Fwd: Linux guru needed to write device driver. Burlington, Massachusetts.

Jerry Feldman gaf at blu.org
Tue Mar 4 14:11:38 EST 2003


On Tue, 4 Mar 2003 13:54:17 -0500
"Wizard" <wizard at neonedge.com> wrote:

> I'm not so sure it's that easy. I haven't written any true OS device
> drivers, but I have written hardware-level code, 8051 embedded code,
> and 80x86 assembly language stuff, and for that I really had to
> understand some things about how the hardware itself worked (I/O
> lines, fetch/store procedures, memory addressing, timings, etc). Much
> of the interfaces to the hardware on the platforms that I have worked
> were basically just simple re-implementations of what the chips on the
> board did. I have an electronics background, so I understood how the
> chips themselves worked and could see it in the interface. I don't
> know that it would be a simple thing for someone lacking that
> experience to just 'pick-up' (I could easily be wrong though). I don't
> think that most high-level languages like Perl, Tcl, C++, or even C
> reflect what the machine is actually doing.
Writing hardware level code and writing device drivers are two different
things. I have not written any Linux device drivers, and only one Unix
one, but I have written them for other systems. The problem with a
device driver is that it is a balancing act. You've got to understand
the hardware (one skill), the OS internals, such as buffering,
multi-tasking stuff like spinlocks and semaphores, and performance. A
poorly written device driver can affect an entire system, even if it is
a slow device like a terminal. 
-- 
Jerry Feldman <gaf at blu.org>
Boston Linux and Unix user group
http://www.blu.org PGP key id:C5061EA9
PGP Key fingerprint:053C 73EC 3AC1 5C44 3E14 9245 FB00 3ED5 C506 1EA9
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.blu.org/pipermail/discuss/attachments/20030304/8fd66daa/attachment.sig>


More information about the Discuss mailing list