• [00:00:00] <raster> interesting is my benchmarks pretty much didnt change
  • [00:00:09] <raster> then again maybe my fill is a tiny portion of it
  • [00:00:18] <raster> i didnt test the ons that push fills
  • [00:00:20] <raster> i properly should
  • [00:00:21] <mru> can you post a disassembly of that function?
  • [00:00:30] <raster> i needto add a way to run a particular test from the cmd-line
  • [00:00:38] <raster> hmm right now - no
  • [00:00:46] * mru can't read intrinsics and doesn't trust the compiler
  • [00:00:58] <raster> i'm on my laptop away from my whole dev setup
  • [00:01:13] <raster> and ian did mention theres gcc intrinsics bugs with pack/unpack
  • [00:01:25] <raster> so he wrote some hand-rolled asm to work around it
  • [00:01:26] <mru> gcc is bugs
  • [00:01:39] <raster> but there's fixes in the pipeline (patches) for gcc
  • [00:01:46] <mru> hahahahaha
  • [00:02:23] <mru> what are those DATA32 types?
  • [00:02:26] <raster> i like intrinsics
  • [00:02:29] <raster> ... if they work
  • [00:02:33] <mru> but they don't
  • [00:02:37] <raster> data32 == unit32_t
  • [00:02:38] <mru> which compiler are you using?
  • [00:02:39] <raster> err
  • [00:02:42] <raster> uint32_t
  • [00:03:10] <raster> i did that code long ago andits stayed - when the standard unit32_t etc. types were not universally available
  • [00:03:25] <raster> i typedeffed it myself in the code
  • [00:03:32] <raster> its typedeffed to that
  • [00:03:41] <raster> effectively
  • [00:03:57] <raster> for me.. it's 32bit
  • [00:03:58] <raster> who cares
  • [00:03:59] <raster> :)
  • [00:04:17] <mru> I do, if I have to understand the code
  • [00:04:21] <raster> (well until u do some math.. then signed vs unsigned may matter)
  • [00:04:28] <raster> oh sure
  • [00:04:39] <raster> its simple
  • [00:04:49] <raster> it takes the solid color c
  • [00:04:52] <raster> 32bit
  • [00:05:07] <raster> duplicates it in a 128bit neon reg 4 times
  • [00:05:12] <raster> uses 4 neon registers
  • [00:05:18] * florian (n=fuchs@g228210198.adsl.alicedsl.de) Quit (Read error: 113 (No route to host))
  • [00:05:19] <raster> (could do more - but for now 4 was good enough)
  • [00:05:31] <raster> then just blasts it out
  • [00:05:32] <raster> actually
  • [00:05:34] <raster> i'm stupig
  • [00:05:40] <raster> why did i dup it 4 times
  • [00:05:44] <raster> in 4 regs
  • [00:05:47] <raster> thats just silly!
  • [00:05:48] <ssvb> mru: symbols with leading '_' are ok for C programs, it is double '_' that are reserved or '_' followed by uppercase letter
  • [00:05:51] * raster kicks himself
  • [00:06:07] <mru> ssvb: read the spec again
  • [00:06:14] <ssvb> mru: which one?
  • [00:06:17] <mru> _<lowercase> is reserved at file scope
  • [00:06:18] <mru> c99
  • [00:07:45] * rsalveti (n=salveti@189.70.60.88) Quit (No route to host)
  • [00:07:52] * rsalveti (n=salveti@189.70.60.88) has joined #beagle
  • [00:10:11] <ssvb> mru: ok, thanks
  • [00:10:37] <raster> hmmm
  • [00:10:57] <mru> raster: how you compile that code?
  • [00:11:00] <mru> compiler? flags?
  • [00:11:00] <raster> i just realised life is going to be ugly doing copies if both ends are unaligned
  • [00:11:26] <raster> -fexpensive-optimizations -frename-registers -fomit-frame-pointer -march=armv7-a -mtune=cortex-a8 -mfpu=neon -mfloat-abi=softfp -O2
  • [00:11:35] <mru> copy smaller pieces until destination is aligned
  • [00:11:37] <mru> then just go
  • [00:11:46] <raster> but if src is not aligned
  • [00:11:53] <raster> u'll never be able to do an aligned fixup
  • [00:11:54] <mru> that's ok
  • [00:12:06] <mru> unaligned loads are ok
  • [00:12:14] <mru> it's the stores that can do silly things
  • [00:12:16] <raster> ooh just the stores?
  • [00:12:33] <mru> yes, read my blog post
  • [00:13:36] <denix> Crofton|irssi: it's fine. I guess it became known earlier than expected... :)
  • [00:13:43] <raster> reading
  • [00:13:44] <mru> raster: the s and m arguments are unused?
  • [00:13:45] <denix> oops, sorry
  • [00:13:47] <raster> (again)
  • [00:14:13] <raster> yeah
  • [00:14:14] <raster> unused
  • [00:14:22] <raster> its a generic span handler
  • [00:14:27] <raster> can have src and mask too
  • [00:14:28] <mru> I figured
  • [00:14:31] <mru> just making sure
  • [00:14:46] <raster> but for this one it just uses color+dest and length (in pixels)
  • [00:15:08] <raster> (and ys i know its a little overhead throwing params onto the stack u wont use)
  • [00:15:26] <raster> i need to do a total retake on the software engine
  • [00:15:41] <raster> despite the fact it still rips the pants of pretty much anything out there
  • [00:15:44] <raster> i'm not happy with it
  • [00:15:45] <raster> :)
  • [00:16:13] <raster> i really want to give tiled memory layout a go
  • [00:16:16] <raster> in software
  • [00:17:31] <mru> I suggest you take a look at the code gcc produces from that
  • [00:17:35] <mru> it's hideous
  • [00:17:47] <raster> whihc bits
  • [00:17:51] <raster> the intrinsics?
  • [00:17:52] <mru> all of it
  • [00:18:07] <raster> i probably should
  • [00:18:16] <raster> for mmx i did the hand-rolled macros
  • [00:18:22] <raster> fordoing the mmx/sse
  • [00:18:30] <raster> so it would gnerate just whatit was told
  • [00:18:38] <raster> so i had todo my own register allocs
  • [00:18:46] <raster> (for the mmx side)
  • [00:18:58] <raster> delete col1/2/3
  • [00:19:00] <mru> register allocation is the fun bit ;-)
  • [00:19:03] <raster> and replace it all with col0
  • [00:19:07] <raster> tat was just stupidity on my part
  • [00:19:08] <mru> gcc did that for you
  • [00:19:20] <raster> i dont know what posessed me to do that
  • [00:19:25] <mru> but then it did other just as stupid things
  • [00:19:46] <raster> i should look
  • [00:20:03] <raster> but fist i shold get alignement fixups
  • [00:20:34] * jrmuizel (n=jrmuizel@mozca02.ca.mozilla.com) Quit ()
  • [00:21:37] <ssvb> btw, if using gcc inline assembly, 64-bit and 128-bit neon/vfp arguments seem to be undocumented: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37188
  • [00:22:06] * mru doesn't use gcc inline assembly
  • [00:22:29] <raster> http://gcc.gnu.org/onlinedocs/gcc/ARM-NEON-Intrinsics.html
  • [00:22:32] <raster> thats as good as u will get
  • [00:22:48] <raster> combine it with the armv7 arm - and u are ok
  • [00:22:54] <ssvb> raster: intrinsics are completely evil :)
  • [00:22:55] <raster> the intrinsics miss sone funcs
  • [00:23:01] <raster> like multiple load/store
  • [00:23:08] <raster> ssvb: i know
  • [00:23:23] <raster> but as is the whole world of hand rolling yourself some assembly :)
  • [00:23:34] <raster> is there enything not evil about it?
  • [00:23:35] <raster> :)
  • [00:23:48] <raster> only a few madmen will subject themselves to such insanity
  • [00:24:00] * gletelli___ (n=chatzill@fw-tnat.cambridge.arm.com) has joined #beagle
  • [00:24:57] * gletelli___ is now known as gletelli
  • [00:25:04] <mru> raster: have you seen my neon code for ffmpeg?
  • [00:25:18] <raster> mru: no - but i was intending to look
  • [00:25:22] * gletelli__ (n=chatzill@fw-tnat.cambridge.arm.com) Quit (Read error: 54 (Connection reset by peer))
  • [00:25:29] <raster> for now i have ian's pixman neon patches
  • [00:25:37] <raster> i prefer to learn by example
  • [00:25:40] <raster> and not reference volume
  • [00:25:41] <raster> :)
  • [00:25:47] <mru> ffmpeg has 2500 lines of pure neon assembler
  • [00:25:54] <raster> ouch
  • [00:25:59] <mru> that should get you started ;-)
  • [00:26:05] <raster> :)
  • [00:26:12] <raster> i will snarf it later
  • [00:26:19] <raster> tho my world is more argb land
  • [00:26:28] <raster> but the asm is the same
  • [00:26:44] <mru> depends
  • [00:26:57] <mru> you're not doing a lot of idct stuff
  • [00:27:38] * hartangas (i=83cfcfc0@gateway/web/ajax/mibbit.com/x-46ce7ac5b41b4e2a) Quit (Remote closed the connection)
  • [00:27:55] <ssvb> mru: gcc inline assembly is mostly ok, if you are only using gcc and don't need many free registers
  • [00:28:05] <raster> mru: thank god
  • [00:28:07] <raster> :)
  • [00:28:24] <raster> but... i will want to port my full super/sub sample scaler
  • [00:28:33] <mru> gcc inline assembly is good if you don't need much speed and don't value your sanity
  • [00:28:33] <raster> tho i may decide to limit it to a # of taps
  • [00:28:40] <raster> i'm going to have to brew someting
  • [00:29:03] <raster> i prefer not doing inline asm - if ic an
  • [00:29:05] <raster> well ok
  • [00:29:11] <ssvb> raster: idct is much easier than some complex compositing function involving scaling and other stuff
  • [00:29:13] <Crofton|irssi> what is "sanity"
  • [00:29:19] * __alanc__ (n=a-campbe@nat/ti/x-92ace136be518716) Quit ()
  • [00:29:25] <raster> i prefer doing asm this way:
  • [00:29:34] <mru> Crofton|irssi: what you get after writing a few k lines of neon assembler
  • [00:29:38] <raster> #define mmx_i2r(op, imm, reg) \
  • [00:29:38] <raster> __asm__ __volatile__ (#op " $" #imm ", %%" #reg \
  • [00:29:38] <raster> : /* nothing */ \
  • [00:29:38] <raster> : /* nothing */);
  • [00:29:40] * mib_gm9ppk (i=18528d55@gateway/web/ajax/mibbit.com/x-e57f39d2e3294940) has joined #beagle
  • [00:29:45] <mru> raster: that's terrible
  • [00:30:03] <raster> mru: it actually produces fine code
  • [00:30:06] <mru> gcc will do lots of silly stuff between your macro invocations
  • [00:30:09] <raster> the #define macros are hell
  • [00:30:17] <raster> in my experience
  • [00:30:18] <raster> it doesnt
  • [00:30:23] <raster> i've checked the output to make sure
  • [00:30:29] <raster> it produces exactly the asm i want
  • [00:30:32] * mib_ive4kr (i=837188eb@gateway/web/ajax/mibbit.com/x-e534afd9ef10c3d2) has joined #beagle
  • [00:30:34] <raster> on x86 at least it does
  • [00:31:15] <raster> #define psllq_i2r(imm, reg) mmx_i2r(psllq, imm, reg)
  • [00:31:17] <raster> for example
  • [00:31:28] * hartangas (i=83cfcfc0@gateway/web/ajax/mibbit.com/x-d65bbc95bbc8cb9d) has joined #beagle
  • [00:31:40] <mru> I know nothing about mmx and tend to keep it that way
  • [00:31:45] <raster> combined with the first macro - produces precisesly 1 mmx asm op and nothing more
  • [00:32:23] <raster> but maybe the x86 back end is smarter
  • [00:32:46] <mru> it's always a gamble
  • [00:32:52] <mru> it might work fine 100 times
  • [00:32:53] <raster> i know
  • [00:32:59] <mru> then it does something totally idiotic
  • [00:33:00] <raster> but i did check the gcc -S output
  • [00:33:08] * dcordes (n=dcordes@unaffiliated/dcordes) Quit (Read error: 60 (Operation timed out))
  • [00:33:17] <ssvb> mru: "pure" assembly has its own drawbacks - potential ABI compatibility issues, while gcc inline assembly is more likely to work correctly if accepted by the compiler
  • [00:33:19] <raster> and it seemed to - in all my use cases, to do the right thing
  • [00:33:35] <raster> tho i really stopped wanting to read the asm hen it wasnt doign the mmx
  • [00:33:38] <raster> ie read normal x86 asm
  • [00:33:41] <raster> that is just hell
  • [00:33:49] <mru> pure assembler is fine as long as you don't have to deal with structs
  • [00:34:16] <ssvb> mru: there are various preprocessors for C structs
  • [00:34:30] <ssvb> mru: you can also easily write one yourself
  • [00:34:42] <mru> nothing guarantees the compiler will have the same idea
  • [00:34:49] <mru> padding is unspecified, remember
  • [00:34:49] <raster> just flatten them before u get into the asm
  • [00:34:56] <raster> ie copy them to a flat int/char/whatever
  • [00:35:13] <raster> depends how "pure" the asm is
  • [00:35:26] <ssvb> mru: there are various compiler specific extensions like packed attribute, etc.
  • [00:35:29] <raster> i'd want to keep my asm to the inner most loops
  • [00:35:38] <raster> and pretty much ONLY write the neon asm
  • [00:35:42] <raster> not the standard arm
  • [00:35:52] <mru> look at my code
  • [00:36:21] <mru> there's not much plain arm there
  • [00:36:33] <mru> mostly some loop control stuff
  • [00:36:46] <raster> i'd still keep my loop ctrls in c
  • [00:36:58] <ssvb> raster: then you will lose to pixman ;)
  • [00:37:03] <mru> that's one more place for gcc to mess up
  • [00:37:08] <raster> ssvb: doubt it
  • [00:37:17] <raster> ssvb: still well ahead of pixman.. ven with neon in pixman
  • [00:37:23] <raster> ie plain ye olde c
  • [00:37:28] <mru> I bet I can out-code gcc by a factor of at least 2 on any non-trivial code
  • [00:37:37] <raster> vs neon enhanced pixman... ye olde c wins overall
  • [00:37:48] <raster> pixman beats the c in only 1 test - the over blends
  • [00:38:29] <ssvb> raster: there could be something wrong in your setup, can you run benchmark of xrender/pixman on N800?
  • [00:38:41] <raster> (despite pixman's linear scaler simply being liner interp / nearest and eva's being a full super/sub sampler)
  • [00:38:49] <ssvb> raster: I know that the standard pixmap is slow, especially for scaling
  • [00:39:04] <raster> ssvb: nup. oe angstrom build fo rn800 is broken atm
  • [00:39:11] <raster> i have run it on my beagleboard
  • [00:39:35] <ssvb> raster: just ignore pixman scaling, it is implemented exceptionally bad
  • [00:39:41] <raster> i have xrender (using pixman) benchmarks of 0.2 vs 0.32+neon
  • [00:39:45] <raster> it shows a definite speedup
  • [00:39:56] <raster> got a xls file here...
  • [00:40:02] <raster> want it?
  • [00:41:03] <ssvb> raster: I only benchmarked imlib2 vs. armv6 optimized xrender using render_bench on N800
  • [00:41:16] <raster> i dont use that anymore
  • [00:41:18] <raster> i use expedite
  • [00:41:26] <raster> tests a much wider range of ops and pipeline
  • [00:41:33] <raster> and its evas's software engine - nto imlib2's
  • [00:41:38] <raster> its had a lot more work done to it
  • [00:42:45] <ssvb> raster: in any case, I don't see any reasons why pixman can be slower
  • [00:43:27] <ssvb> raster: one of the problems of pixman is that it is too complex and not so easy to oplimize, so current implementation is not very good
  • [00:44:33] <raster> http://www.rasterman.com/files/evas-perf.xls
  • [00:44:40] <ssvb> raster: I'll try to check how to run expedite, is it a standalone application?
  • [00:44:52] <raster> expedite is stand alone]
  • [00:44:55] <raster> but it relies on evas
  • [00:44:56] <ssvb> ok
  • [00:45:08] <raster> the good thingis i can directly test against lots of targets
  • [00:45:16] <raster> gl vs xrender vs software client etc.
  • [00:45:26] <raster> as in theory xrender and gl SHOULd be neck and neck
  • [00:45:35] <raster> *IF* drivers were implmented properly
  • [00:45:36] <raster> :)
  • [00:45:54] <raster> and pixman is rather simple
  • [00:45:55] <ssvb> raster: I don't have much trust in HW acceleration
  • [00:45:59] <raster> evas's software is pretty ugly
  • [00:46:06] <raster> no.. very ugly
  • [00:46:10] <raster> it needs a total overhail
  • [00:46:19] <raster> overhaul
  • [00:46:59] <raster> ssvb: i dont either
  • [00:47:16] <raster> ssvb:despite the theory - practice has been that xrender accel has been absymal
  • [00:47:27] <raster> i am very glad i didnt throw my eggs all in that basket
  • [00:47:57] <raster> as such pixman tho is your regular run of the mill linear pixel pisher
  • [00:48:35] <raster> and i know that i want to go ways that will drastically change that
  • [00:48:44] <raster> as i siad - iwan to try out tiled memory layout for software
  • [00:49:16] * ScriptRipper (n=martin@opensuse/member/MartinMohring) Quit (Remote closed the connection)
  • [00:49:22] <raster> mru: ./libavcodec/arm/simple_idct_neon.S
  • [00:49:23] <raster> right?
  • [00:49:32] <mru> that's one of the files
  • [00:49:50] <mru> all files with neon in the name contain neon code
  • [00:49:55] <raster> (i'm looking for the none-pure asm - ie c with inline asm)
  • [00:50:05] <mru> there isn't any
  • [00:50:12] <mru> not neon
  • [00:50:17] <raster> aqaaah
  • [00:50:17] <raster> ok
  • [00:50:51] <raster> have the svn tree up to date now
  • [00:53:24] * geckosenator (n=sean@adsl-68-23-87-156.dsl.dytnoh.ameritech.net) has joined #beagle
  • [00:55:04] <raster> mru: on your blog
  • [00:55:52] <raster> 16/64/other is alignement
  • [00:55:53] <raster> right?
  • [00:56:08] <mru> that's how close the two memory accesses are
  • [00:56:19] <mru> 16 means it's in the same 16-byte block and so on
  • [00:56:27] <mru> maybe I could have been more clear
  • [00:56:36] <raster> and 16byet block is a 16byte aligned block
  • [00:56:38] <raster> right?
  • [00:56:46] <raster> and 16byte block isalso 64byte aligned
  • [00:56:54] <raster> (not a 64bye block 16byte aligned)
  • [00:58:05] <mru> a 16-byte block has an address with the 4 lsb zero
  • [00:58:32] <raster> and similar with 64
  • [00:58:35] * _gm_ (n=_gm_@190.173.84.138) has joined #beagle
  • [00:58:39] <raster> (6bits lsb 0)
  • [00:58:42] <mru> yes
  • [00:58:44] <raster> ok
  • [00:58:45] <raster> cool
  • [00:58:47] <raster> just checking
  • [00:58:59] * rsalveti (n=salveti@189.70.60.88) Quit (Read error: 113 (No route to host))
  • [00:59:02] <mru> my guess is that 16 bytes is the size of some write buffer
  • [00:59:11] * rsalveti (n=salveti@189.70.69.34) has joined #beagle
  • [00:59:15] <mru> cache lines are 64 bytes
  • [00:59:15] <raster> it makes sense
  • [00:59:20] <raster> its also the q reg size
  • [00:59:37] <raster> and neon regs are really just a mem bloock that u can map as d or q reg's
  • [01:00:08] <mru> huh?
  • [01:00:39] <raster> neon registers are really just a blob of ram on the chip
  • [01:00:53] <mru> you could say that of any register file
  • [01:00:56] <raster> q0 CONTAINS D0 and D1
  • [01:01:00] <raster> sure
  • [01:01:27] <raster> but q and d'[s are not separate banks
  • [01:01:29] <raster> its the same
  • [01:01:31] <mru> yes
  • [01:01:39] <raster> so q0 will contain d0 and d1
  • [01:01:43] <mru> yes
  • [01:01:55] <raster> as opposed to have its own separate value
  • [01:02:03] <mru> exactly
  • [01:02:08] <raster> and so the largest unit is 128bits - 16bytes
  • [01:02:21] <raster> and that makes sense for it to be the most optimised native unit
  • [01:02:25] <raster> (for neon)
  • [01:02:37] <mru> 64-bit operations are often faster
  • [01:02:43] <raster> d regs are provides for conveience
  • [01:02:54] <raster> as lots of use cases have really no need for 128bits
  • [01:02:55] <mru> many 128-bit instructions issue as two 64-bit operations
  • [01:03:10] <raster> yup
  • [01:03:18] <raster> but some you can optimise as a ginle 128 op
  • [01:03:20] <raster> single
  • [01:03:21] <mru> that could change in a future chip of course
  • [01:03:25] <raster> sure
  • [01:03:33] <raster> so i'd stick to q ops if i can
  • [01:03:36] <mru> sometimes it's faster to split it
  • [01:03:36] <raster> as opposed to d
  • [01:03:46] <raster> *IF* i can fuly use all the q space
  • [01:03:58] <raster> if i am wasting space - its pointless
  • [01:04:12] <mru> you can dual-issue more efficiently if you split the slow q instructions
  • [01:04:52] <mru> provided you have something to dual-issue with of course
  • [01:05:07] <raster> oh sure
  • [01:05:21] <raster> but for example
  • [01:05:23] <raster> simple one
  • [01:05:25] <raster> the color filler
  • [01:05:41] <raster> i cant see writing 64bit instead of 128bit blobs being.. better
  • [01:05:42] <raster> :)
  • [01:05:44] * _don_ (n=don@nat/ti/x-1072c08a8caac35c) Quit ()
  • [01:06:44] <ssvb> raster: you can have very small pictures to process
  • [01:06:55] <ssvb> raster: for example font glyphs
  • [01:06:59] <raster> thats a major butt-pain
  • [01:07:04] <raster> yeah
  • [01:07:16] <raster> neon seems to lend itself nicely to big fat runs of data
  • [01:07:23] <raster> have to deal with 3 pixels hre
  • [01:07:25] <raster> 4 there
  • [01:07:27] <raster> 2 there
  • [01:07:33] <raster> ... and life begins to suck
  • [01:07:47] * lcuk perks up
  • [01:08:20] * fulgas is now known as FuL|OUT
  • [01:08:29] <raster> also having to do the alignment fixups before and "tail" fills after just leads to uglinss in code
  • [01:08:31] <raster> then again
  • [01:08:32] <raster> its asm
  • [01:08:35] <raster> its ugly by nature
  • [01:08:38] <raster> :)
  • [01:22:40] * mib_gm9ppk (i=18528d55@gateway/web/ajax/mibbit.com/x-e57f39d2e3294940) Quit ("http://www.mibbit.com ajax IRC Client")
  • [01:25:17] * Leon_Nardella (n=leon@200-161-14-111.dsl.telesp.net.br) Quit (Remote closed the connection)
  • [01:26:26] * Crofton (n=balister@63.224.149.101) has joined #beagle
  • [01:26:40] <ssvb> raster: asm is not so ugly, assemblers usually have quite powerful preprocessors and macros, it is possible to develop quite complex software using it
  • [01:27:26] <raster> ssvb: x86 asm... is ugly
  • [01:27:38] <raster> just the nature of the instructionset
  • [01:27:46] <raster> i grew up on 68k asm
  • [01:27:53] <raster> so i'm biased
  • [01:27:54] <raster> :)
  • [01:27:58] <mru> beautiful stuff
  • [01:27:59] <ssvb> raster: yes, x86 is ugly
  • [01:28:11] <raster> 68k was ... so clean
  • [01:28:15] <raster> very beautiful
  • [01:28:22] <raster> i was always resistant to learning c
  • [01:28:33] <raster> as 68kasm+macors seemed almost like c
  • [01:28:37] <mru> I prefer RISC CPUs though
  • [01:28:41] <mru> arm is very nice
  • [01:28:47] <raster> yeah
  • [01:28:52] <raster> arm is my 2nd favorite
  • [01:28:58] <raster> tho it should be my fave technically
  • [01:29:08] <raster> i just dont have trhe emotional attachment to it i had with the 68k
  • [01:29:36] <mru> I've done very little with 68k
  • [01:29:43] <mru> but what I've done was very easy
  • [01:30:11] <raster> yeah - its really easy to get it all in your head
  • [01:30:16] <raster> its nice and symmetrical
  • [01:30:22] <raster> src./dest can be anything to anything
  • [01:30:36] <raster> (ok - address vs data regs... but hey)
  • [01:30:42] <mru> about to say that
  • [01:30:48] <mru> and a zillion different addressing modes
  • [01:30:53] <mru> hard to pick the best one
  • [01:30:58] <raster> yeah
  • [01:30:58] <mru> but thats cisc
  • [01:31:04] <raster> but such choice!
  • [01:31:14] <ssvb> these guys develop self-hosting assembler for x86 (one of the most powerful for this platform): http://flatassembler.net/ :)
  • [01:31:28] * JuanG (n=Juan@nat/ti/x-ccde236f3d95a99b) has joined #beagle
  • [01:31:45] <raster> ssvb: nuts
  • [01:31:51] <raster> x86 asm makes me want to cry
  • [01:31:51] * JuanG (n=Juan@nat/ti/x-ccde236f3d95a99b) has left #beagle
  • [01:31:55] <raster> i like arm
  • [01:32:08] <raster> i have to say being able to bitshift while doing another op... rocks
  • [01:33:21] <raster> mru: ` u know about the 68000 vs the 68020 and up
  • [01:33:22] <raster> ie
  • [01:33:26] <raster> all addreses were 32bit
  • [01:33:31] <raster> despite it being a 16/32bit cpu
  • [01:33:41] <raster> it pretended to be 32bit at the instruction level
  • [01:33:57] <mru> the lower numbered ones have a 32-bit core with a 16-bit interface externally
  • [01:34:05] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [01:34:08] <mru> or 8-bit in the 68008 case
  • [01:34:22] <mru> the alu is always 32-bit
  • [01:34:27] <mru> as are the registers
  • [01:34:39] <raster> yeah
  • [01:34:58] <raster> but on the 68000 only 24bits of the mem bus were addressable
  • [01:35:05] <raster> at all
  • [01:35:16] <raster> so the fools at microsoft who wrote amiga basic
  • [01:35:29] <raster> decided to use the upper 8 bits of address resgisters to store extra data/flags
  • [01:35:42] <raster> and along comes the 68020 whihc wires up all 32 bits for addresses
  • [01:35:43] <raster> and boom
  • [01:36:09] <raster> despie moto being pretyt good about the specs
  • [01:36:23] <raster> and pretty much tellign you to treat it as a 32bit cpu for all intents and purposes
  • [01:36:35] <raster> they just havent wired up all the stuff under the hood yet
  • [01:36:42] <raster> people decided toavoid such advice
  • [01:36:43] <raster> :)
  • [01:37:58] <ssvb> raster: afaik mozilla uses/used lower bits of data pointers as extra flags in java script engine, this confused valgrind quite a bit
  • [01:38:22] <mru> yes, they use 1-3 lsb
  • [01:38:46] <mru> xx1 means the high 31 bits are an integer
  • [01:39:09] <mru> xx0 means different things depending on xx
  • [01:40:01] * guillaum1 (n=gl@AMontsouris-153-1-42-213.w90-2.abo.wanadoo.fr) Quit (Read error: 110 (Connection timed out))
  • [01:41:35] <raster> thats ok
  • [01:41:38] <raster> as they filter it out
  • [01:41:46] <raster> but in the 68k case
  • [01:41:47] <raster> they didnt
  • [01:41:57] <raster> they relied on the hw at the time tojust ignore the upper 8 bits
  • [01:43:47] <ssvb> ARM also had the legacy alignment model with unaligned memory accesses rotating data and other effects
  • [01:45:35] <ssvb> if some software is hardcoded to rely on it, it will not work nowadays on Cortex-A8
  • [01:47:51] <raster> aye
  • [01:47:54] <raster> bad software
  • [01:47:55] <raster> bad
  • [01:47:58] * raster spanks it
  • [01:51:53] * leslie (n=leslie@222.66.141.242) has joined #beagle
  • [01:59:14] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit ()
  • [02:01:36] * mru (n=mru@thrashbarg.mansr.com) Quit (Read error: 104 (Connection reset by peer))
  • [02:10:58] * mru (n=mru@thrashbarg.mansr.com) has joined #beagle
  • [02:11:46] * _gm_ (n=_gm_@190.173.84.138) Quit (Remote closed the connection)
  • [02:13:09] * _gm_ (n=_gm_@190.173.84.138) has joined #beagle
  • [02:16:31] * _gm_ (n=_gm_@190.173.84.138) Quit (Remote closed the connection)
  • [02:19:11] * odesus (n=Omar@201.165.145.213) has joined #beagle
  • [02:19:21] * odesus (n=Omar@201.165.145.213) has left #beagle
  • [02:19:31] <leslie> my beagleboard arrived~~~
  • [02:20:36] * _gm_ (n=_gm_@190.173.72.92) has joined #beagle
  • [02:21:04] <leslie> Could anybody help that what toolchain is used stably? and where can I find it?
  • [02:22:47] * ddompe (n=ddompe@12.108.221.116) has joined #beagle
  • [02:23:03] <ddompe> atin_: ping
  • [02:24:44] * gregoiregentil (n=zonbu@adsl-71-135-114-242.dsl.pltn13.pacbell.net) has joined #beagle
  • [02:24:47] * Wowbagger_ (n=wowbagge@d154-20-130-66.bchsia.telus.net) has joined #beagle
  • [02:25:06] * Wowbagger_ is now known as WowGrumpy
  • [02:38:28] * _gm_ (n=_gm_@190.173.72.92) Quit (Read error: 104 (Connection reset by peer))
  • [02:48:02] <leslie> when I power on beagleboard, do I suppose to get some thing out from UART?
  • [02:50:23] <ds2> yes
  • [02:50:44] <ds2> either the U-boot banner or the 40T magic if you manage to erase your NAND
  • [02:56:56] <leslie> I got nothing...
  • [02:57:26] <ds2> check the cable you are using, make sure it is a "AT/Everex" pin out
  • [02:57:43] <ds2> there are 2 common pin outs used on PCs and the other one doesn't work
  • [02:57:57] <ds2> also, you will need a Null modem DB-9 to DB-9 cable to the PC
  • [03:05:47] <geckosenator> didnt' someone add support for usb?
  • [03:07:32] <jkridner> geckosenator: it isn't quite 100% there yet and requires you to reprogram the u-boot on your board first.
  • [03:08:46] <geckosenator> oh ok
  • [03:08:59] <leslie> ds2: thanks
  • [03:09:01] <geckosenator> well maybe the revC boards will ship with it
  • [03:09:25] * vlad (n=asdf@69-12-240-128.dsl.static.sonic.net) Quit (kubrick.freenode.net irc.freenode.net)
  • [03:11:19] * vlad (n=asdf@69-12-240-128.dsl.static.sonic.net) has joined #beagle
  • [03:11:38] <leslie> I have set /etc/inittab in Angstrom to "l3:3:wait:/sbin/getty 115200 ttyS2", is this setting correct to get tty3 output to UART?
  • [03:13:27] * robtow (n=rtow@nat/montavista/x-3d2dbb42c7ff6bff) Quit (Read error: 110 (Connection timed out))
  • [03:16:10] <geckosenator> why is it you want tty3?
  • [03:16:26] * ddompe (n=ddompe@12.108.221.116) Quit ()
  • [03:17:24] * mru (n=mru@thrashbarg.mansr.com) Quit (Read error: 110 (Connection timed out))
  • [03:17:37] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) Quit (Read error: 131 (Connection reset by peer))
  • [03:17:53] * _gm_ (n=_gm_@190.173.68.29) has joined #beagle
  • [03:20:51] <leslie> geckosenator: maybe I describe error. I have a beagleboard and SD card, but I don't have DVI display device
  • [03:21:31] <leslie> so, I want to have it booted and controled by UART, how to achieve this?
  • [03:21:36] <geckosenator> well from what I recall the consol on uart isn't tty3
  • [03:21:52] <geckosenator> leslie: for me it did that by default
  • [03:22:06] <leslie> from SD card?
  • [03:22:11] <geckosenator> yes
  • [03:22:38] <geckosenator> you might need certain kernel parameters in uboot
  • [03:23:11] <leslie> the supplier has show that it boot from SD card and use init 5
  • [03:23:25] <leslie> I modified it to init 3 when boot
  • [03:23:28] <ds2> ttyS2 is the UART port
  • [03:23:40] <geckosenator> ds2: right, not tty3
  • [03:23:48] <ds2> tty3 is virtual console 3 on the DVI display if you have fbcon enabled
  • [03:23:55] * Strontium (n=Strontiu@118.101.106.2) has joined #beagle
  • [03:24:01] <ds2> people typically use tty1 if they want to have getty on the DVI display running
  • [03:24:12] <leslie> oh, ok
  • [03:24:14] <geckosenator> well you could use tty3 too
  • [03:24:27] <ds2> then you need to do CTRL-ALT-F3
  • [03:24:32] <geckosenator> there might even be a way to hack it to make it over the uart, but I woudln't recommend
  • [03:24:34] * mru (n=mru@thrashbarg.mansr.com) has joined #beagle
  • [03:24:37] <geckosenator> normally just alt-f3
  • [03:24:51] <geckosenator> or alt-<left arrow key> twice :-P
  • [03:24:57] <geckosenator> I mean right
  • [03:25:11] * ssvb (n=ssvb___@a88-114-221-132.elisa-laajakaista.fi) Quit (Read error: 145 (Connection timed out))
  • [03:25:19] <leslie> so what should I set in /etc/inittab? I have just set it to use init 3, and what else?
  • [03:25:21] <Strontium> Hello, I am developing an add on board for the beagle, and I was wondering if the pin out for the new boards digital video interface is available? Also, if the physical dimensions are available so I can design a board to match it.
  • [03:25:56] <geckosenator> Strontium: what kind of board are you making? I have some I just got but I've been working on other stuff so I didn't populate them yet
  • [03:26:42] <leslie> is it to do like: id:S:initdefault: in inittab?
  • [03:27:10] <Strontium> Its a GP board. It has 100Mbs Ethernet, Interface for OLED/TFT panels, RS485, WIFI and Zigbee, RT, Extra SDCARD slot.
  • [03:27:15] <ds2> you are right, ctrl-alt-f3 in X11
  • [03:28:15] <geckosenator> yes
  • [03:28:25] <ds2> probally init default... like id:3:initdefault:
  • [03:28:26] <geckosenator> but if you aren't in X11 the other ones work
  • [03:28:37] <ds2> but that depends on the init you are using
  • [03:28:47] <geckosenator> oh, could be
  • [03:29:22] <ds2> geckosenator: you are completely right... the init comment was about the syntax of the default runlevel
  • [03:29:26] <Strontium> THE LCD will have a touch screen interface.
  • [03:29:49] <geckosenator> Strontium: cool, I have a board here that has SDCARD RT, zigbee, bluetooth, rs232, and tft panel interace with driver, as well as various switching dc/dc regulators
  • [03:29:59] <Strontium> I have a design that uses a TI chip to convert the DVI signal back to RGB, but that is hardly an elegant way to do it.
  • [03:30:09] <geckosenator> Strontium: it has audio and touch screen too.. it's designed for avr32 boards but I'm going to try to wire it to the beagl
  • [03:30:15] <geckosenator> Strontium: I have that too
  • [03:30:16] <ds2> geckosenator: does it all work with the beagle? ;)
  • [03:30:18] <Strontium> nice
  • [03:30:31] <geckosenator> ds2: I'll see.. but right now I'm working on other stuff..
  • [03:30:40] <ds2> heh
  • [03:30:54] <geckosenator> I have dvi->parallel, parallel->parallel level translator
  • [03:31:02] <geckosenator> and parallel->vga
  • [03:31:22] <ds2> geckosenator: you have those dvi to parallel boards built already?
  • [03:31:33] <geckosenator> I have the boards, but I didn't populate them yet
  • [03:32:05] <ds2> I'd be curious to see if it just works
  • [03:32:14] <geckosenator> I'm trying to finish up fixed-point support for avr-gcc..
  • [03:32:16] <ds2> and if it does, I might make a copy ;)
  • [03:32:16] <geckosenator> ds2: me too
  • [03:32:23] <geckosenator> I'd send you a board
  • [03:32:27] <geckosenator> or you could copy it
  • [03:33:01] <ds2> I want something slightly different... DVI to VGA... so i'd you your design as a stepping stone to put on a DAC
  • [03:33:12] * ssvb (n=ssvb___@a88-114-221-132.elisa-laajakaista.fi) has joined #beagle
  • [03:33:20] <geckosenator> yeah, I have both but on separate boards
  • [03:33:26] <geckosenator> so you could plug them together
  • [03:33:29] <ds2> Oh? nice
  • [03:33:33] <geckosenator> and the connector is designed for that
  • [03:33:42] <geckosenator> or you could plug in a board which has a connector for an lcd
  • [03:33:51] <ds2> but then I also want VGA to S-Video, S-Video to DVI and VGA to DVI too
  • [03:34:06] <ds2> that last few combo is for the picoDLP
  • [03:34:09] <geckosenator> well, vga to dvi isn't oo hard
  • [03:34:24] <geckosenator> why do you need that though? the beagle has dvi
  • [03:34:30] <ds2> nope, there are one chip solutions. just need time
  • [03:34:35] <ds2> I can mill PCBs
  • [03:34:42] <geckosenator> heh, ok
  • [03:34:48] <geckosenator> pcbs are pretty cheap to order already
  • [03:35:04] <ds2> not for the quick turrn arounds
  • [03:35:13] <geckosenator> true
  • [03:35:25] <geckosenator> well I didn't do anythign with svideo
  • [03:35:42] <ds2> there are one chip solutions for that too
  • [03:35:48] <geckosenator> I got a dvi to vga converter, and a vga to parallel lcd converter.. but it didn't work with the beagle
  • [03:35:57] <ds2> it is all a matter of laying out a board, making the board, stuffing it, and praying ;)
  • [03:36:04] <geckosenator> heh
  • [03:36:12] <geckosenator> well if mine works you can look at my schematic
  • [03:36:20] <geckosenator> I made like 7 boards
  • [03:36:25] <geckosenator> so I figure not all of them will work
  • [03:36:26] <ds2> it isn't the schematic that is important... it is the layout
  • [03:36:42] <geckosenator> heh, well if the schematic is wrong, the layout doesn't matter
  • [03:36:44] <ds2> at the higher res, there are RF blackmagic involved
  • [03:36:48] <ds2> true
  • [03:37:05] <geckosenator> but the beagle only goes to like 80mhz pixel clock
  • [03:37:23] <ds2> think it goes higher with DSS2
  • [03:37:31] <mru> 86
  • [03:37:34] <geckosenator> ok
  • [03:37:43] <ds2> mru: just 86? hmmmm
  • [03:37:49] <mru> 173/2
  • [03:37:56] <ds2> ah gotcha
  • [03:38:04] <ds2> pesky fixed dividers
  • [03:38:23] <geckosenator> doesn't the omap support dual displays?
  • [03:38:24] <mru> 173 is the max functional clock for the dss and pixel clock can be at most half of that
  • [03:39:02] <ds2> one of the eval boards for the omap has 2 LCDs + S-Video but I have never seen the second LCD working in Linux
  • [03:39:21] <geckosenator> well it would require software and hardware
  • [03:39:39] <Strontium> ds2: DVI to VGA, check out: http://dme.ghost2.net/hdfury/home.php?SID=34&deflang=en
  • [03:39:39] <geckosenator> but anyway, maybe that would allow to double the pixelclock on a single display
  • [03:39:49] <geckosenator> Strontium: I have one of those
  • [03:40:08] <geckosenator> I couldn't make it work yet with the beagle though
  • [03:40:34] <ds2> Strontium:: more fun to build one
  • [03:40:46] <Strontium> true, must not have liked the timing.
  • [03:40:50] <Strontium> ds2, true
  • [03:44:55] <Strontium> So back to my original question. the C rev boards are going to have RGB digital signals, is there a pin out and layout available so I can design a matching board, in anticipation of the new C rev, or do we have to wait till its released?
  • [03:45:56] <ds2> I think you will have to wait
  • [03:50:28] <geckosenator> maybe if you emailed gerald he could tell you if he knows yet or not
  • [03:57:11] * mib_nt31j7 (i=522ddc8b@gateway/web/ajax/mibbit.com/x-9635aaa11e1cafb8) has joined #beagle
  • [03:57:43] <mib_nt31j7> Hello. Can anyone give me help trying to get my beagleboard to do 1024x768 over DVI?
  • [03:58:04] <mib_nt31j7> I've been pulling my hair trying to do this for days now
  • [03:58:18] <mib_nt31j7> I need 24-bit so I'm using DSS2
  • [03:58:25] <ds2> AFAIK, it just works
  • [03:58:46] <mib_nt31j7> I've tried various prebuilt kernels and just built my own 2.6.28
  • [03:58:53] <ds2> but I am using the reference omapfb when I did that
  • [03:59:21] <ds2> what is it doing?
  • [03:59:35] <mib_nt31j7> I can get 1024x768 but when I try and play anything on the framebuffer using mplayer it comes up at 640x480
  • [03:59:48] <ds2> what does fbset say?
  • [04:02:12] * robtow (n=rtow@64.62.142.114) has joined #beagle
  • [04:03:41] * mib_nt31j7 (i=522ddc8b@gateway/web/ajax/mibbit.com/x-9635aaa11e1cafb8) Quit (kubrick.freenode.net irc.freenode.net)
  • [04:03:59] * mib_nt31j7 (i=522ddc8b@gateway/web/ajax/mibbit.com/x-9635aaa11e1cafb8) has joined #beagle
  • [04:04:57] * mib_ive4kr (i=837188eb@gateway/web/ajax/mibbit.com/x-e534afd9ef10c3d2) Quit ("http://www.mibbit.com ajax IRC Client")
  • [04:12:42] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [04:14:23] <mib_nt31j7> hello ds2 are you still there?
  • [04:14:52] <ds2> yes?
  • [04:14:59] <mib_nt31j7> so fbset initially says "640x480-59" on startup
  • [04:15:12] <ds2> that might be why
  • [04:15:28] <mib_nt31j7> When I then call fbset to try and set the mode to 1024x768 it thereafter reports that
  • [04:15:40] <ds2> you need to get that to say 1024x768... it currently doesn't support mode changes (unless that was recently added)
  • [04:16:11] <mib_nt31j7> I have videomode=1024x768@60,vxres=1024,vyres=768 in my u-boot environment
  • [04:16:55] <mru> that doesn't work on the beagle
  • [04:17:06] <mib_nt31j7> I also have a kernel command line parameter of video=omapfb:mode:1024x768@60
  • [04:17:15] <mru> that one has effect
  • [04:17:43] <mib_nt31j7> I can see that parameter appearing in my dmesg
  • [04:18:00] <ds2> I haven't had time to really mess with DSS2... stock l-o omapfb and reference driver omapfb is the stuff I am dealing with and that just works
  • [04:18:02] <mib_nt31j7> I also see OMAP DSS rev 2.0
  • [04:18:19] <mib_nt31j7> unfortunately I need 24-bit
  • [04:18:47] <mib_nt31j7> I've built my own 2.6.28 kernel using open embedded
  • [04:18:58] <mib_nt31j7> ... but I'm not sure how to change the default mode in it
  • [04:19:11] <ds2> 24bit works with those drivers; there is just a bug with the reference driver that I have a patch for
  • [04:19:40] <mib_nt31j7> ... also, now that I've built it whenever I change one of the patches it doesn't rebuild
  • [04:20:03] <mib_nt31j7> What'd the bug you have a patch for?
  • [04:20:44] <ds2> on the reference driver, 24 bit mode is borq cuz the LCDC wasn't configured right...there was a typo
  • [04:20:58] <ds2> soit produced funny colors
  • [04:24:57] * mib_nt31j7 (i=522ddc8b@gateway/web/ajax/mibbit.com/x-9635aaa11e1cafb8) Quit (kubrick.freenode.net irc.freenode.net)
  • [04:43:35] * mib_nt31j7 (i=522ddc8b@gateway/web/ajax/mibbit.com/x-521eb6db77569078) has joined #beagle
  • [04:46:02] * mib_j0tpa4 (i=187e21e5@gateway/web/ajax/mibbit.com/x-6fe20f7790233fc9) has joined #beagle
  • [04:46:54] <mib_j0tpa4> I am having problems partitioning the sd card, everything goes fine until this step: sudo mkfs.ext3 -L LABEL2 /dev/sdc2
  • [04:47:02] <mib_j0tpa4> it says it cannot find the device
  • [04:47:06] * _gm_ (n=_gm_@190.173.68.29) Quit (Remote closed the connection)
  • [04:47:31] * _gm_ (n=_gm_@190.173.68.29) has joined #beagle
  • [04:47:44] <mib_j0tpa4> sudo mkfs.ext3 -L LABEL2 /dev/mmcblk0p2
  • [04:48:34] <mib_j0tpa4> anyone got any ideas???
  • [04:48:36] * emeb (n=ericb@ip72-223-90-212.ph.ph.cox.net) has left #beagle
  • [04:49:48] <geckosenator> hi
  • [04:50:04] <geckosenator> mib_j0tpa4: is it really sdc?
  • [04:50:13] <mib_j0tpa4> no it is sudo mkfs.ext3 -L LABEL2 /dev/mmcblk0p2
  • [04:50:15] <mib_j0tpa4> sorry
  • [04:50:23] <geckosenator> yo0u are making it on the beagle?
  • [04:50:32] <geckosenator> you are runnign that command on the beagle?
  • [04:51:03] <mib_j0tpa4> oh sorry, i am trying to set up my sd card to run angstrom for the first time, and im doing a dual partition
  • [04:51:42] <geckosenator> so
  • [04:51:48] <geckosenator> it isn't /dev/mmcblk0p2
  • [04:53:16] <geckosenator> that was your problem?
  • [04:53:41] <mib_j0tpa4> it says it's /dev/mmcblk0p1p2
  • [04:53:50] <mib_j0tpa4> it doesn't work that way either
  • [04:54:08] <mib_j0tpa4> thats for the seccond partition
  • [04:54:35] <geckosenator> uh
  • [04:54:44] <geckosenator> heres what I suggest
  • [04:54:49] <geckosenator> unplug the card
  • [04:54:53] <geckosenator> then plug it in
  • [04:54:56] <geckosenator> and run dmesg
  • [04:55:03] <geckosenator> and see what device got added
  • [04:55:32] <mib_j0tpa4> mmcblk0:
  • [04:56:10] <mib_j0tpa4> thats the only partition that shows up
  • [04:56:32] <geckosenator> oh
  • [04:56:37] <geckosenator> did you run fdisk?
  • [04:56:52] <mib_j0tpa4> yep.. i've done it several times now
  • [04:57:02] <geckosenator> did you type 'w'
  • [04:57:04] <geckosenator> in fdisk?
  • [04:57:08] <mib_j0tpa4> yep
  • [04:57:19] <geckosenator> did it say writing partitions?
  • [04:57:26] <mib_j0tpa4> yes
  • [04:57:36] <geckosenator> did it say kernel will use old partition table until reboot?
  • [04:58:26] <mib_j0tpa4> don
  • [04:58:27] <mib_j0tpa4> t
  • [04:58:30] <mib_j0tpa4> remember
  • [04:58:42] <mib_j0tpa4> lol, the dmes command wiped it out
  • [04:59:31] <geckosenator> wel you could try runnign again
  • [04:59:33] <geckosenator> or reboot
  • [04:59:40] <geckosenator> just to be sure
  • [05:00:02] <mib_j0tpa4> i will do it again, but both times, before i hit "w", i printed my settings and they were perfect...
  • [05:00:08] <mib_j0tpa4> so maybe i need to reboot?
  • [05:00:23] <mib_j0tpa4> thanks a bunch
  • [05:00:35] <geckosenator> try it
  • [05:00:44] <geckosenator> sorry I need telepathy to do better
  • [05:00:48] * mib_j0tpa4 (i=187e21e5@gateway/web/ajax/mibbit.com/x-6fe20f7790233fc9) Quit ("http://www.mibbit.com ajax IRC Client")
  • [05:11:10] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit ()
  • [05:12:50] * ceyusa (n=ceyusa@189.163.18.92) Quit ("Ex-Chat")
  • [05:14:02] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [05:16:01] * amitu (n=_amit_@cpe-67-49-44-99.socal.res.rr.com) has joined #beagle
  • [05:17:31] <amitu> hey guys i finally got the beagleboard and i've been reading up quite a bit....i needed some clarifications to see if i understood it correctly
  • [05:17:41] <amitu> may i ask the questions?
  • [05:19:55] * vimguy (i=432a64f4@gateway/web/ajax/mibbit.com/x-e1a4462cd24c0121) has joined #beagle
  • [05:20:47] * vimguy (i=432a64f4@gateway/web/ajax/mibbit.com/x-e1a4462cd24c0121) Quit (Client Quit)
  • [05:20:51] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit ()
  • [05:30:07] * moi1234 (n=moi@ip70-179-18-67.sd.sd.cox.net) has joined #beagle
  • [05:32:45] * amitu (n=_amit_@cpe-67-49-44-99.socal.res.rr.com) Quit ("Leaving")
  • [05:47:34] * sh1 (i=3d5fc3cd@gateway/web/ajax/mibbit.com/x-c56080a1dd0408e8) has joined #beagle
  • [05:56:01] * robtow (n=rtow@64.62.142.114) Quit (Read error: 104 (Connection reset by peer))
  • [05:57:50] * WowGrumpy (n=wowbagge@d154-20-130-66.bchsia.telus.net) Quit ("Leaving")
  • [06:20:47] * TAK2004 (n=Administ@dslb-088-072-249-153.pools.arcor-ip.net) has joined #beagle
  • [06:23:46] * ZeZu (n=null@c-98-227-56-17.hsd1.in.comcast.net) Quit (Read error: 104 (Connection reset by peer))
  • [06:30:46] * ZeZu (n=null@c-98-227-56-17.hsd1.in.comcast.net) has joined #beagle
  • [06:32:19] * mib_8v8h9r (i=63c71f4e@gateway/web/ajax/mibbit.com/x-adb02b4e25cd8499) has joined #beagle
  • [06:32:24] * mib_8v8h9r (i=63c71f4e@gateway/web/ajax/mibbit.com/x-adb02b4e25cd8499) Quit (Client Quit)
  • [06:37:30] <koen> good morning all
  • [06:38:21] <sakoman> good morning koen!
  • [06:49:38] * hsi (n=hsi@2001:16d8:ff1a:0:216:3eff:fe5a:59cd) has joined #beagle
  • [06:50:41] * hsi is now known as hugsig
  • [06:54:48] * mib_nt31j7 (i=522ddc8b@gateway/web/ajax/mibbit.com/x-521eb6db77569078) Quit ("http://www.mibbit.com ajax IRC Client")
  • [07:07:29] * pundiramit (n=amit@122.166.2.243) has joined #beagle
  • [07:07:38] * pundiramit (n=amit@122.166.2.243) has left #beagle
  • [07:13:05] * jsync (n=jess@de1-as21224.alshamil.net.ae) has joined #beagle
  • [07:13:18] * koen builds the latest pm branch
  • [07:15:00] <sh1> me too. lol
  • [07:15:05] * garren|work (n=chatzill@mail.dm.co.za) has joined #beagle
  • [07:15:32] <garren|work> morning all
  • [07:20:39] * discoverlego (i=d2124e02@gateway/web/ajax/mibbit.com/x-701d5aff934b2e65) has joined #beagle
  • [07:23:58] <koen> it boots
  • [07:24:48] <geckosenator> yay
  • [07:25:21] <discoverlego> hi all........whenever i give this comand " make CROSS_COMPILE=/root/CodeSourcery/Sourcery_G++_Lite/bin/arm-none-linux-gnueabi- " i get the file x-load.bin but with lots of warnings as well as with omap3530beagle_config i get just configuring the ompa3530_beagle file how can i know that it's correct
  • [07:28:43] <sh1> are you seeing all domains going to ret and off?
  • [07:39:33] * manutk (i=7aa9b085@gateway/web/ajax/mibbit.com/x-1bd43f89da883ba1) has joined #beagle
  • [07:50:41] * raster (n=raster@124-170-202-104.dyn.iinet.net.au) Quit ("Gettin' stinky!")
  • [07:56:01] <Strontium> When a raw beagle board boots (before anything is installed on it straight from digikey), it outputs color bars on the svideo port. Where is the code that does this and where do i find it?
  • [07:56:44] <ds2> NAND and look for the U-boot source code.
  • [07:58:54] * Xerion_ (n=xerion@82-170-197-160.ip.telfort.nl) has joined #beagle
  • [08:00:50] * aleij (n=ad@62.161.153.118) has joined #beagle
  • [08:04:43] <manutk> i have a wierd problem of DVI-D output coming only when i connect serial port to another computer and power on that computer
  • [08:05:14] <manutk> can any one help me on this
  • [08:08:15] * sadhu (i=53eed3a2@gateway/web/ajax/mibbit.com/x-6e461dc78a5c824d) has joined #beagle
  • [08:10:06] <kulve> that sure does sound a bit odd..
  • [08:10:42] <manutk> it happens with USB in device mode also
  • [08:11:06] <manutk> in that case also i need to connect power to the computer where USB A is connected to
  • [08:11:28] <manutk> other wise, DVI display goes blank with "cable not connected " msg
  • [08:14:27] * Xerion (n=xerion@82-170-197-160.ip.telfort.nl) Quit (Read error: 110 (Connection timed out))
  • [08:14:27] * Xerion_ is now known as Xerion
  • [08:19:07] <koen> manutk: what kind of powersupply is connect to the beagle?
  • [08:19:11] <koen> connected*
  • [08:20:09] <manutk> 5V
  • [08:20:15] <manutk> which i got from digi key only
  • [08:20:23] <koen> how many mA?
  • [08:21:39] <manutk> 5V 3.2A
  • [08:21:43] <manutk> is what is shown
  • [08:22:09] <manutk> this actually i ordered along with the board
  • [08:22:44] * ZeZu (n=null@c-98-227-56-17.hsd1.in.comcast.net) Quit (kubrick.freenode.net irc.freenode.net)
  • [08:22:49] <koen> I've seem DVI flake out when you provide ~300mA to the board
  • [08:23:10] * sh1 (i=3d5fc3cd@gateway/web/ajax/mibbit.com/x-c56080a1dd0408e8) Quit (Client Quit)
  • [08:23:25] * ZeZu (n=null@c-98-227-56-17.hsd1.in.comcast.net) has joined #beagle
  • [08:23:45] <manutk> can it be a problem wih the DVI cble
  • [08:23:54] <manutk> actually i bought a cheap one
  • [08:24:07] <manutk> a low quality one
  • [08:26:14] <ds2> what does the LEDs on the board do?
  • [08:27:42] <manutk> power is glowing solid
  • [08:28:05] <manutk> USR0 USR1 are blinking
  • [08:30:03] <ds2> Oh the screen is probally blanked
  • [08:30:17] <ds2> if the LEDs are blinking, the board is running
  • [08:30:28] <ds2> if you got a USB keyboard, hit any key
  • [08:30:55] <ds2> otherwise, try kicking off a process that does something like 'while :;do sleep 200000000 < /dev/fb;done'
  • [08:35:25] * manutk (i=7aa9b085@gateway/web/ajax/mibbit.com/x-1bd43f89da883ba1) Quit ("http://www.mibbit.com ajax IRC Client")
  • [08:45:26] * manutk (i=7aa9b71f@gateway/web/ajax/mibbit.com/x-b0eedcec0063c15e) has joined #beagle
  • [08:48:41] * leslie (n=leslie@222.66.141.242) Quit (Read error: 60 (Operation timed out))
  • [08:49:48] * sadhu (i=53eed3a2@gateway/web/ajax/mibbit.com/x-6e461dc78a5c824d) Quit ("http://www.mibbit.com ajax IRC Client")
  • [08:49:58] * leslie (n=leslie@222.66.141.242) has joined #beagle
  • [09:05:39] * recalcati (i=51485a4f@gateway/web/ajax/mibbit.com/x-eb579fa9a2c4dd72) has joined #beagle
  • [09:06:28] * manutk (i=7aa9b71f@gateway/web/ajax/mibbit.com/x-b0eedcec0063c15e) Quit ("http://www.mibbit.com ajax IRC Client")
  • [09:08:54] * nemequ (n=nemequ@ip68-111-215-155.sd.sd.cox.net) Quit (Connection timed out)
  • [09:09:06] * mib_1sioyv (i=5a0f9024@gateway/web/ajax/mibbit.com/x-6be40cf4cbadd109) has joined #beagle
  • [09:09:23] * [X]Spot (n=stancho@78.90.99.168) has joined #beagle
  • [09:11:34] * ZeZu (n=null@c-98-227-56-17.hsd1.in.comcast.net) Quit (kubrick.freenode.net irc.freenode.net)
  • [09:16:50] * ZeZu (n=null@c-98-227-56-17.hsd1.in.comcast.net) has joined #beagle
  • [09:19:55] * sh1 (i=3d5fc3cd@gateway/web/ajax/mibbit.com/x-14da2dfbcadbdbfa) has joined #beagle
  • [09:20:24] <mru> morning
  • [09:24:48] * sh1 (i=3d5fc3cd@gateway/web/ajax/mibbit.com/x-14da2dfbcadbdbfa) Quit ("http://www.mibbit.com ajax IRC Client")
  • [09:28:28] * _AV500_ (n=AV500@sac91-1-82-232-88-182.fbx.proxad.net) has joined #beagle
  • [09:29:29] * and-ri (n=ubuntu@pd95b4498.dip0.t-ipconnect.de) has joined #beagle
  • [09:34:07] * MostAwesomeDude (n=simpson@c-24-21-147-156.hsd1.or.comcast.net) Quit (Read error: 104 (Connection reset by peer))
  • [09:34:36] * MostAwesomeDude (n=simpson@c-24-21-147-156.hsd1.or.comcast.net) has joined #beagle
  • [09:37:48] * sh1 (i=3d5fc3cd@gateway/web/ajax/mibbit.com/x-f52aececbf3f3112) has joined #beagle
  • [09:37:58] * TAKI (n=Administ@dslb-088-072-249-153.pools.arcor-ip.net) has joined #beagle
  • [09:38:18] * abitos (n=nixgibts@dslb-084-057-155-043.pools.arcor-ip.net) has joined #beagle
  • [09:42:45] * Psychiatras (i=1000@212.122.84.232) Quit ("leaving")
  • [09:43:06] * Psychiatras (i=1000@212.122.84.232) has joined #beagle
  • [09:45:34] <Strontium> I am reading the code in dss_init() in u-boot-revb source, and I gotta say its wierd.
  • [09:46:04] <Strontium> as far as I can tell it sets a couple of IO to outputs and asserts them, and those pins are not connected to anything.
  • [09:46:09] <Strontium> ????
  • [09:46:31] <Strontium> Specifically GPIO 5,8 and 24
  • [09:47:05] <Strontium> Actually 24 is USB, but 5 and 8 are not connected to any logic
  • [09:50:31] * sh1 (i=3d5fc3cd@gateway/web/ajax/mibbit.com/x-f52aececbf3f3112) Quit ("http://www.mibbit.com ajax IRC Client")
  • [09:54:14] * rupeshgujare (n=rupesh@59.160.172.220) has joined #beagle
  • [09:56:00] * TAK2004 (n=Administ@dslb-088-072-249-153.pools.arcor-ip.net) Quit (Read error: 110 (Connection timed out))
  • [09:57:17] * ScriptRipper (n=martin@85.233.34.53.dynamic.cablesurf.de) has joined #beagle
  • [09:57:17] * florian_kc (n=fuchs@217.146.132.69) has joined #beagle
  • [09:59:48] * florian_kc is now known as florian
  • [10:00:30] <florian> good morning
  • [10:00:49] <koen> tomba: do you know if code to support the omapzoom exists for DSS2?
  • [10:00:50] <_AV500_> guten morgen
  • [10:00:51] <koen> florian: moin!
  • [10:01:45] <kulve> koen: do you have a zoom?
  • [10:02:00] <koen> not yet
  • [10:12:26] <recalcati> koan: doing bitbake console-image, when I fetch the kernel it stops, why?
  • [10:12:47] <recalcati> From git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6
  • [10:12:53] <recalcati> * branch master -> FETCH_HEAD
  • [10:13:31] <recalcati> koen: doing bitbake console-image, when I fetch the kernel it stops, why?
  • [10:19:59] <recalcati> koen: sorry, I deleted the linux dir and now it is compiling
  • [10:24:20] * ssvb (n=ssvb___@a88-114-221-132.elisa-laajakaista.fi) Quit ("Leaving")
  • [10:35:17] <tomba> koen: no, I haven't done anything for omapzoom
  • [11:04:01] * sh1 (i=3d5fc3cd@gateway/web/ajax/mibbit.com/x-039c8178febf8dba) has joined #beagle
  • [11:07:48] * moi1234 (n=moi@ip70-179-18-67.sd.sd.cox.net) Quit (Remote closed the connection)
  • [11:20:42] * nemequ (n=nemequ@ip68-111-215-155.sd.sd.cox.net) has joined #beagle
  • [11:38:52] * Leon_Nardella (n=leon@200-161-14-111.dsl.telesp.net.br) has joined #beagle
  • [11:57:22] * rsalveti (n=salveti@189.70.69.34) Quit (Remote closed the connection)
  • [12:00:36] * guillaum1 (n=GZBA4143@AMontsouris-153-1-34-86.w90-2.abo.wanadoo.fr) has joined #beagle
  • [12:12:59] * mib_g0v09p (i=3b93dac7@gateway/web/ajax/mibbit.com/x-6f327f92a99fab1a) has joined #beagle
  • [12:13:47] * abitos (n=nixgibts@dslb-084-057-155-043.pools.arcor-ip.net) Quit (Read error: 148 (No route to host))
  • [12:15:01] * phobbes (n=srussell@user-387ggqq.cable.mindspring.com) has joined #beagle
  • [12:22:47] * sh1 (i=3d5fc3cd@gateway/web/ajax/mibbit.com/x-039c8178febf8dba) Quit ("http://www.mibbit.com ajax IRC Client")
  • [12:28:10] <Strontium> Has anyone used USB booting? Is it normal that Holding the USR button to boot from USB only works at POR, and not if I press the reset button?
  • [12:30:04] * aleij (n=ad@62.161.153.118) has left #beagle
  • [12:35:45] * mib_g0v09p (i=3b93dac7@gateway/web/ajax/mibbit.com/x-6f327f92a99fab1a) Quit ("http://www.mibbit.com ajax IRC Client")
  • [12:38:28] * felipec (i=c0647cdb@gateway/web/ajax/mibbit.com/x-0863088d88aecf92) has joined #beagle
  • [12:40:35] <jkridner> good morning all
  • [12:41:32] <Crofton> gm
  • [12:42:26] <florian> hi jkridner
  • [12:42:35] * rsalveti (n=salveti@200.184.118.130) has joined #beagle
  • [12:42:37] * mru (n=mru@thrashbarg.mansr.com) Quit (kubrick.freenode.net irc.freenode.net)
  • [12:42:37] * Crofton|irssi (n=balister@ossie.wireless.vt.edu) Quit (kubrick.freenode.net irc.freenode.net)
  • [12:43:38] <koen> hey jkridner & Crofton
  • [12:44:11] <Crofton> hopefully, I have more time for kernel booting
  • [12:44:24] <Crofton> koen, where you booting the pm kernel with root on nfs?
  • [12:45:01] * mru (n=mru@thrashbarg.mansr.com) has joined #beagle
  • [12:45:01] * Crofton|irssi (n=balister@ossie.wireless.vt.edu) has joined #beagle
  • [12:46:22] <koen> Crofton: no, ubifs in nand or SD
  • [12:46:32] <koen> Crofton: I don;t use nfs, it's evil and breeds bad QA
  • [12:46:50] * koen uses packages to get changes into his rootfs
  • [12:47:25] <Crofton> hmmm, the build from yesterday was not detecting mmc for me
  • [12:48:45] <Crofton> ten minutes and starbucks should be open
  • [12:48:47] <koen> yes, that's also mentioned in my mails to l-o ml
  • [12:48:55] <koen> but todays build works with SD
  • [12:49:24] <Crofton> I think I'll want to try root on nand and see if that helps lower power
  • [12:51:23] * Crofton wonders why the starbucks website does not list store hours
  • [12:52:45] <koen> because they are franchises
  • [12:53:11] <koen> the subway in wolverhampton behind the hotel was open till 2 am on some days
  • [12:53:36] <Crofton> I'm assuming it opens at 0600
  • [12:56:52] * rupeshgujare (n=rupesh@59.160.172.220) has left #beagle
  • [12:57:46] * cbrake_away is now known as cbrake
  • [12:58:55] <jkridner> koen: does today's build include EHCI, DSS2, LEDs, OTG, and SGX on 2.6.28?
  • [13:00:55] <jkridner> koen: is it possible to build mplayer without X11?
  • [13:00:56] <koen> ECHI, DSS2, OTG and PM
  • [13:01:08] <koen> both EHCI and OTG don't work as expected
  • [13:01:31] <koen> jkridner: it's possible to build it without X11, but I fail to see the point of that
  • [13:01:45] <koen> jkridner: the extra 600kb it saves in the rootfs is neglible
  • [13:02:06] <jkridner> yeah, the PM build I looked at from you didn't work for EHCI for me.
  • [13:02:34] <jkridner> I'm looking to add it to the Beagle validation test, instead of the hack we have in there.
  • [13:02:50] <jkridner> I was looking for something that could be run from console mode.
  • [13:03:04] <koen> you can run the current mplayer from console as well
  • [13:03:22] <koen> mplayer is smart enough to detect whether DISPLAY is set or not
  • [13:05:21] * BThompson (n=BThompso@nat/ti/x-bfd6e83a42b2d41d) has joined #beagle
  • [13:05:26] <jkridner> right, but when I try to build it from OE, it automatically adds X.
  • [13:05:36] <jkridner> I guess I can live with that, but my X failed to build yesterday.
  • [13:05:52] <koen> Crofton: I wish we had starbucks over here (or something resembling it)
  • [13:06:05] * koen loves balzac in berlin
  • [13:06:08] <Crofton> Balzac
  • [13:06:11] <Crofton> heh
  • [13:06:52] <Crofton> Balzac is german for starbucks :)
  • [13:07:26] <koen> I keep saying they should close the "unionbar" at the uni and put a starbucks there
  • [13:08:03] <koen> or something with decent coffee in sizes larger than "coffeecup"
  • [13:08:16] <Crofton> hmm
  • [13:08:27] <Crofton> you don't even have club mate in .nl
  • [13:08:47] <koen> the 750ml starbucks gets you (venti?) is a bit overkill, but nice to have
  • [13:10:31] * koen gets grumpy without coffee in the morning
  • [13:11:01] * garren|work (n=chatzill@mail.dm.co.za) Quit (Read error: 113 (No route to host))
  • [13:13:51] <koen> jkridner: change 'enable-x11' to 'disable-x11' in the mplayer recipe if you don't want x11 support
  • [13:14:06] <koen> jkridner: same goes for other options you want to turn off :)
  • [13:15:21] <jkridner> thanks.
  • [13:16:02] <koen> Crofton: do you have a non-beagle omap3 platform at hand?
  • [13:16:15] <koen> Crofton: I suspect something with ethernet and LCD would be better for PM testing
  • [13:16:32] <koen> since both USB and serial disable anything beyond C1
  • [13:16:40] * Leon_Nardella (n=leon@200-161-14-111.dsl.telesp.net.br) Quit (Read error: 110 (Connection timed out))
  • [13:16:54] <Crofton> there is an evm
  • [13:16:58] <Crofton> I can kill usb :)
  • [13:17:17] <Crofton> we'll see how desperate we get
  • [13:17:34] <koen> try the pm kernel on the evm
  • [13:18:10] <koen> (you need to patch in the rate tables, see beagle-cpufreq.diff, but ignore the gpio stuff, that's for EHCI)
  • [13:19:09] <Crofton> this is the back up plan
  • [13:19:14] * garren|work (n=chatzill@mail.dm.co.za) has joined #beagle
  • [13:20:22] * garren|work (n=chatzill@mail.dm.co.za) Quit (Client Quit)
  • [13:27:30] * phobbes1 (n=srussell@user-387ggve.cable.mindspring.com) has joined #beagle
  • [13:28:56] * garren|work (n=chatzill@mail.dm.co.za) has joined #beagle
  • [13:32:14] * ddompe (n=ddompe@12.108.221.91) has joined #beagle
  • [13:33:15] * phobbes (n=srussell@user-387ggqq.cable.mindspring.com) Quit (Read error: 60 (Operation timed out))
  • [13:44:01] * ddompe_ (n=ddompe@12.108.221.83) has joined #beagle
  • [13:45:47] * Septalicia (n=bakljg@c-98-240-226-129.hsd1.mn.comcast.net) Quit (Read error: 110 (Connection timed out))
  • [13:47:51] * zedstar (n=john@fsf/member/zedstar) Quit (Remote closed the connection)
  • [13:50:53] * [X]Spot (n=stancho@78.90.99.168) Quit (Read error: 110 (Connection timed out))
  • [13:52:02] <jkridner> koen: did you need any more help to get DMAI going?
  • [13:52:37] <jkridner> or anyone else for that matter.... doesn't need to be koen doing all the integration into a Linux distro. :)
  • [13:52:37] * docelic (n=docelic@78.134.195.9) Quit (Read error: 110 (Connection timed out))
  • [13:53:25] * docelic (n=docelic@78.134.201.183) has joined #beagle
  • [13:53:33] <Crofton> Destination Marketing Association International?
  • [13:53:35] <koen> jkridner: yes
  • [13:53:46] <koen> davinci media a interface
  • [13:53:56] <Crofton> ah
  • [13:54:07] <Crofton> adding omap to the google helps
  • [13:54:45] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [13:56:03] * ddompe (n=ddompe@12.108.221.91) Quit (Read error: 110 (Connection timed out))
  • [13:57:29] <jkridner> discoverlelgo: if you are getting zImage, but not uImage, the issue is that you don't have mkimage in your path or didn't type 'make uImage'.
  • [13:58:36] <jkridner> er, discoverlego
  • [14:00:22] * guillaum1 (n=GZBA4143@AMontsouris-153-1-34-86.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
  • [14:00:39] <jkridner> discoverlego: are you getting these messages? I don't like to answer support questions privately.
  • [14:00:58] * guillaum1 (n=gl@AMontsouris-153-1-34-86.w90-2.abo.wanadoo.fr) has joined #beagle
  • [14:01:10] <discoverlego> oh sorry really very sorry jkridner...
  • [14:01:19] <koen> jkridner: http://article.gmane.org/gmane.linux.ports.arm.omap/15683
  • [14:01:49] <discoverlego> jkridner: so now what is the solution to this problem....??? how should i get uImage now...
  • [14:02:04] <jkridner> koen: yeah, that was sent to me a few times this morning.
  • [14:02:12] <koen> :)
  • [14:02:15] <jkridner> discoverlego: do you have mkimage in your path?
  • [14:02:51] <discoverlego> jkridner:no i dont have..so how should i go ahead
  • [14:02:55] <koen> jkridner: it boils down to telling all engineers to work against mainline linux and push patches the regular way
  • [14:03:00] * hartangas (i=83cfcfc0@gateway/web/ajax/mibbit.com/x-d65bbc95bbc8cb9d) Quit ("http://www.mibbit.com ajax IRC Client")
  • [14:03:18] <jkridner> you need to build mkimage with u-boot.
  • [14:03:30] <jkridner> did you build u-boot?
  • [14:03:56] <koen> discoverlego: you'd have less problems if you use OE :)
  • [14:04:29] <discoverlego> jkridner:ya i did even that works fine with the prebuilt images from the site
  • [14:05:02] * ddompe_ (n=ddompe@12.108.221.83) Quit ()
  • [14:05:41] <discoverlego> jkridner: so how should i go there....???
  • [14:06:07] <discoverlego> koen: OE..???:(
  • [14:06:29] * Leon_Nardella (n=leon@200-161-14-111.dsl.telesp.net.br) has joined #beagle
  • [14:07:55] * sh1 (i=3d5fc3cd@gateway/web/ajax/mibbit.com/x-c6a54b3be191d384) has joined #beagle
  • [14:08:08] <jkridner> it should be under the u-boot/tools folder.
  • [14:08:18] <jkridner> discoverlego: you need to put that file in your path.
  • [14:10:29] <jkridner> discoverlogo: do you have u-boot/tools/mkimage?
  • [14:11:12] <discoverlego> yes i have
  • [14:11:32] * mib_dqu3qr (i=74e162a0@gateway/web/ajax/mibbit.com/x-01db9de8566c1b8f) has joined #beagle
  • [14:11:52] * hugsig (n=hsi@2001:16d8:ff1a:0:216:3eff:fe5a:59cd) Quit (Read error: 60 (Operation timed out))
  • [14:12:02] <jkridner> discoverlogo: is it in your path?
  • [14:12:03] <mib_dqu3qr> hello
  • [14:12:04] <jkridner> er, discoverlego.
  • [14:12:25] * mib_dqu3qr (i=74e162a0@gateway/web/ajax/mibbit.com/x-01db9de8566c1b8f) Quit (Client Quit)
  • [14:12:33] <discoverlego> no...i dont have...
  • [14:12:41] * hugsig (n=hsi@2001:16d8:ff1a:0:216:3eff:fe5a:59cd) has joined #beagle
  • [14:14:18] <discoverlego> jkridner: then how should i solve this problem and get uImage...??
  • [14:14:55] <jkridner> discoverlego: put it in your path.
  • [14:15:09] <jkridner> discoverlego: echo $PATH
  • [14:15:36] <jkridner> discoverlego: export PATH=/path/to/u-boot/tools:$PATH
  • [14:16:10] <mru> or copy it to /usr/local/bin/ or similar
  • [14:16:34] <jkridner> assuming you have root permissions.
  • [14:16:52] <discoverlego> ok will try....and will let u know thanks for the help
  • [14:17:55] * newton (i=812ad0b1@gateway/web/ajax/mibbit.com/x-edc5dbd672133707) has joined #beagle
  • [14:23:19] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit ()
  • [14:26:24] * alecrim (n=alecrim@189.2.128.130) has joined #beagle
  • [14:28:16] * simon42 (n=simon@89.238.65.10) Quit ("leaving")
  • [14:29:49] * dcramer (n=davec@dcdsl.ebox.com) Quit ()
  • [14:33:21] * allen (i=574b973e@gateway/web/ajax/mibbit.com/x-dfc489200989e164) has joined #beagle
  • [14:33:55] <allen> hi everyone, allen from salford university in uk
  • [14:34:37] * simon__ (n=simon@89.238.65.10) has joined #beagle
  • [14:34:49] * simon__ is now known as simon42
  • [14:35:15] <mru> hi allen
  • [14:35:42] <allen> Hi mru, just a quick one: does anyone have experience with compiling computer vision code on the beagle? have a ti dev kit coming in on the 23rd for uni use but would love the beagle for my own :-)
  • [14:36:06] <mru> what sort of code?
  • [14:36:29] <mru> any decently written C code should compile and run fine
  • [14:36:34] <allen> mainly C++
  • [14:36:39] <mru> or fortran for that matter
  • [14:36:56] <Strontium> Has anyone used USB booting? Is it normal that Holding the USR button to boot from USB only works at POR, and not if I press the reset button?
  • [14:37:03] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [14:37:17] <mru> if your code works with gcc/glibc on a linux pc, it should work on the beagle
  • [14:37:51] <allen> lovely what compiler is used? TI specific? brilliant I use eclipse ide and the gcc :-)
  • [14:38:03] <mru> most people use gcc
  • [14:38:14] <mru> you can use the ARM compilers too
  • [14:38:35] <mru> the TI compiler is trickier
  • [14:38:39] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit (Client Quit)
  • [14:39:08] <mru> get the codesourcery gcc package
  • [14:39:13] <mru> version 2007q3
  • [14:39:32] <allen> thanks mru i was hoping not to have to use the ti one, not sure if that gcc will work for the davini ive got coming but thats a diff story altogether!
  • [14:40:00] <Strontium> mru why the codesourcery package? Does latest GCC mainline not have some necessary patches (if I build it myself)?
  • [14:40:20] * emeb (n=ericb@ip72-223-90-212.ph.ph.cox.net) has joined #beagle
  • [14:40:21] <mru> allen: for the dsp you need TI compilers
  • [14:40:39] <mru> Strontium: mainline gcc 4.3.2 works but the csl versions are better
  • [14:40:52] <mru> 2007q3 is generally the best
  • [14:41:04] <mru> 2008q3 has more bugs and produces slower code in many cases
  • [14:41:19] <allen> i see! i knew bad news was coming! but i suppose if you want the performance gains...
  • [14:41:47] <_AV500_> allen: gcc works fine for the davinci
  • [14:42:46] <_AV500_> but we do not use mainline, we use a mvista one
  • [14:43:01] <mru> gcc can't compile for the c64x
  • [14:43:11] <allen> thanks av500 have you had much experience?
  • [14:43:27] <mru> gcc of course works fine for the arm9 in the davinci
  • [14:44:22] <allen> so it looks like ill have no problem with the arm but just need to use ti's compiler if i do anything fancy
  • [14:44:38] <koen> /dev/scd0 on /media/dvd type udf (ro,utf8)
  • [14:44:39] <mru> the cortex-a8 is quite powerful
  • [14:44:46] <koen> sweet, dvd on my beagle
  • [14:45:16] <mru> my beagle is *still* in memphis
  • [14:46:32] <kulve> mru: how many times have you verified that? :)
  • [14:46:52] <muriani> koen: dvd playback on beagle?
  • [14:46:56] <muriani> sweet
  • [14:46:59] <_AV500_> allen: TI for the DSP part, but it is free for noncommercial use
  • [14:47:00] <_AV500_> or so
  • [14:47:17] <allen> thanks mru i guess a8 looks good shame ive only got a budget for one dev board
  • [14:48:02] * felipec (i=c0647cdb@gateway/web/ajax/mibbit.com/x-0863088d88aecf92) Quit ("http://www.mibbit.com ajax IRC Client")
  • [14:48:19] <mru> beagles are cheap
  • [14:48:31] <allen> av500 cool im looking forward to completely overhalling my coding for a dsp :-/
  • [14:51:13] * royerfa (n=fabroy01@fw-tnat.cambridge.arm.com) has joined #beagle
  • [14:52:22] * mckoan|away is now known as mckoan
  • [14:52:31] * felipec (i=c0647cda@gateway/web/ajax/mibbit.com/x-a567202a511023d8) has joined #beagle
  • [14:53:23] <allen> hey mru av500 and everyone else just wanted to thank you guys for all your help, maybe see you again next time im on. Many thanks again
  • [14:53:38] <Strontium> Anyone know where I can get a PDF of the Cortex A8 architecture reference, I can;t find it on ARMs site??\
  • [14:53:47] * allen (i=574b973e@gateway/web/ajax/mibbit.com/x-dfc489200989e164) Quit ("http://www.mibbit.com ajax IRC Client")
  • [14:54:34] * discoverlego (i=d2124e02@gateway/web/ajax/mibbit.com/x-701d5aff934b2e65) Quit ("http://www.mibbit.com ajax IRC Client")
  • [14:56:21] <kulve> Strontium: isn't a .html enough?
  • [14:57:05] * Stskeeps (n=chobits@84.238.11.16) Quit ("Reconnecting")
  • [14:57:30] * leslie_ (n=leslie@116.233.164.37) has joined #beagle
  • [14:57:34] <atin_> jkridner: ping?
  • [14:57:40] <jkridner> hi
  • [14:57:53] <atin_> jkridner: hey - take a look at this: http://pastebin.com/m13354dec
  • [14:57:54] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [14:58:19] <atin_> khasim too I guess. that;s the output from my lsusb -v
  • [14:58:34] <atin_> with diego and my patches
  • [14:58:43] <jkridner> guess enumeration looks good.
  • [14:58:59] <atin_> netchip technology?
  • [14:59:08] <Strontium> kulve, you have a html file of the Cortex A8 architecture reference manual?
  • [14:59:09] <atin_> that's what we have?
  • [14:59:13] <jkridner> they allow use of their VID.
  • [14:59:23] <khasim> sorry I joined late,
  • [14:59:29] <khasim> anything for me to confirm
  • [14:59:33] <atin_> http://pastebin.com/m13354dec
  • [14:59:40] <jkridner> at least that is what the Linux documentation says. Linux is free to use their VID.
  • [14:59:41] <atin_> that's the output from my lsusb -v
  • [14:59:45] <atin_> I see.
  • [14:59:49] <kulve> Strontium: http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0344h/index.html
  • [14:59:54] <jkridner> not sure if Netchip really agrees, but it seems to be common practice.
  • [15:00:01] <kulve> that looks like something you are searching for?
  • [15:00:35] <atin_> so I guess the setup code is actually doing something right. I am not sure what my next step should be to confirm this is actually working! :)
  • [15:02:22] <Strontium> kulve, if it wasnt so painfully slow and cumbersome maybe. I opened the link when you posted it, and i am still waiting for the page to load.....
  • [15:02:24] * maelcum (n=quassel@e178144176.adsl.alicedsl.de) Quit (Read error: 110 (Connection timed out))
  • [15:03:48] <jkridner> atin_: have you sent patches to steve?
  • [15:03:55] <jkridner> er, sakoman.
  • [15:03:57] <kulve> Strontium: they opens in two seconds on my home computer..
  • [15:04:11] <atin_> jkridner: not yet, he just has diego's patches
  • [15:04:14] <kulve> well, maybe in 5
  • [15:04:25] <atin_> I wanted to make sure it was doing something first :)
  • [15:04:27] <jkridner> k. do you have /dev/tty.usbmodem000...1?
  • [15:04:38] <atin_> no, but I am on my linux box again.
  • [15:04:50] * maelcum (n=quassel@e178162150.adsl.alicedsl.de) has joined #beagle
  • [15:04:55] <jkridner> well, it is nice to have enumeration without running the ROM USB bootloader, which I believe is what you have done, right?
  • [15:05:05] * Stskeeps (n=chobits@84.238.11.16) has joined #beagle
  • [15:05:13] <atin_> jkridner: right, I just plugged in the usb cable
  • [15:05:31] * JuanG (n=Juan@nat/ti/x-7fe7dbecc8b5a250) has joined #beagle
  • [15:05:37] * JuanG (n=Juan@nat/ti/x-7fe7dbecc8b5a250) has left #beagle
  • [15:06:40] <kulve> Strontium: there's a "PDF version" link at the bottom: http://infocenter.arm.com/help/topic/com.arm.doc.ddi0344h/DDI0344H_cortex_a8_r3p0_trm.pdf
  • [15:07:08] <atin_> jkridner: I'm reading over http://www.linux-usb.org/usbtest/ to see what more I can do before I am convinced this is good.
  • [15:07:49] <khasim> jkridner: I got your rootfs, not tested yet, does this have
  • [15:07:49] <khasim> ALSA and mplayer in it?
  • [15:09:02] <jkridner> it has ALSA, but not mplayer yet.
  • [15:10:19] <jkridner> khasim: this was just an experiment for me. I suggest we put the problem to the community.
  • [15:11:06] <khasim> jkridner: ok, will do that now. ..
  • [15:12:59] * dirk2 (n=dirk@p5B043684.dip0.t-ipconnect.de) has joined #beagle
  • [15:13:17] <jkridner> dirk2: it looks like sakoman applied my pinmux changes.
  • [15:13:52] <jkridner> dirk2: I was going to provide an update based on the feedback, but I was waiting on being given the "go".
  • [15:14:38] <jkridner> dirk2: oh, wait, I was wrong. I was looking at the wrong log. looks like sakoman has not applied it yet.
  • [15:16:01] <sakoman_> jkridner: I was waiting for an update - from the messages it sounded like one was coming
  • [15:16:08] <jkridner> dirk2, khasim, sakoman: I have one conflict left with GPIO172. Do we know if that should be GPIO or SPI1?
  • [15:16:53] <jkridner> sakoman_: I just did a rebase. I can generate a new patch as soon as I resolve this one conflict and built/boot.
  • [15:17:06] <sakoman_> jkridner: depends on how the signal gets used on rev C - don't have a schematic :-)
  • [15:17:13] * leslie_ (n=leslie@116.233.164.37) Quit (Remote closed the connection)
  • [15:17:53] * __alanc__ (n=a-campbe@nat/ti/x-632c874436da3d48) has joined #beagle
  • [15:17:53] * Crofton (n=balister@63.224.149.101) Quit ("Leaving")
  • [15:18:00] <khasim> I think only GPIO_171 is sufficient to detect REV B/C
  • [15:18:04] <dirk2> jkridner: GPIO171 and GPIO 172 are for board detection
  • [15:18:06] <jkridner> k. I don't have one handy either. khasim, is it easy for you to check?
  • [15:18:20] <jkridner> oh, well, I think we want to reserve that one as GPIO.
  • [15:18:28] * jkridner was thinking it was 170 and 171.
  • [15:18:35] <sakoman_> ah, from Gerald's comment he want to use it in the future for rev detection
  • [15:18:42] <dirk2> jkridner: let me see.
  • [15:18:43] <khasim> yes
  • [15:18:43] <jkridner> if 172 is one of the detection pins, I think it should be GPIO and pull-up.
  • [15:18:53] <dirk2> sakoman: yes, we should take both
  • [15:18:54] <sakoman_> agreed
  • [15:19:09] <khasim> jkridner: it should be good
  • [15:19:13] <newton> Hello everybody, I'm thinking of buying a beagleboard rev C1 when it comes out. Does anyone know where to get a reasonably priced LCD that would work with the omap? I was thinking of extracting an old LCD display from an old laptop I have around.
  • [15:19:26] <khasim> jkridner : we should configure it for GPIO172
  • [15:19:48] <sakoman_> did we get agreement on my 2 questions regarding pullup/down on 2 EHCI signals?
  • [15:20:01] <dirk2> jkridner: GPIO171 and GPIO172 as input pull up enable. GPIO171 is already in sakoman's git, so you only have to touch 172
  • [15:20:16] <jkridner> k.
  • [15:20:32] <sakoman_> I think we did on the data line, the control line doesn't seem to have a response
  • [15:20:44] <dirk2> jkridner: did you check the other comments from yesterday, too?
  • [15:21:02] * jrmuizel (n=jrmuizel@mozca02.ca.mozilla.com) has joined #beagle
  • [15:21:07] <jkridner> I looked at yours and modified the patch based on those.
  • [15:21:17] <jkridner> if there have been more comments, I haven't seen them yet.
  • [15:21:20] * jkridner checks mail.
  • [15:21:56] <dirk2> jkridner: GPIO_183 and MMC3_CLK changes are unclear for me
  • [15:21:58] * Strontium (n=Strontiu@118.101.106.2) Quit (Connection timed out)
  • [15:22:29] * Strontium (n=Strontiu@118.101.106.2) has joined #beagle
  • [15:22:53] <dirk2> jkridner: note that you have GPIO172 as pull down in your patch from yesterday, shall be pull up
  • [15:23:12] <jkridner> this patch will have it as a pull-up.
  • [15:24:08] <dirk2> sakoman: If we have jkridner pin mux changes in, I will update beagle board detection to use both pins again ;)
  • [15:24:24] <sakoman_> dirk2: sounds good
  • [15:24:26] <Strontium> kulve, it finally loaded. no thats the Technical Reference Manual, I am after the "ARM Architecture Reference Manual" that describes the "ARMv7-A architecture" thats inside the Cortex A8.
  • [15:24:46] <mru> Strontium: you have to ask ARM for that
  • [15:25:03] <mru> http://infocenter.arm.com/
  • [15:25:23] <mru> you'll find a link to a request form there
  • [15:25:55] <Strontium> mru, I did, they dont respond
  • [15:25:55] <dirk2> everybody interested in having OMAP3 U-Boot v1 patches in mainline will wait for an answer to http://lists.denx.de/pipermail/u-boot/2009-January/045580.html
  • [15:27:05] <atin_> lol!
  • [15:27:08] <newton> Sorry if I'm posting this LCD question in the wrong place. Is there a better place to be asking these sorts of questions?
  • [15:27:17] <atin_> "we don't like omap3"
  • [15:28:00] <dirk2> atin_: I fear we will get some sort of this as answer ;)
  • [15:28:26] <sakoman_> dirk2: that is too funny!
  • [15:28:49] <mru> Strontium: keep pestering them
  • [15:28:54] <sakoman_> my prediction is that there will be no response :-)
  • [15:29:54] <dirk2> jkridner should donate a beagle board for predicting the answer :)
  • [15:30:36] <jkridner> jkridner would only donate a beagle if the patches actually land upstream.
  • [15:32:05] <sakoman_> newton: I think that no one is answering because rev c isn't out, so this is uncharted teritory
  • [15:33:19] <sakoman_> I've gotten a 4.3" LCD & TS working on an overo, so I know it shouldn't be too hard
  • [15:33:40] <sakoman_> and the 35xx evm ships with an lcd/ts
  • [15:35:26] <jkridner> is there a way to rename a remote branch?
  • [15:35:28] * sh1 (i=3d5fc3cd@gateway/web/ajax/mibbit.com/x-c6a54b3be191d384) has left #beagle
  • [15:35:42] <jkridner> I cannot just 'push' a rebased branch.
  • [15:36:56] * tzhau_ (n=tzhau@dsl-245-126-145.telkomadsl.co.za) has joined #beagle
  • [15:38:15] <jkridner> *** Deleting a branch is not allowed in this repository :(
  • [15:38:19] <Crofton|irssi> Strontium: that's funny, my request was approved in a few hours
  • [15:39:16] <mru> maybe Crofton|irssi had a better motivation
  • [15:40:14] <Crofton|irssi> tell them you need it for some cool open source work
  • [15:40:18] <Crofton|irssi> that worked for me
  • [15:40:24] <Crofton|irssi> switching machines
  • [15:40:27] * Crofton|irssi (n=balister@ossie.wireless.vt.edu) Quit ("leaving")
  • [15:40:27] <khasim> sakoman_: do you know why overo's are not shipped to India
  • [15:40:57] * Crofton|irssi (n=balister@pool-96-240-173-146.ronkva.east.verizon.net) has joined #beagle
  • [15:41:15] * phobbes (n=srussell@user-387ggve.cable.mindspring.com) has joined #beagle
  • [15:41:19] * phobbes1 (n=srussell@user-387ggve.cable.mindspring.com) Quit ("Leaving.")
  • [15:41:25] <sakoman_> khasim: I have no clue! You might ask sales@gumstix.com
  • [15:42:22] <sakoman_> Tell them about all the sales they are losing out on!
  • [15:42:59] <Strontium> Crofton|irssi, they didnt like the fact that my companies web site doesnt have nice adds about bunches of products and the like.
  • [15:43:00] <koen> jkridner: push --force ?
  • [15:43:05] * Leon_Nardella1 (n=leon@200-161-14-111.dsl.telesp.net.br) has joined #beagle
  • [15:43:13] <Crofton|irssi> weird
  • [15:43:22] <Strontium> Crofton|irssi, tell me about it
  • [15:43:47] <jkridner> koen: that worked.
  • [15:44:41] <jkridner> dirk2: I'm currently testing http://www.beagleboard.org/gitweb/?p=u-boot-arm.git;a=commitdiff;h=8301f10ca39120020c225defabce9236b88529c5 and will push when tested.
  • [15:44:50] <dirk2> khasim: regarding your mail: Could you give link to your U-Boot git once again?
  • [15:45:28] <jkridner> dirk2: http://gitorious.org/projects/beagleboard-default-u-boot
  • [15:45:30] <koen> jkridner: a sanely configured git repo will disallow non-fastforward merges, unless you give the --force flag
  • [15:45:47] <koen> jkridner: but people pulling your branch will need to to git pull --force as well
  • [15:45:52] <jkridner> koen: that makes good sense.
  • [15:46:20] <jkridner> I really wanted to rename it, not so much delete it.
  • [15:47:54] <dirk2> jkridner: pin mux patch above: looks fine :)
  • [15:48:24] <jkridner> I've also got that up on gitorious http://gitorious.org/projects/beagleboard-default-u-boot/repos/jason-clone/commits/8301f10ca39120020c225defabce9236b88529c5
  • [15:48:58] <jkridner> not sure if any of my other patches need/can be pushed.
  • [15:49:21] <jkridner> build completed. now putting onto SD card.
  • [15:49:26] * robtow (n=rtow@64.62.142.114) has joined #beagle
  • [15:51:17] <khasim> sakoman_: thanks, will ask our sales to get in touch with them
  • [15:52:05] <khasim> koen: you have good web based rootfs generating tool, can we add ALSA and mplayer to it? I need one for validation
  • [15:52:32] <koen> mplayer is already there
  • [15:52:36] * koen adds alsa stuff
  • [15:53:02] <khasim> shall I generate one for beagle now?
  • [15:54:23] <khasim> koen: let me know when I can generate one. I will use the image for board validation
  • [15:54:34] <khasim> with latest 2.6.28 kernel
  • [15:55:21] * rob (i=4a5dd0bb@gateway/web/ajax/mibbit.com/x-f4e1c615c7e4ae74) has joined #beagle
  • [15:55:53] * rob is now known as Guest96974
  • [15:55:55] <jkridner> khasim, koen: we'll want to be sure we can keep a stable copy of the OE repo and all the sources used by the image.
  • [15:56:39] <Guest96974> hey guys, i have a problem formatiing my sd card for use with angstrom! and now i can't get the terminal to show the number of bytes of my card!
  • [15:56:48] <khasim> jkridner: I dont know if we can update a snapshot of the source
  • [15:57:07] <khasim> download/upload snapshot of the source
  • [15:57:20] <Crofton|irssi> jkridner: git branch validated-sw :)
  • [15:57:31] <jkridner> khasim: arago does this.
  • [15:57:52] <jkridner> I know how to do this. it is very straight-forward with OE/Angstrom.
  • [15:58:17] <jkridner> I'm just saying it needs to be done, in addition to pulling down a build from Narc... (can't spell it).
  • [15:58:29] <khasim> jkridner: yest
  • [15:58:32] <khasim> yes
  • [15:58:40] <newton> sakoman_: Thanks for the info. I'm assuming that most people just hook their beagleboards to their TV/computer monitor until c1 comes out?
  • [15:59:04] * phobbes (n=srussell@user-387ggve.cable.mindspring.com) Quit (Read error: 110 (Connection timed out))
  • [15:59:12] * Leon_Nardella (n=leon@200-161-14-111.dsl.telesp.net.br) Quit (Read error: 110 (Connection timed out))
  • [15:59:14] <jkridner> Guest96974: how did you try to format it?
  • [15:59:29] <jkridner> FAT and ext3 should work fine, along with several other formats.
  • [16:00:05] <khasim> jkridner: FYI, I have a backup plan with 2.6.22 for Gerald, now focusing on 2.6.28 migration....
  • [16:00:31] <sakoman_> newton: yes, or just use a console session over the serial port
  • [16:00:35] <jkridner> khasim: did you see the IRC logs about mplayer w/o X11?
  • [16:00:37] <khasim> only missing stuff is Video rendering app based on FBDEV
  • [16:00:38] * bazbell (n=a0192809@nat/ti/x-0f30739325024373) has joined #beagle
  • [16:01:13] <koen> khasim, jkridner: http://amethyst.openembedded.net/~koen/narcissus/
  • [16:01:18] <jkridner> khasim: or is adding X11 fine for the validation image?
  • [16:01:21] <garren|work> cheers all I'm off home
  • [16:01:38] * garren|work (n=chatzill@mail.dm.co.za) Quit ("ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]")
  • [16:01:44] <khasim> jkridner: only requirement from Gerald is time taken to boot and test should be minimal
  • [16:02:33] <newton> satkoman_: Do you think that posting a message to the ti omap forum might yield more information about LCD stuff? The rev c1 of the beagleboard is just going to be exposing the LCD lines from the omap, right?
  • [16:02:33] <khasim> jkridner: we should have a video than a static image to figure out any typical display issues
  • [16:03:16] <jkridner> khasim: do we need to exclude X11 for boot performance reasons?
  • [16:04:04] <sakoman_> newton: Worth a try
  • [16:04:31] <sakoman_> khasim: the gumstix production folks use an x11 based image for validation
  • [16:05:06] <newton> sakoman_: Will do. Thanks again.
  • [16:05:13] * Crofton|irssi builds todays pm kernel
  • [16:05:38] <khasim> jkridner: do we have a sample? I can pass one to Gerald and get his input/suggestion
  • [16:06:57] <jkridner> I believe the best way to get a sample to try is http://amethyst.openembedded.net/~koen/narcissus/. Then, we can finalize with an Angstrom overlay and our own repo branch and downloads directory (that can be public).
  • [16:07:19] <khasim> koen: Can I just select ALSA, mplayer and I2C tools - any dependencies ..?
  • [16:07:20] * JustinLove (n=J_LOVE@64-135-210-33.FoxValley.net) has joined #beagle
  • [16:07:55] <khasim> jkridner: but x11 should also work on S-Video...
  • [16:07:58] <koen> khasim: dependencies are handled automatically
  • [16:08:07] <khasim> koen: thanks
  • [16:08:40] <khasim> jkridner: we need to do similar video rendering test on S-Video.
  • [16:08:48] <khasim> the video file should be very minimal
  • [16:09:01] <khasim> will try koen's image ...
  • [16:10:06] * gregoiregentil (n=zonbu@adsl-71-135-114-242.dsl.pltn13.pacbell.net) has left #beagle
  • [16:10:15] <jkridner> khasim: rendering video on S-video is trivial. you just need to switch output to S-video.
  • [16:11:29] <atin_> jkridner: do you want the patches too? I just sent them to steve and dirk and to the beagle mailing list
  • [16:11:37] <khasim> jkridner: I am wondering if it needs to switch complete screen from DVI to SVideo (LCD to Digital), then the resolution switch must also be handled I dont know how good is the dynamic switch now
  • [16:11:49] <atin_> jkridner: and diego of course
  • [16:12:12] <jkridner> khasim: it is possible to create an additional framebuffer for output.
  • [16:12:24] <jkridner> so, you don't need to have output only on S-video.
  • [16:13:02] * koen would have thought playing a dvd in linux would be a solved problem
  • [16:13:15] * gregoiregentil (n=zonbu@adsl-71-135-114-242.dsl.pltn13.pacbell.net) has joined #beagle
  • [16:13:54] <khasim> jkridner: I feel Mplayer option is good, we can switch between the screens; with single fb. will check and get back
  • [16:14:18] * _don_ (n=don@nat/ti/x-b15a593313bb92ca) has joined #beagle
  • [16:14:27] * khasim thinks 2.6.28 +OE patches is good for Validation kernel
  • [16:14:35] <khasim> any other opinion
  • [16:14:55] <mru> koen: playing dvds in linux is a solved problem
  • [16:15:19] <DJWillis> khasim: seems a very sane point for a validation kernel.
  • [16:15:21] <mru> you might have to write some device drivers of course ;-)
  • [16:15:43] <koen> mru: If i every buy and alpha I'll give you a call :)
  • [16:15:44] * mru had to do that to get full frame rate
  • [16:15:48] <koen> s/and/an/
  • [16:16:09] <khasim> where should be base of from? Sakoman or kernel + OE patches from koen or linux omap + few patches
  • [16:16:32] <khasim> DJWillis: we can take a snap shot
  • [16:16:51] <jkridner> khasim: I believe Arago is where we should base.
  • [16:17:48] <khasim> jkridner: you mean build validation kernel as a arago project? or take a snapshot from arago?
  • [16:17:55] * Crofton|irssi suspects there are issues with arago ....
  • [16:18:06] * khasim too
  • [16:18:07] <DJWillis> khasim: have TI ever considered a validation GIT? i.e. the version used for validation (massive TI hacks and all ;-)) in it's own tree?
  • [16:18:20] * Guest96974 (i=4a5dd0bb@gateway/web/ajax/mibbit.com/x-f4e1c615c7e4ae74) Quit ("http://www.mibbit.com ajax IRC Client")
  • [16:18:24] <koen> DJWillis: have you ever seen a TI git tree?
  • [16:18:24] <florian> khasim: http://wiki.openembedded.net/index.php/BeagleBoard
  • [16:18:28] <koen> DJWillis: I haven't :)
  • [16:18:29] <jkridner> DJWilllis: I believe that is what Arago is.
  • [16:18:43] <florian> khasim: it depends on what you want to validate
  • [16:19:05] <khasim> florian: I have listed the same in recent mail on discussion list
  • [16:19:10] <jkridner> florian: great!
  • [16:19:27] <DJWillis> koen: well I guess only OmapZoom ones and that is not really TI ;-).
  • [16:19:34] * eme1 (n=ericb@ip72-223-90-212.ph.ph.cox.net) has joined #beagle
  • [16:20:02] <Crofton|irssi> I talked to someonoe who built some sdr sw on with arago and it had some unexpected issues
  • [16:20:44] * sakoman_ has no clue what this arago thing is
  • [16:21:08] * eme1 (n=ericb@ip72-223-90-212.ph.ph.cox.net) has left #beagle
  • [16:21:09] <Crofton|irssi> TI's distro based on angstrom/OE
  • [16:21:12] <Crofton|irssi> basically
  • [16:21:14] <khasim> I dont know if we have to get every beagle user to use OE/Arago. I prefer a simple rootfs and kernel snapshot to test his board
  • [16:21:34] * sakoman_ thinks koen's suggestion of 2.6.28 + OE patches is a good one since that is what most Beagle customers end up using
  • [16:21:46] * khasim too
  • [16:22:22] * emeb-eee (n=ericb@ip72-223-90-212.ph.ph.cox.net) has joined #beagle
  • [16:22:30] * Leon_Nardella1 (n=leon@200-161-14-111.dsl.telesp.net.br) Quit ("Leaving.")
  • [16:22:42] <khasim> I think : 2.6.28 + OE patches and Koen's NARCISSUS filesystem image is good starting point for validation software
  • [16:22:54] <khasim> any other opinion
  • [16:23:00] <Crofton|irssi> narcissus?
  • [16:23:10] <DJWillis> sakoman_: I googled Arago and found commits from in someones OE/Arago clone/tree ;-).
  • [16:23:13] <khasim> http://amethyst.openembedded.net/~koen/narcissus/
  • [16:23:17] <DJWillis> jkridner: Took me a while to work out what it was, so why the fork (if that is what it is?)
  • [16:23:39] <sakoman_> khasim: how automated is the validation procedure?
  • [16:23:41] <Crofton|irssi> ah
  • [16:24:17] <Crofton|irssi> "self-admirer'?
  • [16:24:47] <khasim> sakoman_: I am not sure, how Gerald is automating it, they are just simple Linux commands after kernel boots
  • [16:24:52] <Crofton|irssi> DJWillis: I think they had some "special" requirements
  • [16:25:15] <khasim> I am sure, he is running them in a script file
  • [16:25:36] <DJWillis> Crofton|irssi: I did wonder.
  • [16:25:53] <Crofton|irssi> I think it uses CSL tool chains
  • [16:26:05] <Crofton|irssi> I'd try it, but my list of things to try is huge
  • [16:26:08] <koen> Crofton|irssi: http://en.wikipedia.org/wiki/Narcissus_(mythology)
  • [16:26:21] <koen> Crofton|irssi: the guy built an image and fell in love with it
  • [16:26:25] <Crofton|irssi> ah
  • [16:26:25] <khasim> also I need to make sure that tests don't change often, testers will find it diffcult to learn ...
  • [16:26:27] <Crofton|irssi> ok
  • [16:26:41] <sakoman_> gumstix uses the shipping x11 image plus a simple script - they verify the console serial port, hdmi output, functionality of usb keyboard/mouse, audio aoutput
  • [16:26:44] <Crofton|irssi> I read that
  • [16:27:04] <Crofton|irssi> koen: hopefully more people than you :)
  • [16:28:03] <sakoman_> It's all on microSD cards that get inserted in the board, final phase of script writes the onboard flash with x-load, u-boot, kernel, and rootfs
  • [16:28:04] * khasim wants to get back to work now...
  • [16:28:22] <sakoman_> khasim: ttyl
  • [16:28:37] <khasim> sakoman_ : :)
  • [16:30:25] <DJWillis> sakoman_: guess what I have near the top of my Pandora TODO ;-). I may wish to pick your brains on that sometime if you don't mind.
  • [16:35:17] * _AV500_ (n=AV500@sac91-1-82-232-88-182.fbx.proxad.net) Quit (Read error: 113 (No route to host))
  • [16:35:43] * denix0 (n=denys@nat/ti/x-ebd5e1bc74be5832) has joined #beagle
  • [16:38:41] * leslie_ (n=leslie@222.66.141.242) has joined #beagle
  • [16:39:41] * leslie (n=leslie@222.66.141.242) Quit (Read error: 113 (No route to host))
  • [16:42:15] <sakoman_> DJWillis: Sure thing
  • [16:42:56] <atin_> sakoman_: did you get my patches?
  • [16:44:02] * koen has DVD support working in angstrom: http://scap.linuxtogo.org/files/fe545babec8299c8d686ea966d19d594.png
  • [16:44:08] <koen> time for food
  • [16:44:12] <sakoman_> atin_: yes, just noticed the email
  • [16:44:41] <atin_> ok, cool. hoping someone else can run this and make sure the user-button doesn't have to be pressed for the board to be enumerated.
  • [16:45:55] <Crofton|irssi> ok, booting action with pm kernel du jour
  • [16:47:07] <sakoman_> atin: let's see what dirk2 has to say - he may like to have the regulator stuff go in common/power.c
  • [16:47:12] * mib_6rh8i6 (i=4a5dd0bb@gateway/web/ajax/mibbit.com/x-2d9d99d256a66af2) has joined #beagle
  • [16:47:30] * docelic (n=docelic@78.134.201.183) Quit (Read error: 104 (Connection reset by peer))
  • [16:47:34] <mib_6rh8i6> anyone got any ideas on why i cant type in my beagle board console over the serial connection???
  • [16:47:40] <atin_> sakoman_: ok, I was thinking of that too - but wasn't sure because I thought this was also twl4030 specific.
  • [16:47:45] <mib_6rh8i6> i see the boot just fine
  • [16:47:51] <atin_> sakoman_: and not all omap3 stuff would have that I don't think.
  • [16:48:23] <atin_> sakoman_: my next goal was to separate out all the twl4030 code into a common file that gets linked in separately if that is the chip we have.
  • [16:48:38] <atin_> sakoman_: but that would mean stubs for the other boards.
  • [16:48:52] <sakoman_> atin: agreed. actually I think power.c has some issues in that it sets up supplies that aren't needed on Overo, for example
  • [16:49:25] <sakoman_> so this could be a good push for us to resolve those issues
  • [16:49:28] <atin_> when is the rev C board coming? march or so?
  • [16:49:37] <sakoman_> I think that is correct
  • [16:49:39] <muriani> yeah
  • [16:50:00] <atin_> wonder how the better half will like my wanting to buy another beagle . . .
  • [16:50:29] <dirk2> sakoman, atin_: Didn't look at USB patches, yet. My idea was 'get it working first, then clean up'
  • [16:50:44] <atin_> who do I petition to get one since I am actually working on usb? :)
  • [16:51:42] <atin_> dirk2: no problems with that on my end - I am thinking the omap3-dev-usb branch is meant for the people actually working on the code to get it going.
  • [16:51:45] <sakoman_> dirk2: OK, in that case I should just apply atin_'s patch?
  • [16:52:01] <dirk2> sakoman_: yes
  • [16:52:03] <jkridner> atin_: the idea is that Beagle is affordable enough for everybody to buy 2. :)
  • [16:52:19] <atin_> talk to my better half!
  • [16:52:25] <dirk2> atin_: yes, this is my understanding of omap3-dev-usb, too
  • [16:52:37] <atin_> I'll first have to explain why I bought the broken one when the fixed one was only a few months off!
  • [16:53:06] <atin_> (because its cool to work on something that is still being worked on to get off the ground)
  • [16:53:06] <sakoman_> atin_: could you send me your patch as an attachment? gmail mangled it
  • [16:53:28] <atin_> I thought it was an attachment! :) but ok, let me do it again
  • [16:53:40] * Leon_Nardella (n=leon@200-161-14-111.dsl.telesp.net.br) has joined #beagle
  • [16:53:57] * florian (n=fuchs@217.146.132.69) Quit (Remote closed the connection)
  • [16:54:48] * sakoman_ thinks gmail needs a raw feature
  • [16:55:38] <dirk2> atin_: once you think it's working, create one patch from git and send it to beagle list. Then we can discuss clean up by mail review
  • [16:55:43] * nemequ (n=nemequ@ip68-111-215-155.sd.sd.cox.net) Quit (Read error: 60 (Operation timed out))
  • [16:56:17] <atin_> dirk2: I sent a patch to the ML - specified that it was off the omap3-dev-usb branch.
  • [16:56:53] <atin_> tho I think only diego, me and maybe you are the only ones who seem to care about that branch right now.
  • [16:57:13] <atin_> sakoman_: sent
  • [16:57:24] <dirk2> atin_: no, sorry, misunderstanding, I don't mean this patch. I'm talking about clean up sakoman mentioned. we can do clean up after functionality is there
  • [16:57:55] <atin_> dirk2: yes. we're going to be running into issues anyway since first the omap3 stuff has to get into u-boot :)
  • [16:58:44] <atin_> dirk2: then usb stuff. we are also going to run into stuff from the TI folks working on the other OMAP3 boards.
  • [16:59:21] <atin_> dirk2: in fact, I was going to take the changes we've made for TWL4030 init and put them into a separate file so it can be linked into the host side code also.
  • [16:59:29] <sakoman_> atin_: pushed: http://www.sakoman.net/cgi-bin/gitweb.cgi?p=u-boot-omap3.git;a=commit;h=91e13486f599139d1f8cf425b70a6de267087c3f
  • [16:59:50] <atin_> dirk2: but couldn't figure out how to call a platform specific file from the musb code that is generic.
  • [17:00:58] <atin_> dirk2: in fact, usbdcore_musb.c shouldn't be doing that either - at the moment it calls twl4030 code directly
  • [17:01:46] <atin_> dirk2: I guess I'll make that my next to do. figure out a platform_data ptr that can point at stuff like this.
  • [17:02:47] * and-ri (n=ubuntu@pd95b4498.dip0.t-ipconnect.de) has left #beagle
  • [17:03:10] * koen notices cdc ethernet patches for uboot
  • [17:03:18] * recalcati (i=51485a4f@gateway/web/ajax/mibbit.com/x-eb579fa9a2c4dd72) Quit ("http://www.mibbit.com ajax IRC Client")
  • [17:03:27] <atin_> koen: I saw that too. closed my eyes and looked away :)
  • [17:04:28] <atin_> koen: they said they were big changes. I wonder how that will affect us.
  • [17:04:57] <atin_> koen: they must be duplicating work we are trying to do - much more extensively on their side.
  • [17:05:33] <atin_> koen: one of the things on my todo is to take a look at their git and see how we can work with them. :)
  • [17:06:29] <atin_> hey, all you people who read lots of mailing lists. how do you keep track? keep a separate email account that gets the emails?
  • [17:06:38] * nemequ (n=nemequ@ip68-111-215-155.sd.sd.cox.net) has joined #beagle
  • [17:06:41] <atin_> I signed up using my normal gmail account.
  • [17:07:00] <atin_> and then have been trying to write filters to forward all mailing list messages to another gmail account.
  • [17:07:41] <atin_> but there seems to be a problem with gmail filters. either that or I am doing it wrong. but if I mark message as "skip mailbox", "forward to ...", "delete it", then nothing shows up
  • [17:07:47] <adj_> atin_: having multiple mail accounts just for different mailing lists sounds a bit weird
  • [17:07:49] * mib_6rh8i6 (i=4a5dd0bb@gateway/web/ajax/mibbit.com/x-2d9d99d256a66af2) Quit ("http://www.mibbit.com ajax IRC Client")
  • [17:08:02] <atin_> if I dont' say delete it, then it shows up on both my normal account and my forward acct!
  • [17:08:15] <koen> atin_: I read the ones that interest my in thunderbird with nntp via gmane
  • [17:08:29] <atin_> adj_: I have it so that I can read from my forwarded account, but if I want to respond, it'll use my normal email
  • [17:08:29] <koen> atin_: ones that interest me less I read online via gmane
  • [17:08:38] <adj_> i have few procmail rules which sort messages from different lists to different mail directories
  • [17:08:43] * koen is only gets actual mail for a handfull of lists
  • [17:09:11] <atin_> koen: ah. I am only on the linux-usb, u-boot-usb, beagle mailing lists.
  • [17:09:15] <koen> :0
  • [17:09:15] <koen> * ^List-Id:.*mobile-devel-list\.gnome\.org
  • [17:09:15] <koen> $MAILDIR/.Gnome\ mobile/new
  • [17:09:17] <atin_> and even that was swamping me :)
  • [17:09:18] <koen> like that?
  • [17:09:37] <adj_> yeah, pretty much like that
  • [17:09:52] <atin_> adj_: I was going to do that next if I failed at just getting gmail to do it
  • [17:10:16] <atin_> I can't believe that the filter won't work tho! if I don't delete it in my main email account, it works great.
  • [17:10:23] <atin_> tho I also swamp my main acct.
  • [17:11:06] <atin_> the reason I want them in my email, is so I can save patches to a mailbox easily and then apply them with just a simple git am
  • [17:11:45] <atin_> I couldn't figure out how to get gmane to do that.
  • [17:13:39] <koen> gmane doesn't have a 'raw' option
  • [17:13:47] <koen> but reading it using nntp gets you that
  • [17:14:03] * dcramer (n=davec@missdsl.ebox.com) has joined #beagle
  • [17:15:25] <atin_> whats the nntp address for the u-boot list?
  • [17:17:48] * dcramer_ (n=davec@missdsl.ebox.com) has joined #beagle
  • [17:17:48] * gcohler (n=gcohler@nat/ti/x-1b5fb9a3bdefc7b7) has joined #beagle
  • [17:18:51] * dcramer (n=davec@missdsl.ebox.com) Quit (Read error: 104 (Connection reset by peer))
  • [17:19:35] <jkridner> florian: does OTG really work with OE 2.6.28-r3, or just host mode OTG? If it isn't in the defconfig as-is and kernel that is pre-built, then I'd say it isn't there.
  • [17:20:43] <jkridner> florian: at this point, is there a reason that we don't switch to 2.6.28 as being the preferred kernel?
  • [17:21:47] <atin_> jkridner: yeah - why not .28 yet?
  • [17:21:50] <koen> jkridner: 2.6.28 is a bit more flaky than 2.6.27, I want to wait with switching till 2.6.28-omap2 gets taggeg
  • [17:22:11] <koen> jkridner: and dsplink needs to work with 2.6.28 before we switch
  • [17:26:02] <sakoman_> and asla needs to get fixed too
  • [17:26:27] <gregoiregentil> and sgx?
  • [17:26:45] <sakoman_> to name just a few things :-)
  • [17:26:45] <erbo> I'm trying to build a small image based on task-base with an extra lib, but the resulting rootfs is very minimal without even init on it. I'm sure I've made some simple mistake, any care to take a quick look? http://pastebin.com/d7856a912
  • [17:28:08] <koen> sakoman_: one part of the alsa fix is in
  • [17:29:03] <gcohler> sakoman: Any idea why the GWC ASIX Ethernet interface would no longer work with Koen's demo (it used to), and would still work with your feed images on Beagleboard?
  • [17:29:09] <sakoman_> koen: yeah, but I still get a segfault
  • [17:29:49] <sakoman_> gcohler: no clue. you might look at the history for koen's kernel recipe to see why
  • [17:30:07] <koen> erbo: have a look at the testlab dir in deploy/image/beagleboard
  • [17:30:31] <gcohler> sakoman_: Thanks will do.
  • [17:31:19] <royerfa> Hi
  • [17:31:35] <royerfa> how can I get the git tree of the FFmpeg project
  • [17:31:46] <royerfa> git clone URL ?
  • [17:31:47] <erbo> koen: my image build did not create a testlab dir, but it does when I build beagleboard-demo-image
  • [17:32:16] <koen> erbo: if there's no testlab then something went wrong during image creation
  • [17:32:24] <koen> erbo: with would explain the missing bits :)
  • [17:34:05] * garren (n=garren@dsl-243-100-242.telkomadsl.co.za) has joined #beagle
  • [17:34:13] <erbo> I get no error and it builds me a the .jffs2, .tar.gz and .tar.bz2 files + a kernel :/
  • [17:34:21] * valhalla (n=valhalla@81-174-24-85.dynamic.ngi.it) has joined #beagle
  • [17:35:22] <erbo> oh wait
  • [17:35:25] <erbo> I'm blind
  • [17:39:33] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [17:43:56] * mckoan is now known as mckoan|away
  • [17:44:43] <koen> tomba: which branch should I get DSS2 patches from?
  • [17:46:28] <tomba> master
  • [17:47:53] <tomba> it contains same stuff I sent to mailing lists
  • [17:56:06] * mib_supk7c (i=4c00c5cb@gateway/web/ajax/mibbit.com/x-669bc7714516beb9) has joined #beagle
  • [17:58:15] * gcohler (n=gcohler@nat/ti/x-1b5fb9a3bdefc7b7) Quit (Remote closed the connection)
  • [17:58:42] * ldesnogu_ (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) has joined #beagle
  • [17:59:08] * gcohler (n=gcohler@nat/ti/x-0b55b18767f367a1) has joined #beagle
  • [17:59:09] * atin_ is now known as atin_away
  • [18:01:42] * mib_supk7c (i=4c00c5cb@gateway/web/ajax/mibbit.com/x-669bc7714516beb9) Quit ("http://www.mibbit.com ajax IRC Client")
  • [18:02:26] * nemequ (n=nemequ@ip68-111-215-155.sd.sd.cox.net) Quit (Read error: 110 (Connection timed out))
  • [18:04:43] * guillaum1 (n=gl@AMontsouris-153-1-34-86.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
  • [18:05:25] <koen> tomba: does it have the clock changes needed to compile against l-o head?
  • [18:05:46] * mib_g0v09p (i=3b93dac7@gateway/web/ajax/mibbit.com/x-fe3d849b0e3f6daa) has joined #beagle
  • [18:06:09] <tomba> yes, I believe so
  • [18:06:12] * guillaum1 (n=gl@AMontsouris-153-1-34-86.w90-2.abo.wanadoo.fr) has joined #beagle
  • [18:06:38] <koen> Crofton|irssi: btw, the heartbeat led trigger is bad for pm as well
  • [18:08:52] * dcordes (n=dcordes@unaffiliated/dcordes) has joined #beagle
  • [18:10:35] * Xenion (n=robert@p579FC091.dip.t-dialin.net) has joined #beagle
  • [18:11:12] <Xenion> Guten Abend / good evening :-)
  • [18:12:25] * royerfa (n=fabroy01@fw-tnat.cambridge.arm.com) has left #beagle
  • [18:15:12] * mib_g0v09p (i=3b93dac7@gateway/web/ajax/mibbit.com/x-fe3d849b0e3f6daa) Quit ("http://www.mibbit.com ajax IRC Client")
  • [18:15:54] <maelcum> to make sure it's not forgotten, usb (host) does not work reliably in 2.6.27. without rxtx patch the usb port stops working, with rxtx patch the kernel crashes due to a null pointer dereference.
  • [18:16:27] * JoeyBorn (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net) has joined #beagle
  • [18:23:55] <koen> kulve: have you tried xf86-video-omapfb with 2.6.28 and latest dss2?
  • [18:24:02] <koen> kulve: it always fails to init XV
  • [18:29:46] <felipec> how can I see the default CFLAGS of the gcc compiler?
  • [18:35:08] * matt_c (n=mcroydon@pool-71-178-172-179.washdc.fios.verizon.net) Quit ()
  • [18:40:03] <koen> -dumpspecs
  • [18:44:46] * FuL|OUT is now known as fulgas
  • [18:45:00] * emeb-eee (n=ericb@ip72-223-90-212.ph.ph.cox.net) Quit (Read error: 110 (Connection timed out))
  • [18:51:02] <dirk2> sakoman, jkridner: ACK for [PATCH-OMAP3 v3] Modified pinmux to enable HSUSB2 (EHCI) on Beagle
  • [18:52:49] <koen> for one moment I thought you meant that omap3 patches were accepted upstream
  • [18:57:32] <ds2> dirk2: do the U-boot folks enforce having a proper git comment? I seem to recall Linus wants a real git comment for kernel patches
  • [18:59:23] <koen> ds2: they must be silently enforcing it then
  • [18:59:33] * gcohler (n=gcohler@nat/ti/x-0b55b18767f367a1) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [18:59:44] * gcohler (n=gcohler@nat/ti/x-cb8d0dec1805db67) has joined #beagle
  • [18:59:46] <dirk2> ds2: yes, good idea. But as sakoman's u-boot git is 'only' our private git and we manually re-structure patches for U-Boot mailing list, so I don't think that it really matters here. But yes, would be good to improve it next time.
  • [19:01:43] * jsync (n=jess@de1-as21224.alshamil.net.ae) Quit (Read error: 104 (Connection reset by peer))
  • [19:02:35] <ds2> dirk2: 'k be nice to be able to ask people to just pull
  • [19:02:59] <ds2> koen: think he might have said something to that effect at LPC
  • [19:03:30] <dirk2> ds2: wolfgang wants *everything* through mailing list
  • [19:03:44] <ds2> oh :/
  • [19:08:31] * dcramer_ (n=davec@missdsl.ebox.com) Quit ()
  • [19:09:29] * bmxr_laptop (n=chatzill@142.232.129.24) has joined #beagle
  • [19:10:00] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
  • [19:10:44] * rsalveti (n=salveti@200.184.118.130) Quit (Connection timed out)
  • [19:12:11] * dirk2 (n=dirk@p5B043684.dip0.t-ipconnect.de) Quit ("Konversation terminated!")
  • [19:13:01] * GregorR (n=gregor@65.183.185.132) has joined #beagle
  • [19:14:46] <GregorR> I'm semi-considering a BeagleBoard for a project, and I'm looking around for ways to battery-power the beagleboard. I've found a few mentions of powering it with those weird USB-charging battery packs for cell phones and such, but would ideally like to have the USB port free for other uses. Are there any battery packs available that plug into the alternate power port? Also, how long does it last on these (some very rough approximation :) )
  • [19:15:02] * rsalveti (n=salveti@200.184.118.130) has joined #beagle
  • [19:15:44] * dcordes (n=dcordes@unaffiliated/dcordes) Quit (Read error: 131 (Connection reset by peer))
  • [19:16:09] <bmxr_laptop> You would have to determine the size and ma of the battery pack and how much power the usb will be drawing and if you will be using a usb hub
  • [19:16:13] * felipec (i=c0647cda@gateway/web/ajax/mibbit.com/x-a567202a511023d8) Quit ("http://www.mibbit.com ajax IRC Client")
  • [19:17:50] <GregorR> This is why I said "rough" approximation :P
  • [19:19:50] <GregorR> (I want the USB to get Wifi and maybe Bluetooth, which makes me thing a normal 2-AA-powered battery pack would be far too weak to be useful)
  • [19:24:32] * robtow (n=rtow@64.62.142.114) Quit ("Leaving.")
  • [19:26:18] <ds2> You can use a USB to barrel connector cable to power it from the packs
  • [19:27:53] * TehUni (n=cory@c-24-30-35-55.hsd1.ga.comcast.net) Quit (Read error: 104 (Connection reset by peer))
  • [19:28:13] * TehUni (n=cory@c-24-30-35-55.hsd1.ga.comcast.net) has joined #beagle
  • [19:28:17] <ds2> or you can build a simple circuit to do the equiv. Something like what I have: http://www.hy-research.com/Li_power.html
  • [19:28:39] * mib_1sioyv (i=5a0f9024@gateway/web/ajax/mibbit.com/x-6be40cf4cbadd109) Quit (Remote closed the connection)
  • [19:28:39] * newton (i=812ad0b1@gateway/web/ajax/mibbit.com/x-edc5dbd672133707) Quit (Remote closed the connection)
  • [19:28:48] <ds2> You can get the Li charger part from SFE as a board; just tack a switcher to boost it/regulate it
  • [19:29:55] <maelcum> afaics the simplest (if slightly wasteful) way would be to get a four nimh battery pack and a linear 5v regulator.
  • [19:30:50] * mib_1sioyv (i=5a0f9024@gateway/web/ajax/mibbit.com/x-c447b2ac3bf67df4) has joined #beagle
  • [19:31:31] <ds2> why bother with the linear?
  • [19:31:35] <maelcum> make that five nimh cells. the voltage is only 1.2v, i was thinking 1.2.
  • [19:31:36] <ds2> 4 NiMH is 4.8V only
  • [19:31:43] <maelcum> less external components
  • [19:31:55] <maelcum> erm, i was thinking 1.4
  • [19:32:01] <ds2> 5 NiMH gives you 6V so you HAVE to use an LDO, not a 7805
  • [19:32:28] <maelcum> isn't the minimum drop of a linear regulator fairly small?
  • [19:32:30] <ds2> and if you want inplace charging, that is more stuff
  • [19:32:39] <ds2> for a 7805 it can be as bad as 2 to 2.5V
  • [19:32:43] <ds2> a LDO is much smaller
  • [19:34:02] <ds2> bbl
  • [19:34:04] <mru> you could probably get away with 4.8 V unless you need to power usb
  • [19:34:16] <mru> I can't think of anything else on the board that needs more than 3.3 V
  • [19:34:19] <ds2> on a RevB it don't matter
  • [19:34:23] <GregorR> Hmmm ... if I got a battery-powered USB hub (although I don't know if they make any with this much battery power, but ...), plugged it into both the USB port and the power port (via a USB-to-barrel adapter), that would be the simplest method it seems. Although that's under the assumption that these super-battery hubs exist :)
  • [19:34:30] * bmxr_laptop (n=chatzill@142.232.129.24) Quit (Read error: 113 (No route to host))
  • [19:34:32] <ds2> the TWL4030 has a charge pump for USB
  • [19:34:37] <maelcum> the thing about rechargeables is, they keep the voltage until they are almost empty. this is good here.
  • [19:34:41] <ds2> but you need a bit more then 3.3V due to the onboard LDO's
  • [19:34:59] <mru> yes, but 4.8 should be plenty
  • [19:35:05] * greytoque (i=8eb1caa6@gateway/web/ajax/mibbit.com/x-a2f6c0d2147a02a0) has joined #beagle
  • [19:35:18] <mru> I've heard of people running beagles from supplies well under 4 V
  • [19:35:33] <denix0> koen: ping
  • [19:35:35] <ds2> they make power packs that give you a regulated 5V
  • [19:36:01] <adj_> isn't usb vbus powered from VBAT (4.2 volts) by Triton2?
  • [19:36:12] <ds2> mru: yep... think Gerald was saying, the component that would be impacted first as you head toward 3.3V input is the DVI xmitter
  • [19:36:39] <mru> what about the mmc slot?
  • [19:36:40] <adj_> that is, going under 5 volts in input shouldn't affect the usb vbus voltage
  • [19:36:44] <mru> does it provide 3.3V?
  • [19:36:46] <ds2> adj_: on the OTG port, yes, the Triton2 has a charge pump for 5V generation; on RevC and later, the EHCI port uses the raw power
  • [19:37:02] <ds2> donno, check the TPSxxxxxx datasheet ;) or use 1.8V cards }:-)
  • [19:37:12] <greytoque> hello all. does anyone have experence develoing in CCS? I'm trying to get started on the beagleboard - and I don't have any CCS/DSP expereince at all so I have some pretty basic questions.
  • [19:37:21] <ds2> anyways... bbl
  • [19:37:36] * florian (n=fuchs@f048036030.adsl.alicedsl.de) has joined #beagle
  • [19:37:38] <mru> I suspect most cards sold nowadays are 1.8V
  • [19:38:21] <florian> re
  • [19:38:58] <maelcum> has anybody succeeded retrofitting a working host-only usb port to a revB board? i know it's improbable. but it would fix my problems.
  • [19:41:09] <adj_> have there been any public schematics for RevC yet?
  • [19:42:09] <koen> denix0: pong
  • [19:42:39] <koen> maelcum: the port needs to be attached to different balls on the omap
  • [19:43:05] <maelcum> that sounds impossible to do without extremely sophisticated equipment
  • [19:43:37] <maelcum> or very creative hacks :)
  • [19:44:01] <denix0> koen: any ideas why rootfs may want to pull libc6-dev instead of libc6?
  • [19:44:12] * fulgas (n=fn@a83-132-158-61.cpe.netcabo.pt) Quit (Read error: 60 (Operation timed out))
  • [19:44:29] <maelcum> koen: have you stress-tested the usb driver of your patched kernels? the patches didn't really help me...
  • [19:44:59] * fulgas (n=fn@a83-132-158-61.cpe.netcabo.pt) has joined #beagle
  • [19:45:02] <maelcum> also, what's their base? they don't all apply cleanly to v2.6.27 nor v2.6.27-omap1
  • [19:45:08] <kulve> koen: I think I did try it on evm..
  • [19:45:42] <maelcum> (caveat: v2.6.27 from the omap repository, but its changelog looks like the clean 2.6.27)
  • [19:48:53] <koen> denix0: check the testlab dir, that shows what package pulls in others
  • [19:51:23] <denix0> koen: I had to increase the image size, as -dev didn't fit. anyway, there is nothing depending on it specifically in testlab
  • [19:52:17] <sakoman_> jkridner: should I wait for Steve K's ack before applying your mux patch?
  • [19:52:32] <denix0> koen: I also see lots of "libc6-dev: unsatisfied recommendation for ..." in do_rootfs log
  • [19:52:44] <jkridner> I don't think so, since he will go on vacation for a few days soon.
  • [19:52:47] * timtimred (n=meh@92-237-114-21.cable.ubr03.chms.blueyonder.co.uk) has joined #beagle
  • [19:53:00] <timtimred> hi koen
  • [19:53:03] <sakoman_> OK, I'll apply it. Looks safe to me
  • [19:53:05] <jkridner> I know it works for EHCI and all the changes seem sane.
  • [19:53:25] * felipec (n=felipec@a91-153-251-222.elisa-laajakaista.fi) has joined #beagle
  • [19:53:27] <timtimred> spoke to you about oe/foonas over email ...
  • [19:53:28] <maelcum> koen: what's your problem with me? that i'm running debian? i mean, you're in no way obligated to help me but still...
  • [19:53:34] <keesj> elinux down?
  • [19:54:33] <jkridner> sakoman_: do you think I should RFC this to u-boot list (http://www.beagleboard.org/gitweb/?p=u-boot-arm.git;a=commitdiff;h=3fc8ea5c03ee4e292ede63c5c7ee24bce76e479d)?
  • [19:54:42] <koen> hey timtimred
  • [19:54:48] <timtimred> :)
  • [19:54:56] <timtimred> thought i'd pop in and say hi
  • [19:55:04] <timtimred> ok, ive been thinking about foonas/angstrom though
  • [19:55:23] <koen> denix0: if it's in the image, you should be able to see what pulls it in
  • [19:55:41] <maelcum> wtf
  • [19:56:01] <timtimred> it is desirable for me (given the ever decreasing amount of time i have) to remove the foonas stuff from OE
  • [19:56:24] <timtimred> and do like you say and support the NAS properly in angstrom
  • [19:56:25] * geckosenator (n=sean@adsl-68-23-87-156.dsl.dytnoh.ameritech.net) Quit (Read error: 110 (Connection timed out))
  • [19:56:35] <jkridner> dirk2: with the latest pin muxing, my board revision autodetect is no longer working.
  • [19:56:54] <koen> denix0: deploy/glibc/images/beagleboard/Angstrom-Beagleboard-demo-image-glibc-ipk-2009.X-test-20090114-beagleboard-testlab$ grep libc depends.dot
  • [19:57:05] <sakoman_> jkridner: no, you likely wouldn't get a response :-)
  • [19:57:23] <jkridner> k. he usually scans the logs pretty well. :)
  • [19:57:48] <jkridner> he was talking earlier about doing a new version test.
  • [19:57:50] <sakoman_> jkridner: could you send me your patch as an attachment? Taken from the list it doesn't apply
  • [19:58:02] <koen> timtimred: either option would be fine by me, I just want to make sure we're not needlessly diverging
  • [19:58:14] <sakoman_> might have been mangled in the posting (or my email reader)
  • [19:58:16] <koen> timtimred: divergence isn't necessarely bad
  • [19:58:20] <timtimred> i prefer the workload to be shared tbh
  • [19:58:30] <timtimred> ive been doing everything myself for foonas for too long ;)
  • [19:58:52] <jkridner> sakoman_: http://www.beagleboard.org/~arago/0001-Modified-pinmux-to-enable-HSUSB2-EHCI-on-Beagle.patch
  • [19:59:09] <timtimred> 3 arches to test for by yourself is a lot across lots of NAS ....
  • [19:59:17] <timtimred> without the dev work lol
  • [19:59:22] <timtimred> etc
  • [19:59:32] <denix0> koen: ah, right. it's _dev and not -dev in there...
  • [19:59:46] <jkridner> btw, skipisz sent me a private ack. :-|
  • [19:59:53] <koen> denix0: yeah, I couldn't figure out how to escape chars in .dot files sanely
  • [20:00:38] <koen> Crofton|irssi: bon apetit!
  • [20:00:58] <sakoman_> jkridner: I misunderstood your question! I think that patch *should* go to u-boot list
  • [20:01:19] <sakoman_> though who knows what the odds are of getting a response!
  • [20:01:29] <sakoman_> seems really hit & miss :-(
  • [20:01:39] <Xenion> Gute Nacht alle miteinander ! :--)
  • [20:01:43] <koen> heh, I was just about to remark that it will be ignored like the rest of the omap patches
  • [20:01:45] <Xenion> good night folks
  • [20:02:23] <denix0> koen: ok, this is strange: task-base -> udev -> libvolume-id-dev -> libc6-dev
  • [20:02:27] * Xenion (n=robert@p579FC091.dip.t-dialin.net) Quit ("Verlassend")
  • [20:02:38] <jkridner> sakoman_: k. should I send the same original message, or should I provide it as an attachment?
  • [20:03:00] <denix0> koen: it should be libvolume-id0 and not -dev. just checked my other box and it has it right...
  • [20:03:12] <jkridner> Do I manually add ack-by lines?
  • [20:03:46] <koen> denix0: compare the contents of the 2 volumeid packages
  • [20:04:00] <koen> denix0: dpkg-deb -c libvolume-id-dev.ipk
  • [20:04:56] <jkridner> sakoman_: I think I confused myself by having more than one outstanding thread with you.
  • [20:05:38] <sakoman_> jkridner: I pushed your mux patch to omap3-dev and omap3-dev-usb
  • [20:05:53] <sakoman_> http://www.sakoman.net/cgi-bin/gitweb.cgi?p=u-boot-omap3.git;a=commit;h=add630f1625bd8cf2dbe1a87e649b32a1b02423d
  • [20:06:01] <jkridner> sakoman_: do I need to provide that patch to u-boot mailing list?
  • [20:06:20] <sakoman_> jkridner: Yes, I usually add the ack lines by hand
  • [20:06:23] <koen> jkridner: I think that leds are busted in current l-o git
  • [20:06:28] <koen> jkridner: so it's not uboots fault
  • [20:06:52] <koen> jkridner: or disabling the in-kernel mux (by the EHCI patch) breaks things
  • [20:06:55] <jkridner> koen: thanks.
  • [20:07:06] * flo_lap (n=fuchs@f048036030.adsl.alicedsl.de) has joined #beagle
  • [20:07:21] * florian (n=fuchs@f048036030.adsl.alicedsl.de) Quit (Nick collision from services.)
  • [20:07:24] <jkridner> in-kernel mux needs to be fixed anyway.
  • [20:07:37] <jkridner> so, we can assume the problem is with the kernel. that is good.
  • [20:07:38] <denix0> koen: they are identical between boxes
  • [20:08:08] <koen> jkridner: also, I succeed in playing DVDs on my beagle using an usb dvd writer :)
  • [20:08:21] * flo_lap is now known as florian
  • [20:08:24] <jkridner> koen: very nice. love to see a youtube video. :)
  • [20:08:45] <koen> denix0: try bitbake udev -c rebuild, that should fix the glitch
  • [20:09:02] * newton (i=812ad0b1@gateway/web/ajax/mibbit.com/x-64c73576d0faa5b1) has joined #beagle
  • [20:09:07] <sakoman_> jkridner: if you are referring to the "don't use nand environment" patch then I would send it in-line, I don't think the u-boot list likes attachments
  • [20:09:10] <koen> jkridner: I'm currently trying to figure out how to get a 800x600 framebuffer scaled to vga for the pico
  • [20:09:45] <koen> jkridner: since the xv driver rejects source material that is too big (720x576>640x480)
  • [20:10:33] <sakoman_> jkridner: since your patch affects common code it has to go to the u-boot list. I doubt they would take it as an omap3 thing
  • [20:10:40] <koen> kulve: I suspect that the fb init changed a bit in DSS2, which breaks the XV init
  • [20:11:02] <koen> kulve: I see enableplane 1,1, then enableplane 1,0 and XVinit failed
  • [20:14:33] <jkridner> sakoman_: that code is an RFC thing anyway.
  • [20:15:56] * gregoiregentil (n=zonbu@adsl-71-135-114-242.dsl.pltn13.pacbell.net) has left #beagle
  • [20:19:45] <jkridner> sakoman_: RFC is sent.
  • [20:22:00] * gregoiregentil (n=zonbu@adsl-71-135-114-242.dsl.pltn13.pacbell.net) has joined #beagle
  • [20:23:25] <koen> timtimred: send a mail to the angstrom-devel ml with your NAS needs and we'll see how it works out
  • [20:23:28] * koen -> tv now
  • [20:24:03] * JimDonova1 (n=angstrom@c-66-30-221-249.hsd1.ma.comcast.net) has joined #beagle
  • [20:26:08] <koen> btw: http://www.linuxdevices.com/news/NS6362979448.html
  • [20:27:31] <denix0> koen: ok, that worked... can you explain the glitch? :)
  • [20:27:40] <timtimred> koen cool, will have a play with angstrom :)
  • [20:28:06] <denix0> koen: plus, now I don't see all those nasty unsatisfied recommendations messages...
  • [20:29:26] * emeb-eee (n=ericb@ip72-223-90-212.ph.ph.cox.net) has joined #beagle
  • [20:30:03] <DJWillis> koen: have you built mplayer-0.0+1.0rc2+svnr28311-r9 from OE recently for Beagle? allformats.c seems broke.
  • [20:33:48] * greytoque (i=8eb1caa6@gateway/web/ajax/mibbit.com/x-a2f6c0d2147a02a0) Quit ("http://www.mibbit.com ajax IRC Client")
  • [20:33:57] <sakoman_> DJWillis: what is the failure mode?
  • [20:34:46] <DJWillis> sakoman_: i'll let you know it a sec, just tried it on my 'clean' environment to see if it builds
  • [20:35:17] <sakoman_> DJWillis: ah, so build failure as opposed to runtime failure?
  • [20:35:51] <DJWillis> sakoman_: sorry, yep, build not runtime.
  • [20:41:25] <sakoman_> DJWillis: I haven't tried to build that version yet. Perhaps now I'll wait till you report success :-)
  • [20:41:58] * mib_1sioyv (i=5a0f9024@gateway/web/ajax/mibbit.com/x-c447b2ac3bf67df4) Quit ("http://www.mibbit.com ajax IRC Client")
  • [20:44:35] <DJWillis> sakoman_: looks like a patch OE adds may have borked things, oh well, time for a dig and see whats up.
  • [20:53:35] * dcramer (n=davec@dcdsl.ebox.com) has joined #beagle
  • [20:56:34] * mib_1uj6d7 (i=0c05a33e@gateway/web/ajax/mibbit.com/x-5f019f3928b147bc) has joined #beagle
  • [20:58:25] * mib_1uj6d7 (i=0c05a33e@gateway/web/ajax/mibbit.com/x-5f019f3928b147bc) Quit (Client Quit)
  • [21:00:40] * eFfeM (n=Frans@195-241-226-180.ip.telfort.nl) has joined #beagle
  • [21:01:27] <gregoiregentil> How can this patch http://git.mansr.com/?p=u-boot;a=commitdiff;h=caccdb772c3028a3e3e801fb1554788150752ffc be applied to u-boot-2008.10+r22? I can't find the place in the code where this can be modified. Any idea?
  • [21:02:04] <mru> gregoiregentil: the code has been moved around a bit
  • [21:02:15] * garren (n=garren@dsl-243-100-242.telkomadsl.co.za) Quit ("Ex-Chat")
  • [21:02:32] <mru> and that patch isn't entirely recommended
  • [21:02:40] * jkridner|work (n=a0321898@nat/ti/x-d69a9528182ee6cc) has joined #beagle
  • [21:02:49] <mru> it doesn't raise the core voltage so it becomes ever so slightly unstable
  • [21:03:40] <gregoiregentil> mru: OK. Understood. Nevertheless, I'm crazy enough and I still would like to experiment a little bit. Can you say where it has been moved around in u-boot code?
  • [21:03:51] <mru> I don't remember offhand
  • [21:04:02] <mru> search for 0x1f4
  • [21:04:28] <mru> ah, here it is: http://git.mansr.com/?p=u-boot;a=commitdiff;h=ccde41657b4baea9af1e59e91689dd237a7d1c7a
  • [21:05:12] <mru> I haven't updated my u-boot or kernel trees for quite a while
  • [21:05:18] <gregoiregentil> Great! Many thanks!
  • [21:05:26] <mru> my boards are stable and I've been doing other stuff
  • [21:07:17] * Sept (n=bakljg@c-98-240-226-129.hsd1.mn.comcast.net) has joined #beagle
  • [21:09:38] <koen> DJWillis: mplayer built on my machine for c7x0 and beagle, but it might need some tweaking
  • [21:09:51] <koen> DJWillis: mrus NEON patches are in mplayer svn now :)
  • [21:10:07] <koen> (or rather: mplayer updated svn:externals to pull those in from ffmpeg)
  • [21:10:38] <gregoiregentil> mru: similar question about this patch http://git.mansr.com/?p=u-boot;a=commitdiff_plain;h=ef6ee5af8d584bddadb2d45ad4320cef96b8a934;hp=caccdb772c3028a3e3e801fb1554788150752ffc, is there an update somewhere? Is it the only patch concerning 256?
  • [21:11:15] <koen> denix0: OE does something like: for i in DEPENDS of ${PN} add $i-dev as RRECOMMENDS to ${PN}-dev ; done
  • [21:11:16] <mru> upstream u-boot supports 256MB now
  • [21:11:28] <koen> denix0: but those -dev packages might not exists
  • [21:11:37] <koen> denix0: so the warnings are mostly harmless
  • [21:12:19] <denix0> koen: and why udev required libvolume-dev and got fixed by rebuilding it?
  • [21:12:48] <koen> denix0: I have no idea, I have seem it once or twice before, no idea why it happens
  • [21:13:27] <koen> gregoiregentil: the configure portion of your omapfb patch fails, so I removed that portion for the time being
  • [21:13:33] <denix0> koen: ok, thanks!
  • [21:13:44] <tomba> koen: hmm I was running mplayer with xv with the test-branch, and the current master is almost the same as test was
  • [21:14:24] <koen> tomba: I have no idea why it doesn't work any more, XV still works with DSS1
  • [21:14:46] <koen> tomba: but I've been concentrating on PM lately, so I haven't looked into it yet
  • [21:16:08] <tomba> koen: any logs on pastebin? =)
  • [21:17:40] <koen> I can do some tomorrow
  • [21:17:51] <gregoiregentil> koen: I will take a look and fix it.
  • [21:18:34] <koen> tomba: but basically I see 2 overlays getting enabled (dss debug) during X startup and then only 1 later during the same startup
  • [21:18:48] <koen> tomba: so it might be X tickling the overlays in a slightly wrong way
  • [21:19:30] <tomba> during X startup? X should only enable the gfx overlay...
  • [21:19:49] <koen> iirc it probles fb1 for it capabilities
  • [21:20:11] <tomba> perhaps I can test it tomorrow also, let's see. million things to do.
  • [21:21:00] * k2340 (n=k2340@lawn-128-61-17-91.lawn.gatech.edu) has joined #beagle
  • [21:22:11] <k2340> Hi everyone, any word on sgx userspace libraries? I've seen backlogs of talk of OMAP35x_Graphics_SDK_setuplinux_3_00_00_05 and 3_00_00_29.bin. Any word if these work?
  • [21:22:59] <koen> k2340: "not without patching and shuffling"
  • [21:23:15] <koen> tomba: btw, did that new patch apply to your sgx git tree?
  • [21:23:58] <k2340> koen: is it the kind of "patching and shuffling" that strikes fear into the heart of men?
  • [21:24:22] <koen> k2340: no, it's mostly tedious and using non-public bits
  • [21:24:25] <tomba> koen: no, I haven't had time. I briefly looked it and saw that it will cause some conflicts, so I left it for later.
  • [21:24:41] <koen> we're trying to improve things, but everything is a bit scattered and disorganized
  • [21:25:07] <k2340> is there anywhere that lists the steps needed to patch and shuffle?
  • [21:25:22] <k2340> I don't mind tediousness
  • [21:27:12] <koen> the patch has some problems (as tomba just pointed out) and without that patch you need non public userspace :(
  • [21:27:26] <koen> non-public as in "not in an SDK yet"
  • [21:27:35] * koen zzzzz now
  • [21:28:06] <k2340> koen: I'll see if I can get these patches then since my lab is technically funded by TI to do graphics on the omap
  • [21:28:27] <eFfeM> night koen
  • [21:32:38] <Crofton|irssi> gn
  • [21:32:48] <Crofton|irssi> hmmm, this TZ stuff is confusing
  • [21:37:45] * gcohler (n=gcohler@nat/ti/x-cb8d0dec1805db67) Quit (Remote closed the connection)
  • [21:39:52] * eFfeM1 (n=frans@195-241-226-180.ip.telfort.nl) has joined #beagle
  • [21:40:23] <Crofton|irssi> k2340: what kind of graphics?
  • [21:41:22] * zedstar (n=john@fsf/member/zedstar) has joined #beagle
  • [21:41:40] <k2340> AR research/prototyping
  • [21:41:58] <Crofton|irssi> ar?
  • [21:42:05] <k2340> augmented reality
  • [21:42:33] <Crofton|irssi> ah
  • [21:43:08] <k2340> kind of silly that they're paying us to do graphical work when the graphics don't work lol
  • [21:43:11] * valhalla (n=valhalla@81-174-24-85.dynamic.ngi.it) Quit ("Leaving")
  • [21:43:26] <Crofton|irssi> sort of like creating "beer goggle" but without the need to drink a lot of beer
  • [21:43:33] <Crofton|irssi> heh
  • [21:43:39] <k2340> Hah yeah you can do that.
  • [21:43:48] * Crofton|irssi wishes TI would pay him to do some sdr apps :)
  • [21:43:55] * JimDonova1 (n=angstrom@c-66-30-221-249.hsd1.ma.comcast.net) Quit (Read error: 110 (Connection timed out))
  • [21:44:43] <k2340> don't we all lol
  • [21:46:02] * bazbell (n=a0192809@nat/ti/x-0f30739325024373) Quit ("Leaving.")
  • [21:50:00] * fagius (n=fagius@static-71-111-254-119.rlghnc.dsl-w.verizon.net) has joined #beagle
  • [21:51:21] * eFfeM1 (n=frans@195-241-226-180.ip.telfort.nl) Quit ("Leaving.")
  • [21:53:45] * timtimred is now known as l33thaxr
  • [21:54:15] * l33thaxr is now known as timtimred
  • [21:54:15] * timtimred is now known as l33thaxr
  • [21:55:12] * l33thaxr is now known as timtimred
  • [21:55:19] * eFfeM (n=Frans@195-241-226-180.ip.telfort.nl) Quit (Read error: 110 (Connection timed out))
  • [22:01:39] * mib_xezzr1 (i=3fa2c79a@gateway/web/ajax/mibbit.com/x-7f87f43eeedd0f1d) has joined #beagle
  • [22:05:34] * __alanc__ (n=a-campbe@nat/ti/x-632c874436da3d48) Quit (Remote closed the connection)
  • [22:06:57] * newton (i=812ad0b1@gateway/web/ajax/mibbit.com/x-64c73576d0faa5b1) Quit ("http://www.mibbit.com ajax IRC Client")
  • [22:08:06] * DJW|Home (i=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) has joined #beagle
  • [22:11:07] * mib_xezzr1 (i=3fa2c79a@gateway/web/ajax/mibbit.com/x-7f87f43eeedd0f1d) Quit ("http://www.mibbit.com ajax IRC Client")
  • [22:12:24] * docelic (n=docelic@78.134.201.183) has joined #beagle
  • [22:23:19] * BThompson (n=BThompso@nat/ti/x-bfd6e83a42b2d41d) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [22:24:49] * Sept (n=bakljg@c-98-240-226-129.hsd1.mn.comcast.net) Quit ()
  • [22:24:53] * rsalveti (n=salveti@200.184.118.130) Quit (Remote closed the connection)
  • [22:25:54] * DJWillis (n=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) Quit (Read error: 110 (Connection timed out))
  • [22:26:20] * k2340 (n=k2340@lawn-128-61-17-91.lawn.gatech.edu) Quit ("Leaving")
  • [22:29:18] * abitos (n=nixgibts@dslb-084-057-155-043.pools.arcor-ip.net) has joined #beagle
  • [22:34:39] * Sept (n=bakljg@c-98-240-226-129.hsd1.mn.comcast.net) has joined #beagle
  • [22:35:09] * atin_away (n=atin@c-71-232-7-71.hsd1.ma.comcast.net) has left #beagle
  • [22:35:16] * Sept (n=bakljg@c-98-240-226-129.hsd1.mn.comcast.net) Quit (Client Quit)
  • [22:35:43] * Sept (n=bakljg@c-98-240-226-129.hsd1.mn.comcast.net) has joined #beagle
  • [22:35:56] * Sept (n=bakljg@c-98-240-226-129.hsd1.mn.comcast.net) Quit (Client Quit)
  • [22:38:04] * Sept (n=bakljg@c-98-240-226-129.hsd1.mn.comcast.net) has joined #beagle
  • [22:38:20] * Sept (n=bakljg@c-98-240-226-129.hsd1.mn.comcast.net) Quit (Client Quit)
  • [22:38:39] * Sept (n=bakljg@c-98-240-226-129.hsd1.mn.comcast.net) has joined #beagle
  • [22:39:40] * jrmuizel (n=jrmuizel@mozca02.ca.mozilla.com) Quit ()
  • [22:41:48] * JoeyBorn (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net) Quit (Read error: 110 (Connection timed out))
  • [22:44:48] * ldesnogu_ (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) Quit ()
  • [22:51:02] * Crofton|irssi is falling asleep trying to read the clock code
  • [22:59:06] <sakoman_> Crofton|irssi: did you figure out the TZ stuff?
  • [22:59:14] <Crofton|irssi> heh
  • [22:59:32] <Crofton|irssi> I mean the fact my TZ is currently two hours from the normal one
  • [22:59:49] <Crofton|irssi> so everyone's working hours are shifted
  • [22:59:51] <sakoman_> Ah, biological TZ :-)
  • [22:59:55] <Crofton|irssi> right
  • [23:00:44] <sakoman_> I think you are in a perfectly reasonable time zone :-)
  • [23:01:31] <mru> my biological TZ seems to have longer days than the physical one...
  • [23:01:48] * Crofton|irssi noticed mru up very late ....
  • [23:02:07] <sakoman_> mru: I seem to recall studies that show that is actually the case
  • [23:02:21] <mru> I've heard of those too
  • [23:02:40] <sakoman_> If external cues are removed you settle into a biological day that is longer than 24 hours
  • [23:02:51] <mru> probably part of the reason why flying west is so much easier
  • [23:03:43] * uwe_ (n=uwe_@dslb-084-056-023-078.pools.arcor-ip.net) Quit (Read error: 60 (Operation timed out))
  • [23:05:16] <emeb> Crofton|irssi: you out west for work again?
  • [23:07:49] * matt_c (n=mcroydon@c-76-21-131-217.hsd1.md.comcast.net) has joined #beagle
  • [23:07:52] * uwe__ (n=uwe@dslb-084-056-028-182.pools.arcor-ip.net) has joined #beagle
  • [23:07:59] <Crofton|irssi> emeb: yep
  • [23:08:07] <Crofton|irssi> only until next week though
  • [23:08:27] * mib_g0v09p (i=3b93dac7@gateway/web/ajax/mibbit.com/x-fceee1b444007a48) has joined #beagle
  • [23:09:04] * raster (n=raster@124-170-202-104.dyn.iinet.net.au) has joined #beagle
  • [23:13:53] <emeb> Crofton|irssi: Cool - Scottsdale again?
  • [23:14:39] <ds2> what is there to do in Scottsdale?
  • [23:14:45] <ds2> nice library and all but...
  • [23:15:00] <emeb> Rope 'n throw 'n brand-em?
  • [23:15:48] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) has joined #beagle
  • [23:16:00] <Crofton|irssi> yep
  • [23:17:26] * emints (n=jack@postoffice.embeddedalley.com) has joined #beagle
  • [23:19:36] * uwe2 (n=uwe@dslb-084-056-023-078.pools.arcor-ip.net) Quit (Read error: 110 (Connection timed out))
  • [23:20:04] * mib_g0v09p (i=3b93dac7@gateway/web/ajax/mibbit.com/x-fceee1b444007a48) Quit ("http://www.mibbit.com ajax IRC Client")
  • [23:26:15] * mrc3__ (n=mrc3@nat/ti/x-aff7358667b99a90) Quit (Remote closed the connection)
  • [23:26:33] * mrc3__ (n=mrc3@nat/ti/x-f2ce34b21de7d633) has joined #beagle
  • [23:30:28] * matt_c (n=mcroydon@c-76-21-131-217.hsd1.md.comcast.net) Quit ()
  • [23:32:38] * JustinLove (n=J_LOVE@64-135-210-33.FoxValley.net) has left #beagle
  • [23:38:03] * TAKI (n=Administ@dslb-088-072-249-153.pools.arcor-ip.net) Quit (Read error: 104 (Connection reset by peer))
  • [23:38:24] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [23:40:28] * robtow (n=rtow@nat/montavista/x-bc7ece8b70c448c2) has joined #beagle
  • [23:43:27] * robtow1 (n=rtow@nat/montavista/x-b195347b967f36f6) has joined #beagle
  • [23:47:22] * bazbell (n=a0192809@nat/ti/x-69d556f25608eb94) has joined #beagle
  • [23:53:18] * JuanG (n=Juan@nat/ti/x-4a62bcf026a3896f) has joined #beagle
  • [23:53:22] * JuanG (n=Juan@nat/ti/x-4a62bcf026a3896f) has left #beagle
  • [23:54:28] * florian (n=fuchs@f048036030.adsl.alicedsl.de) Quit ("Verlassend")
  • [23:55:29] * uwe___ (n=uwe_@dslb-084-056-028-182.pools.arcor-ip.net) has joined #beagle