[PATCH v2 1/2] Add FDT support for Beaglebone Black.

--- c/src/lib/libbsp/arm/beagle/Makefile.am| 1 + c/src/lib/libbsp/arm/beagle/configure.ac | 9 + c/src/lib/libbsp/arm/beagle/include/bsp.h | 6 +- c/src/lib/libbsp/arm/beagle/startup/bspstart.c | 5 + c/src/lib/libbsp/arm/shared/start/start.S | 5 + 5 f

[PATCH v2 2/2] Add dts file to generate dtb binary file for Beaglebone black.

These dts files import from FreeBSD, git link: https://github.com/freebsd/freebsd/tree/master/sys/gnu/dts The license for these files in beagle/simscripts --- c/src/lib/libbsp/arm/beagle/README | 11 + c/src/lib/libbsp/arm/beagle/simscripts/LICENSE |2 + .../arm/beagle/s

[PATCH 1/6] Add support for multiple memory resources.

From: Hans Petter Selasky --- rtemsbsd/rtems/rtems-kernel-nexus.c | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/rtemsbsd/rtems/rtems-kernel-nexus.c b/rtemsbsd/rtems/rtems-kernel-nexus.c index 008e4b2..1a5a9ef 100644 --- a/rtemsbsd/rtems/rtems-kernel-nexus.c +++ b/rte

[PATCH 2/6] Modify the support for multiple memory resources.

rtems-libbsd only support single resource for a device. But some usb control need two resources, such as musb. So modify support for multiple device resources allocation. --- rtemsbsd/rtems/rtems-kernel-nexus.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rtemsbsd/rtems/rtem

[PATCH 4/6] Port am335x usb driver to RTEMS.

--- freebsd/sys/arm/ti/am335x/am335x_prcm.c | 8 freebsd/sys/arm/ti/ti_cpuid.h | 19 +++ freebsd/sys/arm/ti/ti_prcm.c| 2 ++ freebsd/sys/arm/ti/ti_scm.c | 2 ++ libbsd.py | 30 ++

[PATCH 5/6] Add FDT support for am335x USB driver.

Now RTEMS can detect and mount USB device. --- rtemsbsd/include/bsp/nexus-devices.h | 13 + 1 file changed, 13 insertions(+) diff --git a/rtemsbsd/include/bsp/nexus-devices.h b/rtemsbsd/include/bsp/nexus-devices.h index 1fbf756..2d5694e 100644 --- a/rtemsbsd/include/bsp/nexus-devices

[PATCH 6/6] Add umass support for Beaglebone black.

Now RTEMS can mount and open USB disk. --- rtemsbsd/include/bsp/nexus-devices.h | 1 + 1 file changed, 1 insertion(+) diff --git a/rtemsbsd/include/bsp/nexus-devices.h b/rtemsbsd/include/bsp/nexus-devices.h index 2d5694e..09a4cc3 100644 --- a/rtemsbsd/include/bsp/nexus-devices.h +++ b/rtemsbsd/i

[PATCH 2/7] rtems: Add rtems_scheduler_ident_by_processor_set

Update #3070. --- cpukit/rtems/Makefile.am| 1 + cpukit/rtems/include/rtems/rtems/tasks.h| 25 +++ cpukit/rtems/src/scheduleridentbyprocessorset.c | 60 + testsuites/smptests/smpscheduler02/init.c | 11 + testsuites/sptests

[PATCH 5/7] Add interrupt server set affinity

--- c/src/lib/libbsp/shared/src/irq-server.c | 33 + cpukit/include/rtems/irq-extension.h | 36 2 files changed, 69 insertions(+) diff --git a/c/src/lib/libbsp/shared/src/irq-server.c b/c/src/lib/libbsp/shared/src/irq-server.c inde

[PATCH 4/7] Add interrupt server suspend/resume

This mechanism can be used to safely move the interrupt server from one scheduler instance to another for example. --- c/src/lib/libbsp/shared/src/irq-server.c | 61 ++-- cpukit/include/rtems/irq-extension.h | 39 cpukit/rtems/include/rtems/rtem

[PATCH 7/7] Add interrupt vector set/get affinity

--- c/src/lib/libbsp/arm/shared/arm-gic-irq.c | 24 ++ c/src/lib/libbsp/arm/shared/include/arm-gic-irq.h | 9 +++- .../lib/libbsp/powerpc/qoriq/clock/clock-config.c | 16 +++ c/src/lib/libbsp/powerpc/qoriq/include/irq.h | 8 ++-- c/src/lib/libbsp/powerpc/qoriq/irq/ir

[PATCH 1/7] rtems: Add rtems_scheduler_ident_by_processor()

Update #3069. --- cpukit/rtems/Makefile.am | 1 + cpukit/rtems/include/rtems/rtems/tasks.h | 17 ++ cpukit/rtems/src/scheduleridentbyprocessor.c | 49 testsuites/smptests/smpfatal04/init.c| 6 testsuites/sptests/spschedule

[PATCH 6/7] Add interrupt server move

--- c/src/lib/libbsp/shared/src/irq-server.c | 75 cpukit/include/rtems/irq-extension.h | 26 +++ 2 files changed, 101 insertions(+) diff --git a/c/src/lib/libbsp/shared/src/irq-server.c b/c/src/lib/libbsp/shared/src/irq-server.c index 61d31754c2..a7b

[PATCH 3/7] Create one interrupt server per processor

This allows load balancing of interrupt processing in SMP configurations. --- c/src/lib/libbsp/shared/src/irq-server.c | 263 +-- cpukit/include/rtems/irq-extension.h | 129 --- 2 files changed, 254 insertions(+), 138 deletions(-) diff --git a/c/src/lib

Re: [PATCH 1/7] rtems: Add rtems_scheduler_ident_by_processor()

On Tue, Jul 11, 2017 at 9:41 AM, Sebastian Huber wrote: > Update #3069. > --- > cpukit/rtems/Makefile.am | 1 + > cpukit/rtems/include/rtems/rtems/tasks.h | 17 ++ > cpukit/rtems/src/scheduleridentbyprocessor.c | 49 > > testsuites/sm

Re: [PATCH 2/7] rtems: Add rtems_scheduler_ident_by_processor_set

On Tue, Jul 11, 2017 at 9:41 AM, Sebastian Huber wrote: > Update #3070. > --- > cpukit/rtems/Makefile.am| 1 + > cpukit/rtems/include/rtems/rtems/tasks.h| 25 +++ > cpukit/rtems/src/scheduleridentbyprocessorset.c | 60 > + > testsu

Re: [PATCH 3/7] Create one interrupt server per processor

There should probably be a ticket for this change. Is there a reason someone might have to not want one server per core when using SMP? That is, should this be configurable? On Tue, Jul 11, 2017 at 9:41 AM, Sebastian Huber wrote: > This allows load balancing of interrupt processing in SMP > conf

Re: [PATCH 3/7] Create one interrupt server per processor

On 11/07/17 17:06, Gedare Bloom wrote: Is there a reason someone might have to not want one server per core when using SMP? That is, should this be configurable? Yes, this should be configurable. One option is to add an alternative initialization routine that uses a configuration table. Howev

Re: [PATCH 2/7] rtems: Add rtems_scheduler_ident_by_processor_set

On 11/07/17 16:59, Gedare Bloom wrote: +rtems_status_code rtems_scheduler_ident_by_processor_set( + size_t cpusetsize, + const cpu_set_t *cpuset, + rtems_id*id +); + Also needs doc. I wonder if "identify" or "id" may be better. "ident" seems a bit unusual. All the object

Re: [PATCH 3/7] Create one interrupt server per processor

On 12/07/2017 15:35, Sebastian Huber wrote: > On 11/07/17 17:06, Gedare Bloom wrote: > >> Is there a reason someone might have to not want one server per core >> when using SMP? That is, should this be configurable? > > Yes, this should be configurable. One option is to add an alternative > initi

Re: [PATCH 3/7] Create one interrupt server per processor

On 12/07/17 08:06, Chris Johns wrote: On 12/07/2017 15:35, Sebastian Huber wrote: On 11/07/17 17:06, Gedare Bloom wrote: Is there a reason someone might have to not want one server per core when using SMP? That is, should this be configurable? Yes, this should be configurable. One option is

Re: [PATCH v2 1/2] Add FDT support for Beaglebone Black.

I checked in a slightly modified version: http://git.rtems.org/rtems/commit/?id=f6115d7cd2d1551a05cd257ac9f00476c61e5e0d -- Sebastian Huber, embedded brains GmbH Address : Dornierstr. 4, D-82178 Puchheim, Germany Phone : +49 89 189 47 41-16 Fax : +49 89 189 47 41-09 E-Mail : sebastian.hu

Re: [PATCH 6/6] Add umass support for Beaglebone black.

Please merge with patch 5. -- Sebastian Huber, embedded brains GmbH Address : Dornierstr. 4, D-82178 Puchheim, Germany Phone : +49 89 189 47 41-16 Fax : +49 89 189 47 41-09 E-Mail : sebastian.hu...@embedded-brains.de PGP : Public key available on request. Diese Nachricht ist keine ge

Re: [PATCH 2/6] Modify the support for multiple memory resources.

Is patch 1 and 2 still needed for the later patches? If not, then please drop them. -- Sebastian Huber, embedded brains GmbH Address : Dornierstr. 4, D-82178 Puchheim, Germany Phone : +49 89 189 47 41-16 Fax : +49 89 189 47 41-09 E-Mail : sebastian.hu...@embedded-brains.de PGP : Publ

Re: [PATCH 4/6] Port am335x usb driver to RTEMS.

On 11/07/17 12:53, Sichen Zhao wrote: +#ifndef __rtems__ #include +#endif /* __rtems__ */ https://github.com/RTEMS/rtems-libbsd/blob/master/CONTRIBUTING.md "In general, provide empty header files and do not guard includes." Each #if __rtems__ makes it harder to stay in synchronization wit

Re: [PATCH 5/6] Add FDT support for am335x USB driver.

Please merge with patch 4. -- Sebastian Huber, embedded brains GmbH Address : Dornierstr. 4, D-82178 Puchheim, Germany Phone : +49 89 189 47 41-16 Fax : +49 89 189 47 41-09 E-Mail : sebastian.hu...@embedded-brains.de PGP : Public key available on request. Diese Nachricht ist keine ge

Re: [PATCH v2 1/2] Add FDT support for Beaglebone Black.

Ok, thanks. Best Regards Sichen Zhao From: devel on behalf of Sebastian Huber Sent: Wednesday, July 12, 2017 2:40:28 PM To: Sichen Zhao; devel@rtems.org Cc: punitv...@gmail.com; christian.maude...@embedded-brains.de Subject: Re: [PATCH v2 1/2] Add FDT s