Erwin Rol wrote:
> 
> 
> Kevin Brosius wrote:
> >
> >
> > I suppose we could add an option, something like 'early_sync'  (or maybe
> > something more descriptive) which changes the accelerator sync behavior
> > in 4.0.x drivers to make them behave more like 3.3.6.  I'd be willing to
> > make a trial s3virge driver with a change like this so you could test
> > it.
> 
> I have no problem with being a crash test dummy :-) (will have to switch
> to
> XFree 4.0.1 again though :-)  I will have to give back the G400 on
> monday so
> can't test that anymore,but the super s3virge card is all mine :-). I
> have the
> Xfree 4.0.1 sources here so iyou could just mail me a patch or even just
> say
> what i should change where (if it are small changes) I will try to get
> my hands
> on some other videocards so i can try those too.
> 

If you want to try it yourself, I was going to add the function
S3VAccelSync() to the end of all the functions containing 'Subsequent'
in drivers/s3virge/s3v_accel.c.  (You could make a patch or driver
module available that has this change hard coded, or we could add a
driver option to switch it on and off.  The switching option might add a
minor performance penalty.)

The idea is to simulate 3.3.6 behavior by forcing the accelerator to go
idle at the end of every accelerated graphics primitive.  4.0.x is able
to handle the accel. setup more efficiently, by offering 3 functions
used for a primitive.

They are:
  Setup_xxx      - initialize accel for a primitive
  Subsequent_xxx - perform a primitive
  AccelSync      - wait for accel operation to complete

The normal sequence of operations is:
  Setup, Subsequent (possibly multiple times), AccelSync (only when
required to flush the FIFO).

By adding the AccelSync to every Subsequent operation, you'll flush the
FIFO after all primitive operations, of course paying a performance
penalty.

Kevin
-- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
---
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/

Reply via email to