Hi Martin-Éric,

Hans de Goede wrote:

> Erm, that is quite a bit of work from my side for something which you
> can easily do yourself, edit gspca.c, search for which_bandwidth
> and then under the following lines:
>         u32 bandwidth;
>         int i;
>
> Add a line like this:
>       return 2000 * 2000 * 120;

In case it helps, here are some more complete instructions.

 0. Prerequisites:

        apt-get install git build-essential

 1. Get the kernel history, if you don't already have it:

        git clone \
          git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git

 2. Try linus's master:

        cd linux
        git fetch origin
        git reset --hard origin/master
        cp /boot/config-$(uname -r) .config; # current configuration
        scripts/config --disable DEBUG_INFO
        make localmodconfig; # optional: minimize configuration
        make deb-pkg; # optionally with -j<num> for parallel build
        dpkg -i ../<name of package>; # as root
        reboot
        ... test test test ...

    Hopefully it reproduces the bug.

 3. Try Hans's first suggested change, as described in the quoted text
    above:

        cd linux
        vi drivers/media/video/gspca/gspca.c
        ... make the suggested edits ...
        make deb-pkg; # maybe with -j4
        dpkg -i ../<name of package>; # as root
        reboot
        ... test test test ...

 4. Try Hans's second suggested change, as described in a previous
    message:

        cd linux
        vi drivers/media/video/gspca/gspca.c
        ... make the suggested edits ...
        make deb-pkg; # maybe with -j4
        dpkg -i ../<name of package>; # as root
        reboot
        ... test test test ...

No doubt Jean-François will notice that it is easier to test the
standalone driver because the first build does not have to compile the
whole kernel.  That's fine, too.  The instructions above describe how
to test the in-kernel driver because it's what I'm used to (and
because it means you test the driver against the same version of the
rest of the kernel as would get the fix).

Hope that helps,
Jonathan



--
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20120709203929.GC17301@burratino

Reply via email to