Return-Path: <linux-kernel-owner+willy=40w.ods.org@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id <S268276AbTBWKTb>; Sun, 23 Feb 2003 05:19:31 -0500
Received: (majordomo@vger.kernel.org) by vger.kernel.org
	id <S268292AbTBWKTa>; Sun, 23 Feb 2003 05:19:30 -0500
Received: from phoenix.mvhi.com ([195.224.96.167]:19460 "EHLO
	phoenix.infradead.org") by vger.kernel.org with ESMTP
	id <S268276AbTBWKT2>; Sun, 23 Feb 2003 05:19:28 -0500
Date: Sun, 23 Feb 2003 10:29:37 +0000
From: Christoph Hellwig <hch@infradead.org>
To: Osamu Tomita <tomita@cinet.co.jp>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
       Alan Cox <alan@lxorguk.ukuu.org.uk>
Subject: Re: [PATCH] PC-9800 subarch. support for 2.5.62-AC1 (6/21) FS & partiton
Message-ID: <20030223102937.A15963@infradead.org>
Mail-Followup-To: Christoph Hellwig <hch@infradead.org>,
	Osamu Tomita <tomita@cinet.co.jp>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Alan Cox <alan@lxorguk.ukuu.org.uk>
References: <20030223092116.GA1324@yuzuki.cinet.co.jp> <20030223094325.GG1324@yuzuki.cinet.co.jp>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20030223094325.GG1324@yuzuki.cinet.co.jp>; from tomita@cinet.co.jp on Sun, Feb 23, 2003 at 06:43:25PM +0900
Sender: linux-kernel-owner@vger.kernel.org
X-Mailing-List: linux-kernel@vger.kernel.org
Content-Length: 3119
Lines: 120

On Sun, Feb 23, 2003 at 06:43:25PM +0900, Osamu Tomita wrote:
> -	if (FAT_FIRST_ENT(sb, media) != first) {
> +	if (FAT_FIRST_ENT(sb, media) != first
> +	    && (!pc98 || media != 0xf8 || (first & 0xff) != 0xfe))
> +	{

I think this should be made unconditionally.  There's no reason why
non-pc98 linux machines shouldn't read fat filesystems created on pc98.

> +/* #ifdef CONFIG_BLK_DEV_IDEDISK */
> +#include <linux/ide.h>
> +/* #endif */
> +
> +/* #ifdef CONFIG_BLK_DEV_SD */
> +#include "../../drivers/scsi/scsi.h"
> +#include "../../drivers/scsi/hosts.h"
> +#include <scsi/scsicam.h>
> +/* #endif */

this is really ugly..

> +	int g_head, g_sect;
> +	Sector sect;
> +	const struct nec98_partition *part;
> +	unsigned char *data;
> +	int sector_size = bdev_hardsect_size(bdev);
> +	int major = major(to_kdev_t(bdev->bd_dev));
> +	int minor = minor(to_kdev_t(bdev->bd_dev));

use MAJOR/MINOR here.

> +
> +	switch (major) {
> +#if defined CONFIG_BLK_DEV_HD_ONLY
> +	case HD_MAJOR:
> +	{
> +		extern struct hd_i_struct hd_info[2];
> +
> +		g_head = hd_info[minor >> 6].head;
> +		g_sect = hd_info[minor >> 6].sect;
> +		break;
> +	}
> +#endif /* CONFIG_BLK_DEV_HD_ONLY */
> +#if defined CONFIG_BLK_DEV_SD || defined CONFIG_BLK_DEV_SD_MODULE
> +	case SCSI_DISK0_MAJOR:
> +	case SCSI_DISK1_MAJOR:
> +	case SCSI_DISK2_MAJOR:
> +	case SCSI_DISK3_MAJOR:
> +	case SCSI_DISK4_MAJOR:
> +	case SCSI_DISK5_MAJOR:
> +	case SCSI_DISK6_MAJOR:
> +	case SCSI_DISK7_MAJOR:
> +	{
> +		int diskinfo[3] = { 0, 0, 0 };
> +
> +		pc98_bios_param(bdev, diskinfo);
> +
> +		if ((g_head = diskinfo[0]) <= 0)
> +			g_head = 8;
> +		if ((g_sect = diskinfo[1]) <= 0)
> +			g_sect = 17;
> +		break;
> +	}
> +#endif /* CONFIG_BLK_DEV_SD(_MODULE) */
> +#if defined CONFIG_BLK_DEV_IDEDISK || defined CONFIG_BLK_DEV_IDEDISK_MODULE
> +	case IDE0_MAJOR:
> +	case IDE1_MAJOR:
> +	case IDE2_MAJOR:
> +	case IDE3_MAJOR:
> +	case IDE4_MAJOR:
> +	case IDE5_MAJOR:
> +	case IDE6_MAJOR:
> +	case IDE7_MAJOR:
> +	case IDE8_MAJOR:
> +	case IDE9_MAJOR:
> +	{
> +		ide_drive_t *drive;
> +		unsigned int	h;
> +
> +		for (h = 0; h < MAX_HWIFS; ++h) {
> +			ide_hwif_t  *hwif = &ide_hwifs[h];
> +			if (hwif->present && major == hwif->major) {
> +				unsigned unit = minor >> PARTN_BITS;
> +				if (unit < MAX_DRIVES) {
> +					drive = &hwif->drives[unit];
> +					if (drive->present) {
> +						g_head = drive->head;
> +						g_sect = drive->sect;
> +						goto found;
> +					}
> +				}
> +				break;
> +			}
> +		}
> +	}
> +#endif /* CONFIG_BLK_DEV_IDEDISK(_MODULE) */
> +	default:
> +		printk(" unsupported disk (major = %u)\n", major);
> +		return 0;
> +	}

This is horribly ugly.  Just do an inkernel ioctl instead.  Something
like the following (untested!):

	struct hd_geometry geo;
	int err;

	err = ioctl_by_bdev(bdev, HDIO_GETGEO, &geo);
	if (err)
		return err;

	g_head = geo.heads;
	g_sect = geo.sectors;
	

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/