Subject: Re: Questions about 1.3
To: None <current-users@NetBSD.ORG>
From: der Mouse <mouse@Rodents.Montreal.QC.CA>
List: current-users
Date: 10/05/1997 07:57:17
> I think it can safely be said that there are reasons why somebody
> might want more than eight partitions on a disk,
(Seven, really, because the raw partition bypasses the partition
table.)
> although it can also be argued that such a person really ought to buy
> two smaller spindles rather than one big spindle - it's definitely
> faster.
Taken to its logical conclusion, this argument says I should have a
separate spindle for each filesystem, with partitioning being
completely unnecessary. Even if I were willing to do this (the disks
would cost way more this way), on a system where I need more than 7
partitions per pack, this would run me out of SCSI targets. (Not all
machines are capable of having a wide controller - or even a second
narrow controller - added to them.)
> But none of these things sounds reasonable in the context of a code
> freeze that's three weeks away. This is a change that should be made
> and given a _lot_ of time to settle out, not just three weeks.
Heh. I agree that this should not be in 1.3, though I'm not convinced
it needs all that much settling out; the only downsides to it I've
found are (1) it halves the number of available packs and (2) handling
the interval during which the kernel disagrees with /dev is nontrivial.
Since at least one port _already_ has MAXPARTITIONS>8 (Amiga, IIRC),
anything MI should already have been shaken out. (So why do I think it
doesn't belong in 1.3? Largely because of the upgrade hell, item (2)
above. I believe this is fixable, but not before 1.3 - I dealt with
that interval by hand, something I do not want to ask any significant
fraction of our user base to do.)
der Mouse
mouse@rodents.montreal.qc.ca
7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B