Re: comedi + rtai adeos

From: Bernd Porr <Bernd.Porr_at_cn.stir.ac.uk> · Date: Wed, 03 Mar 2004 23:08:32 +0000

Hi!

I think we should add a define for RTAI. From my knowledge the realtime 
linux variants don't support USB so far.

It would be great if one of the RTAI people would supply a patch for the 
autoconf stuff. I don't use realtime linux just now. So I can't provide 
the patch.

/Bernd

Frank Mori Hess (by way of Frank Mori Hess 
<fmhess_at_users.sourceforge.net>) wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> 
>>Since version 3.0 RTAI copies all headers to $PREFIX/include (defaults
>>to /usr/realtime).  So maybe it's better to use the RTAI _installation_
>>dir instead of the RTAI source dir (at least for RTAI >= 3.0).
> 
> 
> Okay, I changed the configure help to suggest specifying the installation
> directory instead of the source directory, and changed the default rtai
> dir.  The old code should have been able to handle things correctly if you
> gave it the installation directory though (--with-rtaidir=/usr/realtime).
> 
> 
>>==
>>
>>ps.  The current CVS also fails at the compilation of usbdux now:
>>
>>if gcc -DHAVE_CONFIG_H -I. -I../../../comedi/drivers -I../..
>>-I../../../include/ -DKBUILD_MODNAME=usbdux
>>-DKBUILD_BASENAME=../../../comedi/drivers/usbdux -D__KERNEL__
>>-I/usr/src/linux-2.4.24-hal10/include -Wall -Wstrict-prototypes
>>-Wno-trigraphs -O2 -fno-strict-aliasing -fno-common
>>-fomit-frame-pointer -pipe -mpreferred-stack-boundary=2 -march=i686
>>-DMODULE -DMODULE -D__MODULE_KERNEL_i686=1
>>-D__BOOT_KERNEL_ENTERPRISE=0 -D__BOOT_KERNEL_UP=1
>>-D__BOOT_KERNEL_SMP=0 -D__BOOT_KERNEL_BIGMEM=0
>>-D__BOOT_KERNEL_HUGEMEM=0 -I/usr/realtime//include  -g -O2 -MT
>>usbdux.o -MD -MP -MF ".deps/usbdux.Tpo" \
>>  -c -o usbdux.o `test -f '../../../comedi/drivers/usbdux.c' || echo
>>  '../../../comedi/drivers/'`../../../comedi/drivers/usbdux.c; \
>>then mv -f ".deps/usbdux.Tpo" ".deps/usbdux.Po"; \
>>else rm -f ".deps/usbdux.Tpo"; exit 1; \
>>fi
> 
> 
> I've made the hacks that set KBUILD_BASENAME and KBUILD_MODNAME a little
> more robust.  This problem should be gone now.
> 
> 
>>ps2.  You probably would have preferred a one line answer: "yes it
>>works" :)
> 
> 
> A bug report is the next best thing, thanks.
> 
> - --
> Frank
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.0.6 (GNU/Linux)
> Comment: For info see http://www.gnupg.org
> 
> iD8DBQFARR+x5vihyNWuA4URAqBcAJkBPmbjTOOv87T03rvcPMjg1MQOugCg6K4U
> 8h+GCKjLbUAQEErpu6l93YI=
> =r/om
> -----END PGP SIGNATURE-----
> 
> _______________________________________________
> comedi mailing list
> comedi_at_comedi.org
> https://cvs.comedi.org/cgi-bin/mailman/listinfo/comedi

-- 
www:	http://www.cn.stir.ac.uk/~bp1/
         http://www.linux-usb-daq.com/
e-mail:	Bernd.Porr_at_cn.stir.ac.uk
Mobile:	+44 (0)7769-506746
Landl.:	+44 (0)1324 875445
Work:	+44 (0)1786-46-6378
Postal:	Room: 3B141, Department of Psychology, Univ of Stir, FK9 4LA, Scotl