Linus breaks nvidia again
On Sun, Jul 29, 2012 at 6:46 AM, walt <w41ter@gmail.com> wrote:
> For you weirdos who track Linus's git repo every day, and who use the
> proprietary nvidia-drivers, you need to make this change to the nvidia
> code for their kernel module (nvidia.ko):
>
> --- /usr/src/nv/nv-acpi.c.orig 2012-07-29 04:09:03.336096276 -0700
> +++ /usr/src/nv/nv-acpi.c 2012-07-29 04:09:21.260096633 -0700
> @@ -300,7 +300,7 @@
> if (pNvAcpiObject->notify_handler_installed)
> {
> // no status returned for this function
> - acpi_os_wait_events_complete(NULL);
> + acpi_os_wait_events_complete();
>
> // remove event notifier
> status = acpi_remove_notify_handler(device->handle, ACPI_DEVICE_NOTIFY, nv_acpi_event);
>
> ====================================
>
> Note that I've stashed the nvidia code in /usr/src/nv because I do recompile
> it every day after I build the latest kernel from Linus. This way I don't
> have to re-patch the nvidia code every day because I don't have to reinstall
> nvidia-drivers every day. (Just the kernel module part, nvidia.ko)
>
> I know there are some of you who know how to change ebuilds to apply your
> own local patches, so if you'd like to remind me how to do that I'd be
> grateful.
No need to change the ebuild; just put your patch (filename must end
in .patch) in
/etc/portage/patches/x11-drivers/nivida-drivers/
nvidia-drivers uses epatch_user; from the docs of the eclass:
# User patches are intended for quick testing of patches without ebuild
# modifications, as well as for permanent customizations a user might
# desire. Obviously, there can be no official support for arbitrarily
# patched ebuilds. So whenever a build log in a bug report mentions that
# user patches were applied, the user should be asked to reproduce the
# problem without these.
Regards.
--
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México
|
