All of lore.kernel.org
 help / color / mirror / Atom feed
From: Gray Remlin <gryrmln@gmail.com>
To: u-boot@lists.denx.de
Subject: [U-Boot] [PATCH 4/4] UBI/UBIFS: Prevent UBI partition change while UBIFS is mounted
Date: Tue, 02 Nov 2010 00:31:50 +0000	[thread overview]
Message-ID: <4CCF5BF6.2070806@gmail.com> (raw)
In-Reply-To: <20101101190631.D5B191524F4@gemini.denx.de>

On 01/11/10 19:06, Wolfgang Denk wrote:
> Dear Stefan Roese,
> 
> In message <201011011420.35851.sr@denx.de> you wrote:
>>
>>>> Only allow (re-)assignment to an UBI partition/device when UBIFS is
>>>> currently not mounted. Otherwise the following UBIFS commands will
>>>> crash.
>>>>
>>>> Signed-off-by: Stefan Roese <sr@denx.de>
>>>> ---
>>>>
>>>>  common/cmd_ubi.c   |   13 +++++++++++++
>>>>  common/cmd_ubifs.c |    5 +++++
>>>>  2 files changed, 18 insertions(+), 0 deletions(-)
>>>
>>> I'm a bit biased here - from standard Unix command usage it seems
>>> natural that you have to manually umount first, but then we have very
>>> smple device handling in U-Boot, with always only one device in
>>> access.  Would it not make sense to auto-unmount in case the user
>>> switches the device?
>>
>> I can implement it this way if preferred. I'll prepare a new for this later.
> 
> As mentioned - I am not sure what would be best here.
> 
> What is your own position?
> 
> And: anybody else to comment?
> 
> Best regards,
> 
> Wolfgang Denk
> 
As a 'User' what matters the most to me is command user interface
consistency
(yes I want the impossible) regardless of the filesystem\device type.

For example to load uImage, I want 'load <device>:<partition>:<file>'
whether
it be an ide device with ext3 filesystem or a mmc device with ubifs.

If a filesystem needs to be mounted for some devices\commands but not others
then it should 'just happen' when needed. As a 'User' my only interest
is the
desired end result with the minimum of brainpower input. I want all related
commands to take the same arguments in the same order. I want to be able to
use them without understanding (technically) what they do.

An acceptable alternative would be to 'select <device>:<partition>' then
have
all further commands refer to said '<device>:<partition>' without restating.
But I would want all commands to work that way.

And personally, I like terse commands, but I know others hate them....Oh
well.

As a developer I know this is unrealistic...but a 'User' what matters
the most
to me is command user interface consistency!

  reply	other threads:[~2010-11-02  0:31 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-10-28 12:09 [U-Boot] [PATCH 4/4] UBI/UBIFS: Prevent UBI partition change while UBIFS is mounted Stefan Roese
2010-10-29 14:01 ` Ben Gardiner
2010-10-29 21:06 ` Wolfgang Denk
2010-11-01 13:20   ` Stefan Roese
2010-11-01 19:06     ` Wolfgang Denk
2010-11-02  0:31       ` Gray Remlin [this message]
2010-11-02 10:49       ` Stefan Roese
2010-11-02 14:33         ` Ben Gardiner
2010-11-14 21:06         ` Wolfgang Denk

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4CCF5BF6.2070806@gmail.com \
    --to=gryrmln@gmail.com \
    --cc=u-boot@lists.denx.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.