2015-07-27T00:32:36 *** dr__hous1|away is now known as dr__house 2015-07-27T00:32:49 *** dr__house has joined #rtems 2015-07-27T00:34:39 *** Davidbrcz has joined #rtems 2015-07-27T01:26:48 *** Davidbrcz has quit IRC 2015-07-27T03:34:25 *** dr__house is now known as dr__house|away 2015-07-27T04:38:31 *** monstr has joined #rtems 2015-07-27T05:19:08 *** kiwichris has quit IRC 2015-07-27T05:28:31 *** kiwichris has joined #rtems 2015-07-27T05:32:38 *** Davidbrcz has joined #rtems 2015-07-27T07:10:44 *** kiwichris has quit IRC 2015-07-27T07:11:42 *** sebhub has joined #rtems 2015-07-27T07:17:21 *** Davidbrcz has quit IRC 2015-07-27T07:43:33 good morning 2015-07-27T08:03:50 *** ketul has joined #rtems 2015-07-27T08:26:45 *** toxygen has joined #rtems 2015-07-27T08:27:00 hello, is there a way to force agressive round robin scheduling? 2015-07-27T08:27:16 i.e. if there are several tasks in the same priority group, every one is allocated some time on cpu? 2015-07-27T08:27:30 (without yielding from any task) 2015-07-27T08:31:00 you may use the timeslicing option 2015-07-27T08:34:19 ok, thx 2015-07-27T10:19:50 *** peerst_ has joined #rtems 2015-07-27T10:21:55 *** peerst has quit IRC 2015-07-27T10:21:57 *** guerby has quit IRC 2015-07-27T10:22:33 *** peerst_ is now known as peerst 2015-07-27T10:32:57 *** guerby has joined #rtems 2015-07-27T10:56:10 *** crazy3k has joined #rtems 2015-07-27T11:55:49 *** antgreen has quit IRC 2015-07-27T12:34:59 *** beng-nl_ has joined #rtems 2015-07-27T12:36:39 *** beng-nl has quit IRC 2015-07-27T14:03:38 *** gedare has joined #rtems 2015-07-27T14:03:44 *** ChanServ sets mode: +o gedare 2015-07-27T14:08:02 [trac] #2375/defect (new) updated by gedare ( Sebastian or Chris should review this fix. Pulling in all the changes since 4.11 ...) http://devel.rtems.org/ticket/2375 2015-07-27T14:14:14 *** ketul has quit IRC 2015-07-27T14:14:33 *** ketul has joined #rtems 2015-07-27T14:19:10 *** dr__house|away is now known as dr__house 2015-07-27T15:13:48 *** Davidbrcz has joined #rtems 2015-07-27T15:21:10 *** Davidbrcz has quit IRC 2015-07-27T15:35:15 *** monstr has quit IRC 2015-07-27T16:00:53 *** Davidbrcz has joined #rtems 2015-07-27T18:03:45 *** dougk_ has joined #rtems 2015-07-27T18:31:44 *** ketul has quit IRC 2015-07-27T18:34:14 *** ketul has joined #rtems 2015-07-27T19:19:47 [trac] #2376/defect (v:4.10) created by mdavidsaver (close() of tftpfs always returns an error) http://devel.rtems.org/ticket/2376 2015-07-27T19:23:09 [trac] #2375/defect (new) updated by mdavidsaver (Replying to [comment:3 gedare]: > ... Pulling in all the changes since 4.11 ...) http://devel.rtems.org/ticket/2375 2015-07-27T19:56:41 [trac] #2375/defect (new) updated by gedare (Replying to [comment:4 mdavidsaver]: > Replying to [comment:3 gedare]: > > ... ...) http://devel.rtems.org/ticket/2375 2015-07-27T19:57:17 [trac] #2375/defect (new) updated by gedare (P.S. perhaps you can use `git-rebase` to squash the first two commits together? This ...) http://devel.rtems.org/ticket/2375 2015-07-27T20:03:34 *** Davidbrcz has quit IRC 2015-07-27T20:26:47 *** Davidbrcz has joined #rtems 2015-07-27T20:41:26 *** Davidbrcz has quit IRC 2015-07-27T21:04:47 *** ketul has quit IRC 2015-07-27T21:09:24 *** gedare has quit IRC 2015-07-27T21:32:14 *** mumptai has quit IRC 2015-07-27T21:32:51 *** mumptai has joined #rtems 2015-07-27T22:24:05 *** kiwichris has joined #rtems 2015-07-28T00:41:31 [trac] #2375/defect (new) updated by mdavidsaver (Replying to [comment:6 gedare]: > P.S. perhaps you can use `git-rebase` to squash ...) http://devel.rtems.org/ticket/2375 2015-07-28T01:07:43 [trac] #2375/defect (new) updated by mdavidsaver (Replying to [comment:5 gedare]: > Replying to [comment:4 mdavidsaver]: > ... > ...) http://devel.rtems.org/ticket/2375 2015-07-28T02:27:32 *** dr__house is now known as dr__house|away 2015-07-28T03:51:06 *** Davidbrcz has joined #rtems 2015-07-28T04:14:55 *** Davidbrcz has quit IRC 2015-07-28T04:38:50 *** monstr has joined #rtems 2015-07-28T06:36:29 good morning 2015-07-28T06:40:44 *** stryx`_ has quit IRC 2015-07-28T06:41:24 *** stryx`_ has joined #rtems 2015-07-28T07:18:09 *** kiwichris has quit IRC 2015-07-28T07:28:03 *** kiwichris has joined #rtems 2015-07-28T07:34:20 sebhub, hi 2015-07-28T07:35:27 I know the lock patch is simple and "internal" but I still wonder about adding _RTEMS_ as a prefix to ensure we do not clash with anything in future. 2015-07-28T07:36:09 This is my major concern. I see the API and good and powerful and useful. 2015-07-28T07:52:58 all names start with an underscore, this should be enough, e.g. similar to everything else in the score 2015-07-28T07:54:18 in addition the _Package_Name naming convention is quite unique to RTEMS, I have seen it nowhere else 2015-07-28T08:03:31 *** kiwichris has quit IRC 2015-07-28T08:50:24 *** stryx`_ is now known as stryx` 2015-07-28T09:47:24 *** hesham has joined #rtems 2015-07-28T11:16:31 *** asuolen has joined #rtems 2015-07-28T11:20:19 *** ketul has joined #rtems 2015-07-28T13:16:28 *** hesham has quit IRC 2015-07-28T13:16:38 [git] 21789a2 by Sebastian Huber: [score: Rename _POSIX_Absolute_timeout_to_ticks() Rename _POSIX_Absolute_timeout_to_ticks() to _TOD_Absolute_timeout_to_ticks() and move it to the score directory. Delete empty .] http://devel.rtems.org/changeset/21789a2/rtems 2015-07-28T13:44:09 *** stryx` has quit IRC 2015-07-28T13:46:50 *** hesham has joined #rtems 2015-07-28T13:47:07 *** stryx` has joined #rtems 2015-07-28T13:48:14 *** hermann_19829 has joined #rtems 2015-07-28T13:48:34 *** hermann_19829 has left #rtems 2015-07-28T13:55:57 *** hermann_19829_ has joined #rtems 2015-07-28T13:56:10 *** hermann_19829_ has left #rtems 2015-07-28T13:58:27 [trac] SOCIS/2015 edited by hermann19829 (empty comment) http://devel.rtems.org/wiki/SOCIS/2015 2015-07-28T14:27:38 *** gedare has joined #rtems 2015-07-28T14:27:38 *** ChanServ sets mode: +o gedare 2015-07-28T14:50:10 *** anandkp has joined #rtems 2015-07-28T14:50:12 *** gtament has joined #rtems 2015-07-28T14:50:18 Hi!) 2015-07-28T14:50:30 Hi :) 2015-07-28T14:51:02 *** javamonn has joined #rtems 2015-07-28T14:51:29 hello 2015-07-28T14:52:36 *** ragunath has joined #rtems 2015-07-28T14:54:33 @gedare: Hi gedare, once everyone comes, could i go early today as my network is very flaky now and i don't know if it'll last? 2015-07-28T14:54:33 anandkp: Error: "gedare:" is not a valid command. 2015-07-28T14:54:49 * Hi gedare, once everyone comes, could i go early today as my network is very flaky now and i don't know if it'll last? 2015-07-28T14:54:50 *** hermann_19829_ has joined #rtems 2015-07-28T14:54:50 *** jrcatbagan has joined #rtems 2015-07-28T14:55:00 *** salpha has joined #rtems 2015-07-28T14:56:52 *** krohini has joined #rtems 2015-07-28T14:56:54 yes anandkp. 2015-07-28T14:57:18 hi gedare 2015-07-28T14:57:28 Thanks :) 2015-07-28T14:57:29 hi krohini, and everyone else. 2015-07-28T14:57:35 anandkp: you can start. ping verm__. 2015-07-28T14:57:40 hello :) 2015-07-28T14:57:51 hi all! 2015-07-28T14:57:54 Hi. So currently when the user opens the gui, he has to choose which bsps he wants to configure for 2015-07-28T14:58:03 hi gedare 2015-07-28T14:58:06 After that I've coded, using existing functions to get only the required parameters for those bsps - they are currently being displayed. 2015-07-28T14:58:18 hi all 2015-07-28T14:58:25 Now I'm able to get the value from the gui. Working on updating the configuration with it. 2015-07-28T14:58:39 I'm able to create the config.cfg file too 2015-07-28T14:58:45 hi everyone 2015-07-28T15:00:22 ok anandkp, so next up you are going to try to build with that config? 2015-07-28T15:01:14 *** sgworks has joined #rtems 2015-07-28T15:01:32 anandkp: and everyone else too, please make sure to update your status notes on the tracking page 2015-07-28T15:02:04 That's it for now. I haven't got any feedback on last two parts yet. 2015-07-28T15:02:07 *** anandkp has quit IRC 2015-07-28T15:02:18 *** anandkp has joined #rtems 2015-07-28T15:02:33 looks like he is having trouble 2015-07-28T15:02:45 so we'll start on asuolen 2015-07-28T15:02:55 hello 2015-07-28T15:02:58 Yesterday I have posted in the devel ml the v2 of the GPIO patch set for RTEMS and the Pi. In the meantime I have an initial port of my I2C Pi code to the linux I2C framework, which I am testing and ironing out some issues. 2015-07-28T15:03:08 I say initial port because the new API allows for a number of small features that libi2c did not have such as the mangling functionality (i.e.: ignoring acknowledgement for certain messages), sending multiple messages (each may include several data packets) without start conditions in between and such. The device driver I have for my I2C testing device has also been ported to use the new framework's device API so the work can be tes 2015-07-28T15:03:15 SPI is mainly waiting for the GPIO to get in, and I hope to have a go at the SD card during the weekend. 2015-07-28T15:03:56 i skimmed your patches and think it looks good to me, i plan to hold off on merging it though until the 4.11 release settles i think 2015-07-28T15:04:16 ok 2015-07-28T15:04:28 with pin groupings, does that only work within one bank? 2015-07-28T15:04:47 each pin type can be only within a single bank 2015-07-28T15:04:55 so all inputs pins have to be within one bank 2015-07-28T15:05:06 but output may be in another 2015-07-28T15:05:19 ok 2015-07-28T15:05:21 but still belong to the same group 2015-07-28T15:05:30 and can share one isr? 2015-07-28T15:06:25 multiple pins can share the same ISR, by defining the same handler function. This at a GPIO level 2015-07-28T15:06:49 Each bank only has a single ISR "shared" by all the bank's pins 2015-07-28T15:07:08 for the i2c, probably it's ok to start with a basic driver and add the other features incrementally? or does the linux-based i2c framework need to have all features implemented to work 2015-07-28T15:07:21 not at all 2015-07-28T15:07:21 okay 2015-07-28T15:07:40 the linux API leaves for the bus to define which functionalities it suppoorts 2015-07-28T15:07:47 sounds good, if you don't hear anything back on your two patches within a week, ping them 2015-07-28T15:07:53 so a driver can now if the bus has what is need to execute 2015-07-28T15:08:05 ah, ok that makes sense 2015-07-28T15:08:25 you have the SPI code ready? 2015-07-28T15:08:47 you can assume the GPIO code will get committed as-is 2015-07-28T15:08:55 yes the SPI is mainly ready, just some refactoring/cleanup and final testing 2015-07-28T15:09:00 okay 2015-07-28T15:09:06 its pretty much the same as last year 2015-07-28T15:09:10 right. 2015-07-28T15:09:11 it still uses libi2c 2015-07-28T15:09:23 ok thanks asuolen anything else? 2015-07-28T15:09:28 ketul: you can go next 2015-07-28T15:09:35 hi gedare 2015-07-28T15:09:36 no that is all, thanks 2015-07-28T15:09:37 I referred the TRM for I2C interfacing and also got myself used to with libi2c api available. Analysed the work done by Claas Ziemke for beaglebone black. Also did communication with him for my some issues.After that I modified his work as required. So uptil now I almost done with the i2c driver except adding some functions,tests and reviews. So my 2015-07-28T15:09:37 current work can be catched at https://gist.github.com/ketul93/a8cdde82c6d946de1d98 . And for testing purpose I managed one adxl345 sensor which can be interfaced via I2C and it will be cool for demo. I am able to interface that sensor with my raspberry pi (raspbian OS) through I2C and got the desired output. 2015-07-28T15:10:13 and yes gedare I noted your chat with asuolen about i2c :) 2015-07-28T15:10:59 ok it's always good to know your hardware works before testing your software. 2015-07-28T15:11:13 yep I did that 2015-07-28T15:11:25 [trac] SOCIS/2015 edited by salpha (empty comment) http://devel.rtems.org/wiki/SOCIS/2015 2015-07-28T15:11:50 ok so you're going to be finishing your libi2c driver soon? 2015-07-28T15:12:01 yep probably this week 2015-07-28T15:12:11 with testing 2015-07-28T15:12:28 and then i want you to move on to /dev/i2c, and if we get asuolen's GPIO patch in then you can refactor your GPIO code to fit the framework. 2015-07-28T15:12:59 yes after merging his code I will do that 2015-07-28T15:13:08 alright. 2015-07-28T15:13:16 thanks ketul anything else? 2015-07-28T15:13:19 gtament: you can prepare to go next 2015-07-28T15:13:26 no thanks gedare 2015-07-28T15:13:28 I'm typing) 2015-07-28T15:14:19 ok ketul 2015-07-28T15:14:34 [trac] GSoC/2015 edited by jrcatbagan (Added status update for June 21, 2015) http://devel.rtems.org/wiki/GSoC/2015 2015-07-28T15:16:16 So I've runned qemu with testsuites from libbsd in debug mode. Nothing interesting was found, everything is according sources. And that is how it should be, because trouble is with linker. I searched for linker scripts in rtems and libbsd source trees. It seems like libbsd uses rtems' scripts. But I haven't defined which one and how yet. Makefile and wscript have no -T flag and paths to link scripts 2015-07-28T15:17:35 The BSP determines the linker script 2015-07-28T15:18:10 yes 2015-07-28T15:18:11 Please ask on the mailing list when you run into trouble. Now this problem is dragging on for a very long time, I have concerns that only 2.5 weeks remain for you to show progress during the second half of the summer term 2015-07-28T15:18:41 But nobody wants to help me! 2015-07-28T15:19:09 I understand your frustration. Unfortunately, most of the people with the experience lack the time (and there just aren't many with the experience.) 2015-07-28T15:19:46 Even if nobody is helping, it is good for you to show your process and problems on the mailing list so that myself and other mentors are aware of your situation 2015-07-28T15:20:36 Okay, maybe I'll write the blogpost 2015-07-28T15:20:41 yes that might help 2015-07-28T15:20:54 that's all 2015-07-28T15:20:56 I'm having trouble remembering your exact problem now, because it has been a month since we talkeda bout it 2015-07-28T15:21:29 OK send an email after you write a blog post, I want to read what you've been doing, and also to understand what the "trouble is with linker" 2015-07-28T15:21:37 Thanks gtament 2015-07-28T15:21:48 jrcatbagan: you can go now 2015-07-28T15:22:01 hi gedare 2015-07-28T15:22:17 so I was working on SD/MMC interface 2015-07-28T15:22:27 I have the SD interface mostly done 2015-07-28T15:22:39 just need to clean up the implementation 2015-07-28T15:22:49 then moving on to interface to the onboard eMMC 2015-07-28T15:23:06 ok good, I saw the emails and it looks like great work 2015-07-28T15:23:18 As I was in the process of implementing the SD interface, I was able to boot some RTEMS sample applications 2015-07-28T15:23:20 [trac] GSoC/2015 edited by ketul (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-07-28T15:23:39 was that the first time you managed to? 2015-07-28T15:23:47 Yes 2015-07-28T15:23:59 in terms of booting from uMOn? 2015-07-28T15:24:47 yeah 2015-07-28T15:25:06 It's a great milestone. :) 2015-07-28T15:25:07 Yes, first time I was able to boot an application from uMon 2015-07-28T15:25:32 OK, thanks jrcatbagan, anything else? 2015-07-28T15:25:33 I did it by brute force as uMon can't boot without TFS 2015-07-28T15:25:36 ahhh 2015-07-28T15:25:49 So my task list are: 2015-07-28T15:25:54 1. clean up sd interface 2015-07-28T15:26:01 2. implement MMC interface 2015-07-28T15:26:08 3. Implement Ethernet 2015-07-28T15:26:09 and 2015-07-28T15:26:15 4. Implement TFS-less boot 2015-07-28T15:27:06 ...nonetheless, I am able to still boot an application from uMon 2015-07-28T15:27:16 ok 2015-07-28T15:27:26 that is all I have for now 2015-07-28T15:27:29 great, thanks! 2015-07-28T15:27:32 krohini: you're up 2015-07-28T15:27:36 thank you :-) 2015-07-28T15:28:17 As per the mails exchanged, I debugged the Pi 1 and figured out the pending issue with the cach configurations. There was some progress in discussion on that today, so will have that cleared up soon. On the SMP front, there is the interprocessor interrupt that I am adding now. This thing is going slow :( I did ask on devel and got a reply but I have not been able to track where in the source... 2015-07-28T15:28:19 ...code its being called during initialization. Also Sebastian referred to working on to improve some documentation regarding this initialization process. 2015-07-28T15:28:40 SMP... my feeling is that its just something missing, not much that I need to do to run a basic test.. IPI should quickly lead me somewhere now. I didn't even think of it because of the difference that Pi 2 has for starting other cores. 2015-07-28T15:29:15 "its being called" what is "it"? 2015-07-28T15:29:28 (I haven't been able to read the mailing list yet this morning.) 2015-07-28T15:29:40 rather invoked..the IPI 2015-07-28T15:29:43 ah, ok 2015-07-28T15:30:19 ok, so the IPI is your current focus? and submitting code for bug fix? 2015-07-28T15:30:59 I am giving the cores the start address..I am certain they are executing.. The IPI don't know where it comes into picture.. 2015-07-28T15:31:06 Umm yeah.. 2015-07-28T15:31:53 okay 2015-07-28T15:32:17 thanks krohini, ragunath you can go next 2015-07-28T15:32:39 thanks gedare. 2015-07-28T15:33:16 *** ragunath_ has joined #rtems 2015-07-28T15:33:21 *** ragunath has quit IRC 2015-07-28T15:34:03 *** ragunath_ is now known as ragunath 2015-07-28T15:34:15 hi ragunath you're up now 2015-07-28T15:34:31 sorry problem with my n/w 2015-07-28T15:34:38 I worked on porting of lwIP to RTEMS. I studied RTEMS resource builder & understood how third party packages are built using RSB. I created config and bset file for lwip using ntp as reference. I was able to compile lwip using RSB. All the changes made by marcos is added as a patch.A library archive is created which can be used to build apps. 2015-07-28T15:34:55 This is fully using marcos code. The driver is still from TI. Then the next task was to create an app and run it in BBB and check whether everything is working properly. I wrote an app for lwip but could not run it properly. I am planning to ask marcos for details. 2015-07-28T15:35:25 Should we look for drivers other than the TI driver? 2015-07-28T15:36:32 at this point, i don't think so simply because of time 2015-07-28T15:36:55 if you can demo that the lwip builds and runs, that should be enough to get the RSB patches included 2015-07-28T15:37:08 even if users have to take the extra step to get the TI driver themselves. 2015-07-28T15:37:29 so the driver will not be part of the package? 2015-07-28T15:37:41 I don't think kiwichris will accept it. 2015-07-28T15:38:26 But the value of adding lwip is good, and you can write a blog post or documentation on how to get the fully-working stack up 2015-07-28T15:38:40 that is, of course, after you debug the current problems you face :) 2015-07-28T15:39:16 Next steps on lwip are, run a RTEMS app with lwip with TI driver. 2015-07-28T15:39:50 On the cache issue side, I narrowed down the issue to TX part 2015-07-28T15:39:58 After we start DMA on the transmit side, the corrupted ethernet frame is sent. 2015-07-28T15:40:03 I flushed/invalidated data but no effect. 2015-07-28T15:40:30 I am debugging it parallelly .. but still no concrete result 2015-07-28T15:41:06 ok. DMA problems can be tricky to find and fix. 2015-07-28T15:41:20 sounds good, thanks ragunath anything else? 2015-07-28T15:41:23 [trac] GSoC/2015 edited by jrcatbagan (Added status update for July 28, 2015) http://devel.rtems.org/wiki/GSoC/2015 2015-07-28T15:41:40 nothing else thank you! 2015-07-28T15:41:45 ok thanks. hermann_19829_ you can go now 2015-07-28T15:42:04 hi! 2015-07-28T15:42:05 I prepared an initial version of the abstract for the document describing the coverage report generator we are working on. Thanks for the review gedare! 2015-07-28T15:42:06 Currently I try to get rid of hardcoded paths in the config file containing the paths to the static libs for which the coverage reports are generated. 2015-07-28T15:43:10 ok that is an important problem to resolve. 2015-07-28T15:44:05 yes I currently try to solve it by passing the path to the 'cpukit' directory as a command line argument. 2015-07-28T15:44:34 but I will ask in the devel list if this is a propper approach to resolve this. 2015-07-28T15:44:38 you might want to describe the problem and ... yes exactly 2015-07-28T15:44:54 ok :) 2015-07-28T15:45:10 sounds good, thanks hermann_19829_. sgworks you can go next. 2015-07-28T15:45:19 thanks gedare! 2015-07-28T15:46:55 [trac] GSoC/2015 edited by krohini (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-07-28T15:47:00 *** ragunath has quit IRC 2015-07-28T15:47:12 sgworks: are you here? 2015-07-28T15:48:03 *** ragunath has joined #rtems 2015-07-28T15:48:30 ok, salpha you can go 2015-07-28T15:48:46 in this week i've drawn the path ahead and came back to the context. 2015-07-28T15:48:53 1. Trying to solve data race conditions for SMP model. 2. I feel that there are datarace volunerabilities in RTEMS for SMP in regard to strict order mutex. 2015-07-28T15:49:07 I will talk after salpha 2015-07-28T15:49:33 ok, thanks. :) 2015-07-28T15:49:34 i have to focus on Slingshot now which is supposed to generate test cases. 2015-07-28T15:49:46 i've documented the next steps and the process of adapting Slingshot to RTEMS in the blog. 2015-07-28T15:49:48 Is that part of Ballista? 2015-07-28T15:49:58 ok i'll try to have a read. 2015-07-28T15:50:04 salpha: are you done with your final exams now? 2015-07-28T15:50:13 it is based on Ballista. but different. 2015-07-28T15:50:22 my last exam is on Thursday. 2015-07-28T15:50:30 this Friday i'm going to sit by my mentor, Stefan, and he is going to explain Slingshot source code to me. 2015-07-28T15:50:34 i have to change test case template and compilation options to fit to RTEMS. 2015-07-28T15:51:05 *** DrJoel has joined #rtems 2015-07-28T15:51:05 *** DrJoel has joined #rtems 2015-07-28T15:51:05 *** ChanServ sets mode: +o DrJoel 2015-07-28T15:51:13 so hopefully for the next week i'd have more progress. 2015-07-28T15:51:17 sorry.. still messed up timezone wise and out of sorts. 2015-07-28T15:52:05 ok salpha sounds good. 2015-07-28T15:52:18 thanks. :) 2015-07-28T15:52:18 sgworks: you can continue 2015-07-28T15:52:32 1. Trying to solve data race conditions for SMP model. 2. I feel that there are datarace volunerabilities in RTEMS for SMP in regard to strict order mutex. 2015-07-28T15:53:07 I also mailed about the data race problems which I feel exists in RTEMS 2015-07-28T15:53:13 for the given behavior 2015-07-28T15:53:48 sgworks: perhaps you should focus on the UP model and limit the scope of your project since we have limited time left 2015-07-28T15:53:52 As Sebastian pointed out, that code has never worked right single CPU so no review for SMP. 2015-07-28T15:54:11 hmm yeah 2015-07-28T15:54:14 SMP may need a different solution altogether 2015-07-28T15:54:40 you can assume (and even "assert") that STRICT_ORDER is only on when SMP is off. 2015-07-28T15:54:43 but I feel that I can work on SMP this week and see if I can get through it 2015-07-28T15:54:54 okay 2015-07-28T15:55:07 If you don't get through it, you only have 1.5 weeks to convince us the project is a success ;0 2015-07-28T15:55:13 for UP it is done 2015-07-28T15:55:27 yes 2015-07-28T15:55:36 yes, I think it is best to focus on documenting and cleaning up the model, and implementing the necessary code in RTEMS 2015-07-28T15:55:44 yes 2015-07-28T15:55:48 and then we have to be convinced that it is good 2015-07-28T15:56:07 so first I will finish up the UP model and if time permit then work on smp 2015-07-28T15:56:09 okay then we are agreed? send an email to your mentors and CC me on this plan 2015-07-28T15:56:23 yup 2015-07-28T15:56:33 Today I am going to have hangout with Cyrille 2015-07-28T15:56:45 and then I will mail about the next plans 2015-07-28T15:56:54 in evening 2015-07-28T15:57:33 does it sounds good? 2015-07-28T15:57:35 okay great 2015-07-28T15:57:55 alright, I think that's it unless I missed anyone. 2015-07-28T15:58:06 yes!! 2015-07-28T15:58:13 sujay and yang did not make it today I guess 2015-07-28T15:59:23 thanks gedare. until the next week. :) 2015-07-28T15:59:23 ok then I will update on my next plans. 2015-07-28T15:59:27 see you 2015-07-28T15:59:34 thansk 2015-07-28T15:59:37 thanks 2015-07-28T15:59:42 *** sgworks has quit IRC 2015-07-28T16:00:23 *** salpha has quit IRC 2015-07-28T16:02:49 hermann.. since I was late.. once everyone is done, can you give me a very short status 2015-07-28T16:03:12 *** krohini has quit IRC 2015-07-28T16:03:24 ping hermann_19829_ 2015-07-28T16:03:41 hey DrJoel how was the UK 2015-07-28T16:03:45 hi DrJoel 2015-07-28T16:04:07 i recently started working on the hardcoded paths 2015-07-28T16:04:54 I worked ten hour days and was pushed to see anything. Most was at night and a partial Friday out and about. Made it to London Saturday, did a bus/boat tour, Westminister Abbey, and Phantom of the Opera. Went to airport Sunday morning. 2015-07-28T16:04:59 *** ketul has quit IRC 2015-07-28T16:05:00 my plan was to provide an additional command line argument which should contain the path to the cpukit directory 2015-07-28T16:05:13 Lovely other than my debit card getting eaten on the first day and had to get a new one shipped 2015-07-28T16:05:34 hermann_19829_ If that's to the build directory, that sounds reasonable 2015-07-28T16:06:36 DrJoel: looks like you can check-in sebhub's patch-v2. 2015-07-28T16:06:49 on sys/lock 2015-07-28T16:06:58 all the paths in the symbolSet configuration point to a subdirectory in cpukit 2015-07-28T16:08:10 but I do not know yet whether it is possible that a lib outside the cpukit directory is used for analysing coverage 2015-07-28T16:08:10 *** crazy3k has quit IRC 2015-07-28T16:11:24 Should be OK to do so based on the configuration file. I know at least one user is using covoar on their own.] 2015-07-28T16:13:13 ok I will write a mail to the devel list and explain in detail what I am working on 2015-07-28T16:14:09 *** jrcatbagan has quit IRC 2015-07-28T16:14:11 That sounds like a good idea. 2015-07-28T16:14:42 Do you have any patches ready to review? It would be good to start working that. So submit to devel 2015-07-28T16:15:49 ok I will also send the patches to devel 2015-07-28T16:17:58 Did you get the size issue resolved adequately? And the reports look ok on coverage numbers? 2015-07-28T16:20:53 [trac] SOCIS/2015 edited by hermann19829 (empty comment) http://devel.rtems.org/wiki/SOCIS/2015 2015-07-28T16:21:20 *** dr__house|away is now known as dr__house 2015-07-28T16:21:49 Is the coverage itself still high? Or have all the changes negatively impacted things? Rather beyond your project but I am curious. 2015-07-28T16:22:00 And I need to work to duplicate your results when you are ready 2015-07-28T16:23:11 hm yes I removed treating different byte sizes as an error 2015-07-28T16:23:33 *** gtament has quit IRC 2015-07-28T16:23:36 but I did not investigate if that affects the coverage 2015-07-28T16:23:40 OK.. I see that on the wiki page. 2015-07-28T16:23:43 not yet :) 2015-07-28T16:24:11 OK. I was just curious since the SMP work has been a lot of code slinging and there hasn't been a real focus on the coverage since the reports needed to be RSB based. 2015-07-28T16:24:29 * gedare slings code 2015-07-28T16:24:35 Thanks. I will get the abstract back to you as soon as I can. Should be within a couple of hours. 2015-07-28T16:24:59 gedare :) you know the changes have been massive since we last did a coverage push. 2015-07-28T16:25:07 thank you too DrJoel! 2015-07-28T16:25:10 ya 2015-07-28T16:25:15 i know 2015-07-28T16:25:23 no problem.. that's all I have hermann_19829_ 2015-07-28T16:25:42 now i have to debug an ENOMEM problem. malloc failing, *sigh* 2015-07-28T16:26:02 oh DrJoel 2015-07-28T16:26:10 couple things to spring on you ;) 2015-07-28T16:26:24 gedare: I think the next release should focus on tool update, dropping some targets and BSPs, GSOC/SOCIS code merger. Some other stuff will happen but hopefully inside six months. 2015-07-28T16:26:26 are we ready to declare it 4.11? 2015-07-28T16:26:32 okay 2015-07-28T16:27:00 I think so. I am still too fuzzy headed to say yes. I left a build of all BSPs on CentOS 7 and FreeBSD 10 yesterday. I need to review the results. 2015-07-28T16:27:04 there's a bugfix submitted by someone from EPICS community for tftp.close() that should go in 2015-07-28T16:27:08 okay 2015-07-28T16:27:16 Nick Withers had a patch for bootstrap and the perl path. 2015-07-28T16:27:31 Does it have a PR or just on the list? I recall seeing it. 2015-07-28T16:27:34 i'd like to merge the GPIO stuff post-release 2015-07-28T16:27:44 I had him open a PR 2015-07-28T16:27:53 2376 is relevant to 4.11 2015-07-28T16:27:56 Yep on GPIO... part of GSOC 2015-07-28T16:28:00 2375 is a back-port 2015-07-28T16:28:15 *** Davidbrcz has joined #rtems 2015-07-28T16:28:26 it pulls in some of the eval-path changes that were done (by sebhub?) into tftp to fix a bug they observe 2015-07-28T16:28:45 the submitter suggests that users have simply switched from tftp to nfs to circumvent the bug in the past 2015-07-28T16:29:14 also a test case was provided (by link to github) that probably can be merged into network-demos 2015-07-28T16:29:52 OK. So 2376 on master and 4.11, 2375 on 4.10? And a test to add to network-demos. 2015-07-28T16:30:03 I have the windows open and will try to get it done. 2015-07-28T16:30:20 FYI Sebastian goes on holiday today or tomorrow. 2015-07-28T16:30:37 correct 2015-07-28T16:30:43 just a statement 2015-07-28T16:31:05 yeah that's good he's pushed through a bunch of stuff last few months i'm sure he could use it :) 2015-07-28T16:32:40 *** ragunath has quit IRC 2015-07-28T16:49:13 I need one too. :) 2015-07-28T16:49:28 These work trips are not vacation. Some sightseeing but not vacation 2015-07-28T16:49:29 take one! ;) 2015-07-28T16:49:35 yeah i understand 2015-07-28T16:49:45 speaking of, coming to dc anytime soon? 2015-07-28T16:50:04 Any gsoc or socis students need extra help this week? 2015-07-28T16:50:28 I think early December unless something pops up. 2015-07-28T16:51:02 *** hesham has quit IRC 2015-07-28T16:51:54 *** hesham has joined #rtems 2015-07-28T16:53:19 DrJoel: yes, gtament needs a lot of help with debugging libbsd on rpi for USB support 2015-07-28T16:53:44 uggg :( 2015-07-28T16:53:52 i've asked him to write a blog post on his current efforts and send email. he has been struggling for probably about a month without any progress since the mid-term 2015-07-28T16:53:57 or even a bit before it 2015-07-28T16:54:21 on the plus side, it sounds like he is using qemu successfully now 2015-07-28T16:54:33 also, you might like to try to get qemu+rpi up 2015-07-28T16:54:46 per the notes given by jan sommer 2015-07-28T16:58:40 I was bouncing email with him about Ada. I built a sparc Ada toolset before I left but didn't get any chance to use it. 2015-07-28T16:58:45 I will work on getting pi on qemu up 2015-07-28T16:58:59 okay thanks a bunch 2015-07-28T16:59:12 i don't have the bandwidth to help there unfortunately 2015-07-28T16:59:50 I can help gtament build rtems-libbsd but I have never used the USB stack. And I am fighting to get TCP/IP up on qemu pc386 now. 2015-07-28T17:00:28 i'm pretty sure the # of people with usb experience on rtems-libbsd can be counted on a simpson's hand. 2015-07-28T17:00:37 :) Doh! 2015-07-28T17:00:41 Boy!!! 2015-07-28T17:00:51 :D 2015-07-28T17:00:53 Or Mickey Mouse's 2015-07-28T17:03:47 sorting expenses.. ended up with charges across four cards and accidentally used OAR card for personal stuff once. Tedious like taxes to make sure no fraud transaactions, etc. 2015-07-28T17:04:04 yeah 2015-07-28T17:21:08 *** asuolen has quit IRC 2015-07-28T17:23:04 *** hermann_19829_ has quit IRC 2015-07-28T18:45:35 *** monstr has quit IRC 2015-07-28T19:22:42 *** fds_ is now known as fds 2015-07-28T19:36:08 *** Davidbrcz has quit IRC 2015-07-28T20:57:11 *** gedare has quit IRC 2015-07-28T20:57:25 *** hesham has quit IRC 2015-07-28T21:08:00 *** Davidbrcz has joined #rtems 2015-07-28T21:25:13 *** igutek has quit IRC 2015-07-28T21:25:26 *** igutek has joined #rtems 2015-07-28T21:38:27 *** hesham has joined #rtems 2015-07-28T21:40:59 *** hesham has quit IRC 2015-07-28T21:42:50 *** hesham has joined #rtems 2015-07-28T21:45:07 *** hesham1 has joined #rtems 2015-07-28T21:47:03 *** hesham has quit IRC 2015-07-28T22:16:17 *** guerby has quit IRC 2015-07-28T22:25:41 *** edsiper_ has quit IRC 2015-07-28T22:28:29 *** javamonn has quit IRC 2015-07-28T22:39:29 *** kiwichris has joined #rtems 2015-07-28T22:42:53 *** guerby has joined #rtems 2015-07-28T22:59:49 *** edsiper has joined #rtems 2015-07-28T23:46:58 *** hesham1 has quit IRC 2015-07-28T23:55:19 *** hesham has joined #rtems 2015-07-29T00:38:17 *** Davidbrcz has quit IRC 2015-07-29T03:36:47 *** mumptai has quit IRC 2015-07-29T03:49:12 *** mumptai has joined #rtems 2015-07-29T04:23:34 *** dr__house is now known as dr__house|away 2015-07-29T04:32:47 *** igutek has quit IRC 2015-07-29T04:37:38 *** igutek has joined #rtems 2015-07-29T05:05:25 *** monstr has joined #rtems 2015-07-29T06:26:50 *** hesham has quit IRC 2015-07-29T07:09:49 good morning 2015-07-29T07:16:43 [trac] #2375/defect (new) updated by sebastian.huber (In case a certain feature/behaviour/API is important for you, then you should add ...) http://devel.rtems.org/ticket/2375 2015-07-29T07:31:51 sebhub, I understand the _ and naming. It however means it has to sit in this specialised location. In other words it is direct and specific when I think it could be more useful. To use this elsewhere use have to use the _* interface. 2015-07-29T07:32:26 What do you mean by _Package_Name convention ? 2015-07-29T07:32:49 for the Newlib internal locks, libgomp, and the GCC thread model this underscore API is all right 2015-07-29T07:33:03 for the user we can add C11 threads 2015-07-29T07:33:31 _Mutex_Control, _Thread_Get_executing, etc. 2015-07-29T07:34:23 *** kiwichri_ has joined #rtems 2015-07-29T07:34:34 the C11 threads.h header is not part of GCC, we have to add it to Newlib and then implement it in RTEMS (similar to phreads.h) 2015-07-29T07:35:09 I must have misunderstood. The names used are defined in C11 ? 2015-07-29T07:35:42 no, but you can add a simple wrapper to get the C11 functions 2015-07-29T07:36:11 they map one-to-one in terms of parameters 2015-07-29T07:36:13 Are we now exporting functions as an API to newlib with an _ prefix ? 2015-07-29T07:37:05 what do you mean with exporting? 2015-07-29T07:37:25 *** kiwichris has quit IRC 2015-07-29T07:37:40 I see this as an API. It is fixed with external dependences and cannot change. It is not like the score. 2015-07-29T07:38:47 its only more difficult to change since Newlib and GCC depends on it, but only internally, not at a user API level 2015-07-29T07:39:46 I feel we need to treat the interface with respect and like our other APIs. Could I build an application with this interface ? 2015-07-29T07:40:24 if we add a C11 wrapper or use it for the GCC thread model, then yes 2015-07-29T07:40:29 or indirectly via OpenMP 2015-07-29T07:40:42 Where would be add the C11 wrapper ? 2015-07-29T07:40:59 the header file in Newlib, the implementation in cpukit 2015-07-29T07:42:51 the performance gains for OpenMP are sigificant, its not in the range of 5-10% its a factor of more than 50 and 100 in two important microbenchmarks, seehttps://devel.rtems.org/ticket/2274#comment:6 2015-07-29T07:44:46 This is what I am concerned about "the performance gains ... are sigificant" so as a user why can I not see them in a standard application ? 2015-07-29T07:45:36 we can use them for the C11 threads as a third API along with the Classic and pthreads 2015-07-29T07:46:56 i want to use them also for the FreeBSD mutexes, etc. so this would be also an option 2015-07-29T07:49:09 the Classic object based API offers some unique features, but has some weaknesses by designe when it comes to fast object creation/destruction and operations 2015-07-29T07:50:20 its simply not possible to fix this in the Classic objects implementation 2015-07-29T07:53:54 *** zoso has joined #rtems 2015-07-29T07:56:24 I agree, this API is really useful and if it is a performance win user will want it. Is C11 threads able to be done now ? 2015-07-29T07:57:03 not in the current patch set, but its easy to add this later 2015-07-29T07:58:02 I tested this with a modified gthr-rtems.h in GCC to check that the API is all right, the C++11 and C11 threads support is quite similar 2015-07-29T07:58:09 Ok. Will we need other support functions to allow interactions across APIs ? I assume there is no object id. 2015-07-29T07:58:15 C++11 threads is really nice 2015-07-29T07:58:21 Yes it is. 2015-07-29T07:58:51 the threads are still normal pthreads 2015-07-29T07:59:03 Ah ok. 2015-07-29T07:59:09 So an id ? 2015-07-29T07:59:17 there is no interaction with other objects 2015-07-29T07:59:39 for the thread yes, but not for mutexes and condition variables 2015-07-29T08:00:26 Could I get a mutex and ask to have an object constructed to bridge a boundary or is this not worth it ? 2015-07-29T08:00:51 it don't see a use case for this 2015-07-29T08:01:12 the only thing would be condition variables, but it would complicate things a lot 2015-07-29T08:02:03 I suppose it is ok at thread level so a classic thread calling a function that blocks on C11 mutex will be ok. 2015-07-29T08:02:08 the C11 api is not flexible in this respect, the FreeBSD APIs on the other hand would allow this 2015-07-29T08:02:38 yes, the threads are basically all the same in all APIs 2015-07-29T08:03:23 So to bridge a boundary wrap the mutex etc in calls and export those. 2015-07-29T08:04:04 yes 2015-07-29T09:07:40 *** kiwichri_ has quit IRC 2015-07-29T09:20:53 *** mkhoory has joined #rtems 2015-07-29T09:20:57 *** ketul has joined #rtems 2015-07-29T09:45:59 *** ketul has quit IRC 2015-07-29T09:56:02 *** ketul has joined #rtems 2015-07-29T10:00:39 *** peerst has quit IRC 2015-07-29T10:01:02 *** peerst has joined #rtems 2015-07-29T10:05:53 *** guerby has quit IRC 2015-07-29T10:05:54 *** igutek has quit IRC 2015-07-29T10:06:39 *** igutek has joined #rtems 2015-07-29T10:17:10 *** guerby has joined #rtems 2015-07-29T10:34:35 *** mkhoory has quit IRC 2015-07-29T12:47:55 *** monstr has quit IRC 2015-07-29T12:51:51 *** ketul has quit IRC 2015-07-29T12:57:21 *** diginet has quit IRC 2015-07-29T13:02:35 *** monstr has joined #rtems 2015-07-29T13:09:41 *** DrJoel has quit IRC 2015-07-29T13:22:38 *** DrJoel has joined #rtems 2015-07-29T13:22:38 *** DrJoel has joined #rtems 2015-07-29T13:22:38 *** ChanServ sets mode: +o DrJoel 2015-07-29T13:38:21 *** dr__house|away is now known as dr__house 2015-07-29T13:40:01 *** zoso has quit IRC 2015-07-29T13:41:06 *** zoso has joined #rtems 2015-07-29T13:54:12 *** zoso has quit IRC 2015-07-29T14:37:54 *** stryx` has quit IRC 2015-07-29T14:38:24 *** stryx` has joined #rtems 2015-07-29T14:49:28 *** antgreen has joined #rtems 2015-07-29T14:51:20 *** monstr has quit IRC 2015-07-29T15:10:45 *** anandkp has joined #rtems 2015-07-29T15:24:07 *** hesham has joined #rtems 2015-07-29T15:36:19 *** Davidbrcz has joined #rtems 2015-07-29T15:52:44 *** jkridner has joined #rtems 2015-07-29T16:16:06 [trac] #2376/defect (closed) updated by Michael Davidsaver (In [changeset:"74ef2c101a2a8ac2cb4c32ae0da8061e2b76886a/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2376 2015-07-29T16:16:07 [git] 74ef2c1 by Michael Davidsaver: [tftpDriver: close() false error closes #2376.] http://devel.rtems.org/changeset/74ef2c1/rtems 2015-07-29T16:17:26 [trac] #2376/defect (closed) updated by Michael Davidsaver (In [changeset:"8da63adc4b166ad5355f96a96022d4835493f361/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2376 2015-07-29T16:17:26 [git] 8da63ad by Michael Davidsaver: [tftpDriver: close() false error closes #2376.] http://devel.rtems.org/changeset/8da63ad/rtems 2015-07-29T16:21:23 [trac] #2376/defect (closed) updated by Michael Davidsaver (In [changeset:"a4e052b53fb10f0cf2e3c1f4681710ab9912fbcd/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2376 2015-07-29T16:21:23 [git] a4e052b by Michael Davidsaver: [tftpDriver: close() false error closes #2376.] http://devel.rtems.org/changeset/a4e052b/rtems 2015-07-29T16:22:05 *** antgreen has quit IRC 2015-07-29T16:35:57 *** jkridner has quit IRC 2015-07-29T16:37:19 *** jkridner has joined #rtems 2015-07-29T17:01:19 *** jkridner has quit IRC 2015-07-29T17:02:35 *** DrJoel has quit IRC 2015-07-29T17:10:14 *** DrJoel has joined #rtems 2015-07-29T17:10:14 *** DrJoel has joined #rtems 2015-07-29T17:10:14 *** ChanServ sets mode: +o DrJoel 2015-07-29T18:17:58 *** gedare has joined #rtems 2015-07-29T18:17:58 *** ChanServ sets mode: +o gedare 2015-07-29T18:19:21 *** dougk_ has quit IRC 2015-07-29T18:50:11 *** ketul has joined #rtems 2015-07-29T18:57:09 Hey everyone 2015-07-29T18:57:21 I'm quite stuck with adapting the test runner to a new target 2015-07-29T18:57:47 I have it loading the executable via gdb, but it's stuck on the first test. I can't seem to find a way to print out more verbose logs 2015-07-29T18:57:53 Is there an easy way to do this? 2015-07-29T19:03:41 What output do you have? Something like hello world only prints a few lines for test purposes. 2015-07-29T19:08:45 I don't see any console output. I'm using the test runner 2015-07-29T19:10:24 Actually, about 5 minutes later I finally got a timeout and some output 2015-07-29T19:10:57 It looks like it's not actually loading the right code or maybe loading it to the wrong address 2015-07-29T19:14:38 On an embedded PC or qemu? 2015-07-29T19:14:43 What BSP? 2015-07-29T19:32:24 *** Davidbrcz has quit IRC 2015-07-29T19:33:05 I'm using a STM32F4-Discovery 2015-07-29T19:33:21 using a slightly modified BSP 2015-07-29T19:33:47 I copied the beagleboardxm configuration for the test runner, and modified it to connect to the JLink GDB Server 2015-07-29T19:33:56 Does hello world run by hand to completion on it? If it doesn't run back to a gdb prompt, the the rtems-tester won't work. 2015-07-29T19:34:19 Okay, so it needs to return to a gdb prompt? 2015-07-29T19:34:59 I am pretty sure yes. All the BSPs I use with it do so. 2015-07-29T19:35:16 Okay, I'll see if I can put a breakpoint on _exit or something else relevant 2015-07-29T19:35:21 It will then quit gdb and start it again with a different executable. So it has to work once cleanly by hand. 2015-07-29T19:35:55 There is a build option about what to do when the flow returns to the BSP at program shutdown. It may be going into a hard loop instead of resetting the board. 2015-07-29T19:36:11 The bsp has to support bsp_reset() for a reset. 2015-07-29T19:36:38 Breaking on _reset seems to do the trick 2015-07-29T19:37:37 Well, it made it it slightly further 2015-07-29T19:38:03 Is that at the end of program execution? 2015-07-29T19:38:37 It's something in newlib 2015-07-29T19:38:45 I saw the beagleboardxm tester script use it 2015-07-29T19:38:57 I think it is called after main returns 2015-07-29T19:39:35 I think I need a different breakpoint to get gdb to halt after each test 2015-07-29T19:40:17 beng-nl_: awake? 2015-07-29T19:43:06 *** hesham has quit IRC 2015-07-29T19:43:34 It looks like the stm32 BSP has a while(1) loop in bsp_reset 2015-07-29T19:43:54 so for the sake of running the tests, setting a breakpoint on bsp_reset might work. I'll give it a shot 2015-07-29T19:45:35 A BSP can provide its own version of bsp_reset() but that is a stock one that is common. It is libbsp/shared/bspreset_loop.c I think. 2015-07-29T19:46:14 RTEMS_BSP_CLEANUP_OPTIONS() is a macro used in the BSP configure.ac that configures some common behavior in libbsp/shared/bspclean.c 2015-07-29T19:47:00 This is an area where the BSP has a startup/shutdown framework with plugin points. Many BSPs use default versions at some plugin points. 2015-07-29T19:47:26 That makes sense 2015-07-29T19:55:23 So maybe the real answer here is that not all the tests pass :) 2015-07-29T19:55:52 That is quite likely if the RAM is low on this board. 2015-07-29T19:57:21 look at the files *.tcfg in the make/custom directory under your BSP family directory. You can mark some tests as known to fail. Usually these are ones which don't fit into program space on the target or allocate too many resources. If it gets beyond a linking issue, it is wise to know why each test failed before you dismiss it. 2015-07-29T20:03:34 Okay, I'll take a look at that 2015-07-29T20:04:04 It's not entirely clear why they are failing. I'm going to take a break on this and get back to CAN (which is also being difficult :) ) 2015-07-29T20:04:07 Thanks for your help 2015-07-29T20:08:49 *** Davidbrcz has joined #rtems 2015-07-29T20:34:26 no problem 2015-07-29T20:54:59 *** gedare has quit IRC 2015-07-29T21:03:35 *** ketul has quit IRC 2015-07-29T22:01:07 *** dougk_ has joined #rtems 2015-07-29T22:51:00 *** kiwichris has joined #rtems 2015-07-30T01:11:30 *** Davidbrcz has quit IRC 2015-07-30T01:38:26 *** anandkp has quit IRC 2015-07-30T03:53:38 *** mkhoory has joined #rtems 2015-07-30T04:29:15 *** monstr has joined #rtems 2015-07-30T04:46:50 *** dr__house is now known as dr__house|away 2015-07-30T05:07:57 good morning 2015-07-30T05:40:36 *** DrJoel has quit IRC 2015-07-30T05:46:20 *** DrJoel has joined #rtems 2015-07-30T05:46:20 *** DrJoel has joined #rtems 2015-07-30T05:46:20 *** ChanServ sets mode: +o DrJoel 2015-07-30T05:48:50 *** toxygen has quit IRC 2015-07-30T05:49:52 *** toxygen has joined #rtems 2015-07-30T07:17:14 *** DrJoel has quit IRC 2015-07-30T07:23:00 *** DrJoel has joined #rtems 2015-07-30T07:23:00 *** DrJoel has joined #rtems 2015-07-30T07:23:00 *** ChanServ sets mode: +o DrJoel 2015-07-30T08:58:48 *** hesham has joined #rtems 2015-07-30T09:11:14 *** kiwichris has quit IRC 2015-07-30T09:24:44 *** ketul has joined #rtems 2015-07-30T09:46:56 *** kiwichris has joined #rtems 2015-07-30T12:10:27 *** ketul has quit IRC 2015-07-30T13:05:22 *** dr__house|away is now known as dr__house 2015-07-30T13:32:23 *** gedare has joined #rtems 2015-07-30T13:32:24 *** ChanServ sets mode: +o gedare 2015-07-30T13:37:32 hi hesham 2015-07-30T13:41:42 hi gedare 2015-07-30T13:52:49 hesham: are you submitting a paper to EWiLi? 2015-07-30T13:56:57 gedare: No, my supervisor is giving a talk there, so he said it's not a good idea to submit it. 2015-07-30T13:59:44 *** ketul has joined #rtems 2015-07-30T14:22:38 hello DrJoel , gedare :) 2015-07-30T14:22:53 I wrote i2c driver code :- https://gist.github.com/ketul93/a8cdde82c6d946de1d98 2015-07-30T14:26:24 But for compilation when I removed the existing code (written by Claas Ziemke ) under misc/i2c.c and created new one ../beagle/i2c/ . Then I edited Makefile.am and did ./bootstrap still I am getting error "make[6]: *** No rule to make target `../../../../../../../../rtems/c/src/lib/libbsp/arm/beagle/misc/i2c.c', needed by `libbsp_a-i2c.o'. Stop." 2015-07-30T14:26:24 while doing make :( 2015-07-30T14:31:18 I hope beng-nl_ may be here :) 2015-07-30T14:34:55 do bootstrap -c 2015-07-30T14:35:02 and then bootstrap 2015-07-30T14:35:21 and make sure you removed misc/i2c.c from the Makefile.am 2015-07-30T14:35:23 ketul: ^ 2015-07-30T14:36:17 yep gedare I removed that. Let me do that :) 2015-07-30T14:36:19 ketul: Instead of putting "Modified by Ketul Shah"... just add your name with a copyright notice underneath the existing one in the file. 2015-07-30T14:37:09 ok 2015-07-30T14:42:07 hi ketul 2015-07-30T14:42:53 hello DrJoel. Hope u saw the code :) 2015-07-30T14:45:49 ketul: is your code also on your git repo? 2015-07-30T14:46:08 I see you posted a link. Does it work? 2015-07-30T14:46:42 gedare: not yet . After test I will post 2015-07-30T14:46:59 DrJoel: yet to compile 2015-07-30T14:48:14 gedare: or should I do it before it ? 2015-07-30T14:53:02 gedare: same error . I added 1"# I2C 2015-07-30T14:53:02 libbsp_a_SOURCES += i2c/i2c.c 2015-07-30T14:53:02 libbsp_a_SOURCES += i2c/i2c_init.c" to Makefile.am and removed the previous one. Done that u suggested 2015-07-30T14:56:45 Did you add/remove any .h files? 2015-07-30T14:58:21 DrJoel: nop. i2c.h was already there . edited "configure.ac" 2015-07-30T14:58:51 You need to run bootstrap from the BSP directory. 2015-07-30T15:05:03 ketul you should be posting commits to github as you work 2015-07-30T15:05:18 and then you can create a clean branch to rebase and create a submission 2015-07-30T15:05:43 ok gedare let me do it and give u the link :) 2015-07-30T15:29:21 *** edwardk has quit IRC 2015-07-30T15:29:28 gedare: commits pushed . https://github.com/ketul93/rtems/tree/I2C 2015-07-30T15:33:20 thanks. 2015-07-30T15:34:29 gedare: kindly help me with this "You need to run bootstrap from the BSP directory" How to do it ? 2015-07-30T15:36:32 cd .../libbsp/arm/BSP 2015-07-30T15:36:46 ....../bootstrap (script is at the top of RTEMS tree) 2015-07-30T15:37:02 so be in the BSP directory and run the bootstrap script located at the top of the RTEMS tree 2015-07-30T15:37:45 ok DrJoel got it :) 2015-07-30T15:47:01 DrJoel: still not resolved :( . Can you point me any another mistakes done by me (github repo updated ) 2015-07-30T15:51:42 looking. 2015-07-30T15:53:08 look in your libbsp/arm/beagle/Makefile.in see if it include misc/i2c.c 2015-07-30T15:54:01 ketul: ^ 2015-07-30T15:55:55 *** peerst has quit IRC 2015-07-30T16:37:32 *** peerst has joined #rtems 2015-07-30T16:37:40 *** edwardk has joined #rtems 2015-07-30T17:38:22 sorry for a huge delay. gedare I looked into .in . No there is no misc/i2c.c in that file. It has only i2c/i2c.c and i2c/i2c_init.c which I defined in Makefile.am 2015-07-30T17:40:48 hi ketul 2015-07-30T17:40:54 is autoconf and automake from the rtems tools directory? 2015-07-30T17:40:58 maybe you should try deleting your build and starting over 2015-07-30T17:41:15 it may be that ../rtems/configure did not pick up the change 2015-07-30T17:41:34 especially if you have not used --enable-maintainer-mode 2015-07-30T17:42:04 ohh. gedare yes I never used --enable-maintainer-mode 2015-07-30T17:42:06 ketul: so it will be easiest to just rm -rf your build directory and start over from configure 2015-07-30T17:43:01 ok. Thanks 2015-07-30T17:43:59 So gedare first I should delete the b-beagle (build directory). Then again following all the steps 2015-07-30T17:47:40 yea 2015-07-30T17:50:30 *** hesham has quit IRC 2015-07-30T17:52:40 ok thanks gedare. Now error removed :) 2015-07-30T17:53:40 Yeah! Do you build with --enable-maintainer-mode? 2015-07-30T17:54:03 Even if you do, sometimes you need to make clean down near the area you changed files in a Makefile.am for 2015-07-30T17:54:29 nope. Just "CONSOLE_POLLED=1 ../rtems-src/configure --target=arm-rtems4.11 --enable-rtemsbsp="beagleboneblack" --enable-tests" 2015-07-30T17:58:59 --enable-maintainer-mode can help when you are hacking configuration and Makefiles. But nothing is as reliable as a fresh build. 2015-07-30T17:59:53 ohk. 2015-07-30T18:08:02 *** diginet has joined #rtems 2015-07-30T18:34:43 *** edwardk has quit IRC 2015-07-30T18:34:44 *** peerst has quit IRC 2015-07-30T18:34:46 *** guerby has quit IRC 2015-07-30T18:35:45 *** guerby has joined #rtems 2015-07-30T18:37:34 *** dr__house is now known as dr__house|away 2015-07-30T18:37:39 *** monstr has quit IRC 2015-07-30T18:38:56 *** edwardk has joined #rtems 2015-07-30T18:41:31 *** peerst has joined #rtems 2015-07-30T19:25:48 *** ketul has quit IRC 2015-07-30T21:04:54 *** gedare has quit IRC 2015-07-30T21:05:23 *** igutek has quit IRC 2015-07-30T21:05:25 *** DrJoel has quit IRC 2015-07-30T21:05:25 *** mumptai has quit IRC 2015-07-30T21:05:26 *** gigetoo has quit IRC 2015-07-30T21:05:48 *** DrJoel has joined #rtems 2015-07-30T21:05:48 *** DrJoel has joined #rtems 2015-07-30T21:05:48 *** ChanServ sets mode: +o DrJoel 2015-07-30T21:06:06 *** igutek has joined #rtems 2015-07-30T21:06:18 *** mumptai has joined #rtems 2015-07-30T21:06:18 *** gigetoo has joined #rtems 2015-07-30T21:48:12 *** kiwichris has quit IRC 2015-07-30T22:10:04 *** kiwichris has joined #rtems 2015-07-30T22:47:10 [trac] #2375/defect (new) updated by joel.sherrill (Can the proper patches to RTEMS and network-demos be attached to this ticket? I ...) http://devel.rtems.org/ticket/2375 2015-07-30T23:45:36 *** gedare has joined #rtems 2015-07-30T23:45:37 *** ChanServ sets mode: +o gedare 2015-07-30T23:52:18 *** gedare has quit IRC 2015-07-31T04:19:08 *** Davidbrcz has joined #rtems 2015-07-31T04:57:20 *** monstr has joined #rtems 2015-07-31T05:05:01 *** Davidbrcz has quit IRC 2015-07-31T05:05:47 good morning 2015-07-31T05:27:35 [git] 4827470 by Sebastian Huber: [score: Rename struct Thread_Control Add a leading underscore to the structure name to allow forward declarations in standard header files provided by Newlib and GCC.] http://devel.rtems.org/changeset/4827470/rtems 2015-07-31T05:27:35 [git] e421936 by Sebastian Huber: [score: Provide struct _Thread_queue_Heads Use a leading underscore in the structure name to allow forward declarations in standard header files provided by Newlib and GCC.] http://devel.rtems.org/changeset/e421936/rtems 2015-07-31T05:27:35 [git] 10e32a2 by Sebastian Huber: [score: Add potpourri lock statistics These SMP lock statistics are used for all lock objects that lack a storage space for the statistics. Examples are lock objects used in external libraries which are independent of the actual RTEMS build configuration.] http://devel.rtems.org/changeset/10e32a2/rtems 2015-07-31T05:27:35 [git] 0e3c59d by Sebastian Huber: [score: Use a plain ticket lock for thread locks This enables external libraries to use thread locks since they are independent of the actual RTEMS build configuration, e.g. profiling enabled or disabled.] http://devel.rtems.org/changeset/0e3c59d/rtems 2015-07-31T05:27:35 [git] 12f93fb by Sebastian Huber: [score: Add thread queue for self-contained objects] http://devel.rtems.org/changeset/12f93fb/rtems 2015-07-31T05:27:35 [git] 214d8ed by Sebastian Huber: [score: Add self-contained mutex implementation This mutex implementation uses a thread priority queue with a simple priority inheritance mechanism (similar to the object based mutexes). The storage space must be supplied by the user (16 bytes on 32-bit targets).] http://devel.rtems.org/changeset/214d8ed/rtems 2015-07-31T05:27:36 [git] 7237b3e by Sebastian Huber: [score: Add self-contained semaphore implementation] http://devel.rtems.org/changeset/7237b3e/rtems 2015-07-31T05:27:36 [git] 4018871 by Sebastian Huber: [score: Add self-contained futex implementation] http://devel.rtems.org/changeset/4018871/rtems 2015-07-31T05:27:37 [git] a1b4af4 by Sebastian Huber: [score: Add scheduler support] http://devel.rtems.org/changeset/a1b4af4/rtems 2015-07-31T05:27:37 [git] 9e9e61d by Sebastian Huber: [score: Add self-contained condition implementation] http://devel.rtems.org/changeset/9e9e61d/rtems 2015-07-31T05:28:42 [trac] #2274/enhancement (accepted) updated by sebastian.huber (The relevant Newlib patches are checked in: ...) http://devel.rtems.org/ticket/2274 2015-07-31T05:32:11 [trac] #1247/defect (closed) updated by sebastian.huber (It is fixed with this commit in Newlib: ...) http://devel.rtems.org/ticket/1247 2015-07-31T05:42:15 *** crazy3k has joined #rtems 2015-07-31T06:20:01 sebhub, you are going to have to come up with a better name than "Self Contained ..." for this new API :) 2015-07-31T06:56:14 i am open for suggestions ;-) 2015-07-31T07:41:40 *** ketul has joined #rtems 2015-07-31T07:47:54 *** kiwichris has quit IRC 2015-07-31T07:59:31 [git] d0733bb by Sebastian Huber: [bsps/arm: Do not use __ARM_ARCH_7A__ This would lead to link-time errors in case less specialized compiler machine options are used, e.g. to run the GCC test suite.] http://devel.rtems.org/changeset/d0733bb/rtems 2015-07-31T08:30:35 [trac] #2377/defect (v:4.11) created by sebastian.huber (rtems_waf: Tools without a version are not supported) http://devel.rtems.org/ticket/2377 2015-07-31T10:07:43 *** crazy3k has quit IRC 2015-07-31T10:08:44 [trac] GSoC/2015 edited by ragunath (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-07-31T10:48:40 *** mumptai has quit IRC 2015-07-31T10:50:25 *** mumptai has joined #rtems 2015-07-31T11:52:27 *** monstr has quit IRC 2015-07-31T11:57:00 [git] 2fc413a by Sebastian Huber: [i386: Make compile clean] http://devel.rtems.org/changeset/2fc413a/rtems-libbsd 2015-07-31T11:57:01 [git] a3ef8e6 by Sebastian Huber: [e1000: Add missing file] http://devel.rtems.org/changeset/a3ef8e6/rtems-libbsd 2015-07-31T11:57:01 [git] 88205a5 by Sebastian Huber: [waf_generator.py: Move arch header includes Make sure that architecture specific header files are found first.] http://devel.rtems.org/changeset/88205a5/rtems-libbsd 2015-07-31T11:57:01 [git] 3fc15fd by Sebastian Huber: [i386/legacy.c: Disable per CPU stuff] http://devel.rtems.org/changeset/3fc15fd/rtems-libbsd 2015-07-31T11:57:01 [git] 9199609 by Sebastian Huber: [DEVICE.HINTS(5): Enable resource_disabled()] http://devel.rtems.org/changeset/9199609/rtems-libbsd 2015-07-31T11:57:01 [git] 7c33a00 by Sebastian Huber: [Dummy securelevel_ge() and securelevel_gt()] http://devel.rtems.org/changeset/7c33a00/rtems-libbsd 2015-07-31T11:57:01 [git] ad0fb56 by Sebastian Huber: [Disable pcibus_pnp driver module] http://devel.rtems.org/changeset/ad0fb56/rtems-libbsd 2015-07-31T11:57:02 [git] 855c120 by Sebastian Huber: [PCI: Enable some functions] http://devel.rtems.org/changeset/855c120/rtems-libbsd 2015-07-31T11:57:02 [git] 45eb88b by Sebastian Huber: [nexus-devices: Add pc386 nexus device and drivers] http://devel.rtems.org/changeset/45eb88b/rtems-libbsd 2015-07-31T12:23:02 *** gedare has joined #rtems 2015-07-31T12:23:02 *** ChanServ sets mode: +o gedare 2015-07-31T12:29:13 *** gedare has quit IRC 2015-07-31T13:04:04 hi DrJoel 2015-07-31T13:05:57 *** jkridner has joined #rtems 2015-07-31T13:15:16 hi ketul.. never logged off :) 2015-07-31T13:15:40 sebhub: Thanks.. you have just one patch left for newlib, right? 2015-07-31T13:17:46 yes, a stupid typo 2015-07-31T13:18:21 trying to make sure I have commited all that you want before you go on holiday 2015-07-31T13:18:22 only one test was able to catch it 2015-07-31T13:22:05 It should be committed now. Check me. 2015-07-31T13:22:49 ok, thanks 2015-07-31T13:23:44 Thanks for the pc work. If I understand correctly, qemu doesn't redirect com1 to stdio anymore? 2015-07-31T13:23:47 i run currenlty the libstdc++ testsuite with the c++11 threads support 2015-07-31T13:24:18 at least I wasn't able to get it working 2015-07-31T13:24:29 And the code itself is getting to the point where it needs the PCI configured in the BSD code? Or can it just use the support already in the BSP 2015-07-31T13:24:57 the pci probe and attach works, it finds the card 2015-07-31T13:25:12 the card attach (em_attach) needs some resources, and this fails 2015-07-31T13:26:01 it uses the bsd pci support 2015-07-31T13:26:12 Oh.. that is much farther than I got off a quick scan of your email 2015-07-31T13:27:17 Thanks. So the resources need to be attached to the nexus device? or something else? 2015-07-31T13:27:50 the resource managment is in a bad shape, its more or less a hack 2015-07-31T13:28:22 can you point me to a file where it happens and where you think the configured resource needs to be specified? 2015-07-31T13:28:24 I guess the nexus device is responsible for the resources 2015-07-31T13:28:39 I know it is the end of the day there and just want to get some hints before you disappear. 2015-07-31T13:28:59 set a break point to em_attach and debug the call where it allocates the resource 2015-07-31T13:29:41 it goes up in the bus tree, and some bus device must know something about resources and deliver them 2015-07-31T13:30:20 see also rtems-bsd-nexus.c 2015-07-31T13:30:49 OK.. I can probably run it down from there. 2015-07-31T13:31:30 Anything else you know of that needs dealing with before you go? 2015-07-31T13:32:03 no, I have some GCC patches pending for libgomp, but I will ping them occasionally 2015-07-31T13:32:23 I am definitely write after approval on those. :) 2015-07-31T13:32:41 Are you visiting your inlaws? 2015-07-31T13:32:46 yes, jakub jelinek is responsible for this area 2015-07-31T13:32:47 yes 2015-07-31T13:33:13 I thought so even before you mentioned the length of time. 2015-07-31T13:33:57 child care is closed in august 2015-07-31T13:34:01 Hopefully it will be uneventful from a travel perspective. 2015-07-31T13:34:14 i would never travel to shanghai in august if I could choose the date 2015-07-31T13:34:15 ahhh... and the grandparents will be thrilled! 2015-07-31T13:34:52 August is nice in Europe :) 2015-07-31T13:37:50 Which version of qemu were you using? I just double checked using ticker and this command seems to let com1 go to stdio: 2015-07-31T13:37:51 qemu-system-i386 -m 512 -boot a -cpu 486 -drive file=/home/joel/qemu/pc386_fda,index=0,if=floppy,format=raw -drive file=fat:/home/joel/qemu/hd,format=raw -monitor null -nographic -serial stdio --no-reboot 2015-07-31T13:38:20 $ qemu-system-i386 --version 2015-07-31T13:38:20 QEMU emulator version 2.3.50, Copyright (c) 2003-2008 Fabrice Bellard 2015-07-31T13:43:27 QEMU emulator version 2.1.92, Copyright (c) 2003-2008 Fabrice Bellard 2015-07-31T13:45:10 I use the qemu-system-i386 only once a year or so 2015-07-31T13:47:12 I have to go now, cu 2015-07-31T13:47:48 *** sebhub has quit IRC 2015-07-31T13:49:07 bye.. safe travels 2015-07-31T15:28:28 *** ketul has quit IRC 2015-07-31T15:44:42 *** gedare has joined #rtems 2015-07-31T15:44:43 *** ChanServ sets mode: +o gedare 2015-07-31T16:15:58 hi DrJoel 2015-07-31T19:15:26 *** DrJoel has quit IRC 2015-07-31T19:15:42 *** DrJoel has joined #rtems 2015-07-31T19:15:42 *** DrJoel has joined #rtems 2015-07-31T19:15:42 *** ChanServ sets mode: +o DrJoel 2015-07-31T19:24:34 *** ketul has joined #rtems 2015-07-31T19:28:17 hi DrJoel, beng-nl_, gedare . can you give comment(s) on https://github.com/ketul93/rtems/tree/I2C/c/src/lib/libbsp/arm/beagle/i2c . It is free of compilation errors (built). And yes still test on hardware is remaining. As discussed want to do test using adxl345 (some coding for that is remaining). :) 2015-07-31T19:30:58 sorry i can't, but make sure you have fixed any style issues 2015-07-31T19:33:10 ya gedare that I will do make sure. checking again :) 2015-07-31T19:36:18 ok. i noticed a mixture of space and tab 2015-07-31T19:36:54 ok 2015-07-31T19:46:31 *** gedare has quit IRC 2015-07-31T20:44:53 *** DrJoel has quit IRC 2015-07-31T20:44:59 *** DrJoel_ has joined #rtems 2015-07-31T20:46:02 *** DrJoel_ has quit IRC 2015-07-31T21:32:34 *** ketul has quit IRC 2015-07-31T22:06:37 *** kiwichris has joined #rtems 2015-07-31T22:17:35 [trac] #2377/defect (new) updated by chrisj (I do not understand the issue. What does tools without a version mean ?) http://devel.rtems.org/ticket/2377 2015-07-31T22:29:44 *** kiwichris has quit IRC 2015-08-01T00:07:16 *** kiwichris has joined #rtems 2015-08-01T00:51:00 *** kiwichris has quit IRC 2015-08-01T05:54:57 *** kiwichris has joined #rtems 2015-08-01T05:57:28 *** kiwichris has quit IRC 2015-08-01T05:58:16 *** kiwichris has joined #rtems 2015-08-01T06:00:38 *** kiwichris has quit IRC 2015-08-01T06:02:14 *** kiwichris has joined #rtems 2015-08-01T06:05:53 *** kiwichris has quit IRC 2015-08-01T06:07:32 *** kiwichris has joined #rtems 2015-08-01T07:54:14 *** kiwichris has quit IRC 2015-08-01T10:24:36 *** hesham has joined #rtems 2015-08-01T12:24:39 *** hesham has quit IRC 2015-08-01T13:23:57 [trac] 0001-tftpDriver-apply-changes-through-master.patch (apply changes through master) attached to #2375 by mdavidsaver http://devel.rtems.org/ticket/2375 2015-08-01T13:24:15 [trac] 0002-tftpDriver-backport-fixes.patch (backport fixes) attached to #2375 by mdavidsaver http://devel.rtems.org/ticket/2375 2015-08-01T13:24:30 [trac] 0003-tftpDriver-don-t-free-directory-node-s-path-string.patch (don't free directory node's path string) attached to #2375 by mdavidsaver http://devel.rtems.org/ticket/2375 2015-08-01T14:45:41 [trac] GSoC/2015 edited by Asuol (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-08-01T16:54:51 [trac] #2378/defect (v:4.10) created by joel.sherrill (ampolish3 script can't always find perl) http://devel.rtems.org/ticket/2378 2015-08-01T17:00:04 [git] f8cf567 by Nick Withers: [ampolish3: Use env to find perl more reliably Fixes breakage on at least FreeBSD. closes #2379.] http://devel.rtems.org/changeset/f8cf567/rtems 2015-08-01T17:09:08 [git] dc47565 by Nick Withers: [ampolish3: Use env to find perl more reliably Fixes breakage on at least FreeBSD. closes #2379.] http://devel.rtems.org/changeset/dc47565/rtems 2015-08-01T17:12:44 [trac] #2378/defect (new) updated by joel.sherrill (closed by f8cf5670955d3955310c008ce453ff01fc8928ab) http://devel.rtems.org/ticket/2378 2015-08-01T17:12:59 [trac] #2378/defect (closed) updated by joel.sherrill (empty comment) http://devel.rtems.org/ticket/2378 2015-08-01T17:16:12 [git] 3699161 by Nick Withers: [Fixes breakage on at least FreeBSD. closes #2379.] http://devel.rtems.org/changeset/3699161/rtems 2015-08-01T17:19:54 [git] 7e798b4 by Jan Sommer: [Add __synch_synchronize function for arm Fixes link failure when linking Ada programs on the raspberry pi] http://devel.rtems.org/changeset/7e798b4/rtems 2015-08-01T17:20:28 [git] 1cf91e0 by Jan Sommer: [Add __synch_synchronize function for arm Fixes link failure when linking Ada programs on the raspberry pi] http://devel.rtems.org/changeset/1cf91e0/rtems 2015-08-01T17:52:07 *** crazy3k has joined #rtems 2015-08-02T02:52:08 *** kiwichris has joined #rtems 2015-08-02T07:27:26 *** igutek has quit IRC 2015-08-02T07:33:30 *** igutek has joined #rtems 2015-08-02T10:16:51 *** jkridner has quit IRC 2015-08-02T10:43:25 *** jkridner has joined #rtems 2015-08-02T10:52:09 *** jkridner has quit IRC 2015-08-02T11:44:09 *** jkridner has joined #rtems 2015-08-02T11:52:30 *** jkridner has quit IRC 2015-08-02T12:44:43 *** jkridner has joined #rtems 2015-08-02T12:49:23 *** jkridner has quit IRC 2015-08-02T13:45:15 *** jkridner has joined #rtems 2015-08-02T13:49:46 *** jkridner has quit IRC 2015-08-02T14:45:43 *** jkridner has joined #rtems 2015-08-02T14:50:06 *** jkridner has quit IRC 2015-08-02T15:29:55 [trac] #2379/defect (v:4.11) created by joel.sherrill (Ensure ada-tests build) http://devel.rtems.org/ticket/2379 2015-08-02T15:32:21 [trac] #2379/defect (closed) updated by Joel Sherrill (In [changeset:"d138a22b542a98b38afee731a08d6a4896f33e79/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2379 2015-08-02T15:32:22 [git] d138a22 by Joel Sherrill: [ada-tests/support/init.c: Fix compile errors and warnings closes #2379.] http://devel.rtems.org/changeset/d138a22/rtems 2015-08-02T15:33:16 [trac] #2379/defect (closed) updated by Joel Sherrill (In [changeset:"652888c22d3cdf8c326f44ea532584e55a1e3505/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2379 2015-08-02T15:33:16 [git] 652888c by Joel Sherrill: [ada-tests/support/init.c: Fix compile errors and warnings closes #2379.] http://devel.rtems.org/changeset/652888c/rtems 2015-08-02T18:56:21 *** commander has quit IRC 2015-08-02T18:56:29 *** commander has joined #rtems 2015-08-02T22:18:02 *** kiwichris has quit IRC 2015-08-02T22:35:09 *** kiwichris has joined #rtems 2015-08-02T23:59:28 *** mumptai has quit IRC