From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760737AbYEHHl7 (ORCPT ); Thu, 8 May 2008 03:41:59 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751595AbYEHHlv (ORCPT ); Thu, 8 May 2008 03:41:51 -0400 Received: from smtp1.kolej.mff.cuni.cz ([78.128.192.10]:60634 "EHLO smtp1.kolej.mff.cuni.cz" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751403AbYEHHlt (ORCPT ); Thu, 8 May 2008 03:41:49 -0400 X-Greylist: delayed 1365 seconds by postgrey-1.27 at vger.kernel.org; Thu, 08 May 2008 03:41:49 EDT X-Envelope-From: zajio1am@artax.karlin.mff.cuni.cz Date: Thu, 8 May 2008 09:41:34 +0200 From: Ondrej Zajicek To: JosephChan@via.com.tw Cc: akpm@linux-foundation.org, geert@linux-m68k.org, linux-fbdev-devel@lists.sourceforge.net, linux-kernel@vger.kernel.org Subject: Re: [Linux-fbdev-devel] [PATCH 7/9] viafb: VIA Frame Buffer Device Driver Message-ID: <20080508074134.GB11121@localhost.localdomain> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: X-Operating-System: Debian GNU/Linux 4.0 (Etch) User-Agent: Mutt/1.5.13 (2006-08-11) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, May 07, 2008 at 07:09:44PM +0800, JosephChan@via.com.tw wrote: > +static int viafb_ioctl(struct fb_info *info, u_int cmd, u_long arg) > +{ > + switch (cmd) { > + case VIAFB_GET_CHIP_INFO: /*struct chip_information chip_info ; */ > + if (copy_to_user((void __user *)arg, &chip_info, sizeof(chip_info))) > + return -EFAULT; > + break; > + case VIAFB_GET_INFO_SIZE: > + return put_user(sizeof(viainfo), argp); > + case VIAFB_GET_INFO: > + return ioctl_get_viafb_info(arg); ... > + case VIAFB_GET_RESOLUTION: Some of these ioctls looks like they duplicate standard fbdev ioctls. What is a reason for that? > +int get_mode_index(int hres, int vres, int flag) > +{ > + DEBUG_MSG(KERN_INFO "get_mode_index!\n"); > + > + if (flag == 0) { > + if (hres == 480 && vres == 640) { > + resMode = VIA_RES_480X640; > + mode = "480x640"; > + } else if (hres == 640 && vres == 480) { > + resMode = VIA_RES_640X480; > + mode = "640x480"; > + } else if (hres == 800 && vres == 480) { > + resMode = VIA_RES_800X480; > + mode = "800x480"; > + } else if (hres == 800 && vres == 600) { > + resMode = VIA_RES_800X600; > + mode = "800x600"; > + } else if (hres == 1024 && vres == 768) { > + resMode = VIA_RES_1024X768; > + mode = "1024x768"; ... This is particularly ugly function. Some loop through a table is perhaps better than 250 lines of such if statements, isn't it? > + if (!strncmp(this_opt, "mode=", 5)) { > + mode = kmalloc(strlen(this_opt + 4), GFP_KERNEL); > + strcpy(mode, this_opt + 5); > + } else if (!strncmp(this_opt, "mode1=", 6)) { > + mode1 = kmalloc(strlen(this_opt + 5), GFP_KERNEL); > + strcpy(mode1, this_opt + 6); > + } else if (!strncmp(this_opt, "bpp=", 4)) { > + via_fb_bpp = simple_strtoul(this_opt + 4, NULL, 0); > + } else if (!strncmp(this_opt, "bpp1=", 5)) { > + via_fb_bpp1 = simple_strtoul(this_opt + 5, NULL, 0); > + } else if (!strncmp(this_opt, "refresh=", 8)) { > + refresh = simple_strtoul(this_opt + 8, NULL, 0); > + } else if (!strncmp(this_opt, "refresh1=", 9)) { > + refresh1 = simple_strtoul(this_opt + 9, NULL, 0); Most drivers accept default mode specification like 1024x768-16@60 (see Documentation/fb/modedb.txt). Is there a reason to have separate bpp= and refresh= ? -- Elen sila lumenn' omentielvo Ondrej 'SanTiago' Zajicek (email: santiago@crfreenet.org) OpenPGP encrypted e-mails preferred (KeyID 0x11DEADC3, wwwkeys.pgp.net) "To err is human -- to blame it on a computer is even more so."