Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 23 Dec 1998 18:24:01 -0500 (EST)
From:      "John S. Dyson" <dyson@iquest.net>
To:        pfgiffun@bachue.usc.unal.edu.co (Pedro F. Giffuni)
Cc:        freebsd-hackers@FreeBSD.ORG
Subject:   Re: questions/problems with vm_fault() in Stable
Message-ID:  <199812232324.SAA26271@y.dyson.net>
In-Reply-To: <367F9D58.D755DAB0@bachue.usc.unal.edu.co> from "Pedro F. Giffuni" at "Dec 22, 98 08:23:36 am"

next in thread | previous in thread | raw e-mail | index | archive | help
Pedro F. Giffuni said:
> In this thread I've heard good comment on UVM and also that there is not
> anyone really in charge of the VM. 
> 
> Would this indicate that we want UVM ?? Chuck Cranor's papers say a port
> could be done but, as I understand, our VM is faster.
> 
In *no way* is UVM done.  It is still missing merged VM buffer cache and
is not productized (from a performance standpoint.)  As the optimizations
for performance are done, the complexities also arise.  From a tutorial
standpoint, UVM might be good because of it's cleanliness -- however the
performance is the critical thing that distinguishes FreeBSD from the
alternatives.

Almost any negative allegations as to portability regarding the FreeBSD
VM are incorrect and mostly spin.  There are features in the FreeBSD VM that
take advantage of CPU capabilities that are inherently non-portable.  However,
those features are optional, and not necessary for correct operation.  I
suspect that as the Alpha platform is optimized, the VM code will be tuned
to support that super-well also.

Geesh, there is already support in FreeBSD for non-copy read/write to/from
the buffer cache also.  It isn't complete, but is there.  It only takes
someone to finish the job (which I was in the middle of.)  On some machines,
that feature would definitely be a mis-feature, but on the X86, it would
be useful.  Also, the FreeBSD VFS/VM code already supports the ability to
have non-mapped buffers (and has for 2years.)  There is alot in there that
might make the complexity look excessive, but that is only because there
are features in there that are almost ready to go.

I suspect that if there is a concentrated effort on UVM for a couple of
years, by a couple of people, it will be as functional as FreeBSD VM.  At
that point, it might be worthwhile to adopt it.  For an opinion about using
a FreeBSD VM vs. a UVM development environment -- use both for
awhile, and then notice the very significant (noticible) differences
in performance.

-- 
John                  | Never try to teach a pig to sing,
dyson@iquest.net      | it makes one look stupid
jdyson@nc.com         | and it irritates the pig.

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199812232324.SAA26271>