#beagle IRC Log on BeagleBoard.org

Join the chat at beagleboard.org/chat

IRC Log for 2008-05-20

Timestamps are in UTC.

  1. [00:10:54] * jkridner (n=jason@c-76-31-18-64.hsd1.tx.comcast.net) Quit (Read error: 110 (Connection timed out))
  2. [00:57:41] * bazbell (n=a0192809@nat/ti/x-dd0aaf2306a6fff0) has joined #beagle
  3. [02:59:17] * bazbell (n=a0192809@nat/ti/x-dd0aaf2306a6fff0) Quit ("Leaving.")
  4. [03:01:13] * lgentili (n=lgentili@host90.190-224-224.telecom.net.ar) has joined #beagle
  5. [03:14:17] * lgentili (n=lgentili@host90.190-224-224.telecom.net.ar) Quit (Remote closed the connection)
  6. [03:56:40] * BThompson (n=BThompso@cpe-76-185-66-252.tx.res.rr.com) Quit (Read error: 104 (Connection reset by peer))
  7. [05:56:28] * PibbRelay (n=supybot@nat/janrain/x-cfe2c6eaabbc6c81) Quit (SendQ exceeded)
  8. [09:25:20] * PibbRelay (n=supybot@nat/janrain/x-3bf71493a321c2bd) has joined #beagle
  9. [09:44:00] * horizon (i=horizon@141.30.223.123) has joined #beagle
  10. [09:44:24] * horizon_ (i=horizon@141.30.223.123) Quit (Read error: 104 (Connection reset by peer))
  11. [09:46:30] * RogerMonk (n=a0740758@nat/ti/x-32cc178db656ac5b) has joined #beagle
  12. [11:18:16] <koen|uni> ah, tony merged rc3
  13. [11:18:18] * koen|uni builds
  14. [11:18:59] <Crofton> Tony rocks
  15. [11:20:39] * RogerMonk (n=a0740758@nat/ti/x-32cc178db656ac5b) Quit (Remote closed the connection)
  16. [11:21:12] * RogerMonk (n=a0740758@nat/ti/x-26039809eff23534) has joined #beagle
  17. [11:24:54] <DJWillis> koen|uni: Crofton: OE rocks :D
  18. [11:25:07] <Crofton> thanks :)
  19. [11:25:24] <Crofton> well, until you have a bad day
  20. [11:25:43] <Crofton> I'm having a libtool issue building avahi on a 64 bit machine atm :)
  21. [11:25:50] <DJWillis> Crofton: I look forward to those in a sick kind of way ;-)
  22. [11:25:55] <Crofton> 32 bit machine builds ok
  23. [11:26:04] <Crofton> good, we need people like that :)
  24. [11:27:29] <DJWillis> For now, 32bit X84 building for ARMv6/v7 - Built a few noddy helper scripts, worked out some personal conventions and now I have OMAP3EVM and a good chunk of Pandora going, of course the next step is to get a better kernel sorted for Pandora and build the whole lot clean, that will be fun (not) ;-)
  25. [11:28:07] <DJWillis> X86 even ;-), hmmm
  26. [11:28:24] <koen|uni> NOTE: package linux-omap2-2.6.25+2.6.26-rc3+gitf4eb51b909144550048b7922ebd1904a54005394-r15: task do_compile: started
  27. [11:28:28] <koen|uni> there we go
  28. [11:29:33] <koen|uni> heh
  29. [11:29:40] * koen|uni sees DAI driver on linux-omap
  30. [11:31:29] <koen|uni> looks like sakoman could reuse that
  31. [11:36:16] <koen|uni> jkridner_: http://arstechnica.com/articles/paedia/risc-vs-cisc-mobile-era.ars/
  32. [11:36:26] * koen|uni says: death to x86
  33. [11:36:41] * koen|uni heads off to class again
  34. [11:36:53] <Crofton> study hard!
  35. [11:37:07] * RogerMonk (n=a0740758@nat/ti/x-26039809eff23534) has left #beagle
  36. [12:42:09] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
  37. [13:06:08] * NishanthMenon (n=nmenon@nat/ti/x-65d372401eada652) has joined #beagle
  38. [13:30:58] <koen|uni> there, now I officially know what eye-diagrams are
  39. [13:31:45] <Crofton> awesome
  40. [14:00:56] <ali_as> That's a good article on the atom.
  41. [14:01:20] <NishanthMenon> Folks, one question: are we ok to see U-Boot V2 patches as RFC on beagleboard mailing list prior to me sending it to mainline u-Boot list?
  42. [14:02:48] <Crofton> sure
  43. [14:03:06] <Crofton> do we know if u-boot v2 will ever be widely used?
  44. [14:03:29] <sakoman> NishanthMenon: sounds good to me too since I don't subscribe to the u-boot list. too many lists, not enough time.
  45. [14:04:00] <sakoman> Crofton: we can help make it widely used :-)
  46. [14:04:02] <NishanthMenon> Crofton, well.. it is a new thingy.. shrug.. as of date, well... most folks(including me) find u-boot v2 to be nascent.. lots of work to do to get a good feature in
  47. [14:04:26] <Crofton> I haven't looked at it
  48. [14:05:06] <NishanthMenon> but i personally feel that it is better architected (since it copies linux kernel arch) and has better chance of getting stable. further any newbie who has at least a limited knowledge of kernel will find right at home in u-boot v2!
  49. [14:05:40] <NishanthMenon> and ofcourse.. i love being able to use valgrind on sandbox and sparse on arm code...
  50. [14:06:47] <NishanthMenon> will send the patches out in couple of hours..
  51. [14:25:12] <jkridner_> I'll ask the dumb question: what are "valgrind on sandbox" and "sparse on arm"? Do you mean for memory leak detection?
  52. [14:32:55] <NishanthMenon> jkridner_, yes.. u-boot v2 has a build option to build it for your desktop.. so you can write a new feature -> say additional hush commands.. and you'd like to check mem leaks.. yippieee.. valgrind comes to the rescue..
  53. [14:33:43] <NishanthMenon> i was thinking of probably writing a dummy nand,nor and onenand device for sandbox to test fs support.. mebbe sometime in feature..
  54. [14:33:50] <NishanthMenon> s/feature/future
  55. [14:34:00] <jkridner_> k
  56. [14:34:05] * jkridner_ is now known as jkridner
  57. [14:34:50] <NishanthMenon> sparse on arm = make ARCH=arm C=1 CROSS_COMPILE=xyz or make ARCH=arm C=2 CROSS_COMPILE=xyz
  58. [14:35:30] <jkridner> What does "C=x" mean?
  59. [14:37:11] <NishanthMenon> it runs sparse. C=1 runs only on modified files, C=2 runs on all files (without compiling the files if they are already compiled).. sparse is a static analyser.. http://www.kernel.org/pub/software/devel/sparse/
  60. [14:37:42] <NishanthMenon> same thing is present in kernel today too I believe..
  61. [15:43:07] * jkridner|work (n=a0321898@nat/ti/x-e492d2ac3c622caf) has joined #beagle
  62. [16:35:29] <NishanthMenon> Warning.. incoming spam: 14 mails.. on beagleboard.. :(
  63. [16:42:43] <koen|uni> drat, v4l2 is broken in rc3
  64. [16:42:52] * koen|uni is now known as koen
  65. [16:44:34] <suihkulokki> koen: interesting anandtech article, although the writer misses that it was itanium, not ia32 that was supposed to kill risc machines..
  66. [16:45:05] <koen> yeah
  67. [16:45:10] <koen> itanic
  68. [16:46:37] <koen> didn't the lack of proper compilers (and hence linux + windows support) kill itanium?
  69. [16:47:25] <koen> or the excessie power use
  70. [16:47:28] * koen can't remember
  71. [16:47:48] <suihkulokki> those and excessive price
  72. [16:49:02] <suihkulokki> price and halted development also killed the risc processors
  73. [16:49:30] <suihkulokki> "lets stop developing this silly alpha chips, we can soon buy great itanics"
  74. [16:50:42] <koen> that's HP for you
  75. [16:51:21] <koen> I love the design goal of alpha: "make the freaking fastest CPU out there"
  76. [16:51:37] <koen> gotta love the 8MiB cache in the EV7 :)
  77. [16:51:56] <koen> or EV6, I always get confused
  78. [16:52:21] * ldesnogu (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) has joined #beagle
  79. [16:52:26] <ldesnogu> hello
  80. [17:01:37] <NishanthMenon> ldesnogu, hi
  81. [17:16:38] <koen> hey ldesnogu
  82. [17:20:40] <koen> sakoman: did you see the mcbsp patch on linux-omap?
  83. [17:21:06] <sakoman> koen: yes
  84. [17:21:24] <sakoman> I'm using it to replace my own similar patch
  85. [17:21:52] <sakoman> Slowly making progress
  86. [17:21:55] <koen> does mcbsp being a platform device mean I need to register it in my board file?
  87. [17:22:26] <prpplague> koen: refresh my memory, the beagle have ntsc/pal output ?
  88. [17:22:50] <koen> s-video, yes
  89. [17:23:02] <koen> not yet supported in linux-omap git
  90. [17:23:40] <prpplague> ahh
  91. [17:23:45] <sakoman> You need to enable it in defconfig, I don't believe that you need to add anything to your existing board file
  92. [17:24:01] <sakoman> You do need to add an SoC board file though
  93. [17:24:42] <koen> for alsa, right?
  94. [17:25:03] <sakoman> yes in alsa/soc
  95. [17:25:16] <sakoman> that's where the connection to mcbsp happens
  96. [17:25:28] <sakoman> still figuring this all out
  97. [17:25:47] <koen> ok, I;m going to add the mcbsp-to-platformdev patches to linux-omap2_git.bb now
  98. [17:25:59] <sakoman> this next patch set will not be as trivial as the previous ones :-)
  99. [17:31:03] <sakoman> > 1000 lines already
  100. [17:33:05] <koen> is it online somewhere?
  101. [17:39:34] <koen> sweet, the patches still apply to RC3
  102. [17:40:20] * horizon (i=horizon@141.30.223.123) has left #beagle
  103. [17:47:24] <sakoman> koen: no - change rate is too high
  104. [17:48:27] <sakoman> I'll post it once it actually has minimal function
  105. [17:52:02] * BThompson (n=BThompso@nat/ti/x-7ba50542798f3b66) has joined #beagle
  106. [18:00:29] <koen> mcbsp patches are in OE now
  107. [18:25:50] * dirk2 (n=dirk@F3332.f.strato-dslnet.de) has joined #beagle
  108. [18:26:23] * dannyBlue (n=chatzill@a213-22-212-162.cpe.netcabo.pt) has joined #beagle
  109. [18:26:54] <dannyBlue> Hi all
  110. [18:28:22] <koen> hey dannyBlue
  111. [18:29:17] <dannyBlue> Hi Koen, i am picking up on OE amazing work kudos on that.
  112. [18:30:04] <dannyBlue> it's a really great effort
  113. [18:30:20] <dirk2> NishanthMenon: Ping
  114. [18:33:09] <dannyBlue> So did last night discussion about ARM vs MIPS die size get any final conclusions?
  115. [18:36:01] <prpplague> i think it boiled down to "it isn't the size of your processor, but how you use it"
  116. [18:36:08] <dannyBlue> lol
  117. [18:39:31] <dannyBlue> wel there are several conditionants in regards to size maybe mips is smaller but hotter... etc etc...
  118. [18:39:55] <koen> for people interested in bootlogs: http://amethyst.openembedded.net/~koen/beagleboard/bootlog.2.6.23rc3.txt
  119. [18:41:42] <dirk2> koen: .config ?
  120. [18:42:38] <prpplague> koen: wheres apex booting?
  121. [18:43:01] <dannyBlue> looks good i did not know that there was bluetooth support for the beagleboard
  122. [18:43:27] <prpplague> koen: need to specify your LPJ value as part of the kernel args
  123. [18:44:30] <koen> dirk2: http://amethyst.openembedded.net/~koen/beagleboard/defconfig.2.6.26rc3.txt
  124. [18:44:58] <koen> prpplague: u-boot has a 3 seconds timeout atm, so specifying lpj is bogus in that regards
  125. [18:44:59] <prpplague> koen: i assume right now you are just doing dev with a monolithic kernel
  126. [18:45:09] <prpplague> koen: ahh
  127. [18:45:27] <prpplague> koen: i'm freaky about boot time
  128. [18:45:38] <koen> and lowlevel debug is on, so I get console messages twice
  129. [18:45:56] <koen> disabling that and booting with 'quiet' would also make it faster
  130. [18:46:09] <koen> and using rootwait instead of rootdelay
  131. [18:46:10] <koen> etc :)
  132. [18:46:20] <prpplague> yep
  133. [18:46:43] <koen> dannyBlue: beagleboard has usb, so it has bluetooth support :)
  134. [18:46:46] <prpplague> koen: i usually try for power on to usability in 10 seconds or less
  135. [18:48:18] <dannyBlue> I have to go over the specs again... still don't have them memorized
  136. [18:48:47] <koen> dannyBlue: it has an omap3530 + usb, that's all there is :)
  137. [18:49:19] <dannyBlue> ah so the bluetooth is an optional module?
  138. [18:49:29] <koen> usb dongle
  139. [18:49:35] <koen> everything is optional :)
  140. [18:50:15] <NishanthMenon> dirk2, pong. Sorry.. was tied up with meeting (yet again :( )
  141. [18:50:27] <dannyBlue> makes sense, it's a general board
  142. [18:52:06] * dannyBlue is now known as dannyBlue|Away
  143. [18:53:02] <Crofton> anyone know what plans there are to get dsp tools for beagle (beyond buying CCS)?
  144. [18:56:38] <prpplague> Crofton: i wish i had a dollar for ever time i've heard a similiar question with regards to OMAP since 2001
  145. [18:56:55] <Crofton> :)
  146. [18:57:09] <Crofton> there is talk of free (as in free beer) tools
  147. [18:57:26] <Crofton> which is a step in the right direction
  148. [18:58:22] <dirk2> NishanthMenon:
  149. [18:58:26] <dirk2> cp arch/arm/configs/omap3430_sdp3430_per_uart_defconfig .config
  150. [18:58:28] <koen> Crofton: http://open.neurostechnology.com/node/1020
  151. [18:58:29] <dirk2> make menuconfig
  152. [18:58:33] <dirk2> make
  153. [18:59:04] <NishanthMenon> dirk2: alias mymake='make ARCH=arm CROSS_COMPILE=arm-none-eabi- V=1'
  154. [18:59:04] <Crofton> wrong dsp for 6446 and beagle :(
  155. [18:59:16] <NishanthMenon> dirk2, mymake omap3430_sdp3430_per_uart_defconfig;mymake
  156. [18:59:43] <NishanthMenon> dirk2, mymake docs -> you get html documentation (including OMAP one) in Documentation/html
  157. [18:59:53] <NishanthMenon> you need doxygen and graphviz
  158. [19:00:28] <koen> Crofton: it does work on your OSK :)
  159. [19:00:52] <NishanthMenon> dirk2:-rwxr-xr-x 1 nmenon nmenon 28692 2008-05-20 12:05 uboot.bin
  160. [19:01:19] <Crofton> I know
  161. [19:01:34] * Crofton mutters about needing some lackeys again
  162. [19:01:36] <dirk2> NishanthMenon: cc1: error: invalid option ???abi=apcs-gnu???
  163. [19:02:10] <NishanthMenon> wierd... arm-none-eabi-gcc (Sourcery G++ Lite 2008q1-126) 4.2.3 ?
  164. [19:02:12] <Crofton> what really pisses me off is there are several lackeys being wasted in his old research group who could be doing this sort of thing
  165. [19:02:36] <koen> 2008q1 is broken
  166. [19:02:37] <dirk2> gcc version 4.2.3 (Sourcery G++ Lite 2008q1-126)
  167. [19:02:38] <NishanthMenon> dirk2, do you want me to post the quilt set in gz?
  168. [19:02:59] <koen> Crofton: what sort of thing? googling up toolchains?
  169. [19:02:59] <dirk2> NishanthMenon: Patches applied cleanly :)
  170. [19:03:38] <Crofton> no, figuring out how to use the DSP tools and making some samples
  171. [19:04:11] <koen> trick someone into making fftw use the DSP *and* neon :)
  172. [19:04:29] <NishanthMenon> dirk2, my arch/arm/Makefile 24 CPPFLAGS += -mabi=apcs-gnu -DTEXT_BASE=$(TEXT_BASE) -P that looks fine..
  173. [19:05:25] <koen> apcs-gnu?
  174. [19:05:35] <koen> that's so 1990s
  175. [19:06:12] <koen> aapcs is da bomb nowadays
  176. [19:06:16] <dirk2> NishanthMenon: Yes, have same line 24 in arch/arm/Makefile
  177. [19:06:28] <NishanthMenon> building again on my end
  178. [19:06:46] <koen> why does uboot v2 force OABI?
  179. [19:07:07] <koen> (I know it's bare-metal, but still)
  180. [19:07:52] <NishanthMenon> dirk2: http://pastebin.com/d1105fc1b build log (first time i am using pastebin)
  181. [19:08:19] <dirk2> NishanthMenon: Okay, did a quick http://rafb.net/p/GVoOF485.html
  182. [19:08:28] <dirk2> Then it works :)
  183. [19:08:40] <NishanthMenon> aaaaah./..
  184. [19:08:46] <NishanthMenon> mymake alias takes care of it
  185. [19:08:55] <NishanthMenon> u-boot v2 does not handle arch yet properly
  186. [19:09:04] <NishanthMenon> alias mymake='make ARCH=arm CROSS_COMPILE=arm-none-eabi- V=1
  187. [19:09:08] <dirk2> -rwxr-xr-x 1 dirk users 28692 2008-05-20 21:08 uboot.bin
  188. [19:09:19] <NishanthMenon> sorrry... i am so used to using mymake.. :(
  189. [19:10:08] <NishanthMenon> there is MAKEALL which handles every platform.. but not much use... i work around it using mymake alias
  190. [19:10:14] <dirk2> NishanthMenon: Regarding your patches: The first 9 have no numbers in subject
  191. [19:10:36] <NishanthMenon> they dont as they are generic patches.. omap patches build on top of them..
  192. [19:10:43] <NishanthMenon> dirk2, should i be numbering them?
  193. [19:11:20] <dirk2> NishanthMenon: I think so: 00/13, 01/13 etc
  194. [19:11:48] <dirk2> NishanthMenon: Only cosmetics, but one subject has U-BootV2 instead of U-Boot-V2
  195. [19:11:54] <NishanthMenon> Hmmm.. will do.. thanks for pointing it out.. yeah it gets a bit confusing otherwise :( (I am still learning ;) )
  196. [19:12:29] <NishanthMenon> Drat!! the serial patch.. Oops.. thanks for catching..
  197. [19:13:19] <NishanthMenon> dirk2, you can use the uboot.bin I think by downloading over usb and using the result in minicom.. i still need to dig in to the perl script failure
  198. [19:14:10] <NishanthMenon> dirk2, i think it has something to do with using usb to serial converter in linux.. probably the ioctl that Device-Serial pm is sending is not supported.. probably a c code might be better
  199. [19:14:11] <dirk2> NishanthMenon: Yes, thanks, will try that (maybe tomorrow)
  200. [19:14:35] <dirk2> NishanthMenon: Patch 2 has checkpatch errors
  201. [19:15:29] <NishanthMenon> thanks once again.. drat.. i thought quilt did it auto..
  202. [19:15:33] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  203. [19:15:57] <dirk2> NishanthMenon: And patch 8 has: 278 errors, 43 warnings, 1127 lines checked
  204. [19:16:09] <NishanthMenon> yes.. running thru the list now.. :(
  205. [19:16:57] <NishanthMenon> will fix before re-submission. probably will send a patch to add checkpatch to uboot/tools :)
  206. [19:17:50] <dirk2> NishanthMenon: Hmmh, just saw: Use #include <linux/io.h> instead of <asm/io.h>
  207. [19:17:58] <dirk2> Not sure if this applies to uboot ;)
  208. [19:18:46] <NishanthMenon> dirk2, http://pastebin.com/d4886446a
  209. [19:19:17] <NishanthMenon> dunno how linux/io.h got included :(
  210. [19:20:07] <dirk2> No: Above is the warning from Linux checkpatch. It doesn't know that uboot has no linux/
  211. [19:20:32] <dirk2> Seems that Linux checkpatch doesn't apply 1:1 to uboot-v2
  212. [19:21:32] <dirk2> If I remember correctly, Wolfgang states that all uboot *v1* patches have to be checkpatch clean, though
  213. [19:21:44] <NishanthMenon> hmm... but will run thru the checkpatch ones now.. just started cfi.. will hold off that for the timebeing..
  214. [19:21:48] * BThompson (n=BThompso@nat/ti/x-7ba50542798f3b66) Quit ("Trillian (http://www.ceruleanstudios.com")
  215. [19:27:53] <NishanthMenon> drat Lindent KILLS xyzModem.c :(
  216. [19:31:59] <dirk2> Have to go, bye
  217. [19:33:14] * dirk2 (n=dirk@F3332.f.strato-dslnet.de) has left #beagle
  218. [19:33:26] <NishanthMenon> dirk2, thanks. please do ack/nak on list when u do get time ;)
  219. [19:40:22] * BThompson (n=BThompso@nat/ti/x-29b8209547561114) has joined #beagle
  220. [19:40:30] * altex5 (n=x0084519@nat/ti/x-1a98d10f2d31cb8e) has left #beagle
  221. [19:52:04] <koen> Crofton: time to pimp your OSK images on linux-omap ML :)
  222. [20:02:32] <ldesnogu> gn
  223. [20:02:35] * ldesnogu (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) has left #beagle
  224. [21:31:42] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  225. [21:36:06] <prpplague> Crofton: ping
  226. [21:56:39] <Crofton> pong
  227. [22:13:55] * BThompson (n=BThompso@nat/ti/x-29b8209547561114) Quit ("Trillian (http://www.ceruleanstudios.com")
  228. [22:20:46] * dannyBlue|Away (n=chatzill@a213-22-212-162.cpe.netcabo.pt) Quit (Read error: 110 (Connection timed out))
  229. [22:26:24] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
  230. [23:06:03] * BThompson (n=BThompso@cpe-76-185-66-252.tx.res.rr.com) has joined #beagle
  231. [23:45:46] * NishanthMenon (n=nmenon@nat/ti/x-65d372401eada652) Quit ("Ex-Chat")

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.