Re: I/O request ordering

Philippe Strauss (root@urbanet.ch)
Tue, 6 Aug 1996 14:42:31 +0200

Linus Torvalds writes:
>
> On Mon, 5 Aug 1996, Ray Van Tassle-CRV004 wrote:
> >
> > For some time, I have had doubts about the I/O request ordering algorithm in
> > drivers/block/ll_blk_rw. Upon examination, it turned out that (as I
> > suspected) it is a sawtooth rather than an elevator. And it also
> > has a bug.

...
[interesting ordering algorithm discussing deleted]
...

> (Kernel compile times are probably not good indications of disk ordering
> performance, at least not if you have enough memory. I see 90% CPU use for
> them most of the time).

I agree.

Doing a "time make zImage" some while ago on the same computer with
two *really* different hard drive (a SCSI seagate ST31200N on
ncr53c810 and a IDE ATA2 seagate <I DONT REMEMBER WHAT> on a 82371FX
(triton I)) shows me exactly the same elapsed time, on both in the
same second.

Dont need to say that nothing else than the kernel compile was running
in both case.

-- 
Philippe Strauss, CH-1092 Belmont
Email:	philippe.strauss@urbanet.ch


******* Caution! don't use reply without changing "root" to "philippe.strauss"
******* in my email addr. (sendmail masqu. on outbound not yet configured ;*)