2016-07-04T05:20:45 *** tasslehoff has joined #rtems 2016-07-04T05:29:32 *** mumptai has joined #rtems 2016-07-04T05:59:16 *** sebhub has joined #rtems 2016-07-04T06:47:45 sebhub: reorganised the User Guide. I hpoe to have a 4.11 finished soon. 2016-07-04T06:48:00 https://ftp.rtems.org/pub/rtems/people/chrisj/docs/user/index.html 2016-07-04T06:56:37 *** Wonjun_ has joined #rtems 2016-07-04T07:00:00 *** Wonjun__ has quit IRC 2016-07-04T07:04:17 ok, great 2016-07-04T07:04:28 I am currently busy with the OMIP locking protocol 2016-07-04T07:04:38 once this is done, I will work on the documentation 2016-07-04T07:04:45 for 4.12 2016-07-04T08:25:28 Working with a driver that uses rtems_disk_create_phys. Knowing only the devname, how do I open the dev and get an rtems_disk_device* so I can do ioctl calls? 2016-07-04T08:27:10 open(devname) 2016-07-04T08:29:43 sebhub: doesn't that give me a file descriptor that I need to use to get an rtems_disk_device* somehow? 2016-07-04T08:30:41 see testsuites/libtests/block11/init.c 2016-07-04T08:30:49 in the RTEMS sources for an example 2016-07-04T08:31:50 this rtems_disk_create_phys() is legacy code, better use rtems_blkdev_create() 2016-07-04T08:33:13 which RTEMS version do you use? 2016-07-04T08:33:38 ah. rtems_disk_fd_get_disk_device is what I need I see. thanks. 2016-07-04T08:33:41 let's see 2016-07-04T08:34:17 why do you need to access this structure at all? 2016-07-04T08:34:40 its quite dangerous 2016-07-04T08:35:59 *** mumptai has quit IRC 2016-07-04T08:36:16 sebhub: I have a usb mass storage + ramdisk example. I'm trying to rewrite it to expose my sd card instead of the ramdisk. 2016-07-04T08:37:36 seems I'm on 4.10.99.0 2016-07-04T09:10:50 [git] 0057e38 by Sebastian Huber: [libchip: Simplify RTC driver] http://devel.rtems.org/changeset/0057e38/rtems 2016-07-04T10:59:48 *** tasslehoff has quit IRC 2016-07-04T11:54:28 *** gigetoo has quit IRC 2016-07-04T11:58:12 [trac] #2752/enhancement (v:4.10) created by sebastian.huber (Relax execution enviroment for thread begin extensions) http://devel.rtems.org/ticket/2752 2016-07-04T12:03:45 *** gigetoo has joined #rtems 2016-07-04T12:10:51 [git] 137bd37 by Alexander Krutwig: [bsp/atsam: Add RTC driver] http://devel.rtems.org/changeset/137bd37/rtems 2016-07-04T14:04:49 [git] 19a9090 by Pavel Pisa: [arm/score and shared: define ARM hypervisor mode and alternate vector table base access. The main reason for inclusion of minimum hypervisor related defines is that current ARM boards firmware and loaders (U-boot for example) start loaded operating system kernel in HYP mode to allow it take control of virtualization (Linux/KVM for 2016-07-04T14:04:50 example).] http://devel.rtems.org/changeset/19a9090/rtems 2016-07-04T14:04:50 [git] 5812df8 by Pavel Pisa: [bsps/arm: Support recent bootloaders starting kernel in HYP mode When HYP mode is detected at startup then setup HYP mode vectors table (for future extensions) clean exceptions switching to HYP mode and switch CPU to ARM SVC mode. BSPs which want to use this support need to include next option in their configure.ac 2016-07-04T14:04:50 RTEMS_BSPOPTS_SET([BSP_START_IN_HYP_S...] http://devel.rtems.org/changeset/5812df8/rtems 2016-07-04T14:04:50 [git] fe106ad by Pavel Pisa: [arm/raspberrypi: Enable HYP to SVC switch for this BSP. This support is required when newer firmware is used on Raspberry Pi 2 boards.] http://devel.rtems.org/changeset/fe106ad/rtems 2016-07-04T14:04:50 [git] 0cb50ab by Pavel Pisa: [score/arm: Ensure that copile time alignment is 64 bytes for Cortex-A multilib. Some/many Cortex-A cores have data cache line length 64 bytes and maximum value has to be used for system structures alignment.] http://devel.rtems.org/changeset/0cb50ab/rtems 2016-07-04T14:04:50 [git] 2d5902d by Pavel Pisa: [arm/raspberrypi: update VideoCore cache flush workaround to work on RPi2. The arm_cp15_data_cache_clean_and_invalidate leads to hang on RPi2, clean by individual lines works on RPi1 and RPi2.] http://devel.rtems.org/changeset/2d5902d/rtems 2016-07-04T14:04:51 [git] 0e507d5 by Pavel Pisa: [rtems+bsps/cache: Define cache manager operations for code synchronization and maximal alignment. There is need for unambiguous named and defined cache function which should be called when code is updated, loaded or is self-modifying. There should be function to obtain maximal cache line length as well. This function can and 2016-07-04T14:04:51 should be used for allocations...] http://devel.rtems.org/changeset/0e507d5/rtems 2016-07-04T14:04:52 [git] abea02a by Pavel Pisa: [bsp/arm: Report correct maximal cache line length for ARM Cortex-A + L2C-310.] http://devel.rtems.org/changeset/abea02a/rtems 2016-07-04T14:04:52 [git] d431653 by Pavel Pisa: [bsps/arm: Change code to explicit selection of cache implementation for ARM BSPs. The original ARM architecture wide cache_.h is changed to dummy version for targets not implementing/enablig cache at all. The ARM targets equipped by cache should include appropriate implementation. Next options are available for now 2016-07-04T14:04:53 c/src/lib/libbsp/arm/shared/armv467ar-...] http://devel.rtems.org/changeset/d431653/rtems 2016-07-04T14:04:53 [git] 2b885d6 by Pavel Pisa: [libdl/rtl-obj.c: ensure that loaded code is synchronized through caches. Synchronize each cluster of sections of the same type separately to support even cases where text and data are allocated from different areas (for example due allocation from different MPU protection regions). rtems_cache_instruction_sync_after_code_change 2016-07-04T14:04:54 is called even to data sect...] http://devel.rtems.org/changeset/2b885d6/rtems 2016-07-04T15:04:48 *** sendak has joined #rtems 2016-07-04T15:06:14 *** sendak has joined #rtems 2016-07-04T15:09:39 *** verm__ has quit IRC 2016-07-04T17:56:07 *** mumptai has joined #rtems 2016-07-04T18:10:24 [git] 6dc5c03 by Punit Vara: [Beaglebone: Add original BBBIO PWM driver This patch adds the PWM driver code for the Beaglebone Black from BBBIO: https://github.com/VegetableAvenger/BBBIOlib/blob/master/BBBio_lib/BBBiolib_PWMSS.c This commit is for tracking purposes only; the next commit will adapt the code for RTEMS.] 2016-07-04T18:10:25 http://devel.rtems.org/changeset/6dc5c03/rtems 2016-07-04T18:10:25 [git] 5e3096d by Punit Vara: [Beaglebone: Update PWM driver imported from BBBIO This patch adapts the previously added Beaglebone PWM code from BBBIO to RTEMS. This work was done in the context of the Google Summer of Code 2016, and further patches will follow to improve the code quality and documentation.] http://devel.rtems.org/changeset/5e3096d/rtems 2016-07-04T18:38:30 *** Davidbrcz has joined #rtems 2016-07-04T21:48:10 *** Davidbrcz has quit IRC 2016-07-04T21:54:34 *** Davidbrcz has joined #rtems 2016-07-04T22:01:40 *** Davidbrcz has quit IRC 2016-07-04T22:49:49 *** mumptai has quit IRC 2016-07-05T05:50:57 *** mumptai has joined #rtems 2016-07-05T06:10:29 [git] c305847 by Sebastian Huber: [bsps/arm: Fix basic cache support for SMP] http://devel.rtems.org/changeset/c305847/rtems 2016-07-05T07:14:13 *** Wonjun__ has joined #rtems 2016-07-05T07:17:03 *** Wonjun_ has quit IRC 2016-07-05T07:48:51 *** tasslehoff has joined #rtems 2016-07-05T07:53:13 [trac] #2412/defect (assigned) updated by sebastian.huber (empty comment) http://devel.rtems.org/ticket/2412 2016-07-05T07:53:29 [trac] #2412/defect (assigned) updated by sebastian.huber (empty comment) http://devel.rtems.org/ticket/2412 2016-07-05T07:53:37 [trac] #2412/enhancement (assigned) updated by sebastian.huber (empty comment) http://devel.rtems.org/ticket/2412 2016-07-05T07:59:52 [trac] #2684/defect (closed) updated by sebastian.huber ([291945f137d411834c0eaf3b0bb819263efb845b/rtems] No plans to fix this for RTEMS ...) http://devel.rtems.org/ticket/2684 2016-07-05T08:59:51 *** Davidbrcz has joined #rtems 2016-07-05T09:12:14 *** mumptai has quit IRC 2016-07-05T09:47:01 *** darnir has quit IRC 2016-07-05T09:55:26 *** darnir has joined #rtems 2016-07-05T10:28:50 [trac] #2510/enhancement (new) updated by darnir ( == Processor Affinity Set Data Structures == For the scheduler, we need to ...) http://devel.rtems.org/ticket/2510 2016-07-05T10:34:40 *** Davidbrcz has quit IRC 2016-07-05T10:54:19 *** darnir has quit IRC 2016-07-05T10:58:39 *** darnir has joined #rtems 2016-07-05T13:31:05 *** tasslehoff has quit IRC 2016-07-05T14:02:31 How do I access the default idle thread node for the scheduler? 2016-07-05T15:42:42 [git] 116370c by Joel Sherrill: [shared/console-polled.c: Use standard fatal error codes] http://devel.rtems.org/changeset/116370c/rtems 2016-07-05T15:43:44 [git] 5df7499 by Joel Sherrill: [rtems-bsps: Update 4.11 to 4.12] http://devel.rtems.org/changeset/5df7499/rtems 2016-07-05T16:09:30 *** mumptai has joined #rtems 2016-07-05T17:41:52 *** Davidbrcz has joined #rtems 2016-07-05T18:15:28 *** Davidbrcz has quit IRC 2016-07-05T19:11:51 *** danieli has quit IRC 2016-07-05T19:12:53 *** danieli has joined #rtems 2016-07-05T19:55:58 *** danieli has quit IRC 2016-07-05T19:56:58 *** danieli has joined #rtems 2016-07-05T19:58:28 *** danieli has joined #rtems 2016-07-05T20:03:29 *** danieli has joined #rtems 2016-07-05T20:06:15 *** danieli has quit IRC 2016-07-05T20:06:34 *** danieli has joined #rtems 2016-07-05T20:09:27 *** danieli has quit IRC 2016-07-05T20:54:50 *** ita is now known as ita|zzz 2016-07-05T21:25:30 *** danieli has joined #rtems 2016-07-05T21:26:40 *** danieli has quit IRC 2016-07-05T21:26:40 *** danieli has joined #rtems 2016-07-05T21:46:45 *** mumptai has quit IRC 2016-07-05T22:09:49 *** Wonjun_ has joined #rtems 2016-07-05T22:13:31 *** Wonjun__ has quit IRC 2016-07-05T22:48:12 *** Noslac has joined #rtems 2016-07-05T23:18:22 *** Noslac has quit IRC 2016-07-05T23:19:07 *** Wonjun_ has quit IRC 2016-07-05T23:19:30 *** Wonjun_ has joined #rtems 2016-07-05T23:21:37 *** Davidbrcz has joined #rtems 2016-07-05T23:41:52 *** Davidbrcz has quit IRC 2016-07-06T00:57:08 [git] 761ae78 by Chris Johns: [Make the license message viewable with a verbose option. The license details can be viewed if you mount with a -v option: # mount -i nfs -o "-v" 1.2.3.4:/mnt /mnt] http://devel.rtems.org/changeset/761ae78/rtems-libbsd 2016-07-06T03:01:51 [git] 8aa75d0 by Chris Johns: [Config (.cfg) files are only valid if deeper than 5.] http://devel.rtems.org/changeset/8aa75d0/rtems 2016-07-06T05:20:45 *** mumptai has joined #rtems 2016-07-06T07:08:30 [git] b8c7bf9 by Chris Johns: [Set syslog priority to 'debug' for all tests.] http://devel.rtems.org/changeset/b8c7bf9/rtems-libbsd 2016-07-06T07:24:11 *** Davidbrcz has joined #rtems 2016-07-06T07:34:19 [git] 008efaf by Sebastian Huber: [score: Do not disable ISR in _Terminate() This partially reverts 38ee75853f674977609bd078c69fb53420afdd08. Let the calling context decide if interrupts must be disabled or not. The goal is to enable fatal extensions to continue program execution after some fatal errors.] http://devel.rtems.org/changeset/008efaf/rtems 2016-07-06T07:34:19 [git] 6f6cf78 by Sebastian Huber: [score: Postpone SMP shutdown in _Terminate() This enables fatal extensions to continue program execution after some fatal errors.] http://devel.rtems.org/changeset/6f6cf78/rtems 2016-07-06T07:47:56 *** tasslehoff has joined #rtems 2016-07-06T07:54:28 *** Davidbrcz has quit IRC 2016-07-06T07:57:51 *** mumptai has quit IRC 2016-07-06T08:03:10 [git] 3e14594 by Chris Johns: [Add rtems-bsp-builder. This is a testing tool that builds BSPs with a range of configure options for regression testing changes do not break the kernel code. The builds are controlled by an INI file. The INI configuration has profiles which define a specific set of architures and BSP to build. There are architectures which 2016-07-06T08:03:10 contain BSPs and these fu...] http://devel.rtems.org/changeset/3e14594/rtems-tools 2016-07-06T08:09:48 sebhub: hi 2016-07-06T08:18:38 Trying to get a pointer to a disk device, but my ioctl call fails. Is this the right approach: http://pastebin.com/JLJ7QRPT ? 2016-07-06T12:12:06 Nevermind. Had not started clock to my sd/emmc subsystem... 2016-07-06T13:34:12 *** deval has joined #rtems 2016-07-06T13:58:10 *** mudit1729 has joined #rtems 2016-07-06T14:00:02 *** vivekk has joined #rtems 2016-07-06T14:01:24 *** deval has quit IRC 2016-07-06T14:01:59 *** salpha has joined #rtems 2016-07-06T14:04:24 hi 2016-07-06T14:04:32 hi 2016-07-06T14:04:42 do we have a gsoc/socis meeting now? 2016-07-06T14:05:20 or the meeting is canceled this week? 2016-07-06T14:06:30 *** deval has joined #rtems 2016-07-06T14:08:02 sorry I haven't heard any issue 2016-07-06T14:13:50 It seems there are some error at rtems homepage 2016-07-06T14:14:33 yh i also can't access it 2016-07-06T14:26:06 *** deval has quit IRC 2016-07-06T14:36:16 *** vivekk has quit IRC 2016-07-06T14:37:12 *** vivekk has joined #rtems 2016-07-06T14:46:02 *** vivekk has quit IRC 2016-07-06T14:56:04 *** DrJoel has joined #rtems 2016-07-06T14:56:04 *** DrJoel has joined #rtems 2016-07-06T14:56:04 *** ChanServ sets mode: +o DrJoel 2016-07-06T14:56:19 Sorry.. I missed the alarm. 2016-07-06T14:56:26 Any GSoC students still here? 2016-07-06T14:56:49 I'm here. Though I just began my dinner 2016-07-06T15:00:10 So, I've been trying to update the scheduler with a data structure that allows me to efficiently implement the algorithm. The algorithm uses a FIFO queue wherein it pushes the CPU node to the queue 2016-07-06T15:00:45 The closest existing structure I found was a Chain_Control, so I am trying to use that right now. 2016-07-06T15:01:46 One question I had was how do I access the Node for the idle thread that is executing on each CPU during the scheduler initialization 2016-07-06T15:03:00 The idea is that when the scheduler is initialized, I create a mapping of CPUs to the thread it is executing. This structure should start with idle threads and will be updated on each task arrival and exit 2016-07-06T15:03:09 DrJoel: ping 2016-07-06T15:04:33 Sorry.. someone walked into my office. I leave for Scotland tomorrow 2016-07-06T15:05:07 The idle threads don't have affinity and float. 2016-07-06T15:05:14 There are just enough idle threads to cover all the cores. 2016-07-06T15:06:19 So, if I want to represent the fact that the CPU is executing an idle thread, should I just use a NULL? 2016-07-06T15:07:32 THe structure is a static array of Thread_Control pointers 2016-07-06T15:07:34 Sounds reasonable. But IDLE threads always have the lowest priority and are preemptible. So if anything can run on that core, your algorithm should just kick it off. 2016-07-06T15:07:52 That's what the existing algorithm does. It doesn't treat idle threads differently. 2016-07-06T15:08:05 This is simply about setting the base case 2016-07-06T15:08:13 AFAIK none of the current schedulers treat idle differently. 2016-07-06T15:08:29 Oh.. "priming" the scheduler. 2016-07-06T15:08:41 Exactly. 2016-07-06T15:08:49 hold on.. let me look at code a second 2016-07-06T15:11:00 You are talking about the start_idle scheduler entry point 2016-07-06T15:11:48 Umm, I'm not sure. Let me try to explain the exact requirements. 2016-07-06T15:13:07 ok.. brb .. this may be better on the mailing list. sebhub is likely not really here and he should see this also 2016-07-06T15:13:31 Sure. I'll write this down on the mailing list today. 2016-07-06T15:13:33 *** tasslehoff has quit IRC 2016-07-06T15:18:06 *** mudit1729 has quit IRC 2016-07-06T15:18:15 back. 2016-07-06T15:20:28 In threadcreateidle.c, the scheduler hook "_Scheduler_Start_idle( scheduler, idle, cpu );" is called for each idle thread created. This is where you prime the scheduler. This is done before tasking starts. 2016-07-06T15:21:21 I'll try hooking into that then 2016-07-06T15:21:21 At this point, the initialization runs on one core determined by the BSP but RTEMS doesn't know what is on any core really. The scheduler is initialized to nothing and that call does the initial assignment of an idle thread to each core. 2016-07-06T15:21:53 When _Thread_Create_idle() is done executing, it will have assigned an idle thread instance to each core. After that, the scheduling mechanics should work as normal. 2016-07-06T15:22:48 At that point I can update my structure 2016-07-06T15:23:04 Yes. Forcibly with no scheduling because tasking hasn't started yet. 2016-07-06T15:23:26 The idle creation process sets things up so you never have to worry about not having a thread to schedule/assign to a core. 2016-07-06T15:24:20 That is exactly what I want to guarrantee as well. 2016-07-06T15:24:56 Every scheduler in RTEMS depends on that. It was an early design decision. There will always be a thread ready to execute. 2016-07-06T15:25:26 I can hook there then. Thanks! 2016-07-06T15:25:26 When we went to SMP, that guarantee was just extended to having enough idle threads so every core was guaranteed to have one ready to execute. 2016-07-06T15:25:36 But it isn't guaranteed which one it will be. 2016-07-06T15:26:34 As long as there's a thread, it's fine 2016-07-06T15:27:21 That's all any of the schedulers need. 2016-07-06T15:27:54 Then I'm good 2016-07-06T15:29:40 Yep. :) 2016-07-06T15:29:48 Any other gsoc students still around? 2016-07-06T15:48:29 yes i'm here 2016-07-06T15:48:38 from socis though. :) 2016-07-06T15:48:52 this week i made vagrant scripts and tested them. they're on the 'dev' branch of slingshot repository now. 2016-07-06T15:49:21 so it should be possible for others to clone and continue from the latest state now. 2016-07-06T15:49:48 right now i'm changing the test suite template so that after individual test case's crash, it can continue execution from the next test case. 2016-07-06T15:49:48 and this implies that test suite should be re-compiled each time after a test case crash, which might considerably increase test execution time. 2016-07-06T15:49:58 this can be fixed by using rtems libdl but i want to keep it for future steps and focus on automating the process instead, even though it be not optimal. 2016-07-06T15:50:58 Sounds like big progress! 2016-07-06T15:51:40 The recompile will add time but how many test cases will this happen to? 2016-07-06T15:53:29 depends 2016-07-06T15:53:42 for some functions, nearly all test cases might fail 2016-07-06T15:53:52 as it was the case for tanh, for instance 2016-07-06T15:54:27 Would it be more efficient to have many small test cases and avoid recompiles? 2016-07-06T15:54:31 Just asking.. 2016-07-06T15:55:15 i think the overhead of invoking qemu for each small test case is more 2016-07-06T15:55:39 therefore the current method that test cases are put in a bigger test suite seems more efficient 2016-07-06T15:56:07 i think the re-compilation overhead would disappear if i use libdl 2016-07-06T15:57:42 That makes sense. Just requires that the target support it and the base application survives the failure. Proper clean up. 2016-07-06T15:58:22 is the support target dependent? 2016-07-06T15:58:42 is pc386 supported? 2016-07-06T15:59:19 or i misunderstood what you said? 2016-07-06T16:00:55 pc should be supported. arm and powerpc are as well. But say nios2 might not work. libdl requires some architecture specific code and it may not exist for all 18 architectures. 2016-07-06T16:01:10 But it exists for all the major ones AFAIK so this would be OK 2016-07-06T16:01:13 btw i can't access devel.rtems.org. is it down? 2016-07-06T16:01:36 great 2016-07-06T16:01:57 Hmm.. was up earlier.. need to ping someone 2016-07-06T16:02:31 looks like part of OSU OSL is down 2016-07-06T16:03:01 what is it? :) 2016-07-06T16:05:10 We are hosted at Oregon State's Open Source Lab. They host a lot of projects. Ironically, I don't think any of us has ever seen the facility the servers are in. RTEMS was hosted here in Huntsville Alabama for about 20 years. Then we moved to a better and faster facility 2016-07-06T16:05:20 But it just looks like devel is down. git.rtems.org is up 2016-07-06T16:05:24 brb 2016-07-06T16:06:50 okay. 2016-07-06T16:06:59 that was all from my side for this week. 2016-07-06T16:21:06 *** jkridner has quit IRC 2016-07-06T16:33:16 sounds like you make good progress. 2016-07-06T16:33:35 Sorry for the delay. A friend who has been ill popped in. 2016-07-06T17:01:16 *** jkridner has joined #rtems 2016-07-06T17:06:56 *** dr__house|away is now known as dr__house 2016-07-06T17:10:17 *** dr__house has quit IRC 2016-07-06T17:10:17 *** dr__house has joined #rtems 2016-07-06T17:32:30 *** jkridner has quit IRC 2016-07-06T17:33:09 *** jkridner has joined #rtems 2016-07-06T17:57:22 *** DrJoel has quit IRC 2016-07-06T18:22:23 *** Davidbrcz has joined #rtems 2016-07-06T18:30:30 *** salpha has quit IRC 2016-07-06T18:46:38 *** mumptai has joined #rtems 2016-07-06T18:53:32 *** jkridner has quit IRC 2016-07-06T19:36:33 *** Davidbrcz has quit IRC 2016-07-06T19:44:48 *** jkridner has joined #rtems 2016-07-06T21:14:44 *** jkridner has quit IRC 2016-07-06T21:20:24 *** jkridner has joined #rtems 2016-07-06T23:14:00 *** stryx` has quit IRC 2016-07-06T23:14:41 *** stryx` has joined #rtems 2016-07-06T23:26:13 *** stryx` has quit IRC 2016-07-06T23:31:59 *** stryx` has joined #rtems 2016-07-07T03:32:05 *** mumptai has quit IRC 2016-07-07T03:43:41 *** mumptai has joined #rtems 2016-07-07T03:47:12 *** indy has quit IRC 2016-07-07T04:04:18 *** ita|zzz is now known as ita 2016-07-07T04:56:00 *** indy has joined #rtems 2016-07-07T08:34:59 *** mumptai has quit IRC 2016-07-07T12:07:50 *** Wonjun__ has joined #rtems 2016-07-07T12:12:03 *** Wonjun_ has quit IRC 2016-07-07T13:25:25 *** benjamint1 has joined #rtems 2016-07-07T13:26:55 *** benjamint has quit IRC 2016-07-07T13:26:55 *** benjamint1 is now known as benjamint 2016-07-07T14:05:53 [trac] #2753/defect (v:4.9) created by merlo (Wrong path to patches) http://devel.rtems.org/ticket/2753 2016-07-07T14:50:10 *** benjamint1 has joined #rtems 2016-07-07T14:52:04 *** benjamint has quit IRC 2016-07-07T14:52:04 *** benjamint1 is now known as benjamint 2016-07-07T15:14:04 *** benjamint has quit IRC 2016-07-07T15:14:19 *** benjamint has joined #rtems 2016-07-07T16:11:07 *** dougk_ has quit IRC 2016-07-07T16:23:11 *** dougk_ has joined #rtems 2016-07-07T17:32:12 *** mumptai has joined #rtems 2016-07-07T19:48:55 *** Noslac has joined #rtems 2016-07-07T19:52:45 *** Noslac has quit IRC 2016-07-07T22:45:42 *** mumptai has quit IRC 2016-07-08T01:15:49 *** dr__house is now known as dr__house|away 2016-07-08T01:37:02 *** hesham has joined #rtems 2016-07-08T07:09:58 *** tasslehoff has joined #rtems 2016-07-08T07:38:27 *** beng-nl has quit IRC 2016-07-08T07:42:48 *** beng-nl has joined #rtems 2016-07-08T07:57:48 I'm writing an app that exposes SD/eMMC as mass storage. It works, but when I try to partition/format from the host it trapt on a cache invalidate operation. Playing with the CONFIGURE_BDBUF_ defines, but I don't understand how to configure them. 2016-07-08T07:58:01 s/trapt/traps 2016-07-08T07:58:56 If I lower CONFIGURE_BDBUF_CACHE_MEMORY_SIZE rtems_io_register_driver fails with an IO error. 2016-07-08T07:59:46 How do I determine what to set for MAX_READ_AHEAD_BLOCKS, MAX_WRITE_BLOCKS, BUFFER_MIN_SIZE, BUFFER_MAX_SIZE and CACHE_MEMORY_SIZE? 2016-07-08T08:10:49 I fill out my rtems_blkdev_request like this: http://pastebin.com/GpuHGDKU 2016-07-08T08:26:44 *** Davidbrcz has joined #rtems 2016-07-08T09:04:53 [trac] SOCIS/2016 edited by salpha (empty comment) http://devel.rtems.org/wiki/SOCIS/2016 2016-07-08T12:11:16 *** modwizcode has quit IRC 2016-07-08T12:11:46 *** modwizcode has joined #rtems 2016-07-08T12:50:59 *** gigetoo has quit IRC 2016-07-08T12:51:26 *** gigetoo has joined #rtems 2016-07-08T13:27:44 *** tasslehoff has quit IRC 2016-07-08T13:56:13 *** gigetoo has quit IRC 2016-07-08T13:56:56 *** gigetoo has joined #rtems 2016-07-08T14:35:58 *** sebhub has quit IRC 2016-07-08T16:49:35 *** dr__house|away is now known as dr__house 2016-07-08T17:11:42 *** Davidbrcz has quit IRC 2016-07-08T17:44:40 *** benjamint has quit IRC 2016-07-09T04:43:15 [git] ac0e5bf by Chris Johns: [Update to bring in the --show-commands fix.] http://devel.rtems.org/changeset/ac0e5bf/rtems-libbsd 2016-07-09T08:40:44 *** ita has quit IRC 2016-07-09T13:48:49 *** ita has joined #rtems 2016-07-09T23:38:52 *** jkridner has quit IRC 2016-07-10T04:59:55 *** darnir has quit IRC 2016-07-10T05:00:01 *** hesham has quit IRC 2016-07-10T05:00:02 *** dougk_ has quit IRC 2016-07-10T05:00:03 *** danieli has quit IRC 2016-07-10T05:00:04 *** cdesai has quit IRC 2016-07-10T05:00:05 *** edwardk has quit IRC 2016-07-10T05:00:08 *** mithro has quit IRC 2016-07-10T05:00:09 *** kiwichris has quit IRC 2016-07-10T05:00:14 *** bbhoss has quit IRC 2016-07-10T08:53:53 *** darnir has joined #rtems 2016-07-10T08:53:58 *** hesham has joined #rtems 2016-07-10T08:53:59 *** dougk_ has joined #rtems 2016-07-10T08:53:59 *** danieli has joined #rtems 2016-07-10T08:53:59 *** cdesai has joined #rtems 2016-07-10T08:53:59 *** edwardk has joined #rtems 2016-07-10T08:53:59 *** mithro has joined #rtems 2016-07-10T08:53:59 *** kiwichris has joined #rtems 2016-07-10T08:53:59 *** bbhoss has joined #rtems 2016-07-10T20:28:35 *** mumptai has joined #rtems 2016-07-10T20:29:32 *** Davidbrcz has joined #rtems 2016-07-10T20:44:52 *** Davidbrcz has quit IRC 2016-07-10T22:56:45 *** mumptai has quit IRC