• [00:24:58] * hagisbasheruk (n=hagisbas@78.148.135.168) Quit ("Ex-Chat")
  • [00:46:30] * RogerMon1 (n=a0740758@nat/ti/x-e768768f00790a84) Quit (Remote closed the connection)
  • [00:46:59] <mru> hey guys
  • [00:47:12] <mru> someone mentioned mpeg2 decoding speed
  • [00:53:09] <ds2> I was asking about it earlier
  • [00:56:22] <mru> some trivial neon optimisations gives me almost 10% speed increase
  • [00:56:31] <mru> and that's without even touching the idct
  • [00:57:31] <ds2> what's the current frame rate? I saw the times in the thread
  • [00:58:17] <ds2> The TV out + DVI + local LCD would be a nice platform for Video (either ATSC or DVD like media)
  • [00:58:59] <mru> I'm getting 31 fps with the matrixbench mpeg2 sample
  • [00:59:22] <ds2> that's current w/o opt?
  • [00:59:30] <mru> oprofile says 30% or so is spent in idct
  • [00:59:50] <ds2> wonder if the DSP communications is cheap enough to just have it do IDCT
  • [00:59:57] <mru> I reckon that can easily be halved
  • [02:02:43] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [02:40:11] * Ragha_ (n=ragha@pool-71-96-158-8.dfw.dsl-w.verizon.net) has joined #beagle
  • [03:00:34] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [03:40:11] * JoeBorn__ (n=jborn@adsl-75-3-3-238.dsl.chcgil.sbcglobal.net) Quit (Read error: 104 (Connection reset by peer))
  • [03:42:27] * rsalveti (n=salveti@189.70.212.146) Quit (Read error: 104 (Connection reset by peer))
  • [03:43:09] * rsalveti (n=salveti@189.70.212.146) has joined #beagle
  • [03:47:42] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [04:00:19] * Ragha_ (n=ragha@pool-71-96-158-8.dfw.dsl-w.verizon.net) Quit ("Leaving")
  • [04:11:38] * Olipro (n=Olipro@unaffiliated/olipro) Quit (Read error: 104 (Connection reset by peer))
  • [05:14:38] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [05:58:50] * DaQatz (n=db@c-66-30-48-54.hsd1.nh.comcast.net) Quit (Nick collision from services.)
  • [05:59:12] * Lopoi (n=db@c-66-30-48-54.hsd1.nh.comcast.net) Quit (Client Quit)
  • [05:59:39] * DaQatz (n=db@c-66-30-48-54.hsd1.nh.comcast.net) has joined #beagle
  • [06:38:40] * roxfan (n=dunno@58.62-136-217.adsl-dyn.isp.belgacom.be) has joined #beagle
  • [06:45:29] * banderson__ (n=ilinux@69.71.183.7) has joined #beagle
  • [06:47:32] * banderson_ (n=ilinux@69.71.183.7) Quit (Read error: 110 (Connection timed out))
  • [06:53:09] * roxfan2 (n=dunno@137.223-200-80.adsl-dyn.isp.belgacom.be) Quit (Read error: 110 (Connection timed out))
  • [07:06:55] <khasim> all: Floating USB mini B to USB A cable works for me on Beagle
  • [07:07:18] <khasim> but Floating USB mini A to USB A doesn't boot completely
  • [07:08:19] <khasim> the reason is: We have OTG and ethernet gadget enabled, the USB stack is trying to send DHCP over this, as we have it floating we don't receive any responce.
  • [07:09:39] <khasim> If we disable ethernet gadget or select any other mode liek File-backed storage gadget then it boots fine for floating mini A cable as well
  • [07:27:44] * trickie|work (n=trickie@basesoft.demon.nl) has joined #beagle
  • [08:32:19] * RogerMonk (n=a0740758@nat/ti/x-c5bf06bc4b2d94aa) has joined #beagle
  • [10:16:19] * dmatt (i=3b5c74c4@gateway/web/ajax/mibbit.com/x-38930abddd146461) has joined #beagle
  • [11:07:17] * bazbell (n=a0192809@nat/ti/x-b6f6df40be1e439f) Quit (Remote closed the connection)
  • [11:46:14] * moenicke (n=moenicke@unaffiliated/moenicke) has joined #beagle
  • [11:46:26] <moenicke> mornin
  • [11:51:36] * bazbell (n=a0192809@nat/ti/x-b27ff89ecc84cb77) has joined #beagle
  • [11:58:46] * moenicke (n=moenicke@unaffiliated/moenicke) Quit (Remote closed the connection)
  • [12:41:49] * BThompson (n=BThompso@nat/ti/x-9b3639fa123e5072) has joined #beagle
  • [13:01:24] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
  • [13:11:58] <Crofton|work> koen, http://www.upi.com/Odd_News/2008/06/23/Many_Dutch_prepare_for_2012_apocalypse/UPI-40461214263554/
  • [13:12:16] <Crofton|work> do you have a life raft, or are you speculating in future beach front property
  • [13:18:01] * robclark (n=robclark@nat/ti/x-57cf3d84945a0bbc) has joined #beagle
  • [13:27:26] <Crofton|work> sakoman, http://rafb.net/p/SZqAT442.html
  • [13:27:40] <Crofton|work> I suspect I need another patch to make this work?
  • [13:27:51] * Olipro (n=Olipro@unaffiliated/olipro) has joined #beagle
  • [13:31:49] * Crofton|work hopes sakoman has not evacuated
  • [13:37:47] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [13:41:47] * dschaeffer (n=daniel@timesys-gw0.cust.expedient.net) has joined #beagle
  • [13:47:30] <dmatt> does the beagle board support image capture hardware
  • [13:52:52] * johnnybug is now known as jconnolly
  • [13:54:42] <ali_as> That's a bit non specific.
  • [13:56:53] <dmatt> security camera monitoring
  • [14:01:13] <dmatt> i dont think it provides. sorry
  • [14:02:06] * dmatt (i=3b5c74c4@gateway/web/ajax/mibbit.com/x-38930abddd146461) has left #beagle
  • [14:04:46] <sakoman> Crofton: no, no evacuation yet
  • [14:04:59] <sakoman> 2400 acres, no estimated containment
  • [14:05:35] * NishanthMenon (n=nmenon@nat/ti/x-d9d81eb57238bcd8) has joined #beagle
  • [14:06:08] <sakoman> they've renamed it from Bear Creek fire to Kirkman fire for some unknown reason
  • [14:06:23] <sakoman> Crofton: did you update your defconfig?
  • [14:08:21] <sakoman> should have:
  • [14:08:25] <sakoman> CONFIG_SND_OMAP_SOC=y
  • [14:08:25] <sakoman> CONFIG_SND_OMAP_SOC_MCBSP=y
  • [14:08:25] <sakoman> CONFIG_SND_OMAP_SOC_OMAP3BEAGLE=y
  • [14:08:25] <sakoman> CONFIG_SND_SOC_TWL4030=y
  • [14:09:08] <Crofton|work> updated some let me check
  • [14:09:26] * dmatt (i=3b5c74c4@gateway/web/ajax/mibbit.com/x-38930abddd146461) has joined #beagle
  • [14:10:20] <Crofton|work> sakoman, yes
  • [14:10:44] <sakoman> OK, let me try to reproduce the problem here
  • [14:11:13] <Crofton|work> I took the SoC patch and dropped it into .dev, bumped the rev to git head
  • [14:11:26] <sakoman> koen updated the OE recipe and I'm not sure I've tried a build with the current stuff
  • [14:11:27] <Crofton|work> and dropped the SmartReflex patches :)
  • [14:11:39] <Crofton|work> hmm, I haven't pushed this though
  • [14:12:09] <sakoman> oh, OK
  • [14:12:49] <sakoman> well, I'll do the same by hand
  • [14:13:22] <Crofton|work> I can push
  • [14:13:30] <Crofton|work> bbl need to go check on sick cat
  • [14:13:38] <sakoman> k
  • [14:27:02] * dmatt (i=3b5c74c4@gateway/web/ajax/mibbit.com/x-38930abddd146461) has left #beagle
  • [15:05:05] * dmatt (i=3b5c74c4@gateway/web/ajax/mibbit.com/x-38930abddd146461) has joined #beagle
  • [15:07:23] * robclark (n=robclark@nat/ti/x-57cf3d84945a0bbc) Quit ()
  • [15:09:09] <dante> dmatt, if you have a usb camera, you could probably hook that up.
  • [15:10:02] * robclark (n=robclark@nat/ti/x-fcf13b72f8e433fe) has joined #beagle
  • [15:11:54] <dmatt> i am thinking about the camera controller in omap
  • [15:11:54] * bazbel1 (n=a0192809@nat/ti/x-e2b3b07c5219fa16) has joined #beagle
  • [15:12:56] * bazbell (n=a0192809@nat/ti/x-b27ff89ecc84cb77) Quit (Remote closed the connection)
  • [15:16:16] * bazbel1 (n=a0192809@nat/ti/x-e2b3b07c5219fa16) Quit (Client Quit)
  • [15:22:42] * trickie|work (n=trickie@basesoft.demon.nl) Quit ("Leaving")
  • [15:34:45] <NishanthMenon> dmat http://www.beagleboard.org/uploads/Beagle_HW_Reference_Manual_A_5.pdf see page 118 the CSI2 and CPP lines are DNIed. but if you would like to use OMAP camera maybe you might be interested in Zoom MDK
  • [15:41:45] * NishanthMenon (n=nmenon@nat/ti/x-d9d81eb57238bcd8) Quit ("Ex-Chat")
  • [15:42:34] * robclark (n=robclark@nat/ti/x-fcf13b72f8e433fe) Quit (Remote closed the connection)
  • [15:42:51] * Bari|_ (n=ly@c-24-12-190-159.hsd1.il.comcast.net) has joined #beagle
  • [15:43:09] * robclark (n=robclark@nat/ti/x-b2008c90c8e15e93) has joined #beagle
  • [15:45:51] <sakoman> sadly I have not been able to reproduce either jkridner's or khasim's usb results:
  • [15:46:16] <sakoman> using 1.1.4 u-boot with my kernel still results in a hang at boot
  • [15:47:29] <sakoman> switching to usb host mode for musb (ie no OTG, no ethernet gadget) also still results in a hang at boot
  • [15:47:44] <sakoman> the latter was tested with both 1.1.4 and 1.3.3 u-boot
  • [15:51:40] <dmatt> thanks
  • [15:55:57] * Bari| (n=ly@c-24-12-190-159.hsd1.il.comcast.net) Quit (Read error: 110 (Connection timed out))
  • [15:55:57] * banderson__ (n=ilinux@69.71.183.7) has left #beagle
  • [16:04:25] * RogerMonk (n=a0740758@nat/ti/x-c5bf06bc4b2d94aa) Quit (Remote closed the connection)
  • [16:10:54] <sakoman> hmm . . . interesting new data point on musb in host mode
  • [16:11:35] <sakoman> it will still hang with an unconnected usb otg cable
  • [16:12:31] <sakoman> it will still hang with usb otg cable connected to iogear 4 port usb hub
  • [16:13:26] <sakoman> but it will boot and properly recognize devices with a Cyberpower 4 port hub!
  • [16:14:04] <sakoman> interestingly same kernel options on EVM works is all cases above
  • [16:15:33] <sakoman> s/is/in/
  • [16:36:58] * NishanthMenon (n=gnat@nat/ti/x-ca9a3a69a182da4f) has joined #beagle
  • [16:37:22] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [16:38:03] <NishanthMenon> symbian on beagle anyone? http://news.cnet.com/8301-13505_3-9975902-16.html http://www.linuxdevices.com/news/NS8056168201.html
  • [16:46:56] * dirk2 (n=dirk@F3146.f.strato-dslnet.de) has joined #beagle
  • [16:55:00] <koen> sakoman: that might explain why I'm not seeing failed boots with musb
  • [16:55:47] <sakoman> koen: indeed. what brand of hub are you using?
  • [16:57:58] <dirk2> koen: Have you already tried http://marc.info/?l=linux-omap&m=121429149109160&w=2 ? Sounds good. Maybe some more stability?
  • [17:06:01] <koen> sakoman: http://catalog.belkin.com/IWCatProductPage.process?Product_Id=198333
  • [17:06:10] <koen> dirk2: haven't rtied that yet
  • [17:07:21] <sakoman> koen: quite interesting that the iogear hub works with evm, but not beagleboard
  • [17:12:22] * bazbell (n=a0192809@nat/ti/x-94742338534903e9) has joined #beagle
  • [17:24:41] <sakoman> koen: have you ever tried to put beagle demo image in flash?
  • [17:24:47] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [17:24:52] <sakoman> nand that is
  • [17:25:22] <ds2> has anyone looked if there are significant differences in how the TWL4030 is initialized on the beagle vs the EVM? this sounds like the transceiver on there is unhappy about something
  • [17:25:56] <sakoman> ds2: not yet, it is on my "todo" list
  • [17:26:04] <koen> sakoman: I have console-image in NAND, but I don't use it yet
  • [17:26:12] <ds2> heh those never ending lists ;) heh
  • [17:26:40] <sakoman> ds2: no kidding!
  • [17:33:38] * robclark (n=robclark@nat/ti/x-b2008c90c8e15e93) Quit (Remote closed the connection)
  • [17:33:41] <dirk2> sakoman: I think to remember that you somewhere have a short "howto put something to beagle NAND and how to boot from it then". Do you like to resend the pointer again? ;)
  • [17:34:12] * robclark (n=robclark@nat/ti/x-2098c9a5339c81d2) has joined #beagle
  • [17:43:09] <sakoman> dirk2: http://www.sakoman.net/omap3/flash%20procedure.txt
  • [17:47:17] <dirk2> thanks
  • [17:48:35] <dirk2> sakoman: which linux terminal program do you at PC to send data for loadb?
  • [17:48:48] <sakoman> kermit
  • [17:49:13] <sakoman> you could load things with mmc too
  • [17:49:36] <sakoman> I use both methods depending on mood
  • [17:49:41] <dirk2> yes, I thought about this. Even for the large parts
  • [17:50:40] <dirk2> In my experience it's always better to write jffs2 rootfs with kernel jffs2 driver than with uboot
  • [17:55:45] <sakoman> dirk2: agreed. u-boot jffs2 writes result in a lot of warnings at boot
  • [17:56:53] <sakoman> mtd-utils tend to work a bit better
  • [18:23:28] <koen> mru: is you NEON stuff in mplayer svn or somewhere else online?
  • [18:25:28] * Crofton|work is annoyed koen can do stuff faster then he can ....
  • [18:25:35] <Crofton|work> multitasking is bad :)
  • [18:28:53] <Crofton|work> koen, did I remove the other patches properly?
  • [18:32:02] <koen> define properly
  • [18:32:05] * RogerMonk (n=a0740758@nat/ti/x-9709d21861d5f337) has joined #beagle
  • [18:35:44] <dirk2> I read the logs and wonder what's the USB OTG boot issue conclusion so far
  • [18:35:51] <dirk2> It seems to depend on uboot version (TI 1.1.4 vs git 1.3.3), the cable and the termination (e.g. hub)
  • [18:39:30] <sakoman> dirk2: seems different people get different results
  • [18:40:15] <sakoman> my results seem to be independent of u-boot version
  • [18:40:39] <sakoman> for evm: naked cable and both brands of hub always work
  • [18:40:47] <koen> dvb-usb: Hauppauge Nova-T Stick successfully initialized and connected.
  • [18:40:58] * koen wonders how well the cortex can decode mpeg2 :)
  • [18:41:18] <dirk2> okay, the uboot version thing was introduced by jkridner
  • [18:41:32] <sakoman> for beagle: naked cable always fails, iogear hub always fails CyberPower hub always works
  • [18:41:33] <dirk2> or I misunderstood it ;)
  • [18:42:26] <koen> belkin also always works
  • [18:42:39] <sakoman> so my conclusion is that there is something different about beagle setup (probably sw) that causes it to fail with naked cable and iogear
  • [18:42:54] * Crofton|work wonders if he should go to Best Buy, or somewhere that stocks Belkin ...
  • [18:43:06] <ds2> Hmmmm
  • [18:43:10] <koen> I bought mine from store.apple.com
  • [18:43:24] <Crofton|work> crazy apple fans :)
  • [18:43:29] <ds2> does a mini-B connected to it cause it to fail?
  • [18:45:21] <dirk2> NishanthMenon: ping
  • [18:46:50] <sakoman> ds2: I always get messed up with mini-A/B. The one that terminates in an A receptacle fails (ie the OTG adaptor), the one that terminates in an A plug does not fail
  • [18:47:11] <ds2> gotcha
  • [18:47:51] <ds2> what that says is if the ID pin on the connector is grounded, things become upset
  • [18:48:16] <sakoman> yup
  • [18:48:36] <ds2> wonder if there is an expectation of current draw on the other pins or if the PHY confused if it sees something grounded but the D-/D+ are still in the default states
  • [18:49:14] <ds2> is the cyberpower hub the self powered (battery included) hub?
  • [19:02:05] <dirk2> Do we have schematic of EVM? Then we could compare ID pin connection of Beagle with EVM. Beagle OTG ID pin schematic: http://www.beagleboard.org/uploads/Beagle_HW_Reference_Manual_A_5.pdf page 56 figure 33
  • [19:03:57] <dirk2> And while talking about possible uboot difference (removed logo and audio): http://www.sakoman.net/cgi-bin/gitweb.cgi?p=u-boot-omap3.git;a=commitdiff;h=af115b1c0d3423f9f0c707969fa241d29e8885d5
  • [19:04:11] <dirk2> There are a lot of uncommented registers writes :(
  • [19:17:08] * dirk2 (n=dirk@F3146.f.strato-dslnet.de) has left #beagle
  • [19:17:46] <sakoman> ds2: no the cyberpower hub is plugged into the wall
  • [19:18:00] <ds2> what about the other hubs? are they plugged into the wall also?
  • [19:18:05] <sakoman> yes
  • [19:18:48] <ds2> not sure how to go about answer this but - is the pull up/downs on the hub driven by the Vbus line or by the local power source? (for each hub)
  • [19:19:19] <ds2> I know some of the cheaper hubs will violate the USB spec and drive the pull's with local power (useful if you are using a Vbus-less setup like the Nokia 770's)
  • [19:19:39] * BThompson (n=BThompso@nat/ti/x-9b3639fa123e5072) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [19:20:09] <sakoman> ds2: I have no idea!
  • [19:21:13] <ds2> okay - what about trying this: instead of just having a dangling cable, put a cheap dollar store variety USB LED lamp on there
  • [19:21:31] <ds2> current draw should be minimal but those things only pull from Vbus and not the D+/- lines
  • [19:21:50] <sakoman> ds2: don't have one of those :-(
  • [19:22:12] <ds2> sakoman: any other USB powered things like a fan handy?
  • [19:22:25] <sakoman> no, no gadgets at all :-)
  • [19:22:44] <ds2> worse case, you can make one.. get a dollar store cable, chop it up, and put a 470ohm resistor between Vbus and Gnd
  • [19:22:53] <ds2> (or an LED if that's handy)
  • [19:23:28] <sakoman> i've tried directly connecting thumb drives to the cable and they do not work
  • [19:23:58] <ds2> Oh? so it is just just a dangling cable that causes problems? thumbdrives on there will cause the hang also?
  • [19:24:01] <koen> http://amethyst.openembedded.net/~koen/beagleboard/uImage-2.6.25+2.6.26-rc7+r29+git6604ac6c8a14d583a463159f3a1601427dab7e05-r29-beagleboard.bin
  • [19:24:10] <koen> 2.6.26rc7 + NAND + sound
  • [19:24:21] <sakoman> next time I go out I'll pick up a usb lamp
  • [19:25:12] <ds2> of course if jkridner is right, and there is no Vbus power at all...
  • [19:25:20] <Crofton|work> koen, I just built one of thise
  • [19:25:39] <Crofton|work> alsamixer runs, but I get i2c timeouts when I try to adjust the volume
  • [19:25:43] <ds2> is the mini-AB connector accessible enough to probe it and see if it has Vbus?
  • [19:26:09] <sakoman> ds2: sure do you know which pins?
  • [19:26:31] <ds2> sakoman: probe them all.. ground to all 5 pins... one of them should register about 4.5V-5.5V
  • [19:26:40] <ds2> the rest will be 0V or around 3.3V
  • [19:29:06] <keesj> prpplague: I have my FlySwatter
  • [19:29:15] <prpplague> keesj: good deal
  • [19:29:43] * hagisbasheruk (n=hagisbas@78.148.135.168) has joined #beagle
  • [19:29:48] <prpplague> keesj: when will you be done with adding cortex-a8 support to openocd? hehehe
  • [19:29:56] <keesj> haha
  • [19:30:44] <keesj> I have a few steps to take first
  • [19:30:48] <ds2> in case you need the pin #, Gnd to Pin 1 should be vbus
  • [19:31:21] <keesj> first get openocd working on a few arm11 boards would be great, get the hang of it
  • [19:33:36] <sakoman> ds2: OK on beagle pin 1 is .45 volts, on EVM pin 1 is 5 volts
  • [19:33:40] * moenicke (n=moenicke@unaffiliated/moenicke) has joined #beagle
  • [19:33:45] <ds2> whoa
  • [19:33:47] <sakoman> the other 3 pins are 0v
  • [19:33:49] <ds2> there is no Vbus
  • [19:33:52] <ds2> yikes
  • [19:34:25] <sakoman> is vbus provided by the PHY?
  • [19:34:29] <ds2> the lack of Vbus might be the problem
  • [19:34:32] <ds2> normally, yes
  • [19:34:44] <ds2> or pin 1 can be directly wired to a 5V source
  • [19:34:54] <sakoman> might be a 4030 config problem, I'll look into that
  • [19:35:06] <sakoman> I have the evm schematic
  • [19:35:33] <ds2> or could be the board... jkridner thinks there is no Vbus but I don't see why if there is provision of for the EHCI host only --- that implies that is 5V on the board somewhere
  • [19:36:01] <ds2> (or are there some cheating going on?) sorry, don't have time tos tudy the schematics in detail... might be a thing for the 4th
  • [19:36:32] <sakoman> vbus on evm comes from vbus pin of ISP1504 phy acording to the evm schematic
  • [19:37:23] <ds2> is R43 populated?
  • [19:37:48] <ds2> the beagle manual page 120 (PDF reader #) shows a 0 ohm jumper from the triton to the Vbus
  • [19:39:05] * Olipro (n=Olipro@unaffiliated/olipro) Quit (Read error: 104 (Connection reset by peer))
  • [19:39:15] <sakoman> r43 on beagle is loaded with 0 ohm
  • [19:39:35] <ds2> Hmmm maybe there just a lack of 5V on the board?
  • [19:43:14] <sakoman> well, it seems strange that it is .45 volts
  • [19:43:57] <hagisbasheruk> so how is it progressing then Crofton
  • [19:44:35] <ds2> that seems like a diode drop
  • [19:49:23] <sakoman> maybe d3 is loaded backwards
  • [19:50:05] <sakoman> hmmm . . . I take that back
  • [19:50:24] <ds2> where's D3 go?
  • [19:50:35] <sakoman> vbus to ground
  • [19:50:55] <sakoman> looks like transient suppression device
  • [19:51:03] <ds2> ah yes I found it
  • [19:51:04] <sakoman> page 116
  • [19:51:10] <koen> Crofton|work, sakoman: I've added Pauls DPLL patches to OE
  • [19:51:15] <ds2> nice that the manual is searchable unlike thoe #@$@!#$%$#@@!#$#@@# EVM docs
  • [19:52:05] <ds2> wth... page 56 - R43 is shown as 10K!?
  • [19:52:32] <ds2> lunch time bbl... might be useful to try to follow where the 5V comes from (or should come from)
  • [19:52:52] <Crofton|work> how many more patches are yu going to add?
  • [19:53:09] <sakoman> ds2: I measure .44 v on both sides of r43
  • [19:53:18] <sakoman> and it has a 0 marking
  • [19:53:28] <ds2> so the manual is wrong :/
  • [19:54:19] <koen> Crofton|work: I think I'm thru my patch queue
  • [19:54:26] <Crofton> ok
  • [19:55:58] <sakoman> ds2: still there?
  • [20:11:10] <bazbell> sakoman: Are you trying to get VBUS enabled for the USB Host port on P7?
  • [20:11:32] <sakoman> perhaps :-)
  • [20:12:01] <koen> grrrrrr
  • [20:12:04] <sakoman> it seems to be a difference between the working evm and the non-working beagle :-)
  • [20:12:07] <koen> dvb breaks musb dma
  • [20:13:22] <sakoman> bazbell: do you have any wisdom for me?
  • [20:14:14] <bazbell> The voltage to VBUS on P7 is controlled by a current-limited power distribution switch (TPS2061) which is enabled by LEDA/VIBRA.P on the TWL4030. You need to set this to logic 0.
  • [20:18:52] <sakoman> bazbell: I'm sorry! Actually we are trying to get VBUS for P1 not P7!
  • [20:19:07] <sakoman> I misread your initial question
  • [20:19:34] <sakoman> So we are trying to use musb as a host or otg port
  • [20:19:55] <sakoman> On the evm things are configured to deliver 5V on pin 1
  • [20:20:05] <sakoman> On beagle pin 1 is 0.44 volts
  • [20:20:35] <sakoman> On evm host/otg works, on beagle it doesn't
  • [20:20:59] <sakoman> Our suspicion is that it may be realted to the improper VBUS voltage
  • [20:23:58] * hagisbasheruk (n=hagisbas@78.148.135.168) Quit ("Ex-Chat")
  • [20:25:34] <koen> Top causes for wakeups: 70.6% ( 40.0) <interrupt> : DMA 16.5% ( 9.3) <interrupt> : gp timer
  • [20:25:55] <koen> DMA does 80% of the wakeups
  • [20:26:05] * koen wonders what it's DMA'ing
  • [20:36:34] * Bari|_ is now known as Bari|
  • [20:36:39] <bazbell> So by using an OTG configuration, you are supplying power to the beagle board via the DC_POWER J1, correct? Just checking and sorry in advance if this is a dumb question.
  • [20:36:55] * DaQatz (n=db@c-66-30-48-54.hsd1.nh.comcast.net) Quit (Connection timed out)
  • [20:39:57] <ds2> sakoman: back now
  • [20:40:00] <Crofton|work> bazbell, yes. sakoman has a rev B board and some of us with rev A have fed power into the pads where the connector was supposed to go
  • [20:40:08] * Bari| is now known as Bari\
  • [20:40:20] * BThompson (n=BThompso@nat/ti/x-30f6fa8d7bcc120c) has joined #beagle
  • [20:41:39] * Bari\ is now known as bari-
  • [20:41:52] * bari- is now known as bari_
  • [20:41:58] * bari_ (n=ly@c-24-12-190-159.hsd1.il.comcast.net) Quit ("left")
  • [20:43:52] <sakoman> bazbell: correct, power via J1
  • [20:44:08] <sakoman> sorry, went outside for a fire check
  • [20:44:23] <Crofton> how close is it?
  • [20:44:51] <sakoman> Crofton: couple of miles
  • [20:45:01] <ds2> are you just walking out the door or you are climbing up on a tower and surveying the area with a binocular or even an IR camera? :)
  • [20:45:23] * Bari| (n=ly@c-24-12-190-159.hsd1.il.comcast.net) has joined #beagle
  • [20:45:37] * Bari| is now known as Bari\
  • [20:45:39] <sakoman> ds2: walking outside & down about 800' of driveway
  • [20:45:56] <ds2> oh nothing too sophisticated
  • [20:45:57] <sakoman> I live out in the middle of nowhere
  • [20:46:04] <Crofton> you should make a tower
  • [20:46:14] <sakoman> in my spare time :-)
  • [20:46:19] <ds2> I kind of had the image of you climbing up an observation tower
  • [20:46:25] <sakoman> heh
  • [20:46:33] <ds2> like what the FS uses
  • [20:46:41] <sakoman> much lower tech
  • [20:46:51] <ds2> and keeping an hourly log of what is observed at what angles
  • [20:47:36] <sakoman> it's close enough that it will be fairly obvious if it heads this way
  • [20:48:01] <ds2> got a vehicle all gased and packed ready to go?
  • [20:48:32] <sakoman> listening to the scanner. I feel so bad for the firefighters. they are stretched so thin there isn't anyone to bring them MRE's for lunch
  • [20:48:47] * Crofton thinks that may be a good thing ....
  • [20:48:57] <sakoman> I hear them requesting more resources and the incident commander telling them to make do
  • [20:49:02] <Crofton> urg
  • [20:49:03] <koen> sakoman: no gumstix based UAVs to drop off lunches?
  • [20:49:05] <Crofton> that is not good
  • [20:49:18] <ds2> what about the local ham groups or RACES?
  • [20:49:27] <sakoman> shoot, I'd bring them luch if they would let me back in there
  • [20:49:42] <Crofton|work> much better then MRE's
  • [20:49:52] <sakoman> they are wisely keeping people out of the area
  • [20:50:16] <ds2> isn't this a bit early for fires this year?
  • [20:50:22] <sakoman> very
  • [20:50:38] <sakoman> and to make things worse it is 100F and about 15% RH
  • [20:51:04] <Crofton|work> if they can put it out, it should minimize your risk for the next few years ...
  • [20:51:18] <sakoman> from that direction, yeah :-)
  • [20:51:38] <Crofton|work> "look on the bright side of life"
  • [20:52:14] <sakoman> too smokey to find the bright side today ;-)
  • [20:52:33] <sakoman> sun is just a dim red ball
  • [20:55:14] <ds2> things are smokey down here in the bay area today but the good thing is temp is down
  • [20:57:09] <sakoman> looks pretty much like this: http://www.redding.com/photos/galleries/2008/jun/22/shingletown-fire/5964/
  • [20:58:08] <bazbell> sakoman: If 0.44V is measured on both sides of R43, I would say that the TWL4030 is not configured to enable the USB charge pump for VBUS (pin R8 of TWL4030). The bit 5 of register address 0x0A needs to be set to logic 1 to enable the charge pump.
  • [20:58:19] <Crofton|work> ok, last beagle boot for the day
  • [20:58:48] <koen> rc7 with pauls patches is a lot more stable than earlier kernels
  • [20:58:58] <sakoman> bazbell: thanks! I was just searching the evm uboot to see where that was done
  • [20:59:40] <sakoman> I'll add that to u-boot and see if the situation improves
  • [21:00:03] <bazbell> sakoman: I will log off now, but should be back in a couple of hours.
  • [21:00:23] <sakoman> bazbell: thanks for the help
  • [21:00:38] * bazbell (n=a0192809@nat/ti/x-94742338534903e9) Quit (Remote closed the connection)
  • [21:03:13] <sakoman> koen: I look forward to a stable beagle! it's been quite frustrating for a while
  • [21:04:06] * DaQatz (n=db@c-66-30-48-54.hsd1.nh.comcast.net) has joined #beagle
  • [21:04:30] <ds2> interesting, did not know the twl4030 has a charge pump
  • [21:05:08] <koen> sakoman: it's less annoying than going over a gazillion jumpers on a blackfin board to debug a kernel boot crash
  • [21:05:39] * koen hates "unbreak me" switches
  • [21:05:41] <Crofton|work> bother
  • [21:05:48] <Crofton|work> Uncompressing Linux............................................................................................................................................ done, booting the kernel.
  • [21:05:52] <Crofton|work> is all I am getting
  • [21:06:10] <koen> try enabling debug_ll again
  • [21:06:21] <Crofton|work> it went that time
  • [21:06:27] <Crofton|work> after a warm boot
  • [21:06:54] * moenicke (n=moenicke@unaffiliated/moenicke) Quit ()
  • [21:07:02] <sakoman> ds2: uses vbatt to generate vbus
  • [21:07:34] <sakoman> will provide +5 for USB till vbatt gets down to about 2.7 V
  • [21:07:43] <sakoman> pretty handy1
  • [21:07:51] <Crofton|work> awesome, no i2c timeouts when I adjust volume at this time
  • [21:08:20] <Crofton|work> this moves boost with gcc 4.3 up my list
  • [21:09:47] <sakoman> Crofton: do you actually get audio out too?
  • [21:10:01] <Crofton> I haven;t connected speakers
  • [21:10:20] <Crofton> what's an easy way to make noise?
  • [21:11:26] <ali_as> Clap?
  • [21:11:44] * khilman (n=khilman@deeprooted.net) has joined #beagle
  • [21:12:55] <koen> Crofton: cat /dev/urandom > /dev/dsp
  • [21:12:55] <Crofton|work> http://rafb.net/p/Rm0H8f45.html
  • [21:13:11] <Crofton|work> cat /dev/random > /dev/audio didn't work so well :(
  • [21:13:30] <ali_as> You are supposed to say "from the beagle." and then I can say "put it in a blender".
  • [21:14:31] <Crofton|work> rebooting did this ....
  • [21:14:45] <Crofton|work> http://rafb.net/p/Z8sxPa93.html
  • [21:16:05] * BThompson (n=BThompso@nat/ti/x-30f6fa8d7bcc120c) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [21:17:41] <sakoman> Crofton: isn't that the same issue you saw earlier today?
  • [21:18:17] <Crofton|work> not sure
  • [21:18:27] <Crofton|work> I'm just pasting them as they appear
  • [21:18:40] <Crofton|work> urandom to dsp also crashed
  • [21:19:01] <koen> irq56 is a sign of broken boot
  • [21:19:23] <sakoman> yeah, irq56 has nothing to do with audio
  • [21:19:49] <Crofton|work> yeah
  • [21:19:52] <Crofton|work> same problem
  • [21:20:03] <Crofton|work> checked scroll back for earlier paste
  • [21:20:43] <Crofton|work> http://rafb.net/p/Rm0H8f45.html does seem sound related
  • [21:22:13] <sakoman> I have no idea what the expected behavious for writing to /dev/audio is
  • [21:22:19] <sakoman> try /dev/dsp
  • [21:22:39] <ds2> Ooops is not it :)
  • [21:22:53] <Crofton|work> same problem
  • [21:23:00] <ds2> SoC driver or the normal one?
  • [21:24:06] <ds2> check the McBSP code, I think there is some stale code that did not get updated at the last McBSP cleanup
  • [21:41:17] * NishanthMenon (n=gnat@nat/ti/x-ca9a3a69a182da4f) Quit ("For Sale: Parachute. Only used once, never opened, small stain.")
  • [21:42:43] * dschaeffer (n=daniel@timesys-gw0.cust.expedient.net) Quit ("Leaving.")
  • [21:43:57] * NishanthMenon (n=nmenon@nat/ti/x-3e3e376b87f923af) has joined #beagle
  • [21:47:52] * bazbell (n=a0192809@nat/ti/x-60342f20b3e6d3a5) has joined #beagle
  • [21:53:41] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
  • [22:02:08] <NishanthMenon> kinda late to say pong to dirk2.. :( icechat hung on me.. :(...
  • [22:04:11] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) has joined #beagle
  • [22:30:51] * JonHunter (n=a0741266@nat/ti/x-11854ebbbe8272e5) has joined #beagle
  • [22:33:41] * Olipro (n=Olipro@unaffiliated/olipro) has joined #beagle
  • [22:43:29] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [22:46:01] * JonHunter (n=a0741266@nat/ti/x-11854ebbbe8272e5) has left #beagle
  • [22:53:52] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  • [22:58:04] * RogerMonk (n=a0740758@nat/ti/x-9709d21861d5f337) Quit (Remote closed the connection)
  • [23:06:25] <sakoman> ds2: after looking at vbus with a scope I am convinced that the usb issue is hw!
  • [23:07:11] <sakoman> when linux boots the usb code does turn on the charge pump on the 4030 to generate vbus
  • [23:07:58] <sakoman> my cheap dvm was reading 5.2 V (as expected) for the EVM and 6.3 V for the beagle!
  • [23:08:44] <sakoman> I then hooked up the scope. EVM is a nice clean 5.2V. Beagle has several volts of AC noise riding on the DC
  • [23:10:25] <sakoman> when the cyberpower hub is connected, vbus immediately settles to 5 V -- they likely have a cap on vbus input
  • [23:11:05] <sakoman> with the iogear hub vbus stays ugly -- still huge voltage swing!
  • [23:11:49] <sakoman> so my theory is that beagle is missing sufficient charge pump output filter caps
  • [23:20:15] <ali_as> Is this just OTG related, nothing to do with the other port?
  • [23:31:50] * cwz (n=chris@mekaneck.topdogpc.com) has joined #beagle
  • [23:42:30] <Crofton> yeah, the ehci host port apparently has bad phy issues :(
  • [23:47:28] <sakoman> to add a bit more data, EVM has 4.7uF + 0.1uF caps on vbus. beagle has no caps as far as I can tell
  • [23:47:49] * robclark (n=robclark@nat/ti/x-2098c9a5339c81d2) Quit ()
  • [23:48:06] <sakoman> After dinner I'll solder a cap on vbus and hopefully that will put this to bed
  • [23:49:49] <Crofton> don't forget to climb the tower ...
  • [23:50:11] <cwz> the host connector on beagle doesn't have any series resistors for DM/DP either
  • [23:51:19] <cwz> neither does the OTG port :(
  • [23:52:35] <cwz> For VBUS, there's typically a small (eg. 1 ohm) resistor in series with the caps to prevent ringing -- just in case anyone who has edit capability to the schematic is listening
  • [23:53:04] <sakoman> cwz: yeah I noticed that too
  • [23:53:50] <sakoman> off to dinner
  • [23:57:24] <ds2> or put a small load on the... say 1K
  • [23:59:34] <cwz> a small load on what?