Changeset 6bcd464


Ignore:
Timestamp:
01/23/2019 11:05:19 AM (3 years ago)
Author:
Ken Moffat <ken@…>
Branches:
10.0, 10.1, 11.0, 8.4, 9.0, 9.1, bdubbs/svn, ken/refactor-virt, lazarus, qt5new, trunk, xry111/git-date, xry111/git-date-for-trunk, xry111/git-date-test
Children:
d0331db
Parents:
31ee0ed
Message:

Rust: I was mistaken about the Ryzen problem with the testsuite, traps and a few segfaults are normal but nobody usually looks at the syslog.

git-svn-id: svn://svn.linuxfromscratch.org/BLFS/trunk/BOOK@21021 af4574ff-66df-0310-9fd7-8a98e5e911e0

Files:
2 edited

Legend:

Unmodified
Added
Removed
  • general/prog/rust.xml

    r31ee0ed r6bcd464  
    202202    </para>
    203203
    204     <caution>
     204    <note>
    205205      <para>
    206         On AMD Ryzen processors (family 17h), the non-optimized version of libstd
    207         which is compiled at the start of the tests contains one or more opcodes
    208         which are not implemented on this CPU family. These will be logged in the
     206        The testsuite will generate some messages in the
    209207        <phrase revision="sysv">system log</phrase>
    210208        <phrase revision="systemd">systemd journal</phrase>
    211         and will be followed a few minutes later by segmentation faults. Despite
    212         that, the tests continue to run, apparently normally. But the system may
    213         reboot before the tests have completed. The normal optimized libraries
    214         run without this problem.
     209        for traps on invalid opcodes, and for segmentation faults.
     210        In themselves these are nothing to worry about, although if the
     211        output from the testsuite reports tests which FAIL with such faults
     212        then there may be a problem.
    215213      </para>
    216       <para>
    217         A mitigation is to install gdb
    218         <!-- systemd apparently handles this with systemd-coredump -->
    219         <phrase revision="sysv">and to run the tests with 'ulimit -C disabled'</phrase>
    220         but this does not always prevent the system rebooting.
    221       </para>
    222     </caution>
     214    </note>
    223215
    224216    <para>
  • introduction/welcome/changelog.xml

    r31ee0ed r6bcd464  
    4545      <para>January 23rd, 2018</para>
    4646      <itemizedlist>
     47        <listitem>
     48          <para>[ken] - Change the 'Caution' on the rustc page (about running
     49          the testsuite on AMD Ryzens) to a 'Note', the reboot problem was
     50          almost certainly kernel triple-faults on an -rc kernel, although
     51          the testsuite seemed to reliably trigger it at that time. The
     52          supposed mitigation (now removed) was clearly irrelevant, and to
     53          top it all the invalid opcodes also appear on other machines, so
     54          they too are probably deliberate.</para>
     55        </listitem>
    4756        <listitem>
    4857          <para>[pierre] - Update dependencies for libtasn1 and GTK+3:
Note: See TracChangeset for help on using the changeset viewer.