Re: Changes in disk layout in VAX?????
Hugo Villeneuve <[email protected]> Tue, 12 Aug 2003 01:03:24 -0400
| Newsgroups | gmane.os.openbsd.vax |
|---|---|
| Message-ID | <[email protected]> |
On Mon, Aug 11, 2003 at 11:36:05PM +0200, [email protected] wrote: > > On Mon, Aug 11, 2003 at 03:54:50PM -0400, [email protected] wrote: > > > I was installing a snapshot from several days ago onto an > > > RZ25 drive to use as a cloning drive onto other machines > > > and noticed that the disklabel would not let me have the > > > first 518 sectors of the disk, but seemed to reserve that > > > for some reason. Is this new, and what is this doing? > > > Thanks > > > > It isn't new. The first cylinder is reserved for the boot blocks. > > > > (On many platform, the first cylinder is reserved for boot blocks. > > That's how it is on my vax and hppa computers.) > > > > (It won't need all the cylinder but anyway, the FFS format doesn't > > like not starting or ending on cylinder boundary.) > > > An ffs filesystem actually leaves the first 8k on a partition free, > that also includes the first partition on disk. And the vax only need > a little less than 8k for its boot block, so reserving a cylinder for > it sounds like a waste of disk space. Oh. I never questionned why actually. It said it was for the boot blocks and I beleived it. I may do some test in 2 days. Currently my vax is busy doing make build of 3.3-stable for the realpath patch. Maybe the cylinder is reserved just to be safe. Is that 8k free space at the beginning is also true for the swap partition? Does disklabel force a FFS partition at the beginning of the disk? (Like on sparc "a" must start at sector 0). (I've been known to be lasy and put the swap at the start of the disk because I know the size I want and let "a" have the rest without me having to do any calculation. (But I don't think this was on VAX)) How is it on NetBSD by default? > > -- Ragge -- Hugo Villeneuve <[email protected]> http://EINTR.net/