2015-08-17T00:01:09 *** yangqiao has joined #rtems 2015-08-17T00:16:02 [trac] GSoC/2015 edited by yangqiao (revert the mistake of deletion in rev 197) http://devel.rtems.org/wiki/GSoC/2015 2015-08-17T00:29:10 *** Davidbrcz has joined #rtems 2015-08-17T00:47:41 *** crazy3k has joined #rtems 2015-08-17T00:58:49 *** gedare has joined #rtems 2015-08-17T00:58:49 *** ChanServ sets mode: +o gedare 2015-08-17T01:07:31 *** gedare has quit IRC 2015-08-17T01:37:42 *** gedare has joined #rtems 2015-08-17T01:37:42 *** ChanServ sets mode: +o gedare 2015-08-17T01:49:40 [trac] #2390/defect (new) updated by gedare (I also prefer -1, i.e. ~0 to indicate an error. It's an easy thing to check against. ...) http://devel.rtems.org/ticket/2390 2015-08-17T01:54:24 *** gedare has quit IRC 2015-08-17T03:01:25 *** dr__house is now known as dr__house|away 2015-08-17T03:49:00 *** Davidbrcz has quit IRC 2015-08-17T04:42:13 *** ketul has joined #rtems 2015-08-17T05:15:19 *** Davidbrcz has joined #rtems 2015-08-17T05:57:13 *** monstr has joined #rtems 2015-08-17T06:56:44 *** kiwichris has quit IRC 2015-08-17T07:03:04 *** mumptai has joined #rtems 2015-08-17T07:38:12 *** Davidbrcz has quit IRC 2015-08-17T08:13:54 *** ketul has quit IRC 2015-08-17T08:42:11 *** mumptai has quit IRC 2015-08-17T08:45:55 *** ketul has joined #rtems 2015-08-17T08:47:25 hi beng-nl 2015-08-17T08:47:29 hi ketul! 2015-08-17T08:48:00 yes ! in irq.c I just reordered bsp_interrupt_handler_dispatch(irq); 2015-08-17T08:48:01 uint32_t psr = _ARMV4_Status_irq_enable(); 2015-08-17T08:48:57 ok 2015-08-17T08:49:00 any other debug info? 2015-08-17T08:49:23 like print in the dispatch function to verify it is indeed infinite recursion? 2015-08-17T08:49:27 (as i think) 2015-08-17T08:49:45 and the change not merged yet ! just checked github/rtems 2015-08-17T08:50:14 yes I included print in bsp_interrupt_dispatch() 2015-08-17T08:50:24 and it is going infinite printing 2015-08-17T08:50:43 infinite printing? and then reset? or? 2015-08-17T08:51:26 infinite printing though not sending signal on int pin. After applying signal gets reboot 2015-08-17T08:51:48 oh.. 2015-08-17T08:52:05 *headscratch* interesting 2015-08-17T08:52:14 :) 2015-08-17T08:52:33 I am not much aware about irq.c so 2015-08-17T08:52:46 do you also print the irq so that you can see that you are getting the gpio irq number? 2015-08-17T08:53:39 are you talking about vector no. 2015-08-17T08:59:50 yeah 2015-08-17T09:00:06 then you can tell the gpio hardware is actually doing something 2015-08-17T09:00:09 that is a good start 2015-08-17T09:01:30 yep I am getting vector no. 98 which is for GPIO bank 1 2015-08-17T09:02:29 I wrote print in rtems_gpio_bsp_get_vector() and getting it 2015-08-17T09:03:50 but beng-nl I think the problem is in irq.c right. I think BBB hardware is doing well for interrupt generation. 2015-08-17T09:04:27 oh yes 2015-08-17T09:04:31 too well :) 2015-08-17T09:05:16 ok, so what is the gpio configuration? when is it supposed to generate an interrupt? 2015-08-17T09:05:55 ahh. I wrote at HIGH_LEVEL on BBB_P8_15. 2015-08-17T09:06:02 aha 2015-08-17T09:06:51 so it can be found on AM335X_GPIO_IRQSTATUS_0 2015-08-17T09:07:37 but it's already generating interrupts without you applying the gpio signal though? 2015-08-17T09:08:44 umm. Then should it not reboot over infinite time. It only reboots when I apply the signal 2015-08-17T09:11:05 beng-nl: I found something interesting ! When I wrote "printk("%d\n", irq);" I got 67 2015-08-17T09:11:23 in bsp_interrupt_dispatch(void) @irq.c 2015-08-17T09:12:06 and in rtems_gpio_bsp_get_vector() printk gives me 98 @bbb-gpio.c 2015-08-17T09:12:42 and 67 is printed in infinite loop. 2015-08-17T09:16:46 aha 2015-08-17T09:17:14 so you are not getting the irq number you are expecting? 67 instead of 98? 2015-08-17T09:19:17 I think so .. but not sure. I think can you test the program on your bbb if possible ? 2015-08-17T09:46:55 beng-nl: I checked also pin is detecting proper interrupt too :) . When applied LOW_LEVEL doesn't respond. so only a thing is left to stop reboot. 2015-08-17T09:48:31 and in TRM also @ARM Cortex-A8 Interrupts Table 98 Int Number is assigned for "GPIOINT1A" 2015-08-17T10:50:44 [trac] 0001-Closes-ticket-2390-and-also-updates-the-RPI-implemen.patch (Added ben and gedare suggestions.) attached to #2390 by Asuol http://devel.rtems.org/ticket/2390 2015-08-17T12:11:07 *** gedare has joined #rtems 2015-08-17T12:11:08 *** ChanServ sets mode: +o gedare 2015-08-17T12:25:47 *** gedare has quit IRC 2015-08-17T12:27:37 [trac] #2390/defect (new) updated by beng (Assuming the patch works as intended (LGTM) I'm in favour of merging it asap as it's ...) http://devel.rtems.org/ticket/2390 2015-08-17T12:37:55 beng-nl: INTC_SIR_IRQ should be updated or not from gpio bsp functions ? I noticed irq.c is reading from that register 2015-08-17T12:40:12 *** ketul has quit IRC 2015-08-17T12:49:10 is there a register in the gpio register set to clear the irq btw? 2015-08-17T12:49:18 if that isn't done yet that is the first place to look 2015-08-17T12:53:10 kulp: interesting, you can trigger the irq by writing to GPIO_IRQSTATUS_RAW_0 - that should be very convenient for debugging 2015-08-17T12:54:39 aha, a write to GPIO_IRQSTATUS_0.n clears the irq 2015-08-17T13:35:52 *** dougk_ has quit IRC 2015-08-17T14:07:33 *** ketul has joined #rtems 2015-08-17T14:46:15 *** sgworks has joined #rtems 2015-08-17T14:46:41 hi beng-nl ! 2015-08-17T14:47:06 read your mail. trying to follow :) . Nice approach suggested 2015-08-17T14:47:43 *** sgworks has quit IRC 2015-08-17T14:55:30 [trac] #2391/defect (v:4.11) created by ricclaus ('python import rtems' fails in gdb) http://devel.rtems.org/ticket/2391 2015-08-17T14:56:18 [trac] 0001-Added-mutex-rbtrees-and-time.patch () attached to #2391 by ricclaus http://devel.rtems.org/ticket/2391 2015-08-17T14:58:07 *** hermann_19829 has joined #rtems 2015-08-17T15:01:16 [trac] SOCIS/2015 edited by hermann19829 (empty comment) http://devel.rtems.org/wiki/SOCIS/2015 2015-08-17T15:18:02 beng-nl: I amazed to see print (infinitely) from bsp_interrupt_dispatch() @irq.c in this simple application :- http://pastebin.com/ppkRm8Y1 2015-08-17T16:04:14 *** jkridner has joined #rtems 2015-08-17T16:07:25 *** igutek2 has quit IRC 2015-08-17T17:00:57 *** Davidbrcz has joined #rtems 2015-08-17T17:10:47 ketul: doesn't that mean that the irq is not getting cleared? 2015-08-17T17:10:53 in the interrupt handler? 2015-08-17T17:29:05 hi beng-nl. but shouldn't it be cleared when BBB reboots ? 2015-08-17T17:30:07 even when I have a app "rtems_test_begin(); rtems_test_end(); 2015-08-17T17:30:07 exit(0); " bsp_interrupt_dispatch() is called and printing 2015-08-17T17:59:29 *** hermann_19829_ has joined #rtems 2015-08-17T18:02:08 *** hermann_19829 has quit IRC 2015-08-17T18:25:50 ketul: aha 2015-08-17T18:27:17 beng-nl: can you tell me when bsp_interrupt_dispatch() would be called ? Is it the task given to cpu runs in parallel? 2015-08-17T18:28:57 ketul: yes, it should be cleared on reset. but it could be the gpio interrupt is triggered at the current (low) level 2015-08-17T18:30:17 ok. Then according to you no bug at irq.c side. right ? So we need not to see at that or ...? 2015-08-17T18:31:01 sent one mail of the recently tested code beng-nl 2015-08-17T18:32:10 ketul: i am honestly not sure where the bug is.. i know there are some shortcomings in irq.c. 2015-08-17T18:32:32 but the e.g. the timer interrupt is handled normally so *something* must be right :) 2015-08-17T18:33:59 ok. 67 is the vector no. for TINT1_1MS so that we are getting 2015-08-17T18:34:22 aha, that is fine 2015-08-17T18:34:29 yes that is normal then :) 2015-08-17T18:35:05 yup. 2015-08-17T18:35:29 beng-nl: can you check GPIO_SYSCONFIG and GPIO_SYSSTATUS 2015-08-17T18:35:59 register from TRM. Is that creating any reboot ? 2015-08-17T18:37:21 I am trying to poll them continuously by print in my app. Do you think that may solve our issue ? 2015-08-17T18:40:17 *** mumptai has joined #rtems 2015-08-17T18:43:13 *** mumptai has quit IRC 2015-08-17T18:44:56 *** dr__house|away is now known as dr__house 2015-08-17T18:45:43 *** mumptai has joined #rtems 2015-08-17T18:48:07 *** mumptai has joined #rtems 2015-08-17T18:53:20 *** Davidbrcz_ has joined #rtems 2015-08-17T18:56:04 *** monstr has quit IRC 2015-08-17T18:56:40 *** Davidbrcz has quit IRC 2015-08-17T18:57:29 ketul: filter out irc 67 for printing, and see if you get some better debug prints when gpio irq happens 2015-08-17T18:58:37 *** hermann_19829_ has quit IRC 2015-08-17T19:00:45 beng-nl: I am getting something. Let me do pastebin :) 2015-08-17T19:02:06 *** Davidbrcz_ has quit IRC 2015-08-17T19:03:32 beng-nl: this app doesn't reboot BBB :) -good news. But it is in infinite loop printing. That should be because I didn't cleared the irq_status register. http://pastebin.com/CgNtc9ii 2015-08-17T19:04:45 so according to me reboot is happening by mmio_read() or print() functions. I still need to figure out :) 2015-08-17T19:07:48 beng-nl: hooray ! finally working it ! So problem is created by printk() according to me :) . consumed a lot of time :( 2015-08-17T19:08:57 beng-nl: final code working :- http://pastebin.com/gGLj0cqn 2015-08-17T19:17:55 *** Davidbrcz_ has joined #rtems 2015-08-17T20:05:41 *** Davidbrcz_ has quit IRC 2015-08-17T20:06:49 *** gedare has joined #rtems 2015-08-17T20:06:49 *** ChanServ sets mode: +o gedare 2015-08-17T20:13:10 ketul: really! 2015-08-17T20:15:30 hmm. Just sent a mail :) 2015-08-17T20:18:02 really excellent detective work 2015-08-17T20:18:17 :D 2015-08-17T20:18:29 i don't think we are there yet.. judging from your email 2015-08-17T20:18:35 but you have found out a lot of tricky stuff, well done 2015-08-17T20:18:54 yahh. thanks a lot. 2015-08-17T20:19:03 if commenting out function calls stops a reset, could the stack be too small? 2015-08-17T20:19:36 yes high possibility I think 2015-08-17T20:19:44 is there a big variable on the stack in the generic gpio code or in the bbb bsp? 2015-08-17T20:19:51 like a big buffer or state struct? 2015-08-17T20:20:54 beng-nl: I tested it without gpio api too. same problem is also there 2015-08-17T20:21:49 still.. 2015-08-17T20:21:53 it only takes 1 or 2 functions 2015-08-17T20:22:00 currently testing with many no. of print statements. So that I can be able to come to conclusion that printk is the only function works !! 2015-08-17T20:22:40 ok 2015-08-17T20:22:51 so for this idea (stack) let's try a bigger stack size 2015-08-17T20:23:24 yes. but how can we increase stack size ? 2015-08-17T20:23:43 and yes Steve B also got the same problem because of puts() 2015-08-17T20:24:51 it seems #define CONFIGURE_INIT_TASK_STACK_SIZE should do it, like e.g. in fstests 2015-08-17T20:25:53 yep good point. Let me do it. and yes all the printk works :) 2015-08-17T20:26:07 interesting :) 2015-08-17T20:26:11 you are getting close i think 2015-08-17T20:26:52 yes ! beng-nl currently it is 2 * RTEMS_MINIMUM_STACK_SIZE 2015-08-17T20:27:11 should it be 10 ? or any guess ;) 2015-08-17T20:27:21 there is plenty of memory so let's go crazy 2015-08-17T20:27:28 yeah 10 why not :) 2015-08-17T20:27:38 yaah ok :D 2015-08-17T20:27:50 ketul: do you have a fast way to iterate (recompile & run)? or are you reshuffling microsd cards? 2015-08-17T20:28:09 reshuffling microsd cards :( 2015-08-17T20:29:24 but different tabs in terminal so up arrow will serve all the things faster :) 2015-08-17T20:30:22 you are learning fast :-) 2015-08-17T20:30:31 beng-nl: you can push the patch on 2390 if you like it 2015-08-17T20:30:37 :) 2015-08-17T20:31:34 gedare: aha, ok; don't we need the ok from the rpi owner? (or is that andrei) 2015-08-17T20:31:52 (andre*) 2015-08-17T20:32:15 no there isn't one, so i'm giving it. 2015-08-17T20:32:34 Alan Cudmore is the main RPi person, plus the slew of GSoC students 2015-08-17T20:32:34 hmm 2015-08-17T20:32:56 i'll post on the trac to 2015-08-17T20:32:59 gedare: i see. hmm.. normally i'd do a build test before i push someting myself.. ehr 2015-08-17T20:33:03 [trac] #2390/defect (new) updated by gedare (go ahead) http://devel.rtems.org/ticket/2390 2015-08-17T20:33:22 that's a good idea. ;) 2015-08-17T20:35:41 beng-nl: :( . got reboot comment removed from val = rtems_gpio_get_value(SW1); should CONFIGURE_EXTRA_TASK_STACKS be increased ? 2015-08-17T20:35:57 *** Davidbrcz_ has joined #rtems 2015-08-17T20:36:36 try it yeah. fstests doesn't but.. yes let's try 2015-08-17T20:36:52 ok 2015-08-17T20:40:40 ok let's see how to build a rpi bsp :) 2015-08-17T20:41:16 beng-nl: unfortunately didn't work. I went up to 100 2015-08-17T20:41:23 drat. 2015-08-17T20:41:26 ok, nice try :) 2015-08-17T20:41:38 does another macros make effect ? 2015-08-17T20:41:54 "CONFIGURE_MAXIMUM_TASKS " ? 2015-08-17T20:41:56 ketul: can you commit & merge (in your github) your current state so i can experiment too a bit? 2015-08-17T20:42:03 ketul: noo that's just about threads 2015-08-17T20:42:11 ohh ya. 2015-08-17T20:42:18 sure just a minute :) 2015-08-17T20:42:45 i still think this is likely to be an infinite recursion problem - so still stack related, just unlimited; not sure of course, but let's test this hypothesis 2015-08-17T20:43:15 beng-nl: should I commit test cases too ? 2015-08-17T20:43:19 I mean app 2015-08-17T20:45:05 sure 2015-08-17T20:45:07 yes please 2015-08-17T20:45:08 :) 2015-08-17T20:46:42 beng-nl: done :) . https://github.com/ketul93/rtems/tree/GPIO_BBB 2015-08-17T20:46:48 ok 2015-08-17T20:46:54 give me some time, i have to rebuild a lot of stuff 2015-08-17T20:47:00 *** igutek has joined #rtems 2015-08-17T20:47:29 aha ok. here 2:17 a.m.. But yes I can work for several time :) 2015-08-17T20:47:37 :-) 2015-08-17T20:49:12 beng-nl: one question is like then why printk() works ? Is that because it has nothing to do with bbb registers ? So cpsr couldn't be affected ? 2015-08-17T20:51:54 why printk works at all? 2015-08-17T20:52:03 or does the whole app work perfectly once you add printk()'s? 2015-08-17T20:52:55 app works perfectly when commenting the functions like - mmio_read() 2015-08-17T20:52:55 - rtems_gpio_get_value() 2015-08-17T20:52:55 - rtems_gpio_set() 2015-08-17T20:52:55 - rtems_gpio_clear() 2015-08-17T20:53:19 if printk() is only there then bbb doesn't reboot 2015-08-17T20:54:05 I mean printk() (as many you want)can be included rtems_gpio_irq_state fun() 2015-08-17T20:57:09 beng-nl: can you test the other 2 patches Andre posted? [PATCH 1/2] Removed GPIO API from the libbsp/shared Makefile. [PATCH 2/2] Added GPIO API sources to rpi Makefile 2015-08-17T20:57:26 they look pretty straightforward but since you're spinning up tools... 2015-08-17T20:58:37 beng-nl: But it is true that with the help of printk() we can be able to go to the root of the "reboot" issue ;) 2015-08-17T21:00:25 gedare: i'm happy to try to build but have no hardware 2015-08-17T21:00:46 rpi that is ;-) 2015-08-17T21:02:57 that's ok it's a build system patch 2015-08-17T21:03:04 any way if it break rpi someone will complain 2015-08-17T21:03:10 *eventually* 2015-08-17T21:03:12 gtg by 2015-08-17T21:03:13 e 2015-08-17T21:03:17 ok gedare 2015-08-17T21:03:20 bye 2015-08-17T21:03:38 I do have rpi :) . But don't know how to build bsp and all 2015-08-17T21:03:56 gedare: good bye 2015-08-17T21:05:47 *** igutek has quit IRC 2015-08-17T21:08:01 *** gedare has quit IRC 2015-08-17T21:09:29 * ketul is writing the gsoc post for my project :) 2015-08-17T21:11:33 well done :) 2015-08-17T21:33:20 beng-nl: I think I need to sleep down as yoga session after 4 hours. Is it ok ? Can you please send an email for your updated work 2015-08-17T21:33:34 after getting it over :) 2015-08-17T21:33:54 ketul: i won't promise any progress but yes, please sleep for your yoga session :0 2015-08-17T21:33:57 :) 2015-08-17T21:34:47 ok np. Let's see how much detective work should we supposed to do ? :D 2015-08-17T21:34:59 yes :- 2015-08-17T21:35:00 yes :-) 2015-08-17T21:35:09 ok good bye 2015-08-17T21:36:26 *** ketul has quit IRC 2015-08-17T22:08:27 *** mumptai has quit IRC 2015-08-17T22:38:47 *** gedare has joined #rtems 2015-08-17T22:38:48 *** ChanServ sets mode: +o gedare 2015-08-17T22:43:21 *** gedare has quit IRC 2015-08-17T23:25:50 *** kiwichris has joined #rtems 2015-08-18T00:30:50 [trac] #2390/defect (new) updated by Andre Marques (In [changeset:"b09a578e8ad4583a97349f866e981980f0ff6ade/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2390 2015-08-18T00:30:51 [git] b09a578 by Andre Marques: [Closes ticket #2390, and also updates the RPI implementation. makes rtems_gpio_bsp_get_value return uint32_t. Motivation: simplify beagle gpio implementation for common gpio apio.] http://devel.rtems.org/changeset/b09a578/rtems 2015-08-18T02:15:50 *** Davidbrcz_ has quit IRC 2015-08-18T03:26:45 *** Davidbrcz_ has joined #rtems 2015-08-18T03:46:20 *** verm__ has quit IRC 2015-08-18T03:46:20 *** verm__ has joined #rtems 2015-08-18T04:00:00 *** stryx`_ has quit IRC 2015-08-18T04:00:08 *** stryx`__ has joined #rtems 2015-08-18T04:02:45 *** dr__house is now known as dr__house|away 2015-08-18T04:06:31 *** crazy3k has quit IRC 2015-08-18T04:37:19 *** monstr has joined #rtems 2015-08-18T04:42:08 *** stryx`__ has quit IRC 2015-08-18T04:48:50 *** stryx` has joined #rtems 2015-08-18T06:49:23 *** Davidbrcz_ has quit IRC 2015-08-18T06:51:02 *** crazy3k has joined #rtems 2015-08-18T07:35:24 *** mumptai has joined #rtems 2015-08-18T08:00:35 *** kiwichris has quit IRC 2015-08-18T08:13:37 *** mumptai has quit IRC 2015-08-18T10:06:55 *** crazy3k has quit IRC 2015-08-18T10:42:01 *** ketul has joined #rtems 2015-08-18T11:38:09 *** peerst has quit IRC 2015-08-18T12:43:02 hi beng-nl ! 2015-08-18T12:44:45 *** peerst has joined #rtems 2015-08-18T12:47:31 hi ketul 2015-08-18T12:47:51 ketul: i didn't get around to our gpio problem, sorry again. i just ran out of time after looking at the rpi stuff & the gpio patch 2015-08-18T12:48:14 ketul: but at least the gpio api change is merged so we could merge our gpio work so far, what do you think? 2015-08-18T12:48:23 beng-nl: It's ok. :) I read your mail 2015-08-18T12:48:29 sure we should merge it ! 2015-08-18T12:48:58 ok, i'll pull and rebase your work and do some build tests etc 2015-08-18T12:49:09 so I am testing it resistor mode 2015-08-18T12:49:42 if I set it as pull_up should it give me 3.3 v right ? 2015-08-18T12:50:50 beng-nl: Should I push my work at new branch that can be merged easily ? 2015-08-18T12:50:50 *** peerst has quit IRC 2015-08-18T12:51:26 ketul: yes 3.3v 2015-08-18T12:51:37 ketul: i'll do it (the commit) 2015-08-18T12:52:19 :( . Not getting 3.3 v .So still 2 things remaining resistor mode and iqr 2015-08-18T12:52:28 irq* 2015-08-18T12:52:39 what are you getting? 2015-08-18T12:52:43 0 2015-08-18T12:52:46 aha 2015-08-18T12:52:53 which is totally not fine 2015-08-18T12:52:58 :) 2015-08-18T12:53:08 ketul: cpukit/include/rtems/gpio.h, that should be somewhere else right? 2015-08-18T12:54:12 ah. Are u talking about the gpio api ? 2015-08-18T12:54:21 yeah cpukit/include/rtems/gpio.h in your change 2015-08-18T12:54:30 that should be beagle-specific, or? 2015-08-18T12:55:06 aha. That is too old ! After that andre committed new api 2015-08-18T12:55:13 beng-nl: let me do one thing 2015-08-18T12:55:33 I think I should make a new branch and push all the updated code 2015-08-18T12:55:46 okay, i'll wait for you then 2015-08-18T12:55:58 sure. I will post it asap :) 2015-08-18T12:56:31 let's try and merge it today, before or during the meeting 2015-08-18T12:57:08 ok sure, But I think today there is no meeting 2015-08-18T12:57:14 ah ok 2015-08-18T12:57:17 right i remember 2015-08-18T12:57:22 ok let's try and merge it today 2015-08-18T12:57:28 and then also work on irq & pullup 2015-08-18T12:57:51 sure ! 2015-08-18T12:58:43 so currently I will write RTEMS_NOT_DEFINED in irq and resistor mode function. right ? 2015-08-18T12:59:42 if that builds, fine 2015-08-18T12:59:47 for bbxm & bbb 2015-08-18T13:00:07 ok. for bbxm it is already RTEMS_NOT_DEFINED 2015-08-18T13:00:12 iok 2015-08-18T13:00:13 ok 2015-08-18T13:08:08 ketul: btw there is a user button on the bbb that is a gpio input, good test target perhaps if you hadn't considered it yet 2015-08-18T13:09:05 ohh yep. Didn't take look at that one 2015-08-18T13:41:43 ketul: so the gpio.h in cpukit/ shouldn't be there right? 2015-08-18T13:41:59 we should use the generic gpio.h 2015-08-18T13:42:00 ? 2015-08-18T13:43:44 Definitely ! 2015-08-18T13:43:49 beng-nl: ^ 2015-08-18T13:48:25 ok 2015-08-18T13:48:29 so you'll remove that file? 2015-08-18T13:48:33 from your commit? 2015-08-18T13:48:52 yes I will 2015-08-18T13:50:32 beng-nl: and yes it is also not there in current branch i.e. https://github.com/ketul93/rtems/tree/GPIO_BBB/cpukit/include/rtems 2015-08-18T13:50:48 oh.. ehm 2015-08-18T13:50:56 i was using GPIO_API branch 2015-08-18T13:50:59 GPIO_BBB is current? 2015-08-18T13:51:06 yep. 2015-08-18T13:51:09 ok 2015-08-18T13:51:19 But new one is coming in a short :) 2015-08-18T13:51:22 ok :) 2015-08-18T13:51:23 let me know :) 2015-08-18T13:52:03 sure. beng-nl what about andre's patch on DrJoel's issue of bsp broke 2015-08-18T13:52:21 eh? where's that? 2015-08-18T13:52:46 *** igutek has joined #rtems 2015-08-18T13:52:47 "Added GPIO API sources to rpi Makefile" 2015-08-18T13:53:16 oh, you think i should merge that? 2015-08-18T13:53:44 i will when i have some time, it isn't blocking us, is it? 2015-08-18T13:54:01 I think so as DrJoel mailed about the bsp broke for some 3 boards because of same name. 2015-08-18T13:54:33 beng-nl: not kind of blocking. But if so I can add that source of api in bsp Makefile 2015-08-18T13:55:27 aha, i see it 2015-08-18T13:55:33 hmm let's leave that for now and get our own stuff done first 2015-08-18T13:55:41 hmm. ok 2015-08-18T13:57:35 *** peerst has joined #rtems 2015-08-18T14:08:28 *** ketul has quit IRC 2015-08-18T14:13:54 *** ketul has joined #rtems 2015-08-18T14:33:36 beng-nl: https://github.com/ketul93/rtems/tree/GPIO_DRV_BBB 2015-08-18T14:39:14 Built successfully for both BB-xM and BBB :) Kindly test it so can be merged asap 2015-08-18T14:41:53 *** gtament has joined #rtems 2015-08-18T14:43:30 branch is GPIO_DRV_BBB ? 2015-08-18T14:46:05 Hello, everyone! 2015-08-18T14:48:21 hi gtament 2015-08-18T14:49:53 We have meeting today, aren't we? 2015-08-18T14:53:02 *** sgworks has joined #rtems 2015-08-18T14:59:08 [git] f795d62 by Hesham Almatary: [doc: zlib1g-dev package needs to be installed on Linux Mint] http://devel.rtems.org/changeset/f795d62/rtems-source-builder 2015-08-18T14:59:29 *** EdSut has joined #rtems 2015-08-18T14:59:40 *** ketul has quit IRC 2015-08-18T14:59:58 [git] 0d59530 by Hesham Almatary: [doc: zlib1g-dev package needs to be installed on Linux Mint] http://devel.rtems.org/changeset/0d59530/rtems-source-builder 2015-08-18T15:02:37 We have meeting today, don't we? 2015-08-18T15:03:04 I also don't know 2015-08-18T15:03:07 may be not 2015-08-18T15:04:44 It seems like no. Gedare anŠ² Joel are not online. But I don't remember, they said something about no meeting on next tuesday last time 2015-08-18T15:06:26 I also don't remember but other people are also not online 2015-08-18T15:06:42 so I guess last meeting was the concluding meeting 2015-08-18T15:06:56 ok then see you. 2015-08-18T15:06:59 *** sgworks has quit IRC 2015-08-18T15:07:23 *** EdSut has quit IRC 2015-08-18T15:07:35 *** gtament has quit IRC 2015-08-18T15:13:55 *** ketul has joined #rtems 2015-08-18T15:14:05 hi beng-nl 2015-08-18T15:14:07 yes ! 2015-08-18T15:21:05 *** jrcatbagan has joined #rtems 2015-08-18T15:33:09 *** sgworks has joined #rtems 2015-08-18T15:36:39 *** peerst has quit IRC 2015-08-18T15:40:17 *** anandkp92_ has joined #rtems 2015-08-18T16:07:05 *** peerst has joined #rtems 2015-08-18T16:13:28 *** dr__house|away is now known as dr__house 2015-08-18T16:15:01 *** hermann_19829 has joined #rtems 2015-08-18T16:52:16 *** hermann_19829 has quit IRC 2015-08-18T16:53:05 *** mumptai has joined #rtems 2015-08-18T17:12:32 *** Davidbrcz has joined #rtems 2015-08-18T17:16:50 [git] 151e53f by Ketul Shah: [Beagle: GPIO support (for BBB) GPIO Driver Development for BeagleBone Black based on the generic GPIO API] http://devel.rtems.org/changeset/151e53f/rtems 2015-08-18T17:16:56 ketul: congratulations :-) 2015-08-18T17:23:51 :D beng-nl Thanks a lot ! 2015-08-18T17:25:19 a great motivational event for me :) 2015-08-18T17:25:54 beng-nl: as a curiosity is it merged on github ? 2015-08-18T17:31:46 ketul: well deserved :-) 2015-08-18T17:31:50 ketul: github? 2015-08-18T17:32:08 yup I mean https://github.com/RTEMS/rtems 2015-08-18T17:32:43 ketul: it seems it is a little bit behind. guess it'll catch up 2015-08-18T17:33:04 umm. ok :) 2015-08-18T17:33:14 it is on https://git.rtems.org/rtems/ :) 2015-08-18T17:33:51 hooray ! 2015-08-18T17:34:07 now let's do continue our remaining work ! 2015-08-18T17:34:12 yes :-) 2015-08-18T17:34:59 ketul: are you on twitter? 2015-08-18T17:35:04 i wanna tweet about it :) 2015-08-18T17:35:09 umm. No 2015-08-18T17:35:12 :( 2015-08-18T17:35:16 ok np 2015-08-18T17:35:43 fb and LinkeIn :) 2015-08-18T17:40:46 beng-nl: just signed up :) https://twitter.com/ketul1shah 2015-08-18T17:46:24 ketul: i @-mentioned you, your star is rising :-) 2015-08-18T17:46:38 :D yep 2015-08-18T17:54:27 *** andrei___ has joined #rtems 2015-08-18T18:26:15 *** anandkp92_ has quit IRC 2015-08-18T18:54:16 *** monstr has quit IRC 2015-08-18T19:37:18 *** andrei___ has quit IRC 2015-08-18T20:06:06 beng-nl: sent a mail about gsoc post :) 2015-08-18T20:11:19 *** Davidbrcz has quit IRC 2015-08-18T20:36:05 *** ketul has quit IRC 2015-08-18T20:38:58 *** Davidbrcz has joined #rtems 2015-08-18T20:49:06 *** Davidbrcz has quit IRC 2015-08-18T20:51:29 got it & handled :) 2015-08-18T21:43:36 *** jkridner has quit IRC 2015-08-18T21:43:49 *** guerby has quit IRC 2015-08-18T21:44:17 *** jkridner has joined #rtems 2015-08-18T21:44:54 *** guerby has joined #rtems 2015-08-18T22:12:07 *** kiwichris has joined #rtems 2015-08-18T22:28:42 *** Davidbrcz has joined #rtems 2015-08-18T22:59:17 *** Davidbrcz has quit IRC 2015-08-18T23:50:39 *** Davidbrcz has joined #rtems 2015-08-19T01:25:11 *** gedare has joined #rtems 2015-08-19T01:25:11 *** ChanServ sets mode: +o gedare 2015-08-19T01:30:46 *** gedare has quit IRC 2015-08-19T02:33:17 *** Davidbrcz has quit IRC 2015-08-19T03:39:18 *** notbmatt has quit IRC 2015-08-19T04:14:07 *** Davidbrcz has joined #rtems 2015-08-19T04:49:47 *** guerby has quit IRC 2015-08-19T04:56:40 *** monstr has joined #rtems 2015-08-19T06:30:05 *** Davidbrcz has quit IRC 2015-08-19T06:59:49 *** guerby has joined #rtems 2015-08-19T08:20:28 *** kiwichris has quit IRC 2015-08-19T08:28:15 *** zoso has joined #rtems 2015-08-19T08:56:00 *** mumptai has quit IRC 2015-08-19T09:46:33 *** zoso has quit IRC 2015-08-19T09:47:14 *** ketul has joined #rtems 2015-08-19T10:17:50 *** jkridner has quit IRC 2015-08-19T10:19:27 *** jkridner has joined #rtems 2015-08-19T10:22:18 *** zoso has joined #rtems 2015-08-19T10:29:22 *** ketul has quit IRC 2015-08-19T10:41:18 *** ketul has joined #rtems 2015-08-19T10:58:27 *** zoso has quit IRC 2015-08-19T11:13:56 *** hesham has joined #rtems 2015-08-19T11:54:36 [trac] #2392/defect (v:4.10) created by karagjahan9 (Do You Know What Natural Penis Exercises Can Do in Your Sex Life with Testosterone Boosters) http://devel.rtems.org/ticket/2392 2015-08-19T13:54:17 *** ketul has quit IRC 2015-08-19T14:14:42 *** hermann_19829 has joined #rtems 2015-08-19T15:07:05 [trac] #2393/defect (v:4.10) created by dcb (rtems/c/src/lib/libbsp/sparc/shared/1553/b1553brm.c:992: bad if test ?) http://devel.rtems.org/ticket/2393 2015-08-19T15:18:24 *** hermann_19829 has quit IRC 2015-08-19T15:25:24 *** antgreen has joined #rtems 2015-08-19T16:03:51 *** hesham has quit IRC 2015-08-19T16:30:03 *** Davidbrcz has joined #rtems 2015-08-19T16:52:02 *** Davidbrcz has quit IRC 2015-08-19T17:00:36 *** mumptai has joined #rtems 2015-08-19T17:22:01 *** antgreen has quit IRC 2015-08-19T17:24:15 *** andrei_ has joined #rtems 2015-08-19T17:47:48 *** monstr has quit IRC 2015-08-19T17:51:41 *** zoso has joined #rtems 2015-08-19T17:55:55 *** Davidbrcz has joined #rtems 2015-08-19T17:59:42 *** arvind_k has joined #rtems 2015-08-19T18:02:51 *** zoso has quit IRC 2015-08-19T18:24:53 *** krohini has joined #rtems 2015-08-19T19:08:32 *** crazy3k has joined #rtems 2015-08-19T19:09:30 *** krohini has quit IRC 2015-08-19T19:11:07 *** arvind_k has quit IRC 2015-08-19T19:32:49 *** hermann_19829 has joined #rtems 2015-08-19T19:33:06 *** krohini has joined #rtems 2015-08-19T19:33:49 *** krohini has quit IRC 2015-08-19T20:00:21 *** Davidbrcz has quit IRC 2015-08-19T20:38:49 *** Davidbrcz has joined #rtems 2015-08-19T20:41:44 *** Davidbrcz has quit IRC 2015-08-19T20:42:08 *** Davidbrcz has joined #rtems 2015-08-19T20:59:03 *** hermann_19829 has quit IRC 2015-08-19T20:59:17 [trac] #2392/defect (closed) updated by beng (empty comment) http://devel.rtems.org/ticket/2392 2015-08-19T21:02:10 *** gedare has joined #rtems 2015-08-19T21:02:10 *** ChanServ sets mode: +o gedare 2015-08-19T21:06:43 *** gedare has quit IRC 2015-08-19T21:18:21 *** mumptai has quit IRC 2015-08-19T21:21:14 *** Davidbrcz has quit IRC 2015-08-19T21:41:56 *** Davidbrcz has joined #rtems 2015-08-19T21:47:20 *** kiwichris has joined #rtems 2015-08-19T22:29:24 *** andrei_ has quit IRC 2015-08-19T23:05:11 *** Davidbrcz has quit IRC 2015-08-19T23:18:36 *** Davidbrcz has joined #rtems 2015-08-19T23:58:50 *** gedare has joined #rtems 2015-08-19T23:58:50 *** ChanServ sets mode: +o gedare 2015-08-20T00:07:06 *** gedare has quit IRC 2015-08-20T02:42:40 *** Davidbrcz has quit IRC 2015-08-20T03:29:41 *** dr__house is now known as dr__house|away 2015-08-20T04:45:13 *** ketul has joined #rtems 2015-08-20T05:02:18 *** Davidbrcz has joined #rtems 2015-08-20T05:45:59 *** monstr has joined #rtems 2015-08-20T05:59:01 *** mumptai has joined #rtems 2015-08-20T06:03:46 *** zoso has joined #rtems 2015-08-20T06:19:21 *** Davidbrcz has quit IRC 2015-08-20T07:42:23 *** mumptai has quit IRC 2015-08-20T07:55:55 *** ketul has quit IRC 2015-08-20T08:08:31 *** kiwichris has quit IRC 2015-08-20T08:33:26 *** zoso has quit IRC 2015-08-20T08:45:41 *** hesham has joined #rtems 2015-08-20T09:26:27 *** guillaum1 has joined #rtems 2015-08-20T09:28:09 * guillaum1 just got a promising HelloWorld in Java on RTEMS/sis 2015-08-20T10:15:20 * guillaum1 just got a Java garbage collector working 2015-08-20T10:39:43 *** guillaum1 has quit IRC 2015-08-20T10:41:48 *** gedare has joined #rtems 2015-08-20T10:41:48 *** ChanServ sets mode: +o gedare 2015-08-20T10:50:26 *** gedare has quit IRC 2015-08-20T11:37:13 :) 2015-08-20T12:20:50 *** guillaum1 has joined #rtems 2015-08-20T12:47:17 *** guillaum1 has quit IRC 2015-08-20T13:08:14 *** guillaum1 has joined #rtems 2015-08-20T13:24:29 *** guillaum1 has quit IRC 2015-08-20T13:29:40 *** guillaum1 has joined #rtems 2015-08-20T13:35:56 *** dougk_ has joined #rtems 2015-08-20T14:42:30 *** monstr has quit IRC 2015-08-20T14:47:00 *** ketul has joined #rtems 2015-08-20T14:51:11 hi beng-nl ! 2015-08-20T14:53:22 hi ketul ! 2015-08-20T14:54:17 beng-nl: Now we can add in Black/White and Green too :) 2015-08-20T14:54:24 Am335X Soc 2015-08-20T14:54:26 haha, the beaglebone green? 2015-08-20T14:54:29 yep 2015-08-20T14:54:35 yeah, quite cool and amazingly cheap 2015-08-20T14:54:48 :-) 2015-08-20T14:54:55 oh why hello jkridner :-) 2015-08-20T14:55:05 hi jkridner :) 2015-08-20T14:55:13 hi ketul, beng-nl 2015-08-20T14:55:43 jkridner: gsoc is nearly over, all our 3 beagle students did fantastic work 2015-08-20T14:56:08 jkridner: a lot more support for the beagles on rtems now.. your donation was a great investment :) 2015-08-20T14:57:18 in fact the bootloader project is good for more than only rtems most likely 2015-08-20T14:57:26 beng-nl: agreed ! We also got a amazing community support and mentor like beng-nl :) 2015-08-20T14:57:36 haha *blushes* 2015-08-20T14:57:56 hi all 2015-08-20T14:58:00 hi guillaum1 2015-08-20T14:58:11 ketul: i sure wish we have a jvm on rtems 2015-08-20T14:58:19 ketul: if only some hero were working on that 2015-08-20T14:58:24 :-D 2015-08-20T14:58:41 ;-) 2015-08-20T14:59:38 +1 2015-08-20T15:02:59 beng-nl: reading "REAL-TIME JAVA IN SPACE: POTENTIAL..." 2015-08-20T15:03:05 :) 2015-08-20T15:05:37 beng-nl: wish I'd come through with some of the software license stuff. 2015-08-20T15:06:16 oh, starterware? 2015-08-20T15:06:45 i you're up against a gaggle of lawyers there who see risk and no benefit for TI 2015-08-20T15:06:53 i guess* 2015-08-20T15:14:06 *** dr__house|away is now known as dr__house 2015-08-20T15:17:59 *** guillaum1 has quit IRC 2015-08-20T15:34:55 *** ketul has quit IRC 2015-08-20T16:41:12 *** Davidbrcz has joined #rtems 2015-08-20T16:58:14 *** Davidbrcz has quit IRC 2015-08-20T17:21:54 *** mumptai has joined #rtems 2015-08-20T17:39:06 *** Davidbrcz has joined #rtems 2015-08-20T18:54:50 *** hesham has quit IRC 2015-08-20T19:06:27 *** hermann_19829 has joined #rtems 2015-08-20T19:27:16 *** edsiper has quit IRC 2015-08-20T19:27:43 *** edsiper has joined #rtems 2015-08-20T19:28:49 *** dougk_ has quit IRC 2015-08-20T19:29:20 *** cdesai has quit IRC 2015-08-20T19:30:16 *** cdesai has joined #rtems 2015-08-20T19:42:57 *** dougk_ has joined #rtems 2015-08-20T20:12:24 *** Davidbrcz has quit IRC 2015-08-20T20:28:30 *** dougk_ has quit IRC 2015-08-20T20:47:56 [trac] GSoC/2015/RaspberryPi2Support edited by krohini (empty comment) http://devel.rtems.org/wiki/GSoC/2015/RaspberryPi2Support 2015-08-20T20:49:15 *** Davidbrcz has joined #rtems 2015-08-20T20:54:46 *** hermann_19829 has quit IRC 2015-08-20T22:06:58 *** crazy3k has quit IRC 2015-08-20T22:16:50 *** mumptai has quit IRC 2015-08-20T23:19:08 *** Davidbrcz has quit IRC 2015-08-20T23:19:17 *** Davidbrcz has joined #rtems 2015-08-20T23:23:41 *** Davidbrcz has quit IRC 2015-08-20T23:45:24 hi all 2015-08-21T00:08:38 *** Davidbrcz has joined #rtems 2015-08-21T00:45:31 *** kiwichris has joined #rtems 2015-08-21T02:07:47 *** geheimni2` has quit IRC 2015-08-21T02:52:00 *** Davidbrcz has quit IRC 2015-08-21T03:56:13 *** Davidbrcz has joined #rtems 2015-08-21T04:07:56 *** Davidbrcz has quit IRC 2015-08-21T04:45:21 *** mumptai has joined #rtems 2015-08-21T05:10:36 *** ketul has joined #rtems 2015-08-21T05:19:53 *** monstr has joined #rtems 2015-08-21T05:24:30 *** monstr has quit IRC 2015-08-21T05:30:39 *** zoso has joined #rtems 2015-08-21T05:37:35 *** monstr has joined #rtems 2015-08-21T06:01:10 *** mumptai has quit IRC 2015-08-21T06:28:06 *** Davidbrcz has joined #rtems 2015-08-21T06:47:22 *** mumptai has joined #rtems 2015-08-21T07:01:44 *** Davidbrcz has quit IRC 2015-08-21T07:10:03 *** mumptai has quit IRC 2015-08-21T07:48:51 *** hesham has joined #rtems 2015-08-21T08:15:10 *** kiwichris has quit IRC 2015-08-21T09:05:56 *** geheimnis` has joined #rtems 2015-08-21T09:28:44 *** ketul has quit IRC 2015-08-21T09:30:26 *** ketul has joined #rtems 2015-08-21T09:44:12 *** zoso has quit IRC 2015-08-21T09:59:58 *** zoso has joined #rtems 2015-08-21T11:11:16 hi beng-nl ! 2015-08-21T11:28:36 beng-nl: tested pull up/down and got output successfully :) . Test app :- http://pastebin.com/PpELxKGi 2015-08-21T11:48:49 *** zoso has quit IRC 2015-08-21T12:02:49 [trac] GSoC/2015/RaspberryPi_peripherals_and_SD_card created http://devel.rtems.org/wiki/GSoC/2015/RaspberryPi_peripherals_and_SD_card 2015-08-21T12:03:46 [trac] GSoC/2015 edited by Asuol (Added wiki page) http://devel.rtems.org/wiki/GSoC/2015 2015-08-21T12:05:30 [trac] GSoC/2015 edited by Asuol (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-08-21T12:39:21 *** rokka has quit IRC 2015-08-21T12:49:36 *** rokka has joined #rtems 2015-08-21T13:06:49 *** beng-nl has quit IRC 2015-08-21T13:10:51 *** beng-nl has joined #rtems 2015-08-21T13:14:22 ketul: nice 2015-08-21T13:44:27 *** monstr has quit IRC 2015-08-21T14:04:54 *** gedare has joined #rtems 2015-08-21T14:04:54 *** ChanServ sets mode: +o gedare 2015-08-21T14:08:05 verm__: gtament has complained he can't make edits because of being the SPAM group 2015-08-21T14:08:14 verm__: was there maybe a problem with the re-configuration? 2015-08-21T14:08:27 verm__: I can't see the admin/general/perm page to make changes 2015-08-21T14:10:20 verm__: email sent. 2015-08-21T14:14:28 beng-nl: :) . have you tested ? 2015-08-21T14:15:53 gedare: hi. One student mailed me from blog about one issue. His pastebin :- http://pastebin.com/PC2Vz72U . Can you give some solution for that 2015-08-21T14:34:03 *** dougk_ has joined #rtems 2015-08-21T14:38:37 hi ketul 2015-08-21T14:39:06 ketul: advise the student to sign up for users mailing list and post questions for help there 2015-08-21T14:39:24 ok. That I did :) 2015-08-21T14:40:16 nothing jumps out at me 2015-08-21T14:40:22 is "beagleboneblack" the right name for the bsp? 2015-08-21T14:41:04 yep . copied from the blog of beng-nl . I told him to send mail but not yet sent 2015-08-21T14:41:14 yes, that name s ok 2015-08-21T14:41:15 is* ok 2015-08-21T14:41:38 I think qemu is making this issue. 2015-08-21T14:42:56 gedare, beng-nl anything else required from our side as the gsoc deadline approaching :) 2015-08-21T14:43:52 the gpio irq thing seems like a appropriate & appropriately sized thing to try to still solve 2015-08-21T14:45:00 beng-nl: can you suggest approach for that ? 2015-08-21T14:45:16 As per TRM I found all the things fine but .. 2015-08-21T14:46:22 have you tried to compare to the minix code? it implements the same functionality 2015-08-21T14:46:37 you can see if the registers programming is reasonable 2015-08-21T14:47:20 ketul: fix the spacing on the line: 2015-08-21T14:47:24 + if (gpio_pad_conf[bank][pin] == CONF_NOT_DEFINED){ 2015-08-21T14:47:30 re: BBB Resistor mode 2015-08-21T14:48:05 also ketul, if it the case that PULL_DOWN and NO_PULL have the identical code, you should combine those cases 2015-08-21T14:48:07 e.g. 2015-08-21T14:48:21 case PULL_DOWN: case NO_PULL_RESISTOR: mmio_clear(...); break; 2015-08-21T14:49:58 ketul: (beng-nl) no more code is required. please get your documentation finished, prepare code for upload to melange, add your entry to the GDoc report if you haven't, and prepare/submit patches. I hope you will continue with RTEMS after the deadilne, and working on the GPIO IRQ would be a nice little project to chip away at as your time permits. 2015-08-21T14:50:17 +1 2015-08-21T14:50:29 ketul has submitted a nice entry for the post, it's in the gdoc 2015-08-21T14:50:34 ragunath too 2015-08-21T14:50:38 ketul: the email came from Hardik into devel mailing list I think 2015-08-21T14:50:57 ok great thanks beng-nl i haven't had time to look. started a new job this week. 2015-08-21T14:51:06 tenure track professor. :0 2015-08-21T14:51:39 gedare: good point committed a small mistake in that switch case .. Thanks 2015-08-21T14:52:07 Couldn't find in hardware test as both give 0v. 2015-08-21T15:00:34 beng-nl: looked at https://github.com/minix3/minix/blob/master/minix/lib/libgpio/gpio_omap.c . Same thing using set32() 2015-08-21T15:01:07 good start 2015-08-21T15:02:28 ketul: you also do a write32(bank->base_address + regs->IRQSTATUS, 0xffffffff); (or similar) to clear the interrupts in the irq handler? 2015-08-21T15:03:25 IRQSTATUS is offset 0x02C btw 2015-08-21T15:04:36 beng-nl: same effect is here :- /* Retrieve the interrupt event status. */ 2015-08-21T15:04:36 event_status = mmio_read(bbb_reg(bank_nr, AM335X_GPIO_IRQSTATUS_0)); 2015-08-21T15:04:36 /* Clear the interrupt line. */ 2015-08-21T15:04:36 mmio_write( 2015-08-21T15:04:36 (bbb_reg(bank_nr, AM335X_GPIO_IRQSTATUS_0)), event_status); 2015-08-21T15:04:52 in rtems_gpio_bsp_interrupt_line() 2015-08-21T15:06:35 beng-nl: and code updated as gedare suggested :) 2015-08-21T15:09:53 0x02C and 0x2C would be same as type casting could be there in bbb_conf_reg() . So I think problem is not at the register addressing and all. Because print() works perfectly with the same code :) 2015-08-21T15:19:24 ok, so i see that interrupts are on for HIGH_LEVEL 2015-08-21T15:20:32 i can see how that will turn on an infinite number of interrupts; that may be the problem as soon as that irq is turned on again, it'll re-occur 2015-08-21T15:21:14 could you try GPIO_FALLINGDETECT (0x14c) instead? 2015-08-21T15:28:09 or did you already 2015-08-21T15:28:20 (noticed the commented out BOTH_EDGES) 2015-08-21T15:28:48 but beng-nl in gpio layer you can look at rtems_gpio_enable_interrupt(). In that threaded_handling=false (in our case) int. can occur with this condition :- if ( gpio_bank_state[bank].interrupt_counter == 0 ) 2015-08-21T15:31:18 so HIGH_LEVEL won't create any problem I assume. With printk I tested with infinite time too. It started printed for infinite time. And when I removed the signal it stopped :) 2015-08-21T15:32:06 it worked (printing) for as long as the signal was high? 2015-08-21T15:32:15 and then the bbb reset when you removed the signal? 2015-08-21T15:32:37 yep ! Nop not reset as only print() function was there 2015-08-21T15:33:03 As I told when adding another tasks like gpio_set() and all it resets 2015-08-21T15:33:22 aha 2015-08-21T15:33:56 so everything actually works fine if you only have a print() in the irq handling function? 2015-08-21T15:33:59 off on off on etc? 2015-08-21T15:34:17 Yes ! with printk() only 2015-08-21T15:34:26 ok, i didn't realize that 2015-08-21T15:34:38 and then it crashes if you also have a gpio_set_etc. in the handler function? 2015-08-21T15:34:47 immediately? or after the signal is removed? 2015-08-21T15:35:04 Immediately it is 2015-08-21T15:35:21 before executing the gpio_set() 2015-08-21T15:35:31 just before executing ... 2015-08-21T15:36:02 That I came to know When I put print() statement just before gpio_set() 2015-08-21T15:36:08 prints then reboot 2015-08-21T15:36:15 fascinating 2015-08-21T15:36:42 ;) 2015-08-21T15:37:10 beng-nl: If you will get some time you can have a test of that so you will get the exact picture and then we can proceed further :) 2015-08-21T15:37:45 i wish i had more of that :) 2015-08-21T15:37:48 but i should make some time for this yes 2015-08-21T15:38:10 ketul: which branch should i take that has a test program that illustrates this? 2015-08-21T15:38:11 Sure.. and please :) 2015-08-21T15:38:33 let me fetch a link 2015-08-21T15:40:19 beng-nl: https://github.com/ketul93/rtems/commit/b6384069913ae8983d0f85b3b8175329f12cbbe0 2015-08-21T15:41:25 Also fine if you find time as per your convenience. Kindly mail .. :) 2015-08-21T15:43:46 eh 2015-08-21T15:44:20 what do you want me to mail? i didn't follow 2015-08-21T15:56:40 *** kulp is now known as KULP 2015-08-21T15:58:19 *** andrei_ has joined #rtems 2015-08-21T16:14:37 gah. i'm starting to think the onboard buttons are not gpio's after all 2015-08-21T16:25:08 *** Davidbrcz has joined #rtems 2015-08-21T16:40:59 *** Davidbrcz has quit IRC 2015-08-21T17:13:17 *** Davidbrcz has joined #rtems 2015-08-21T17:15:27 ahh. Yep ! beng-nl good point. With mail I wanted to tell you to mail the approach if I am not on IRC :) 2015-08-21T17:19:00 aha 2015-08-21T17:19:34 google-summer-of-code-students-list flooded today :D 2015-08-21T17:19:38 i have a test program that reads a button state & sets a led, using RISING_EDGE 2015-08-21T17:19:55 ohh. Is it working fine ? 2015-08-21T17:20:00 the irq is constantly invoked, and everything hangs once i press the button 2015-08-21T17:21:23 is that the same? 2015-08-21T17:21:53 *** ketul has quit IRC 2015-08-21T17:22:25 *** ketul has joined #rtems 2015-08-21T17:23:32 did I miss anything ? 2015-08-21T17:26:06 did you leave? :) 2015-08-21T17:28:39 yep .. But ok.. so your bbb just got hang ? 2015-08-21T17:33:29 yep 2015-08-21T17:33:34 different program perhaps 2015-08-21T17:36:06 *** gedare has quit IRC 2015-08-21T17:36:35 beng-nl: ohk. Have tried with only printk() ? 2015-08-21T17:37:49 yep that's all i'm doing 2015-08-21T17:38:03 (y) 2015-08-21T17:38:08 this really looks exactly like the irq generator is set at low state instead of edge 2015-08-21T17:38:58 HMM 2015-08-21T17:39:01 mmio_set(bbb_reg(bank, AM335X_GPIO_IRQSTATUS_SET_0), BIT(pin)); 2015-08-21T17:39:17 that is done before the mode set 2015-08-21T17:39:24 then again.. the rest of the code is executed too 2015-08-21T17:39:30 so nevermind 2015-08-21T17:40:59 mmio_set(bbb_reg(bank, AM335X_GPIO_IRQSTATUS_SET_0), BIT(pin)); is for I think to enable interrupt on particular pin. Then "AM335X_GPIO_FALLINGDETECT" and all will decide which type of interrupt . 2015-08-21T17:41:06 Is it so ? or.. 2015-08-21T17:41:22 yep 2015-08-21T17:41:44 so shouldn't we configure the mode first, then turn the irq generating on? otherwise you might get unwanted irq's (for the mode that happens to be on now) 2015-08-21T17:43:43 ohkk. Good point. We should give a try for that .. 2015-08-21T17:43:54 i don't think it'll do much really 2015-08-21T17:43:59 but i'm adding a few prints there anyway 2015-08-21T17:44:42 hmm but for try it's ok. Because in our case all things are done regardless of order 2015-08-21T18:03:18 ketul: i'm out of time debugging this for now :( 2015-08-21T18:03:28 we'll chat by email perhaps ok? 2015-08-21T18:04:26 ok. sure. Me too busy for patch generation :) 2015-08-21T18:04:56 beng-nl: Hope we will get some solution asap 2015-08-21T18:18:14 [trac] GSoC/2015/BeagleBSPPortingUmon edited by jrcatbagan (Updated wiki for Beagle BSP Improvements - Porting uMon to the Beaglebone Black) http://devel.rtems.org/wiki/GSoC/2015/BeagleBSPPortingUmon 2015-08-21T18:19:08 *** andrei_ has quit IRC 2015-08-21T19:01:29 [trac] GSoC/2015/RaspberryPi2Support edited by krohini (empty comment) http://devel.rtems.org/wiki/GSoC/2015/RaspberryPi2Support 2015-08-21T19:04:28 [trac] GSoC/2015/RaspberryPi2Support edited by krohini (empty comment) http://devel.rtems.org/wiki/GSoC/2015/RaspberryPi2Support 2015-08-21T20:18:42 *** ketul has quit IRC 2015-08-21T20:22:14 *** Davidbrcz has quit IRC 2015-08-21T20:26:44 *** mumptai has joined #rtems 2015-08-21T20:45:57 *** dougk_ has quit IRC 2015-08-21T20:47:51 *** ketul has joined #rtems 2015-08-21T20:49:32 *** ketul has quit IRC 2015-08-21T20:51:12 *** Davidbrcz has joined #rtems 2015-08-21T20:55:54 *** Davidbrcz has quit IRC 2015-08-21T20:56:02 *** Davidbrcz has joined #rtems 2015-08-21T21:38:21 *** hesham has quit IRC 2015-08-21T22:55:08 *** mumptai has quit IRC 2015-08-22T02:43:21 *** Davidbrcz has left #rtems 2015-08-22T03:07:07 [trac] #2394/defect (v:4.10) created by kevinpat88 (Choosing the Best Weight loss Powder Almased) http://devel.rtems.org/ticket/2394 2015-08-22T03:09:32 [trac] #2394/infra (assigned) updated by joel.sherrill (empty comment) http://devel.rtems.org/ticket/2394 2015-08-22T03:21:23 [trac] #2394/infra deleted (Choosing the Best Weight loss Powder Almased) http://devel.rtems.org/ticket/2394 2015-08-22T08:42:07 *** ketul has joined #rtems 2015-08-22T09:01:57 *** hesham has joined #rtems 2015-08-22T09:19:11 *** zoso has joined #rtems 2015-08-22T09:39:28 *** zoso has quit IRC 2015-08-22T10:14:16 *** ketul has quit IRC 2015-08-22T10:53:32 *** ketul has joined #rtems 2015-08-22T11:42:27 *** sepi has joined #rtems 2015-08-22T14:11:16 *** guerby has quit IRC 2015-08-22T14:12:10 *** guerby has joined #rtems 2015-08-22T15:22:19 *** ketul has quit IRC 2015-08-22T16:30:05 [git] 91be006 by Jarielle Catbagan: [README: Fix directory name references and remove reference to nonexistent main/README] http://devel.rtems.org/changeset/91be006/umon 2015-08-22T16:30:06 [git] 63834b9 by Jarielle Catbagan: [BBB: Add proper attributions to the files modified] http://devel.rtems.org/changeset/63834b9/umon 2015-08-22T16:30:06 [git] c900821 by Jarielle Catbagan: [BBB: README: Add section on installing uMon into the eMMC from an SD card] http://devel.rtems.org/changeset/c900821/umon 2015-08-22T17:18:56 [trac] #2395/defect (v:4.10) created by davidravi9 (Buy online Supplements Calriphen diet) http://devel.rtems.org/ticket/2395 2015-08-22T17:26:44 [trac] #2395/infra (assigned) updated by joel.sherrill (empty comment) http://devel.rtems.org/ticket/2395 2015-08-22T17:30:56 [trac] #2395/infra deleted (Buy online Supplements Calriphen diet) http://devel.rtems.org/ticket/2395 2015-08-22T19:06:04 *** ita_ has joined #rtems 2015-08-22T19:06:33 *** hesham has quit IRC 2015-08-22T19:08:12 *** ita has quit IRC 2015-08-22T21:18:39 hi all 2015-08-22T22:21:53 *** crazy3k has joined #rtems 2015-08-23T07:12:07 *** zoso has joined #rtems 2015-08-23T08:40:47 *** hesham has joined #rtems 2015-08-23T12:19:00 *** ketul has joined #rtems 2015-08-23T15:32:02 *** arvind_k has joined #rtems 2015-08-23T15:35:04 *** zoso has quit IRC 2015-08-23T19:33:05 *** hesham has quit IRC 2015-08-23T19:47:50 *** jkridner has quit IRC 2015-08-23T19:48:52 *** jkridner has joined #rtems 2015-08-23T20:01:25 *** ketul has quit IRC 2015-08-23T22:24:37 *** arvind_k has quit IRC 2015-08-23T23:03:53 *** kiwichris has joined #rtems