Re: [rtl] RTAI latency (Athlon 800 - XFree 4.0.1)

[yodaiken@fsmlabs.com]

On Mon, Nov 20, 2000 at 02:43:38PM -0700, John Regehr wrote:
> > > My suggestion is that you recompile X with this macro set to false - so that X doesn't
> > > try to disable irqs and see if that fixes your problem. I suspect that this is not
> > > your problem.
> > 
> > You are correct that it is probably not the dissabling of interruprts
> > that is the problem, bacuse that is only used in a few places (the ATI
> > driver and in SupperProbe) and only to do the clockprobing. 
> 
> Video drivers can hose RTLinux/RTAI by writing to a command FIFO on the
> video board without first checking if it's full.  In the past some
> Windows drivers have done this as an "optimization" (it speeds up the

This is possible, however, I suspect that Edwin's problem is not with X.
Before speculating too much, he should see if a big I/O load without X
gets the same behavior.


> common case by eliminating a PCI read).  A write to a full FIFO will
> stall the PCI bus (and the rest of the machine) - this condition can
> last for many milliseconds.

Many milliseconds? 
Do  you have an example hardware to test?