[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Gnu-arch-users] Corrupt pristine tree
From: |
Miles Bader |
Subject: |
Re: [Gnu-arch-users] Corrupt pristine tree |
Date: |
Mon, 15 Mar 2004 17:57:25 -0500 |
User-agent: |
Mutt/1.3.28i |
On Tue, Mar 16, 2004 at 09:32:13AM +1100, address@hidden wrote:
> > What benefit is there in including the device number in the inode
> > signatures?
>
> Presumably because inode is not unique, only (device, inode) is unique.
This is true, but almost certainly irrelevant -- a revlib entry isn't going
to cross device boundaries, nor is a revision library expected to continue
working if copied to a different device.*
As far as I can figure, the device number is pretty pointless in inode sigs
(though it is harmful).
* Note that inode-numbers alone allow _detecting_ such copied revlibs, with
99.999% accuracy, and will properly trigger the existing corrupted revlib
case.
It might be useful to have some method for regenerating library inode-sigs
to allow doing this, for use in rare circumstances, e.g., if you just
upgraded your hard disk, and copied the old contents over.
-Miles
--
Suburbia: where they tear out the trees and then name streets after them.
- [Gnu-arch-users] Corrupt pristine tree, C. R. Oldham, 2004/03/13
- Re: [Gnu-arch-users] Corrupt pristine tree, Aaron Bentley, 2004/03/13
- Re: [Gnu-arch-users] Corrupt pristine tree, John Meacham, 2004/03/15
- Re: [Gnu-arch-users] Corrupt pristine tree, Charles Duffy, 2004/03/15
- Re: [Gnu-arch-users] Corrupt pristine tree, Miles Bader, 2004/03/15
- Re: [Gnu-arch-users] Corrupt pristine tree, John Meacham, 2004/03/15
[Gnu-arch-users] Re: Corrupt pristine tree, Greek0, 2004/03/13
Re: [Gnu-arch-users] Corrupt pristine tree, Anders Rune Jensen, 2004/03/13