From mboxrd@z Thu Jan  1 00:00:00 1970
From: Richard Weinberger <richard@sigma-star.at>
Date: Tue, 16 Jan 2018 14:22:57 +0100
Subject: [U-Boot] [PATCH] ubi: Fix filesystem corruption on detach when
	fastmap enabled
In-Reply-To: <CABatt_xtfoXRpfOYBPZwD=PmTx_+H+X5gqbYGGLY=KtxVbCQJw@mail.gmail.com>
References: <CABatt_wEa+jiSmutzZrjyRFXqypPTbfdo_Z8CVDhuuRMUXjA6w@mail.gmail.com>
 <2306db96-4988-dfa3-c1d3-d830fa03ada5@denx.de>
 <CABatt_xtfoXRpfOYBPZwD=PmTx_+H+X5gqbYGGLY=KtxVbCQJw@mail.gmail.com>
Message-ID: <8925017.tvIN1CM1GY@blindfold>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

Heiko, Martin,

Am Dienstag, 16. Januar 2018, 10:11:41 CET schrieb Martin Townsend:
> Ah. Must be the mail client. Sorry about that I'll setup git send-mail
> for v2.  I'll wait to see what Richard has to say in case there is a
> better fix.

Thanks for letting me know!
Indeed, there is a problem. I'm a little astonished that nobody noticed so 
far.
Most likely because while detaching UBI in Linux mostly no Fastmap work is 
scheduled,
and therefore in most cases flush_work(&ubi->fm_work) does nothing.

As you noticed in U-Boot the story is different, you always update the Fastmap 
upon detach.

Martin, can you please explain what corruption you see?
>From reading the code I'd assume that you miss volumes but a full scan would 
recover everything.

For Linux I suggest this fix: