Re: Detection of HTT

From: Brooks Davis (brooks_at_one-eyed-alien.net)
Date: 11/16/04

  • Next message: Scott Long: "Re: Detection of HTT" 
    Date: Mon, 15 Nov 2004 20:11:03 -0800
To: "M. Warner Losh" <imp@bsdimp.com>

    On Mon, Nov 15, 2004 at 08:46:08PM -0700, M. Warner Losh wrote:
    > In message: <20041116022926.GB57056@squash.dsto.defence.gov.au>
    > "Wilkinson, Alex" <alex.wilkinson@dsto.defence.gov.au> writes:
    > : Yes you're right ! This sux ! What the hell is Intel claiming that I have a
    > : HTT capable CPU, but really I have only 1 core !
    >
    > I seem to recall at the time HTT was introduced that you could put an
    > HTT into a non-HTT ready mobo. Maybe that's what's happend?

    I think that's usually a BIOS issue, but not always. There are also
    some early P4s that set the HTT flag but only have one core. That's
    valid since all the flag means is that you have the extra instruction
    which says how many cores you have. I'm pretty sure that doens't apply
    to 2.8GHz CPUs though.

    -- Brooks 

    -- 
Any statement of the form "X is the one, true Y" is FALSE.
PGP fingerprint 655D 519C 26A7 82E7 2529  9BF0 5D8E 8BE9 F238 1AD4

  • Next message: Scott Long: "Re: Detection of HTT"

    Relevant Pages

    • Re: MySQL 5.0.22 , FreeBSD 6.1-STABLE: Benchmark
      ... This server is a Dual Xeon 2.8GHz, 4GB of RAM and 2x73GB SCSI disks that do 320MB/s ... Switching from HTT to no HTT was achieved by machdep.hyperthreading_allowed, and switching from/to libpthread/libthr was done via libmap.conf. ... I think that this, does show impressive scaling to actually see performance increase with HTT enabled, from what I have seen on benchmarks on many hardware sites testing on MS Windows is that on the average best you get is an extra 5% performance out of HTT per core. ... The other promising benchmark I saw on per CPU scaling was a few months ago with a posted super smack benchmark on a -current box that was getting a score of around 60,000 on a slightly better Quad core AMD64 machine which proves consistent scaling per core, which as far as my memory goes shows good scaling when entering the 4+ core arena on MySQL. ...
      (freebsd-performance)
    • Re: MySQL 5.0.22 , FreeBSD 6.1-STABLE: Benchmark
      ... This server is a Dual Xeon 2.8GHz, 4GB of RAM and 2x73GB SCSI disks that do 320MB/s ... Switching from HTT to no HTT was achieved by machdep.hyperthreading_allowed, and switching from/to libpthread/libthr was done via libmap.conf. ... I think that this, does show impressive scaling to actually see performance increase with HTT enabled, from what I have seen on benchmarks on many hardware sites testing on MS Windows is that on the average best you get is an extra 5% performance out of HTT per core. ... The other promising benchmark I saw on per CPU scaling was a few months ago with a posted super smack benchmark on a -current box that was getting a score of around 60,000 on a slightly better Quad core AMD64 machine which proves consistent scaling per core, which as far as my memory goes shows good scaling when entering the 4+ core arena on MySQL. ...
      (freebsd-performance)
    • Re: MySQL 5.0.22 , FreeBSD 6.1-STABLE: Benchmark
      ... This server is a Dual Xeon 2.8GHz, 4GB of RAM and 2x73GB SCSI disks that do 320MB/s ... Switching from HTT to no HTT was achieved by machdep.hyperthreading_allowed, and switching from/to libpthread/libthr was done via libmap.conf. ... I think that this, does show impressive scaling to actually see performance increase with HTT enabled, from what I have seen on benchmarks on many hardware sites testing on MS Windows is that on the average best you get is an extra 5% performance out of HTT per core. ...
      (freebsd-performance)
    • Re: machdep.hyperthreading_allowed does not affect SMT cores
      ... HTT is a half core. ... HTT is the idea to work on the last thing while new data will transfer from ram to cpu cache. ... they make a mistake to use it as a normale core. ... On a dualcore core 2 must use, bevor core 1 is used. ...
      (freebsd-current)
    • Re: Detection of HTT
      ... November 2004 05:11, Brooks Davis wrote: ... > some early P4s that set the HTT flag but only have one core. ... P4s without HTT are/were available in the 1.3-2.8 GHz range (HTT actually was ...
      (freebsd-current)