Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 7 Jan 1999 02:31:04 +0000 (GMT)
From:      Terry Lambert <tlambert@primenet.com>
To:        Don.Lewis@tsc.tdk.com (Don Lewis)
Cc:        dyson@iquest.net, tlambert@primenet.com, dillon@apollo.backplane.com, pfgiffun@bachue.usc.unal.edu.co, freebsd-hackers@FreeBSD.ORG
Subject:   Re: questions/problems with vm_fault() in Stable
Message-ID:  <199901070231.TAA08668@usr09.primenet.com>
In-Reply-To: <199901062244.OAA01922@salsa.gv.tsc.tdk.com> from "Don Lewis" at Jan 6, 99 02:44:47 pm

next in thread | previous in thread | raw e-mail | index | archive | help
> } Subject: Re: questions/problems with vm_fault() in Stable
> 
> } Any of the problems with the existing VFS/VM scheme have been with the
> } intricacies of dealing with VFS special cases, and dealing with the
> } I/O abstraction of buffers as a cache.  Forget "files" and think
> } "blobs of memory."  Once the notion of file is forgotten, then shadowing,
> } invalidation and aliasing of memory become very obvious...

[ ... ]

> Maybe the thing to do is to turn all the filesystems into stacking
> layers, including ffs.

UFS *is* a stacking layer that imposed directory hierarchy on FFS.

The complication doesn't arise (i.e., FFS works despite generalized
stacking being shot in the head from day one of the 4.4 integration)
because *the same set of vnodes are used at the same level for both
layers*.

This is tantamount to there being implicit object coherency between
the layers.

We know that implicit object coherency between stacking layers would
work because we have this example where it *does* work.

Put another way, any method that results in a umapfs with *any*
implementations for anything other than credential-containing
VOP descriptors is wrong (the VOP_BYPASS stuff is a crock; it
results from some side-effects that are gratuitous.  See the
null_bypass() code for details on this; if SunOS can do it right,
*BSD can damn well do it right).


					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.

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?199901070231.TAA08668>