Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   "Namjae Jeon" <namjae.jeon@samsung.com>
To:     =?windows-1257?Q?'Valdis_Kl=E7tnieks'?= <valdis.kletnieks@vt.edu>
Cc:     <linux-fsdevel@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
        <gregkh@linuxfoundation.org>, <hch@lst.de>,
        <sj1557.seo@samsung.com>, <pali.rohar@gmail.com>, <arnd@arndb.de>,
        <viro@zeniv.linux.org.uk>, "'Namjae Jeon'" <linkinjeon@gmail.com>,
        "'Sasha Levin'" <sashal@kernel.org>
In-Reply-To: <89603.1581722921@turing-police>
Subject: RE: [PATCH] exfat: tighten down num_fats check
Date:   Mon, 17 Feb 2020 09:37:55 +0900
Message-ID: <001b01d5e52a$7f029340$7d07b9c0$@samsung.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="windows-1257"
Content-Transfer-Encoding: 7bit
Thread-Index: AQEsiKNbA3f6Ud8ZBzvjODngDpGbQQJ4Yw3zqV1wpoA=
Content-Language: ko
CMS-TYPE: 101P
DLP-Filter: Pass
References: <CGME20200214232853epcas1p241e47cdc4e0b9b5c603cc6eaa6182360@epcas1p2.samsung.com>
        <89603.1581722921@turing-police>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

> Change the test for num_fats from != 0 to a check for specifically 1.
> 
> Although it's theoretically possible that num_fats == 2 for a TexFAT
> volume (or an implementation that doesn't do the full TexFAT but does
> support 2 FAT tables), the rest of the code doesn't currently DTRT if it's
> 2 (in particular, not handling the case of ActiveFat pointing at the
> second FAT area), so we'll disallow that as well, as well as dealing with
> corrupted images that have a trash non-zero value.
> 
> Signed-off-by: Valdis Kletnieks <valdis.kletnieks@vt.edu>
> 
> --- a/fs/exfat/super.c	2020-02-14 17:45:02.262274632 -0500
> +++ b/fs/exfat/super.c	2020-02-14 17:46:37.200343723 -0500
> @@ -450,7 +450,7 @@ static int __exfat_fill_super(struct sup
>  	}
> 
>  	p_bpb = (struct pbr64 *)p_pbr;
> -	if (!p_bpb->bsx.num_fats) {
> +	if (p_bpb->bsx.num_fats  != 1) {
>  		exfat_msg(sb, KERN_ERR, "bogus number of FAT structure");
Could you please update error message for the reason why num_fats is allowed
only 1?
>  		ret = -EINVAL;
>  		goto free_bh;
Let's remove exfat_mirror_bh(), FAT2_start_sector variable and the below
related codes together.

sbi->FAT2_start_sector = p_bpb->bsx.num_fats == 1 ?
                sbi->FAT1_start_sector :
                        sbi->FAT1_start_sector + sbi->num_FAT_sectors;

Thanks for your patch!
> 
> 
> 
>