Changeset eae92372


Ignore:
Timestamp:
08/26/2024 10:59:31 PM (3 weeks ago)
Author:
Bruce Dubbs <bdubbs@…>
Branches:
12.2, multilib, trunk, xry111/arm64, xry111/clfs-ng, xry111/loongarch, xry111/loongarch-12.2, xry111/mips64el, xry111/multilib
Children:
68c63ae
Parents:
04befc5e
Message:

Add more detail to the About SBUs page.

Files:
2 edited

Legend:

Unmodified
Added
Removed
  • chapter04/aboutsbus.xml

    r04befc5e reae92372  
    2727
    2828  <para>For example, consider a package whose compilation time is 4.5
    29   SBUs. This means that if your system took 10 minutes to compile and
     29  SBUs. This means that if your system took 4 minutes to compile and
    3030  install the first pass of binutils, it will take
    31   <emphasis>approximately</emphasis> 45 minutes to build the example package.
     31  <emphasis>approximately</emphasis> 18 minutes to build the example package.
    3232  Fortunately, most build times are shorter than one SBU.</para>
    3333
     
    3737  numbers can vary by as much as dozens of minutes in some cases.</para>
    3838
    39   <para>Before measuring the build time of any package (no matter Binutils
    40   pass 1 or a package of which the SBU is being measured), make sure a
    41   system power profile suitable to make the system running with the
    42   maximum performance (and the maximum power consumption) is selected.  Or
    43   the measured SBU value may be severly inaccurate because the system may be
    44   operated differently building Binutils pass 1 and the other package.
     39  <para>On some newer systems, the motherboard is capable of contolling
     40  the system clock speed. This can be controlled with a command such as
     41  <command>powerprofilesctl</command>. This is not available in LFS, but
     42  may be available on the host system. After LFS is complete, it can be
     43  added to a system with the procedures at the
     44  <ulink url='&blfs-book;sysutils/power-profiles-daemon.html'>
     45  BLFS power-profiles-daemon</ulink> page.
     46
     47  Before measuring the build time of any package it is advisable to use a
     48  system power profile set for maximum performance (and maximum power
     49  consumption).
     50
     51  Otherwise the measured SBU value may be inaccurate because the
     52  system may react differently when building <xref linkend='ch-tools-binutils-pass1'/>
     53  or other packages.
     54
    4555  Be aware that a significant inaccuracy can still show up even if the same
    46   profile (except one maximizing the performance) is used for both packages:
    47   the system may respond slower for <quote>saving the power</quote> building
    48   Binutils pass 1, because the system load seems only about 25% of the load
    49   building the other package (with <parameter>-j4</parameter>).</para>
     56  profile is used for both packages because the system may respond slower if
     57  the system is idle when starting the build procedure. Setting the power
     58  profile to "performance" will minimize this problem.</para>
    5059
    5160  <para>On most distros the power profile can be managed with either
     
    7180    single processor build to properly analyze the error messages.</para>
    7281
    73     <para>The times presented here are based upon using four cores (-j4). The
     82    <para>The times presented here for all packages
     83    (except <xref linkend='ch-tools-binutils-pass1'/> which is based on one core)
     84    are based upon using four cores (-j4). The
    7485    times in Chapter 8 also include the time to run the regression tests for
    7586    the package unless specified otherwise.</para>
  • chapter05/binutils-pass1.xml

    r04befc5e reae92372  
    66]>
    77
    8 <sect1 id="ch-tools-binutils-pass1" role="wrap">
     8<sect1 id="ch-tools-binutils-pass1" role="wrap" xreflabel="binutils-pass1">
    99  <?dbhtml filename="binutils-pass1.html"?>
    1010
Note: See TracChangeset for help on using the changeset viewer.