[rescue] Re: SM[567]1/HyperSPARC comparison

[Greg A. Woods]

[ On Tuesday, February 5, 2002 at 12:41:41 (-0500), Loomis, Rip wrote: ]
> Subject: SM[567]1/HyperSPARC comparison (was RE: [rescue] SS10/20 death)
>
> As for me, I love having unbalanced systems--my Axil has
> one SM51 and one SM61 and it's always fun to see the look
> on PC-type techies' faces when I explain that.

Hmmm.... does that really work?  I've got what I seem to recall is an
SM51 here that I think still works, and an Axil-320 with an SM71 already
running in it.  I'd always been told that on the SS20's and their clones
you had to keep to the same speed CPUs.  If it does work then the next
question is going to be whether it'll work in NetBSD with the SMP
branch....

> [1] SM-61 should be SuperSPARC architecture, while SM71/81 are actually
>     SuperSPARC-II.  *Supposedly* SuperSPARC-II is significantly faster--
>     which is why an SM-71 might actually be worth half again as much
>     as an SM-61.  I've never done any benchmarking on the systems I
>     have, but if anyone cares let me know a good freely-available
>     benchmark suite to run.  I do know that in limited testing a single
>     SM71 was *way* faster than a single 90MHz HyperSPARC, before I
>     sold the HyperSPARC to someone concerned only about MHz.

I like the SM71, but it's not really what I'd call "fast" except in
comparison with the SM51 and of course anything lesser like my SS1's and
SS2's.  ;-)

NetBSD 1.5W (SOMETIMES) #23: Wed Aug 22 19:36:42 EDT 2001
    woods at sometimes:/build/NetBSD-obj/arch/sparc/compile/SOMETIMES
total memory = 287 MB
nbuf at 7350 is too large for VM_MAX_KERNEL_BUF... adjusted to 896
avail memory = 248 MB
using 896 buffers containing 29400 KB of memory
bootpath: /iommu at f,e0000000/sbus at f,e0001000/espdma at f,400000/esp at f,800000/sd at 0,0
mainbus0 (root): SUNW,Axil-320
mainbus0 (root): SUNW,Axil-320
cpu0 at mainbus0: TMS390Z50 v0 or TMS390Z55 @ 75 MHz, on-chip FPU
cpu0: physical 20K instruction (64 b/l), 16K data (32 b/l), 1024K external (32 b/l): cache enabled

The cool thing about this CPU, and this much RAM is that the total
throughput of the machine is quite awsome.  With the right disk
subsystem it could easily do mail/news/web/dns for several thousand
people and their web sites.  Indeed this very machine actually used to
do that, and with only the SM51 and less memory at the time too!

I had more RAM, 383MB total, but it was flaky and finaly gave up the
ghost:

/netbsd: memory error:
/netbsd:      EFSR: 231<CE,DW=3,SYNDROME=2>
/netbsd:      MBus transaction: 8fffcd30<VAH=0,TYPE=3,SIZE=5,C,VA=ff,S,MID=8>
/netbsd:      address: 0x0ef89600

-- 
								Greg A. Woods

+1 416 218-0098;  <gwoods at acm.org>;  <g.a.woods at ieee.org>;  <woods at robohack.ca>
Planix, Inc. <woods at planix.com>; VE3TCP; Secrets of the Weird <woods at weird.com>