Created attachment 24780 [details]
xorg.conf with ModeDebug

The log says:
(II) intel(0): Skipping LVDS from driver feature BDB's LVDS config info.

I think this should be fixed by below commit. It should be in 2.6.99.903. Could you double check if your source really includes it:

commit a03feab1b7ce511e1704bee96c83290d1c7b2785
Author: Zhenyu Wang <zhenyu.z.wang@intel.com>
Date:   Wed Mar 25 11:13:52 2009 +0800

    Disable LVDS config parsing from VBT for now
    
    As wider tests showed that this doesn't work for all VBIOS, so
    disable it for now and reenable it after we get reliable method.
    (cherry picked from commit 375b2e40fcb17e94538a75392950e2533c1bb031)

diff --git a/src/i830_bios.c b/src/i830_bios.c
index 9b13bf4..4f2355d 100644
--- a/src/i830_bios.c
+++ b/src/i830_bios.c
@@ -181,6 +181,12 @@ parse_driver_feature(I830Ptr pI830, struct bdb_header *bdb)
     if (!IS_I9XX(pI830))
 	return;
 
+    /* XXX Disable this parsing, as it looks doesn't work for all
+       VBIOS. Reenable it if we could find out the reliable VBT parsing
+       for LVDS config later. */
+    if (1)
+	return;
+
     feature = find_section(bdb, BDB_DRIVER_FEATURES);
     if (!feature)
 	return;

It is 2.6.99.903 and it does have that new code.

Are you able to git bisect the xf86-video-intel master branch to see which commit causes this regression? (we can only provide limited support on NetBSD as we are only working on Linux, but the git bisect info will be quite helpful)

git bisect is fun! Thank you for having me use it.

$ git bisect good                                                              
f6d8ae69b0f97e696c142f06c8038f336ed024f9 is first bad commit
commit f6d8ae69b0f97e696c142f06c8038f336ed024f9
Author: Zhenyu Wang <zhenyu.z.wang@intel.com>
Date:   Wed Feb 25 09:57:00 2009 +0800

    Use LVDS config in Driver feature BDB for integrated LVDS check
    
    The LVDS config bits in VBT driver feature block is used by vendor
    to identify the board implement of integrated LVDS/eDP or SDVO LVDS.
    And video bios uses these bits for LVDS enabling or not. So check
    these bits for integrated LVDS might eliminate more quirks.

:040000 040000 eb0c7e78bd6dae7037076dfbcd74da57e8b4cbad bd0a7e774f3260589000e6d8cc9a8b948a0e6361 M      src

I saved the Xorg.0.log for every good and bad test. Let me know if you want me to post these for you. Also if you want to login to this NetBSD/amd64 system, please let me know.

Anything else I can provide? I tested latest code again today and still not working. A couple weeks ago, I provided the bisect to show where it stopped working for me.

The version that does detect my video crashes frequently (as reported in my other ticket here).

Any other suggestions on what I can do?

Is there another source of Xorg drivers for G45 other than from http://intellinuxgraphics.org/  (since that says for "Linux" and I don't have Linux)?

I have good experiences with xf86-video-intel on other non-Linux systems in the past (but haven't tried recent versions yet).

Does current xf86-video-intel 2.8.0 or git tip work for you now?

Jeremy, I assume NetBSD has fixed GM45 support. Correct me if I'm wrong.

Jeremy, I assume NetBSD has fixed GM45 support. Correct me if I'm wrong.