[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: heads up

From: Ognyan Kulev
Subject: Re: heads up
Date: Mon, 21 Mar 2005 08:51:40 +0200
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.5) Gecko/20050105 Debian/1.7.5-1

Alfred M. Szmidt wrote:
   The reason that filesystems do not have user context is because I
   was not sufficiently far-sighted at the time to realize the full
   flexibility of the translator concept I had created.

No, it is because they can't have "user context", but feel free to
disprove me with some code.

In L4 port, user provides resources to server to do work. For example, program provides region of its own memory to ext2fs so that ext2fs can fulfill read request. (This can extend to providing memory pages to ext2fs that can't be used by user but hold ext2fs data structures that concern only the user. (Wow, I started lecturing about L4 ;-) )

But you probably know all of this. My point is that the same concept can be applied to "interactive context" passed to server.

Personally, I think that we need "context" abstraction. In particular, I'm interested in "transaction context" that may allow atomic change in multiple filesystems. I think the only context we have now is "auth context".

I was surprised to see that you talk about code -- we are talking about change in Hurd design now.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]