On my NEC Versa SX laptop with a trident cyber9388 video chipset, any horizontal line of pure white has only tbe right-most pixel actually white, and all others are a drab green color. Other colors are not affected, and it appears that this is the only corruption, but it's a big one. The majority of text displays are borderline unreadable, since most use either white text or a white background. The last working version of XFree86 on my laptop was 4.1.0-3. All newer versions have displayed either the technicolor distortion described in bug #61834 or this white-field distortion. I tried the alternate trident_drv.o drivers available at www.xfree86.org/~alanh and people.redhat.com/~mharris, and both of them displayed the same problem.
Cc'ing Trident driver maintainer Please attach your X server log, and X config file. Also, for troubleshooting purposes, does the problem disappear if you use Option "noaccel"?
Created attachment 79405 [details] XF86Config file
Created attachment 79406 [details] XFree86 log file when using trident driver
Problem is unchanged when I set Option "NoAccel". However, I did recently figure out that the display is fine under the (unaccelerated, slow) "vesa" driver. By the way, I live in Durham and could stand being without my laptop for a week or so if it would help you to see the actual machine. Let me know if you're interested and I could bring it by the Redhat offices.
I found that using the vesa driver in RH 8.0 fixed my display problems as well. It's a Fujitsu Lifebook 735DX. The video card is a Trident 9385 and redhat- config-xconfig discovers it as a TGUI9660. When I run startx, the display looks bad for the first several seconds, with vertical lines of various colors. The screen then clears and Xwindows starts normally.
upgrade to XFree86 4.3.0 and if the problem persists log it in the XFree86 Bugzilla database. I'm removing myself as a CC.
I agree with Alan H's suggestion, if the problem still exists in Red Hat Linux 9 using XFree86 4.3.0, please open a new bug report in XFree86 bugzilla at http://bugs.xfree86.org and feel free to add me as a CC <mharris.org.uk> The 'vesa' driver, while suboptimal, does work as a workaround as indicated above, so this problem is unlikely to ever be fixed in XFree86 4.2.x in Red Hat Linux 8.0. Closing as CURRENTRELEASE. Reopen and attach upstream bug report URL if this recurs in RHL 9.