On suggestions received elsewhere, I tried this with Option "noaccel" set, but it made no difference. The results, and the meaningless pixel pattern on the screen, remained the same.
On another suggestion, I tried the vesa driver, which probably should work with this card as well. That driver (current version is the same for both etch and lenny as far as I can tell) also insists that the "V_BIOS is not found" and that there are no usable modes. It doesn't even try to initialize the display. I notice in the Xorg.0.log for both drivers that the driver reports the BIOS address as 80800000. However, the lspci command reports it as 09c10000. In the available resource listing, the 80800000 address block always appears, but 09c10000 does not. Is there an address translation going on that is not quite working? The setpci command, when used to query the ROM address register of the 86C764/765 chipset, returns 80800000 rather than the 09c10000 returned by lspci. There were several similar bug reports in the freedesktop.org reporting system dating to late 2006 and early 2007, but they are all reported as fixed. These involved missing assigns or equates in the source code. Is it possible that those fixes never filtered into the Alpha source stream? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
