Join the chat at beagleboard.org/chat
IRC Log for 2010-06-29
Timestamps are in UTC.
- [00:06:59] <bkinman>
Who are you with Redb3ard?
- [00:08:58] <Redb3ard>
Smartfield.com... and my skills are probably closer to a sysadmin's. No degree.
- [00:09:17] * pablo_ (~pablo@201.250.186.175) has left #beagle
- [00:14:51] * jrmuizel (~jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit (Quit: jrmuizel)
- [00:15:43] * jrmuizel (~jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
- [00:20:09] * whittenburg (~whittenbu@ip72-213-154-11.ok.ok.cox.net) has joined #beagle
- [00:21:37] * cwillu (~cwillu@cwillu.com) Quit (Remote host closed the connection)
- [00:22:21] * ppoudel (~chatzilla@129.114.246.141) Quit (Quit: ChatZilla 0.9.86 [Firefox 3.0.17/2010010604])
- [00:27:12] * bkinman (~bkinman@soenat3.cse.ucsc.edu) Quit (Quit: bkinman)
- [00:30:18] * whittenburg (~whittenbu@ip72-213-154-11.ok.ok.cox.net) Quit (Read error: Operation timed out)
- [00:32:30] * cwillu (~cwillu@cwillu.com) has joined #beagle
- [00:34:34] <djlewis>
needed rain this time and it all skirts around me. :(
- [00:42:32] <djlewis>
emeb, people are getting ship notifications from mouser for launchpad
- [00:46:12] <emeb>
djlewis: interesting - so mouser is fulfilling orders for them...
- [00:46:32] <djlewis>
so it appears
- [00:46:48] <emeb>
I got my bo notice last Thurs.
- [00:47:02] <emeb>
so I'm prolly _way_ down the list...
- [00:47:05] <djlewis>
got my bo notice last wed
- [00:47:13] <djlewis>
from TI
- [00:47:45] <emeb>
ditto
- [00:49:36] <djlewis>
ti order status is still a simple BO
- [00:58:25] * drinkcat (~nicolas@181-162.4-85.cust.bluewin.ch) Quit (Ping timeout: 240 seconds)
- [00:59:57] <ds2>
djlewis: did you move to PHX?
- [01:00:20] <djlewis>
Redb3ard, ping
- [01:00:29] <Redb3ard>
Hey.
- [01:00:41] <djlewis>
i found a windoze pinmux tool
- [01:00:51] <Redb3ard>
Which city is TI in again? Houston or Dallas?
- [01:00:55] <djlewis>
g
- [01:00:58] <djlewis>
Dallas
- [01:01:10] <Redb3ard>
Willing to move for a job?
- [01:01:21] <djlewis>
you tease
- [01:01:29] <Redb3ard>
Serious.
- [01:01:42] <Redb3ard>
I can't promise, I don't have hiring authority. But we're looking for engineers.
- [01:01:53] <Redb3ard>
It's tough, we're in Lubbock, and not so many are willing to move here.
- [01:02:16] <Redb3ard>
Mail me your resume and I'll give it to my boss tomorrow... we don't have an HR department.
- [01:02:20] <ds2>
what state is Lubbock?
- [01:02:27] <Redb3ard>
Texas,
- [01:02:36] <djlewis>
I'm a mostly self taught hdwre/softwre desigh with some college associated to it.
- [01:03:10] <Redb3ard>
If you're good, the degree doesn't matter so much.
- [01:03:49] <Redb3ard>
My email is john.oyler@smartfield.com.
- [01:04:19] <djlewis>
I lived in lubbock as a child in about 1960 :)
- [01:04:33] <ds2>
wow
- [01:05:02] <Redb3ard>
What's wow?
- [01:05:04] <djlewis>
make that 63, I was 10
- [01:05:28] <ds2>
the time
- [01:06:25] <djlewis>
ooh Lubbock is in a cool area
- [01:06:30] <Redb3ard>
Well, anyone who's savvy enough to be here in this channel is welcome to mail a resume to me. Just in case someone else sees it.
- [01:07:08] <djlewis>
how bib is lubbock now?
- [01:07:20] <djlewis>
and did you want the pinmux tool?
- [01:07:51] <Redb3ard>
I really want to be about 10 times smarter than I am. Though, even twice as talented would work.
- [01:08:16] <djlewis>
i'd like to earn more income without having to move.
- [01:08:57] <Redb3ard>
I don't know that we could necessarily pay more. And moving's usually part of the deal as well.
- [01:13:18] * ppotera_ (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) has joined #beagle
- [01:16:09] * espindola (~espindola@CPE001a704e2b6d-CM001225dd5348.cpe.net.cable.rogers.com) Quit (Quit: Ex-Chat)
- [01:16:53] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) Quit (Ping timeout: 260 seconds)
- [01:16:56] * ppotera_ is now known as ppotera
- [01:17:48] * cmurillo (~cmurillo@186.32.57.22) Quit (Quit: Leaving.)
- [01:26:04] * tconant (~tconant@68.111.35.226) Quit (Quit: tconant)
- [01:27:59] <djlewis>
it runs in windoze but will save for linux
- [01:28:07] <djlewis>
http://processors.wiki.ti.com/index.php/Pin_Mux_Utility_for_ARM_MPU_Processors
- [01:28:42] <djlewis>
it expects you to modify u-boot but it should still be useful, maybe.
- [01:29:10] * Zoxc (~zoxc@ti0128a340-dhcp0372.bb.online.no) Quit ()
- [01:38:29] * cmurillo1 (~cmurillo@186.32.57.22) has joined #beagle
- [01:43:20] * cmurillo1 (~cmurillo@186.32.57.22) Quit (Client Quit)
- [01:44:14] * khasim (~a0393720@192.163.20.231) has joined #beagle
- [01:52:05] <ds2>
6
- [01:52:33] <ds2>
the pinmux is not complicated to require utilites... read a table, stuff a number in a register. Done.
- [01:55:00] * whittenburg (~whittenbu@ip72-213-154-11.ok.ok.cox.net) has joined #beagle
- [01:56:16] <djlewis>
agreed, just offering more in case it was wanted
- [01:56:27] * Ceriand|work (~Ceriand@unaffiliated/ceriand) Quit (Quit: Leaving.)
- [02:05:40] * karan (~karan@202.3.77.11) Quit (Quit: Leaving)
- [02:15:38] * Ceriand|desktop (~Ceriand@unaffiliated/ceriand) has joined #beagle
- [02:25:11] * cmurillo (~cmurillo@186.32.57.22) has joined #beagle
- [02:25:59] * sombunall (~sawdis@modemcable203.209-56-74.mc.videotron.ca) has joined #beagle
- [02:26:00] * cmurillo (~cmurillo@186.32.57.22) Quit (Client Quit)
- [02:27:08] * maria1 (~mrodrigue@186.32.57.22) Quit (Quit: Leaving.)
- [02:28:50] * kanru (~kanru@61-30-10-70.static.tfn.net.tw) has joined #beagle
- [02:31:45] * espindola (~espindola@CPE001a704e2b6d-CM001225dd5348.cpe.net.cable.rogers.com) has joined #beagle
- [02:34:48] * calculus (~calculus@gentoo/user/calculus) Quit (Ping timeout: 252 seconds)
- [02:45:55] <djlewis>
turning this heat generator off.
- [02:51:24] * duffolonious (~bryan@75-168-95-114.mpls.qwest.net) Quit (Ping timeout: 265 seconds)
- [02:53:27] * duffolonious (~bryan@75-168-95-114.mpls.qwest.net) has joined #beagle
- [03:00:38] * cody_ (~cody@dslb-084-056-103-108.pools.arcor-ip.net) has joined #beagle
- [03:02:39] * calculus (~calculus@gentoo/user/calculus) has joined #beagle
- [03:03:58] * cody (~cody@dslb-084-056-090-113.pools.arcor-ip.net) Quit (Ping timeout: 276 seconds)
- [03:07:52] * whittenburg (~whittenbu@ip72-213-154-11.ok.ok.cox.net) Quit (Ping timeout: 260 seconds)
- [03:19:21] * soman (~somnath@118.102.130.6) has joined #beagle
- [03:24:46] * djlewis (~djlewis@75.15.64.195) Quit (Quit: Clolur me gone)
- [03:26:52] * _don_ (~don@nat/ti/x-jbwsdawpslwbdxcx) has joined #beagle
- [03:29:43] * hitlin37 (6eea0302@gateway/web/freenode/ip.110.234.3.2) has joined #beagle
- [03:40:17] * hitlin37 (6eea0302@gateway/web/freenode/ip.110.234.3.2) Quit (Quit: Page closed)
- [03:40:38] * mpoirier (~quassel@S0106002369de4dac.cg.shawcable.net) has joined #beagle
- [03:45:15] * hitlin37 (6eea0302@gateway/web/freenode/ip.110.234.3.2) has joined #beagle
- [03:45:53] * bkinman (~bkinman@c-67-188-116-157.hsd1.ca.comcast.net) has joined #beagle
- [03:45:58] * mpoirier (~quassel@S0106002369de4dac.cg.shawcable.net) Quit (Ping timeout: 264 seconds)
- [03:46:12] * aholler (~aholler@p57B22CDC.dip0.t-ipconnect.de) Quit (Ping timeout: 260 seconds)
- [03:46:51] * _don_ (~don@nat/ti/x-jbwsdawpslwbdxcx) Quit (Ping timeout: 252 seconds)
- [03:47:38] * aholler (~aholler@p57B21EBD.dip0.t-ipconnect.de) has joined #beagle
- [03:50:09] * whittenburg (~whittenbu@ip72-213-154-11.ok.ok.cox.net) has joined #beagle
- [03:52:04] <TheUni>
mru: got you a rebased branch. ready for my lashes and flames.
- [03:52:52] * emeb (~ericb@ip72-223-81-194.ph.ph.cox.net) Quit (Ping timeout: 276 seconds)
- [03:52:53] * khasim (~a0393720@192.163.20.231) Quit (Ping timeout: 240 seconds)
- [03:53:31] <CruNcher>
is the Mali 200 more efficient then a SGX 530/40 core ?
- [03:55:04] * calculus (~calculus@gentoo/user/calculus) Quit (Ping timeout: 260 seconds)
- [04:02:08] * calculus (~calculus@gentoo/user/calculus) has joined #beagle
- [04:03:21] * GrizzlyAdams1 (~Grizzly@ip98-184-88-41.mc.at.cox.net) Quit (Ping timeout: 252 seconds)
- [04:10:29] <TheUni>
mru: incase you read backlog: http://gitorious.org/theuni/ffmpeg/commits/rebased
- [04:10:31] <TheUni>
will pm you tomorrow
- [04:12:16] <_av500_>
TheUni: thx
- [04:12:48] <TheUni>
_av500_: ?
- [04:12:57] <_av500_>
ffmpeg patches
- [04:13:18] <TheUni>
oh. those won't help anyone but xbmc :)
- [04:13:48] <TheUni>
well, i suppose they'll help some.
- [04:14:15] <_av500_>
we'll see
- [04:16:53] <TheUni>
yea, there are a few reasonable fixes in there i think. others are probably poorly implemented but maybe can get some attention so they're done right
- [04:19:49] * GrizzlyAdams (~Grizzly@ip98-184-88-41.mc.at.cox.net) has joined #beagle
- [04:29:25] * sombunall (~sawdis@modemcable203.209-56-74.mc.videotron.ca) Quit (Quit: Leaving.)
- [04:32:19] * espindola (~espindola@CPE001a704e2b6d-CM001225dd5348.cpe.net.cable.rogers.com) Quit (Quit: Ex-Chat)
- [04:42:28] * notzed (~notzed@ppp118-210-117-15.lns20.adl2.internode.on.net) has joined #beagle
- [04:42:30] * emeb|mac (~ericb@ip72-223-81-194.ph.ph.cox.net) has joined #beagle
- [04:56:32] * hvaibhav (~a0393758@192.163.20.231) has joined #beagle
- [05:00:43] * mylesche (3d94388a@gateway/web/freenode/ip.61.148.56.138) has joined #beagle
- [05:01:09] * CruNcher (~luls_lol@dslb-094-219-152-017.pools.arcor-ip.net) Quit ()
- [05:01:57] * jonmasters (~jcm@dallas.jonmasters.org) Quit (Read error: Operation timed out)
- [05:03:55] * mylesche_ (3d94388a@gateway/web/freenode/ip.61.148.56.138) has joined #beagle
- [05:04:01] * mylesche (3d94388a@gateway/web/freenode/ip.61.148.56.138) Quit (Client Quit)
- [05:04:26] * niclas (~niclas@nat/ti/x-ywrfkuddfaldjnvu) Quit (Remote host closed the connection)
- [05:04:26] * mrc3_ (~mrc3@nat/ti/x-nhdwqarnfrmvnxrr) Quit (Remote host closed the connection)
- [05:04:26] * _koen_ (~x0115699@nat/ti/x-fipuavzvxqiewuea) Quit (Remote host closed the connection)
- [05:05:03] * mrc3_ (~mrc3@nat/ti/x-yhggrcshrxyffvvd) has joined #beagle
- [05:05:06] * niclas (~niclas@nat/ti/x-alfvznnrhllhfxuf) has joined #beagle
- [05:05:09] * _koen_ (~x0115699@nat/ti/x-gbnyfghxdheicanc) has joined #beagle
- [05:08:20] <mylesche_>
hi,there,is anyone help bout beagleboard nand flash driver?
- [05:08:21] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) Quit (Quit: ppotera)
- [05:09:38] <av500>
depens
- [05:09:40] <av500>
depends
- [05:11:18] * mrc3_ (~mrc3@nat/ti/x-yhggrcshrxyffvvd) Quit (Remote host closed the connection)
- [05:11:24] <ds2>
flat head, phillips, or torx?
- [05:11:36] * mrc3_ (~mrc3@nat/ti/x-bvvvrthgjgbhqkyu) has joined #beagle
- [05:12:04] * jonmasters (~jcm@dallas.jonmasters.org) has joined #beagle
- [05:13:00] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) has joined #beagle
- [05:16:44] <av500>
ds2: tri-wing of course
- [05:17:15] <mylesche_>
hello,is anyone here?
- [05:17:25] <av500>
ask, dont ask to ask
- [05:18:47] <mylesche_>
in uboot for beaglboard(3530), CFG_NAND_LEGACY is defined or not?
- [05:28:59] <mylesche_>
there are some actions like "nand->hwcontrol = omap_nand_hwcontrol;" in board_nand_init(struct nand_chip *nand)(locate in omap3530beagleboar\nand.c).but if CFG_NAND_LEGACY is defined ,there is no hwcontrol() in the struct nand_chip.If CFG_NAND_LEGACY is not defined,uboot will use mtd's nand.c,another nand.c for beagleboard3530 is useless ,why?
- [05:29:01] * Belna (~Thomas@DSL01.212.114.252.242.ip-pool.NEFkom.net) has joined #beagle
- [05:30:03] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) Quit (Quit: ppotera)
- [05:32:35] * bkinman (~bkinman@c-67-188-116-157.hsd1.ca.comcast.net) Quit (Quit: bkinman)
- [05:38:38] * khasim (~a0393720@192.163.20.231) has joined #beagle
- [05:46:25] * lag (~lag@cpc2-aztw21-0-0-cust264.aztw.cable.virginmedia.com) has joined #beagle
- [05:54:23] * amitk (~amit@a91-154-124-12.elisa-laajakaista.fi) has joined #beagle
- [05:55:14] * khasim (~a0393720@192.163.20.231) has left #beagle
- [05:58:11] * mylesche_ (3d94388a@gateway/web/freenode/ip.61.148.56.138) Quit (Ping timeout: 252 seconds)
- [06:01:52] * pting (~pting@adsl-99-32-245-41.dsl.lsan03.sbcglobal.net) Quit (Quit: Ex-Chat)
- [06:03:33] * GrizzlyAdams (~Grizzly@ip98-184-88-41.mc.at.cox.net) Quit (Quit: Leaving.)
- [06:04:20] * Sept (~bakljg@c-24-131-138-21.hsd1.mn.comcast.net) Quit (Ping timeout: 260 seconds)
- [06:06:04] * Redb3ard (~SF0010MAC@75.110.202.83) Quit (Quit: Redb3ard)
- [06:12:34] * niclas (~niclas@nat/ti/x-alfvznnrhllhfxuf) Quit (Remote host closed the connection)
- [06:12:34] * berco (~a0919194@nat/ti/x-pbimonseihheyvls) Quit (Remote host closed the connection)
- [06:12:34] * _koen_ (~x0115699@nat/ti/x-gbnyfghxdheicanc) Quit (Remote host closed the connection)
- [06:12:35] * rcranetx (~rcranetx@nat/ti/x-xhujxapcrahihebl) Quit (Remote host closed the connection)
- [06:12:52] * niclas (~niclas@nat/ti/x-ulowzttiwehboygg) has joined #beagle
- [06:13:15] * rcranetx (~rcranetx@nat/ti/x-lirejnxaeorbdyvg) has joined #beagle
- [06:13:28] * berco (~a0919194@nat/ti/x-gxgeuktlilufudpl) has joined #beagle
- [06:13:43] * _koen_ (~x0115699@nat/ti/x-napikitvvncqstbx) has joined #beagle
- [06:14:09] * topfs2 (~topfs2@xbmc/staff/topfs2) has joined #beagle
- [06:17:14] * tsjsieb (~tsjsieb@dejongbeheer.nl) has joined #beagle
- [06:24:35] * emeb|mac (~ericb@ip72-223-81-194.ph.ph.cox.net) Quit (Ping timeout: 265 seconds)
- [06:29:10] * virals (~Viral_Sac@122.166.2.35) has joined #beagle
- [06:30:39] <av500>
TheUni: ping
- [06:31:16] * hitlin37 (6eea0302@gateway/web/freenode/ip.110.234.3.2) Quit (Quit: Page closed)
- [06:31:33] * naam_ (6eea0302@gateway/web/freenode/ip.110.234.3.2) has joined #beagle
- [06:34:56] * screwgoth (~raseel@122.170.63.16) has joined #beagle
- [06:36:09] <ds2>
less then 12hours before we can heckle the students ;)
- [06:36:14] <screwgoth>
Hi, while trying to flash the BB's flash, as per : http://code.google.com/p/beagleboard/wiki/BeagleNANDFlashing , before going past "fatload mmc 0 0x80200000 flash-uboot.bin" the board rebooted
- [06:36:39] <screwgoth>
Not, I'm not seeing the uboot prompt ... I see messages only till "I2C: Ready"
- [06:36:46] <screwgoth>
Any way for me to recover from there
- [06:37:02] * lamp_ (bc9e4165@gateway/web/freenode/ip.188.158.65.101) has joined #beagle
- [06:37:02] <av500>
boot from SD
- [06:37:06] <av500>
hold the user button
- [06:37:44] <ds2>
sigh... all this mucking with u-boot is going to cause someone to fry theirs :/
- [06:38:53] <lamp_>
hi
- [06:40:30] <av500>
lo
- [06:40:34] <screwgoth>
ds2: :-)
- [06:40:44] <screwgoth>
av500: You mean reset to factory , right ?
- [06:41:06] <ds2>
screwgoth: no, he means boot the U-boot from SD
- [06:41:12] <lamp_>
conecting beagle to linux with com to usb convertor and config minicom but not see enything then try with hyperterminal in winxp and config
- [06:41:32] <av500>
lamp_: check cable
- [06:41:36] <lamp_>
but not see enything in hyperterminal
- [06:42:06] <ds2>
lamp_: Do cable diag per SRM.
- [06:42:07] <lamp_>
cheked cable true
- [06:42:40] <av500>
check cable more
- [06:42:50] <screwgoth>
ds2: hmm.... but shouldn't I get the you-boot prompt first , so that I enter the appropriate bootargs to boot from SD ?
- [06:42:57] <av500>
no
- [06:43:03] <ds2>
screwgoth: no.
- [06:43:46] <ds2>
av500: do you know if today's event will be simulcasted in shoutcast/icecast format?
- [06:44:08] <lamp_>
installing ubuntu 10.4 in sd an boot it but my kaboard not work and not see enything in com port
- [06:44:11] <av500>
ds2: ino nothin
- [06:44:14] <screwgoth>
ds2: ok thanks, lemme try
- [06:46:08] * mobidev (~mobidev@94.127.205.30) has joined #beagle
- [06:52:09] * franktango (~a0746747@nat/ti/x-vaqnnlcbhooappwj) has joined #beagle
- [06:54:02] * lirtex (~liorc@89-138-158-93.bb.netvision.net.il) has joined #beagle
- [06:55:35] <lirtex>
Hi, do i need to do anything "special" to get sound working on the bealeboard? I have alsa installed, i'm trying to play a wav file and it seems to play fine, but I cant hear anything. I'm using headphones, and I tried setting the volume high in the mixer (although I could not find a PCM channel there). Any idea?
- [06:55:54] * mobidev (~mobidev@94.127.205.30) Quit (Quit: I go offline...)
- [06:56:10] * calculus (~calculus@gentoo/user/calculus) Quit (Ping timeout: 264 seconds)
- [06:57:19] <lamp_>
da2_: cable cheked
- [06:57:38] <av500>
check how?
- [06:58:28] <ds2>
is that like 'checking' in hockey?
- [06:58:36] <lamp_>
whit ohmeter
- [06:59:11] <ds2>
how many oh's did it register? ;)
- [07:00:04] <lamp_>
0
- [07:01:03] <lamp_>
i test it again
- [07:01:12] <lamp_>
]& icheck howts ok
- [07:01:20] * screwgoth (~raseel@122.170.63.16) Quit (Quit: Leaving.)
- [07:01:21] <lamp_>
it si ok
- [07:02:43] <lamp_>
do u have any garanty?
- [07:03:03] <lamp_>
i think its about product
- [07:06:54] * jrmuizel (~jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit (Quit: jrmuizel)
- [07:07:57] <ds2>
i think you need to read the srm
- [07:08:22] * Proxyles (~henrik@c-f893e255.56-4-64736c14.cust.bredbandsbolaget.se) has joined #beagle
- [07:10:08] * plars (~plars@75-27-138-126.lightspeed.austtx.sbcglobal.net) Quit (Read error: Operation timed out)
- [07:13:20] * courville (~courville@archos.rain.fr) has joined #beagle
- [07:13:59] * lamp_ (bc9e4165@gateway/web/freenode/ip.188.158.65.101) Quit (Quit: Page closed)
- [07:15:39] * hrw|gone is now known as hrw
- [07:17:18] * plars (~plars@75-27-138-126.lightspeed.austtx.sbcglobal.net) has joined #beagle
- [07:17:44] * eFfeM_work (~frans@atwork-193.r-212.178.107.atwork.nl) Quit (Quit: Leaving.)
- [07:18:58] * topfs2 (~topfs2@xbmc/staff/topfs2) Quit (Ping timeout: 264 seconds)
- [07:23:30] * jpsaman (~jpsaman@videolan/developer/jpsaman) has joined #beagle
- [07:27:59] * topfs2 (~topfs2@xbmc/staff/topfs2) has joined #beagle
- [07:31:11] * Ceriand|desktop (~Ceriand@unaffiliated/ceriand) Quit (Read error: Connection reset by peer)
- [07:40:02] * _courvil1e_ (~marc@courville.org) Quit (Quit: leaving)
- [07:40:11] * _courville_ (~marc@courville.org) has joined #beagle
- [07:43:37] * florian_kc (~fuchs@Maemo/community/contributor/florian) has joined #beagle
- [07:43:48] * florian_kc is now known as florian
- [07:45:15] * naam_ (6eea0302@gateway/web/freenode/ip.110.234.3.2) Quit (Ping timeout: 252 seconds)
- [07:46:04] * KosiNuss (~tom@p4FD12276.dip0.t-ipconnect.de) has joined #beagle
- [07:50:09] * jpirko (~jirka@ip-89-103-91-100.net.upc.cz) has joined #beagle
- [07:52:12] * jpirko (~jirka@ip-89-103-91-100.net.upc.cz) Quit (Client Quit)
- [07:52:25] * jpirko (~jirka@ip-89-103-91-100.net.upc.cz) has joined #beagle
- [07:53:41] * _roger_ (~a0740758@nat/ti/x-wykkczazmmycdphu) has joined #beagle
- [08:07:43] * t_s_o (~tso@183.84-49-135.nextgentel.com) has joined #beagle
- [08:08:23] * neo01124 (~neo@122.177.3.3) has joined #beagle
- [08:10:43] * drinkcat (~nicolas@181-162.4-85.cust.bluewin.ch) has joined #beagle
- [08:11:18] * Try`0xff is now known as Tryum
- [08:16:13] * Artanis (Artanis@159.108.165.83.dynamic.mundo-r.com) Quit (Ping timeout: 264 seconds)
- [08:16:30] * Phrog (~chatzilla@ip-87-82-198-210.easynet.co.uk) has joined #beagle
- [08:18:16] * ceyusa (~vjaquez@155.99.117.91.static.mundo-r.com) has joined #beagle
- [08:19:20] * lirtex (~liorc@89-138-158-93.bb.netvision.net.il) Quit (Remote host closed the connection)
- [08:19:47] * lirtex (~liorc@89-138-158-93.bb.netvision.net.il) has joined #beagle
- [08:21:32] * xpc (~z@87-126-142-238.btc-net.bg) has joined #beagle
- [08:32:26] * chrisUSB (d456d2d3@gateway/web/freenode/ip.212.86.210.211) Quit (Quit: Page closed)
- [08:32:33] * cbenzUSB (d456d2d3@gateway/web/freenode/ip.212.86.210.211) has joined #beagle
- [08:32:53] * cbenzUSB is now known as chrisUSB
- [08:36:58] * neo01124 (~neo@122.177.3.3) Quit (Remote host closed the connection)
- [08:46:01] * Phrog (~chatzilla@ip-87-82-198-210.easynet.co.uk) Quit (Ping timeout: 245 seconds)
- [08:46:02] * tasslehoff (~Mich@147.84-49-231.nextgentel.com) has joined #beagle
- [08:52:35] * bearsh (~quassel@adsl-245-48-fixip.datacomm.ch) Quit (Read error: Connection reset by peer)
- [08:54:01] * bearsh (~quassel@adsl-245-48-fixip.datacomm.ch) has joined #beagle
- [08:54:18] * Phrog (~chatzilla@ip-87-82-198-210.easynet.co.uk) has joined #beagle
- [08:55:26] * jpirko (~jirka@ip-89-103-91-100.net.upc.cz) Quit (Remote host closed the connection)
- [08:58:00] * Artanis (Artanis@159.108.165.83.dynamic.mundo-r.com) has joined #beagle
- [08:59:15] * Phrog (~chatzilla@ip-87-82-198-210.easynet.co.uk) Quit (Ping timeout: 265 seconds)
- [09:01:18] * raster (raster@enlightenment/developer/raster) Quit (Quit: Gettin' stinky!)
- [09:03:01] * Artanis (Artanis@159.108.165.83.dynamic.mundo-r.com) Quit (Ping timeout: 264 seconds)
- [09:06:27] * Artanis (Artanis@159.108.165.83.dynamic.mundo-r.com) has joined #beagle
- [09:08:58] * ant_work (~andrea@host214-85-static.34-85-b.business.telecomitalia.it) has joined #beagle
- [09:12:12] <chrisUSB>
Hi, i am having kernel-panic lokups when running omapfbplay... could anyone have a look at my pastebin: http://pastebin.com/47h3Q5CS
- [09:13:01] <av500>
nice
- [09:14:14] <chrisUSB>
you think so? I just want it to play any video :-)
- [09:16:32] <av500>
I think I have seen that before here, but I cant remember
- [09:16:58] <av500>
are you running X?
- [09:17:23] <koen>
chrisUSB: what's the resoloution you are using?
- [09:17:40] * Proxyles (~henrik@c-f893e255.56-4-64736c14.cust.bredbandsbolaget.se) Quit (Remote host closed the connection)
- [09:18:08] * eFfeM_work (~frans@atwork-193.r-212.178.107.atwork.nl) has joined #beagle
- [09:18:12] <chrisUSB>
the thing is: I am not having an monitor attached. I just want it to play a video to have the cpu load..
- [09:18:19] <chrisUSB>
yes there is a x running;
- [09:19:45] <chrisUSB>
do not know the resolution now; bb is rebooting.. will have a look
- [09:22:52] * Proxyles (~henrik@c-f893e255.56-4-64736c14.cust.bredbandsbolaget.se) has joined #beagle
- [09:22:56] <chrisUSB>
is the screen resolution important?
- [09:23:37] <av500>
try without X
- [09:26:51] <chrisUSB>
if only i knew how to stop x :-); i remember having used stopx under debian or so....
- [09:33:47] <av500>
init 3
- [09:38:45] <chrisUSB>
ok thats what i did just before starting omapfbplay... so if init 3 suceeded ( i didn't get an error) i should have stopped x before...
- [09:39:38] <av500>
ps should tell you
- [09:40:33] <chrisUSB>
the process xorg .... should be missing?
- [09:40:45] <av500>
ps aux
- [09:40:52] <av500>
should show nothing x related
- [09:41:07] <av500>
still, what is your resolution?
- [09:41:32] <chrisUSB>
couldn't figure out what is configured by now...
- [09:42:17] * jpirko (~jirka@nat/redhat/x-kchjoustsycfaiym) has joined #beagle
- [09:42:27] <av500>
cat /proc/cmdline
- [09:43:24] <chrisUSB>
av500: will have a look soon; meanwhile did a run with an image without any module loaded and omapfbplay seems to run without crashing.
- [09:43:37] <chrisUSB>
but the 'crashing' bb is up now...
- [09:44:04] * Entasis (~Jarred@ppp118-210-194-202.lns20.adl6.internode.on.net) has joined #beagle
- [09:46:15] <chrisUSB>
cmdline: console=ttyS2,115200n8 mpurate=720 vram=12M omapfb.mode=dvi:1024x768MR-16@60 omapdss.def_disp=dvi root=/dev/mmcblk0p2 rw rootfstype=ext3 rootwait
- [09:46:32] <av500>
looks ok
- [09:47:47] <chrisUSB>
i executed init 3, but still see x11vnc, Xorg -br..., gpe-dm; is it ok having them still running?
- [09:48:09] <av500>
no
- [09:48:14] <chrisUSB>
:-)
- [09:48:22] <chrisUSB>
ok ill try killing
- [09:48:52] <koen>
why no do /etc/init.d/gpe-dm stop?
- [09:49:27] <chrisUSB>
didn't know about that parameter...
- [09:49:28] <av500>
koen: meh, newfangled stuff...
- [09:51:30] <chrisUSB>
koen: gpe-dm cat: /var/run/gpe-dm.pid: No such file or directory sh: you need to specify whom to kill
- [09:52:15] <chrisUSB>
koen: sorry incomplete: /etc/init.d/gpe-dm stop Stopping GPE display manager: gpe-dm cat: /var/run/gpe-dm.pid: No such file or directory sh: you need to specify whom to kill
- [09:53:35] * maltanar (d5a5598d@gateway/web/freenode/ip.213.165.89.141) has joined #beagle
- [09:53:40] <maltanar>
hi everyone
- [09:53:46] <av500>
hi
- [09:53:57] * negril_ is now known as negril
- [09:54:24] <maltanar>
I'm a bit confused with my timezones as usual - it's ~ two hours until the lightning talks q&a session, right?
- [09:55:03] <av500>
its UGT here, so no idea... :)
- [09:56:01] <koen>
it's almost noon here
- [09:56:09] <koen>
so, good morning maltanar!
- [09:56:19] <chrisUSB>
av500: any hints how to stop x, init3 seems not to be enough
- [09:56:52] <av500>
chrisUSB: no idea
- [10:01:16] <chrisUSB>
it seems as if gpe-dm stop is looking for pid's in /var/run/gpe-dm.pid... do you have this file?
- [10:02:33] <av500>
me?
- [10:02:36] <av500>
no
- [10:02:39] <av500>
I dont run a BB
- [10:02:42] <av500>
and not X either
- [10:03:10] <chrisUSB>
av500: ok thx
- [10:03:26] <chrisUSB>
with you i meant all those using bb with x
- [10:03:26] * maltanar remembers how it's always supposed to be morning while greeting people in IRC
- [10:03:46] * xpc (~z@87-126-142-238.btc-net.bg) Quit (Quit: Leaving)
- [10:06:00] <chrisUSB>
kill -9 all processes suceeded...
- [10:06:27] <av500>
kill -9 1
- [10:07:18] <chrisUSB>
av500: maybe i try it when i'm completely out of ideas... *g
- [10:07:44] <chrisUSB>
after having killed all x-things omapfbplay runs continuously..
- [10:07:52] <chrisUSB>
av500,koen: thx
- [10:07:53] <av500>
good
- [10:09:36] * ssvb (~ssvb___@a88-112-120-50.elisa-laajakaista.fi) Quit (Quit: Leaving)
- [10:12:40] * KosiNuss_ (~tom@p4FD11C94.dip0.t-ipconnect.de) has joined #beagle
- [10:14:48] * mcgeagh (~mcgeagh@xbmc/staff/mcgeagh) Quit (Quit: Ex-Chat)
- [10:15:11] * KosiNuss (~tom@p4FD12276.dip0.t-ipconnect.de) Quit (Ping timeout: 245 seconds)
- [10:24:13] * maltanar (d5a5598d@gateway/web/freenode/ip.213.165.89.141) Quit (Quit: Page closed)
- [10:26:50] * akumar (~mm4@122.170.63.16) has joined #beagle
- [10:28:23] <akumar>
i am using angstrom on beagle board c4
- [10:28:31] <akumar>
and kernel 29
- [10:29:04] <akumar>
but error, when "opkg update" command is fired on root prompt.
- [10:29:10] <akumar>
details: http://pastebin.com/uMyhZXuU
- [10:29:32] <akumar>
please any one suggest me, thanks
- [10:30:28] <_koen_>
what is kernel 29?
- [10:30:48] <_koen_>
I don't know what that is, but the 2.6.32 kernel from angstrom works great on my C4
- [10:31:48] <akumar>
_koen_: 2.6.32 kernel, it is giving error for opkg update.
- [10:32:42] <akumar>
_koen_: _koen_: 2.6.29 kernel, it is giving error for opkg update.
- [10:33:31] <_koen_>
well, stop using that old 2.6.29 kernel
- [10:34:01] <akumar>
_koen_: but i want to used 2.6.29 kernel
- [10:36:16] <_koen_>
then you're on your own
- [10:36:30] <av500>
akumar: why?
- [10:41:32] <lool>
Hey, I've got a beagle rev C3 and it hung overnight; I suspect it might be related to the rootfs on an USB key on an USB hub on the host port; I think there are some hardware issues with this port, but I don't know whether they are severe to the point they would hang the board
- [10:42:30] <akumar>
av500: it is client request, so that i am using.
- [10:45:33] * KosiNuss_ (~tom@p4FD11C94.dip0.t-ipconnect.de) Quit (Remote host closed the connection)
- [10:57:37] <ogra>
lool, there are some soldering instructions to work around that afaik
- [10:58:08] <ogra>
needs some capacitor to be soldered somewhere, i think hrw knows how exactly it works
- [10:59:49] <av500>
lool: yes, the EHCI on <C4 can have issues
- [11:00:09] <av500>
needs a 22uF soldered at the right place, but is not a 100% fix, works for some, not for others
- [11:03:03] <lool>
Hmm
- [11:03:13] <lool>
And the mini USB is stable as a host port?
- [11:03:27] <lool>
ogra: Do you know if the mini USB is well supported in maverick kernels?
- [11:03:51] <lool>
I'm moving to a SD card for the rootfs but it's quite limitating
- [11:03:55] <ogra>
lool, nope, havent tested it, GrueMaster does daily kernel tests, he could probably tell
- [11:04:01] <lool>
(in terms of performance and size)
- [11:04:07] <lool>
ogra: thanks
- [11:04:08] <ogra>
indeed
- [11:04:18] <hrw>
lool: 0.22uF capacitor between GND and 1.8V lines in expansion header
- [11:04:41] * ogra hands everyone a ?? char :)
- [11:05:01] <av500>
ogra: your h is upside down!
- [11:05:04] <hrw>
ogra: better hand keycombo
- [11:05:17] * lool tends to avoid hardware patches when he only has a single board and relies on it for critical work
- [11:05:46] <ogra>
hrw, lazy germans have a key for it ;)
- [11:05:49] <av500>
lool: you might also try to do some logging to see if it is really ehci....
- [11:05:56] <hrw>
ogra: ;D
- [11:06:04] <av500>
ALT-m
- [11:06:08] <av500>
ALT GR-m
- [11:06:45] <hrw>
av500: gracias
- [11:06:47] <hrw>
??
- [11:08:26] * t_s_o (~tso@183.84-49-135.nextgentel.com) Quit (Remote host closed the connection)
- [11:10:15] * theholyduck (~holyduck@ip-108-138-106-77.eidsiva.net) Quit (Read error: Connection reset by peer)
- [11:18:45] * akumar (~mm4@122.170.63.16) Quit (Ping timeout: 240 seconds)
- [11:20:25] * courville (~courville@archos.rain.fr) Quit (Ping timeout: 260 seconds)
- [11:21:29] * xMDKx (~mdkcore@nat/mandriva/x-vpjqzfqrvfzitfik) has joined #beagle
- [11:26:54] * raster (~raster@enlightenment/developer/raster) has joined #beagle
- [11:36:43] * akumar (~mm4@122.170.63.16) has joined #beagle
- [11:37:34] * arun (~arun@unaffiliated/sindian) Quit (Ping timeout: 264 seconds)
- [11:41:06] <chrisUSB>
could it be that omapfbplay is somewhat buggy? I try to to use the -t parameter with 1; it returns an error saying: omapfbplay BigBuckBunny_640x360.m4v -t 1 Invalid size/count 'SER=***'
- [11:41:35] <chrisUSB>
Somehow the 'U'from User is being interpreted from somewhere...
- [11:44:04] <av500>
what User?
- [11:45:58] <chrisUSB>
USER=*** --> 'SER=***'
- [11:46:08] <chrisUSB>
the *** represents my current username at the beagleboard
- [11:46:10] * new2bb (a4a4fa0a@gateway/web/freenode/ip.164.164.250.10) has joined #beagle
- [11:46:34] <av500>
what is your command line?
- [11:47:08] <chrisUSB>
omapfbplay BigBuckBunny_640x360.m4v -t 1
- [11:48:46] * ant_work (~andrea@host214-85-static.34-85-b.business.telecomitalia.it) Quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.6/20100625231939])
- [11:54:16] * cwillu_at_work (~cwillu@cwillu.com) Quit (Remote host closed the connection)
- [11:54:39] * Tryum is now known as Try`0xff
- [11:56:07] <av500>
chrisUSB: t needs w h and n
- [11:56:24] <av500>
if (!w || !h || !n) { fprintf(stderr, "Invalid size/count '%s'\n", size); ...
- [11:56:41] <av500>
and this only test the display speed, not the video decode
- [11:59:01] <chrisUSB>
av500: oh, shame on me... but i still do not understand why size includes 'SER=***"...
- [11:59:06] <mru>
that's a bug
- [11:59:08] <mru>
sorry
- [11:59:15] <av500>
%s
- [11:59:24] <mru>
no, that's not the bug
- [11:59:41] <av500>
right
- [11:59:53] <av500>
size?size:"--"
- [12:00:11] <av500>
hmm, no
- [12:01:45] <chrisUSB>
something wrong with the pointer?
- [12:01:57] <mru>
git pull
- [12:03:33] <av500>
mru: -usage? :)
- [12:03:58] <mru>
remember who the target audience for this app is
- [12:04:03] * gustavoz (~gustavoz@host212.200-82-38.telecom.net.ar) has joined #beagle
- [12:09:00] <koen>
trolls?
- [12:12:41] * maltanar (d5a5598d@gateway/web/freenode/ip.213.165.89.141) has joined #beagle
- [12:13:11] <mru>
well, _troll_ to be precise
- [12:13:36] * screwgoth (~raseel@122.170.63.16) has joined #beagle
- [12:13:46] * jpsaman (~jpsaman@videolan/developer/jpsaman) Quit (Quit: Leaving)
- [12:14:55] * maltanar wonders what time the lightning talks q&a session will be
- [12:16:25] <TheUni>
_av500_: pong
- [12:17:39] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) has joined #beagle
- [12:22:15] * rsalveti (~rsalveti@187.115.155.46) Quit (Ping timeout: 265 seconds)
- [12:22:38] * maqr (~maqr@httpcraft/hax) Quit (Read error: Connection reset by peer)
- [12:22:58] * maqr (~maqr@httpcraft/hax) has joined #beagle
- [12:25:17] * shoragan (~shoragan@debian/developer/shoragan) Quit (Quit: shoragan)
- [12:25:45] * maltanar (d5a5598d@gateway/web/freenode/ip.213.165.89.141) Quit (Ping timeout: 252 seconds)
- [12:26:20] * shoragan (~shoragan@sicherheitsschwankung.de) has joined #beagle
- [12:28:08] * calculus (~calculus@gentoo/user/calculus) has joined #beagle
- [12:28:59] * mpoirier (~quassel@S0106002369de4dac.cg.shawcable.net) has joined #beagle
- [12:33:27] * akumar (~mm4@122.170.63.16) Quit (Remote host closed the connection)
- [12:33:54] * shoragan (~shoragan@sicherheitsschwankung.de) Quit (Quit: shoragan)
- [12:35:00] * shoragan (~shoragan@sicherheitsschwankung.de) has joined #beagle
- [12:36:35] <drinkcat>
maltanar: same as the meetings (i.e. in 1h25')
- [12:39:31] * Try`0xff is now known as Tryum
- [12:41:40] * akumar (~mm4@122.170.63.16) has joined #beagle
- [12:44:36] * new2bb (a4a4fa0a@gateway/web/freenode/ip.164.164.250.10) Quit (Quit: Page closed)
- [12:46:22] * maltanar (d5a5598d@gateway/web/freenode/ip.213.165.89.141) has joined #beagle
- [12:48:17] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) Quit (Quit: ppotera)
- [12:55:45] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) has joined #beagle
- [12:58:25] * screwgoth (~raseel@122.170.63.16) Quit (Ping timeout: 260 seconds)
- [12:58:30] * akumar (~mm4@122.170.63.16) Quit (Ping timeout: 252 seconds)
- [13:00:29] * t_s_o (~tso@183.84-49-135.nextgentel.com) has joined #beagle
- [13:03:34] * Phrog (~chatzilla@ip-87-82-198-210.easynet.co.uk) has joined #beagle
- [13:05:26] * Crazymik3 (~Crazymik3@CPE00259c601d5d-CM00080da43848.cpe.net.cable.rogers.com) Quit (Quit: Leaving)
- [13:06:29] * whittenburg (~whittenbu@ip72-213-154-11.ok.ok.cox.net) Quit (Ping timeout: 240 seconds)
- [13:11:49] * screwgoth (~raseel@122.169.85.105) has joined #beagle
- [13:13:17] * rsalveti (~rsalveti@200.184.118.130) has joined #beagle
- [13:14:20] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) Quit (Quit: ppotera)
- [13:15:24] <koen>
TheUni: btw, I just crosscompiled xbmc in OE
- [13:15:41] <TheUni>
nice, that's great news
- [13:15:49] * maqr (~maqr@httpcraft/hax) Quit (Read error: Connection reset by peer)
- [13:15:49] * akumar (~mm4@122.169.85.105) has joined #beagle
- [13:15:58] <mru>
TheUni: looking at your patches
- [13:16:05] * maqr (~maqr@httpcraft/hax) has joined #beagle
- [13:16:19] <TheUni>
koen: though OE gave me a headache last time i tried ;)
- [13:16:27] * maqr (~maqr@httpcraft/hax) Quit (Read error: Connection reset by peer)
- [13:17:35] <av500>
mru: there is even one for the container format you don't know :)
- [13:17:47] <TheUni>
mru: i merely broke them out, didn't spend too much time hunting down descriptions and reasonings. please let me know if you need more info on them and i'll track down the authors. i realize that some like "asf hacks" are probably worthless as-is.
- [13:17:52] <mru>
av500: yeah, it's working too
- [13:18:33] <mru>
scratch that
- [13:18:52] <av500>
TheUni: do you have a test file for: http://gitorious.org/theuni/ffmpeg/commit/9a14ef3bb7805ccc5b7a7ec2a565984a77bf5431 ?
- [13:19:14] * Redb3ard (~SF0010MAC@lbb-dslst.69616810.amaonline.com) has joined #beagle
- [13:19:15] <mru>
it's getting the right frame rate but timestamps are still fucked
- [13:20:04] <av500>
time to get tcvp polished...
- [13:21:05] * maqr (~maqr@httpcraft/hax) has joined #beagle
- [13:23:01] * cfriedt (~cfriedt@modemcable178.225-21-96.mc.videotron.ca) has joined #beagle
- [13:24:22] * soman (~somnath@118.102.130.6) Quit (Ping timeout: 264 seconds)
- [13:24:37] * raster (~raster@enlightenment/developer/raster) Quit (Quit: Gettin' stinky!)
- [13:27:57] * emeb|mac (~ericb@ip72-223-81-194.ph.ph.cox.net) has joined #beagle
- [13:28:53] <TheUni>
av500: sec, phone
- [13:30:33] * kevinsc (~a0214685@nat/ti/x-bouuesslpobzujno) has joined #beagle
- [13:33:06] <jkridner1>
30 minutes to the GSoC Q&A.
- [13:39:00] * [-ip-] (~ts@p57987A82.dip.t-dialin.net) has joined #beagle
- [13:40:42] * brijesh (~brijesh@nat/ti/x-ounwxeioktnyvfls) has joined #beagle
- [13:42:34] * Meizirkki (~Meizirkki@YKMDLXXVII.gprs.sl-laajakaista.fi) has joined #beagle
- [13:42:58] * akumar (~mm4@122.169.85.105) Quit (Ping timeout: 264 seconds)
- [13:43:03] * screwgoth (~raseel@122.169.85.105) Quit (Ping timeout: 240 seconds)
- [13:47:51] * virals (~Viral_Sac@122.166.2.35) Quit (Remote host closed the connection)
- [13:48:12] * whittenburg (~whittenbu@mail.tacticalelectronics.com) has joined #beagle
- [13:48:15] <TheUni>
av500: back. i think i can dig one up. one of the devs remembered that patch specifically and said that it's still needed
- [13:50:43] * billmar (48d089fe@gateway/web/freenode/ip.72.208.137.254) has joined #beagle
- [13:50:48] <TheUni>
av500: seems to reference this ticket: http://trac.xbmc.org/ticket/5245 . (unverified) sample here: http://rocky.eld.leidenuniv.nl/ta-sample.avi
- [13:52:31] <maltanar>
jkridner1: will there be a separate chatroom for the QnA session?
- [13:52:40] <jkridner>
no, we'll chat here.
- [13:53:07] * djlewis (~djlewis@75.15.64.195) has joined #beagle
- [13:53:49] <maltanar>
okay
- [13:56:07] <av500>
TheUni: thx
- [13:57:26] * ppoudel (~chatzilla@129.114.246.141) has joined #beagle
- [13:57:28] * screwgoth (~raseel@122.170.16.253) has joined #beagle
- [13:58:32] <jkridner>
hope some people are starting to show up for the GSoC Q&A.
- [13:58:55] <ppoudel>
Yep.
- [13:58:56] <av500>
we can ask gsoc?
- [13:59:52] <jkridner>
you can *ask* anything you'd like. ;)
- [14:00:07] * akumar (~mm4@122.170.16.253) has joined #beagle
- [14:00:20] <jkridner>
Jefro probably provided the best summary: http://jefro.wordpress.com/2010/06/27/texas-instruments-etech-days-on-tuesday/
- [14:02:12] <jkridner>
we'll probably go back to full lightning talks on September 29th and December 7th, but for today, we are just talking about the Google Summer of Code projects for the lightning talks.
- [14:03:16] <jkridner>
the lightning talks are intended to be a way for more people to get their work spoken about, instead of just what TI puts on the agenda...
- [14:04:20] <jkridner>
so, if there is something you want to talk about, we can use http://tinyurl.com/etechlightning to get a space reserved for September 29th or December 7th.
- [14:04:41] <jkridner>
There are 6 videos uploaded to http://beagleboard.org/gsoc....
- [14:05:26] * neo01124 (~neo@122.163.219.254) has joined #beagle
- [14:05:56] <jkridner>
the first one talks about why students would even get involved with the Google Summer of Code.
- [14:08:08] <topfs2>
jkridner, was the audio better in the file I sent than on vimeo?
- [14:08:31] <jkridner>
topfs2: I'll check my e-mail now.
- [14:08:54] <_koen_>
topfs2: what's the current version of xbmc?
- [14:09:03] <_koen_>
the svn rev, or 2010.x or 6.3?
- [14:09:42] <jkridner>
topfs2: did you send them to @ti.com or @beagleboard.org?
- [14:09:46] * _koen_ currently has "0.0", which is most likely incorrect
- [14:09:47] * BThompson (~a0193480@nat/ti/x-lposrfazzdiukfvm) has joined #beagle
- [14:09:53] <topfs2>
jkridner, @ti.com
- [14:11:05] <TheUni>
mru / av500: anything else i can do to help? am about to head out
- [14:11:27] <topfs2>
jkridner, sorry @beagleboard
- [14:11:39] <jkridner>
topfs2: I got a .mkv file to @ti.com
- [14:11:53] <jkridner>
not exactly sure what .mkv is.
- [14:12:01] <topfs2>
yeah and then I should have sent a .zip today if it went through
- [14:12:08] <av500>
TheUni: nothing atm I guess
- [14:12:33] <TheUni>
k. cya
- [14:13:39] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) has joined #beagle
- [14:13:41] <jkridner>
k, in addition to the 1 video on why students are doing something with Google Summer of Code, there are introductory videos talking about 5 of the on-going projects.
- [14:14:08] <jkridner>
The first project listed is the "USB Sniffer"...
- [14:14:41] <jkridner>
that project (http://www.youtube.com/watch?v=wcGk4VmviU0) attempts to turn the BeagleBoard into a USB bus analyzer.
- [14:14:45] * emeb|mac (~ericb@ip72-223-81-194.ph.ph.cox.net) Quit (Ping timeout: 240 seconds)
- [14:15:04] <florian>
sounds like an interesting idea
- [14:15:07] * jkridner keeps talking until getting interrupted by questions (which I hope happens soon).
- [14:15:08] <_koen_>
I only just realized how funny "usb sniffer" is for a "beagle"
- [14:15:17] <jkridner>
:)
- [14:15:20] <maltanar>
ahahaha
- [14:15:25] <maltanar>
indeed :)
- [14:15:27] <neo01124>
:)
- [14:15:39] * notzed wonders what usb's smell like
- [14:15:39] <topfs2>
lol _koen_ I haven't thought about that before either
- [14:15:54] <_chase_>
jkridner: It was a little unclear from the video if you had to configure the beagleboard for which devices you wanted to present and sniff or if that was automatic?
- [14:15:56] <drinkcat>
.-)
- [14:16:00] * kanru (~kanru@61-30-10-70.static.tfn.net.tw) Quit (Ping timeout: 260 seconds)
- [14:16:13] <jkridner>
Nicolas?
- [14:16:25] <_koen_>
so, using the scrollbar in outlook makes it crash
- [14:16:52] <drinkcat>
_chase_: well, the goal is to be automatic.
- [14:17:45] <drinkcat>
_chase_: and you can only sniff one device at a time (and that device cannot be a hub)
- [14:17:58] <_chase_>
drinkcat: So do you take all connected devices on the beagleboard and present them to the USB host? Or do I need a configuration file saying what to present to the host?
- [14:18:01] <_chase_>
OK
- [14:18:02] <jkridner>
drinkcat: if you have a hub in the middle, how do you select which device is proxied?
- [14:18:59] <drinkcat>
jkridner: the way it works now, is that you "unbind" the device that you want to sniff, so that it can be claimed by the proxy.
- [14:19:08] * jkridner continues to introduce the rest of the projects while drinkcat (Nicolas) continues to answer questions on the USB Sniffer project.
- [14:19:10] <_chase_>
drinkcat: Do you have any plans to proxy multiple devices so you can track all the USB traffic going through the system?
- [14:20:02] <_chase_>
I'm thinking along the lines of issues that have been seen only when multiple devices are connected in the system. In that case it would be nice to see all the USB traffic going on.
- [14:20:06] * arunjoseph (~arun@192.163.20.231) Quit (Remote host closed the connection)
- [14:20:09] * hvaibhav (~a0393758@192.163.20.231) Quit (Remote host closed the connection)
- [14:20:55] <drinkcat>
_chase_: that cannot be done. If you have >1 device, you have multiple USB addresses. And the MUSB block (i.e., the gadget controller), only supports listening on one USB address.
- [14:21:04] <jkridner>
The DSP Ease of Use or DSP-RPC-POSIX project using C6Run (http://www.youtube.com/watch?v=c8z-UxFwsQA) is making the C6000 much easier to program.
- [14:21:13] * hvaibhav (~a0393758@192.163.20.231) has joined #beagle
- [14:21:16] * arunjoseph (~arun@192.163.20.231) has joined #beagle
- [14:21:46] <koen>
ppoudel: have you looked at c6accell for your opencv project?
- [14:21:56] <koen>
(that's actually _roger_'s question)
- [14:21:58] <drinkcat>
_chase_: that would have been great though, and I considered doing that at first (i.e., appearing as a virtual USB hub, so that you can connect real devices+virtual devices).
- [14:22:01] <jkridner>
this project essentially makes programming the DSP as easy as programming for the ARM only.
- [14:22:21] <ppoudel>
No, I haven't.
- [14:22:22] <koen>
jkridner: the hidden assumption is that programming the arm is easy :)
- [14:22:48] <_chase_>
drinkcat: Interesting. I'm not a USB expert, but would it be possible to sniff every transaction that goes down to the low level hardware (i.e. through the driver)?
- [14:23:16] <jkridner>
koen: it isn't that it is easy, just that it matches a skill-set that is widely held--the ability to create applications in C utilizing POSIX APIs.
- [14:23:31] <ppoudel>
koen: I am actually unknown what c6accell is about it?
- [14:23:52] <jkridner>
koen: even if it is a small subset, like fopen and printf.
- [14:23:55] <koen>
ppoudel: it's like c6run, making the usage of accell'd bits easier
- [14:24:11] <ppoudel>
OK
- [14:24:11] <koen>
ppoudel: I think they are planning a demo to show canny edge detect
- [14:24:12] <jkridner>
koen: is c6accell the same as c6runlib?
- [14:24:20] <koen>
jkridner: AIUI, no
- [14:24:33] * jkridner hasn't heard of c6accell
- [14:24:36] <maltanar>
jkridner: IIRC it's a CE-based TI project
- [14:24:45] <koen>
jkridner: there's c6flo (the gui stuff), c6accell (the accell bits), and c6run
- [14:24:51] <drinkcat>
_chase_: not really, there is a lot of processing done on the chip (for example, data coming from different endpoints are put in buffers, etc...), so you cannot access it at a low level...
- [14:25:10] <jkridner>
oh, I think I know something about that project---before it got its name.
- [14:25:28] * rhk (~rhk@adsl-75-44-25-53.dsl.sfldmi.sbcglobal.net) has joined #beagle
- [14:25:30] <_chase_>
drinkcat: OK. thanks.
- [14:25:57] <drinkcat>
_chase_: to draw a parallel with networking, the view we have from the kernel is similar to a TCP application: you have no idea about what's happening at lower level (IP, ethernet), like errors, retransmissions, etc...
- [14:26:07] <jkridner>
I think that one (C6Accell) is making a specific set of functions available on the ARM that are accelerated by the DSP, whereas C6Run is focused on being able to create new routines on the C6x.
- [14:26:26] * _roger_ (~a0740758@nat/ti/x-wykkczazmmycdphu) has left #beagle
- [14:26:46] <ppoudel>
koen:where can I about this c6accell? google doesn't show anything for this word.
- [14:26:47] <maltanar>
that's how I've understood it to be, as well
- [14:26:56] <maltanar>
ppoudel: ?? think it might be c6accel
- [14:27:05] * jkridner wants to ask the question to maltanar (Yaman) if DSP-RPC-POSIX will ever be included in the Angstrom SDKs from Narcissus.
- [14:27:25] <maltanar>
definitely!
- [14:27:34] <maltanar>
I haven't been able to delve into bitbake recipe making so far
- [14:27:45] * Beagle8 (~Beagle8@69.241.25.0) Quit (Ping timeout: 240 seconds)
- [14:27:45] <koen>
jkridner: when someone fixes that *CENSORED* thing masquearading as buildsystem for that
- [14:28:00] <ppoudel>
thanks maltanar
- [14:28:12] <koen>
jkridner: I made opencv-dsp build by doing echo > Rules.make
- [14:28:13] <maltanar>
but bb recipes should be in place before the end of the summer
- [14:28:26] <jkridner>
maltanar: OK, that is the main thing I've been looking for... and hoped you might fix... so that I can just install my toolchain and start coding away for my Angstrom-built file system.
- [14:28:28] <maltanar>
the buildsystem is also a bit problematic, yes :)
- [14:28:28] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) Quit (Quit: ppotera)
- [14:28:47] <jkridner>
maltanar: ideally, toolchains would also be built for other distros, like Ubuntu.
- [14:28:56] <maltanar>
jkridner: exactly. otherwise we don't really have "ease" of development if one has to set up all those dependencies by hand
- [14:29:15] <jkridner>
maltanar: but, working from the C6Run distro itself is way too limiting.
- [14:29:17] <koen>
or if like it is currently, it doesn't integrate with anything
- [14:29:19] * _roger_ (~a0740758@nat/ti/x-hnswqxlcrfrraciu) has joined #beagle
- [14:29:42] * koen reboots TI laptop to fix outlook
- [14:30:10] <Redb3ard>
TI?
- [14:30:18] <jkridner>
The 3rd project with a video on http://beagleboard.org/gsoc is the Pulse Width Modulation (PWM) on BeagleBoard project...
- [14:30:18] <eFfeM_work>
outlook ???
- [14:30:24] <Redb3ard>
I thought they stopped making laptops around 1989.
- [14:31:03] <jkridner>
http://www.youtube.com/watch?v=d_vYMTuSGrM is the link for the PWM video.
- [14:31:06] <mru>
look out, he's running outlook!
- [14:31:22] <topfs2>
*shrugs*
- [14:31:26] <koen>
embarassingly, I don't have a working TI calculator anymore
- [14:31:29] <mru>
Redb3ard: koen works for TI, I guess he meant the work laptop
- [14:31:39] * 20QABBIG8 (~cmurillo@186.32.57.22) has joined #beagle
- [14:31:41] * mru has a TI calculator
- [14:31:45] <Redb3ard>
Oh. Duh.
- [14:31:46] <mru>
no idea if it still works
- [14:31:55] <ogra>
does it run outlook ?
- [14:31:57] * _koen_ (~x0115699@nat/ti/x-napikitvvncqstbx) Quit (Ping timeout: 248 seconds)
- [14:32:03] <av500>
TI calcs are for wimps
- [14:32:05] <mru>
no non-volatile storage on those things sucked
- [14:32:06] <Redb3ard>
I was imagining him running Outlook on some 1980s monstrosity that weighed 30 pounds.
- [14:32:17] <jkridner>
ogra: do you mean lookout?
- [14:32:22] <av500>
Redb3ard: it feels like that
- [14:32:46] * jkridner started TI with a TI desktop computer.
- [14:32:58] * jkridner also had experience running a TI mainframe.
- [14:33:00] * mcookcomcast (44572a6e@gateway/web/freenode/ip.68.87.42.110) has joined #beagle
- [14:33:01] * djlewis (~djlewis@75.15.64.195) Quit (Quit: Leaving.)
- [14:33:18] * _koen_ (~x0115699@nat/ti/x-lghsswkdzqbupedp) has joined #beagle
- [14:33:25] <Redb3ard>
My first computer ever was the TI 99/4a.
- [14:33:39] <jkridner>
er, I guess the 990 was a "minicomputer".
- [14:33:42] <Redb3ard>
Good old TMS 9900... they don't make em like that any more.
- [14:34:02] * emeb (~ericb@ip72-223-81-194.ph.ph.cox.net) has joined #beagle
- [14:34:03] <jkridner>
http://en.wikipedia.org/wiki/TI-990
- [14:34:11] <Redb3ard>
And by "make em like that" I mean hobble a 16bit cpu in an 8 bit board with just a few hundred bytes of ram.
- [14:34:17] <mru>
Redb3ard: sometimes that's a good thing
- [14:35:02] <jkridner>
there's nothing quite like setting the toggle switches to the desired address of your program counter and then flipping the reset toggle switch.
- [14:35:35] <Redb3ard>
Hey, was that mini the same cpu?
- [14:35:38] <Redb3ard>
Seems like it.
- [14:35:47] <Redb3ard>
Or it's at least in the same family.
- [14:35:59] <jkridner>
the 990 used the TMS9900.
- [14:36:04] <Redb3ard>
Kickass.
- [14:36:14] * jrmuizel (~jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
- [14:36:20] <av500>
jkridner: I guess it has a borken MUSB too....
- [14:36:47] <mcookcomcast>
http://oldcomputers.net/ads/80s/ti992-a.jpg
- [14:37:02] <jkridner>
av500: if you didn't mind carrying the packets in the other room, it worked just fine.
- [14:37:25] <Redb3ard>
Never saw that one. I had the stainless steel one I think.
- [14:37:29] <av500>
yeah, you had to solder a tiny 22F cap to make it work :)
- [14:38:03] <mcookcomcast>
my first computer: http://oldcomputers.net/trs80i.html
- [14:38:14] <Redb3ard>
My uncle had one of those.
- [14:38:24] <jkridner>
Well, Varun's PWM project might also require a capacitor, if you really wanted to use it for audio.
- [14:38:33] <Redb3ard>
I think mom only bought us the TI because it was so horrible they had it in bargain bins that year.
- [14:38:34] <av500>
my 1st was a portable: http://www.old-computers.com/museum/computer.asp?c=1000
- [14:38:48] <Redb3ard>
Lord knows she couldn't have afforded anything better.
- [14:39:39] * maltanar has to run in 10 minutes, any more questions on C6Run and dsp-rpc-posix?
- [14:39:39] <jkridner>
mcookcomcast: we share the same 1st computer heritage.
- [14:39:48] <mcookcomcast>
nice :-)
- [14:40:10] <Redb3ard>
Thanks, now you make me feel uncool about my first computer.
- [14:40:29] <Redb3ard>
25 years later, and I still feel like the kid who has to wear hand-me-downs.
- [14:40:36] <mcookcomcast>
there's no such thing as an un-cool computer... unless it's a MAC
- [14:40:39] <billmar>
my job out of college used a 9900 for the SR-71 air data sensor (triple modular redundant)
- [14:41:26] <jkridner>
how many SR-71s can you make with a BeagleBoard?
- [14:41:41] <Redb3ard>
They made the originals with slide-rules.
- [14:41:45] <Redb3ard>
So I imagine quite a few.
- [14:42:04] * CHRIS_7 (3fa44003@gateway/web/freenode/ip.63.164.64.3) has joined #beagle
- [14:42:23] <jkridner>
everybody get the basic idea of the DSP-RPC-POSIX project then?
- [14:42:33] <jkridner>
folks watching the PWM video?
- [14:43:04] <Redb3ard>
What's the basic idea?
- [14:45:23] <mcookcomcast>
another level of abstration?
- [14:45:33] <mcookcomcast>
abstraction
- [14:45:39] <jkridner>
Redb3ard: as http://www.youtube.com/watch?v=c8z-UxFwsQA discusses, the basic idea is to provide scripts and libraries to make programming on the C6000 like programming on the ARM with GCC.
- [14:46:05] <mcookcomcast>
there must be a boat load of marshalling going on
- [14:46:07] <jkridner>
mcookcomcast: well, there are all sorts of abstractions for programming the DSP....
- [14:46:07] <Redb3ard>
That would be pretty cool.
- [14:46:12] <maltanar>
so you can compile and run #include <stdio.h> void main() { printf("Hello world!\n");} on the DSP
- [14:46:37] <jkridner>
not so much adding another one as it is choosing a simple one by making some basic assumptions.
- [14:46:49] <Redb3ard>
I don't honestly know that I'd have much use for it... I'm still trying to figure out what I can do now that our product will have a microphone in port. Seems like it ought to be good for something.
- [14:46:59] <av500>
maltanar: can you compile and run ffmpeg this way?
- [14:47:18] <maltanar>
av500: no, probably not at this stage :)
- [14:47:33] <mcookcomcast>
jkridner: would it be able to substantially reduce BOM costs?
- [14:47:43] <maltanar>
the RPC layer will be able to eventually access any GPP-side function residing in any library, though
- [14:47:45] <jkridner>
maltanar: would it work before the end of the project?
- [14:47:58] <maltanar>
jkridner: 'it' being ffmpeg?
- [14:48:21] <av500>
or openoffice, you chose :)
- [14:48:31] <jkridner>
maltanar: yes, ffmpeg, if you didn't care about the interprocessor communication performance.
- [14:49:05] <jkridner>
maltanar: if you just wanted to be able to test it functionally and start profiling a few routines.
- [14:49:41] <maltanar>
jkridner: I honestly can't answer that right now since I've never compiled ffmpeg on the ARM either and don't know about its dependencies
- [14:50:08] <maltanar>
but I don't see why not
- [14:50:19] <jkridner>
mcookcomcast: how would BOM costs be affected? main way I'd see is to reduce the need to buy a more expensive processor or add other processing elements because you'd get more functionality out of the same device.
- [14:50:26] <maltanar>
though C6RunLib would be more suitable for that task
- [14:50:31] <maltanar>
and not C6RunApp
- [14:50:38] <jkridner>
if building ffmpeg as a lib.
- [14:50:48] <jkridner>
which is what would most likely make sense.
- [14:50:53] <maltanar>
c6runlib = gpp->dsp rpc calls
- [14:50:59] <mcookcomcast>
jkridner: that's where I was going... if the wrapper allows you to do more with less, sounds interesting
- [14:51:15] <maltanar>
so you can have the interesting parts on the DSP and plain regular stuff that involve dependencies etc. on the GPP
- [14:51:58] <jkridner>
as I mentioned the PWM project has a video, as do the OpenCV and XBMC optimization projects.
- [14:52:17] * sjhill (~sjhill@home.bethel-hill.org) has joined #beagle
- [14:52:24] * kevinsc (~a0214685@nat/ti/x-bouuesslpobzujno) Quit (Quit: Leaving.)
- [14:52:32] * Zoxc (~zoxc@ti0128a340-dhcp0372.bb.online.no) has joined #beagle
- [14:52:32] <jkridner>
the PWM project allows you to do motor control, etc.
- [14:53:05] <jkridner>
etc. = LED brightness controls, audio, and I can hardly even imagine what.
- [14:53:40] <jkridner>
any low-frequency D/A conversion, if used in conjunction with the right filter/amp.
- [14:54:24] <koen>
jkridner: or use http://www.adafruit.com/index.php?main_page=product_info&cPath=37&products_id=252 as cpu meter
- [14:54:33] <jkridner>
the project adds a kernel module to make it simpler than trying to just program the GPIO, serial port or timer registers.
- [14:54:46] * djlewis (~djlewis@75.15.64.195) has joined #beagle
- [14:55:01] <mcookcomcast>
jkridner... that is very interesting. any arduino integrations?
- [14:56:05] <jkridner>
mcookcomcast: this makes it less necessary to use an Arduino. There have been many projects that integrated an Arduino and I mostly wonder why it is really necessary, but folks don't understand the low-levels of sensor integration.
- [14:56:23] * jrmuizel (~jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit (Quit: jrmuizel)
- [14:56:46] <mcookcomcast>
I think it has to do with the assemblies and examples
- [14:57:06] <mcookcomcast>
there's a well understood model for how to integrate 'shields
- [14:57:11] <mcookcomcast>
' in arduino
- [14:57:16] <mcookcomcast>
for example
- [14:57:21] * jayabharath (~a0866114@nat/ti/x-qqlaveezjpgnpfiu) has joined #beagle
- [14:57:38] <jkridner>
http://www.youtube.com/watch?v=TSgAzdaJ1Xo is the OpenCV optimizations (using DSP) video introduction.
- [14:58:07] <jkridner>
http://www.youtube.com/watch?v=gvJ32T-W3Gw is the XBMC optimizations video introduction.
- [14:58:41] <jkridner>
OpenCV is a library focused on PCs initially for computer vision.
- [14:59:12] <mcookcomcast>
i use opencv
- [14:59:15] <jkridner>
There are some libraries for vision focused on the TI vision devices...
- [14:59:24] <mcookcomcast>
lots of amazing features
- [14:59:33] * jrmuizel (~jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
- [14:59:36] * topfs2 hopes people can hear the audio on the xbmc video :)
- [14:59:49] <jkridner>
The TI libraries are used in real vision/security systems.
- [14:59:50] <mcookcomcast>
i hear it...
- [15:00:20] <jkridner>
but, lots of research is going on using the OpenCV libraries....
- [15:00:51] <jkridner>
so, that project attempts to bring a few of the performance advantages of the C6000 to the OpenCV library.
- [15:02:10] * maltanar really has to run now, but will be glad to answer any C6Run-related questions personally on IRC or via mail at maltanar@gmail.com later
- [15:02:40] * maltanar (d5a5598d@gateway/web/freenode/ip.213.165.89.141) has left #beagle
- [15:03:27] <jkridner>
my experience writing C code for the C6000 has been that you get ARM/x86 comparable performance per cycle with simple compilation, but to take advantage of the VLIW architecture requires giving the compiler some optimization hints.
- [15:03:48] * jkridner stops to remind everyone that TI eTech Day is starting now: http://www.ti.com/etechdays
- [15:04:25] <jkridner>
there are several tracks depending on your subject of interest and live Q&A.
- [15:05:57] <jkridner>
For me, the MSP430 Chronos watch is a bit fun for wireless sensor integration with the BeagleBoard.
- [15:05:59] * RoHS (~RoHS@68-64-214-18.static.forethought.net) has joined #beagle
- [15:06:31] <jkridner>
The AM37x is discussed in the "DSPs & MPUs in Application" track.
- [15:06:43] * alancam (~a-campbel@nat/ti/x-bzynztjryepazwhk) has joined #beagle
- [15:06:44] <mru>
jkridner: "give the compiler some hints" == write asm yourself?
- [15:07:10] <jkridner>
mru: no, tell it that a loop is going to run a minimum number of times is a big one.
- [15:07:43] <mru>
yeah, I know there are pragmas for those things
- [15:07:48] * maria (~mrodrigue@186.32.57.22) has joined #beagle
- [15:07:48] <jkridner>
when I was doing C6000 programming frequently, I could pretty much guess what assembly the C compiler was going to give me.
- [15:08:23] <jkridner>
http://processors.wiki.ti.com/index.php/C6000_CGT_Optimization_Lab_-_1 has some good C compiler hints.
- [15:08:53] <jkridner>
the MUST_ITERATE one has a huge impact.
- [15:09:15] <jkridner>
otherwise, you never really get into a well-scheduled innerloop.
- [15:09:29] <mru>
of course
- [15:09:41] <mru>
I'm surprised the compiler makes good use of such info though
- [15:09:50] <jkridner>
the nice thing about the C6000 is that you can have a very complex inner-loop as the addressing modes are very flexible and you have a good variety of instruction units.
- [15:10:04] <mru>
circular addressing ftw
- [15:10:17] <notzed>
how does it compare in performance per clock on good code?
- [15:10:28] * hvaibhav (~a0393758@192.163.20.231) Quit (Quit: Leaving.)
- [15:10:29] <jkridner>
mru: it is almost like learning another syntax to provide the compiler with just the right hints.
- [15:10:43] <jkridner>
notzed: to what? writing assembly?
- [15:11:18] * cwillu_at_work (~cwillu@cwillu.com) has joined #beagle
- [15:11:23] * mru is sure writing asm by hand still beats the compiler by a good margin
- [15:11:35] * tsolox (~tsolox@124.6.157.10) has joined #beagle
- [15:12:05] <notzed>
i mean the dsp vs cpu
- [15:12:11] * Kmus (~askme@217.33.179.86) has joined #beagle
- [15:12:12] <jkridner>
depends on how experienced you are in writing the assembly.
- [15:12:27] <jkridner>
notzed: depends on the operation.
- [15:12:29] <notzed>
hmm, it wasn't meant to be a philosphical question
- [15:12:47] <mru>
jkridner: suppose I'm doing after a year's practice
- [15:13:07] <notzed>
"<jkridner:#beagle> my experience writing C code for the C6000 has been that you get ARM/x86 comparable performance per cycle with simple compilation, but to take
- [15:13:08] <notzed>
+advantage of the VLIW architecture requires giving the compiler some optimization hints.
- [15:13:18] <notzed>
... and then ...
- [15:13:36] <jkridner>
I believe the C6000 running at 430MHz is comparable to the ARM running at 600MHz for "typical" code with a mix of style of operations.
- [15:13:42] <mru>
notzed: for any pair of processors, you can always construct code that whichever one you want performs better on
- [15:13:50] <mru>
it's a process called benchmarking
- [15:14:07] <tsolox>
rephrase: can a process(es) be executed by a kernel after boot-up, without the support of some shell?
- [15:14:27] <mru>
/sbin/init comes to mind
- [15:14:39] <jkridner>
mru, notzed: for certain benchmarks, you can let people knowledgeable in each "code-it-out"...
- [15:14:40] <notzed>
indeed, but that wasn't the question. nevermind.
- [15:15:06] <jkridner>
like what BDTI folks do.
- [15:15:11] <mru>
if you pick a problem perfectly tailored to one of the instruction sets, it will probably win
- [15:15:13] * _don_ (~don@nat/ti/x-hxswneobhrifcinq) has joined #beagle
- [15:15:22] <jkridner>
mru: right.
- [15:15:43] <mru>
benchmarking is the art of finding those problems
- [15:16:29] * neo01124 (~neo@122.163.219.254) Quit (Remote host closed the connection)
- [15:16:48] <jkridner>
mru: yeah, I won't argue that a skilled assembly programmer will squeeze out another 20-30% off even the best compilers.
- [15:17:14] * franktango (~a0746747@nat/ti/x-vaqnnlcbhooappwj) has left #beagle
- [15:17:44] <jkridner>
but, if you are talking within the first several months of programming on the C6000, applying the C pragmas is the way to go.... and it will get you a long way.
- [15:18:12] <jkridner>
well, seems we are losing students and I'm doing way too much talking....
- [15:18:17] <jkridner>
there is also the FFTW project....
- [15:18:25] <jkridner>
and I think cfriedt might still be here.
- [15:18:49] * Entasis (~Jarred@ppp118-210-194-202.lns20.adl6.internode.on.net) Quit (Quit: Leaving)
- [15:19:03] <mru>
ah yes, he is...
- [15:19:07] * tsolox (~tsolox@124.6.157.10) Quit (Quit: leaving)
- [15:19:10] * mru reaches for cattle prod...
- [15:19:14] <ppoudel>
If someone has any question regarding OpenCV acceleration, I am here too.
- [15:19:33] <topfs2>
same for xbmc, I'm here too :)
- [15:19:48] <Redb3ard>
Heh... we're going to start playing with OpenCV soon.
- [15:19:53] * jrmuizel (~jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit (Quit: jrmuizel)
- [15:20:28] <cfriedt>
hi, sorry.
- [15:20:42] <cfriedt>
my autoscroll wasn't kicking in
- [15:20:58] <cfriedt>
is there a question for me?
- [15:21:12] <jkridner>
cfriedt: can you explain quickly what FFTW is?
- [15:21:28] * kanru (~kanru@118-168-235-45.dynamic.hinet.net) has joined #beagle
- [15:22:14] <cfriedt>
yes. FFTW is a library that aims to provide the fastest fourier transform in platform-independent C.
- [15:22:22] <_koen_>
in the west
- [15:22:27] <_koen_>
but not in the east
- [15:22:57] <cfriedt>
exactly
- [15:23:03] <jkridner>
notzed: BTDIsimmark2000 seems to agree with me.
- [15:24:16] <jkridner>
cfriedt: so, you won't introduce any ARM NEON assembly?
- [15:24:33] * b7500af1 (~GH@2001:468:c80:4280:21c:bfff:fe8b:90b4) has joined #beagle
- [15:24:47] <jkridner>
topfs2: I think everyone knows what XBMC is already. :)
- [15:25:11] <cfriedt>
well, as anyone can see from a quick google search, gcc's neon optimization flags for C still produce machine code that lacks in performance
- [15:25:26] <topfs2>
hehe jkridner if they don't I'm very sad :P
- [15:25:27] <cfriedt>
fftw's main niche has been with x86 hardware which have the major advantage of hardware reschedulers (out-of-order execution), which NEON lacks
- [15:25:37] * Belna (~Thomas@DSL01.212.114.252.242.ip-pool.NEFkom.net) Quit (Ping timeout: 276 seconds)
- [15:25:49] <cfriedt>
one supported arch without OOE is power pc, and the C codelet interface does suffer on that architecture somewhat.
- [15:26:43] <cfriedt>
so its a bit non-traditional for fftw, but by introducing some (lots!) of hand-coded NEON routines, there will be a major speedup beyond what is capable with the C codelet interface
- [15:28:18] <jkridner>
well, if there are no more questions, I think we best turn the students free.
- [15:28:37] <jkridner>
a couple that got questions earlier already left.
- [15:28:47] <jkridner>
thanks to those of you who asked questions.
- [15:28:58] <jkridner>
please, spread the word about the great work these students are doing.
- [15:29:16] <mru>
I will when I see it
- [15:30:18] <av500>
ppoudel: the stuff you run on the dsp, does it run in series or in parallel to the arm?
- [15:30:36] * jkridner looks at gitorious.org/gsoc2010-fftw-neon
- [15:30:43] <mru>
or maybe in a parallel series?
- [15:30:44] <ppoudel>
In series, so far.
- [15:31:48] <drinkcat>
ppoudel: what happens if you have some (Linux) context switch while you are running code on the dsp?
- [15:32:14] <av500>
drinkcat: nothing bad
- [15:32:22] <av500>
in fact u want that
- [15:32:35] <drinkcat>
sure, then you got some paralelism
- [15:32:38] <av500>
other processes can run while u wait for the dsp
- [15:33:13] <ppoudel>
drinkcat: No problem regarding that, I think it is taken care by OS.
- [15:33:31] <drinkcat>
but, then, what if you have 2 apps using the dsp? the second one just waits?
- [15:33:42] <av500>
drinkcat: no
- [15:33:53] <av500>
dsplink and CE take care of that
- [15:34:03] <av500>
you can run the dsp from 2 apps
- [15:34:07] <drinkcat>
take care how?
- [15:34:25] <av500>
there is a small OS on the DSP that multitasks
- [15:34:31] <drinkcat>
ok
- [15:34:39] <av500>
and CE allows several processe/threads to call dsp algos
- [15:34:51] <ppoudel>
called DSP/BIOS
- [15:34:57] * jconnoll1 is now known as jconnolly
- [15:34:58] <av500>
each dsp call ends up in a syscall in the end
- [15:35:10] <av500>
at that point userspace waits for the syscall to end
- [15:35:22] <av500>
so kernel can schedule other processes in the meantime
- [15:35:25] <drinkcat>
okay
- [15:35:49] <av500>
and, you can also call stuff "async"
- [15:36:00] <av500>
you make the call, then you can checlk for completion
- [15:36:44] <av500>
that can save you one separate thread....
- [15:37:51] <drinkcat>
and how are the tasks scheduled on the dsp? is it preemptive? cooperative?
- [15:38:05] <av500>
jkridner: ?
- [15:38:12] <av500>
there are priorities you can set
- [15:38:23] <av500>
I guess it is preemptive
- [15:38:36] <jkridner>
DSP/BIOS provides both.
- [15:38:48] <jkridner>
typically, you'd do cooperative for real-time stuff.
- [15:38:50] * kevinsc (~a0214685@nat/ti/x-mfxdbqzdagxqdifc) has joined #beagle
- [15:38:55] <av500>
unlike gnome, there are a lot of settings :)
- [15:39:37] <av500>
for audio/video, you would set the audio process to higher prio
- [15:39:38] <jkridner>
the simplest ideal scheduling (assuming no transition overhead) is to run the task with the next deadline first....
- [15:39:49] <jkridner>
but, that isn't a typical method.
- [15:39:50] <av500>
as you can live with losing a frame, but not with broken audio...
- [15:40:04] <av500>
losing->being late
- [15:40:27] <drinkcat>
and where does the DSP/BIOS save the tasks states? main memory?
- [15:40:31] <av500>
but then, on the omap3 you would run audio on the arm anyway,,,
- [15:40:37] <av500>
drinkcat: part of main mem
- [15:40:38] <jkridner>
by "cooperative", I mean in reference to tasks running at the same priority level...
- [15:40:53] <jkridner>
tasks at a higher priority level always interrupt other running tasks in DSP/BIOS.
- [15:41:14] <jkridner>
it is possible to have "preemptive", where the scheduler runs one task for a while then runs another one for a while....
- [15:41:22] <jkridner>
but, you wouldn't typically do that for real-time tasks.
- [15:41:37] <av500>
so not really for gpp driven sutff like CE...
- [15:42:31] <jkridner>
drinkcat: DSP/BIOS gets task stacks from various pools depending on configuration.
- [15:42:55] <jkridner>
part of the headache with DSP/BIOS, Link, CE, etc. is that everything is a configuration option.
- [15:43:17] <jkridner>
at some point, we just need to define some defaults that work.
- [15:43:58] * tasslehoff (~Mich@147.84-49-231.nextgentel.com) Quit (Ping timeout: 276 seconds)
- [15:44:18] <drinkcat>
I guess that finding the best combination to get max perf for a given app is not necessarily trivial...
- [15:44:21] * jconnolly (~jconnolly@firebug.buglabs.net) Quit (Remote host closed the connection)
- [15:44:29] * XorA|gone (~XorA@www.xora.org.uk) Quit (Ping timeout: 248 seconds)
- [15:44:57] <av500>
drinkcat: for audio and video decoding on the dsp, it was not hard
- [15:44:59] <jkridner>
drinkcat: as av500 says, it is typical to pull out of main (DDR) memory, but that requires the GPP (ARM) to share some memory with the DSP for it to use.
- [15:45:26] <jkridner>
yeah, it doesn't have to be hard.
- [15:45:26] <av500>
drinkcat: it is hard if you have real time constraints on the dsp side
- [15:45:33] <jkridner>
you just have to get above the noise.
- [15:46:05] <av500>
S/N++
- [15:46:20] <drinkcat>
jkridner: can you bypass the DDR for ARM-DSP communication?
- [15:46:46] * jconnolly (~jconnolly@firebug.buglabs.net) has joined #beagle
- [15:46:47] <jkridner>
drinkcat: sure, but only for small messages/tasks.
- [15:47:05] <jkridner>
there is a 64kB on-chip memory.
- [15:47:21] <jkridner>
there is the L1/L2 $ on the DSP.
- [15:47:29] * jpirko (~jirka@nat/redhat/x-kchjoustsycfaiym) Quit (Remote host closed the connection)
- [15:47:31] <jkridner>
there are the mailboxes.
- [15:47:35] * mobidev (~mobidev@85.26.186.108) has joined #beagle
- [15:49:19] <drinkcat>
jkridner: if you access content from that 64K RAM, does it get cached again in L1/L2 (either on DSP or ARM sides)?
- [15:49:59] * jkridner isn't sure, but believes it is configurable.
- [15:50:27] <av500>
dsp has caches that can be configured
- [15:50:44] <av500>
dsp has own L1/L2 iirc
- [15:50:54] <av500>
i dont think the 64k are cached
- [15:51:19] <av500>
drinkcat: but hey, there is a nice TRM :)
- [15:51:46] <drinkcat>
av500: TRM?
- [15:52:19] <mru>
which 64k are you talking about?
- [15:52:27] <mru>
the L3 SRAM?
- [15:52:42] <djlewis>
I enjoyed the presentations, thanks guys :)
- [15:53:01] <drinkcat>
mru: http://focus.ti.com/docs/prod/folders/print/omap3530.html, I see some 64K "RAM", didn't look further...
- [15:53:16] <mru>
yeah, that's the L3 ram
- [15:53:54] <mru>
the IVA has more internal ram
- [15:53:56] <mru>
L1 and L2
- [15:53:59] * Ceriand|work (~Ceriand@pc41.cs.ucdavis.edu) has joined #beagle
- [15:54:01] <drinkcat>
L3? so it's a cache? I understand it as a scratchpad...
- [15:54:05] * Ceriand|work (~Ceriand@pc41.cs.ucdavis.edu) Quit (Changing host)
- [15:54:05] * Ceriand|work (~Ceriand@unaffiliated/ceriand) has joined #beagle
- [15:54:06] <mru>
both can be accessed directly from ARM too
- [15:54:24] * Proxyles (~henrik@c-f893e255.56-4-64736c14.cust.bredbandsbolaget.se) Quit (Remote host closed the connection)
- [15:54:42] <mru>
the IVA L1 and L2 are configurably split cache/regular ram
- [15:55:02] * screwgoth (~raseel@122.170.16.253) Quit (Quit: Leaving.)
- [15:55:47] <jkridner>
the L3 RAM cannot act as cache
- [15:55:50] * _roger_ (~a0740758@nat/ti/x-hnswqxlcrfrraciu) Quit (Quit: Leaving.)
- [15:56:00] <jkridner>
L1/L2 are configurable in IVA.
- [15:56:07] <mru>
isn't that what I said?
- [15:56:12] <jkridner>
yeah.
- [15:56:26] * akumar (~mm4@122.170.16.253) has left #beagle
- [15:57:05] <jkridner>
except I don't think you answered if L3 was cache.
- [15:57:41] <jkridner>
I just didn't read the log well. sorry.
- [15:58:22] <jkridner>
this audio from topfs2 is messed up in some way that should be easier for me to catch than it has been.
- [15:58:29] <jkridner>
it is as if there are bytes swapped.
- [15:59:03] <jkridner>
topfs2: did you send the wave file to me, or just this mkv?
- [15:59:11] <topfs2>
Weird. Well I sent both
- [15:59:15] <topfs2>
I can try again though
- [15:59:27] <topfs2>
which @ti or @beagle?
- [15:59:38] <jkridner>
k, I just found the .zip.
- [15:59:40] <av500>
@tiagle
- [16:00:07] <jkridner>
those aren't mic artifacts.
- [16:00:19] <jkridner>
they are encoding artifacts.
- [16:01:03] <topfs2>
its mp3 in the mkv shouldn't be noticable I would guess?
- [16:01:08] <topfs2>
I have a flac aswell
- [16:01:35] <topfs2>
the ogg sounds fine for me :)
- [16:04:45] * duffolonious (~bryan@75-168-95-114.mpls.qwest.net) Quit (Ping timeout: 240 seconds)
- [16:07:13] * duffolonious (~bryan@75-168-95-114.mpls.qwest.net) has joined #beagle
- [16:09:52] <topfs2>
jkridner, you want me to send the flac?
- [16:09:57] <jkridner>
sure
- [16:10:15] * jrmuizel (~jrmuizel@mozca02.ca.mozilla.com) has joined #beagle
- [16:13:34] * _koen_ (~x0115699@nat/ti/x-lghsswkdzqbupedp) Quit (Ping timeout: 264 seconds)
- [16:17:27] * Jujo (4522ff5c@gateway/web/freenode/ip.69.34.255.92) has joined #beagle
- [16:18:41] <Jujo>
so I talked to digikey yesterday to see if my xM was still shipping this week. found out they're backordered to NOVEMBER?
- [16:19:01] <av500>
dont listen to DK
- [16:19:07] <av500>
they have no idea
- [16:19:07] <topfs2>
jkridner, should be sent now
- [16:19:13] <av500>
and no XM has yet been shipped
- [16:19:15] <Jujo>
lol, really?
- [16:19:21] <jkridner>
yeah, I'm 3/4 through downloading.
- [16:19:31] <av500>
no XM has been shipped yet, there are still issues
- [16:19:49] <Jujo>
i see
- [16:20:36] <jkridner>
right, memories only working right on half the boards. we are waiting on new memories right now and should be able to report on that in a few days.
- [16:21:57] <Jujo>
cool, glad to hear that. i had just ordered a bunch of cabling and whatnot for it
- [16:22:44] <koen>
the beagle mailinglist pretty much has daily updates on it
- [16:22:48] <kblin>
too bad it doesn't stack with the old beagles. I see why, of course. connectors galore :)
- [16:23:08] <kblin>
but there goes my plan of building a 3" rack ;)
- [16:23:44] * Crofton (~balister@host188-253-static.88-82-b.business.telecomitalia.it) has joined #beagle
- [16:24:16] <Jujo>
ahh there is a goog groups
- [16:24:38] <jkridner>
topfs2: wow, the flac sounds *really* bad.
- [16:24:57] * jkridner looks for some command-line tool rather than Audacity to help edit the bytes.
- [16:25:00] <av500>
jkridner: maybe you can try on a non TI pc? :)
- [16:25:04] <jkridner>
topfs2: what type of computer are you using?
- [16:25:49] * jayabharath (~a0866114@nat/ti/x-qqlaveezjpgnpfiu) Quit (Ping timeout: 264 seconds)
- [16:26:40] <koen>
Crofton: bon giorno
- [16:26:41] <Jujo>
do u know of http://sox.sourceforge.net/
- [16:26:52] <koen>
I'm wearing a pair now
- [16:27:20] <av500>
sox still exists??
- [16:27:22] <av500>
sox still exists?
- [16:27:29] <mru>
is it red?
- [16:27:36] <Jujo>
SoX 14.3.1 was released on April 11, 2010. Highlights include:
- [16:27:39] * prpplague (~danders@nat/ti/x-wgwvpjwspxvzxnci) has joined #beagle
- [16:27:45] <av500>
ho prpplague
- [16:28:37] <topfs2>
jkridner, normal PC with ubuntu :)
- [16:28:50] <topfs2>
the recording though is on a silly crappy mic :)
- [16:28:52] <jkridner>
k
- [16:29:00] <topfs2>
x86 nothing weird at all
- [16:29:18] <topfs2>
I really can't see how it can be different :S
- [16:29:33] <prpplague>
av500: greetings earthling
- [16:29:49] * jayabharath (~a0866114@nat/ti/x-lxbfgnpnxzsroecw) has joined #beagle
- [16:30:38] <jkridner>
my mac produces the same noise.
- [16:30:45] * mrc3_ (~mrc3@nat/ti/x-bvvvrthgjgbhqkyu) Quit (Remote host closed the connection)
- [16:30:52] <Crofton>
greetings
- [16:31:03] * mrc3_ (~mrc3@nat/ti/x-zoaskbiiwswwegzm) has joined #beagle
- [16:33:01] * bala (cf0d7c08@gateway/web/freenode/ip.207.13.124.8) has joined #beagle
- [16:34:02] * bala (cf0d7c08@gateway/web/freenode/ip.207.13.124.8) Quit (Client Quit)
- [16:35:05] <TheUni>
mru: thanks for that
- [16:35:14] * kanru (~kanru@118-168-235-45.dynamic.hinet.net) Quit (Read error: Operation timed out)
- [16:36:33] * mcookcomcast (44572a6e@gateway/web/freenode/ip.68.87.42.110) Quit (Ping timeout: 252 seconds)
- [16:38:50] * ssvb (~ssvb___@a88-112-120-50.elisa-laajakaista.fi) has joined #beagle
- [16:42:05] * xMDKx (~mdkcore@nat/mandriva/x-vpjqzfqrvfzitfik) Quit (Ping timeout: 240 seconds)
- [16:43:04] * xMDKx (~mdkcore@nat/mandriva/x-gkdkeovycfchohyc) has joined #beagle
- [16:43:35] * Phrog (~chatzilla@ip-87-82-198-210.easynet.co.uk) Quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.6/20100625231939])
- [16:48:11] * florian (~fuchs@Maemo/community/contributor/florian) Quit (Remote host closed the connection)
- [16:49:29] * lifeeth (~praneeth@unaffiliated/lifeeth) has joined #beagle
- [16:56:29] * lirtex (~liorc@89-138-158-93.bb.netvision.net.il) Quit (Remote host closed the connection)
- [16:57:44] * Phrog (~chatzilla@87.113.145.220) has joined #beagle
- [17:00:08] * djlewis (~djlewis@75.15.64.195) Quit (Ping timeout: 260 seconds)
- [17:01:08] * dl9pf_ (~quassel@p5B214231.dip.t-dialin.net) Quit (Read error: Connection reset by peer)
- [17:02:46] * neo01124 (~neo@122.163.219.254) has joined #beagle
- [17:03:06] * notzed (~notzed@ppp118-210-117-15.lns20.adl2.internode.on.net) Quit (Ping timeout: 245 seconds)
- [17:09:46] * dodeca (~aki@mini.brewt.org) has joined #beagle
- [17:12:23] * t_s_o (~tso@183.84-49-135.nextgentel.com) Quit (Remote host closed the connection)
- [17:16:45] * duffolonious (~bryan@75-168-95-114.mpls.qwest.net) Quit (Ping timeout: 248 seconds)
- [17:18:57] * duffolonious (~bryan@75-168-95-114.mpls.qwest.net) has joined #beagle
- [17:19:51] * maqr (~maqr@httpcraft/hax) Quit (Read error: Connection reset by peer)
- [17:20:11] * maqr (~maqr@httpcraft/hax) has joined #beagle
- [17:20:31] * maqr (~maqr@httpcraft/hax) Quit (Read error: Connection reset by peer)
- [17:21:48] * dl9pf (~quassel@opensuse/member/dl9pf) has joined #beagle
- [17:22:43] * lag (~lag@cpc2-aztw21-0-0-cust264.aztw.cable.virginmedia.com) Quit (Quit: Leaving)
- [17:24:37] * duffolonious (~bryan@75-168-95-114.mpls.qwest.net) Quit (Ping timeout: 264 seconds)
- [17:25:13] * maqr (~maqr@httpcraft/hax) has joined #beagle
- [17:25:17] * maqr (~maqr@httpcraft/hax) Quit (Read error: Connection reset by peer)
- [17:26:42] * duffolonious (~bryan@75-168-95-114.mpls.qwest.net) has joined #beagle
- [17:27:19] * Crofton (~balister@host188-253-static.88-82-b.business.telecomitalia.it) Quit (Read error: Connection reset by peer)
- [17:29:16] * jpirko (~jirka@ip-89-103-91-100.net.upc.cz) has joined #beagle
- [17:30:11] * maqr (~maqr@httpcraft/hax) has joined #beagle
- [17:30:44] * mrc3_ (~mrc3@nat/ti/x-zoaskbiiwswwegzm) Quit (Remote host closed the connection)
- [17:31:11] * mrc3_ (~mrc3@nat/ti/x-necogpmyuxhuvped) has joined #beagle
- [17:36:34] <b7500af1>
koen, question: i am trying to build dsplink.. was going to use the angstrom kernel from oe, that would be located at oe/build/tmp-angstrom_2008_1/sysroots/beagle-ang-linux-gnueabi/, right?
- [17:39:05] <koen>
yes
- [17:39:32] <b7500af1>
thanks
- [17:43:34] * neo01124 (~neo@122.163.219.254) Quit (Ping timeout: 276 seconds)
- [17:44:01] * neo01124 (~neo@122.163.123.120) has joined #beagle
- [17:45:15] * arunjoseph (~arun@192.163.20.231) Quit (Remote host closed the connection)
- [17:45:46] * arunjoseph (~arun@192.163.20.231) has joined #beagle
- [17:47:13] * Xerion (~xerion@82-170-197-160.ip.telfort.nl) has joined #beagle
- [17:50:21] * ang_ (d1dd0802@gateway/web/freenode/ip.209.221.8.2) has joined #beagle
- [17:51:10] * ang_ (d1dd0802@gateway/web/freenode/ip.209.221.8.2) Quit (Client Quit)
- [17:54:14] * djlewis_ (41401e0d@gateway/web/freenode/ip.65.64.30.13) has joined #beagle
- [17:54:19] <djlewis_>
gm guys
- [17:54:39] * hrw is now known as hrw|gone
- [17:55:02] <mru>
morning djlewis_
- [17:56:16] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) has joined #beagle
- [18:00:42] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) Quit (Client Quit)
- [18:04:49] <_av500_>
morning djlewis_
- [18:08:58] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) has joined #beagle
- [18:10:15] * ceyusa (~vjaquez@155.99.117.91.static.mundo-r.com) Quit (Quit: leaving)
- [18:16:01] * Meizirkki (~Meizirkki@YKMDLXXVII.gprs.sl-laajakaista.fi) Quit (Ping timeout: 245 seconds)
- [18:22:45] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) Quit (Quit: ppotera)
- [18:24:31] * florian (~fuchs@Maemo/community/contributor/florian) has joined #beagle
- [18:32:53] * Meizirkki (~Meizirkki@ZYMMDCCXLI.gprs.sl-laajakaista.fi) has joined #beagle
- [18:33:24] * t_s_o (~tso@183.84-49-135.nextgentel.com) has joined #beagle
- [18:33:40] <emeb>
djlewis_: just tried that davicom usb/eth dongle (http://www.dealextreme.com/details.dx/sku.2797) - seems to work fine.
- [18:34:17] <emeb>
this is on the beagle r78 angstrom kernel.
- [18:34:36] <emeb>
gets kinda warm tho...
- [18:39:20] * arunjoseph (~arun@192.163.20.231) Quit (Remote host closed the connection)
- [18:40:20] * arunjoseph (~arun@192.163.20.231) has joined #beagle
- [18:46:12] * dl9pf_ (~quassel@p5B215CD9.dip.t-dialin.net) has joined #beagle
- [18:46:51] * dl9pf (~quassel@opensuse/member/dl9pf) Quit (Ping timeout: 245 seconds)
- [18:53:45] <djlewis_>
tnx emeb
- [18:54:17] <emeb>
djlewis_: np.
- [18:54:40] <djlewis_>
so is that 18k builtin a buffer maybe?
- [18:55:27] <djlewis_>
_av500_: pizza tuesday, :)
- [18:55:55] <djlewis_>
av500: ^^^^
- [18:58:31] <emeb>
djlewis_: Probably. Can't imagine what else they're using it for
- [18:59:06] <emeb>
OTOH, the circuit board layout I can see through the tranparent blue case is rather different from what's shown in the pix
- [18:59:28] <emeb>
so I imagine there is some diff between the text and actual capabilities.
- [19:00:08] <djlewis_>
hmm, heat means power draw.. not for the weak of power source
- [19:01:54] <_av500_>
djlewis_: pizza, who'd have thought....
- [19:02:07] * Phrog (~chatzilla@87.113.145.220) Quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.6/20100625231939])
- [19:05:12] <emeb>
djlewis_: yep - I'm running it via a powered hub right now. Need to see if beagle can power it from musb port
- [19:05:47] * neo01123 (~neo@122.163.112.240) has joined #beagle
- [19:05:55] * neo01124 (~neo@122.163.123.120) Quit (Ping timeout: 260 seconds)
- [19:06:41] * neo01123 is now known as neo01124
- [19:08:01] <djlewis_>
_av500_: :) and Staurdays with left overs on Sunday was good too :)
- [19:08:28] <koen>
emeb: not likely
- [19:08:43] <emeb>
koen: yep - not expecting much.
- [19:09:25] <emeb>
that said, the musb was able to power my pegasus dongle (before that gave up the ghost)
- [19:10:18] * mobidev (~mobidev@85.26.186.108) Quit (Ping timeout: 252 seconds)
- [19:12:02] <emeb>
tried it - no joy. big surprise.
- [19:12:10] <koen>
shocked, I am
- [19:13:29] <koen>
emeb: there's an omapl138 board coming up with an fpga and lots of io, is that interesting to you?
- [19:13:47] <emeb>
koen: yes it is!
- [19:14:04] <emeb>
l138 is same as hawk, right?
- [19:14:07] <koen>
yes
- [19:14:10] * pting (~pting@adsl-99-32-245-41.dsl.lsan03.sbcglobal.net) has joined #beagle
- [19:14:20] <emeb>
That's not bad.
- [19:14:42] * emeb hasn't done anything with his hawkboard tho...
- [19:14:59] * eFfeM (~frans@j200125.upc-j.chello.nl) has joined #beagle
- [19:15:35] <koen>
it will be targeted at industrial folks, so I don't know the pricepoint
- [19:15:51] * lifeeth (~praneeth@unaffiliated/lifeeth) Quit (Quit: Up and at 'em, Atom Ant!)
- [19:15:52] <koen>
I hope they have an edition for consumer temperatures as well
- [19:16:29] <emeb>
big questions are: 1) what kind of FPGA, 2) how hard is it to interface (fiddly connectors?)
- [19:18:17] <koen>
"altera"
- [19:18:36] <koen>
I'll poke the people involved to give some press kits
- [19:18:59] <koen>
the animal will be 'caribou', btw
- [19:20:03] <emeb>
Altera could be good - depends on what family & what size part.
- [19:20:31] <emeb>
If part is too big then you might not be able to use the free tools. Don't know Altera's policy on that.
- [19:21:23] <emeb>
caribou - isn't that just just another name for reindeer?
- [19:23:39] <emeb>
Yep - Altera's free design tools only run on WinXX and are limited to smaller / cheaper FPGA parts.
- [19:23:53] <emeb>
(Xilinx is similar, but their free tools will run on Linux)
- [19:23:53] * Phrog (~chatzilla@87.113.145.220) has joined #beagle
- [19:25:23] * [-ip-] (~ts@p57987A82.dip.t-dialin.net) has left #beagle
- [19:27:07] * isbric_ (~isbric@c-989472d5.020-216-73746f3.cust.bredbandsbolaget.se) Quit (Quit: Lost terminal)
- [19:28:25] * RoHS (~RoHS@68-64-214-18.static.forethought.net) Quit (Quit: RoHS)
- [19:28:40] * iWill (~kvirc@unaffiliated/iwill) has joined #beagle
- [19:28:54] * neo01124 (~neo@122.163.112.240) Quit (Remote host closed the connection)
- [19:31:35] <koen>
emeb: it's a reindeer that likes the cold
- [19:31:46] <koen>
you know, industrial temps and all
- [19:32:43] <emeb>
cold - check. Now need an explanation for the antlers...
- [19:33:24] * Lioric (be98c2a9@gateway/web/freenode/ip.190.152.194.169) has joined #beagle
- [19:33:40] <_av500_>
koen: caribouard?
- [19:34:52] <koen>
emeb: it makes more sense if you know the internal codenames for the upcoming chips, but those are nda (both names and chips)
- [19:35:09] <koen>
_av500_: I guess you could call it that if you're canadian
- [19:36:21] <Lioric>
Hi, I want to purchase the Beagle Vga Board form BeagleBoardToys.com, but there the Paypal method is not enabled and there is no contact email or any other contact to ask this, is there anybody here that have purchased from them?
- [19:36:34] <_av500_>
no
- [19:36:39] <_av500_>
does not seem so
- [19:36:45] <_av500_>
nobody ever confessed
- [19:40:43] <Lioric>
or is there any way to purchase the ICETEK-OMAP3530-VGA module from outside of asia?
- [19:42:18] <prpplague>
Lioric: what is the problem with the ordering?
- [19:45:17] * mcookcomcast (45f11900@gateway/web/freenode/ip.69.241.25.0) has joined #beagle
- [19:45:45] <mcookcomcast>
topfs2: xbmc is running at 98% cpu
- [19:45:56] <mcookcomcast>
is that to be expected?
- [19:46:15] <koen>
mcookcomcast: yes, it's a gameloop
- [19:46:43] <koen>
which means "rerender static picture at 60fps"
- [19:47:12] <mcookcomcast>
i see... right now I get a blank/black screen
- [19:47:29] <_av500_>
mcookcomcast: it is a feature
- [19:47:32] <koen>
you need to run it from inside an x session
- [19:47:35] <mcookcomcast>
is that the static picture being painted at 60 fps?
- [19:48:04] <koen>
no, 12
- [19:48:19] <koen>
or 38 if you have an xm
- [19:48:35] <koen>
btw, 'opkg install xbmc' now works on angstrom
- [19:48:49] <mcookcomcast>
get the heck out of here -)
- [19:49:00] <mcookcomcast>
that's toooooooo funny
- [19:49:05] <koen>
http://www.angstrom-distribution.org/repo/?pkgname=xbmc
- [19:49:08] <mcookcomcast>
nice
- [19:49:14] <koen>
got it to crosscompile this afternoon
- [19:49:40] <koen>
nice to compile it in like 10 minutes instead of 2 hours
- [19:49:47] <mcookcomcast>
i simply ran xbmc from an xterm within X...
- [19:50:01] <mcookcomcast>
is that what I should be doing?
- [19:50:11] <koen>
that should work
- [19:50:17] <koen>
it works here
- [19:51:04] <mcookcomcast>
I must have screwed something up...
- [19:51:19] <mcookcomcast>
maybe I'll update the code with opkg now that it is out there
- [19:51:35] <mcookcomcast>
then, I'll check the MR-16 in u-boot
- [19:51:50] <mcookcomcast>
I think that I read that it has to be set at 16 anyway...
- [19:51:52] * eFfeM (~frans@j200125.upc-j.chello.nl) Quit (Quit: Leaving.)
- [19:52:05] <koen>
I wonder if 16 bit will be faster
- [19:52:09] * koen should try tomorrow
- [19:52:57] <mcookcomcast>
thank you for getting the package to work
- [19:53:27] <mcookcomcast>
was kind of fun compiling the whole thing... but boy did it take some time
- [19:53:44] <koen>
2 hours on an xm
- [19:53:49] <koen>
the extra ram helps a lot
- [19:54:18] <koen>
I wonder how it runs on a panda, with 300MHz sgx540, dualcore a9 and even more ram
- [19:54:23] <mcookcomcast>
so, once you launch xbmc, what should you see?
- [19:54:47] <koen>
that's an excellent question
- [19:54:55] <koen>
(translation: I have a youtube for that)
- [19:55:00] <mcookcomcast>
what i mean is, I have the thing running on linux on x86
- [19:55:03] <koen>
mcookcomcast: http://www.youtube.com/watch?v=80Uia6FkvnA
- [19:57:50] * theholyduck (~holyduck@ip-199-138-106-77.eidsiva.net) has joined #beagle
- [20:07:01] * Lioric (be98c2a9@gateway/web/freenode/ip.190.152.194.169) Quit (Ping timeout: 252 seconds)
- [20:07:14] <mcookcomcast>
koen: nice videp
- [20:07:16] <mcookcomcast>
video
- [20:07:38] <mcookcomcast>
i updated the xbmc to the one in the package
- [20:07:44] <mcookcomcast>
and I set the MR to 16
- [20:07:47] <mcookcomcast>
i rebooted
- [20:08:01] <mcookcomcast>
i get an error on starting the xbmc app
- [20:08:58] <mcookcomcast>
weirdness... line 1 syntax error word unexpected (expecting ")" )
- [20:11:19] <mcookcomcast>
i'm using E for windows manager, if that makes a diff
- [20:13:59] <prpplague>
mcookcomcast: if the app you are starting is a binary, then most likely it is compiled incorrectly
- [20:14:31] <mcookcomcast>
yep... it's the one i just got from opkg, though
- [20:14:54] <mcookcomcast>
heading out on vacation.... will check back next week
- [20:14:57] <mcookcomcast>
thanks
- [20:15:15] * Phrog (~chatzilla@87.113.145.220) Quit (Remote host closed the connection)
- [20:15:22] * mcookcomcast (45f11900@gateway/web/freenode/ip.69.241.25.0) Quit (Quit: Page closed)
- [20:15:27] * tconant (~tconant@68.111.35.226) has joined #beagle
- [20:21:01] * arun (~arun@unaffiliated/sindian) has joined #beagle
- [20:24:58] * amitk (~amit@a91-154-124-12.elisa-laajakaista.fi) Quit (Quit: leaving)
- [20:32:18] * Meizirkki (~Meizirkki@ZYMMDCCXLI.gprs.sl-laajakaista.fi) Quit (Remote host closed the connection)
- [20:33:53] <djlewis_>
hmmm, and I was going to offer to carry mcooks bags.
- [20:34:16] <_av500_>
djlewis_: i have a driveway to pav
- [20:34:19] <_av500_>
pave
- [20:34:36] <djlewis_>
That might be a nice vacation area :)
- [20:36:29] <_av500_>
yes
- [20:36:32] <_av500_>
its nice
- [20:36:38] <_av500_>
once paved
- [20:37:31] <djlewis_>
all the beer and ppizza I can want for?
- [20:37:35] <djlewis_>
pizza
- [20:37:41] * cfriedt (~cfriedt@modemcable178.225-21-96.mc.videotron.ca) Quit (Ping timeout: 245 seconds)
- [20:38:28] <_av500_>
hmm
- [20:38:30] <_av500_>
tempting
- [20:43:42] <djlewis_>
here in the US we hire a company that sends a big hoss dump truck and a road masher to lay pavement
- [20:44:10] <djlewis_>
pavement is mixed and in a hot gooey state
- [20:44:42] <djlewis_>
the road masher spreades it out like pizza dough :)
- [20:44:58] <_av500_>
pizza again
- [20:45:08] <djlewis_>
hehee
- [20:45:44] <djlewis_>
so I'll come over there, eat my pizza, drink the beer and call the pavement co for you :)
- [20:47:14] <djlewis_>
_av500_: just email me your credit card numbers and we'll get this show on the road.
- [20:47:41] * topfs2 (~topfs2@xbmc/staff/topfs2) Quit (Ping timeout: 240 seconds)
- [20:55:01] * gustavoz (~gustavoz@host212.200-82-38.telecom.net.ar) Quit (Quit: Leaving)
- [20:57:04] * topfs2 (~topfs2@xbmc/staff/topfs2) has joined #beagle
- [20:57:13] * raster (~raster@enlightenment/developer/raster) has joined #beagle
- [21:00:18] * b7500af1 (~GH@2001:468:c80:4280:21c:bfff:fe8b:90b4) Quit (Quit: Leaving)
- [21:00:21] * b7500af1 (~GH@2001:468:c80:4280:21c:bfff:fe8b:90b4) has joined #beagle
- [21:05:06] * b7500af1 (~GH@2001:468:c80:4280:21c:bfff:fe8b:90b4) Quit (Ping timeout: 252 seconds)
- [21:06:45] * Xerion (~xerion@82-170-197-160.ip.telfort.nl) Quit (Quit: )
- [21:08:07] * awozniak (~awozniak@adsl-76-205-222-173.dsl.snlo01.sbcglobal.net) Quit (Ping timeout: 240 seconds)
- [21:12:05] * arun__ (~arun@212-123-133-164.ip.telfort.nl) has joined #beagle
- [21:12:30] * arun (~arun@unaffiliated/sindian) Quit (Ping timeout: 260 seconds)
- [21:18:55] * jpirko (~jirka@ip-89-103-91-100.net.upc.cz) Quit (Quit: Leaving)
- [21:22:18] * ceyusa (~vjaquez@95.61.238.167) has joined #beagle
- [21:23:48] * raster (~raster@enlightenment/developer/raster) Quit (Quit: Gettin' stinky!)
- [21:24:45] * xMDKx (~mdkcore@nat/mandriva/x-gkdkeovycfchohyc) Quit (Read error: Connection reset by peer)
- [21:30:21] * awozniak (~awozniak@adsl-76-205-222-173.dsl.snlo01.sbcglobal.net) has joined #beagle
- [21:30:42] * mrc3_ (~mrc3@nat/ti/x-necogpmyuxhuvped) Quit (Remote host closed the connection)
- [21:31:07] * mrc3_ (~mrc3@nat/ti/x-fwvmviyewsvtycha) has joined #beagle
- [21:33:07] * sifr (~sifr@host86-163-223-16.range86-163.btcentralplus.com) has joined #beagle
- [21:33:40] <sifr>
question: how can I use the beagle board in conjunction with a bread baord? Is it possible or am I talking nonsense? Ta.
- [21:35:15] * emeb (~ericb@ip72-223-81-194.ph.ph.cox.net) has left #beagle
- [21:36:03] <prpplague>
sifr: you can purchase a trainer board: http://www.elinux.org/BeagleBoard_Trainer
- [21:36:10] <prpplague>
sifr: it works well with breadboard
- [21:36:29] <sifr>
prpplague: the trainer board or the actual beagle?
- [21:37:02] <prpplague>
sifr: trainerboard is designed to work with the beagle to enable developers to easy interface to some of the I/O
- [21:37:38] <sifr>
prpplague: ah i see
- [21:37:50] <prpplague>
sifr: the trainer can easily be used with a small breadboard like this - http://www.sparkfun.com/commerce/product_info.php?products_id=8802
- [21:38:36] <prpplague>
sifr: in addition the trainer also has an atmega328 which is arduino compatible interfaced to the beagle via a uart
- [21:39:25] <prpplague>
sifr: ( a note, i designed the trainer so i am biased, but it sounds like that is what you need)
- [21:40:18] <sifr>
prpplague: cool, but I am buying beagle to avoid adruino if i can, I want something that would let me play with components and is linux compatible to let me utilise all that software already outthere
- [21:40:39] <prpplague>
sifr: the trainer will do that
- [21:40:52] <prpplague>
sifr: the atmega328 is just there as an additional feature
- [21:41:29] <sifr>
prpplague: hmm ok and do I use the trainer board on its own or do I use it in conjunction with beagleboard?
- [21:41:43] <prpplague>
sifr: it is designed to mate to the beagleboard
- [21:41:53] <sifr>
oh cool
- [21:41:55] * topfs2 (~topfs2@xbmc/staff/topfs2) Quit (Read error: Connection reset by peer)
- [21:42:25] <sifr>
all i need to do now is find a uk ditributor now :)
- [21:42:50] <prpplague>
sifr: tincantools ships overseas, just use the USPS as the shipping option, it is pretty cheap
- [21:43:09] <sifr>
k cheers. i'll have a look see.
- [21:43:16] <prpplague>
http://www.tincantools.com
- [21:43:26] * arun__ is now known as arun
- [21:43:34] <sifr>
yes I've been there before
- [21:44:40] <prpplague>
sifr: here is an example where koen in controlling a stepper motor using the trainer and beagle - http://www.youtube.com/watch?v=sz0erpdWtFY&feature=related
- [21:44:52] <sifr>
thanks, quick question is there a limit to how many components I could attach to it using the trainer board? Or can I chain multiple things together?
- [21:45:40] <sifr>
prpplague: wow thanks alot man. I've spent a good few hours trying to decide between beagle and adruino i think you have convinced me.
- [21:48:47] <prpplague>
sifr: you could add a header and pass the signals out to other boards
- [21:50:06] * PBansal (~pbansal@nat/ti/x-qgxithhtrchmpuev) has joined #beagle
- [21:53:32] * scrp3l_ (~scrp3l@201.250.140.237) has joined #beagle
- [21:56:26] * rsalveti (~rsalveti@200.184.118.130) Quit (Remote host closed the connection)
- [21:57:22] * scrp3l (~scrp3l@201.250.186.175) Quit (Ping timeout: 264 seconds)
- [22:09:41] * billmar (48d089fe@gateway/web/freenode/ip.72.208.137.254) Quit (Quit: Page closed)
- [22:21:16] * rhk (~rhk@adsl-75-44-25-53.dsl.sfldmi.sbcglobal.net) Quit (Quit: ChatZilla 0.9.86 [Firefox 3.6.6/20100625231939])
- [22:28:45] * calculus (~calculus@gentoo/user/calculus) Quit (Ping timeout: 248 seconds)
- [22:30:36] * jayabharath (~a0866114@nat/ti/x-lxbfgnpnxzsroecw) Quit (Quit: Leaving.)
- [22:31:32] * brijesh (~brijesh@nat/ti/x-ounwxeioktnyvfls) Quit (Quit: Leaving)
- [22:35:12] * scrp3l_ (~scrp3l@201.250.140.237) Quit (Quit: Ex-Chat)
- [22:36:30] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) has joined #beagle
- [22:36:33] * iWill (~kvirc@unaffiliated/iwill) Quit (Ping timeout: 252 seconds)
- [22:36:55] * ceyusa (~vjaquez@95.61.238.167) Quit (Quit: leaving)
- [22:37:03] * prpplague (~danders@nat/ti/x-wgwvpjwspxvzxnci) Quit (Quit: Leaving)
- [22:40:56] * ppotera (~ppotera@99-100-130-116.lightspeed.stlsmo.sbcglobal.net) Quit (Client Quit)
- [22:42:52] * kmargar (~markos@athedsl-423319.home.otenet.gr) has joined #beagle
- [22:43:43] * markos_ (~markos@athedsl-426106.home.otenet.gr) Quit (Ping timeout: 260 seconds)
- [22:45:25] * Kayin_ (mversteege@shell2.skyberate.net) Quit (Remote host closed the connection)
- [22:49:04] * sifr (~sifr@host86-163-223-16.range86-163.btcentralplus.com) Quit (Quit: Lost terminal)
- [22:50:48] * maria (~mrodrigue@186.32.57.22) Quit (Quit: Leaving.)
- [22:57:37] * raster (raster@enlightenment/developer/raster) has joined #beagle
- [23:00:14] * florian (~fuchs@Maemo/community/contributor/florian) Quit (Quit: Verlassend)
- [23:01:28] * _don_ (~don@nat/ti/x-hxswneobhrifcinq) Quit ()
- [23:01:32] * calculus (~calculus@gentoo/user/calculus) has joined #beagle
- [23:02:15] * Redb3ard (~SF0010MAC@lbb-dslst.69616810.amaonline.com) Quit (Quit: Redb3ard)
- [23:04:00] * buZz (~buzz@bydogen.stoned-it.com) Quit (Ping timeout: 260 seconds)
- [23:05:24] * buZz (~buzz@bydogen.stoned-it.com) has joined #beagle
- [23:05:25] <djlewis_>
later....
- [23:05:28] * djlewis_ (41401e0d@gateway/web/freenode/ip.65.64.30.13) Quit ()
- [23:07:49] * kanru (~kanru@118-168-235-45.dynamic.hinet.net) has joined #beagle
- [23:12:11] * utd_student (816e055a@gateway/web/freenode/ip.129.110.5.90) has joined #beagle
- [23:14:17] * _don_ (~don@nat/ti/x-aienzsfavjsspsmj) has joined #beagle
- [23:15:30] <utd_student>
How do i start writing code for DSP? Do I use CCS v4?
- [23:16:11] * _don_ (~don@nat/ti/x-aienzsfavjsspsmj) Quit (Client Quit)
- [23:17:03] <utd_student>
Is there a wiki for DSP programming of BeagleBoard?
- [23:17:18] * CHRIS_7 (3fa44003@gateway/web/freenode/ip.63.164.64.3) Quit (Quit: Page closed)
- [23:18:38] <tharvey>
utd_student, have you googled pixhawk? may be where you want to start
- [23:18:44] * _don_ (~don@nat/ti/x-dtkjlqiadrgpsevs) has joined #beagle
- [23:18:45] * _don_ (~don@nat/ti/x-dtkjlqiadrgpsevs) Quit (Client Quit)
- [23:19:02] * kanru (~kanru@118-168-235-45.dynamic.hinet.net) Quit (Ping timeout: 265 seconds)
- [23:19:22] <utd_student>
tharvey : I've gone thru pixhawk. it is quite outdated. but if that is the best then i'll follow that
- [23:20:13] <tharvey>
its the only thing I've seen but I haven't looked recently - I'm assuming you actually want to use the DSP for your own code as opposed to say use it for standard video encoding in which case you would go a different path
- [23:20:39] <utd_student>
yes. i have a computer vision algorithm written in C
- [23:20:48] <utd_student>
it works great on the ARM but is slow
- [23:21:01] <utd_student>
so i want to leverage the DSP side to speed up certain functions
- [23:21:21] <utd_student>
i cant get my head around gpp side and dsp side programming methods
- [23:22:17] <tharvey>
I hear ya. I took a 3 day course from TI involving the boundary and it didn't even touch doing anything on the dsp side - was more high level
- [23:22:32] <tharvey>
and more focused on the linux side
- [23:23:34] <tharvey>
TI has a framework that is typically used to handle the complexities of resources that are shared by both the GPP and DSP and the framework itself is pretty complex, as is the toolset that's evolved around it
- [23:24:03] * kanru (~kanru@118-168-235-45.dynamic.hinet.net) has joined #beagle
- [23:25:25] <utd_student>
There are too many things that I cant put together. dsplink, dspbridge, codec engine, dsplibrary. I mean which of these do I use? There is too much documentation!
- [23:26:14] <tharvey>
yup, those are all the things covered at a high level in the course I took - if you end up finding a good site that goes over this please mention it here. The TI site may have some documentation there
- [23:26:45] <utd_student>
Hmm..ok. Guess my search continues..
- [23:27:12] <utd_student>
Thank you!
- [23:27:47] <tharvey>
dspbridge and dsplink are whats used to move data between the GPP and DSP iirc, codec engine is the framework that exists both on the GPP and the DSP side to manage codecs in a way that they can share resources, dsplibrary is a library you can use for code on the DSP itself
- [23:28:49] <tharvey>
you don't have to operate in the codec engine framework if you don't want to - that was a layer added so that codecs from various sources could interoperate and one codec couldn't necessarily stomp on another - I think of it as an OS layer managing resources
- [23:29:17] <utd_student>
Yeah I dont need any codecs
- [23:29:28] <tharvey>
some of the terminology used seems strange to me, they call the blob of DSP code that gets loaded onto the DSP from codec engine the 'server' for example
- [23:29:32] <utd_student>
So I installed dsplink on the beagleboard
- [23:30:54] <tharvey>
dsplink is the userland component iirc and dsbpridge is the kernel module that dsplink interfaces with
- [23:30:58] * theholyduck (~holyduck@ip-199-138-106-77.eidsiva.net) Quit (Read error: Connection reset by peer)
- [23:31:41] <tharvey>
you would use it to load your dsp blob and to shuttle data to/from it from the GPP - you would in that case build your dsp blob using the c64x compiler and dsplib
- [23:32:06] <utd_student>
Oh ok. Can I use CCS for that?
- [23:32:25] <tharvey>
I imagine thats the same model that pixhawk uses and I would imagine all the info on their site is still up to date - I don't know that they use the codec engine framework
- [23:32:35] <tharvey>
ccs?
- [23:32:57] <tharvey>
afaik you would use ti_cgt_c6000_6.1.9
- [23:33:14] * 20QABBIG8 (~cmurillo@186.32.57.22) Quit (Quit: Leaving.)
- [23:33:40] <tharvey>
looks like pixhawk does fit into the codec engine
- [23:33:42] <utd_student>
I installed ti_cgt_c6000_6.1.9 but am confused as to how to use it. Do I just compile using the cl6x
- [23:34:09] <tharvey>
are you using openembedded to build your firmware for beagle?
- [23:35:19] <utd_student>
well no
- [23:35:21] <utd_student>
i'
- [23:35:38] <utd_student>
i've just installed from the angstrom distribution page
- [23:35:58] <tharvey>
you may need to build your own
- [23:36:07] <utd_student>
i have a version that i built using oe. but i was not able to install the dsplink examples due to md5sum error
- [23:36:37] <utd_student>
so koen instructed that the dsplink versions are in the angstrom feeds and just need to opkg install dsplink
- [23:37:00] * Redb3ard (~SF0010MAC@75.110.202.83) has joined #beagle
- [23:37:00] <tharvey>
not sure if angstrom packages exist for all the pieces - there are some ti recipes that need sources you have to download directly form their site
- [23:37:58] <tharvey>
currently it appears that OE recipes that angstrom is built from have progressed to the point that the only thing you need to download manually is ti_cgt_c6000, so you may be fine
- [23:38:26] * BThompson (~a0193480@nat/ti/x-lposrfazzdiukfvm) Quit (Remote host closed the connection)
- [23:38:33] <tharvey>
you may learn quite a bit by disecting the pixhaw recipes and see what they build and how they do it
- [23:38:54] <tharvey>
but you will have to learn a bit of bitbake (what Openembedded uses) likely to understand them
- [23:39:53] <utd_student>
I have a limited undestanding of bitbake because I used it to build the beagleboard firmware, which I did not use in the end however
- [23:39:57] <tharvey>
another example I can think of us the gst-ti project that uses codec engine to build gstreamer plugins to access dsp based encode/decode on the TI processors - those recipes are in OE as well and could be educational
- [23:39:59] <utd_student>
I also build the dsplink examples
- [23:40:20] * jrmuizel (~jrmuizel@mozca02.ca.mozilla.com) Quit (Quit: jrmuizel)
- [23:40:36] <utd_student>
Do I just go through the recipes?
- [23:41:24] <tharvey>
the recipes are what bitbake uses to build the packages that you downloaded or installed from your angstrom feed
- [23:41:59] <tharvey>
those recipes show you how the dsplink examples were built, thus you can try to make sense of how they work I assume
- [23:42:21] <tharvey>
myself, I haven't built those so I'm not sure if the examples do anything useful
- [23:43:12] * mpoirier (~quassel@S0106002369de4dac.cg.shawcable.net) Quit (Ping timeout: 260 seconds)
- [23:43:33] <utd_student>
I thought the recipes are just like makefiles, telling the compiler how to link and build all the files
- [23:44:16] <tharvey>
yes, thats pretty much what they are, instructions on how to download sources, patch them, configure them, build them, and package them
- [23:44:50] <tharvey>
I'm just saying if you want to understand how the dsplink examples work you'll need to find the source code etc - look in the recipes for that info
- [23:46:10] <utd_student>
oh ok. thanks!
- [23:48:26] <tharvey>
no prob
- [23:55:47] * utd_student (816e055a@gateway/web/freenode/ip.129.110.5.90) has left #beagle
- [23:59:21] * GrizzlyAdams (~Grizzly@ip98-184-88-41.mc.at.cox.net) has joined #beagle
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.