2016-05-23T03:50:35 *** mkhoory has joined #rtems 2016-05-23T04:14:21 *** monstr has joined #rtems 2016-05-23T05:17:48 *** mumptai has joined #rtems 2016-05-23T05:49:47 *** sebhub has joined #rtems 2016-05-23T05:57:09 sebhub: hi 2016-05-23T06:31:03 sebhub: hi 2016-05-23T06:37:01 hi chris 2016-05-23T06:37:16 Sorted, you remove smpfatal03 :) 2016-05-23T06:37:32 The printk is a good idea but it has generated a lot of wanrings. 2016-05-23T06:37:46 The attribute that is to check warnings 2016-05-23T06:37:53 yes, no surprise 2016-05-23T06:38:08 I have covered most cases I hope. 2016-05-23T06:38:16 nice 2016-05-23T06:38:23 The testsuite has some spots I am addressing. 2016-05-23T06:38:47 The mixing of the tmacors and the *_begink/(_endk was fragile. 2016-05-23T06:40:07 I turned off the warnings check for the jffs2 2016-05-23T06:57:51 sebhub: this line https://git.rtems.org/rtems/tree/cpukit/libmisc/stackchk/check.c#n235 generates a warning on i386 cause PRIxPTR is 'lx' 2016-05-23T06:58:47 you need a (uintptr_t) cast 2016-05-23T08:10:38 *** mumptai has quit IRC 2016-05-23T10:17:12 *** mkhoory has quit IRC 2016-05-23T11:40:37 *** mkhoory has joined #rtems 2016-05-23T12:54:34 *** mkhoory has quit IRC 2016-05-23T13:29:08 *** mkhoory has joined #rtems 2016-05-23T13:40:45 *** mkhoory has quit IRC 2016-05-23T13:44:42 *** mkhoory has joined #rtems 2016-05-23T14:11:51 *** mkhoory has quit IRC 2016-05-23T14:24:57 *** monstr has quit IRC 2016-05-23T15:07:37 *** gedare has joined #rtems 2016-05-23T15:07:38 *** ChanServ sets mode: +o gedare 2016-05-23T15:51:48 *** mumptai has joined #rtems 2016-05-23T16:09:50 *** MiD has quit IRC 2016-05-23T17:31:52 *** guerby has quit IRC 2016-05-23T17:37:43 *** guerby has joined #rtems 2016-05-23T17:59:10 *** guillaum1 has quit IRC 2016-05-23T18:07:37 *** guillaum1 has joined #rtems 2016-05-23T18:26:44 *** guillaum1 has quit IRC 2016-05-23T20:31:57 *** mumptai has quit IRC 2016-05-23T20:32:52 *** dipupo has joined #rtems 2016-05-23T21:06:24 *** gedare has quit IRC 2016-05-23T21:27:54 *** guillaum1 has joined #rtems 2016-05-23T22:49:07 *** lcpfnvcy_ has joined #rtems 2016-05-23T22:52:34 *** shelf has joined #rtems 2016-05-23T22:53:14 *** cdesai_ has joined #rtems 2016-05-23T22:53:35 *** lcpfnvcy has quit IRC 2016-05-23T22:53:35 *** cdesai has quit IRC 2016-05-23T22:53:36 *** indy has quit IRC 2016-05-23T22:53:38 *** timothyh has quit IRC 2016-05-23T22:53:40 *** indy_ has joined #rtems 2016-05-24T03:15:40 *** dougk_ has quit IRC 2016-05-24T03:28:16 *** dougk has joined #rtems 2016-05-24T04:19:51 *** monstr has joined #rtems 2016-05-24T06:17:14 *** indy_ is now known as indy 2016-05-24T07:15:45 *** mkhoory has joined #rtems 2016-05-24T07:27:10 hi kiwichris_, does RSB work with 4.10? I'm tyring to use it on Linux, downloading one of the patches for binutils fails 2016-05-24T07:27:19 incorrect URL it sems 2016-05-24T10:17:47 *** cdesai_ is now known as cdesai 2016-05-24T11:11:28 *** mkhoory has quit IRC 2016-05-24T12:06:40 [git] 7af4d67 by Sebastian Huber: [mpci: Add missing return statements] http://devel.rtems.org/changeset/7af4d67/rtems 2016-05-24T12:06:40 [git] 16dbcdc by Sebastian Huber: [score: _CORE_message_queue_Seize() Delete unused parameter.] http://devel.rtems.org/changeset/16dbcdc/rtems 2016-05-24T13:30:17 [trac] #2718/defect (v:4.10) created by sebastian.huber (Blocking _CORE_message_queue_Submit() may lead to unpredictable results) http://devel.rtems.org/ticket/2718 2016-05-24T13:37:51 [trac] #2718/defect (closed) updated by Sebastian Huber (In [changeset:"4b623d655bb4f4853a6ce385ae17e505dddbe7ce/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2718 2016-05-24T13:37:52 [git] 7088340 by Sebastian Huber: [psxtests/psxmsgq01: Enable test case] http://devel.rtems.org/changeset/7088340/rtems 2016-05-24T13:37:52 [git] 4b623d6 by Sebastian Huber: [score: Fix blocking _CORE_message_queue_Submit() Close #2718.] http://devel.rtems.org/changeset/4b623d6/rtems 2016-05-24T13:39:34 [git] b1860df by Sebastian Huber: [psxtests/psxmsgq01: Fix typo] http://devel.rtems.org/changeset/b1860df/rtems 2016-05-24T13:46:00 *** benjamint has quit IRC 2016-05-24T13:50:21 *** benjamint has joined #rtems 2016-05-24T14:42:38 *** guillaum1 has quit IRC 2016-05-24T14:42:39 *** shelf has quit IRC 2016-05-24T14:45:59 *** dougk has quit IRC 2016-05-24T14:46:20 *** dougk has joined #rtems 2016-05-24T15:40:16 *** shelf has joined #rtems 2016-05-24T15:45:52 *** guillaum1 has joined #rtems 2016-05-24T15:57:33 *** guillaum1 has quit IRC 2016-05-24T16:06:15 *** guillaum1 has joined #rtems 2016-05-24T16:28:04 *** gedare_ has joined #rtems 2016-05-24T16:37:30 *** gedare_ has quit IRC 2016-05-24T16:50:23 *** gedare has joined #rtems 2016-05-24T16:50:23 *** ChanServ sets mode: +o gedare 2016-05-24T17:06:10 *** jkridner has joined #rtems 2016-05-24T17:28:46 *** gedare has quit IRC 2016-05-24T18:16:41 *** guillaum1 has quit IRC 2016-05-24T18:17:38 *** guillaum1 has joined #rtems 2016-05-24T18:27:43 *** monstr has quit IRC 2016-05-24T19:22:31 *** gigetoo has quit IRC 2016-05-24T22:27:52 *** gigetoo has joined #rtems 2016-05-25T01:02:11 *** intern has joined #rtems 2016-05-25T01:04:26 Good afternoon. I have been tasked with creating a new board support package for RTEMS and I have no experience with the software. I've been reading about RTOS's and looking through RTEMS documentation, but I'm a bit lost. Where do I begin creating a BSP? 2016-05-25T01:16:04 I am currently reading through the source code in the git repository. 2016-05-25T02:08:05 *** intern has quit IRC 2016-05-25T02:08:18 *** intern has joined #rtems 2016-05-25T02:35:05 *** jkridner has quit IRC 2016-05-25T02:37:19 *** Wonjun has joined #rtems 2016-05-25T03:30:26 *** intern has quit IRC 2016-05-25T03:30:38 *** intern has joined #rtems 2016-05-25T03:54:08 *** intern has quit IRC 2016-05-25T04:11:08 *** jkridner has joined #rtems 2016-05-25T04:25:10 *** monstr has joined #rtems 2016-05-25T05:06:14 *** intern2 has joined #rtems 2016-05-25T05:13:43 *** intern2 has quit IRC 2016-05-25T06:32:32 [trac] #2719/defect (v:4.10) created by sebastian.huber (pthread_spin_unlock() returns wrong error status in case the lock is unlocked) http://devel.rtems.org/ticket/2719 2016-05-25T06:54:53 sebhub: hi 2016-05-25T06:55:07 Has OBJECTS_LOCAL gone? 2016-05-25T06:55:30 I am seeing .. ../../freebsd/sys/kern/sys_generic.c:1927:8: error: 'OBJECTS_LOCAL' undeclared (first use in this function) 2016-05-25T06:56:51 [git] 24d0ee5 by Chris Johns: [cpukit, testsuite: Add rtems_printf and rtems_printer support. This change adds rtems_printf and related functions and wraps the RTEMS print plugin support into a user API. All references to the plugin are removed and replaced with the rtems_printer interface. Printk and related functions are made to return a valid number of 2016-05-25T06:56:51 characters formatted and outp...] http://devel.rtems.org/changeset/24d0ee5/rtems 2016-05-25T06:56:51 [git] 1503c1c by Chris Johns: [i386/pc386: Fix printk formatting warnings.] http://devel.rtems.org/changeset/1503c1c/rtems 2016-05-25T06:56:51 [git] c16e54a by Chris Johns: [libfs/jff2: Avoid printk format warnings. This change does not fix the printk format warnings rather it ignores them.] http://devel.rtems.org/changeset/c16e54a/rtems 2016-05-25T06:56:51 [git] 8f9d07b by Chris Johns: [bsp/shared: Fix printk warnings.] http://devel.rtems.org/changeset/8f9d07b/rtems 2016-05-25T06:56:52 [git] 75e3e0e by Chris Johns: [libcsupport: Fix printk warnings.] http://devel.rtems.org/changeset/75e3e0e/rtems 2016-05-25T06:56:52 [git] 458e4c4 by Chris Johns: [score/i386: Fix printk format warnings.] http://devel.rtems.org/changeset/458e4c4/rtems 2016-05-25T06:56:53 [git] 62cba15 by Chris Johns: [libmisc: Fix printk format warnings.] http://devel.rtems.org/changeset/62cba15/rtems 2016-05-25T06:56:53 [git] 61ea345 by Chris Johns: [testsuite: Fix printk formating warning.] http://devel.rtems.org/changeset/61ea345/rtems 2016-05-25T06:56:54 [git] addf1aa by Chris Johns: [testsuite: Fix printk format warnings.] http://devel.rtems.org/changeset/addf1aa/rtems 2016-05-25T07:04:27 yes, I have to update libbsd 2016-05-25T07:06:49 OK. I am using it all the time at the moment. 2016-05-25T07:06:52 i will fix this in the next couple of hours 2016-05-25T07:07:30 ok, I will try to fix it along with the rtems changes next time 2016-05-25T08:20:11 *** punitvara has joined #rtems 2016-05-25T08:51:40 *** punitvara has quit IRC 2016-05-25T10:45:44 [trac] #2719/defect (closed) updated by Sebastian Huber (In [changeset:"9ec7d492b951efec32520bff7106e21e16e9c74d/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2719 2016-05-25T10:45:44 [git] 25e28d8 by Sebastian Huber: [confdefs.h: Fix heap alloc size estimate Account for the heap minimum block size.] http://devel.rtems.org/changeset/25e28d8/rtems 2016-05-25T10:45:44 [git] 0e9d5b6 by Sebastian Huber: [mpci: Fix thread queue flush method We must call the MP callout for proxies if we unblock them after a thread queue extraction. This was missing in _Thread_queue_Flush_critical(). Move thread remove timer and unblock code to new function _Thread_Remove_timer_and_unblock().] http://devel.rtems.org/changeset/0e9d5b6/rtems 2016-05-25T10:45:44 [git] 8866e62 by Sebastian Huber: [score: Move thread queue object support] http://devel.rtems.org/changeset/8866e62/rtems 2016-05-25T10:45:44 [git] c3d8d9e by Sebastian Huber: [score: Get rid of mp_id parameter Get rid of the mp_id parameter used for some thread queue methods. Use THREAD_QUEUE_QUEUE_TO_OBJECT() instead.] http://devel.rtems.org/changeset/c3d8d9e/rtems 2016-05-25T10:45:44 [git] 631b3c8 by Sebastian Huber: [score: Move thread queue MP callout to context Drop the multiprocessing (MP) dependent callout parameter from the thread queue extract, dequeue, flush and unblock methods. Merge this parameter with the lock context into new structure Thread_queue_Context. This helps to gets rid of the conditionally compiled method call 2016-05-25T10:45:45 helpers.] http://devel.rtems.org/changeset/631b3c8/rtems 2016-05-25T10:45:45 [git] 9ec7d49 by Sebastian Huber: [posix: Fix pthread_spin_unlock() error status Close #2719.] http://devel.rtems.org/changeset/9ec7d49/rtems 2016-05-25T10:48:17 [trac] #2719/defect (closed) updated by Sebastian Huber (In [changeset:"3f72dda6ee518d3ea04341ad4df079ecb1895ef7/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2719 2016-05-25T10:48:17 [git] 3f72dda by Sebastian Huber: [posix: Fix pthread_spin_unlock() error status Update #2719.] http://devel.rtems.org/changeset/3f72dda/rtems 2016-05-25T10:49:19 [trac] #2719/defect (closed) updated by sebastian.huber (empty comment) http://devel.rtems.org/ticket/2719 2016-05-25T11:12:38 [git] 9546d90 by Sebastian Huber: [rtems-bsd-mutex: Update due to API changes] http://devel.rtems.org/changeset/9546d90/rtems-libbsd 2016-05-25T11:12:38 [git] 92e9493 by Sebastian Huber: [testsuite: Update due to API changes] http://devel.rtems.org/changeset/92e9493/rtems-libbsd 2016-05-25T11:12:38 [git] 43a14eb by Sebastian Huber: [Update due to API changes] http://devel.rtems.org/changeset/43a14eb/rtems-libbsd 2016-05-25T11:59:30 [trac] #2720/defect (v:4.10) created by sebastian.huber (sem_post() with overflow returns an unusual error status) http://devel.rtems.org/ticket/2720 2016-05-25T12:11:08 [trac] #2721/defect (v:4.10) created by sebastian.huber (sem_init() does not honour SEM_VALUE_MAX) http://devel.rtems.org/ticket/2721 2016-05-25T12:13:58 [trac] #2722/defect (v:4.10) created by sebastian.huber (SEM_VALUE_MAX is unusually small on RTEMS) http://devel.rtems.org/ticket/2722 2016-05-25T13:01:44 *** gedare has joined #rtems 2016-05-25T13:01:44 *** ChanServ sets mode: +o gedare 2016-05-25T13:24:15 *** benjamint has quit IRC 2016-05-25T13:26:36 *** benjamint has joined #rtems 2016-05-25T13:32:40 *** igutek has joined #rtems 2016-05-25T13:36:22 [trac] GSoC/2016 edited by deval (empty comment) http://devel.rtems.org/wiki/GSoC/2016 2016-05-25T13:38:53 *** sacha has joined #rtems 2016-05-25T13:40:40 *** _sambeet has joined #rtems 2016-05-25T13:41:29 *** deval has joined #rtems 2016-05-25T13:44:40 <_sambeet> Hi..... the weekly irc meeting is supposed to start now? 2016-05-25T13:47:37 *** mudit1729 has joined #rtems 2016-05-25T13:50:22 *** Wonjun has quit IRC 2016-05-25T13:50:34 *** punitvara has joined #rtems 2016-05-25T13:50:39 *** Wonjun has joined #rtems 2016-05-25T13:54:57 [trac] GSoC/2016 edited by deval (empty comment) http://devel.rtems.org/wiki/GSoC/2016 2016-05-25T13:55:43 _sambeet, yes, in about 4 minutes 2016-05-25T13:55:59 *** ppisa has joined #rtems 2016-05-25T13:56:48 yes _sambeet we'll start soon 2016-05-25T13:57:54 *** sebhub has quit IRC 2016-05-25T14:00:15 Welcome to our first weekly meeting. I hope you have all had a pleasant start to the summer and are diving into coding! 2016-05-25T14:00:23 These meetings are intended to help make sure everyone is making good progress on their projects, and to see if any particular assistance is needed. 2016-05-25T14:00:29 I'll call on each of you in turn. Since I don't know your IRC names yet, if I miss you please speak up before the meeting ends. If you need to leave early, let me know so I can call on you sooner. 2016-05-25T14:00:47 So today, let's start with a roll call. What students are here? 2016-05-25T14:00:59 Darshit Shah 2016-05-25T14:01:04 Punit Vara 2016-05-25T14:01:05 Mudit Jain 2016-05-25T14:01:06 Deval Shah 2016-05-25T14:01:08 Sai Charan Sane 2016-05-25T14:01:13 Wonjun Hwang 2016-05-25T14:01:16 <_sambeet> Sambeet Panigrahi 2016-05-25T14:02:35 Thanks. We'll follow the same format for everyone. First, briefly introduce your project to everyone, state what you've done so far, and what your plan is for the upcoming week. Then, if you have any questions, problems, or concerns you can mention them. 2016-05-25T14:02:42 We'll start with darnir, please go ahead. 2016-05-25T14:03:55 Late introduction: Isaac Gutekunst, co-mentor for Improving Trace 2016-05-25T14:04:05 Hi. The concept of the project is to implement a new scheduling policy for RTEMS that supports Arbitrary Processor Affinities in SMP configurations. While the existing implementation can do the same, it is pretty inefficient. 2016-05-25T14:04:16 Habeeb Olufowobi 2016-05-25T14:04:19 And before I forget, you should each add an update to https://devel.rtems.org/wiki/GSoC/2016 today. 2016-05-25T14:04:35 ^ https://devel.rtems.org/wiki/GSoC/2016#StudentStatusUpdates 2016-05-25T14:04:38 So, I will be implementing a new scheduler called StringAPA based on the paper linked to in my proposal. 2016-05-25T14:05:51 Till now, I've been looking into the Schedulign Simulator and trying to get it to compile. However that has been a large task with a few breakages that I haven't been able to fix yet. Based on my last conversatio with sebhub, it is okay to push this task to a little later since we don't need schedsim till the end of the project anyways 2016-05-25T14:06:24 I've been looking into the existing scheduler implementations and working on a draft for how I will implement the new scheduler as well. I have most of the specifics charted out now. 2016-05-25T14:06:50 ok darnir you also have noted that sebhub pushed a dummy implementation of the strongapa scheduler? 2016-05-25T14:07:02 gedare: Yes, I have seen that. 2016-05-25T14:07:27 ok great. one thing I'd like you to do is set up your blog and write an introductory post, basically you can convert your proposal into a blog post 2016-05-25T14:07:29 Over the next week, I will start implementing a basic version of the iterative MVM algorithm in the paper. 2016-05-25T14:07:34 *** sacha has quit IRC 2016-05-25T14:07:43 gedare: Sure, I'll do that ASAP. 2016-05-25T14:07:47 this is most useful since the proposals are private, and contain some confidential/private information 2016-05-25T14:08:05 ok thanks darnir, any questions? 2016-05-25T14:08:11 punitvara: you'll be next 2016-05-25T14:08:29 I just had one question, I was unable to find a high level description of the scheduler API. That is what are the required entry points and the expectations of those methods 2016-05-25T14:08:31 (I will usually indicate the next person as the previous is wrapping up.) 2016-05-25T14:08:37 I'll be missing for 10 minutes, my appologies 2016-05-25T14:09:10 I've seen the dummy implementation, but am not sure about the exact API that I should follow 2016-05-25T14:09:29 Hmm, darnir if that information is not in the doxygen for the cpukit, then there may be some (outdated) information I can send you 2016-05-25T14:09:43 i'll take a quick look. 2016-05-25T14:09:49 I was unable to find it, will take a look again as well. 2016-05-25T14:10:13 It should be in the new documentation that Chris has been working on right? 2016-05-25T14:10:16 https://docs.rtems.org/doxygen/cpukit/html/structScheduler__Operations.html 2016-05-25T14:10:46 well, this is internal kernel documentation, which isn't maintained the same as the user-facing docs 2016-05-25T14:11:08 that links to the function table used by each scheduler 2016-05-25T14:11:18 Aah, I've been searching in the wrong place then. Thanks! 2016-05-25T14:11:32 also, https://docs.rtems.org/doxygen/cpukit/html/group__ScoreScheduler.html 2016-05-25T14:11:37 that is the higher-level view. 2016-05-25T14:11:47 ok. punitvara are you ready? 2016-05-25T14:11:55 yes 2016-05-25T14:11:56 go ahead 2016-05-25T14:11:57 My project goal is to improve Beagle bone Black Board Support Package. First I am focusing to support Pulse Width Modulation peripherals to BBB. There is no code in RTEMS for PWM right now. 2016-05-25T14:12:18 thanks darnir send any other questions to the ml or mentors 2016-05-25T14:12:33 Thanks gedare 2016-05-25T14:13:25 (The meeting will go quicker if you each prepare your status updates ahead of time.) 2016-05-25T14:13:33 TI Startware open source package has PWM bare-bone driver. Till now I am successful to add that code to RTEMS 2016-05-25T14:13:44 But that patch is of 11000 lines 2016-05-25T14:14:18 I sent to mailing list as well but it required moderator approval but I didn't get till now 2016-05-25T14:14:44 I have some concerns about bringing in that code, partly because I'm not convinced the licensing issue is clear 2016-05-25T14:15:35 also, why do we need 11k LoC to operate a PWM? Perhaps there should be a simpler approach to take. 2016-05-25T14:15:55 Yes there 2016-05-25T14:16:06 I tested that code with around 500 lines 2016-05-25T14:16:15 For the first overview od a such large contribution, you can send link to the commit in personal GitHub repo. 2016-05-25T14:16:37 I sent on mailing list but I will send it again for sure 2016-05-25T14:16:47 +1 on github code reviews. put your commits on the top of a branch and send the link to the branch 2016-05-25T14:17:07 Martin told me to test PWM with TI SW code only 2016-05-25T14:17:17 ok 2016-05-25T14:17:18 So that's why I am focusing on it 2016-05-25T14:17:50 Sounds good, anything else? 2016-05-25T14:18:05 But recently both are working on different way also. There is some issue with TI SW code 2016-05-25T14:18:06 mudit1729 will be next 2016-05-25T14:18:20 so this week I will be working to resolve that issue with mentors 2016-05-25T14:18:28 hmm, ok. that is with the "open" version 2016-05-25T14:18:42 As for the Ti or other company HAL, there is usually significant code redundancy with common RTEMS code and have it in multiple copies can be nightmare, but other solution can be hard as well. 2016-05-25T14:19:15 Yes that is too open source. I have edited that code according to RTEMS 2016-05-25T14:19:54 and tested successfully as well. But again I am not able to test generating frequency accurately 2016-05-25T14:20:37 the best for that is to use another microcontroller or some other waveform generator 2016-05-25T14:20:52 there are some solutions that can work through your PC/desktop/laptop too 2016-05-25T14:20:55 You can usually use some capture unit to record exact time time of edges arival. 2016-05-25T14:20:56 Yes some one suggested another BBB as logic analyzer 2016-05-25T14:21:16 so I will purchase another BBB next week 2016-05-25T14:21:19 The most generic is Zynq, there you can get all what you need. 2016-05-25T14:21:21 yeah, if you don't have access to any oscilloscope/LA 2016-05-25T14:22:06 I can get access to oscilloscope this week only for two days. Because this is final week of my masters 2016-05-25T14:22:08 ok, creating that logic analyzer as an RTEMS app eventually would make for a nice side contribution of your project, too. 2016-05-25T14:22:41 but you can probably find some open source linux software to do it, for getting started. 2016-05-25T14:22:44 sounds good. anything else? 2016-05-25T14:22:48 mudit1729: you're up. 2016-05-25T14:22:58 Hi everyone. 2016-05-25T14:23:05 I would love to contribute 2016-05-25T14:23:14 If you want a cheap logic analizer, I've had a pretty good experience with this one: http://www.seeedstudio.com/depot/preorder-open-workbench-logic-sniffer-p-612.html?cPath=75 2016-05-25T14:23:14 My project is to provide Low Level Peripherals and SD Card support for RPi BSP. 2016-05-25T14:23:20 I will try my best for that too 2016-05-25T14:23:24 Thanks 2016-05-25T14:23:38 My first deliverable is to provide SD card support for the RPi BSP. The code has already been ported to rtems-libbsd, however it has to be tailored to RPi by adding the quirks and logics unique to the SD host controller on Rpi 2016-05-25T14:24:05 Present Status : I have cherry picked two commits from Andre's branch and I have built rtems-libbsd using waf. I had initial errors of redefinition however they were resolved.nexus-devices.h was being called twice. In addition to that I was working with Pavel to test his series of changes related to RPI2. I have downloaded and succesfully built rki.bin from his branch and was testing the same on RPi2 2016-05-25T14:24:33 I will personally ping you @gedare for logic analyzer app for more suggestions 2016-05-25T14:24:51 punitvara: check igutek or use google. i have no great insight. 2016-05-25T14:25:13 Next Week Schedule : I would be going through the present implementation of the driver in FreeBSD using the FreeBSD documentation to learn/understand what are the different APIs that are used in the driver, how it interacts with SD host controllers, and the general driver interface that is provided to the user. I will be updating the blog with a write up with my understanding of the present implementation. Then I would be looking a 2016-05-25T14:25:58 very good mudit1729. sounds like you're off to a nice start. 2016-05-25T14:26:57 Once that is done, I would be testing various functions and debug teh ones that are causing kernel panic 2016-05-25T14:27:19 ok. sounds good, any questions? deval you'll be next 2016-05-25T14:27:23 Thanks gedare 2016-05-25T14:27:33 ya I had two qquestions 2016-05-25T14:27:38 Hello, Main aim of my project is to add USB and Networking support to Raspberry PI. Since Ethernet controller of Raspberry PI is a USB one my first task would be to add USB root hub support. For that I need USB DWC OTG drivers and hardware specific drivers (for bcm283x) in the right place. USB DWC OTG drivers are already there in the codebase. To continue with Yurii's last year's work, I cherry picked his commits for USB roothub dri 2016-05-25T14:27:43 go ahead mudit1729 2016-05-25T14:27:54 1. I have built the rki.bin, however rtems is not booting on Rpi2 2016-05-25T14:28:10 With the same firmware raspbian boots up 2016-05-25T14:29:10 Have you tried my branch? If it does not boot, can you switch to mainline and test that? 2016-05-25T14:29:38 mudit1729: as far as i know, rtems on rpi2 has not been very stable, hard to debug and test. 2016-05-25T14:29:40 okay, Yes I have tried your branch, I will be doing that and give you the update. 2016-05-25T14:29:47 If you have problems then I have hacks, suggestion how to print early from RTEMS to check place where it stucks. 2016-05-25T14:30:08 ok deval you should try to reach out to yurii, and ragunath, about working with those drivers--what doc helps, etc 2016-05-25T14:31:01 There is a second question. Are there any implemented drivers available that can be taken as reference ? 2016-05-25T14:31:08 okay I will do that 2016-05-25T14:31:10 Is it possible to use JTAG/SWD on the Pi? 2016-05-25T14:31:27 sure deval I just sent their emails in case you did not have them. 2016-05-25T14:31:31 I use next hack to get early print output 2016-05-25T14:31:32 If so, there are very inexpensive JTAG probes that might help with the printing early issue 2016-05-25T14:31:33 +void ll_chrout(char ch)+{+ bcm2835_usart_fns.deviceWritePolled(0, ch);+}++void ll_strout(char *str)+{+ char ch;+ while ((ch = *(str++)) != 0) {+ if (ch == '\n')+ bcm2835_usart_fns.deviceWritePolled(0, '\r');+ bcm2835_usart_fns.deviceWritePolled(0, ch);+ }+}+ 2016-05-25T14:31:47 @ppisa I like that! 2016-05-25T14:31:47 igutek: Error: "ppisa" is not a valid command. 2016-05-25T14:31:55 mudit1729: SD card drivers? I think sebastian did one a long time ago 2016-05-25T14:32:01 Yes I received that. Thank you 2016-05-25T14:32:39 Default testsuits which are related to networking are having compilation/linking errors. Later I realised that the driver (bcm283x) itself is not compling. So I will now read documentation regarding adding drivers in rtems-libbsd. 2016-05-25T14:32:45 That should help me to verify the code or if required write that from scratch. I should be able to add the hardware specific driver before the next status update. 2016-05-25T14:33:02 So after understanding the documentation I will decide wether I should go ahead with Yurii's code or add drivers on my own. (Depends upon number of changes I have to perform on Yurii's code to get it working.) 2016-05-25T14:33:03 okay, Ill contact him over mail. I have a RPI2, if RTEMS is not stable on RPI2, what would be the method for testing it, QEMU ? 2016-05-25T14:33:10 http://cmp.felk.cvut.cz/~pisa/rtems/rpi-u-boot/rtems-4.12-rpi-u-boot-debug-160501.patch 2016-05-25T14:33:17 ok deval. I know yurii got stuck on some compiler problem for a long time. if you don't make progress, ask for more help; 2016-05-25T14:33:26 there are full set of hak to find where RTEMS stucks 2016-05-25T14:33:41 mudit1729: the stability problem is that not many people are testing it, and I recall it was hard to debug 2016-05-25T14:33:55 mudit1729: i don't know if there exists a qemu for it, you should look. 2016-05-25T14:34:09 Sure. I will keep everyone posted. 2016-05-25T14:34:21 ok thanks deval, anything else? 2016-05-25T14:34:27 one more thing 2016-05-25T14:34:31 I am a bit confused about how drivers interact with HID and hardware here. It would be great if I get some links to understand the concept. 2016-05-25T14:34:45 Wonjun: you'll be next 2016-05-25T14:35:17 deval: I personally don't know. Maybe others who have worked with it will be able to point you in the right direction. 2016-05-25T14:35:39 As for my RPi2 QEMU, send me email through list and I try to summary my knowledge or put it in the wiki. 2016-05-25T14:36:03 Okay 2016-05-25T14:36:11 thanks 2016-05-25T14:36:19 ppisa : Ill do that. Is a qemu available for RPi ? 2016-05-25T14:36:48 RPi1 with RSB newlib on 4.12 locks in strlen. NewwLib git solves that. 2016-05-25T14:37:22 QEMU 2.6 has RPi2 support able to run Windows. 2016-05-25T14:37:26 Wonjun: go ahead, are you here? 2016-05-25T14:37:32 yesh 2016-05-25T14:37:40 Hi. The concept of the project is to port RTEMS for Jailhouse hypervisor 2016-05-25T14:37:41 Initial version of Jailhouse developed x86 BSP for Linux and QEMU. 2016-05-25T14:37:41 So We will make RTEMS be inmated by jailhouse. 2016-05-25T14:37:57 Till now I am trying to execute initial version of Jailhouse with instructions in Git. I am working on VMware To execute jailhouse with Ubuntu 14.04.4-amd64 and QEMU. 2016-05-25T14:38:01 Okay, Ill keep that into consideration while setting up qemu, will put on the group if I have any erros. 2016-05-25T14:38:12 errors/issues. 2016-05-25T14:38:13 On the other hand, when Jailhouse is enabled by command, I got some error. I am checking with Jailhouse google group. 2016-05-25T14:38:20 Thanks for your answers 2016-05-25T14:38:24 @mudit this qemu works with RPi https://github.com/Torlus/qemu/tree/rpi 2016-05-25T14:38:24 deval: Error: "mudit" is not a valid command. 2016-05-25T14:38:50 deval : Thanks. 2016-05-25T14:39:40 This month I was busy to write a paper, I will use more time. 2016-05-25T14:39:41 ok Wonjun, this is during following the steps in "Demonstration in QEMU/KVM"? 2016-05-25T14:39:50 yes 2016-05-25T14:40:59 alright, if you don't get a good answer for your problem, bring it up again soon by email. 2016-05-25T14:41:12 sometimes the problem can be in the hw/host OS configuration 2016-05-25T14:41:35 especially when dealing with multiple levels of virtualization here: vmware, qemu/kvm, jailhouse 2016-05-25T14:42:00 yes. that is risk. 2016-05-25T14:42:13 if at all possible, you might want to try a native Linux host 2016-05-25T14:42:26 ok thanks Wonjun, anything else? 2016-05-25T14:43:00 Ok. If I install successfully, I will be updating the blog. thanks 2016-05-25T14:43:19 great. up next is _sambeet 2016-05-25T14:43:33 <_sambeet> This project aims to improve the existing port of Rock on RTEMS.The existing port of Rock is 4 years old and there has been a lot of updating in both software's.I would try to automate the entire build process using build systems autoproj and rtems source builder.I will also add relevant rock pages to rtems. 2016-05-25T14:44:45 <_sambeet> I have started my work by taking up the old scripts of the port as the baseline. Presently I am updating each script with the updated version of software. I have already added rsb support for rtems in the script.Before that I took some time to build all individual rock dependencies and set up an environment for each. 2016-05-25T14:45:22 <_sambeet> Presently I am going through the rtems port of OmniORB which is still shipped with the Tar file but has not been maintained. I have been in touch with OmniORB project maintainers and they have agreed to help me with the update. 2016-05-25T14:46:21 _sambeet: it would be worth it to understand if you can "re-write" the various scripts within the RSB to make it control building rock and the other projects. 2016-05-25T14:46:35 by understand, i mean to scope out what it needs, and whether it is feasible 2016-05-25T14:47:04 good news regarding omniorb 2016-05-25T14:47:24 are you able to test that part in isolation? 2016-05-25T14:48:33 <_sambeet> Yes I have tested in isolation.But yet to try the old port. 2016-05-25T14:48:45 <_sambeet> I will update it soon. 2016-05-25T14:48:52 ok 2016-05-25T14:49:01 thanks _sambeet any questions now? 2016-05-25T14:49:06 dipupo: you'll be next 2016-05-25T14:49:21 <_sambeet> There's one about tab? 2016-05-25T14:49:31 <_sambeet> Rsb 2016-05-25T14:50:28 What do you mean? 2016-05-25T14:50:42 oh, go ahead 2016-05-25T14:50:55 tab=rsb, got it. 2016-05-25T14:52:01 go ahead dipupo you can start while _sambeet asks his question. 2016-05-25T14:52:26 <_sambeet> Suppose in a given prefix I used rsb to build the toolset.And if I give the command again rsb doesn't detect that toolset is already built in the prefix and rebuilds it totally again. 2016-05-25T14:52:43 <_sambeet> Can there be a fix to it? 2016-05-25T14:52:43 Hi 2016-05-25T14:52:49 The objective of my project is to port RTEMS to TI TM4C129. I will also be adding lwip for networking if time permits. 2016-05-25T14:53:12 _sambeet: RSB will always over-write whatever is in the given prefix. 2016-05-25T14:53:25 Presently, I have build the BSP for the board family but I have not been able to test it on qemu. 2016-05-25T14:53:25 I have been having several issues with the version of qemu used to build the BSP and the newer version. 2016-05-25T14:53:29 _sambeet: this is purposeful, by design. it will also only do the install if there are no errors in the entire build. 2016-05-25T14:54:18 _sambeet: there are some commands you can add to shorten the re-build time, such as not to clean, not to re-download. but generally, the design of RSB is to fully re-build from clean every time, so you know you have nothing left over from prior builds. 2016-05-25T14:55:07 As discussed yesterday with Gedare, I tried building older qemu version closest to what was they used but the build failed with or without the patches that was provided. 2016-05-25T14:55:10 ok dipupo I saw your email. the best long-term solution would be to try to map the patches from the old version on to the current version 2016-05-25T14:55:32 well, especially if you can't build the older version. 2016-05-25T14:56:53 <_sambeet> Ok.i will go through commands.the main problem is I want to save some time rebuilding the entire thing again for a small change.is there any other way possible? 2016-05-25T14:58:14 _sambeet: read through the cmd line args you can pass to see, some of them will save time to avoid re-downloading files. i don't think there is an option to just "re-compile" though. ask Chris Johns if he has any good suggestions. I know this can be time-consuming when developing 2016-05-25T14:58:51 dipupo: you may want to reply to the email by sebhub to ask him for assistance to find the version of qemu referenced in the README or by the patches, to help solidfy your starting point 2016-05-25T14:59:32 Okay 2016-05-25T14:59:35 <_sambeet> Thanks a lot.I will forward my doubts to Chris Johns. 2016-05-25T15:00:31 dipupo: meanwhile, you should try to match what the patches changed in the old version, with where the code exists in the version currently being built by RSB 2016-05-25T15:01:04 if you can't find the matching code, ask for help. I suggest you create some kind of table to map each patch/patch hunk from the old to the new. 2016-05-25T15:02:27 When I applied the patches to one of the old version, it was requesting for git submodule (stated in the config file) but could not see any in the cloned version 2016-05-25T15:03:46 ok, you may need to populate the git submodules before you can apply the patch 2016-05-25T15:04:20 https://git-scm.com/book/en/v2/Git-Tools-Submodules 2016-05-25T15:04:35 Ok thanks 2016-05-25T15:04:43 anything else dipupo? 2016-05-25T15:05:19 That's all for now 2016-05-25T15:05:20 ok, looks like sacha and vivekk are missing, so i'll send them email. thanks for coming everyone see you next week! 2016-05-25T15:05:33 please remeber to update your tracking status 2016-05-25T15:05:35 Thank you 2016-05-25T15:05:50 https://devel.rtems.org/wiki/GSoC/2016#StudentStatusUpdates 2016-05-25T15:05:57 thank you 2016-05-25T15:06:22 gedare: Could you please discuss TI SW licence issue ? Christ agreed last time on mailing list 2016-05-25T15:06:34 http://processors.wiki.ti.com/index.php/StarterWare 2016-05-25T15:06:58 I will be using code from here. Official TI wiki 2016-05-25T15:10:17 *** ppisa has left #rtems 2016-05-25T15:10:28 *** _sambeet has left #rtems 2016-05-25T15:11:21 punitvara: my problem is that it is a wiki, and not a web site, and the link was entered by the same person who posted the open starterware 2016-05-25T15:12:03 so, i have not got a good sense if the open starterware has a clean foundation, or if someone just took the am335x starterware distributed by TI, posted it to sourceforge, and put up a link there. 2016-05-25T15:12:52 So what do you suggest me to do in this case ? 2016-05-25T15:13:05 the issue here is that to get the source code from TI officially seems to require filling some form. it may be worth it for you to ask about the legality of this open starterware on the E2E forums, where a TI rep might see and respond 2016-05-25T15:13:32 Thanks I will do that. 2016-05-25T15:13:34 https://github.com/VegetableAvenger/BBBIOlib/blob/master/BBBio_lib/BBBiolib_PWMSS.c 2016-05-25T15:14:08 This is the another repo I referred PWM calculation 2016-05-25T15:14:15 and code as well. 2016-05-25T15:15:34 Some one has any other reference for PWM code for BBB like for freeBSD ported code. Most welcome to suggest 2016-05-25T15:41:59 that lib looks reasonable. you should glance at freebsd, and ask others who have worked with the BB where they might have found resources. 2016-05-25T16:14:15 Sure I will. 2016-05-25T16:33:36 *** punitvara has quit IRC 2016-05-25T16:45:51 *** ketul has joined #rtems 2016-05-25T16:54:14 *** ketul has quit IRC 2016-05-25T17:55:03 bdfgtyhtgfdasyfrtg6y'hynyu'] 2016-05-25T17:55:08 +1 2016-05-25T17:55:15 oops 2016-05-25T17:55:20 cleaning me keyboard :) 2016-05-25T18:28:01 *** deval has quit IRC 2016-05-25T18:55:07 *** monstr has quit IRC 2016-05-25T19:09:04 *** gedare has quit IRC 2016-05-25T19:29:21 *** gedare has joined #rtems 2016-05-25T19:29:21 *** ChanServ sets mode: +o gedare 2016-05-25T20:34:06 *** gedare has quit IRC 2016-05-25T22:39:34 *** Davidbrcz has quit IRC 2016-05-25T23:49:18 *** Davidbrcz has joined #rtems 2016-05-26T01:57:55 [trac] GSoC/2016 edited by hwj1593 (empty comment) http://devel.rtems.org/wiki/GSoC/2016 2016-05-26T01:58:48 [trac] GSoC/2016 edited by hwj1593 (empty comment) http://devel.rtems.org/wiki/GSoC/2016 2016-05-26T02:11:13 beng-nl: BeagleBone work this year? Need boards? 2016-05-26T03:46:26 *** Davidbrcz has quit IRC 2016-05-26T03:51:10 *** Davidbrcz has joined #rtems 2016-05-26T05:45:17 *** monstr has joined #rtems 2016-05-26T05:51:51 *** mudit1729 has quit IRC 2016-05-26T05:53:18 [git] 3039e18 by Chris Johns: [testsuite: Fix networking samples to use the RTEMS printer.] http://devel.rtems.org/changeset/3039e18/rtems 2016-05-26T08:26:45 jkridner: yes i am a mentor of a BBB project with punitvara. i have boards but other mentors of the same project do not and they would be welcome yes! 2016-05-26T08:29:29 jkridner: i'll email you and cc martin.galvan@tallertechnologies.com and marcos.diaz@tallertechnologies.com 2016-05-26T08:44:42 *** davee_ has joined #rtems 2016-05-26T11:32:35 *** hesham has quit IRC 2016-05-26T11:35:38 *** hesham has joined #rtems 2016-05-26T12:01:03 *** davee_ has quit IRC 2016-05-26T12:30:27 *** lcpfnvcy_ has quit IRC 2016-05-26T12:46:13 *** lcpfnvcy has joined #rtems 2016-05-26T13:25:39 *** lcpfnvcy has quit IRC 2016-05-26T14:09:23 *** lcpfnvcy has joined #rtems 2016-05-26T14:23:28 [trac] GSoC/2016 edited by Mudit (empty comment) http://devel.rtems.org/wiki/GSoC/2016 2016-05-26T15:05:05 *** gedare has joined #rtems 2016-05-26T15:05:05 *** ChanServ sets mode: +o gedare 2016-05-26T19:27:20 *** jkridner has quit IRC 2016-05-26T19:42:26 [git] dbedcf9 by Sebastian Huber: [testsuites: Fix locked_printf() test printer] http://devel.rtems.org/changeset/dbedcf9/rtems 2016-05-26T19:50:14 *** jkridner has joined #rtems 2016-05-26T19:59:02 *** monstr has quit IRC 2016-05-26T20:34:59 *** gedare has quit IRC 2016-05-26T21:31:02 *** commander has joined #rtems 2016-05-26T21:38:31 [trac] GSoC/2016 edited by darnir (Student updates) http://devel.rtems.org/wiki/GSoC/2016 2016-05-26T22:07:15 *** gedare has joined #rtems 2016-05-26T22:07:15 *** ChanServ sets mode: +o gedare 2016-05-26T22:13:42 *** gedare has quit IRC 2016-05-26T23:59:15 *** gigetoo has quit IRC 2016-05-27T00:00:29 *** gigetoo has joined #rtems 2016-05-27T02:20:03 *** davee_ has joined #rtems 2016-05-27T04:41:48 *** monstr has joined #rtems 2016-05-27T05:42:50 *** sebhub has joined #rtems 2016-05-27T06:14:21 [trac] #2720/defect (closed) updated by Sebastian Huber (In [changeset:"39bcf7417ea39806e4817a9ce72cfc20c060c4bf/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2720 2016-05-27T06:14:22 [git] 39bcf74 by Sebastian Huber: [Fix semaphore post overflow status Close #2720.] http://devel.rtems.org/changeset/39bcf74/rtems 2016-05-27T06:14:32 [trac] #2721/defect (closed) updated by Sebastian Huber (In [changeset:"d887c1b5b30733fd2d6cf07a4a264f57ab9f4b4b/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2721 2016-05-27T06:14:33 [git] d887c1b by Sebastian Huber: [posix: Fix sem_init() with too large initial value Close #2721.] http://devel.rtems.org/changeset/d887c1b/rtems 2016-05-27T06:14:42 [git] dce4879 by Sebastian Huber: [score: Add Status_Control for all APIs Unify the status codes of the Classic and POSIX API to use the new enum Status_Control. This eliminates the Thread_Control::Wait::timeout_code field and the timeout parameter of _Thread_queue_Enqueue_critical() and _MPCI_Send_request_packet(). It gets rid of the status code translation 2016-05-27T06:14:43 tables and instead uses s...] http://devel.rtems.org/changeset/dce4879/rtems 2016-05-27T06:29:59 *** monstr has quit IRC 2016-05-27T06:41:58 *** monstr has joined #rtems 2016-05-27T08:19:36 *** dipupo has quit IRC 2016-05-27T08:26:18 [trac] SOCIS/2016 created http://devel.rtems.org/wiki/SOCIS/2016 2016-05-27T08:30:49 *** dipupo has joined #rtems 2016-05-27T09:06:14 *** davee_ has quit IRC 2016-05-27T09:07:56 *** benjamint has quit IRC 2016-05-27T09:09:37 *** benjamint has joined #rtems 2016-05-27T09:21:29 *** dipupo has quit IRC 2016-05-27T09:44:53 *** dipupo has joined #rtems 2016-05-27T10:01:20 [trac] GSoC/2016 edited by Sambeet (empty comment) http://devel.rtems.org/wiki/GSoC/2016 2016-05-27T11:46:25 *** cdesai has quit IRC 2016-05-27T11:48:59 *** cdesai has joined #rtems 2016-05-27T13:30:59 *** monstr has quit IRC 2016-05-27T13:58:54 *** sebhub has quit IRC 2016-05-27T17:00:10 *** dipupo has quit IRC 2016-05-27T18:34:52 *** mumptai has joined #rtems 2016-05-27T19:23:25 [trac] GSoC/2016 edited by vivekkukreja (empty comment) http://devel.rtems.org/wiki/GSoC/2016 2016-05-28T00:47:14 *** igutek has quit IRC 2016-05-28T00:59:27 *** mumptai has quit IRC 2016-05-28T01:14:39 *** darnir has quit IRC 2016-05-28T01:18:49 *** darnir has joined #rtems 2016-05-28T07:47:46 *** mumptai has joined #rtems 2016-05-28T11:49:57 *** jkridner has quit IRC 2016-05-28T11:50:47 *** jkridner has joined #rtems 2016-05-28T11:59:10 *** jkridner|work has joined #rtems 2016-05-28T12:02:03 *** jkridner has quit IRC 2016-05-28T13:42:19 *** jkridner has joined #rtems 2016-05-28T13:43:38 *** jkridner|work has quit IRC 2016-05-28T16:33:43 *** jkridner has quit IRC 2016-05-28T16:33:56 *** jkridner has joined #rtems 2016-05-28T19:42:06 *** jkridner|work has joined #rtems 2016-05-28T19:42:17 *** jkridner has quit IRC 2016-05-28T23:24:07 *** jkridner|work has quit IRC 2016-05-28T23:24:29 *** jkridner has joined #rtems 2016-05-29T03:57:22 *** mumptai has quit IRC 2016-05-29T04:09:53 *** mumptai has joined #rtems 2016-05-29T14:26:29 I'm trying to write a SMP test that uses a specific scheduler. Namely, the strongapascheduler that I am working on. How do I configure rtems to select it? 2016-05-29T14:31:27 Found it, by browsing through a few tests 2016-05-29T18:33:23 *** jkridner has quit IRC 2016-05-29T18:33:44 *** jkridner has joined #rtems 2016-05-29T22:03:08 *** benjamint has quit IRC 2016-05-29T22:03:24 *** benjamint has joined #rtems