18: do you get a network interface that is really the beagle?
[02:33:43] <atin_> yes, mru - but I must have mucked something. let me try the fbset method
[02:33:50] <geckosenator> ok
[02:33:51] <mru> fbset (and fb.modes) uses yet another way of measuring
[02:33:56] <geckosenator> run fbset on ubuntu
[02:34:05] <geckosenator> and see what timings work
[02:34:14] <mru> and fbset doesn't work with l-o git anyway
[02:34:17] <mike18> geckosenator: there's a listing for eth0 which i'm guessing is local ethernet and one for l0 "Link encap: Local Loopback"
[02:34:26] <mru> fbset uses nanoseconds or picoseconds
[02:34:41] <mru> xorg and omapfb internals use pixels
[02:35:00] <geckosenator> ok
[02:35:16] <atin_> oh man :)
[02:35:21] <geckosenator> so you are saying the modeline is right after all
[02:35:41] <atin_> no, the modeline can be converted to the omap settings.
[02:35:51] <atin_> http://www.tkk.fi/Misc/Electronics/faq/vga2rgb/calc.html
[02:35:54] <mru> xorg modelines use absolute pixel numbers
[02:36:12] * TAK2004 (n=Administ@dslb-088-072-200-222.pools.arcor-ip.net) Quit ("Verlassend")
[02:36:13] <geckosenator> the only timging that should really matter is the pixelclock right?
[02:36:15] <atin_> you can enter the xorg modeline into it and it gives back the front porch, back porch and sync pulse values
[02:36:17] <mru> omapfb internally uses lengths of various parts
[02:36:24] <geckosenator> I get the refresh rates
[02:36:43] <mru> atin_: that calculation is simple subtraction
[02:36:54] <atin_> is 64560 too large for a pixel clock?
[02:37:09] <geckosenator> no
[02:37:15] <geckosenator> well, what units?
[02:37:32] <mru> atin_: that's normal for omapfb
[02:37:36] <mru> the unit is kHz
[02:37:48] <atin_> http://pastebin.com/ddf5c107
[02:38:00] <atin_> that's what I have in omapfb_main.c now to test my monitor
[02:38:03] <geckosenator> I think you can do up to 92000
[02:38:16] <atin_> the modeline was Modeline "1024x768" 64.56 1024 1056 1296 1328 768 783 791 807
[02:38:35] <atin_> but darn monitor didn't like it tho the specs say that's the modeline to use.
[02:38:37] <geckosenator> how come fbset reports 44.79 hz?
[02:38:58] <atin_> yeah - that one I don't know - the pixelclock gets changed somehow?
[02:39:01] <mru> the highest pixel clock supported by the chip is 86.5 MHz
[02:39:18] <geckosenator> ok
[02:39:33] <mru> if you try to run it higher the picture becomes unstable
[02:39:54] <atin_> oh, so I could have gone higher? my monitor says the highest pixel clock for 1924x768 is 78.7
[02:39:57] <atin_> Mhz
[02:40:03] <atin_> err, 1024x768
[02:40:07] <geckosenator> then you don't need to go higher
[02:40:11] <geckosenator> than 86.5
[02:40:40] <mru> atin_: what kernel are you using?
[02:40:44] <atin_> guess I can try the higher number.
[02:40:49] <atin_> 2.6.27-r4
[02:40:55] <mru> from where?
[02:41:01] <mru> l-o git directly or OE?
[02:41:01] <atin_> with my edits to omapfb_main.c
[02:41:04] <atin_> OE
[02:41:15] <mru> so you have my patches, good
[02:41:17] <atin_> and then I copied the sources and have been modifying those
[02:41:39] <atin_> (I finally figured out how to just use make directly rather than bitbake if I want to make my own changes)
[02:42:00] <mru> you can't get any arbitrary pixel clock
[02:42:13] <atin_> my monitor spec sheet says:
[02:42:15] <mru> with the new dss2 drivers you will
[02:42:24] <atin_> VESA, 1024 x 768 60.023 75.029 78.700 +/+
[02:42:40] <atin_> for hfreq,vfreq,pixclock
[02:43:12] <atin_> all the other pixclocks are higher than 100, so can't use higher than 1024x768 it seems
[02:43:31] <mru> it depends on how picky the monitor is
[02:44:03] <mru> you won't get 78.7 MHz with the current driver either
[02:44:17] <atin_> ah
[02:44:31] <mru> the OE kernel can give you 432MHz divided by any integer
[02:45:11] <mru> if you ask for 78 you'll probably get 72
[02:45:22] <atin_> is there a newer kernel than 2.6.27-r4 I should be using now?
[02:45:36] <atin_> ok, there is also VESA,1024 x 768 56.476 70.069 75.000
[02:46:48] <atin_> which is a modeline of Modeline "1024x768" 78.36 1024 1056 1352 1384 768 783 792 807
[02:47:12] <atin_> is here a newer kernel than 2.26.7-r4 that has the monitor stuff working b etter?
[02:47:26] <atin_> (why I care I don't know - I am not doing graphics anyway)
[02:47:51] <atin_> just want to make my monitor work :)
[02:48:06] <atin_> it was working until my last git pull
[02:48:10] <atin_> then it stopped working
[02:56:42] <atin_> woohoo!!
[02:56:51] <atin_> this last modeline worked :)
[02:58:49] * geckosenator (n=sean@c-24-8-193-190.hsd1.co.comcast.net) Quit (Read error: 110 (Connection timed out))
[03:09:55] <atin_> I can't get over this little tiny thing running firefox and looking not much different from my ubuntu desktop
[03:12:13] * keesj (n=keesj@ip49-193-210-87.adsl2.static.versatel.nl) Quit (Read error: 60 (Operation timed out))
[03:16:55] * keesj (n=keesj@ip49-193-210-87.adsl2.static.versatel.nl) has joined #beagle
[03:36:56] * atin_ (n=atin@c-71-232-7-71.hsd1.ma.comcast.net) Quit ()
[03:47:34] * _AV500_ (n=av500@p50996ded.dip0.t-ipconnect.de) has joined #beagle
[03:47:52] * emeb_mac (n=ericb@ip72-223-90-212.ph.ph.cox.net) has joined #beagle
[03:58:30] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
[03:58:38] * calculu5 (n=calculus@gentoo/user/calculus) has joined #beagle
[03:59:41] * AV500 (n=av500@p50996ded.dip0.t-ipconnect.de) Quit (Read error: 110 (Connection timed out))
[04:02:27] * RyoKimball1 (n=Ryo_Kimb@c-69-247-180-145.hsd1.ms.comcast.net) has joined #beagle
[04:02:58] <RyoKimball1> Does anyone have a good suggestion for a portable display I can get with/for the Beagle?
[04:03:59] <RyoKimball1> (Even better would be a capacitive touch screen...)
[04:06:32] * RyoKimball1 .oO(is there a command that can tell whether a room is awake or not?)
[04:11:39] * calculus (n=calculus@gentoo/user/calculus) Quit (Read error: 110 (Connection timed out))
[04:13:13] * mrc3_ (n=ddiaz@189.157.115.100) has joined #beagle
[04:13:39] <mrc3_> hello! i'm a new happy user of the beagle
[04:13:59] <Pavlov> nice
[04:14:01] <mrc3_> i have angstrom and poky running on it, but how do i connect a keyboard or mouse? i can't get it to work
[04:16:08] * ScriptRipper (n=martin@opensuse/member/MartinMohring) Quit (Read error: 104 (Connection reset by peer))
[04:22:05] * RyoKimball1 (n=Ryo_Kimb@c-69-247-180-145.hsd1.ms.comcast.net) has left #beagle
[04:27:59] * Beagle7 (n=Beagle7@adsl-69-152-209-101.dsl.fyvlar.swbell.net) has joined #beagle
[04:28:18] <Beagle7> Hello, I have a question about the Beagle Board
[04:29:29] <Beagle7> I need to do some benchmarking on some mathematical functions I have written in C. Can the Beagle Board compile and run C code, or would I have to do it through linux?
[04:30:02] <Beagle7> I need to have a timing function that I can access in my C code, get the clock cycles or something like that
[04:32:05] <mrc3_> Beagle7, you can do that as much as you can do it on a pc
[04:33:19] <mrc3_> you can crosscompile on your host machine (be it windows or linux, though it's a lot easier with linux), or directly in the beagle if you have an sdk (a compiler that runs natively in arm)
[04:34:28] * Beagle8 (n=Beagle8@c-67-180-119-206.hsd1.ca.comcast.net) has joined #beagle
[04:34:38] <Beagle7> so if I compiled my code for the arm, it would be able to run on the beagle board without using an operating system to back it up?
[04:34:41] * Beagle8 is now known as kc8apf
[04:36:35] * straggler77 (n=dehort@h26.102.89.75.static.ip.windstream.net) has joined #beagle
[04:36:40] <kc8apf> I'm working on BeagleBoard support for OpenOCD. To get started, I need the details on the JTAG scan chain and JRC.
[04:37:05] <kc8apf> I've heard there is a document that describes this that is currently unreleased but can be had by asking the appropriate person
[04:37:11] <kc8apf> who would that person be?
[04:37:27] * straggler77 (n=dehort@h26.102.89.75.static.ip.windstream.net) Quit (Client Quit)
[04:39:28] <mrc3_> Beagle7, sure you need an operating system, something that runs that program you compiled
[04:39:38] <mrc3_> there are many freely available options for you to use
[04:40:50] <mrc3_> kc8apf, if you don't get an answer, please try again on morning times (gmt-0600)
[04:41:46] <kc8apf> Would jkridner be the right contact?
[04:44:20] <kc8apf> I'll try back in the morning
[04:44:24] <kc8apf> thanks mrc3
[04:44:26] * kc8apf (n=Beagle8@c-67-180-119-206.hsd1.ca.comcast.net) Quit ()
[04:58:30] * supo (n=supo@s-inf-pc117.oulu.fi) has joined #beagle
[05:03:04] * soman (n=somnath@61.16.248.242) has joined #beagle
[05:20:51] * khasim (n=a0393720@192.163.20.231) has joined #beagle
[05:26:31] <ds2> Pico DLP + laser pointer
[05:35:46] * mrc3_ (n=ddiaz@189.157.115.100) Quit (Read error: 104 (Connection reset by peer))
[05:42:31] * emeb_mac (n=ericb@ip72-223-90-212.ph.ph.cox.net) Quit ()
[05:43:34] * shri (i=3d5fc3cd@gateway/web/ajax/mibbit.com/x-c3235f582253474f) has joined #beagle
[05:44:09] * jsync (n=jess@59.160.172.220) has joined #beagle
[05:45:25] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit ()
[05:52:29] * bazbell (n=a0192809@nat/ti/x-6a4c85addf2b2157) Quit (Remote closed the connection)
[06:07:09] * TehUni (n=cory@c-24-30-35-55.hsd1.ga.comcast.net) has joined #beagle
[06:07:09] * TheUni (n=cory@c-24-30-35-55.hsd1.ga.comcast.net) Quit (Read error: 104 (Connection reset by peer))
[06:08:34] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
[06:19:38] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit ()
[06:19:51] * mrc3 (n=ddiaz@189.157.115.100) has joined #beagle
[06:43:50] <kulve> mrc3: any news about the tiopenmax 0.4?
[06:44:23] <mrc3> kulve, yes: no news
[06:45:22] <mrc3> i will have some more free time myself these days (and now i have a beagle to pet!), so i'll try to gather as many public pieces as possible to form something
[06:45:44] <kulve> I understood that the issue is the non-public pieces..
[06:45:45] <mrc3> i don't think i'll be able to push anyone during the holidays, but we'll get something, i'm sure
[06:47:12] <mrc3> kulve, of course. the audio and video guys were overwhelmed with stuff in the past two crazy months. i just needed those guys to validate the contents before releasing anything
[06:47:55] <mrc3> i was myself way too busy pokyfying everything in the release (meta-ti, as you can see in github)
[06:49:16] <mrc3> kulve, btw, felipec is coming to monterrey tomorrow. are you coming anytime soon? him and ceyusa will be in town, so we can have some good discussions
[06:50:49] <kulve> sorry, no. Is there some official openmax meeting or..?
[06:52:59] <mrc3> not really. we're just a bunch of geeks who happen to be born in monterrey
[06:53:02] <mrc3> (and in ceyusa's case, who worked and still has friends in monterrey)
[06:53:30] <mrc3> i believe beer will be weighing factor in the discussion table
[06:53:35] <kulve> Monterrey seems to be a bit far from Finland just to attend a meeting ;)
[06:55:39] <kulve> I managed to get the gst-openmax working with ti's openmax implementation, but I needed to take the DSP binaries from a non-public TI release. I'm hoping everyting works nicely from the public releases after the tiopenmax 0.4 release (and I understood that felipe had the same issue)
[06:56:13] <mrc3> yes, i heard that too
[06:56:39] <mrc3> kulve, did i show you this before? http://github.com/mrchapp/gst-goo/tree/master (and libgoo is there too)
[06:57:36] <kulve> I did found it some time ago. But didn't actually try it because I got the gst-openmax working with TI's dsp as well..
[06:58:21] <kulve> for the next MWC (in February) I need to get something working again. I'm not sure yet which libs I'll take then..
[06:58:43] <mrc3> very good. at least that is now totally open source and development will continue on public git too
[06:59:09] * robclark (n=robclark@nat/ti/x-bd0944db2140e00e) has joined #beagle
[06:59:12] <kulve> yes, that's a very good thing.
[06:59:48] <mrc3> so, do you have a beagle barking and all?
[07:00:08] <kulve> mrc3: do you have some comparison about the differencies using the TI's DSP stuff through gst-openmax or through gst-goo?
[07:00:32] <kulve> mrc3: yeah, we have couple of beagles at the office and I got the gst-openmax working on top of TI's DSP stuff
[07:01:27] <mrc3> kulve, not really. we haven't tried gst-openmax much since... early this year, i think
[07:01:35] <mrc3> those would be very interesting numbers
[07:02:03] <kulve> I'm mostly interested about the features they provide
[07:02:09] <mrc3> i was asking about getting the beagle working because i just received mine, but can't get a mouse to work with it
[07:02:54] <kulve> I have a power usb hub (and power beagle) and I have all the usb peripherals working that I connect to the hub (mouse, keyboard, gamepad, etc)
[07:03:05] <kulve> s,power,powered,g
[07:03:55] <kulve> for me the tricky parts were to power both the hub and the beagle and to use a mini-a usb cable (those seems to be quite rare)
[07:04:49] <mrc3> the usb cable i have *looks* like a mini-a usb cable. let me try again using an external power source
[07:05:04] <mrc3> so, powering from the usb + having peripherals connected on the usb is a no-no?
[07:05:15] <kulve> the visible differencies between the mini-a and mini-b are quite small
[07:05:37] <kulve> I didn't get the beagle to work in usb host mode without powerin the beagle too
[07:06:45] <kulve> http://en.wikipedia.org/wiki/Universal_Serial_Bus#Types_of_USB_connector
[07:07:23] <kulve> in mini-a cable the pin4 is connected to the ground and in mini-b it's not connected to anything
[07:07:38] <kulve> on the beagle side connector
[07:14:43] <mrc3> no luck
[07:14:53] <mrc3> must be the cable, then
[07:15:02] <mrc3> kulve, what kernel sources are you using with your beagle?
[07:15:29] * robclark (n=robclark@nat/ti/x-bd0944db2140e00e) Quit ()
[07:17:30] <kulve> mrc3: I usually take the revision from linux-omap that OE uses and then I patch the OE patches there.. Koen does good job in testing that version :)
[07:17:50] <kulve> requires some manual work though..
[07:20:51] <mrc3> excellent
[07:21:08] <mrc3> kulve, thanks for all the pointers! i think i'm gonna call it a night
[07:21:28] <mrc3> see ya!
[07:27:14] * aanishn (n=aanishn@59.160.172.220) has joined #beagle
[07:34:02] <ds2> 9
[07:48:53] * leslie (n=leslie@222.66.141.182) has joined #beagle
[07:55:43] * Wowbagger_ (n=wowbagge@d154-20-138-76.bchsia.telus.net) has joined #beagle
[08:05:13] * ZeZu (n=null@c-98-227-56-17.hsd1.in.comcast.net) Quit ()
[08:20:47] * nemequ (n=nemequ@ip68-111-215-155.sd.sd.cox.net) has joined #beagle
[08:23:53] * Wowbagger_ (n=wowbagge@d154-20-138-76.bchsia.telus.net) Quit (Client Quit)
[08:33:16] * RogerMonk (n=a0740758@nat/ti/x-dda32c9502769294) has joined #beagle
[08:34:51] * svante (n=svante@77.233.226.4) has joined #beagle
[09:30:41] * abitos (n=nixgibts@dslb-084-057-157-130.pools.arcor-ip.net) has joined #beagle
[09:33:18] * guillaum1 (n=gl@AMontsouris-153-1-20-180.w86-212.abo.wanadoo.fr) Quit ("Leaving.")
[09:33:52] * guillaum1 (n=GZBA4143@AMontsouris-153-1-20-180.w86-212.abo.wanadoo.fr) has joined #beagle
[09:36:29] <koen> oh joy
[09:36:37] <koen> someone blew up my mailbox
[09:37:04] <koen> the snailmail one with fireworks
[09:38:07] <_AV500_> koen: I have you did not have any beagleboard pending :-)
[09:38:12] <_AV500_> koen: I hope you did not have any beagleboard pending :-)
[09:42:42] <koen> luckily not :)
[09:43:20] <ds2> fireworks are still allowed in the EU!?
[09:43:32] <_AV500_> yep
[09:43:36] <mru> fireworks are great
[09:43:48] <_AV500_> there are sold a few days before the 31st
[09:44:07] <koen> ds2: I live in this city: http://en.wikipedia.org/wiki/Vuurwerkramp
[09:44:08] <_AV500_> but only to adults (in my case my grandmother was the proxy :-)
[09:46:09] <tomba> koen: I wonder what he meant with "OMAP do not have backlight class support". I'm using the backlight framework in dss2 just fine...
[09:47:26] <koen> tomba: the existing lcd driver for the evm doesn't have it
[09:48:09] <koen> tomba: but the evm DSS2 driver doesn't hook into the backlight class AFAIK
[09:48:35] <tomba> no, I've implemented it only for our own display
[09:49:01] <tomba> but the backlight framework is totally separate from the FB or DSS. so I don't understand why "OMAP does not have..."
[09:50:01] <tomba> I guess it's more like "we've done it like this, we're not gonna change it"
[09:52:06] * zedstar (n=john@fsf/member/zedstar) has joined #beagle
[09:54:52] * ssvb (n=ssvb___@a88-114-221-132.elisa-laajakaista.fi) Quit ("Leaving")
[09:54:54] <koen> I think so to
[09:54:57] <koen> hence my mails :)
[09:55:10] * koen hates patching userspace for lazy vendors
[09:55:48] <koen> I like kernel 2.6 so much because drivers tend to use the standard interfaces
[10:03:19] * ScriptRipper (n=martin@85.233.34.53.dynamic.cablesurf.de) has joined #beagle
[10:13:44] <koen> heh
[10:14:18] <koen> it looks like it's just a matter of renaming the magic in the machine file to set_bl_intensity and passing that to e.g. corgi_bl.c
[10:33:44] * rsalveti (n=salveti@189.70.38.175) has joined #beagle
[10:38:39] * leslie (n=leslie@222.66.141.182) Quit (Remote closed the connection)
[10:39:08] * leslie (n=leslie@222.66.141.242) has joined #beagle
[10:44:57] * leslie (n=leslie@222.66.141.242) Quit ("Konversation terminated!")
[10:52:25] * abitos (n=nixgibts@dslb-084-057-157-130.pools.arcor-ip.net) Quit (Nick collision from services.)
[10:52:41] * abitos (n=nixgibts@dslb-084-057-166-200.pools.arcor-ip.net) has joined #beagle
[10:54:58] * soman (n=somnath@61.16.248.242) Quit (Connection timed out)
[10:55:29] * soman (n=somnath@61.16.248.242) has joined #beagle
[10:56:12] * florian_ (n=fuchs@217.146.132.69) has joined #beagle
[10:57:07] * florian_ is now known as florian
[11:01:15] * shri (i=3d5fc3cd@gateway/web/ajax/mibbit.com/x-c3235f582253474f) Quit ("http://www.mibbit.com ajax IRC Client")
[11:13:50] * philippe (n=fille@a83-245-252-47.elisa-laajakaista.fi) has joined #beagle
[11:14:56] * rupeshgujare (n=Administ@59.160.172.220) has joined #beagle
[11:15:52] * khasim looking at Neuros OSD2, the packaging is very beautiful...
[11:22:21] <koen> I guess I must have a different package :)
[11:22:37] <koen> "white cardboard box with board + remote + cables"
[11:23:01] <koen> I wasn't lucky in the casing-or-not lottery for the devboards
[11:25:43] * supo (n=supo@s-inf-pc117.oulu.fi) Quit ("Ex-Chat")
[11:26:27] <florian> good morning
[11:28:33] <_AV500_> florian: hello
[11:29:51] <_AV500_> koen: did you convince Neuros to upgrade your OSD2 with an OMAP3 CPU module?
[11:31:57] * soman (n=somnath@61.16.248.242) Quit (Connection reset by peer)
[11:32:13] * soman (n=somnath@61.16.248.242) has joined #beagle
[11:34:03] <koen> _AV500_: heh, no
[11:36:29] <_AV500_> in the end, they just went X86: http://www.neurostechnology.com/neuros-link
[11:37:47] <koen> I noticed
[11:38:12] <koen> dunno if that's the fault of the davinci SoCd
[11:38:25] <koen> neuros has a weird development process
[11:39:32] <koen> I suspect they couldn't get enough bananas for a million monkeys and settled for 5
[11:39:38] <_AV500_> the idea was good, but neither the DM320 nor the DV have reached critical mass
[11:40:02] <_AV500_> with the OMAP3 it would have been different, due to BB the critical mass is there...
[11:40:35] <koen> I know which SoC I'd want on a neuros type device
[11:41:09] <_AV500_> well, we made products with DM320, DV and OMAP3, so all of them can "work"
[11:41:32] <koen> you haven't made products with the TI SoC I want :)
[11:41:41] <_AV500_> which one
[11:41:43] <koen> since that hasn't been released
[11:41:49] <_AV500_> which one
[11:43:07] <koen> dunno if I'm allowed to utter that name in public :)
[11:43:17] <koen> I'll wait for TI to announce it first
[11:43:24] <koen> dunno when that happens
[11:43:32] <koen> I guess I don't know a lot :)
[11:44:07] * TAK2004 (n=Administ@dslb-088-074-033-167.pools.arcor-ip.net) has joined #beagle
[12:04:51] * and-ri (n=ubuntu@pd95b4498.dip0.t-ipconnect.de) has joined #beagle
[12:05:07] <and-ri> hello :-)
[12:05:53] <and-ri> i hope i can find here some good souls who can help me
[12:06:27] <and-ri> i play around with the Angstrom fs (Date 05.12.08)
[12:08:02] <and-ri> and tried the kernels which are available for it (uImage 2.6.26-omap1)
[12:08:29] <and-ri> i try to play a movie with omapfbplay with this fs and kernel
[12:08:43] <and-ri> but it didn't work
[12:11:01] <and-ri> then i compiled my own kerne (2.6.26-omap1) with the configuration of the original kernel which i mentioned above.....and i recieved the error "FBIOPAN_DISPLAY: Invalid argument"
[12:11:28] <and-ri> when i tried the omapfbplay
[12:13:02] <and-ri> any idea how to change the kernel to kill this error
[12:14:06] * cbrake_away is now known as cbrake
[12:21:03] * FuL|OUT is now known as fulgas
[12:21:59] * abitos (n=nixgibts@dslb-084-057-166-200.pools.arcor-ip.net) Quit (Read error: 113 (No route to host))
[12:22:53] <koen> and-ri: you need the correct bootargs
[12:23:02] * khasim is not convinced with DVD archiver. Why is this used for ?
[12:23:02] <koen> and-ri: and angstrom moved to 2.6.27
[12:23:30] <koen> khasim: for lazy people not wanting to get up and put a dvd in their dvd player :)
[12:23:39] <khasim> :)
[12:23:58] <khasim> but truely is OSD2 meant only for that application - can't believe
[12:34:18] * Openfree (n=df@222.65.143.118) has joined #beagle
[12:34:40] * philippe (n=fille@a83-245-252-47.elisa-laajakaista.fi) Quit (Read error: 113 (No route to host))
[12:36:20] * RobertK (i=3edc0622@gateway/web/ajax/mibbit.com/x-89032a0bc39d8e75) has joined #beagle
[12:36:30] <RobertK> good morning
[12:36:39] <RobertK> tomba: *ping*
[12:37:49] <RobertK> tomba: I not able to change the virtual resolution of the framebuffer. I am using the newest kernel from your git. Any idea? (I send some Code to the ml).
[12:39:52] <koen> khasim: AIUI the osd2 only converts stuff from it's input, the archivers reads the mpeg from the disc
[12:40:47] <khasim> koen: then quality will not be as good as an archiver right?
[12:41:07] <koen> right
[12:41:24] <koen> you might be able to connect an IDE dvd drive or usb drive to the osd2
[12:41:41] <khasim> none of my ramdisk images are getting mounted on OSD2
[12:41:47] <khasim> with the 2.6.23 kernel
[12:42:56] <koen> OE can build images for the OSD2
[12:43:13] <khasim> Do you have a filesystem image for OSD2
[12:43:24] <khasim> I want to port latest Davinci GIT on to OSD2
[12:45:23] * JuanG (n=Juan@nat/ti/x-9b5ee1d7b8dc9a26) has joined #beagle
[12:45:34] * JuanG (n=Juan@nat/ti/x-9b5ee1d7b8dc9a26) has left #beagle
[12:46:31] <koen> great!
[12:46:38] <koen> I'm tired of using 2.6.23 on it
[12:47:20] <koen> khasim: see PM for the image
[12:47:39] <khasim> koen: got it
[12:50:37] <koen> zuh, kulve: http://lists.freedesktop.org/archives/xorg/2008-December/041629.html implies you could use pixman to convert stuff
[12:54:08] <tomba> RobertK: hmm. it may be broken, I don't think I have tested virtual resolutions for long time
[12:55:08] <RobertK> tomba: it worked two weeks ago (I do not exactly remember). Its also not possible to change the virt. res with fbset ('fbset -fb /dev/fb0 -vyres 490')
[12:55:09] <koen> I'd like to use virtual resolution for the pico
[12:55:12] <tomba> RobertK: oh, you first need to allocate more memory for it
[12:55:48] <RobertK> hmmm, I will check this. This code worked for weeks ;-)
[12:56:40] <tomba> there's a certain amount of mem allocated for the framebuffer. by default it's the size of the screen. OMAPFB_SETUP_MEM (or something like that) ioctl can be used to free and allocate that
[12:57:33] <tomba> in the older version, that used OMAPTAG_FBMEM, the memory was allocated a bit differently, and on beagle by default it allocated a bit more
[12:58:40] <koen> tomba: would it be possible to realloc it in the fly (within the bounds of the DMA size)?
[12:58:45] <tomba> a simple thing to test is to change from 24/32bpp to 16bpp. that'll give you some extra memory and you can then test virtual x/y easily
[13:00:37] <tomba> koen: it would. but I'm not sure if it's good. I'll have to think about it
[13:01:08] <koen> on the beagle it would be handy if I switch from e.g. monitor to pico dlp
[13:01:49] <LoneMeow> hmm, does it pre-reserve the pages (or even need physically contiguous pages)? since large physcontig allocations on a running system are quite failureprone...
[13:02:06] <tomba> yep. I'm just not sure if it's the task of the omapfb driver to handle these automatically, or should there be some user space component that will manage the displays and setup the memories accordingly
[13:02:41] <tomba> LoneMeow: I implemented this video ram manager, it allocates the memory at boot time
[13:03:08] <khasim> koen: I get Warning: unable to open an initial console.
[13:03:13] <LoneMeow> tomba, hmm, how does it choose how much total to allocate? commandline parameter?
[13:03:26] <tomba> LoneMeow: it's defined in the board file
[13:03:44] <khasim> koen: I doubt if device nodes are getting created
[13:04:00] <LoneMeow> tomba, ah makes sense, I guess
[13:04:31] <tomba> LoneMeow: at least partially. I haven't gotten to implement any cmdline parameters yet. if I had, I think that would be one of them =)
[13:05:14] <LoneMeow> tomba, mostly thinking about situations where the boardfile specifies a rather large max memory but you really *know* you won't need all of that... would be neat to be able to bypass it and save some pages
[13:05:32] <tomba> LoneMeow: but the main points in the vram allocator are: 1) it can allocate memory from SRAM 2) it can allocate very large regions, dma_alloc can't 3) it can allocate the memory at defined position, which can be told by the bootloader
[13:05:51] <and-ri> koen: thx, but i used the bootargs with setenv already which were mentioned i the readme....and i move also to 2.6.27.... but unfortunately ...my own compile kernel has still some trouble with "omapfbplay" , he said "mmap:invalid argument" or "Illegal Instruction "
[13:07:02] <tomba> joy, DSS2 runs on N810 now also =)
[13:12:29] <koen> khasim: ah, I guess that makes sense
[13:12:35] <koen> khasim: I'll look into it
[13:14:13] <koen> tomba: great, now zuh can make xf86-video-omapfb work better with it :)
[13:14:51] <RobertK> tomba: I changed the bpp - no success: http://www.mibbit.com/pb/Sk54O6
[13:15:53] <tomba> RobertK: I mean switch it to 16bit, then set vxres. what is this pan program doing?
[13:16:46] <RobertK> tomba: Setting PAN failed (ioctl:FBIOPAN_DISPLAY)
[13:16:48] <zuh> koen: re pixman: Not really, at least not without major work. I only can use YUV formats as sources (ie. it has FetchPixel impls for them, which convert to RGB32) and even if the corresponding store functions would exist, it would do a yuy2->rgb32->yv12 conversion which is NOT really a good idea :)
[13:17:05] <koen> heh
[13:17:21] <koen> that's indeed not what we want
[13:17:59] <LoneMeow> zuh, I still have the v6 YUYV->YV12 code around btw
[13:18:08] <zuh> Does it work?
[13:18:16] <LoneMeow> though it requires some touching up... like supporting non-8-pixel aligned dimensions and such
[13:18:23] <zuh> details
[13:18:25] <LoneMeow> and it's slower on v7 than a naive byte-wide loop
[13:18:31] <LoneMeow> faster on v6, though
[13:18:44] <tomba> RobertK, you can't pan if the virtual resolution is the same as the real resolution
[13:19:02] <LoneMeow> especially on the v6s that are really bandwidth limited and have crappy write buffers
[13:19:13] <zuh> but the dss2 sounds great, now that the wlan driver can be used without silly closed-source binaries I can start using linux-omap kernels for real.
[13:19:25] <LoneMeow> still considering a hand-written generic assembler version
[13:19:49] <LoneMeow> and a NEON version would be great for v7s, I tried to adapt mru's code but all I got out of it was infinite loop hangs
[13:19:49] <tomba> zuh, usb ethernet is all you need ;)
[13:20:03] <jkridner> good morning all
[13:20:18] <RobertK> tomba: I know. Thats why I set the vyres: http://www.mibbit.com/pb/xyCa98 and that failed! I did not get an error code but the vyres does not change...
[13:20:54] <jkridner> kc8apf: yt?
[13:21:09] <tomba> RobertK, it's because there's no memory to make it larger. switch first to 16bpp, and then run it
[13:21:56] <jkridner> koen: quick question, are the Zoom libs working for you?
[13:21:57] <ScriptRipper> jmkridner: hi Jason.
[13:22:10] <jkridner> hi ScriptRipper
[13:22:29] <RobertK> tomba: I thought I already did that. I chnaged the bpp to 16 with fbset and then run my program: http://www.mibbit.com/pb/eG4Gud
[13:23:24] <tomba> RobertK: hmm ok. in the earlier copy paste you switched back to 32bpp. what happens if you try to set the virt res with fbset?
[13:24:08] <khasim> jkridner: good morning
[13:24:17] <jkridner> hi khasim
[13:25:22] <khasim> jkridner: I was validating my Neuros box
[13:25:43] <jkridner> great. should I see you in #neuros?
[13:25:52] <khasim> jkridner: I have there GIT kernel working, will have to see if davinci GIT kernel boots on the same
[13:25:55] * bazbell (n=a0192809@nat/ti/x-8677740cb1210d47) has joined #beagle
[13:28:14] <RobertK> tomba: It does not work: http://www.mibbit.com/pb/cgqb0k
[13:28:18] * _AV500_ is now known as AV500
[13:28:54] <tomba> RobertK: how about fbset -yres 400 -vyres 480
[13:29:36] <RobertK> tomba: no success: http://www.mibbit.com/pb/c2u6j8
[13:29:56] <tomba> RobertK: it worked fine =)
[13:30:17] <RobertK> tomba: you are right :-)
[13:30:23] <tomba> at least the fbset output is ok. perhaps there's something that prevents setting virtual resolution larger than the display. I'll have to check
[13:31:14] <RobertK> So I could set the res to 1024*768 and then change it to 640x480 with vyres=960? I will try that.
[13:33:00] <tomba> hmm I didn't understand that. but you can try this: fbset -xres 100 -vxres 100 -yres 500 -vyres 500
[13:33:46] <tomba> that at least worked on N810, so I don't think there's a limit regarding the display size
[13:35:15] <RobertK> tomba: I set it (fbset -fb /dev/fb0 -xres 100 -vxres 100 -yres 500 -vyres 500) but the display does not change (my monitor still show the linux penguin @640x480)
[13:35:45] <tomba> yes, it doesn't change the resolution. it (xres&yres) changes the size of the graphics plane
[13:36:07] <tomba> the only way to change the resolution currently is to change the kernel config
[13:36:36] <koen> khasim: try the image tool again, it now generates initial devnodes now
[13:36:37] <tomba> but anyway, if that works, it's probably about memory. so you could try the setupmem ioctls to allocate more memory for the framebuffer
[13:36:50] <khasim> koen: thanks, will give a try
[13:37:46] <koen> khasim: thanks for finding that bug :)
[13:37:50] <RobertK> tomba: or can I reserve more memory while booting - that means change the kernel source? Where do you reserve the memory?
[13:38:49] <tomba> RobertK: you still need the ioctls. omapfb always allocates just enough memory for the framebuffer. what you can change at boot time is the total video ram size, from with omapfb allocates the memory
[13:39:18] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
[13:40:16] <RobertK> tomba: okay, I will try it with the ioctl's and OMAPFB_SETUP_MEM and will report any failure or success.
[13:50:59] * skipisz (n=a0867407@nat/ti/x-8a5aca685fdc6d2b) has joined #beagle
[13:51:28] * gcohler (n=gcohler@nat/ti/x-ce2e98701958edc7) has joined #beagle
[13:52:21] <khasim> koen: same issue
[13:52:28] <khasim> VFS: Mounted root (ext2 filesystem).
[13:52:28] <khasim> Freeing init memory: 124K
[13:52:28] <khasim> Warning: unable to open an initial console.
[13:52:42] <tomba> I wish there was a nice test program for omapfb to test all these virtual resolutions, pannings, overlays etc. currently it's a bit difficult to test them
[13:52:52] <khasim> koen: I tried with a NFS image of mine (it is a MV filesystem) that booted fine
[13:53:43] <khasim> koen: I am actually deleting the contents of opkg in /var/lib, to fit into a ramdisk
[13:53:49] <khasim> koen: will that be an issue
[13:54:54] <LoneMeow> tomba, write a test program!
[13:54:56] <LoneMeow> ;)
[13:55:13] <tomba> LoneMeow: No, I'll concentrate on the wishing part!
[13:57:20] <suihkulokki> wii LoneMeow
[13:57:36] <suihkulokki> why not wolf :)
[13:58:20] <LoneMeow> long story, but it has to do with QuakeNet and someone else also wanting to use that nickname
[13:59:59] <koen> khasim: did you extract it as root?
[14:00:45] <khasim> yes
[14:00:50] <khasim> with all permissions
[14:01:46] <koen> deleting /var/lib/opkg isn't an issue
[14:01:54] <khasim> ok
[14:02:05] <RobertK> tomba: are you working on that virt. res. issue? Does it makes sense that I try it also with OMAPFB_SETUP_MEM?
[14:02:44] <khasim> will have to drop off, will test again later
[14:02:48] <khasim> thanks
[14:03:33] <tomba> RobertK: you will have to use setupmem anyway. the question is that will setup mem fix the problem or not. if not, then there's a bug somewhere
[14:03:55] <tomba> RobertK: but if you meant automatic memory realloc on virtual res change, no, I'm not working on that
[14:04:37] <RobertK> tomba: okay, I will give it a try.
[14:06:02] * Beagle7 (n=Beagle7@adsl-69-152-209-101.dsl.fyvlar.swbell.net) Quit ()
[14:09:57] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
[14:12:28] <RobertK> tomba: getting the memory information works but setting not. The code: http://www.mibbit.com/pb/44VitS
[14:12:36] <RobertK> any idea?
[14:15:26] * aanishn (n=aanishn@59.160.172.220) has left #beagle
[14:19:58] <Crofton|work> does codec engine build and work for the Davinci EVM?
[14:20:16] <koen> last time we tried it did
[14:20:26] <Crofton|work> ok
[14:20:44] <Crofton|work> someone is asking about building root file system for davinci :)
[14:23:20] <tomba> RobertK: disable the plane first with OMAPFB_QUERY_PLANE, plane_info.enabled = 0, OMAPFB_SETUP_PLANE
[14:23:21] * rupeshgujare (n=Administ@59.160.172.220) has left #beagle
[14:23:31] <tomba> RobertK: I think you can't free the memory if the plane is in use
[14:25:01] <tomba> and enable it afterwards, of course =)
[14:26:29] <Crofton|work> I hate the DV list. replies go to the sender, not the list, so my attempt to pimp OE on the list failed :)
[14:28:12] * flo_lap (n=fuchs@217.146.132.69) has joined #beagle
[14:28:12] * RogerMonk (n=a0740758@nat/ti/x-dda32c9502769294) Quit (Read error: 60 (Operation timed out))
[14:28:31] * RogerMonk (n=a0740758@nat/ti/x-5108be1ada12c772) has joined #beagle
[14:29:58] <RobertK> tomba: I will try that!
[14:30:29] <tomba> at least my test program worked, I was able to get larger virtual resolution than the screen
[14:33:58] * svante (n=svante@77.233.226.4) Quit ("Ex-Chat")
[14:37:06] <RobertK> tomba: after dis/enabling the planes I can setup the memory. But when I try to set the vyres to 960 it changes to 481?!?! The code: http://www.mibbit.com/pb/pYgANv
[14:37:23] <RobertK> I increased the mem with: "meminfo.size+=480;"
[14:38:00] * nemequ (n=nemequ@ip68-111-215-155.sd.sd.cox.net) Quit (Read error: 60 (Operation timed out))
[14:38:10] <tomba> well, you need to increase the memory more. 480 bytes is not enough for even one line
[14:38:25] * and-ri (n=ubuntu@pd95b4498.dip0.t-ipconnect.de) Quit (Connection timed out)
[14:38:38] <tomba> and don't add to the memory, set it to something. if you want x*y virt res, then set it to x*y*bytesperpixel
[14:41:05] <RobertK> tomba: of course you are right. Thank you, now it seems to work (I will do some more testings). I will send an email to the list for the archives.
[14:42:56] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
[14:49:43] * nemequ (n=nemequ@ip68-111-215-155.sd.sd.cox.net) has joined #beagle
[14:55:45] * emeb (n=ericb@ip72-223-90-212.ph.ph.cox.net) has joined #beagle
[14:58:03] <RobertK> tomba: :-) tested and verified - everything works again! Thank you so much!
[15:03:21] * RogerMon1 (n=a0740758@nat/ti/x-a9760efc0a3b58dc) has joined #beagle
[15:06:53] * RobertK (i=3edc0622@gateway/web/ajax/mibbit.com/x-89032a0bc39d8e75) Quit ("http://www.mibbit.com ajax IRC Client")
[15:10:57] * RogerMonk (n=a0740758@nat/ti/x-5108be1ada12c772) Quit (Remote closed the connection)
[15:15:21] * soman (n=somnath@61.16.248.242) Quit (Remote closed the connection)
[15:21:38] * fr3men (n=fr3men@ip-5-146.sn2.eutelia.it) has joined #beagle
[15:22:32] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit ()
[15:22:33] * jsync (n=jess@59.160.172.220) has left #beagle
[15:32:04] * Crofton|irssi (n=balister@66-207-66-26.black.dmt.ntelos.net) Quit (Read error: 60 (Operation timed out))
[15:32:52] * n6pfk (n=mike@c-76-104-40-104.hsd1.va.comcast.net) has joined #beagle
[15:34:32] * Crofton|irssi (n=balister@66-207-66-26.black.dmt.ntelos.net) has joined #beagle
[15:37:52] * gcohler (n=gcohler@nat/ti/x-ce2e98701958edc7) Quit ("Trillian (http://www.ceruleanstudios.com")
[15:40:48] * Openfree (n=df@222.65.143.118) Quit (Remote closed the connection)
[15:57:18] * khasim (n=a0393720@192.163.20.231) has joined #beagle
[16:11:27] * scruggs (n=chris@72-161-120-114.dyn.centurytel.net) Quit ("Ex-Chat")
[16:16:37] * Crofton (n=balister@66-207-66-26.black.dmt.ntelos.net) Quit (Read error: 113 (No route to host))
[16:21:04] * scruggs (n=chris@72-161-120-114.dyn.centurytel.net) has joined #beagle
[16:30:09] * fr3men (n=fr3men@ip-5-146.sn2.eutelia.it) Quit ("Leaving.")
[16:44:37] * jkridner|work (n=a0321898@nat/ti/x-5cbedddd70cb9591) has joined #beagle
[16:44:59] <mrc3> hello, all! quick question about usb keyboard
[16:45:19] <mrc3> (though most likely quick answer is "mini-a usb cable, check it!")
[16:46:08] <mrc3> i connected my all-new all-great beagle to a usb hub with a usb mouse in it
[16:46:13] <mrc3> i am powering the beagle using a) an external power source, and b) the usb hub
[16:46:47] <mrc3> (btw, s/keyboard/mouse, i still don't have a good usb keyboard)
[16:47:04] * calculu5 is now known as calculus
[16:47:20] <mrc3> i cannot get the mouse to move using angstrom and several different kernels (2.6.22, 2.6.26, from here, from there, compiled by myself, etc.)
[16:47:48] <mrc3> as soon as i connected the beagle via usb to my laptop, it said "usb0: high speed config #1: 8 mA, Ethernet Gadget, using CDC Ethernet"
[16:48:16] <mrc3> is that in any way interfering with having the mouse plugged via usb? or is it definitely my cable? (because it really looks like a mini-a)
[16:48:46] * flo_lap (n=fuchs@217.146.132.69) Quit (Read error: 113 (No route to host))
[16:50:05] * and-ri (n=ubuntu@pd95b4498.dip0.t-ipconnect.de) has joined #beagle
[16:50:41] <koen> tomba: how far along is dss2 in the linux-fbdev review process?
[16:50:52] <and-ri> how can i install a package? i tried "ipkg Package.ipk" but he said, ipkg not found
[16:51:02] <koen> and-ri: opkg
[16:51:18] <koen> mrc3: try 'lsusb' and see if it detects the mouse and/or hub
[16:51:35] <and-ri> thx
[16:53:41] <and-ri> a lot of thanks to you koen, i read that you do the whole stuff with angstrom for omap.... this is pretty cool
[16:54:39] <and-ri> may, can you tell me whether you are able to run omapfbplay with the newest angstrom fs, which you have made
[16:56:38] <mrc3> koen, `lsusb -v` shows only one entry, for "MUSB HDRC host driver"
[16:57:34] <koen> mrc3: it seems it doesn't detect the mouse
[16:57:44] <koen> mrc3: check 'dmesg' for any hints
[16:58:00] <koen> and-ri: you can use omapfbplay if you don't start X
[16:58:22] <koen> and-ri: put 'exit' in line 2 of /etc/init.d/gpe-dm
[16:58:33] <koen> the XV driver and omapfbplay don't get along too wel
[16:58:38] <kulve> mrc3: that message on you laptop means that the beagle is in usb slave mode acting as a ethernet device (and usb slave cannot use other usb slave (peripheral) devices)
[16:59:50] <mrc3> i went through dmesg a couple of times already. can you confirm your setup looks like mine?: i'm powering the beagle via the usb hub with a (supposedly) mini-a cable. i tried kulve's suggestion of using an external power source and keeping the usb cable connected to the hub, and still no joy
[17:00:23] <koen> I don't think a mini-A would allow backpower
[17:00:39] <mrc3> instead of plugging the usb cable to the laptop (and thus getting the ethernet messages) i am now connecting to the usb hub
[17:00:58] <mrc3> koen, ahh, so it is a disguised mini-b cable?
[17:01:06] <kulve> koen: it does.. (most usb hubs doesn't power the host, but one of our models do)
[17:01:54] <koen> kulve: ah, didn't know that
[17:01:58] <kulve> mrc3: you can check if it's mini-a with a voltage meter (or what ever those are called). Check if the pin4 and pin5 in the beagle side connector are connected together
[17:02:12] <jkridner|work> just found out about: http://www.ti.com/dummiesbook
[17:02:13] <koen> but the beagle doesn't take power in host mode AFAIK
[17:02:54] <kulve> koen: in our setup, I first connect the usb cable and the beagle powers up. Then I connect the power cable and the beagle is resetted using the power cable. Annoying, but works..
[17:03:09] <koen> QualiTI?
[17:03:58] <jkridner|work> forgive us, please.
[17:04:03] <jkridner|work> :)
[17:04:07] <mrc3> koen, kulve, thanks for the suggestions. i don't think i can get a mini-a cable in this town, but i think that's what i'm needing now
[17:04:20] <jkridner|work> small entertainment where we can get it. :)
[17:05:14] <koen> jkridner|work: QualiTI sounds like a suggesting I would make expecting it to get shot down :)
[17:05:34] <koen> jkridner|work: I've learned not to make such suggestions since they usually don't get shot down ;)
[17:05:54] <jkridner|work> exactly.
[17:06:26] <jkridner|work> the mild amusement turns into a form of acceptance.
[17:06:43] <and-ri> thx koen
[17:06:47] * and-ri (n=ubuntu@pd95b4498.dip0.t-ipconnect.de) has left #beagle
[17:07:00] <jkridner|work> we think these things will live internally and the name reflects the purpose of the tool.
[17:07:14] <kulve> mrc3: we "modified" our mini-b to mini-a by opening the connector and soldering the pins 4 and 5 together. Real ugly looking thing, but it worked :)
[17:07:27] <jkridner|work> then we must be slightly embarrassed when everyone reads it.
[17:07:35] <jkridner|work> anyway, the book should be good.
[17:08:03] <jkridner|work> it looks like there is a sample MPEG4 decode library included.
[17:08:11] <jkridner|work> in the sample code that is.
[17:08:15] <jkridner|work> for OMAP3530.
[17:08:31] <jkridner|work> I haven't reviewed the book, but I respect the authors.
[17:10:11] <jkridner|work> looking into how to order the book, I seem to have placed an order.
[17:10:51] <koen> heh
[17:10:53] <koen> same here
[17:11:10] <koen> looks like a usefull book
[17:11:28] <koen> although I'm slightly embarrased by needing a dummy book
[17:11:49] <koen> well, I still have 6-8 weeks to set my pride aside :)
[17:13:26] * soman (n=somnath@122.169.161.41) has joined #beagle
[17:19:17] * JoeyBorn (n=jborn@dsl017-022-252.chi1.dsl.speakeasy.net) has joined #beagle
[17:20:28] * garren (n=garren@dsl-243-85-62.telkomadsl.co.za) has joined #beagle
[17:22:24] <garren> hi all
[17:33:09] * torus (n=nils@c130105.adsl.hansenet.de) has joined #beagle
[17:33:13] * Ragha_ (n=Ragha@nat/ti/x-a9473017b9e9a1e0) has joined #beagle
[17:33:14] <torus> hi folks.
[17:33:23] * uwe2 (n=uwe@dslb-088-064-073-163.pools.arcor-ip.net) has joined #beagle
[17:33:43] <torus> could anyone please help me for a simple cross-compile job? (description follows)
[17:34:20] * guillaum1 (n=GZBA4143@AMontsouris-153-1-20-180.w86-212.abo.wanadoo.fr) Quit ("Leaving.")
[17:34:36] <torus> I want to cross-compile libogg. I run configure --target=arm-none-gnueabi-linux and configure does not complain. If I run make however I always get x86 binaries.. the compiler is in the path..
[17:34:39] <torus> any ideas?
[17:35:29] <skipisz> Try --target=arm-none-gnueabi-linux --host=arm-none-linux-gnueabi --build=i386-linux
[17:37:08] <torus> will do so..
[17:37:21] <Ragha_> you can modify the Makefiles to change CC, LD etc
[17:37:39] <torus> Ragha_, tried that (CC is set for example) but it seems to make no difference.
[17:38:41] <torus> hm... skipisz. it still calls the native gcc.
[17:39:49] <torus> oh! I got it!
[17:39:59] <torus> ./configure --target=arm-none-linux-gnueabi --host=arm-none-linux-gnueabi --build=i386-linux <--- that's the right one (just for the irclog)
[17:40:16] <torus> thanks guys!
[17:43:03] <kulve> I like the SB. There I run: ./configure [enter]
[17:43:22] <torus> what's the SB? Never heared of that.
[17:43:27] <kulve> scratchbox.org
[17:43:51] <torus> kulve, linux image with native compiler on the arm?
[17:44:15] <kulve> torus: SB hides the cross compiling part. So you can pretend you are in a real native arm environment
[17:44:41] <kulve> it do have it's own issues, but generally it works nicely
[17:45:17] <kulve> it uses qemu run run arm binaries where needed
[17:45:25] <torus> looks nice..
[17:45:27] <kulve> e.g. when running ./configure
[17:46:01] <kulve> torus: you might want to check linux.onarm.com if you want to try it out..
[17:46:09] * uwe__ (n=uwe@dslb-084-056-046-156.pools.arcor-ip.net) Quit (Read error: 110 (Connection timed out))
[17:46:42] <torus> kulve, I will... that cross-compilation stuff drives me crazy.. Now I have to find out where the damn compiled lib is. :-)
[17:47:06] <kulve> yeah, in SB everything (well, almost) looks like a native environment
[17:47:40] * uwe_ (n=uwe_@dslb-084-056-046-156.pools.arcor-ip.net) Quit (Read error: 110 (Connection timed out))
[17:47:58] <torus> I already thought about compiling and installing gcc onto the bb. I don't mind the performance..
[17:48:14] <torus> but cross compiling gcc is not easy either.
[17:48:24] <kulve> yeah, some people are doing that, but I find SB a much more convenient
[17:49:01] * torus goes back to work...
[17:49:57] * Xerion (i=xerion@82-170-197-160.ip.telfort.nl) Quit (" ")
[17:52:52] <garren> if i've run OE console image and then later run the demo window manager one... if I then later run the console image will it have to download everything again?
[18:02:02] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
[18:04:34] * garren (n=garren@dsl-243-85-62.telkomadsl.co.za) Quit ("Ex-Chat")
[18:06:59] * atin_ (n=atin@c-71-232-7-71.hsd1.ma.comcast.net) has joined #beagle
[18:07:15] * garren (n=garren@dsl-243-85-62.telkomadsl.co.za) has joined #beagle
[18:11:12] * geckosenator (n=sean@c-75-71-72-89.hsd1.co.comcast.net) has joined #beagle
[18:13:14] * uwe_ (n=uwe_@dslb-088-064-073-163.pools.arcor-ip.net) has joined #beagle
[18:14:41] * nemequ (n=nemequ@ip68-111-215-155.sd.sd.cox.net) Quit ("Ex-Chat")
[18:15:15] * florian (n=fuchs@217.146.132.69) Quit (Remote closed the connection)
[18:17:54] * erbo_ (n=erik@c-7b7de455.115-16-64736c14.cust.bredbandsbolaget.se) Quit (Read error: 60 (Operation timed out))
[18:19:38] * Sept (n=bakljg@c-98-240-226-129.hsd1.mn.comcast.net) has joined #beagle
[18:20:55] <koen> garren: no
[18:22:57] * koen sets fire to python.distutils
[18:23:16] * ldesnogu_ (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) has joined #beagle
[18:26:56] * robtow (n=rtow@nat/montavista/x-50670df498ba18df) has joined #beagle
[18:29:43] * geckosenator (n=sean@c-75-71-72-89.hsd1.co.comcast.net) Quit (Read error: 110 (Connection timed out))
[18:30:55] * MAD|chores is now known as MostAwesomeDude
[18:31:28] * rsalveti (n=salveti@189.70.38.175) Quit (Remote closed the connection)
[18:31:53] * DJWillis (n=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) has joined #beagle
[18:31:58] * DJWillis (n=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) Quit (Read error: 104 (Connection reset by peer))
[18:40:29] * dirk2 (n=dirk@92.117.36.240) has joined #beagle
[18:40:58] <jkridner> might be a bit late in the day to bring this up, but I'd like to start planning a software design contest for Beagle.
[18:41:12] <jkridner> first thing to do is to select some judges.
[18:41:26] <jkridner> being a judge will disqualify you from submitting an entry.
[18:41:44] <jkridner> the prize will be a Rev C prototype with working USB host.
[18:42:10] <jkridner> thought I'd chat about this here before getting the discussion going on the mailing list.
[18:42:27] <koen> what kind of software design?
[18:42:59] <koen> would one get points for cross-compiling something or for properly porting or for a cool idea?
[18:43:05] <koen> or for integration cool things?
[18:43:18] <jkridner> a full SD card demo image. judges to decide the criteria for judgement.
[18:43:34] <jkridner> I'd like to see ease-of-use be a top criteria, but I'll let the judges conclude.
[18:44:06] <koen> "good, although it lacks flubber, mhee"
[18:44:24] <jkridner> I think there can be several categories for points. moving something from native builds or just having a cool idea, whatever.
[18:45:15] <jkridner> we can set some requirements that all source must be open or that any binary dependencies must be freely available, or stick to a small handful of called-out allowed binary blobs.
[18:45:39] <jkridner> anyway, I want most of the judges to be non-TI.
[18:46:09] <jkridner> koen: I'm hopeful you'd be a judge and be disqualified from the contest itself.
[18:47:03] <dirk2> I think koen has already a Rev C prototype ;))
[18:48:24] * Crofton|1rssi (n=balister@pool-71-171-23-169.nwrknj.east.verizon.net) has joined #beagle
[18:49:02] <jkridner> but not one with a working USB host, which would be the point. :)
[18:49:11] <jkridner> I should maybe say a C1 prototype.
[18:49:22] <jkridner> dirk2: I'd think you should be a judge as well.
[18:50:05] <jkridner> I think having 7 judges should be good. perhaps khasim and I from TI, then the rest from the community?
[18:50:39] <dirk2> jkridner: If you focus on "a full SD card demo image" that's fine. But you should state this clearly. E.g. what's about people having a perfect kernel patch making XXX in kernel working perfectly (MUSB? DSS2?)?
[18:51:27] <jkridner> yeah, having the most peripherals working with a certain performance level should definitely be a criteria.
[18:51:55] <koen> I'd accept an invitation to be a judge :)
[18:52:05] <jkridner> maybe we should collect a wishlist on a wiki for the criteria and then move them in or out of the approved category?
[18:52:41] <jkridner> given the involvement of each of you, keon and dirk2, I don't think many people would object to you being on the judges.
[18:52:51] <dirk2> whishlist on a wiki sounds good
[18:52:52] <jkridner> I don't expect this first competition to run very long.
[18:53:15] <jkridner> I'd like to do the give-away in early January.
[18:53:42] <jkridner> maybe even have the demo available for CES.
[18:54:04] <jkridner> (there aren't currently any plans to show Beagle at CES, as far as I've heard. I'm skipping this year.)
[18:54:19] <dirk2> CES is when?
[18:55:15] <koen> 8-11 januari iirc
[18:55:37] <jkridner> Jan 8.
[18:55:39] <jkridner> 3 weeks.
[18:55:41] <jkridner> so, this would be a *very* short contest. :)
[18:55:43] <jkridner> we can probably re-use the committee and process if it goes well.
[18:55:58] * Crofton (n=balister@pool-71-171-23-169.nwrknj.east.verizon.net) has joined #beagle
[18:56:17] <koen> hey Crofton
[18:56:18] <jkridner> so, dirk2, will you accept as well and help pick out the other judges?
[18:56:24] <koen> Crofton: I fixed gnuradio build in OE :)
[18:56:37] <jkridner> so, koen, jkridner, and dirk2 + 4 more?
[18:56:38] <dirk2> dirk2: yes. I propose sakoman
[18:56:58] <jkridner> sakoman ping.
[18:57:13] <sakoman> pong
[18:57:13] <jkridner> we may have everyone picked out on IRC before even getting to the mailing list. :)
[18:57:35] <jkridner> would you be interested in judging a software design contest for Beagle to give away a Rev C?
[18:57:38] <koen> ds2, robertk, geoff C?
[18:57:42] <sakoman> of course!
[18:57:43] <jkridner> does that conflict with Overo activities?
[18:57:48] <jkridner> k, great!
[18:58:37] <jkridner> that's 4. (dirk2 + jkridner vote yea, koen doesn't get chance to speak up first) :)
[18:59:10] <jkridner> ds2 ping
[18:59:26] <Crofton> urg
[18:59:35] <Crofton> need percocet to kick in ...
[18:59:39] * balister_ (n=balister@pool-71-171-23-169.nwrknj.east.verizon.net) has joined #beagle
[18:59:43] <Crofton> I want to win
[18:59:55] * Crofton|irssi (n=balister@66-207-66-26.black.dmt.ntelos.net) Quit (Read error: 110 (Connection timed out))
[19:00:19] * erbo_ (n=erik@c-7b7de455.115-16-64736c14.cust.bredbandsbolaget.se) has joined #beagle
[19:00:25] <jkridner> does that mean you want to stay out of the judging Crofton? or are you just looking to win pain relief?
[19:00:33] <Crofton> heh
[19:00:40] <Crofton> what ever works for you:)
[19:00:49] <dirk2> Crofton: Send some beer to judge koen :))
[19:01:00] <balister_> hehe
[19:01:20] * Crofton|work (n=balister@66-207-66-26.black.dmt.ntelos.net) Quit (Read error: 110 (Connection timed out))
[19:01:22] <jkridner> so, jkridner, sakoman, koen, dirk2, + 3 (I expect khasim to be one of those)
[19:01:39] <jkridner> is that the right number of judges?
[19:01:55] <sakoman> seems like plenty to me
[19:02:01] <balister_> make sure there are an odd number
[19:02:08] <jkridner> plenty as in too much or plenty as in just right?
[19:02:16] <jkridner> yeah, for sure an odd number.
[19:02:25] <jkridner> I'm going for 7.
[19:02:28] <koen> jkridner: ds2, robertk, geoff C?
[19:02:32] <sakoman> who's this balister_ guy and what did he do with Crofton?
[19:02:36] <balister_> heh
[19:02:37] <koen> jkridner: unless they want to go for the prize :)
[19:02:44] <balister_> changin dsl providers
[19:02:52] <jkridner> koen: would you put Geoff C in over khasim?
[19:03:01] <ds2> pong
[19:03:16] <koen> jkridner: I have no real opinion on that
[19:03:27] <dirk2> jkridner, koen: mru?
[19:03:32] <jkridner> I don't want it to be to TI-heavy. If we put in Geoff C, I'd be tempted to add ...
[19:03:35] <jkridner> yeah, like mru.
[19:03:47] <koen> mru of course :)
[19:04:02] <ds2> jkridner: pong
[19:04:12] <jkridner> ds2: you interested in being a judge for a short-time-fuse software design competition to give away a Rev C proto with working USB host?
[19:04:36] * ZeZu (n=null@c-98-227-56-17.hsd1.in.comcast.net) has joined #beagle
[19:04:36] <ds2> jkridner: sure
[19:04:43] * Beagle0 (n=Beagle0@brln-4db9ffa4.pool.einsundeins.de) has joined #beagle
[19:05:06] <jkridner> want to try to execute this in 3 weeks so that I can give away the board. I'm expecting to have 2 boards I can give away on Jan 6 or so.
[19:05:08] * Beagle0 is now known as RobertK
[19:05:15] <jkridner> I thought it would be best to make it a competition.
[19:05:27] * ds2 scrolls up
[19:05:29] <jkridner> judges are excluded from making entries.
[19:05:47] <sakoman> jkridner: is that enough time to get entries?
[19:05:53] <RobertK> I would do the job (I am RobertK with a web based IRC client)
[19:06:25] <RobertK> Oh, I see, "/nick" worked...
[19:06:28] <jkridner> we are up to: sakoman, dirk2, koen, jkridner ds2, ... nominations for RobertK, khasim, Geof C
[19:06:34] <jkridner> hi RobertK.
[19:06:42] <ds2> <--- willing to chip in to BUY a board if someone could make MUSB work perfectly in the next 3 week!
[19:06:49] <jkridner> oh, and mru.
[19:07:14] <jkridner> ds2: put it on the wishlist! (URL pending)
[19:07:37] <ds2> 'k
[19:07:44] <jkridner> dirk2: shall I make the wiki page for the wishlist or do you already have it covered? ;)
[19:08:10] * balister_ is now known as Crofton|work
[19:08:19] <ds2> fixing up musb is on my todo list since the things I want to do seem to always run into musb issues :(
[19:08:21] <jkridner> RobertK: I guess you've seen you've been nominated.
[19:08:28] <dirk2> what about: Create a wiki page. Everybody can propose every project/patch/xxx. This will be listed at the wiki page. Then everybody can vote (not sure how to do this technically). Then the judges will select 2 from the top ~20 ?
[19:08:37] * geckosenator (n=sean@71.237.94.78) has joined #beagle
[19:08:45] <RobertK> jkridner: yes. And what does this means? (I just came in...)
[19:08:49] <Crofton|work> I'm assuming you are accepting ports of existing sw to the beagle
[19:09:04] <jkridner> Crofton|work: I'd suggest we do, yes.
[19:09:14] <jkridner> I'd like to see an entire SD-card image.
[19:09:18] <Crofton|work> it will be hard to do original works on a short fuse
[19:09:24] <jkridner> it may only have 1 feature that is new.
[19:09:29] * koen wonders why google dislikes mail from certain TI people
[19:09:35] <jkridner> no reason not to use what already exists.
[19:09:37] <Crofton|work> of course you need something that highlights omap3 capability
[19:09:43] <koen> some people always end up in 'spam'
[19:10:05] * ds2 gets flash back from ESC of the made scramble to build SD cards
[19:10:15] <dirk2> jkridner: Yes, create a page ;)
[19:10:16] <jkridner> RobertK: can you scroll-back?
[19:10:34] <RobertK> jkridner: I will have a look at the logs.
[19:11:20] <jkridner> back to sakoman's concerns.... we are looking at a very short time frame and a lot of judges (now up to maybe 9).
[19:11:48] <jkridner> I think there is enough out there that someone could do something interesting in 2-3 weeks.
[19:12:10] <jkridner> this also goes along with a lot of people's winter breaks.
[19:12:26] <ds2> and company shutdowns ;)
[19:12:37] <jkridner> I'd like to see something that shows off OMAP3530 core value, but we can likely repeat this competition again.
[19:12:44] <jkridner> ds2: and that is a good thing!
[19:14:22] * RobertK_ (n=rkuhn@brln-4db9ffa4.pool.einsundeins.de) has joined #beagle
[19:14:32] * RobertK (n=Beagle0@brln-4db9ffa4.pool.einsundeins.de) Quit ()
[19:14:41] * RobertK_ is now known as RobertK
[19:14:57] * ds2 waits for the image has full suspend/resume work as verified by a meter ;)
[19:15:08] <geckosenator> that would be nice
[19:15:19] * torus (n=nils@c130105.adsl.hansenet.de) Quit (Remote closed the connection)
[19:15:29] <geckosenator> well the new pm stuff doesn't it use so low power idle that most people won't need to suspend?
[19:15:34] <jkridner> so, do we let sakoman's concerns stand?
[19:15:50] <ds2> and there is someone that can do it... just don't think a board is worth his time though ;)
[19:15:55] <jkridner> are we ok with 9 judges?
[19:16:38] <jkridner> dirk2: where should I look for the wiki entry?
[19:17:34] <dirk2> jkridner: Sorry, I mean: yes, please create a page.
[19:17:42] <jkridner> k. will do.
[19:17:47] <sakoman> jkridner: it will be obvious if the time frame is too short
[19:17:59] <sakoman> by the number of entries
[19:18:01] <ds2> does projects posted prior to today count?
[19:18:02] <jkridner> :)
[19:18:06] <koen> geckosenator: the new pm stuff needs work to do low power properly with usb and mmc
[19:18:18] <Crofton|work> I tend to agree with time frame issue
[19:18:28] <jkridner> yes, I think that projects already registered can count.
[19:18:43] <ds2> Hmmm
[19:18:50] <jkridner> I think we can require that all entries make a project registration.
[19:19:07] * soman (n=somnath@122.169.161.41) Quit (Remote closed the connection)
[19:19:13] <ds2> posted as in released... thinking specifically of the Android port
[19:19:15] <jkridner> so, I think some tweaks to something that exists would be good enough.
[19:19:19] <geckosenator> koen: ah ok
[19:19:28] <sakoman> a short time frame will limit how agressive people will be with what they attempt
[19:19:45] <ds2> IMO, 3 weeks is more then plenty
[19:20:27] <jkridner> sakoman: that's why I'm looking for fit-and-finish to be a top criteria--make the project about putting something together, rather than creating it.
[19:20:42] <sakoman> i don't have a fundamental issue with 3 weeks, just asking the question
[19:21:19] <RobertK> I think 3 weeks are okay.
[19:21:42] <ds2> there are complete trade show demos done in less then that time
[19:22:01] <RobertK> ds2: I know???
[19:22:06] <ds2> is there a write up requirement, i.e. docss on what to look for
[19:22:59] <jkridner> we are creating a wish-list of features on a wiki page and we will select which ones will be used as giving points for the contest
[19:23:14] <ds2> I mean from the contestant
[19:24:07] <jkridner> ah, excellent point.
[19:24:58] <ds2> don't want each entry to be a easter egg hunt
[19:25:56] <dirk2> an other question: Do we accept applications on OMAP3 non-Beagle boards? E.g. cool Overo stuff?
[19:26:54] <RobertK> I think running on bb is a must.
[19:27:55] <jkridner> here's my start at the wiki entry: http://elinux.org/BeagleBoard/contest
[19:28:00] <mru> hi guys
[19:28:01] <ds2> maybe dual boot images that work perfectly for both overo and bb?
[19:28:17] <jkridner> hi mru.
[19:28:19] <atin_> is there something fundamental wrong with wanting to have net/usb drivers in u-boot? so we can boot off tftp, mount nfs etc?
[19:28:50] * chelli (n=chelli@p54B86897.dip.t-dialin.net) has joined #beagle
[19:28:56] <jkridner> mru: I think it is clear you have first and second nominations to be a judge for this contest.
[19:29:02] <jkridner> would you accept?
[19:29:12] <jkridner> can I get a second for RobertK?
[19:29:40] <mru> I'd be honoured
[19:29:53] <jkridner> mru: great.
[19:30:02] <atin_> what does stable musb entail?
[19:30:12] * mru gets to judge people, bwahahaha
[19:30:28] <mru> is there an anti-prize for the loser?
[19:30:35] <ldesnogu_> mru, what a change :p
[19:30:37] <jkridner> I think we are at: sakoman, ds2, koen, dirk2, jkridner, mru, and pending: RobertK, khasim, Geof C (did I miss anyone)?
[19:30:53] <mru> ldesnogu_: oh, that was my evil twin
[19:31:00] <ldesnogu_> :)
[19:31:44] <jkridner> atin_: check with ds2 on that one. I'll be looking for him to expand on that wishlist item since he brought it up.
[19:31:53] <jkridner> I probably missed some wishlist items from this conversation.
[19:32:01] <ldesnogu_> jkridner, you put SGX in the wishlist; has the release made any progress?
[19:32:07] <atin_> ok - I seem to have time right now and want something to work on :)
[19:32:21] <jkridner> progress, yes, released, no.
[19:32:39] <jkridner> I'll leave it on the wishlist, but don't expect it to move to the approved category.
[19:32:55] <atin_> trying to get net/usb drivers into u-boot is feeling like too big a job right now, maybe I should first figure out usb itself.
[19:32:59] <ds2> atin_: 1. ISOC devices enumerate and work (test here is a hand full of GSPCA cameras); 2. USB2 TV tuners work (again, I have a test device)
[19:33:02] <ldesnogu_> oh I had missed the approved criteria section :)
[19:33:30] <ds2> and a general catch all - any device that works on x86 should work on the beagle as long as there is no x86 specific binary peice
[19:35:10] <ds2> and one other detail - hot plug must work and basic OTG functionality must be there (can plug in a B cable and expect gadget to work; plug in A cable and expect host to work)
[19:35:16] <ds2> that should be all ;)
[19:35:34] <atin_> gah - I have like 5 pieces of usb hardware, one really old vicam camera, a bluetooth dongle, an ethernet dongle and keyboard, mouse. I haven't tried the vicam, but the rest work so I can't actually test this...
[19:35:49] <atin_> tho hotplug I can I guess.
[19:36:30] <dirk2> atin_: Do you talk about OMAP3 MUSB support for U-Boot?
[19:36:33] <ds2> atin_: lack of hardware in the right hands seems to play a large role inthis unfortunately... and I seem to run in the troublesome hw too often :(
[19:36:54] <jkridner> If we can get one more non-TI judge, we'll have an odd number.
[19:36:59] <jkridner> RobertK?
[19:37:16] <RobertK> jkridner: here
[19:37:24] <atin_> dirk2 - that is what I was thinking, it is a pain to have to keep writing the SD card - I wish the ethernet dongle would work in u-boot so I can config_dhcp, config_net and have it boot off tftp
[19:37:32] <jkridner> btw, everyone has been very polite. no negative back-channel comments about why these people shouldn't be a judge yet. :)
[19:38:05] <atin_> you're kidding - these are the people who are writing all of the software that makes bb do anything at all, who is going to question that! :)
[19:38:08] <mru> maybe people know what the consequences would be...
[19:38:10] <jkridner> dirk2, ds2, sakoman, mru: can I get a second for RobertK?
[19:38:25] * guillaum1 (n=Guillaum@AMontsouris-153-1-20-180.w86-212.abo.wanadoo.fr) has joined #beagle
[19:38:32] <dirk2> atin_: TI is already working on this: http://lists.denx.de/pipermail/u-boot/2008-December/044905.html
[19:39:01] <jkridner> RobertK: are you willing?
[19:39:09] <RobertK> yes
[19:39:15] <atin_> cool! when tho? where did they get submitted?
[19:39:28] <atin_> and how do I get them :)
[19:39:48] <dirk2> atin_: They are not submitted :( Only second version of DaVinci patches.
[19:39:57] <jkridner> k. you make #7. Geof C and Khasim are less critical, since they work at TI. :)
[19:40:08] <jkridner> just didn't want to pause on an even number.
[19:41:15] <dirk2> atin_: We discussed a little and idea was to not do it only for Davinci, but include OMAP3, too. Sounds like this would be some additional work they will do beginning on January. Best you ask them directly before doing anything your self.
[19:41:19] <jkridner> RobertK: am I right to assume you are Robert Kuhn?
[19:41:30] <RobertK> yes, i am.
[19:41:35] <sakoman> jkridner: yes, you have my second
[19:41:36] <atin_> ok, thanks dirk
[19:42:11] <ldesnogu_> jkridner, you could perhaps categorize the criteria, for instance a section dedicated to applications (use of DSP, SGX, unless you meant support)
[19:42:26] <ds2> jkridner: seconded
[19:42:40] <dirk2> atin_: Best you ask Thomas Abraham, see his mail from today in the U-Boot archives.
[19:42:57] <jkridner> ldesnogu_: I'd like to see the judges help with organizing the wishlist. :)
[19:42:59] <atin_> ok - I was just going thru the archives for that.
[19:43:09] * Wowbagger_ (n=wowbagge@d154-20-138-76.bchsia.telus.net) has joined #beagle
[19:43:43] <ldesnogu_> jkridner, yeah they have to realize this is some real work :)
[19:43:43] <jkridner> since we have good turn-out, would having a phone call maybe 3-4 hours earlier than now on tomorrow be suitable for discussing details and sending out an announcement on the mailing list?
[19:44:01] <mru> we shouldn't be too strict with criteria
[19:44:11] <mru> I'm on an airplane that time tomorrow
[19:44:29] <jkridner> mru: I agree with that, but we should let people know some things that will win them points.
[19:44:39] <jkridner> cool factor should be #1.
[19:44:45] <mru> open source should be an absolute requirement
[19:44:57] <ds2> email thread prehaps? since we are distributed over many timezones
[19:44:59] <ldesnogu_> I hope we'll see some demo like in the old days!
[19:45:06] <jkridner> wow us and you'll win. :)
[19:46:17] <sakoman> either phone call or email is ok with me
[19:46:17] <dirk2> sorry, have to leave, bbl
[19:46:50] <jkridner> bye dirk2
[19:47:07] <jkridner> k, I'll start an e-mail thread on the mailing list.
[19:47:07] <RobertK> email would be better - I will have a christmas party in my baby's kindergarten
[19:47:17] <jkridner> will schedule a call only if needed to tie up loose ends.
[19:47:58] * dirk2 (n=dirk@92.117.36.240) has left #beagle
[19:48:16] * Xerion (i=xerion@82-170-197-160.ip.telfort.nl) has joined #beagle
[19:49:05] <ldesnogu_> jkridner, I am not sure I will have time to enter though I'd love to
[19:49:21] <koen> ldesnogu_: http://linux.slashdot.org/article.pl?sid=08/12/18/1458221
[19:49:29] <jkridner> I would love for you to as well. simple entries could very well win.
[19:49:59] <ldesnogu_> yeah and I will put in all the things I've said I'd work on and had no time to do :p
[19:50:19] <jkridner> There is a vendor looking to sell programmed SD cards. I'm going to encourage them to sell cards with the winning image.
[19:50:50] <ldesnogu_> koen, Linux Year has been announced for almost ten years now :)
[19:50:53] <jkridner> Hopefully that will add to the loot for the winner.
[19:51:46] <jkridner> well, I have other errands too. it was great getting so much accomplished so fast. I'm very glad you are all on the IRC!
[19:52:06] <jkridner> I'll send out a work-in-progress announcement later today.
[19:54:12] <ldesnogu_> koen, this blog is so dumb, but I am so biased too...
[19:54:38] <koen> ldesnogu_: I haven't even read that blog :)
[19:54:51] <ldesnogu_> however I agree with his conclusion that 2009 won't be the year of Linux
[19:55:05] <ldesnogu_> this will never be the year of Linux I'm afraid
[19:55:28] * scruggs (n=chris@72-161-120-114.dyn.centurytel.net) Quit (Read error: 110 (Connection timed out))
[19:55:52] * scruggs (n=chris@72-161-125-188.dyn.centurytel.net) has joined #beagle
[19:56:43] * adj_ (i=antti@hervanta.com) Quit (kubrick.freenode.net irc.freenode.net)
[19:56:43] * hadara_ (i=hadara@bsd.ee) Quit (kubrick.freenode.net irc.freenode.net)
[19:56:43] * geist (n=geist@c-69-181-145-146.hsd1.ca.comcast.net) Quit (kubrick.freenode.net irc.freenode.net)
[19:58:00] * mike18 (n=Beagle3@CPE-76-85-180-121.neb.res.rr.com) Quit ()
[19:58:56] * Beagle9 (n=Beagle9@76.85.180.121) has joined #beagle
[20:00:16] * geist (n=geist@c-69-181-145-146.hsd1.ca.comcast.net) has joined #beagle
[20:00:16] * adj_ (i=antti@hervanta.com) has joined #beagle
[20:00:16] * hadara_ (i=hadara@bsd.ee) has joined #beagle
[20:00:44] * mike18 (n=asdf@76.85.180.121) has joined #beagle
[20:01:59] <RobertK> bye
[20:02:04] * RobertK (n=rkuhn@brln-4db9ffa4.pool.einsundeins.de) Quit ("Tschau")
[20:04:31] * fille_ (n=fille@a83-245-252-47.elisa-laajakaista.fi) has joined #beagle
[20:04:48] <bkero> The last couple years have already been the year of linux
[20:05:09] <bkero> The measurements are all just wrong :)
[20:05:28] * fille_ is now known as philippe
[20:06:50] <ldesnogu_> bkero, for me it's all been the year of Linux for 10 years :-)
[20:07:14] <bkero> I gauge how much of a 'year of linux' it is by how much Microsoft fucks me on a daily basis.
[20:07:22] <bkero> That stopped a couple years ago, so I consider it a success.
[20:07:42] <ldesnogu_> bkero, hope you'll never get a job where you must use MS stuff
[20:08:04] <ldesnogu_> as a developer it's a real nightmare
[20:08:30] <geckosenator> year of linux?
[20:08:47] <geckosenator> I have used linux for 9 years
[20:08:53] <bkero> ldesnogu_: Already have a job I love at a company where people despise it as much as I do.
[20:08:56] <n6pfk> Is there a forth system for the beagle?
[20:09:02] <geckosenator> n6pfk: of course
[20:09:13] <n6pfk> url?
[20:09:16] <geckosenator> well it isn't going to be beagle specific
[20:09:25] <geckosenator> just run a forth
[20:09:27] <n6pfk> Pygmy?
[20:09:39] <geckosenator> I wrote a forth in scheme once
[20:09:52] <geckosenator> sure
[20:10:04] <ldesnogu_> bkero, I had to fight 3 years to get Linux but finally got it thanks in part for my BeagleBoard needs :)
[20:10:07] <geckosenator> oh you want it to compile to native arm instructions
[20:10:16] <sakoman> before I go digging into the TRM . . .
[20:10:28] <bkero> ldesnogu_: that was not the eyar of linux for you then :)
[20:10:50] <geckosenator> when I was working, I ported everything to compile on linux
[20:10:54] <ldesnogu_> bkero, it was through X :)
[20:10:55] <sakoman> does anyone know if there is an easy way to tell whether you are running on a 3503 or a 3530?
[20:10:59] <n6pfk> I want to port thr OFW vocabularies and develope device tree for the beagle for diagnosics an driver development.
[20:11:02] <bkero> ldesnogu_: It's gotten to the point where I have taken all windows experience off my resume. I won't work at a company that makes me use it.
[20:11:28] <geckosenator> bkero: that's good
[20:11:59] <n6pfk> And write some videos games in forth for fun.
[20:12:08] * Fl1pFl0p (n=fl1pfl0p@ip68-8-230-62.sd.sd.cox.net) Quit (Read error: 110 (Connection timed out))
[20:12:28] <geckosenator> I got in trouble for writing programs that ran fine on linux but crashed on windows
[20:12:41] <geckosenator> fortunately I quit that job
[20:12:58] * RobotGuy (n=robotguy@pool-96-225-210-142.ptldor.fios.verizon.net) has joined #Beagle
[20:13:09] <geckosenator> you can actually do things in windows that hard-reset the machine
[20:13:35] <bkero> There's no way to protect windows against a forkbomb
[20:13:39] <bkero> You -cant- lock it down
[20:13:42] <geckosenator> heh
[20:13:56] <geckosenator> and the way it handles shared libraries is broken
[20:14:10] <ldesnogu_> yeah but now with Vista it opens 10 dialog boxes to make sure you really want it to crash
[20:14:18] <ldesnogu_> that's progress
[20:14:43] <geckosenator> I know!
[20:14:56] <geckosenator> and it autorespawns
[20:15:01] <geckosenator> if a program crashes
[20:15:26] <sakoman> koen: just did my weekly-ish build of beagleboard-demo-image. looks like file manager is broken again. may be theme related, since theme dowloads & selection also see broken :-(
[20:15:43] <ldesnogu_> I found XP not bad for running games, but the PS3 OS is better at that
[20:15:47] <koen> the e-wm filemanager?
[20:15:52] <sakoman> yes
[20:16:07] <koen> I noticed that as well, thought it was a bug on my side
[20:16:18] <sakoman> a number of things don't seem right
[20:16:54] <sakoman> system menu isn't populated with shutdown, restart, etc any more
[20:17:39] <sakoman> guess it is time to look at checkins to see what changed
[20:17:56] <sakoman> have to run into town first though
[20:19:28] <koen> sakoman: the SRCREV got updated and a 'sysactions' thingy got added
[20:20:12] <sakoman> revert! revert!
[20:20:35] <sakoman> I'll look when I get back :-)
[20:20:50] <sakoman> unless you beat me to it ;-)
[20:22:33] <Crofton|work> anyone here using verizon dsl?
[20:23:29] <koen> can you hear me now?
[20:24:19] <Crofton|work> yes :)
[20:24:57] * RobotGuy (n=robotguy@pool-96-225-210-142.ptldor.fios.verizon.net) Quit (Remote closed the connection)
[20:26:25] * rsalveti (n=salveti@200.184.118.130) has joined #beagle
[20:31:37] * nemequ (n=nemequ@ip68-111-215-155.sd.sd.cox.net) has joined #beagle
[20:34:05] * MostAwesomeDude is now known as MAD|sickness
[20:40:36] * garren (n=garren@dsl-243-85-62.telkomadsl.co.za) Quit ("Ex-Chat")
[20:49:57] * Xenion (n=robert@p579FCC6C.dip.t-dialin.net) has joined #beagle
[20:50:42] <Xenion> Guten Abend jungs ! :-)
[20:51:13] * nemequ (n=nemequ@ip68-111-215-155.sd.sd.cox.net) Quit ("Ex-Chat")
[20:58:45] * florian (n=fuchs@g227178217.adsl.alicedsl.de) has joined #beagle
[21:02:19] <florian> re
[21:03:17] * robtow (n=rtow@nat/montavista/x-50670df498ba18df) Quit ("Leaving.")
[21:03:42] <koen> florian: http://lists.linuxtogo.org/pipermail/angstrom-distro-devel/2008-December/002830.html
[21:03:52] * Wowbagger_ (n=wowbagge@d154-20-138-76.bchsia.telus.net) Quit ("Leaving")
[21:04:01] <koen> florian: patches to add more GPE packages accepted :)
[21:06:34] <florian> koen: very nice...
[21:06:43] <florian> patches... oh dear, I need more time
[21:07:01] <florian> first I need to fix gpe-info
[21:07:02] * robtow (n=rtow@nat/montavista/x-b741fb31d006540d) has joined #beagle
[21:07:14] <koen> the crashes?
[21:07:27] <koen> I can't find why it sometimes workes and sometimes doesn't
[21:07:28] <florian> yes
[21:07:51] <koen> everytime I sit down and plan to gdb it it starts working :(
[21:08:03] <geckosenator> heh
[21:08:15] <florian> I guess its some stack corruption... maybe on ARM devices it does not know only.
[21:08:44] <koen> I have screenshots of it running on the beagle, so it works at least sometimes :)
[21:09:11] <Vegar> have you tried valgrind?
[21:09:36] <koen> valgrind is an x86 emulator, so it won't work on arm
[21:09:39] <Vegar> err.. perhaps valgrind doesn't run on arm
[21:09:41] <koen> but you could use it on the x86 binary
[21:09:57] <koen> well, valgrind runs on arm, but as I said, it's an x86 emulator :)
[21:09:58] <Vegar> I should learn to think before speaking
[21:10:25] <koen> it took me a while to learn that about valgrind :)
[21:10:47] <Vegar> hehe
[21:16:53] <mru> what? valgrind runs on arm?
[21:17:45] <florian> koen: hum well, but in this case it won't help us much ;)
[21:18:53] * RyoKimball (n=Ryo_Kimb@75-120-204-2.dyn.centurytel.net) has joined #beagle
[21:20:15] <RyoKimball> I probably won't be here to get the answer, but...
[21:20:15] <RyoKimball> Can anyone suggest a good display to purchase with the Beagle? I am wanting to make it a portable device. Touchscreen is a plus; capacitive/multitouch doubly so.
[21:22:01] * philippe (n=fille@a83-245-252-47.elisa-laajakaista.fi) Quit (Read error: 60 (Operation timed out))
[21:26:15] * n6pfk (n=mike@c-76-104-40-104.hsd1.va.comcast.net) Quit ("Leaving")
[21:32:42] * Adventus (i=82384119@gateway/web/ajax/mibbit.com/x-a503131694713a16) has joined #beagle
[21:34:47] <geckosenator> hi
[21:34:49] * skipisz (n=a0867407@nat/ti/x-8a5aca685fdc6d2b) has left #beagle
[21:35:00] <geckosenator> RyoKimball: you want to detect pressure on the touchscreen?
[21:35:23] * JoeyBorn (n=jborn@dsl017-022-252.chi1.dsl.speakeasy.net) Quit (Read error: 145 (Connection timed out))
[21:35:45] <RyoKimball> Huh... I never thought about it, but that could be useful. So, yes if it's available.
[21:36:01] <geckosenator> well you said you wanted capacitive
[21:36:06] <geckosenator> most touchscreens are resistive
[21:36:12] <RyoKimball> multi-touch capability is more important to me.
[21:36:26] <RyoKimball> Which resistive doesn't have, to my understanding.
[21:38:24] * philippe (n=fille@a83-245-252-47.elisa-laajakaista.fi) has joined #beagle
[21:39:24] <Adventus> Hiya
[21:39:37] <Adventus> At the gp32x forums were still getting very low FP nbench scores.
[21:40:36] <mru> guess they haven't tweaked things properly
[21:41:03] <mru> to get any decent FP performance you need to use single precision only and enable runfast mode
[21:41:12] <mru> the details are in the manual
[21:41:21] <Adventus> They've done that, the threads here http://www.gp32x.com/board/index.php?showtopic=45583&st=40
[21:41:49] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
[21:42:11] <Crofton|work> http://valgrind.org/info/platforms.html
[21:42:12] <Adventus> They changed it single precision.... enabled runfast and compiled libm with vfp.
[21:43:42] <Adventus> We get a max FP score of 1.2
[21:44:00] <Adventus> at 500mhz
[21:44:10] <mru> where would I go to download the code?
[21:44:45] <mru> getting good floating point performance from cortex-a8 is a bit of a black art
[21:44:59] <Adventus> I guess you would have to contact notaz, I'm sure he would give it to you.
[21:45:07] <Adventus> haha yea it seems that way
[21:45:12] <geckosenator> isn't there a way to use genetic algorithms to generate optimal assembly?
[21:45:18] <geckosenator> it's called a superoptimizer
[21:46:09] <mru> isn't the code public?
[21:47:06] <Adventus> Possibly, i don't know where the Pandora repo is though or if notaz has uploaded it.
[21:47:12] * Wowbagger_ (n=wowbagge@d154-20-138-76.bchsia.telus.net) has joined #beagle
[21:49:10] * RyoKimball (n=Ryo_Kimb@75-120-204-2.dyn.centurytel.net) has left #beagle
[21:53:32] <mru> seems to be this: http://www.tux.org/~mayer/linux/bmark.html
[21:55:36] <Adventus> Oh yea thats probably the original source code, i thought you meant the notaz modified version
[21:55:42] <Adventus> sorry
[21:55:54] <mru> what modifications did he make?
[21:56:40] <Adventus> On the thread he says he just changed it all to floats, and i guess used sinf() etc.
[21:59:28] <Adventus> so nothing major.
[21:59:58] <mru> wow, that code is ugly
[22:00:13] <florian> koen: I really wonder why this gpe-info segfault happens in Angstrom. The line gdb points to should never run since there is no log buffer in Angstroems busybox.
[22:01:34] * koen wonders about -fsingle-precision-constant
[22:01:34] <koen> Treat floating point constant as single precision constant instead of implicitly converting it to double precision constant.
[22:01:50] <mru> does what it says I guess
[22:01:59] <koen> ldesnogu_: did you say quake2 ran faster with -fsingle-precision-constant ?
[22:02:07] * NishanthMenon (n=nMenon@nat/ti/x-a81b318837ec7dbb) has joined #beagle
[22:02:58] * NishanthMenon (n=nMenon@nat/ti/x-a81b318837ec7dbb) Quit (Client Quit)
[22:06:21] <ldesnogu_> koen, I said that based on my experience on PS2 Linux which does not support HW DP FP
[22:06:50] <ldesnogu_> you should give it a try :-)
[22:07:11] <ldesnogu_> but I guess that won't make things much faster on Cortex-A8
[22:07:50] * a0133052 (n=a0133052@nat/ti/x-0d48f88a578a0ef0) has joined #beagle
[22:07:53] <koen> ldesnogu_: -fsingle-precision-constant + -ftree-vectorize should emit NEON instructions, right?
[22:10:45] * a0133052 (n=a0133052@nat/ti/x-0d48f88a578a0ef0) Quit (Remote closed the connection)
[22:12:33] <ldesnogu_> koen, you may also have to change calls to libm functions that take double (cos -> cosf for instance)
[22:14:06] * philippe (n=fille@a83-245-252-47.elisa-laajakaista.fi) Quit (Read error: 148 (No route to host))
[22:19:46] * Raja (n=Raja@nat/ti/x-58ae2d1f76985400) has joined #beagle
[22:22:14] <mru> Adventus: add -fno-math-errno to the flags too
[22:22:53] * RobotGuy (n=robotguy@pool-96-225-210-142.ptldor.fios.verizon.net) has joined #Beagle
[22:24:41] <Adventus> yea thats a possibility, although i think they tried -ffast-math.
[22:24:50] <mru> they do different things
[22:25:01] <ldesnogu_> mru, notaz already added the flag as I hinted
[22:25:11] <mru> oh, missed that
[22:25:20] * mru can't comprehend web forums
[22:25:26] <ldesnogu_> that benchmark is useless anyway
[22:25:30] <ldesnogu_> look at Fourier code
[22:25:40] <ldesnogu_> it uses cos, sin and pow ad nauseum
[22:25:55] <mru> looking at any of the code makes me want to kill someone
[22:25:56] * ssvb (n=ssvb___@a88-114-221-132.elisa-laajakaista.fi) has joined #beagle
[22:26:14] * ldesnogu_ wonders that will happen to the participants of the contest
[22:26:30] <Adventus> haha
[22:26:55] <mru> they'd be wise to check their front door locks
[22:27:07] <ldesnogu_> :)
[22:27:49] <mru> I'm sure we'll get some worthy contestants
[22:28:03] * Wowbagger_ (n=wowbagge@d154-20-138-76.bchsia.telus.net) Quit ("Leaving")
[22:28:06] * nm_ (n=nmn@nat/ti/x-7d4749ddd6cd12e3) has joined #beagle
[22:28:10] * nm_ (n=nmn@nat/ti/x-7d4749ddd6cd12e3) Quit (Client Quit)
[22:28:22] * RobotGuy is now known as ke7yut
[22:28:52] <Adventus> has any BB fp benchmarks been made?
[22:28:55] <ldesnogu_> I hope some people will use the DSP
[22:29:27] <ldesnogu_> Adventus, why are people obsessed by FP on Cortex-A8? It's just bad at it unless you use SIMD NEON code
[22:29:46] <mru> ldesnogu_: because you haven't finished the A9 yet
[22:30:05] <Adventus> Because were lazy game programmers, who like floats. :)
[22:30:07] <ldesnogu_> mru, heh a chip is never finished :)
[22:30:26] <mru> ldesnogu_: true, but we haven't even seen r0p0 silicon yet
[22:30:34] <ldesnogu_> I have :)
[22:30:43] <mru> that's different
[22:30:44] * NishanthMenon (n=nmenon@nat/ti/x-b3bb272c0c8caa2d) has joined #beagle
[22:31:04] <mru> would it be correct to assume there will be a realview board with A9?
[22:31:13] <ldesnogu_> probably
[22:31:23] <mru> not that I could afford one or anything
[22:31:27] <ldesnogu_> I think every core has been put on a board or another by ARM
[22:31:36] <ldesnogu_> yeah they're way too expensive
[22:31:36] * mru assumes anything without a published price is expensive
[22:31:59] <mru> how much are they?
[22:32:02] <mru> roughly
[22:32:20] <ldesnogu_> hum I only know the price it costs to ARM so I'm afraid I can say
[22:32:23] <ldesnogu_> can't*
[22:32:44] <mru> 4 digits?
[22:32:57] <ldesnogu_> let's say it's more expensive than an RVDS Professional package
[22:33:13] <mru> I don't know the price of that either
[22:33:23] <ldesnogu_> so definitely >= 4 digits
[22:33:44] <ldesnogu_> at least for boards with high perf cores on them
[22:33:54] * NishanthMenon (n=nmenon@nat/ti/x-b3bb272c0c8caa2d) Quit (Client Quit)
[22:33:58] <mru> I'm not interested in the other ones
[22:34:03] <prpplague> atleast 4 digits, maybe 5
[22:34:30] <mru> guess we'll have to wait for TI to make something
[22:34:38] <prpplague> the new s3c6400 sdk board is $5k and its not even that interesting
[22:34:42] <ldesnogu_> TI or some other licensee
[22:35:06] <mru> someone that can be persuaded to give me a board...
[22:35:33] <ldesnogu_> well you can't afford to accept any board
[22:35:47] <ldesnogu_> it might come with an A9 without NEON :)
[22:36:32] <mru> right then, wait for someone to make a board *with* neon and give me one
[22:36:36] * mrc3__ (n=mrc3@nat/ti/x-b22b94a1d4e54e76) has joined #beagle
[22:36:52] <mru> the first part is bound to happen
[22:37:18] <ldesnogu_> the second is less obvious
[22:37:21] <ldesnogu_> :)
[22:37:55] <ldesnogu_> RealView PB A9 is mentioned as planned here: http://www.arm.com/pdfs/RealView_HW_Platforms_Selector_v10.pdf
[22:38:12] <ldesnogu_> it's the big expensive board
[22:39:21] * mik18 (n=asdf@76.85.180.121) has joined #beagle
[22:44:51] * mike18 (n=asdf@76.85.180.121) Quit (Read error: 60 (Operation timed out))
[22:48:13] * ke7yut is now known as RobotGuy
[22:48:49] * nemequ (n=nemequ@ip68-111-215-155.sd.sd.cox.net) has joined #beagle
[22:51:40] * RobotGuy is now known as ke7yut
[22:57:11] * Beagle9 (n=Beagle9@76.85.180.121) Quit (Read error: 110 (Connection timed out))
[22:58:27] <jkridner|work> ldesnogu_: where is the beagleboard on that list?:)
[22:59:39] <ldesnogu_> jkridner|work, those are only ARM Ltd boards :-)
[22:59:40] * NishanthMenon (n=nmenon@nat/ti/x-deab898b29387ba5) has joined #beagle
[23:00:01] <jkridner|work> Beagle is open hardware. ARM is welcome to make them. :)
[23:00:31] <ldesnogu_> I don't even know who designs these boards
[23:00:45] <ldesnogu_> but many engineers are crying to get low cost boards
[23:00:58] <ldesnogu_> I have started to convert people to Beagle ;)
[23:01:02] <mru> jkridner|work: I could afford beagle boards even I didn't get them for free
[23:01:28] <ldesnogu_> so true
[23:01:41] <ldesnogu_> I hope that will motivate other companies to make low cost boards
[23:01:44] <mru> if someone made an A9 based board for under $500 I'd buy it
[23:01:53] <jkridner> you got lucky the first time and have earned every board since. :)
[23:02:11] <mru> jkridner: no luck involved
[23:02:15] <jkridner> no?
[23:02:17] <ldesnogu_> mru, are you sure you want to take the chance with a first generation silicon? :)
[23:02:33] <jkridner> Thomas knows what is up?
[23:02:34] <mru> jkridner: no, I asked nicely and they gave me one
[23:02:54] <jkridner> ah. then I should say that we were lucky.
[23:02:59] <mru> ldesnogu_: debugging is fun ;-)
[23:04:01] <ldesnogu_> mru, really? look at A8 errata for pre r1p2 and tell me that you'd have liked that :p
[23:04:22] <mru> the smiley is there for a reason
[23:04:47] <mru> and yes, some of those errata do look scary
[23:05:21] <mru> most of the errors seem to be in the integration of various parts
[23:05:22] <ldesnogu_> sorry, when time is getting close to go sleeping I lose my irony detector
[23:05:58] <ldesnogu_> the problem for all processors in general is in the memory subsystem
[23:06:07] <ldesnogu_> this part has become extremely complex
[23:06:25] <ldesnogu_> with dozens of requests running at the same time
[23:07:09] <ldesnogu_> and then add SMP to the equation and you'll feel the pain of the validation team :)
[23:07:25] <mru> I can sort of imagine
[23:07:41] <mru> debugging race conditions in software is bad enough
[23:09:25] <mru> but someone has to run ffmpeg on the chip or you'll never find all the bugs ;-)
[23:09:36] * JoeyBorn (n=jborn@h-68-164-11-5.chcgilgm.dynamic.covad.net) has joined #beagle
[23:13:31] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
[23:16:58] * guillaum1 (n=Guillaum@AMontsouris-153-1-20-180.w86-212.abo.wanadoo.fr) Quit ("Leaving.")
[23:17:16] * guillaum1 (n=gl@AMontsouris-153-1-20-180.w86-212.abo.wanadoo.fr) has joined #beagle
[23:18:11] <atin_> dang it. why the hell won't uinput show up ... I have built it as a kernel driver - have a new uImage, do I need to do something else?
[23:18:50] * ke7yut (n=robotguy@pool-96-225-210-142.ptldor.fios.verizon.net) Quit (Remote closed the connection)
[23:18:55] <atin_> if I build it as a module, I can insmod it
[23:19:27] <atin_> but if I build it into the kernel (CONFIG_INPUT_UINPUT=y), it won't show up
[23:19:45] <atin_> I don't need to modify the rootfs too do I?
[23:22:12] * abitos (n=nixgibts@dslb-084-057-166-200.pools.arcor-ip.net) has joined #beagle
[23:25:06] * RogerMon1 (n=a0740758@nat/ti/x-a9760efc0a3b58dc) has left #beagle
[23:26:32] * Xenion (n=robert@p579FCC6C.dip.t-dialin.net) Quit (Remote closed the connection)
[23:29:40] * ldesnogu_ (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) Quit ()
[23:35:03] * jackster (n=chatzill@76-191-209-171.dsl.dynamic.sonic.net) has joined #beagle
[23:35:06] * jackster (n=chatzill@76-191-209-171.dsl.dynamic.sonic.net) Quit (Client Quit)
[23:35:17] * chelli (n=chelli@debian/developer/tschmidt) Quit ("IRC is just multiplayer notepad")
[23:37:35] <koen> atin_: you mean the devnode doesn't show up when you build it into the kernel?
[23:40:49] * Adventus (i=82384119@gateway/web/ajax/mibbit.com/x-a503131694713a16) Quit ("http://www.mibbit.com ajax IRC Client")
[23:49:33] * Raja (n=Raja@nat/ti/x-58ae2d1f76985400) Quit (Remote closed the connection)

These logs were automatically created by BeagleLogBot on irc.freenode.net using the Java IRC LogBot.

See http://microformats.org/wiki/mflogbot for more information.