[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <83075.1581903681@turing-police>
Date: Sun, 16 Feb 2020 20:41:21 -0500
From: "Valdis Klētnieks" <valdis.kletnieks@...edu>
To: "Namjae Jeon" <namjae.jeon@...sung.com>
Cc: linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
gregkh@...uxfoundation.org, hch@....de, sj1557.seo@...sung.com,
pali.rohar@...il.com, arnd@...db.de, viro@...iv.linux.org.uk,
"'Namjae Jeon'" <linkinjeon@...il.com>,
"'Sasha Levin'" <sashal@...nel.org>
Subject: Re: [PATCH] exfat: tighten down num_fats check
On Mon, 17 Feb 2020 09:37:55 +0900, "Namjae Jeon" said:
> Could you please update error message for the reason why num_fats is allowed
> only 1?
Sure.. No problem..
> > 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;
You might want to hold off on that part for a bit - I've asked Sasha Levin for
input on what exactly Windows does with this, and Pali has a not-obviously-wrong
suggestion on using the second FAT table. The code tracking FAT2_start_sector
looks OK - what would be missing is doing a similar versioning on the FAT
the rest of the code references.
We may end up heaving that code over the side in the end, but let's make
sure we're doing it with more information in hand....
Content of type "application/pgp-signature" skipped
Powered by blists - more mailing lists