Re: top performance sucks (cpu cycles of course) :)

Rob Hagopian (hagopiar@vuser.vu.union.edu)
Fri, 6 Feb 1998 13:21:32 -0500 (EST)

It's not very efficient becaue it uses /proc to get all of the info in
human readable form, parses it, then displays it all again. There was talk
at one point about putting all the data in a binary format too, but I
haven't heard about it lately...
-Rob H.

On Fri, 6 Feb 1998, David Burrows wrote:

> This is 2.1.85 compiled UP, on a 486dx33 with 8mb acting as a masquerading
> ppp gateway for a small lan (with caching nameserver too <g>).
> Everything is working perfectly for once as it should. I have this
> kernel also working on two P133s compiled with pgcc 1.01. Anyway Check
> this out on the 486..
>
> PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME COMMAND
> 1323 root 19 0 588 588 460 R 0 7.6 8.6 0:04 top
>
> Of course this agrees with Heisenberg's uncertainty principal, at least in
> the manner you cannot analyze something without affecting the experiment.
> But 7.6%!? How is this value calculated? It doesn't seem very efficient,
> or is it?
>
> Regards,
>
> Dave.
>
>
>
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.rutgers.edu
>

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu