Re: Performance degradation over time on VAX...

=0A=
A little late, but I took the challenge and I did some tests that I hope fo=
llow what you have done.=0A=
=0A=
Here are the results, FWIW...=0A=
=0A=
My simh 3.12-5 boot config I used with the ISOs.  All testing done using th=
e same host system.=0A=
=0A=
-----------------------------------------------------=0A=
=0A=
set cpu 128m=0A=
=0A=
set cpu SIMHALT=0A=
set cpu idle=3DNETBSD=0A=
=0A=
attach nvr ./nvram.bin=0A=
=0A=
set rq0 RAUSER=3D1024=0A=
att rq0 RA1G.0.dsk=0A=
=0A=
set rq1 disabled=0A=
=0A=
set rq2 cdrom=0A=
att rq2 /home/sim312/vax/iso/LEGACY/{vaxcd-*.iso or NetBSD-*.iso}=0A=
=0A=
set rq3 disabled=0A=
=0A=
set xq disabled=0A=
=0A=
set dz lines=3D4=0A=
set dz 7b=0A=
attach -am dz 1144=0A=
=0A=
set  tq disabled=0A=
set lpt disabled=0A=
set  rl disabled=0A=
set  ts disabled=0A=
=0A=
boot cpu=0A=
=0A=
-----------------------------------------------------=0A=
Run on each release ISO selecting a shell after booting=0A=
=0A=
dd if=3D/dev/rra0c of=3D/dev/null bs=3D1024=0A=
=0A=
-----------------------------------------------------=0A=
=0A=
2.0.0.txt:1024458752 bytes transferred in 92.210 secs (11110061 bytes/sec)=
=0A=
2.0.2.txt:1024458752 bytes transferred in 92.630 secs (11059686 bytes/sec)=
=0A=
2.1.0.txt:1024458752 bytes transferred in 93.000 secs (11015685 bytes/sec)=
=0A=
3.1.1.txt:1024458752 bytes transferred in 97.490 secs (10508347 bytes/sec)=
=0A=
4.0.0.txt:1024458752 bytes transferred in 114.550 secs (8943332 bytes/sec)=
=0A=
4.0.1.txt:1024458752 bytes transferred in 114.230 secs (8968386 bytes/sec)=
=0A=
5.0.0.txt:1024458752 bytes transferred in 157.420 secs (6507805 bytes/sec)=
=0A=
5.0.2.txt:1024458752 bytes transferred in 160.880 secs (6367844 bytes/sec)=
=0A=
5.1.2.txt:1024458752 bytes transferred in 162.350 secs (6310186 bytes/sec)=
=0A=
5.1.5.txt:1024458752 bytes transferred in 161.820 secs (6330853 bytes/sec)=
=0A=
5.2.3.txt:1024458752 bytes transferred in 158.070 secs (6481044 bytes/sec)=
=0A=
6.0.6.txt:1024458752 bytes transferred in 179.890 secs (5694917 bytes/sec)=
=0A=
6.1.2.txt:1024458752 bytes transferred in 176.960 secs (5789210 bytes/sec)=
=0A=
6.1.3.txt:1024458752 bytes transferred in 178.940 secs (5725152 bytes/sec)=
=0A=
7.0.0.txt:1024458752 bytes transferred in 197.520 secs (5186607 bytes/sec)=
=0A=
7.1.2.txt:1024458752 bytes transferred in 197.030 secs (5199506 bytes/sec)=
=0A=
7.2.0.txt:1024458752 bytes transferred in 197.720 secs (5181361 bytes/sec)=
=0A=
8.0.0.txt:1024458752 bytes transferred in 204.430 secs (5011293 bytes/sec)=
=0A=
8.1.0.txt:1024458752 bytes transferred in 204.420 secs (5011538 bytes/sec)=
=0A=
8.2.0.txt:1024458752 bytes transferred in 211.010 secs (4855024 bytes/sec)=
=0A=
9.0.0.txt:1024458752 bytes transferred in 199.950 secs (5123574 bytes/sec)=
=0A=
9.1.0.txt:1024458752 bytes transferred in 201.240 secs (5090731 bytes/sec)=
=0A=
9.2.0.txt:1024458752 bytes transferred in 204.750 secs (5003461 bytes/sec)=
=0A=
9.3.0.txt:1024458752 bytes transferred in 201.380 secs (5087192 bytes/sec)=
=0A=
10.0..txt:1024458752 bytes transferred in 224.340 secs (4566545 bytes/sec)=
=0A=
10.1..txt:1024458752 bytes transferred in 223.530 secs (4583092 bytes/sec)=
=0A=
11rc5.txt:1024458752 bytes transferred in 226.920 secs (4514625 bytes/sec)=
=0A=
=0A=
-----------------------------------------------------=0A=
Run on already installed images that had rebuilt themselves from source=0A=
=0A=
dd if=3D/dev/rra0c of=3D/dev/null bs=3D1024 count=3D1000448=0A=
=0A=
-----------------------------------------------------=0A=
=0A=
1.2.1.txt:1024458752 bytes transferred in 64 secs (16007168 bytes/sec)=0A=
1.3.3.txt:1024458752 bytes transferred in 64 secs (16007168 bytes/sec)=0A=
1.4.3.txt:1024458752 bytes transferred in 63 secs (16261250 bytes/sec)=0A=
1.5.3.txt:1024458752 bytes transferred in 113 secs (9066006 bytes/sec)=0A=
1.6.2.txt:1024458752 bytes transferred in 125.800 secs (8143551 bytes/sec)=
=0A=
=0A=
=0A=
________________________________________=0A=
From: [email protected] <[email protected]> on behalf of Jo=
hnny Billquist <[email protected]>=0A=
Sent: Wednesday, April 29, 2026 10:42 AM=0A=
To: [email protected]=0A=
Subject: Performance degradation over time on VAX...=0A=
=0A=
Had a bit of discussions about NetBSD on VAXen (on discord of all=0A=
places), and the perceived performance degradation over time. After a=0A=
bunch of poking I finally caved in and did some simple tests that show=0A=
that something at least seems not entirely right.=0A=
=0A=
The setup and test done is as follows. Running on simh (a couple of=0A=
years old, but I can't see this really being relevant here).=0A=
=0A=
Fetched iso images of several versions of NetBSD. Just the installation=0A=
iso, booted it, mounted a file system where I had a 1G file. Ran=0A=
=0A=
dd if=3Dx.y of=3D/dev/null bs=3D1024=0A=
=0A=
and recorded the time for the execution.=0A=
=0A=
To be more clear - this is just booting the installation, and quitting=0A=
from the installation system. So it's really single user, with the=0A=
install system memory file system. And every time with the exact same=0A=
simh setup on the same machine with the same load.=0A=
=0A=
The emulated machine is a VAX 8650. 32MB of memory, with 4 MSCP disks.=0A=
Nothing else (well, there was ethernet configured in simh, but this is=0A=
never started or anything in NetBSD as I never went beyond just getting=0A=
a shell prompt).=0A=
One mount and one dd. The install iso is attached as rq0 (in simh) which=0A=
is ra0 in NetBSD. I had a second disk on ra1, which I just mounted under=0A=
/mnt, and that file system wasn't otherwise in any way interacted with,=0A=
apart from the reading of the file x.y=0A=
=0A=
The time for doing this dd under different versions of NetBSD then:=0A=
=0A=
2.1: 181s=0A=
4.0.1: 190s=0A=
5.0: 259s=0A=
6.0: 301s=0A=
7.0: 415s=0A=
10.99: 467s=0A=
=0A=
10.99 was not done with the installation CD, I should mention. That's my=0A=
"current" system, but I just booted it to single user, and did the same=0A=
mount and dd.=0A=
=0A=
Noticeable is that time only slightly increased from 2.1 to 4.0.1, but=0A=
after that it's constantly been getting worse. Current compared to 2.1=0A=
is between 2 and 3 times slower.=0A=
=0A=
And this for a simple dd on a system not doing anything else. There is=0A=
no good reason why disk I/O would have slowed down this much in newer=0A=
versions that I can think of.=0A=
=0A=
I think this shows a very limited picture of the performance, though. I=0A=
think that in a properly set up system, the impact is much worse, but in=0A=
this test, nothing else needed any CPU, so we can't tell if any aspects=0A=
beyond I/O was also hurting in this simple test.=0A=
=0A=
Anyway - this is perhaps food for thought for someone who is interested=0A=
in digging into it more. If I have plenty of time, I might dig more, but=0A=
I'm sortof busy with other things as well. :)=0A=
=0A=
   Johnny=0A=
=0A=