#beagle IRC Log on BeagleBoard.org

Join the chat at beagleboard.org/chat

IRC Log for 2008-06-08

Timestamps are in UTC.

  1. [00:26:21] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  2. [01:22:09] * NeuroTech (n=NeuroTec@adsl-75-51-176-138.dsl.lsan03.sbcglobal.net) has joined #Beagle
  3. [02:10:12] * GeneralAntilles (n=GeneralA@pdpc/supporter/active/generalantilles) has joined #beagle
  4. [03:03:46] * GeneralAntilles (n=GeneralA@pdpc/supporter/active/generalantilles) Quit (Read error: 104 (Connection reset by peer))
  5. [05:44:21] * NeuroTech (n=NeuroTec@adsl-75-51-176-138.dsl.lsan03.sbcglobal.net) Quit (Read error: 110 (Connection timed out))
  6. [08:41:23] * ldesnogu (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) has joined #beagle
  7. [08:51:45] <ldesnogu> hi
  8. [08:52:09] <ldesnogu> did anyone try the c6x tools? it's taking a bit of time to download :)
  9. [09:53:48] <koen> not yet
  10. [12:02:55] <ldesnogu> the disas works
  11. [12:03:30] <ldesnogu> but no simulator, which makes me want BeagleBoard even more :)
  12. [12:17:37] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  13. [12:35:02] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  14. [12:37:07] <ali_as> Is the fbset pixclock in picoseconds?!
  15. [12:43:08] <sakoman> ali_as: no, not picoseconds
  16. [12:43:28] <sakoman> khz IIRC
  17. [12:45:22] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  18. [12:45:24] <ali_as> I'm getting a lot of references to,
  19. [12:45:25] <ali_as> -pixclock <value>
  20. [12:45:25] <ali_as> set the length of one pixel (in picoseconds). Note that the frame buffer device may only support some pixel lengths
  21. [12:46:01] <sakoman> used to be picoseconds a few revisions back
  22. [12:46:17] <ali_as> Aha, so this is just out of date documentation. Ok.
  23. [12:46:33] <sakoman> also there is a value used internally that *is* picoseconds
  24. [12:46:47] <sakoman> forget which struct it is
  25. [12:46:56] <ali_as> Well, that's ok, I hate for this to be confusing ;)
  26. [12:48:27] <sakoman> koen: I can play an mp3 now and actually hear the sound!
  27. [12:49:07] <sakoman> well, the right channel anyway. the left channel sounds severely clipped :-(
  28. [12:49:31] <sakoman> need to go through all the setup registers one more time :-(
  29. [12:50:12] <sakoman> perhaps a gain screwup somewhere, or maybe I2S format is slightly off
  30. [13:04:50] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  31. [13:09:03] * RogerMonk (n=a0740758@nat/ti/x-e8fc335a57e96769) has joined #beagle
  32. [13:21:34] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  33. [13:25:54] * DJWillis (n=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) has joined #beagle
  34. [13:40:58] <koen> sakoman: it is progress :)
  35. [13:45:09] <koen> blech
  36. [13:45:16] <koen> my hand smell like fixer again
  37. [13:45:27] <koen> so much for the 'odorless' claim on the flask
  38. [13:55:14] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  39. [14:02:50] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  40. [14:12:17] * NeuroTech (n=NeuroTec@adsl-75-51-176-138.dsl.lsan03.sbcglobal.net) has joined #Beagle
  41. [14:15:43] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  42. [14:37:56] * JoeBorn__ (n=jborn@cpe-66-1-5-91.il.sprintbbd.net) has joined #beagle
  43. [14:39:36] * JoeBorn__ is now known as JoeBorn
  44. [14:44:38] * NeuroTech (n=NeuroTec@adsl-75-51-176-138.dsl.lsan03.sbcglobal.net) Quit (Read error: 110 (Connection timed out))
  45. [15:00:46] <jkridner> koen: can I ask you a monotone question?
  46. [15:00:50] <koen> sure
  47. [15:00:51] <jkridner> I did 'mtn --db=~/stuff/OE.mtn up'
  48. [15:01:01] <jkridner> and got "mtn: misuse: workspace required but not found"
  49. [15:01:12] <koen> cd org.openembedded.dev ; mtn up
  50. [15:01:35] <koen> 'up' is a workspace operation and needs a checkout
  51. [15:02:01] <koen> monotone seperated database and workspace operations cleanly
  52. [15:02:01] <jkridner> mtn: misuse: database /root/stuff/OE.mtn does not exist
  53. [15:02:03] <jkridner> I must have something wrong in my config file...
  54. [15:02:07] <koen> so 'mtn pull' works outside your checkout
  55. [15:02:12] <koen> and don't use OE as root
  56. [15:02:17] <jkridner> I moved my database to /home/angstrom/stuff/OE.mtn
  57. [15:02:48] <koen> vi _MTN/options and change the path
  58. [15:02:55] <koen> (inside org.oe.dev)
  59. [15:03:12] * NishanthM (n=Nishanth@cpe-24-175-68-203.tx.res.rr.com) Quit (Read error: 104 (Connection reset by peer))
  60. [15:03:48] <koen> bah, I've become so used to uboot that I'm starting to dislike OpenFirmware
  61. [15:04:11] <jkridner> k. I guess you could consider that separation an advantage over git.
  62. [15:07:22] <JoeBorn> hi all.
  63. [15:07:29] <jkridner> hi JoeBorn!
  64. [15:09:04] <JoeBorn> hows it going?
  65. [15:10:40] <koen> the weather is a bit too nice outside :)
  66. [15:10:57] <koen> 28 degrees C is a bit too hot for me
  67. [15:11:03] <koen> (unless it's a holiday)
  68. [15:14:21] <jkridner> I'm still trying to cut my teeth on OE. Seems like it is showing up everywhere and it is my best hope for getting a nice demo system.
  69. [15:14:37] <koen> it has a very steep learning curve
  70. [15:15:07] <koen> I realised yesterday that gremlin and I will need to prepare some tutorials for the neuros people
  71. [15:15:46] * koen curses the ppc->powerpc mess in the kernel
  72. [15:15:49] * dirk2 (n=dirk@F309e.f.strato-dslnet.de) has joined #beagle
  73. [15:15:57] <koen> dirk2: moin!
  74. [15:16:16] <JoeBorn> koen: bless you for that!!!
  75. [15:16:26] <dirk2> koen: moin, moin
  76. [15:18:18] <jkridner> Is OE already out for the OSD?
  77. [15:18:38] <koen> http://amethyst.openembedded.net/~koen/neuros/
  78. [15:18:43] <koen> that has a barebones images
  79. [15:18:55] <jkridner> ah. osd2.
  80. [15:19:10] <jkridner> I'm anxiously awaiting one of those. :)
  81. [15:19:16] <koen> we are currently working on getting the neuros specific stuff working (msp430, rtc, dsplink)
  82. [15:20:08] <koen> jkridner: judging from the kernel defconfig it should work on both osd1 and osd2
  83. [15:20:39] <dirk2> jkridner: Thanks for the DSP compiler hint! This would be a nice new beagle blog entry, no?
  84. [15:21:56] * NishanthM (n=Nishanth@cpe-24-175-68-203.tx.res.rr.com) has joined #beagle
  85. [15:23:54] <jkridner> k. I'd be happy if it was showing up on other blogs and I could just get it put into the Beagle RSS feed.
  86. [15:24:22] <jkridner> Inspiration for a blog post hasn't really struck me yet.
  87. [15:24:41] <jkridner> I'm simply relieved that one of the promises I've been making for months now has finally come true. :)
  88. [15:25:13] <jkridner> I always knew it would, but I feel a bit beaten down in the waiting process.
  89. [15:26:04] <jkridner> of course, it is a bit strange that I could find inspiration in blinking lights and less in a compiler...
  90. [15:26:18] <jkridner> but blinking lights under your control are so cool! :)
  91. [15:26:56] <JoeBorn> well, give me a bit of information about what it is and I'm happy to post on OdNT about it.
  92. [15:29:01] <dirk2> JoeBorn: We are talking about
  93. [15:29:03] <dirk2> http://groups.google.com/group/beagleboard/browse_thread/thread/ae0ed508f4f46cd1
  94. [15:29:03] <ali_as> Like the plaque.....not with the snappen die springwerk unt watch das blinkenlights.
  95. [15:30:56] <JoeBorn> dirk2: yeah, understood, but I'd like to post some bounties for the debugging tools needed, etc.
  96. [15:31:26] <JoeBorn> jkridner and I had a long discussion of this some time ago
  97. [15:31:32] <ali_as> Has anyone tried to overclock a beagle, or is the speed fixed on the omap silicon?
  98. [15:32:43] <JoeBorn> and we ran into this when we posted http://open.neurostechnology.com/node/1020
  99. [15:33:35] <ldesnogu> ali_as, Pandora guys have overclocked OMAP3530 to 900 MHz
  100. [15:33:47] <ali_as> Woah.
  101. [15:34:05] <ldesnogu> it was on an EVM
  102. [15:34:19] <JoeBorn> in the artic tundra :)
  103. [15:34:24] <ldesnogu> :)
  104. [15:34:56] <ali_as> EVM?
  105. [15:35:00] <ldesnogu> I guess doing that in a real boxed Pandora could indeed bring some trouble
  106. [15:35:42] <ldesnogu> ali_as, http://focus.ti.com/docs/toolsw/folders/print/tmdxevm3503.html
  107. [15:35:50] <ldesnogu> except they have a 3530
  108. [15:35:53] <ali_as> Ahh, the dev board?
  109. [15:35:59] * NishanthM (n=Nishanth@cpe-24-175-68-203.tx.res.rr.com) Quit ("Aloha")
  110. [15:42:08] <DJWillis> JoeBorn: nope, in a warm room in the English Home Counties ;)
  111. [15:43:21] <koen> linux-omap git lacks all the good DVFS stuff to make overclocking possible
  112. [15:43:47] <koen> and I'm unsure how the DSP reacts to the cortex clock
  113. [15:46:00] <koen> I remember nokia people being grumpy about it
  114. [15:48:06] <JoeBorn> DJWillis: yeah, I was just teasing, I've lived with an overclocked DSP on the DM320 for over a year now with no trouble.
  115. [15:49:02] <dirk2> koen: git omap3_beagle_defconfig has CONFIG_DEBUG_LL enabled. How stable is git kernel for you nowadays? Is it already time to disable it?
  116. [15:49:37] <koen> it's pretty stable
  117. [15:49:57] <koen> I keep it turned on "just in case"
  118. [15:50:07] <koen> you never know what another RC breaks :)
  119. [15:50:20] <koen> but feel free to disable it, I'm not using that defconfig
  120. [15:50:39] <dirk2> stability: :) ah, yes, I thought that I read something like this in the logs
  121. [15:51:01] <dirk2> any idea what made the change from quite flaky to pretty stable?
  122. [15:51:18] <koen> My bet is on the tony's multi-omap stuff
  123. [15:51:30] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  124. [15:52:18] <dirk2> omap3_beagle_defconfig: I think this what a newcomer will try first
  125. [15:52:44] <dirk2> s/this/this is/
  126. [15:55:42] <koen> will a newcomer compile his/her own kernel or try the binaries?
  127. [15:56:34] <dirk2> Depends ;) If I would be a beagle newcomer, I would try to compile my own kernel ;)
  128. [15:57:15] <koen> and I always make my own defconfig :)
  129. [15:57:45] <dirk2> using omap3_beagle_defconfig as starting point?
  130. [15:58:06] <koen> can't remember what I used
  131. [15:58:19] <ldesnogu> koen, dirk2: you are certainly not representative :)
  132. [15:58:20] <jkridner> after the free compiler, the discussion on the OSD quickly moved to the debugger.
  133. [15:58:46] <jkridner> seems to me there'd be more interest in using the clocks you got, rather than making more clocks. :)
  134. [15:59:30] <ldesnogu> koen: as a newcomer I would first try a pre-built kernel, then start from a beagle defconfig
  135. [16:00:28] <koen> I've seen too many defconfigs that disabled AF_PACKET
  136. [16:00:38] <koen> so I have a huge distrust against them :)
  137. [16:02:12] <jkridner> still, we should try to get the defconfig right for beagle.
  138. [16:02:28] <dirk2> kernel stability: Should we close beagle issue 8 & 14 http://code.google.com/p/beagleboard/issues/list then?
  139. [16:02:39] <koen> I care more for getting the beagle kernel right :)
  140. [16:04:34] <jkridner> I think 8 is closed for sure.
  141. [16:05:34] <koen> what is currenly missing from git compared to wtbu?
  142. [16:05:54] <jkridner> I think much of the power management is still missing.
  143. [16:05:56] <koen> s-video and dvfs for sure
  144. [16:06:10] <koen> sakoman has the audio stuff pretty much finished
  145. [16:06:43] <koen> and git has rtc and led support, which wtbu lacks :)
  146. [16:07:24] <jkridner> koen: I never saw the SD problems. Are you confident we can mark 14 as fixed?
  147. [16:08:30] <jkridner> dirk2: thanks for watching these issues.
  148. [16:08:50] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  149. [16:08:52] <koen> I haven't seen SD problems in a long time
  150. [16:09:03] <koen> where "long" is a week or 2
  151. [16:09:46] <jkridner> I don't know how heavy your usage is.
  152. [16:10:09] <koen> >3 reboots/day
  153. [16:10:12] <jkridner> It seems Philip originally reported this on the .22 kernel.
  154. [16:13:19] <jkridner> well, it looks to me like the problem was pervasive and there were TWL4030 fixes, so I'll consider this as having found the root cause and mark as fixed.
  155. [16:13:37] <jkridner> Is this different than defect #1?
  156. [16:16:31] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) has joined #beagle
  157. [16:21:13] <sakoman> koen: alsa output is close. haven't tried input yet, but I expect after a bit of register tweaking it will work
  158. [16:21:46] <koen> sakoman: great!
  159. [16:22:07] <jkridner> I used to be pretty good at manipulating McBSP configurations, but it has been years.
  160. [16:22:09] <koen> I've removed my usb audio thingy in anticipation of your patch :)
  161. [16:22:13] <sakoman> too damn many registers in the 4030!
  162. [16:22:27] <jkridner> if you share a bit, some of it might come back to me.
  163. [16:22:38] <jkridner> problems more in the 4030 config than the serial port?
  164. [16:23:22] <sakoman> yes, serial port is fine. right channel is perfect, volume control works, sample rate negotiation works
  165. [16:24:00] <sakoman> left channel sounds bad though
  166. [16:24:01] <jkridner> well, sometimes serial misconfig can show up in funny ways, such as on only one channel.
  167. [16:24:14] <sakoman> yes, that was my first suspicion
  168. [16:24:22] <jkridner> if you are shifted by one bit, funny things happen.
  169. [16:24:35] <sakoman> but the signals seem OK
  170. [16:24:52] <jkridner> because the MSB may not often change, but LSB changes frequently.
  171. [16:25:15] <sakoman> I may try the left & right shifted configs just to be sure
  172. [16:25:18] <koen> that would explain the clipping
  173. [16:25:44] <jkridner> the working channel could be wrong amplitude by the shift, whereas the other channel could pick up the frequently changing LSB in an MSB position and sound really bad.
  174. [16:26:11] <sakoman> yeah, I've seen that many times
  175. [16:28:00] <sakoman> heading out to play for a few hours, will try modifying serial configs a bit later
  176. [16:28:16] <koen> is the current patch online somewhere?
  177. [16:28:32] * koen would love to see http://www.kokkinizita.net/linuxaudio/jaaa-pict.html implememted in the DSP
  178. [16:29:15] <sakoman> no, not yet. just in a local git branch
  179. [16:29:54] <sakoman> back in a few hours
  180. [16:31:19] <dirk2> Btw, you all know
  181. [16:31:21] <dirk2> http://nishanthmenon.blogspot.com/2008/06/horrah-first-omap-uboot-v2-support.html
  182. [16:38:58] <jkridner> I'm looking forward to trying the uboot v2. I wonder how easy it will be to move in a USB framework.
  183. [16:39:49] <jkridner> koen: some FFTs for your audio analyzer should be very easy, once we get link support into the kernel.
  184. [16:40:22] * jkridner waits and waits on OE builds.
  185. [16:41:26] <jkridner> I am doing my OE builds on an EC2 machine with the hope that I can eventually move it to distcc and parallel build like crazy.
  186. [16:42:13] <jkridner> has anyone built a dev machine using OE that has all the right tools setup for distcc?
  187. [16:44:13] * JoeBorn (n=jborn@cpe-66-1-5-91.il.sprintbbd.net) Quit ("out for a bit")
  188. [16:44:47] <koen> jkridner: OE developers prefer icecream over distcc
  189. [16:44:57] <koen> distcc is very stupid when it comes to toolchains
  190. [16:47:16] <jkridner> can you recommend a distribution that builds icecream for a desktop and is otherwise nicely setup to perform the parallel build tasks?
  191. [16:47:32] <koen> icecream is KDE ware
  192. [16:47:49] <koen> so it changes protocol more than an average person changes socks
  193. [16:48:04] <jkridner> http://oe.linuxtogo.org/wiki/IceCC
  194. [16:48:25] <koen> that one
  195. [16:48:29] <koen> I don't use
  196. [16:48:33] <koen> it
  197. [16:50:12] <jkridner> how do you stand the wait?
  198. [16:51:37] <koen> I only have to build things once, so I start builds from scratch before I go to bed
  199. [16:52:19] <koen> and this afternoon I took the change to print some negatives in the dark room while the powerpc build started
  200. [16:52:27] <jkridner> but you are messing with a bunch of different platforms.
  201. [16:53:45] <koen> right
  202. [16:53:51] <koen> which can share a lot of packages
  203. [16:54:39] <koen> the neuros is armv5te, which means it shares 90% of its packages with my at91 board
  204. [16:55:27] <koen> and I use bitbake's parallel build facility since it's building on a raid5 array on an SMP machine
  205. [16:55:28] <jkridner> well, there must be someone out there that sympathizes with me. :)
  206. [16:55:56] <jkridner> well, I could take some advantage of that on my Mac Pro.
  207. [16:59:04] <koen> PARALLEL_MAKE = " -j3 "
  208. [16:59:05] <koen> BB_NUMBER_THREADS = "3"
  209. [16:59:24] <koen> the first is a regular parallel make (it passes -j3 to makefiles)
  210. [16:59:50] <koen> the second is the number of bitbake threads, so it will start unpacking the kernel while it's compiling gcc, etc
  211. [17:23:20] * dirk2 (n=dirk@F309e.f.strato-dslnet.de) has left #beagle
  212. [17:23:52] <DJWillis> koen: interesting, so if you have a single CPU box with some reasonable (hardware cached RAID) storage you could up the BB_NUMBER_THREADS and keep PARALLEL_MAKE set to a nothing? I assumed they had to go hand in hand.
  213. [17:28:02] <koen> DJWillis: correct
  214. [17:28:37] <koen> while you are building glibc (which is a beast, uclibc builds in 3 minutes) you can unpack and patch the rest
  215. [17:30:05] * mru builds glibc in not much more than 3 minutes ;-)
  216. [17:39:41] * JoeBorn (n=jborn@adsl-75-3-3-180.dsl.chcgil.sbcglobal.net) has joined #beagle
  217. [19:19:05] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  218. [19:30:25] <ds2> is anyone working on a opensource codec package for the DSP now that there are tools available? things I am thinking of are vorbis,theora, and SBC... others would be nice but those are the main ones
  219. [19:31:55] <suihkulokki> ds2: see http://lists.maemo.org/pipermail/maemo-developers/2008-June/033651.html
  220. [19:32:28] <ds2> I know about the maemo efforts but they use a 2420 which has a different DSP then the 35xx
  221. [19:33:49] <koen> ds2: http://wiki.neurostechnology.com/index.php/Summer_of_Code_2008/Ogg_Theora_Codec
  222. [19:34:44] <koen> with the code at http://github.com/marceloguedes/theora-davinci/tree/master
  223. [19:35:36] <ds2> thought the divinci uses yet another DSP?
  224. [19:35:53] <jkridner> dm6446 and omap3 use the same DSP.
  225. [19:35:59] <ds2> Oh cool
  226. [19:36:07] <jkridner> they have some different accelerators, but the basic DSP ISA is the same.
  227. [19:36:21] <jkridner> DM6446 DSP runs at 600MHz, vs 430MHz on the OMAP3.
  228. [19:36:33] <ds2> thinking more of apps in open pandora
  229. [19:36:37] <suihkulokki> ah, so omap3 has the VLIW architecture now?
  230. [19:36:42] <koen> but the cortex runs at 600MHz vs the 300 on the DM :)
  231. [19:36:43] <jkridner> yes.
  232. [19:37:06] <koen> (the dm has an arm926, not a cortex)
  233. [19:37:15] <jkridner> and Cortex is almost 2:1 instructions per cycle.
  234. [19:37:19] <ds2> so the DM is more like a 2430 then?
  235. [19:37:31] <jkridner> so DM6446 is heavy on DSP performance and OMAP3 is heaving on ARM performance.
  236. [19:37:36] <ds2> oh wait... not even a 2430
  237. [19:37:39] <jkridner> heaving=heavy.
  238. [19:37:39] <koen> 1710
  239. [19:38:11] <ds2> the 2430 has the same DSP as the OMAP3 which is why I thought of it but the 926... Hmmm
  240. [19:38:34] <jkridner> DM6446, in ARM performance, is more like 1710.
  241. [19:38:44] <jkridner> but, DM6446 has more DSP performance than even OMAP3.
  242. [19:38:49] <ds2> *nod* the 926 core
  243. [19:39:37] <ds2> wish there was a cheap atom board available.. be interesting to do a side by side power per compute power comparism
  244. [19:39:50] <koen> $80 isn't cheap enough?
  245. [19:39:57] <ds2> there is an $80 atom board?!
  246. [19:40:00] <ali_as> They should split the atom.
  247. [19:40:52] <koen> ds2: http://linuxdevices.com/news/NS6489450984.html
  248. [19:40:58] <ds2> for the people that have the beagle, what's the typical current draw when powered by the USB port... 100mA? less?
  249. [19:41:07] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
  250. [19:41:07] * BeagleLogBotTest changes topic to 'Welcome to #Beagle | Discussion about the OMAP3 Beagle Board - http://beagleboard.org | Beagle search tools are on #dashboard at irc.gimp.org, NOT here ;) | Log is at http://beagleboard.org/irclog'
  251. [19:41:44] <ds2> wtf a fan on an atom platform? is that a joke
  252. [19:41:58] <jkridner> note the big fan and heatsink on the atom motherboard.
  253. [19:42:00] <jkridner> :)
  254. [19:42:16] <koen> ds2: the fan is for the northbridge
  255. [19:42:33] <ds2> uh.. 12 to 29 Watt?! intel is being funny here
  256. [19:42:45] <ali_as> HAhahahahahahahaaa.
  257. [19:42:49] <jkridner> :)
  258. [19:43:12] <ds2> I think that comparism just ended without buying a single part
  259. [19:43:25] <ali_as> "specifications suggest power consumption ranging down to five Watts in sleep mode"
  260. [19:43:31] <ali_as> AAAaaahahahahahahhahahaaaahahhhahahahaha.
  261. [19:43:58] <ds2> 5watt sleep mode? come back when you have to change the range on your scope to see power consumption in sleep mode
  262. [19:44:14] * ali_as hahahaaaa, cough, ahem.
  263. [19:44:35] <ds2> at that power level, I wonder if a small array of OMAP3s can emulate an x86 faster ;)
  264. [19:44:44] <koen> the beagle does 1.8W at console prompt
  265. [19:45:22] <suihkulokki> I presume without dyntick enabled?
  266. [19:45:23] <jkridner> much of that is the DVI framer and lack of good power management code.
  267. [19:45:28] <ds2> hmmm that's almost 400mA at 5V
  268. [19:45:54] <ds2> <--- thinking of wild things like a hand crank powered system but 400mA is a tad much to hand crank
  269. [19:46:19] <ds2> now if only Linux ran on the bigger MSP430s ;)
  270. [19:46:49] <jkridner> there you go.
  271. [19:47:07] <jkridner> http://mspgcc.sourceforge.net/
  272. [19:47:49] <ds2> I want linux to run on it, not develope for it from Linux
  273. [19:48:10] <ds2> it would be nice to be able to get a reasonable usage on a CR2032 Li cell powered system
  274. [19:48:15] <jkridner> well, but having gcc is a critical step to getting Linux.
  275. [19:48:25] <koen> suihkulokki: with dyntick, but linux-omap git doesn't have any worthwile PM yet
  276. [19:48:34] <ds2> Timex does it with their datalink stuff, alas, it is a epson CPU
  277. [19:48:42] <koen> (for omap3, that is)
  278. [19:48:58] <jkridner> koen: have you tried measuring with the .22 kernel?
  279. [19:49:04] <koen> jkridner: nope
  280. [19:49:16] <jkridner> the power management there isn't complete either, but you might see a few things improve.
  281. [19:49:20] <koen> jkridner: and the .345A was sakoman measurement :)
  282. [19:49:41] <koen> turning of unused clocks saved 0.005 A iirc
  283. [19:49:51] <jkridner> getting all the power management right is not on the top of my priorities right now.
  284. [19:50:12] <koen> the beagle has a powercord, not a battery :)
  285. [19:50:26] <ds2> it "should" just be a matter of having someone unencumbered porting the code from the TI tree into the OMAP tree
  286. [19:50:47] <koen> ds2: that's what nokia seems to be doing these days
  287. [19:50:49] <ds2> the TWL4030 can do battery management
  288. [19:50:52] <jkridner> ARM going into memory retention mode and getting the DRAM into self-refresh should make a nice drop in the power consumption.
  289. [19:51:01] <koen> they just submitted the smartreflex stuff based on the TI code
  290. [19:51:36] <ds2> a port of that code exists but it is encumbered :/
  291. [19:52:27] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  292. [19:53:01] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) Quit (Remote closed the connection)
  293. [19:53:05] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
  294. [20:37:51] * bazbell (n=a0192809@nat/ti/x-e7f9511ac4678380) has joined #beagle
  295. [21:10:58] * JoeBorn (n=jborn@adsl-75-3-3-180.dsl.chcgil.sbcglobal.net) Quit (Read error: 104 (Connection reset by peer))
  296. [21:12:38] * JoeBorn (n=jborn@adsl-75-3-11-133.dsl.chcgil.sbcglobal.net) has joined #beagle
  297. [21:17:19] * ldesnogu (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) Quit ("Leaving")
  298. [21:21:35] * Lopoi (i=db@c-66-30-48-54.hsd1.nh.comcast.net) Quit (Client Quit)
  299. [21:22:10] * DaQatz (i=Stones@c-66-30-48-54.hsd1.nh.comcast.net) has joined #beagle
  300. [21:30:29] <jkridner> ds2: what does it mean to be "encumbered"?
  301. [21:32:03] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) Quit (Remote closed the connection)
  302. [21:32:47] * BeagleLogBotTest (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
  303. [21:46:18] <ali_as> Strings are attached.
  304. [23:37:58] <Crofton|work> urg, two days in deepest darkest WV, during a heat wave

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.