I'm not saying I disagree with your conclusion about touching less
memory (certainly today this is very true), but there was so much other
work going on at that time to optimize the X server at that time (some
of which I did, some of which had to do with touching memory less and
some of which didn't) that I'm not sure you can really say that the
speed improvements were the direct result of touching less memory as
opposed to simply executing less code, or touching the frame buffer
(often slower than memory) less. CPU's were pretty slow in those days
(compared to the relative speed of today's CPU's and memory) and it
sometimes made sense even to touch memory *more* using lookup tables and
such in order to save instructions.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/