Re: [PATCH] 498+ days uptime

Juha Virtanen (jiivee@hut.fi)
Mon, 24 Aug 1998 09:54:45 +0300


David Luyer writes:
:>beat it, I believe). Generally on highly multi-user machines, the first one
:>to wrap is the context switch count.

It's not. Quickest wrappers are network interface byte
counters. Do have some real value they should be 64 bits wide.

:>had? Is there some low priority job scheduler in the kernel to do this
:>kind of thing (weekly or so tasks which belong in kernel space)?

Could idle task do this?

:>We can be fairly certain that the stats won't wrap within a week, or we

Depends on counter. 4 GB of network traffic isn't that much at
all (my home box has about 0.5 GB network traffic daily and it
mostly consists of running X apps over network).

Theoretically, it takes only 344 seconds to push over 4 GB stuff
out to 100 Mb/s ethernet.

Also, IDE devices tend to generate huge amounts of interrupts on
busy machine.

:>for a more active Ultrix box). If I had a Linux box like that, I'd like an
:>accurate uptime report (but again, it's not vital, just nice).

I'd like to see accurate counters especially for network traffic.
But, as we're now in code-freeze, I suppose 64-bit counters will
be one of the new features in 2.3.x series.

Juha

-- 
<URL:http://www.iki.fi/jiivee/>

Säästä paperia - älä tulosta tätä meiliä. Save paper - do not print this e-mail.

- 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.altern.org/andrebalsa/doc/lkml-faq.html