RE: v4l2 for 2.4.x kernels replaces videodev.h so that kern el no longer compiles

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]



kernel (2.4.5, patched for i2c 2.6.0) still won-t compile. 
videodevX still working fine together with bttv2 out onto xawtv. (tryed
multiple times insmod rmmod switching
back and forth between bttv and bttv2).

[root@pippo root]#tar -xvBf videodevX-20010613.tar

[root@pippo videodevX]# patch -p1 < v4l2_owner.patch
(Stripping trailing CRs from patch.)
patching file videodev.h
(Stripping trailing CRs from patch.)
patching file videodevX.c

[root@pippo videodevX]# make
make: /scripts/pathdown.sh: Command not found
/usr/share/locale/en/LC_MESSAGES/make.moDIR=`pwd`; (cd
/lib/modules/2.4.5/build; make SUBDIRS=$DIR modules)
make[1]: Entering directory `/usr/src/linux-2.4.5'
make -C  /root/videodevX CFLAGS="-D__KERNEL__ -I/usr/src/linux-2.4.5/include
-Wall -Wstrict-prototypes -O2 -foms
make[2]: Entering directory `/root/videodevX'
gcc -D__KERNEL__ -I/usr/src/linux-2.4.5/include -Wall -Wstrict-prototypes
-O2 -fomit-frame-pointer -fno-strict-c
videodevX.c: In function `videodev_proc_read':
videodevX.c:1148: warning: `d' might be used uninitialized in this function
videodevX.c: At top level:
videodevX.c:1069: warning: `video_build_proc' defined but not used
gcc -D__KERNEL__ -I/usr/src/linux-2.4.5/include -Wall -Wstrict-prototypes
-O2 -fomit-frame-pointer -fno-strict-c
make[2]: Leaving directory `/root/videodevX'
make[1]: Leaving directory `/usr/src/linux-2.4.5'

[root@pippo videodevX]# make install
...blah blah.....

[root@pippo videodevX]# cd /usr/src/linux-2.4.5/
[root@pippo linux-2.4.5]# make modules
....blah blah.....
make -C video modules
make[3]: Entering directory `/usr/src/linux-2.4.5/drivers/media/video'
gcc -D__KERNEL__ -I/usr/src/linux-2.4.5/include -Wall -Wstrict-prototypes
-O2 -fomit-frame-pointer -fno-strict-c
videodev.c:436: conflicting types for `video_register_device_Recfe1c4b'
/usr/src/linux-2.4.5/include/linux/videodev.h:1327: previous declaration of
`video_register_device_Recfe1c4b'
videodev.c:564: warning: static declaration for `videodev_init' follows
non-static
make[3]: *** [videodev.o] Error 1
make[3]: Leaving directory `/usr/src/linux-2.4.5/drivers/media/video'
make[2]: *** [_modsubdir_video] Error 2
make[2]: Leaving directory `/usr/src/linux-2.4.5/drivers/media'
make[1]: *** [_modsubdir_media] Error 2
make[1]: Leaving directory `/usr/src/linux-2.4.5/drivers'
make: *** [_mod_drivers] Error 2

[root@pippo linux-2.4.5]#

Michele Slocovich

> -----Original Message-----
> From:	Justin Schoeman [SMTP:justin@xxxxxxxxxxxxxxxxxxxx]
> Sent:	mercoledì 27 giugno 2001 10.59
> To:	video4linux-list@xxxxxxxxxx; Bill Dirks
> Subject:	Re:  v4l2 for 2.4.x kernels replaces videodev.h so that
> kernelno  long er compiles
> 
> Erik Walthinsen wrote:
> > 
> > On Tue, 26 Jun 2001, Michele Slocovich wrote:
> > 
> > > v4l2 for 2.4.x kernels replaces videodev.h so that my kernel wont
> compile
> > > anymore (either patched with new 2.6.0 i2c or not)
> > > sorry for my bloathed ignorance but may I ask whether this is by
> design ?
> > > and for my curiosity why so ? *maybe for v4l2 drivers to compile
> against it
> > > ?*
> > 
> > I stated a long time ago that I believe that calling the v4l2 header
> > videodev.h in direct conflict with v4l's videodev.h is a major flaw in
> > v4l2.  As the kernel progresses, it's getting harder and harder to
> bodily
> > yank all of v4l out so you can safely insert v4l2.  Changing the header
> > name to videodev2.h to match the 'version' of the API is natural,
> simple,
> > and correct, IMO.  It's called 'versioning', and lack of it is the
> > fundamental reason why Windoze sucks so much.  Highly incompatible
> headers
> > by the same name is the absolute worst possible case.
> > 
> > My $0.02,
> >         Omega
> 
> videodevX is no longer v4l1 or v4l2, but both, containing all the
> functionality of both.  Unfortunately, this means that videodevX needs
> to track changes made to the kernel v4l1 driver, and this got a bit out
> of synch.  The attached patch should fix this problem.  Please test it
> (insmod/rmmod drivers a couple of times, and see if you don't get any
> errors), and let us know.
> 
> Thanks,
> 
> -justin << File: v4l2_owner.patch >> 





[Index of Archives]     [Linux DVB]     [Video Disk Recorder]     [Asterisk]     [Photo]     [DCCP]     [Netdev]     [Xorg]     [Util Linux NG]     [Xfree86]     [Free Photo Albums]     [Fedora Users]     [Fedora Women]     [ALSA Users]     [ALSA Devel]     [Linux USB]

Powered by Linux