#beagle IRC Log on BeagleBoard.org

Join the chat at beagleboard.org/chat

IRC Log for 2008-08-19

Timestamps are in UTC.

  1. [00:03:11] * RogerMonk (n=a0740758@nat/ti/x-0f2a6cad7b3fdc73) Quit (Remote closed the connection)
  2. [00:07:32] * Olipro (n=Olipro@unaffiliated/olipro) has joined #beagle
  3. [00:14:04] <cian> DDevine: I have been looking at the small LCD problem
  4. [00:14:25] <cian> I have concluded that the best solution is to build your own
  5. [00:14:46] <cian> the 4.3" PSP LCD on sparkfun with a TFS101a from TI should do just fine
  6. [00:15:54] <dcordes> cian: how to connect the controller?
  7. [00:16:19] <cian> dcordes: what controller?
  8. [00:16:28] <cian> sorry I meant TFP101a
  9. [00:16:45] <cian> it's a DVI to 24 pin parallel LCD interface
  10. [00:17:19] <cian> so you go 24 pin parallel on the OMAP to the TFP410 on the beagle via DVI to a TFP101a to the LCD
  11. [00:18:22] <dcordes> I guess TFP101a is what I asked for
  12. [00:18:45] <cian> ok
  13. [00:18:47] <cian> http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=NHD-4.3-480272YF-ATXI%23-T-ND
  14. [00:18:53] <cian> is a nice option if you want touch as well
  15. [00:19:23] <cian> If you feel like cheating you could string the ADC for the touch sensor onto the HDMI ports I2C lines
  16. [00:19:31] <cian> and avoid using a 2nd cable
  17. [00:19:43] <dcordes> dope
  18. [00:20:10] <dcordes> otherwise one would use the expansion i2c?
  19. [00:20:27] <cian> yes, or a usb device
  20. [00:20:39] <dcordes> cian: did you read sound worked for me?
  21. [00:20:47] <dcordes> to a certain extent
  22. [00:20:49] <cian> yes, I saw that
  23. [00:20:53] <cian> good news!
  24. [00:21:13] <dcordes> it's still a mystery to me why the other uboot gets rid of panic
  25. [00:21:13] <cian> incidentally tsc2004 is probably what you need for the touch screen
  26. [00:22:01] <ds2> hmmm 100 pin QFP Hmmmmm
  27. [00:22:12] <ds2> you fabbing a board?
  28. [00:22:44] <cian> ds2: yes, I've got access to CNC milling machines
  29. [00:22:48] <cian> so I can do PCBs in house
  30. [00:22:53] <ds2> I see
  31. [00:23:14] <ds2> what kind of bits do you use to route traces for these high pitch devices?
  32. [00:23:42] <cian> One sec, I'll pull up the site...
  33. [00:24:11] * mru hasn't done PCBs since the university days
  34. [00:24:11] <ds2> micro100?
  35. [00:24:21] <mru> and we had no luxuries like cnc
  36. [00:24:32] * Olipro_ (n=Olipro@unaffiliated/olipro) Quit (Connection timed out)
  37. [00:24:57] <ds2> if there is enough interest, it may be cheaper to get a batch of boards made up
  38. [00:25:22] <ds2> the TQFP can be oven assembled and the rest can be part of a kit
  39. [00:25:44] <cian> http://www.thinktink.com/stack/volumes/voli/store/mechmill.htm is what I'm using
  40. [00:25:54] <cian> http://www.millpcbs.com/index.php?page=cutters has a discussion on bits
  41. [00:26:03] <cian> ds2: I'd love to see it become a kit
  42. [00:27:09] <ds2> cian: but are there more interest then just me and you =)
  43. [00:27:23] <cian> ds2: who knows?
  44. [00:27:35] <ds2> cian: any problems with burrs on these bits? also how fast do you run them?
  45. [00:28:07] <cian> ds2: We run them a 10,000rpm with a feed rate of 128
  46. [00:28:37] <cian> and yes we've had trouble with burrs, I forget how we solved it
  47. [00:28:54] <cian> I'm a medic, not an engineer so I don't get too involved with the actual fabrication
  48. [00:29:34] <dcordes> oh the TFP101a has 100 pins. now I get what you guys talk about
  49. [00:29:48] <cian> yes, but you can actually ignore 24 of them
  50. [00:29:53] <dcordes> great :D
  51. [00:30:08] <cian> it's designed to be able to do 2 pixels per clock
  52. [00:30:17] <cian> but the samsung panels only want 1 pixel per clock
  53. [00:30:33] <dcordes> is there no easier way?
  54. [00:30:54] <cian> any 4.3" LCD products I could find were S-Video in only
  55. [00:31:09] <dcordes> s-video?
  56. [00:31:10] <cian> in an ideal world we'd be able to access the 24 parallel lines on the OMAP processor
  57. [00:31:27] <ds2> cian: ah, I ask cuz the they look quite different form the two bits I have used before
  58. [00:31:46] <cian> dcordes: yup, designed for cars and such
  59. [00:32:08] <ds2> there might be other hobbiest use for a DVI-D to raw LCD converter boards
  60. [00:32:31] <ds2> a rough estimate is the bom cost around $30 in quantity
  61. [00:33:00] <cian> ds2: tbh, it might just be best to give a layout to sparkfun and let them do the rest...
  62. [00:33:53] <ds2> cian: I know how to get it through the process and wouldn't mind doing it as a side project, just donno if even 100 boards would sell
  63. [00:34:13] <cian> ds2: are you @ti?
  64. [00:34:16] <dcordes> if you would get a link on bb.org :)
  65. [00:34:33] <dcordes> I doubt it would be hard to sell such devices
  66. [00:34:35] <ds2> cian: nope
  67. [00:34:49] <cian> ok, just curious
  68. [00:35:00] <ds2> dcordes: famous last words of all those other businesses that went bankrupt
  69. [00:35:05] <cian> ds2: I agree that 100 boards would easily sell
  70. [00:35:25] <cian> Incidentally I've got parts for 10 on my desk at the moment
  71. [00:35:36] <ds2> nice
  72. [00:35:44] <dcordes> we talk about dvi-d to whatever the small lcds use?
  73. [00:35:51] <ds2> you got the right DVI-D connector?
  74. [00:36:13] <cian> ds2: I actually did the same thing as the beagle board, I got a HDMI connector to save space
  75. [00:36:25] <ds2> ah
  76. [00:36:53] <cian> I wonder what the DVI spec says about putting random crap on the I2C bus
  77. [00:37:32] <ds2> do we really care about the real spec?
  78. [00:37:33] <ds2> :)
  79. [00:37:36] <cian> for myself I like the idea of having a tsc2004 on the I2C bus, but I don't think it would be a good idea to market a board with that
  80. [00:37:56] <dcordes> why not?
  81. [00:38:15] <dcordes> I have a device that uses a tsc2003 or similar
  82. [00:38:17] <ds2> do you envision the board sitting in a seperate box then the beagle board?
  83. [00:38:21] <dcordes> at least it's driven by tsc2003.c
  84. [00:38:47] <cian> also, I haven't bothered to put an actual I2C EEPROM on my design, so EDID doesn't work
  85. [00:39:27] <ds2> cian: do you have a working prototype?
  86. [00:39:42] <cian> ds2: wirewrapped, no touch yet
  87. [00:39:51] <ds2> wirewrapped?!?!?!?!?!?!?!?!?!
  88. [00:39:59] <vlad_> ah sweet, I forgot my power adapter at home, and I found I have a 5V/4A adapter sitting under my desk
  89. [00:40:07] <ds2> 65MHz signals though through wirewrapped circuits? Hmmm
  90. [00:40:09] <cian> going to mill in about 7 hours
  91. [00:40:34] <ds2> I'd be very interseted to see how well it turned out
  92. [00:40:36] <cian> ds2: it only barely works & I clocked it waaay lower
  93. [00:40:47] <cian> It's not connected to the beagle at the moment
  94. [00:41:03] <cian> because I don't have a HDMI to HDMI cable to spare
  95. [00:41:07] <ds2> was just looking at these parts last week and figure it is too much effort for a 1 one off and no one else is crazy enough to try it
  96. [00:41:26] <dcordes> is that when you have the chips and all connected with 1000 little cables?
  97. [00:41:44] <cian> ds2: DDevice was looking for something like this about an hour and a half ago
  98. [00:41:53] <cian> that's why I bothered to mention it
  99. [00:41:57] <ds2> wire wrapping is when you have thin 30G wire twisted onto posts to do interconnect
  100. [00:42:18] <cian> dcordes: I used a QFP breakout board
  101. [00:42:25] <dcordes> what's that?
  102. [00:42:27] <ds2> cian: If you don't mind, I'd be interested in carrying on the design
  103. [00:43:03] <cian> ds2: I can send you eagle files in about 24 hours if you want
  104. [00:43:11] <dcordes> I will buy one or two.
  105. [00:43:15] <ds2> cian: please do
  106. [00:43:51] <cian> and I'd love to see someone carry on the design
  107. [00:44:11] <ds2> cian: are you in the US?
  108. [00:44:22] <cian> nope
  109. [00:44:48] <cian> I am currently doing medical engineering research in Trinity College Dublin, Ireland
  110. [00:45:06] <cian> from October I will be back in the University of Bristol Medical School in England
  111. [00:45:23] <Crofton|work> my parents went to Bristol
  112. [00:45:30] <cian> small world!
  113. [00:45:47] <Crofton|work> yeah
  114. [00:46:03] <Crofton|work> I walked by Trinity in May :)
  115. [00:46:23] <cian> Crofton|work: where are you based?
  116. [00:46:36] <Crofton|work> Blacksburg Vrginia, USA
  117. [00:47:04] <cian> what did you think of Dublin?
  118. [00:48:34] <cian> ds2: on a totally off topic note, have you ever used anything in TI's http://www.ti.com/litv/pdf/mtss002d (TSSOP (DA)) package
  119. [00:51:34] <Crofton|work> Dublin was a nice ppalce
  120. [00:51:46] <ds2> cian: not from TI, but I have used other TSSOP's (glue logic/74xx stuff)
  121. [00:51:52] <cian> The msp430f2274 is currently driving me insane, because it's a 38 pin TSSOP but has a pin pitch of .65mm, where most TSSOP >32 pins are .5mm pitch
  122. [00:52:44] <ds2> time to become friends with the eagle library editor
  123. [00:53:43] <cian> ds2: i've got packages on eagle, that's not the problem, it's having to build up boards for it because I can't get a commercially available breakout board at .65mm pitch
  124. [00:53:54] <cian> also can't get a zif socket for programming them
  125. [00:54:14] <ds2> is the f2274 biwire capable?
  126. [00:54:22] <ds2> I rarely use breakout boards
  127. [00:55:07] <cian> ds2: yes, so i've put test points on my board to program it after soldering
  128. [00:56:20] <ds2> or add pins to mate it with the ez430
  129. [00:57:14] <cian> done than too on some boards, I had to hunt down an alternative mating product for the ez430 on friday, digikey are out of stock on the maxmill part that TI uses
  130. [00:58:39] * rsalveti (n=salveti@200.184.118.132) Quit (Read error: 113 (No route to host))
  131. [00:59:56] <ds2> checked mouser?
  132. [01:03:03] <cian> mouser international shipping is slow and expensive
  133. [01:03:14] <cian> sourcing from ireland is a pain
  134. [01:03:34] <cian> ds2: looks like I'm not the only one that uses DVI I2C for weird stuff http://www.paintyourdragon.com/uc/i2c/index.html
  135. [01:05:16] * rsalveti (n=salveti@189.70.119.44) has joined #beagle
  136. [01:05:47] <ds2> nice
  137. [01:06:18] <JJS> i like the software/hardware compatibility images
  138. [01:08:18] <cian> yes, they're good
  139. [01:08:20] <vlad_> haha
  140. [01:09:41] <DDevine> lol
  141. [01:09:46] <DDevine> i like this site.
  142. [01:10:39] * DDevine is going to go afk again. I have to go buy a damn roller ruler because the school spent all their money on micrsoft licenses/
  143. [01:12:20] <cian> DDevine: have you been following our thoughts on a DVI 4.3" LCD?
  144. [01:12:36] <DDevine> I quickly read before
  145. [01:12:49] <DDevine> ill copy the info into a text file now.
  146. [01:13:18] <JJS> the dvi interface adds an unexpected wrinkle to a car pc application
  147. [01:14:14] <JJS> i did find a 7" lcd with dvi in but they're around $400
  148. [01:14:27] <JJS> i think 4" would be perfect
  149. [01:15:11] <DDevine> i agree JJS.
  150. [01:15:43] <DDevine> OK I have to go buy the $7 piece of plastic that the school was meant to provide me with...
  151. [01:17:00] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  152. [01:17:36] <cian> ds2: I'll keep you posted on my progress with the TFP101a, etc
  153. [01:17:39] * cian (n=cianh@cian.ws) Quit ("Foo!")
  154. [01:24:14] * nelson (n=nelson@rrcs-72-43-17-175.nys.biz.rr.com) has joined #beagle
  155. [01:52:01] <dcordes> In order to change my resoluttion from 1024x768 to 1280x720, I set the according omapfb kconfig (FB_OMAP_092M9R=y). now in the demo image, fbset reports mode "640x480-144"
  156. [01:52:04] <dcordes> is this a bug?
  157. [01:52:46] <ds2> sure you are booting the right image?
  158. [01:58:15] <dcordes> yes absolutely, else the resolution wouldn't have changed
  159. [01:58:34] <dcordes> (in fbset)
  160. [01:58:54] <ds2> did you also the clock?
  161. [01:59:14] <ds2> maybe you up'ed the resolution enough to result in an invalid refresh
  162. [02:00:10] <dcordes> Xfbdev draws a 640x480 frame in the top right corner
  163. [02:00:44] <dcordes> and it's the right alignment
  164. [02:01:02] <dcordes> it looks like if it would draw fullscreen nowm this would be the 1280x720
  165. [02:03:14] <dcordes> ds2: I didn't change anything about clock. only the mentioned kconfig
  166. [02:05:45] <ds2> dcordes: has anyone tried that resolution?
  167. [02:06:44] <dcordes> ds2: let me check the ml
  168. [02:09:51] <dcordes> ds2: hm I only see people speaking of 720p video playback, not using it as fb resolution
  169. [02:13:59] <ds2> dcordes: it may need a the clock patch
  170. [02:14:55] <dcordes> ds2: is there a bugtracker or so for the kernel?
  171. [02:28:14] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  172. [02:28:17] * chakie_work (n=jeh@vpn-pub2.hard.ware.fi) Quit (Read error: 60 (Operation timed out))
  173. [02:28:46] * chakie_work (n=jeh@vpn-pub2.hard.ware.fi) has joined #beagle
  174. [02:33:56] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) Quit (Read error: 60 (Operation timed out))
  175. [02:48:17] <ds2> nope
  176. [02:49:48] <dcordes> ds2: I doubt it's a problem with bus speed or clock or anythign like that, because it draws fine, just in that small 640x480 window. wouldn't an insufficient clock cause display corruption?
  177. [02:50:15] <ds2> dcordes: the driver might have backed off
  178. [02:50:19] <dcordes> tbh, I have no clue what you mean with clock
  179. [02:55:17] * lgentili (i=bd2fa60e@gateway/web/ajax/mibbit.com/x-fb6724598e37415d) has joined #beagle
  180. [02:59:01] <ds2> to generate the display, it scans horizontally then moves vertically and repeats
  181. [02:59:16] <ds2> once it reaches the end, it goes back to the top and repeats
  182. [02:59:38] <ds2> well, each pixel is one tick of the pixel clock
  183. [02:59:49] <ds2> so for more dots, it takes longer to draw the entire display
  184. [03:00:23] <ds2> eventually, you get sufficient number of pixels, the time it takes to draw the entire display is too long; the time to draw and repeat is the refresh rate
  185. [03:00:31] * JuanG (n=Juan@nat/ti/x-80902fc00e87ee4f) has joined #beagle
  186. [03:00:47] * JuanG (n=Juan@nat/ti/x-80902fc00e87ee4f) has left #beagle
  187. [03:01:13] <ds2> so getting back to what I am saying, there is a by default a single pixel clock and I suspect if the refresh rate gets too low (i.e takes too long), the driver may bail and try a different mode all together
  188. [03:05:06] <dcordes> ds2: ok thanks. is it complicated/dangerous to alter the clock?
  189. [03:22:20] <ds2> nope
  190. [03:22:35] <ds2> there is a patch going around to bump up the clock
  191. [03:34:27] * lgentili (i=bd2fa60e@gateway/web/ajax/mibbit.com/x-fb6724598e37415d) Quit ("http://www.mibbit.com ajax IRC Client")
  192. [03:36:44] * FWMiller (n=fwmiller@c-67-162-139-200.hsd1.co.comcast.net) has joined #beagle
  193. [03:36:50] <FWMiller> greetings
  194. [03:37:00] <FWMiller> i hate to ask this again, somebody already told me once
  195. [03:37:09] <FWMiller> where can i get a serial cable with a ribbon connector?
  196. [03:39:03] <dcordes> FWMiller: you must look for d-sub to idc10. but look out, there are different variants of how they wire the d-sub. make sure it looks like this: http://www.pccables.com/07120.htm
  197. [03:39:48] <dcordes> if you're in the us it would be save to order it from that website I think
  198. [03:44:17] <dcordes> FWMiller: you can also make one out of a normal ide cable and a d-sub 9 pin cable
  199. [03:45:29] * JoeBorn (n=jborn@h-68-165-56-138.chcgilgm.dynamic.covad.net) has joined #beagle
  200. [03:50:43] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  201. [03:51:51] <FWMiller> don't think i have the tools to make my own
  202. [03:51:55] <FWMiller> anymore
  203. [03:53:00] <FWMiller> thanks for the link!
  204. [04:02:21] * JoeBorn (n=jborn@h-68-165-56-138.chcgilgm.dynamic.covad.net) Quit (Read error: 104 (Connection reset by peer))
  205. [04:04:39] * rsalveti (n=salveti@189.70.119.44) Quit (Read error: 104 (Connection reset by peer))
  206. [04:04:57] * rsalveti (n=salveti@189.70.119.44) has joined #beagle
  207. [04:06:13] <dcordes> FWMiller: you're welcome
  208. [04:06:20] <dcordes> bbl
  209. [04:15:58] * ZetaSyanthis (n=Administ@c-98-226-61-207.hsd1.il.comcast.net) has joined #beagle
  210. [04:16:02] * ZetaSyanthis (n=Administ@c-98-226-61-207.hsd1.il.comcast.net) has left #beagle
  211. [04:16:37] * JoeBorn (n=jborn@h-68-165-56-138.chcgilgm.dynamic.covad.net) has joined #beagle
  212. [04:26:29] * JJS is now known as JJS[out]
  213. [04:35:13] * FWMiller (n=fwmiller@c-67-162-139-200.hsd1.co.comcast.net) Quit ()
  214. [04:38:30] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  215. [04:39:50] * JoeBorn (n=jborn@h-68-165-56-138.chcgilgm.dynamic.covad.net) Quit (Read error: 104 (Connection reset by peer))
  216. [04:40:05] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  217. [04:43:35] * neilthereildeil (n=Administ@151.200.243.155) has joined #beagle
  218. [04:47:36] * JoeBorn (n=jborn@h-69-3-16-60.chcgilgm.dynamic.covad.net) has joined #beagle
  219. [04:48:54] * neilthereildeil (n=Administ@151.200.243.155) has left #beagle
  220. [05:03:12] * JoeBorn (n=jborn@h-69-3-16-60.chcgilgm.dynamic.covad.net) Quit (Read error: 60 (Operation timed out))
  221. [05:09:04] * Olipro_ (n=Olipro@unaffiliated/olipro) has joined #beagle
  222. [05:11:21] * nooomem (n=chatzill@ip68-100-201-32.dc.dc.cox.net) Quit ("ChatZilla 0.9.83 [Firefox 3.0.1/2008070208]")
  223. [05:11:28] * Olipro (n=Olipro@unaffiliated/olipro) Quit (Nick collision from services.)
  224. [05:11:28] * Olipro_ is now known as Olipro
  225. [05:15:40] * JoeBorn (n=jborn@h-68-164-228-217.chcgilgm.dynamic.covad.net) has joined #beagle
  226. [05:16:09] * t_s_o (n=tso@173.84-49-129.nextgentel.com) has joined #beagle
  227. [05:16:42] * Guest62082 (n=root@rrcs-208-125-50-2.nys.biz.rr.com) Quit (Read error: 110 (Connection timed out))
  228. [05:46:06] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  229. [05:55:52] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  230. [05:55:52] * JoeBorn (n=jborn@h-68-164-228-217.chcgilgm.dynamic.covad.net) Quit (Read error: 104 (Connection reset by peer))
  231. [06:11:18] * JoeBorn (n=jborn@h-68-164-11-248.chcgilgm.dynamic.covad.net) has joined #beagle
  232. [06:47:28] <khasim> JoeBorn: hi
  233. [06:47:58] * trickie (n=trickie@basesoft.xs4all.nl) has joined #beagle
  234. [07:04:07] * methril|laptop (n=methril@229.72.220.87.dynamic.jazztel.es) Quit (brown.freenode.net irc.freenode.net)
  235. [07:04:07] * denix (n=denys@nat/ti/x-292ca2ca61732e6c) Quit (brown.freenode.net irc.freenode.net)
  236. [07:04:07] * JJS[out] (n=josh@CPE-70-92-19-222.wi.res.rr.com) Quit (brown.freenode.net irc.freenode.net)
  237. [07:04:07] * [g2] (n=tom@nslu2-linux/g2) Quit (brown.freenode.net irc.freenode.net)
  238. [07:06:16] * [g2] (n=tom@nslu2-linux/g2) has joined #beagle
  239. [07:18:47] * methril|laptop (n=methril@87.220.72.255) has joined #beagle
  240. [07:31:28] * JoeBorn (n=jborn@h-68-164-11-248.chcgilgm.dynamic.covad.net) Quit (Killed by ballard.freenode.net (Nick collision))
  241. [07:32:47] * sweetlilm (n=sweet@196.37.229.15) Quit ()
  242. [07:41:05] * denix (n=denys@nat/ti/x-64625e2a2ac9cad0) has joined #beagle
  243. [08:07:25] * guillaum1 (n=gl@AMontsouris-153-1-67-145.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
  244. [08:15:50] * geraint (n=geraint@217.207.128.219) has joined #beagle
  245. [08:16:26] * geraint is now known as gnorth
  246. [08:39:46] <gnorth> Probably a bit early for most of the folks here, but I'm having some difficulty bringing up my Beagle Board with Angstrom Linux - I'm seeing the board reset at the "done, booting the kernel" message. To start with, I've got a couple of questions: 1) Is it considered reliable to use power-over-USB, or should I start using a dedicated 5V power supply, and 2) How do I determine the load and boot addresses to use for my particular uImage? (I'm using the stoc
  247. [08:40:47] * sweetlilm (n=sweet@196.37.229.15) has joined #beagle
  248. [08:41:33] <sweetlilm> can anyone help me out? I am getting the bootloader output on serial term, but can't seem to enter any characters, is there any specific term setup?
  249. [08:41:56] <kulve> sweetlilm: do you have hw flow control off?
  250. [08:42:12] <sweetlilm> I believe so, let me check
  251. [08:42:20] <kulve> gnorth: afaik, the power-over-usb isn't 100% reliable, but I'm not sure
  252. [08:42:42] <gnorth> Thanks - I'll try switching. I'm also getting resets randomly on "mmcinit", which is suspicious.
  253. [08:42:42] <kulve> gnorth: and your sentence was clipped: ...(I'm using the stoc
  254. [08:42:50] <gnorth> (I'm using the stock one downloaded from http://www.angstrom-distribution.org/demo/beagleboard/)
  255. [08:43:28] <sweetlilm> flow control is set to NONE
  256. [08:43:40] <sweetlilm> 115200 8N1
  257. [08:45:29] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  258. [08:45:55] <sweetlilm> hmmm nothing... weird
  259. [08:46:48] <gnorth> Are your characters having an effect, but not appearing, or not even being registered? (e.g. try "help" <enter> at the uboot prompt)
  260. [08:47:02] <sweetlilm> no effect at all
  261. [08:47:18] <sweetlilm> I get to: "OMAP3 beagleboard.org #"
  262. [08:47:48] <sweetlilm> maybe my cable is dodgy?
  263. [08:48:23] <DJWillis> sweetlilm: RX a problem? Or software setup, got another term soft you can try?
  264. [08:48:46] <sweetlilm> tried SecureCRT and HyperTerm, both exhibit the same behaviour
  265. [08:49:22] <DJWillis> sweetlilm: Putty? To be honest it should cable mind you.
  266. [08:51:07] <gnorth> What would the expected behavior be if your cable wasn't a full null-modem?
  267. [08:53:39] * sweetlilm is off to check the damn cable again
  268. [08:54:27] <DJWillis> gnorth: pass, 'weirdness' would get my vote as I never get consistent errors with cable probs ;)
  269. [09:04:10] * banderson (n=oe@69-71-183-7.mammothnetworks.com) Quit (Read error: 110 (Connection timed out))
  270. [09:11:48] * JJS[out] (n=njosh@CPE-70-92-19-222.wi.res.rr.com) has joined #beagle
  271. [09:20:47] <sweetlilm> bollocks, cable is fine :(
  272. [09:25:33] <gnorth> Are you using a USB->Serial bridge on the PC side? I've had problems with them in the past because often even if the cable is fully wired, the bridge isn't.
  273. [09:26:15] <sweetlilm> no I haven't even got there yet. Just trying to boot the Angstrom dist off SD (board is being powered by USB only)
  274. [09:27:34] <sweetlilm> what term should I use in linux?
  275. [09:30:52] <keesj> picocom!
  276. [09:31:09] <keesj> picocom -b 115200 /dev/ttyUSB0
  277. [09:32:54] <sweetlilm> hmm got cutecom, same behaviour as in windows, I think my RX line is screwed on the board
  278. [09:50:58] * methril (n=Methril@213.27.233.98) has joined #beagle
  279. [10:10:02] * Guest62082 (n=root@rrcs-208-125-50-2.nys.biz.rr.com) has joined #beagle
  280. [10:37:46] * sweetlilm (n=sweet@196.37.229.15) Quit ()
  281. [10:49:37] <gnorth> Having followed the instructions on http://elinux.org/BeagleBoard, I can't get my BeagleBoard to boot from the SD card. I have partition 1 DOS FAT32 and bootable, and I've tried with and without the User button pressed. Anything else I should try?
  282. [10:50:17] <gnorth> I thought it was booting MLO/uboot from the SD, but have just realised that I'm not getting the "Starting on with MMC" message
  283. [10:53:57] * Beagle6 (n=Beagle7@hunter.netmodule.com) has joined #beagle
  284. [10:53:58] <Beagle6> bb
  285. [10:54:47] * Beagle6 is now known as sprotch
  286. [10:57:21] * banderson (n=oe@69.71.183.7) has joined #beagle
  287. [11:03:41] * sweetlilm (n=sweet@196.37.229.15) has joined #beagle
  288. [11:13:13] <sweetlilm> still not able to send any input via serial to the board. RX must be broken :( *sob*
  289. [11:15:07] * ade_ (n=ade@80.177.64.36) has joined #beagle
  290. [11:26:41] <sweetlilm> is there any reason not to power the board off of USB. i.e. would that interfere with the serial comms at all?
  291. [11:39:41] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  292. [11:42:04] * DoZo2 (n=danielde@20-74-ftth.onsneteindhoven.nl) has joined #beagle
  293. [11:43:13] * JuanG (n=Juan@nat/ti/x-5f258676912f5ffa) has joined #beagle
  294. [11:43:55] * JuanG (n=Juan@nat/ti/x-5f258676912f5ffa) has left #beagle
  295. [11:59:59] <dcramer> sweetlilm: you were asking yest about the serial connector, no ?
  296. [12:01:33] <dcramer> http://www.pccables.com/07120.htm
  297. [12:02:21] <sweetlilm> I build one, and I am getting serial output, but when I type in the term, nothing happens, I think I have killed the RX line :(
  298. [12:02:27] <sweetlilm> err built
  299. [12:02:59] <sweetlilm> 3 different cable, 2 different machines 4 different terminal progs, same thing
  300. [12:05:05] <DoZo2> i've got 2 brackets, one pulled from an old machine, one bought. One works, the other doesn't.
  301. [12:05:39] <DoZo2> and HyperTerminal as the terminal prog
  302. [12:06:29] <DoZo2> <- happy because the ZHost usb cable works, so *finally* a working keyboard and mouse
  303. [12:07:40] <sweetlilm> lucky you :) I'm probably going to have to order another board :(
  304. [12:08:58] <DDevine> Gah, damn time zones...
  305. [12:09:08] <DDevine> I am in the middle of the EU and the US...
  306. [12:09:36] <DoZo2> isn't that kind of... wet?
  307. [12:10:00] <DDevine> theres a place... they call it australia....
  308. [12:10:08] <DoZo2> ah, the other way round :-)
  309. [12:10:18] <DDevine> i guess theres china just up a few thousand k as well.
  310. [12:10:44] <DDevine> I think i talk to the US in the morning and EU at night... i think...
  311. [12:11:02] <DDevine> ah well, we started talking about small screens for the beagle
  312. [12:11:18] <DoZo2> i like my big screen ;-)
  313. [12:11:38] <DDevine> I think when we find out what works and what doesnt we should write a small guide/large article on them.
  314. [12:11:50] <DDevine> because it is something that so many people have/will go through
  315. [12:12:16] <DDevine> not everybody wants to just plug it into a DVI monitor and leave it at that
  316. [12:12:34] <DDevine> there is many questions that people want simple answers for
  317. [12:12:37] <DoZo2> why would you want a small screen? Mobility?
  318. [12:12:53] <DDevine> Mobility and Pimpage.
  319. [12:13:03] <DDevine> small screens are all the rage these days
  320. [12:13:20] * cbrake_away is now known as cbrake
  321. [12:13:51] <DDevine> so if i get some time soon i will let both time zones know so they can pitch in some notes for me to work with.
  322. [12:14:33] <DoZo2> do you have a url that features a screen you are thinking of?
  323. [12:15:09] <DDevine> i know of a few screens
  324. [12:15:27] * nooomem (n=chatzill@ip68-100-201-32.dc.dc.cox.net) has joined #Beagle
  325. [12:15:52] <DDevine> personally I am going to buy a display from crystal fontz.
  326. [12:16:01] <DDevine> but most people want the colour LCD screens
  327. [12:17:16] <DDevine> http://www.crystalfontz.com/phpthumb/phpThumb.php?id=877&w=430 <-- I like this because i can also just make a cover and it turns into a great case.
  328. [12:17:37] <DDevine> it is sorta big, but that is ok because i want to have a battery pack inside it and stuff.
  329. [12:17:49] * robclark (n=robclark@nat/ti/x-23c23b000fb7eb41) has joined #beagle
  330. [12:18:05] <DDevine> also it will allow me to dock it right in the front of my computer
  331. [12:19:29] <DDevine> So yeah, i want to write a guide that gets all the facts in one place.
  332. [12:19:50] <DoZo2> what do you want to show on the lcd?
  333. [12:20:02] <DoZo2> what does it connect to on the beatle
  334. [12:20:06] <DDevine> it will be a mp3 player initially
  335. [12:20:10] <DoZo2> ok
  336. [12:20:15] <DDevine> so obviously artist song and album
  337. [12:20:29] <DDevine> and i can use the buttons for play/stop/pause/skip
  338. [12:20:52] <DDevine> but it could do much more obviously.
  339. [12:21:14] * DDevine is off to have a shower.
  340. [12:21:44] <DoZo2> so
  341. [12:21:49] <DoZo2> keyboard and mouse work
  342. [12:22:00] <DoZo2> next subject... OpenGL
  343. [12:26:27] * BThompson (n=BThompso@nat/ti/x-d5373fec04e361a8) has joined #beagle
  344. [12:39:41] <DDevine> DoZo2: that one connects through USB - they can also connect through RS232
  345. [12:40:24] <DDevine> but most other small tft screens connect through S-Video or VGA
  346. [12:40:33] <DDevine> which is a little problem...
  347. [13:11:56] * sprotch (n=Beagle7@hunter.netmodule.com) Quit ()
  348. [13:12:31] * DDevine (n=devine@123-243-187-85.static.tpgi.com.au) Quit (Read error: 60 (Operation timed out))
  349. [13:21:34] * DDevine (n=devine@123-243-187-85.static.tpgi.com.au) has joined #beagle
  350. [13:24:03] * _apk (n=kows@85-18-136-98.fastres.net) has joined #beagle
  351. [13:28:13] * rsalveti (n=salveti@189.70.119.44) Quit (Read error: 110 (Connection timed out))
  352. [13:29:28] <gnorth> Looks like my booting problems were due to bad power from the USB hub - powering it directly from my desktop's USB port works as expected, although I still can't boot u-boot off SD (but don't currently care)
  353. [13:31:30] <Crofton|work> gnorth, what brand hub?
  354. [13:33:25] <gnorth> Some generic piece of crap - our sysadmins did warn me that it might be rubbbish.
  355. [13:36:03] * Beagle4 (n=Beagle4@hunter.netmodule.com) has joined #beagle
  356. [13:36:23] * Beagle4 is now known as sprotch
  357. [13:46:24] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
  358. [13:47:03] * NishanthMenon (n=gnat@nat/ti/x-73d746a3b341554d) has joined #beagle
  359. [13:59:27] * dcordes (n=dcordes@unaffiliated/dcordes) Quit ("leaving")
  360. [13:59:27] * sprotch (n=Beagle4@hunter.netmodule.com) Quit (Read error: 104 (Connection reset by peer))
  361. [14:11:15] * rsalveti (n=salveti@200.184.118.132) has joined #beagle
  362. [14:20:13] * NishanthMenon (n=gnat@nat/ti/x-73d746a3b341554d) Quit ("Do fish get thirsty?")
  363. [14:24:56] * NishanthMenon (n=gnat@nat/ti/x-d15ace35d82ce52b) has joined #beagle
  364. [14:47:17] * DoZo2 (n=danielde@20-74-ftth.onsneteindhoven.nl) Quit (Read error: 110 (Connection timed out))
  365. [14:51:23] * [-ip-] (n=ts@mnhm-590e08cb.pool.einsundeins.de) has joined #beagle
  366. [14:57:30] <sweetlilm> OMG it boots!!!!! Lesson learned for today, NEVER, NEVER, NEVER, EVER! use a USB<->Serial adapter
  367. [14:58:14] * sweetlilm is off to hunt down a suitable 5V supply so I can plug in a keyboard
  368. [14:58:38] <uberfry> huh?
  369. [14:58:43] <uberfry> I thought usb host isn't working?
  370. [14:58:48] <sweetlilm> OTG
  371. [15:01:20] <nelson> sweetlilm: yeah, USB <--> Serial adapters aren't as compatible as one might wish.
  372. [15:01:43] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  373. [15:02:01] <sweetlilm> nelson: that is an understatement :) built 10 different serialcables until I figured that out :)
  374. [15:02:52] <nelson> RS-232 .... ya gotta love it (or NOT!)
  375. [15:03:22] <nelson> The truly sad thing is that we had a great RS-232 replacement in HP-IL, back in 1983.
  376. [15:04:08] <sweetlilm> you'll have to fill me in on that I was 9 :)
  377. [15:05:42] <nelson> Kids these days ...... get off my lawn!
  378. [15:07:15] <BThompson> there are probably people here who were not born yet ;)
  379. [15:08:41] <sweetlilm> Love being called a kid, do it again! :)
  380. [15:13:05] * nooomem (n=chatzill@ip68-100-201-32.dc.dc.cox.net) Quit ("ChatZilla 0.9.83 [Firefox 3.0.1/2008070208]")
  381. [15:16:57] * JoeBorn (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net) has joined #beagle
  382. [15:27:15] * trickie (n=trickie@basesoft.xs4all.nl) Quit ("Leaving")
  383. [15:38:38] * robclark (n=robclark@nat/ti/x-23c23b000fb7eb41) Quit ()
  384. [15:40:13] * robclark (n=robclark@nat/ti/x-089f3162d338fbe6) has joined #beagle
  385. [15:40:35] * methril is now known as methril|away
  386. [15:43:41] * _kamera2_ (n=opera@ipn36372-c52397.cidr.lightship.net) has joined #beagle
  387. [15:43:54] * sweetlilm (n=sweet@196.37.229.15) Quit (Read error: 104 (Connection reset by peer))
  388. [15:44:35] <_kamera2_> i read somewhere something about firefox ported to the QT environment. just wondering if QT has optimizations for the arm core, hence can FF potentially run faster on the beagle board.. any takers?
  389. [15:45:13] <chakie> i don't think it has that many optimizations
  390. [15:46:22] <chakie> i will try both the x11 and the embedded versions though when i get a board, i really like qt
  391. [15:46:53] <_kamera2_> what about opera.. i know they do make browsers for mobile devices.. although they do tend to have licenses for mobile devices..
  392. [15:47:41] <chakie> no idea about that
  393. [15:47:56] * Fragguel (n=rafa@alhakem.uco.es) has joined #beagle
  394. [15:47:58] <chakie> arora could be a candidate for a simplish browser
  395. [15:48:15] <chakie> not aUrora
  396. [15:49:05] <chakie> http://code.google.com/p/arora/
  397. [15:49:35] <chakie> dinner time
  398. [15:52:02] <uberfry> you guys wanna see how much I rule?
  399. [15:52:02] <uberfry> http://forums.maxconsole.net/showthread.php?t=121359
  400. [15:57:19] * gnorth (n=geraint@217.207.128.219) Quit (Read error: 113 (No route to host))
  401. [16:00:14] * _kamera2_ (n=opera@ipn36372-c52397.cidr.lightship.net) has left #beagle
  402. [16:03:51] <vlad_> plugging in a usb keyboard seems to kill my usb ethernet adapter. not ideal.
  403. [16:04:33] <DDevine> not very ideal at all.
  404. [16:10:21] <vlad_> of course, plan A was to use Synergy to just share my mouse/keyboard from another computer.. mouse works fine, but for some reason X doesn't get any keyboard input
  405. [16:27:19] * max_fr (i=max@212-198-79-163.rev.numericable.fr) has joined #beagle
  406. [16:27:24] <max_fr> Hi all
  407. [16:29:16] <vlad_> woo, working keyboard
  408. [16:29:33] <max_fr> I have to recompile the kernel, do you advise me 2007q3-51 or 2007q3-53 toolchain ?
  409. [16:39:13] * valhalla (n=valhalla@81-174-21-16.dynamic.ngi.it) has joined #beagle
  410. [16:53:07] * t_s_o (n=tso@173.84-49-129.nextgentel.com) Quit ("Konversation terminated!")
  411. [16:58:51] <DJWillis> max_fr: The ARM Linux one (53 IIRC), the other one is bare metal, not a lot of use for a kernel build ;)
  412. [17:02:06] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  413. [17:12:30] * guillaum1 (n=Guillaum@AMontsouris-153-1-67-145.w90-2.abo.wanadoo.fr) has joined #beagle
  414. [17:14:25] <max_fr> DJWillis: Thank you :-)
  415. [17:21:38] * dirk2 (n=dirk@F32e5.f.strato-dslnet.de) has joined #beagle
  416. [17:28:14] * JuanG (n=Juan@nat/ti/x-3f1586fb54984916) has joined #beagle
  417. [17:28:19] * JuanG (n=Juan@nat/ti/x-3f1586fb54984916) has left #beagle
  418. [17:29:37] <dirk2> max_fr, DJWillis: Sourcery G++ Lite 2007q3-51 for ARM GNU/Linux http://www.codesourcery.com/gnu_toolchains/arm/portal/release313 is the correct one for everything: U-Boot, kernel and user space (unless you avoid -Os)
  419. [17:31:45] <_apk> Hi! :)
  420. [17:32:51] <_apk> I'm going to buy a beagleboard....I have just one question: the only way that I'll have to connect it to a monitor, is using the s-video out. Is there a way do make the beagleboard to boot with s-video out enabled?
  421. [17:33:10] <max_fr> dirk2: Ok
  422. [17:33:25] <_apk> I was reading that debian will output video on the dvi-d out as default...
  423. [17:38:42] <ds2> should be possible but you can't use the l-o driver though
  424. [17:40:58] <khasim> _apk: It should be possible to get boot messages on s-video, you have change two lines of code in s/w
  425. [17:41:51] <ds2> know what those to lines are off the top of your head?
  426. [17:42:10] * ade_ (n=ade@80.177.64.36) Quit ()
  427. [17:42:24] <khasim> in display.c make TV for GFX pipeline instead of LCD
  428. [17:42:31] <khasim> I think only one line :)
  429. [17:42:33] <ds2> DOH
  430. [17:42:42] <ds2> and I was looking in drivers/video/omap @2S$##$%@#!#!@#
  431. [17:42:42] <_apk> khasim, I'm seeing that it'll be a good idea buying a null modem cable and use that!
  432. [17:42:58] <_apk> btw thank you for your answer :)
  433. [17:43:04] <ds2> seems all the intresting stuff is in display.c :(
  434. [17:43:13] <khasim> ds2: yes :)
  435. [17:43:50] <khasim> I hope we will port similar cool stuff to GIT as well, I
  436. [17:44:06] <ds2> I hope so too
  437. [17:44:25] <ds2> as long as it isn't the big mess I saw in one attempt
  438. [17:44:57] <khasim> actually it can be made more simple, the issue started with a bad requirement/design may be
  439. [17:45:24] * Fragguel (n=rafa@alhakem.uco.es) Quit ("Ex-Chat")
  440. [17:45:40] <_apk> I'll have to buy a usb wifi dongle too..as anyone some suggestion for one working "easy" with beagleboard (and maybe debian for arm)? :)
  441. [17:46:50] <vlad_> I would suggest running wired if you can, wifi's pretty finicky
  442. [17:47:52] <ds2> I'd tentatively start with a rt73 based WiFi module as that's what i have had luck with on OMAP boards in general
  443. [17:52:09] * geraint (n=gnorth@tu008.demon.co.uk) has joined #beagle
  444. [17:52:44] <_apk> oh vlad :(
  445. [17:53:33] <_apk> I have the idea (and this is why i'm buying this board) to use it near my tv to watch films (and this is why I have only a s-video input)
  446. [17:54:15] <_apk> thank you ds2! i'm going to search for that :)
  447. [18:03:31] <DJWillis> dirk2: ta, I always get the minor numbers round the wrong way.
  448. [18:14:56] <dirk2> DJWillis: Some days ago we found that wiki was wrong about the minor numbers, too ;) Therefore I still remember and now know what's correct...
  449. [18:18:41] * geraint_ (n=gnorth@tu008.demon.co.uk) has joined #beagle
  450. [18:19:50] * geraint (n=gnorth@tu008.demon.co.uk) Quit (Read error: 101 (Network is unreachable))
  451. [18:21:35] * BThompson (n=BThompso@nat/ti/x-d5373fec04e361a8) Quit ("Trillian (http://www.ceruleanstudios.com")
  452. [18:28:42] * robclark (n=robclark@nat/ti/x-089f3162d338fbe6) Quit ()
  453. [18:31:28] * robclark (n=robclark@nat/ti/x-533847b3e336a06e) has joined #beagle
  454. [18:36:25] * robclark (n=robclark@nat/ti/x-533847b3e336a06e) Quit (Client Quit)
  455. [18:38:44] * Olipro (n=Olipro@unaffiliated/olipro) Quit (No route to host)
  456. [18:42:21] * asciiforever (n=Jon@74-131-26-229.dhcp.insightbb.com) has joined #beagle
  457. [18:46:06] <DJWillis> dirk2: :), oh on another topic, I wanted to say thanks for the u-boot work (already thanked Steve), following that with interest as it should get the Pandora board files down to next to nothing other then the common stuff and a few defines :). Nice and easy to support.
  458. [18:48:50] * JoeBorn__ (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net) has joined #beagle
  459. [18:53:41] <dirk2> DJWillis: If Mani likes the stuff I just sent him in the longish mail, stuff needed for a new board/omap3/xxx will be quite small (assuming you don't have to change too much in common code)
  460. [18:55:34] <dirk2> DJWillis: Do you already have an idea what will be different/needs to be changed regarding the common code for Pandora?
  461. [19:01:42] * Olipro (n=Olipro@unaffiliated/olipro) has joined #beagle
  462. [19:04:40] * jkridner|work (n=a0321898@nat/ti/x-45f179a99756dafe) Quit (Remote closed the connection)
  463. [19:09:38] <Crofton|work> jkridner, looks like the Sugar guy will get some help from a guy on the OE list
  464. [19:12:10] * dcordes (n=dcordes@unaffiliated/dcordes) has joined #beagle
  465. [19:13:22] * valhalla (n=valhalla@81-174-21-16.dynamic.ngi.it) Quit ("Leaving")
  466. [19:19:24] * zaphand1 (n=Z@85.64.205.76.dynamic.barak-online.net) has joined #beagle
  467. [19:20:37] * jkridner_ (n=jason@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
  468. [19:21:47] * jkridner (n=jason@c-76-31-18-64.hsd1.tx.comcast.net) Quit (Read error: 104 (Connection reset by peer))
  469. [19:21:51] * jkridner|work (n=a0321898@nat/ti/x-22579c21f837c574) has joined #beagle
  470. [19:25:00] * khasi1 (n=a0393720@192.163.20.231) has joined #beagle
  471. [19:25:21] * dirk2 (n=dirk@F32e5.f.strato-dslnet.de) has left #beagle
  472. [19:25:37] * jkridner_ is now known as jkridner
  473. [19:30:07] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  474. [19:32:20] <dcordes> did somebody ever try to set the omapfb to 1280x720? I did that yesterday and ended up with a small 640x480 window in the top left and the rest was black. fbset reported 640x480. ds2 suggested to raise the clock because probably it would be to low in order to draw the bigger fb. any thoughts on this?
  475. [19:34:55] <mru> use my kernel patches
  476. [19:35:13] <mru> the oe kernel should have them, I think
  477. [19:35:23] <mru> fbset doesn't work very well with omapfb
  478. [19:36:00] <mru> actually.. it seems to report the correct values now
  479. [19:36:17] <sakoman> dcordes: I've had mine set to 1280x720. If you are getting a 640x480 image that likely means something is messed up in your defconfig -- if the code ca't find a defconfig resolution setting it understands then it defaults to 640x480
  480. [19:36:18] <mru> I was certain I'd seen it report bogus values before
  481. [19:36:29] <sakoman> See: http://www.sakoman.net/cgi-bin/gitweb.cgi?p=linux-omap-2.6.git;a=blob;f=drivers/video/omap/lcd_omap3beagle.c;h=e6d2ba59cc8524648f5080d7c237cfe99b8252fa;hb=1a1be60b0b4060662dc97d2a37c78f9504a59f65
  482. [19:36:42] * dcordes_ (n=dcordes@unaffiliated/dcordes) has joined #beagle
  483. [19:37:09] * dcordes (n=dcordes@unaffiliated/dcordes) Quit (Read error: 104 (Connection reset by peer))
  484. [19:37:16] <sakoman> for a look at the code to see why this would happen. So in short, make sure that your kernel config has CONFIG_FB_OMAP_092M9R defined
  485. [19:37:52] <mru> those timings probably won't work with many TVs
  486. [19:38:13] <sakoman> mru: those settings are intended for lcd computer monitors
  487. [19:38:33] <sakoman> they can certainly be extended to support tv compatible timings too
  488. [19:38:40] <mru> of course they can
  489. [19:38:48] <mru> just commenting
  490. [19:38:49] <sakoman> but my assumption was that most folks will be uning lcd monitors
  491. [19:39:34] <sakoman> mru: would you like me to add a few tv compatible options?
  492. [19:39:56] * khasi1 (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  493. [19:39:59] * jkridner|work (n=a0321898@nat/ti/x-22579c21f837c574) Quit (Remote closed the connection)
  494. [19:40:24] <mru> don't do it for my sake
  495. [19:40:55] <mru> I have some half-finished work to support selection from the kernel command line
  496. [19:41:25] <sakoman> mru: have you tried using fbset to change them?
  497. [19:41:31] <mru> that won't work
  498. [19:41:39] * jkridner_ (n=jason@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
  499. [19:41:40] <mru> the ioctl for that doesn't do anything
  500. [19:41:56] <sakoman> :-)
  501. [19:42:18] <sakoman> someone ought to fix that!
  502. [19:42:29] <mru> that someone might be me...
  503. [19:43:37] * jkridner|work (n=a0321898@nat/ti/x-37a8a48c9bfd327e) has joined #beagle
  504. [19:45:39] <koen> vlad_: I'm testing your mozconfig changes, but I'm pretty busy with non-arm stuff this week, so progress will be slow
  505. [19:46:08] <vlad_> koen: no worries -- I just got things building on beagle, so I'll be doing a bunch of tweaking
  506. [19:46:17] <vlad_> I may bug you for some OE help in a few days though
  507. [19:47:08] <Crofton|work> vlad_, I may be some help also
  508. [19:47:20] <Crofton|work> I have gnuradio trunk building again
  509. [19:47:25] <vlad_> ok, great
  510. [19:47:35] <Crofton|work> Eric knows too much about autofoo I fear
  511. [19:48:20] <koen> vlad_: Crofton|work, sakoman and likewise are all OE gurus too :
  512. [19:48:23] <koen> )
  513. [19:48:47] * jkridner (n=jason@c-76-31-18-64.hsd1.tx.comcast.net) Quit (Read error: 110 (Connection timed out))
  514. [19:49:03] <jkridner_> OE experts: how do I find why a package is being built? For some reason, I'm building cacao, even though I only want to build jamvm.
  515. [19:49:10] * jkridner_ is now known as jkridner
  516. [19:49:27] <jkridner> of course, I am RTFMing now and will probably figure it out.
  517. [19:49:27] <koen> bitbake -g jamvm then look in depends.dot
  518. [19:49:37] <Crofton|work> bitbake -g and look at depends.dot
  519. [19:49:47] <Crofton|work> echo ,,,,
  520. [19:51:53] <sakoman> you should bitbake -g . . .
  521. [19:52:21] <jkridner> :)
  522. [19:52:36] <sakoman> we need likewise to chime in ;-)
  523. [19:54:18] <jkridner> hmmm... didn't expect that... jamvm depends on cacao.
  524. [19:54:29] <jkridner> I thought they were alternatives.
  525. [20:00:29] <Crofton|work> does anything on here help?
  526. [20:00:30] <Crofton|work> PYTHON_CPPFLAGS='-I${STAGING_INCDIR}/python${PYTHON_VERSION}' \
  527. [20:00:38] <Crofton|work> http://wiki.openembedded.net/index.php/Java
  528. [20:00:49] <Crofton|work> grr, needed to clear the pastebuffer
  529. [20:10:11] <jkridner> Crofton|work: ?
  530. [20:11:01] <Crofton|work> jkridner, the OE wiki page
  531. [20:12:30] <koen> jkridner: I'm putting some more developement stuff (git, svn, etc) and python modules in the angstrom feeds for the sugar people
  532. [20:12:38] <Crofton|work> the PYTHON vars was my mistake
  533. [20:13:26] <jkridner> Crofton|work: it seems to point me towards Cacao over Jamvm, but doesn't seem to tell me why Jamvm pulls in Cacao.
  534. [20:13:52] <jkridner> I'd picked Jamvm due to performance, but I think I may just go with Cacao after reading that wiki page.
  535. [20:13:58] <Crofton|work> I was hoping the answer was there :(
  536. [20:14:27] <dcordes_> sakoman: that's what I did, switching the resolution in the kernel configuzration (CONFIG_FB_OMAP_092M9R)
  537. [20:14:36] <dcordes_> mru: so you think the problem is rooted in fbset?
  538. [20:14:42] <mru> no
  539. [20:14:47] <jkridner> I've read elsewhere that there were considerations of halting JamVM development, but that some people were interested in seeing it continue due to performance.
  540. [20:14:53] <dcordes_> mru: I am using the OE kernel.
  541. [20:15:00] <jkridner> koen: great!
  542. [20:15:05] <dcordes_> mru: so I should have those patches you mentioned
  543. [20:15:20] <jkridner> I'll halt my attempts to native compile git and just pull it from OE.
  544. [20:15:29] <mru> dcordes_: I'm not certain what's in that kernel
  545. [20:15:35] <mru> koen is the expert there
  546. [20:15:39] * mru looks at koen
  547. [20:15:49] <jkridner> should I set PREFERRED_PROVIDER in a local conf file or in the angstrom conf file?
  548. [20:15:50] <dcordes_> koen: did you see my problem with CONFIG_FB_OMAP_092M9R?
  549. [20:16:24] <dcordes_> jkridner: I usually set such thing in local.conf
  550. [20:16:31] <dcordes_> jkridner: I think that's the convenient
  551. [20:16:39] <dcordes_> * convenient way
  552. [20:16:47] <koen> dcordes_: the defconfig works on my screen :)
  553. [20:16:56] <koen> and that's the only dvi screen I have
  554. [20:17:07] <dcordes_> koen: I have super 16:10 panel
  555. [20:17:17] <dcordes_> it looks shity with 1024x768
  556. [20:18:07] <koen> jkridner: btw, AMD had a complete sugar OE overlay a while ago
  557. [20:18:25] <koen> jkridner: they loathed fedora and kept using OE built rootfs on their XOs
  558. [20:18:35] <sakoman> dcordes_: check to make sure you *really* have set it to CONFIG_FB_OMAP_092M9R in the OE recipe defconfig
  559. [20:18:44] <koen> jkridner: the first thing that got booted on a XO was oe built :)
  560. [20:18:53] * JoeBorn__ (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net) Quit ("open.neurostechnology.com")
  561. [20:19:04] <koen> zcat /proc/config.gz | grep CONFIG_FB_OMAP
  562. [20:19:18] <dcordes_> sakoman: I let OE setup the source, then I reconfigured and built manually. that's what I awlays do to edit OE kernel.
  563. [20:19:19] <sakoman> dcordes_: I've been known to edit the wrong file on occasion ;-)
  564. [20:19:29] <dcordes_> sakoman: so bitbake doesn't to any witchcraft I don't know about
  565. [20:19:43] <koen> jkridner: I don't know how secret and NDA those recipes are
  566. [20:19:45] <jkridner> koen: is there an existing recipe to use as a starting point?
  567. [20:19:56] <jkridner> hmmm...
  568. [20:20:08] <dcordes_> sakoman: also it is in the config.gz
  569. [20:20:10] <jkridner> strange they wouldn't share those.
  570. [20:20:48] <dcordes_> sakoman: the strange thing is, Xfbdev draws the 640x480 screen in the top left corner, om top of a black 1280x720 area
  571. [20:21:01] <dcordes_> it looks like the whole screen is painted black with the right resolution
  572. [20:21:03] <Crofton> jkridner, ask CosmicPenguin
  573. [20:21:31] <sakoman> dcordes_: ah, then perhaps X is confused. Can your monitor tell you what resolution it thinks it is running at?
  574. [20:21:42] <dcordes_> sakoman: let me check..
  575. [20:22:12] <dcordes_> sakoman: 1280x720 :)
  576. [20:22:25] <sakoman> Ah, so you are looking in the wrong place ;-)
  577. [20:22:36] <sakoman> Better look at x setup!
  578. [20:22:39] <dcordes_> it's userspace problem
  579. [20:22:43] <sakoman> yup
  580. [20:23:08] <koen> Xfbdev?
  581. [20:23:19] <koen> beagleboard is using Xorg for a while now
  582. [20:23:29] <dcordes_> oh
  583. [20:23:57] <dcordes_> koen: Xfbdev is running on my beagleboard though.
  584. [20:24:03] <dcordes_> inside the demo image
  585. [20:24:15] <koen> must be an old demo :)
  586. [20:24:47] <dcordes_> it's the one you announced on angstrom mainpage
  587. [20:25:50] <dcordes_> koen: any chance the newer one will get rid of the 640x480 problem?
  588. [20:27:30] <Crofton|work> hmm, my wired nic came up at boot time
  589. [20:29:49] <dcordes_> I take it as just try
  590. [20:35:02] <koen> jkridner: NOTE: package sugar-toolkit-0.82.1: completed
  591. [20:35:25] <jkridner> sweet. simplest port I've ever done. ;)
  592. [20:35:46] <jkridner> I guess that is just a part of it, or is that the whole thing?
  593. [20:36:00] <dcordes_> koen: http://www.sakoman.net/feeds/omap3/glibc/images/beagleboard/ <- shall I try one of those?
  594. [20:36:55] <dcordes_> what's overo?
  595. [20:37:02] <Crofton|work> a horse
  596. [20:37:08] <Crofton|work> a funny coloured horse
  597. [20:37:20] * max_fr (i=max@212-198-79-163.rev.numericable.fr) has left #beagle
  598. [20:37:28] <koen> jkridner: it's the first file mentioned int he http://sugarlabs.org/go/ReleaseTeam/CurrentRelease/Sucrose source list :)
  599. [20:37:47] <Crofton|work> http://www.linuxdevices.com/news/NS3581886124.html
  600. [20:38:03] <sakoman> dcordes_: you could try one of those if you like to live dangerously :-)
  601. [20:38:23] <sakoman> Overo is the new Gumstix product based on omap3
  602. [20:38:38] <Crofton|work> sakoman, eventually I gave a straight answer
  603. [20:38:47] <dcordes_> Crofton|work: man I'm so sorry
  604. [20:38:54] <Crofton|work> this sugar theme for kids I find disturbing :)
  605. [20:39:42] <sakoman> koen: don't you have a beagleboard-demo-image that dcordes_ can use?
  606. [20:39:52] <dcordes_> Crofton|work: I didn't mean to bother you
  607. [20:40:02] <Crofton|work> the first time I saw the overo name, I googled and all I got was the horse reference
  608. [20:40:19] <koen> sakoman: I can upload one
  609. [20:41:09] <Crofton|work> dcordes_, no problem
  610. [20:41:10] <sakoman> it's a good name -- lots of o's ;-)
  611. [20:41:24] <jkridner> koen: k. so there is still some time for me to get *something* done.
  612. [20:41:47] <koen> http://ewi546.ewi.utwente.nl/~koen/Angstrom-Beagleboard-demo-image-glibc-ipk-2008.1-test-20080814-beagleboard.rootfs.tar.bz2
  613. [20:41:56] <jkridner> I want to get another complete OE build some day.
  614. [20:45:35] * koen still hasn't looked at the ffmpeg v4l build problem :(
  615. [20:45:36] <dcordes_> jkridner: did you ever get your task-native-sdk problem fixed?
  616. [20:46:23] <jkridner> no. I just decided to do a complete rebuild and take control over my situation. I haven't managed to complete that rebuild yet. :(
  617. [20:47:32] <jkridner> I believe there is some data corruption going on with my host, but I'm still looking to confirm it.
  618. [20:47:46] <jkridner> it is alarming, since I'm using Amazon EC2.
  619. [20:47:51] <koen> NOTE: package sugar-0.82.0: completed
  620. [20:48:10] <mru> koen: what ffmpeg v4l problem?
  621. [20:48:11] <jkridner> my Mac Pro is in the shop and it never would build everything completely.
  622. [20:48:20] <mru> errors from video4linux2.h?
  623. [20:48:24] <dcordes_> jkridner: backup your extra package list, clear temp, rebuild?
  624. [20:48:40] <jkridner> I don't have an OE setup other than to run on BeagleBoard.org itself right now.
  625. [20:49:08] <jkridner> that's what I've been doing. last failure was in cacao: http://tinderbox.openembedded.net/public/logs/785898.txt
  626. [20:49:12] <koen> In file included from /OE/angstrom-dev/work/armv7a-angstrom-linux-gnueabi/ffmpeg-0.4.9+r26+gitra1b91f5e7bf5612ea3a5c63579f220b0af8da1f8-r26/git/libavdevice/v4l2.c:38:/OE/angstrom-dev/staging/arm-angstrom-linux-gnueabi/usr/include/linux/videodev2.h:599: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'v4l2_std_id'/OE/angstrom-dev/staging/arm-angstrom-linux-gnueabi/usr/include/linux/videodev2.h:691: error: expected specifier-qualifier-li
  627. [20:49:23] <jkridner> http://tinderbox.openembedded.net/packages/103712/
  628. [20:50:39] <mru> koen: yes, we've seen that on a few systems
  629. [20:50:41] <dcordes_> maybe it's fixed by now, did you try a pull already?
  630. [20:51:28] <mru> koen: the trouble is that asm/types.h fails to define __u64 in C99/POSIX mode
  631. [20:51:33] <mru> incorrectly, of course
  632. [20:51:50] <mru> some distros still haven't fixed it
  633. [20:53:03] <mru> you could just disable v4l2 in ffmpeg for now
  634. [20:53:21] <koen> mru: funny thing is that it only started failing after I told it to build a new rev from your git tree
  635. [20:53:33] <koen> jkridner: sugar-artwork-0.82.0/cursor/cursorthemegen/.libs/lt-cursorthemegen: cannot execute binary file
  636. [20:53:40] <mru> we added -std=c99 to CFLAGS recently
  637. [20:53:40] * Olipro_ (n=Olipro@unaffiliated/olipro) has joined #beagle
  638. [20:54:05] <koen> jkridner: x86 crapware ;)
  639. [20:54:14] <koen> mru: that must be it
  640. [20:54:14] <Crofton> koen, I have similar issues with some gnuradio stuff :)
  641. [20:54:23] <Crofton> I run the lt thing on the build system
  642. [20:54:24] <mru> jkridner: libtool crap too
  643. [20:54:30] <koen> mru: since v4l2.c didn't get any updates recently
  644. [20:56:07] <sakoman> koen: I haven't been able to build the beagle demo image for quite a few days now :-(
  645. [20:58:21] <jkridner> any reason I should not delete org.openembedded.dev and move to fetching it with git, rather than mtn?
  646. [20:58:34] <Crofton|work> I just started a demo image build
  647. [20:58:35] * koen was hoping other people would fix it since many OE devs have beagleboards now
  648. [20:58:40] * Crofton|work curses usb
  649. [20:59:01] <koen> jkridner: nothing about from it lagging 2 days or so every now and then
  650. [20:59:07] <Crofton|work> now ./test_usrp_standard_rx runs and hangs
  651. [20:59:19] <Crofton|work> the entire board :(
  652. [20:59:33] <jkridner> so, mtn is still the check-in method?
  653. [20:59:45] <Crofton|work> yeah
  654. [21:02:09] <koen> our git dude just had his compsny
  655. [21:02:15] <koen> bought by intel
  656. [21:02:30] <koen> so he's probably busy with that
  657. [21:02:51] <dcordes_> koen: thank you for the upload. I just booted it and Xorg draws to fullscreen, also fbset reports 1280x720 now. but my monitor seems to get wrong info, so the image is cropped off like 20 pixels or so on all sides. (like if it was zoomed a bit)
  658. [21:03:03] * mru gave up on mtn
  659. [21:03:15] * sakoman did to
  660. [21:03:20] <sakoman> too
  661. [21:03:21] <mru> nice idea, bad implementation, and too many shortfalls
  662. [21:03:43] <mru> but then it is c++, so what else is there to expect
  663. [21:06:25] * koen gently asks the sugar dudes to fix their shit to not run arm binaries on x86
  664. [21:08:07] <Crofton|work> hmm, I bet I need to turn LL debug back on the catch the usb crash ....
  665. [21:08:17] * Olipro (n=Olipro@unaffiliated/olipro) Quit (Connection timed out)
  666. [21:24:19] * asciiforever (n=Jon@74-131-26-229.dhcp.insightbb.com) Quit ("Leaving")
  667. [21:26:17] <koen> jkridner: http://www.angstrom-distribution.org/repo/?name=sugar&arch=&action=search
  668. [21:27:09] <koen> jkridner: the packages probably needs some tweaking to add the right python modules to RDEPENDS
  669. [21:27:12] <dcordes_> koen: I just played around a bit in the e17 d
  670. [21:27:20] * dcordes_ needs to get new keyboard
  671. [21:28:04] <dcordes_> koen: and I found that if you remove some 'modules' from it, you get a much faster windowmanager experience. I would suggest to remove the dropshadow option from the builds
  672. [21:29:01] <koen> ah, right
  673. [21:29:08] * koen still thinks e17 is crap
  674. [21:29:19] <koen> but people seem to like the bling
  675. [21:29:53] <dcordes_> the e17 in demo is the first one I tried
  676. [21:30:14] <dcordes_> I've seen one on a playstation3 (yellow dog linux)
  677. [21:33:11] <vlad_> e17 is kinda crap, but it's also great bling :)
  678. [21:34:11] <koen> highlight of its crappiness are that it always has icons for firefox, xmms and xterm, even if those aren't installed
  679. [21:34:26] <koen> .desktop files exist for a reson
  680. [21:34:30] <koen> +a somewhere
  681. [21:35:46] <vlad_> it gets even better, it uses the firefox icon and "Firefox" even if you don't actually get firefox when you launch it, since it just launches web-browser
  682. [21:36:32] <dcordes_> hm still better than using matchbox or so desktop on a device with mouse and keyboard
  683. [21:36:40] <koen> well, I symlinked /usr/bin/epiphany to /usr/bin/firefox fot 'fix' e17 :)
  684. [21:36:47] <dcordes_> or is there something else in OE that one could use?
  685. [21:37:06] <koen> twm :)
  686. [21:37:16] <ds2> nothing wrong with matchbox :P
  687. [21:37:32] <ds2> koen: let's get with the times... tvtwm at least!
  688. [21:37:56] <dcordes_> koen: is there an image.bb that uses it?
  689. [21:38:49] <koen> dcordes_: dunno
  690. [21:38:55] <koen> dcordes_: angstrom has packagemanagement
  691. [21:39:10] <vlad_> icewm isn't atrocious
  692. [21:39:15] <vlad_> lightweight at least
  693. [21:39:59] * koen uses the default wm in osx and commandline on linux
  694. [21:41:38] <vlad_> '-march=armv7-a -Wa,-march=armv7-a' is what I want to tell gcc to optimize for armv7?
  695. [21:41:43] <vlad_> (-a)
  696. [21:41:50] <mru> -mcpu=cortex-a8
  697. [21:41:53] <mru> -fpu=neon
  698. [21:41:56] <mru> -mfpu=neon
  699. [21:42:00] <vlad_> mmm
  700. [21:42:00] <vlad_> ok
  701. [21:42:31] <koen> -march=armv7-a -mtune=cortex-a8 -mfpu=neon -mfloat-abi=softfp
  702. [21:42:34] <koen> that's what I use
  703. [21:42:48] <mru> -mcpu=cortex-a8 is equivalent to the first two of those
  704. [21:43:02] <ds2> doesn't softfp negate the benefits of having neon?
  705. [21:43:14] <koen> no
  706. [21:43:19] <mru> partially
  707. [21:43:25] <ds2> what does softfp do then?
  708. [21:43:40] <mru> it passes floating point arguments to functions in integer registers
  709. [21:43:50] <koen> hardfloat code with softfloat calling conventions
  710. [21:44:07] <mru> it's a bit suboptimal
  711. [21:44:12] <mru> it wastes integer registers
  712. [21:44:25] <ds2> so instead of generating direct neon instructions, it makes a function call to something that uses neon?
  713. [21:44:26] <mru> and moving data from neon to arm registers is *really* slow
  714. [21:44:31] <mru> ds2: no
  715. [21:44:44] <ds2> hmmm
  716. [21:44:47] <mru> but you might call a function with float arguments
  717. [21:44:58] <koen> ds2: softfp is basically a code-word for eabi
  718. [21:45:05] <mru> not true
  719. [21:45:17] <mru> eabi specifies a hardfloat calling convention
  720. [21:45:29] <ds2> what mru says isn't as expensive as what koen is saying
  721. [21:45:40] <ds2> koen: I thought modern tool chains default to eabi?
  722. [21:46:36] <mru> eabi specifies calling conventions for both softfloat and hardfloat
  723. [21:46:42] <koen> modern toolchain default to what you put in the spec file
  724. [21:46:51] <mru> gcc is stupid and can only use softfloat calls
  725. [21:47:04] <ds2> hmm okay
  726. [21:47:18] <ds2> and just to be clear, neon appears as yet another coprocessor right?
  727. [21:47:32] <mru> so when you call a function taking a float argument, gcc transfers the data from float registers to integer registers
  728. [21:47:37] <mru> yes, neon is a coprocessor
  729. [21:47:57] <mru> the eabi has a section about coprocessor arguments
  730. [21:48:08] <ds2> I see
  731. [21:53:08] * Beagle7 (n=Beagle7@79-69-120-243.dynamic.dsl.as9105.com) Quit (Read error: 110 (Connection timed out))
  732. [21:53:53] * Olipro__ (n=Olipro@unaffiliated/olipro) has joined #beagle
  733. [21:56:08] * geraint_ (n=gnorth@tu008.demon.co.uk) Quit ()
  734. [21:56:28] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
  735. [21:58:03] <mru> what makes softfp bad is that a transfer from neon to arm takes 20 cycles
  736. [21:58:36] <davr> ouch
  737. [21:58:45] <ds2> hmmm isn't an in cache memory xfer quicker?
  738. [21:58:54] <jkridner> this is odd to me: http://tinderbox.openembedded.net/packages/?name=asio
  739. [21:58:57] <mru> not in this case
  740. [21:59:04] <jkridner> sometimes it builds, sometimes it doesn't.
  741. [21:59:08] <mru> the arm and neon pipelines aren't interlocked
  742. [22:00:09] <mru> the neon coprocessor receives instructions from the arm pipeline through a 16-entry fifo
  743. [22:00:11] <jkridner> haven't changed anything. it seems to build when I do 'bitbake asio', but it took a really long time.
  744. [22:00:21] * cbrake is now known as cbrake_away
  745. [22:01:13] <mru> for the instruction transfer to happen, everything in the neon pipeline and instruction fifo has to complete
  746. [22:01:42] <jkridner> no, it is just failing now. somehow it built once.
  747. [22:02:25] <mru> going through memory, the same delay would apply
  748. [22:02:56] <mru> I'm not sure how coherency is maintained between arm and neon
  749. [22:03:13] <ds2> thought they shared the MMU, sounds like they don't?
  750. [22:03:23] <mru> and neither are ARM... there are bugs in that area ;-)
  751. [22:03:37] <mru> they share everything memory related
  752. [22:04:30] <Crofton|work> jkridner, what are you building?
  753. [22:05:42] <ds2> as for why building jamvm pulls in cacao is probally cuz of class path, it wants a java compiler and jikes nonlonger suffice
  754. [22:06:08] <jkridner> I'm trying to build a bigger demo image with task-native-sdk, java, and several other items that I have previously added via 'opkg install'.
  755. [22:06:37] <jkridner> bash, screen, vim.
  756. [22:06:57] * JJS[out] is now known as JJS
  757. [22:07:03] <jkridner> mono (yikes)
  758. [22:07:14] <Crofton|work> what is asio?
  759. [22:07:30] <jkridner> I don't know, but it uses boost. :)
  760. [22:07:34] <ds2> ewwwwwwww vb clone
  761. [22:07:38] * mru had no idea jkridner was such a masochist
  762. [22:07:39] <Crofton|work> heh
  763. [22:07:54] <jkridner> indeed.
  764. [22:08:25] <jkridner> part of a paper I'm trying to write and I want to survey across multiple managed languages.
  765. [22:10:36] * Olipro_ (n=Olipro@unaffiliated/olipro) Quit (Success)
  766. [22:11:36] <jkridner> tar rsync less gnash scummvm firefox and others as well.
  767. [22:11:47] <jkridner> someone is pulling in asio
  768. [22:12:10] <Crofton|work> http://www.boost.org/doc/libs/1_36_0/doc/html/boost_asio.html
  769. [22:12:14] <ds2> gnash just builds?
  770. [22:12:20] <Crofton|work> asio is part of boost it seems
  771. [22:12:36] <ds2> nevermind, you are using OE so no good answer there
  772. [22:15:19] <Crofton|work> I also see the u64 problem in ffmpeg
  773. [22:16:13] <mru> it's easy to fix the system headers
  774. [22:16:14] <jkridner> abiword-plugins depends on asio
  775. [22:25:42] * denix (n=denys@nat/ti/x-64625e2a2ac9cad0) Quit ("Leaving")
  776. [22:30:07] * methril|laptop (n=methril@87.220.72.255) Quit (Read error: 60 (Operation timed out))
  777. [22:30:31] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) has joined #beagle
  778. [22:33:37] * methril|laptop (n=methril@61.73.220.87.dynamic.jazztel.es) has joined #beagle
  779. [22:38:22] <_apk> damn! digikey has finished the beagle board :(((
  780. [22:40:04] <jkridner> finished?
  781. [22:41:04] <ds2> eh?
  782. [22:43:05] <jkridner> _apk: finished? you mean out-of-stock? they should get some more this week.
  783. [22:43:49] <ds2> anyone know if it is acceptable to populate the pads on the expansion connector from the bottom instead of the top?
  784. [22:44:16] <mru> doesn't sound pretty...
  785. [22:44:27] <mru> electrically I can't see how it would make a difference
  786. [22:44:40] <jkridner> well, to give a lower clearance, it seems useful.
  787. [22:44:47] <jkridner> that is part of why we didn't populate the header.
  788. [22:45:13] <jkridner> it is easier to add a header to the top, than to remove a header and add a new one to the bottom.
  789. [22:45:37] <mru> koen: remember the neuros board?
  790. [22:45:38] <ds2> I am more thinking of mechanical issues like maybe the copper rings are smaller on the bottom then on top or vice versa
  791. [22:45:46] <_apk> oh good jkridner :) I've just sent them an email :)))
  792. [22:46:14] <jkridner> they are plated through with same size rings on top and bottom.
  793. [22:46:34] <ds2> great, saves me from pulling out the microscope and graticle ;)
  794. [22:46:58] <ds2> one of the few instances where having through holes is better then ust a series of pads
  795. [22:48:22] * likewise (n=chatzill@82-171-51-231.ip.telfort.nl) Quit ("ChatZilla 0.9.83 [Firefox 3.0/2008061015]")
  796. [22:49:08] <mru> holes are much easier to hand-solder...
  797. [22:50:18] <ds2> SMT is easier to add and remove
  798. [22:53:00] <jkridner> how does OE decide to not rebuild something but to use what has already been built?
  799. [22:55:38] <jkridner> given that I already had asio built, I don't understand why it rebuilt and failed without changing the metadata
  800. [22:56:23] <jkridner> i do have parallel make, but I pre-built the dependency (boost) on which it (asio) was failing.
  801. [22:59:50] * Beagle8 (n=Beagle8@63.127.251.27) has joined #beagle
  802. [23:00:16] * ds2 resists with all his might from commenting more about OE ;)
  803. [23:01:07] <Beagle8> Got a beginner question - I'm trying to plug in a powered USB hub into my new beagle, though I seem to only be able to either power the beagle or power the devices... any suggestions? Is there some special cable I need? thanks in advance.
  804. [23:01:40] <ds2> Beagle8: are you using the power connector (2.1mm barrel/coax) ?
  805. [23:02:05] <Beagle8> no... just the usb connector....
  806. [23:02:09] <jkridner> you need a mini-A to std-A cable like: http://www.vernier.com/accessories/access.html?usb-mini&template=basic.html
  807. [23:02:21] <jkridner> then, you will need to power the board using the barrel connector.
  808. [23:02:43] <Beagle8> ah... OK so that mini-A to std-A cable wont power the board then... right?
  809. [23:02:49] <Beagle8> mine is I believe
  810. [23:03:13] * harper (n=harper@24-148-50-204.stk-bsr1.chi-stk.il.cable.rcn.com) has joined #beagle
  811. [23:03:29] * harper is now known as harperrr
  812. [23:06:28] * Beagle1 (n=Beagle1@ppp-71-138-48-90.dsl.sndg02.pacbell.net) has joined #beagle
  813. [23:12:08] * Beagle8 (n=Beagle8@63.127.251.27) Quit ()
  814. [23:21:09] * NishanthMenon (n=gnat@nat/ti/x-d15ace35d82ce52b) Quit ("If you think nobody cares, try missing a few payments")
  815. [23:22:00] * rsalveti (n=salveti@200.184.118.132) Quit (Read error: 113 (No route to host))
  816. [23:22:47] * odesus (n=Widmung@201.165.141.79) has joined #beagle
  817. [23:23:02] <odesus> Hi to everyone..
  818. [23:23:10] <mru> hi
  819. [23:23:20] <odesus> Does anybody have testing ubuntu arm?
  820. [23:23:25] <odesus> Is it stable enough?
  821. [23:23:44] <mru> sorry, I don't know
  822. [23:24:01] <odesus> oh...
  823. [23:24:01] <mru> I think someone was talking about running it
  824. [23:24:11] <odesus> oh...
  825. [23:24:30] <odesus> Why is kind of difficult to install OE ?
  826. [23:24:38] <odesus> bitbake, monotone.. etc, etc..
  827. [23:24:57] <mru> what do you run on your pc?
  828. [23:24:58] <dcordes_> you can use git instead of mtn
  829. [23:25:25] <odesus> thanks dcordes..
  830. [23:25:47] <odesus> I'm trying to install OE in my pc (mru)
  831. [23:28:04] * guillaum1 (n=Guillaum@AMontsouris-153-1-67-145.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
  832. [23:28:22] * guillaum1 (n=gl@AMontsouris-153-1-67-145.w90-2.abo.wanadoo.fr) has joined #beagle
  833. [23:30:05] * Beagle5 (n=Beagle5@ppp-71-138-48-90.dsl.sndg02.pacbell.net) has joined #beagle
  834. [23:32:29] * odesus (n=Widmung@201.165.141.79) has left #beagle
  835. [23:39:30] <Beagle5> I've just got my board and have a couple of questions... One is, will going through a HDMI-->DVI-->VGA chain work properly? I'm not getting any output on my monitor.
  836. [23:39:40] <Beagle5> (Mind you, I may be at fault.)
  837. [23:41:03] <ds2> it could work
  838. [23:41:26] <ds2> as long as they are what they should be and you are generating the right timing
  839. [23:42:04] <mru> unless that's an expenstive dvi to vga converter with clocks and dacs, it won't work
  840. [23:42:20] <mru> there are no analogue signals from the beagle
  841. [23:42:24] <mru> or indeed in hdmi
  842. [23:42:52] <ds2> expensive is relative... you can get one for around $200
  843. [23:43:03] <mru> that's more than a beagle
  844. [23:43:10] <Beagle5> oi!
  845. [23:43:17] <ds2> but cheaper then the other boards
  846. [23:43:19] <mru> and more than the $2 you pay for the plug adapter
  847. [23:43:30] <mru> dvi-a to vga
  848. [23:45:40] <Beagle5> I'm totally not up on all of these video connectors. Tell me if I've got anything wrong: So DVI (typically) has analog and digital pins on it, hdmi has no analog, thus I've gotta get something active that'll go from DVI's digital pins (or hdmi's pins) to vga?
  849. [23:46:18] <jkridner> Beagle5: yes.
  850. [23:46:22] <davr> if you're gonna pay $200, might as well buy a nice LCD :P
  851. [23:46:27] <jkridner> a typical DVI-D to HDMI cable will do.
  852. [23:46:34] <Beagle5> Seriously.
  853. [23:47:22] <mru> DVI-D is digital, DVI-A is analogue, DVI-I has both
  854. [23:47:39] <Beagle5> Hopefully I can get the boss man to go for the 25" LCD, then when the project is over...new monitor in my office ;[
  855. [23:47:39] <mru> there's also dual-link DVI-D
  856. [23:47:41] <ds2> and the connectors are suppose to be keyed appropriately
  857. [23:47:48] <jkridner> http://www.vernier.com/accessories/access.html?usb-mini&template=basic.html should work.
  858. [23:47:51] <Beagle5> mru: thanks
  859. [23:48:14] <mru> hdmi is single-link dvi
  860. [23:48:47] <Beagle5> mru: what do you mean by single link? just the digital lines
  861. [23:48:48] <Beagle5> ?
  862. [23:49:03] <Beagle5> ah...
  863. [23:49:03] <jkridner> one digital channel.
  864. [23:49:16] <jkridner> DVI can have 2 digital channels.
  865. [23:49:20] <Beagle5> nevermind, wikipedia's got good page on it.
  866. [23:49:36] <ds2> the beagle's display doesn't generate a fast enough signal to need dual channels
  867. [23:49:51] <Beagle5> k
  868. [23:50:10] <mru> the hdmi spec doesn't include any modes that need it
  869. [23:50:48] <ds2> except, strictly speaking, the beagle isn't hdmi, but DVI-D just happen to be using the hdmi connector
  870. [23:50:57] <Beagle5> sure
  871. [23:51:09] <Beagle5> that's what I had been gathering.
  872. [23:51:16] <mru> you could also say it's hdmi without the sound
  873. [23:51:48] <Beagle5> I figured that much as well too, cause why waste board space for audio connectors if you've got it out the hdmi port?
  874. [23:52:08] <Beagle5> but anyhow, thanks very much for clearing all of that up...I'm out
  875. [23:52:22] <ds2> have you priced a HDMI to DVI-D + RCA jacks cable? :)
  876. [23:52:47] <Beagle5> My boss blew 100$ on an hdmi cable the other day...
  877. [23:52:48] <mru> there is no such thing
  878. [23:52:51] <Beagle5> it's a goddamned racket
  879. [23:53:21] <ds2> you are paying too much
  880. [23:53:33] <ds2> HDMI-DVI-D cables can be had for $5US
  881. [23:53:43] <Beagle5> Nono....
  882. [23:53:48] <Beagle5> I got the adapter for about that.
  883. [23:54:20] * jkridner is changing the link on http://beagleboard.org/hardware from a $33 cable to a $15 cable, but still using Digi-Key.
  884. [23:54:28] <Beagle5> My boss blew 100$ on some monster cable for his home theatre system...He knows just as well as I do that a cable's a cable (at least in the commercial world) but couldn't find anything cheaper....
  885. [23:54:52] <jkridner> Let me know if anyone knows a really good supply (book mark them using delicious.com with beagleboard and peripheral and verified tags.
  886. [23:55:08] <ds2> jkridner: Amazon has cheaper cables
  887. [23:55:10] <Beagle5> like a wall wart?
  888. [23:55:56] <Beagle5> never mind.....misread.
  889. [23:56:06] <Beagle5> I need to sleep. Bye.
  890. [23:56:13] * Beagle5 (n=Beagle5@ppp-71-138-48-90.dsl.sndg02.pacbell.net) Quit ()
  891. [23:56:29] <harperrr> what is openembedded?
  892. [23:56:53] <ds2> $8.99
  893. [23:57:08] <ds2> there are even cheaper cables if you want to go with direct import sellers
  894. [23:57:45] * odesus (n=Widmung@201.165.141.79) has joined #beagle
  895. [23:58:15] * DDevine (n=devine@123-243-187-85.static.tpgi.com.au) Quit (Read error: 104 (Connection reset by peer))
  896. [23:58:44] * DDevine (n=devine@123-243-187-85.static.tpgi.com.au) has joined #beagle
  897. [23:59:14] <odesus> hi.. anybody with experience with open embedded who can help me?
  898. [23:59:41] <jkridner> I think I need to program the bot to provide wiki links to "what is ..." questions. :) harperr: it is a build system. http://en.wikipedia.org/wiki/OpenEmbedded

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.