Re: Lost file-system story

["Valeriy E. Ushakov" <uwe%stderr.spb.ru@localhost>]

On Sun, Dec 11, 2011 at 23:23:33 -0500, Donald Allen wrote:

> On Sun, Dec 11, 2011 at 9:53 PM, Greg A. Woods <woods%planix.ca@localhost> 
> wrote:
>
> > Perhaps this sentence from McKusick's memo about fsck will help you to
> > understand:  "fsck is able to repair corrupted file systems using
> > procedures based upon the order in which UNIX honors these file system
> > update requests."  This is true for all Unix-based filesystems.
> 
> I'm not going to put words in McKusick's mouth, but I think you have
> misinterpreted this to mean that without ordering, recovery is
> impossible. If that's what you think (and you've said so, except when
> you've contradicted yourself), then you are wrong. Why? Because the
> evidence (e.g., my experiments) says  that recovery *is* possible. Not
> guaranteed. Possible.

What you are arguing is effectively isomorphic to:

1. I have C code that does i = i++ + i++;
2. When I use compiler C1 it always give me this specific result for i.
3. When I use compiler C2 it sometimes (or always) gives me some
   different result.
4. B/c of #2 C2 compiler must be wrong

-uwe