Hi all,
a lot of people seem to be confused about (non-working) notebook display
blanking, so...
- add verbose explanation about the whole notebook screen blanking situation
and a possible development direction to prominent place
I might actually be tempted to unify graphics hardware backlight handling
myself, since having a console that doesn't shut down the backlight properly
is mildly annoying...
Patch against vanilla 2.5.70.
Andreas Mohr
diff -urN linux-2.5.70.orig/drivers/video/console/vgacon.c linux-2.5.70/drivers/video/console/vgacon.c
--- linux-2.5.70.orig/drivers/video/console/vgacon.c 2003-05-28 17:57:07.000000000 +0200
+++ linux-2.5.70/drivers/video/console/vgacon.c 2003-05-29 03:13:53.000000000 +0200
@@ -661,6 +661,22 @@
}
}
+/*
+ * vgacon_blank() is able to simply blank the screen, or to even switch
+ * the monitor off in case of "Green", "Energy Star" monitors
+ * (by switching off the sync signals using VESA mechanisms).
+ * However, it is NOT able to switch off the *backlight* of (most?) notebooks.
+ * For these cases, we usually make use of APM display blanking methods.
+ * However if you chose to use ACPI on your notebook instead of APM,
+ * then you're screwed for now, since ACPI expects the operating system itself
+ * to manage video device power state (there's no convenient APM call that
+ * does it for you). Since we don't have VGA adapter specific support for
+ * that (only XFree86 has it), you're only able to switch off your backlight
+ * in XFree86, and only if the driver supports that ('xset dpms force off').
+ * ppc already has config PMAC_BACKLIGHT for several graphics cards,
+ * so backlight handling should simply be generalized here instead,
+ * to also support i386 notebook backlight etc.
+ */
static int vgacon_blank(struct vc_data *c, int blank)
{
switch (blank) {
Hi!
> a lot of people seem to be confused about (non-working) notebook display
> blanking, so...
>
> - add verbose explanation about the whole notebook screen blanking situation
> and a possible development direction to prominent place
Please submit it to Rusty trivial patch monkey Russell
<[email protected]>.
> I might actually be tempted to unify graphics hardware backlight handling
> myself, since having a console that doesn't shut down the backlight properly
> is mildly annoying...
Yep, solving that somehow would be good.
> Patch against vanilla 2.5.70.
>
> Andreas Mohr
>
> diff -urN linux-2.5.70.orig/drivers/video/console/vgacon.c linux-2.5.70/drivers/video/console/vgacon.c
> --- linux-2.5.70.orig/drivers/video/console/vgacon.c 2003-05-28 17:57:07.000000000 +0200
> +++ linux-2.5.70/drivers/video/console/vgacon.c 2003-05-29 03:13:53.000000000 +0200
> @@ -661,6 +661,22 @@
> }
> }
>
> +/*
> + * vgacon_blank() is able to simply blank the screen, or to even switch
> + * the monitor off in case of "Green", "Energy Star" monitors
> + * (by switching off the sync signals using VESA mechanisms).
> + * However, it is NOT able to switch off the *backlight* of (most?)
> notebooks.
Acurate, AFAICS. Why can't that stupid notebook manufacturers make it
turn backlight off on VESA blank?!
Pavel
--
When do you have a heart between your knees?
[Johanka's followup: and *two* hearts?]