On 2000-09-06T12:52:29,
Linus Torvalds <torvalds@transmeta.com> said:
I do agree with your assessment.
Except for a single point:
> And quite frankly, for most of the real problems (as opposed to the stupid
> bugs - of which there are many, as the latest crap with "truncate()" has
> shown us) a debugger doesn't much help. And the real problems are what I
> worry about. The rest is just details. It will get fixed eventually.
I want these to get fixed _soon_. Not later. I want fixing these to be made as
easy as possible, because they are annoying and don't have to be, because the
"simple" problems are easier spotted with a debugger.
If a real problem gets spotted easier and fixed correctly, all the better. If
some people are able to better understand the code and what it does with a
debugger, all the better.
The community peer review will weed out incorrect patches. I doubt there will
be too many of them or that people will suddenly start patch bombing l-k with
crap because a debugger has been added.
Hell, if a debugger only supplies a single clue to make people go "Duh, I was
looking in the entirely wrong direction, better get to understand how the
networking code works instead of hunting it in the VM layer", it is well worth
it.
I don't need quick and shallow fixes any more than you do. I want the code to
stabilize, and do The Right Thing(tm). For all "fixes" which lead to clumsy,
harder to understand code - see figure 1.
For those fixes which can be spotted because we can easily get a consistent
view of what happened on the system the time it crashed - no more "What
processes were you running, try disabling <making a more-or-less-educated
guess> and see if it persists" -, debugger output which makes you go "Duh, I
have a typo there", I _do_ care for a debugger.
> I do realize that others disagree. And I'm not your Mom. You can use a
> kernel debugger if you want to, and I won't give you the cold shoulder
> because you have "sullied" yourself. But I'm not going to help you use
> one, and I wuld frankly prefer people not to use kernel debuggers that
> much. So I don't make it part of the standard distribution, and if the
> existing debuggers aren't very well known I won't shed a tear over it.
>
> Because I'm a bastard, and proud of it!
>
> Linus
This makes for a good Think Geek shirt, and I will personally buy one and try
to get it signed by Alan.
But for trying to shepherd a project the size of the Linux, I do think it is
the wrong attitude if taken to the extremes. Not trying to make it easier for
the masses to provide better QA data, or even fix the stupid errors
themselves, so that the core team can focus on the real hard bugs and design
issues, I call that wrong.
Sincerely,
Lars Marowsky-Brée <lmb@suse.de>
Development HA
-- Perfection is our goal, excellence will be tolerated. -- J. Yahl- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Sep 07 2000 - 21:00:28 EST