**** BEGIN LOGGING AT Sun Jun 22 02:59:58 2014
Jun 22 06:15:21 <karki> any PRU - timer examples flying about?
Jun 22 07:14:54 <Abhishek_> mranostay: ping
Jun 22 07:19:18 <mranostay> Abhishek_: never mind got it to load
Jun 22 14:32:13 <Abhishek_> Final series of cleanups & fixes to BeagleLogic module: http://git.io/Z2V8ng
Jun 22 16:09:18 <karki> panto : how do I work with the timer in the PRU? I need to make my wait() timer based.
Jun 22 16:11:26 <Abhishek_> karki: Have you looked at section 9.6 of the PRU reference guide?
Jun 22 16:11:47 <Abhishek_> err, just section 9
Jun 22 16:21:40 <Abhishek_> jkridner: Is the timing of tomorrow's BotSpeak hangout shown on our calendars confirmed for tomorrow?
Jun 22 16:22:25 <Abhishek_> jkridner: And do we have a meeting at 0130 EDT tomorrow?
Jun 22 16:29:50 * karki forgot about the meeting! :p
Jun 22 16:35:15 <Abhishek_> karki: how's your health now?
Jun 22 16:35:30 <karki> It's okay :)
Jun 22 16:35:46 <karki> I'm more or less back to normal.
Jun 22 16:36:13 * karki managed to give the rest of his family dengue ;)
Jun 22 16:36:27 <Abhishek_> has the platelet count normalized yet?
Jun 22 16:38:05 <karki> idk! should have. I stopped taking blood tests after the count started increasing (and was very close to normal)
Jun 22 16:39:10 <karki> Abhishek_ : looks like I will be using the IEP for the timer stuff!
Jun 22 16:39:40 <karki> I had looked at it at the morning, just didn't catch on to it.
Jun 22 16:40:08 <karki> going through panto's code, lookes like even he used IEP for the timer stuff!
Jun 22 16:40:17 <karki> *looks
Jun 22 16:40:38 * Abhishek_ has implemented hard cycle-counted delay loops for the BeagleLogic sample loop
Jun 22 17:52:23 <jkridner> hi Abhishek_
Jun 22 17:52:29 <jkridner> 1:30 AM EDT tonight.
Jun 22 17:52:57 <Abhishek_> hi jkridner
Jun 22 18:01:04 * Abhishek_ has cleaned up the BeagleLogic repo (removed previous libprussdrv example and code)
Jun 22 18:52:01 <jkridner> Abhishek_: would be nice if you had a single script to do the setup/build.
Jun 22 18:52:15 <jkridner> you don't even give links to the specific repos for the dependent projects.
Jun 22 18:52:38 <Abhishek_> jkridner: I'm updating documentation just now
Jun 22 18:53:22 <Abhishek_> Writing build instructions and uploading the bin files containing the device tree and ELF for the PRUs to be loaded in
Jun 22 18:54:09 <Abhishek_> [It's all in the wiki on my GitHub repo]
Jun 22 18:54:40 <jkridner> Then make sure you point to that in your README.md
Jun 22 18:54:52 <Abhishek_> yup, that is to be updated too
Jun 22 18:55:43 <Abhishek_> jkridner: which kernel are you running on your BBB?
Jun 22 18:56:02 <Abhishek_> (I mean: rcn-ee's kernel orkoen's kernel)
Jun 22 18:56:08 <jkridner> 3.8.x from http://github.com/beagleboard/kernel
Jun 22 18:56:34 <jkridner> rcn-ee's. people still using koen's kernels a lot?
Jun 22 18:56:49 <jkridner> I mean, koen did most of the legwork for the current rcn-ee kernels....
Jun 22 18:57:02 <jkridner> but, I didn't know people were still using the old ones or koen's latest developments
Jun 22 18:57:11 <Abhishek_> I am currently running koen's kernel [for mranostay's lighting examples]
Jun 22 18:57:34 <Abhishek_> will switch back to rcn-ee's soon; preparing a patch set for inclusion
Jun 22 18:59:05 <Abhishek_> jkridner: could you send me /drivers/remoteproc/pru_rproc.c from your kernel tree?
Jun 22 18:59:33 <Abhishek_> (the ones in koen's kernel and rcn-ee's are different)
Jun 22 18:59:44 <jkridner> I'm not currently rebuilding the kernel I'm running.
Jun 22 18:59:53 <Abhishek_> k
Jun 22 19:00:25 <jkridner> currently build 3.8.13-bone50
Jun 22 19:09:15 <Abhishek_> I'll send you a patchset which should apply on top of the bone50 and which should enable BeagleLogic support in the kernel
Jun 22 19:15:20 <mranostay> Abhishek_: so which PRU pins are you using?
Jun 22 19:15:41 <Abhishek_> P8_39 to P8_46
Jun 22 19:16:28 <Abhishek_> that's mentioned in the device tree file
Jun 22 19:21:41 <Abhishek_> https://github.com/abhishek-kakkar/BeagleLogic/wiki/Getting-Started
Jun 22 22:23:06 <Abhishek_> jkridner: I'm sending you a patchset for the kernel to enable BeagleLogic support
Jun 22 22:23:38 <jkridner> great. send it to the beagleboard@... list.
Jun 22 22:28:53 <jkridner> karki: does pru0_firmware need to be in /lib/firmware?
Jun 22 22:29:08 <jkridner> I get a kernel error when I try to 'rmmod pru_speak'.
Jun 22 22:29:55 <jkridner> Abhishek_: the direct message window for me is an annoyance.
Jun 22 22:30:01 <jkridner> (too many windows)
Jun 22 22:30:25 <Abhishek_> jkridner: the pru_rproc driver on which that driver is based cannot be unloaded without rebooting the system
Jun 22 22:30:42 <jkridner> Abhishek_: ugh. :(
Jun 22 22:30:51 <jkridner> I wonder when the firmware is loaded.
Jun 22 22:30:54 <Abhishek_> I separated my beaglelogic code out of pru_rproc for this very reason
Jun 22 22:31:13 <jkridner> can yours be unloaded?
Jun 22 22:31:13 <Abhishek_> the firmware is loaded when the device tree overlays are loaded
Jun 22 22:31:19 <jkridner> ah, k.
Jun 22 22:31:44 <Abhishek_> the firmware can be reloaded using a sysfs attr
Jun 22 22:32:14 * Abhishek_ just tested his BeagleLogic code by compiling it into the kernel
Jun 22 22:33:03 <jkridner> karki: looks like the file should be /lib/firmware/pru_firmware for now?
Jun 22 22:33:13 <jkridner> Abhishek_: *blech*
Jun 22 22:33:34 <jkridner> recompiling kernel to test out your code is not a great experience.
Jun 22 22:33:58 <Abhishek_> I know, just trying it out as mranostay was having some issues
Jun 22 22:34:38 <Abhishek_> it seems to work for me correctly in spite of that
Jun 22 22:35:09 <jkridner> *ugh* karki, I wasn't able to unload your DT file. :(
Jun 22 22:47:25 <jkridner> Abhishek_: so, how are you reloading firmware?
Jun 22 22:47:37 * jkridner wonders if karki is doing the same
Jun 22 22:47:44 <Abhishek_> yup, it's the same
Jun 22 22:48:08 <Abhishek_> echo 0:beaglelogic-pru0,1:beaglelogic-pru1 > /sys/devices/ocp.3/4a300000.prurproc/load
Jun 22 22:48:42 <jkridner> k. and those need to be in /lib/firmware?
Jun 22 22:48:48 <Abhishek_> yep
Jun 22 22:49:11 <Abhishek_> the PRUs will be reset after FW is written into it
Jun 23 01:49:10 <jkridner> Abhishek_: have you seen "omap_hwmod: pruss: failed to hardreset"?
Jun 23 01:49:26 <Abhishek_> jkridner: That is normal
Jun 23 01:50:06 <Abhishek_> occurs whenever the device tree overlay is loaded
Jun 23 01:53:36 <jkridner> karki: http://paste.debian.net/106312
Jun 23 01:53:46 <jkridner> karki: I don't seem to be able to get the PRU firmware to respond.
Jun 23 01:54:43 <Abhishek_> jkridner: why is pruss_uio being loaded?
Jun 23 01:55:14 <jkridner> because I'm trying to run cape-universal in parallel and it seems to load that... I then promptly unload it and load pruspeak.ko
Jun 23 01:55:58 <jkridner> there's a chance some things accidentally fell out-of-order.
Jun 23 01:56:02 <Abhishek_> you might have to prevent it from loading on startup.
Jun 23 01:58:21 * Abhishek_ has updated the README of BeagleLogic and created the BeagleLogic wiki at https://github.com/abhishek-kakkar/BeagleLogic/wiki
Jun 23 01:58:46 <jkridner> Abhishek_: why is that? shouldn't the rproc driver clean up the state of the PRU?
Jun 23 01:59:28 <Abhishek_> I don't really know, panto would be able to shed more light on this
Jun 23 01:59:59 <Abhishek_> or it might even be the uio_pruss driver at fault, if it doesn't clear things up
Jun 23 02:00:03 <jkridner> Abhishek_: k. I think using cape-universal for all of the pinmuxing is better than having custom .dts files for all of that.
Jun 23 02:01:16 <Abhishek_> hmm yes, though I'll have to look into how BeagleLogic reacts to the universal cape
Jun 23 02:01:29 <jkridner> Abhishek_: that would be very helpful.
Jun 23 02:02:01 <Abhishek_> and I think the solution might be in modifying the universal cape
Jun 23 02:02:51 <Abhishek_> jkridner: this one? https://github.com/jadonk/beaglebone-universal-io
Jun 23 02:03:17 <jkridner> https://github.com/cdsteinkuehler/beaglebone-universal-io is more upstream
Jun 23 02:03:48 <jkridner> not sure if I ever made any changes...
Jun 23 02:04:07 <Abhishek_> jkridner: could you try removing this line and reinstalling the universal cape? https://github.com/cdsteinkuehler/beaglebone-universal-io/blob/master/cape-universal-00A0.dts#L156
Jun 23 02:04:27 <jkridner> I've tried to switch to cape-universal for pruspeak: https://github.com/jadonk/pruspeak/commit/ee119d1b50ab98cdfc240a25ac38cce278f9cb96
Jun 23 02:05:17 <jkridner> Abhishek_: this too, no https://github.com/cdsteinkuehler/beaglebone-universal-io/blob/master/cape-universal-00A0.dts#L1719 ?
Jun 23 02:05:38 <Abhishek_> yup, I think so
Jun 23 02:05:59 <Abhishek_> perhaps that is what triggers the loading of the uio_pruss driver
Jun 23 02:19:26 <DiegoTc> jkridner, is to confirm. I am going to display the tutorials this way: /bone101/Support/GSOC/app/views/tutorial.html?gistid=10611021
Jun 23 02:20:54 <jkridner> I don't like making it a query, personally
Jun 23 02:21:18 <jkridner> I also don't understand what the app/views/tutorial.html adds for value
Jun 23 02:21:44 <jkridner> also, why do a variable=? ie., gist=
Jun 23 02:21:59 <DiegoTc> app/view was because I was working on node, haven't change that
Jun 23 02:22:14 <jkridner> but it isn't needed, right? just legacy?
Jun 23 02:22:25 <DiegoTc> correct
Jun 23 02:22:44 <DiegoTc> I have to remove that, but haven't had time on doing that. Not such importance right now
Jun 23 02:23:16 <jkridner> If it was possible, I'd get rid of the # or ? when pointing to the top of the deck, but I don't see a good way to do that without some (possibly ugly) server hacks.
Jun 23 02:24:17 <DiegoTc> jkridner, the variable on query is for sharing: e.g hey check this new tutorial: /bone101/Support/GSOC/app/views/tutorial.html?gistid=10611021
Jun 23 02:26:14 <jkridner> What is wrong with /bone101/Support/Tutorial#10611021 ?
Jun 23 02:26:46 <jkridner> hash vs. query
Jun 23 02:27:05 <DiegoTc> none.
Jun 23 02:27:26 <DiegoTc> that's why I asked, I see you share me something last time
Jun 23 02:27:43 <DiegoTc> thanks :)
Jun 23 02:28:04 <jkridner> I'm just trying to think it out... I really don't  know which is better and thought you might want to say why one was better than the other.
Jun 23 02:28:38 <DiegoTc> I have always work with query
Jun 23 02:28:52 <DiegoTc> but there's not a special reason too
Jun 23 02:28:55 <jkridner> for me, /bone101/Support/Tutorial/10611021 would be ideal. Maybe that is a simple change that could be made to the beagleboard.org server? it is not running node.js however.
Jun 23 02:29:11 <jkridner> .... and support on static servers is still required.
Jun 23 02:33:07 <Abhishek_> jkridner: Upcoming patch in your inbox
Jun 23 02:34:54 <DiegoTc> jkridner, you're the boss! Which one them. 3 options sounds good. But want to know which one will be the best option
Jun 23 02:36:52 <jkridner> DiegoTc: I wouldn't get too hung up on it as long as the URLs look clean for now and can be shared easily, including referencing an individual card.
Jun 23 02:37:56 <DiegoTc> ok, thanks jkridner
Jun 23 02:40:00 <DiegoTc> jkridner, when you said individual card you referred to the complet gist tutorial https://gist.github.com/DiegoTc/10611021 or one file of the gist?
Jun 23 02:40:23 <DiegoTc> I guessing is the first option I say, but want to confirm
Jun 23 02:41:36 <jkridner> one file of the gist
Jun 23 02:42:07 <jkridner> actually, 1 card, because I could see one card having multiple files such as an image and an HTML or markdown file.
Jun 23 02:59:56 <Abhishek_> jkridner: is the patch I sent you clear for sending to the beagleboard list?
**** ENDING LOGGING AT Mon Jun 23 02:59:58 2014