Actually, Apple laptops are fine at least with recent 2.6 kernels as those
always switch to a special VT during sleep. I assume the same is true with ACPI,
by way of either the kernel or suspend scripts. Given that and the issues with
ACPI support, that missing might be considered a feature at this point. ;)

However, this is causing real problems for APM users, see e.g.
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=363120 .

Created attachment 5480 [details] [review]
Linux patch

Patch to fix this for Linux. Also removes the last traces of ACPI from the
server on platforms that don't support it.

applied to 1.1 branch and head, thanks!

Created attachment 5592 [details] [review]
correctly detect arch

not build acpi for i*86. refer to correctly detect arch patch

ACPI as well as  not work

fragment Xorg.0.log

(**) RgbPath set to "/usr/share/X11/rgb"
(==) ModulePath set to "/usr/lib/X11/modules"
(**) Option "DontZap"
(**) Option "AllowMouseOpenFail"
(II) No APM support in BIOS or kernel
(II) Module ABI versions:
        X.Org ANSI C Emulation: 0.3
        X.Org Video Driver: 1.0
        X.Org XInput driver : 0.6
        X.Org Server Extension : 0.3
        X.Org Font Renderer : 0.5
(II) Loader running on linux
(II) LoadModule: "bitmap"

SRPMS inhere ftp://ftp.altlinux.ru/pub/people/shrek/Xorg-7.1

Created attachment 5598 [details] [review]
Include xorg-config.h in lnx_acpi.c

ACPI is disabled because lnx_acpi.c doesn't include xorg-config.h, so it
doesn't get the correct definition of xf86InfoRec.

Thanks for testing this, and sorry I couldn't test it earlier on my AMD64 box.

The arch detection patch looks good to me as well.

> ACPI is disabled because lnx_acpi.c doesn't include xorg-config.h, so it
> doesn't get the correct definition of xf86InfoRec.

yes, ACPI work
Please, pay your attention to #5877

re-fixed, thanks!