Subject: Re: Compiler timings on varous MVII NetBSDs etc.
To: None <>
From: NetBSD Bob <>
List: port-vax
Date: 01/25/2001 13:42:39
> > Interesting.  My take on it, is that like the PDP11 and the 4M ram
> > limitations, we are reaching a 16mb ram limitation in the old VAXentoyz.
> > I am sensing that gcc may be running out of ram.  Especially in the
> > 25 hour kernel compile arena and the neverending perl compile thicket.
> > Isildur got a kernel compile in 30 minutes on a KA650 box.
> Only the MVII, MS2000 have the 16mb limit.  LAter machines are at least
> capable of 32mb or much more.

Well, most of the early 3100 line seem stuck at 16mb, too.  The M76
can handle 32M.  Someone correct me if I am misinterpreting the ram
capability on my MV3100/10, and MV3100/M38 boxes, cuz, IFF they will
hold more ram than 16mb, I gotta find some.  All mine have 16mb,
except for the M76 at 32mb and the MVII's at 13mb and 7mb.

> personally I find it ghastly to need more than 16mb to compile reasonably.

Consider my old AIX 1 box that had 4mb of ram, and it loaded AIX
and ran fine and compiled fine.  With 6mb of ram it served as our
departmental email box, for years, with 150 accounts.  Food for thought.

> Sound much more like disk flogging (not enough or large enough IO
> buffers?).

Well, what can one do to increase buffering all around in the right
places to speed this kind of thing up?  The machine in question is
set up with 4x1g scsi drives, partitioned at 32mb root, 32mb swap,
and the rest usr, for testing.  Do they need more swap than that
on a 13mb ram MVII critter?

Interestingly, the disks don't seem to be flogging all that much
differently from my M76 box.  NetBSD-1.2 seems to audibly flog
the disks a bit more than 1.3.2, 1.4.3, and 1.5, but not noticeably
different on kernel compiles.

Is the generic disk flogging you are speaking of a function of
available ram (13mb), I/O buffering, temp space, or what?