Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 8 Nov 1996 11:12:08 +0100 (MET)
From:      sos@freebsd.org
To:        proff@suburbia.net (Julian Assange)
Cc:        sos@freebsd.org, hackers@freebsd.org
Subject:   Re: patches for nice text modes (170x48x16 etc) (syscons.c)
Message-ID:  <199611081012.LAA29180@ra.dkuug.dk>
In-Reply-To: <199611080939.UAA03898@suburbia.net> from "Julian Assange" at Nov 8, 96 08:39:27 pm

next in thread | previous in thread | raw e-mail | index | archive | help
In reply to Julian Assange who wrote:
> 
> > In reply to Julian Assange who wrote:
> > 
> > OK, I'll just inform that this is not the way it will be done in
> > the official syscons. I have discussed this with Julian, but he
> > hasn't the time to change this to the proposed interface (which
> > isn't in the official sources yet).
> 
> Your "proposed interface" doesn't work, and never will work without
> substantial modifications.  Despite my [previously stated] objections to
> it I actually tried very hard implimenting the patch as an lkm using
> your new sc_user_ioctl() hook as you insisted. It was simply impossible,
> given the variables used and routines that are called when the screen is
> (physically) resized.

Now wait a minute, the ioctl thing is there and it works, as I
also stated to you in private mail was that some other _minor_
changes to syscons would be nessesary (mostly removing some
static declarations), and that I would help out with that part.
You flatly refused to change the code, due to lack of time as I 
remember, maybe you tried later I don't know.

> It is beyond me, that you insist that something so fundemental to the
> opperations of the console as changing the number of rows/cols on the
> screen should be a module. Even so, I would be accepting of your decision if
> your interface was actually capable of doing the job.

In your case it might be simple, but IMHO what you have done
is only a half solution, in that it does not support switching
screens to other modes etc. If you did care to take the time
a do this in a general manner, I would have no trouble in
including it, but putting in X hacks for Y half solutions
is just not my idea of keeping the quitlity of our sources
in a resonable shape.
> 
> As I recall my fal words on the matter were that you should convert my
> patch to your [moving target] interface.

And there will be a solution to this, but I have only so many hours
in a day...

> I'd have thought the maintainer of syscons would have been delighted to
> see a patch that substantially increased functionality. All I have got
> out of it is negativity, complications and a complete waste of my time.

Sorry about that, but if we don't do things like this in a general
way, we will end up with a multiheaded monster, that noone cares
to maintain. If you would take the task of maintaining syscons, and
devote the time needed (which is considerable, I've been there)
then I'd gladly hand it over to you, I have plenty to do in
other areas of the system, plus a full time job, vife and kids.

> If this is FreeBSD evangelism, God save us.

No, its called common sense and responsibility to our users..
 

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Soren Schmidt             (sos@FreeBSD.org)             FreeBSD Core Team
               So much code to hack -- so little time.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199611081012.LAA29180>