On Tue, 4 Sep 2007 09:19:44 -0500, Big Iron wrote: >IIRC, wipe may not function as expected because, when you rewrite parts >of an HFS, I think that new pages will be assigned within the HFS and a >new copy of the data is written. The old pages are not over-written. > When the programmer does an update-in-place and new pages are assigned, are any cached copies of such pages marked invalid, or at least updated in place in cache while being assigned new disk addresses?
Does the FS nowadays resparsify when zeroes are logically written over previously nonzero pages? If so, only pointer blocks would be updated. It might be better to write nonzero data but with a repetitive pattern which is readily compressed. -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
