From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932279AbWEEBeA (ORCPT ); Thu, 4 May 2006 21:34:00 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932383AbWEEBeA (ORCPT ); Thu, 4 May 2006 21:34:00 -0400 Received: from web.bloglines.com ([65.214.39.152]:56012 "HELO blw09bos.io.askjeeves.info") by vger.kernel.org with SMTP id S932279AbWEEBd7 (ORCPT ); Thu, 4 May 2006 21:33:59 -0400 Message-ID: <1146792838.2129823533.18589.sendItem@bloglines.com> Date: 5 May 2006 01:33:58 -0000 From: grfgguvf.29601511@bloglines.com To: adaplas@gmail.com CC: linux-kernel@vger.kernel.org Subject: Re: Weird framebuffer bug? MIME-Version: 1.0 Content-Type: text/plain;charset="utf-8" Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org --- Antonino A. Daplas wrote: > Does the same thing happens with different > color depths? That was it. With mode vga=0x318 (1024x768 "16m colours") no matter what colour depth I set X to the display will be incorrect. Set to 24bits every 4th (not 5th, that was my bad estimation) vertical line is omitted. This is because the framebuffer is really using 32bits/pixel. Using mode vga=0x317 (1024x768 "64k colours") and X set to 16bits colour depth everything is OK. This is really an X fbdev driver problem... It should use 32bits/pixel by padding the last byte. I will write Xorg about this. > Can you try using the "vesa" driver with X? If > the same thing happens, it might > be a problem with the BIOS. [I tried this first, and as expected X's vesa driver works.] Thanks Antonino!