Automatic merge of trunk into multilib

This commit is contained in:
Thomas Trepl 2023-02-04 00:30:13 +01:00
commit 607ca5a486
9 changed files with 80 additions and 63 deletions

View File

@ -39,6 +39,32 @@
<listitem revision="sysv"> or <listitem revision="systemd"> as
appropriate for the entry or if needed the entire day's listitem.
-->
<listitem>
<para>2023-02-02</para>
<itemizedlist>
<listitem>
<para>[xry111] - Update to glibc-2.37. Fixes
<ulink url="&lfs-ticket-root;5203">#5203</ulink>.</para>
</listitem>
<listitem>
<para>[xry111] - Update to bc-6.2.3. Fixes
<ulink url="&lfs-ticket-root;5204">#5204</ulink>.</para>
</listitem>
<listitem>
<para>[xry111] - Update to linux-6.1.9. Fixes
<ulink url="&lfs-ticket-root;5205">#5205</ulink>.</para>
</listitem>
<listitem>
<para>[xry111] - Update to vim-9.0.1273. Addresses
<ulink url="&lfs-ticket-root;4500">#4500</ulink>.</para>
</listitem>
<listitem>
<para>[xry111] - Remove
<option>--disable-exec-static-tramp</option> for libffi.</para>
</listitem>
</itemizedlist>
</listitem>
<listitem>
<para>2023-02-01</para>
<itemizedlist>

View File

@ -67,11 +67,6 @@ ln -sfv ../lib/ld-linux-x86-64.so.2 $LFS/lib64/ld-lsb-x86-64.so.3</userinput></s
</para>
</note>
<para>Fix an issue building Glibc with parallel jobs and make-4.4
or later:</para>
<screen><userinput remap="pre">sed '/MAKEFLAGS :=/s/)r/) -r/' -i Makerules</userinput></screen>
<para>Some of the Glibc programs use the non-FHS-compliant
<filename class="directory">/var/db</filename> directory to store their
runtime data. Apply the following patch to make such programs store their

View File

@ -87,6 +87,19 @@ lassign [wait] _ _ _ value
exit $value
EOF</userinput></screen>
<para>The test suite uses <command>diff</command> to detect the
difference between test script output and the expected output. Any
output from <command>diff</command> (prefixed with
<computeroutput>&lt;</computeroutput> and
<computeroutput>&gt;</computeroutput>) indicates a test failure, unless
there is a message saying the difference can be ignored.
<!-- Some host distros set core file size hard limit < 1000, then the
test "ulimit -c -S 1000" attempts to set soft limit > hard limit
and fail. -->
One test named <filename>run-builtins</filename> is known to fail on
some host distros with a difference on the first line of the
output.</para>
<para>Install the package:</para>
<screen><userinput remap="install">make install</userinput></screen>

View File

@ -184,22 +184,7 @@ cd build</userinput></screen>
<para>Twelve tests fail in the gold test suite when the
<option>--enable-default-pie</option> and
<option>--enable-default-ssp</option> options are passed to GCC.
<!-- Caused by egrep deprecation. Note that we don't "patch" temp grep.
And it seems unworthy to add a sed into temp grep just for one test
failure. (I don't really agree to "patch" grep in the first place,
anyway.)
Will be fixed in 2.40 (upstream commit 67d1991). -->
The test named <filename>ar with versioned solib</filename> is also
known to fail.
<!-- https://sourceware.org/bugzilla/show_bug.cgi?id=29107 -->
Because of a race between the gas test suite and the gprofng test suite,
some gas tests may fail if parallel jobs are used. If you are not
assured of the correctness of gas, invoke
<command>make check-gas</command> to run the gas test suite separately and
all tests should pass. <!--Three gprofng tests yield the
<literal>ERROR</literal> result, and they are counted as
<literal>UNRESOLVED</literal> in the summary.--></para>
<option>--enable-default-ssp</option> options are passed to GCC.</para>
<para>Install the package:</para>
@ -208,7 +193,7 @@ cd build</userinput></screen>
<para>Remove useless static libraries and an empty man page:</para>
<screen><userinput remap="install">rm -fv /usr/lib/lib{bfd,ctf,ctf-nobfd,sframe,opcodes}.a
rm -fv /usr/share/man/man1/gprofng.1</userinput></screen>
rm -fv /usr/share/man/man1/{gprofng,gp-*}.1</userinput></screen>
</sect2>

View File

@ -43,11 +43,6 @@
<sect2 role="installation">
<title>Installation of Glibc</title>
<para>First, fix an issue building Glibc with parallel jobs and Make-4.4
or later:</para>
<screen><userinput remap="pre">sed '/MAKEFLAGS :=/s/)r/) -r/' -i Makerules</userinput></screen>
<para>Some of the Glibc programs use the non-FHS compliant
<filename class="directory">/var/db</filename> directory to store
their runtime data. Apply the following patch to make such programs
@ -55,6 +50,12 @@
<screen><userinput remap="pre">patch -Np1 -i ../&glibc-fhs-patch;</userinput></screen>
<!-- https://sourceware.org/bugzilla/show_bug.cgi?id=30068 -->
<para>Fix an issue identified upstream:</para>
<screen><userinput remap="pre">sed '/width -=/s/workend - string/number_length/' \
-i stdio-common/vfprintf-process-arg.c</userinput></screen>
<para>The Glibc documentation recommends building Glibc
in a dedicated build directory:</para>
@ -172,12 +173,6 @@ esac</userinput></screen>
is known to fail in the LFS chroot environment.</para>
</listitem>
<listitem>
<para>The <emphasis>nss/tst-nss-files-hosts-long</emphasis>
test is known to fail if the system has no non-loopback IP
addresses.</para> <!-- PR 24816 -->
</listitem>
<!-- https://sourceware.org/pipermail/libc-alpha/2022-August/141567.html -->
<listitem>
<para>The <emphasis>stdlib/tst-arc4random-thread</emphasis>

View File

@ -58,8 +58,9 @@
<screen><userinput remap="test">make check</userinput></screen>
<para>One test named <filename>run-low_high_pc.sh</filename> is known to
fail on 32-bit x86 systems.</para>
<!-- https://sourceware.org/bugzilla/show_bug.cgi?id=30047 -->
<para>The test named <filename>run-native-test.sh</filename> is known to
fail.</para>
<para>Install only Libelf:</para>

View File

@ -62,8 +62,7 @@
<screen><userinput remap="configure">./configure --prefix=/usr \
--disable-static \
--with-gcc-arch=native \
--disable-exec-static-tramp</userinput></screen>
--with-gcc-arch=native</userinput></screen>
<variablelist>
<title>The meaning of the configure option:</title>
@ -81,17 +80,6 @@
the x86 options in the GCC manual</ulink>.</para>
</listitem>
</varlistentry>
<varlistentry>
<!-- To editors: try to remove it once GJS fixed (at GNOME 43) -->
<term><parameter>--disable-exec-static-tramp</parameter></term>
<listitem>
<para>Disable static trampoline support, a new security
feature in Libffi. Some BLFS packages (notably
<application>GJS</application>) have not been adapted for it.</para>
</listitem>
</varlistentry>
</variablelist>
<para>Compile the package:</para>

View File

@ -41,6 +41,20 @@
<sect2 role="installation">
<title>Installation of MPFR</title>
<!-- https://sympa.inria.fr/sympa/arc/mpfr/2023-01/msg00002.html
https://gitlab.inria.fr/mpfr/mpfr/-/commit/5172494c
Note that it's a different issue from Glibc #30068 we've fixed in
chapter08/glibc.xml with a sed. Normally we just document the
test failure, but in this case the particular subtest also serves
as an additional guard against Glibc #30068. So we apply the
upstream fix here. -->
<para>Fix a test case based on a bug of old Glibc releases:</para>
<screen><userinput remap="pre">sed -e 's/+01,234,567/+1,234,567 /' \
-e 's/13.10Pd/13Pd/' \
-i tests/tsprintf.c</userinput></screen>
<para>Prepare MPFR for compilation:</para>
<screen><userinput remap="configure">./configure --prefix=/usr \
@ -58,7 +72,7 @@ make html</userinput></screen>
Do not skip it under any circumstances.</para>
</important>
<para>Test the results and ensure that all 183 tests passed:</para>
<para>Test the results and ensure that all 197 tests passed:</para>
<screen><userinput remap="test">make check</userinput></screen>

View File

@ -58,11 +58,11 @@
<!ENTITY bash-fin-du "50 MB">
<!ENTITY bash-fin-sbu "1.4 SBU">
<!ENTITY bc-version "6.2.2">
<!ENTITY bc-size "446 KB">
<!ENTITY bc-version "6.2.3">
<!ENTITY bc-size "448 KB">
<!ENTITY bc-url "https://github.com/gavinhoward/bc/releases/download/&bc-version;/bc-&bc-version;.tar.xz">
<!ENTITY bc-md5 "778d889cbaa7dbc07aa1661950ca88fb">
<!ENTITY bc-home "https://git.yzena.com/gavin/bc">
<!ENTITY bc-md5 "897adcfd53f06135b884d71874a63069">
<!ENTITY bc-home "https://git.gavinhoward.com/gavin/bc">
<!ENTITY bc-fin-du "7.4 MB">
<!ENTITY bc-fin-sbu "less than 0.1 SBU">
@ -255,10 +255,10 @@
<!ENTITY gettext-fin-du "235 MB">
<!ENTITY gettext-fin-sbu "2.7 SBU">
<!ENTITY glibc-version "2.36">
<!ENTITY glibc-size "18,175 KB">
<!ENTITY glibc-version "2.37">
<!ENTITY glibc-size "18,244 KB">
<!ENTITY glibc-url "&gnu;glibc/glibc-&glibc-version;.tar.xz">
<!ENTITY glibc-md5 "00e9b89e043340f688bc93ec03239b57">
<!ENTITY glibc-md5 "e89cf3dcb64939d29f04b4ceead5cc4e">
<!ENTITY glibc-home "&gnu-software;libc/">
<!ENTITY glibc-tmp-du "821 MB">
<!ENTITY glibc-tmp-sbu "4.4 SBU">
@ -433,12 +433,12 @@
<!ENTITY linux-major-version "6">
<!ENTITY linux-minor-version "1">
<!ENTITY linux-patch-version "8">
<!ENTITY linux-patch-version "9">
<!--<!ENTITY linux-version "&linux-major-version;.&linux-minor-version;">-->
<!ENTITY linux-version "&linux-major-version;.&linux-minor-version;.&linux-patch-version;">
<!ENTITY linux-size "131,631 KB">
<!ENTITY linux-size "131,632 KB">
<!ENTITY linux-url "&kernel;linux/kernel/v&linux-major-version;.x/linux-&linux-version;.tar.xz">
<!ENTITY linux-md5 "c201fad8846646a21161a1a3fb39fbf5">
<!ENTITY linux-md5 "ab1ac5556bd2c808e62bde0db04915f6">
<!ENTITY linux-home "https://www.kernel.org/">
<!-- measured for 5.13.4 / gcc-11.1.0 on x86_64 : minimum is
allnoconfig rounded down to allow for ongoing cleanups,
@ -726,13 +726,13 @@
<!ENTITY util-linux-fin-du "283 MB">
<!ENTITY util-linux-fin-sbu "1.0 SBU">
<!ENTITY vim-version "9.0.1060">
<!ENTITY vim-version "9.0.1273">
<!-- <!ENTITY vim-majmin "90"> -->
<!ENTITY vim-docdir "vim/vim90">
<!ENTITY vim-size "16,488 KB">
<!ENTITY vim-size "10,892 KB">
<!--<!ENTITY vim-url "https://github.com/vim/vim/archive/v&vim-version;/vim-&vim-version;.tar.gz">-->
<!ENTITY vim-url "&anduin-sources;/vim-&vim-version;.tar.gz">
<!ENTITY vim-md5 "cee12f38abdac37edeced59461bb04d4">
<!ENTITY vim-url "&anduin-sources;/vim-&vim-version;.tar.xz">
<!ENTITY vim-md5 "9c80755d2d95ec4ef713f66e57671797">
<!ENTITY vim-home "https://www.vim.org">
<!ENTITY vim-fin-du "217 MB">
<!ENTITY vim-fin-sbu "2.5 SBU">