Re: Ideas for reducing memory copying and zeroing time
Linus Torvalds (torvalds@cs.helsinki.fi)
Fri, 19 Apr 1996 12:53:50 +0300 (EET DST)
> > where it won't write allocate (which makes a lot of sense on small
> > caches and produces a better hit pattern, the PC memory subsystem is
> > pretty fast anyways though...)
>
> Pardon ?. If the PC one is "fast" the sparc one must be absolutely awful to
> make it look that. Triton isnt bad but neptune definitely leaves me with
> bandwidth problems, even single CPU.
PC memory subsystems _are_ reasonably fast, at least when done well. Most of
the newer Pentium machines get pretty good performance out of the memory
subsystem (despite the lack of write buffers etc on the CPU). My P166 actually
has _half_ the memory latency when compared to my Alpha Ev5 running at 300MHz..
That's the sync cache and triton, obviously. The bad PC memory subsystems
really are bad..
Linus