Re: RSB documentation

2019-01-17 Thread Sebastian Huber

On 17/01/2019 07:36, Chris Johns wrote:

On 17/1/19 5:19 pm, Sebastian Huber wrote:

On 17/01/2019 06:45, Chris Johns wrote:

On 14/1/19 5:36 pm, Sebastian Huber wrote:

I added a build of the partially restructured user manual here:

https://ftp.rtems.org/pub/rtems/people/sebh/user.pdf


Any HTML?

https://ftp.rtems.org/pub/rtems/people/sebh/user/index.html


Looks good. I like the arrangement and flow.



Thanks for the review, I checked in the patches. I will now continue to 
work on the Quick Start chapter.


With respect to an example RTEMS application. Should we

(1) suggest to check out the examples repository and start from here, or

(2) should we provide a very simple Makefile based project?

I tend to option (2) and just mention that the examples repository exists.

--
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 geschäftliche Mitteilung im Sinne des EHUG.

___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

FreeBSD 12 stable branch for libbsd

2019-01-17 Thread Sebastian Huber

Hello,

I would like to add a "freebsd-12" branch to libbsd which tracks the 
FreeBSD 12 stable branch instead of the trunk. I would like to figure 
out if it is possible to maintain this branch more easily for production 
systems. The libbsd master will still track the trunk.


--
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 geschäftliche Mitteilung im Sinne des EHUG.

___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

Re: [PATCH 5/5] Add bsp riscv/grlib

2019-01-17 Thread Jiri Gaisler

On 1/17/19 8:04 AM, Sebastian Huber wrote:
> Hello Jiri,
>
> it would be nice if we could use the new license for this new BSP:
>
> https://git.rtems.org/rtems/tree/LICENSE.BSD-2-Clause
>
OK.


> BSP names should be unique even across architectures. So, I think the name 
> grlib is a bit too unspecific. What about griscv?

Very clever - I like that  ...!

I will rework the patch set according to the various comments - stay tuned ...

Jiri.

>
> On 16/01/2019 15:44, Jiri Gaisler wrote:
>>   bsps/riscv/grlib/clock/clockdrv.c | 203 +
>>   bsps/riscv/grlib/config/grlib.cfg |   9 +
>>   bsps/riscv/grlib/console/console.c    | 162 
>>   bsps/riscv/grlib/console/printk_support.c | 126 ++
>>   bsps/riscv/grlib/headers.am   |  16 +
>>   bsps/riscv/grlib/include/amba.h   | 143 +++
>>   bsps/riscv/grlib/include/bsp.h    |  68 
>>   bsps/riscv/grlib/include/bsp/irq.h    |  73 
>>   bsps/riscv/grlib/include/bsp/riscv.h  |  57 +++
>
> Do you need this riscv.h?
>
>>   bsps/riscv/grlib/include/dev/serial/htif.h    |  54 +++
>
> Do you need this htif.h?
>
>>   bsps/riscv/grlib/include/tm27.h   |   1 +
>>   bsps/riscv/grlib/irq/irq.c    | 145 +++
>>   bsps/riscv/grlib/start/amba.c | 115 ++
>>   bsps/riscv/grlib/start/bsp_fatal_halt.c   |  34 ++
>>   bsps/riscv/grlib/start/bsp_specs  |   9 +
>>   bsps/riscv/grlib/start/bspsmp.c   |  98 +
>>   bsps/riscv/grlib/start/bspstart.c |  71 
>>   bsps/riscv/grlib/start/linkcmds.base  | 385 ++
>
> The linkcmds.base should be definitely shared with the generic existing BSP 
> family and move to bsps/riscv/shared/start/linkcmds.base.
>
>>   bsps/riscv/grlib/start/linkcmds.in    |  46 +++
>>   bsps/riscv/grlib/start/start.S    | 142 +++
>
> The start.S should be shared with the existing BSP family and move to 
> bsps/riscv/shared/start/start.S.
>
___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

[PATCH] Change first copyright year to 1988

2019-01-17 Thread Sebastian Huber
The OAR copyright goes back to 1988.
---
 common/conf.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/common/conf.py b/common/conf.py
index 5be4568..c658113 100644
--- a/common/conf.py
+++ b/common/conf.py
@@ -24,7 +24,7 @@ master_doc = 'index'
 
 # General information about the project.
 project = u'RTEMS Documentation Project'
-copyright = u'2018, 2019 RTEMS Project and contributors'
+copyright = u'1988, 2019 RTEMS Project and contributors'
 
 # The version info for the project you're documenting, acts as replacement for
 # |version| and |release|, also used in various other places throughout the
-- 
2.16.4

___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel


Re: RSB documentation

2019-01-17 Thread Chris Johns
On 17/1/19 7:15 pm, Sebastian Huber wrote:
> 
> With respect to an example RTEMS application. Should we
> 
> (1) suggest to check out the examples repository and start from here, or
> 
> (2) should we provide a very simple Makefile based project?
> 
> I tend to option (2) and just mention that the examples repository exists.
> 

This is a complicated problem to solve because of the ABI related issues and the
need for specific flags for each BSP. If you provide explicit values they will
be right for some and wrong for others.

I think we need a section on this topic and we provide the support for build
systems users are willing to submit and support however we need to consider
tools to help this happen.

The long term plan is to provide a command called 'rtems-config' in rtems-tools
that it like pkgconfig. You query it for the flags you need. It is basically the
internals of rtems_waf but brought out to be a specific command. This can then
be used to create a simple Makefile based application we know will work for all
BSPs and architectures.

After the kernel we have 3rd party packages such as libbsd and libdl so we need
to consider how we expand what we support. With this in mind I have recently
been adding support to rtems_waf ...

 https://git.rtems.org/chrisj/rtems_waf.git/?h=libbsd-libdl

This should not be just about waf and rtems_waf, all solutions should be built
on a common framework that encapsulates the RTEMS complexity.

We need to be careful with these examples because users will copy what we do and
this means what we do becomes copied and magnified.

Chris
___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel


[PATCH 1/2] Change Linux default prefix to "/opt"

2019-01-17 Thread Sebastian Huber
http://refspecs.linuxbase.org/FHS_3.0/fhs/ch03s13.html#purpose14
---
 source-builder/sb/linux.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/source-builder/sb/linux.py b/source-builder/sb/linux.py
index f270121..367532a 100644
--- a/source-builder/sb/linux.py
+++ b/source-builder/sb/linux.py
@@ -63,6 +63,7 @@ def load():
 '_host_arch':   ('none','none', cpu),
 '_usr': ('dir', 'required', '/usr'),
 '_var': ('dir', 'required', '/var'),
+'_prefix':  ('dir', 'optional', '/opt'),
 '__bzip2':  ('exe', 'required', '/usr/bin/bzip2'),
 '__gzip':   ('exe', 'required', '/bin/gzip'),
 '__tar':('exe', 'required', '/bin/tar')
-- 
2.16.4

___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel


[PATCH 2/2] Change default prefix

2019-01-17 Thread Sebastian Huber
Use OS prefix + "rtems" + $rtems_version as the default prefix to
automatically separate different RTEMS versions.

Close #3675.
---
 source-builder/sb/options.py | 6 ++
 1 file changed, 6 insertions(+)

diff --git a/source-builder/sb/options.py b/source-builder/sb/options.py
index c5fcabe..e08b088 100644
--- a/source-builder/sb/options.py
+++ b/source-builder/sb/options.py
@@ -307,6 +307,12 @@ class command_line:
 # Must have a host
 if self.defaults['_build'] == self.defaults['nil']:
 raise error.general('--build not set')
+# Default prefix
+prefix = self.parse_args('--prefix')
+if prefix is None:
+value = path.join(self.defaults['_prefix'], 'rtems', 
str(version.version()))
+self.opts['prefix'] = value
+self.defaults['_prefix'] = value
 # Manage the regression option
 if self.opts['regression'] != '0':
 self.opts['no-install'] = '1'
-- 
2.16.4

___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel


Re: [PATCH 1/2] Change Linux default prefix to "/opt"

2019-01-17 Thread Chris Johns
Ok to push.

On 17/1/19 9:34 pm, Sebastian Huber wrote:
> http://refspecs.linuxbase.org/FHS_3.0/fhs/ch03s13.html#purpose14
> ---
>  source-builder/sb/linux.py | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/source-builder/sb/linux.py b/source-builder/sb/linux.py
> index f270121..367532a 100644
> --- a/source-builder/sb/linux.py
> +++ b/source-builder/sb/linux.py
> @@ -63,6 +63,7 @@ def load():
>  '_host_arch':   ('none','none', cpu),
>  '_usr': ('dir', 'required', '/usr'),
>  '_var': ('dir', 'required', '/var'),
> +'_prefix':  ('dir', 'optional', '/opt'),
>  '__bzip2':  ('exe', 'required', '/usr/bin/bzip2'),
>  '__gzip':   ('exe', 'required', '/bin/gzip'),
>  '__tar':('exe', 'required', '/bin/tar')
> 
___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel


Re: [PATCH 5/5] Add bsp riscv/grlib

2019-01-17 Thread Jiri Gaisler


On 1/17/19 8:04 AM, Sebastian Huber wrote:
> Hello Jiri,
>
> it would be nice if we could use the new license for this new BSP:
>
I realized that the most of my new bsp (>90%) is actually copied/tweaked from 
the other riscv and leon3 bsps. I don't think it is right to change the license 
since I am for the most part not the original author. I have maintained the 
original license headers in all files without adding myself as contributor. I 
suggest we keep it like that until the original authors agree to the BSD 
license ..?

Jiri.


___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel


Re: [PATCH] Change first copyright year to 1988

2019-01-17 Thread Chris Johns
On 17/1/19 7:45 pm, Sebastian Huber wrote:
> The OAR copyright goes back to 1988.

Yes, this is a good idea.

Thanks
Chris
___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel