**** BEGIN LOGGING AT Sun Mar 15 02:59:59 2015
Mar 15 04:42:03 <kiran4399> sidbh_: ping
Mar 15 08:51:26 <webglider> does anyone know when jkridner will be coming back here?
Mar 15 09:16:37 <kiran4399> sidbh__: there ?
Mar 15 09:16:54 <sidbh__> kiran4399, yes
Mar 15 09:17:30 <kiran4399> sidbh__: You've heard of pruspeak right ??
Mar 15 09:17:38 <sidbh__> yes
Mar 15 09:19:38 <kiran4399> sidbh__: So I am trying to access gpio pins other than those bound to pru using starterware..
Mar 15 09:20:14 <sidbh__> ok and ...
Mar 15 09:20:59 <kiran4399> sidbh__: I replaced the R30 write with GPIOPinwrite() with necessary changes.. But ..it is not working..
Mar 15 09:21:23 <kiran4399> sidbh__: sometimes the beaglebone is ending up in a deadlock..
Mar 15 09:22:07 <Abhishek__> hmm, probably due to conflict with the Linux GPIO subsystem?
Mar 15 09:22:09 <sidbh__> point me to your code and result, via gist, pastbin or github
Mar 15 09:22:24 <sidbh__> s/pastbin/pastebin/
Mar 15 09:22:41 <kiran4399> sidbh__: I somehow feel that I have to enable the pin, which I want to access,  in the device tree overlays.. What do you think ?
Mar 15 09:23:05 <kiran4399> OK...
Mar 15 09:31:30 <kiran4399> sidbh__: this is a small section in the source code pru0_firmware.c which I have modified.. https://gist.github.com/kiran4399/ab3f3f31c4d07e5f0837  Please look at it..
Mar 15 09:32:53 <kiran4399> sidbh__: one more thing.. presently I am testing only with gpio1 module.. so that is the reason I kept by default, gpio1_instance address as the base address in every api..
Mar 15 09:33:09 <kiran4399> val1 is the pin number..
Mar 15 09:33:55 <sidbh__> kiran4399, also give me all the #defines you used
Mar 15 09:35:35 <kiran4399> sidbh__: I only added the GPIO's list https://github.com/kiran4399/starterware_PRU/blob/master/examples/pru_speak_integrate/pru_firmware.h#L19
Mar 15 09:42:04 <webglider> julianduque: are you here by any chance?
Mar 15 09:42:11 <kiran4399> sidbh__: BTW pruspeak repo is :https://github.com/deepakkarki/pruspeak/tree/master/src/pru-firmware
Mar 15 09:49:25 <sidbh__> kiran4399, try giving a delay before you disable global access, use empty for loop for now
Mar 15 09:51:20 <kiran4399> sidbh__: OK.. I will do that.. I have one more doubt.. I am trying to access GPIO1_7... but that is not enabled in the device tree. Are you sure that it really does not matter ?
Mar 15 09:51:40 <sidbh__> nope
Mar 15 09:52:03 <sidbh__> you just follow the steps mentioned in the datasheet
Mar 15 09:52:44 <sidbh__> of am335x to access peripherals and make sure there aren't access clashes from linux space.
Mar 15 09:56:53 <kiran4399> sidbh__: tried what you told me.. the system is not ending in deadlock.. but I am unable to access the pin..
Mar 15 09:57:53 <sidbh__> ok push the changes to ur git let me try it on my BBB
Mar 15 09:58:18 <sidbh__> I'll see what I can find
Mar 15 10:11:35 <kiran4399> sidbh__: thttps://github.com/kiran4399/pruspeak ..
Mar 15 10:11:47 <kiran4399> sidbh__: https://github.com/kiran4399/pruspeak
Mar 15 10:14:15 <sidbh__> kiran4399, what is the reference for this line: https://gist.github.com/kiran4399/ab3f3f31c4d07e5f0837#file-dio_handler-c-L52
Mar 15 10:17:28 <kiran4399> sidbh__: https://github.com/kiran4399/pruspeak/blob/master/src/pru-firmware/pru0_firmware.c#L131
Mar 15 10:18:52 <sidbh__> that will not disable the global access, instead would initiate the standby mode
Mar 15 10:20:08 <sidbh__> which I think wasn't ur objective
Mar 15 10:24:01 <sidbh__> so in my opinion lose that line
Mar 15 10:24:03 <kiran4399> sidbh__: oops..
Mar 15 10:24:22 <kiran4399> sidbh__: so shall I remove the line ?
Mar 15 10:24:27 <sidbh__> yes
Mar 15 10:25:30 <kiran4399> sidbh__: I actually got that from here.. https://github.com/kiran4399/pruspeak/blob/master/src/pru-firmware/pru_firmware.h#L62
Mar 15 10:26:04 <sidbh__> no that's wrong
Mar 15 10:27:00 <sidbh__> and do SYSCFG stuff during init, it doesn't belong in a function
Mar 15 10:29:39 <kiran4399> sidbh__: Oh you want me to do https://github.com/BeaglePilot/PRUSS-C/blob/master/PRUSS_LIB/example/blinkled.c#L15 again in pru0_firmware.c ?
Mar 15 10:31:09 <sidbh__> no, I just want you to remove the line I mentioned
Mar 15 10:31:43 <kiran4399> sidbh__: Oh.. sorry .. ok I understood.. :-)
Mar 15 10:37:42 <kiran4399> sidbh__: OK.. I did that and I recompiled everything again.. but the led is connected to gpio1_39 is not lightning up.. I even tried those user-led's (21,22,23,24) but none of them is responding .. :-( Are you sure that my code is correct ?
Mar 15 10:40:29 <sidbh__> ok I'll try it out on my BBB and see
Mar 15 10:44:43 <kiran4399> sidbh__: OK thanks..
Mar 15 13:23:02 <soulslicer> Hello is anyone around?
Mar 15 13:24:51 <anujdeshpande> soulslicer: you have to knock before asking
Mar 15 13:25:03 <soulslicer> Sorry I am new to this.
Mar 15 13:25:08 <anujdeshpande> :D
Mar 15 13:25:09 <soulslicer> How exactly do I do that
Mar 15 13:25:19 * anujdeshpande is trolling
Mar 15 13:25:30 <soulslicer> Fantastic..anyway..I would like to ask
Mar 15 13:25:37 <soulslicer> Are you a lead dev on the beaglebone black?
Mar 15 13:25:45 <soulslicer> I am looking for someone who has worked on the DSP
Mar 15 13:28:52 <anujdeshpande> soulslicer: this is a community channel. you should ask your question and someone who knows the answer, or someone else who might know the answer, will reply.
Mar 15 13:29:42 <soulslicer> Ic right
Mar 15 13:30:09 <soulslicer> I am trying to see if I can do a GSOC project with some ideas I have for the DSP
Mar 15 13:31:39 <anujdeshpande> soulslicer: sure you can, you just need to post your idea on the ideas wiki and then try finding someone who will mentor here
Mar 15 13:31:54 <soulslicer> Post my idea on the ideas wiki?
Mar 15 13:32:00 <soulslicer> This one here? http://elinux.org/BeagleBoard/GSoC/Ideas
Mar 15 13:32:22 <anujdeshpande> soulslicer: that’s the one.
Mar 15 13:32:32 <soulslicer> So you mean..actually edit the thing?
Mar 15 13:33:50 <_av500_> soulslicer: the beaglebone has no DSP
Mar 15 13:34:10 <soulslicer> It does
Mar 15 13:34:15 <_av500_> for gsoc ideas, please sign up on the mailing list
Mar 15 13:34:16 <soulslicer> http://elinux.org/BeagleBoard/DSP_Howto
Mar 15 13:34:27 <_av500_> soulslicer: thats the beagleboard
Mar 15 13:34:29 <_av500_> not the bone
Mar 15 13:34:33 <soulslicer> Ah right
Mar 15 13:34:36 <soulslicer> my bad
Mar 15 13:35:02 <soulslicer> Whats the status of the Beaglebone Black?
Mar 15 13:35:06 <soulslicer> I mean the beagleboard?
Mar 15 13:35:36 <_av500_> most people are on the bone these days
Mar 15 13:35:56 <soulslicer> right right
Mar 15 13:36:51 <soulslicer> This might sound ridiculous, but how likely is it to say, do image processing on the Beaglebone's 3D Graphics accelerator..
Mar 15 13:37:18 <soulslicer> It doesn't have OpenCL support, but Im sure it supports GLSL
Mar 15 13:37:19 <_av500_> its a more or less stock GPU
Mar 15 13:37:32 <_av500_> with some kind of GL
Mar 15 13:38:34 <soulslicer> Are the graphic drivers open source?
Mar 15 13:39:33 <_av500_> soulslicer: nope
Mar 15 13:43:53 <soulslicer> right..
Mar 15 13:44:22 <_av500_> I dont see much value in inmage processing on the bone
Mar 15 13:47:30 <soulslicer> Ic, I'm trying to find a project that I might be able to work under for GSOC..
Mar 15 13:47:40 <soulslicer> i recently spoke to one of the lead devs on the PRU Project
Mar 15 13:47:53 <soulslicer> Unfortunately it seems his  project places are completely full
Mar 15 13:50:30 <soulslicer> Is it possible to propose my own project ideas?
Mar 15 14:17:20 <sidbh__> kiran4399, I have some suggestions for sorting out your problem
Mar 15 14:18:29 <kiran4399> sidbh__: First of all thank you for taking out some time for me.. I am very grateful to you.. :-D
Mar 15 14:18:38 <kiran4399> sidbh__: please go ahead..
Mar 15 14:20:34 <sidbh__> I ran the code with only blinkled part inside the main and it ran just fine, meaning the library works, the issue must be either a lot of writes to PRU_CFG, I think OCP master should be set only once during init before while(1), remove the remaining sets done here and there, especially the once initialising standby
Mar 15 14:21:23 <sidbh__> so only one "PRUCFG_SYSCFG = PRUCFG_SYSCFG & (~SYSCFG_STANDBY_INIT);" in the whole code
Mar 15 14:24:20 <sidbh__> second you should understand that writing to global registers might not be instantaneous, it is quite possible that the program pointer is onto the next instruction and the write is still being done, so check if the register is written before proceeding. I might be wrong with this statement, so please verify docs.
Mar 15 14:27:29 <sidbh__> try resolving for these two issues and see what happens.
Mar 15 14:28:16 <kiran4399> sidbh__: you mean to say after the timer_init() in the main ?
Mar 15 14:28:23 <sidbh__> yes
Mar 15 14:29:05 <sidbh__> also how are you debugging your code
Mar 15 14:29:36 <kiran4399> since there is no access to stdio .. I am unable to debug..
Mar 15 14:30:45 <shubhangi> sidbh__: pruspeak enable disables OCP master several times. so i guess that shouldn't be a problem
Mar 15 14:31:02 <sidbh__> I think there are some print upcalls in place, contact karki to know how they work
Mar 15 14:31:11 <shubhangi> so it kind of narrows it does to the other reason
Mar 15 14:31:19 <shubhangi> down*
Mar 15 14:32:05 <sidbh__> shubhangi, the code doesn't disables OCP master, it initiates standby
Mar 15 14:34:16 <shubhangi> sidbh__: ok
Mar 15 14:34:41 <shubhangi> sidbh__: PRUCFG_SYSCFG = PRUCFG_SYSCFG & (~SYSCFG_STANDBY_INIT); this has been quiet a few times though
Mar 15 14:35:20 <sidbh__> that code is not the one I think causes the issue as much as its counterpart that is being used: PRUCFG_SYSCFG = PRUCFG_SYSCFG | (SYSCFG_STANDBY_INIT);
Mar 15 14:35:21 <shubhangi> followed by PRUCFG_SYSCFG = PRUCFG_SYSCFG | SYSCFG_STANDBY_INIT); to restrict global access
Mar 15 14:35:31 <shubhangi> yes
Mar 15 14:35:53 <sidbh__> that code as it clearly states will initiate STANDBY
Mar 15 14:36:23 <sidbh__> now if you write 0 to PRU's SYSCFG bit no. 4 you enable OCP master
Mar 15 14:36:44 <shubhangi> writing to memory must be enveloped between these two pieces of code
Mar 15 14:36:45 <kiran4399> sidbh__: right..
Mar 15 14:36:52 <sidbh__> but if you write 0 it initiates standby and "Does not" disables OCP
Mar 15 14:36:54 <sidbh__> master
Mar 15 14:37:23 <kiran4399> sidbh__: what is the difference between both of them ??
Mar 15 14:38:00 <shubhangi> one sets the bit, other clears it
Mar 15 14:43:59 <sidbh__> kiran4399, ans. to ur question: Standby mode: Standby mode disables most of the clocks of the device, while keeping the design powered up. This reduces the power drawn to the static leakage current, plus a tiny clock power overhead required to enable the device to wake up from the standby state.
Mar 15 14:44:33 <sidbh__> OCP master: the PRUs have access to all resources on the SoC through the Interface/OCP master port, and the external host processors can access the PRU-ICSS resources through the Interface/OCP Master port.
Mar 15 14:45:46 <sidbh__> now as we can see they are totally different things. We are lucky that PRUCFG_SYSCFG = PRUCFG_SYSCFG | SYSCFG_STANDBY_INIT only initiates the standby not goes so far as to enable it.
Mar 15 14:46:50 <sidbh__> shubhangi, yes one clears the bit and other sets the bit, but clearly doing both of those does not have same result
Mar 15 14:47:11 <sidbh__> s/same/complementary/
Mar 15 14:47:17 <shubhangi> sidbh__: i didnt say they are same
Mar 15 14:47:33 <sidbh__> sorry not same complementary
Mar 15 14:47:58 <shubhangi> ok
Mar 15 14:49:42 <sidbh__> for example in many micro controllers you will find separate registers for set and reset, and say if you write 0 to set register doesn't necessarily mean the pin will go low, if it does it would be clearly stated in the dateasheet/user manual
Mar 15 14:49:58 <sidbh__> s/dateasheet/datasheet/
Mar 15 14:51:59 <abh33> hello there
Mar 15 14:52:02 <sidbh__> which in the case of PRU it no where stated that there is a way to change OCP master's enable state once you have activated it, so playing with the register all over the code doesn't seem like a good idea, it should be activated once at initiation
Mar 15 14:53:38 <sidbh__> although shubhangi you probably might be correct that its possible that this is be the issue kiran4399 has been having problem with, but its bad practice anyway :)
Mar 15 14:54:50 <sidbh__> s/this is/this is not the issue/
Mar 15 14:55:42 <shubhangi> okay :)
Mar 15 14:57:03 <shubhangi> sidbh__: i was facing the same issue when accessing gpio module registers in memory. got it sorted it by clearing the standy bit
Mar 15 14:57:19 <shubhangi> sidbh__: thanks, will take care to do it only once
Mar 15 15:29:28 <right-hearted> Hi, in order to apply for GSoC do I have to only submit my proposal?
Mar 15 15:30:34 <_av500_> right-hearted: getting in contact with possible mentors helps
Mar 15 15:30:44 <_av500_> as does introducing yourself on the mailing list
Mar 15 16:00:34 <kiran4399> sidbh: there ?
Mar 15 16:04:42 <sidbh> kiran4399, yes
Mar 15 16:06:40 <kiran4399> sidbh: I do not know why this problem is coming.. but whenever I fiddle with the SYS_CFG stuff.. it is getting deadlocked and 2 of the user led's are lightening up and the other 2 or not lightning..
Mar 15 16:07:30 <kiran4399> sidbh: I mean no flashing of led's is taking place ..
Mar 15 16:07:49 <kiran4399> sidbh: when I type SET DIO[7], 1
Mar 15 16:08:10 <kiran4399> sidbh: it is going for an infinite loop..
Mar 15 16:10:18 <kiran4399> sidbh: I removed all the SYS_CFG's and just kept the enabling SYS_CFG i.e. PRUCFG_SYSCFG = PRUCFG_SYSCFG & (~SYSCFG_STANDBY_INIT); after timer_init()
Mar 15 16:10:28 <kiran4399> sidbh_: I removed all the SYS_CFG's and just kept the enabling SYS_CFG i.e. PRUCFG_SYSCFG = PRUCFG_SYSCFG & (~SYSCFG_STANDBY_INIT); after timer_init()
Mar 15 18:08:49 <kiran4399> fa
Mar 15 18:51:59 <right-hearted> How would common bootloader be tested?
Mar 15 18:52:29 <vvu> same file ran on the 2 boards maybe
Mar 15 18:52:46 <vvu> and if we get until the kernel => win
Mar 15 18:53:28 <right-hearted> community will be used for testing then
Mar 15 18:54:50 <right-hearted> have there been any progress made on that task? Or is it new one?
Mar 15 18:56:34 <vvu> it is a new one
Mar 15 18:58:51 <right-hearted> How do I reach a particular mentor? Just call his name?
Mar 15 18:59:02 <vvu> ds2: ^^
Mar 15 20:15:51 <ds2> send a mail to the gsoc list or try to catch me on weekdays...less running around
Mar 15 20:31:42 <webglider> jkridner
Mar 15 21:07:17 <webglider> jkridner: Hello, nice to see you
Mar 15 21:07:43 <webglider> jkridner: Could you please check my reply to our conversation: http://bit.ly/1Evqpf9
Mar 15 21:42:40 <shubhangi> alexanderhiam_: added gpio0 modules support to pruspeak https://github.com/shubhi1407/pruspeak . sent pull request to karki :)
Mar 15 21:42:47 <shubhangi> module*
Mar 15 21:43:15 <shubhangi> can be easily expanded to other modules
Mar 15 23:55:42 <vvu> alexanderhiam_: how does your conversion work for GPIOs P8.10 is translated in which GPIO ?
Mar 15 23:56:17 <vvu> nvmind found them in the wiki
Mar 16 01:26:43 <anonymous_> have great ideas of how the Demo Android app with BBB Android can look like
Mar 16 02:03:19 <hendersa> Evening, all. Or morning, as the case may be.
Mar 16 02:22:27 <vvu> morning !
Mar 16 02:25:18 <vvu> alexanderhiam_: i am using PyBBIO for a small project to monitor some plastic injection machines
Mar 16 02:26:03 <vvu> i am doing mainly a while loop in which I am calling digitalRead on a GPIO pin...after some point my RAM just keeps going up without being freed
Mar 16 02:36:26 <ds2> what kind of plastic injection machine?
Mar 16 02:37:39 <vvu> to make buckets :)
Mar 16 02:38:09 <ds2> oh big ones
Mar 16 02:38:22 <ds2> thought you were messing with the smaller stuff
Mar 16 02:39:06 <vvu> neee, it's for a friend of my dad
Mar 16 02:39:16 <vvu> people at the factory closing machines during night
Mar 16 02:39:25 <vvu> and when he is there in the morning production is low
Mar 16 02:39:37 <vvu> and he wants a way to check what actually happen
Mar 16 02:48:53 <anonymous_> sounds interesting
**** ENDING LOGGING AT Mon Mar 16 02:59:58 2015