2015-08-10T01:38:30 *** gedare has joined #rtems 2015-08-10T01:38:30 *** ChanServ sets mode: +o gedare 2015-08-10T01:49:23 *** gedare has quit IRC 2015-08-10T04:07:30 *** crazy3k has quit IRC 2015-08-10T04:25:28 [trac] #2388/defect (v:4.10) created by nick.withers ([PATCH] [NFS client] Remove old CVS keywords) http://devel.rtems.org/ticket/2388 2015-08-10T04:27:01 [trac] nfs-cvs-keywords.patch () attached to #2388 by nick.withers http://devel.rtems.org/ticket/2388 2015-08-10T04:27:21 [trac] #2388/defect (new) updated by nick.withers (empty comment) http://devel.rtems.org/ticket/2388 2015-08-10T04:31:59 *** crazy3k has joined #rtems 2015-08-10T05:03:29 *** danieli has quit IRC 2015-08-10T05:08:46 *** monstr has joined #rtems 2015-08-10T05:56:16 *** danieli has joined #rtems 2015-08-10T09:09:31 *** ketul has joined #rtems 2015-08-10T10:05:51 *** crazy3k has quit IRC 2015-08-10T11:11:35 *** monstr has quit IRC 2015-08-10T11:11:53 *** monstr has joined #rtems 2015-08-10T11:45:54 hi beng-nl :) 2015-08-10T12:12:11 *** jkridner has joined #rtems 2015-08-10T12:21:25 *** stryx` has quit IRC 2015-08-10T12:22:25 *** ketul has quit IRC 2015-08-10T12:22:26 *** stryx` has joined #rtems 2015-08-10T12:48:42 *** danieli has quit IRC 2015-08-10T12:52:01 *** danieli has joined #rtems 2015-08-10T12:57:36 [trac] GSoC/2015 edited by yangqiao (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-08-10T12:58:10 *** danieli has quit IRC 2015-08-10T13:10:01 *** danieli has joined #rtems 2015-08-10T13:15:10 *** danieli has quit IRC 2015-08-10T13:39:37 *** gedare has joined #rtems 2015-08-10T13:39:37 *** ChanServ sets mode: +o gedare 2015-08-10T13:54:28 *** ketul has joined #rtems 2015-08-10T13:59:02 *** MrBIOS has joined #rtems 2015-08-10T14:08:05 *** monstr has quit IRC 2015-08-10T14:11:56 *** MrBIOS has quit IRC 2015-08-10T14:12:17 *** dr__house|away is now known as dr__house 2015-08-10T14:13:17 *** DrJoel has joined #rtems 2015-08-10T14:13:17 *** DrJoel has joined #rtems 2015-08-10T14:13:17 *** ChanServ sets mode: +o DrJoel 2015-08-10T14:33:07 hello ! currently writing code for gpio. Want some input for setting resistor (pull up/pull down). In BBB there is no sequential pad config register's offsets. And for gpio code I need to map it with pin number . So what are the preferred ways (thinking about array). Reference :- http://www.valvers.com/embedded-linux/beaglebone-black/step04-gpio/ 2015-08-10T14:44:33 Are you looking at using a device tree? 2015-08-10T14:45:39 *** MrBIOS has joined #rtems 2015-08-10T14:46:47 DrJoel: nop. Reference link for that table(Pin Register Description GPIO Designation Register Offset) 2015-08-10T14:48:06 *** DrJoel_ has joined #rtems 2015-08-10T14:48:30 *** DrJoel has quit IRC 2015-08-10T14:48:40 *** DrJoel_ has quit IRC 2015-08-10T14:48:56 *** DrJoel has joined #rtems 2015-08-10T14:48:57 *** DrJoel has joined #rtems 2015-08-10T14:48:57 *** ChanServ sets mode: +o DrJoel 2015-08-10T14:49:16 DrJoel: So that table (on that reference link) and TRM shows me that I can't map that offsets with simple mathematical manipulation . 2015-08-10T14:49:48 that type of setup at the BSP/driver level looks useful. You want to present the pins in banks based on how they appear physically on connectors. So letting the BSP "present" that view is a good thing 2015-08-10T14:55:58 umm. DrJoel . I need to update the pad_config register for setting resistor setting (pull up / down) . So for every gpio pin offset is given in TRM (or that table ) . So for that I need to do mapping of that offsets with the pin_no. So which way is preferred (array or ).. because simple mathematical operation is not possible as it is not linear 2015-08-10T14:59:18 Is the user presented a logical pin number that you could look up? 2015-08-10T15:00:27 yep DrJoel . pin_no varies from 1 to 128 (4 gpio banks * 32 pins each) 2015-08-10T15:04:59 Can you use that to look up things in the table? 2015-08-10T15:06:56 Yes ! But that table is not having linear mapping so for look up I need to have one look up table. Which method would be used to define that look up table. Is it array a only option or any others ? 2015-08-10T15:07:51 At the BSP level, it is the only option. But can't you call pins 0-31 and map them arbitrarily in your table. So a connector with 32 pins could map to various registers? 2015-08-10T15:13:05 But how as offsets are fixed for the specified gpio pin and that is not a linear. 2015-08-10T15:15:33 and DrJoel the table should be declared in gpio.c right ? or gpio.h would be preferred ? 2015-08-10T15:15:59 I may be out of sync but don't the user requests get mapped to hardware? Is there a need to do it the other way? 2015-08-10T15:17:10 I would think the table was private to the implementation on a system (e.g. board + add-on cards). Number of pins is generic info, pin # to HW mapping, resistor, direction, etc. is internal to implementation. 2015-08-10T15:20:19 Yep DrJoel . GPIO API is merged developed by andre. And that implements what you thought about generic api. 2015-08-10T15:21:33 so I was talking about bbb-gpio.h and bbb-gpio.c :) 2015-08-10T15:25:01 Try to map those from user to hw as easily as you can. But make it logically easy to user. Pain should be on hw adaption/mapping. 2015-08-10T15:26:03 ok sure. Thanks DrJoel :) 2015-08-10T15:33:51 *** ketul has quit IRC 2015-08-10T15:49:21 *** hesham has joined #rtems 2015-08-10T16:23:55 *** ragunath has joined #rtems 2015-08-10T16:24:06 hi DrJoel 2015-08-10T16:27:47 hello.. going through the graphics packages now. Then back to yours. :) 2015-08-10T16:36:20 DrJoel:you need beaglebone black to run the app :) 2015-08-10T16:36:55 DrJoel:you were going to just compile it or want to run an app 2015-08-10T16:37:47 I am in the process of cleaning the drivers patch .. I can write a blog post explaining how to run it 2015-08-10T16:51:18 *** krohini has joined #rtems 2015-08-10T16:53:39 Did you see the question on the mailing list? 2015-08-10T16:53:55 First step is just to get it compiling. 2015-08-10T16:54:07 The question of credit needs to be addressed. 2015-08-10T16:57:07 Most of the work is from what Daniel/Marcos has done 2015-08-10T16:57:23 Their work creates a library including both lwip & driver 2015-08-10T16:57:50 If the credit is properly attributed in the files, then OK. Just reply to their post on the devel list. :) 2015-08-10T16:58:49 Actually their files did not have any of their names.. it just had lwip copyright notice .. no specific author names 2015-08-10T16:58:53 I will reply them 2015-08-10T17:01:16 [git] 80f3b2b by YANG Qiao: [add t1lib patch] http://devel.rtems.org/changeset/80f3b2b/rtems-source-builder 2015-08-10T17:01:17 [git] 0b80ff5 by YANG Qiao: [add libpng bset] http://devel.rtems.org/changeset/0b80ff5/rtems-source-builder 2015-08-10T17:01:17 [git] c22fc6d by YANG Qiao: [add microwindows patches] http://devel.rtems.org/changeset/c22fc6d/rtems-source-builder 2015-08-10T17:01:17 [git] d2ee14c by YANG Qiao: [add tiff-4.0.2 bset] http://devel.rtems.org/changeset/d2ee14c/rtems-source-builder 2015-08-10T17:01:17 [git] d9ee8ae by YANG Qiao: [add freetype2 bset] http://devel.rtems.org/changeset/d9ee8ae/rtems-source-builder 2015-08-10T17:01:17 [git] 457c7ec by YANG Qiao: [add t1lib bset (patch url should be changed after it is merged into rtems git server)] http://devel.rtems.org/changeset/457c7ec/rtems-source-builder 2015-08-10T17:01:18 [git] a7d2db0 by YANG Qiao: [add nanoX bset] http://devel.rtems.org/changeset/a7d2db0/rtems-source-builder 2015-08-10T17:01:18 [git] 68e81c7 by YANG Qiao: [add nxlib bset] http://devel.rtems.org/changeset/68e81c7/rtems-source-builder 2015-08-10T17:01:19 [git] 01f8d0c by YANG Qiao: [fix libjpeg conf] http://devel.rtems.org/changeset/01f8d0c/rtems-source-builder 2015-08-10T17:01:34 Thanks. The attribution must be correct. I will focus on building it. 2015-08-10T17:01:57 Daniel is right that moving it from the POSIX API to Classic will reduce footprint a little more. 2015-08-10T17:02:38 Maybe it is possible to have two backend/ports for RTEMS in their tree. I have seen this on other packages ported to RTOSes. Use POSIX or native API. 2015-08-10T17:03:40 oh ok I didnt know that .. it is currently using posix API 2015-08-10T17:03:54 1) get it working. 2) optimize 2015-08-10T17:04:07 +1 2015-08-10T17:04:10 also, i'd rather that you can help yurii since he is badly stuck 2015-08-10T17:04:11 we need a baseline. 2015-08-10T17:04:28 let users who care pay ;) 2015-08-10T17:04:36 gedare:yep that is good 2015-08-10T17:04:49 He is on my list. And the one I meant to say "blind leading the blind" about. I still am fighting the configuration myself 2015-08-10T17:05:10 gedare:I will look into yurri's changes 2015-08-10T17:05:14 ok great thanks 2015-08-10T17:05:16 i have to run now 2015-08-10T17:05:19 see you tomorrw 2015-08-10T17:05:26 gedare:bye 2015-08-10T17:06:15 gedare: I am prone to commit the lwip RSB changes if they build. The tag and patch source can be adjusted. What do you think? 2015-08-10T17:07:16 DrJoel: the lwip does not have any driver 2015-08-10T17:07:56 Does it build without a driver? That was one of my questions on gmail. Is it a base package that a user can build, then locally add a driver to when compiling their application? 2015-08-10T17:08:14 DrJoel: as per the discussion with chris,we can include the driver in separate repo & compile it separately 2015-08-10T17:08:23 Yes it builds without a driver 2015-08-10T17:08:44 So the RSB lwip package is just a base... should build for any BSP.. then add driver for real use? 2015-08-10T17:08:59 *** MrBIOS has quit IRC 2015-08-10T17:09:25 *** MrBIOS has joined #rtems 2015-08-10T17:09:44 yes thats the goal 2015-08-10T17:09:59 finally we will have 3 bset files 2015-08-10T17:10:23 OK.. but the base one is what I have now. 2015-08-10T17:10:24 *** gedare has quit IRC 2015-08-10T17:10:27 beagle-lwip.bset which will call lwip.bset which builds the base lwip 2015-08-10T17:10:52 Where are the tests a user could check this out with. 2015-08-10T17:10:52 then it calls the bsp specific bset file which builds the driver 2015-08-10T17:12:24 tests have to be written separately using the lwip api 2015-08-10T17:15:03 *** MrBIOS has quit IRC 2015-08-10T17:24:28 What test did you use? 2015-08-10T17:25:37 hello DrJoel 2015-08-10T17:26:00 I have a basic application, which gets the ip address using dhcp. 2015-08-10T17:26:12 So ARP, DHCP and ping are working 2015-08-10T17:32:03 *** mumptai has joined #rtems 2015-08-10T17:44:57 *** ketul has joined #rtems 2015-08-10T17:45:08 That should be enough to proof it works. 2015-08-10T17:45:36 Have you looked at the ttcp application from network demos? That is the next thing I recommend trying to run. It measures throughput and puts some stress on the stack and driver 2015-08-10T17:47:16 ** checking tcp app 2015-08-10T17:52:53 we have to check whether the app is compatible with RTEMS 2015-08-10T17:53:25 rtems-libbsd apps are not compatible with lwip 2015-08-10T17:54:10 DrJoel: are we clear about the attribution for the port? Is there anything else should I mail? 2015-08-10T18:00:54 There is a version of ttcp in network-demos. 2015-08-10T18:01:25 I haven't checked the files/readme. Are the RTEMS specific files and log entries clear about the attribution? 2015-08-10T18:01:38 I think I understand it but that's not the same as a permanent record. 2015-08-10T18:04:17 brb 2015-08-10T18:06:45 *** hesham has quit IRC 2015-08-10T19:26:18 *** DrJoel has quit IRC 2015-08-10T19:48:44 *** krohini has quit IRC 2015-08-10T19:49:39 *** ragunath has quit IRC 2015-08-10T20:00:00 *** krohini has joined #rtems 2015-08-10T20:01:56 *** ketul has quit IRC 2015-08-10T20:03:23 *** ketul has joined #rtems 2015-08-10T20:14:56 *** ankur has joined #rtems 2015-08-10T20:27:27 *** krohini has quit IRC 2015-08-10T20:31:04 *** ankur has quit IRC 2015-08-10T20:31:44 *** ankur has joined #rtems 2015-08-10T20:41:20 *** ankur has quit IRC 2015-08-10T21:23:40 *** mirspace has joined #rtems 2015-08-10T21:38:51 *** mirspace has left #rtems 2015-08-10T21:42:02 *** mirspace has joined #rtems 2015-08-10T21:42:33 *** mirspace has left #rtems 2015-08-10T22:46:19 *** ketul has quit IRC 2015-08-10T22:54:03 *** danieli has joined #rtems 2015-08-10T23:01:58 *** gedare has joined #rtems 2015-08-10T23:01:58 *** ChanServ sets mode: +o gedare 2015-08-10T23:12:29 *** gedare has quit IRC 2015-08-11T00:07:46 *** mumptai has quit IRC 2015-08-11T02:27:55 *** danieli has quit IRC 2015-08-11T03:28:24 *** crazy3k has joined #rtems 2015-08-11T04:00:55 *** ketul has joined #rtems 2015-08-11T04:36:22 *** monstr has joined #rtems 2015-08-11T05:40:47 *** danieli has joined #rtems 2015-08-11T06:21:04 *** sujayraaj has joined #rtems 2015-08-11T06:52:33 *** sujayraaj has quit IRC 2015-08-11T07:33:30 *** dspace has joined #rtems 2015-08-11T07:34:50 *** ita_ has joined #rtems 2015-08-11T07:35:43 *** ita has quit IRC 2015-08-11T08:06:03 *** hesham has joined #rtems 2015-08-11T08:42:56 [trac] GSoC/2015 edited by ketul (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-08-11T09:31:29 *** ketul has quit IRC 2015-08-11T11:24:42 *** ketul has joined #rtems 2015-08-11T11:40:55 *** sujayraaj has joined #rtems 2015-08-11T11:48:55 *** sujayraaj has quit IRC 2015-08-11T12:21:22 *** igutek has quit IRC 2015-08-11T12:26:51 [trac] GSoC/2015 edited by ragunath (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-08-11T12:40:19 *** gedare has joined #rtems 2015-08-11T12:40:19 *** ChanServ sets mode: +o gedare 2015-08-11T12:57:00 *** hesham has quit IRC 2015-08-11T13:01:27 *** asuolen has joined #rtems 2015-08-11T13:05:04 ketul: put else { on same line as }, e.g. 2015-08-11T13:05:06 + } 2015-08-11T13:05:06 + else { 2015-08-11T13:05:12 + } else { 2015-08-11T13:05:23 just one line 2015-08-11T13:05:50 ok gedare. will do. 2015-08-11T13:07:54 ketul: if you only use arm_delay for a 5-cycle delay, I doubt you need it 2015-08-11T13:08:05 5 cycles will elapse by the time the function returns a value. 2015-08-11T13:08:29 ya right. But it is listed in TRM. 2015-08-11T13:08:43 just leave a comment there, or insert 5 nops directly 2015-08-11T13:08:48 you don't need a function with a loop 2015-08-11T13:09:49 ok. In TRM "When the clocks are enabled by writing to the GPIO_LEVELDETECT0, 2015-08-11T13:09:50 GPIO_LEVELDETECT1, GPIO_RISINGDETECT, and GPIO_FALLINGDETECT registers, 2015-08-11T13:09:50 the detection starts after 5 clock cycles. This period is required to clean the synchronization 2015-08-11T13:09:50 edge/level detection pipeline." 2015-08-11T13:12:56 yeah i'd just put in 5 nops then 2015-08-11T13:13:13 and remove the arm_delay function 2015-08-11T13:13:23 except... 2015-08-11T13:13:35 how come it only gets used at rtems_bsp_enable_interrupt 2015-08-11T13:13:38 and not in disable_interrupt? 2015-08-11T13:17:21 I thought in TRM this is given for after "clocks are enabled by ...." . Anyway good catch. Will put into disable() too. 2015-08-11T13:20:06 *** ketul has quit IRC 2015-08-11T13:36:04 [git] 2dfb410 by Joel Sherrill: [Revert "add microwindows patches" This reverts commit c22fc6df4e972bd0cb8852421ba73475f443c461. Patches should be in rtems-tools.] http://devel.rtems.org/changeset/2dfb410/rtems-source-builder 2015-08-11T13:36:04 [git] 9092cce by Joel Sherrill: [Revert "add t1lib patch" This reverts commit 80f3b2bb5ae85fdd13065e3d93ca0a809874337c. Patches should be in rtems-tools.] http://devel.rtems.org/changeset/9092cce/rtems-source-builder 2015-08-11T13:37:57 [git] 45d4773 by YANG Qiao: [add microwindows patches] http://devel.rtems.org/changeset/45d4773/rtems-tools 2015-08-11T13:41:14 *** ketul has joined #rtems 2015-08-11T13:50:42 *** jkridner has quit IRC 2015-08-11T13:52:04 *** jkridner has joined #rtems 2015-08-11T14:01:13 *** jkridner has quit IRC 2015-08-11T14:02:03 *** jkridner has joined #rtems 2015-08-11T14:03:42 *** dougk_ has quit IRC 2015-08-11T14:11:15 *** hermann_19829_ has joined #rtems 2015-08-11T14:12:05 *** hermann_19829_ has quit IRC 2015-08-11T14:13:13 *** hermann_19829 has joined #rtems 2015-08-11T14:32:50 *** anandkp has joined #rtems 2015-08-11T14:34:02 *** sujayraaj has joined #rtems 2015-08-11T14:40:30 *** sujayraaj has quit IRC 2015-08-11T14:43:05 [trac] GSoC/2015 edited by krohini (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-08-11T14:43:21 *** gtament has joined #rtems 2015-08-11T14:43:27 Hi! 2015-08-11T14:43:35 Hello, everybody! 2015-08-11T14:44:16 hi! 2015-08-11T14:45:33 [git] 3a3069e by ragunath: [lwIP configuration & build files Patch to build lwip with RTEMS Resource builder. It builds the lwip sources and creates a library.] http://devel.rtems.org/changeset/3a3069e/rtems-source-builder 2015-08-11T14:49:53 *** ragunath has joined #rtems 2015-08-11T14:50:15 *** anandkp has quit IRC 2015-08-11T14:52:56 *** jrcatbagan has joined #rtems 2015-08-11T14:52:59 *** salpha has joined #rtems 2015-08-11T14:53:20 hi everyone! 2015-08-11T14:54:09 *** sgworks has joined #rtems 2015-08-11T14:54:11 *** sujayraaj has joined #rtems 2015-08-11T14:56:30 Hello, all ! 2015-08-11T14:57:37 *** EdSut has joined #rtems 2015-08-11T14:57:45 *** anandkp has joined #rtems 2015-08-11T14:58:18 *** anandkp92 has joined #rtems 2015-08-11T14:58:21 hi ketul 2015-08-11T14:58:32 hello beng-nl 2015-08-11T14:58:41 hi beng-nl 2015-08-11T14:58:45 hi ragunath 2015-08-11T14:58:47 hi beng-nl 2015-08-11T14:58:50 ragunath: i'm sorry i didn't reply to your email 2015-08-11T14:59:00 ragunath: how are you doing? 2015-08-11T14:59:11 no problem .. good 2015-08-11T14:59:38 I continued with lwIP 2015-08-11T15:00:30 Good morning! 2015-08-11T15:00:33 And Evening! 2015-08-11T15:00:36 hi gedare :) 2015-08-11T15:00:42 Hi, Gedare! 2015-08-11T15:00:46 and afternoon if anyone is in the central-atlantic region 2015-08-11T15:00:49 hello gedare and all 2015-08-11T15:00:51 which I don't believe is the case. 2015-08-11T15:00:55 hi gedare 2015-08-11T15:01:00 hey beng-nl welcome back 2015-08-11T15:01:05 thank you :) 2015-08-11T15:01:12 hey folks! 2015-08-11T15:01:15 hi Gedare 2015-08-11T15:01:16 hi EdSut 2015-08-11T15:01:24 hello EdSut 2015-08-11T15:01:30 Hi gedare 2015-08-11T15:01:36 gedare: hello 2015-08-11T15:02:14 *** krohini has joined #rtems 2015-08-11T15:02:25 hi gedare 2015-08-11T15:02:29 hello 2015-08-11T15:02:40 I'm glad to see we have a good turnout 2015-08-11T15:03:27 Umm yeah, there has been some progress :) 2015-08-11T15:03:29 I have a bunch of announcements before we start 2015-08-11T15:03:58 Announcements for GSoC Students: 2015-08-11T15:04:01 1. Work on your section in https://docs.google.com/document/d/10WTXnliwO5w5pIxwCWSOgfmMZsqV7EwkTDkiQUqGDW4/edit?usp=sharing 2015-08-11T15:04:05 2. "Pencils Down" is on Monday August 17. Identify any final push you want to make, and start to plan your code submission and documentation efforts. 2015-08-11T15:04:09 3. GSoC coding ends on August 21. Do NOT submit any code to Google that is written after that period. You should take a snapshot (push a git branch) of your code at or near the end of the term. 2015-08-11T15:04:12 4. However, you ARE encouraged to continue working on your projects and submitting. We hope you stick around and continue to participate with RTEMS as your time permits. 2015-08-11T15:04:15 5. For submitting code to Google, I recommend a zip of patches. If you get all your code ready for submission, you just can upload that to Google in addition to sending to devel mailing list. 2015-08-11T15:04:19 6. This is our last "all hands" meeting. Address further questions to the mailing list or by individual contact by email or on IRC. 2015-08-11T15:04:29 SOCIS Students: 2015-08-11T15:04:32 1. Work on your section in https://docs.google.com/document/d/1aX7JgHYJjtrAjW3ufgOZfnZraY9U0yZcLyt7zhuykqE/edit?usp=sharing 2015-08-11T15:04:35 2. This is our last "all hands" meeting. Stay in touch by e-mail, and you should also be looking at code cleanup and final deliverables for the coding period. 2015-08-11T15:05:23 Note: the wrap-up reports are different gdocs for GSoC and SOCIS. We expect you to do some preliminary writing in there, and to finalize it before the evaluation period ends. 2015-08-11T15:05:49 (Let's use the posted evaluation period for GSoC, which is Friday August 28) 2015-08-11T15:07:24 Since this is our last meeting, I would like each of you to focus on reporting what you will be doing for your final push and then your cleanup/deliverables. 2015-08-11T15:07:50 Let's begin with anandkp92. Hi anandkp92. 2015-08-11T15:07:55 Hi ! 2015-08-11T15:08:06 This week I have included event handlers for the remaining menu like view BSP chosen etc. 2015-08-11T15:08:21 However, the main part of the GUI is complete in the sense that it is able to choose BSP, obtain corresponding options, modify the values and create a config.cfg file with these values. 2015-08-11T15:08:59 Currently I am trying to do 2 things: 1) run waf build from within the gui - Thomas had suggested something to do this. 2015-08-11T15:09:03 I'll be trying it now 2015-08-11T15:09:15 2) Translation of GUI. 2015-08-11T15:09:34 To be clear, the waf build works fine if done from the cmd line after creating the cfg? 2015-08-11T15:09:40 y es 2015-08-11T15:09:47 What do you mean "translation"? 2015-08-11T15:10:09 aah thomas asked me to look into translating the gui into a different language 2015-08-11T15:10:12 oh, gotcha 2015-08-11T15:10:17 hmmm 2015-08-11T15:10:36 Ok yes, it should be investigated to be made easy to do 2015-08-11T15:10:58 yeah i will check it out 2015-08-11T15:11:27 other than that there are a few minor bugs - like we get an error when we choose a specific pair of BSPs 2015-08-11T15:11:52 we are looking into it right now 2015-08-11T15:12:01 OK if you can't fix them make sure you document them clearly 2015-08-11T15:12:19 amar said he'll also check it out cause it's from the original code 2015-08-11T15:12:27 yes. 2015-08-11T15:12:39 OK thanks anandkp92 it sounds like you're right on track 2015-08-11T15:13:03 next up gtament, you can go 2015-08-11T15:13:13 i guess thats it for now.. 2015-08-11T15:13:18 unless you had something else? 2015-08-11T15:13:27 no.. thats it :) thanks 2015-08-11T15:14:03 I finally got driver loaded. And of course, it's not working as expected, so I'm debuging right now. And I'm going to work for days and nights to make up for lost time. Since not much time left, I'm going to submit only ported driver. But I won't stop on it 2015-08-11T15:16:31 ping 2015-08-11T15:16:33 ok gtament thanks I know you had problems 2015-08-11T15:16:55 I'm glad you are through them. have you had time to write your blog post about your efforts? 2015-08-11T15:17:49 actually there is nothing to wrtite about. As I expected the problem was in one missing line of code 2015-08-11T15:18:20 ok, well what I really also mean to ask is whether you have anything to write regarding the effort, but yes I understand 2015-08-11T15:18:21 [trac] GSoC/2015 edited by anandkp92 (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-08-11T15:18:45 gtament: I would also like you to update in https://devel.rtems.org/wiki/GSoC/2015#YuriiShevtsov 2015-08-11T15:19:11 Ok, and also I have to clean up my repo and update wiki. I remember 2015-08-11T15:19:14 i know there didn't feel like much to write about because you were struggling, but it is important to at least capture that 2015-08-11T15:19:16 yes, good 2015-08-11T15:19:41 ok thanks gtament anything else? 2015-08-11T15:19:57 I'm done. Will go debugging 2015-08-11T15:20:01 thx! 2015-08-11T15:20:03 hermann_19829: you can go now 2015-08-11T15:20:19 Hi! 2015-08-11T15:20:21 This week I sent the patches which are required to run the coverage report generator to the devel list. Then I investigated the frontend of the couverture project, which is used to create coverage reports from execution traces as well. They support statement, decision, and MC/DC coverage analysis for their reports but only for programs written in Ada. For that they create Ada library information files which are generated by adding 2015-08-11T15:20:21 the -fdump-scos switch when compiling the Ada code with the GNAT compiler. 2015-08-11T15:20:21 I investigated these things because I think we also need something similar to the Ada library information file format. We need these annotated source files to reconstruct decisions and conditions from the executed branches. 2015-08-11T15:21:51 ok, when you make some progress there please write what you discover.. 2015-08-11T15:22:37 yes! I will have to discuss this plan also with DrJoel. 2015-08-11T15:23:15 good. i wonder if there are other tools that might be better supported out of the box 2015-08-11T15:23:53 alright, thanks hermann_19829. you should also discuss with Joel about your next steps. i know he wants to get some improvements to the modularity of coverage testing 2015-08-11T15:24:15 ok! 2015-08-11T15:24:22 Thanks gedare! 2015-08-11T15:25:27 alright 2015-08-11T15:25:34 salpha: you can go now 2015-08-11T15:25:51 i've gone deeper through Slingshot in this week to know how does it generate and execute test suites. 2015-08-11T15:25:57 now i've manually ported two test programs (one for "abs", one for "free") to rtems. 2015-08-11T15:26:04 after compilation and execution, these modified test programs are running inside QEMU. 2015-08-11T15:26:11 now my biggest problem is how to deal with different process/thread models between linux and rtems. 2015-08-11T15:26:17 since Slingshot is using linux' process model hence utilizing functions like fork, waitpid, etc. to execute and monitor child which is running module under test. 2015-08-11T15:26:23 i have to figure out how to adapt rtems to rtems' process model and set up appropriate signal handlers and find a proper way to run module under test. 2015-08-11T15:27:16 ahh.. 2015-08-11T15:27:27 *adapt Slingshot 2015-08-11T15:27:46 yeah. have you upated your blog with info on Slingshot? 2015-08-11T15:27:52 I haven't had a chance to look at blogs recently 2015-08-11T15:28:06 *** anandkp92 has quit IRC 2015-08-11T15:28:30 not the Slingshot itself since like GRINDER, i think this belong's to Slingshot's repository (which doesn't publicly exist yet). 2015-08-11T15:28:43 but how to adapt it to rtems has been partly covered. 2015-08-11T15:28:51 i mean documented. 2015-08-11T15:29:27 I think it is also useful for you to capture your thoughts and understanding in the blog. it may be useful for yourself later, or others. even if only related to slingshot/GRINDER 2015-08-11T15:29:56 ok, sure. i'll do it. almost a lot to document about that. 2015-08-11T15:30:17 Yeah. I need to get a clear picture of what we can accomplish for sure by the end of the SOCIS term, or what should be the claimed deliverables 2015-08-11T15:30:25 if you have some ideas to help clarify that topic, please send by email 2015-08-11T15:30:44 ok 2015-08-11T15:30:59 good idea to clarify it more. 2015-08-11T15:31:04 thanks salpha. up next is ketul, hi ketul 2015-08-11T15:31:06 hi beng-nl 2015-08-11T15:31:12 hi gedare and beng-nl 2015-08-11T15:31:14 thanks gedare. 2015-08-11T15:31:23 This week I mainly worked on gpio driver code using the GPIO API merged. Finally, I have come up with some good code. Now will wait for some time for review. And currently doing all demo and finding the bugs if in that code. For I2C part I almost done with the code but not getting desired output by hardware testing. Anyway for that I need to debug 2015-08-11T15:31:24 it once more. I will try my best for merging gpio driver in upcoming time ! 2015-08-11T15:31:45 +1 2015-08-11T15:31:56 i'll help to get that done asap 2015-08-11T15:31:59 ok ketul thanks i gave you my feedback on your code already 2015-08-11T15:32:08 beng-nl: sure thanks. 2015-08-11T15:32:19 gedare: yep changed accordingly 2015-08-11T15:32:36 i didn't read closely for correctness of course, mostly style. Fix that "} else {" issue and replace arm_delay with explicit nops, and add to the disable func 2015-08-11T15:32:38 ok 2015-08-11T15:32:52 gedare: yep done 2015-08-11T15:33:01 thanks ketul 2015-08-11T15:33:12 thanks gedare and beng-nl :) 2015-08-11T15:33:15 ragunath: you can go now 2015-08-11T15:33:21 hi all 2015-08-11T15:33:31 I continued my work on integrating lwip to RTEMS. I had a discussion with chris on how to integrate the changes. 2015-08-11T15:33:34 [trac] SOCIS/2015 edited by salpha (empty comment) http://devel.rtems.org/wiki/SOCIS/2015 2015-08-11T15:33:41 He suggested we separate the lwip base and driver part. This way the lwip base remains stable while the driver part 2015-08-11T15:33:41 can be frequently updated without affecting the package. 2015-08-11T15:33:59 The ideal scenario is have a lwip bset file for a bsp which will build lwip & then it will choose the respective driver & 2015-08-11T15:33:59 build it. 2015-08-11T15:34:06 I separated the lwip base and built it using RSB and the changes has been submitted. 2015-08-11T15:34:17 I am preparing the patch for driver side. I am not sure if it can be integrated because of licensing issues. 2015-08-11T15:34:26 *** hermann_19829 has quit IRC 2015-08-11T15:34:35 *** krohini has quit IRC 2015-08-11T15:34:42 gedare: what can be done about driver side patch? 2015-08-11T15:35:03 ragunath: the licensing issue matters the most for code we put into rtems.git 2015-08-11T15:35:24 for patches that can be pulled in via the RSB, the problem is a bit different 2015-08-11T15:35:38 there we instead need to focus on providing the clear identification of the licensing requirements imposed on the end user 2015-08-11T15:36:08 what chris suggested is we have repo for lwip drivers and build it as per bsp requireents 2015-08-11T15:36:12 in other words, we make it clear to the user what the license is on the code they ask to build, and they can agree to use it or not based on those terms. 2015-08-11T15:36:12 *requirements 2015-08-11T15:36:25 *** hermann_19829 has joined #rtems 2015-08-11T15:36:28 Yeah that should be fine 2015-08-11T15:36:34 fine by me ... I will send the patch soon 2015-08-11T15:36:46 We just want to avoid mixing licenses in one repo, or creating a confusing situation 2015-08-11T15:36:55 if that makes sense. 2015-08-11T15:37:17 Then I can write a blog post on how to run lwip based app on BBB 2015-08-11T15:37:45 yeah that will be great to round out the project 2015-08-11T15:38:08 and getting more doc done, in the repos or on the trac wiki 2015-08-11T15:38:20 ok thanks ragunath anything else? 2015-08-11T15:38:27 Sure I will be happy to do it 2015-08-11T15:38:35 thats it from me thanks gedare 2015-08-11T15:38:52 ok jrcatbagan you can go now. Hi EdSut 2015-08-11T15:39:02 Hi gedare, EdSut, beng-nl, and all! 2015-08-11T15:39:03 *** salpha has quit IRC 2015-08-11T15:39:09 Since the last meeting, I cleaned up the mmc interface to the onboard eMMC that I implemented, submitted the patches, and have been merged with uMon master. Furthermore, I cleaned up the BBB port as much as possible with astyle to be consistent with the coding conventions established. While I was doing this, I was simultaneously looking into Ethernet and having been using the AM335x TRM, Section 14 "Ethernet Subsystem", IEEE 802.3 2015-08-11T15:39:10 Specification, and the datasheet for the LAN8710A PHY on the board as reference. 2015-08-11T15:39:10 2015-08-11T15:39:10 Up until the final push, I will be focusing on cleaning up the BBB port and developing documentation. I will continue working on implementing Ethernet, which I reckon that I'll have it ready after GSOC. 2015-08-11T15:39:13 2015-08-11T15:39:16 Should the BBB port of uMon be integrated with RSB with the functionality that it currently has or should it wait untl the port is completely finished? 2015-08-11T15:39:22 *** krohini has joined #rtems 2015-08-11T15:40:33 i'd like to see it done 2015-08-11T15:40:59 it is mature enough to be useful I believe, so we should push it forward and get it into shape that makes it easy to use 2015-08-11T15:41:34 One interesting test would be to launch a simple RTEMS app using what is there. 2015-08-11T15:41:35 Chris and Joel should be good assets for this, and also some of your fellow students who have worked with RSB this summer. 2015-08-11T15:42:00 Oh, yes, that too should be a goal even before ethernet 2015-08-11T15:42:17 the infrastructure is sufficient for the proof-of-concept port we wanted to have 2015-08-11T15:42:24 If that works, then it "could" be used as is. 2015-08-11T15:42:27 Exactly 2015-08-11T15:42:39 I would like to see, in the following order: 2015-08-11T15:42:53 1) Cleanup and Documentation [never-ending struggle] 2015-08-11T15:43:06 to boot an app, this would require TFS-less mechanisms or implementing TFS, am I right, which both dont exist yet 2015-08-11T15:43:07 2) RTEMS Samples booting 2015-08-11T15:43:21 3) RSB support 2015-08-11T15:43:27 4) Ethernet in uMon for BBB 2015-08-11T15:44:04 Well, we can discuss the requirements to get #2 accomplished over email 2015-08-11T15:44:09 Jrcatbagan: I would start by just testing from the command line. 2015-08-11T15:44:15 #1 is necessary step for your GSoC wrap-up 2015-08-11T15:44:42 EdSut: I see, ok 2015-08-11T15:44:46 gedare: ok, will do 2015-08-11T15:44:48 Install an RTEMS image in mmc somewhere, then just launch it with mmc read and "call" 2015-08-11T15:44:56 The rest are possible future paths that can be explored in any order, but I prefer the order I gave. 2015-08-11T15:45:11 It will be good for you to include these future works in your final report/deliverables 2015-08-11T15:45:21 That order makes sense. Just don't forget about cache. 2015-08-11T15:45:26 ok thanks jrcatbagan anything else? 2015-08-11T15:45:40 gedare: that is all I have, thanks 2015-08-11T15:45:47 EdSut: will do, thanks 2015-08-11T15:46:19 sgworks: you can go next 2015-08-11T15:46:21 hi sgworks 2015-08-11T15:46:26 1. I will submit patch for uniprocessor implementation by tomorrow on devel-list. 2. Also submit the test cases results for the patch. 3. I have SMP implementation under JPF but not sure whether I will be able to finish its rtems implementation by the deadline so just focusing on uniprocessor implementation. 4. Next week I completely focusing on the documentation of JPF implementation for both architectures. 2015-08-11T15:46:31 Hi Gedare 2015-08-11T15:46:55 That sounds like a great plan to me 2015-08-11T15:47:00 Do you have any questions? 2015-08-11T15:47:25 Umm may be if I get stuck with implementation I will post on devel-list 2015-08-11T15:47:32 For your patches, please provide a patch that applies against 4.11, and one against master. Or just one patch if it applies cleanly to both. 2015-08-11T15:47:33 but I feel that I should be fine 2015-08-11T15:47:52 and it should not take time as we know the system in and out 2015-08-11T15:47:56 There shouldn't be much difference between 4.11 and master 2015-08-11T15:48:03 famous last words 2015-08-11T15:48:10 but you do have until Monday. :) 2015-08-11T15:48:19 yeah no problem 2015-08-11T15:48:29 (For GSoC purpose. Just keep pushing on it 'til you get it.) 2015-08-11T15:48:33 asuolen: you can go now 2015-08-11T15:48:38 hello 2015-08-11T15:48:43 SPI and I2C bus work is finished, but I still haven't heard from my mentors about it. Currently I am working within rtems-libbsd using the freebsd/sys/dev/mmc implementation of the SD simplified specs and tying with the Pi's EMMC module. A similar situation seems to be in use for the altera cyclone at rtemsbsd/sys/dev/dw_mmc. 2015-08-11T15:48:48 The plan is to have something working on the SD card by the soft pencils down so I can focus on cleanup and documentation (specially for the GPIO API, which currently is mostly doxygen and some blog posts) on a wiki page. 2015-08-11T15:50:17 Your mentors are quite busy, so you might want to send to devel if you feel comfortable with the code and the testing you have done on it. 2015-08-11T15:50:43 ok will do that then 2015-08-11T15:50:58 ok, good. The GPIO documentation is the most important. 2015-08-11T15:51:35 Thanks asuolen 2015-08-11T15:51:58 anything else asuolen? sujayraaj: you're up 2015-08-11T15:52:14 Okay. Sure. 2015-08-11T15:52:29 no that is all, thanks gedare. 2015-08-11T15:53:12 1. Most of things regarding monkey are done except for a bug that is somehow making it run too slow for regular use 2. Code has been cleaned, and a README.rtems added 2015-08-11T15:53:16 3. Use of monkey as a static library with which an application can link against has been tested. I created an example application ( that can be pushed into examples-v2 ). The user can let monkey run as a background webserver task and do other work or something along these lines. 2015-08-11T15:53:26 4. Cmake support was cleaned so that a user can build it in an easier way ( it still ain't that easy. ) 2015-08-11T15:53:36 Last Push : I am working now on RSB recipe for monkey. I started today afternoon and hopefully will be done by tonight. I can attempt building libbsd by RSB too. It is a bit more straightforward to do. 2015-08-11T15:53:45 I will probably have a discussion with chris tomorrow morning on what else I can do before the deadline, except for working on that bug. All that will be left would be Documentation. 2015-08-11T15:54:25 ok sujayraaj that sounds great 2015-08-11T15:54:40 and if you don't get the bug fixed, make sure you describe it and what you have found out about it. 2015-08-11T15:55:04 a bug report should be created probably in our Trac for it, if the bug remains after you finish 2015-08-11T15:55:36 Okay. I will make sure I do that. I also had plans on making network-demos build through waf. I will do that if I get time. 2015-08-11T15:56:17 hmm, doesn't it? 2015-08-11T15:56:41 *** EdSut has quit IRC 2015-08-11T15:56:50 ah, i guess we didn't deal with that one 2015-08-11T15:56:53 ok, thanks sujayraaj 2015-08-11T15:56:59 example-v2 does, network-demos don't 2015-08-11T15:57:00 krohini: you're up 2015-08-11T15:57:15 oh. Thanks gedare. :) 2015-08-11T15:57:35 sujayraaj: sure, it should be pretty easy. i wouldn't put it as high priority. if we participate in GCI again it is a good set of tasks for high schoolers. 2015-08-11T15:57:46 ok so I have updated my blog with everything that I have tried..problems and fixes.. 2015-08-11T15:57:46 although i don't know if we will, GCI takes a lot of work... 2015-08-11T15:57:59 speaking of which, any GSoC/SOCIS students are welcome to pitch-in as mentors for GCI. 2015-08-11T15:58:09 *** sgworks has quit IRC 2015-08-11T15:58:18 thanks krohini, i also saw your email, and you can include some of that in your summary of your project 2015-08-11T15:58:51 I am implementing my mailbox approach for IPI.. there are some queries here..So will discuss that 2015-08-11T15:59:16 And I am looking to bring up the code refactor patch 2015-08-11T15:59:22 again 2015-08-11T15:59:50 ok 2015-08-11T15:59:53 If what I am doing is right should have SMP working in the next two days 2015-08-11T16:00:02 joel asked if you tried the rwbarrier change 2015-08-11T16:00:08 which i think i asked you before and you said you had 2015-08-11T16:00:13 but go ahead and respond to that email 2015-08-11T16:00:29 Yes, that didn't work.. Ya I will respond :) 2015-08-11T16:00:35 and let me know if the work-around i gave you for the bss init works for you 2015-08-11T16:00:43 rather, reply to that email too on devel ml 2015-08-11T16:01:03 Yup! will try that and report the result.. 2015-08-11T16:01:29 There are no other updates really.. Have to update the wiki as well 2015-08-11T16:02:03 *** gtament has quit IRC 2015-08-11T16:02:49 ok, thanks krohini i think you have plenty to keep chugging away at 2015-08-11T16:03:01 I am going to continue trying to get SMP working 2015-08-11T16:03:20 even if you don't get it by the GSoC deadline, I hope you can keep trying to work at it afterward 2015-08-11T16:03:25 hmm.. ok thanks gedare. Thats it for me 2015-08-11T16:03:28 same goes for everyone really. 2015-08-11T16:03:32 Ok thanks krohini. 2015-08-11T16:03:37 I think that is everyone. Did I miss anyone? 2015-08-11T16:03:40 Yes certainly! 2015-08-11T16:03:48 going once 2015-08-11T16:03:52 going twice 2015-08-11T16:04:14 Ok, we're all done. Thanks everyone and I'll watch for email questions as we get closer to deadlines 2015-08-11T16:04:29 ok :) 2015-08-11T16:06:51 *** sujayraaj has quit IRC 2015-08-11T16:07:01 Thanks gedare and everyone 2015-08-11T16:07:17 bye everyone 2015-08-11T16:07:21 *** ragunath has quit IRC 2015-08-11T16:07:57 *** hermann_19829 has quit IRC 2015-08-11T16:08:30 *** jrcatbagan has quit IRC 2015-08-11T16:08:45 *** krohini has quit IRC 2015-08-11T16:10:37 *** asuolen has quit IRC 2015-08-11T17:07:26 *** danieli has quit IRC 2015-08-11T17:21:11 *** danieli has joined #rtems 2015-08-11T17:58:42 *** krohini has joined #rtems 2015-08-11T18:05:08 *** ketul has quit IRC 2015-08-11T18:08:18 *** igutek has joined #rtems 2015-08-11T18:43:51 *** monstr has quit IRC 2015-08-11T19:22:06 *** krohini has quit IRC 2015-08-11T19:33:51 *** mumptai has joined #rtems 2015-08-11T19:57:54 *** dougk_ has joined #rtems 2015-08-11T21:10:00 *** gedare has quit IRC 2015-08-11T21:27:07 *** mumptai has quit IRC 2015-08-11T21:56:10 *** gedare has joined #rtems 2015-08-11T21:56:11 *** ChanServ sets mode: +o gedare 2015-08-11T22:04:31 *** igutek has quit IRC 2015-08-12T01:17:06 *** gedare has quit IRC 2015-08-12T01:22:16 *** dspace has quit IRC 2015-08-12T01:56:54 *** sujayraaj has joined #rtems 2015-08-12T01:57:43 *** sujayraaj has quit IRC 2015-08-12T02:00:58 *** sujayraaj has joined #rtems 2015-08-12T03:50:04 *** dr__house is now known as dr__house|away 2015-08-12T04:07:13 *** crazy3k has quit IRC 2015-08-12T04:27:59 *** monstr has joined #rtems 2015-08-12T05:51:53 *** danieli has quit IRC 2015-08-12T06:33:44 *** sujayraaj has joined #rtems 2015-08-12T06:36:49 *** danieli has joined #rtems 2015-08-12T07:20:06 *** sujayraaj has quit IRC 2015-08-12T11:04:19 *** hesham has joined #rtems 2015-08-12T13:22:03 *** danieli has quit IRC 2015-08-12T13:46:33 *** danieli has joined #rtems 2015-08-12T14:35:01 *** ketul has joined #rtems 2015-08-12T14:53:16 *** ketul has quit IRC 2015-08-12T16:01:57 *** dr__house|away is now known as dr__house 2015-08-12T16:30:39 *** sujayraaj has joined #rtems 2015-08-12T16:56:47 *** Davidbrcz has joined #rtems 2015-08-12T16:57:57 *** sujayraaj has quit IRC 2015-08-12T17:32:19 *** krohini has joined #rtems 2015-08-12T18:08:03 *** monstr has quit IRC 2015-08-12T20:03:03 *** hesham has quit IRC 2015-08-12T20:15:06 *** Davidbrcz has quit IRC 2015-08-12T20:19:12 *** gedare has joined #rtems 2015-08-12T20:19:12 *** ChanServ sets mode: +o gedare 2015-08-12T21:07:02 *** gedare has quit IRC 2015-08-12T21:20:49 *** Davidbrcz has joined #rtems 2015-08-12T21:25:27 *** Davidbrcz has quit IRC 2015-08-12T21:25:35 *** Davidbrcz has joined #rtems 2015-08-12T22:10:03 *** yangqiao has joined #rtems 2015-08-12T22:12:26 *** yangqiao has quit IRC 2015-08-12T23:23:14 *** danieli has quit IRC 2015-08-12T23:32:53 *** antgreen has joined #rtems 2015-08-13T00:15:07 *** gedare has joined #rtems 2015-08-13T00:15:07 *** ChanServ sets mode: +o gedare 2015-08-13T00:25:00 *** gedare has quit IRC 2015-08-13T01:03:21 *** rtems-irc has joined #rtems 2015-08-13T01:06:51 *** Davidbrcz has quit IRC 2015-08-13T03:58:49 *** dr__house is now known as dr__house|away 2015-08-13T04:26:21 *** monstr has joined #rtems 2015-08-13T05:45:43 *** danieli has joined #rtems 2015-08-13T07:12:40 *** jkridner has quit IRC 2015-08-13T07:13:35 *** jkridner has joined #rtems 2015-08-13T07:20:40 *** monstr has quit IRC 2015-08-13T07:24:05 *** monstr has joined #rtems 2015-08-13T07:30:58 *** monstr has quit IRC 2015-08-13T07:32:46 *** monstr has joined #rtems 2015-08-13T07:41:24 *** antgreen has quit IRC 2015-08-13T07:42:45 *** antgreen has joined #rtems 2015-08-13T07:46:34 *** antgreen` has joined #rtems 2015-08-13T07:46:47 *** antgreen has quit IRC 2015-08-13T07:54:52 *** hesham has joined #rtems 2015-08-13T07:55:07 *** monstr_ has joined #rtems 2015-08-13T07:56:23 *** monstr has quit IRC 2015-08-13T08:07:18 *** danieli has quit IRC 2015-08-13T08:16:34 *** monstr_ has quit IRC 2015-08-13T08:31:42 *** monstr_ has joined #rtems 2015-08-13T09:11:20 *** danieli has joined #rtems 2015-08-13T09:11:46 *** danieli has joined #rtems 2015-08-13T09:13:46 *** jkridner has quit IRC 2015-08-13T09:33:18 [trac] #2389/defect (v:4.10) created by yarkzone99 (Buying Cheap Weight Advana Cleanse) http://devel.rtems.org/ticket/2389 2015-08-13T09:52:05 *** ketul has joined #rtems 2015-08-13T09:55:28 *** ketul has quit IRC 2015-08-13T10:00:05 *** gedare has joined #rtems 2015-08-13T10:00:05 *** ChanServ sets mode: +o gedare 2015-08-13T10:34:04 *** ketul has joined #rtems 2015-08-13T11:14:17 *** monstr_ has quit IRC 2015-08-13T11:25:36 [trac] GSoC/2015/rpi_graphic created http://devel.rtems.org/wiki/GSoC/2015/rpi_graphic 2015-08-13T11:26:46 *** monstr_ has joined #rtems 2015-08-13T11:30:43 hi gedare 2015-08-13T11:33:13 hi ketul 2015-08-13T11:33:59 hi beng-nl :) . I was stuck at gpio_input and today get it resolved . Actually found bug in GPIO API. 2015-08-13T11:34:48 very nice 2015-08-13T11:35:00 gpio digital in? 2015-08-13T11:35:47 In uint8_t rtems_gpio_bsp_get_value(uint32_t bank, uint32_t pin); just found out for BBB uint8_t is not satisfied. As in BBB pin number varies from 0-32 . So uint32_t would be there. 2015-08-13T11:35:58 yes beng-nl in digital_in 2015-08-13T11:37:18 Tried all the debugging and at last found ahh problem in api :) . 2015-08-13T11:40:07 ketul: good job :) 2015-08-13T11:40:45 how do you test digital in? 2015-08-13T11:41:09 thanks :) . umm beng-nl . Let me gist it. 2015-08-13T11:45:43 [trac] GSoC/2015/rpi_graphic edited by yangqiao (empty comment) http://devel.rtems.org/wiki/GSoC/2015/rpi_graphic 2015-08-13T11:46:39 beng-nl: https://gist.github.com/ketul93/c67e6752bb36c6c91716 2015-08-13T11:47:22 [trac] GSoC/2015 edited by yangqiao (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-08-13T12:13:35 hello ketul and beng-nl 2015-08-13T12:13:54 get_value returns the pin number? 2015-08-13T12:20:51 hi gedare. No. It returns the value for particular pin value. According API rtems_gpio_get_value(uint32_t pin_number) will call rtems_gpio_bsp_get_value(bank, pin). 2015-08-13T12:20:59 Where rtems_gpio_bsp_get_value() would return the value in DATAIN(32 bit) register and &nding with the mask of pin no. i.e. BIT(pin). So it returns 32 bit value (as pin no. 0-32). 2015-08-13T12:21:06 And rtems_gpio_get_value() has "return ( rv > 0 ) ? 1 : rv;" where rv =rtems_gpio_bsp_get_value(bank, pin); 2015-08-13T12:24:28 gedare: In current api rtems_gpio_bsp_get_value() has uint8_t return type. But when I tested with BBB(for pin no.15) I always got value 0. Because it can only return 8 bits which is not well to get the correct value. So I tried to change it to uint32_t and got the output on hardware 2015-08-13T12:38:33 oh, i understand now. 2015-08-13T12:38:42 umm 2015-08-13T12:38:49 if pin numbers go from 0 to 32, don't you need 33 bits? 2015-08-13T12:39:14 what does it mean to & with the mask of pin number 0? 2015-08-13T12:39:17 ketul: ^ 2015-08-13T12:39:42 and what does it mean to & with the mask of pin 32? 2015-08-13T12:41:44 umm. gedare sorry committed a mistake about pin no. (0-31) . return (mmio_read(bank_addr + AM335X_GPIO_DATAIN)) & BIT(pin)); 2015-08-13T12:43:39 ok, and BIT(x) is just (1< will give a value for the particular pin's bit from the DATAIN register . That is what I would like to tell you by &ing with the mask of pin no. :) 2015-08-13T12:44:00 yaa 2015-08-13T12:44:17 ok. write this up in your bug report. 2015-08-13T12:44:35 ok creating a ticket 2015-08-13T12:58:18 *** jkridner has joined #rtems 2015-08-13T13:37:51 *** dr__house|away is now known as dr__house 2015-08-13T13:38:47 *** igutek has joined #rtems 2015-08-13T14:04:05 *** jkridner|work has joined #rtems 2015-08-13T14:07:46 *** jkridner has quit IRC 2015-08-13T14:20:50 [trac] #2390/defect (v:4.10) created by ketul (GPIO API bug) http://devel.rtems.org/ticket/2390 2015-08-13T14:22:34 [trac] gpio_api_bug.patch () attached to #2390 by ketul http://devel.rtems.org/ticket/2390 2015-08-13T14:23:19 *** monstr_ has quit IRC 2015-08-13T14:26:50 *** jkridner|work is now known as jkridner 2015-08-13T14:42:52 *** guillaum1 has joined #rtems 2015-08-13T14:44:47 hi all 2015-08-13T14:47:52 Do you know if some people could be interested in a small java 5 interpreter for RTEMS ? I'm trying to port my JVM to OS without official Java support. 2015-08-13T15:00:15 * guillaum1 will try and post the info on the forum :-) 2015-08-13T15:06:49 guillaum1: Yes you should on users@rtems.org 2015-08-13T15:09:58 ketul: Thanks. Will do 2015-08-13T15:11:27 guillaum1: I think there would be some interest. There have been some Java projects before, notably FijiVM comes to mind. 2015-08-13T15:11:43 [trac] 0001-Fixes-ticket-2390-and-also-updates-the-RPI-implement.patch () attached to #2390 by Asuol http://devel.rtems.org/ticket/2390 2015-08-13T15:11:57 guillaum1: there is also support for opengcj i believe, rudimentary that it may be 2015-08-13T15:12:30 [trac] #2390/defect (new) updated by Asuol (Although the BSP function rtems_gpio_bsp_get_value() may return the register value ...) http://devel.rtems.org/ticket/2390 2015-08-13T15:12:45 guillaum1: however, if your interpreter is (1) lightweight and (2) permissive-licensed, there might be interest. 2015-08-13T15:13:30 peerst: Are you around? 2015-08-13T15:14:20 gedare: yep 2015-08-13T15:15:29 peerst: PM if you please 2015-08-13T15:23:22 gedare: The license is "GPL with linking exception", so more permissive than GPL. As the JVM is provided with a Java 5 standard library, it can't be considered as lightweight but I have planned to provide a CLDC 1.1 package (far smaller). 2015-08-13T15:28:58 I read some documentations about FijiVM. This VM seems very good but its "academic licensing" could be a major issue for users. 2015-08-13T15:31:00 gedare: FYI, the VM already works on Minix and HelenOS (with GUI) 2015-08-13T15:33:28 *** ketul has quit IRC 2015-08-13T15:38:32 ok guillaum1 i think it could be of interest. 2015-08-13T15:38:43 are you from Charles U? 2015-08-13T15:39:14 * guillaum1 looks at the GCC toolchain 2015-08-13T15:44:23 guillaum1: ah i thought i recognized your nick :) 2015-08-13T15:44:34 :-D 2015-08-13T16:02:10 gedare: no, just a java/jvm developper hobbyist :-) 2015-08-13T16:08:45 *** guillaum1 has quit IRC 2015-08-13T16:11:31 *** antgreen` has quit IRC 2015-08-13T16:19:26 *** gedare has quit IRC 2015-08-13T17:23:59 *** Davidbrcz has joined #rtems 2015-08-13T18:09:39 *** danieli has quit IRC 2015-08-13T18:12:27 *** antgreen has joined #rtems 2015-08-13T18:19:29 *** hesham has quit IRC 2015-08-13T18:33:53 *** ketul has joined #rtems 2015-08-13T18:37:05 *** stryx` has quit IRC 2015-08-13T18:38:04 *** stryx`_ has joined #rtems 2015-08-13T19:31:28 *** gedare has joined #rtems 2015-08-13T19:31:28 *** ChanServ sets mode: +o gedare 2015-08-13T19:59:35 *** ketul has quit IRC 2015-08-13T20:10:09 *** Davidbrcz has quit IRC 2015-08-13T21:02:22 [trac] GSoC/2015 edited by gtament (empty comment) http://devel.rtems.org/wiki/GSoC/2015 2015-08-13T21:07:44 *** Davidbrcz has joined #rtems 2015-08-13T21:11:18 *** gedare has quit IRC 2015-08-13T21:19:01 added 'gtament' to the SPAM group 2015-08-13T21:21:46 *** Davidbrcz has quit IRC 2015-08-13T21:22:30 *** Davidbrcz has joined #rtems 2015-08-13T21:29:25 verm__: i registered for the gsoc summit standby list :) 2015-08-13T22:17:22 *** dougk_ has quit IRC 2015-08-13T22:32:48 *** Davidbrcz has quit IRC 2015-08-13T22:33:10 *** Davidbrcz has joined #rtems 2015-08-13T23:32:02 beng-nl: nice! hope to see you there! 2015-08-13T23:32:26 yeah would be great! 2015-08-13T23:32:38 i guess it's a long shot, 4 for rtems, but worth a try 2015-08-13T23:32:44 i always loved the gsoc summit in the past 2015-08-14T00:04:02 *** gedare has joined #rtems 2015-08-14T00:04:02 *** ChanServ sets mode: +o gedare 2015-08-14T00:12:15 *** gedare has quit IRC 2015-08-14T01:29:30 *** Davidbrcz has quit IRC 2015-08-14T04:35:00 *** monstr has joined #rtems 2015-08-14T05:43:14 *** dr__house is now known as dr__house|away 2015-08-14T08:27:19 *** ketul has joined #rtems 2015-08-14T09:25:40 hi beng-nl 2015-08-14T09:53:34 hi ketul 2015-08-14T09:53:55 beng-nl: :) . replied to mail 2015-08-14T09:56:19 yeah :) 2015-08-14T09:56:28 beng-nl: I think the problem is at rtems_gpio_bsp_set_resistor_mode(). Can you check and give your input 2015-08-14T09:56:42 the irq problem? 2015-08-14T09:59:44 yep :( 2015-08-14T10:00:07 Actually in TRM for resistor mode nothing is given in gpio section 2015-08-14T10:17:09 what is the relation between the resistor mode and the irq problem? 2015-08-14T10:23:25 beng-nl: I don't think any relation. But as in code first I am configuring it to PULL UP. So just wanted the first resistor mode to be successful . Then we can look forward about irq 2015-08-14T10:54:19 *** zoso has joined #rtems 2015-08-14T12:31:03 *** antgreen has quit IRC 2015-08-14T12:32:26 *** antgreen has joined #rtems 2015-08-14T12:54:18 *** pakkinen has joined #rtems 2015-08-14T12:55:13 hello 2015-08-14T12:56:35 does anybody know if there are any issues with using C++11 features when developing code for RTEMS? especially for powerpc platforms 2015-08-14T13:27:13 *** monstr has quit IRC 2015-08-14T13:40:07 *** verm__ has quit IRC 2015-08-14T13:47:51 *** verm__ has joined #rtems 2015-08-14T13:58:51 *** dougk_ has joined #rtems 2015-08-14T14:22:13 pakkinen: hi :) . If you don't get your answer on IRC. You can always ask on users@rtems.org 2015-08-14T14:23:37 alright thanks :) 2015-08-14T15:04:01 *** jkridner has quit IRC 2015-08-14T15:12:51 *** zoso has quit IRC 2015-08-14T15:24:50 [trac] #2390/defect (new) updated by beng (Personally I am not in favour of a magic constant such as 0xDEADBEEF. Who is to say ...) http://devel.rtems.org/ticket/2390 2015-08-14T15:26:19 [trac] Debugging/OpenOCD/Xilinx_Zynq edited by ricclaus (empty comment) http://devel.rtems.org/wiki/Debugging/OpenOCD/Xilinx_Zynq 2015-08-14T15:37:41 *** ketul has quit IRC 2015-08-14T15:52:04 *** dr__house|away is now known as dr__house 2015-08-14T16:09:02 *** igutek1 has joined #rtems 2015-08-14T16:11:21 *** igutek2 has joined #rtems 2015-08-14T16:11:43 *** igutek has quit IRC 2015-08-14T16:13:34 *** igutek1 has quit IRC 2015-08-14T16:15:33 *** igutek has joined #rtems 2015-08-14T16:17:22 *** igutek2 has quit IRC 2015-08-14T16:35:37 *** igutek1 has joined #rtems 2015-08-14T16:38:04 *** igutek2 has joined #rtems 2015-08-14T16:38:40 *** igutek has quit IRC 2015-08-14T16:40:00 *** igutek1 has quit IRC 2015-08-14T16:53:53 *** danieli has joined #rtems 2015-08-14T17:41:16 *** ketul has joined #rtems 2015-08-14T17:47:05 hello beng-nl 2015-08-14T17:51:06 hi ketul 2015-08-14T18:05:28 *** ketul has quit IRC 2015-08-14T18:08:19 *** ketul has joined #rtems 2015-08-14T18:10:59 *** Davidbrcz has joined #rtems 2015-08-14T18:33:16 *** pakkinen has quit IRC 2015-08-14T18:35:50 beng-nl: have you looked at irq code ? 2015-08-14T20:18:31 *** Davidbrcz has quit IRC 2015-08-14T20:26:54 *** ketul has quit IRC 2015-08-14T21:18:38 *** Davidbrcz has joined #rtems 2015-08-14T21:41:40 *** danieli has quit IRC 2015-08-15T00:58:13 *** Davidbrcz has quit IRC 2015-08-15T01:11:56 *** gedare has joined #rtems 2015-08-15T01:11:56 *** ChanServ sets mode: +o gedare 2015-08-15T01:17:38 *** gedare has quit IRC 2015-08-15T01:23:54 *** antgreen has quit IRC 2015-08-15T01:24:07 *** antgreen` has joined #rtems 2015-08-15T01:35:14 *** gedare has joined #rtems 2015-08-15T01:35:14 *** ChanServ sets mode: +o gedare 2015-08-15T01:43:26 *** gedare has quit IRC 2015-08-15T02:53:54 *** Davidbrcz has joined #rtems 2015-08-15T03:47:48 *** ketul has joined #rtems 2015-08-15T06:21:46 *** Davidbrcz has quit IRC 2015-08-15T06:30:53 *** dr__house is now known as dr__house|away 2015-08-15T09:00:55 *** ketul has quit IRC 2015-08-15T09:07:54 *** ketul has joined #rtems 2015-08-15T09:27:49 hi beng-nl 2015-08-15T10:48:36 *** danieli has joined #rtems 2015-08-15T11:36:28 ketul: no not yet 2015-08-15T11:38:34 beng-nl: sorry I didn't get you :) 2015-08-15T11:38:53 ketul: i mean no i haven't look at the code yet, i'll try to do it today 2015-08-15T11:39:36 ahh. ok please I tried a lot for debugging but I think some silly mistake is still there 2015-08-15T11:44:18 *** antgreen` has quit IRC 2015-08-15T11:46:01 ok 2015-08-15T11:46:30 so what is the problem? you're expecting to get an irq but you're not getting one? 2015-08-15T11:46:42 yes. 2015-08-15T11:47:20 also tried to debug by printk.. 2015-08-15T11:49:49 beng-nl: at bsp level looks fine. I mean register addressing and all 2015-08-15T11:56:23 ketul: very nice test app 2015-08-15T11:56:34 :) 2015-08-15T12:01:20 beng-nl: also when I make threaded_handling = false. While applying interrupt on the defined pin. BBB gets reboot. Same problem some days ago raised by Steve B. But with threaded_handling = true not able to detect the interrupt event. 2015-08-15T12:02:56 where is this threaded_handling? why does it change the behaviour? 2015-08-15T12:04:20 rtems_gpio_enable_interrupt. You can look at that function in libbbsp/shared/gpio.c . Reason is still need to be searched. 2015-08-15T12:05:23 it may be a problem with the irq server system and the beagle bsp i think 2015-08-15T12:05:33 we have a similar problem with ragunath 2015-08-15T12:06:13 the beagle reset may indicate infinite recursion when an irq happens 2015-08-15T12:06:13 ohk seems like. So ragunath found any hack on it? 2015-08-15T12:06:22 probably caused by this: 2015-08-15T12:06:23 uint32_t psr = _ARMV4_Status_irq_enable(); 2015-08-15T12:06:28 in bsp_interrupt_dispatch 2015-08-15T12:06:45 it wants to allow interrupts happening while an interrupt is being handled, but it doesn't work in the irq server case 2015-08-15T12:07:15 so first, try moving _ARMV4_Status_irq_enable to below bsp_interrupt_handler_dispatch, so the irq stays masked 2015-08-15T12:07:29 or don't enable it at all and see if you get just 1 irq 2015-08-15T12:07:54 the reset is "good" because it means the trickiest part is done: configuring the hardware to trigger an irq :-) 2015-08-15T12:08:25 this irq server business tripping up the bsp is my fault i think :) 2015-08-15T12:08:52 something i still need to figure out 2015-08-15T12:09:20 ketul: does the above help you to continue experimenting? 2015-08-15T12:09:24 yep. What would be if we make threaded_handling = true 2015-08-15T12:11:52 let's try that once the = false case works, it looks simpler look int at gpio.c 2015-08-15T12:11:54 ok? 2015-08-15T12:12:39 ok. 2015-08-15T12:19:12 so beng-nl with threaded_handling = false on irq it calls at least - rtems_gpio_bsp_interrupt_line() (got printk() in that function ) and after that it reboots. 2015-08-15T12:19:58 So now trying to move _ARMV4_Status_irq_enable . Will tell you what happens :) 2015-08-15T12:22:08 and beng-nl have you looked at the part after #if IS_DM3730 . Is it good to keep it in that way ? 2015-08-15T12:31:44 *** ita has joined #rtems 2015-08-15T12:33:38 *** ita_ has quit IRC 2015-08-15T12:34:11 beng-nl: Tried to re order _ARMV4_Status_irq_enable. But it still reboots. 2015-08-15T12:45:31 *** danieli has quit IRC 2015-08-15T13:35:47 *** danieli has joined #rtems 2015-08-15T14:13:55 beng-nl: ^ 2015-08-15T14:13:58 *** ketul has quit IRC 2015-08-15T14:18:41 *** ketul has joined #rtems 2015-08-15T14:19:38 beng-nl: ^ 2015-08-15T14:59:29 *** ketul_ has joined #rtems 2015-08-15T14:59:41 *** ketul has quit IRC 2015-08-15T14:59:51 *** ketul_ is now known as ketul 2015-08-15T15:18:32 *** ketul has quit IRC 2015-08-15T15:55:35 *** Davidbrcz has joined #rtems 2015-08-15T16:00:35 *** Davidbrcz has quit IRC 2015-08-15T16:13:20 *** mumptai has joined #rtems 2015-08-15T16:27:41 *** jrcatbagan has joined #rtems 2015-08-15T16:51:32 *** jrcatbagan has quit IRC 2015-08-15T17:23:29 *** ketul has joined #rtems 2015-08-15T17:31:15 *** danieli has quit IRC 2015-08-15T17:31:39 *** mumptai has quit IRC 2015-08-15T19:26:52 ketul: bummer :( 2015-08-15T19:29:50 *** krohini has joined #rtems 2015-08-15T19:29:52 beng-nl: why ? :( 2015-08-15T19:29:57 *** krohini has quit IRC 2015-08-15T19:36:02 bummer that it didn't help you much 2015-08-15T19:37:24 np :) . beng-nl can you suggest more about to stop reboot and let bbb go to the isr function 2015-08-15T19:40:19 *** dr__house|away is now known as dr__house 2015-08-15T20:05:38 *** gedare has joined #rtems 2015-08-15T20:05:38 *** ChanServ sets mode: +o gedare 2015-08-15T20:15:26 *** gedare has quit IRC 2015-08-15T21:05:41 *** ketul has quit IRC 2015-08-15T22:35:32 *** mumptai has joined #rtems 2015-08-15T22:54:43 *** Davidbrcz has joined #rtems 2015-08-15T23:07:35 *** Davidbrcz has quit IRC 2015-08-15T23:24:24 *** Davidbrcz has joined #rtems 2015-08-16T00:04:41 *** Davidbrcz has quit IRC 2015-08-16T03:05:09 *** Davidbrcz has joined #rtems 2015-08-16T04:28:06 *** mumptai has quit IRC 2015-08-16T04:41:05 *** mumptai has joined #rtems 2015-08-16T05:04:54 *** kiwichris has joined #rtems 2015-08-16T05:07:01 *** crazy3k has joined #rtems 2015-08-16T06:30:46 *** dr__house is now known as dr__house|away 2015-08-16T07:05:59 *** geheimni2` has joined #rtems 2015-08-16T07:06:21 *** sirfrank1in has joined #rtems 2015-08-16T07:06:51 *** geheimni1` has quit IRC 2015-08-16T07:06:53 *** sirfranklin has quit IRC 2015-08-16T07:38:34 *** Davidbrcz has quit IRC 2015-08-16T10:06:10 *** crazy3k has quit IRC 2015-08-16T14:18:22 *** ketul has joined #rtems 2015-08-16T14:23:33 *** ketul has quit IRC 2015-08-16T17:34:28 *** Davidbrcz has joined #rtems 2015-08-16T18:13:27 *** Davidbrcz has quit IRC 2015-08-16T18:43:34 *** dr__house|away is now known as dr__house 2015-08-16T18:47:19 *** mumptai has quit IRC 2015-08-16T19:00:21 *** mumptai has joined #rtems 2015-08-16T19:22:12 *** ketul has joined #rtems 2015-08-16T20:07:36 *** ketul has quit IRC 2015-08-16T20:57:51 *** mumptai has quit IRC 2015-08-16T21:09:37 *** kiwichris has quit IRC 2015-08-16T22:15:19 *** gedare has joined #rtems 2015-08-16T22:15:19 *** ChanServ sets mode: +o gedare 2015-08-16T22:24:30 *** gedare has quit IRC 2015-08-16T23:29:44 *** kiwichris has joined #rtems