linux-kernel - Re: [PATCH 2.6.23] SysRq: print hotkey info while pressing undef key, try 2

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] [thread-next>] [day] [month] [year] [list]

Date:	Wed, 17 Oct 2007 16:09:08 -0400
From:	Valdis.Kletnieks@...edu
To:	Yi Yang <yang.y.yi@...il.com>
Cc:	linux-kernel@...r.kernel.org, akpm@...l.org,
	mj@...ey.karlin.mff.cuni.cz, crutcher+kernel@...astacks.com,
	pavel@....cz, schwab@...e.de
Subject: Re: [PATCH 2.6.23] SysRq: print hotkey info while pressing undef key, try 2

On Wed, 17 Oct 2007 23:22:58 +0800, Yi Yang said:

> SysRq has already provided a similiar help before this patch, but it
> is not so clear that the user doesn't know what happened and what
> he/she should do.

The person is in one of two states:

1) He has been told "recreate the problem, hit alt-sysreq-cokebottle,
and send me the results".  He has a mission, and the only feedback he
needs is (a) that he hit cokebottle and not pepsibottle, and (b) the
resulting output.

2) He's already read the file in Documentation/ and just needs a reminder.
In this case, the fact it's only 2 or 3 lines and doesn't scroll other stuff
out of sight is more important.

> In addition, that funtion has a big loop with another big loop
> embedded which is very inefficient, it is intended to skip some hot

You're optimizing code that hopefully never gets executed, and even if
it does, you have the optimization *backwards*.  If you're worried about
the efficiency, trim it down to output 3 lines - do you realize how many
instructions it takes in the VGA and fb drivers to actually *output* all
these lines? (Seriously - I had a 1.6Ghz P4 laptop, where scrolling the
screen with vga=791 actually ran so slowly that it horqued up the timer
initialization code. *That* was a fun bug to figure out..)

> key help info for such a function as "Changing Loglevel", just print
> a help info for this, that is very unnecessary. In fact, the key '0'
>  - '8' have different results the user should know.

And ironically enough, you then output the same exact text for all levels.

> +static char __read_mostly sysrq_hot_key_table[36] = {
> +	'0', '1', '2', '3', '4', '5', '6', '7', '8', '9',
> +	'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j',
> +	'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't',
> +	'u', 'v', 'w', 'x', 'y', 'z'

The lists of 'Not defined' tends to scroll the screen away. The old code
instead focuses on listing the things you *can* do.  If I'm looking at the
help output, I don't care that 'g' is not defined.  I need to be reminded
that 'p' is 'showPc' and D is show-all-locks.

Content of type "application/pgp-signature" skipped