[lkml]   [1996]   [Aug]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: interrupt counts
On Mon, 19 Aug 1996, Mikael Abrahamsson wrote:

> On Mon, 19 Aug 1996, Matti E Aarnio wrote:
> > HOWEVER, I might want to have it as a configuration option.
> > At low-level 386/25 (my home machine) I am entirely happy
> > with 32-bit counters.
> >
> > bool 'Do you want unwrappable (64 bit) network counters ?' \
> I dont know how much handling 64bit instead of 32 bit decereases
> performance, but this is definately an option.

This seems a little excessive to me. On my 486-33 (DX) I can
increment a long long or a short ten million times with a time
difference of only about 1.25 seconds, total time is 5 - 6
seconds. (This is with the 8k cache). Note this is also with
no optimization.

So if you lose 1.25 seconds in the week or so that the counter
flips over on a 386...

Pascal, n.: A programming language named after a man who would
turn over in his grave if he knew about it.

 \ /
  Last update: 2005-03-22 13:38    [W:0.040 / U:0.248 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site