[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#595805: ITP: devmem2 -- Simple program to read/write from/to any hardware address



Dear all,

On 09/07/2010 12:54 AM, Ben Hutchings wrote:
> On Mon, 2010-09-06 at 21:37 +0200, Michael Opdenacker wrote:
>   
>> Package: wnpp
>> Severity: wishlist
>> Owner: Michael Opdenacker <michael.opdenacker@free-electrons.com>
>>
>>
>> * Package name    : devmem2
>>   Version         : 1.0.0
>>   Upstream Author : Jan-Derk Bakker <jdb@lartmaker.nl>
>> * URL             : http://www.lartmaker.nl/lartware/port/devmem2.c 
>> * License         : GPLv2 or later
>>   Programming Lang: C
>>   Description     : Simple program to read/write from/to any hardware address
>>
>> devmem2 can be used to access physical addresses 
>> in your system, when allowed by the kernel.
>> It releaves you from having to access
>> /dev/mem by yourself with a C program, and
>> allows for quick experiments with the hardware
>> from the command line, before implementing a
>> real device driver.
>>     
> This is a quick hack, too trivial to deserve a package.  A more useful
> tool would be one that allows you to define named registers and fields
> in the various spaces they may exist (memory, I/O, PCI config space,
> indirect...).  I've written and used something like that in my day job -
> though it wasn't general enough to attempt to release.
>   
Thank you very much for your feedback about devmem2. It's true that it's
a very quick hack, and I now understand that it's not generic and
portable enough to be package in Debian.

However, I do believe something providing this kind of functionality
will be useful. That's all the more true as Debian GNU/Linux is
increasingly used on embedded platforms, and that's the reason why
BusyBox has its own devmem applet (see
http://free-electrons.com/issues/debian/wnpp/595805/devmem.c)

What would you advise me to do next? May I reuse the BusyBox code (which
is an improved version of the original, and probably deserves more to be
included) and propose a package again? Or should we extend the
configuration of the BusyBox package to support the devmem applet (if
needed), and tell users to run "busybox devmem" to get this functionality?

The second choice would be the easiest solution, though less user friendly.

Thanks again,

Cheers,

Michael.

-- 
Michael Opdenacker, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com
+ 33 621 604 642




Reply to: