Re: rsb to build 4.10 tools with modern compilers (ie under MSYS2)

On 28/05/2016 1:04 AM, Worth Burruss wrote: Chris, If you would prefer this as a ticket let me know. Yes a ticket would be good. Thanks. Chris ___ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel

Re: GSOC2016: Improvements in RTEMS Tracing Framework

On 26/05/2016 10:46 PM, vivek kukreja wrote: I'm facing a problem with function tracing. I made changes to rtems-tld Can you please send me a diff off-line? Thanks. command to ouput a CTF metadata file but i need sizes of arguments and return values at compile-time to be put in the metadata f

Re: GSOC2016: Improvements in RTEMS Tracing Framework

On 27/05/2016 4:22 PM, vivek kukreja wrote: Now the ctrace command returns the following error: trace read failed: RTEMS_UNSATISFIED. Is there enough resources defined in your init.c where confdefs.h is included? I think the problem lies in rtems_capture_read function call in capture_suppo

Re: [PATCH v2 0/7] Raspberry Pi VideoCore support for console on HDMI output

Hi Pavel, I will try to run the tests with the HDMI output before Monday. If I get time, I’ll try the Raspberry Pi zero and also see if I can narrow down the firmware issue. Eventually I would like to try the u-boot option as well as JTAG debugging. Alan > On May 27, 2016, at 10:18 AM, Pavel

Re: PWM driver tested in RTEMS with RGB

On Thu, May 12, 2016 at 6:55 PM, Martin Galvan wrote: > Hi Punit! Sorry I didn't answer, I was off sick for a couple days. Marcos > and I are a bit crowded so we can't review every bit of your patch right now > (plus we don't have our BBB anymore). Could you make the RGB test work using > SW code

Re: LM3S69xx BSP Qemu Error

Hi Martin, Many thanks for your response. I have build qemu by hand and also rebuild rtems with the configure line provided ybut still getting the same message. Can you please also specify the command line argument used to run hello.exe. Let me try that to see if I have some parameters in my com

Re: [civetweb] Interface for setting stack size and scheduler options

I updated the Interface (header) in my draft. Could you take a look at the improved version? https://github.com/c-mauderer/civetweb/compare/1e4a9ee...a6bea13 Please note: I only updated the interface not the implementation. I will do that as soon as we have agreed on the interface.

Re: LM3S69xx BSP Qemu Error

Hi Martin, I am able to run HelloWorld with what you provided. I also noticed that the process does not terminate by itself and I have to manually kill it. I noticed that for the HelloWorld and ticker. Does it behave like that when you run it? Below the output of my arm-rtems4.12-gcc --version.

Re: LM3S69xx BSP Qemu Error

On Fri, May 27, 2016 at 12:01 PM, Olufowobi, Habeeb wrote: > Hi Martin, > > I am able to run HelloWorld with what you provided. I also noticed that the > process does not terminate by itself and I have to manually kill it. I > noticed that for the HelloWorld and ticker. Does it behave like that wh

Re: Ada support on gcc-6 broken again

- Joel Sherrill schrieb: > Did it again. :( > > Two days of moving with a rental truck must have be rusty. :( > > On Fri, May 27, 2016 at 9:08 AM, Jan Sommer > wrote: > > > Am 2016-05-27 14:44, schrieb Joel Sherrill: > > > >> On May 27, 2016 7:17 AM, "Jan Sommer" [...] > > It probably sh

rsb to build 4.10 tools with modern compilers (ie under MSYS2)

Attached are 3 patches I needed to build 4.10 tools for coldfire processors (m68k) under windows (7 and 10) using MSYS2. I have been meaning to send them for awhile. The primary patch is for GCC and is needed to build older GCC with newer GCC. I found this patch on the GCC mailing list and

Re: LM3S69xx BSP Qemu Error

On Fri, May 27, 2016 at 11:15 AM, Olufowobi, Habeeb wrote: > Can you please also specify the command line argument used to run hello.exe. > Let me try that to see if I have some parameters in my command line that was > not supposed to be there. Certainly. I'm running the following: ./bin/qemu-sy

Re: Ada support on gcc-6 broken again

Did it again. :( Two days of moving with a rental truck must have be rusty. :( On Fri, May 27, 2016 at 9:08 AM, Jan Sommer wrote: > Am 2016-05-27 14:44, schrieb Joel Sherrill: > >> On May 27, 2016 7:17 AM, "Jan Sommer" >> wrote: >> >>> >>> Hello, >>> >>> building the current tools (gcc-6) from

Re: [PATCH v2 0/7] Raspberry Pi VideoCore support for console on HDMI output

Hello all, patches seems to not change situation/make problems so if there is no remark I would push them on Monday. Do you agree? The RPi1 and RPi2 testing state: Tested to work on RPi1 with serial and graphic output and direct boot and boot over U-boot. Reported to work on RPi2 by Alan Cudmore

Re: Ada support on gcc-6 broken again

Am 2016-05-27 14:44, schrieb Joel Sherrill: On May 27, 2016 7:17 AM, "Jan Sommer" wrote: Hello, building the current tools (gcc-6) from master with Ada support fails. I will try to post the patch for that to gcc over the weekend (@Joel I hope you don't mind me putting you CC for that). Plea

[PATCH 03/12] score: Use thread queue lock for MrsP

Replace the ISR lock in MRSP_Control with a thread queue. This simplifies the Classic semaphore implementation. Only the lock part of the thread queue is used. --- cpukit/rtems/src/semobtain.c| 7 +++ cpukit/score/include/rtems/score/mrsp.h | 15 +-- cpukit/s

[PATCH 08/12] rtems: Simplify rtems_semaphore_delete()

Due to the unified status codes, we can now use a common flush and destroy method for semaphore and mutex variants. --- cpukit/rtems/src/semdelete.c | 17 ++--- 1 file changed, 6 insertions(+), 11 deletions(-) diff --git a/cpukit/rtems/src/semdelete.c b/cpukit/rtems/src/semdelete.c in

[PATCH 06/12] score: Simplify CORE mutex

Remove superfluous support for simple binary semaphores. With this we can get rid of the CORE_MUTEX_NESTING_BLOCKS variant. --- cpukit/rtems/include/rtems/rtems/semimpl.h | 1 + cpukit/rtems/src/semcreate.c | 43 cpukit/rtems/src/semdelete.c

[PATCH 07/12] rtems: Simplify rtems_semaphore_flush()

Due to the unified status codes, we can now use a common flush method for semaphore and mutex variants. --- cpukit/rtems/src/semflush.c | 15 +-- 1 file changed, 5 insertions(+), 10 deletions(-) diff --git a/cpukit/rtems/src/semflush.c b/cpukit/rtems/src/semflush.c index 2410ed2..0db8

[PATCH 12/12] score: Add CORE mutex variants

Add CORE_recursive_mutex_Control and CORE_ceiling_mutex_Control to avoid the run-time evaluation of attributes to figure out how a particular mutex methods should behave. Start with the no protocol variants. This eliminates the CORE_MUTEX_DISCIPLINES_FIFO and CORE_MUTEX_DISCIPLINES_PRIORITY. ---

[PATCH 10/12] posix: Avoid use of internal mutex methods

Avoid use of internal mutex methods for pthread_mutex_setprioceiling(). --- cpukit/posix/src/mutexsetprioceiling.c | 31 ++- 1 file changed, 10 insertions(+), 21 deletions(-) diff --git a/cpukit/posix/src/mutexsetprioceiling.c b/cpukit/posix/src/mutexsetprioceiling.c

[PATCH 02/12] score: _CORE_mutex_Check_dispatch_for_seize()

Move the safety check performed by _CORE_mutex_Check_dispatch_for_seize() out of the performance critical path and generalize it. Blocking on a thread queue with an unexpected thread dispatch disabled level is illegal in all system states. Add the expected thread dispatch disable level (which may

[PATCH 01/12] score: Add _Thread_queue_Context_set_MP_callout()

Add _Thread_queue_Context_set_MP_callout() to simplify _Thread_queue_Context_initialize(). This makes it possible to more easily add additional fields to Thread_queue_Context. --- cpukit/libnetworking/rtems/rtems_glue.c | 4 +- cpukit/posix/include/rtems/posix/barrierimpl.h | 2 +- c

[PATCH 09/12] posix: Avoid use of internal mutex methods

Avoid use of internal mutex methods for condition variables. --- cpukit/posix/src/condwaitsupp.c | 36 +--- 1 file changed, 13 insertions(+), 23 deletions(-) diff --git a/cpukit/posix/src/condwaitsupp.c b/cpukit/posix/src/condwaitsupp.c index 7dff27f..ebcb3c4 10064

[PATCH 11/12] posix: Delete POSIX_Mutex_Protocol::process_shared

--- cpukit/posix/include/rtems/posix/mutex.h | 2 -- cpukit/posix/src/mutexinit.c | 2 -- 2 files changed, 4 deletions(-) diff --git a/cpukit/posix/include/rtems/posix/mutex.h b/cpukit/posix/include/rtems/posix/mutex.h index bc8e529..e1dfa34 100644 --- a/cpukit/posix/include/rtems/po

[PATCH 05/12] score: Add semaphore variants

--- cpukit/posix/include/rtems/posix/semaphoreimpl.h | 2 + cpukit/posix/src/semaphorecreatesupp.c | 6 +-- cpukit/posix/src/semaphoredeletesupp.c | 6 ++- cpukit/posix/src/semaphorewaitsupp.c | 1 + cpukit/posix/src/sempost.c | 1 + cpuki

[PATCH 04/12] rtems: Simplify rtems_semaphore_delete()

--- cpukit/rtems/include/rtems/rtems/sem.h | 5 cpukit/rtems/src/semdelete.c | 46 +- cpukit/rtems/src/semobtain.c | 5 3 files changed, 28 insertions(+), 28 deletions(-) diff --git a/cpukit/rtems/include/rtems/rtems/sem.h b/cpukit/

Re: Ada support on gcc-6 broken again

Used the wrong from address and it bounced to the list. :( On Fri, May 27, 2016 at 7:16 AM, Jan Sommer wrote: > Hello, > > building the current tools (gcc-6) from master with Ada support fails. > Is this the only branch that is broken? Or is the gcc master broken as well. > I will try to post

Ada support on gcc-6 broken again

Hello, building the current tools (gcc-6) from master with Ada support fails. I will try to post the patch for that to gcc over the weekend (@Joel I hope you don't mind me putting you CC for that). However as gcc-6.1 was just released a month ago, I guess it won't be part of an official gcc-rel

Re: Skip Building a few packages by RSB

If you are build RTEMS multiple times, then you should just build it by hand. The tools themselves news to get built much less frequently. Why do you need to build RTEMS so much is my question. I think you should not and install RTEMS and then focus on Rock. You should only need to rebuild RTEMS

Skip Building a few packages by RSB

Hi, I am working on porting Rock to RTEMS as a Summer of code project.I have to modify the code and build RTEMS using RSB a lot many times. RSB is designed to clean install every time but building newlib and gcc each time is a very time consuming task. Is there a way I can select what packages to b