Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 22 Mar 1995 02:40:17 -0800 (PST)
From:      "Rodney W. Grimes" <rgrimes@gndrsh.aac.dev.com>
To:        bde@zeta.org.au (Bruce Evans)
Cc:        hasty@star-gate.com, phk@ref.tfs.com, faq@FreeBSD.org, hackers@FreeBSD.org
Subject:   Re: Why IDE is bad
Message-ID:  <199503221040.CAA08681@gndrsh.aac.dev.com>
In-Reply-To: <199503221019.UAA09977@godzilla.zeta.org.au> from "Bruce Evans" at Mar 22, 95 08:19:58 pm

next in thread | previous in thread | raw e-mail | index | archive | help
> 
> >I don't think you will find many ide's faster than the WDC, I know
> >it can run faster on a better IDE-controller, but this is a fairly
> >standard IDE controller so it's a good indication.
> >And yes, but FreeBSD doesn't support DMA in IDE (yet ?)...
> 
> The WDC specs that I've seen show DMA speeds slightly slower than
> IDE speeds so DMA may not be worth using.
> 
> I forgot to think about the possibility of overlapping DMA cycles
> with cpu cycles in my previous mail about this.  I reported a 28%
> slowdown for counting to 100000000.  The cpu mostly executes 2
> instructions in a loop.  There should be few bus accesses to slow
> the cpu down.  Maybe busmastering is broken.

It may also be caused by the fact that your CPU has to run SNOOP
cycles during the bus master DMA.  I do not know if the 486 can
run a SNOOP cycle at the same time it does a regular cache access,
somehow I really doubt it.

I do know that the Pentium can run SNOOPS and access both the
instruction and the data caches all in the same cycle.  Thats
one heck of a lot of data moving all at once!!!

> Bruce


-- 
Rod Grimes                                      rgrimes@gndrsh.aac.dev.com
Accurate Automation Company                   Custom computers for FreeBSD



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199503221040.CAA08681>