From mboxrd@z Thu Jan  1 00:00:00 1970
From: Stephen Warren <swarren@wwwdotorg.org>
Date: Thu, 06 Sep 2012 12:57:40 -0600
Subject: [U-Boot] gpt: GUID/UUID - GPT restoration - open questions
In-Reply-To: <20120906161443.0a884e29@amdc308.digital.local>
References: <1345795995-24656-1-git-send-email-l.majewski@samsung.com>
	<20120824104836.22dfc647@amdc308.digital.local>
	<504896CF.6050505@gmail.com>
	<20120906161443.0a884e29@amdc308.digital.local>
Message-ID: <5048F224.5090009@wwwdotorg.org>
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

On 09/06/2012 08:14 AM, Lukasz Majewski wrote:
> Hi Rob,
> 
>> The question I'm asking is why we need partition creation support in
>> u-boot in the first place? 
> 
> Please consider following scenarios:
> 
> 1. eMMC content is exported to the user (via UMS USB Mass Storage). Then
> user by chance or on purpose will corrupt MBR/GPT.
> 
> 2. misuse of "mmc" command - user by chance writes mmc erase 0.
> 
> To unbrick the device one would simply use "gpt/part restore" command
> with already defined at ./include/config/{target}.h default settings
> for partitions.
> 
>> Next you need filesystem creation too?
> 
> I can use UMS to export eMMC to host PC (with restored partitions) and
> then use mkfs.* to create proper file system.

In that case, wouldn't you be exposing the entire eMMC device over UMS,
and hence the user could just run fdisk/parted/... on the host, just
like you're proposing they run mkfs on the host?