Subject: Re: Puzzling questions about FFS
To: None <tech-kern@NetBSD.ORG>
From: Raul Deluth Miller <rockwell@nova.umd.edu>
List: tech-kern
Date: 10/28/1994 15:13:59
Kim Andersen:
. >As someone pointed out, the only way to ensure file system
. >integrity is to write out first data blocks, then indirect blocks
. >and then the inode. FFS does the opposite which is at best useless,
. >at worst wrong.

. How can this model be better then the opposite .  

If you write data before writing pointers to the data, your pointers
will always point to valid data.  If you write pointers to where the
data will be before writing the data then you have dangling pointers
if the process in prematurely terminated.

-- 
Raul D. Miller           n =: p*q             NB. 9<##:##:n [.large prime p, q
<rockwell@nova.umd.edu>  y =: n&|&(*&x)^:e 1  NB. -.1 e.e e.&factors<:p,q [.e<n
                         NB.  public e, n, y
                         x -: n&|&(*&y)^:d 1  NB. 1=(d*e)+.p*&<:q