Re: GGI Project Unhappy On Linux

Jes Degn Soerensen (jds@kom.auc.dk)
28 Mar 1998 13:38:07 +0100


>>>>> "Chris" == Chris Evans <chris@ferret.lmh.ox.ac.uk> writes:

Chris> On 28 Mar 1998, Jes Degn Soerensen wrote:

>> There is nothing the prevents you from pipelining the blit
>> operations in user space as well and/or start doing other things
>> while `waiting' for a blit operation to finish.

Chris> The `waiting' as you put it is the precise problem, as it is a
Chris> polling operation, maybe something like

Chris> (while !(inw(gfx_base+status) & status_acceldone)

I am aware of that but if you have two operations to perform (A and B)
and they do not conflict then its pretty silly to do

issue(A)
waitblit()
issue(B)
waitblit()

In fact you normally want just to do that waitblit() right before you
start messing with the video memory and not one instruction before.

This does not solve the interrupt generation option, but in many cases
it will be faster than sleeping waiting for an interrupt.

Anyway, I don't see why we couldn't set up an event generation system
for graphics boards with fbcon if people want it.

Jes

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu