Ok, after the amazingly fast bugfix/commit by Chuq & Ross, the Multia is
once again a usable X terminal. :)  For the curious, here is the previous
benchmarks, adding the '1.4.1 (UVM fix)' tests.  These last two were done
with a local 1GB SCSI with 1.4.1 (actually 'release' branch) kernel and
userland, with Chuq's UVM fix.

NetBSD 1.3.3 + stock Xserver
   120 reps @  57.7834 msec (    17.3/sec): Copy 500x500 from window to window
  2000 reps @   4.6569 msec (   215.0/sec): 500x500 rectangle
    70 reps @  81.7023 msec (    12.2/sec): (xor) 500x500 rectangle
  5000 reps @   1.2423 msec (   805.0/sec): Fill 100x100 equivalent cmplx poly
160000 reps @   0.0403 msec ( 24800.0/sec): Char in 80-char image line (6x13)
300000 reps @   0.0221 msec ( 45300.0/sec): 10-pixel line segment

NetBSD 1.4 + stock Xserver
    32 reps @ 164.3807 msec (     6.1/sec): Copy 500x500 from window to window
    50 reps @ 122.5877 msec (     8.2/sec): 500x500 rectangle
     2 reps @ 3764.8863 msec (     0.3/sec): (xor) 500x500 rectangle
   200 reps @  42.6437 msec (    23.5/sec): Fill 100x100 equivalent cmplx poly
  2400 reps @   2.1210 msec (   471.0/sec): Char in 80-char image line (6x13)
  4000 reps @   1.2572 msec (   795.0/sec): 10-pixel line segment

NetBSD 1.4 + R. C. Dowdeswell's XalphaNetBSD
   800 reps @   8.8409 msec (   113.0/sec): Copy 500x500 from window to window
  2000 reps @   2.5193 msec (   397.0/sec): 500x500 rectangle
  1000 reps @   5.3003 msec (   189.0/sec): (xor) 500x500 rectangle
  8000 reps @   0.6773 msec (  1480.0/sec): Fill 100x100 equivalent cmplx poly
 56000 reps @   0.0946 msec ( 10600.0/sec): Char in 80-char image line (6x13)
100000 reps @   0.0514 msec ( 19500.0/sec): 10-pixel line segment

[New Tests]
NetBSD 1.4.1 (UVM patch) + stock XalphaNetBSD
   120 reps @  57.4746 msec (    17.4/sec): Copy 500x500 from window to window
  2000 reps @   4.6990 msec (   213.0/sec): 500x500 rectangle
    70 reps @  81.6900 msec (    12.2/sec): (xor) 500x500 rectangle
  5000 reps @   1.1911 msec (   840.0/sec): Fill 100x100 equivalent cmplx poly
160000 reps @   0.0394 msec ( 25400.0/sec): Char in 80-char image line (6x13)
300000 reps @   0.0217 msec ( 46100.0/sec): 10-pixel line segment

NetBSD 1.4.1 (UVM patch) + R. C. Dowdeswell XalphaNetBSD
   800 reps @   8.8409 msec (   113.0/sec): Copy 500x500 from window to window
  3000 reps @   2.4917 msec (   401.0/sec): 500x500 rectangle
  1000 reps @   5.2983 msec (   189.0/sec): (xor) 500x500 rectangle
  8000 reps @   0.6773 msec (  1480.0/sec): Fill 100x100 equivalent cmplx poly
 56000 reps @   0.0992 msec ( 10100.0/sec): Char in 80-char image line (6x13)
100000 reps @   0.0509 msec ( 19600.0/sec): 10-pixel line segment

So, 1.4.x is as fast as (and might be just slightly faster than) 1.3.3, and
this is _great_ news! (well ok, the fix needs to be pulled up for 1.4.x)

-Andrew
-- 
-----------------------------------------------------------------
Andrew Gillham                            | This space left blank
gillham@whirlpool.com                     | inadvertently.
I speak for myself, not for my employer.  | Contact the publisher.