2013-02-25T00:02:12 *** kiwichris has joined #rtems 2013-02-25T01:44:16 *** sebhub has joined #rtems 2013-02-25T01:46:04 good morning 2013-02-25T02:21:53 *** MegaAlex|away is now known as MegaAlex 2013-02-25T04:43:39 *** arvind_k is now known as arvind_khadri 2013-02-25T04:44:09 *** the9a3eedi has quit IRC (Ping timeout: 252 seconds) 2013-02-25T05:49:03 *** MegaAlex is now known as MegaAlex|away 2013-02-25T07:35:30 *** MegaAlex|away has quit IRC (Ping timeout: 256 seconds) 2013-02-25T07:37:23 *** MegaAlex|away has joined #rtems 2013-02-25T08:05:01 *** gedare has joined #rtems 2013-02-25T08:46:31 hi sebhub 2013-02-25T10:05:38 *** arvind_khadri has quit IRC (Ping timeout: 246 seconds) 2013-02-25T10:18:21 *** MegaAlex|away is now known as MegaAlex 2013-02-25T10:54:54 *** sebhub has quit IRC (Remote host closed the connection) 2013-02-25T11:32:49 *** arvind_khadri has joined #rtems 2013-02-25T13:20:00 *** heath has joined #rtems 2013-02-25T13:20:05 * heath waves hello 2013-02-25T13:20:14 i didn't know rtems was based out of huntsville :) 2013-02-25T13:21:02 <- birmingham 2013-02-25T13:24:57 *** gedare has quit IRC (Ping timeout: 240 seconds) 2013-02-25T14:38:22 *** DrJoel has joined #rtems 2013-02-25T14:38:22 *** ChanServ sets mode: +o DrJoel 2013-02-25T15:45:56 Hi Joel. 2013-02-25T16:01:23 hello kiwichris 2013-02-25T16:01:30 trying out a new irc client on windows 2013-02-25T16:01:52 Oh ok. Have you managed to look over the Source Builder doco ? 2013-02-25T16:02:45 not yet .. but has anyone noticed the two figures in the user's guide (pdf) which are incorrectly rotated 90 degrees? 2013-02-25T16:03:47 We need a group review to ensure we haven't added services that are undocumented 2013-02-25T16:04:23 Do the RTEMS doco or the Source Builder ? 2013-02-25T16:05:07 the C User's Guide that exists.. has not had a thorough review in a while and my updating to texinfo 5.0 has let me trip across some things. 2013-02-25T16:05:21 I have fixed some but I worry a little attention would be a good thing 2013-02-25T16:05:33 it would be easy to split by chapter if folks would volunteer 2013-02-25T16:29:29 Ok 2013-02-25T16:30:02 I need to read the source builder docs and start to construct the VM 2013-02-25T16:30:49 Not sure I understand the question. The docs explain how to use it , well I hope it does :) 2013-02-25T16:31:09 I wanted to make sure the doco is at a point it is useful for someone using the VM 2013-02-25T16:33:00 If they tell me how to build the tools for the VM, then they are in good shape. I want to pre-build sparc and i386. They will be on the VM. Then students can update themselves later. 2013-02-25T16:33:26 The Quick Start should be all you need 2013-02-25T16:33:32 Did you look at this kit? http://www.adafruit.com/products/1014 2013-02-25T16:33:44 Yes. It is nice. 2013-02-25T16:34:10 http://www.rtems.org/ftp/pub/rtems/people/chrisj/source-builder/source-builder.html#_quick_start 2013-02-25T16:34:21 http://www.rtems.org/ftp/pub/rtems/people/chrisj/source-builder/source-builder.html#_centos_6 2013-02-25T16:34:32 I just noticed it has gpio pins. If they are accessible, then this may be the class HW kit I have been looking for 2013-02-25T16:35:28 It's ability to hook up a JTAG debugger is not simple which is a shame 2013-02-25T16:35:48 It means you have to debug with printf and load kernel on SD etc 2013-02-25T16:35:53 :( my other alternative is a beagle bone 2013-02-25T16:36:08 The Beaglebone is nice but not has cheap. 2013-02-25T16:36:26 Pandaboard ? 2013-02-25T16:36:29 http://www.adafruit.com/products/703 2013-02-25T16:36:45 That is a nice board. 2013-02-25T16:37:03 The breadboard parts with it are what interests me for class projects 2013-02-25T16:37:11 It is OMAP4 and is in the FreeBSD kernel as well. 2013-02-25T16:39:12 I think it has the whole JTAG stuff on board so all you do is plug in to USB and connect OpenOCD 2013-02-25T16:39:31 Ooo comes with a sticker 2013-02-25T16:40:33 I can provide stickers :) 2013-02-25T16:41:36 The Beagleboard-xM is an OMAP3 and I think the Beagleboard is OMAP2, so nothing is the same which is a same. 2013-02-25T16:41:45 same -> shame 2013-02-25T16:43:09 I think Claas Ziemke had Beaglebone in good shape.. we just need to get it merged 2013-02-25T16:43:19 or at least in the shape of the Pi. :) 2013-02-25T16:44:00 Beaglebone or Beagleboard ? The github project I looked at I thought it was just a Beagleboard 2013-02-25T16:44:26 At the kernel level they are different. 2013-02-25T16:45:11 i had a libi2c email from home that included bone in it. That's where I just nicked his email from :) 2013-02-25T16:45:33 Nice 2013-02-25T16:46:58 We just need to push on that getting merged 2013-02-25T16:47:25 Sure. If you decide to take on the Beaglebone I will arrange to get one also. 2013-02-25T16:50:28 Hmm the subject says Beaglebone and the contents say beagleboard 2013-02-25T16:52:04 we need to push on him. That one needs to be in the tree by 4.11 if it runs hello and ticker 2013-02-25T16:54:58 Pandaboards and far to costly 2013-02-25T16:55:50 RaspberryPi vs Beaglebone for RTEMS ?. Beaglebone cause the JTAG support is better, no extra TTL level coverters for serial console access 2013-02-25T17:01:53 That breadboard kit with the Bone looks like a great hobby combination 2013-02-25T17:02:07 Yeah it looks great. 2013-02-25T17:05:06 Yeah, the RaspberryPi's HDMI port is very nice, but is this a big things for RTEMS+hobbie ? 2013-02-25T17:09:18 Does it require a binary blob on Linux? 2013-02-25T17:09:32 On the Pi ? 2013-02-25T17:09:34 Graphics aren't the RTEMS application domain :) 2013-02-25T17:09:36 yes on linux 2013-02-25T17:09:53 Yes and then you have an API to use it. 2013-02-25T17:10:19 The blob is closed how-ever my understanding is the API is available. 2013-02-25T17:10:28 Yes I agree 2013-02-25T17:12:18 We have never dealt with such a blob before.. and RTEMS users are more likely to be interested in adding a cool IO board. :) 2013-02-25T17:12:56 The integrated jtag interface is nice. Easy to debug and manage. 2013-02-25T17:13:33 that's a better feature.. 2013-02-25T17:13:53 I don't know what the digital and analog I/O options are for the two. That's where it would be interesting to our community 2013-02-25T19:41:46 *** arvind_khadri has quit IRC (Read error: Operation timed out) 2013-02-25T19:41:55 *** MegaAlex is now known as MegaAlex|away 2013-02-25T21:13:59 *** arvind_khadri has joined #rtems 2013-02-25T21:21:59 *** arvind_khadri has quit IRC (Ping timeout: 252 seconds) 2013-02-25T21:23:05 *** arvind_khadri has joined #rtems 2013-02-25T21:32:49 *** arvind_khadri has quit IRC (Ping timeout: 248 seconds) 2013-02-25T21:44:14 *** arvind_khadri has joined #rtems 2013-02-25T21:56:47 *** the9a3eedi has joined #rtems 2013-02-25T22:33:16 *** heath has left #rtems ("http://quassel-irc.org - Chat comfortably. Anywhere.") 2013-02-25T23:44:16 *** lcpfnvcy has quit IRC (Read error: Connection reset by peer) 2013-02-25T23:44:29 *** lcpfnvcy has joined #rtems 2013-02-26T01:12:41 I have to say... the RTEMS documentation is one of the best documentations of software I've ever seen 2013-02-26T01:12:42 great job guys 2013-02-26T01:13:02 properly organized, everything is explained properly with few "exceptions" and "gotchas" from what I've seen 2013-02-26T02:02:23 *** the9a3eedi has quit IRC (Read error: Connection reset by peer) 2013-02-26T02:02:49 *** the9a3eedi has joined #rtems 2013-02-26T02:14:40 *** sebhub has joined #rtems 2013-02-26T02:16:02 good morning 2013-02-26T04:51:26 *** MegaAlex|away is now known as MegaAlex 2013-02-26T05:03:16 *** the9a3eedi has quit IRC (Ping timeout: 248 seconds) 2013-02-26T05:15:00 *** omytea has joined #rtems 2013-02-26T05:49:04 hi all 2013-02-26T05:50:59 +1 for BeagleBone … I would so use this for building a Erlang+RTEMS SD card image on 2013-02-26T05:51:23 s/on/for/ 2013-02-26T05:51:30 I think it is a good choice 2013-02-26T05:51:41 If we can support them all we will 2013-02-26T05:52:03 The Pi has more mindshare at the moment though 2013-02-26T05:52:34 The Pi is nice. I have been building tools on as a test of the Source Builder 2013-02-26T05:52:58 http://www.rtems.org/ftp/pub/rtems/people/chrisj/source-builder/source-builder.html 2013-02-26T05:53:19 This is the latest doco for the Source Builder if anyone cares to review it 2013-02-26T05:54:56 asciidoc is really nice 2013-02-26T05:55:18 Nice and easy for this type of documentation 2013-02-26T05:55:42 *** MegaAlex is now known as MegaAlex|away 2013-02-26T05:55:45 I have included the asciidoc code in projects in the past 2013-02-26T05:56:03 Make generating doco simpler. 2013-02-26T05:56:40 for README files etc it is great 2013-02-26T05:57:19 I have done full release documentation with embedded versions, revisions etc 2013-02-26T05:57:54 By release doco I mean what a release contains, fixes, etc 2013-02-26T05:58:24 In general I think user manuals need something stronger. 2013-02-26T06:06:22 I installed asciidoc after I saw what you are doing with it kiwichris … plan to use it 2013-02-26T06:07:47 hehe great 2013-02-26T06:09:36 Check this out http://www.youtube.com/user/RtemsProject 2013-02-26T06:09:40 I little bit of fun 2013-02-26T06:14:25 bye all, I am off for the night 2013-02-26T07:42:07 *** MegaAlex|away has quit IRC (Ping timeout: 245 seconds) 2013-02-26T07:43:21 *** MegaAlex|away has joined #rtems 2013-02-26T08:13:39 *** arvind_khadri has quit IRC (Ping timeout: 260 seconds) 2013-02-26T10:15:49 *** sebhub has quit IRC (Remote host closed the connection) 2013-02-26T11:19:09 *** arvind_khadri has joined #rtems 2013-02-26T11:34:45 *** omytea has quit IRC (Ping timeout: 248 seconds) 2013-02-26T11:57:08 *** arvind_khadri has quit IRC (Ping timeout: 256 seconds) 2013-02-26T13:30:50 *** gedare has joined #rtems 2013-02-26T13:52:50 hey gedare 2013-02-26T13:53:01 howdy DrJoel 2013-02-26T13:56:07 lots of hacking this week. :) 2013-02-26T13:56:15 How are things in DC/VA? 2013-02-26T13:56:33 yeah a lot of activity. pretty good, busy of course 2013-02-26T13:57:16 i have a new project underway... a lot of learning, and pretty soon some hacking too. 2013-02-26T13:57:24 :) RTEMS related 2013-02-26T13:57:33 unfortunately not 2013-02-26T13:58:08 oh well.. it pays the bills 2013-02-26T13:58:19 exactly 2013-02-26T13:58:22 *** DrJoel sets mode: +o gedare 2013-02-26T13:58:31 I don't know if that sticks or not. 2013-02-26T13:58:40 no clue.. i could test it 2013-02-26T13:58:47 *** gedare has left #rtems ("Leaving") 2013-02-26T13:58:52 *** gedare has joined #rtems 2013-02-26T13:58:57 looks like not 2013-02-26T13:59:00 Is you nickname registered? 2013-02-26T13:59:03 ya 2013-02-26T13:59:16 hmmm.. then we need to find a faq 2013-02-26T14:03:28 If you find it, let me know.. all I see is temporay 2013-02-26T14:05:12 i see: /msg ChanServ ACCESS #rtems ADD gedare OP 2013-02-26T14:07:11 that doesn't work. weird errors about templates 2013-02-26T14:08:46 Do you think we should pull the BeagleBone project in from github and try to get it building? 2013-02-26T14:08:55 Better to be in the main tree than rot outside it 2013-02-26T14:09:14 http://www.adafruit.com/products/703 looks awesome for class use 2013-02-26T14:09:38 ping the student to see how outdated his github is from what he had working.. but it would be better to have something that we at least have some idea "should work" on the hardware 2013-02-26T14:10:07 i don't know that his bsp works for beagle bone 2013-02-26T14:10:10 i think he was targetting beagle board 2013-02-26T14:10:16 and i think those have different cpu 2013-02-26T14:19:18 They do have different cpus and I pinged him yesterday... 2013-02-26T14:22:40 *** antgreen has joined #rtems 2013-02-26T14:23:28 Does RTEMS really support vfork? 2013-02-26T14:23:38 No 2013-02-26T14:24:07 We may have stubs that return -1/ENOSYS 2013-02-26T14:24:15 For a lot of methods we have those 2013-02-26T14:24:32 oh, ok. I'm reworking how exception handling happens on moxie (mostly driven by vfork() on uClinux) 2013-02-26T14:24:52 and a quick google shows references for vfork and RTEMS, but I didn't dig 2013-02-26T14:25:05 If the port is working, we shuld get it in the tree. Chris is building a toolset now 2013-02-26T14:25:26 There may have been an issue with one tool. I don't recall. 2013-02-26T14:25:29 We would like it in 4.11 2013-02-26T14:25:39 what is the deadline for that? 2013-02-26T14:26:33 no hard deadline yet. Just working "blocker issues" now 2013-02-26T14:26:38 There was an issue with gdb until he installed dtc 2013-02-26T14:26:52 Moxie is "desirable" so better to be in than squeak. :) 2013-02-26T14:28:21 hmmm.. ok 2013-02-26T14:29:01 I don't think we are less than a month from branching though. 2013-02-26T14:29:26 I was going to rework the exception model to separate user and super-mode stacks in order to support vfork on uClinux 2013-02-26T14:29:49 gedare do you want to propose renaming the ChangeLog's to ChangeLog-pre-2012? or ChangeLog-pre-git? 2013-02-26T14:29:53 but if RTEMS has no vfork(), then I'll just assume we always run in super mode, and I can make the change later without breaking things 2013-02-26T14:30:04 sounds great! 2013-02-26T14:30:07 i want to propose deleting them ;) 2013-02-26T14:30:23 i know that won't fly though soooo 2013-02-26T14:30:31 Go ahead.. I just want to keep my karma intact for a little longer this week. 2013-02-26T14:30:45 my second choice is to name them .ChangeLog 2013-02-26T14:30:56 The documentation changes are going well... and I definitely could have gotten arrows 2013-02-26T14:31:00 yea 2013-02-26T14:31:05 lol.. awesome choice 2013-02-26T14:31:40 But it is a milestone with them building on texinfo 5.0 and old texinfo plus improvements internally. 2013-02-26T14:31:57 Now for a group users guide review sprint.. would be beneficial 2013-02-26T14:31:57 antgreen: we don't bother with user/super mode separation with any arch afaik 2013-02-26T14:32:05 ight 2013-02-26T14:32:09 right 2013-02-26T14:32:59 vfork() would require an interrupt handling stack, so I was going to make interrupts put you in super mode on an alternate stack 2013-02-26T14:33:11 but I'll just avoid the whole issue for now. 2013-02-26T14:33:22 many(most?) architectures in rtems do use an interrupt stack 2013-02-26T14:33:28 architectures without hardware support for switching stacks do it manually 2013-02-26T14:33:29 but all code runs in privileged mode 2013-02-26T14:33:49 same disclaimer I give all (most?) 2013-02-26T14:33:55 do you allow for nested interrupts? 2013-02-26T14:33:58 yea 2013-02-26T14:34:16 if the arch permits it i think.. i forget if any do not 2013-02-26T14:34:26 basically... 2013-02-26T14:34:33 the arch should provide some ISR_Handler function 2013-02-26T14:34:40 that sets up a C environment with a safe stack to use 2013-02-26T14:34:46 and vectors to the user interrupt handler 2013-02-26T14:35:09 this is probably described somewhere...hmmm 2013-02-26T14:35:47 ok, I'm going to submit my patch in two phases. One will be a basic port that runs on the gdb simulator with no ISR_Handler, but it runs on the gdb sim. 2013-02-26T14:35:51 I can do that today. 2013-02-26T14:36:12 I think the "Porting" guide has a section explaining interrupts.. but feel free to keep asking too :) 2013-02-26T14:36:15 Then the phase 2 will add a RTC and an interrupt handler 2013-02-26T14:36:27 that should be fine, especially for "limited release" 2013-02-26T14:36:41 i mean... we can bug fix the bsp after release easier than adding it after release 2013-02-26T14:36:50 and by bug fix i mean add interrupt handling :) 2013-02-26T14:36:54 that is an important baseline. It is enough code to review. BSP can use the sim-clock-idle task fake clock tick. It is enough to run most tests and ensure tools are workimng 2013-02-26T14:37:30 I submitted my qemu port upstream a few days ago, which can better simulate the hardware 2013-02-26T14:37:46 If the CPU does not support a dedicated interrupt stack, then the porter 2013-02-26T14:37:46 has two options: (1) execute interrupts on the stack of the interrupted 2013-02-26T14:37:46 task, and (2) have RTEMS manage a dedicated interrupt stack. 2013-02-26T14:37:46 NOTE: If CPU_HAS_SOFTWARE_INTERRUPT_STACK is TRUE, then the macro 2013-02-26T14:37:46 CPU_ALLOCATE_INTERRUPT_STACK should also be set to TRUE. 2013-02-26T14:37:47 Only one of CPU_HAS_SOFTWARE_INTERRUPT_STACK and 2013-02-26T14:37:49 CPU_HAS_HARDWARE_INTERRUPT_STACK should be set to TRUE. 2013-02-26T14:37:53 yay qemu 2013-02-26T14:42:06 *** kiwichris has quit IRC (Quit: This computer has gone to sleep) 2013-02-26T14:43:03 *** kiwichris has joined #rtems 2013-02-26T14:45:12 *** kiwichris has quit IRC (Client Quit) 2013-02-26T14:57:42 *** MegaAlex|away has quit IRC (Quit: ZNC - http://znc.in) 2013-02-26T15:00:06 antgreen there are scripts in rtems-testing/sim-scripts to assist in invoking simulators. Sounds like Moxie will have two scripts. There is a framework for qemu and anotehr for gdb based simulators 2013-02-26T15:00:41 *** MegaAlex|away has joined #rtems 2013-02-26T15:01:16 *** gedare has quit IRC (Ping timeout: 272 seconds) 2013-02-26T15:11:10 Ok - patches are ready. 2013-02-26T15:11:17 rtems-devel? 2013-02-26T15:11:21 is that where they go? 2013-02-26T15:12:37 *** kiwichris has joined #rtems 2013-02-26T15:23:05 patches away 2013-02-26T15:23:16 oops - sent one too many emails 2013-02-26T15:23:20 oh we;; 2013-02-26T15:23:22 ll 2013-02-26T15:27:07 *** kiwichris has quit IRC (Quit: This computer has gone to sleep) 2013-02-26T15:49:55 *** MegaAlex|away has quit IRC (Ping timeout: 248 seconds) 2013-02-26T15:51:16 *** MegaAlex|away has joined #rtems 2013-02-26T15:53:55 *** kiwichris has joined #rtems 2013-02-26T16:29:17 what is the smallest size heap I can reasonably expect to run a trivially small RTEMS program with? 2013-02-26T16:30:00 I'm wondering if I can make do with embedded RAM on my FPGA for a simple test 2013-02-26T16:30:32 how big is that ram ? 2013-02-26T16:34:56 I'm trying to figure that out. Docs say "kb" but I think they may mean "kiloBITS". 2013-02-26T16:35:41 I was able to get the text size for helloworld down to about 32k using -ffunction-sections and section gc. 2013-02-26T16:35:52 Nice. 2013-02-26T16:35:55 overall the minimum stack size has an impact on how much memory a program takes. 2013-02-26T16:36:09 why is ffunction-sections not default? 2013-02-26T16:36:15 FYI I have added moxie tool building the RTEMS Source Builder 2013-02-26T16:36:25 It didn't work on all targets when I looked into it initially. 2013-02-26T16:36:35 It does make sense to re-investigate it as a default option 2013-02-26T16:36:41 I figured out the dtc tool issue for gdb 2013-02-26T16:36:54 How much did it save you? 2013-02-26T16:37:05 it almost halved the text size 2013-02-26T16:37:10 IIRC 2013-02-26T16:37:18 I'll test again tonight 2013-02-26T16:37:39 RTEMS is generally one file per function. So I am curious what got left out. It is a hint that we have unnecessary dependencies in files or newlib or somewhere else. 2013-02-26T16:37:56 libc ? 2013-02-26T16:37:56 I can do a diff of symbols 2013-02-26T16:38:03 building now 2013-02-26T16:38:06 But if the linker can do that much for you, then we need to turn it on everywhere it works. Computers do dirty, repetitive work well 2013-02-26T16:40:22 hmm.. I think that patch I posted is missing something 2013-02-26T16:40:31 cd ../rt 2013-02-26T16:40:37 oops 2013-02-26T16:43:45 one line missing from cpukit/configure.ac 2013-02-26T16:55:07 Congratulations on your admittance to the elite RTEMS Facebook group :) 2013-02-26T16:57:54 hehe - thanks. 2013-02-26T16:58:12 * DrJoel is going home 2013-02-26T16:58:12 Got to welcome the toddlers home. later. 2013-02-26T17:14:00 *** DrJoel has quit IRC (Ping timeout: 256 seconds) 2013-02-26T21:21:26 *** krispin1 has joined #rtems 2013-02-26T21:28:56 *** krispin1 has quit IRC (Quit: Leaving.) 2013-02-26T21:53:38 *** antgreen has quit IRC (*.net *.split) 2013-02-26T21:53:40 *** rokka has quit IRC (*.net *.split) 2013-02-26T21:55:41 *** antgreen has joined #rtems 2013-02-26T21:55:41 *** rokka has joined #rtems 2013-02-26T22:12:47 *** the9a3eedi has joined #rtems 2013-02-26T22:16:23 bah. I'll repost my moxie port patches tomorrow. They were against an older rtems. I am testing the newer patches now. 2013-02-26T22:49:31 ok, I will wait 2013-02-26T23:44:39 *** MegaAlex|away has quit IRC (Ping timeout: 260 seconds) 2013-02-26T23:56:43 *** MegaAlex|away has joined #rtems 2013-02-27T00:43:25 *** arvind_khadri has joined #rtems 2013-02-27T00:48:09 *** arvind_khadri has quit IRC (Read error: Connection reset by peer) 2013-02-27T00:52:30 *** arvind_khadri has joined #rtems 2013-02-27T01:32:03 moorning 2013-02-27T01:32:31 anyone here know how the hell u remove a shared mem segment that cant be removed by ipcrm? 2013-02-27T01:32:36 dont find the pid 2013-02-27T01:32:40 either with lsof 2013-02-27T01:32:43 or ps 2013-02-27T01:32:44 :( 2013-02-27T01:32:50 dont want to reboot the box 2013-02-27T01:32:55 this is linux 2013-02-27T01:33:03 sorry if it's (off) topic 2013-02-27T01:34:19 key shmid owner perms bytes nattch status 2013-02-27T01:34:20 0x00000000 4227075 root 644 80 2 dest 2013-02-27T01:34:25 i've tried as root 2013-02-27T01:34:27 no joy 2013-02-27T01:34:36 says nattach... 2013-02-27T01:34:41 =2 2013-02-27T02:23:11 *** sebhub has joined #rtems 2013-02-27T02:47:30 good morning 2013-02-27T02:50:15 solved my issue with a reboot 2013-02-27T02:50:16 :( 2013-02-27T02:50:25 still dont know what happened or why! 2013-02-27T02:50:57 reboot of what? 2013-02-27T03:29:04 sebhub: was a linux box... question was a little off topic... sorry 2013-02-27T03:29:26 sebhub: tried to ipcrm a shared mem segment but couldn't 2013-02-27T03:29:30 reboot fixed it 2013-02-27T03:29:34 still dont know why 2013-02-27T04:16:32 *** lcpfnvcy has quit IRC (Read error: Connection reset by peer) 2013-02-27T04:27:45 *** lcpfnvcy has joined #rtems 2013-02-27T05:07:14 *** the9a3eedi has quit IRC (Ping timeout: 246 seconds) 2013-02-27T05:29:25 *** lcpfnvcy has quit IRC (Ping timeout: 248 seconds) 2013-02-27T05:30:25 *** lcpfnvcy has joined #rtems 2013-02-27T05:53:54 *** arvind_khadri has quit IRC (Ping timeout: 240 seconds) 2013-02-27T06:11:51 *** arvind_khadri has joined #rtems 2013-02-27T08:13:27 ok, just resubmitted moxie port patches 2013-02-27T09:36:33 *** arvind_khadri has quit IRC (Ping timeout: 248 seconds) 2013-02-27T09:47:47 *** DrJoel has joined #rtems 2013-02-27T10:26:22 *** DrJoel has left #rtems 2013-02-27T10:26:57 *** DrJoel has joined #rtems 2013-02-27T10:27:26 *** the9a3eedi has joined #rtems 2013-02-27T10:30:11 *** sebhub has quit IRC (Remote host closed the connection) 2013-02-27T10:41:27 *** the9a3eedi has quit IRC (Ping timeout: 252 seconds) 2013-02-27T11:13:31 antgreen walking by the keyboard .. do the comments so far make sense? 2013-02-27T11:20:17 going to read them now 2013-02-27T11:21:27 DrJoel ... so ChangeLog files have been deprecated? 2013-02-27T11:21:43 there is a script in rtems-testing/merge-helpers which checks for some mundane issues. CVS Id's being one of them. It may or may nor check the right direction.. forbidden or requjire on them.. but I amhappy to update it as we spot trivial things a script should catch.. it takes load off subimtters and reviewers 2013-02-27T11:21:45 I shouldn't add to them, or create new ones. 2013-02-27T11:22:33 and I run check_submission on my patch file? 2013-02-27T11:23:01 no on a source directory 2013-02-27T11:23:15 ok, trying that now 2013-02-27T11:24:48 hmm.. ok. I'll fix these up. 2013-02-27T11:25:10 /home/green/hack/rtems-submit 2013-02-27T11:25:14 oops 2013-02-27T11:25:27 http://paste.fedoraproject.org/3912 2013-02-27T11:25:29 that one 2013-02-27T11:32:30 DrJoel is says :"RTEMS_BSP_CLEANUP_OPTIONS is NOT in ./configure.ac" 2013-02-27T11:32:44 but I have "RTEMS_BSP_CLEANUP_OPTIONS(0, 0)" in configure.ac 2013-02-27T11:34:52 It's also complaining about missing RTEMS_BSP_BOOTCARD_OPTIONS, but I don't see that being used anywhere 2013-02-27T11:39:51 DrJoel: better, but not sure what to do about last two warnings: http://paste.fedoraproject.org/3915/ 2013-02-27T12:13:53 Version 3 of the patch series just submitted 2013-02-27T12:15:55 sorry working from home today.. family duties 2013-02-27T12:16:29 From sparc/erc32/configure.ac: RTEMS_BSP_CLEANUP_OPTIONS(0,1,1) 2013-02-27T12:16:57 gah - I just saw Gedare's note. Version 4 in the works. 2013-02-27T12:16:58 Every BSP needs that in configure.ac. They configure the behavior of the shared code when you are on the shutdown path. 2013-02-27T12:17:20 I have that already 2013-02-27T12:17:25 so why is it complaining? 2013-02-27T12:18:17 ./c/src/aclocal/bsp-bspcleanup-options.m4 2013-02-27T12:18:33 Bug in the script. 2013-02-27T12:18:40 It also needs three arguments. 2013-02-27T12:20:13 I actually don't see why it is complaining.. may be a slight difference in the name it is checking for... 2013-02-27T12:20:28 set -x and see why it barks 2013-02-27T12:28:54 *** sevikkk has quit IRC (Ping timeout: 252 seconds) 2013-02-27T12:55:08 I just ran it on the erc32 BSP and it didn't complain on it. You have to run it in the BSP directory sparc/erc32 in my case. Check that it works there. 2013-02-27T12:59:27 ok 2013-02-27T12:59:41 I just cleaned up everything Gedare found. 2013-02-27T13:00:32 btw - what does the name "score" mean or where does it come from? 2013-02-27T13:00:41 I want a good check_submission script since I believe that for job X, people have Y time. They can spend it nitpicking or giving substantive feedback. Let a script do the easy crap 2013-02-27T13:00:46 SuperCore 2013-02-27T13:01:09 http://www.rtems.org/wiki/images/b/b9/RTEMSArchitecture.png 2013-02-27T13:01:31 All APIs tasking, communication and synchronization APIs are a thin wrapper over it 2013-02-27T13:03:19 I just rant check_submission on the erc32 BSP and got this: http://paste.fedoraproject.org/3928 2013-02-27T13:03:59 but your run was clean? 2013-02-27T13:04:46 not 100% clean. :) Mostly looks the same 2013-02-27T13:04:52 === Checking for RTEMS_BSP_BOOTCARD_OPTIONS in BSP configure.ac 2013-02-27T13:04:52 RTEMS_BSP_BOOTCARD_OPTIONS is NOT in ./configure.ac 2013-02-27T13:04:52 === Checking for RTEMS_BSP_CLEANUP_OPTIONS in BSP configure.ac 2013-02-27T13:04:52 === Checking for not using -O2 2013-02-27T13:04:52 === Checking for turning on extra GCC warning checks 2013-02-27T13:04:53 === Checking for timing information 2013-02-27T13:05:44 This script is not that old.. and only 1-2 BSPs have run it before being submitted. It hasn't even been vetted by the community. If you see false positives, ignore them and suggest ways to improve the script. 2013-02-27T13:06:02 tabs in .S are a topic for discussion. That triggers a lot 2013-02-27T13:06:29 ok, I've added the last parameter to RTEMS_BSP_CLEANUP_OPTIONS 2013-02-27T13:06:37 I think I've done everything people have asked so far 2013-02-27T13:07:45 At the risk of spamming the list again, I'll send in version 5 in a few minutes 2013-02-27T13:09:23 DrJoel: actually, should I send the whole thing again, or an incremental patch that deals with recent comments? 2013-02-27T13:09:31 it's a lot smaller 2013-02-27T13:10:26 I'll do the latter 2013-02-27T13:15:17 UnI wish gedare or kiwichris were awake. 2013-02-27T13:15:33 My preference would be to keep getting large replacement patches until code is in git. 2013-02-27T13:17:01 Ok, I can resubmit tonight. 2013-02-27T13:17:13 kiwichris is in NZ? 2013-02-27T13:17:25 and Gedare in Europe? 2013-02-27T13:20:50 Chris is from NZ but in Sidney .. he has screwed me up on recognizing NZ and Aussie accents :) 2013-02-27T13:21:00 Gedare is a postdoc student at GWU in DC 2013-02-27T13:21:05 I am in Huntsville AL 2013-02-27T13:21:08 Where are you? 2013-02-27T13:21:31 Toronto, these days 2013-02-27T13:21:44 I've spend a lot of time in Hunstville - nice place! 2013-02-27T13:21:54 Isn't that where Jeff Johnson is also? 2013-02-27T13:22:01 Yes 2013-02-27T13:22:11 We're both ex-Cygnus 2013-02-27T13:22:20 I love Huntsville. Large enough with small town feel. :) 2013-02-27T13:22:22 but I only recently moved back to toronto 2013-02-27T13:22:40 I didn't realize you were a Cygnus relic .. lol 2013-02-27T13:22:45 we bought a company in Huntsville ~10years ago and still have an office there 2013-02-27T13:23:07 yep.. we were in the same complex until about 3 years ago. 2013-02-27T13:23:08 mostly ex-intergraph people 2013-02-27T13:23:15 oh, wow 2013-02-27T13:23:50 Our office is now on the other side of Research Park across from Bridge Street (if you know what that is) 2013-02-27T13:24:14 On a random tangent.. since you did the gdb simulator.. do you know if it shows console output when debugging with Eclipse? 2013-02-27T13:24:35 hm... no idea 2013-02-27T13:24:43 I guess I could try that 2013-02-27T13:24:56