Re: Kernel 2.2.14, dirty buffers, stalls in interactivity of system/NFS-clients ...

From: Richard B. Johnson (root@chaos.analogic.com)
Date: Wed Mar 29 2000 - 12:46:50 EST


On Wed, 29 Mar 2000, Daniel J Blueman wrote:

> Hi Richard,
>
> > I found that the large flushes, taking a long time, made my workstation
> > a bit temperamental so I run a little daemon that does what bdflush used
> > to do:
> >
> > for(;;)
> > {
> > sync();
> > sleep(30);
> > }
>
> Calling sync() is a Bad Thing as far as performance goes. It holds the
> global kernel lock for quite a long time. On a production server, that
> _would_ cause things to grind to a halt for the duration of the flush. On a
> workstation, it'll not do too much harm. The kernel thread kflushd does (or
> should do) what bdflush does anyway.

Yes. But what it does it gets done in a a few hundred milliseconds instead
of a 30 second flush.

Cheers,
Dick Johnson

Penguin : Linux version 2.3.41 on an i686 machine (800.63 BogoMips).

-
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.tux.org/lkml/



This archive was generated by hypermail 2b29 : Fri Mar 31 2000 - 21:00:25 EST