vm86 in kernel [was: vesafb...]

[Aki M Laukkanen]

On Tue, 25 Jan 2000, Kendall Bennett wrote:
> Yes, I am talking about text framebuffer console acceleration and 
> mode sets. Graphics mode consoles can benefit tremendously from 
> hardware accelerated blitting (for scrolling) and hardware 
> accelerated text rendering for drawing the fonts. 

Ok, I'll see what you are getting at considering the statements about
XFree 4 modularised driver support later. 

Here are some timing results which I consider enlightening. I'll have to
say I was surpised. These results were gotten by cycling the available
modes in X and panning around in the virtual screen with a mouse.

                     Name|Time (ms)|Times
-------------------------+---------+-----
pan_display              |   13.587|  268
encode_fix               |    0.011|   21
set_var                  |  464.447|   21

As you can see encode_fix does not call LRMI_int() and the time it takes
might be mostly gettimeofday() overhead (I have to change my timing
code to use rdtsc some time). However even pan_display which I know to be
just a couple of outb-instructions on CT6555x takes 13.5 ms. The time for
set_var() which might call SET_MODE, SET_LOGICAL_SCAN_LINE_LENGTH and
SET_DISPLAY_START depending of the mode is almost half a seconds. I have
to double-check that it does not try to clear the screen by itself but I'm
sure I got it right. Scheduling policy of the daemon was SCHED_RR.

> framebuffer console support that can utilise XFree86 4.0 loadable 
> modules to implement the mode set and blitting functions. Why re-
> write code when it could be re-used (even at the binary level!).

Which makes sense. I haven't researched this myself but according to Matan
Ziv-Av (svgalib maintainer) supporting XFree 4 loadable modules will not
be a easy task. There are apparently hundreds of support functions which
these drivers need.
 
> Forget the VBE 3.0 PMI interface functions, at least for now. 
> Unfortunately a number of well known graphics cards (NVIDIA boards 
> from Diamond) shipped with a broken VBE 3.0 PMI interface. NVIDIA 

I had a feeling that this would be the case. Hence the idea concerning
black listing BIOSes. Thanks for the information. I knew that the BIOS of
my laptop (Toshiba) would not be alone in this respect. :(

> I will email you copies of the code that you can use, under the GPL 
> license. Feel free to cut and paste what stuff you need, but make 
> sure you retain my copyright in your code somewhere.

Thank you. I certainly will.

-- 
D.


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo en vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/