On Thu, 8 May 1997 14:06:04 -0400 (EDT) 
 wonko@blackhole (Brian Hechinger) wrote:

...ah, for port-i386, which didn't see the original message, it appears
as if the last kinks are worked out of the bounce buffer code.
Juergen Hannken-Illjes has reported that the kernel runs very well ...
In his test, of 34807 dma map loads for ISA, 22521 of them resulted in
bounces.  No errors or warnings were reported.  There were 36 loads
per second, and 27 bounces per second.

I'm currently working on converting the isadma controller code (for
non-busmasters) to the bus dma framework, and creating a better API
for isadma controller users to use.

 > please ignore my broken email headers, they will be fixed this weekend.
 > 
 > > Excellent!  Thank you very much for the good news!  I can't even begin
 > > to describe how happy this makes me :-)
 > 
 > makes you feel proud of the work you've done.
 > 
 > > I think that the extra memory is an acceptable trade-off.  After all,
 > > you've installed extra memory in the machine, yes?  :-)
 > 
 > am i willing to give up 1 or 2M of ram to add RAM over 16M??  you betcha.
 > even if i add 8M of ram and have to sacrifice 2M for bounce buffers, that's
 > an extra 6M of RAM for the system i wouldn't have had otherwise. (and every
 > Meg counts, especially when you run Solaris <g>)  :)
 > 
 > congratulations on the excellent work.
 > 
 > -brian
 > 
 > -- 
 > "Yes, evil comes in many forms, whether it be a man-eating cow or
 >  Joseph Stalin, but you can't let the package hide the pudding!  Evil
 >  is just plain bad!  You don't cotton to it.  You gotta smack it in the
 >  nose with the rolled-up newspaper of goodness!  Bad dog!  Bad dog!"
 >                                 --The Tick
 > 

Jason R. Thorpe                                       thorpej@nas.nasa.gov
NASA Ames Research Center                               Home: 408.866.1912
NAS: M/S 258-6                                          Work: 415.604.0935
Moffett Field, CA 94035                                Pager: 415.428.6939