/dev/vcs: stuck with dimensions <255?

/dev/vcs: stuck with dimensions <255?

[Mario Lang]

Hi.

I just realized, that /dev/vcsa behaviour is broken when
used with lines or columns more than 255.  I can easily
get this behaviour by using the built-in 4x6 font on a
1024xsomething resolution.  That results in 256 columns.  However,
the vcs devices expose dimensions and cursor position in the first
4 bytes.

My question is now:  Are we stuck with this now?  I found that
it is at least possible to use TIOCGWINSZ on the corresponding
/dev/tty%d device to get correct dimensions, but then again, how
would I optain the cursor position?

I am wondering why a char was choosen at all.  I see no gain
by "saving" space there...

-- 
CYa,
  Mario

Re: /dev/vcs: stuck with dimensions <255?

[Samuel Thibault]

Le mar 30 sep 2003 à 11:10:11 +0200, Mario Lang a tapoté sur son clavier :
> I just realized, that /dev/vcsa behaviour is broken when
> used with lines or columns more than 255.  I can easily
> get this behaviour by using the built-in 4x6 font on a
> 1024xsomething resolution.  That results in 256 columns.  However,
> the vcs devices expose dimensions and cursor position in the first
> 4 bytes.
> 
> My question is now:  Are we stuck with this now?  I found that
> it is at least possible to use TIOCGWINSZ on the corresponding
> /dev/tty%d device to get correct dimensions, but then again, how
> would I optain the cursor position?
> 
> I am wondering why a char was choosen at all.  I see no gain
> by "saving" space there...

Well, vcsa was added a long time ago, I guess such dimensions weren't
possible at that time.

Of course we aren't stuck: we just need to add /dev/vcsx* which would
hold 32-bits integers, hoping this is sufficient once for all...

Regards,
Samuel