2015-03-23T00:49:21 *** dr__house is now known as dr__house|away 2015-03-23T01:08:04 *** monstr has joined #rtems 2015-03-23T01:18:43 *** waleedziad has quit IRC (Ping timeout: 250 seconds) 2015-03-23T01:25:01 [git] 8531776 by Chris Johns: [rtemstoolkit: Add an array operator to return recs in a section.] http://devel.rtems.org/changeset/8531776/rtems-tools 2015-03-23T01:25:01 [git] 29ad92c by Chris Johns: [trace-linker: Add options, names, enables, and triggers. Move the options to a section so an option can be a single line. This gives the user the ability to localise specific configurations in a top level configuration file. Provide support for names, enables and triggers. Names is an array of names of the trace functions. The 2015-03-23T01:25:01 table is sorted and y...] http://devel.rtems.org/changeset/29ad92c/rtems-tools 2015-03-23T01:50:34 *** kiwichris has quit IRC () 2015-03-23T02:04:36 *** jrcatbagan has quit IRC (Quit: nah nah nah nah, nah nah nah nah, hey hey hey, goodbye) 2015-03-23T02:38:12 *** ketul has joined #rtems 2015-03-23T03:17:17 *** hermann_19829 has joined #rtems 2015-03-23T03:37:20 *** andrei__ has joined #rtems 2015-03-23T03:37:35 *** andrei__ is now known as andrei_ 2015-03-23T03:37:44 *** andrei_ has joined #rtems 2015-03-23T03:37:44 *** andrei_ has joined #rtems 2015-03-23T03:46:30 *** Davidbrcz has joined #rtems 2015-03-23T03:49:36 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-23T03:50:21 *** ketul has joined #rtems 2015-03-23T03:54:35 *** Davidbrcz has quit IRC (Ping timeout: 252 seconds) 2015-03-23T04:04:00 is there any way to download rtems sourcecode(https://git.rtems.org/rtems/) over http 2015-03-23T04:04:21 my http_proxy won t let git clone for some reason 2015-03-23T04:08:26 *** Hesham has joined #rtems 2015-03-23T04:28:46 *** monstr has quit IRC (Ping timeout: 256 seconds) 2015-03-23T04:41:19 *** monstr has joined #rtems 2015-03-23T05:35:06 *** mkhoory-2 has joined #rtems 2015-03-23T05:38:46 *** mkhoory has quit IRC (Ping timeout: 264 seconds) 2015-03-23T05:39:27 *** mkhoory-2 has quit IRC (Ping timeout: 252 seconds) 2015-03-23T05:39:27 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-23T05:40:20 *** ketul has joined #rtems 2015-03-23T06:11:53 *** dibya has joined #rtems 2015-03-23T06:14:22 any one interested to be a co-mentor for the project "heart beat measure with arduino and sending it to drupal website". it is a gsoc15 proposal for drupal organisation.they need a comentor as they are not much excellent in arduino coding part 2015-03-23T06:15:24 any one interested to be a co-mentor for the project "heart beat measure with arduino and sending it to drupal website". it is a gsoc15 proposal for drupal organisation.they need a comentor as they are not much excellent in arduino coding part interested mentors should mail to dibya.dev94@gmail.com 2015-03-23T06:18:57 *** hermann_ has joined #rtems 2015-03-23T06:18:57 *** hermann_19829 has quit IRC (Read error: Connection reset by peer) 2015-03-23T06:24:59 Any one interested to be a co-mentor for the project "heart beat measure with arduino and sending it to drupal website". it is a google summer code 15 proposal for drupal organisation.they need a co-mentor as they are not much excellent in arduino coding part. Any one interested should mail to dibya.dev94@gmail.com or jingsheng.wang@insready.com 2015-03-23T06:32:19 *** dibya has quit IRC (Ping timeout: 246 seconds) 2015-03-23T07:18:21 *** Hesham has quit IRC (Quit: Leaving.) 2015-03-23T08:11:14 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-23T08:15:51 *** ragunath has joined #rtems 2015-03-23T08:22:13 *** ragunath has quit IRC (Ping timeout: 246 seconds) 2015-03-23T08:31:34 *** ragunath has joined #rtems 2015-03-23T08:48:31 *** Hesham has joined #rtems 2015-03-23T08:51:28 *** krohini has joined #rtems 2015-03-23T08:54:26 *** gedare has joined #rtems 2015-03-23T08:54:27 *** ChanServ sets mode: +o gedare 2015-03-23T09:06:26 hello, gedare ! 2015-03-23T09:07:41 [trac] GSoC/2015 edited by anandkp92 (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-23T09:07:49 I have been looking through code, and most synchrnoziation primitives look like this 2015-03-23T09:07:50 void _CORE_spinlock_Initialize( CORE_spinlock_Control *the_spinlock, CORE_spinlock_Attributes *the_spinlock_attributes ) 2015-03-23T09:09:00 basically : a control and some attributes 2015-03-23T09:09:45 looking in posix/src/cond.c ( the posix part of condvar implementation) 2015-03-23T09:09:58 there are some objects that are passed as parameters 2015-03-23T09:13:00 DrJoel mentioned the aspet about not being able to lock on different level mutexes. So, I got some questions(which I can probably find an answer to myself, but might take a lot longer0 2015-03-23T09:13:50 When refactoring the posix implementation to SuperCore: does the score have all the capabilities required to call the posix API? 2015-03-23T09:15:49 I'm not entirely sure, but this https://github.com/RTEMS/rtems/blob/master/cpukit/score/src/corebarrierrelease.c (line 28 - 34 ) looks like a callback mechanism, does that make sense? 2015-03-23T09:17:45 hi andrei_ 2015-03-23T09:18:46 yup 2015-03-23T09:19:17 not sure what you mean by "capabilities required" 2015-03-23T09:20:14 rtems has a single address space. if you have a function pointer you can invoke the pointed-to function from anywhere in the system. 2015-03-23T09:20:58 hm, I was thinking of references 2015-03-23T09:21:09 my understanding is that when you call an score function, it calls in itself 2015-03-23T09:21:18 the according posix api function 2015-03-23T09:21:31 I was aking if the score has access to the object on which to call the function 2015-03-23T09:21:35 asking 2015-03-23T09:24:04 I'm trying to figure if refactoring posix to supercore involves just writting a correct score api that calls posix accordingly or if I would need to expose aditional data structures/functions 2015-03-23T09:36:27 Hi Gedare! 2015-03-23T09:37:11 What is the current work for the raspberry pi 2 support? 2015-03-23T09:41:42 *** monstr has quit IRC (Remote host closed the connection) 2015-03-23T09:47:10 hi krohini i just sent back email 2015-03-23T09:47:33 *** mkhoory has joined #rtems 2015-03-23T09:47:43 andrei_: normally, all "objects" should exist within the score layer, and be accessed through score-layer functions. 2015-03-23T09:48:16 the score should not call to the classic (rtems_) or posix layers. but, in this case, it might be best to have a call-out to do that 2015-03-23T09:48:40 because otherwise you would have some complicated mess to figure out the right way to invoke the underlying score sempahore object to do mutex lock/unlock in the condvar 2015-03-23T09:48:58 because the classic and posix layers have different options to configure their semaphores... 2015-03-23T09:55:04 gedare: It doesn't matter (from timing and behavior perspectives) whether _Thread_Dispatch was called from ISR_Handler or rtems_task_delete/ enable dispatch(), right? 2015-03-23T09:55:34 Assuming that dispatch_necessary = 1 2015-03-23T09:59:25 what do you mean timing/behavior... 2015-03-23T10:00:49 *** DrJoel has joined #rtems 2015-03-23T10:00:49 *** DrJoel has joined #rtems 2015-03-23T10:00:49 *** ChanServ sets mode: +o DrJoel 2015-03-23T10:01:39 gedare: hi gedare 2015-03-23T10:02:01 from the point of the call to _Thread_Dispatch() and _Thread_Enable_dispatch() the path through the code is pretty similar, so yes the behavior will be alike.. 2015-03-23T10:02:04 gedare: I am trying to get the timing of _Thread_Dispatch (in cycles) 2015-03-23T10:02:04 gedare: We are about to leave for Birmingham for the appointment. If you have time to bump the newlib hash, that would be appreciated. I was testing against gcc, binutils, gdb ,and newlib head 2015-03-23T10:02:16 I can then start a big build sweep when I get back. 2015-03-23T10:02:39 gedare: Thanks, that's what I need to know. 2015-03-23T10:02:47 DrJoel: ef23a12ff8f840cc571e47870cd5f4ad6bca4553 ? 2015-03-23T10:02:48 Hesham: tm26 or tm27 does some of that. Manually sets up cases and then calls it. May be worth splitting into multiple cases if there are more now than then 2015-03-23T10:03:12 Yep. That's the one that updates newlib.hin. 2015-03-23T10:03:33 I only have 4 warnings left and one false positive if I ignore the m32c 2015-03-23T10:03:47 I haven't looked at the avr but it even says 0 warnings 2015-03-23T10:04:09 thanks. about to leave. appointment in 2.5 hours 100 miles away 2015-03-23T10:04:15 thanks again 2015-03-23T10:04:17 *** DrJoel has quit IRC (Client Quit) 2015-03-23T10:04:53 DrJoel: Yeah I provided the functionality to execute tm27 interrupt, the problem is that this line is returning with garbage instead of the next thread to switch to https://github.com/RTEMS/rtems/blob/master/testsuites/tmtests/tm27/task1.c#L195 2015-03-23T10:05:11 hi ragunath. 2015-03-23T10:05:52 Hesham: you may like to ask on devel ml if there might be a bug in the test now. 2015-03-23T10:06:05 the scheduling/thread code has a lot of churn 2015-03-23T10:06:10 gedare: i have added my proposal to wiki.. if possible pls review and give your comments 2015-03-23T10:06:10 *** andrei_ has quit IRC (Ping timeout: 246 seconds) 2015-03-23T10:06:31 ragunath: i got a quick skim of it in melange yesterday. overall it is good, but may be a bit ambitious 2015-03-23T10:07:09 which is fine, as long as expectations are set at the right level. 2015-03-23T10:08:30 gedare: I know, but I needed to get timing analysis quickly, and already made a workaround to get (for _Thread_Dispatch). I will also post about this issue once I feel comfortable with the performance analysis part of the paper. 2015-03-23T10:08:34 also, you should identify more explicit milestones where you plan to submit code "upstream" to devel ml to get included in rtems.git 2015-03-23T10:11:04 ok Hesham. you might consider instrumenting at Thread_Do_dispatch() if you can 2015-03-23T10:11:26 gedare: ok thank you very much... i will fine tune the proposal... still i am not sure how much effort it will take for lwIP integration 2015-03-23T10:12:14 ragunath: me either. i view this as a parallel track though that you can work on 2015-03-23T10:12:42 developing drivers for peripherals can be done in parallel to integrating the network stacks and trying to get RSB to build them. 2015-03-23T10:14:31 gedare: hmmm ... i downloaded the lwIP code marcos provided.... have to check them to get an idea 2015-03-23T10:14:51 sure don't worry too much about it 2015-03-23T10:14:57 we often re-scope projects after accepting them. 2015-03-23T10:15:12 and sometimes you find some things are much easier, other much harder, than you originally anticipated. 2015-03-23T10:15:53 gedare: yeah i have experienced that before :) 2015-03-23T10:17:15 gedare: you were able to view the proposal on melange? if we edit it do you get mail or any notification? 2015-03-23T10:18:55 ragunath: we don't get notifications, we can see the history if we look 2015-03-23T10:19:31 ragunath: we'll end up re-reading the proposals carefully after the submissions close. right now we want to make sure you get the proposal in the system, and give some quick feedback at a high-level 2015-03-23T10:19:31 *** hermann_ has quit IRC (Read error: Connection reset by peer) 2015-03-23T10:19:44 *** hermann_19829 has joined #rtems 2015-03-23T10:23:37 gedare: oh ok ...i will come up with details about when to submit code to rtems.git 2015-03-23T10:24:20 great 2015-03-23T10:31:17 *** ragunath is now known as ragunath|away 2015-03-23T10:38:12 *** dr__house|away is now known as dr__house 2015-03-23T10:40:53 [trac] GSoC/2015 edited by gadia (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-23T10:42:27 [git] 6523533 by Gedare Bloom: [Bump newlib git head] http://devel.rtems.org/changeset/6523533/rtems-source-builder 2015-03-23T10:55:01 Antti is trying to reply to the mailing list (he's not a member) however his reply needs an approval. 2015-03-23T10:55:37 verm__: can you "let it through"? 2015-03-23T10:58:14 *** mkhoory has quit IRC (Ping timeout: 252 seconds) 2015-03-23T11:05:40 *** ragunath|away has quit IRC (Ping timeout: 246 seconds) 2015-03-23T11:11:33 Hesham: is there a qualitative difference between using rump kernels via POSIX and using them via a custom hypercall if? 2015-03-23T11:13:15 gedare: Antti said that the hypercall implementation is small enough not to consider POSIX-interface stability. It will also be more efficient for performance. 2015-03-23T11:13:34 *** krohini has quit IRC (Ping timeout: 245 seconds) 2015-03-23T11:13:35 yes i wondered about that. 2015-03-23T11:13:35 Both can be ported anyway. 2015-03-23T11:13:44 ok you should consider both then 2015-03-23T11:14:17 But the issue is about how the integration would take place, and whether it worth the effort 2015-03-23T11:15:11 i wonder if it can all be done at "application" level, maybe it makes sense to build more of an application packager like https://github.com/alanc98/rki 2015-03-23T11:15:20 or to integrate something into RSB 2015-03-23T11:15:24 Most likely I'd consider both. 2015-03-23T11:16:36 It will be easier for me to work on the application level. 2015-03-23T11:17:27 We can work on both application and cpukit variants 2015-03-23T11:18:55 The rumpkernel project can generate a totally independent rump kernel library (that assumes the platform gives the hypercall implementation). 2015-03-23T11:19:37 *** hermann_19829 has quit IRC () 2015-03-23T11:20:21 What I suggested is to deal with rumpkernel project as a separate module/package, build it from RSB and we can then build the rump kernel library (that depends on the provided RTEMS/hypercall) to be linked with RTEMS. 2015-03-23T11:36:36 gedare: Should I update this page? https://devel.rtems.org/wiki/Developer/Projects/Open/MicroblazePort 2015-03-23T11:36:37 Some of my colleagues here needs the microblaze port. I have hello world working, ticker just needs the timer driver. 2015-03-23T11:42:03 *** jrcatbagan has joined #rtems 2015-03-23T11:42:21 *** Davidbrcz has joined #rtems 2015-03-23T11:42:28 Hi gedare 2015-03-23T11:42:39 hi Davidbrcz 2015-03-23T11:42:44 I think you were right 2015-03-23T11:42:50 there is something wrong with my board 2015-03-23T11:42:54 http://comments.gmane.org/gmane.comp.hardware.opencores.leon-sparc/17159 2015-03-23T11:43:31 tl;dr : get a new version of the lib and flash the FPGA onboard 2015-03-23T11:43:34 Hesham: sure if there is some progress go ahead and update. also I think the approach you suggest for rump kernels makes sense. if we need to add a 'rump' subdir in cpukit that is fine 2015-03-23T11:43:48 Davidbrcz: that's always a possibility when dealing with hardware. 2015-03-23T11:45:28 yep =/ 2015-03-23T11:46:39 [trac] Developer/Projects/Open/MicroblazePort edited by heshamelmatary (empty comment) http://devel.rtems.org/wiki/Developer/Projects/Open/MicroblazePort 2015-03-23T11:58:31 *** krohini has joined #rtems 2015-03-23T12:02:04 *** Hesham has quit IRC (Ping timeout: 255 seconds) 2015-03-23T12:14:26 Hello! Just curious, is there an inherent latency between posting a message at devel and seeing archived? I posted a message 14 hours ago and wanted to confirm that it was sent by checking the archives but with no luck. Please excuse me if this is coming off as a nuisance. 2015-03-23T12:15:31 hi jrcatbagan 2015-03-23T12:16:23 i see messages i posed just half an hour ago 2015-03-23T12:16:28 probably your email didn't go through for some reason 2015-03-23T12:17:06 hmmm, ok I'll try to resend it 2015-03-23T12:17:08 thanks! 2015-03-23T12:19:45 *** anandkp92 has joined #rtems 2015-03-23T12:20:31 *** krohini has quit IRC (Ping timeout: 255 seconds) 2015-03-23T12:21:58 *** krohini has joined #rtems 2015-03-23T12:22:50 gedare: I was trying to post my proof of completing the hello world for gsoc. I am currently in process of finishing my proposal. I am interested in the Beagle BSP improvements, or more specifically booting withou u-boot. Do you have any adivce that I should keep in mind? Thanks! 2015-03-23T12:23:19 ping beng-nl 2015-03-23T12:24:14 thanks again! 2015-03-23T12:24:49 jrcatbagan: the BB expert in the house is beng-nl so if he gets around to seeing it maybe he'll have some advice for you 2015-03-23T12:25:11 make sure you do the steps suggested on the beagle bsp project page about getting BB running 2015-03-23T12:26:37 *** ragunath has joined #rtems 2015-03-23T12:53:28 *** ragunath has quit IRC (Ping timeout: 246 seconds) 2015-03-23T12:54:40 *** ragunath has joined #rtems 2015-03-23T13:02:36 *** Hesham has joined #rtems 2015-03-23T13:07:38 *** ketul has joined #rtems 2015-03-23T14:04:18 *** commander has quit IRC (*.net *.split) 2015-03-23T14:07:18 *** rtemsLogger has joined #rtems 2015-03-23T14:10:28 *** commander has joined #rtems 2015-03-23T14:17:06 *** Davidbrcz has quit IRC (Ping timeout: 265 seconds) 2015-03-23T14:20:56 *** Hesham has quit IRC (Ping timeout: 272 seconds) 2015-03-23T14:21:46 *** andrei__ has joined #rtems 2015-03-23T14:31:18 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-23T14:37:30 *** ragunath has quit IRC () 2015-03-23T14:43:25 *** hermann_19829 has joined #rtems 2015-03-23T14:44:22 jrcatbagan: did you ever figure out why your email didn't go through? 2015-03-23T14:56:25 gedare: yes, it was my mistake. haven't used my gmail extensively before so all emails from devel were hidden from the default inbox 2015-03-23T14:57:54 hence why I didn't get the devel list confirmation 2015-03-23T14:58:06 sorry about the inconvenience 2015-03-23T15:07:26 okay great you got it cleared up 2015-03-23T15:10:20 as far as your project, you'll have to do some investigating to look for alternatives to try out. 2015-03-23T15:10:39 *** krohini has quit IRC (Ping timeout: 245 seconds) 2015-03-23T15:34:41 jrcatbagan: got it now thanks 2015-03-23T15:36:26 *** krohini has joined #rtems 2015-03-23T15:57:30 *** gedare has quit IRC (Quit: Leaving) 2015-03-23T16:02:22 *** krohini has quit IRC (Ping timeout: 256 seconds) 2015-03-23T16:19:16 *** anandkp92 has quit IRC (Ping timeout: 246 seconds) 2015-03-23T16:23:11 *** kiwichris has joined #rtems 2015-03-23T16:38:34 *** Davidbrcz has joined #rtems 2015-03-23T16:46:06 *** hermann_19829 has left #rtems ("http://quassel-irc.org - Chat comfortably. Anywhere.") 2015-03-23T17:16:23 *** waleedziad has joined #rtems 2015-03-23T17:30:25 *** waleedziad has quit IRC (Ping timeout: 244 seconds) 2015-03-23T18:13:25 *** Davidbrcz has quit IRC (Ping timeout: 256 seconds) 2015-03-23T18:34:22 *** andrei__ has quit IRC (Ping timeout: 246 seconds) 2015-03-23T19:37:04 [trac] Developer/Tracing/Trace_Linker edited by chrisj (Update.) http://devel.rtems.org/wiki/Developer/Tracing/Trace_Linker 2015-03-23T19:39:49 [trac] Developer/Tracing/Trace_Linker edited by chrisj (Minor fixes.) http://devel.rtems.org/wiki/Developer/Tracing/Trace_Linker 2015-03-23T19:40:21 [trac] Developer/Tracing/Trace_Linker edited by chrisj (More TOC detail.) http://devel.rtems.org/wiki/Developer/Tracing/Trace_Linker 2015-03-23T20:31:01 *** krohini has joined #rtems 2015-03-23T21:40:39 *** krohini has quit IRC (Ping timeout: 245 seconds) 2015-03-23T22:09:52 *** krohini has joined #rtems 2015-03-23T22:30:59 *** ketul has joined #rtems 2015-03-23T22:38:41 [git] 1703041 by Chris Johns: [rtems-tld: Add lock and buffer allocator support to generators. Generators can control a lock and buffer allocation so a single alloc can happen and a lock released.] http://devel.rtems.org/changeset/1703041/rtems-tools 2015-03-23T23:45:20 *** dr__house is now known as dr__house|away 2015-03-23T23:59:06 *** mkhoory has joined #rtems 2015-03-24T00:00:34 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-24T00:30:28 *** jrcatbagan has quit IRC (Quit: nah nah nah nah, nah nah nah nah, hey hey hey, goodbye) 2015-03-24T00:31:29 *** krohini has quit IRC (Ping timeout: 245 seconds) 2015-03-24T01:30:35 *** monstr has joined #rtems 2015-03-24T01:45:55 *** krohini has joined #rtems 2015-03-24T01:48:07 *** kiwichris has quit IRC () 2015-03-24T02:07:25 *** krohini has quit IRC (Ping timeout: 252 seconds) 2015-03-24T02:31:55 *** kiwichris has joined #rtems 2015-03-24T03:52:44 *** Charith has joined #rtems 2015-03-24T04:07:05 *** Charith has quit IRC (Quit: Page closed) 2015-03-24T04:16:28 *** Hesham has joined #rtems 2015-03-24T04:27:04 *** guerby has quit IRC (Ping timeout: 272 seconds) 2015-03-24T04:29:17 *** guerby has joined #rtems 2015-03-24T05:49:35 *** mkhoory has quit IRC (Ping timeout: 264 seconds) 2015-03-24T06:19:41 *** noselasd has joined #rtems 2015-03-24T06:22:11 *** ketul has joined #rtems 2015-03-24T07:39:54 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-24T07:43:00 [trac] Developer/Projects/Open/ImproveRaspberryPiBSP edited by gtament (empty comment) http://devel.rtems.org/wiki/Developer/Projects/Open/ImproveRaspberryPiBSP 2015-03-24T08:12:00 *** ragunath has joined #rtems 2015-03-24T08:18:20 [git] 5ce776e by Joel Sherrill: [cpukit/libmisc/shell/main_df.c: Eliminate printf() format warning] http://devel.rtems.org/changeset/5ce776e/rtems 2015-03-24T08:18:20 [git] 5cd910b by Joel Sherrill: [cpukit/libmisc/shell/hexdump-conv.c: Eliminate printf() format warning] http://devel.rtems.org/changeset/5cd910b/rtems 2015-03-24T08:18:20 [git] 6442d54 by Joel Sherrill: [tools/build/rtems-test-check: Support multiple include files] http://devel.rtems.org/changeset/6442d54/rtems 2015-03-24T08:18:20 [git] f061f2f by Joel Sherrill: [testsuites/testdata: Add sets to disable all jffs2 and mrfs tests] http://devel.rtems.org/changeset/f061f2f/rtems 2015-03-24T08:18:20 [git] 1d09732 by Joel Sherrill: [avrtest: Disable jffs2 and mrfs tests] http://devel.rtems.org/changeset/1d09732/rtems 2015-03-24T08:18:20 *** rtems-bot has quit IRC (Excess Flood) 2015-03-24T08:18:28 *** rtems-bot has joined #rtems 2015-03-24T08:27:53 *** ragunath has quit IRC (Quit: Page closed) 2015-03-24T08:34:58 *** ragunath has joined #rtems 2015-03-24T08:35:42 *** ragunath has left #rtems 2015-03-24T08:35:48 *** ragunath has joined #rtems 2015-03-24T08:36:37 [git] 2e9c3d5 by Sebastian Huber: [score: Add thread priority change handler Since the thread current priority change and thread queue requeue is performed in one critical section it is possible to simplify the thread queue requeue procedure. Add a thread queue agnostic thread priority change handler so that we are able to use alternative thread queue 2015-03-24T08:36:37 implementations. Update #2273.] http://devel.rtems.org/changeset/2e9c3d5/rtems 2015-03-24T08:36:37 [git] 57947f1 by Sebastian Huber: [score: Add thread lock Update #2273.] http://devel.rtems.org/changeset/57947f1/rtems 2015-03-24T08:36:37 [git] 3134eb8 by Sebastian Huber: [score: Use a dedicated ISR lock for thread queues This makes it possible to use the standard thread queues in combination with objects using fine grained locking. There is no change for uni-processor configurations. Update #2273.] http://devel.rtems.org/changeset/3134eb8/rtems 2015-03-24T08:36:37 [git] 4054c91 by Sebastian Huber: [score: Add scheduler acquire/release This is currently a global lock for all scheduler instances. It should get replaced with one lock per scheduler instance in the future. Update #2273.] http://devel.rtems.org/changeset/4054c91/rtems 2015-03-24T08:36:38 [git] ad57a59 by Sebastian Huber: [score: Fix MrsP ISR/task sychronization] http://devel.rtems.org/changeset/ad57a59/rtems 2015-03-24T08:36:38 [git] a54866e by Sebastian Huber: [monitor: Fix notepad dump] http://devel.rtems.org/changeset/a54866e/rtems 2015-03-24T08:36:39 [git] 4f10ed6 by Sebastian Huber: [smptests/smpfatal08: Workaround for QorIQ BSPs] http://devel.rtems.org/changeset/4f10ed6/rtems 2015-03-24T08:50:53 *** gedare has joined #rtems 2015-03-24T08:50:53 *** ChanServ sets mode: +o gedare 2015-03-24T09:16:22 *** ragunath has quit IRC (Ping timeout: 246 seconds) 2015-03-24T09:28:59 [git] 67e7ea5 by Sebastian Huber: [Special case for ] http://devel.rtems.org/changeset/67e7ea5/rtems-libbsd 2015-03-24T09:29:00 [git] 20b7d3d by Sebastian Huber: [Update due to Newlib etc. changes] http://devel.rtems.org/changeset/20b7d3d/rtems-libbsd 2015-03-24T09:29:00 [git] f661c79 by Sebastian Huber: [SLEEP(9): Import from FreeBSD] http://devel.rtems.org/changeset/f661c79/rtems-libbsd 2015-03-24T09:29:00 [git] 53b03a1 by Sebastian Huber: [SLEEP(9): Port to RTEMS] http://devel.rtems.org/changeset/53b03a1/rtems-libbsd 2015-03-24T09:29:00 [git] 9975839 by Sebastian Huber: [SLEEPQUEUE(9): Import from FreeBSD] http://devel.rtems.org/changeset/9975839/rtems-libbsd 2015-03-24T09:29:00 [git] 8475e7a by Sebastian Huber: [SLEEPQUEUE(9): Port to RTEMS] http://devel.rtems.org/changeset/8475e7a/rtems-libbsd 2015-03-24T09:29:00 [git] 50553bc by Sebastian Huber: [CONDVAR(9): Use FreeBSD implementation] http://devel.rtems.org/changeset/50553bc/rtems-libbsd 2015-03-24T09:29:01 [git] 5bcdc25 by Sebastian Huber: [smp01: New test] http://devel.rtems.org/changeset/5bcdc25/rtems-libbsd 2015-03-24T09:57:16 *** hermann_19829 has joined #rtems 2015-03-24T10:03:48 *** andrei__ has joined #rtems 2015-03-24T10:04:07 *** krohini has joined #rtems 2015-03-24T10:05:37 *** sgworks has joined #rtems 2015-03-24T10:06:39 good morning 2015-03-24T10:06:49 *** sgadia has joined #rtems 2015-03-24T10:08:36 good morning 2015-03-24T10:11:02 hi sgworks 2015-03-24T10:11:10 Hello Sir 2015-03-24T10:11:15 I wanted to discuss on like what aspects of multicore should we be carefull of while solving the problem of nested mutexes 2015-03-24T10:13:07 sgworks: you might like to read about SMP locking a bit here https://devel.rtems.org/wiki/Developer/SMP#SMPLocks 2015-03-24T10:13:51 the basic problem i anticipate is that lock scalability becomes a problem, as you get more threads executing, plus finer-grained locking, which means more lock acquire/release 2015-03-24T10:15:09 also we have to consider the implications of potential parallelism/contention on the locks 2015-03-24T10:15:29 iterating data structures in order to acquire/release could become a serious bottleneck 2015-03-24T10:17:04 the primary task for the project is to get it working correctly for uniprocessor, but we also should be forward-thinking about the situation of nested mutexes in multicore also. 2015-03-24T10:17:59 Yes True that 2015-03-24T10:18:12 *** ragunath has joined #rtems 2015-03-24T10:20:53 *** krohini has quit IRC (Ping timeout: 246 seconds) 2015-03-24T10:23:02 gedare: Okay, I will skim over MCS locking aspects 2015-03-24T10:23:26 good, and for smp locking you should also look into bjorn brandenburgs work 2015-03-24T10:23:33 for real-time considerations 2015-03-24T10:24:02 however that is a lot of material, so you need to prioritize and focus on getting your basic proposal written and uploaded to Melange 2015-03-24T10:25:56 I will also go through paper: Non-Scalable Locks are dangerous - http://pdos.csail.mit.edu/6.828/2014/readings/linux-lock.pdf by Boyd-wickizer et al. 2015-03-24T10:27:09 Due to shortage of time what should I mention in proposal regarding multicore aspect of project. 2015-03-24T10:27:31 apart from the solution mentioned over mail 2015-03-24T10:29:33 *** dr__house|away is now known as dr__house 2015-03-24T10:30:24 de3scribe the multicore aspects as research and optional "stretch goals" 2015-03-24T10:30:45 the main contribution for your project is to produce a robust implementation for uniprocessor 2015-03-24T10:34:55 hello 2015-03-24T10:35:08 hi 2015-03-24T10:36:24 gedare: did you have any time to look over what I changed? 2015-03-24T10:36:29 not yet. 2015-03-24T10:36:33 ah, no worries 2015-03-24T10:37:51 gedare: Thanks a lot. 2015-03-24T10:38:19 sure sgworks i look forward to seeing your proposal. 2015-03-24T10:46:19 [trac] GSoC/2015 edited by anandkp92 (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-24T10:53:38 *** ketul has joined #rtems 2015-03-24T10:57:16 *** sgworks has quit IRC (Ping timeout: 255 seconds) 2015-03-24T10:57:50 *** hermann_19829 has quit IRC (Ping timeout: 272 seconds) 2015-03-24T11:01:43 *** andrei__ has quit IRC (Ping timeout: 246 seconds) 2015-03-24T11:10:36 *** krohini has joined #rtems 2015-03-24T11:23:41 *** charith has joined #rtems 2015-03-24T11:25:38 *** charith has quit IRC (Client Quit) 2015-03-24T11:50:41 *** jrcatbagan has joined #rtems 2015-03-24T12:04:53 [trac] GSoC/2015 edited by krohini (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-24T12:06:41 [trac] GSoC/2015 edited by krohini (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-24T12:26:39 *** charith has joined #rtems 2015-03-24T12:31:52 [trac] Developer/Projects/Open/x86_64_BSP created http://devel.rtems.org/wiki/Developer/Projects/Open/x86_64_BSP 2015-03-24T12:32:38 [trac] Developer/Projects/Open/x86_64_BSP edited by amar (Fix references.) http://devel.rtems.org/wiki/Developer/Projects/Open/x86_64_BSP 2015-03-24T12:34:32 [trac] Developer/OpenProjects edited by amar (Add x86_64 BSP.) http://devel.rtems.org/wiki/Developer/OpenProjects 2015-03-24T12:34:58 *** anandkp92 has joined #rtems 2015-03-24T12:34:58 hi charith 2015-03-24T12:35:24 hi gedare! 2015-03-24T12:36:27 if you have any more questions feel free, and verm__ is Amar. 2015-03-24T12:37:14 I'm going through all the materials you all provided 2015-03-24T12:37:56 good. generally i think there is plenty to chew on. 2015-03-24T12:38:35 Ah how can I add myself to the tracking table. I think we are talking about this https://devel.rtems.org/wiki/GSoC/2015#StudentsSummerofCodeTrackingTable right? 2015-03-24T12:38:48 create an account 2015-03-24T12:38:51 then you can edit the page 2015-03-24T12:40:19 ah thanks. I should thank you all for giving me a lot of pointers to focus my research. 2015-03-24T12:40:57 we want all our students to be successful. 2015-03-24T12:41:09 in our experience it means getting the strongest proposals possible. 2015-03-24T12:43:37 *** monstr has quit IRC (Ping timeout: 244 seconds) 2015-03-24T12:44:51 I asked my mentor to join with melange. He said he will. 2015-03-24T12:45:21 I started to work on getting clang to built. Then I am planing to try compiling rtems with it 2015-03-24T12:46:06 I will definitely run into troubles and I'll ask here. 2015-03-24T12:46:52 ok, well you might want to try out the "waf branch" and building RTEMS that way, I think it will be the preferred path forward 2015-03-24T12:51:03 Ok sure. I'll come and clarify things as I proceed. Thanks for the guidance. 2015-03-24T13:03:04 gedare: hi 2015-03-24T13:03:48 hi ragunath 2015-03-24T13:04:22 gedare: i have a doubt on source builder 2015-03-24T13:05:11 gedare: how RSB decides how to build a package.... some may use configure some may not ? how RSB decides how to build a package? 2015-03-24T13:05:25 hmmm 2015-03-24T13:06:53 ragunath: the build instructions are encapsulated in scripts located atrtems-source-builder/source-builder/config 2015-03-24T13:07:39 the language for the scripts i think it should be documented in RSB docs 2015-03-24T13:07:46 excuse my poor english 2015-03-24T13:08:50 gedare: no, that's unforgivable! 2015-03-24T13:09:19 it's true though, i'm just a lazy native typer. 2015-03-24T13:09:37 gedare: got it ... thank you :) 2015-03-24T13:09:43 great. 2015-03-24T13:13:46 [trac] GSoC/2015 edited by Charith Eranga (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-24T13:14:07 hi 2015-03-24T13:14:21 hi anandkp92 2015-03-24T13:14:22 why was waf build introduced ? 2015-03-24T13:14:44 to decrease the build time so we can have continuous integration 2015-03-24T13:14:57 waf is better able to exploit thread-level parallelism than make 2015-03-24T13:15:07 ohh! 2015-03-24T13:15:27 if thats so why are we still sticking to make? 2015-03-24T13:15:55 i mean why is make still being used 2015-03-24T13:16:34 it also has automatic dependency discovery (the old build is tied to gcc...), dependencies based on signatures, kernel output, automatic rebuilds when the configuration is modified... 2015-03-24T13:16:57 anandkp92: rtems 5 has not been released yet 2015-03-24T13:16:59 *** andrei_ has joined #rtems 2015-03-24T13:18:52 mm i'll check it out then! 2015-03-24T13:21:04 anandkp92: replacing the build system is a large effort being done by volunteers. we're getting closer though. 2015-03-24T13:23:40 gedare: hope to be one of the volunteers soon then! 2015-03-24T13:23:51 or better yet paid 2015-03-24T13:25:41 *** charith_ has joined #rtems 2015-03-24T13:27:00 we use the waf build to build the build-set from the rsb for specific boards then, right? 2015-03-24T13:27:07 *** charith_ has quit IRC (Remote host closed the connection) 2015-03-24T13:27:22 *** charith_ has joined #rtems 2015-03-24T13:34:10 anandkp92: RSB is a tool that helps to build software from source. We use it to build compiler tool sets, and it can also build RTEMS itself, and some other packages such as Qemu 2015-03-24T13:34:55 anandkp92: the "waf build" of RTEMS is a modified version of the rtems.git repository that can be built using Waf 2015-03-24T13:35:16 I don't know that RSB knows how to build the "waf build" of RTEMS or not. 2015-03-24T13:35:30 You still require the tools built with RSB in order to build RTEMS using Waf 2015-03-24T13:35:42 oh ok 2015-03-24T13:35:58 i follow you now 2015-03-24T13:36:04 *** andrei_ has quit IRC (Ping timeout: 246 seconds) 2015-03-24T13:36:08 thanks! 2015-03-24T13:43:23 *** charith_ has quit IRC (Remote host closed the connection) 2015-03-24T13:45:37 *** andrei_ has joined #rtems 2015-03-24T13:52:38 *** antgreen has quit IRC (Ping timeout: 272 seconds) 2015-03-24T13:57:34 *** charith_ has joined #rtems 2015-03-24T13:57:43 *** charith has quit IRC (Ping timeout: 255 seconds) 2015-03-24T13:59:02 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-24T14:01:33 *** charith has joined #rtems 2015-03-24T14:02:13 *** charith_ has quit IRC (Ping timeout: 264 seconds) 2015-03-24T14:07:40 *** hermann_19829 has joined #rtems 2015-03-24T14:09:40 *** anandkp92 has quit IRC (Ping timeout: 246 seconds) 2015-03-24T14:13:37 *** charith_ has joined #rtems 2015-03-24T14:14:13 *** charith has quit IRC (Ping timeout: 264 seconds) 2015-03-24T14:15:09 *** krohini has quit IRC (Quit: ChatZilla 0.9.91.1 [Firefox 35.0.1/20150125222139]) 2015-03-24T14:15:35 *** anandkp92 has joined #rtems 2015-03-24T14:26:15 *** krisalpha has joined #rtems 2015-03-24T14:27:23 *** Hesham has quit IRC (Ping timeout: 264 seconds) 2015-03-24T14:30:40 *** anandkp92 has quit IRC (Ping timeout: 246 seconds) 2015-03-24T14:33:41 *** ragunath has left #rtems 2015-03-24T14:38:32 [trac] GSoC/2015 edited by gtament (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-24T14:44:02 *** charith has joined #rtems 2015-03-24T14:44:11 *** charith_ has quit IRC (Ping timeout: 264 seconds) 2015-03-24T14:53:04 andrei_: mp_support has to do with "multiprocessing support", so you should think on some better name for the callout 2015-03-24T14:54:35 hmm, let me open the proposal 2015-03-24T14:54:41 andrei_: also, it might make sense to initialize the condition variable with the callout, rather than providing it to each invocation of wait() and signal(). Instead you could store the callout and args in the Condition_variable_Control structure. 2015-03-24T14:56:09 *** Hesham has joined #rtems 2015-03-24T14:57:29 first 2015-03-24T14:57:44 CORE_mutex_API_mt_support_callout api_mutex_mt_support sounds better? 2015-03-24T14:57:51 multithreaded support 2015-03-24T14:57:58 keeping the same ... format 2015-03-24T15:00:46 it's not a strong reason, but I put the call-outs in each as I wasn't sure it s exactly the same call-out each time 2015-03-24T15:02:08 i'd just say CORE_mutex_API_callout probably. 2015-03-24T15:02:22 the callout intends to give access to the API layer of the mutex 2015-03-24T15:02:39 we pointed out the mp_support callouts just as an example of score callouts 2015-03-24T15:03:21 but these callouts for the condvar don't have much to do with multiprocessing, but rather how to block on the API-specific lock type 2015-03-24T15:03:24 right? 2015-03-24T15:03:58 yeah 2015-03-24T15:06:39 actually 2015-03-24T15:06:53 the initialize should contain the 2015-03-24T15:06:58 ifdef'ed callout 2015-03-24T15:07:00 right? 2015-03-24T15:09:25 hm, or actually not. I think "wait" needs the callout either way 2015-03-24T15:35:02 *** krisalpha has quit IRC (Quit: Leaving) 2015-03-24T16:01:23 *** Hesham has quit IRC (Ping timeout: 250 seconds) 2015-03-24T16:07:45 *** gedare has quit IRC (Ping timeout: 246 seconds) 2015-03-24T16:25:34 *** kiwichris has quit IRC (Ping timeout: 252 seconds) 2015-03-24T16:27:48 *** gedare has joined #rtems 2015-03-24T16:27:49 *** ChanServ sets mode: +o gedare 2015-03-24T16:29:26 gedare: hi, sorry for the ping, did you get my last replies? 2015-03-24T16:29:43 andrei_: yeah i have no more feedback at the moment. 2015-03-24T16:30:10 ah, ok, I'll investigate on my own, then. Thanks ! 2015-03-24T16:30:48 except, you don't want to use that ifdef 2015-03-24T16:31:12 well, maybe you do. 2015-03-24T16:31:22 but that would be for a different callout if you needed it 2015-03-24T16:31:45 RTEMS_MULTIPROCESSING is a way to use multiple RTEMS on different processors that communicate via something like RPC 2015-03-24T16:41:16 *** kiwichris has joined #rtems 2015-03-24T16:48:24 *** Schoumi has quit IRC (*.net *.split) 2015-03-24T16:49:50 *** Schoumi has joined #rtems 2015-03-24T16:51:54 I'm trying to build rtems with waf. Can anyone suggest a good guideline or a instruction source 2015-03-24T16:55:34 *** andrei_ has quit IRC (Ping timeout: 246 seconds) 2015-03-24T16:57:17 *** hermann_19829 has quit IRC () 2015-03-24T16:57:37 gedare, hi 2015-03-24T16:57:54 hey kiwichris 2015-03-24T16:58:07 Hmmm .... https://git.rtems.org/rtems-libbsd/tree/Makefile#n4 2015-03-24T16:58:16 This has no future 2015-03-24T16:58:53 i've never looked closely 2015-03-24T16:59:05 Good thing being it is generated. 2015-03-24T16:59:22 hi gadare 2015-03-24T16:59:53 This is a pain to move to the RSB with this file. 2015-03-24T17:00:19 I am not interested in hacks going in to just make this work. 2015-03-24T17:00:29 hi charith 2015-03-24T17:00:39 sure i could see that. 2015-03-24T17:00:53 'm trying to build rtems with waf. Can you suggest a good guideline or a instruction source 2015-03-24T17:01:22 I found some on the net. but it seems they are bit out dated. 2015-03-24T17:04:11 verm__ sent some emails to the devel and i think users mailing lists umm.. last month i think 2015-03-24T17:04:36 search for Waf branch - Overview [2/6] 2015-03-24T17:04:41 in february archives 2015-03-24T17:09:46 charith, have you cloned the repo ? 2015-03-24T17:09:51 thanks gadare 2015-03-24T17:10:08 yes I cloned the repo 2015-03-24T17:10:40 and cloned the waf from the google code also 2015-03-24T17:11:11 I tried to config>configure>build 2015-03-24T17:12:21 failed with an error about 'microtime'. I'll go thorough the mailing archive once again and try to find a clue 2015-03-24T17:13:54 charith, what was the command line ? 2015-03-24T17:14:05 *** waleedziad has joined #rtems 2015-03-24T17:14:40 u mean the error? 2015-03-24T17:15:39 error: static declaration of 'microtime' follows non-static declaration 2015-03-24T17:18:28 the version of rtems that the waf branch uses may need a specific version of the compiler tool set that is different than what you have built 2015-03-24T17:18:36 in fact i would bet that is the case 2015-03-24T17:19:10 oh the tool set is 4.11 2015-03-24T17:20:20 yep it is 2015-03-24T17:20:36 kiwichris: i've never successfully built any rtems networking stack. :) 2015-03-24T17:21:01 gedare, really, not even the one in the main tree ? 2015-03-24T17:21:02 rtems on waf is 4.10 thank you very much 2015-03-24T17:21:20 kiwichris: actually i've compiled it i guess, and run loopback, but never run any other net code 2015-03-24T17:21:23 to be more specific 2015-03-24T17:21:23 charith, I doubt that. 2015-03-24T17:21:32 charith: no you need 4.11 tools, but you need tools from a few months ago 2015-03-24T17:21:40 the 4.11 tools are not yet stable 2015-03-24T17:21:49 gedare, ah ok that makes sense. 2015-03-24T17:23:04 i gotta go. 2015-03-24T17:23:17 waf branch's rtems is 4.10.99.0 2015-03-24T17:23:51 do i need to use the exact version of toolchain 2015-03-24T17:24:45 ok gadare. thanks for your help. I'll find my way through. 2015-03-24T17:25:03 charith, you need 4.11, the version labelling is such that is the pre-4.11 do we do not actually have 4.11 out until it is released. 2015-03-24T17:27:14 *** gedare has quit IRC (Ping timeout: 246 seconds) 2015-03-24T17:27:57 kiwichris: is there a way to know the exact matching toolchain 2015-03-24T17:28:24 charith, waf should be on the latest 4.11 tools. 2015-03-24T17:29:00 charith, there have been some changes which may not have been picked. We need to wait for verm__ to let us know. 2015-03-24T17:29:25 I have the latest tools. I cloned it from git and built it myself. 2015-03-24T17:29:49 ok then. I'll wait for verm_ to respond. 2015-03-24T17:43:47 *** charith has quit IRC (Read error: Connection reset by peer) 2015-03-24T17:46:49 [trac] GSoC/2015 edited by waleedziad (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-24T18:09:38 oh, charith left 2015-03-24T18:28:28 https://waf.io/op/poll <- express your preferences there 2015-03-24T19:30:11 *** ketul has joined #rtems 2015-03-24T19:32:58 ita: doesn't work 2015-03-24T19:35:49 *** waleedziad has quit IRC (Remote host closed the connection) 2015-03-24T20:56:26 [trac] GSoC/2015 edited by Asuol (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-24T21:02:00 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-24T21:51:24 [trac] GSoC/2015 edited by jrcatbagan (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-24T21:54:36 *** jrcatbagan has quit IRC (Quit: nah nah nah nah, nah nah nah nah, hey hey hey, goodbye) 2015-03-24T22:03:49 *** hemant271990 has joined #rtems 2015-03-24T22:24:00 verm__, I sometimes feel elections work the same way 2015-03-24T22:36:56 *** mkhoory has joined #rtems 2015-03-24T23:03:59 heh 2015-03-24T23:33:43 *** dr__house is now known as dr__house|away 2015-03-25T00:07:24 *** hemant271990 has quit IRC (Quit: Page closed) 2015-03-25T01:12:18 *** kiwichris has quit IRC () 2015-03-25T01:22:29 *** monstr has joined #rtems 2015-03-25T02:27:50 *** hermann_19829 has joined #rtems 2015-03-25T03:12:26 *** andrei_ has joined #rtems 2015-03-25T03:50:37 *** Hesham has joined #rtems 2015-03-25T03:57:25 *** andrei_ has quit IRC (Ping timeout: 246 seconds) 2015-03-25T04:08:34 *** kiwichris has joined #rtems 2015-03-25T05:00:02 *** ketul has joined #rtems 2015-03-25T05:09:46 *** andrei__ has joined #rtems 2015-03-25T05:16:40 [trac] GSoC/2015 edited by abd_elhamid_ (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-25T05:47:22 [git] e7e86ba by Sebastian Huber: [smpmigration01: Delete timing dependent assert The cycles per runner depends too heavily on the hardware timing making the assert unreliable. Delete runner tasks before we print out the status.] http://devel.rtems.org/changeset/e7e86ba/rtems 2015-03-25T05:47:22 [git] 8aa967c by Sebastian Huber: [bsp/altera-cyclone-v: Typo] http://devel.rtems.org/changeset/8aa967c/rtems 2015-03-25T05:47:22 [git] 10a699d by Sebastian Huber: [testsupport: Generate unique worker names] http://devel.rtems.org/changeset/10a699d/rtems 2015-03-25T06:29:16 *** hermann_19829 has quit IRC (Ping timeout: 256 seconds) 2015-03-25T07:18:25 [git] cb4dd48 by Sebastian Huber: [sparc: Ensure interrupt service after ISR enable] http://devel.rtems.org/changeset/cb4dd48/rtems 2015-03-25T07:24:11 *** hermann_19829 has joined #rtems 2015-03-25T07:38:13 *** charith has joined #rtems 2015-03-25T07:40:43 \history 2015-03-25T07:41:08 *** charith has left #rtems 2015-03-25T07:43:55 *** stryx` has quit IRC (Ping timeout: 252 seconds) 2015-03-25T07:46:48 *** stryx`_ has joined #rtems 2015-03-25T08:15:43 [trac] GSoC/2015 edited by bremby (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-25T08:52:51 *** CaptHindisght has joined #rtems 2015-03-25T08:53:06 *** ragunath has joined #rtems 2015-03-25T09:04:18 *** gedare has joined #rtems 2015-03-25T09:04:18 *** ChanServ sets mode: +o gedare 2015-03-25T09:12:24 good morning 2015-03-25T09:14:20 gedare: good morning 2015-03-25T09:15:28 *** DrJoel has joined #rtems 2015-03-25T09:15:28 *** DrJoel has joined #rtems 2015-03-25T09:15:28 *** ChanServ sets mode: +o DrJoel 2015-03-25T09:18:06 DrJoel: aren't you on break this week? 2015-03-25T09:18:42 *** charith has joined #rtems 2015-03-25T09:18:46 more or less. It is the boys' spring break and I am trying to stay home as much as possible. Still answering email and tending long builds. 2015-03-25T09:18:56 ;) 2015-03-25T09:19:01 hi 2015-03-25T09:19:07 hi all 2015-03-25T09:19:13 I have gcc 4.9 building with the newlib snapshot. Now working on the 4.8 targets 2015-03-25T09:19:55 hi gadare 2015-03-25T09:19:55 Also think I have the sparc gdb 7.9 patches issue winding down. RSB applies patches in the OPPOSITE order they are listed in the file which is a problem for a patch set with over twenty patches 2015-03-25T09:20:52 hi charith 2015-03-25T09:21:09 @gadare it's about the yerterday issue 2015-03-25T09:21:09 charith: Error: "gadare" is not a valid command. 2015-03-25T09:22:04 DrJoel: good to hear. that is interesting about the patch ordering. i'm pretty sure 'sort' can reverse the list for you quickly though. did you see the u-boot replacement project? This could be an interesting spin-off for RTEMS 2015-03-25T09:22:10 charith: which issue? 2015-03-25T09:22:22 I couldn't contact verm__ for clarification. Any clues for solveing that issue 2015-03-25T09:24:00 It could be. I wonder if we want to do that from scratch or leverage micromonitor. I need to email Ed Sutter. It looks like his hosting has an issue. 2015-03-25T09:25:53 *** charith has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-25T09:26:06 Micromonitor is a really capable monitor and already supports a lot of hardware. If it is feasible (technical and license) to adopt as a baseline, we could grow more from it than starting from scratch 2015-03-25T09:26:26 *** starprince has joined #rtems 2015-03-25T09:29:43 yes do that. 2015-03-25T09:30:23 looks like i'll be serving on my first NSF panel this year 2015-03-25T09:31:22 and i'll wrap up my taxes today, not too shabby if i can get 'em done before april! 2015-03-25T09:31:28 ESA SoC has opened today, should I copy/paste the proposal or just through a link to the Google doc there? 2015-03-25T09:31:35 I'm looking back into using RTEMS on an STM32F design to control engines in autos (RCU) 2015-03-25T09:31:51 Hesham: i'd copy-paste and make a note at the top that is whaty ou did, and also link tot he gdoc is nice 2015-03-25T09:32:19 hi CaptHindisght, i think there has been some use of stm32f but i don't know in what software domain 2015-03-25T09:32:25 we are going through the usual arguments of RTOS vs just using libs 2015-03-25T09:32:46 congrats on the NSF panel gedare! 2015-03-25T09:32:57 * DrJoel still needs to wrap up his taxes. Just a few more items to enter. 2015-03-25T09:33:21 I asked this a few years ago bit what is the size of the kernel? IIRC ~4K 2015-03-25T09:33:24 well CaptHindisght if you don't know what you get, RTEMS is mature software with lots of features, and lots of users. 2015-03-25T09:33:30 compiled size? 2015-03-25T09:33:34 yes 2015-03-25T09:33:45 CaptHindisght: It is easy to start out not thinking you need much and then see that grow. 2015-03-25T09:33:51 there are various tricks that have to be played to shrink it down 2015-03-25T09:33:53 gedare: And should adjust the timeline according to ESA program? I think it's the same as GSoC. 2015-03-25T09:33:56 Hesham: how small did you get RTEMS? 2015-03-25T09:34:07 27KB 2015-03-25T09:34:13 yeah that sounds about right 2015-03-25T09:34:21 * DrJoel that's a full statically linked executable though, right? 2015-03-25T09:34:37 includes complete BSP, print, clock, etc? 2015-03-25T09:34:38 did you have printf? 2015-03-25T09:34:50 the kernel itself... hmm 2015-03-25T09:35:04 Yes, but printk wasn't working on though 2015-03-25T09:35:39 DrJoel: yeah thats what usually bites the kids 2015-03-25T09:36:02 even the small devices have 1MB of Flash and 192KB RAM 2015-03-25T09:36:03 we never really measure "kernel size". We tend to focus on the size of a statically linked executable that brings up a single task and then shuts the system down. Or hello world. 2015-03-25T09:36:11 ah ok 2015-03-25T09:36:40 gedare: No printf of course 2015-03-25T09:36:40 yeah. the base you might then add your application on. 2015-03-25T09:37:11 ticker took 27KB 2015-03-25T09:37:24 minimum was about 17KB 2015-03-25T09:38:04 and for a few $ more 320KBytes of SRAM 2015-03-25T09:39:37 You shouldn't need that kind of RAM for RTEMS. I have ticker down to < 8k I think of OS workspace even on SPARC. And that's got 4 tasks in it 2015-03-25T09:39:48 DrJoel: ever heard anyone try to use busybox in RTEMS? or seen the toybox alternative, which has a usable license? 2015-03-25T09:40:02 So that includes their (obviously minimal) stacks 2015-03-25T09:40:33 gedare: Not that I know of. Chris mentions that sometimes and another BSD project which automatically turns the commands into "my_main" type things which can be linked in. 2015-03-25T09:40:50 could be an interesting addition someday. 2015-03-25T09:41:00 ok i have to focus and figure out my problem for the day. 2015-03-25T09:41:19 https://devel.rtems.org/wiki/TBR/BSP/STM32_F4 just a template page 2015-03-25T09:41:28 yeah.. chris and I also noted that we have to worry about name collision and static data. It isn't enough just to fix main() 2015-03-25T09:41:55 CaptHindisght: Ask on the devel@ list. I know Embedded Brains was actually giving away an RTEMS Ruler which was an F4 board. :) 2015-03-25T09:42:13 will try it on this board in the next few days 2015-03-25T09:42:21 ok thanks for the tip 2015-03-25T09:43:03 I'm not locked into stm32 2015-03-25T09:44:21 I thought I actually had a picture of that to pass along but it isn't showing up on Facebook quickly. I did an interview for the embedded.fm podcast and they wanted a picture of my "messy desk". I thought I would make sure the RTEMS Ruler was visible 2015-03-25T09:44:26 there's been some active development with the tms570 lately, that might be an interesting target. 2015-03-25T09:44:57 my desk swallows pens 2015-03-25T09:44:59 * DrJoel notes the patches have to be sorted in numeric order. 2015-03-25T09:45:12 i.e. 0001, 0002, 0003? 2015-03-25T09:45:51 I think this thread may help with the reducing RTEMS size https://lists.rtems.org/pipermail/devel/2014-December/009635.html 2015-03-25T09:46:04 Yep. "ls -1 | sort -n -r" is the order I am trying now 2015-03-25T09:46:38 that should work 2015-03-25T09:46:45 but uhh 2015-03-25T09:46:49 Also in examples-v2 is a set of ticker variants which get smaller and smaller by tweaking configuration settings and the code itself to do the same thing with lighter methods 2015-03-25T09:47:03 can't you do ls -lr 2015-03-25T09:47:25 I wrote a script that prints the lines the RSB needs including the hash for each patch. :) 2015-03-25T09:48:18 CaptHindisght: for size, the BSP has a few things it can do to really help. (1) compile with -Os, (2) turn on per function/data item sections. The SPARC BSPs do this and some of the RTEMS tests dropped by almost 50% in size. 2015-03-25T09:49:05 DrJoel: ls -lr? 2015-03-25T09:49:30 er, -1r i guess :0 2015-03-25T09:51:37 *** starprince has quit IRC (Ping timeout: 246 seconds) 2015-03-25T09:52:47 *** CaptHindisght is now known as CaptHindsight 2015-03-25T09:53:04 (in case anyone is trying to find me) 2015-03-25T09:53:28 CaptHindisght: DrJoel: Yeah for Epiphany too, it reduced the size immensely https://github.com/heshamelmatary/rtems-epiphany/blob/squeezed-rtems/c/src/lib/libbsp/epiphany/parallella/make/custom/parallella.cfg 2015-03-25T09:54:47 Hesham: do you recall the footprint before and after? 2015-03-25T09:55:20 DrJoel: what is the value in carrying around the old stack, if we can get the new stack and lwIP integrated to RSB? 2015-03-25T09:55:37 kiwichris had some concerns about the new stack though, and the hacks it uses to get to compile. 2015-03-25T09:55:52 CaptHindsight: No, I did a lot of other source code hacking besides these flags to aggressively reduce the size. 2015-03-25T09:56:00 There isn't any except for those who don't have HW support in either lwip or new stack. 2015-03-25T09:56:32 I really would like to see RTEMS + (RSB network stack of old, LWIP, new) + (network services from stack) 2015-03-25T09:56:43 sorry the last is just using stack 2015-03-25T09:56:58 so RTEMS git itself doesn't have any networking in it. 2015-03-25T09:57:30 CaptHindsight: In general, ticker took about 2-3 MB before any optimization. When I used the flags + examples-v2 (ticker) + source code hacking, I managed to get ticker working with 27 KB size. 2015-03-25T09:58:02 Hesham: something was badly wrong at 2-3Mb.. unless that is the .exe with symbols size 2015-03-25T09:58:22 even that is too big 2015-03-25T09:58:24 It is normally max about 100k code/data size even when the BSP doesn't care about size and you have large stacks 2015-03-25T09:58:37 DrJoel: Yes, it was the normal file size. 2015-03-25T09:58:38 there must have been something getting compiled in as an object 2015-03-25T09:58:43 oh 2015-03-25T09:58:47 printf() itself is often between 10-20K added. 2015-03-25T09:58:57 you have to use the size binutils command to get an accurate measure 2015-03-25T09:59:07 printing the time requires gmtime() which is ~10K code 2015-03-25T09:59:20 sizes based on sparc 2015-03-25T09:59:35 Ah, it wasn't the size of the to-be-loaded sections only 2015-03-25T10:02:15 hello ! 2015-03-25T10:02:46 gedare, as we discussed yesterday. I looked over the _control struct, I found out that it usually contains atrributes and the thread queue 2015-03-25T10:03:09 epiphany-rtems4.11-size ticker.exe 2015-03-25T10:03:09 text data bss dec hex filename 2015-03-25T10:03:09 178176 1272 16598720 16778168 10003b8 ticker.exe 2015-03-25T10:03:38 Something it linked into the .text that shouldn't be 2015-03-25T10:04:06 and it's thread queue 2015-03-25T10:04:08 $ arm-rtems4.11-size ticker-executables/arm-stm32f* 2015-03-25T10:04:09 text data bss dec hex filename 2015-03-25T10:04:09 59596 1548 63976 125120 1e8c0 ticker-executables/arm-stm32f105rc-ticker.ralf 2015-03-25T10:04:09 59608 1548 129512 190668 2e8cc ticker-executables/arm-stm32f4-ticker.ralf 2015-03-25T10:04:36 DrJoel: Maybe some e-lib (epiphany library) old stuff that I copied during early porting 2015-03-25T10:04:42 But I am not using it anymore 2015-03-25T10:04:53 andrei__: if the condvar invokes a semaphore, then the condvar doesn't need the threadq, since the semaphore has it 2015-03-25T10:04:54 right? 2015-03-25T10:05:07 ah, I was going to ask if it makes sense to put the callout in atribute 2015-03-25T10:05:09 but it doesn't 2015-03-25T10:05:16 I figured out why 2015-03-25T10:05:41 think of the Control structure has a "Class" definition in OOP sense 2015-03-25T10:05:52 the instance of it is the object 2015-03-25T10:06:00 so you store whatever persistent state is needed in there 2015-03-25T10:06:25 both mutex and semaphore have the thread queue 2015-03-25T10:06:54 the key question is: do threads block in the condvar, or in a mutex invoked by the condvar? 2015-03-25T10:07:20 that has the IMFS, printf, and appears to have POSIX enabled. So a lot of stuff to trim plus -O2 so 7-11% likely to be lost there. 2015-03-25T10:07:44 they block on a CV and when released compete for the mutex 2015-03-25T10:07:57 * DrJoel walks away from the keyboard but is not logging off 2015-03-25T10:08:40 ok, hold on 2015-03-25T10:09:57 I was about to say 2015-03-25T10:09:59 that it locks on a mutex 2015-03-25T10:10:35 more reading must be done! 2015-03-25T10:10:36 I thought that's exactly why you'd want a condvar instead of a spin lock 2015-03-25T10:10:49 p.s. the freebsd condvar implementation recently appeared in libbsd 2015-03-25T10:11:05 https://git.rtems.org/rtems-libbsd/commit/?id=50553bcd8e47f80e7e41438b45f41ddecb607754 2015-03-25T10:11:28 It is an API level implementation of a Classic monitor from old school languages like Modula-2, Concurrent Pascal, etc. 2015-03-25T10:12:14 thank you DrJoel, next time perhaps you can wait until you're called on 2015-03-25T10:12:18 ;-) 2015-03-25T10:12:58 I'll look over it 2015-03-25T10:13:18 another question is with the callouts(specifically, the ifdef) 2015-03-25T10:13:35 So far I m certain that I need a callout, without ifdef guards 2015-03-25T10:14:03 well, you might, i just want you to know what it means 2015-03-25T10:14:14 if the RTEMS_MULTIPROCESSING case must be handled differently that is. 2015-03-25T10:14:38 By the way, I have my suspicions about whether RTEMS_MULTIPROCESSING even still works, i don't know that it gets much testing in development.... 2015-03-25T10:15:22 but maybe in that case, you need to provide a different call-out, to access a remote synchronization primitive 2015-03-25T10:15:31 I was about to say that if cv relies on semaphores 2015-03-25T10:15:45 semaphores already handle it differently for RTEMS_MULTIPROCESSING 2015-03-25T10:15:56 so I would need the ifdef only if I need to do something different "inside" 2015-03-25T10:15:59 makes sense? 2015-03-25T10:16:01 yup 2015-03-25T10:16:21 the RTEMS_MP is to deal with RPC-like semantics 2015-03-25T10:16:29 where operations act on remote objects 2015-03-25T10:16:33 in a separate instance of RTEMS 2015-03-25T10:16:59 I might be wrong, but I believe it's difficult to figure this out until I have a detailed implementation plan 2015-03-25T10:17:09 is it ok to mention it with an "if necessary" in the proposal? 2015-03-25T10:17:12 yep 2015-03-25T10:17:17 some fluidity is expected 2015-03-25T10:18:03 awesome ! I'll take a look over the cv implementation 2015-03-25T10:18:08 thanks! 2015-03-25T10:23:36 *** hermann_19829 has quit IRC (Ping timeout: 250 seconds) 2015-03-25T10:25:22 *** shipre has joined #rtems 2015-03-25T10:28:00 *** shipre has quit IRC (Client Quit) 2015-03-25T10:41:48 * Hesham Submitted the application to ESA 2015-03-25T10:52:06 good 2015-03-25T10:52:16 andrei__: submit your application to esa socis also please 2015-03-25T10:56:37 *** gwty has joined #rtems 2015-03-25T11:06:09 hello 2015-03-25T11:07:48 hi gwty 2015-03-25T11:11:34 hi gedare, I'm gowtham. I had introduced myself on the mailing list sometime ago 2015-03-25T11:12:04 hi gowtham. 2015-03-25T11:13:56 have you had a chance to follow-up on my response yet? 2015-03-25T11:14:51 gedare: I will, thanks for pointing out ! 2015-03-25T11:14:53 gwty: i believe i suggested that you do our GSoC Getting Started, and you should be working to figure out details of a potential project to work on 2015-03-25T11:15:04 *** andrei__ has quit IRC (Quit: Page closed) 2015-03-25T11:16:10 I'm sorry I haven't been able to reply to the mail. I am working on it. 2015-03-25T11:17:57 ok. are you having any troubles? 2015-03-25T11:18:35 I'm having trouble in getting RTEMS to compile properly on my computer 2015-03-25T11:22:24 did you get the rtems-source-builder to compile the Compiler Tool Set? 2015-03-25T11:22:39 i did 2015-03-25T11:23:43 what step of the tutorial are you stuck on? 2015-03-25T11:25:35 *** ragunath has quit IRC (Quit: Page closed) 2015-03-25T11:30:19 *** ketul has joined #rtems 2015-03-25T11:31:13 *** jrcatbagan has joined #rtems 2015-03-25T11:32:49 hey jrcatbagan 2015-03-25T11:33:09 DrJoel and I were talking about your proposal, and thought we'd mention micromonitor as a possible starting point for you to consider. 2015-03-25T11:33:39 hi gedare! 2015-03-25T11:34:19 ok I will definitely look into that right now, thanks for the advice 2015-03-25T11:35:24 i will try to incorporate some more specifics into my proposal (code, diagrams, etc.) 2015-03-25T11:35:50 good. time is tight so don't be overly meticulous 2015-03-25T11:35:53 *** gwty has quit IRC (Ping timeout: 244 seconds) 2015-03-25T11:36:00 will do! 2015-03-25T11:56:25 *** gwty has joined #rtems 2015-03-25T11:57:13 *** sujayraaj has joined #rtems 2015-03-25T12:12:20 *** sujayraaj has left #rtems 2015-03-25T12:15:59 *** jrcatbagan has quit IRC (Quit: nah nah nah nah, nah nah nah nah, hey hey hey, goodbye) 2015-03-25T12:27:39 *** ragunath has joined #rtems 2015-03-25T12:35:51 *** jrcatbagan has joined #rtems 2015-03-25T12:41:53 *** andrei__ has joined #rtems 2015-03-25T12:53:05 beng-nl : hi are you there? I have generated .exe for gpio interface but getting some error while interfacing /dev/mem on my BBB. 2015-03-25T12:54:13 *** ankur has joined #rtems 2015-03-25T13:09:27 *** charith_ has joined #rtems 2015-03-25T13:30:30 *** dr__house|away is now known as dr__house 2015-03-25T13:31:40 *** ankur has left #rtems 2015-03-25T13:39:54 gedare: I haven't looked yet into the cv implementation but I will, it just struck me as I was walking home 2015-03-25T13:40:08 does the score lock on the cv and not on a mutex, because it locks on the (rtems / posix) api? 2015-03-25T13:40:20 and if it'd lock on a mutex it wouldn't behave similar to a cv but to a .. mutex? 2015-03-25T13:40:35 score cv lock on the* 2015-03-25T13:41:43 i think you've just about got the gist of it 2015-03-25T13:42:04 score layer locks on a cv internally, and when that 'releases', the thread contends for the higher-layer mutex. 2015-03-25T13:46:31 awesome, I ll look over it with that in mind, thanks 2015-03-25T13:47:48 *** Hesham has quit IRC (Ping timeout: 265 seconds) 2015-03-25T13:50:02 :q 2015-03-25T13:50:08 oops, this isn't vim. 2015-03-25T13:53:40 *** monstr has quit IRC (Ping timeout: 255 seconds) 2015-03-25T14:18:19 *** Hesham has joined #rtems 2015-03-25T14:19:21 ketul: hi 2015-03-25T14:19:38 hi 2015-03-25T14:20:12 generated gpio.exe with mmap but getting /dev/mem could not be opened. 2015-03-25T14:25:53 jrcatbagan: hi 2015-03-25T14:31:32 benag-nl and gedare: done editing in proposal. please can you look at it. 2015-03-25T14:31:42 beng-nl* 2015-03-25T14:32:23 ketul: i don't have ready-made answer there - perhaps the node should exist? 2015-03-25T14:33:05 beng-nl : didn't get "perhaps the node should exist? " 2015-03-25T14:33:20 the device node, /dev/mem 2015-03-25T14:34:12 ok. Then I flashed emmc with debian and tried to open /dev/mem but said could not be opened 2015-03-25T14:35:43 ketul: i actually don't totally get what you're doing; i imagine implementing the gpio interface would actually be a c level thing with c level calls, without the filesystem interface being needed 2015-03-25T14:38:57 ketul: i've made a few comments. the proposal is much improved from before 2015-03-25T14:39:12 gedare: hello! 2015-03-25T14:40:26 beng-nl: let me explain in details. 2015-03-25T14:40:41 gedare : thanks for your comments 2015-03-25T14:41:19 jrcatbagan: we've contacted the author of micromonitor and he is interested in being a co-mentor if your project gets accepted. he is going to work on getting the current code base ready for use, there are a few small issues to resolve 2015-03-25T14:41:20 beng-nl: First i have done the register mapping of all the gpio registers as done in minix3 2015-03-25T14:41:54 jrcatbagan: with that in mind, you're strongly encouraged to look into using micromonitor as the basis for your gsoc project. ;) 2015-03-25T14:42:45 beng-nl: then I have created gpio.h file containing all the declarations of the useful functions. 2015-03-25T14:43:08 gedare: awesome! yeah I was looking through the micromonitor documentation for the past few hours trying to see how I can tweak the proposal to focus more on micromonitor. Thanks so much! 2015-03-25T14:44:43 beng-nl: After that created api through init.c . I am getting error in that code. https://gist.github.com/9e84e075384949414e59.git 2015-03-25T14:44:47 sure. ed said we might want to consider starting from release 1.19 as being a stable code base to work from 2015-03-25T14:45:53 jrcatbagan: there isn't currently a public repository, i've suggested he consider hosting it on github, but in the worst case you can just upload the 1.19 tarball to github and work from there 2015-03-25T14:47:17 gedare: sounds good! I definitely start keeping in touch with the author. is there any way you recommend for me to notify him? 2015-03-25T14:47:27 I'll* 2015-03-25T14:50:34 yeah umm 2015-03-25T14:52:13 beng-nl: while booting that I image I got all the user leds remained on and it got hanged I thought. 2015-03-25T14:52:39 jrcatbagan: i've sent an email to you and ed to make an introduction. 2015-03-25T14:53:06 gedare: Thank you so much! 2015-03-25T14:54:08 so by this exercise I got familiar with writing driver and api in RTEMS and compilation. 2015-03-25T14:54:26 and generation of .exe file. 2015-03-25T14:55:59 ketul: is that gist in an RTEMS source code you wrote? 2015-03-25T14:56:19 or is it a linux application 2015-03-25T14:56:28 no its in RTEMS 2015-03-25T14:56:55 successfully compiled in rtems 2015-03-25T14:57:24 ohh thats the part of my code where I am getting stuck. 2015-03-25T14:57:27 ketul: RTEMS does not set-up a Unix-like filesystem for you 2015-03-25T14:57:41 there isn't going to be a /dev/mem 2015-03-25T14:57:55 and mmap is probably not going to do anything 2015-03-25T14:59:15 gedare: but I think /dev/mem is inside the BBB. 2015-03-25T15:00:27 http://superuser.com/questions/71389/what-is-dev-mem 2015-03-25T15:00:57 it's a sort of file that linux creates 2015-03-25T15:01:46 ok then I need to change init.c accordingly 2015-03-25T15:02:13 gedare: am I right ? 2015-03-25T15:02:55 need to find out about initialization of gpio registers 2015-03-25T15:03:21 yeah.. 2015-03-25T15:03:58 ok. kernel space interfacing is required. 2015-03-25T15:04:09 hmmm 2015-03-25T15:04:19 maybe you should go read about the raspberry pi gpio support in rTEMS 2015-03-25T15:04:39 ya that I already referred 2015-03-25T15:04:50 and I got much out of it 2015-03-25T15:04:57 did you read andre's blog? 2015-03-25T15:05:24 yep 2015-03-25T15:05:37 ok, did you see his test case for GPIO? 2015-03-25T15:05:43 yes 2015-03-25T15:06:10 really I got the specified path from that 2015-03-25T15:06:24 ok, maybe follow it through. i don't know how you got to using /dev/mem or mmap() from there 2015-03-25T15:07:00 ok. he didn't use mmap and /dev/mem 2015-03-25T15:08:01 and one of your comment about function name. That I got from his blog and work done. 2015-03-25T15:08:58 gedare: one thing I want to ask you why his work hasn't merged in current rtems.? 2015-03-25T15:10:53 there were a few lingering issues having to do with the generality of the interface, and we had hoped to have something more generic but wind up thinking it is better to leave it as rpi-specific for now 2015-03-25T15:11:01 but if you follow the same interface as him, then this could help 2015-03-25T15:11:17 eventually, RTEMS needs to have a more well-defined, consistent driver framework 2015-03-25T15:12:47 oh.. ok 2015-03-25T15:15:31 gedare: hi 2015-03-25T15:16:02 and gedare: just as enthusiastic I want to know that about the python support in rtems provided not knowing much about both.:) 2015-03-25T15:16:02 heyr agu 2015-03-25T15:16:05 hey ragunath . 2015-03-25T15:16:59 gedare: do we write access to rtems.git if selected for gsoc to commit changes ? 2015-03-25T15:17:19 *get 2015-03-25T15:17:25 ragunath: no, you will submit patches to devel@rtems.org for commit 2015-03-25T15:17:40 preferably after you and your mentor agree they're ready for review and inclusion 2015-03-25T15:18:11 gedare: okay .. then any of you guys will pick it up 2015-03-25T15:18:28 right. 2015-03-25T15:19:11 gedare: so as far as pushing changes upstream is concerned can we say it is enough if i make a patch , get reviewed and send to devel 2015-03-25T15:19:30 yes, when upstream is rtems.git that is the correct flow 2015-03-25T15:20:56 gedare: okay thanks 2015-03-25T15:21:57 ragunath: sure. we also expect the students will post their code to github continuosly, so your mentor and others may review as you progress. but ultimately, code gets into rtems.git by going through the devel mailing list. 2015-03-25T15:24:25 gedare: i wanted to know how much time is needed to commit changes to rtems.git.. sure i will be posting my code to github 2015-03-25T15:25:11 ragunath: it depends on lots of factors. usually within a day or two 2015-03-25T15:25:44 if no changes are needed after code review 2015-03-25T15:27:50 *** hermann_19829_ has joined #rtems 2015-03-25T15:28:17 gedare: better to have some time separately allocated to commit changes 2015-03-25T15:30:26 i'd put in about a week or so before the midterm to clean-up, code-review with mentor, and submit, and then again before the final 2015-03-25T15:30:51 if you get code submitted by the midterm, it's pretty hard to fail. :) 2015-03-25T15:30:52 jrcatbagan: I have used micromonitor in the past on both some mips and arm boards. Don't have access to any of the real hardware any more though. If one of the targets he supports runs on qemu you could get a feel for the user interface that way. 2015-03-25T15:31:07 gedare: want to ask that if we found code that can be freely distributed or regenerated according to its licence term. whether can we use in rtems or not? 2015-03-25T15:31:09 good idea DrJoel. 2015-03-25T15:31:21 ketul: yes, but check first 2015-03-25T15:31:42 *** charith_ has quit IRC (Ping timeout: 250 seconds) 2015-03-25T15:31:45 ketul: "freely distributed" is not always as free as you'd think. 2015-03-25T15:32:01 RTEMS prefers free as in beer. 2015-03-25T15:32:08 ragunath: keep posting code as it is ready for review and it will happen naturally along the way. Hopefully you won't have to block out time just to get stuff committed. You should be able to work on step N+1 while step N is getting reviewed 2015-03-25T15:32:08 ;) 2015-03-25T15:32:49 and freely distributed may have obligations or restrictions on the end user. Best to just ask about any potential code source 2015-03-25T15:33:01 DrJoel: Thank you, I'll definitely keep that in mind as I am currently browsing through the micromonitor docs/sources 2015-03-25T15:33:31 yeah ok :) so that licence of the code should satisfy the terms of the rtems' licence.? 2015-03-25T15:34:07 the short version yes. 2015-03-25T15:34:30 but the real rule is that we don't take code with viral licenses or that push obligations on users. 2015-03-25T15:35:03 ya that I heard in one of your lectures :) 2015-03-25T15:35:28 DrJoel: sure DrJoel.... i would plan as you mentioned 2015-03-25T15:40:00 *** Hesham has quit IRC (Ping timeout: 264 seconds) 2015-03-25T15:45:50 gedare: seen your profile. Can you explain more on " LockdownOS" 2015-03-25T15:46:04 what will be its functionaity 2015-03-25T15:47:17 ketul: it is a research project with the composite os: http://composite.seas.gwu.edu/ 2015-03-25T15:48:44 ohh thats really looks awesome 2015-03-25T15:50:39 *** CaptHindsight has left #rtems ("not here") 2015-03-25T15:50:54 so gedare : this research project will be helpful for United States Navy and Marine Corps. Is it so? 2015-03-25T15:51:28 hopefully 2015-03-25T15:51:38 ok great. 2015-03-25T15:51:42 what's the best way of linking a paste of code that s > 10 lines. is pastebin suitable? 2015-03-25T15:51:48 yes 2015-03-25T15:52:28 Research like that is funny. It could really be useful for them but they seem very prone never to use much done as research. Hard to get from lab to field 2015-03-25T15:53:30 http://pastebin.com/K2EErhcF 2015-03-25T15:53:35 at (paste's) line 175 2015-03-25T15:53:45 DrJoel : I think thats the universal thinking in all country. 2015-03-25T15:54:06 does it check if the giant lock is owned by it and then drops it? 2015-03-25T15:54:37 if it does, is there any reason it does this except running into a deadlock (I hold the giant lock and I lock on some other stuff ) 2015-03-25T15:55:31 don't know. 2015-03-25T15:56:22 ok, I ll look into it 2015-03-25T16:00:53 Look at examples-v2/led/posix_cond_server for how it is supposed to be used in practice 2015-03-25T16:01:28 lock.. wait.. unlock.. implicitly unlocked inside wait and relocked before returning 2015-03-25T16:03:23 headed out. bye 2015-03-25T16:03:32 bye 2015-03-25T16:06:48 beang-nl: are you there? 2015-03-25T16:06:53 beng-nl 2015-03-25T16:07:22 *** k12 has joined #rtems 2015-03-25T16:08:08 *** k12 has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-25T16:08:13 *** gedare has quit IRC (Ping timeout: 264 seconds) 2015-03-25T16:08:26 * DrJoel is going away from the kb 2015-03-25T16:09:55 DrJoel: want your advice. 2015-03-25T16:11:58 DrJoel: can you suggest any materials may be videos for practical embedded systems concepts. provided I studied theoretical part of it. 2015-03-25T16:20:53 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-25T16:20:57 *** kiwichris has quit IRC () 2015-03-25T16:23:34 *** ragunath has left #rtems 2015-03-25T16:28:20 *** gedare has joined #rtems 2015-03-25T16:28:21 *** ChanServ sets mode: +o gedare 2015-03-25T16:29:55 hey DrJoel i think you cc'd the wrong person on the umon email. 2015-03-25T16:45:41 *** kiwichris has joined #rtems 2015-03-25T17:01:02 DrJoel: you did it again 2015-03-25T17:18:44 *** hermann_19829_ has left #rtems ("http://quassel-irc.org - Chat comfortably. Anywhere.") 2015-03-25T17:37:22 yep.. autocomplete wasn't my friend 2015-03-25T17:37:34 AC wasn't working and gut was on way. I was melting. :( 2015-03-25T17:41:46 DJoel( or anyone, really) I ve been looking at the cv implementation in https://git.rtems.org/rtems-libbsd/commit/?id=50553bcd8e47f80e7e41438b45f41ddecb607754 , can I ask a few questions? Thanks 2015-03-25T17:42:31 DrJoel * sorry, typo 2015-03-25T17:45:19 sure.. 2015-03-25T17:45:48 That's the FreeBSD implementation. I don't know anything about that. 2015-03-25T17:46:09 I thought we are talking about POSIX cv's in rtems/cpukit/posix/src and the POSIX mutexes 2015-03-25T17:46:19 I can try to answer on that code 2015-03-25T17:46:25 we were, I'm not interested that much in code specificity 2015-03-25T17:46:28 as in the general idea 2015-03-25T17:46:39 OK .. ask 2015-03-25T17:46:44 you said score cv locks on a cv and initially it made no sense 2015-03-25T17:47:14 so, basically the reason is that in score you lock the threadcore on the appropriate cv(posix or rtems) 2015-03-25T17:47:24 so that you get a cv behaviour, if you'd lock on a mutex 2015-03-25T17:47:24 yeah.. the entire cv application programming pattern is somewhat insane at first glance. :) 2015-03-25T17:47:27 (like I thought) 2015-03-25T17:47:35 you'd have to wait the unlock 2015-03-25T17:47:54 [trac] GSoC/2015 edited by abd_elhamid_ (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-25T17:48:25 threadqueue, not threadcore, sorry 2015-03-25T17:48:32 right.. you have to unlock the associated mutex before blocking on the cv because another thread must be able to enter the "monitor/critical section" and release it 2015-03-25T17:48:42 Did you look at the application example I posted earlier? 2015-03-25T17:49:16 sorry, not sure if I was around, I missed it 2015-03-25T17:50:03 examples-v2/led/*cond* is a set of example programs showing how the user must use the APIs 2015-03-25T17:52:00 *** gedare has quit IRC (Ping timeout: 252 seconds) 2015-03-25T17:52:58 regarding the programming pattern, I'm just having trouble coping with multiple level of mutexes/ synchronization 2015-03-25T17:54:21 for example, right now I believe that wait ...locks on a mutex, adds the thread to the queue, releases, then locks on the cv? 2015-03-25T17:55:14 *** gedare has joined #rtems 2015-03-25T17:55:14 *** ChanServ sets mode: +o gedare 2015-03-25T17:57:46 * DrJoel waves to gedare 2015-03-25T17:58:19 not locks on the cv "blocks" on the cv. The mutex is unlocked at that point so the cv can be signaled. 2015-03-25T17:59:37 Here are some slides on the computer science monitor programming language structure this is trying to emulate: http://www.cs.utexas.edu/users/witchel/372/lectures/08.Semaphore-Monitors.pdf 2015-03-25T18:02:33 thanks, I ll look into it 2015-03-25T18:05:51 * gedare waves back, and leaves again 2015-03-25T18:09:30 andrei__: Do not feel bad about being confused about CVs. Jennifer and I implemented them initially and I couldn't figure out how to use them in an application. LOL 2015-03-25T18:10:36 *** gedare has quit IRC (Ping timeout: 272 seconds) 2015-03-25T18:12:11 DrJoel: Thanks, I'm trying to propose an approach that's as closest to reality as possible and it's proving to be more difficult than I expected 2015-03-25T18:13:50 I'll take some time to go through everything that you and gedare said and come up with a conclusion, I think I got enough info to provide at least a better approach than I currently have 2015-03-25T18:28:25 *** gwty_ has joined #rtems 2015-03-25T18:28:34 *** gwty has quit IRC (Ping timeout: 245 seconds) 2015-03-25T18:35:00 *** jrcatbagan has quit IRC (Ping timeout: 246 seconds) 2015-03-25T18:37:57 *** jrcatbagan has joined #rtems 2015-03-25T18:49:34 *** andrei__ has quit IRC (Ping timeout: 246 seconds) 2015-03-25T19:14:57 [git] 2e54abc by Chris Johns: [sb: Make the source and patch labels have sortable label. This patch changes 'patch1' to 'patch0001' so sorting of the macro keys in the patch macro map returns the order the patches are declared. The helps mamage patch sets that need a specific order.] http://devel.rtems.org/changeset/2e54abc/rtems-source-builder 2015-03-25T19:52:45 *** geheimnis` has quit IRC (Ping timeout: 246 seconds) 2015-03-25T19:54:16 *** geheimnis` has joined #rtems 2015-03-25T21:14:43 *** sgworks has joined #rtems 2015-03-25T22:35:27 *** sgworks has quit IRC (Ping timeout: 246 seconds) 2015-03-25T22:36:11 *** gwty_ has quit IRC (Ping timeout: 244 seconds) 2015-03-25T22:42:29 *** ketul has joined #rtems 2015-03-25T22:54:08 *** gwty has joined #rtems 2015-03-25T22:55:34 *** sgworks has joined #rtems 2015-03-25T23:11:22 [trac] GSoC/2015 edited by gadia (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-25T23:18:18 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-25T23:18:18 *** gwty has quit IRC (Read error: Connection reset by peer) 2015-03-25T23:19:22 *** sgworks has quit IRC (Remote host closed the connection) 2015-03-25T23:24:17 *** gwty has joined #rtems 2015-03-25T23:54:55 *** gwty has quit IRC (Ping timeout: 256 seconds) 2015-03-26T00:17:54 *** jrcatbagan has quit IRC (Quit: nah nah nah nah, nah nah nah nah, hey hey hey, goodbye) 2015-03-26T00:20:02 *** gwty has joined #rtems 2015-03-26T00:38:29 *** sgworks has joined #rtems 2015-03-26T00:51:35 [trac] rtems-trace-buffering.png (RTEMS Trace Buffering) attached to #Developer/Tracing by chrisj http://devel.rtems.org/ticket/Developer/Tracing 2015-03-26T00:57:47 [trac] Developer/Tracing edited by chrisj (Add trace buffering support.) http://devel.rtems.org/wiki/Developer/Tracing 2015-03-26T01:01:15 *** sgworks has quit IRC (Remote host closed the connection) 2015-03-26T01:02:49 [trac] Developer/Tracing/Trace_Buffering created http://devel.rtems.org/wiki/Developer/Tracing/Trace_Buffering 2015-03-26T01:03:07 [trac] Developer/Tracing/Trace_Buffering edited by chrisj (empty comment) http://devel.rtems.org/wiki/Developer/Tracing/Trace_Buffering 2015-03-26T01:03:56 *** gwty has quit IRC (Ping timeout: 252 seconds) 2015-03-26T01:05:19 [trac] Developer/Tracing/Trace_Buffering edited by chrisj (Minor fixes.) http://devel.rtems.org/wiki/Developer/Tracing/Trace_Buffering 2015-03-26T01:05:22 *** gwty has joined #rtems 2015-03-26T01:09:26 [git] b988768 by Chris Johns: [trace-linker: Add Trace Buffering support. Trace buffering traces into a static buffer complete with timestamp and the executing context. A shell command provides access to the data.] http://devel.rtems.org/changeset/b988768/rtems-tools 2015-03-26T01:35:43 *** monstr has joined #rtems 2015-03-26T01:55:01 [trac] GSoC/2015 edited by jrcatbagan (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-26T02:17:33 *** gwty has quit IRC (Ping timeout: 250 seconds) 2015-03-26T02:23:44 [git] b69cd3f by Chris Johns: [trace-linker: Add API and POSIX interfaces.] http://devel.rtems.org/changeset/b69cd3f/rtems-tools 2015-03-26T02:27:35 *** kiwichris has quit IRC () 2015-03-26T03:14:05 *** dr__house is now known as dr__house|away 2015-03-26T04:08:59 *** Hesham has joined #rtems 2015-03-26T04:13:11 *** shipre has joined #rtems 2015-03-26T04:23:37 *** andrei__ has joined #rtems 2015-03-26T04:43:11 [trac] GSoC/2015 edited by bremby (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-26T04:43:36 [trac] GSoC/2015 edited by bremby (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-26T06:20:39 *** stryx`_ has quit IRC (Ping timeout: 264 seconds) 2015-03-26T06:26:09 *** mkhoory has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** shipre has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** monstr has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** andrei__ has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** rtems-bot has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** guerby has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** beng-nl has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** cdesai has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** timothyh has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** dr__house|away has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** geheimnis` has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** gigetoo has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** mazzoo has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** ChanServ has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** noselasd has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** Hesham has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** ita has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** diginet has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** peerst has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** DrJoel has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** commander has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** fds has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** rokka has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** verm__ has quit IRC (*.net *.split) 2015-03-26T06:26:09 *** Schoumi has quit IRC (*.net *.split) 2015-03-26T06:28:42 *** rtemsLogger has joined #rtems 2015-03-26T06:29:17 *** Schoumi has joined #rtems 2015-03-26T06:30:30 *** shipre has quit IRC (*.net *.split) 2015-03-26T06:30:30 *** monstr has quit IRC (*.net *.split) 2015-03-26T06:33:42 *** stryx`__ has joined #rtems 2015-03-26T06:40:38 *** stryx`_ has joined #rtems 2015-03-26T06:44:10 *** shipre_ has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-26T07:01:25 *** mkhoory has quit IRC (Read error: Connection reset by peer) 2015-03-26T07:06:47 *** waleedziad has joined #rtems 2015-03-26T07:21:57 *** waleedziad has quit IRC (Ping timeout: 265 seconds) 2015-03-26T08:05:49 [git] 9bfcb86 by Sebastian Huber: [rtems-bsd-mutex: Support priority changes] http://devel.rtems.org/changeset/9bfcb86/rtems-libbsd 2015-03-26T08:05:49 [git] 024e064 by Sebastian Huber: [nexus-devices: Fix for QorIQ T variants] http://devel.rtems.org/changeset/024e064/rtems-libbsd 2015-03-26T08:05:49 [git] fc9e83b by Sebastian Huber: [mmc: Import MMC/SDCard support from libusb] http://devel.rtems.org/changeset/fc9e83b/rtems-libbsd 2015-03-26T08:05:49 [git] 9ea1778 by Sebastian Huber: [mmc: Remove RTEMS special cases] http://devel.rtems.org/changeset/9ea1778/rtems-libbsd 2015-03-26T08:05:49 [git] 8179d4c by Sebastian Huber: [dw_mmc: Implement host acquire/release This should avoid the "mmc: host bridge didn't serialize us." panic.] http://devel.rtems.org/changeset/8179d4c/rtems-libbsd 2015-03-26T08:05:50 [git] d0ecc91 by Sebastian Huber: [media01: New test] http://devel.rtems.org/changeset/d0ecc91/rtems-libbsd 2015-03-26T08:25:21 *** ragunath has joined #rtems 2015-03-26T08:29:47 *** gedare has joined #rtems 2015-03-26T08:29:47 *** ChanServ sets mode: +o gedare 2015-03-26T08:34:54 good morning 2015-03-26T08:39:25 *** charith has joined #rtems 2015-03-26T08:40:27 * DrJoel waves back. hacking until everyone wakes up here. Slow spring break staycation 2015-03-26T08:50:44 hey DrJoel take a min to answer Abd's question please 2015-03-26T08:50:49 about stack analysis, on devel ml 2015-03-26T08:53:40 Answering now 2015-03-26T08:54:28 *** andrei__ has quit IRC (Ping timeout: 246 seconds) 2015-03-26T09:07:51 * gedare tries wrangling some more mentors 2015-03-26T09:09:45 *** krohini has joined #rtems 2015-03-26T09:10:26 *** ketul has joined #rtems 2015-03-26T09:11:58 Hi! 2015-03-26T09:12:58 I was looking into understanding the SMP aspect of Raspberry PI 2 support project. 2015-03-26T09:13:38 hi krohini 2015-03-26T09:16:08 I get that there are initializations which need to be done with respect to memory management, cache management and then synchronizations in the kernel for scheduling, interrupt handling, critical section handling and so 2015-03-26T09:17:23 the kernel parts should be (hopefully) not part of your project 2015-03-26T09:17:33 krohini: Look at arm/zynq for what another another BSP was required to do. 2015-03-26T09:17:52 mostly the BSP has to worry about correct boot-up, and providing the hooks for SMP features like interprocessor interrupts (IPIs) 2015-03-26T09:17:58 [joel@rtbf64a arm]$ grep HAS_SMP */*am 2015-03-26T09:17:58 altera-cyclone-v/Makefile.am:if HAS_SMP 2015-03-26T09:17:58 realview-pbx-a9/Makefile.am:if HAS_SMP 2015-03-26T09:17:58 xilinx-zynq/Makefile.am:if HAS_SMP 2015-03-26T09:18:27 a lot of it might already be done if the rpi2 uses arm cp15 2015-03-26T09:18:29 There are three ARM examples. If the ARM CPU model revision is the same, you get lucky and get to share that part. 2015-03-26T09:18:31 i don't know if this is the case or not. 2015-03-26T09:19:14 Those three give you an idea of what you have to provide. arm cpu model specific code is shared. And as gedare noted, you may reuse or you may have o write it. Neither of us know. 2015-03-26T09:20:13 beng-nl: hi 2015-03-26T09:20:38 bbl 2015-03-26T09:22:30 ok.. so bsp related changes only..I'll first try these suggestions..thanks! 2015-03-26T09:31:17 hi devs, 2015-03-26T09:31:32 hi charith 2015-03-26T09:31:46 I'm particularly interested about building rtems with clang 2015-03-26T09:32:43 Can you update me on what are the areas we have covered up doing that. I think there is a project going on with Amar. 2015-03-26T09:32:53 hi gadare! 2015-03-26T09:33:26 i thought that you have been speaking with him about it? 2015-03-26T09:33:43 me too 2015-03-26T09:34:11 mmh..yeah I sent some chats to amar through this. 2015-03-26T09:34:15 me too 2015-03-26T09:34:24 BSPs : 191 2015-03-26T09:34:24 BSPs with Zero : 186 2015-03-26T09:34:24 BSPs with only in shared: 5 2015-03-26T09:34:49 But didn't had much chance to catch him here. May be the timezone difference :) 2015-03-26T09:34:52 If you ignore the avr, there is a known false positive on one BSP, one on or1k, and one overflow on 16 bit targets 2015-03-26T09:35:44 I was trying to figure things on my own last few days As I thought it will help me to understand better. 2015-03-26T09:37:51 gedare: hi. you were right about mmap. recently found https://github.com/petermancuso/bbb designed for Angstrom Distribution or can be applied to linux too 2015-03-26T09:38:19 i see charith well, i have sent as much info about the project as I know out 2015-03-26T09:39:14 ketul, RTEMS does not have the mmap interface implemented, this won't work. 2015-03-26T09:40:34 yep I came to know that. but I have done AM335X register address mapping in header file 2015-03-26T09:41:15 thank you very much gadare for that info. I went through them and get most of the things set up. I even started to hack on clang/llvm too. I just wanted to get get some facts to focus my proposal on. I discuss with amar. 2015-03-26T09:41:18 i see. well, your best bet in terms of code reuse would be something that has a permissive license and that works on "baremetal" BB 2015-03-26T09:41:33 RTEMS is a flat address space with no protection so the mmap and user/kernel space issues are just not there. We do have and encourage strong layering so random user code isn't twiddling hardware bits 2015-03-26T09:41:45 ok charith by the way, he is verm__ here, though i don't know if he's here. you might try to private-message him 2015-03-26T09:42:09 mmap code may be useful for references if you can map it mentally back to the real addresses and direct operations 2015-03-26T09:42:45 yeah DrJoel 2015-03-26T09:42:49 yep i know. I already contacted few times. I will continue to do so. 2015-03-26T09:52:17 gedare: should I mail to beng-nl for reviewing proposal or should I on mailing list as about 1 day is remaining 2015-03-26T09:55:13 ketul: you do what you think is right for you. at this point there's no guarantee anyone has time to give it another review before the deadline 2015-03-26T09:55:56 i think we gave you some useful feedback already, and you should be focusing on improving the details of the proposal especially in regards to API design and testing 2015-03-26T09:56:38 ok. ya according to that I done changes 2015-03-26T09:56:53 ok 2015-03-26T09:58:19 and gedare you asked about i2c . but I covered it in proposal. so exactly didn't get about "What about the i2c driver?" 2015-03-26T09:58:59 *** krohini has quit IRC (Ping timeout: 245 seconds) 2015-03-26T09:59:04 *** anandkp92 has joined #rtems 2015-03-26T10:01:16 what i2c API are you implementing? 2015-03-26T10:05:48 *** jrcatbagan has joined #rtems 2015-03-26T10:29:27 *** antgreen has joined #rtems 2015-03-26T10:44:41 *** charith_ has joined #rtems 2015-03-26T10:45:34 *** charith has quit IRC (Ping timeout: 255 seconds) 2015-03-26T10:47:16 *** hermann_19829 has joined #rtems 2015-03-26T10:49:56 DrJoel: would it be ok to forward the message to devel or would you recommend just starting a new thread? 2015-03-26T10:51:27 jrcatbagan: start a new thre 2015-03-26T10:51:34 give a wee bit of context 2015-03-26T10:51:54 generally speaking it is a bit of a faux pas to forward "private" email to "public" lists. 2015-03-26T10:52:13 gedare: ok, thank you! will do! 2015-03-26T10:52:26 +1 2015-03-26T10:52:44 gedare: I am enjoying working AC today. :) 2015-03-26T10:53:10 heh. DrJoel i have an electrician here now, to fix an outlet that blew 2015-03-26T10:53:20 and i have to write 2 papers... 2015-03-26T10:53:54 Hmmm... I have replaced a few over the years where dust built up in them until they shorted. Apparently that failure mode is not uncommon 2015-03-26T10:54:13 interesting. i had to do that for a light switch a few years ago. 2015-03-26T10:54:16 One was a dryer plug and it smelled awful. Another was a regular outlet and it just burnt a bit 2015-03-26T10:54:53 Hmm.. I also had a bathroom switch that did that. I have gotten away from putting in the cheapest and moving up a couple of grades. Seems to help some 2015-03-26T10:58:08 *** waleedziad has joined #rtems 2015-03-26T11:00:52 *** Hesham has quit IRC (Ping timeout: 252 seconds) 2015-03-26T11:04:42 *** charith__ has joined #rtems 2015-03-26T11:04:50 *** charith_ has quit IRC (Ping timeout: 244 seconds) 2015-03-26T11:20:13 verm__: in https://git.rtems.org/amar/waf.git/tree/rtems_waf/builder.py, in the function _get_id(name), a counter is kept for each name_id, which keeps incrementing. Why? 2015-03-26T11:20:23 *** gwty has joined #rtems 2015-03-26T11:23:00 hi 2015-03-26T11:25:46 hi gwty 2015-03-26T11:29:09 hi gedare, i successfully built rtems-source-builder. turns out the git protocol was blocked by proxy. 2015-03-26T11:30:52 ahhh 2015-03-26T11:31:32 gwty: did you find instructions in the doc for getting around the issue or did you figure it out yourself 2015-03-26T11:32:11 i figured it out. 2015-03-26T11:33:45 *** gwty has quit IRC (Read error: Connection reset by peer) 2015-03-26T11:35:05 fwiw I can't build with RSB right now. rtems-tools is failing. :( 2015-03-26T11:36:21 *** gwty has joined #rtems 2015-03-26T11:40:07 gedare: hi 2015-03-26T11:41:55 hi ragunath 2015-03-26T11:42:59 gedare: i have updated my proposal in melange with more details on deliverables 2015-03-26T11:43:23 *** krohini has joined #rtems 2015-03-26T11:43:57 If a sample program is to be used for testing SMP, what are the considerations? 2015-03-26T11:45:28 gedare: i added more details in project schedule section... can you pls take a look? 2015-03-26T11:45:44 What are the existing smp tests? 2015-03-26T11:46:01 testsuites/smptests 2015-03-26T11:46:21 you should be able to run them on the arm/zynq BSP on qemu AFAIK 2015-03-26T11:46:49 ragunath: i suspect the lwIP changes are mostly going to go up to the RSB 2015-03-26T11:47:55 ragunath: and the driver probably goes into libbsd, if it targets the new tcp/ip stack. 2015-03-26T11:48:03 gedare: oh i missed that .. i was going to change that ... i will correct it 2015-03-26T11:48:26 but the extra details are good. the schedule is aggressive i think optimistic, but gives a good layout and progression for work to do. 2015-03-26T11:48:46 although, i wonder about putting gpio at the tail end. seems like this is "low-hanging fruit" 2015-03-26T11:50:42 anandkp92: that is used in 'waf list' 2015-03-26T11:50:50 to automatically enumerate the target so they're unique 2015-03-26T11:56:15 Also, I have learnt that the performance hints that cache configurations are not right. Can you tell me how this was identified.. 2015-03-26T11:57:48 The performance in the drhystone and whetstone benchmarks under rtems was significantly lower than with Linux 2015-03-26T11:58:00 The CPU model is different so the cache control is different 2015-03-26T11:58:18 Ihaven't looked at the code but you might even see this area of initialization disabled 2015-03-26T11:58:53 verm__: oh ok 2015-03-26T11:59:17 i'm guessing Alan has an RKI package with dhrystone and whetstone in it 2015-03-26T11:59:50 Yes. May make sense to add them to examples-v2 as well. 2015-03-26T12:01:17 umm..what is the RKI package? 2015-03-26T12:02:00 https://github.com/alanc98/rki 2015-03-26T12:02:18 i don't know if it's up-to-date 2015-03-26T12:04:34 *** Hesham has joined #rtems 2015-03-26T12:05:54 that's definitely worth a post on an rtems mailing list and cc'ing Alan to ask. 2015-03-26T12:08:34 *** charith__ has quit IRC (Ping timeout: 245 seconds) 2015-03-26T12:12:40 anandkp92: eventually i'll change it to something that makes more sense i want to name the targets 2015-03-26T12:12:54 the reason it's designed this way is to hide the logic and all the python from users 2015-03-26T12:13:43 *** charith has joined #rtems 2015-03-26T12:15:35 i got the unique name part, but could you explain how it helps in hiding the python? 2015-03-26T12:17:13 verm__: are you talking about each class inheriting from rtems_base and hence hiding code? 2015-03-26T12:17:28 *** andrei_ has joined #rtems 2015-03-26T12:25:05 anandkp92: no i'km talking about all the wscript files in cpukit/ and c/ 2015-03-26T12:25:15 *** gwty has quit IRC (Read error: Connection reset by peer) 2015-03-26T12:26:16 *** charith has quit IRC (Ping timeout: 272 seconds) 2015-03-26T12:26:29 *** charith has joined #rtems 2015-03-26T12:27:39 *** gwty has joined #rtems 2015-03-26T12:35:14 *** monstr_ has quit IRC (Ping timeout: 256 seconds) 2015-03-26T12:37:04 verm__: ok.. 2015-03-26T12:43:13 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-26T12:44:46 *** anandkp92 has quit IRC (Ping timeout: 246 seconds) 2015-03-26T12:47:51 *** dr__house|away is now known as dr__house 2015-03-26T12:50:37 *** anandkp92 has joined #rtems 2015-03-26T12:52:01 *** Hesham has quit IRC (Ping timeout: 264 seconds) 2015-03-26T12:53:53 *** anandkp92 has quit IRC (Client Quit) 2015-03-26T12:54:10 *** anandkp92 has joined #rtems 2015-03-26T13:04:28 *** charith has quit IRC (Read error: Connection reset by peer) 2015-03-26T13:04:53 *** charith has joined #rtems 2015-03-26T13:26:57 verm__: Just sent a message to devel that is just over the size limit. Please let it through. It is a patch set from Jiri Gaisler that needs to be added to rtems-tools 2015-03-26T13:31:30 ok 2015-03-26T13:32:23 This patch set lets gdb simulator run leon2 and leon3 executables as well as eliminate the reason sis exists as a BSP variant separate from erc32. 2015-03-26T13:32:38 done 2015-03-26T13:32:39 The sis simulator didn't initialize the hw in the same state as the real HW. 2015-03-26T13:33:10 So now we get to figure out the impact of eliminating the sis BSP variant (after 4.11). I am sure it will impact more places than any other BSP. 2015-03-26T13:33:15 * DrJoel is heading outside for a while 2015-03-26T13:33:20 Thanks for letting the message through 2015-03-26T14:03:25 *** charith has quit IRC (Ping timeout: 264 seconds) 2015-03-26T14:52:10 *** ragunath has quit IRC (Quit: Page closed) 2015-03-26T14:58:59 *** krohini has quit IRC (Ping timeout: 245 seconds) 2015-03-26T14:59:32 *** Hesham has joined #rtems 2015-03-26T15:21:44 hello 2015-03-26T15:21:54 hi andrei_ 2015-03-26T15:23:36 I kind of feel bad to ask, since you guys have already been helping me out a lot. I went through what DrJoel provided yesterday and I wrote an rtems API pseudocode for wait and signal, can I paste it here to get some opinions? 2015-03-26T15:26:08 i don't have time to take a look today 2015-03-26T15:26:51 no probs 2015-03-26T15:38:41 [git] 1a6683a by Sebastian Huber: [libcsupport: Include missing header file] http://devel.rtems.org/changeset/1a6683a/rtems 2015-03-26T15:44:37 *** krohini has joined #rtems 2015-03-26T15:53:14 *** krohini has quit IRC (Ping timeout: 246 seconds) 2015-03-26T16:05:56 *** gwty_ has joined #rtems 2015-03-26T16:06:18 *** gedare has quit IRC (Ping timeout: 252 seconds) 2015-03-26T16:06:31 *** gwty has quit IRC (Ping timeout: 252 seconds) 2015-03-26T16:07:38 *** kiwichris has joined #rtems 2015-03-26T16:17:58 *** alaeddine has joined #rtems 2015-03-26T16:19:21 [trac] GSoC/2015 edited by hermann19829 (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-26T16:22:47 [trac] GSoC/2015 edited by hermann19829 (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-26T16:40:40 *** kiwichris has quit IRC () 2015-03-26T16:41:09 *** gwty_ has quit IRC (Ping timeout: 248 seconds) 2015-03-26T16:43:34 *** waleedziad has quit IRC (Ping timeout: 245 seconds) 2015-03-26T16:56:31 [git] 20feafb by Joel Sherrill: [Add rtems-tester configurations for simsh family of BSPs] http://devel.rtems.org/changeset/20feafb/rtems-tools 2015-03-26T17:00:59 *** jrcatbagan has quit IRC (Quit: see ya) 2015-03-26T17:01:54 *** Hesham has quit IRC (Quit: Leaving.) 2015-03-26T17:10:27 *** jrcatbagan has joined #rtems 2015-03-26T17:19:27 *** hermann_19829 has left #rtems ("http://quassel-irc.org - Chat comfortably. Anywhere.") 2015-03-26T17:37:46 *** gwty has joined #rtems 2015-03-26T17:40:07 *** gwty has quit IRC (Read error: Connection reset by peer) 2015-03-26T17:41:41 *** gwty has joined #rtems 2015-03-26T17:42:30 *** hermann_19829 has joined #rtems 2015-03-26T17:45:25 *** andrei_ has quit IRC (Ping timeout: 246 seconds) 2015-03-26T17:53:34 *** gwty_ has joined #rtems 2015-03-26T17:53:59 *** gwty has quit IRC (Ping timeout: 250 seconds) 2015-03-26T18:00:58 *** gwty_ has quit IRC (Read error: Connection reset by peer) 2015-03-26T18:01:22 *** gwty has joined #rtems 2015-03-26T18:05:11 *** gwty_ has joined #rtems 2015-03-26T18:06:10 *** gwty has quit IRC (Ping timeout: 256 seconds) 2015-03-26T18:10:57 *** gwty_ has quit IRC (Read error: Connection reset by peer) 2015-03-26T18:11:33 *** gwty has joined #rtems 2015-03-26T18:18:48 *** gwty has quit IRC (Ping timeout: 265 seconds) 2015-03-26T18:36:25 *** hermann_19829 has quit IRC (Remote host closed the connection) 2015-03-26T18:47:09 *** alaeddine has quit IRC () 2015-03-26T19:59:56 *** ketul has joined #rtems 2015-03-26T20:05:13 *** gedare has joined #rtems 2015-03-26T20:05:13 *** ChanServ sets mode: +o gedare 2015-03-26T20:14:54 *** gedare has quit IRC (Ping timeout: 252 seconds) 2015-03-26T20:36:19 *** dr__house is now known as dr__house|away 2015-03-26T20:59:57 *** yuanbin has joined #rtems 2015-03-26T21:13:06 *** krohini has joined #rtems 2015-03-26T21:25:22 *** krohini has quit IRC (Ping timeout: 265 seconds) 2015-03-26T21:48:59 *** kiwichris has joined #rtems 2015-03-26T22:23:41 [trac] #2312/defect (accepted) updated by chrisj (empty comment) http://devel.rtems.org/ticket/2312 2015-03-26T22:24:33 [trac] #2312/defect (closed) updated by Chris Johns (In [changeset:"f5b639bf7267b04efebe45dbd7047b573e7abe5c/rtems-tools"]: {{{ ...) http://devel.rtems.org/ticket/2312 2015-03-26T22:24:34 [git] f5b639b by Chris Johns: [rtems-tester: Remove repeated install line in wscript. Fixes #2312.] http://devel.rtems.org/changeset/f5b639b/rtems-tools 2015-03-26T22:26:12 *** krohini has joined #rtems 2015-03-26T22:32:06 *** fark has joined #rtems 2015-03-26T22:33:37 *** fark has quit IRC (Quit: Konversation terminated!) 2015-03-26T22:37:13 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-26T22:39:50 *** ketul has joined #rtems 2015-03-26T23:17:49 *** yuanbin has quit IRC (Remote host closed the connection) 2015-03-26T23:31:39 *** yuanbin has joined #rtems 2015-03-26T23:43:54 *** timothyh has quit IRC (Ping timeout: 256 seconds) 2015-03-26T23:46:33 *** shelf has joined #rtems 2015-03-26T23:51:51 *** yuanbin has quit IRC (Remote host closed the connection) 2015-03-26T23:53:16 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-26T23:53:31 *** ketul has joined #rtems 2015-03-26T23:53:33 *** shelf is now known as timothyh 2015-03-26T23:54:24 *** yuanbin has joined #rtems 2015-03-27T00:03:26 *** yuanbin has quit IRC (Remote host closed the connection) 2015-03-27T00:05:08 *** yuanbin has joined #rtems 2015-03-27T00:06:15 *** yuanbin has joined #rtems 2015-03-27T00:11:20 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-27T00:12:25 *** ketul has joined #rtems 2015-03-27T00:17:08 *** ketul has quit IRC (Client Quit) 2015-03-27T00:17:58 *** ketul has joined #rtems 2015-03-27T00:31:19 *** punitvara has joined #rtems 2015-03-27T00:33:22 *** punit_ has joined #rtems 2015-03-27T00:41:55 *** punit_ has quit IRC (Ping timeout: 246 seconds) 2015-03-27T01:00:00 *** yuanbin has quit IRC (Remote host closed the connection) 2015-03-27T01:00:30 *** yuanbin has joined #rtems 2015-03-27T01:07:23 *** tripsh has joined #rtems 2015-03-27T01:23:47 *** monstr has joined #rtems 2015-03-27T01:29:24 *** krohini has quit IRC (Ping timeout: 245 seconds) 2015-03-27T01:30:03 *** yuanbin has quit IRC (Remote host closed the connection) 2015-03-27T01:30:52 *** yuanbin has joined #rtems 2015-03-27T01:34:04 *** punitvara has quit IRC (Ping timeout: 246 seconds) 2015-03-27T02:03:35 *** gwty_ph has joined #rtems 2015-03-27T02:07:14 Hello 2015-03-27T02:16:11 *** jrcatbagan has quit IRC (Quit: nah nah nah nah, nah nah nah nah, hey hey hey, goodbye) 2015-03-27T02:42:25 *** gwty has joined #rtems 2015-03-27T02:42:28 *** gwty_ph has quit IRC (Read error: Connection reset by peer) 2015-03-27T02:48:37 *** yuanbin has quit IRC (Remote host closed the connection) 2015-03-27T02:50:27 *** yuanbin has joined #rtems 2015-03-27T02:54:13 *** gwty has quit IRC (Read error: Connection reset by peer) 2015-03-27T02:55:08 *** gwty has joined #rtems 2015-03-27T03:17:17 *** hermann_19829 has joined #rtems 2015-03-27T03:24:51 *** yuanbin has quit IRC (Remote host closed the connection) 2015-03-27T03:25:12 *** yuanbin has joined #rtems 2015-03-27T03:31:40 *** krohini has joined #rtems 2015-03-27T03:31:44 *** tripsh has quit IRC (Quit: Page closed) 2015-03-27T03:42:04 *** beng-nl has quit IRC (Ping timeout: 272 seconds) 2015-03-27T03:51:33 *** yuanbin has quit IRC (Remote host closed the connection) 2015-03-27T03:53:10 *** yuanbin has joined #rtems 2015-03-27T04:06:03 *** ketul has joined #rtems 2015-03-27T04:23:06 [git] d499e6d by Chris Johns: [trace-linker: Add the trace function signatures to the wrapper code. This gives the tools the ability to extract all needed data from the executable.] http://devel.rtems.org/changeset/d499e6d/rtems-tools 2015-03-27T04:23:18 gwty, hi 2015-03-27T04:24:30 Hi kiwichris 2015-03-27T04:24:45 *** waleedziad has joined #rtems 2015-03-27T04:24:58 gwty, how can I help ? 2015-03-27T04:26:10 Do you have an idea about LEON support in RTEMS? I'd like to improve it during this GSoC. 2015-03-27T04:40:08 *** yuanbin has quit IRC (Remote host closed the connection) 2015-03-27T04:41:38 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-27T04:41:59 *** ketul has joined #rtems 2015-03-27T04:56:18 *** Hesham has joined #rtems 2015-03-27T05:28:07 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-27T05:33:10 *** ketul has joined #rtems 2015-03-27T05:34:40 *** ketul has joined #rtems 2015-03-27T05:40:51 *** yuanbin has joined #rtems 2015-03-27T06:02:52 *** andrei_ has joined #rtems 2015-03-27T06:03:52 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-27T06:06:01 *** ketul has joined #rtems 2015-03-27T06:19:16 *** by_sauka has joined #rtems 2015-03-27T06:21:17 *** yuanbin has quit IRC (Remote host closed the connection) 2015-03-27T06:23:37 *** by_sauka has quit IRC (Client Quit) 2015-03-27T06:24:00 *** yuanbin has joined #rtems 2015-03-27T06:29:59 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-27T06:31:52 *** ketul has joined #rtems 2015-03-27T06:38:26 *** andrei__ has joined #rtems 2015-03-27T06:38:34 *** andrei_ has quit IRC (Ping timeout: 246 seconds) 2015-03-27T06:41:00 *** gwty has quit IRC (Read error: Connection reset by peer) 2015-03-27T06:46:57 *** andrei_ has joined #rtems 2015-03-27T06:49:04 *** andrei__ has quit IRC (Ping timeout: 246 seconds) 2015-03-27T06:56:12 *** gwty has joined #rtems 2015-03-27T07:07:25 *** krohini has quit IRC (Ping timeout: 250 seconds) 2015-03-27T07:09:44 [trac] GSoC/2015 edited by hermann19829 (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-27T07:27:22 *** ragunath has joined #rtems 2015-03-27T07:29:34 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-27T07:38:06 *** hermann_19829 has quit IRC () 2015-03-27T07:50:04 *** monstr has quit IRC (Ping timeout: 255 seconds) 2015-03-27T08:03:52 *** gedare has joined #rtems 2015-03-27T08:03:53 *** ChanServ sets mode: +o gedare 2015-03-27T08:11:36 Hi gedare 2015-03-27T08:13:05 hey gwty 2015-03-27T08:16:03 Ive sent you the hello world patch 2015-03-27T08:16:58 *** krohini has joined #rtems 2015-03-27T08:18:40 *** yuanbin has quit IRC (Quit: Leaving) 2015-03-27T08:20:17 gwty: i saw it and commented 2015-03-27T08:21:44 *** hermann_19829 has joined #rtems 2015-03-27T08:25:49 hi all... i have uploaded proof of enrollment and submitted my proposal ... anything else we need to do? 2015-03-27T08:25:56 hi ragunath 2015-03-27T08:26:10 hi gedare 2015-03-27T08:26:19 i don't think so, not from Google's OSPO perspective anyway 2015-03-27T08:27:09 Okay thanks gedare. I have e-mailed Daniel yesterday, he has yet to reply. I have gone through the patches and their comments, compiled his port and have an idea of the project will be. I am currently writing the proposal and will submit soon. 2015-03-27T08:27:41 gedare: having spent so much time on gsoc do not want to miss anything 2015-03-27T08:30:57 ok gwty have you been able to run his modified version? for this project, if you want to avoid buying hardware, you may have to focus your efforts on how to do some proper testing of drvmgr/pci libraries 2015-03-27T08:31:30 ragunath: understood. i think you're covered by proof + application. OSPO will contact if they need more proof 2015-03-27T08:32:55 [git] bfd2b7d by Joel Sherrill: [Add Jiri Gaisler SIS patch set for gdb 7.9 This patch set adds a lot of new capability including support for the leon2 and leon3. It also eliminates the difference between the initial state of the simulated erc32 and the real hardware.] http://devel.rtems.org/changeset/bfd2b7d/rtems-tools 2015-03-27T08:33:02 thanks gedare 2015-03-27T08:33:26 the next part is annoying, all the waiting 2015-03-27T08:33:58 you should focus on clearing your schedule of pending obligations, and wrapping up your school term, and stay in touch :) 2015-03-27T08:34:04 gwty: askign about the leon? 2015-03-27T08:34:18 gedare: to add to it i also have my semester exams during this time :) 2015-03-27T08:36:36 yeah, my advice, go focus on school 2015-03-27T08:39:34 As long as you have the paperwork in place. And you do. :) 2015-03-27T08:39:57 Yes DrJoel 2015-03-27T08:41:20 DrJoel: I've adviced gwty to talk to DanielH, but I don't know if he'll get back in time. In terms of scoping out a project around this Merge Leon stuff, my intuition is the best approach is to define testing for drvmgr and pci drivers 2015-03-27T08:42:28 I think that's the jist of it. The patch set is ~50 patches and I expect us to just suck it up as sparc specific code before 4.11 branches. 2015-03-27T08:43:46 I have been able to run it gedare. Thanks for the heads up. Would it be okay to write basic testing support for LEON in rtems-tools during the project? 2015-03-27T08:45:59 gwty: no. we're talking about writing some new test cases that use the drvmgr. You might need to implement a "dummy device" and "dummy driver" so that you can hook the dummy driver up to drvmgr and make calls to the dummy device through it. 2015-03-27T08:47:18 Sorry ... had to walk away 2015-03-27T08:47:36 Yes. The code needs tests and the code needs exercising on architectures beyond the sparc 2015-03-27T08:48:46 *** andrei_ has quit IRC (Ping timeout: 246 seconds) 2015-03-27T08:49:47 hermann_19829: I have commented on your proposal in Melange. 2015-03-27T08:51:45 awk 2015-03-27T08:52:01 gedare: wow that was fast! thanks a lot 2015-03-27T08:53:10 Yes gedare. I understand it. I asked if it would be better to add general testing code for LEON also (using the TSIM) , or if it would be too much to do. 2015-03-27T08:54:12 hermann_19829: if i don't put out fires as they come, pretty soon the entire forest is burning. 2015-03-27T08:54:45 gwty: you mean to extend the "tester" to use tsim? I think it would be better if it works with qemu actually. 2015-03-27T08:54:58 That should be a matter of about an hour of coding though... 2015-03-27T08:55:02 if it doesn't already 2015-03-27T08:56:32 [trac] GSoC/2015 edited by bremby (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-27T08:57:06 Thanks. I shall include it then, and perhaps try to get a running patch for it out soon :) 2015-03-27T09:00:56 *** krohini has quit IRC (Ping timeout: 256 seconds) 2015-03-27T09:12:31 *** fark has joined #rtems 2015-03-27T09:12:40 hello 2015-03-27T09:13:56 my name is Carlos 2015-03-27T09:14:16 i have a proposal for GSoC 2015-03-27T09:16:31 *** krohini has joined #rtems 2015-03-27T09:19:08 hi fark. 2015-03-27T09:20:12 hello! 2015-03-27T09:21:46 I know it's so late, but i have a proposal for rasberry Pi BSP layer improve 2015-03-27T09:22:28 The mentor is Alan Cudmore 2015-03-27T09:22:56 [trac] GSoC/2015 edited by abd_elhamid_ (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-27T09:23:48 I think i should send my proposal to rtems-devel@rtems.org, and add it to the list in the wiki 2015-03-27T09:24:49 I should provide access to Alan cudmor for gdoc proposal, but i can't find his Google account... 2015-03-27T09:25:11 Do you know were i can find it? 2015-03-27T09:25:13 hi fark what aspect interests you? 2015-03-27T09:25:46 fark: i doubt alan will have time to look today anyways, just send it to devel ml 2015-03-27T09:26:18 I'm interesting in complete low level peripheral and SD Card project group 2015-03-27T09:26:27 oh :( 2015-03-27T09:26:41 fark: I doubt you can be competitve with the other student who has already proposed doing that project 2015-03-27T09:27:39 I want to try 2015-03-27T09:27:45 [trac] #2314/enhancement (v:4.11) created by joel.sherrill (Integrate Coverage Reporting into rtems-tools) http://devel.rtems.org/ticket/2314 2015-03-27T09:27:49 hmm.. verm__ you added x86_64 BSP with high priority? do you know someone who would mentor such a project? 2015-03-27T09:28:25 [trac] 0001-Coverage-analysis-added-to-rtems-tester.patch (Patch as it stands today) attached to #2314 by joel.sherrill http://devel.rtems.org/ticket/2314 2015-03-27T09:28:37 ok fark. you should start by reviewing the prior work done by Andre, and his blog posts and code on github 2015-03-27T09:29:13 *** beng-nl has joined #rtems 2015-03-27T09:29:42 [trac] #2314/enhancement (new) updated by joel.sherrill (empty comment) http://devel.rtems.org/ticket/2314 2015-03-27T09:31:03 fark, you'll need to do our "getting started" guide too, and for this project, you need to demo it for the raspberry pi 2015-03-27T09:33:36 I already see it. I haven't started to review it yet, but I know there is the beginning. 2015-03-27T09:33:43 Do you think i should add my proposal to the wiki list and send it to Google? The time limit expires in some hours. 2015-03-27T09:34:29 Make your proposal to google-melange and include a link to the google doc along with a project overview. That way edits can easily happen during the review period 2015-03-27T09:35:32 thank you very much! 2015-03-27T09:35:32 yep. the important thing is to get the "header" (abstract, title) correct in Melange, and the body can link to your gdoc, but its better if you can fill in some of the body in Melange too. 2015-03-27T09:35:55 you should work on the Google application first, since you have to do your enrollment proof 2015-03-27T09:36:36 I already sent enrollment proof 2015-03-27T09:42:32 [trac] GSoC/2014 edited by fark (Added my proposal to the list) http://devel.rtems.org/wiki/GSoC/2014 2015-03-27T09:53:27 *** gwty has quit IRC (Read error: Connection reset by peer) 2015-03-27T09:54:11 *** gwty has joined #rtems 2015-03-27T09:59:21 [trac] GSoC/2015 edited by krohini (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-27T10:04:03 [trac] GSoC/2014 edited by fark (So sorry) http://devel.rtems.org/wiki/GSoC/2014 2015-03-27T10:04:04 [git] c13abcc by Joel Sherrill: [Update newlib to 20150223 snapshot tarball This includes Joel's patch to dynamically probe for the proper setting for intptr_t and uintptr_t. This eliminates many printf() format warnings due to them being incorrectly defined.] http://devel.rtems.org/changeset/c13abcc/rtems-source-builder 2015-03-27T10:04:04 [git] 3742749 by Joel Sherrill: [Update 4.11 gdb to 7.9 for sparc and use Jiri Gaisler's patch set It is combined into one until Chris can work around using the unorderd dict container on the patch set.] http://devel.rtems.org/changeset/3742749/rtems-source-builder 2015-03-27T10:05:18 [trac] GSoC/2015 edited by fark (Added my proposal to the list) http://devel.rtems.org/wiki/GSoC/2015 2015-03-27T10:06:05 *** punitvara has joined #rtems 2015-03-27T10:06:10 hello 2015-03-27T10:06:37 I need some assistance about gsoc proposal 2015-03-27T10:06:37 hi punitvara 2015-03-27T10:06:47 I have idea 2015-03-27T10:07:18 but I didnt able to complete task given by company 2015-03-27T10:07:24 I ahve written my roposal 2015-03-27T10:07:26 have 2015-03-27T10:07:47 What should I do ? I have to directly submit to google or RTEMS 2015-03-27T10:07:48 ? 2015-03-27T10:07:50 any idea? 2015-03-27T10:07:54 To google 2015-03-27T10:08:04 https://www.google-melange.com/gsoc/homepage/google/gsoc2015 2015-03-27T10:08:08 through their system 2015-03-27T10:08:24 you should also add yourself to the table on our wiki https://devel.rtems.org/wiki/GSoC/2015 2015-03-27T10:08:26 I didnt complete the task of init hello word 2015-03-27T10:08:32 [trac] GSoC/2015 edited by krohini (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-27T10:08:33 and read our instructions athttps://devel.rtems.org/wiki/GSoC 2015-03-27T10:08:57 punitvara: you can finish that after the deadline for submission of the proposal, we have some time before we make our choice on what to accept/reject 2015-03-27T10:09:09 it is mandatory to be done before we will accept though 2015-03-27T10:09:26 right now you should focus on getting your proposal into Melange 2015-03-27T10:09:58 Ohk so according to you I should send my proposal to google 2015-03-27T10:10:14 and then i should think about init hello word tas 2015-03-27T10:10:16 task 2015-03-27T10:10:19 right? 2015-03-27T10:10:33 fark: also work on getting your proposal into melange 2015-03-27T10:10:40 yes punitvara 2015-03-27T10:12:22 how to add myself to 2015-03-27T10:12:32 https://devel.rtems.org/wiki/GSoC/2015 2015-03-27T10:12:36 this sote? 2015-03-27T10:12:38 site 2015-03-27T10:15:48 you need to register there for an account 2015-03-27T10:18:25 *** hermann_19829 has quit IRC (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.) 2015-03-27T10:36:16 *** krohini has quit IRC (Ping timeout: 272 seconds) 2015-03-27T10:46:55 *** ketul has joined #rtems 2015-03-27T10:51:17 hi gedare 2015-03-27T10:52:46 hi ketul 2015-03-27T10:53:42 gedare: I have added work flow can you look at my proposal once? 2015-03-27T10:55:26 *** fark has quit IRC (Quit: Konversation terminated!) 2015-03-27T10:56:54 *** jrcatbagan has joined #rtems 2015-03-27T11:08:13 ketul: where? 2015-03-27T11:09:06 on melange before Project Deliverables 2015-03-27T11:12:03 gedare: got it ? 2015-03-27T11:14:19 ketul: yes. can any of the code be shared between the rpi and bb? 2015-03-27T11:14:36 ketul: what do you mean by "should be at kernel space"? 2015-03-27T11:14:58 not directly it can be shared 2015-03-27T11:15:22 kernel space - means like not using any user space (example mmap) :) 2015-03-27T11:16:05 what should i write into content ? 2015-03-27T11:16:10 into melang site 2015-03-27T11:16:13 [trac] GSoC/2015 edited by gwty (gwty submits GSOC 2015 proposal) http://devel.rtems.org/wiki/GSoC/2015 2015-03-27T11:16:21 only project discription 2015-03-27T11:16:33 punitvara: copy and paste your proposal from the gdoc template 2015-03-27T11:17:11 gedare: am I correct? 2015-03-27T11:17:45 all means relavant experience 2015-03-27T11:17:49 dates planning? 2015-03-27T11:17:52 everything? 2015-03-27T11:17:56 Hello gedare, i have submitted by draft proposal in melange and docs. Do take a look. 2015-03-27T11:18:07 gedare: rpi and bb has different processors and so almost all the things will be different 2015-03-27T11:18:24 gedare:but approach and testing would be similar. 2015-03-27T11:19:14 *** anandkp92 has quit IRC (Quit: Page closed) 2015-03-27T11:21:42 *** anandkp92 has joined #rtems 2015-03-27T11:22:04 *** punitvara has quit IRC (Ping timeout: 246 seconds) 2015-03-27T11:30:16 ketul: can't they share the same header files? that would help ensure consistency between interfaces 2015-03-27T11:31:17 gedare: no they can't. I have gone through the patch of last year . because both have different registers for different things. 2015-03-27T11:31:42 *** krohini has joined #rtems 2015-03-27T11:31:48 gwty: i've sent an email to Daniel Hellstrom about your project. what you need to do is to fill in more details about the deliverables and schedule. try to think about what the testing framework might be like. find links to the submitted patches and the code reviews for ideas of what else needs to be done. 2015-03-27T11:32:38 gwty: my opinion is that the key aspect of your project is really got to be about "After it, I plan to write tests for libdrvmgr. Writing them would involve creating dummy devices and drivers [as pointed out by Gedare]. 2015-03-27T11:32:38 I would also write basic tester code for LEON BSP and other BSPs if time permits." 2015-03-27T11:32:52 so you really ought to have some good details about how you will do that. 2015-03-27T11:33:08 merging the code is pretty much almost done. 2015-03-27T11:33:27 which means what you proposed to do in your first half is very little 2015-03-27T11:33:47 gwty.. writing a special test fixture device driver changes based on the type of driver but I know the termios tests use one I wrote. If I remember correctly, it reads an array of structure elements which tell it how to respond or what event to generate at a particular point in time. 2015-03-27T11:34:22 The details will vary but the concept might help. The manager is portable. The driver being managed can be anything 2015-03-27T11:44:14 gedare: 2015-03-27T11:47:29 gedare: anything you want me to add? 2015-03-27T11:48:06 hi ketul, i think you have a sufficient level of detail now. it is up to you whether you wuold like to further tweak or try to improve it 2015-03-27T11:48:39 [trac] RTEMS_Gsoc2015_Proposal_Punitvara.odt () attached to #GSoC/2015 by punitvara http://devel.rtems.org/ticket/GSoC/2015 2015-03-27T11:50:36 ok thanks. gedare. but I am sure I will be a ahead of what I proposed. As in some time will have nice demo of user leds. 2015-03-27T11:52:15 gedare, beng-nl: thank you for your additional comments, I have taken it into consideration while updating my proposal; please feel free to make any more additional comments/suggestions 2015-03-27T11:55:18 Thank you gedare. I am making changes. I was under the impression that merging leon code had not yet started. I thought of making most of the changes in rtems-tools to add support to LEON BSPs and also extend it to testing drivers 2015-03-27T11:58:50 gwty: my suggestion is more about testing the drvmgr, not the drivers. to test the drivers, you need to have the devices == hw or simulator+device model 2015-03-27T11:59:12 [trac] GSoC/2015 edited by punitvara (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-27T11:59:57 jrcatbagan: sure thing. the umon direction is exciting if we can get the code base sorted out. even if not, it will be well worth your time studying. you should definitely think "bigger" than even the other BB variants too. 2015-03-27T12:00:05 [trac] GSoC/2015 edited by punitvara (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-27T12:00:56 [trac] GSoC/2015 edited by punitvara (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-27T12:01:48 gedare: definitely! 2015-03-27T12:01:54 Yes gedare, my mistake 2015-03-27T12:27:04 Thank you DrJoel. I went through the termios test code. I guess you were referring to the termios01 testcase. It gave me a better idea of what to do. 2015-03-27T12:29:17 gwty: you will definitely want to try including some sketch of what test cases might look like, how you'll go about implementing them, so on. 2015-03-27T12:49:30 *** hermann_19829 has joined #rtems 2015-03-27T12:58:11 Is it better to change my proposal title, for now it is not mainly merging code, or should i leave it as it is? 2015-03-27T13:06:14 i'd suggest leaving it 2015-03-27T13:06:24 not worth mucking about too much with Melange right now2. 2015-03-27T13:08:37 gwty on the leon code. I just merged patches to gdb that let the simulator in it run all sparc BSPs. It doesn't simulate any drivers but I wanted to pass that along 2015-03-27T13:09:02 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-27T13:10:10 gwty: yes.. all the termios0* tests use one of a couple of test drivers. One simulates the interrupt driven part, the other polled. At least that's what I remember. For sure, maybe not a perfect example but at least it is one 2015-03-27T13:10:32 melange is soon to get overloaded if history teaches us anything. :) 2015-03-27T13:11:16 *** DrJoel has quit IRC (Quit: Leaving) 2015-03-27T13:18:02 *** hermann_19829 has quit IRC (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.) 2015-03-27T13:19:44 *** DrJoel has joined #rtems 2015-03-27T13:19:44 *** ChanServ sets mode: +o DrJoel 2015-03-27T13:20:33 gedare: You probably noticed but I wanted to make sure you knew I pushed the RSB changes to move to the newlib snapshot and the changes to move sparc gdb to 7.9 with Jiri's patches 2015-03-27T13:20:56 Everything builds very cleanly. Only a few warnings left if you ignore avr. One 16-bit warning left 2015-03-27T13:21:27 i saw. that's good 2015-03-27T13:21:32 I am cleaning up rtems-testing now since I had a number of local changes. I want to add support to sim-scripts and rtems-tester for the leon2/leon3 using gdb 2015-03-27T13:21:35 my feeling is we should cut and run ;) 2015-03-27T13:22:03 I was going to post asking that Monday. Friday email gets ignored. 2015-03-27T13:22:11 I don't know of much else that needs doing. 2015-03-27T13:22:19 moxie will get fixed when moxie gets fixed 2015-03-27T13:22:21 it's either that, or push the leon code in and re-do the sparc build/test 2015-03-27T13:22:46 If it gets pushed, it won't take long to do that. 2015-03-27T13:23:10 I honestly would rather have it in. 2015-03-27T13:23:12 *** ketul has joined #rtems 2015-03-27T13:27:54 *** DrJoel has quit IRC (Quit: Leaving) 2015-03-27T13:37:57 Do you mean this DrJoel https://devel.rtems.org/changeset/bfd2b7d/rtems-tools ? Thanks again for pointing me to termios tests 2015-03-27T13:46:38 [trac] GSoC/2015 edited by waleedziad (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-03-27T13:55:21 too bad nobody picked up the x86_64 bsp :( 2015-03-27T13:55:22 oh well 2015-03-27T13:55:48 verm__: heh, someone just commented on it on the mailing list, too little too late i suspect 2015-03-27T13:55:57 oh? 2015-03-27T13:56:02 the devel list? 2015-03-27T13:56:08 i don't think it's too late to get a good proposal in! 2015-03-27T13:56:23 eh, 3 minutes... 2015-03-27T13:56:29 feel free to respond there 2015-03-27T13:57:21 :) 2015-03-27T13:57:33 oh, strange i didn't get that email 2015-03-27T13:58:19 or maybe i did and deleted it, nuts 2015-03-27T13:58:51 5th year student is pretty good 2015-03-27T14:01:50 I couldn't update my proposal in time :( 2015-03-27T14:02:33 are submissions closed already? 2015-03-27T14:02:34 gwty: you can continue to improve your gdoc version. 2015-03-27T14:02:36 verm__: yes. 2015-03-27T14:02:38 gedare: deadline is reached .... 2015-03-27T14:02:40 oh 2015-03-27T14:02:40 2 minutes ago 2015-03-27T14:02:48 oh well so no x86_64 bsp then 2015-03-27T14:03:56 gedare: do you guys also check gdoc version? 2015-03-27T14:04:24 it's the only one we look at 2015-03-27T14:04:42 oh ok 2015-03-27T14:04:53 there is no way to track changes using the melange site 2015-03-27T14:05:21 hmmm then fine 2015-03-27T14:05:48 yup. the melange one we use for some reference, and it can be used by Google, but the gdoc is the "living proposal" 2015-03-27T14:06:25 cool :) 2015-03-27T14:06:54 thanks gedare for patiently answering our questions 2015-03-27T14:07:11 Ok gedare, thank you for that information 2015-03-27T14:07:13 hey sure, be sure to mention how awesome I am when you do surveys. ;) 2015-03-27T14:09:26 Yes sure :) 2015-03-27T14:09:47 gedare: definitely :) 2015-03-27T14:15:18 totally 4425 applications 2015-03-27T14:17:12 time to go to sleep ... bye 2015-03-27T14:17:16 *** ragunath has quit IRC () 2015-03-27T14:18:31 heh yeah 2015-03-27T14:18:38 sigh i wish the intel galileo was 64bit 2015-03-27T14:18:59 feel free to talk to the student about re-orienting toward x86_64 2015-03-27T14:19:10 oh? nice 2015-03-27T14:19:11 emailing 2015-03-27T14:19:12 galileo has...problems. 2015-03-27T14:19:15 yeah 2015-03-27T14:19:25 yup go ahead, CC me and Joel 2015-03-27T14:19:28 verm__: ^ 2015-03-27T14:23:26 verm__: intel edison is 64 bits 2015-03-27T14:23:54 doesn't edison suffer from the same problem as galileo? poor documentation and hardware debugging? 2015-03-27T14:24:43 also when making an x86_64 bsp it won't matter what board is used 2015-03-27T14:24:51 edison and galileo are both problematic. 2015-03-27T14:24:51 it should just work on all of them 2015-03-27T14:25:15 ok 2015-03-27T14:29:39 just curious, are we allowed to improve our proposal in gdoc even though submissions are closed? 2015-03-27T14:30:15 yes 2015-03-27T14:30:32 *** krohini has quit IRC (Quit: ChatZilla 0.9.91.1 [Firefox 35.0.1/20150125222139]) 2015-03-27T14:30:39 if org allow us. 2015-03-27T14:30:54 verm__: thanks 2015-03-27T14:31:00 you can improve your gdoc's. we won't open edits in Melange for various reasons 2015-03-27T14:35:37 verm__: there are 2 waf repos, one in your git and other in chris. are they different? 2015-03-27T14:36:46 verm__: i don't see the clang project in melange, the student may have missed the deadline.. 2015-03-27T14:37:06 i sent an email reminder 2 days ago. 2015-03-27T14:37:12 seriously? oh well 2015-03-27T14:37:37 she had someone from clang to mentor her, too 2015-03-27T14:37:57 yeah 2015-03-27T14:40:25 that happens sometimes though, students completely miss the deadline not realising when it is 2015-03-27T14:40:33 or that it's midnight or don't know when 6pm UTC is 2015-03-27T14:40:50 maybe a lesson is learned. 2015-03-27T14:44:58 *** gwty has quit IRC (Remote host closed the connection) 2015-03-27T14:45:06 *** gwty has joined #rtems 2015-03-27T14:49:01 18 applications. uptick from last year. 2015-03-27T14:49:14 makes sense given the smaller pool of orgs. 2015-03-27T14:49:34 gedare: Hi, I just want to make sure that I'm not the one who withdrawn his proposal. I accidentally hit the withdraw button 2015-03-27T14:51:31 ha! no it's still there 2015-03-27T14:51:41 i'm guessing you have to confirm it withdrawn 2015-03-27T14:52:13 i don't see the widthdrawn one anymore 2015-03-27T14:52:39 Good, the button coverted to "Yes", and immediately I returned it back no "No" 2015-03-27T14:52:45 to* 2015-03-27T14:58:53 what is a rump kernel? 2015-03-27T14:59:13 heshams project is that. 2015-03-27T14:59:25 anandkp92: Yeah 2015-03-27T15:00:11 Hesham: sorry, im unaware of it. could you give a base idea? 2015-03-27T15:00:28 anandkp92: Sure http://heshamelmatary.blogspot.co.uk/2015/02/thoughts-on-supporting-rump-kernels-on.html 2015-03-27T15:04:15 Hesham: thanks! 2015-03-27T15:04:32 Welcome 2015-03-27T15:18:28 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-27T15:29:58 *** hermann_19829 has joined #rtems 2015-03-27T15:48:26 *** Hesham has quit IRC (Ping timeout: 264 seconds) 2015-03-27T15:48:45 *** hermann_19829 has quit IRC (Remote host closed the connection) 2015-03-27T15:55:56 *** gedare has quit IRC (Ping timeout: 252 seconds) 2015-03-27T15:56:11 *** jrcatbagan has quit IRC (Quit: nah nah nah nah, nah nah nah nah, hey hey hey, goodbye) 2015-03-27T16:23:26 *** gedare has joined #rtems 2015-03-27T16:23:26 *** ChanServ sets mode: +o gedare 2015-03-27T16:27:28 *** gwty has quit IRC (Ping timeout: 256 seconds) 2015-03-27T16:42:21 *** gedare has quit IRC (Ping timeout: 256 seconds) 2015-03-27T17:35:51 *** jrcatbagan has joined #rtems 2015-03-27T18:01:22 [git] da801b0 by Chris Johns: [rtems: Change the base URL for RTEMS git to https to avoid the redirects.] http://devel.rtems.org/changeset/da801b0/rtems-source-builder 2015-03-27T18:01:24 gedare, hi 2015-03-27T18:04:55 *** anandkp92 has quit IRC (Ping timeout: 246 seconds) 2015-03-27T18:57:29 *** dr__house|away is now known as dr__house 2015-03-27T20:24:54 *** waleedziad has quit IRC (Ping timeout: 256 seconds) 2015-03-27T21:11:54 *** gedare has joined #rtems 2015-03-27T21:11:54 *** ChanServ sets mode: +o gedare 2015-03-27T21:12:02 hi kiwichris 2015-03-27T21:12:13 hi 2015-03-27T21:12:36 Well done on getting all the proposals we have in. 2015-03-27T21:12:54 thanks. it was a sprint 2015-03-27T21:13:03 i'm actually most proud of the number of mentors registered ;) 2015-03-27T21:13:11 registered...wrangled 2015-03-27T21:16:12 should be a great, hectic summer. especially if we get as many slots as we asked for. 2015-03-27T21:19:49 *** gwty has joined #rtems 2015-03-27T21:22:39 gedare: yes great work you did a really great job especially with the crush near the end 2015-03-27T21:22:57 i'm just glad we have 3 weeks to sort it all out! 2015-03-27T21:23:05 yeah 2015-03-27T21:23:10 well a lot of projects got a ton of students 2015-03-27T21:23:20 gsoc is still fairly popular but less orgs this year 2015-03-27T21:23:48 yeah the stats show consistent number of students, about 4400 2015-03-27T21:24:00 but fewer orgs than the past 5 years i think 2015-03-27T21:24:32 way fewer 2015-03-27T21:24:37 "numapps" is 2015: 6409 proposals (by 4425 students). 137/416 orgs. 2014: 190/371 orgs; 1304/6313 proposals (by 4420 students). 2013: 177/417 orgs; 1192/5999 proposals (by 4144 students). 2012: 180/406 orgs; 1212/6685 proposals (by 4258 students). 2011: 175/417 orgs; 1116/5474 proposals (by 3731 students). 2010: 151/367 orgs; 1026/5539 proposals (by 3464 students). 2015-03-27T21:44:59 fairly large drop for the amount of students 2015-03-27T21:45:30 when? 2015-03-27T21:45:36 there are 5 more students than last year 2015-03-27T21:45:41 i mean orgs 2015-03-27T21:45:43 ahh. 2015-03-27T21:45:46 gotcha 2015-03-27T21:45:54 some projects that used to get 1-2 students got 10+ this year 2015-03-27T21:46:00 yes it will be interesting to see if slots are proportional to students or to orgs 2015-03-27T21:46:11 maybe that was the point i don't know we'll have to see how many slots are out that may drop by a lot, too 2015-03-27T21:46:19 exactly 2015-03-27T22:18:48 *** gedare has quit IRC (Ping timeout: 256 seconds) 2015-03-27T23:51:39 *** waleedziad has joined #rtems 2015-03-28T01:04:05 *** gwty has quit IRC (Quit: AndroIRC - Android IRC Client ( http://www.androirc.com )) 2015-03-28T01:04:27 *** gwty has joined #rtems 2015-03-28T01:04:30 *** gwty has left #rtems 2015-03-28T01:04:48 *** gwty has joined #rtems 2015-03-28T02:13:56 *** hermann_19829 has joined #rtems 2015-03-28T03:04:58 *** gwty has quit IRC (Remote host closed the connection) 2015-03-28T03:11:08 *** hermann_19829 has quit IRC (Remote host closed the connection) 2015-03-28T03:11:32 *** gwty has joined #rtems 2015-03-28T04:00:41 *** ketul has joined #rtems 2015-03-28T04:52:34 *** anandkp92 has joined #rtems 2015-03-28T04:54:31 *** Hesham has joined #rtems 2015-03-28T05:37:40 *** Hesham has quit IRC (Quit: Leaving.) 2015-03-28T05:58:13 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-28T05:58:55 *** ketul has joined #rtems 2015-03-28T06:44:10 *** gedare has joined #rtems 2015-03-28T06:44:11 *** ChanServ sets mode: +o gedare 2015-03-28T07:06:00 *** gedare has quit IRC (Ping timeout: 252 seconds) 2015-03-28T07:12:18 *** gedare has joined #rtems 2015-03-28T07:12:18 *** ChanServ sets mode: +o gedare 2015-03-28T08:00:52 *** gedare has quit IRC (Ping timeout: 255 seconds) 2015-03-28T08:19:12 *** ragunath has joined #rtems 2015-03-28T08:58:34 *** gwty has quit IRC (Ping timeout: 245 seconds) 2015-03-28T09:00:15 *** gwty has joined #rtems 2015-03-28T09:49:18 *** gwty has quit IRC (Ping timeout: 244 seconds) 2015-03-28T09:58:31 *** gwty has joined #rtems 2015-03-28T10:30:03 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-28T10:30:09 *** ragunath has quit IRC (Quit: Page closed) 2015-03-28T13:22:43 *** anandkp92 has quit IRC (Ping timeout: 246 seconds) 2015-03-28T13:30:59 *** ketul has joined #rtems 2015-03-28T13:37:50 *** gedare has joined #rtems 2015-03-28T13:37:51 *** ChanServ sets mode: +o gedare 2015-03-28T14:05:24 *** anandkp92 has joined #rtems 2015-03-28T14:20:17 *** gedare has quit IRC (Ping timeout: 246 seconds) 2015-03-28T14:30:48 *** gedare has joined #rtems 2015-03-28T14:30:48 *** ChanServ sets mode: +o gedare 2015-03-28T14:43:37 *** gedare has quit IRC (Ping timeout: 264 seconds) 2015-03-28T14:52:02 *** gwty has quit IRC (Ping timeout: 252 seconds) 2015-03-28T15:11:14 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-28T15:37:26 *** jrcatbagan has joined #rtems 2015-03-28T17:16:31 *** anandkp92 has quit IRC (Ping timeout: 246 seconds) 2015-03-28T19:22:48 *** jrcatbagan has joined #rtems 2015-03-28T19:35:09 *** jrcatbagan has quit IRC (Quit: nah nah nah nah, nah nah nah nah, hey hey hey, goodbye) 2015-03-28T20:31:25 *** gedare has joined #rtems 2015-03-28T20:31:25 *** ChanServ sets mode: +o gedare 2015-03-28T20:38:13 *** gedare has quit IRC (Ping timeout: 255 seconds) 2015-03-28T22:08:02 *** gwty has joined #rtems 2015-03-28T23:37:28 [git] 3237c8e by Chris Johns: [sb: Provide an unverified SSL context to the URL. The RTEMS servers are causing an exception when downloading patches. The solution is provided in PEP-0476 (https://www.python.org/dev/peps/pep-0476/#opting-out).] http://devel.rtems.org/changeset/3237c8e/rtems-source-builder 2015-03-28T23:41:34 [trac] WikiStart edited by chrisj (Add tracing.) http://devel.rtems.org/wiki/WikiStart 2015-03-28T23:50:12 [trac] History/Timeline edited by chrisj (Reformatted) http://devel.rtems.org/wiki/History/Timeline 2015-03-28T23:51:25 [trac] WikiStart edited by chrisj (Update Timeline to Historical Timeline.) http://devel.rtems.org/wiki/WikiStart 2015-03-28T23:57:35 [trac] Mission_Statement edited by chrisj (Clean up) http://devel.rtems.org/wiki/Mission_Statement 2015-03-28T23:58:03 *** gwty has quit IRC (Read error: Connection reset by peer) 2015-03-28T23:58:09 [trac] WikiStart edited by chrisj (Update Mission statement link.) http://devel.rtems.org/wiki/WikiStart 2015-03-28T23:59:16 [trac] WikiStart edited by chrisj (Split line.) http://devel.rtems.org/wiki/WikiStart 2015-03-29T00:02:59 *** gwty has joined #rtems 2015-03-29T00:04:21 [trac] Developer/Tracing edited by chrisj (Add link to Trace Buffering page.) http://devel.rtems.org/wiki/Developer/Tracing 2015-03-29T00:24:56 *** ketul has joined #rtems 2015-03-29T00:28:28 [trac] Developer/Tracing/Trace_Linker edited by chrisj (Update the generator documentation.) http://devel.rtems.org/wiki/Developer/Tracing/Trace_Linker 2015-03-29T00:37:23 *** dr__house is now known as dr__house|away 2015-03-29T00:37:32 [git] 5032d71 by Chris Johns: [trace-linker: Update comments.] http://devel.rtems.org/changeset/5032d71/rtems-tools 2015-03-29T01:48:31 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-29T01:49:16 *** ketul has joined #rtems 2015-03-29T02:07:44 [git] b249516 by Chris Johns: [rtemstoolkit: Add support to return the system path split as paths. Seacch the path for the program name if not found and set it as an absolute path. This allow the prefix to be found.] http://devel.rtems.org/changeset/b249516/rtems-tools 2015-03-29T02:26:11 *** gwty has quit IRC (Ping timeout: 265 seconds) 2015-03-29T02:28:44 *** gwty has joined #rtems 2015-03-29T03:00:44 [trac] Developer/Tracing/Trace_Linker edited by chrisj (Update limitations.) http://devel.rtems.org/wiki/Developer/Tracing/Trace_Linker 2015-03-29T04:26:03 *** gwty has quit IRC (Read error: Connection reset by peer) 2015-03-29T04:55:23 *** Hesham has joined #rtems 2015-03-29T05:08:58 [trac] Developer/Tracing/Trace_Buffering edited by chrisj (Add the fileio example.) http://devel.rtems.org/wiki/Developer/Tracing/Trace_Buffering 2015-03-29T05:10:00 [trac] Developer/Tracing/Trace_Buffering edited by chrisj (Minor fixes.) http://devel.rtems.org/wiki/Developer/Tracing/Trace_Buffering 2015-03-29T05:14:58 [trac] Developer/Tracing/Trace_Buffering edited by chrisj (Add extra trace options.) http://devel.rtems.org/wiki/Developer/Tracing/Trace_Buffering 2015-03-29T05:19:56 [trac] fileio-trace.ini (Trace buffering fileio configuration file.) attached to #Developer/Tracing/Trace_Buffering by chrisj http://devel.rtems.org/ticket/Developer/Tracing/Trace_Buffering 2015-03-29T05:22:48 [trac] Developer/Tracing/Trace_Buffering edited by chrisj (Add the attachment link) http://devel.rtems.org/wiki/Developer/Tracing/Trace_Buffering 2015-03-29T05:26:59 [trac] Developer/Tracing/Trace_Buffering edited by chrisj (Minor fixes.) http://devel.rtems.org/wiki/Developer/Tracing/Trace_Buffering 2015-03-29T05:39:51 *** Hesham has quit IRC (Quit: Leaving.) 2015-03-29T05:53:50 *** punitvara has joined #rtems 2015-03-29T05:53:59 hello 2015-03-29T05:55:15 for creating new BSP , at the time of building environment which .bset file should choose ? 2015-03-29T05:55:52 as I am creating new BSP for x86 ,there no supported file 2015-03-29T05:56:03 so i choose rtems-base file 2015-03-29T05:56:13 but it shows error 2015-03-29T05:56:14 any help ? 2015-03-29T06:01:26 *** punitvara has quit IRC (Quit: Page closed) 2015-03-29T06:01:49 *** punitvara has joined #rtems 2015-03-29T06:01:56 hello [03:25] for creating new BSP , at the time of building environment which .bset file should choose ? [03:25] as I am creating new BSP for x86 ,there no supported file [03:26] so i choose rtems-base file [03:26] but it shows error [03:26] any help ? 2015-03-29T06:25:04 *** punitvara has quit IRC (Ping timeout: 246 seconds) 2015-03-29T07:00:20 *** waleedziad has quit IRC (Quit: Leaving) 2015-03-29T08:18:21 *** ragunath has joined #rtems 2015-03-29T08:21:38 gravex34 2015-03-29T08:37:01 *** ragunath has quit IRC (Ping timeout: 246 seconds) 2015-03-29T09:10:57 *** Hesham has joined #rtems 2015-03-29T09:14:56 *** ragunath has joined #rtems 2015-03-29T09:29:16 *** ketul has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2015-03-29T09:30:48 *** Hesham has quit IRC (Ping timeout: 272 seconds) 2015-03-29T09:54:32 *** Hesham has joined #rtems 2015-03-29T10:39:10 *** ragunath has quit IRC (Ping timeout: 246 seconds) 2015-03-29T10:41:47 *** Hesham has quit IRC (Ping timeout: 246 seconds) 2015-03-29T11:29:21 *** Hesham has joined #rtems 2015-03-29T11:35:10 *** ragunath has joined #rtems 2015-03-29T11:35:19 *** ragunath has quit IRC (Client Quit) 2015-03-29T11:37:23 [git] e98d5fa by Joel Sherrill: [Add configurations for erc32, leon2, and leon3 in GDB SIS These require a GDB >= 7.9 with Jiri Gaisler's patches applied. These are currently in the RSB and are in the process of being merged by the GDB project.] http://devel.rtems.org/changeset/e98d5fa/rtems-tools 2015-03-29T11:39:36 [git] e197b8b by Joel Sherrill: [bit_all_bsps: Add or1k and fix issue for distributed builds These scripts support using GNU parallel to distribute the build across multiple computers with a shared source directory.] http://devel.rtems.org/changeset/e197b8b/rtems-testing 2015-03-29T11:39:36 [git] 22a9856 by Joel Sherrill: [bit_all_multilib: Add multiple CPUs including moxie and or1k] http://devel.rtems.org/changeset/22a9856/rtems-testing 2015-03-29T11:39:36 [git] 96c37b3 by Joel Sherrill: [bit_rtems: Fix bug for exit status and make cleaning build directory optional] http://devel.rtems.org/changeset/96c37b3/rtems-testing 2015-03-29T11:39:36 [git] 3c7ecee by Joel Sherrill: [common.sh: Update list of CPUs and clean up] http://devel.rtems.org/changeset/3c7ecee/rtems-testing 2015-03-29T11:39:36 [git] b88ed3f by Joel Sherrill: [rtems/user.cfg: Add options for cleaning on exit and parallel build options] http://devel.rtems.org/changeset/b88ed3f/rtems-testing 2015-03-29T11:39:37 [git] 630dda1 by Joel Sherrill: [sim-scripts/or1ksim.in: Update name of simulator] http://devel.rtems.org/changeset/630dda1/rtems-testing 2015-03-29T11:39:37 [git] 045bcdb by Joel Sherrill: [sim-scripts/gdb-sim-run.in: Silence a useless message] http://devel.rtems.org/changeset/045bcdb/rtems-testing 2015-03-29T11:39:38 [git] 6568994 by Joel Sherrill: [Rename leon2 and leon3 scripts to indicate they use tsim] http://devel.rtems.org/changeset/6568994/rtems-testing 2015-03-29T11:39:38 [git] 12c31d1 by Joel Sherrill: [Add support for erc32, leon2 and leon3 in GDB SIS] http://devel.rtems.org/changeset/12c31d1/rtems-testing 2015-03-29T11:43:00 *** DrJoel has joined #rtems 2015-03-29T11:43:00 *** DrJoel has joined #rtems 2015-03-29T11:43:00 *** ChanServ sets mode: +o DrJoel 2015-03-29T11:43:17 *** DrJoel has quit IRC (Client Quit) 2015-03-29T12:26:17 *** Hesham has quit IRC (Ping timeout: 250 seconds) 2015-03-29T12:58:14 *** gwty has joined #rtems 2015-03-29T15:09:28 *** dr__house|away is now known as dr__house 2015-03-29T16:35:08 *** kiwichris has quit IRC (Ping timeout: 272 seconds) 2015-03-29T16:48:52 *** kiwichris has joined #rtems 2015-03-29T17:43:23 [trac] #2315/defect (v:4.11) created by joel.sherrill (rtems-tester support for pc386) http://devel.rtems.org/ticket/2315 2015-03-29T18:10:12 [trac] #2315/defect (new) updated by chrisj (A key part of doing this is getting a single boot loader. It mean both sides of the ...) http://devel.rtems.org/ticket/2315 2015-03-29T22:14:49 *** gwty has quit IRC (Quit: AndroIRC - Android IRC Client ( http://www.androirc.com )) 2015-03-29T22:31:43 [git] bd16849 by Chris Johns: [sb: Update the downloader for 2.7.8 and earlier without a context.] http://devel.rtems.org/changeset/bd16849/rtems-source-builder 2015-03-29T23:15:35 *** dr__house is now known as dr__house|away