Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 28 Oct 1999 18:27:35 -0700
From:      "Ronald F. Guilmette" <rfg@monkeys.com>
To:        Kazutaka YOKOTA <yokota@zodiac.mech.utsunomiya-u.ac.jp>
Cc:        freebsd-bugs@freebsd.org
Subject:   Re: kern/14566: Non-kernel programs have little/no control over VGA/VESA devices 
Message-ID:  <3401.941160455@i180.value.net>
In-Reply-To: Your message of Fri, 29 Oct 1999 09:00:14 %2B0900. <199910290000.JAA06014@zodiac.mech.utsunomiya-u.ac.jp> 

next in thread | previous in thread | raw e-mail | index | archive | help

In message <199910290000.JAA06014@zodiac.mech.utsunomiya-u.ac.jp>, you wrote:

>
>> Yes, the vgl(3) library contains functions which implement the kinds of
>> things that I was talking about, and yes, there should be a reference to
>> vgl(3) _somewhere_ on the vga(4) man page.
>> 
>> However, I think that my bug report is still very valid.
>[...]
>> The vgl(3) library is an ordinary, user-level library, properly documented
>> in Section 3 of the manual.  But the functionality it provides *must* be
>> implemented in terms of some lower-level (kernel) capabilities.  Those
>> lower level (direct hardware control) capabilities are obviously available
>> in user land (or else the vgl(3) library could not have been built) but I
>> still don't know how _I_, as a programmer, can get at those lower level
>> capabilities directly.
>[...]
>
>I don't know how closely you want to control the video hardware.  But,
>I have a version of slightly enhanced vgl library which can do VESA
>modes as well as the standard VGA modes (the version of vgl included
>in -CURRENT and -STABLE cannot handle the VESA modes *sigh*).
>
>Soeren and I developed it, but have not yet committed to the source
>tree.  I can send you the new version for review if you like.


Thank you.  It is a generous offer, however I think you're not understanding
what I am really driving at.

I am merely arguing in favor of strict adherence to the general principal
that low-level hardware functionality ought to be accessed in all cases
via some well-documented user/kernel interface that is both documented and
also available to _all_ applications.

In fact, I do not have _anything_ specific that I want to do to the video
hardware _today_.  I do however worry about the things that I may want to do
in the future, and about how I will find out how to acomplish those things.
And also, as I mentioned (above) I think it is Important to stick to first
principals and to provide a comprehensive _abstraction_ of the actual hard-
ware at the user/kernel interface.  (This could be important if, for example,
someone somday wants to write a new OS and wants to have it provide full
FreeBSD emulation, just as FreeBSD now provides Linux emulation.)


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-bugs" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3401.941160455>