Leopard improves SIMH performance - VMS

This is a discussion on Leopard improves SIMH performance - VMS ; Since a number of folks here use Mac OS X, I thought I'd mention that after upgrading to 10.5 (aka Leopard), my simulated VAX hosted on a dual 2.0 GHz Power Mac G5 running OpenVMS v7.3 on SIMH v3.7-3 sped ...

+ Reply to Thread
Results 1 to 3 of 3

Thread: Leopard improves SIMH performance

  1. Leopard improves SIMH performance

    Since a number of folks here use Mac OS X, I thought I'd mention that
    after upgrading to 10.5 (aka Leopard), my simulated VAX hosted on a
    dual 2.0 GHz Power Mac G5 running OpenVMS v7.3 on SIMH v3.7-3 sped up
    by over a third. A long-running job that took just over three days
    before the upgrade took just under two days after. I did not recompile
    SIMH, so whatever made the difference is available at run time. ISTR
    that SIMH does a lot with 64-bit integers, so the increased 64-bitness
    of Leopard may be a factor.

    --
    Posted via a free Usenet account from http://www.teranews.com


  2. Re: Leopard improves SIMH performance

    On Jan 4, 7:14 pm, "Craig A. Berry"
    wrote:
    > Since a number of folks here use Mac OS X, I thought I'd mention that
    > after upgrading to 10.5 (aka Leopard), my simulated VAX hosted on a
    > dual 2.0 GHz Power Mac G5 running OpenVMS v7.3 on SIMH v3.7-3 sped up
    > by over a third. A long-running job that took just over three days
    > before the upgrade took just under two days after. I did not recompile
    > SIMH, so whatever made the difference is available at run time. ISTR
    > that SIMH does a lot with 64-bit integers, so the increased 64-bitness
    > of Leopard may be a factor.
    >
    > --
    > Posted via a free Usenet account fromhttp://www.teranews.com


    Thanks for posting that. I've been thinking about running that on my
    wife's Mac at home (dual 2.5GHz G5) to take the VS3100m76 offline (it
    eats power and the Mac is always on anyway). Can you provide a rough
    equivalency to a real VAX on the performance you are getting?

    Thanks!

    Rich

  3. Re: Leopard improves SIMH performance

    In article
    <757e775c-5ace-4543-a9b1-b6ad4d6c398c@h11g2000prf.googlegroups.com>,
    Rich Jordan wrote:

    > On Jan 4, 7:14 pm, "Craig A. Berry"
    > wrote:
    > > Since a number of folks here use Mac OS X, I thought I'd mention that
    > > after upgrading to 10.5 (aka Leopard), my simulated VAX hosted on a
    > > dual 2.0 GHz Power Mac G5 running OpenVMS v7.3 on SIMH v3.7-3 sped up
    > > by over a third. A long-running job that took just over three days
    > > before the upgrade took just under two days after. I did not recompile
    > > SIMH, so whatever made the difference is available at run time. ISTR
    > > that SIMH does a lot with 64-bit integers, so the increased 64-bitness
    > > of Leopard may be a factor.


    > Thanks for posting that. I've been thinking about running that on my
    > wife's Mac at home (dual 2.5GHz G5) to take the VS3100m76 offline (it
    > eats power and the Mac is always on anyway). Can you provide a rough
    > equivalency to a real VAX on the performance you are getting?


    After the Leopard upgrade I'm getting just under 7 VUPs using a very
    simple DCL test procedure that's been posted more than once in this
    forum (and reposted below). I don't know what it was before the
    upgrade. It's worth noting that the benefits of the host system's disk
    and file caching make a pure CPU comparison less than the full story.
    It boots OpenVMS v7.3 in 80 seconds. I've got TCP/IP Services but no
    DECWindows installed.



    $ type calculate_vups.com
    $! CALCULATE_VUPS:
    $!
    $ set noon
    $ orig_privs = f$setprv("ALTPRI")
    $ process_priority = f$getjpi(0,"PRIB")
    $ cpu_multiplier = 10 ! VAX = 10 - Alpha/AXP =
    40
    $ cpu_round_add = 1 ! VAX = 1 - Alpha/AXP =
    9
    $ cpu_round_divide = cpu_round_add + 1
    $ init_counter = cpu_multiplier * 525
    $ init_loop_maximum = 205
    $ start_cputime = f$getjpi(0,"CPUTIM")
    $ loop_index = 0
    $ 10$:
    $ loop_index = loop_index + 1
    $ if loop_index .ne. init_loop_maximum then goto 10$
    $ end_cputime = f$getjpi(0,"CPUTIM")
    $ init_vups = ((init_counter / (end_cputime - start_cputime) + -
    cpu_round_add) / cpu_round_divide) * cpu_round_divide
    $ loop_maximum = (init_vups * init_loop_maximum) / 10
    $ base_counter = (init_counter * init_vups) / 10
    $ vups = 0
    $ times_through_loop = 0
    $ 20$:
    $ start_cputime = f$getjpi(0,"CPUTIM")
    $ loop_index = 0
    $ 30$:
    $ loop_index = loop_index + 1
    $ if loop_index .ne. loop_maximum then goto 30$
    $ end_cputime = f$getjpi(0,"CPUTIM")
    $ new_vups = ((base_counter / (end_cputime - start_cputime) + -
    cpu_round_add) / cpu_round_divide) * cpu_round_divide
    $ if new_vups .eq. vups then goto 40$
    $ vups = new_vups
    $ times_through_loop = times_through_loop + 1
    $ if times_through_loop .le. 5 then goto 20$
    $ 40$:
    $ new_privs = f$setprv(orig_privs)
    $ set message /nofacility/noidentification/noseverity/notext
    $ ASSIGN/SYSTEM/EXEC 'vups' MACHINE_VUPS_RATING
    $ set message /facility/identification/severity/text
    $ write sys$output "Approximate System VUPs Rating : ", -
    vups / 10,".", vups - ((vups / 10) * 10)
    $ exit

    --
    Posted via a free Usenet account from http://www.teranews.com


+ Reply to Thread