2015-07-13T01:16:24 *** Davidbrcz has quit IRC 2015-07-13T01:20:49 *** sujayraaj has joined #rtems 2015-07-13T01:24:55 kiwichris : hi 2015-07-13T01:25:06 sujayraaj, hi 2015-07-13T01:29:06 Finally feeling well enough to work.Even though I lost around 6 days, I think I can make it up. :) 2015-07-13T01:29:50 by the way kiwichris ,did you get time to compile and run monkey from my tree on your system? 2015-07-13T01:30:04 No I have not. 2015-07-13T01:30:31 I will try this week. 2015-07-13T01:34:09 Basically since we were out of options I had tried using sockets in place of file descriptors. the mypipe function ( pipe is just in the name ) the code is currently using , uses sockets and binds them to a port on the system. 2015-07-13T01:34:32 https://github.com/sujayraaj/monkey/blob/master/mk_core/mk_event_kqueue.c#L207 2015-07-13T01:54:24 *** sujayraaj has quit IRC 2015-07-13T02:02:29 *** Davidbrcz has joined #rtems 2015-07-13T02:05:42 *** sujayraaj has joined #rtems 2015-07-13T03:27:33 *** sujayraaj has quit IRC 2015-07-13T03:47:57 *** Davidbrcz has quit IRC 2015-07-13T04:53:11 [trac] GSoC/2015/NestedMutex edited by gadia (empty comment) http://devel.rtems.org/wiki/GSoC/2015/NestedMutex 2015-07-13T04:53:51 [trac] GSoC/2015/NestedMutex edited by gadia (empty comment) http://devel.rtems.org/wiki/GSoC/2015/NestedMutex 2015-07-13T05:23:37 *** sujayraaj has joined #rtems 2015-07-13T05:41:30 *** Davidbrcz has joined #rtems 2015-07-13T05:43:42 *** sujayraaj has quit IRC 2015-07-13T05:48:03 *** mumptai has joined #rtems 2015-07-13T06:25:21 *** sebhub has joined #rtems 2015-07-13T06:25:34 good morning 2015-07-13T07:32:32 *** kiwichris has quit IRC 2015-07-13T07:48:53 *** Davidbrcz has quit IRC 2015-07-13T08:08:08 [git] 9341773 by Sebastian Huber: [score: Typo] http://devel.rtems.org/changeset/9341773/rtems 2015-07-13T08:08:09 [git] f5d4570 by Sebastian Huber: [score: Simplify _Thread_Lock_set() Exploit the fact that the current thread lock must be the default thread lock and interrupts are disabled if we call _Thread_Lock_set().] http://devel.rtems.org/changeset/f5d4570/rtems 2015-07-13T08:46:59 *** cdesai has quit IRC 2015-07-13T09:52:02 *** hesham has joined #rtems 2015-07-13T10:18:47 *** geheimnis` has joined #rtems 2015-07-13T10:56:08 *** mkhoory-2 has joined #rtems 2015-07-13T10:59:51 *** mkhoory has quit IRC 2015-07-13T11:11:21 *** mkhoory-2 has quit IRC 2015-07-13T11:25:08 *** mkhoory has joined #rtems 2015-07-13T12:19:22 *** mkhoory has quit IRC 2015-07-13T12:37:06 *** cdesai__ has joined #rtems 2015-07-13T12:38:08 *** cdesai has joined #rtems 2015-07-13T13:30:14 *** hesham has quit IRC 2015-07-13T14:20:53 *** sujayraaj has joined #rtems 2015-07-13T14:23:23 *** gedare has joined #rtems 2015-07-13T14:23:23 *** ChanServ sets mode: +o gedare 2015-07-13T14:25:12 *** sujayraaj has quit IRC 2015-07-13T14:27:00 *** ketul has joined #rtems 2015-07-13T14:55:12 *** beng-nl has quit IRC 2015-07-13T15:24:52 [trac] GSoC/2015/NestedMutex edited by gadia (empty comment) http://devel.rtems.org/wiki/GSoC/2015/NestedMutex 2015-07-13T15:26:13 [trac] GSoC/2015/NestedMutex edited by gadia (empty comment) http://devel.rtems.org/wiki/GSoC/2015/NestedMutex 2015-07-13T15:34:11 *** beng-nl has joined #rtems 2015-07-13T15:36:50 *** sirfranklin has quit IRC 2015-07-13T15:36:51 *** Schoumi has quit IRC 2015-07-13T15:36:57 *** fds has quit IRC 2015-07-13T15:36:58 *** dr__house|away has quit IRC 2015-07-13T15:37:02 *** sirfranklin has joined #rtems 2015-07-13T15:37:03 *** dr__house|away has joined #rtems 2015-07-13T15:37:06 *** Schoumi has joined #rtems 2015-07-13T15:37:08 *** fds has joined #rtems 2015-07-13T15:42:01 hi all 2015-07-13T15:45:58 *** beng-nl has quit IRC 2015-07-13T16:07:15 *** Davidbrcz has joined #rtems 2015-07-13T16:18:01 *** Davidbrcz has quit IRC 2015-07-13T16:53:10 *** antgreen has joined #rtems 2015-07-13T16:53:45 *** hermann_19829 has joined #rtems 2015-07-13T17:13:33 *** Davidbrcz has joined #rtems 2015-07-13T17:32:52 *** antgreen has quit IRC 2015-07-13T17:45:40 hi gedare. How was your vacation ? 2015-07-13T17:51:48 *** ketul has quit IRC 2015-07-13T17:56:48 *** ketul has joined #rtems 2015-07-13T18:31:15 *** sujayraaj has joined #rtems 2015-07-13T18:31:25 *** gedare has quit IRC 2015-07-13T18:33:49 *** jkridner has joined #rtems 2015-07-13T18:56:50 *** ketul has quit IRC 2015-07-13T19:04:42 *** ketul has joined #rtems 2015-07-13T19:08:52 *** hesham has joined #rtems 2015-07-13T19:40:43 *** dr__house|away is now known as dr__house 2015-07-13T19:40:52 *** dr__house has joined #rtems 2015-07-13T20:05:53 *** Davidbrcz has quit IRC 2015-07-13T20:24:14 *** edsiper has quit IRC 2015-07-13T20:24:21 *** edsiper has joined #rtems 2015-07-13T20:30:20 *** Davidbrcz has joined #rtems 2015-07-13T20:32:22 *** Davidbrcz has quit IRC 2015-07-13T20:32:30 *** Davidbrcz has joined #rtems 2015-07-13T20:38:18 *** sujayraaj has quit IRC 2015-07-13T21:11:36 *** hermann_19829 has quit IRC 2015-07-13T22:04:46 *** kiwichris has joined #rtems 2015-07-13T22:09:28 *** kiwichris has quit IRC 2015-07-13T22:12:31 *** ketul has quit IRC 2015-07-13T22:14:57 *** kiwichris has joined #rtems 2015-07-13T22:16:16 [git] 083e6d6 by Chris Johns: [libnetworking: Send the hostname if set in the network configuration. This allows a suitably configured DHCP server with DDNS to enter the name into the DNS table making it addressiable via it's host name.] http://devel.rtems.org/changeset/083e6d6/rtems 2015-07-13T22:18:13 [trac] #2341/enhancement (closed) updated by chrisj (empty comment) http://devel.rtems.org/ticket/2341 2015-07-13T23:13:40 *** beng-nl has joined #rtems 2015-07-13T23:13:57 hi all 2015-07-13T23:32:43 [git] b919871 by Jarielle Catbagan: [BBB: sd_setup.sh: Change /bin/bash -> /bin/sh to solely adhere to POSIX shell constructs] http://devel.rtems.org/changeset/b919871/umon 2015-07-13T23:47:22 *** mumptai has quit IRC 2015-07-13T23:52:46 hi beng-nl 2015-07-14T00:00:32 *** mumptai has joined #rtems 2015-07-14T00:32:13 [git] 5bed147 by Ed Sutter: [remove files leftover from csb740 port] http://devel.rtems.org/changeset/5bed147/umon 2015-07-14T00:36:02 [git] 8107331 by Ed Sutter: [replace bashrc with envsetup to eliminate implication that it is bash specific] http://devel.rtems.org/changeset/8107331/umon 2015-07-14T00:36:33 *** jkridner|work has joined #rtems 2015-07-14T00:37:31 *** jkridner has quit IRC 2015-07-14T01:45:36 [trac] #2369/defect (new) updated by nick.withers (Sorry Sebastian, I'm still trying to work through this... I seem to have problems ...) http://devel.rtems.org/ticket/2369 2015-07-14T02:00:56 [trac] #2369/defect (new) updated by nick.withers (Replying to [comment:6 nick.withers]: > I seem to have problems with my toolset and ...) http://devel.rtems.org/ticket/2369 2015-07-14T02:14:25 [git] b8ffb40 by Ed Sutter: [BBB: Changes to support runtime blinking LED.] http://devel.rtems.org/changeset/b8ffb40/umon 2015-07-14T03:10:46 *** Davidbrcz has quit IRC 2015-07-14T04:22:41 *** dr__house is now known as dr__house|away 2015-07-14T04:50:23 [trac] #2369/defect (new) updated by nick.withers (If I build with "CPU_CFLAGS = -mcpu=8540 -meabi -msdata=sysv -msoft-float -mno-spe ...) http://devel.rtems.org/ticket/2369 2015-07-14T04:51:36 *** Davidbrcz has joined #rtems 2015-07-14T06:28:03 [trac] #2369/defect (new) updated by sebastian.huber (We can still use your original patch and use the SPR 268. I don't care that much ...) http://devel.rtems.org/ticket/2369 2015-07-14T06:41:58 *** sujayraaj has joined #rtems 2015-07-14T06:57:12 *** kiwichris has quit IRC 2015-07-14T06:58:51 *** sujayraaj has quit IRC 2015-07-14T07:10:39 *** ketul has joined #rtems 2015-07-14T07:28:47 *** kiwichris has joined #rtems 2015-07-14T07:30:31 *** hesham has quit IRC 2015-07-14T07:48:50 *** sujayraaj has joined #rtems 2015-07-14T07:49:59 kiwichris : hi 2015-07-14T07:50:18 sujayraaj, hello. 2015-07-14T07:51:58 kiwichris , That problem is proving too troublesome for me and I am not able to make way ahead. Please tell me how should I proceed. 2015-07-14T07:52:30 sujayraaj, is this the pipe work around ? 2015-07-14T07:52:47 kiwichris : Yes. 2015-07-14T07:53:53 sujayraaj, sorry but I have been on a mad deadline this past week. It is easing up now. I need to take a detailed look at monkey and figure out the real purpose of the pipe. 2015-07-14T07:55:07 I suspect the client thread sync method used may need to be changed with an abstraction of the process made. 2015-07-14T07:55:35 "process" I mean the way Monkey does it. 2015-07-14T07:55:53 kqueue and pipes are fine but not a clean match for operating systems. 2015-07-14T07:56:05 for all operating systems 2015-07-14T07:56:45 Monkey's real value is being a web server and not a way to manage events on operating systems. 2015-07-14T07:58:33 Yeah. Pretty much. By the way, Note: when on linux (while using epoll) monkey uses 'eventfd' , for the channel create function. 2015-07-14T07:59:15 Ok; I need to head off for a bit. 2015-07-14T07:59:25 *** kiwichris has quit IRC 2015-07-14T08:02:54 *** sujayraaj has quit IRC 2015-07-14T08:12:35 *** Davidbrcz has quit IRC 2015-07-14T08:49:51 *** kiwichris has joined #rtems 2015-07-14T08:58:33 *** kiwichris has quit IRC 2015-07-14T09:12:13 hi beng-nl :) 2015-07-14T09:21:37 *** kiwichris has joined #rtems 2015-07-14T09:32:13 *** kiwichris has quit IRC 2015-07-14T09:33:51 *** kiwichris has joined #rtems 2015-07-14T09:57:48 *** kiwichris has quit IRC 2015-07-14T09:58:16 *** kiwichris has joined #rtems 2015-07-14T10:31:25 *** krohini has joined #rtems 2015-07-14T11:00:47 *** ketul has quit IRC 2015-07-14T12:12:15 *** timothyh has quit IRC 2015-07-14T12:27:59 *** ketul has joined #rtems 2015-07-14T12:49:26 *** ketul has quit IRC 2015-07-14T13:16:59 *** asuolen has joined #rtems 2015-07-14T13:17:48 *** shelf has joined #rtems 2015-07-14T13:33:40 *** jkridner|work has quit IRC 2015-07-14T14:05:05 *** jkridner has joined #rtems 2015-07-14T14:23:55 *** gedare has joined #rtems 2015-07-14T14:23:55 *** ChanServ sets mode: +o gedare 2015-07-14T14:34:08 *** ketul has joined #rtems 2015-07-14T14:40:08 *** anandkp92 has joined #rtems 2015-07-14T14:43:34 *** DrJoel has joined #rtems 2015-07-14T14:43:34 *** DrJoel has joined #rtems 2015-07-14T14:43:34 *** ChanServ sets mode: +o DrJoel 2015-07-14T14:43:39 morning DrJoel 2015-07-14T14:44:35 *** jrcatbagan has joined #rtems 2015-07-14T14:47:40 *** EdSut has joined #rtems 2015-07-14T14:48:22 *** gtament has joined #rtems 2015-07-14T14:48:33 Hi, everybody!) 2015-07-14T14:49:01 *** ragunath has joined #rtems 2015-07-14T14:50:25 hi gtament :) and gedare and DrJoel and kiwichris 2015-07-14T14:50:39 *** anandkp92 has quit IRC 2015-07-14T14:51:01 *** ragunath has quit IRC 2015-07-14T14:51:42 Hey folks! 2015-07-14T14:51:46 EdSut! 2015-07-14T14:52:37 hello everyone :) 2015-07-14T14:52:39 *** ragunath has joined #rtems 2015-07-14T14:52:58 *** javamonn has joined #rtems 2015-07-14T14:53:11 hi ketul :) 2015-07-14T14:53:21 hi ben 2015-07-14T14:53:24 hi ragunath 2015-07-14T14:53:25 hello 2015-07-14T14:53:32 hi gedare 2015-07-14T14:53:33 Hi EdSut! 2015-07-14T14:53:36 Hi beng-nl! 2015-07-14T14:53:37 hi jrcatbagan! 2015-07-14T14:53:40 Hi gedare! 2015-07-14T14:53:57 :-) 2015-07-14T14:54:00 *** yangqiao has joined #rtems 2015-07-14T14:55:03 Good to "see" everyone. :-) 2015-07-14T14:55:15 :-) 2015-07-14T14:56:26 we'll start a bit early today i think, since EdSut and jrcatbagan are both here, why don't we start with jrcatbagan. 2015-07-14T14:56:40 jrcatbagan: i saw your email on progress, great work on the ddr3 2015-07-14T14:56:50 hi gedare! 2015-07-14T14:56:50 yes 2015-07-14T14:57:00 so I would say two milestones 2015-07-14T14:57:05 uMon is now booting from SD 2015-07-14T14:57:11 and DDR3 is finally initialized 2015-07-14T14:57:18 wow, great 2015-07-14T14:57:30 currently looking into running some memory diagnostics on DDR3 2015-07-14T14:57:38 with "mt" command 2015-07-14T14:57:53 Also there is a script that sets up an SD 2015-07-14T14:57:59 to boot Umon either 2015-07-14T14:58:04 ok, do you also need to move ddr3 init earlier? 2015-07-14T14:58:07 in raw mode or from a FAT filesystem 2015-07-14T14:58:12 yes 2015-07-14T14:58:15 Good stuff!!! 2015-07-14T14:58:23 Definitely! 2015-07-14T14:58:35 ok, what will come after that, since that's mostly clean-up work? 2015-07-14T14:58:48 getting the MMC interface initialized 2015-07-14T14:58:58 that way uMon can boot an application image 2015-07-14T14:59:07 from SD and ultimately from the onboard eMMC 2015-07-14T14:59:18 and then Ethernet after MMC 2015-07-14T14:59:53 ok super 2015-07-14T15:00:02 keep up the good work 2015-07-14T15:00:05 Thanks! 2015-07-14T15:00:11 Great! The MMC needs to be split between device-specific code and the 'mmc' command front end we can talk more about that 2015-07-14T15:00:16 gtament: would you like to go now? 2015-07-14T15:00:22 Definitely! 2015-07-14T15:00:30 Looking forward 2015-07-14T15:00:31 *** hermann_19829 has joined #rtems 2015-07-14T15:00:46 *** sujayraaj has joined #rtems 2015-07-14T15:01:11 *** krohini has joined #rtems 2015-07-14T15:01:24 ok maybe he isn't here right now. how about ketul? 2015-07-14T15:01:27 *** Davidbrcz has joined #rtems 2015-07-14T15:01:27 *** irnhorse4 has joined #rtems 2015-07-14T15:01:33 hi gedare 2015-07-14T15:01:40 hi ketul, beng-nl 2015-07-14T15:01:49 yo 2015-07-14T15:02:06 Now I have started working on i2c part. reading part almost completed (TRM) 2015-07-14T15:02:40 and gedare me and beng-nl decided not to work in TSC as of now 2015-07-14T15:02:41 ok, how is the code looking for the gpio and adc? 2015-07-14T15:02:46 ok 2015-07-14T15:03:03 yep, too big for the remaining gsoc time 2015-07-14T15:03:06 ketul: res1 ==> reserved1 .. shouldn't be abbreviated 2015-07-14T15:03:07 have you familiarized yourself with the i2c framework? 2015-07-14T15:03:29 DrJoel: ok 2015-07-14T15:03:36 *** anandkp92 has joined #rtems 2015-07-14T15:03:37 *** sgworks has joined #rtems 2015-07-14T15:03:45 gedare: my intent is to do final build & functionality test of gpio and merge it (modulo community feedback), and harmonize if/when rpi implementation merges. good idea? 2015-07-14T15:04:18 *** ragunath has quit IRC 2015-07-14T15:04:19 gedare: currently working on it as i2c.h is already there. 2015-07-14T15:04:39 beng-nl: yes that works for me. 2015-07-14T15:04:45 hello everybody 2015-07-14T15:04:47 gedare: ok great. 2015-07-14T15:04:50 gedare: i will do that soonish. 2015-07-14T15:04:56 :) 2015-07-14T15:05:09 ketul: i get that bb has i2c functionality, but we need to re-write it to fit with cpukit/dev/i2c framework. 2015-07-14T15:05:30 you may like to study other implementations for that i2c support. 2015-07-14T15:05:50 *** ragunath has joined #rtems 2015-07-14T15:06:07 gedare: bbb currently doesn't have the i2c functions' definitions. only api is ready 2015-07-14T15:06:17 gedare: it's nice to see how clean gpio will be compared to the current test cases using led.h in examples-v2 btw :) 2015-07-14T15:06:30 beng-nl: agreed. 2015-07-14T15:06:45 ketul: OK, but you understand that you need to make the implementation suitable for use with the code in cpukit/dev/i2c? 2015-07-14T15:07:20 not with the code in c/src/libchip/i2c, or as a standalone i2c 2015-07-14T15:07:34 ok i see 2015-07-14T15:08:16 ok, thanks ketul. anything else? 2015-07-14T15:08:19 ragunath: you can go next 2015-07-14T15:08:22 *** sgworks has quit IRC 2015-07-14T15:08:39 gedare: thanks. will discuss with beng-nl after meeting if required :) 2015-07-14T15:08:39 Hello everyone 2015-07-14T15:08:54 Networking in BBB is working with cache disabled. I am trying to make it work with cache enabled. 2015-07-14T15:09:00 The problem is we are able to receive the packets, but the ALE entries are corrupted. Because of which we are sending junk packets. 2015-07-14T15:09:03 Hi everyone, gedare. Add me to the queue too, please :) 2015-07-14T15:09:08 Packets are not getting corrupted when received. During the transmission too the packet is in good condition. It is messed up because of the ALE. 2015-07-14T15:09:14 I am trying to find why ALE entries are corrupted. As per the datasheet, the switch will automatically update the ALE on receiving a packet. I am checking that. 2015-07-14T15:09:19 I also checked the suggestions given by sebastian. 2015-07-14T15:09:27 I also tried flushing/invalidating cache but it did not solve the issue. I have also tried allocate non cache memory but it did not work. May be I can send a patch to sebastian on the changes I made for this and see if it the right way to do. 2015-07-14T15:09:38 Remind me what ALE means 2015-07-14T15:09:57 sorry .. Address Lookup Engine 2015-07-14T15:10:08 ok. do you have another task to work on while you debug this cache issue? 2015-07-14T15:10:29 not now.. i was about to ask the same 2015-07-14T15:10:40 you can write an email about the various things you have tried, and maybe others will have suggestions. 2015-07-14T15:10:41 can i start lwIP porting paralelly 2015-07-14T15:10:44 one idea would be to start on lwIP 2015-07-14T15:10:50 and then it should have the same problem 2015-07-14T15:10:52 *parallely 2015-07-14T15:10:54 and you can debug both together 2015-07-14T15:10:59 it may help. 2015-07-14T15:11:07 yeah I can do that 2015-07-14T15:11:15 ok 2015-07-14T15:11:24 this will be good for making progress too 2015-07-14T15:11:30 sounds good, anything else now? 2015-07-14T15:11:39 As I recall, there was a driver for the BBB so you should be able to bring it up quickly and have something to compare against. 2015-07-14T15:11:51 sujayraaj: you can get ready to go next, and then it will be yangqiao. 2015-07-14T15:12:02 ok 2015-07-14T15:12:05 DrJoel: they also have working with cache disable 2015-07-14T15:12:07 yes tallertech has one 2015-07-14T15:12:07 sure. 2015-07-14T15:12:30 ``it uses some vendor files though so care must be taken 2015-07-14T15:12:35 if i remember right. 2015-07-14T15:12:45 for them too cache has to be disabled, as per the discussion i had they did debug much on this 2015-07-14T15:12:56 * did not debug 2015-07-14T15:13:06 thanks ragunath so work on writing up what you have tried out, and get started on lwIP? 2015-07-14T15:13:09 +1 on licensing.. 2015-07-14T15:13:15 :( on them not solving it either. :( 2015-07-14T15:13:26 :( 2015-07-14T15:13:27 meh, it was a "works-for-me" situation i'm sure. 2015-07-14T15:13:41 ok gedare i will send a write up and start on lwIP 2015-07-14T15:13:43 ok thanks ragunath. sujayraaj you're up, hi edsiper DrJoel 2015-07-14T15:13:56 hello.. have been here .. but lrked a while 2015-07-14T15:14:15 yup just pinging cognizant mentors. 2015-07-14T15:14:31 hi DrJoel 2015-07-14T15:14:54 I wonder if kiwichris is up too. Anyways. 2015-07-14T15:14:58 Last work was completely unproductive as I was down with a high fever. ( Its rainy season here -_- ). The work is presently stuck on an issue which is taking its time. Pipe isn't supported by kqueue. And we need pipe for some communication between threads. 2015-07-14T15:15:43 get rest and fluids. 2015-07-14T15:16:00 +1 and get better 2015-07-14T15:16:08 so this is a problem on RTEMS side--libbsd? 2015-07-14T15:16:57 *** sujayraaj_ has joined #rtems 2015-07-14T15:17:08 hi sujayraaj_. 2015-07-14T15:17:27 sujayraaj_, Last work was completely unproductive as I was down with a high fever. ( Its rainy season here -_- ). The work is presently stuck on an issue which is taking its time. Pipe isn't supported by kqueue. And we need pipe for some communication between threads. 2015-07-14T15:17:27 get rest and fluids. 2015-07-14T15:17:27 +1 and get better 2015-07-14T15:17:27 so this is a problem on RTEMS side--libbsd? 2015-07-14T15:17:42 thanks gedare. 2015-07-14T15:17:56 Yes. the problem is on rtems side 2015-07-14T15:18:05 the file descriptors generated by pipe 2015-07-14T15:18:17 don't get processed by kevent 2015-07-14T15:18:19 ok. this would be worth sending to devel ml 2015-07-14T15:18:22 and we get an EINVAL 2015-07-14T15:18:37 *** sgworks has joined #rtems 2015-07-14T15:18:39 just to document the issue at least. and even opening a ticket in trac would be appropriate 2015-07-14T15:18:52 creating the ticket would be sufficient actually. 2015-07-14T15:19:06 ok. I will create a ticket today then. 2015-07-14T15:19:21 *** sujayraaj has quit IRC 2015-07-14T15:19:29 Hi, I've ported the type1 library microwindows and nxlib into rsb, the libraries are able to be built but I need to build more tests to ensure. On the same time I'm trying to figure out what code corrupt the atag area, trying to make the atag works. The graphic console works well on my rpis, I havn't received any more feedback yet. That's all. 2015-07-14T15:19:38 great. do you know how to proceed sujayraaj_, or are you stuck? 2015-07-14T15:19:44 Sebastian suggested porting the BSD pipe code. 2015-07-14T15:19:45 So, presently, we were discussing minimal changes we can make on monkey side 2015-07-14T15:20:24 ok yangqiao i think ppisa is mainly testing your work, but you should submit patches for RSB soon. 2015-07-14T15:20:44 excuse-me, I saw sujayraaj_ is disconnected. 2015-07-14T15:20:46 sujayraaj_: i think hammering out the pipe issue is good. 2015-07-14T15:20:46 Well, I am stuck with that DrJoel, someone would have to show me the ropes on getting things ported from freebsd. . a bit documentation to read would be good. ( which I have found none ) 2015-07-14T15:21:01 it's ok yangqiao, you can keep going too we'll manage both overlap a bit 2015-07-14T15:21:20 sujayraaj_: that will be a good thing to ask (complain) about on devel ml 2015-07-14T15:21:34 sujayraaj_: there is libbsd.txt which give some details 2015-07-14T15:21:57 oh! Thanks ragunath. I will have a look at it then. 2015-07-14T15:22:31 ok, I didn't send the patches because ppisa said he has problem with it. I've got to wait until he get it works. Since you said the it's ok to use dectection of display size as a temporary fix. 2015-07-14T15:23:19 Well, gedare, that will be it from me for today. Let me try the ml and some docs and see if I get through. 2015-07-14T15:23:42 yangqiao: well, he is busy too, but i don't want a large backlog of patches. if the RSB patches work they can be submitted even without the RPi support available, right? 2015-07-14T15:24:03 ok, thanks sujayraaj_ good luck, it's messy business working at the bleeding edge of development sometimes. 2015-07-14T15:24:03 sujayraaj_: There is likely none on porting. You would have to find the appropriate files in the corresponding freebsd source and integrate them into the rtems-libbsd tree 2015-07-14T15:24:33 sujayraaj_: and as ragunath said, look at that file for instructions on how to go about doing such a port. 2015-07-14T15:24:34 yes, rsb is independent. I'll send patches for rsb and rtems-tools (for build patches) 2015-07-14T15:24:57 ok yangqiao that would be good so we can get them committed soon 2015-07-14T15:25:02 Detecting that odd resolution as an indicator is probably OK. It didn't look like any I had ever seen before. Any magic non-useful value that indicates a state is OK. If there turn out to be more magic values or different ones, the code can change 2015-07-14T15:25:09 ok. no problem 2015-07-14T15:25:18 yangqiao: are you able to see the values of atag memory in a good state? or are they always corrupted? 2015-07-14T15:25:40 yangqiao: If it helps, I posted patches to port another couple of libraries to RSB building. I did the libjpeg one. 2015-07-14T15:26:53 anandkp92: you'll be up next and can start when you're ready, we'll overlap a bit with yangqiao 2015-07-14T15:26:55 always corrupted. I don't think it's my mistake of picking the wrong format. I was planning to run the same atag scan program on a linux distribution 2015-07-14T15:27:33 *** EdSut has quit IRC 2015-07-14T15:27:34 ok yangqiao that is a good idea, and compare your code perhaps to see how it aligns 2015-07-14T15:27:41 in fact, the atag parameter shows just a part of it , correctly 2015-07-14T15:27:48 yangqiao: does there need to be an initialization done on it? 2015-07-14T15:27:53 yeah i saw that... 2015-07-14T15:27:53 ok 2015-07-14T15:28:19 Hi all... so I have implemented String type options in the GUI and its almost complete now. 2015-07-14T15:28:32 Currently I am working on cleaning up the classes and create a smoother flow for GUI. Thomas has given suggestions on improving the code and the documentation. 2015-07-14T15:28:45 And so I will be working on the documentation of what I have completed so far. 2015-07-14T15:28:54 Next step would be to actually create the config.cfg file using all the values that have been entered - this would happen upon clicking Configure menu item. 2015-07-14T15:29:13 Normally, the atag is in the area befor 0x8000, where the kernel placed. And it's set by gpu whe the system boot. No more initialization is necessary I think 2015-07-14T15:29:16 anandkp92: ok great, will that be a fully working demo then? 2015-07-14T15:29:34 yangqiao: ok, but then does the boot code touch that memory perhaps? 2015-07-14T15:29:37 But I've got to find out why it preserve only part of it... A bit anoying 2015-07-14T15:29:40 no, after that we'll have to run Build option from the gui and then it will be a fully working demo 2015-07-14T15:30:10 but i think after getting the cfg file, the rest wouldn't be so time consuming 2015-07-14T15:30:31 anandkp92: ok thanks, any questions right now? 2015-07-14T15:30:34 rpi use the default arm boot code start.S . I'm looking into the code, it's a bit difficult but I didn't come up with a better idea. 2015-07-14T15:30:47 no.. nothing else :) 2015-07-14T15:31:22 yangqiao: yes it can be hard to understand. it may be a case that the region of memory for the atag needs to be properly set in the linkcmds file to avoid the shared code from using it. i'm not certain really. 2015-07-14T15:31:25 anand are you using baseconfigparser in python for config file changes? 2015-07-14T15:31:34 ok thanks anandkp92. asuolen you can go now 2015-07-14T15:31:39 thanks :) 2015-07-14T15:31:44 hello 2015-07-14T15:32:06 I have added to the GPIO API support for pin groupings and multi-pin operations. These are useful to operate on a GPIO pin group as a if it was a single entity (i.e.: can be useful to emulate/bit-bang interfaces using GPIO pins), and also allow for fewer hardware calls. I also updated the existing data structures to optimize the memory requirements. The current state and description of the API is currently in my blog (which I have u 2015-07-14T15:32:17 I have tested all the features with the Pi, and am now updating the existing API test cases. At this point the API is almost ready for another round of code review. In the meantime I want to work in the I2C porting, which hopefully can be done by the next week. 2015-07-14T15:32:28 I tried to run the scan code in the bspstart.c . It seems to be corrupted. So maybe I can locate the problem in the early boot stage. 2015-07-14T15:32:57 asuolen: are you satisfied with the IRQ support in the GPIO framework? 2015-07-14T15:33:30 I'm planning to do some tests as I said . I'll report in time 2015-07-14T15:33:42 yangqiao: yeah i think you're going to have to push in earlier. and compare the atag memory region with the regions defined in the linkcmds 2015-07-14T15:33:51 thanks yangqiao 2015-07-14T15:33:52 My main concern is that it currently relies only on threaded handlers. Maybe that feature can become optional 2015-07-14T15:33:56 thanks 2015-07-14T15:34:15 *** yangqiao has quit IRC 2015-07-14T15:34:27 asuolen: i have concerns with how the framework is designed as well. There is also the "interrupt server" approach that might be more usable to tie into 2015-07-14T15:34:47 i suspect we need to go back to the drawing board a bit on the IRQ handling portion of the API yet. 2015-07-14T15:35:03 meanwhile, doing the new i2c support is a good way to make progress. 2015-07-14T15:35:24 ok. Agree with the interrupt server. 2015-07-14T15:35:38 also gpio & i2c huh :-) 2015-07-14T15:36:34 *** anandkp92_ has joined #rtems 2015-07-14T15:36:51 *** anandkp92 has quit IRC 2015-07-14T15:37:06 Will evaluate the interrupt server, and post something in the ml. Also it should allow some interrupts to execute in ISR context if desired 2015-07-14T15:37:15 beng-nl: i know, right. 2015-07-14T15:37:29 ok asuolen that is a good plan 2015-07-14T15:37:59 thanks asuolen. krohini you're up now 2015-07-14T15:38:17 hi everybody 2015-07-14T15:38:48 I have pushed some smp support code to my github repository. 2015-07-14T15:39:17 It took me quite some time to get the boot sequence to be able to identify the implementation 2015-07-14T15:39:35 *** Davidbrcz has quit IRC 2015-07-14T15:39:43 krohini: is your SMP code ready for review, or are you still working on it? 2015-07-14T15:39:54 krohini: i just sent an email response on your configure issue 2015-07-14T15:40:03 we can keep working on that through email 2015-07-14T15:40:13 ok gedare. 2015-07-14T15:40:39 I have only a basic implementation of some required functions. 2015-07-14T15:41:29 ok do you know how to proceed, after you figure out how to actually build with SMP turned on? 2015-07-14T15:41:34 though a review of that would be helpful. I did figure out the functions where implementation for pi2 becomes different from existing a9-mpcore support and implemented those 2015-07-14T15:42:28 I suppose I would want to see th epi2 boot succesfully. 2015-07-14T15:42:40 ok sounds reasonable 2015-07-14T15:42:48 *** anandkp92_ has quit IRC 2015-07-14T15:42:50 then, let's keep working on the build system problems. 2015-07-14T15:42:51 sounds like you broke through the configureissues 2015-07-14T15:42:55 anything else krohini? 2015-07-14T15:43:02 There are aspects like interrupt support which I saw 2015-07-14T15:43:19 sorry, which I need to look into 2015-07-14T15:43:44 But then I think I would move on to that after I have a correct boot up. 2015-07-14T15:44:29 DrJoel: I have been following it up. From my side it is yet to be resolved. But ya, guess I am close 2015-07-14T15:44:54 And the code refactor also I suppose is close to required 2015-07-14T15:45:24 alright krohini thanks. hermann_19829 you're up 2015-07-14T15:45:33 hi 2015-07-14T15:45:36 *** sgworks has quit IRC 2015-07-14T15:45:44 I prepared 2 blog posts explaining the setup to generate coverage reports for pc386 and leon2. Now everybody should be able to reproduce my reports. Further I generated a coverage report for pc386 that I also published on my blog (not finished to set the correct links yet for all results). 2015-07-14T15:45:56 Next step is to correct the diagnostic output messages for mismatched sizes of covoar and changing the behavior such that a size mismatch is not treated as an error, as suggested by DrJoel. 2015-07-14T15:46:00 Thanks. 2015-07-14T15:46:38 ok hermann_19829, that and getting the hard-coded paths and any other environment-related dependencies fixed? 2015-07-14T15:47:34 no the hard-coded paths are not fixed yet. but they are next. 2015-07-14T15:48:05 *** sgworks has joined #rtems 2015-07-14T15:48:07 i already asked Krzysztof whether he is still working on that. otherwise I will fix that 2015-07-14T15:48:36 If he hasn't gotten to it by now, then you can assume it is your issue 2015-07-14T15:48:58 ok DrJoel 2015-07-14T15:49:17 What's after that? 2015-07-14T15:49:28 I want to use this to get coverage runs for 4.11 :) 2015-07-14T15:50:17 well DrJoel you probably can if you fix the paths for your local checkout 2015-07-14T15:50:18 after that we should have a running report generator which we could integrate into the rtems-tools repo 2015-07-14T15:50:53 is separating the report into smaller modules on your horizon? 2015-07-14T15:51:02 gedare: want it merged and no hard-coded paths. :) 2015-07-14T15:52:01 dure. 2015-07-14T15:52:02 sure. 2015-07-14T15:52:16 sgworks: you're up we'll overlap a bit with hermann_19829 2015-07-14T15:52:31 well, hermann_19829 and DrJoel can discuss later or by email. 2015-07-14T15:52:37 hi sgworks, javamonn. 2015-07-14T15:52:38 1. Identify data race conditions in RTEMS 2. Find whether biglock can be broken in RTEMS C code. 3. May be I will try to code the RTEMS based on our JPF solution. And see what are potential data race conditions. (Need to confirm with Cyrille) 4. Documentation. We are good with biglock implementation in JPF although we are struggling with JPF's verify() to get enough resources to 2015-07-14T15:52:40 yes gedare. I will have to write some ideas down how I would work on that and then coordinate that with you 2015-07-14T15:53:13 complete the test 2015-07-14T15:53:46 right trying to find different solutions for implementation of locks 2015-07-14T15:54:01 ^now 2015-07-14T15:54:02 sgworks: there is support for fine-grained locks in RTEMS, but you have to know what structure is being locked. also, acquiring a lock in order to deal with priority inheritance issues due to acquiring a lock should be handled carefully to avoid dependency between the locks involved. 2015-07-14T15:54:42 which fine-grained locks? 2015-07-14T15:54:47 but the fine-grained locks are mainly involved for SMP support 2015-07-14T15:54:50 see the spinlock implementations 2015-07-14T15:54:54 yes 2015-07-14T15:55:10 is it MCS one? 2015-07-14T15:55:31 anyways I will have a look at it 2015-07-14T15:55:32 i don't know if we care yet, since the resource sharing is different in SMP. the problem you're working on is mainly relevant in UP 2015-07-14T15:55:51 and in UP, these locks just get turned into isr_disable / isr_enable, since that is most efficient 2015-07-14T15:55:59 ok gedare, let's discuss later DrJoel 2015-07-14T15:56:14 ok 2015-07-14T15:56:47 hermann_19829 +1 and let's stick to the devel list as much as possible 2015-07-14T15:57:28 yes we are just disabling the interrupts 2015-07-14T15:57:44 sgworks: the problem to consider then is how long is the lock held 2015-07-14T15:58:01 yes 2015-07-14T15:58:08 FWIW while everyone is here, it looks like I will be in Hastings UK next week. If that's close to anyone, email me privately 2015-07-14T15:58:10 so we need to know how much time the critical section lasts, i.e. how big are the linked lists 2015-07-14T15:58:14 yes DrJoel I will use it more often from now 2015-07-14T15:58:15 *** Davidbrcz has joined #rtems 2015-07-14T15:58:57 Critical sections need to be fixed time, bounded by a design constant, or at least predictable "big-O" 2015-07-14T15:59:52 sgworks: so we need you to work out the parameters that control how much time is spent holding the global lock, ok? 2015-07-14T16:00:19 yes I guess its just about updating all nested/interconnected mutex 2015-07-14T16:00:23 and, my half a banana is not holding me over through lunchtime, so i'd like to wrap this up. :) did I miss anyone? 2015-07-14T16:00:31 and changing the threads current priority 2015-07-14T16:00:52 sgworks: it will be a function of mutexes and/or threads, and possibly on how many mutexes can be held by threads. 2015-07-14T16:00:57 i need you to figure out the details 2015-07-14T16:01:04 Okay I will analyse on points you mentioned and get back to you on main 2015-07-14T16:01:04 mail 2015-07-14T16:01:12 and how to make it as "tight" a bound as possible 2015-07-14T16:01:22 yes 2015-07-14T16:01:22 i.e. don't just say nthreads * nmutexes, if it can be smaller. 2015-07-14T16:01:24 ok great 2015-07-14T16:01:40 oh, gtament are you herenow? 2015-07-14T16:02:13 gedare: ok, just deleting the earlier build and doing it again worked right. some learning alon the way, but certainly annoying.. 2015-07-14T16:02:19 krohini: i don't see your configure.ac changes in your repo yet 2015-07-14T16:02:25 alright krohini 2015-07-14T16:02:36 if you add --enable-maintainer-mode to your configure cmdline it may just work better 2015-07-14T16:02:46 the autotools build system is a bit flukey sometimes 2015-07-14T16:03:09 gtament: if you get back and get this, can you just send an email update? thanks. 2015-07-14T16:04:15 *** Davidbrcz has quit IRC 2015-07-14T16:04:18 umm, ok. will try it the next time. I do keep coming across quite a few build issues.. 2015-07-14T16:06:34 *** sgworks has quit IRC 2015-07-14T16:07:04 ok thanks everyone. until next week! 2015-07-14T16:07:07 hmm 2015-07-14T16:07:28 DrJoel: can you run the meeting next week? 2015-07-14T16:08:07 thanks gedare, EdSut, beng-nl, DrJoel, and everyone 2015-07-14T16:08:17 Not likely I will be in the UK 2015-07-14T16:08:19 let's just cancel next week. 2015-07-14T16:08:24 OK 2015-07-14T16:08:24 NO MEETING NEXT WEEK 2015-07-14T16:08:47 ok gedare. Thanks everyone :) 2015-07-14T16:08:51 sure, i'll be at NIST all day. 2015-07-14T16:09:02 Sounds fun! Mine is work also 2015-07-14T16:11:13 *** jrcatbagan has quit IRC 2015-07-14T16:12:17 *** irnhorse4 has quit IRC 2015-07-14T16:12:58 *** hermann_19829 has quit IRC 2015-07-14T16:16:45 *** ragunath has quit IRC 2015-07-14T16:18:24 *** krohini has quit IRC 2015-07-14T17:05:44 DrJoel, I had a question, does rtems support unix domain sockets, AF_UNIX ? 2015-07-14T17:08:46 It should. That should just be naturally supported in either stack. 2015-07-14T17:16:51 *** Davidbrcz has joined #rtems 2015-07-14T17:23:57 *** gtament has quit IRC 2015-07-14T17:33:03 *** sujayraaj_ has quit IRC 2015-07-14T17:45:26 *** asuolen has quit IRC 2015-07-14T17:51:31 *** krohini has joined #rtems 2015-07-14T17:51:44 *** krohini has quit IRC 2015-07-14T17:59:40 *** ketul has quit IRC 2015-07-14T18:24:36 *** hesham has joined #rtems 2015-07-14T18:25:45 *** fds_ has joined #rtems 2015-07-14T18:26:19 *** gigetoo_ has joined #rtems 2015-07-14T18:29:43 *** kulp_ has joined #rtems 2015-07-14T18:30:50 *** mumptai has quit IRC 2015-07-14T18:30:51 *** fds has quit IRC 2015-07-14T18:30:56 *** gigetoo has quit IRC 2015-07-14T18:30:57 *** kulp has quit IRC 2015-07-14T18:30:57 *** edwardk has quit IRC 2015-07-14T18:30:59 *** gigetoo_ is now known as gigetoo 2015-07-14T18:58:26 *** javamonn has quit IRC 2015-07-14T19:06:41 *** kulp_ has quit IRC 2015-07-14T19:06:41 *** kulp_ has joined #rtems 2015-07-14T19:06:45 *** kulp_ is now known as kulp 2015-07-14T19:49:21 *** mumptai has joined #rtems 2015-07-14T19:51:07 *** Davidbrcz has quit IRC 2015-07-14T19:52:51 *** edwardk has joined #rtems 2015-07-14T20:14:21 [trac] #2370/defect (v:4.12) created by dcb (rtems/c/src/lib/libbsp/powerpc/mvme5500/network/if_1GHz/if_wm.c:1560: possible bad if ?) http://devel.rtems.org/ticket/2370 2015-07-14T20:19:51 [trac] #2370/defect (new) updated by joel.sherrill (I didn't check this file but assume it is based on the one in NetBSD. ...) http://devel.rtems.org/ticket/2370 2015-07-14T20:23:03 *** hermann_19829_ has joined #rtems 2015-07-14T20:37:34 [trac] #2370/defect (new) updated by gedare (Problem is that masking off 0xe removes the 1 bit, so anything & 0xe != 1 ever.) http://devel.rtems.org/ticket/2370 2015-07-14T20:48:13 *** hermann_19829_ has quit IRC 2015-07-14T20:50:21 [trac] #2370/defect (new) updated by joel.sherrill (So the code is still broken on the NetBSD head?) http://devel.rtems.org/ticket/2370 2015-07-14T20:57:57 [trac] #2370/defect (new) updated by gedare (Well, the second part of the condition is a nop. I have no idea what the intent of ...) http://devel.rtems.org/ticket/2370 2015-07-14T21:04:54 *** gedare has quit IRC 2015-07-14T21:23:17 [trac] #2370/defect (new) updated by joel.sherrill (This driver is for an Intel NIC. The driver (by name) is not in FreeBSD. The NIC is ...) http://devel.rtems.org/ticket/2370 2015-07-14T21:23:22 *** Davidbrcz has joined #rtems 2015-07-14T21:36:42 *** dr__house|away is now known as dr__house 2015-07-14T21:55:36 [trac] #2371/enhancement (v:4.11) created by joel.sherrill (Add support for POSIX robust mutexes) http://devel.rtems.org/ticket/2371 2015-07-14T22:16:43 *** Davidbrcz has quit IRC 2015-07-14T23:21:55 [trac] GSoC/2015 edited by jrcatbagan (Weekly Status Update for July 14, 2015) http://devel.rtems.org/wiki/GSoC/2015 2015-07-14T23:39:03 *** mumptai has quit IRC 2015-07-15T00:44:58 *** kiwichris has quit IRC 2015-07-15T01:06:16 *** kiwichris has joined #rtems 2015-07-15T04:04:33 *** ketul has joined #rtems 2015-07-15T04:43:14 *** dr__house is now known as dr__house|away 2015-07-15T04:58:57 *** gedare has joined #rtems 2015-07-15T04:58:57 *** ChanServ sets mode: +o gedare 2015-07-15T04:59:18 hi gedare 2015-07-15T05:04:12 *** gedare has quit IRC 2015-07-15T05:38:59 *** Davidbrcz has joined #rtems 2015-07-15T05:44:45 *** hesham has quit IRC 2015-07-15T05:54:57 [trac] #2369/defect (new) updated by nick.withers (https://sourceware.org/ml/libc-alpha/2008-04/msg00036.html and ...) http://devel.rtems.org/ticket/2369 2015-07-15T06:06:33 *** Davidbrcz has quit IRC 2015-07-15T06:24:08 [trac] ppc-mfspr.patch (Proposed patch to use mfspr from the main time base universally; does not address MVME3100 CFLAGS issues) attached to #2369 by nick.withers http://devel.rtems.org/ticket/2369 2015-07-15T06:47:06 [git] c5d0ca9 by Sebastian Huber: [libcsupport: Workaround for GCC 5.1 and later Disable an optimization which would lead to a recursive calloc() call in calloc().] http://devel.rtems.org/changeset/c5d0ca9/rtems 2015-07-15T06:56:47 sebhub, hi 2015-07-15T06:58:50 hi chris 2015-07-15T07:03:19 [trac] #2369/defect (closed) updated by Nick Withers (In [changeset:"e2fcb7dc64c040b8298684148390aba6bf4f4912/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2369 2015-07-15T07:03:20 [git] e2fcb7d by Nick Withers: [powerpc: Fix _CPU_Counter_read() The mftb is not available on Book E processors. Use SPR 268 instead. Close #2369.] http://devel.rtems.org/changeset/e2fcb7d/rtems 2015-07-15T07:04:46 sebhub, is the 5.1 calloc issue been raised as a bug ? 2015-07-15T07:05:15 [trac] #2369/defect (closed) updated by sebastian.huber (Sorry, I didn't noticed your patch. the Alternate Time Base may run with a much ...) http://devel.rtems.org/ticket/2369 2015-07-15T07:05:45 a gcc bug? no 2015-07-15T07:06:26 there are similar issues with memset() etc. it is solved in glibc via compiler options and attributes 2015-07-15T07:06:46 i don't think gcc will fix this, but you can open a PR if you feel better ;-) 2015-07-15T07:06:54 Why is this happening ? 2015-07-15T07:07:26 Huh, us changing optmisation is a hack and I did not approve the patch. 2015-07-15T07:07:28 GCC knows what the C library functions do and replaces code patterns that resemble a C library call 2015-07-15T07:07:53 Then it has a bug. 2015-07-15T07:08:30 If it wants to be this smart and compete then surely it can see the function itself is call the name it is replacing. How hard is that ? 2015-07-15T07:08:41 The attribute hack sure, that is a hack. 2015-07-15T07:09:26 i don't know how hard this is 2015-07-15T07:09:43 They created the problem so why no fix it. 2015-07-15T07:09:51 before this is fixed in GCC I don't see another way to fix this 2015-07-15T07:10:06 there is a fix, the GCC option 2015-07-15T07:10:17 A work around once fix sure but that is a different converstion. 2015-07-15T07:10:58 Really and truly gcc is happy to be known as the compiler that works at -O and lower ? 2015-07-15T07:10:58 so you would open a GCC PR? 2015-07-15T07:11:37 I would if I used 5.1 and cared but I do not use it and nor does RTEMS 4.11. 2015-07-15T07:11:58 So now 4.11 with its released compiler is hit by this. 2015-07-15T07:12:19 On the master after 4.11 that is also a different converstion. 2015-07-15T07:12:24 conversation. 2015-07-15T07:13:35 this option has no impact on GCC 4.9 2015-07-15T07:13:49 I think gcc and what the people who develop it do is fantastic but they need to be mindful of all their users and my concern for 5.1 and our users is how this sort of transformation effects users. 2015-07-15T07:14:03 users don't write calloc 2015-07-15T07:14:05 And does 4.9 have the same bug ? 2015-07-15T07:14:18 no, its new to 5.1 and later 2015-07-15T07:14:25 No and we do and yet an option can bypass it. 2015-07-15T07:15:07 Why build it when using 5.1 ? 2015-07-15T07:15:54 Does this change effect RTEMS 4.11 releases with GCC 4.9 ? 2015-07-15T07:16:14 no 2015-07-15T07:16:46 the -fno-builtin option is available for a long time 2015-07-15T07:17:13 if you use other compilers, then this is a problem 2015-07-15T07:18:00 if you feel uncomfortable with this patch, we can revert it, but in this case GCC 5.1 or later cannot be used 2015-07-15T07:19:27 I have no problem with this. We upport gcc 4.9 for RTEMS 4.11. Using 5.1 with RTEMS 4.11 is a user issue to resolve. 2015-07-15T07:19:47 I have a range of calloc calls in my executable with 4.9 2015-07-15T07:19:55 Is calloc as a builtin new in 5.1 ? 2015-07-15T07:20:12 And we are not using --no-builtin 2015-07-15T07:20:14 [trac] #2369/defect (closed) updated by sebastian.huber (I would leave the c/src/lib/libcpu/powerpc/shared/include/powerpc-utility.h as is.) http://devel.rtems.org/ticket/2369 2015-07-15T07:21:10 the problem are not the calloc calls 2015-07-15T07:21:40 the problem is that GCC replaces a malloc() + memset() with a call to calloc() 2015-07-15T07:22:26 so if you implement calloc() with a malloc() + memset() (as we do in RTEMS) you and up with calloc() calling calloc() 2015-07-15T07:22:33 which we now have turn off optimisation. It would be better to add a config to RTEMS gcc to stop the transformation and to have malloc and memset get used. 2015-07-15T07:23:03 no, for user code this is all right 2015-07-15T07:23:24 I understand the bug and what I feel gcc should check for, ie doing in the call itself. 2015-07-15T07:23:48 i cannot change the GCC behaviour 2015-07-15T07:24:02 except using the right options 2015-07-15T07:24:07 But the calloc call that is made is to our code which is now built with which options ? 2015-07-15T07:24:32 -O2 -fno-builtin 2015-07-15T07:24:38 only for calloc.c 2015-07-15T07:27:00 Do you have a cortexA9 code sample for calloc handy ? 2015-07-15T07:27:08 with 5.1 ? 2015-07-15T07:28:54 https://lists.rtems.org/pipermail/devel/2015-March/010710.html 2015-07-15T07:30:23 I mean with -O2 -fno-builtin 2015-07-15T07:30:31 What we will have with this change. 2015-07-15T07:31:15 I write my own infinite loops without the compiler helping out :) 2015-07-15T07:32:03 i checked it only on powerpc, and here it works with this option 2015-07-15T07:32:16 its not only an arm problem 2015-07-15T07:33:37 The issue is what happens to the exist uses on 4.9. If there is no change then I have no objection. 2015-07-15T07:33:48 existing users that is 2015-07-15T07:34:18 ie if the code for RTEMS 4.11 and GCC 4.9 is the same then the patch can be considered a NOP. 2015-07-15T07:35:04 ok, I can check this on powerpc 2015-07-15T07:35:33 Great. I will try and check ARM tonight. 2015-07-15T07:38:56 no code change due to -fno-builtin on powerpc 2015-07-15T07:39:21 the calloc function is so simple that this would be a real surprise 2015-07-15T07:49:45 It is easy to check to makes sure there are no surprises. 2015-07-15T07:55:29 for me this infinite loop is a GCC bug, but I don't think its worth the time to push this further 2015-07-15T07:58:26 [git] a57dfa0 by Sebastian Huber: [libbsd.txt: Clarify rules to modify FreeBSD code] http://devel.rtems.org/changeset/a57dfa0/rtems-libbsd 2015-07-15T08:03:03 sebhub, I have found machine/clock.h for the i386 needs to be copied in from freebsd-org. 2015-07-15T08:03:48 i didn't work with the x86 2015-07-15T08:04:29 I suspect you are right about the gcc 5.1 bug. I think the fact they have "developer" only attributes available is not great. 2015-07-15T08:04:53 There are ARM and other versions. 2015-07-15T08:05:40 this bug is probably archtitecture independent 2015-07-15T08:05:59 i only noticed it on arm and powerpc, since this is what i use most 2015-07-15T08:07:24 I mean machine/clock.h for ARM and others. 2015-07-15T08:09:33 the arm version is empty? 2015-07-15T08:13:14 I have not lookde. 2015-07-15T08:13:16 I have not looked 2015-07-15T08:14:14 Powerpc is not. 2015-07-15T08:16:30 Do I just add 'sys/i386/include/clock.h' and then get a suitable path for machine/clock.h to point to it ? 2015-07-15T08:17:26 ie before rtemsbsd/include/machine/clock.h 2015-07-15T08:18:06 I would avoid dependencies to the FreeBSD platform dependent clock code 2015-07-15T08:19:25 we have a addCPUDependentHeaderFiles() in libbsd.py, this should be used for such header files 2015-07-15T08:19:29 but this means patching altq_subr.c 2015-07-15T08:19:46 and the code is already in with sys/i386/i386/legacy.c 2015-07-15T08:20:12 [trac] #2369/defect (closed) updated by nick.withers (Replying to [comment:12 sebastian.huber]: > Sorry, I didn't noticed your patch. ...) http://devel.rtems.org/ticket/2369 2015-07-15T08:20:28 In libbsd.py sys/i386/include/legacyvar.h is listed twice, is this an error ? 2015-07-15T08:20:55 i cannot say anything about the i386 2015-07-15T08:21:03 i never compile libbsd on this arch 2015-07-15T08:21:11 Did Joel do this ? 2015-07-15T08:26:31 [trac] #2369/defect (closed) updated by sebastian.huber (Replying to [comment:14 nick.withers]: > Replying to [comment:12 sebastian.huber]: ...) http://devel.rtems.org/ticket/2369 2015-07-15T08:27:46 yes, this stuff was done by Joel and Jennifer 2015-07-15T08:27:58 i may have broken it in the mean time 2015-07-15T08:28:26 I want to avoid hacking around issues cause they tend to come back in some way. 2015-07-15T08:28:48 yes, like my hacks to get the user space commands working 2015-07-15T08:29:01 they are now the blocker for an easy update to freebsd 10 2015-07-15T08:29:28 on the other hand i don't have an infinite budget 2015-07-15T08:33:06 [trac] #2369/defect (closed) updated by nick.withers (Replying to [comment:15 sebastian.huber]: > Replying to [comment:14 nick.withers]: ...) http://devel.rtems.org/ticket/2369 2015-07-15T08:39:14 [trac] #2369/defect (closed) updated by sebastian.huber (Replying to [comment:16 nick.withers]: > Replying to [comment:15 sebastian.huber]: ...) http://devel.rtems.org/ticket/2369 2015-07-15T08:41:13 [trac] #2369/defect (closed) updated by nick.withers (Replying to [comment:16 nick.withers]: > Replying to [comment:15 sebastian.huber]: ...) http://devel.rtems.org/ticket/2369 2015-07-15T08:43:24 *** kiwichris has quit IRC 2015-07-15T08:45:37 [trac] #2369/defect (closed) updated by sebastian.huber (Replying to [comment:18 nick.withers]: > Replying to [comment:16 nick.withers]: > ...) http://devel.rtems.org/ticket/2369 2015-07-15T08:53:26 [trac] #2369/defect (closed) updated by Sebastian Huber (In [changeset:"93f5adb644363d321ddfcbf48586296ed7add189/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2369 2015-07-15T08:53:26 [git] 93f5adb by Sebastian Huber: [powerpc: Do not use the ATB for e500 multilib The e500v1 has no support for the ATB. Update #2369.] http://devel.rtems.org/changeset/93f5adb/rtems 2015-07-15T10:08:28 *** mumptai has joined #rtems 2015-07-15T12:00:33 *** shelf is now known as timothyh 2015-07-15T12:07:46 *** edsiper has quit IRC 2015-07-15T12:08:14 *** edsiper has joined #rtems 2015-07-15T16:20:06 *** edwardk has quit IRC 2015-07-15T16:22:31 *** edwardk has joined #rtems 2015-07-15T18:07:55 *** gedare has joined #rtems 2015-07-15T18:07:55 *** ChanServ sets mode: +o gedare 2015-07-15T18:19:21 *** gedare has quit IRC 2015-07-15T18:34:10 *** gedare has joined #rtems 2015-07-15T18:34:11 *** ChanServ sets mode: +o gedare 2015-07-15T19:01:28 *** fds_ is now known as fds 2015-07-15T19:12:47 *** gedare has quit IRC 2015-07-15T19:15:06 *** ketul has quit IRC 2015-07-15T19:15:59 *** ketul has joined #rtems 2015-07-15T19:34:39 *** ketul has quit IRC 2015-07-15T20:23:35 *** gedare has joined #rtems 2015-07-15T20:23:35 *** ChanServ sets mode: +o gedare 2015-07-15T20:35:10 *** gedare has quit IRC 2015-07-15T22:10:26 *** kiwichris has joined #rtems 2015-07-16T01:21:04 *** kiwichri_ has joined #rtems 2015-07-16T01:21:24 *** kiwichris has quit IRC 2015-07-16T02:55:33 *** dr__house|away is now known as dr__house 2015-07-16T03:46:22 *** edwardk has quit IRC 2015-07-16T03:50:27 *** peerst has quit IRC 2015-07-16T04:10:16 *** edwardk has joined #rtems 2015-07-16T04:24:45 *** peerst has joined #rtems 2015-07-16T05:06:07 *** Davidbrcz has joined #rtems 2015-07-16T05:25:59 *** sujayraaj has joined #rtems 2015-07-16T05:29:16 kiwichri_ : hi 2015-07-16T05:29:28 sujayraaj, hi 2015-07-16T05:31:25 Yesterday I spent time reading about Message Queue and related things from the rtems c user guide. 2015-07-16T05:31:59 There is just one thing bugging me, that since we haven't reached a decision yet, won't it mess up our deadlines? 2015-07-16T05:41:05 kiwichri_ : ? 2015-07-16T05:41:29 Sorry, something local took me away. 2015-07-16T05:41:44 We have a technical issue to sort out and this is part of it. 2015-07-16T05:41:49 It is important to get sorted. 2015-07-16T05:42:14 If you think you can write a Monkey event module using the RTEMS classic message queue please try. 2015-07-16T05:43:04 okay. I will give it a try today. 2015-07-16T05:49:45 *** kiwichri_ has quit IRC 2015-07-16T06:02:06 *** dr__house is now known as dr__house|away 2015-07-16T06:22:06 *** Davidbrcz has quit IRC 2015-07-16T06:39:10 *** ketul has joined #rtems 2015-07-16T07:30:54 hi beng-nl :) 2015-07-16T07:34:44 *** ketul has quit IRC 2015-07-16T07:39:51 *** ketul-droid has joined #rtems 2015-07-16T07:47:58 *** sujayraaj has quit IRC 2015-07-16T07:50:00 *** ketul has joined #rtems 2015-07-16T07:53:06 *** ketul-droid has quit IRC 2015-07-16T07:53:21 *** ketul-droid has joined #rtems 2015-07-16T07:54:22 *** ketul-droid has quit IRC 2015-07-16T07:54:34 *** ketul-droid has joined #rtems 2015-07-16T07:55:57 *** ketul-droid has quit IRC 2015-07-16T08:14:44 [trac] libgomp-parallel-bench-posix-no-malloc.png () attached to #2274 by sebastian.huber http://devel.rtems.org/ticket/2274 2015-07-16T08:24:27 *** DrJoel has quit IRC 2015-07-16T08:32:00 *** ketul has quit IRC 2015-07-16T08:32:03 [trac] #2274/enhancement (accepted) updated by sebastian.huber (The malloc() problem is solved in the GCC 6.0: ...) http://devel.rtems.org/ticket/2274 2015-07-16T08:33:23 [trac] init.c (Microbench) attached to #2274 by sebastian.huber http://devel.rtems.org/ticket/2274 2015-07-16T08:57:45 *** kiwichris has joined #rtems 2015-07-16T09:05:10 *** kiwichris has quit IRC 2015-07-16T09:15:24 *** kiwichris has joined #rtems 2015-07-16T09:19:59 *** ketul has joined #rtems 2015-07-16T11:00:08 *** ketul has quit IRC 2015-07-16T11:53:53 *** ketul has joined #rtems 2015-07-16T11:58:06 *** ketul has quit IRC 2015-07-16T13:07:32 *** gedare has joined #rtems 2015-07-16T13:07:33 *** ChanServ sets mode: +o gedare 2015-07-16T13:15:31 verm__: can you see if a [PATCH 2/3] is held up for devel due to size? 2015-07-16T13:22:44 [trac] #2372/defect (v:4.11) created by ppisa ([Patch] bsp/tms570: fix get time resolution after infrastructure change to timecounter.) http://devel.rtems.org/ticket/2372 2015-07-16T13:25:00 [trac] 0001-bsp-tms570-fix-get-time-resolution-after-infrastruct.patch () attached to #2372 by ppisa http://devel.rtems.org/ticket/2372 2015-07-16T13:57:12 *** DrJoel has joined #rtems 2015-07-16T13:57:13 *** DrJoel has joined #rtems 2015-07-16T13:57:13 *** ChanServ sets mode: +o DrJoel 2015-07-16T14:47:21 *** jkridner has quit IRC 2015-07-16T14:53:03 *** jkridner has joined #rtems 2015-07-16T14:54:09 *** jkridner|work has joined #rtems 2015-07-16T14:57:24 *** jkridner has quit IRC 2015-07-16T16:40:19 *** ketul has joined #rtems 2015-07-16T17:38:11 * DrJoel is doing a build sweep of all BSPs.. then pushing a couple of patches and branching. 2015-07-16T17:52:21 beng-nl: are the patches just posted for BBB OK to merge? 2015-07-16T17:52:58 I am doing a build sweep and close to branching. If you want them in before branching, speak up. 2015-07-16T18:04:51 eHmm.. look like Umon so no impact 2015-07-16T18:16:49 [git] 9cff822 by Joel Sherrill: [Most bsp.h: Switch to LIBBSP_@CPU@_@BSP_FAMILY@_H for guard This was done by the following script run from libbsp: find * -name bsp.h | xargs -e grep -l "#ifndef.*_BSP_H" | while read b do echo $b cpu=`echo $b | cut -d'/' -f1 | tr '[:lower:]' '[:upper:]' ` bsp=`echo $b | cut -d'/' -f2 | tr '[:lower:]' '[:upper:]' ` 2015-07-16T18:16:49 g="LIBBSP_${cpu}_${bsp}_BSP_H...] http://devel.rtems.org/changeset/9cff822/rtems 2015-07-16T18:16:49 [git] 4784214 by Joel Sherrill: [remaining bsp.h: Fix by hand to LIBBSP_@CPU@_@BSP_FAMILY@_BSP_H These files were left after running the script in the previous patch.] http://devel.rtems.org/changeset/4784214/rtems 2015-07-16T18:16:49 [git] bea49c9 by Premysl Houdek: [bsp/tms570: New/generated header files for TMS570 SoC peripherals registers. The header files are generated by script make_header.py. Current script's version can be found at: https://github.com/AoLaD/rtems-tms570-utils/tree/headers/headers/python Registers offsets and fields have been extracted from reference manual. 2015-07-16T18:16:49 Signed-off-by: Premysl Houdek <...] http://devel.rtems.org/changeset/bea49c9/rtems 2015-07-16T18:16:49 [git] 602e395 by Pavel Pisa: [bsp/tms570: fix get time resolution after infrastructure change to timecounter. The update fixes breakage of TMS570 support after Alexander Krutwig switch of RTEMS time read to timecounter mechanism bsps: Convert clock drivers to use a timecounter Mechanism to specify odd (non 1 Mhz) time base update frequencies implemented 2015-07-16T18:16:49 after objections of Martin Ga...] http://devel.rtems.org/changeset/602e395/rtems 2015-07-16T18:16:50 [git] 069560a by Premysl Houdek: [bsp/tms570: source changes reflecting new headers. Signed-off-by: Premysl Houdek ] http://devel.rtems.org/changeset/069560a/rtems 2015-07-16T18:16:50 [git] 7e14385 by Joel Sherrill: [sh/shared/startup/bspstart.c: Add include of percpu.h] http://devel.rtems.org/changeset/7e14385/rtems 2015-07-16T18:16:51 [git] db01d0c by Joel Sherrill: [sparc64/rtems/score/cpu.h: Delete dead declaration] http://devel.rtems.org/changeset/db01d0c/rtems 2015-07-16T18:56:05 *** ketul has quit IRC 2015-07-16T21:10:47 *** gedare has quit IRC 2015-07-16T21:28:54 *** hesham has joined #rtems 2015-07-16T21:44:47 *** kiwichris has quit IRC 2015-07-16T21:53:14 *** kiwichris has joined #rtems 2015-07-16T21:56:34 *** DrJoel has quit IRC 2015-07-16T22:02:12 [git] 6201756 by Chris Johns: [Fix buiding for the i386.] http://devel.rtems.org/changeset/6201756/rtems-libbsd 2015-07-16T22:07:58 [trac] #2373/defect (v:4.11) created by joel.sherrill (PowerPC BSPs that do not build) http://devel.rtems.org/ticket/2373 2015-07-16T22:08:23 [trac] #2373/defect (new) updated by joel.sherrill (empty comment) http://devel.rtems.org/ticket/2373 2015-07-16T22:22:35 [git] 5ed4dc3 by Joel Sherrill: [sim-scripts/pc386.in: Update command line to new qemu] http://devel.rtems.org/changeset/5ed4dc3/rtems-testing 2015-07-16T22:53:09 *** mumptai has quit IRC 2015-07-17T01:56:16 *** Davidbrcz has joined #rtems 2015-07-17T03:08:52 *** krohini has joined #rtems 2015-07-17T03:58:20 *** hesham has quit IRC 2015-07-17T05:23:19 *** guerby_ has joined #rtems 2015-07-17T05:23:27 *** guerby has quit IRC 2015-07-17T05:36:18 *** ketul has joined #rtems 2015-07-17T05:56:48 *** Davidbrcz has quit IRC 2015-07-17T06:00:03 [trac] #2373/defect (closed) updated by Sebastian Huber (In [changeset:"e5a79e54d9949c8bdbab03ce800c781b5d859fea/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2373 2015-07-17T06:00:03 *** Davidbrcz has joined #rtems 2015-07-17T06:00:04 [git] e5a79e5 by Sebastian Huber: [bsp/mpc83xx: Update due to header guard change Close #2373.] http://devel.rtems.org/changeset/e5a79e5/rtems 2015-07-17T06:07:04 *** Davidbrcz has quit IRC 2015-07-17T06:29:23 *** zoso has joined #rtems 2015-07-17T07:22:42 [git] b56ddbb by Sebastian Huber: [doc: Fix SMP task variables section] http://devel.rtems.org/changeset/b56ddbb/rtems 2015-07-17T08:24:18 *** kiwichris has quit IRC 2015-07-17T08:24:57 *** kiwichris has joined #rtems 2015-07-17T08:27:23 *** ketul has quit IRC 2015-07-17T08:57:02 *** zoso has quit IRC 2015-07-17T09:19:06 *** kiwichris has quit IRC 2015-07-17T09:23:14 *** zoso has joined #rtems 2015-07-17T09:54:19 *** mumptai has joined #rtems 2015-07-17T09:57:29 [git] 3bb342c by Sebastian Huber: [doc: Clarify interrupts disable problems on SMP] http://devel.rtems.org/changeset/3bb342c/rtems 2015-07-17T10:06:30 *** ketul has joined #rtems 2015-07-17T10:50:53 *** ketul has quit IRC 2015-07-17T10:57:06 *** zoso has quit IRC 2015-07-17T11:19:46 *** hesham has joined #rtems 2015-07-17T11:44:21 *** ketul has joined #rtems 2015-07-17T12:09:50 *** guerby_ is now known as guerby 2015-07-17T13:26:25 hi all 2015-07-17T13:29:18 halo beng-nl :) 2015-07-17T13:38:40 *** gedare has joined #rtems 2015-07-17T13:38:40 *** ChanServ sets mode: +o gedare 2015-07-17T13:45:29 *** Davidbrcz has joined #rtems 2015-07-17T13:51:02 *** Davidbrcz has quit IRC 2015-07-17T13:52:11 [git] 9154c3f by Sebastian Huber: [doc: Add thread dispatch details for SMP] http://devel.rtems.org/changeset/9154c3f/rtems 2015-07-17T14:14:27 *** sebhub has quit IRC 2015-07-17T14:42:49 [git] 3f923fd by Joel Sherrill: [Update date] http://devel.rtems.org/changeset/3f923fd/rtems 2015-07-17T14:43:44 [git] ac7d5ef by Joel Sherrill: [Initial revision] http://devel.rtems.org/changeset/ac7d5ef/rtems 2015-07-17T14:43:45 [git] bf61e45 by Joel Sherrill: [Forcing RCS revision to 1.2 I hope] http://devel.rtems.org/changeset/bf61e45/rtems 2015-07-17T14:43:45 [git] 88d594a by Joel Sherrill: [Fully tested on all in-house targets] http://devel.rtems.org/changeset/88d594a/rtems 2015-07-17T14:43:45 [git] 5b9d6dd by Joel Sherrill: [test force of revision] http://devel.rtems.org/changeset/5b9d6dd/rtems 2015-07-17T14:43:45 *** rtems-trac_ has quit IRC 2015-07-17T14:44:11 *** rtems-trac has joined #rtems 2015-07-17T14:47:03 *** gigetoo has quit IRC 2015-07-17T14:47:30 *** geheimnis` has quit IRC 2015-07-17T14:48:11 *** geheimnis` has joined #rtems 2015-07-17T14:52:42 *** gigetoo has joined #rtems 2015-07-17T15:18:47 [git] bf13d27 by Chris Johns: [Initial import.] http://devel.rtems.org/changeset/bf13d27/rtems-source-builder 2015-07-17T15:18:47 [git] 19622ea by Chris Johns: [Merged in SpecBuilder FreeBSD support.] http://devel.rtems.org/changeset/19622ea/rtems-source-builder 2015-07-17T15:18:47 [git] 864360e by Chris Johns: [Add a force option.] http://devel.rtems.org/changeset/864360e/rtems-source-builder 2015-07-17T15:18:47 [git] ed0de00 by Chris Johns: [Create the sources path if not present and the user has forced.] http://devel.rtems.org/changeset/ed0de00/rtems-source-builder 2015-07-17T15:18:47 [git] 5260449 by Chris Johns: [Support modules other than bintuls and gcc. Added support to build autoconf and automake plus other packages such as libusb and the stlink project. This includes support to build a specific github version downloaded from github.] http://devel.rtems.org/changeset/5260449/rtems-source-builder 2015-07-17T15:18:47 [git] f107171 by Chris Johns: [Fix the error message on a download failure.] http://devel.rtems.org/changeset/f107171/rtems-source-builder 2015-07-17T15:18:47 [git] 4861d2e by Chris Johns: [Add expat and gdb support.] http://devel.rtems.org/changeset/4861d2e/rtems-source-builder 2015-07-17T15:18:48 *** rtems-trac has quit IRC 2015-07-17T15:18:55 *** rtems-trac has joined #rtems 2015-07-17T15:20:04 [git] bf13d27 by Chris Johns: [Initial import.] http://devel.rtems.org/changeset/bf13d27/rtems-source-builder 2015-07-17T15:20:04 [git] 19622ea by Chris Johns: [Merged in SpecBuilder FreeBSD support.] http://devel.rtems.org/changeset/19622ea/rtems-source-builder 2015-07-17T15:20:04 [git] 864360e by Chris Johns: [Add a force option.] http://devel.rtems.org/changeset/864360e/rtems-source-builder 2015-07-17T15:20:04 [git] ed0de00 by Chris Johns: [Create the sources path if not present and the user has forced.] http://devel.rtems.org/changeset/ed0de00/rtems-source-builder 2015-07-17T15:20:05 [git] 5260449 by Chris Johns: [Support modules other than bintuls and gcc. Added support to build autoconf and automake plus other packages such as libusb and the stlink project. This includes support to build a specific github version downloaded from github.] http://devel.rtems.org/changeset/5260449/rtems-source-builder 2015-07-17T15:20:05 *** rtems-trac has quit IRC 2015-07-17T15:20:11 *** rtems-trac has joined #rtems 2015-07-17T15:20:31 [git] 78c9ac0 by Chris Johns: [Create RTEMS Tools.] http://devel.rtems.org/changeset/78c9ac0/rtems-tools 2015-07-17T15:20:31 [git] 2dc8560 by Chris Johns: [This commit was generated by cvs2svn to compensate for changes in r2, which included commits to RCS files with non-trunk default branches.] http://devel.rtems.org/changeset/2dc8560/rtems-tools 2015-07-17T15:20:31 [git] 7231f49 by Chris Johns: [2010-08-09 Chris Johns * specbuilder, specbuilder/sb-build, specbuilder/sb-crossgcc, specbuilder/sb-setup, specbuilder/sb-specdump, specbuilder/sb-status, specbuilder/specbuilder/.cvsignore, specbuilder/specbuilder/build.py, specbuilder/specbuilder/config.sub, 2015-07-17T15:20:31 specbuilder/specbu...] http://devel.rtems.org/changeset/7231f49/rtems-tools 2015-07-17T15:20:31 [git] dc2ed4f by Chris Johns: [2010-08-09 Chris Johns * specbuilder/perl.prov, specbuilder/perl.req: New. * specbuilder/defaults.py: The config.sub path is a sub-dir of the command. * specbuilder/setup.py: Fix make bug. Install RPMLIB files when creating the directory. * 2015-07-17T15:20:31 *** rtems-trac has quit IRC 2015-07-17T15:20:38 *** rtems-trac has joined #rtems 2015-07-17T15:24:08 [git] 6480e09 by Joel Sherrill: [Initial import.] http://devel.rtems.org/changeset/6480e09/rtems-testing 2015-07-17T15:24:09 [git] e200336 by Joel Sherrill: [This commit was generated by cvs2svn to compensate for changes in r2, which included commits to RCS files with non-trunk default branches.] http://devel.rtems.org/changeset/e200336/rtems-testing 2015-07-17T15:24:09 [git] 5e35b21 by Joel Sherrill: [2008-11-26 Joel Sherrill * README, VERSIONS: Augment and update. * ChangeLog: New file.] http://devel.rtems.org/changeset/5e35b21/rtems-testing 2015-07-17T15:24:09 [git] f8617e3 by Joel Sherrill: [2008-11-26 Joel Sherrill * README: Update as trying this on gcc CFARM. * test_driver: If newlib, binutils or gdb is checked out from CVS (e.g. source directory is (*-cvs), the update before building.] http://devel.rtems.org/changeset/f8617e3/rtems-testing 2015-07-17T15:24:09 [git] cca9b69 by Joel Sherrill: [2008-12-10 Joel Sherrill * gcc/testsuite/ada/acats/Makefile.rtems: Invoke simsh correctly.] http://devel.rtems.org/changeset/cca9b69/rtems-testing 2015-07-17T15:24:09 *** rtems-trac has quit IRC 2015-07-17T15:24:15 *** rtems-trac has joined #rtems 2015-07-17T15:24:45 [git] 3083373 by Joel Sherrill: [Initial import of reorganized examples.] http://devel.rtems.org/changeset/3083373/examples-v2 2015-07-17T15:24:46 [git] 1400868 by Joel Sherrill: [This commit was generated by cvs2svn to compensate for changes in r2, which included commits to RCS files with non-trunk default branches.] http://devel.rtems.org/changeset/1400868/examples-v2 2015-07-17T15:24:46 [git] f65262f by Joel Sherrill: [2009-08-06 Joel Sherrill * ChangeLog, Makefile, cxx_throw/MAIL, cxx_throw/Makefile, cxx_throw/init.cc, libcpp/Makefile, libcpp/README, libcpp/foo.cc, libcpp/foo.h, libcpp/foo1.cpp: New files.] http://devel.rtems.org/changeset/f65262f/examples-v2 2015-07-17T15:24:46 [git] ab7caff by Joel Sherrill: [2009-08-06 Joel Sherrill * ChangeLog, Makefile, fdopen/.cvsignore, fdopen/Makefile, fdopen/README, fdopen/test.c, filerdback/.cvsignore, filerdback/Makefile, filerdback/README, filerdback/test.c, repeated_opens/.cvsignore, repeated_opens/Makefile, repeated_opens/README, 2015-07-17T15:24:46 repeated_opens/test.c: New files.] http://devel.rtems.org/changeset/ab7caff/examples-v2 2015-07-17T15:24:46 *** rtems-trac has quit IRC 2015-07-17T15:24:52 *** rtems-trac has joined #rtems 2015-07-17T15:25:19 [git] c87143a by Joel Sherrill: [base from Eric Norum -- Demos.30May1998.tar.gz] http://devel.rtems.org/changeset/c87143a/network-demos 2015-07-17T15:25:19 [git] f444fb8 by Joel Sherrill: [This commit was generated by cvs2svn to compensate for changes in r2, which included commits to RCS files with non-trunk default branches.] http://devel.rtems.org/changeset/f444fb8/network-demos 2015-07-17T15:25:19 [git] 481950e by Joel Sherrill: [Added notes from Eric to README. Added CVS $Id$ to all files.] http://devel.rtems.org/changeset/481950e/network-demos 2015-07-17T15:25:19 [git] 497f13d by Joel Sherrill: [Switched to using "../user.cfg"] http://devel.rtems.org/changeset/497f13d/network-demos 2015-07-17T15:25:19 [git] 2ddb68f by Joel Sherrill: [New files.] http://devel.rtems.org/changeset/2ddb68f/network-demos 2015-07-17T15:25:19 *** rtems-trac has quit IRC 2015-07-17T15:25:26 *** rtems-trac has joined #rtems 2015-07-17T15:43:39 *** dr__house|away is now known as dr__house 2015-07-17T15:50:54 *** hesham has quit IRC 2015-07-17T16:27:19 *** DrJoel has joined #rtems 2015-07-17T16:27:19 *** DrJoel has joined #rtems 2015-07-17T16:27:19 *** ChanServ sets mode: +o DrJoel 2015-07-17T16:28:29 *** DrJoel changes topic to "Welcome to RTEMS (http://www.rtems.org). We have been ACCEPTED as an organization in both the 2015 Google Summer of Code and ESA Summer of Code in Space student programs. The 4.1 branch has been cut. Please test!!!" 2015-07-17T16:35:34 oops ! DrJoel *4.11 :) 2015-07-17T16:48:45 *** DrJoel changes topic to "Welcome to RTEMS (http://www.rtems.org). We have been ACCEPTED as an organization in both the 2015 Google Summer of Code and ESA Summer of Code in Space student programs. The 4.11 branch has been cut. Please test!!!" 2015-07-17T16:48:50 That should be better now. :) 2015-07-17T16:49:19 Soon we will have to discuss bumping the major, deleting some targets, deleting some BSPs, ... clean up. 2015-07-17T16:50:56 and now it is time for me to go see the Minions movie with my sons and finish getting ready to go to the UK next week 2015-07-17T16:51:14 I should be sporadically answering email still 2015-07-17T16:51:44 have a great week :) 2015-07-17T16:53:41 Thanks. 2015-07-17T16:53:45 *** DrJoel has quit IRC 2015-07-17T17:16:46 *** ketul has quit IRC 2015-07-17T19:27:22 *** krohini has joined #rtems 2015-07-17T20:07:34 *** krohini has quit IRC 2015-07-17T20:09:19 *** krohini has joined #rtems 2015-07-17T20:50:28 *** krohini has quit IRC 2015-07-17T20:53:45 *** gedare has quit IRC 2015-07-17T23:23:48 *** kiwichris has joined #rtems 2015-07-17T23:26:37 *** mumptai has quit IRC 2015-07-17T23:37:47 [trac] #2374/defect (v:4.11) created by chrisj (RSB builds 3rd party packages as Canadian Cross (Cxc) packages) http://devel.rtems.org/ticket/2374 2015-07-17T23:42:31 [trac] Packages/SNMP edited by chrisj (Add formatting to make the code section readable.) http://devel.rtems.org/wiki/Packages/SNMP 2015-07-18T02:02:35 *** gedare has joined #rtems 2015-07-18T02:02:35 *** ChanServ sets mode: +o gedare 2015-07-18T02:14:50 *** gedare has quit IRC 2015-07-18T03:23:26 *** kiwichris has quit IRC 2015-07-18T03:30:10 *** kiwichris has joined #rtems 2015-07-18T03:35:32 *** kiwichris has quit IRC 2015-07-18T03:53:02 *** kiwichris has joined #rtems 2015-07-18T04:03:20 *** kiwichris has quit IRC 2015-07-18T04:25:48 *** kiwichris has joined #rtems 2015-07-18T04:34:17 *** ketul has joined #rtems 2015-07-18T05:05:29 *** kiwichris has quit IRC 2015-07-18T05:05:44 *** kiwichris has joined #rtems 2015-07-18T05:46:10 *** dr__house is now known as dr__house|away 2015-07-18T07:18:46 *** Schoumi has quit IRC 2015-07-18T09:26:05 *** ankur has joined #rtems 2015-07-18T10:41:23 *** ankur has quit IRC 2015-07-18T12:07:41 *** mumptai has joined #rtems 2015-07-18T12:18:51 *** hesham has joined #rtems 2015-07-18T15:20:49 *** jkridner|work is now known as jkridner 2015-07-18T15:29:44 [git] 82f6941 by Jarielle Catbagan: [BBB: cpuio.c: Fixed indentation] http://devel.rtems.org/changeset/82f6941/umon 2015-07-18T15:29:45 [git] 90c2701 by Jarielle Catbagan: [BBB: am335x.h: Fix invalid macro names and add definitions/redefinitions for DDR PHY, Control Module, CM_PER, CM_WKUP, and EMIF0 registers] http://devel.rtems.org/changeset/90c2701/umon 2015-07-18T15:29:45 [git] 9f5d36e by Jarielle Catbagan: [BBB: config.h: Enable INCLUDE_MEMCMDS] http://devel.rtems.org/changeset/9f5d36e/umon 2015-07-18T15:29:45 [git] b9f0b9e by Jarielle Catbagan: [BBB: am335x.h/cpuio.c: Add CM_PER_GPIO1_CLKCTRL in am335x.h and change CM_DIV_M3_DPLL_PER to CM_PER_GPIO1_CLKCTRL in cpuio.c:initCPUio()] http://devel.rtems.org/changeset/b9f0b9e/umon 2015-07-18T15:29:45 [git] 273af8f by Jarielle Catbagan: [BBB: cpuio.c/rom_reset.S: Configure the MPU, Core, DDR, and Per PLLs and initialize the DDR3 memory] http://devel.rtems.org/changeset/273af8f/umon 2015-07-18T15:29:45 [git] 8a80d54 by Jarielle Catbagan: [BBB: rom_reset.S: Initialize the SP at the top of SRAM prior to invoking pll_init and ddr_init] http://devel.rtems.org/changeset/8a80d54/umon 2015-07-18T15:29:45 [git] d55aa7c by Jarielle Catbagan: [BBB: Makefile: Map uMon RAM in DDR3] http://devel.rtems.org/changeset/d55aa7c/umon 2015-07-18T15:29:46 [git] 01aa78a by Ed Sutter: [INCLUDE_MEMCMDS eliminates the need for INCLUDE_DM/PM] http://devel.rtems.org/changeset/01aa78a/umon 2015-07-18T15:29:46 [git] 723a2e2 by Ed Sutter: [fix uninitialized variable issue in mpu_pll_init()] http://devel.rtems.org/changeset/723a2e2/umon 2015-07-18T17:07:22 *** mumptai has quit IRC 2015-07-18T18:42:09 *** ketul has quit IRC 2015-07-18T21:54:06 *** gedare has joined #rtems 2015-07-18T21:54:06 *** ChanServ sets mode: +o gedare 2015-07-18T22:14:22 *** gedare has quit IRC 2015-07-18T22:16:17 *** hesham has quit IRC 2015-07-18T23:58:20 *** dr__house|away is now known as dr__house 2015-07-19T00:03:31 *** mumptai has joined #rtems 2015-07-19T01:20:30 *** igutek has quit IRC 2015-07-19T01:21:12 *** igutek has joined #rtems 2015-07-19T03:18:06 *** Davidbrcz has joined #rtems 2015-07-19T03:47:56 *** dr__house is now known as dr__house|away 2015-07-19T05:57:10 *** ketul has joined #rtems 2015-07-19T07:24:42 *** Davidbrcz has quit IRC 2015-07-19T07:28:18 *** jkridner has quit IRC 2015-07-19T07:28:58 *** jkridner has joined #rtems 2015-07-19T07:33:46 *** jkridner has quit IRC 2015-07-19T08:22:14 *** ketul has quit IRC 2015-07-19T12:45:24 *** jkridner has joined #rtems 2015-07-19T13:33:50 *** antgreen has joined #rtems 2015-07-19T16:29:15 *** hesham has joined #rtems 2015-07-19T17:34:58 *** Davidbrcz has joined #rtems 2015-07-19T17:42:24 *** Davidbrcz has quit IRC 2015-07-19T17:46:39 *** schoumi has joined #rtems 2015-07-19T17:48:51 *** schoumi is now known as Schoumi 2015-07-19T18:01:15 *** jkridner has quit IRC 2015-07-19T18:02:56 *** jkridner has joined #rtems 2015-07-19T18:09:18 *** dr__house|away is now known as dr__house 2015-07-19T18:59:07 *** ketul has joined #rtems 2015-07-19T20:57:11 *** ketul has quit IRC 2015-07-19T21:49:27 *** Davidbrcz has joined #rtems 2015-07-19T22:07:01 *** Davidbrcz has quit IRC 2015-07-19T22:30:11 *** kiwichris has quit IRC 2015-07-19T22:54:47 *** gedare has joined #rtems 2015-07-19T22:54:47 *** ChanServ sets mode: +o gedare 2015-07-19T23:01:09 *** gedare has quit IRC 2015-07-19T23:07:47 *** kiwichris has joined #rtems 2015-07-19T23:28:24 *** mumptai has quit IRC 2015-07-19T23:30:09 *** ankur has joined #rtems 2015-07-19T23:30:16 *** ankur has left #rtems