Re: Pgcc in Deb
[sorry for carrying this back into the mailing list without asking Michael
first, but it seems appropriate ...]
Michael Stone wrote:
>
> No, the problem is that your pentium optimizations haven't been shown to
> have a noticable impact in anything but fringe cases for people using
> debian. (I.e., disk or network waits are so much larger than cpu waits
> that optimizing cpu doesn't make a significant difference.) If you do
> benchmarks that disprove this, it might be revisted. (But be aware that
> many people have already tried this and were not impressed with the
> results.)
When I took an OS course, we were asked to write a round-robin scheduler that
simulates processes with I/O blocking. It wasn't fun, but what I remember
from that experience is that it will make a difference when you are running
a lot of processes. That's the whole point of a good scheduling algorithm
right? Keeping the CPU loaded, fairness, etc. Normally, your CPU shouldn't
be doing any idle loops except in the case of badly written programs, and
doing spin-locks for multi-threading. In that case, if I were to think that
my computer is waiting for I/O all the time, I could also fancy turning down
all the optimizations would be a good idea to introduce more stability in
my system. That is, it seems to me too conservative to favor an argument
that talks about I/O waits when there are so many apps running on a typical
desktop.
orion:gcc-2.95.2$ ps aux | wc -l
91
Do you see my point?
About the previous attempts: I wonder if they've simply compiled the whole
distro, and tried it that way. If you could get a running pentium-pro
optimized system on an *ordinary* PII/PIII box, I'm pretty sure you'd get
some difference.
If I do benchmarks, would this be revisited? I'm eager to burn my cluster
for a pentium-pro optimized Debian. I used to optimize every single piece
of code when I wrote assembly, and I feel irritated when I think of the
possible optimizations on my system.
> It is also very important to be aware that pentium
> optimizations will not improve (and may actually lower) performance on
> other architectures, such as K6 or PentiumIII. The archive costs
> associated with architecture-specifc optimizations have historically
> seemed much higher than the benefits of those optimizations. (Barring
> the gee-whiz factor, which we try to keep out of this sort of decision.)
>
Are you sure about the PentiumIII? I would think that it's backwards
compatible with PentiumII, and implementing the same optimization paths.
However, I do know that you need to compile separately for the K6, but
in K7 you should be able to go -march=pentium-pro.
I guess there wouldn't be much archive cost if it were done just once, or
in a limited fashion. It could be, done for example, when the potato comes
out of the freeze. [ though I plan to do it single handedly as soon as I
finish installing my cluster ]
Regards,
--
++++-+++-+++-++-++-++--+---+----+----- --- -- - -
+ Eray "eXa" Ozkural . . . . . .
+ CS, Bilkent University, Ankara ^ . o . .
| mail: erayo@cs.bilkent.edu.tr . ^ . .
Reply to: