After 14 builds (never tought that PXE and Core2Quad were so useful!!).... the bad commit is:

01c66889c14aa163c49355b3be2ccfb214500599 is the first bad commit
commit 01c66889c14aa163c49355b3be2ccfb214500599
Author: Zhao Yakui <yakui.zhao@intel.com>
Date:   Wed Oct 28 05:10:00 2009 +0000

    drm/i915: Add ACPI OpRegion support for Ironlake

    Add the support of ACPI opregion on Ironlake so that the backlight
    brightness can be adjusted by using ACPI interface
       >/sys/class/backlight/acpi_video0/brightness

    Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
    Tested-by: Zhao Yakui <yakui.zhao@intel.com>
    [zhenyuw: cleanups, fix typo for checking GSE irq and convert to
    current irq handling logic.]
    Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
    Signed-off-by: Eric Anholt <eric@anholt.net>

:040000 040000 79f96cd37f779fb2b33f0e6545e83f32cd5c1372 3b9f77adfeed13386058a46a17e3f4facd8d862c M      drivers

The problem is half-solved...
connecting by ssh (as root) and typing
# echo -n 10 >/sys/class/backlight/acpi_video0/brightness
revives the screen.
Starting X gives me a fully working KWin compositing (with OpenGL)
I will temporarily put this line in an init script, but I think that this bug must be solved as soon as possible!
Another question:
How can I be sure that the ATI gpu isn't wasting power while PC is booted in Linux?
Best regards,
Massimo Maggi
massimo@mmmm.it

Il 18/07/2010 14:57, Chris Wilson ha scritto:

On Sun, 18 Jul 2010 14:29:34 +0200, Massimo Maggi <massimo@mmmm.it> wrote:

I've set up a very minimal kernel config (the resulting kernel is not
useful for anything except showing the problem), and built versions:
2.6.32.16
2.6.33.6
2.6.34.1
2.6.35-rc5
Only 2.6.32.16 is able to correctly initialize the framebuffer.

Excellent, as you have managed to reproduce this with a minimal kernel,
can you do a bisect between 2.6.33.16 and 2.6.33.6 and see which patch
causes the regression? It should only take a couple of hours. My bet is
one of the Arrandale enabling patches...