i.MX6ULL BSP ?

[Ian Caddy]

Hi All,

We are starting a new board development using an i.MX6ULL (MCIMX6Y2) 
with a single ARM Cortex-A7 core and was wondering if anyone has already 
created a BSP for this or something similar to this, and would be 
willing to share some or all of it?


regards,

Ian Caddy

--
Ian Caddy
Goanna Technologies Pty Ltd
+61 8 9444 2634

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

Re: Re: how to use system function to execute the script

[bin.w...@qkmtech.com]

thanks，it works



bin.w...@qkmtech.com
 
From: Chris Johns
Date: 2018-06-22 13:07
To: bin.w...@qkmtech.com; Users
Subject: Re: how to use system function to execute the script
On 22/06/2018 14:44, bin.w...@qkmtech.com wrote:
> hi everyone:
> 
> i want to use the "system()" to execute the script. but do not has output
> 
> first i want to use "system()" to execute the shell cmd, but alse have no 
> output
> 
> the test code is as followed:
> 
> system("ls -l");
> system("joel /flash/script/spdloop_1");//
>  "/flash/script/spdloop_1"--this is the  script file path in the 
> filesystem 
>   
 
RTEMS is a single address space, single process OS so you can not exec another
process. You can call the shell command handler from any thread. Make sure you
have enough stack for the thread.
 
Try something like:
 
int run_shell_command(const char* cmd)
{
  #define CMD_MAX_ARGS (20)
  char* cmd_argv = strdup(cmd);
  int   argc;
  char* argv[CMD_MAX_ARGS];
  int   r;
 
  if (cmd_argv == NULL) {
fprintf(stderr, "shell: command: no memory: %s", cmd);
return 1;
  }
 
  if (rtems_shell_make_args(cmd_argv, , argv, CMD_MAX_ARGS) < 0) {
fprintf(stderr, "shell: command: arg parse: %s", cmd);
free(cmd_argv);
return 1;
  }
 
  r = rtems_shell_execute_cmd(argv[0], argc, argv);
  free(cmd_argv);
  return r;
}
 
Chris
 
___
users mailing list
users@rtems.org
http://lists.rtems.org/mailman/listinfo/users

Re: Error building libbsd

[Nicolas Tsiogkas]

Hi Christian,

I will go with python3 indeed as I have to finally close a patch for
Ethercat SOEM working with RTEMS 5. In parallel I will try to investigate
with the waf examples and report back.

Thanks for the prompt reply and effort. :)

Regards,
Niko

On Mon, Jun 25, 2018 at 10:15 AM Christian Mauderer <
christian.maude...@embedded-brains.de> wrote:

> Hello Niko,
>
> if the patch didn't do anything to your output, the error has to happen
> quite early. The first one should have been printed when the build()
> function of the wscript has been reached. So I would assume that it
> already happens before that.
>
> If you have time and desire to find the bug, you might could try to
> build some other tools that use waf as a build system (for example the
> waf examples from the waf website). Otherwise I would suggest to go with
> python3 for now.
>
> Note that it shouldn't really be a problem with python2 in general: My
> OpenSUSE machine is using python2 too and has no problems with the
> current libbsd. As far as I know most distributions still use python2 as
> the default case. I only know Arch Linux that does use python3.
>
> Best regards
>
> Christian Mauderer
>
> Am 25.06.2018 um 10:08 schrieb Nicolas Tsiogkas:
> > Hi,
> >
> > sorry for the late reply.
> >
> > So, the patch didn't do a thing. Got the same print.
> >
> > But I tried to run waf with python3
> >
> > python3 /home/niko/.local/bin/waf-2.0.8
> >
> > and
> >
> > python3 /home/niko/.local/bin/waf-2.0.8 install
> >
> > That was successful. From that I can assume that there is some issue
> > with the python2 shipped with ubuntu? I can try at home as well in case
> > there is something specific with the python configuration on my current
> > pc and report back.
> >
> > Thanks for the help.
> >
> > Regards,
> > Niko
> >
> > On Fri, Jun 22, 2018 at 9:56 PM Christian Mauderer  > > wrote:
> >
> > Hello Niko,
> >
> > Am 22.06.2018 um 13:23 schrieb Nicolas Tsiogkas:
> > > Hi,
> > >
> > > waf was downloaded by the website (waf.io 
> > )
> > >
> > > niko@im-desktop-021:~/sandbox/rtems-libbsd$ which waf-2.0.8
> > > /home/niko/.local/bin/waf-2.0.8
> >
> > Should be OK.
> >
> > >
> > > running also with waf-2.0.4 which used to work gives the same error
> > >
> > > ~/sandbox/rtems-libbsd$ waf-2.0.8 -v
> > > Could not create the directory ///h
> > >
> > > ~/sandbox/rtems-libbsd$ waf-2.0.4 -v
> > > Could not create the directory ///h
> > >
> > > that's pretty weird.
> >
> > Yes it is. The build scripts do a lot of stuff in preparation for the
> > real build. But as far as I know nothing of it should create a
> > directory. The first operation that creates a directory should print
> the
> > following line:
> >
> > [ 1/99] Compiling
> > testsuite/include/rtems/bsd/test/network-config.h.in
> > 
> >
> > Anyway: To find out how far you get into the build scripts, I added a
> > few debug messages. Could you try to apply the attached patch and see
> > whether it prints some messages. Note that the messages are not
> really
> > clean and it seems that there is a bug that prints them two times.
> But
> > they should give a hint when the build process breaks.
> >
> > Best Regards
> >
> > Christian Mauderer
> >
> > >
> > > Regards,
> > > Niko
> > >
> > > On Fri, Jun 22, 2018 at 12:23 PM Christian Mauderer
> > >  > 
> > >  > >> wrote:
> > >
> > > Hello Niko,
> > >
> > > from a quick glance, everything till the waf call looks good.
> > What is
> > > really odd is the output of the last waf-2.0.8 call. There
> > should be a
> > > lot more output starting with something like
> > >
> > > 
> > > Waf: Entering directory
> > >
> >
>   
> `/home/EB/christian_m/Projekte/some-project/rtems-libbsd/build/arm-rtems5-atsamv-everything'
> > > [  1/122] Creating
> > >
> >  build/arm-rtems5-atsamv-everything/build-include/rtems/bsd/modules.h
> > > [  2/122] Compiling rtemsbsd/rtems/generate_kvm_symbols
> > > [  3/122] Compiling
> > > testsuite/include/rtems/bsd/test/network-config.h.in
> > 
> > > 
> > > [  4/122] Compiling freebsd/sys/crypto/blowfish/bf_locl.h
> > > [  5/122] Compiling freebsd/sys/crypto/blowfish/bf_pi.h
> > > ...
> > > 
> > >
> > > You are even missing the "Waf: Entering directory". So that
> > seems like a
> > > quite early bug.
> > >
> > > You can try whether a `waf -v` provides more output. Beneath
> > that: Where
> > 

Re: Error building libbsd

[Christian Mauderer]

Hello Niko,

if the patch didn't do anything to your output, the error has to happen
quite early. The first one should have been printed when the build()
function of the wscript has been reached. So I would assume that it
already happens before that.

If you have time and desire to find the bug, you might could try to
build some other tools that use waf as a build system (for example the
waf examples from the waf website). Otherwise I would suggest to go with
python3 for now.

Note that it shouldn't really be a problem with python2 in general: My
OpenSUSE machine is using python2 too and has no problems with the
current libbsd. As far as I know most distributions still use python2 as
the default case. I only know Arch Linux that does use python3.

Best regards

Christian Mauderer

Am 25.06.2018 um 10:08 schrieb Nicolas Tsiogkas:
> Hi,
> 
> sorry for the late reply.
> 
> So, the patch didn't do a thing. Got the same print.
> 
> But I tried to run waf with python3
> 
> python3 /home/niko/.local/bin/waf-2.0.8
> 
> and
> 
> python3 /home/niko/.local/bin/waf-2.0.8 install
> 
> That was successful. From that I can assume that there is some issue
> with the python2 shipped with ubuntu? I can try at home as well in case
> there is something specific with the python configuration on my current
> pc and report back.
> 
> Thanks for the help.
> 
> Regards,
> Niko
> 
> On Fri, Jun 22, 2018 at 9:56 PM Christian Mauderer  > wrote:
> 
> Hello Niko,
> 
> Am 22.06.2018 um 13:23 schrieb Nicolas Tsiogkas:
> > Hi,
> >
> > waf was downloaded by the website (waf.io 
> )
> >
> > niko@im-desktop-021:~/sandbox/rtems-libbsd$ which waf-2.0.8 
> > /home/niko/.local/bin/waf-2.0.8
> 
> Should be OK.
> 
> >
> > running also with waf-2.0.4 which used to work gives the same error
> >
> > ~/sandbox/rtems-libbsd$ waf-2.0.8 -v
> > Could not create the directory ///h
> >
> > ~/sandbox/rtems-libbsd$ waf-2.0.4 -v
> > Could not create the directory ///h
> >
> > that's pretty weird.
> 
> Yes it is. The build scripts do a lot of stuff in preparation for the
> real build. But as far as I know nothing of it should create a
> directory. The first operation that creates a directory should print the
> following line:
> 
> [ 1/99] Compiling
> testsuite/include/rtems/bsd/test/network-config.h.in
> 
> 
> Anyway: To find out how far you get into the build scripts, I added a
> few debug messages. Could you try to apply the attached patch and see
> whether it prints some messages. Note that the messages are not really
> clean and it seems that there is a bug that prints them two times. But
> they should give a hint when the build process breaks.
> 
> Best Regards
> 
> Christian Mauderer
> 
> >
> > Regards,
> > Niko
> >
> > On Fri, Jun 22, 2018 at 12:23 PM Christian Mauderer
> >  
> >  >> wrote:
> >
> >     Hello Niko,
> >
> >     from a quick glance, everything till the waf call looks good.
> What is
> >     really odd is the output of the last waf-2.0.8 call. There
> should be a
> >     lot more output starting with something like
> >
> >     
> >     Waf: Entering directory
> >   
>  
> `/home/EB/christian_m/Projekte/some-project/rtems-libbsd/build/arm-rtems5-atsamv-everything'
> >     [  1/122] Creating
> >   
>  build/arm-rtems5-atsamv-everything/build-include/rtems/bsd/modules.h
> >     [  2/122] Compiling rtemsbsd/rtems/generate_kvm_symbols
> >     [  3/122] Compiling
> >     testsuite/include/rtems/bsd/test/network-config.h.in
> 
> >     
> >     [  4/122] Compiling freebsd/sys/crypto/blowfish/bf_locl.h
> >     [  5/122] Compiling freebsd/sys/crypto/blowfish/bf_pi.h
> >     ...
> >     
> >
> >     You are even missing the "Waf: Entering directory". So that
> seems like a
> >     quite early bug.
> >
> >     You can try whether a `waf -v` provides more output. Beneath
> that: Where
> >     does your waf come from? You call it with a version number.
> >
> >     Best regards
> >
> >     Christian Mauderer
> >
> >     Am 22.06.2018 um 11:21 schrieb Nicolas Tsiogkas:
> >     > Hi,
> >     >
> >     > I did the process again with no luck. I am attaching a zip with
> >     command
> >     > outputs/logs.
> >     >
> >     > Regards,
> >     > Niko
> >     >
> >     > On Fri, Jun 22, 2018 at 9:44 AM Nicolas Tsiogkas
> >     mailto:lou.n...@gmail.com>
> 

Re: Error building libbsd

[Nicolas Tsiogkas]

Hi,

sorry for the late reply.

So, the patch didn't do a thing. Got the same print.

But I tried to run waf with python3

python3 /home/niko/.local/bin/waf-2.0.8

and

python3 /home/niko/.local/bin/waf-2.0.8 install

That was successful. From that I can assume that there is some issue with
the python2 shipped with ubuntu? I can try at home as well in case there is
something specific with the python configuration on my current pc and
report back.

Thanks for the help.

Regards,
Niko

On Fri, Jun 22, 2018 at 9:56 PM Christian Mauderer 
wrote:

> Hello Niko,
>
> Am 22.06.2018 um 13:23 schrieb Nicolas Tsiogkas:
> > Hi,
> >
> > waf was downloaded by the website (waf.io )
> >
> > niko@im-desktop-021:~/sandbox/rtems-libbsd$ which waf-2.0.8
> > /home/niko/.local/bin/waf-2.0.8
>
> Should be OK.
>
> >
> > running also with waf-2.0.4 which used to work gives the same error
> >
> > ~/sandbox/rtems-libbsd$ waf-2.0.8 -v
> > Could not create the directory ///h
> >
> > ~/sandbox/rtems-libbsd$ waf-2.0.4 -v
> > Could not create the directory ///h
> >
> > that's pretty weird.
>
> Yes it is. The build scripts do a lot of stuff in preparation for the
> real build. But as far as I know nothing of it should create a
> directory. The first operation that creates a directory should print the
> following line:
>
> [ 1/99] Compiling testsuite/include/rtems/bsd/test/network-config.h.in
>
> Anyway: To find out how far you get into the build scripts, I added a
> few debug messages. Could you try to apply the attached patch and see
> whether it prints some messages. Note that the messages are not really
> clean and it seems that there is a bug that prints them two times. But
> they should give a hint when the build process breaks.
>
> Best Regards
>
> Christian Mauderer
>
> >
> > Regards,
> > Niko
> >
> > On Fri, Jun 22, 2018 at 12:23 PM Christian Mauderer
> >  > > wrote:
> >
> > Hello Niko,
> >
> > from a quick glance, everything till the waf call looks good. What is
> > really odd is the output of the last waf-2.0.8 call. There should be
> a
> > lot more output starting with something like
> >
> > 
> > Waf: Entering directory
> >
>  
> `/home/EB/christian_m/Projekte/some-project/rtems-libbsd/build/arm-rtems5-atsamv-everything'
> > [  1/122] Creating
> > build/arm-rtems5-atsamv-everything/build-include/rtems/bsd/modules.h
> > [  2/122] Compiling rtemsbsd/rtems/generate_kvm_symbols
> > [  3/122] Compiling
> > testsuite/include/rtems/bsd/test/network-config.h.in
> > 
> > [  4/122] Compiling freebsd/sys/crypto/blowfish/bf_locl.h
> > [  5/122] Compiling freebsd/sys/crypto/blowfish/bf_pi.h
> > ...
> > 
> >
> > You are even missing the "Waf: Entering directory". So that seems
> like a
> > quite early bug.
> >
> > You can try whether a `waf -v` provides more output. Beneath that:
> Where
> > does your waf come from? You call it with a version number.
> >
> > Best regards
> >
> > Christian Mauderer
> >
> > Am 22.06.2018 um 11:21 schrieb Nicolas Tsiogkas:
> > > Hi,
> > >
> > > I did the process again with no luck. I am attaching a zip with
> > command
> > > outputs/logs.
> > >
> > > Regards,
> > > Niko
> > >
> > > On Fri, Jun 22, 2018 at 9:44 AM Nicolas Tsiogkas
> > mailto:lou.n...@gmail.com>
> > > >> wrote:
> > >
> > > Hi Christian,
> > >
> > > it is a fresh clone. Although I had a previous build with 4.11
> I
> > > decided try from scratch given the changes.
> > >
> > > Error is during the build. I'm configuring normally and when I
> try
> > > to waf it then it fails.
> > >
> > > Is there any way to increase the verbosity? I will repeat all
> the
> > > process from scratch keeping logs so I can send them and try
> > to find
> > > the issue.
> > >
> > > Thanks for your time!
> > >
> > > Regards,
> > > Niko
> > >
> > > On Thu, Jun 21, 2018 at 6:49 PM Christian Mauderer
> > > mailto:l...@c-mauderer.de>
> > >> wrote:
> > >
> > > Am 21.06.2018 um 16:26 schrieb Nicolas Tsiogkas:
> > > > And apparently I forgot the host.
> > > >
> > > > It is ubuntu 16.04:
> > > > Linux im-desktop-021 4.13.0-45-generic
> > #50~16.04.1-Ubuntu SMP
> > > Wed May 30
> > > > 11:18:27 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
> > > > gcc version 5.4.0 20160609 (Ubuntu
> 5.4.0-6ubuntu1~16.04.9)
> > > >
> > > > N.
> > > >
> > > >
> > > > On Thu, Jun 21, 2018 at 4:19 PM Nicolas Tsiogkas
> > > mailto:lou.n...@gmail.com>
> >