Changeset 406e5ffe

04/05/2018 10:31:16 PM (5 years ago)
Ken Moffat <ken@…>
10.0, 10.1, 11.0, 11.1, 11.2, 8.3, 8.4, 9.0, 9.1, basic, bdubbs/svn, elogind, lazarus, perl-modules, plabs/python-mods, qt5new, trunk, upgradedb, xry111/intltool, xry111/soup3, xry111/test-20220226

Firefox-59.0.2 and reword its note on how the SBUs vary, also extend the SBU comments in the Conventions section.

git-svn-id: svn:// af4574ff-66df-0310-9fd7-8a98e5e911e0

4 edited


  • introduction/welcome/changelog.xml

    rb81acc38 r406e5ffe  
    4646      <itemizedlist>
    4747        <listitem>
     48          <para>[ken] - Update to firefox-59.0.2 (security fix). Fixes
     49          <ulink url="&blfs-ticket-root;10557">#10557</ulink>.</para>
     50        </listitem>
     51        <listitem>
    4852          <para>[ken] - Work around missing quotes in c_rehash from
    4953          openssl-1.1.0h so that make-ca will work again. Fixes
  • introduction/welcome/conventions.xml

    rb81acc38 r406e5ffe  
    235235    depends on the individual package and specific hardware used.  </para>
     237    <para>For packages which use ninja (e.g. anything using meson) or rust, by
     238    default all cores are used so similar comments will be seen on such packages
     239    even when the build time is minimal.</para>
    237241    <para>Where even a parallel build takes more than 15 SBU, on certain
    238242    machines the time may be considerably greater even when the build does not
    239     use swap.</para>
     243    use swap. In particular, different micro-architectures will build some
     244    files at different relative speeds and this can introduce delays when
     245    certain make targets wait for another file to be created. Where a large
     246    build uses a lot of C++ files, processors with Simultaneous Multi Threading
     247    will share the Floating Point Unit and can take 45% longer than when using
     248    four 'prime' cores (measured on an intel i7 using taskset and keeping the
     249    other cores idle).</para>
    241251    <para>Some packages do not support parallel builds and using -j1 for the
  • packages.ent

    rb81acc38 r406e5ffe  
    733733<!ENTITY chromium-version             "64.0.3282.186">
    734734<!ENTITY epiphany-version             "">
    735 <!ENTITY firefox-version              "59.0.1">
     735<!ENTITY firefox-version              "59.0.2">
    736736<!ENTITY flashplayer-version          "">
    737737<!ENTITY qupzilla-version             "2.2.6">
  • xsoft/graphweb/firefox.xml

    rb81acc38 r406e5ffe  
    88  <!ENTITY firefox-download-http "&mozilla-http;/firefox/releases/&firefox-version;/source/firefox-&firefox-version;.source.tar.xz">
    10    <!ENTITY firefox-hgversion "3db9e3d52b17563efca181ccbb50deb8660c59ae">
     10   <!ENTITY firefox-hgversion "239e434d6d2b8e1e2b697c3416d1e96d48fe98e5">
    1111  <!-- use tinyurl to get a much shorter link for the wget example -->
    12   <!ENTITY firefox-tiny "">
     12  <!ENTITY firefox-tiny "">
    1313  <!ENTITY firefox-download-http "&mozilla-hg;&firefox-hgversion;.tar.bz2">
    1414  <!ENTITY firefox-download-ftp  " ">
    15   <!ENTITY firefox-md5sum        "5aff0df363d778ebe0e572d6c9f13dbe">
     15  <!ENTITY firefox-md5sum        "fa4f7a5a7e2b2c7900369530b396b74a">
    1616  <!ENTITY firefox-size          "291 MB">
    1717  <!-- NB with stylo, much of the build uses rust, and therefore cargo files.
    130130      <para>
    131         The build times for this version of firefox (using stylo, the new CSS
    132         rendering code) seem to vary widely between different machines. On one
    133         old Intel i3 with only 4GB of RAM the build took
    134         <emphasis>considerably</emphasis> longer. The reason for this is
    135         not understood. A build with stylo disabled is typically 1 or 2 SBU
    136         quicker and needs 0.8GB less disk space.
     131        As with other large packages which use C++ (or rust), the SBU times
     132        to build this vary more widely than you might expect. Also, almost 6GB
     133        of real memory is used during the final link and the SBUs can increase
     134        significantly if the machine has to swap to do this.
    137135      </para>
Note: See TracChangeset for help on using the changeset viewer.