Re: FreeBSD Security Advisory FreeBSD-SA-05:09.htt [REVISED]

From: Garrett Wollman (wollman_at_csail.mit.edu)
Date: 05/14/05

  • Next message: Drew B. [Security Expertise/Freelance Security research].: "Re: different ways to disable https in apache..." 
    Date: Fri, 13 May 2005 22:33:30 -0400
To: "Poul-Henning Kamp" <phk@phk.freebsd.dk>

    <<On Sat, 14 May 2005 04:20:19 +0200, "Poul-Henning Kamp" <phk@phk.freebsd.dk> said:

    > The political problem is that if all operating systems do that,
    > Intel has a pretty dud feature on their hands, and they are not
    > particularly eager to accept that fact.

    Intel already had a pretty dud feature on their hands; just ask anyone
    in the architecture community (probably including those who work for
    Intel). Pentium 4 CPUs simply don't have enough I/O bandwidth to
    maintain two simultaneous, independent instruction streams. The value
    to the feature can't be realized until you have enough cache (in both
    size and bandwidth) to be able to partition it among logical CPUs in
    exactly the manner that Colin has suggested. (The fundamental problem
    in computer architecture for the past several years has been how to
    deal with the fact that gates are cheap and easy to make, but wires --
    particularly external I/O wires -- are expensive and hard.)

    The only way to get full performance out of an HTT processor today is
    for both threads to be running out of L1 cache. Multimedia and
    numerical benchmarks are often parallelizable in this way (assuming
    the OS provides gang scheduling); general-purpose applications rarely
    are.

    -GAWollman

    _______________________________________________
    freebsd-security@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-security
    To unsubscribe, send any mail to "freebsd-security-unsubscribe@freebsd.org"

  • Next message: Drew B. [Security Expertise/Freelance Security research].: "Re: different ways to disable https in apache..."

    Relevant Pages

    • Re: FreeBSD Security Advisory FreeBSD-SA-05:09.htt [REVISED]
      ... > The political problem is that if all operating systems do that, ... > Intel has a pretty dud feature on their hands, ... > particularly eager to accept that fact. ...
      (FreeBSD-Security)
    • Re: FreeBSD Security Advisory FreeBSD-SA-05:09.htt [REVISED]
      ... >> The political problem is that if all operating systems do that, ... >Intel already had a pretty dud feature on their hands; ...
      (FreeBSD-Security)
    • Re: Dell vs. eMachines T6420
      ... I still maintain that Intel "Core Duo" chips are multicore, ... Allowing communication across the L2 cache limits the use ... Intel itself has admitted several things about its "Core Duo" philosophy: ... cannot see or communicate with each other...Intel itself admits it does ...
      (alt.sys.pc-clone.dell)
    • Re: CPU questions...
      ... chosen to run with the ABIT IP35 Pro LGA 775 Intel P35 ATX Intel ... multiple physical processors, multiple cores on the same processor and ... each of which has four hyperthreads. ... versus dedicated cache. ...
      (comp.os.linux.hardware)
    • Re: 64 bit atomic reads on x86
      ... cache-line boundary and each CPU holds exclusively one of the two cache ... "Accesses to cacheable memory that are split ... Intel Xeon, P6 family, Pentium and Intel486 processors. ... Assuming that LOCK# is used, and though my knowledge of this area ...
      (comp.programming.threads)