**** BEGIN LOGGING AT Wed Aug 05 02:59:58 2015
Aug 05 05:10:26 * nerdboy hopes everyone is off writing tests...
Aug 05 11:56:46 <shubhangi> Abhishek : ping
Aug 05 14:22:25 <shubhangi> Abhishek__: ping
Aug 05 14:22:34 <Abhishek__> hi shubhangi
Aug 05 14:23:12 <shubhangi> Abhishek__: Issues:
Aug 05 14:23:12 <shubhangi> One major optimization remains i.e I am able to get different transfer rates depending on what frequency one kicks (ie interrupts) the ARM. Kicking the ARM after writing to each 512byte buffer followed by a kick results in stalling the ARM because of too many interrupts within too short a time. Kicking after filling all 512 buffers have been filled in bu the
Aug 05 14:23:12 <shubhangi> PRU leads it to wait for ARM to consume buffers first before PRU can start using them again. So, in order to attain maximum throughput, an optimum value needs to be found out after which the ARM is kicked. The user also has freedom to decide this value on his own (in the pru firmware) depending upon his requirements.
Aug 05 14:23:29 <shubhangi> Abhishek__: from todays report.
Aug 05 14:23:36 <shubhangi> need some guidance
Aug 05 14:24:28 <Abhishek__> You would want to kick the ARM core at not more than 1000 times a second
Aug 05 14:25:42 <shubhangi> okay. and how do i test for what transfer rates i am getting ?
Aug 05 14:26:40 <shubhangi> i thought of writing test data to a file for a fixed time interval and then getting the file size
Aug 05 14:26:44 <Abhishek__> two points at which you note down nanoseconds / or microseconds elapsed, and then divide by the number of bytes transferred
Aug 05 14:27:09 <Abhishek__> You must be using copy_to_user somewhere?
Aug 05 14:27:25 <shubhangi> yes. read file op for the misc dev
Aug 05 14:28:01 <shubhangi> before and after that ?
Aug 05 14:28:35 <Abhishek__> yes. You could also do it on the userspace side
Aug 05 14:29:28 <Abhishek__> So you could read a fixed number of buffers and see the timings
Aug 05 14:29:49 <shubhangi> okay
Aug 05 14:30:05 <shubhangi> i also need to prepare some docs as to how the user has to code the vring part at pru side
Aug 05 14:30:31 <shubhangi> and what all goes under the hood ... it took me a long time to get it all together
Aug 05 14:30:54 <Abhishek__> try to document things in parallel with the code... last week would be a lot of hurrying.
Aug 05 14:31:08 <shubhangi> virtio,vring,virtqueues .. it gets very confusing
Aug 05 14:31:14 <shubhangi> okay
Aug 05 14:32:45 <shubhangi> finally i have realized how rpmsg would just have been an overhead on top of vrings.
Aug 05 14:33:08 <shubhangi> i'll put it all down somewhere
Aug 05 14:43:08 <Abhishek__> Please do
Aug 05 15:09:19 <_av500_> alo
Aug 05 15:09:23 <_av500_> 1h to go
Aug 05 15:31:38 <vvu> yep
Aug 05 15:32:21 * vvu is on the train from palo alto to sf...packed like a sardine
Aug 05 15:48:06 * moto-timo hands vvu some oil
Aug 05 15:48:57 * karki_ lurks 
Aug 05 15:59:43 <nerdboy> moin
Aug 05 16:00:06 <rohith> gm
Aug 05 16:00:13 <ankur> gm
Aug 05 16:00:28 <vvu> mornin
Aug 05 16:00:29 <alexanderhiam> howdy
Aug 05 16:00:37 <shubhangi> Morning
Aug 05 16:00:56 <vvu> mornin azizulhakim
Aug 05 16:01:23 <azizulhakim> morning vvu
Aug 05 16:01:36 <jkridner> anyone from BeaglePilot/ErleCopter here today?
Aug 05 16:01:40 <_av500_> aloha
Aug 05 16:01:45 <_av500_> neemo: report?
Aug 05 16:01:47 <neemo> mornin
Aug 05 16:01:51 <_av500_> did I miss it?
Aug 05 16:01:51 <neemo> up in a sec
Aug 05 16:02:00 <_av500_> it was due tuesday
Aug 05 16:02:02 <rohith> jkridner: vmayoral is travelling
Aug 05 16:02:08 <jkridner> k. :(
Aug 05 16:02:11 <_av500_> next year, it will be due monday
Aug 05 16:02:23 <neemo>  _av500_ yeah, sorry bout that :C
Aug 05 16:02:31 <jkridner> hi mdp... you been following the PRU developments?
Aug 05 16:02:48 <moto-timo> next year daily commits will be required from the beginning
Aug 05 16:02:54 <_av500_> no
Aug 05 16:03:09 <_av500_> but I really want the reports not to come in last minute
Aug 05 16:03:25 <moto-timo> I'd like to not get two commits in the entire GSoC
Aug 05 16:03:30 <mdp> jkridner: not really, somebody just pinged me about omap pru resets upstream but I haven't had a chance to respond yet
Aug 05 16:03:35 <_av500_> that too
Aug 05 16:03:49 <ds2> hmmm
Aug 05 16:03:54 <_av500_> +1
Aug 05 16:04:44 <vvu> so plan for today?
Aug 05 16:05:00 <arianepaola> hi everyone
Aug 05 16:05:21 <shubhangi> Hi
Aug 05 16:05:29 <vvu> hello arianepaola
Aug 05 16:05:48 <_av500_> 1) welcome everybody
Aug 05 16:05:49 <arianepaola> moto-timo: jkridner: _av500_: I have been ill. as requested I have added the issue list and also creating pull requests for all items. before shouting for it....
Aug 05 16:06:05 <jkridner> arianepaola: you are alive!
Aug 05 16:06:24 <moto-timo> arianepaola: I figured you had been abducted by aliens
Aug 05 16:06:49 <vvu> those pesky aliens
Aug 05 16:07:12 <_av500_> 2) who's here?
Aug 05 16:07:18 <rohith> here
Aug 05 16:07:18 <neemo> yup
Aug 05 16:07:19 <apaar> here
Aug 05 16:07:24 <ankur> here
Aug 05 16:07:31 <shubhangi> Here
Aug 05 16:07:38 <arianepaola> vvu moto-timo you probably watched too much the x files... but hey who did not
Aug 05 16:07:55 <azizulhakim> here
Aug 05 16:08:18 <_av500_> that makes 7
Aug 05 16:09:10 <_av500_> ehab excused himself
Aug 05 16:09:13 <_av500_> as did Diego
Aug 05 16:09:37 <moto-timo> ebadawy missing
Aug 05 16:10:09 <alexanderhiam> neemo: no report and 5 days without pushing anything, what's going on??
Aug 05 16:10:56 <_av500_> moto-timo: ehab sent a mail
Aug 05 16:11:06 <moto-timo> _av500_: got it
Aug 05 16:11:17 <_av500_> its roughly 1.5 weeks until "soft pencil down"
Aug 05 16:11:18 <moto-timo> missed that comment while I was looking through my list
Aug 05 16:11:19 <_av500_> Aug 17th
Aug 05 16:11:38 <moto-timo> that's 10 commits for those of you counting
Aug 05 16:11:39 <neemo> alexanderhiam: got about 4 commits in the backlog, can push it now but still digging through the MPU to fix the damn thing, not happy with the documentation one bit
Aug 05 16:12:23 <nerdboy> push early, push often
Aug 05 16:12:25 <neemo> alexanderhiam: though the MPU is feature complete now (when I squash the bugs) and on to the API
Aug 05 16:12:31 <nerdboy> just like voting...
Aug 05 16:12:42 <moto-timo> push, amend, squash, push, push
Aug 05 16:12:42 <alexanderhiam> I thought we made it pretty clear that we needed to see commits daily no matter how you felt about the state of it...
Aug 05 16:12:47 <_av500_> 3) finals are approaching fast
Aug 05 16:12:55 <_av500_> about 12 days left
Aug 05 16:13:02 <_av500_> then another week to clean up
Aug 05 16:13:13 <moto-timo> even if all you commit is a doc change, it'
Aug 05 16:13:17 <_av500_> so everybody should know exactly what to do
Aug 05 16:13:17 <neemo> alexanderhiam: fair enough, still don't feel comfortable with pushing something in pre-alpha phase :/
Aug 05 16:13:24 <moto-timo> it's better than no commit
Aug 05 16:13:28 <neemo> alexanderhiam: I'll push what I've got though
Aug 05 16:13:39 <alexanderhiam> then push it to a different branch
Aug 05 16:13:48 <neemo> point
Aug 05 16:13:59 <moto-timo> that's what work-in-progress branches are for
Aug 05 16:14:34 <_av500_> moto-timo: neemo nerdboy: pls continue after
Aug 05 16:14:55 <_av500_> 4) as usual, is anybody blocked from working on his task(s) during the final ~2weeks?
Aug 05 16:15:03 <_av500_> asking students and mentors
Aug 05 16:15:25 <shubhangi> Dont feel so
Aug 05 16:15:41 <ankur> no
Aug 05 16:15:43 <rohith> nope
Aug 05 16:15:50 <shubhangi> Have to write example apps
Aug 05 16:15:50 <azizulhakim> not blocking
Aug 05 16:16:00 <neemo> no blockers
Aug 05 16:16:02 <apaar> nope
Aug 05 16:16:45 <_av500_> arianepaola: ?
Aug 05 16:16:58 <arianepaola> (06:15:53 PM) Ariane: no blocking
Aug 05 16:17:35 <_av500_> oops
Aug 05 16:17:47 <_av500_> my irc ate that
Aug 05 16:17:48 <_av500_> damn
Aug 05 16:17:52 <_av500_> ok
Aug 05 16:17:55 <moto-timo> mine too
Aug 05 16:18:07 <_av500_> arianepaola: type harder
Aug 05 16:18:31 <arianepaola> _av500_: :-)
Aug 05 16:19:08 <_av500_> oo
Aug 05 16:19:10 <_av500_> ok
Aug 05 16:19:15 <_av500_> 5) reports
Aug 05 16:19:21 <_av500_> this weeks its azizulhakim and rohith
Aug 05 16:19:38 <_av500_> next week its neemo and ehab
Aug 05 16:19:56 <_av500_> arianepaola: is the internet that back in Luxembourg?
Aug 05 16:20:21 <arianepaola> its 1gbit/s, split for student apartments 10mb
Aug 05 16:20:34 <_av500_> you seem to cut out a lot
Aug 05 16:20:54 <arianepaola> probably lots of people downloading things
Aug 05 16:20:59 <neemo> that's not to shabby, at least on paper
Aug 05 16:21:13 <_av500_> aynanyeqany
Aug 05 16:21:22 <_av500_> gah
Aug 05 16:21:24 <_av500_> anyway
Aug 05 16:21:26 <Abhishek__> shubhangi: not to skip documentation while focusing on examples.
Aug 05 16:21:27 <neemo> then again I'm on a 200mb connection and get what 15 kb :/ (going insane)
Aug 05 16:21:27 <_av500_> lets move on
Aug 05 16:21:30 <neemo> y
Aug 05 16:21:39 <_av500_> rohith: want to start?
Aug 05 16:21:48 <rohith> _av500_: sure
Aug 05 16:21:49 <neemo> _av500_: btw, should I do a redo next week then?
Aug 05 16:21:55 <_av500_> rohith: 15 min as usual?
Aug 05 16:21:59 <rohith> yup
Aug 05 16:22:00 <_av500_> neemo: a redo?
Aug 05 16:22:04 <shubhangi> Abhishek__: will make sure of that
Aug 05 16:22:07 <neemo> I had one last week
Aug 05 16:22:11 <_av500_> oops
Aug 05 16:22:16 <neemo> sorry rohith  for the disturbance
Aug 05 16:22:34 <rohith> neemo: I haven't started :P
Aug 05 16:22:36 <_av500_> neemo: sorry, not you
Aug 05 16:22:37 <neemo> _av500_: I mean, I don't object, but I think others are on the list
Aug 05 16:22:43 <_av500_> ankur: its you next week
Aug 05 16:22:49 <_av500_> neemo: I mixed it up
Aug 05 16:22:52 <_av500_> anyway,
Aug 05 16:22:53 <ankur> ok
Aug 05 16:22:55 <_av500_> rohith: now
Aug 05 16:22:57 <neemo> _av500_: hanks for the correction, forgot to send an email
Aug 05 16:23:02 <neemo> rohith: go go :)
Aug 05 16:23:08 <rohith> _av500_: Okay. Here I go...
Aug 05 16:23:28 <rohith> So, as you all know, my project is Beaglepilot2
Aug 05 16:23:46 <rohith> and involves making an underwater vehicle class
Aug 05 16:23:59 <rohith> that is compatible with the APM autopilot
Aug 05 16:24:27 <rohith> Since my last presentation, I have pushed a lot of commits
Aug 05 16:24:45 <rohith> and now the APMsubmarine vehicle code compiles :)
Aug 05 16:25:08 <rohith> I have also cross-compiled the same so that it works on the BBB
Aug 05 16:25:23 <rohith> and no compilation errors encountered
Aug 05 16:25:44 <rohith> The base code structure is now in place
Aug 05 16:26:03 <rohith> although it hasn't been tested on an actual vehicle
Aug 05 16:26:30 <anujdeshpande> hey ankur
Aug 05 16:26:37 <rohith> You can find the project at https://github.com/BeaglePilot2/ardupilot/
Aug 05 16:26:56 <rohith> and the relevant vehicle folder is the APMsubmarine
Aug 05 16:27:16 <rohith> https://github.com/BeaglePilot2/ardupilot/tree/master/APMsubmarine
Aug 05 16:27:46 <rohith> You can try and compile the same yourself by following the instructions in the README
Aug 05 16:27:48 <ankur> hi anujdeshpande: rohith is giving report. lets talk after its over.
Aug 05 16:28:02 <rohith> Any questions so far?
Aug 05 16:29:26 <rohith> ?
Aug 05 16:29:49 <_av500_> rohith: besides making code compile, what did you achieve?
Aug 05 16:30:14 <rohith> _av500_: The base code structure is in place
Aug 05 16:30:32 <rohith> So the vehicle can interface with the existing sensors
Aug 05 16:30:42 <rohith> such as baro, imu etc.
Aug 05 16:31:14 <rohith> All the files that I have added so far ar compiled into a single .elf file
Aug 05 16:31:28 <_av500_> and you could run it?
Aug 05 16:31:34 <rohith> Which runs at statup on the Erle Brain
Aug 05 16:31:49 <rohith> I have to test the sensors
Aug 05 16:32:02 <rohith> Will be doing so in the next couple of days
Aug 05 16:32:09 <ds2> what kind of sensors?
Aug 05 16:32:28 <rohith> Also, the goals for next week shall be updated after discussion with my mentors
Aug 05 16:32:52 <rohith> ds2: Onboard sensors such as imu,baro...
Aug 05 16:33:17 <rohith> Also, external sensors that are supported by ardupilot should be supported too
Aug 05 16:33:32 <rohith> Since the same set of libraries have been used
Aug 05 16:34:07 <rohith> The library folder can be found here : https://github.com/BeaglePilot2/ardupilot/tree/master/libraries
Aug 05 16:34:26 <_av500_> rohith: so its the set of sensors that defines the submarine from other vehicles?
Aug 05 16:34:58 <rohith> _av500_: The libraries are common to all vehicles
Aug 05 16:35:12 <_av500_> so what defines a submarine?
Aug 05 16:35:13 <rohith> It is the vehicle folder that differentiates them
Aug 05 16:35:26 <rohith> https://github.com/BeaglePilot2/ardupilot/tree/master/APMsubmarine
Aug 05 16:35:39 <rohith> This is different for each vehicle
Aug 05 16:36:15 <rohith> Each vehicle also has a different header file and config
Aug 05 16:36:31 <rohith> inside the respective vehicle folder
Aug 05 16:36:49 <rohith> I will be testing the onboard sensors in the upcoming week
Aug 05 16:36:56 <ds2> what sensors are unique to a sub?
Aug 05 16:37:13 <rohith> Since there is only one .elf executable for the whole vehicle
Aug 05 16:37:46 <rohith> ds2: Right now the existing libraries have been utilized
Aug 05 16:38:13 <rohith> Although support for more submarine specific sensors can be added later on
Aug 05 16:38:26 <rohith> Like the DVL...
Aug 05 16:39:07 <rohith> The APMsubmarine is fairly basic at this stage
Aug 05 16:39:33 <rohith> Any other questions?
Aug 05 16:39:34 <_av500_> rohith: ic
Aug 05 16:39:45 <_av500_> what hardware are you testing on?
Aug 05 16:40:10 <rohith> I have cross compiled for the Erle Brain
Aug 05 16:40:20 <_av500_> ok
Aug 05 16:40:22 <rohith> I will be testing the code on the same
Aug 05 16:40:29 <_av500_> understood
Aug 05 16:40:41 <rohith> Erle Brain is officially supported by ardupilot
Aug 05 16:40:46 <_av500_> right
Aug 05 16:41:03 <rohith> vmayoral and others did the work to make it APM compatible
Aug 05 16:41:14 <_av500_> rohith: thanks for the report
Aug 05 16:41:23 <rohith> Which was his project last year in GSoC
Aug 05 16:41:39 <rohith> _av500_: Thanks :)
Aug 05 16:41:43 <_av500_> azizulhakim: your're up next
Aug 05 16:42:04 <azizulhakim> thanks _av500_
Aug 05 16:42:25 <azizulhakim> My project is android-based remote display
Aug 05 16:42:46 <azizulhakim> Driver Code: https://github.com/azizulhakim/beagleusb
Aug 05 16:43:08 <azizulhakim> Android App Code: https://github.com/azizulhakim/bard-droid
Aug 05 16:43:52 <azizulhakim> On my last report I mentioned the work on audio and input driver
Aug 05 16:44:29 <azizulhakim> until my last report I had two big issues to solve
Aug 05 16:44:44 <azizulhakim> which I managed to solve and is added in the code base
Aug 05 16:45:22 <azizulhakim> Issue #1 Audio driver crash on bone
Aug 05 16:45:53 <azizulhakim> 2. Video driver performance
Aug 05 16:46:27 <azizulhakim> my audio driver was working okay in my ubuntu laptop, but it had some crash issue on bone
Aug 05 16:47:10 <azizulhakim> it could serve only one run of some audio stream. after the installation of the driver, if I run some audio file twice, the driver would crash
Aug 05 16:48:05 <azizulhakim> it was kind of a confusing issue that took me a lot of time to solve. my current code has that issue fixed
Aug 05 16:48:34 <_av500_> azizulhakim: good
Aug 05 16:48:42 <azizulhakim> 2nd big issue was video driver performance
Aug 05 16:48:52 <azizulhakim> thanks _av500_
Aug 05 16:49:19 <azizulhakim> we've limited cache and ram on bone, which causes a lot of page fault and our driver works on the concept of page fault
Aug 05 16:49:41 <azizulhakim> whenever there is page fault, our driver traces the dirty pages and update those page on application side
Aug 05 16:50:10 <azizulhakim> because of large number of page fault, we have to send huge data over USB
Aug 05 16:51:03 <azizulhakim> beaglebone usb transfer speed is 3.5Mbps which is good enough compared to other systems, but not good enough to send this huge data
Aug 05 16:51:33 <azizulhakim> so I added lazy update method
Aug 05 16:51:50 <azizulhakim> I don't put pressure on usb connection each time there is a page fault
Aug 05 16:52:04 <azizulhakim> rather I keep track of which pages are modified
Aug 05 16:52:27 <azizulhakim> I created a kernel thread which periodically checks for dirty pages and update those rather than doing update on each page fault
Aug 05 16:52:35 <_av500_> ic
Aug 05 16:52:59 <Abhishek__> 3.5 MB per second or 3.5 Mbits / second ?
Aug 05 16:53:00 <azizulhakim> so this way we have 10 FPS video now
Aug 05 16:53:06 <azizulhakim> Mbits
Aug 05 16:53:14 <_av500_> that seems awfully low
Aug 05 16:53:23 <Abhishek__> isn't 3.5 mbits low compared to 480 MBps
Aug 05 16:53:30 <Abhishek__> not even 1%
Aug 05 16:53:54 <_av500_> is that from raw measurements?
Aug 05 16:54:01 <Abhishek__> 10fps at what resolution?
Aug 05 16:54:12 <rohith> azizulhakim: bits or bytes?
Aug 05 16:54:45 <rohith> Oops. Didn't see. Its bits
Aug 05 16:55:28 <ds2> is this a kernel mode driver or userland driver?
Aug 05 16:56:22 <_av500_> ds2: kernel
Aug 05 16:56:43 <nerdboy> embedded usb can be a little different...  stephanie has some good points in here that might be helpful => https://github.com/VCTLabs/scale13x-presentations
Aug 05 16:56:55 <nerdboy> the USB one obviously...
Aug 05 16:57:05 <_av500_> nerdboy: still he should get at least 10MB/s
Aug 05 16:57:13 <nerdboy> true
Aug 05 16:57:37 <nerdboy> so maybe a couple of things to "fix"
Aug 05 16:57:39 <ds2> seems almost as if it is in FS mode
Aug 05 16:57:47 <_av500_> I dont expect embedded usb to go over 20MB/s
Aug 05 16:57:59 <_av500_> ds2: hmm, good point
Aug 05 16:58:15 <ds2> MUSB is bad but not that bad
Aug 05 16:58:23 * nerdboy can remember having issues with musb on TI parts before
Aug 05 16:58:25 <_av500_> its not musb
Aug 05 16:58:29 <_av500_> its the host port
Aug 05 16:58:41 <_av500_> BBB is host, phone is device
Aug 05 16:58:51 <ds2> BBB host is MUSB
Aug 05 16:58:51 <nerdboy> normally it's gadget driver
Aug 05 16:58:55 <_av500_> nope
Aug 05 16:59:00 <_av500_> ADK mode is host
Aug 05 16:59:16 <_av500_> its the USB-A connector
Aug 05 16:59:26 <nerdboy> oh yeah
Aug 05 16:59:29 <ds2> you confusing it with the BBC/BBX
Aug 05 16:59:36 <ds2> there is no EHCI on the BBB/BBW
Aug 05 16:59:36 * nerdboy still on only one coffeee
Aug 05 16:59:54 <nerdboy> the host port (A) should be musb
Aug 05 16:59:59 <_av500_> ds2: right
Aug 05 17:00:00 <_av500_> sorry
Aug 05 17:00:03 <_av500_> indeed
Aug 05 17:00:10 <_av500_> too many BBs :)
Aug 05 17:00:22 <nerdboy> too many keys on the keyboard...
Aug 05 17:00:28 <_av500_> anyway, thats something to look into
Aug 05 17:00:34 <_av500_> but we lost him
Aug 05 17:00:40 <_av500_> and my laptop is going down
Aug 05 17:00:45 <_av500_> and time is up
Aug 05 17:00:54 <_av500_> I will follow up wih him by mail
Aug 05 17:00:59 <_av500_> thanks everybody
Aug 05 17:01:02 <nerdboy> let's all review that presentation again...
Aug 05 17:01:07 <_av500_> see you all next week
Aug 05 17:01:26 <_av500_> keep coding and comitting
Aug 05 17:01:40 <ds2> and keep in mind of the schedule
Aug 05 17:01:42 <nerdboy> <cough> neemo <cough>
Aug 05 17:01:45 <ds2> the end is near
Aug 05 17:02:02 * neemo hands nerdboy a napkin :)
Aug 05 17:02:08 <_av500_> +1^111
Aug 05 17:02:12 <neemo> also will commit, at least to a WIP branch
Aug 05 17:02:36 <ds2> 1^X for all X is 1 :P
Aug 05 17:02:49 <_av500_> ds2: my X is > INFTY
Aug 05 17:02:56 <neemo> that way, I don't feel like i corrupted my soul by commiting junk to master and you guys can view what I've done
Aug 05 17:03:06 <nerdboy> it could not be worse than some master branches i have seen...
Aug 05 17:03:11 <neemo> lol
Aug 05 17:03:12 <ds2> lim X->INF 1^X = 1 :P
Aug 05 17:03:27 <_av500_> ds2: to INFTY and beyond!
Aug 05 17:03:31 <moto-timo> arianepaola: you understand that I expect commits every two days at a minimum for the rest of GSoC?
Aug 05 17:03:39 <neemo> well, it works, apart from the part that don't and they're pissing me off, but I'm working on it, should be all done by tonight = +2/3 hours
Aug 05 17:03:43 <ds2> neemo: just stick it in a different branch. visibily is essential
Aug 05 17:03:53 <_av500_> ok, signing off
Aug 05 17:03:57 * nerdboy hopes _av500_ is ieee-compliant
Aug 05 17:04:01 <neemo> ds2: roger roger
Aug 05 17:04:07 <Abhishek__> ​IEEE 754
Aug 05 17:04:16 <arianepaola> moto-timo: Aye aye, sir!
Aug 05 17:04:26 <ds2> neemo: going through the exact same arguement with some developers new to git at work
Aug 05 17:04:29 <nerdboy> otherwise it's not defined...
Aug 05 17:04:30 <_av500_> nerdboy: compliancy is only for standard people
Aug 05 17:04:59 <nerdboy> can't even get a NaN
Aug 05 17:05:07 <moto-timo> arianepaola: even if it is only comments or documentation... that is also valuable
Aug 05 17:05:17 <neemo> ds2: I love git, don't get me wrong. Think it's the best thing since sliced bread
Aug 05 17:05:29 <neemo> I'm just a bit, erm... code shy ?
Aug 05 17:05:40 <nerdboy> -mfp16-format=ieee
Aug 05 17:05:48 <neemo> Don't like putting out a piece of code which I think is well, not there yet
Aug 05 17:06:08 <arianepaola> moto-timo: yes :)
Aug 05 17:06:24 <neemo> but alas, help and comments are more useful then my self-correction
Aug 05 17:06:26 <neemo> :)
Aug 05 17:06:41 <moto-timo> arianepaola: ok then. Looking forward to it :)
Aug 05 17:10:51 <nerdboy> put it on another branch and push it
Aug 05 17:11:11 <nerdboy> have to tell a guy at work the same thing...
Aug 05 17:31:12 <neemo> https://github.com/nvisnjic/PyBBIO/blob/WIP/bbio/libraries/MPU9250/MPU9250.py
Aug 05 17:31:29 <neemo> nerdboy: alexanderhiam ^^ thar she blows
Aug 05 17:31:57 <neemo> I'll be working on closing that issue and merging with the master, then making a pull request
Aug 05 17:32:00 <neemo> by tonight
Aug 05 17:32:19 <alexanderhiam> neemo: thanks. And how's the Beaglesat Python package going? I was expecting to see that coming along by this point
Aug 05 17:33:51 <neemo> started it, but came back to finish this library to the end
Aug 05 17:34:16 <neemo> I'll be switching back to it tonight, if I manage to sort the bugs in this or no
Aug 05 17:34:32 <azizulhakim> sorry all, _av500_
Aug 05 17:34:41 <azizulhakim> there was some mishap
Aug 05 17:36:17 <azizulhakim> Abhishek__: 10 fps at 1024x768 resolution
Aug 05 17:36:59 <azizulhakim> Abhishek__: and it is 3.5 MBytes , not bits
Aug 05 17:40:45 <neemo> alexanderhiam: nerdboy: In other news, an interesting piece of information for the AK8963
Aug 05 17:40:57 <neemo> when I dropped the docs from InvenSense and went straight to AK
Aug 05 17:41:36 <neemo> they show that the AK8963 can run at 100Hz without a problem, by putting it in MODE2
Aug 05 17:42:52 <ds2> :)
Aug 05 17:43:13 <alexanderhiam> slick
Aug 05 17:43:29 <alexanderhiam> can you get the data off it that quick?
Aug 05 17:44:50 <neemo> should be able to, I'll add that in a bit, now that I've pinpointed the issue with the calibration
Aug 05 17:45:16 <neemo> ds2: thanks for the cue to go digging :)
Aug 05 17:45:49 <neemo> also sidequestion, alexanderhiam , how do I use the SPI in PyBBIO if I want to read, say 12 bytes from the same register, or 480 for that matter
Aug 05 17:46:12 <alexanderhiam> a loop
Aug 05 17:46:16 <neemo> I've fixed it with doing a for loop and a single byte read on the same address, but I'm sure there's a better way, I just didn't look in the right place
Aug 05 17:46:21 <alexanderhiam> there's nothing built in to the API to do that
Aug 05 17:46:26 <neemo> hmmm
Aug 05 17:46:41 <neemo> can't SPI read form the same address continuously?
Aug 05 17:46:47 <alexanderhiam> the API just provides the functions for the standard API spec
Aug 05 17:46:54 <alexanderhiam> that's up to user implementation
Aug 05 17:47:08 <alexanderhiam> "reading from registers" isn't part of the SPI spec
Aug 05 17:47:21 <neemo> I mean, I don't neeed it to badly for the calibration, but the FIFO is faster now that i've set it up
Aug 05 17:47:36 <alexanderhiam> that's the kind of thing that can be moved to a C extension later to speed it up
Aug 05 17:47:39 <neemo> the point is the fifo is one register which I should read continuously
Aug 05 17:47:45 <neemo> yeah
Aug 05 17:48:10 <neemo> read the spi code python calls, couldn't find a same address continous read
Aug 05 17:48:18 <neemo> we can add that later, when speed is of essence
Aug 05 17:48:38 <neemo> anyway, for loop is good enough for now, off to do more testing/debugging
Aug 05 18:02:00 <Abhishek__> jkridner: just heard on Hackaday that the BeagleBone green is out!
Aug 05 18:05:08 <ds2> look at autoincrement. some chips can disable it some can't
Aug 05 18:07:42 <jkridner> Abhishek__: sweet. Just reading the hackaday article now.
Aug 05 18:08:27 <Abhishek__> jkridner: do you think you could get me one now ;) ?
Aug 05 18:09:12 <jkridner> I can certainly drop a line to the Seeed folks to see if they want to send you one. I don't have a discounted access or anything.
Aug 05 18:09:56 <Abhishek__> jkridner: It'd be great if you could introduce them to the BeagleLogic project and if they'd be interested in distributing capes.
Aug 05 18:12:42 <ds2> jkridner: back from MF/Detroit?
Aug 05 18:54:59 <neemo> nerdboy: alexanderhiam: Phew, finally
Aug 05 18:56:48 <neemo> perfectly zeroed out
Aug 05 19:24:51 <neemo> nerdboy: alexanderhiam -> and so it was pushed -> https://github.com/nvisnjic/PyBBIO/blob/master/bbio/libraries/MPU9250/MPU9250.py
Aug 05 19:25:22 * neemo feels relieved that abomination was patched up
Aug 05 19:28:14 <nerdboy> naming your branch fubar-WIP makes things pretty clear...
Aug 05 20:14:21 <nerdboy> plus modern "agile" says push a test with just a stub, make it fail, always broken until minimal implementation starts making tests pass...
Aug 05 22:56:33 <vvu> :qall
Aug 05 22:56:36 <vvu> oh well
Aug 05 22:56:39 <vvu> not here
**** ENDING LOGGING AT Thu Aug 06 02:59:58 2015