2016-07-18T00:06:49 *** gedare has joined #rtems 2016-07-18T00:06:49 *** ChanServ sets mode: +o gedare 2016-07-18T00:13:04 *** gedare has quit IRC 2016-07-18T00:19:53 *** Wonjun__ has joined #rtems 2016-07-18T00:23:07 *** Wonjun_ has quit IRC 2016-07-18T01:49:31 *** Wonjun__ has quit IRC 2016-07-18T01:50:21 *** Wonjun__ has joined #rtems 2016-07-18T04:25:18 *** mkhoory-3 has joined #rtems 2016-07-18T05:39:38 *** monstr has joined #rtems 2016-07-18T05:45:50 *** Davidbrcz has joined #rtems 2016-07-18T06:32:51 *** Davidbrcz has quit IRC 2016-07-18T06:47:04 *** Davidbrcz has joined #rtems 2016-07-18T10:07:45 *** Davidbrcz has quit IRC 2016-07-18T10:27:53 *** Wonjun_ has joined #rtems 2016-07-18T10:30:59 *** Wonjun__ has quit IRC 2016-07-18T10:54:15 *** monstr has quit IRC 2016-07-18T11:37:22 *** Davidbrcz has joined #rtems 2016-07-18T11:48:26 *** Davidbrcz has quit IRC 2016-07-18T12:23:24 [git] 55bde66 by Punit Vara: [beagle: pwm polishing . added a README to pwm . added select_pwmss() to select pwmss-generic registers, as opposed to PWM-specific registers . added pwmss_clock_en_status(), beagle_pwmss_is_running() and pwmss_tb_clock_check() . other API improvements . style improvements] 2016-07-18T12:23:25 http://devel.rtems.org/changeset/55bde66/rtems 2016-07-18T12:43:20 kiwichris: hi 2016-07-18T13:50:11 lo 2016-07-18T14:28:28 *** mkhoory-3 has quit IRC 2016-07-18T14:28:46 *** mkhoory-3 has joined #rtems 2016-07-18T16:35:52 *** dr__house|away is now known as dr__house 2016-07-18T17:04:46 *** Wonjun_ has quit IRC 2016-07-18T17:05:13 *** Wonjun_ has joined #rtems 2016-07-18T18:41:51 *** cdesai has quit IRC 2016-07-18T18:42:11 *** cdesai has joined #rtems 2016-07-18T18:46:15 *** cdesai has quit IRC 2016-07-18T18:46:35 *** cdesai has joined #rtems 2016-07-18T20:15:17 *** gedare has joined #rtems 2016-07-18T20:15:17 *** ChanServ sets mode: +o gedare 2016-07-18T20:15:24 *** gedare has quit IRC 2016-07-18T20:57:01 *** mumptai has joined #rtems 2016-07-18T21:01:53 *** gedare has joined #rtems 2016-07-18T21:01:54 *** ChanServ sets mode: +o gedare 2016-07-18T21:02:07 *** dr__house is now known as dr__house|away 2016-07-18T21:06:42 *** gedare has quit IRC 2016-07-18T22:12:56 [git] 9510012 by Chris Johns: [config: Fix telnetd configuration warning.] http://devel.rtems.org/changeset/9510012/rtems-libbsd 2016-07-18T22:13:27 beng-nl: hi, sorry I did not see you say hi yesterday 2016-07-18T22:56:11 :-) 2016-07-19T03:10:50 *** ita has quit IRC 2016-07-19T03:10:51 *** ita has joined #rtems 2016-07-19T05:22:13 *** sebhub has joined #rtems 2016-07-19T05:28:11 [trac] #2758/enhancement (new) updated by sebastian.huber (Please send patches to the devel@rtems.org mailing list.) http://devel.rtems.org/ticket/2758 2016-07-19T05:35:47 *** monstr has joined #rtems 2016-07-19T05:40:17 [trac] #2760/defect (v:4.10) created by sebastian.huber (AltiVec enable is borken on some PowerPC BSPs) http://devel.rtems.org/ticket/2760 2016-07-19T05:40:59 [trac] #2760/defect (closed) updated by Sebastian Huber (In [changeset:"c6c4fce942d52a9ea21069ef71a7ef0b454b91f2/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2760 2016-07-19T05:41:00 [git] c6c4fce by Sebastian Huber: [bsps/powerpc: Fix AltiVec enable There must be an isync after the mtmsr to ensure that the AltiVec is enabled for subsequent instructions. Close #2760.] http://devel.rtems.org/changeset/c6c4fce/rtems 2016-07-19T05:42:47 [trac] #2761/defect (v:4.10) created by sebastian.huber (AltiVec context switch is broken on some PowerPC BSPs) http://devel.rtems.org/ticket/2761 2016-07-19T05:43:05 [trac] #2760/defect (closed) updated by sebastian.huber (empty comment) http://devel.rtems.org/ticket/2760 2016-07-19T06:12:51 [trac] #2761/defect (closed) updated by Sebastian Huber (In [changeset:"8d785f72d9610fb80a65d7848404f0f7507e026c/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2761 2016-07-19T06:12:52 [git] 8d785f7 by Sebastian Huber: [bsps/powerpc: Fix AtliVec context switch Properly pass the stack aligned context to _CPU_Context_switch_altivec() since _CPU_altivec_ctxt_off defined via ppc_context. Close #2761.] http://devel.rtems.org/changeset/8d785f7/rtems 2016-07-19T06:15:15 [trac] #2761/defect (closed) updated by Sebastian Huber (In [changeset:"dc0f537b27ec5ca221ea67fde1d262c26ca3ec2e/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2761 2016-07-19T06:15:15 [git] dc0f537 by Sebastian Huber: [bsps/powerpc: Fix AtliVec context switch Properly pass the stack aligned context to _CPU_Context_switch_altivec() since _CPU_altivec_ctxt_off defined via ppc_context. Update #2761.] http://devel.rtems.org/changeset/dc0f537/rtems 2016-07-19T06:33:12 *** Davidbrcz has joined #rtems 2016-07-19T07:31:14 *** dspace has joined #rtems 2016-07-19T07:40:37 *** monstr has quit IRC 2016-07-19T07:42:20 [trac] #2762/defect (v:4.11) created by sebastian.huber (Shared PowerPC console driver broken due to Termios API changes) http://devel.rtems.org/ticket/2762 2016-07-19T07:43:49 [trac] #2762/defect (closed) updated by Sebastian Huber (In [changeset:"c2a4b8bf37de516468cf73e039f0d4b1b4921d84/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2762 2016-07-19T07:43:49 [git] c2a4b8b by Sebastian Huber: [bsps/powerpc: Fix shared console driver The Termios modes are now an emum. Do not use them in #if expressions. Close #2762.] http://devel.rtems.org/changeset/c2a4b8b/rtems 2016-07-19T07:44:32 [trac] #2762/defect (closed) updated by Sebastian Huber (In [changeset:"7b24946c2bd2abeba0da9983a07385271b4667b7/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2762 2016-07-19T07:44:33 [git] 7b24946 by Sebastian Huber: [bsps/powerpc: Fix shared console driver The Termios modes are now an emum. Do not use them in #if expressions. Update #2762.] http://devel.rtems.org/changeset/7b24946/rtems 2016-07-19T07:49:52 *** benjamint has quit IRC 2016-07-19T07:54:38 *** monstr has joined #rtems 2016-07-19T08:18:38 *** mumptai has quit IRC 2016-07-19T08:31:05 *** BBD has joined #rtems 2016-07-19T08:33:10 Good Morning everybody ! I'm pleased to be part of the discussion 2016-07-19T08:34:22 *** BBD has quit IRC 2016-07-19T08:34:40 *** BBD has joined #rtems 2016-07-19T08:36:27 *** Davidbrcz has quit IRC 2016-07-19T08:39:30 I'm a student interested with RTEMS and I might need the help of the community regarding my project: I am trying to enable the network interface with RTEMS and RTEMS-libbsd on the Beaglebone Black. To do 2016-07-19T08:49:30 To do so, I tried to add the cpsw driver from Texas Instruments from FreeBSD to RTEMS-libbsd. However, when I try to run some examples of the RTEMS-libbsd testsuite with u-boot, It seems that RTEMS do not configure the network interface successfully and is stuck in a boot loop. May I ask you to give me some leads on how to bypass this issue ? 2016-07-19T08:54:29 Please find below link of the log file I obtain after booting RTEMS with RTEMS-libbsd on the Beaglebone Black : 2016-07-19T08:54:31 http://textuploader.com/5e9vw 2016-07-19T08:56:57 *** Davidbrcz has joined #rtems 2016-07-19T10:32:33 *** BBD has quit IRC 2016-07-19T10:32:47 *** BBD has joined #rtems 2016-07-19T10:59:30 *** mkhoory-3 has quit IRC 2016-07-19T12:25:34 *** benjamint has joined #rtems 2016-07-19T13:00:51 *** benjamint has quit IRC 2016-07-19T13:10:16 *** benjamint has joined #rtems 2016-07-19T13:11:30 *** mkhoory-3 has joined #rtems 2016-07-19T13:37:40 *** Davidbrcz has quit IRC 2016-07-19T14:11:09 *** benjamint1 has joined #rtems 2016-07-19T14:13:40 *** benjamint has quit IRC 2016-07-19T14:13:40 *** benjamint1 is now known as benjamint 2016-07-19T14:14:53 *** Davidbrcz has joined #rtems 2016-07-19T14:22:18 *** gedare has joined #rtems 2016-07-19T14:22:18 *** ChanServ sets mode: +o gedare 2016-07-19T14:34:24 *** dspace has quit IRC 2016-07-19T15:01:54 *** m-ric has joined #rtems 2016-07-19T15:06:05 Hi rtemers, I run RTEMS-4.10.99 on a custom board. I transfer data over ethernet. My board can receive packets successfully (RX) but stalls sending packets (TX). I'd like to debug that issue, but have limited experience dbg network stack. I have a JTAG and can break into RTEMS code. What important info (mbuf, interrupt) should I check first to debug my problem? 2016-07-19T15:22:12 in particular, I read this https://docs.rtems.org/doc-current/share/rtems/html/networking/Debug-Output.html#Debug-Output but i don't see any printf in mbuf.h nor if.h 2016-07-19T15:36:51 hi m-ric 2016-07-19T15:37:04 hi gedare 2016-07-19T15:37:36 if you can you might want to update to the 4.11 release branch 2016-07-19T15:39:39 looks like you ran into some old doc. you should consider inquiring about this on the users@rtems.org mailing list (subscribe first) 2016-07-19T15:54:35 *** dr__house|away is now known as dr__house 2016-07-19T16:02:09 gedare, updating to 4.11 requires a lots of custom changes to support the board I have. It is too much work as of now. Ok, I'll send an email to @users. 2016-07-19T16:03:03 gedare, anyhow do you know what debug options I can enable in the network stack of 4.10.99? 2016-07-19T16:04:28 *** BBD has quit IRC 2016-07-19T16:07:39 *** Davidbrcz has quit IRC 2016-07-19T16:39:29 *** mumptai has joined #rtems 2016-07-19T18:02:29 is there any difference between printk and printf in RTEMS? 2016-07-19T18:36:49 *** Davidbrcz has joined #rtems 2016-07-19T19:02:52 *** Davidbrcz has quit IRC 2016-07-19T19:04:41 *** monstr has quit IRC 2016-07-19T21:19:54 *** m-ric has quit IRC 2016-07-19T23:39:02 *** dr__house is now known as dr__house|away 2016-07-19T23:40:00 *** hesham has joined #rtems 2016-07-20T04:26:39 *** gedare has quit IRC 2016-07-20T04:49:28 *** gedare has joined #rtems 2016-07-20T04:49:29 *** ChanServ sets mode: +o gedare 2016-07-20T05:41:36 *** monstr has joined #rtems 2016-07-20T06:22:33 *** Davidbrcz has joined #rtems 2016-07-20T06:24:52 *** dspace has joined #rtems 2016-07-20T06:43:39 *** Davidbrcz has quit IRC 2016-07-20T06:57:47 *** Davidbrcz has joined #rtems 2016-07-20T07:08:34 *** Davidbrcz has quit IRC 2016-07-20T07:49:20 *** mumptai has quit IRC 2016-07-20T07:58:02 *** Davidbrcz has joined #rtems 2016-07-20T08:21:27 *** Davidbrcz has quit IRC 2016-07-20T08:35:44 *** Davidbrcz has joined #rtems 2016-07-20T11:50:52 *** darnir has quit IRC 2016-07-20T11:53:37 *** m-ric has joined #rtems 2016-07-20T11:56:41 *** darnir has joined #rtems 2016-07-20T12:08:06 [trac] #2763/defect (v:4.12) created by vivekkukreja (Incorrect exit if condition for capture_flags_global in cpukit/libmisc/capture/capture.c) http://devel.rtems.org/ticket/2763 2016-07-20T12:08:27 *** dspace has quit IRC 2016-07-20T12:09:53 *** Davidbrcz has quit IRC 2016-07-20T12:16:51 *** idark has joined #rtems 2016-07-20T12:36:31 *** idark has quit IRC 2016-07-20T12:40:34 *** idark has joined #rtems 2016-07-20T13:08:16 *** Davidbrcz has joined #rtems 2016-07-20T13:42:29 *** deval has joined #rtems 2016-07-20T13:45:48 *** _sambeet has joined #rtems 2016-07-20T13:46:32 *** idark has left #rtems 2016-07-20T13:57:27 good morning 2016-07-20T13:57:38 *** salpha has joined #rtems 2016-07-20T13:57:40 *** mudit1729 has joined #rtems 2016-07-20T13:57:45 good morning ! 2016-07-20T13:57:53 <_sambeet> Good morning 2016-07-20T13:57:53 hi! 2016-07-20T13:58:18 *** punitvara has joined #rtems 2016-07-20T13:59:00 Hi 2016-07-20T13:59:25 Good morning 2016-07-20T13:59:51 *** vivekk has joined #rtems 2016-07-20T14:00:08 we'll start with salpha today, go ahead salpha 2016-07-20T14:00:38 This week i ran a test suite with 630 test cases. A total 44 test cases crashed, accounting for 8 functions. 2016-07-20T14:00:54 Slingshot proceeded to the next test case when a test case was crashed so the test suite execution ran automatically. However, current approach on automatization is not scalable and i’ll update the blog regarding that. 2016-07-20T14:01:05 I went through the evaluation report for rtems 4.5.0 robustness (and stress) testing, which was done by Critical Software. I couldn’t find how they dealt with test cases that crash the system but their approach is also somewhat different so might not be very useful in my case. But the interesting point is that their result can be used as a base for 2016-07-20T14:01:06 comparison on how slingshot performs. 2016-07-20T14:01:28 Next i’ll start changing test suite template from the current static model to the dynamic one via libdl. 2016-07-20T14:02:35 ok.. 2016-07-20T14:02:54 sorry i have to look at your project plan quick, been awhile since we chatted 2016-07-20T14:03:04 *** DrJoel has joined #rtems 2016-07-20T14:03:04 *** DrJoel has joined #rtems 2016-07-20T14:03:04 *** ChanServ sets mode: +o DrJoel 2016-07-20T14:03:22 where do you sit with respect to your midterm goals? 2016-07-20T14:03:51 according to the proposal, what i estimated for the whole period is over now. 2016-07-20T14:03:58 ah, ok 2016-07-20T14:03:59 * DrJoel waves to gedare.. just made it to the office .. but barely. :) 2016-07-20T14:04:12 therefore i've started extra tasks proposed. 2016-07-20T14:04:13 hey joel, you can chelax 2016-07-20T14:04:16 ok good 2016-07-20T14:04:28 i'm glad, i thought your plan was a bit unambitious in the proposal. :) 2016-07-20T14:04:32 it was quick! :) 2016-07-20T14:04:48 no, a lot in my mind to do. 2016-07-20T14:05:03 ok sounds good, we need to make sure to go over your code to see what needs to be published 2016-07-20T14:05:08 slingshot should become way more better. 2016-07-20T14:05:36 i've written vagrant scripts so it should be easier to re-produce 2016-07-20T14:05:50 you can find them in the 'dev' branch of the repo 2016-07-20T14:05:51 agreeing with gedare.. get what is useful published and reproduced. 2016-07-20T14:06:12 ok. well i'm not great at "finding" things. much better at following directions. :) 2016-07-20T14:06:13 you mean the results or test suites? 2016-07-20T14:06:25 you had to create the standalone slingshot 2016-07-20T14:06:38 yeah that's done 2016-07-20T14:06:41 i want to understand if that is something that needs to be released and maintained 2016-07-20T14:06:47 and how that project will persist 2016-07-20T14:06:56 or is it just a code dump on github 2016-07-20T14:06:57 i see 2016-07-20T14:07:22 i feel a little offended. what do you mean by code dump? 2016-07-20T14:07:40 code dump means someone puts code in a github repository, and it just sits there 2016-07-20T14:07:41 of course it's stand-alone and usable now. 2016-07-20T14:07:58 a github repository is not the same thing as a maintained project, that is what is meant 2016-07-20T14:08:04 ok we can talk about this later 2016-07-20T14:08:05 so start to make a plan for how to create "deliverable products" 2016-07-20T14:08:18 from now on, it depends on the community. and i have to produce some interesting results/reports to motivate people look at slingshot 2016-07-20T14:08:25 yes exactly 2016-07-20T14:08:41 okay 2016-07-20T14:08:45 ok thank you, i look forward to your next blog update 2016-07-20T14:09:01 deval: you're up now 2016-07-20T14:09:12 hello 2016-07-20T14:09:14 Latest updates on my projects are in this email. https://lists.rtems.org/pipermail/devel/2016-July/015609.html 2016-07-20T14:09:20 I have sent this mail in order to get the help from community (that would speed up my project, if someone already knowes the ans) and parallely I am also looking to the questions raised after those observations. 2016-07-20T14:09:25 As far as I can uderstand, the problem can be in two places. Interrupt handler or mailbox because that handles power functions. So I am looking in that direction. 2016-07-20T14:10:12 Are you gettting the interrupt? 2016-07-20T14:10:40 What would be the way to certainly know that ? 2016-07-20T14:11:09 Simplest is a printk in the IRQ handler. Even a single character.. progress report 2016-07-20T14:11:19 If you don't get it, then you know it was in the interrupt setup 2016-07-20T14:11:39 If you don't see but one, then you know you didn't acknowledge it correctly or something 2016-07-20T14:12:54 deval: you should also create a new mailing list post for new problems 2016-07-20T14:13:11 Okay I would look in there. Today 2016-07-20T14:13:23 when you reply to a thread with a new topic, often your question will be "lost" in the rest of the discussion 2016-07-20T14:13:46 +1 focused email threads work best. 2016-07-20T14:14:09 okay. I will make a new one after the interrupt observation today 2016-07-20T14:14:19 will keep that in mind 2016-07-20T14:14:37 i would guess that the USB is being re-initialized or disabled during the libbsd initialization, and not getting turned back on. i don't have any idea how to read the freebsd boot log so I can't help with that. 2016-07-20T14:15:33 yes you got that correct. 2016-07-20T14:15:37 you should post the boot log you get with your test suite, too 2016-07-20T14:15:38 I will keep digging. 2016-07-20T14:15:45 for comparison purposes 2016-07-20T14:15:54 alright. thanks deval 2016-07-20T14:16:10 _sambeet: you can go now 2016-07-20T14:16:18 yes I have marked the lines which are in my testsuit with "*" mark before them 2016-07-20T14:16:22 <_sambeet> Yes sir 2016-07-20T14:17:00 <_sambeet> This week I worked on running a component on rtt.To test a file on rtt apart from Orogen is also required but cross compiling orogen will delay the RSB package for RTT.So Thomas suggested I write the component by hand and compile it using rtems i386 gcc.So I am now compiling the component to test it.I had configured rtt for static building but boost is giving serialization problems so I am working to make it work. 2016-07-20T14:17:20 Thank you. 2016-07-20T14:17:48 deval: that's a bit hard to comprehend. it's much simpler to compare two outputs one after the other, i think. 2016-07-20T14:19:20 ok _sambeet, not sure why you should get any kind of problem like that due to static build. boost is headers only isn't it? 2016-07-20T14:19:30 *** salpha has quit IRC 2016-07-20T14:19:57 sure I will take care of that in next update. 2016-07-20T14:20:16 *** Davidbrcz has quit IRC 2016-07-20T14:21:54 _sambeet: are you starting on the RSB part of the project? And, did you have to patch upstream projects that you should submit patches to them? 2016-07-20T14:22:28 <_sambeet> No sir it's not due to static building.I might have made some code changes for static building and I am getting problems because of that. 2016-07-20T14:22:42 brb.. badly need to reboot 2016-07-20T14:22:47 *** DrJoel has quit IRC 2016-07-20T14:23:01 ah, ok 2016-07-20T14:23:02 <_sambeet> The testing of Rtt is not yet over. 2016-07-20T14:23:21 <_sambeet> I should finish it tomorrow. 2016-07-20T14:23:57 ok. provide an update on your blog when you finish that 2016-07-20T14:24:20 *** deval has quit IRC 2016-07-20T14:24:36 <_sambeet> Then I will work on RSB recipe.Yes sir, I am waiting only because I have not got enough results. 2016-07-20T14:24:40 *** DrJoel has joined #rtems 2016-07-20T14:24:40 *** DrJoel has joined #rtems 2016-07-20T14:24:40 *** ChanServ sets mode: +o DrJoel 2016-07-20T14:24:51 and you should start to identify the code changes you should submit to other projects 2016-07-20T14:25:14 ok you should start to study how to do the RSB part though 2016-07-20T14:25:41 <_sambeet> I am doing that now sir. 2016-07-20T14:26:17 ok good, thanks _sambeet 2016-07-20T14:26:32 mudit1729: you can go now 2016-07-20T14:26:38 <_sambeet> Thank you sir 2016-07-20T14:26:43 I am writing the DMA library for the mainline using the logic defined in the DMA driver in the FreeBSD. I will most likely be sharing the code in two days mostly. Previously, the question was also which devices would be used. For testing, I am thinking of using MCP23008 for I2C and microchip 23k256 for SPI. The drivers have already been written for the devices by asuol for testing purposes during his time. The DMA API will be us 2016-07-20T14:26:50 For the SD card driver, I think I got to know where the issue was occurring, The previous driver checks for the Capababilities register in the SDHCI controller and then accordingly starts using DMA. I am forcefully disabling DMA which i think is the root cause and running them in normal/PIO mode. 2016-07-20T14:26:58 With these understandings, I dont think I am blocked on anything presently. Just need to speed up my work. 2016-07-20T14:27:32 The DMA API will be used at the SPI/I2C bus level and mostly I am expecting that the driver wont require any major change. [ sentence got cut 2016-07-20T14:28:10 ok seems reasonable, happy hacking. :) 2016-07-20T14:28:17 thanks 2016-07-20T14:28:28 *** deval has joined #rtems 2016-07-20T14:28:32 Wonjun_: you can go now 2016-07-20T14:28:42 hello 2016-07-20T14:28:49 This week I wrote a config file for loading rtems. as I sent message by email,I can load rtems and a loader file now. 2016-07-20T14:28:55 but there are still some error about executing rtems as inmate. 2016-07-20T14:29:02 It was caused by pci and pio configuration. I showed my config files to Maxim. He said that he found nothing strange in my cell configuration. 2016-07-20T14:29:17 He also gave me example files about hpet and ioapic. even though I followed the configurations I got a error about invaild ioapic read. 2016-07-20T14:29:35 I thought I need to check pio and pci devices much more deeply. I will try more about that. 2016-07-20T14:30:11 Where did you send that email? 2016-07-20T14:30:13 *** deval has left #rtems 2016-07-20T14:30:50 *** _sambeet has quit IRC 2016-07-20T14:31:03 I sent to Maxim he worked on jailhouse for his bachelor paper 2016-07-20T14:31:08 hmm. is it a problem with RTEMS trying to use some pci/pio? 2016-07-20T14:31:09 ah, ok 2016-07-20T14:31:40 it is a problem with jailhouse configuation. 2016-07-20T14:32:48 I also asked Jan some documents about configuration but there ism 2016-07-20T14:33:09 there isn't document about configuation. 2016-07-20T14:33:46 ok, i guess you were not able to get the configuration he used before? 2016-07-20T14:33:58 and he said that the student who worked on jailhouse - rtems before is out of reach 2016-07-20T14:34:08 i see 2016-07-20T14:34:30 (this is why I stress so much about writing good blog posts and pushing out your code!) 2016-07-20T14:34:33 I couldn't get a used file from jan 2016-07-20T14:34:37 *** deval_ has joined #rtems 2016-07-20T14:36:20 ok, while you work on this, i also would like you to try to prepare a "cleaner" version of your patched RTEMS 2016-07-20T14:36:56 because the version you created uses a single patch that combines all the patches Jan had sent you, plus whatever other changes you maybe had made, plus you also git-add'ed a bunch of library files and other auto-generated files that shouldn't be included 2016-07-20T14:37:38 also, you should be using the rtems-4.12 tools if you are working from the current or recent 'master' branch 2016-07-20T14:37:54 if you work from the 4.11 branch, then you should use the rtems-4.11 tools. 2016-07-20T14:38:39 to make the "cleaner" version, you should take either 4.11 branch, or current master branch, and apply the patches from Jan, then make corrections that you need to get them merged or able to compile. 2016-07-20T14:38:41 ok Wonjun_ ? 2016-07-20T14:38:57 can I check again? you mean 2016-07-20T14:40:19 I should prepare cleaner version like I should use only one patch file not all the patches. 2016-07-20T14:40:31 Wonjun_ RTEMS will need legacy support on the PIC. That may be the root of the issue 2016-07-20T14:40:37 no, like you should try to keep the patches that Jan had sent you as separate commits 2016-07-20T14:40:54 and only commit changes you make to source in the rtems tree 2016-07-20T14:41:04 and don't git-add other files like binary and library files 2016-07-20T14:41:31 Wonjun_: I recommend highly switching to the master. the pc386 BSP has had a LOT of work and it is likely to be very painful for anyone else to update the work. 2016-07-20T14:41:38 also, i'm not sure why you have these binary and library files in your repository at all. they should be in a separate directory where you are building RTEMS, so there may be some problem with the workflow there. 2016-07-20T14:41:47 I worry it won't ever get merged if you don't update it and get it merged into RTEMS soon 2016-07-20T14:42:01 *** Davidbrcz has joined #rtems 2016-07-20T14:42:03 i think we just need to make sure to get you on a good fresh path 2016-07-20T14:42:46 so the first thing I want you to do is... 2016-07-20T14:43:03 update your master branch to the current HEAD 2016-07-20T14:43:07 (by git pull) 2016-07-20T14:43:14 and then create your jailhouse branch from the master 2016-07-20T14:43:22 rebuild your 4.12 compiler tools using the RSB 2016-07-20T14:43:34 and build/run the pc386 BSP on qemu without making any changes at all 2016-07-20T14:43:46 that should give confidence you have the good stable baseline for development 2016-07-20T14:44:09 then you should apply the patches Jan had sent you, and fix the merge-conflicts for each one as you go 2016-07-20T14:44:37 the patches that he sent you were properly made with git, so you should use git-am to apply them 2016-07-20T14:44:49 that will retain the original code authorship and commit message information 2016-07-20T14:45:06 probably there will be some conflicts since the pc386 bsp underwent a lot of changes 2016-07-20T14:45:24 if you have trouble with the merging you should send out email to the devel mailing list, and you can ask me, Joel, and Chris Johns directly as well 2016-07-20T14:45:30 yes probably.. 2016-07-20T14:45:37 Ok 2016-07-20T14:45:42 Getting it to run from master with no changes may require changes to your boot args on the kernel command. Ask on the devel@ list 2016-07-20T14:46:06 Yeah.. lots of changes to support some non-legacy features but no one has gotten to the APIC yet 2016-07-20T14:46:36 Firt i will create my jailhouse branch from the master then apply the patches 2016-07-20T14:46:44 alright Wonjun_ send an email with your status updates to Joel tomorrow and Friday regarding building the current master and then about merging Jan's patches. 2016-07-20T14:46:58 correct, but before you do that, you should make sure you can successfully compile and run the master 2016-07-20T14:47:12 to give you confidence in your compiler tools and qemu version 2016-07-20T14:47:36 what dose it mean 'run'? 2016-07-20T14:47:55 execute the ticker.exe sample 2016-07-20T14:47:56 where should i run rtems? qemu? 2016-07-20T14:48:00 I see 2016-07-20T14:48:03 within qemu 2016-07-20T14:48:30 https://devel.rtems.org/wiki/Developer/Simulators/QEMU 2016-07-20T14:48:39 there are a couple different options to try 2016-07-20T14:49:21 ok thanks Wonjun_, you should build your 4.12 tools and compile master for the i386/pc386 BSP and run ticker.exe without any trouble 2016-07-20T14:49:41 if you run into problems you should definitely ask, since that should "just work". 2016-07-20T14:50:15 after that you will create your jailhouse branch and try to get the patches from Jan to merge into RTEMS, which will probably require you to fix some merge conflicts. 2016-07-20T14:50:32 Ok I will sen an email to Dr Jeol with my status updates about 20 hours after. 2016-07-20T14:51:16 ok thanks Wonjun_ 2016-07-20T14:51:25 Thank you for your advices. 2016-07-20T14:51:34 punitvara: you can go next. hi beng-nl 2016-07-20T14:51:49 "Just work" but --console now should be --console=/dev/XXX not --console=XXX as before. Addressing that should be enough 2016-07-20T14:52:31 I will check that. 2016-07-20T14:53:30 punitvara: are you here? 2016-07-20T14:53:33 vivekk: hi 2016-07-20T14:53:50 Hi Gedare, I could not put in many hours this week since I was ill. I opened a ticket for the capture engine bug and i'm understanding the commit process now. 2016-07-20T14:54:03 I've created a branch on my forked repo and pushed my changes for CTF tracing. Its in progress. 2016-07-20T14:54:18 I'm looking for ways to transport trace from target to host. For this, I built libbsd for xilinx_zynq_a9_qemu bsp and contacted Isaac today for help with setting up a TFTP server. I'm lookin into samples now. Once we have trace on host, I can start testing immediately. 2016-07-20T14:54:19 [trac] Developer/Simulators/QEMU edited by gedare (Fix link to CANEmulation page.) http://devel.rtems.org/wiki/Developer/Simulators/QEMU 2016-07-20T14:54:39 I'll also start pushing my changes for Capture Engine this week. 2016-07-20T14:55:11 ok. any particular reason for using zynq_qemu? 2016-07-20T14:56:22 No i built libbsd for sparc/sis too. I was advised to use this bsp early on for better testing 2016-07-20T14:57:13 sis doesn't have any network or external connections so you have to use somethign like pc386 or zynq which has a network interface simulated 2016-07-20T14:57:20 correct 2016-07-20T14:57:41 i don't know if anyone uses libbsd with sis? 2016-07-20T14:58:03 regardless, i get it now 2016-07-20T14:58:08 I may have compiled and done some initial debug but it isn't worth the effort. 2016-07-20T14:58:17 libbsd is probably best used with zynq 2016-07-20T14:58:50 *** dspace has joined #rtems 2016-07-20T14:58:58 but you need to double-check with chris if that is the best platform to use for this, or if he might want you to use something else, like pc386/libbsd or even pc386 with the in-tree network stack 2016-07-20T14:59:04 I just followed the libbsd documentation for all bsps i installed. Im trying to run a TFTP server on qemu for arm 2016-07-20T14:59:13 OK ill write a mail now 2016-07-20T14:59:53 ok thanks vivekk. if you get stalled on the networking aspect, you should discuss alternative methods for transporting the trace data 2016-07-20T15:00:21 because the transport aspect is somewhat uninteresting compared with the other coding challenges you face 2016-07-20T15:00:43 Sure I'll write a mail by tomorrow on devel. Thanks 2016-07-20T15:00:54 ok thanks vivekk 2016-07-20T15:01:00 darnir: are you here today? 2016-07-20T15:01:05 punitvara: are you here yet? 2016-07-20T15:01:11 yes 2016-07-20T15:01:16 hi punitvara beng-nl 2016-07-20T15:01:21 go ahead punitvara 2016-07-20T15:01:26 Last week as you suggested me to use /dev/i2c framework I asked opinion of commu nity on ml. Sebastian suggested to use /dev/i2c and also suggested i2c drivers w ith same framework. I have read those drivers and started coding for I2C as per /dev/i2c way. For first time Ben suggested me to create node with I2C0 bus then I can work further. I am working on it 2016-07-20T15:02:14 Next week I will try to complete whatever Ben suggested 2016-07-20T15:02:37 ok 2016-07-20T15:02:44 you have i2c devices for testing? 2016-07-20T15:04:39 alright, thanks punitvara 2016-07-20T15:04:44 did i miss anyone? 2016-07-20T15:05:08 EEPROM is on BBB board it self 2016-07-20T15:05:18 I have also purchased BM085 2016-07-20T15:05:19 ah, good 2016-07-20T15:05:46 But Ben suggested me to use EEPROM. But first I have to test small codes then I have to go further as Ben suggeste 2016-07-20T15:05:49 *suggested 2016-07-20T15:05:58 alright, sounds good 2016-07-20T15:06:56 [trac] SOCIS/2016 edited by salpha (empty comment) http://devel.rtems.org/wiki/SOCIS/2016 2016-07-20T15:07:34 [trac] SOCIS/2016 edited by salpha (empty comment) http://devel.rtems.org/wiki/SOCIS/2016 2016-07-20T15:08:11 good meeting everyone please remember to update your tracking page status entries 2016-07-20T15:08:54 [trac] SOCIS/2016 edited by salpha (empty comment) http://devel.rtems.org/wiki/SOCIS/2016 2016-07-20T15:11:32 *** DrJoel has quit IRC 2016-07-20T15:13:24 *** sebhub has quit IRC 2016-07-20T15:14:53 *** bbd has joined #rtems 2016-07-20T15:15:53 Hello everybody, pleased to be part of this discussion 2016-07-20T15:17:33 I am currently studying the Real-Time Operating System RTEMS in which I try to enqble the network interface. To do so, I followed several parts of your tutorial with success but notwithstanding I have an issue : In fact, I tried to port the cpsw driver from FreeBSD to libbsd and monitor the behaviour with the examples of the testsuite. However, despite the modifications made on the files, it seems that the network interface is not atta 2016-07-20T15:18:32 May I ask you some leads to help me to bypass this issue ? 2016-07-20T15:19:32 Please find below the link of the log file generated during the boot process of RTEMS on the Beaglebone Black : 2016-07-20T15:19:43 http://textuploader.com/5e9vw 2016-07-20T15:20:07 NB: I am using the waf version of rtems-libbsd and RTEMS v4.12 Kind Regards !!! 2016-07-20T15:24:38 hi bbd, "the network interface is not att", what do you mean? 2016-07-20T15:25:30 I mean attached 2016-07-20T15:26:26 It seems that the PHY register has some problems but I am not able to understand where 2016-07-20T15:27:14 did you check how the autonegotiation ended up b/w your PHY and MAC? 2016-07-20T15:29:02 Sorry I am not that familiar with this autonegiciation tool, can you give me a quick explanation on how it works ? 2016-07-20T15:30:12 https://en.wikipedia.org/wiki/Autonegotiation 2016-07-20T15:32:22 your MAC driver should have some debug info about the autoneg procedure. Debug that and make sure PHY and MAC agreed. If 1000Mb is negociated but does not work, try to direct/hard-code the autoneg to be 100Mb first 2016-07-20T15:33:01 e.g. in rtems/.../greth.c I can enable some DEBUG macros to get autoneg printf'ed 2016-07-20T15:37:41 Thanks for the hint. One question: where can I hard code the negiciaion size to run some tests with different values ? 2016-07-20T15:38:00 *** deval_ has quit IRC 2016-07-20T15:39:33 *** gedare has quit IRC 2016-07-20T15:39:56 in your MAC driver code. For what platform do you build RTEMS? 2016-07-20T15:40:24 what's your platform ethernet MAC chip? Do you know the manufacturer and device name? 2016-07-20T15:40:37 Beaglebone Black 2016-07-20T15:41:54 *** vivekk has quit IRC 2016-07-20T15:41:58 check the doc available on the web and find its MAC and PHY devices. Look in RTEMS source to find the driver file corresponding to this device, and read the code. 2016-07-20T15:42:28 *** Davidbrcz has quit IRC 2016-07-20T15:43:08 Isn't the BBB NIC supported out of the box ? 2016-07-20T15:43:19 in RTEMS? 2016-07-20T15:43:24 yes, 2016-07-20T15:44:01 BBB is community-based, "supported" does not mean fully working 2016-07-20T15:44:29 according to different websites, it seems the guys successfully used the network interface. 2016-07-20T15:45:12 ok, then bbd should contact him 2016-07-20T15:45:43 Ok. we tried, still waiting for answer. Tnx for you support. 2016-07-20T15:46:59 Our goal was to use RTEMS and add IoT protocol (MQTT) features to it. But this requires a working network interface. 2016-07-20T15:47:32 welcome. Ok, interesting project! :-) 2016-07-20T15:48:13 yes. indeed. bbd is master student in my team. 2016-07-20T15:49:05 I need to leave the chat now. I'll online again tomorrow. cheers 2016-07-20T15:51:15 *** dspace has quit IRC 2016-07-20T15:52:37 So basically, how can I use the greth.c program to check PHY and MAC ? 2016-07-20T15:54:47 *** monstr has quit IRC 2016-07-20T15:56:58 greth.c is my ethernet device, it is probably the one used on BBB 2016-07-20T15:57:09 it is probably not* 2016-07-20T15:57:46 you have to _find_ the BBB net device first 2016-07-20T15:57:53 and find its driver source code. 2016-07-20T15:57:57 in RTEMS 2016-07-20T16:02:50 The BBB is MII-friendly, thus I downloaded the cpsw driver from Texas Instruments for BBB on FreeBSD 2016-07-20T16:03:08 *The BBB network interface 2016-07-20T16:13:06 *** punitvara has quit IRC 2016-07-20T16:21:07 OK it seems that the BBB has no network driver in the compiled BSP folder (beaglebone/arm-rtems4.12/c/beagleboneblack/libchip/network) 2016-07-20T16:23:00 but uses SMC LAN8710A PHY 2016-07-20T16:32:55 *** mumptai has joined #rtems 2016-07-20T16:33:28 bbd, then you lack a MAC driver in your RTEMS-BBB 2016-07-20T16:35:00 is memalign the proper call to alloc aligned data in RTEMS-4.10.99 or is there any more recent syscalls? 2016-07-20T16:37:27 I don't know, I started with the version 4.12 2016-07-20T16:59:02 *** bbd has quit IRC 2016-07-20T17:03:45 *** benjamint1 has joined #rtems 2016-07-20T17:03:54 *** benjamint has quit IRC 2016-07-20T17:05:58 *** benjamint has joined #rtems 2016-07-20T17:08:27 *** benjamint1 has quit IRC 2016-07-20T17:18:55 *** mumptai has quit IRC 2016-07-20T17:29:34 *** benjamint has quit IRC 2016-07-20T17:31:57 *** dr__house|away is now known as dr__house 2016-07-20T17:37:28 *** benjamint has joined #rtems 2016-07-20T18:03:18 *** benjamint1 has joined #rtems 2016-07-20T18:03:40 *** benjamint has quit IRC 2016-07-20T18:04:30 *** benjamint has joined #rtems 2016-07-20T18:07:31 *** benjamint1 has quit IRC 2016-07-20T18:49:03 *** m-ric has quit IRC 2016-07-20T19:59:49 *** ita is now known as itazzz 2016-07-20T20:17:07 [trac] libdl-strtab-test.tar.gz () deleted from #2754 by pggauvin http://devel.rtems.org/ticket/2754 2016-07-20T20:17:08 [trac] libdl-strtab-test.tar.gz (fixed linking and made sure all symbols resolved) attached to #2754 by pggauvin http://devel.rtems.org/ticket/2754 2016-07-20T20:17:38 [trac] rtems-libdl-obj-cache-disable.patch (workaround for this issue) attached to #2754 by pggauvin http://devel.rtems.org/ticket/2754 2016-07-20T20:26:35 [trac] #2754/defect (new) updated by pggauvin (I've narrowed down the problem to a corrupted section name for `.strtab` being ...) http://devel.rtems.org/ticket/2754 2016-07-20T20:34:15 *** mumptai has joined #rtems 2016-07-20T20:42:13 *** m-ric has joined #rtems 2016-07-20T20:48:13 *** dspace has joined #rtems 2016-07-20T20:50:58 *** dspace has quit IRC 2016-07-20T20:51:46 *** dspace has joined #rtems 2016-07-20T20:56:36 *** dspace has quit IRC 2016-07-20T22:41:59 *** mumptai has quit IRC 2016-07-21T00:08:49 *** dr__house is now known as dr__house|away 2016-07-21T00:40:29 [trac] #2438/defect (closed) updated by Pavel Pisa (In [changeset:"8709aa0459217822ce4118416c8547ed08d96ac1/rtems"]: {{{ ...) http://devel.rtems.org/ticket/2438 2016-07-21T00:40:30 [git] a48c052 by Pavel Pisa: [arm/raspberrypi: cache manager can be used for mailbox synchronization now. Remove workarounds. Signed-off-by: Pavel Pisa ] http://devel.rtems.org/changeset/a48c052/rtems 2016-07-21T00:40:30 [git] 3338121 by Pavel Pisa: [bsps/arm: do not disable MMU during translation table management operations. Disabling MMU requires complex cache flushing and invalidation operations. There is almost no way how to do that right on SMP system without stopping all other CPUs. On the other hand, there is documented sequence of operations which should be used 2016-07-21T00:40:30 according to ARM manual and it g...] http://devel.rtems.org/changeset/3338121/rtems 2016-07-21T00:40:30 [git] 577e7fb by Pavel Pisa: [arm/bsps: CP15 and basic cache support entire cache clean for more architecture variants now. Next cache operations should work on most of cores now rtems_cache_flush_entire_data() rtems_cache_invalidate_entire_data() rtems_cache_invalidate_entire_instruction() Instruction cache invalidate works on the first level for now 2016-07-21T00:40:30 only. Data cacache operati...] http://devel.rtems.org/changeset/577e7fb/rtems 2016-07-21T00:40:30 [git] bef2b73 by Pavel Pisa: [arm/raspberrypi: use cache manager operations to flush/invalidate all cache levels. This fix strange behavior where some stale content has been stored in level 2 cache before RTEMS has been start from U-boot which has reappeared after MMU enable and shadow vector table at start of SDRAM.] 2016-07-21T00:40:31 http://devel.rtems.org/changeset/bef2b73/rtems 2016-07-21T00:40:31 [git] dcf806e by Pavel Pisa: [arm/raspberrypi: reorder and update MMU config table to nor force RW section later to RO. Enable even the first megabyte of SDRAM to be cache-able after problems with stale cache content has been resolved by previous commit. Because major part of application usually fits to the first megabyte this speedups test dhrystone 2016-07-21T00:40:32 application by factor 40.] http://devel.rtems.org/changeset/dcf806e/rtems 2016-07-21T00:40:32 [git] 8709aa0 by Pavel Pisa: [libdl/rtl-obj.c: synchronize cache after code relocation. Memory content changes caused by relocation has to be propagated to memory/cache level which is used/snooped during instruction cache fill. Closes #2438] http://devel.rtems.org/changeset/8709aa0/rtems 2016-07-21T03:25:52 reading the network stack in 4.11, in rtems_glue.c I have rtems_bsdnet_semaphore_release(); rtems_event_system_receive (event_in, option_set, ticks, event_out); rtems_bsdnet_semaphore_obtain(); 2016-07-21T03:26:38 wouldn't it be the contrary? sem_obtain(); ... ; sem_release() ? 2016-07-21T03:26:47 m-ric: hi 2016-07-21T03:27:04 kiwichris, hi 2016-07-21T03:27:43 m-ric: I do not follow the question or issue you are seeing. 2016-07-21T03:28:37 long story short: I have a custom Leon4 with greth MAC running RTEMS-4.11.99 2016-07-21T03:28:50 OK 2016-07-21T03:28:59 my network stalls in RX/TX randomly 2016-07-21T03:29:24 ping traffic or something else? 2016-07-21T03:29:41 recv(4MB) 2016-07-21T03:29:51 data or bit rate? 2016-07-21T03:30:19 sometimes it can issue 50x recv(4MB); send(16MB); sometime just stalling after few KB recv'd 2016-07-21T03:30:45 is the greth a scatter/gather DMA MAC? 2016-07-21T03:30:56 single core? 2016-07-21T03:31:32 yes scatter gather MAC, but only in TX AFAIRead the doc 2016-07-21T03:31:40 yes single core 2016-07-21T03:31:53 SO full size rx buffers chained? 2016-07-21T03:32:56 yes, I can have 64 8-byte recv descriptors 2016-07-21T03:33:06 Great 2016-07-21T03:33:10 1024B-aligned 2016-07-21T03:33:31 When it stalls can you walk the descriptor table and are they all full, partially full, or all empty? 2016-07-21T03:34:51 partially full 2016-07-21T03:34:56 good 2016-07-21T03:35:09 Is the PHY always reporting a solid link? 2016-07-21T03:36:06 don't know how to check that. I access it thru i2c 2016-07-21T03:36:26 got to see how to read it from my debugger 2016-07-21T03:36:43 interesting they are normally from the MAC via the MII interface. 2016-07-21T03:37:01 Do you have leds on the connector or somewhere? 2016-07-21T03:40:22 yes, I got led, but right I am remotely debugging frome home :-) 2016-07-21T03:40:49 right now * 2016-07-21T03:40:59 Ah 2016-07-21T03:41:07 The stall could be one of ... 2016-07-21T03:41:28 1. link drops out; this can happen if some termination is not 100% and with load something trips. 2016-07-21T03:41:49 2. level or edge interrupts not configured correctly 2016-07-21T03:42:00 3. edge interrupt ack not correct 2016-07-21T03:42:25 4. descriptor hand back to the hardware not handled correctly and the driver and hardware get out of step 2016-07-21T03:42:42 5 some form of cache flush issue 2016-07-21T03:42:42 1. I would see a led go off on the ethernet connector then? 2016-07-21T03:42:51 You would hope so 2016-07-21T03:43:00 2. interrupt b/w PHY and MAC? 2016-07-21T03:43:07 MAC and cpu 2016-07-21T03:43:34 *** mudit1729 has quit IRC 2016-07-21T03:44:49 my datasheet - section GRETH MAC- mentions "the core also supports status change interrupts from the PHY" 2016-07-21T03:45:06 good 2016-07-21T03:45:27 is the core supposed to be the CPU or the MAC here? 2016-07-21T03:45:32 cpu 2016-07-21T03:45:40 sorry mac 2016-07-21T03:46:51 ok, then PHY can interrupt the MAC, which interrupt the CPU, which sends RTEMS events b/w network thread and driver thread, right? 2016-07-21T03:50:46 5. I ran a custom greth driver from my manufacturer, which hacked a lot about Leon cache... If problem comes from the cache, disabling it would hide my issue, am i correct? 2016-07-21T03:51:02 yes 2016-07-21T03:51:21 yes to cache question 2016-07-21T03:51:28 thanks :-) 2016-07-21T03:53:16 4. how can I know that happens? 2016-07-21T03:54:06 my RX descriptors are not full, seem to contain valid data: no error, IP/TCP detected, EN Off, IE Off, WR Off 2016-07-21T03:54:18 depends on the MAC. They have pointers in them you can read and if the hardware pointer and the driver pointer do not align something has gone wrong. 2016-07-21T03:54:49 ie driven thinks 3 is next and the hardware is stuck at 5 2016-07-21T03:54:59 driven -> driver 2016-07-21T03:56:05 i see 2016-07-21T03:56:09 easy to check 2016-07-21T03:57:43 arf, eclipse is so slow making JTAG mem accesses thru my Wifi/VPN connection :-) 2016-07-21T04:21:09 kiwichris, ok I have been able to read my registers 2016-07-21T04:23:11 the driver RX desc points to 0x800e0000 (greth.rxdesc) and accesses the second element (greth.rx_ptr = 1) of a 16-byte struct, thus 0x800e0008 2016-07-21T04:23:39 and? 2016-07-21T04:24:31 the MAC hardware register GBETH_RX_PTR points to 0x800e0010 2016-07-21T04:25:31 which is non empty 2016-07-21T04:25:41 but disabled 2016-07-21T04:27:01 It looks to me like an issue updating or acknowledging the descriptors. If they are all working and interrupts have being lost you would see an overflow error as the data comes in and uses all the descriptors up. 2016-07-21T04:28:15 let me check the number of desc the driver defined 2016-07-21T04:29:57 I have 16 RX desc 2016-07-21T04:31:54 *** mkhoory-2 has joined #rtems 2016-07-21T04:32:13 all my RX desc are full in fact 2016-07-21T04:32:35 Then it looks like a lost interrupt. 2016-07-21T04:33:00 so the PHY works correctly because data is there 2016-07-21T04:33:14 but MAC > CPu interrupt is broken. 2016-07-21T04:33:37 I'll check that tomorrow :-) 2016-07-21T04:33:39 I would spend some time looking here. 2016-07-21T04:33:46 where do you leave kiwichris ? 2016-07-21T04:33:56 Make sure the level vs edge set up is ok 2016-07-21T04:34:03 live* 2016-07-21T04:34:18 level/edge: sure 2016-07-21T04:34:40 its midnight in Montreal 2016-07-21T04:34:57 Ah yes it would be late. 2016-07-21T04:35:07 and street skunks stinks over my window :-) 2016-07-21T04:35:09 Mid afternoon here in Sydney 2016-07-21T04:35:11 *** mkhoory-3 has quit IRC 2016-07-21T04:35:18 an aussie guy! 2016-07-21T04:35:35 Kiwi living in Australia 2016-07-21T04:35:39 I finished my studies in Wollongong long time ago :-) 2016-07-21T04:35:51 From here? 2016-07-21T04:36:25 what do you mean? 2016-07-21T04:37:06 *** mkhoory-2 has quit IRC 2016-07-21T04:37:34 Did you come to Aus to study or a local who has left? 2016-07-21T04:38:36 I'm french, came to Oz to finish my MEng, then went back to France, worked few years with poor mngt, then flew to Quebec 2016-07-21T04:39:38 Ah nice. 2016-07-21T04:40:08 what employs you in Sydney? 2016-07-21T04:40:36 Amar (_verm) is in Toronto. 2016-07-21T04:40:56 I work for myself consulting to people here and around the world 2016-07-21T04:41:54 good 2016-07-21T04:42:11 Amar? is that a company? 2016-07-21T04:42:27 No that is his name. He is part of the RTEMS team 2016-07-21T04:42:52 i see that on github :-) 2016-07-21T04:42:59 amar takhar 2016-07-21T04:43:02 interesting 2016-07-21T04:43:04 yeap 2016-07-21T04:43:43 thanks heap Chris, have a good afternoon! 2016-07-21T04:44:03 cya and all the best 2016-07-21T04:49:33 *** m-ric has quit IRC 2016-07-21T05:21:29 *** mumptai has joined #rtems 2016-07-21T05:53:22 *** dspace has joined #rtems 2016-07-21T06:25:00 *** monstr has joined #rtems 2016-07-21T06:55:44 *** Davidbrcz has joined #rtems 2016-07-21T07:21:27 *** Davidbrcz has quit IRC 2016-07-21T07:35:19 *** Davidbrcz has joined #rtems 2016-07-21T08:05:13 *** BBD has joined #rtems 2016-07-21T08:16:45 *** Davidbrcz has quit IRC 2016-07-21T08:50:38 *** BBD has quit IRC 2016-07-21T09:09:12 *** mumptai has quit IRC 2016-07-21T09:35:07 *** BDD has joined #rtems 2016-07-21T09:47:45 *** mrd_ has joined #rtems 2016-07-21T09:50:38 *** davee_ has joined #rtems 2016-07-21T09:51:13 *** davee_ has quit IRC 2016-07-21T09:51:20 *** mrd_ has quit IRC 2016-07-21T10:03:26 *** BDD has quit IRC 2016-07-21T11:56:46 *** monstr has quit IRC 2016-07-21T13:30:39 [git] 3cdda03 by Sebastian Huber: [posix: Fix double chain extract] http://devel.rtems.org/changeset/3cdda03/rtems 2016-07-21T13:50:25 *** m-ric has joined #rtems 2016-07-21T14:12:19 my 16 RX desc are full of data, MAC mentions in its STATUS reg that a successful RX has occurred. 2016-07-21T14:13:24 my app only recv() on a STREAM socket, but I see UDP packet in my RX desc. Is this expected? Is it the MAC driver responsibility to drop them? 2016-07-21T14:52:55 *** dspace has left #rtems 2016-07-21T14:55:03 *** dspace has joined #rtems 2016-07-21T14:55:52 *** dspace has joined #rtems 2016-07-21T14:58:13 *** dspace has quit IRC 2016-07-21T14:58:40 *** dspace has joined #rtems 2016-07-21T14:59:06 *** dspace has joined #rtems 2016-07-21T15:00:14 *** dspace has quit IRC 2016-07-21T15:03:08 *** dspace has joined #rtems 2016-07-21T15:23:15 *** dspace has quit IRC 2016-07-21T15:24:03 *** dspace has joined #rtems 2016-07-21T15:33:51 *** dspace has quit IRC 2016-07-21T15:38:04 *** dspace has joined #rtems 2016-07-21T15:40:38 *** dspace has quit IRC 2016-07-21T15:42:16 *** dspace has joined #rtems 2016-07-21T15:44:20 *** dspace has quit IRC 2016-07-21T15:44:29 *** dspace has joined #rtems 2016-07-21T15:56:37 *** dspace has quit IRC 2016-07-21T15:56:47 *** dspace has joined #rtems 2016-07-21T15:57:29 *** dspace has joined #rtems 2016-07-21T15:58:10 *** dspace has joined #rtems 2016-07-21T15:58:52 *** dspace has quit IRC 2016-07-21T17:28:09 *** dr__house|away is now known as dr__house 2016-07-21T18:31:21 *** mumptai has joined #rtems 2016-07-21T19:54:38 hi kiwichris, I sent another email to the mailing-list, but forgot to Cc you. Subject is "How to debug 4.10.99 network stack?" 2016-07-21T21:19:41 *** dr__house is now known as dr__house|away 2016-07-21T22:03:04 *** mumptai has quit IRC 2016-07-21T22:08:51 *** Davidbrcz has joined #rtems 2016-07-21T22:13:39 *** Davidbrcz has quit IRC 2016-07-21T22:27:55 *** Davidbrcz has joined #rtems 2016-07-21T22:35:51 *** Davidbrcz has quit IRC 2016-07-22T05:34:32 [trac] Release/4.11 edited by sebastian.huber (empty comment) http://devel.rtems.org/wiki/Release/4.11 2016-07-22T06:05:44 *** sebhub has joined #rtems 2016-07-22T06:21:02 *** Davidbrcz has joined #rtems 2016-07-22T06:45:49 *** lcpfnvcy has quit IRC 2016-07-22T06:46:04 *** Davidbrcz has quit IRC 2016-07-22T06:52:37 *** lcpfnvcy has joined #rtems 2016-07-22T07:00:50 *** dspace has joined #rtems 2016-07-22T07:05:46 *** monstr has joined #rtems 2016-07-22T07:13:07 *** Wonjun__ has joined #rtems 2016-07-22T07:16:03 *** Wonjun_ has quit IRC 2016-07-22T07:25:23 *** Wonjun__ has quit IRC 2016-07-22T07:25:49 *** Wonjun__ has joined #rtems 2016-07-22T07:28:44 *** Wonjun__ has quit IRC 2016-07-22T07:29:09 *** Wonjun__ has joined #rtems 2016-07-22T07:30:43 *** Wonjun__ has quit IRC 2016-07-22T07:31:08 *** Wonjun__ has joined #rtems 2016-07-22T08:12:02 *** diginet2 has quit IRC 2016-07-22T08:12:03 *** dr__house|away has quit IRC 2016-07-22T08:12:11 *** dr__house|away has joined #rtems 2016-07-22T08:12:36 *** diginet2 has joined #rtems 2016-07-22T09:19:12 [git] 059529e by Sebastian Huber: [score: Add debug support to chains This helps to detect * double insert, append, prepend errors, and * get from empty chain errors.] http://devel.rtems.org/changeset/059529e/rtems 2016-07-22T09:25:21 [git] 64ed0bb by Sebastian Huber: [score: Assert proper seqlock writer state Helps to detect invalid concurrent writes.] http://devel.rtems.org/changeset/64ed0bb/rtems 2016-07-22T09:36:46 *** modwizcode has quit IRC 2016-07-22T09:37:28 *** modwizcode has joined #rtems 2016-07-22T10:37:36 [git] 992494e by Sebastian Huber: [score: Fix for RTEMS_DEBUG] http://devel.rtems.org/changeset/992494e/rtems 2016-07-22T13:40:58 *** benjamint has quit IRC 2016-07-22T13:41:46 *** benjamint has joined #rtems 2016-07-22T13:56:11 *** sebhub has quit IRC 2016-07-22T14:06:06 *** monstr has quit IRC 2016-07-22T14:59:52 *** mumptai has joined #rtems 2016-07-22T15:39:45 *** indy has quit IRC 2016-07-22T15:44:21 *** indy has joined #rtems 2016-07-22T15:46:08 *** dspace has quit IRC 2016-07-22T17:02:51 *** dr__house|away is now known as dr__house 2016-07-22T17:02:59 *** dr__house has joined #rtems 2016-07-22T17:31:43 *** Davidbrcz has joined #rtems 2016-07-22T17:40:49 *** Davidbrcz has quit IRC 2016-07-22T19:16:34 [git] 0055b65 by Pavel Pisa: [bsp/tms570: ensure that change of SCI baudrate is not applied in the middle of character Tx. The rtems_monitor_task() setups/updates termios attributes of the opened TTY and if there is ongoing some other output it leads to the stuck. It would be better to use some termios API function which would call drainOutput() in 2016-07-22T19:16:34 rtems/cpukit/libcsupport/src/termios...] http://devel.rtems.org/changeset/0055b65/rtems 2016-07-22T19:55:52 *** sendak is now known as verm__ 2016-07-22T19:55:58 *** verm__ has joined #rtems 2016-07-22T20:00:32 *** Davidbrcz has joined #rtems 2016-07-22T20:38:44 *** Davidbrcz has quit IRC 2016-07-22T22:18:31 *** m-ric has quit IRC 2016-07-22T23:00:07 *** dr__house is now known as dr__house|away 2016-07-23T03:09:34 *** lcpfnvcy has quit IRC 2016-07-23T06:47:45 *** Davidbrcz has joined #rtems 2016-07-23T06:56:29 *** Davidbrcz has quit IRC 2016-07-23T07:43:10 *** Davidbrcz has joined #rtems 2016-07-23T08:00:54 *** Davidbrcz has quit IRC 2016-07-23T10:11:42 [git] 1c62f7e by Deval Shah: [Added USB interrupt handler. And defined some macros regarding USB.] http://devel.rtems.org/changeset/1c62f7e/rtems 2016-07-23T10:11:43 [git] d548318 by Pavel Pisa: [arm/raspberrypi: remove duplicate setup of IRQ handler in the main ARM exception table. Exception table setup is processed by common CPU architecture support. For ARM architecture, it can be found in the file rtems/c/src/lib/libbsp/arm/shared/start/start.S and ends by bsp_vector_table_copy_done label. The actual tabel content 2016-07-23T10:11:43 can be found at bsp_start...] http://devel.rtems.org/changeset/d548318/rtems 2016-07-23T10:11:43 [git] d216c5d by Pavel Pisa: [arm/raspberrypi: change interrupt dispatch and enable to generic vector id based approach. Using conditional branches to find bits is extremely inefficient and for asynchronous delivery of different interrupt sources lead to total confusion of branch prediction unit.] http://devel.rtems.org/changeset/d216c5d/rtems 2016-07-23T10:44:32 *** Davidbrcz has joined #rtems 2016-07-23T11:45:49 *** Davidbrcz has quit IRC 2016-07-23T12:32:12 *** lcpfnvcy has joined #rtems 2016-07-23T13:16:02 [git] ba0471d by Pavel Pisa: [arm/raspberrypi: add Secure Digital (SD) controller interrupt source ID from Mudit Jain'a tree.] http://devel.rtems.org/changeset/ba0471d/rtems 2016-07-23T13:23:34 *** Davidbrcz has joined #rtems 2016-07-23T14:16:01 *** Davidbrcz has quit IRC 2016-07-23T23:11:42 *** mumptai has quit IRC 2016-07-24T06:17:59 *** mumptai has joined #rtems 2016-07-24T09:05:17 *** dspace has joined #rtems 2016-07-24T12:46:19 *** dspace has quit IRC 2016-07-24T14:36:32 *** Davidbrcz has joined #rtems 2016-07-24T14:46:54 *** Davidbrcz has quit IRC 2016-07-24T15:01:51 *** Davidbrcz has joined #rtems 2016-07-24T17:46:57 *** Davidbrcz has quit IRC 2016-07-24T18:33:13 *** Davidbrcz has joined #rtems 2016-07-24T19:56:22 [trac] #2764/defect (v:4.11) created by snob-wolpike (Typo in c/src/lib/libbsp/i386/pc386/include/edid.h) http://devel.rtems.org/ticket/2764 2016-07-24T20:02:29 *** Davidbrcz has quit IRC 2016-07-24T20:05:26 *** Davidbrcz has joined #rtems 2016-07-24T22:12:34 *** Davidbrcz has quit IRC 2016-07-24T22:44:09 *** mumptai has quit IRC 2016-07-24T23:38:42 *** gedare has joined #rtems 2016-07-24T23:38:54 *** ChanServ sets mode: +o gedare 2016-07-24T23:49:22 *** gedare has quit IRC