2016-05-30T04:38:05 *** monstr has joined #rtems 2016-05-30T05:30:52 *** sebhub has joined #rtems 2016-05-30T08:45:51 *** mumptai has quit IRC 2016-05-30T13:08:51 [git] 3da4d0e by Sebastian Huber: [4.12/rtems-default.bset: Use GCC 6-20160526 In addition use Newlib snapshot 2.4.0.20160527.] http://devel.rtems.org/changeset/3da4d0e/rtems-source-builder 2016-05-30T13:53:42 *** monstr has quit IRC 2016-05-30T13:55:47 *** monstr has joined #rtems 2016-05-30T14:18:08 [git] fd6fde8 by Sebastian Huber: [score: Add libatomic support Close #2695.] http://devel.rtems.org/changeset/fd6fde8/rtems 2016-05-30T14:18:08 [git] cb2651d by Sebastian Huber: [network: Align with Newlib type definitions] http://devel.rtems.org/changeset/cb2651d/rtems 2016-05-30T14:18:08 [git] 868cd24 by Sebastian Huber: [score: Use Newlib provided ] http://devel.rtems.org/changeset/868cd24/rtems 2016-05-30T14:18:08 [git] dfcc8bb by Sebastian Huber: [score: Adjust thread queue layout Adjust thread queue layout according to Newlib. This makes it possible to use the same implementation for and CORE mutexes in the future.] http://devel.rtems.org/changeset/dfcc8bb/rtems 2016-05-30T14:18:08 [git] 0e1d11f by Sebastian Huber: [score: Add _Thread_queue_Context_set_MP_callout() Add _Thread_queue_Context_set_MP_callout() to simplify _Thread_queue_Context_initialize(). This makes it possible to more easily add additional fields to Thread_queue_Context.] http://devel.rtems.org/changeset/0e1d11f/rtems 2016-05-30T14:18:09 [git] 9330605 by Sebastian Huber: [score: _CORE_mutex_Check_dispatch_for_seize() Move the safety check performed by _CORE_mutex_Check_dispatch_for_seize() out of the performance critical path and generalize it. Blocking on a thread queue with an unexpected thread dispatch disabled level is illegal in all system states. Add the expected thread dispatch 2016-05-30T14:18:09 disable level (which may be 1 or...] http://devel.rtems.org/changeset/9330605/rtems 2016-05-30T14:18:10 [git] af746b0 by Sebastian Huber: [score: Use thread queue lock for MrsP Replace the ISR lock in MRSP_Control with a thread queue. This simplifies the Classic semaphore implementation. Only the lock part of the thread queue is used.] http://devel.rtems.org/changeset/af746b0/rtems 2016-05-30T14:18:10 [git] 3ca6e61 by Sebastian Huber: [rtems: Simplify rtems_semaphore_delete()] http://devel.rtems.org/changeset/3ca6e61/rtems 2016-05-30T14:18:11 [git] 2581a56 by Sebastian Huber: [score: Add semaphore variants] http://devel.rtems.org/changeset/2581a56/rtems 2016-05-30T14:18:11 [git] 09c5ca4 by Sebastian Huber: [score: Simplify CORE mutex Remove superfluous support for simple binary semaphores. With this we can get rid of the CORE_MUTEX_NESTING_BLOCKS variant.] http://devel.rtems.org/changeset/09c5ca4/rtems 2016-05-30T14:18:12 [git] 8a77ef6 by Sebastian Huber: [rtems: Simplify rtems_semaphore_flush() Due to the unified status codes, we can now use a common flush method for semaphore and mutex variants.] http://devel.rtems.org/changeset/8a77ef6/rtems 2016-05-30T14:18:12 [git] 3c1ad25 by Sebastian Huber: [rtems: Simplify rtems_semaphore_delete() Due to the unified status codes, we can now use a common flush and destroy method for semaphore and mutex variants.] http://devel.rtems.org/changeset/3c1ad25/rtems 2016-05-30T14:18:13 [git] bac50cf by Sebastian Huber: [rtems: Remove superfluous includes] http://devel.rtems.org/changeset/bac50cf/rtems 2016-05-30T14:18:24 [git] 70d52b8 by Sebastian Huber: [rtems-bsd-mutex: Update due to API changes] http://devel.rtems.org/changeset/70d52b8/rtems-libbsd 2016-05-30T14:52:28 *** monstr has quit IRC 2016-05-30T17:07:32 *** mumptai has joined #rtems 2016-05-30T21:58:39 *** gigetoo has quit IRC 2016-05-30T21:59:56 *** gigetoo has joined #rtems 2016-05-30T22:53:03 *** mumptai has quit IRC 2016-05-30T22:54:48 *** gigetoo has quit IRC 2016-05-30T22:56:20 *** gigetoo has joined #rtems 2016-05-30T23:20:01 *** gigetoo has quit IRC 2016-05-30T23:22:46 *** gigetoo has joined #rtems 2016-05-30T23:50:27 [git] f9798ad by Chris Johns: [Add a stats report command. The report shows the level of changes we have made to the FreeBSD code.] http://devel.rtems.org/changeset/f9798ad/rtems-libbsd 2016-05-31T01:05:18 *** gigetoo has quit IRC 2016-05-31T01:09:19 *** gigetoo has joined #rtems 2016-05-31T03:14:59 *** dougk has quit IRC 2016-05-31T03:27:55 *** dougk has joined #rtems 2016-05-31T04:31:13 *** monstr has joined #rtems 2016-05-31T06:16:04 *** mumptai has joined #rtems 2016-05-31T06:29:20 *** jkridner has quit IRC 2016-05-31T06:59:18 *** jkridner has joined #rtems 2016-05-31T07:17:54 [git] ab5cd63 by Chris Johns: [tests: Fix compile errors after tools upgrade.] http://devel.rtems.org/changeset/ab5cd63/rtems-libbsd 2016-05-31T07:19:10 sebhub: should the arphole write have an assert? 2016-05-31T08:09:28 [trac] #2723/enhancement (v:4.12) created by sebastian.huber (CPUINFO command to report per-processor information) http://devel.rtems.org/ticket/2723 2016-05-31T08:12:02 [trac] #2723/enhancement (new) updated by Sebastian Huber (In [changeset:"01cb5540bd7b391f9a19c79b284fdfc4385afc06/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2723 2016-05-31T08:12:03 [git] ccda945 by Sebastian Huber: [bsps: Sort some sections first by alignment This helps to avoid alignment padding and thus may reduce some memory waste.] http://devel.rtems.org/changeset/ccda945/rtems 2016-05-31T08:12:03 [git] 0546e65 by Sebastian Huber: [bsps: Add defines for some linker subsections The GNU ld sort by name or alignment needs distinct input sections.] http://devel.rtems.org/changeset/0546e65/rtems 2016-05-31T08:12:03 [git] a9cff51 by Sebastian Huber: [bsp/qoriq: Fix nocache region placement Avoid memory waste due to alignment padding. Provide enough space for the DPAA driver infrastructure.] http://devel.rtems.org/changeset/a9cff51/rtems 2016-05-31T08:12:03 [git] 3242614 by Sebastian Huber: [Fix rtems_print_printer_fprintf() We must use vfprintf().] http://devel.rtems.org/changeset/3242614/rtems 2016-05-31T08:12:03 [git] 5e072f6 by Sebastian Huber: [cpuuse: Hide implementation details] http://devel.rtems.org/changeset/5e072f6/rtems 2016-05-31T08:12:03 [git] 01cb554 by Sebastian Huber: [shell: Add CPUINFO command Update #2723.] http://devel.rtems.org/changeset/01cb554/rtems 2016-05-31T08:21:13 strange, I didn't got errors in the arphole and loopback01 tests 2016-05-31T08:22:01 I added the arphole test to test the ipv4 link local addressing of the dhcpcd 2016-05-31T08:22:51 runnig it on a real network is not a good idea 2016-05-31T08:27:13 [git] dd2c57d by Sebastian Huber: [libchip: Remove dwmac network interface driver The network interface driver for this module is located in the new network stack (libbsd), see file "if_dwc.c".] http://devel.rtems.org/changeset/dd2c57d/rtems 2016-05-31T08:46:30 *** mumptai has quit IRC 2016-05-31T09:53:27 [git] 4c59eeb by Pavel Pisa: [arm/raspberrypi: cmdline enhancement and early access workaround. cmdline argument is returned without option name. The calls to command line parsing can be called earlier before BSS is cleaned out.] http://devel.rtems.org/changeset/4c59eeb/rtems 2016-05-31T09:53:27 [git] de378ad by Pavel Pisa: [arm/raspberrypi: ensure that correct RPI_PERIPHERAL_BASE is provided by raspberrypi.h If the raspberrypi.h has been included without preceding inclussion of bsp.h then BSP_IS_RPI2 has not been set for Raspberry Pi 2 BSP variant and bad things happen later. The patch includes bspopts.h by raspberrypi.h and even includes bsp.h in 2016-05-31T09:53:27 critical peripherals support.] http://devel.rtems.org/changeset/de378ad/rtems 2016-05-31T09:53:27 [git] b96defd by Pavel Pisa: [arm/raspberrypi: Raspberry Pi v2 ALLOCATE_BUFFER VC4 operation returns direct address.] http://devel.rtems.org/changeset/b96defd/rtems 2016-05-31T09:53:27 [git] c1a9f6a by YANG Qiao: [arm/raspberrypi: add VideoCore framebuffer without initialization] http://devel.rtems.org/changeset/c1a9f6a/rtems 2016-05-31T09:53:28 [git] c64d5f0 by Pavel Pisa: [arm/raspberrypi: move MMU in front of application image to respect variable memory size. The page table is placed at address 0x00004000 which provides required 16 kB space till the start of application image. The RAM size specified in a linker script is upper limit address of RAM utilized for the work area initialization. If 2016-05-31T09:53:28 VideoCore reports to use lowe...] http://devel.rtems.org/changeset/c64d5f0/rtems 2016-05-31T09:53:29 [git] c12069e by YANG Qiao: [arm/raspberrypi: add video outchar support for rpi bsp] http://devel.rtems.org/changeset/c12069e/rtems 2016-05-31T09:53:29 [git] 6b5df95 by YANG Qiao: [arm/raspberrypi: add fbcons support for rpi bsp] http://devel.rtems.org/changeset/6b5df95/rtems 2016-05-31T12:58:22 *** darnir has quit IRC 2016-05-31T13:02:52 *** darnir has joined #rtems 2016-05-31T14:17:05 [trac] #2724/defect (v:4.10) created by burrusse (Compiling GCC 4.4.7 for rtems 4.10 branch fails when compiled with newer GCC) http://devel.rtems.org/ticket/2724 2016-05-31T14:24:10 [trac] gcc-g++-4.4.7-rtems4.10-20160413.diff () attached to #2724 by burrusse http://devel.rtems.org/ticket/2724 2016-05-31T14:24:27 [trac] gdb-7.3.1-rtems4.10-20160414.diff () attached to #2724 by burrusse http://devel.rtems.org/ticket/2724 2016-05-31T14:24:44 [trac] rsb-15april2016.diff () attached to #2724 by burrusse http://devel.rtems.org/ticket/2724 2016-05-31T14:56:55 *** gedare has joined #rtems 2016-05-31T14:56:55 *** ChanServ sets mode: +o gedare 2016-05-31T15:39:56 *** gedare has quit IRC 2016-05-31T16:23:22 *** gedare has joined #rtems 2016-05-31T16:23:22 *** ChanServ sets mode: +o gedare 2016-05-31T16:23:23 *** jkridner|work has joined #rtems 2016-05-31T16:25:44 *** jkridner has quit IRC 2016-05-31T16:53:06 *** gedare has quit IRC 2016-05-31T17:43:41 *** mumptai has joined #rtems 2016-05-31T19:56:21 *** monstr has quit IRC 2016-05-31T21:36:18 *** mumptai has quit IRC 2016-05-31T22:51:13 *** cdesai has quit IRC 2016-05-31T22:53:10 *** cdesai has joined #rtems 2016-06-01T05:38:08 *** monstr has joined #rtems 2016-06-01T07:54:25 [git] a9cc6a8 by Sebastian Huber: [smptests/smpatomic01: New test cases Demonstrate that a read-modify-write atomic operation may be necessary on some archtitectures to observe the latest value written.] http://devel.rtems.org/changeset/a9cc6a8/rtems 2016-06-01T09:27:16 [git] f807b84 by Sebastian Huber: [score: Fix _Thread_Lock_acquire() A read-modify-write operation is necessary to read the last value written. See for example C11 standard or Power ISA 2.07, Book II: Power ISA Virtual Environment Architecture, Section 1.6.3 Memory Coherence Required [Category: Memory Coherence] and Section 1.7.3 Atomic Update.] 2016-06-01T09:27:16 http://devel.rtems.org/changeset/f807b84/rtems 2016-06-01T12:50:41 *** salpha has joined #rtems 2016-06-01T13:07:32 [trac] SOCIS edited by salpha (empty comment) http://devel.rtems.org/wiki/SOCIS 2016-06-01T13:08:12 *** vivekk has joined #rtems 2016-06-01T13:14:58 *** salpha has quit IRC 2016-06-01T13:36:47 *** gedare has joined #rtems 2016-06-01T13:36:47 *** ChanServ sets mode: +o gedare 2016-06-01T13:42:23 *** deval has joined #rtems 2016-06-01T13:46:02 *** salpha has joined #rtems 2016-06-01T13:48:53 hello everyone 2016-06-01T13:55:20 *** _sambeet has joined #rtems 2016-06-01T13:56:26 good morning 2016-06-01T13:56:40 salpha you can get us started a bit early today, if you like. 2016-06-01T13:56:53 introduce yourself, and give a status briefing, since other students don't know your project yet 2016-06-01T13:58:00 * gedare pokes salpha 2016-06-01T13:58:29 sorry 2016-06-01T13:58:53 i'm saeed ehteshamifar, working on rtems fault injection project during this SOCIS 2016-06-01T13:59:44 it's about adapting two fault injection tools to rtems so in the end, it can automatically generate faulty rtems applications to evaluate how strong rtems is, against faults. 2016-06-01T14:00:21 *** DrJoel has joined #rtems 2016-06-01T14:00:22 *** DrJoel has joined #rtems 2016-06-01T14:00:22 *** ChanServ sets mode: +o DrJoel 2016-06-01T14:00:54 *** mudit1729 has joined #rtems 2016-06-01T14:01:16 *** salpha2 has joined #rtems 2016-06-01T14:01:26 i've not done technical stuff this week but as a first step, i wanna distribute Slingshot as a stand-alone tool (not as a patch to the original version). 2016-06-01T14:01:43 this also entails recapping my work from the last year. 2016-06-01T14:01:56 good, i hope you can get these "wrap-up" pieces done quickly, they should rightly have been done by you during the intervening months. :) 2016-06-01T14:02:04 the task seems to be straight-forward, i hope. 2016-06-01T14:02:37 right. :) 2016-06-01T14:02:43 'll then update Slingshot's repository on github with new build instructions. 2016-06-01T14:02:59 *** dipupo has joined #rtems 2016-06-01T14:03:17 good, it may be worth it to provide an RSB script at some point 2016-06-01T14:03:58 ok thanks salpha2 any questions? 2016-06-01T14:04:12 no, not yet. 2016-06-01T14:04:13 thanks 2016-06-01T14:04:21 ok. deval you ready? 2016-06-01T14:04:25 yes 2016-06-01T14:04:29 ok go ahead 2016-06-01T14:04:35 *** salpha has quit IRC 2016-06-01T14:04:42 Hello, My first task is to add USB support (from libbsd) to raspberry pi. I have resolved previous errors in building the drivers. So now the bcm283x_dwcotg driver compiles along with testsuits. Now I tried running the testsuits, it just shows "nexus0: " and hangs. Also, if I plug-in any usb device to Raspberry pi they are not getting powered on. 2016-06-01T14:04:50 I asked Yurii, he worked on the same problem last year, but he doesnt remember much about the errors or issues, however I have taken care whatever he has mentioned in his blog. 2016-06-01T14:04:55 RPI cant work with qemu > https://lists.rtems.org/pipermail/devel/2016-May/015101.html So only option left is JTAG debugger. Which I have ordered but that would take about a week to get shipped to my place. 2016-06-01T14:05:56 ok, if you're stuck meanwhile, perhaps you can look into the qemu problem 2016-06-01T14:06:02 *** _sambeet has quit IRC 2016-06-01T14:06:06 it appears you could fix the problem by changing the clock source? 2016-06-01T14:06:15 *** beng-nl has quit IRC 2016-06-01T14:06:29 I will have to see that. 2016-06-01T14:06:41 yes please do, send a reply to the email thread related to it 2016-06-01T14:06:42 I dont have much idea into that 2016-06-01T14:06:46 Verify the Pi actually has a Nexus device configured in rtems-libbsd. That is a tricky thing to get correct even if it is there. We have been battling it on the PC. 2016-06-01T14:06:47 but I will try 2016-06-01T14:06:56 *** beng-nl has joined #rtems 2016-06-01T14:06:57 *** salpha2 has quit IRC 2016-06-01T14:07:03 send email reply to say you might like to look into it, and ask for extra guidance 2016-06-01T14:07:25 okay 2016-06-01T14:07:27 sure 2016-06-01T14:07:39 thanks deval. mudit1729 you're up next 2016-06-01T14:07:48 thank you 2016-06-01T14:08:01 *** g_ has joined #rtems 2016-06-01T14:09:14 darnir: you here? 2016-06-01T14:09:19 yes 2016-06-01T14:09:26 ok you can go i gues mudit1729 is asleep a bit. 2016-06-01T14:09:31 Okay 2016-06-01T14:09:33 I've spent the last week working on creating the base for the iterative MVM solver. I've started coding it up, but I'm currently doing it out-of-tree since I had a few issues with understanding how to debug the scheduler code. Maybe we can have a meeting later this week (friday?) to go through that. Till then I'm working on implementing the algorithm outside in C99 and it shouldn't be too hard to just port 2016-06-01T14:09:39 it to RTEMS as soon as I have a better grip of step tracing through it. 2016-06-01T14:10:19 Send an email to me, joel, and sebastian about planning a meeting to discuss the scheduler. 2016-06-01T14:10:24 Okay. 2016-06-01T14:10:32 *** _sambeet has joined #rtems 2016-06-01T14:10:35 if you have some particular questions that may help 2016-06-01T14:10:45 Sure. 2016-06-01T14:10:56 what BSP you use? 2016-06-01T14:11:10 realview 2016-06-01T14:11:12 arm 2016-06-01T14:11:35 It's the only one with good SMP support, I've been told. 2016-06-01T14:11:35 you can hook gdb? 2016-06-01T14:11:49 *** _sambeet_ has joined #rtems 2016-06-01T14:11:55 Yes. But I had some issues with writing a test application and stepping through it. 2016-06-01T14:12:03 ok report on those problems 2016-06-01T14:12:16 And one question that may be of general interest.. 2016-06-01T14:12:52 Whenever I make a small change to the rtems source and try to compile it using make, the process takes a fairly long time. Though make should just compile the changed files and link them together. 2016-06-01T14:13:22 This means I end up waiting a lot between minor changes. I'd like to see if I'm doing something wrong. 2016-06-01T14:14:27 there are a couple of tricks you can do to speed up compilation time 2016-06-01T14:14:29 I use a script that invokes "make preinstall; autoreconf; make" in the build dir to build the source 2016-06-01T14:14:30 and yes this is of general interest 2016-06-01T14:14:42 *** _sambeet has quit IRC 2016-06-01T14:14:48 if you add/remove header files, you have to re-run bootstrap -p 2016-06-01T14:14:59 if you modify config.ac/makefile.am you ahve to re-run bootstrap 2016-06-01T14:15:05 and then configure&&make 2016-06-01T14:15:26 but only from the directory you changed. And if you build with --enable-maintainer-mode adding files is generally OK. 2016-06-01T14:15:30 otherwise, if all you change is source file contents, you don't need to re-run bootstrap or configure, assuming you still want the same configure options 2016-06-01T14:15:46 in such case, you can just run 'make' from your build tree, BUT 2016-06-01T14:15:50 But often just removing the executables in the build tree and doing a make is OK 2016-06-01T14:15:53 this will not re-make the testsuites 2016-06-01T14:16:06 so usually what has to be done is... 2016-06-01T14:16:26 `find . -name "*.exe" | xargs rm` && make 2016-06-01T14:16:29 from a build directory: rm -f `find . -name "*.exe"` ; make -jXXX ... where XXX is a number about 1.5-2x the number of cores. 2016-06-01T14:16:47 or what he said. 2016-06-01T14:16:54 Looks like we do the same thing. :) 2016-06-01T14:17:30 this should cut your build time. if you only use one particular test application, you can also just rm that .exe 2016-06-01T14:17:36 I usually just run "make", but that seems to take a couple of minutes to execute as it compiles the entire tree. I'll check again to see if I was doing something wrong. 2016-06-01T14:17:42 and your build time should be about 30s or less 2016-06-01T14:18:12 And make preinstall is required in which particular cases? 2016-06-01T14:18:13 just remember that you have to rm your .exe since the make won't pick up the dependency, there is a bug in the build system 2016-06-01T14:18:35 i never use that one. 2016-06-01T14:18:58 Aah. Probably why my tests weren't working as expected as well. And why I often end up doing a fresh compile. 2016-06-01T14:19:29 Oh. I remember someone mentioned to me once that I should run make preinstall after a git pull when building in maintainer mode. 2016-06-01T14:19:48 *** _sambeet has joined #rtems 2016-06-01T14:20:39 But I'll consider these and try to bring my build times down. That'll help with productivity. Thanks! 2016-06-01T14:20:41 maybe. anytime you git pull, you should just nuke your build tree. 2016-06-01T14:20:42 if you have multiple cores, there is sb-bootstrap in the rtems-source-builder repo which parallelizes the bootstrap. On a 12 core machine, it bootstraps the entire tree in ~45 sec. I do it fresh after each serious pull 2016-06-01T14:20:54 and re-run bootstrap 2016-06-01T14:21:03 ok thanks darnir. _sambeet would you like to go? 2016-06-01T14:21:55 or mudit1729 if you're around 2016-06-01T14:22:09 _sambeet seems to be having connectivity issues, 2016-06-01T14:22:22 _sambeet_ or _sambeet I don't know 2016-06-01T14:22:30 *** _sambeet_ has quit IRC 2016-06-01T14:22:42 Hi 2016-06-01T14:22:45 <_sambeet> Here is my status report. 2016-06-01T14:22:52 there was a power outage here. 2016-06-01T14:22:54 <_sambeet> I was down with fever this week.So I could not work more. To speed up I am using an already patched version of OmniORB 4.1.Till now its not giving problems.I also corrected the codes in remaining all Rock dependencies except libxmll.The Patches doesn't seem to work.Once I have an already working Autoproj port I will update all the dependencies. Once I am done with a working version, I ll start updating the dependencies again. 2016-06-01T14:23:35 Hi everyone. 2016-06-01T14:23:36 For what purpose was OmniORB patched? 2016-06-01T14:23:43 hey mudit1729 wait a sec until _sambeet is done 2016-06-01T14:23:53 sure. 2016-06-01T14:24:58 <_sambeet> It was used for the previous SOCIS project for porting rock to rtems.Thomas provided me the scripts. 2016-06-01T14:25:00 _sambeet: the patches for libxml don't work? double check all version numbers/commit hashes if available. otherwise you will need to look at the libxml source and figure out where to make the manual edits for the patches. 2016-06-01T14:25:04 And are you able to test each of the dependent packages? 2016-06-01T14:25:24 <_sambeet> Yes sir.exc 2016-06-01T14:26:15 Yes.. you can test? 2016-06-01T14:26:31 <_sambeet> except libxml I made manual corrections to all dependencies 2016-06-01T14:27:04 <_sambeet> Eigen boost OmniORB all have been tested 2016-06-01T14:27:44 _sambeet: are you running these on your computer? 2016-06-01T14:27:45 It would be really nice to move the build scripts to RSB recipes. That will simplify making the port work on other architectures and BSPs. But it has to work on one first. 2016-06-01T14:28:28 <_sambeet> I would be testing on i386 this week. 2016-06-01T14:29:15 <_sambeet> Once I am done testing the port with Autoproj I ll make rsb recipes. 2016-06-01T14:29:30 ok, so i'm not sure what you mean above when you say the dependencies "all have been tested", please be clear what you have done on your PC, and what is done on an RTEMS target. 2016-06-01T14:30:25 Dig into the libxml issue, and then as you try to get the whole package to build for RTEMS target it would be a good time to write a blog report on this process. 2016-06-01T14:30:28 mudit1729: you're up next 2016-06-01T14:30:35 _sambeet: anything else to add? 2016-06-01T14:31:09 <_sambeet> It means sir I have run test examples individually but yet to build Rock 2016-06-01T14:31:31 <_sambeet> Sure sir I would update my blog 2016-06-01T14:31:49 ok thanks _sambeet. hi mudit1729 2016-06-01T14:32:00 Hi 2016-06-01T14:32:22 updates : I have done an initial port of the driver adding all the dependencies. waf builds everything succesfully. I have sent a mail regarding the changes that I had made. There were few changes suggested by people, I am presently working on that. 2016-06-01T14:33:26 In the mail I had asked regarding the review and the merge process. 2016-06-01T14:33:57 yeah 2016-06-01T14:35:07 *** beng-nl has quit IRC 2016-06-01T14:35:21 ok anything else mudit1729? 2016-06-01T14:35:27 I wanted to know about that. Second was there was a code of bcm2835 specific code ported. I wanted to discuss regarding the placement of the same 2016-06-01T14:35:27 dipupo: you'll be up after mudit1729. 2016-06-01T14:36:17 You're already going through the review and merge process. it is about posting patches or link to github to devel mailing list and asking people to review 2016-06-01T14:36:31 okay, 2016-06-01T14:36:58 i don't know about the specific code, that you should ask back to the mailing list. does that code go into libbsd? 2016-06-01T14:37:31 you need to see if there are other examples you can mimic, or else ask for guidance3. 2016-06-01T14:37:44 Okay. 2016-06-01T14:38:02 and yes the code goes into libbsd 2016-06-01T14:38:49 alright. thanks mudit1729. also take some time to write in your blog, and think about how you might consider improving the libbsd tocumentation. 2016-06-01T14:39:00 *doc. 2016-06-01T14:39:04 hi dipupo 2016-06-01T14:39:09 Yes, I will be doing that this week 2016-06-01T14:39:22 Hello, I was able to resolve the issues I was having with building and running LM3S69XX BSP last week thanks to Martin. 2016-06-01T14:39:51 Presently, am working on adding bsp_reset to this BSP because it doesnt have its own and running program doesn't terminate. 2016-06-01T14:39:51 I have not been able to test the LM4F BSP because qemu doesn't support the board. So, I have ordered one that will be delivered by Friday.The LM4F is a benchmark for my port and I needed to see its behavior on actual board since there is no emulator that support it. 2016-06-01T14:40:16 *** _sambeet has quit IRC 2016-06-01T14:40:29 I have also make all the code changes for my BSP port but the build failed when I ran it last night. I am currently debugging and any issues am not able to resolve with the sent to the mailing list. 2016-06-01T14:41:06 For bsp_reset(), have you gotten far enough to tell if the reset "device" is part of the SoC or the board it is on? Hopefully it is part of the SoC so there on all boards with the SoC. 2016-06-01T14:42:26 From what I have read, it is part of the board 2016-06-01T14:43:40 ok thanks dipupo. 2016-06-01T14:44:03 Thank you 2016-06-01T14:44:11 dipupo, and everyone else, please make sure you are pushing your code up to github as you work. 2016-06-01T14:44:30 Wonjun: you're up next 2016-06-01T14:46:16 Can i start now? 2016-06-01T14:46:20 yes go ahead 2016-06-01T14:46:31 Last week, I tryied to install jailhouse. To install jailhouse, I used ubuntu 14.04 LTS to begin with. but that version has old kernel version. 2016-06-01T14:46:31 so I changed Debian 8.4 but it was also old. 2016-06-01T14:46:46 and then I am using 16.04 LTS and QEMU version 2.5. It is Ok. 2016-06-01T14:46:46 I enabled jailhouse on QEMU successfully, but when I create cell, guest freeze. 2016-06-01T14:46:58 Jan said, those host and QEMU setting is ok, It should works. 2016-06-01T14:46:58 so I am trying to find a solution at jailhouse google group site. 2016-06-01T14:47:49 Now I am updating QEMU to latest version. 2016-06-01T14:47:50 ok. guest = cellmate? 2016-06-01T14:47:55 yes 2016-06-01T14:48:22 is the cellmate the same version of Linux as in the Ubuntu? 2016-06-01T14:48:55 yes they recommand. 2016-06-01T14:49:31 ok. can you attach gdb to your qemu instance? 2016-06-01T14:51:06 I will try. 2016-06-01T14:51:27 ok thanks Wonjun anything else? vivekk you're up next 2016-06-01T14:52:18 touching with Jan is very active. so it can be solved 2016-06-01T14:52:23 thanks 2016-06-01T14:52:39 Thanks. Hello everyone i am Vivek. my GSOC project title is 'Improvements in Tracing framework'. In the first phase i will obtain traces in CTF format. I forwarded an experimental patch for function tracing to Chris and he has advised many changes. 2016-06-01T14:52:54 First, we are yet to decide on an interface that lets user choose standard vs CTF trace. Second, we need to define a configuration for tracing like YAML file in barectf but higher-level. Third, we need variable types/sizes for CTF metadata file. We can get function signatures from DWARF info in ELF files but as Chris said, libDWARF is yet to be added to the toolkit so we are looking at C++ alternatives. 2016-06-01T14:52:56 great Wonjun. 2016-06-01T14:53:32 Wonjun.. just be persistent and keep asking for help getting it to work at all. 2016-06-01T14:53:35 vivekk: note that it's not just "standard" vs "CTF", but about any kind of trace format one might care to add natively 2016-06-01T14:54:29 ok vivekk that is a lot, where will you begin? 2016-06-01T14:55:30 Sure. I'll open a mail chain to discuss architectural issues. I'm currently improving on the code and trying to create the CTF metadata file 2016-06-01T14:56:19 ok, thanks vivekk. 2016-06-01T14:56:23 hmm, anyone else? 2016-06-01T14:56:38 i know punit is on travel 2016-06-01T14:57:02 sacha appears to be missing. 2016-06-01T14:57:02 There is a mail chain regarding the same. Mathieu and other barectf maintainers have some interesting solutions. Finally i'm still encountering problems running capture.exe example for ARM/Xilinx_zynq_a9_qemu. I will write a mail to the previous chain regarding the same. 2016-06-01T14:57:04 thanks 2016-06-01T14:57:26 try other BSPs too, then 2016-06-01T14:57:57 for your project it will eventually be important to be able to test and work across different architectures to ensure your solutions are portable. 2016-06-01T14:58:18 sparc/sis, powerpc/psim and mips/jmr3904 all use simulators in gdb, are easy to run, and changing architectures can often make bugs very explicit 2016-06-01T14:58:41 ok thanks everyone, remember to update https://devel.rtems.org/wiki/GSoC/2016#StudentStatusUpdates 2016-06-01T14:59:01 gedare: One more thing. 2016-06-01T14:59:02 as that provides a concise history of your project statuses in the aggregate 2016-06-01T14:59:16 sure darnir, 2016-06-01T14:59:24 yes i'm cooncurrently testing my changes on sparc/sis. will update my blog today. thanks everyone 2016-06-01T14:59:30 meanwhile, meeting adjourned 2016-06-01T14:59:33 It's just a minor annoyance, but on running bootstrap, configure and make, the rtems source tree is always modified 2016-06-01T14:59:38 If you need help, don't be afraid to reach out and ask. It is summer, folks are busy .. just ping us. 2016-06-01T14:59:43 The c/src/lib/libbsp/bfin/acinclude.m4 and c/src/lib/libbsp/powerpc/acinclude.m4 files are changed with one line moved around in the file. 2016-06-01T14:59:59 try using rtems-source-builder/source-builder/sb-bootstrap 2016-06-01T15:00:18 @darnir are the rtems tools at the front of your PATH? That sounds like the sorting problem. 2016-06-01T15:00:18 DrJoel: Error: "darnir" is not a valid command. 2016-06-01T15:00:22 darnir are the rtems tools at the front of your PATH? That sounds like the sorting problem. 2016-06-01T15:00:36 Yes, they are. 2016-06-01T15:00:42 Which host? 2016-06-01T15:00:59 we've seen this nuisance crop up from time to time. 2016-06-01T15:01:02 Using sb-bootstrap, I don't see the issue though. Just ./bootstrap used to cause it. 2016-06-01T15:01:27 Machine is currently running configure, after a fresh pull + bootstrap. Let's see. 2016-06-01T15:01:30 kiwichris_ (Chris Johns) has spent time investigating and fixing this multiple times. File a ticket and assign it to him. Provide details. It is a sorting issue which won't die. 2016-06-01T15:01:39 *** Wonjun has quit IRC 2016-06-01T15:02:23 Okay. What details would be useful / expected? 2016-06-01T15:02:27 I'll do that. 2016-06-01T15:03:06 *** Wonjun has joined #rtems 2016-06-01T15:04:11 host and version info for your autotools 2016-06-01T15:05:10 Okay. I'll do it today itself. Apparently the compilation failed as well. Looking into that first. Thanks! 2016-06-01T15:06:59 Just email the list and file tickets. I am trying to be more responsive. 2016-06-01T15:07:29 Installing updates brb 2016-06-01T15:07:32 *** DrJoel has quit IRC 2016-06-01T15:08:39 *** DrJoel has joined #rtems 2016-06-01T15:08:54 *** DrJoel has joined #rtems 2016-06-01T15:08:54 *** ChanServ sets mode: +o DrJoel 2016-06-01T15:12:04 *** deval has quit IRC 2016-06-01T15:28:01 *** igutek has joined #rtems 2016-06-01T15:32:52 *** DrJoel has quit IRC 2016-06-01T15:34:29 *** DrJoel has joined #rtems 2016-06-01T15:34:30 *** DrJoel has joined #rtems 2016-06-01T15:34:30 *** ChanServ sets mode: +o DrJoel 2016-06-01T15:37:30 *** DrJoel has quit IRC 2016-06-01T15:39:57 Hey everyone, before I hit the email list, does anyone (probably sebhub), have an example file bringing up the network on the ATSAMV bsp? 2016-06-01T16:17:46 *** g_ has quit IRC 2016-06-01T16:36:36 *** jkridner|work is now known as jkridner 2016-06-01T16:46:08 *** beng-nl has joined #rtems 2016-06-01T17:38:49 *** vivekk has quit IRC 2016-06-01T18:05:14 *** mudit1729 has quit IRC 2016-06-01T19:34:57 *** mumptai has joined #rtems 2016-06-01T20:10:30 *** igutek has quit IRC 2016-06-01T20:26:24 *** monstr has quit IRC 2016-06-01T21:07:49 *** gedare has quit IRC 2016-06-01T21:11:21 *** igutek has joined #rtems 2016-06-01T22:48:19 *** gigetoo has quit IRC 2016-06-01T22:49:08 *** gigetoo has joined #rtems 2016-06-01T22:56:37 *** mumptai has quit IRC 2016-06-01T23:42:54 *** gigetoo has quit IRC 2016-06-01T23:43:31 *** gigetoo has joined #rtems 2016-06-02T03:57:21 *** guillaum1 has quit IRC 2016-06-02T04:01:52 *** guillaum1 has joined #rtems 2016-06-02T04:49:08 *** Wonjun has quit IRC 2016-06-02T05:16:05 *** ita has quit IRC 2016-06-02T05:28:22 *** monstr has joined #rtems 2016-06-02T05:45:20 [git] c6556e2 by Sebastian Huber: [score: Maybe fix _Thread_Lock_acquire() The approach with the generation number was broken. The load/store of the current lock, the thread queue and the thread queue operations were not properly synchronized. Under certain conditions on a PowerPC T4240 old thread queue operations operated on a new thread queue (NULL 2016-06-02T05:45:20 pointer).] http://devel.rtems.org/changeset/c6556e2/rtems 2016-06-02T06:01:38 *** mumptai has joined #rtems 2016-06-02T08:39:17 *** mumptai has quit IRC 2016-06-02T12:01:59 *** monstr has quit IRC 2016-06-02T12:11:25 *** monstr has joined #rtems 2016-06-02T13:47:30 *** sacha23 has joined #rtems 2016-06-02T13:55:52 Hi ,is there weekly meetup now? 2016-06-02T14:02:55 *** sacha23 has quit IRC 2016-06-02T14:10:51 *** gedare has joined #rtems 2016-06-02T14:10:51 *** ChanServ sets mode: +o gedare 2016-06-02T14:46:44 *** monstr has quit IRC 2016-06-02T15:33:04 *** benjamint has quit IRC 2016-06-02T20:39:20 *** gedare has quit IRC 2016-06-02T22:03:14 *** mumptai has joined #rtems 2016-06-02T23:07:33 *** mumptai has quit IRC 2016-06-03T00:20:34 darnir: what host are you on? 2016-06-03T00:21:11 igutek: which networking stack? 2016-06-03T05:46:15 *** mumptai has joined #rtems 2016-06-03T05:48:04 *** monstr has joined #rtems 2016-06-03T06:16:16 [trac] #2725/defect (v:4.10) created by sebastian.huber (Classic binary semaphores without a locking protocol can be release by everyone) http://devel.rtems.org/ticket/2725 2016-06-03T06:22:30 sebhub: is the #2725 first sentence saying without a locking protocol there is no owner check in the release path? 2016-06-03T06:24:37 yes, without a locking protocol 2016-06-03T06:24:47 [trac] #2725/defect (new) updated by sebastian.huber (empty comment) http://devel.rtems.org/ticket/2725 2016-06-03T06:25:17 [trac] #2725/defect (new) updated by chrisj (empty comment) http://devel.rtems.org/ticket/2725 2016-06-03T06:46:33 sebhub: untar patch ok to push? 2016-06-03T08:12:32 [git] 2321313 by Chris Johns: [arm: Fix printk warnings.] http://devel.rtems.org/changeset/2321313/rtems 2016-06-03T08:12:32 [git] 560db81 by Chris Johns: [libmisc/stackchk: Fix printk warnings.] http://devel.rtems.org/changeset/560db81/rtems 2016-06-03T08:14:23 *** mumptai has quit IRC 2016-06-03T08:38:01 *** Davidbrcz_ has joined #rtems 2016-06-03T08:39:06 *** Davidbrcz has quit IRC 2016-06-03T11:59:14 kiwichris_: Linux. I'm not sure exactly what other data you're looking for. 2016-06-03T13:00:38 *** gedare has joined #rtems 2016-06-03T13:00:39 *** ChanServ sets mode: +o gedare 2016-06-03T13:31:04 *** gedare has quit IRC 2016-06-03T16:22:22 *** mumptai has joined #rtems 2016-06-03T18:34:32 *** monstr has quit IRC 2016-06-03T23:01:33 *** dipupo has quit IRC 2016-06-04T00:11:07 darnir: the distro, version of the distro and the output, please put in a ticket as Joel suggested 2016-06-04T00:11:31 [trac] #2207/enhancement (closed) updated by Chris Johns (In [changeset:"d84e346b26017f021c1a7d5c8ad078c7264240ab/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2207 2016-06-04T00:11:32 [trac] #2415/defect (new) updated by Chris Johns (In [changeset:"d84e346b26017f021c1a7d5c8ad078c7264240ab/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2415 2016-06-04T00:11:32 [git] d84e346 by Chris Johns: [libmisc/untar: Support directory create and overwrites. Share the common code. Support creating directories for files with a path depth greater than 1. Some tar files can have files with a path depth greater than 1 and no directory entry in the tar file to create a directory. Support overwriting existing files and directories 2016-06-04T00:11:32 failing in a similar way to ...] http://devel.rtems.org/changeset/d84e346/rtems 2016-06-04T03:53:52 *** mumptai has quit IRC 2016-06-04T04:06:27 *** mumptai has joined #rtems 2016-06-04T06:35:41 *** guillaum1_mobile has joined #rtems 2016-06-04T06:38:33 *** mumptai has quit IRC 2016-06-04T06:42:49 *** guillaume_mobile has joined #rtems 2016-06-04T06:46:24 *** guillaum1_mobile has quit IRC 2016-06-04T06:55:00 *** guillaume_mobile has quit IRC 2016-06-04T07:12:17 *** mumptai has joined #rtems 2016-06-04T07:43:56 *** ita has joined #rtems 2016-06-04T09:55:43 [trac] SOCIS/2015 edited by salpha (empty comment) http://devel.rtems.org/wiki/SOCIS/2015 2016-06-04T09:56:52 [trac] SOCIS/2015 edited by salpha (empty comment) http://devel.rtems.org/wiki/SOCIS/2015 2016-06-04T22:13:03 *** mumptai has quit IRC 2016-06-04T23:51:17 *** ita is now known as ita|zzz 2016-06-05T03:37:21 *** indy has quit IRC 2016-06-05T03:45:24 *** indy has joined #rtems 2016-06-05T08:46:41 *** mumptai has joined #rtems 2016-06-05T10:53:37 *** mumptai has quit IRC 2016-06-05T10:54:18 *** mumptai has joined #rtems 2016-06-05T11:00:01 *** danieli_ has quit IRC 2016-06-05T11:00:58 *** danieli has joined #rtems 2016-06-05T11:30:13 *** danieli has quit IRC 2016-06-05T11:30:14 *** danieli has joined #rtems 2016-06-05T12:14:57 *** ita|zzz is now known as ita 2016-06-05T16:21:46 *** ita_ has joined #rtems 2016-06-05T16:23:48 *** ita has quit IRC 2016-06-05T20:40:03 *** Boquete has quit IRC 2016-06-05T20:49:49 *** Boquete has joined #rtems 2016-06-05T22:20:50 *** danieli has quit IRC 2016-06-05T22:58:11 *** danieli has joined #rtems 2016-06-05T23:07:17 *** mumptai has quit IRC