lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date:   Tue, 17 Jan 2023 16:51:26 +0100
From:   Fabio Fantoni <fantonifabio@...cali.it>
To:     Christoph Hellwig <hch@...radead.org>
Cc:     corbet@....net, axboe@...nel.dk, linux-block@...r.kernel.org,
        linux-doc@...r.kernel.org, linux-kernel@...r.kernel.org,
        Sergei Shtepa <sergei.shtepa@...am.com>
Subject: Re: [PATCH v2] documentation: fix Generic Block Device Capability

Il 17/01/2023 07:05, Christoph Hellwig ha scritto:
> On Mon, Jan 16, 2023 at 07:14:41PM +0100, Fabio Fantoni wrote:
>> Il 16/01/2023 18:53, Christoph Hellwig ha scritto:
>>> On Tue, Jan 10, 2023 at 02:21:04PM +0100, Fabio Fantoni wrote:
>>>> - * ``GENHD_FL_REMOVABLE``: indicates that the block device gives access to
>>>> + * ``GENHD_FL_REMOVABLE`` (0x01): indicates that the block device gives access to
>>> The numberic values really do not belong into the documentation.  They
>>> are just implementation details.
>>>
>> Thanks for reply, if values are not into the documentation see from
>> /sys/block/<disk>/capability output what flags are enabled will require look
>> to source code of include/linux/blkdev.h and
>> Documentation/block/capability.rst will be less useful, or I'm wrong?
> Urgg.  I think this file is generally just a bad idea.  The flags are
> kernel internal and not isolated from userspace.  It seems nothing broke
> with the various renumbering lately, but we need to isolate it from
> the implementation details.  And if it really should be a user API
> we need a text version of it.

Up to 5.13 the page was "ok" 
(https://www.kernel.org/doc/html/v5.13/block/capability.html) from 5.14 
doc comments of other functions was appaired, from 5.17 the flags had 
"major" changes and values was removed from doc comment 
(https://www.kernel.org/doc/html/v5.17/block/capability.html)

With this patch return to have only block capability flags in this file 
and including the values: https://snipboard.io/QrN6hg.jpg

But as you tell that this file is a bad idea, 
Documentation/block/capability.rst should be removed?

About if should be a user API, I don't have enough experience to be able 
to say if and in which cases it could be useful on the user side. From 
what I could see with the documentation page correct is that you can at 
least immediately see (after a very fast search) for example if the 
block device is removable and if it is partitionable. And even if the 
values was changed pointing to the doc page of the used kernel version 
and having the values in doc is still possible see what flags are 
enabled from doc page.

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ