Changeset c093e6b
- Timestamp:
- 10/01/2022 10:30:12 PM (19 months ago)
- Branches:
- multilib
- Children:
- f3b2f16
- Parents:
- f29824e (diff), 4349661 (diff)
Note: this is a merge changeset, the changes displayed below correspond to the merge itself.
Use the(diff)
links above to see all the changes relative to each parent. - Files:
-
- 1 added
- 47 edited
Legend:
- Unmodified
- Added
- Removed
-
Makefile
rf29824e rc093e6b 27 27 PDF_OUTPUT ?= LFS-BOOK.pdf 28 28 NOCHUNKS_OUTPUT ?= LFS-BOOK.html 29 DUMPDIR ?= ~/ cross-lfs-commands29 DUMPDIR ?= ~/lfs-commands 30 30 else 31 31 BASEDIR ?= ~/lfs-systemd … … 151 151 --noent \ 152 152 --postvalid \ 153 153 -o $(RENDERTMP)/lfs-full.xml \ 154 154 $(RENDERTMP)/lfs-html2.xml 155 155 … … 162 162 $(Q)xsltproc --nonet \ 163 163 --stringparam profile.condition html \ 164 164 --output $(RENDERTMP)/lfs-html.xml \ 165 165 stylesheets/lfs-xsl/profile.xsl \ 166 166 $(RENDERTMP)/lfs-full.xml 167 167 168 168 wget-list: $(BASEDIR)/wget-list $(BASEDIR)/wget-list-$(REV) … … 173 173 $(Q)xsltproc --xinclude --nonet \ 174 174 --output $(BASEDIR)/wget-list \ 175 175 stylesheets/wget-list.xsl \ 176 176 chapter03/chapter03.xml 177 177 … … 213 213 $(Q)./git-version.sh $(REV) 214 214 215 #dump-commands: validate 216 # @echo "Dumping book commands..." 217 # $(Q)xsltproc --nonet \ 218 # --output $(RENDERTMP)/lfs-html.xml \ 219 # --stringparam profile.revision $(REV) \ 220 # stylesheets/lfs-xsl/profile.xsl \ 221 # $(RENDERTMP)/lfs-full.xml 222 223 # $(Q)rm -rf $(DUMPDIR) 224 225 # $(Q)xsltproc --output $(DUMPDIR)/ \ 226 # stylesheets/dump-commands.xsl \ 227 # $(RENDERTMP)/lfs-html.xml 228 # @echo "Dumping book commands complete in $(DUMPDIR)" 229 230 all: book nochunks pdf # dump-commands 215 dump-commands: validate 216 @echo "Dumping book commands..." 217 218 $(Q)rm -rf $(DUMPDIR) 219 220 $(Q)xsltproc --output $(DUMPDIR)/ \ 221 stylesheets/dump-commands.xsl \ 222 $(RENDERTMP)/lfs-full.xml 223 @echo "Dumping book commands complete in $(DUMPDIR)" 224 225 all: book nochunks pdf dump-commands 231 226 232 227 .PHONY : all book dump-commands nochunks pdf profile-html tmpdir validate md5sums wget-list version -
appendices/dependencies.xml
rf29824e rc093e6b 1187 1187 <seglistitem> 1188 1188 <seg> 1189 <ulink url="&blfs-book;general/pcre .html">PCRE</ulink>1189 <ulink url="&blfs-book;general/pcre2.html">PCRE2</ulink> 1190 1190 and 1191 1191 <ulink url="&blfs-book;general/libsigsegv.html">libsigsegv</ulink> … … 2357 2357 <segtitle>&runtime;</segtitle> 2358 2358 <seglistitem> 2359 <seg>Glibc and Patch</seg>2359 <seg>Glibc</seg> 2360 2360 </seglistitem> 2361 2361 </segmentedlist> -
bootscripts/ChangeLog
rf29824e rc093e6b 1 2022-09-30 Bruce Dubbs <bdubbs@linuxfromscratch.org> 2 * Mount /dev/shm as a tmpfs. 3 1 4 2022-07-23 Thomas Trepl <thomas@linuxfromscratch.org> 2 5 * Mark an raid array clean when root (/) has been remounted -
bootscripts/lfs/init.d/mountvirtfs
rf29824e rc093e6b 39 39 fi 40 40 41 mkdir -p /run/lock /run/shm42 chmod 1777 /run/ shm /run/lock41 mkdir -p /run/lock 42 chmod 1777 /run/lock 43 43 44 44 log_info_msg "Mounting virtual file systems: ${INFO}/run" … … 59 59 fi 60 60 61 ln -sfn /run/shm /dev/shm 61 mkdir -p /dev/shm 62 log_info_msg2 " ${INFO}/dev/shm" 63 mount -o nosuid,nodev /dev/shm || failed=1 62 64 63 65 (exit ${failed}) -
bootscripts/lfs/init.d/swap
rf29824e rc093e6b 21 21 # Default-Start: S 22 22 # Default-Stop: 0 6 23 # Short-Description: Mounts and unmounts swap partitions.24 # Description: Mounts and unmounts swap partitions defined in23 # Short-Description: Activates and deactivates swap partitions. 24 # Description: Activates and deactivates swap partitions defined in 25 25 # /etc/fstab. 26 26 # X-LFS-Provided-By: LFS -
chapter01/changelog.xml
rf29824e rc093e6b 40 40 appropriate for the entry or if needed the entire day's listitem. 41 41 --> 42 43 <listitem> 44 <para>2022-10-01</para> 45 <itemizedlist> 46 <listitem> 47 <para>[bdubbs] - Update to iana-etc-20220922. Addresses 48 <ulink url="&lfs-ticket-root;5006">#5006</ulink>.</para> 49 </listitem> 50 <listitem> 51 <para>[bdubbs] - Update to tzdata-2022d. Fixes 52 <ulink url="&lfs-ticket-root;5119">#5119</ulink>.</para> 53 </listitem> 54 <listitem> 55 <para>[bdubbs] - Update to readline-8.2. Fixes 56 <ulink url="&lfs-ticket-root;5121">#5121</ulink>.</para> 57 </listitem> 58 <listitem> 59 <para>[bdubbs] - Update to linux-5.19.12. Fixes 60 <ulink url="&lfs-ticket-root;5115">#5115</ulink>.</para> 61 </listitem> 62 <listitem> 63 <para>[bdubbs] - Update to libffi-3.4.3. Fixes 64 <ulink url="&lfs-ticket-root;5116">#5116</ulink>.</para> 65 </listitem> 66 <listitem> 67 <para>[bdubbs] - Update to libcap-2.66. Fixes 68 <ulink url="&lfs-ticket-root;512">#5120</ulink>.</para> 69 </listitem> 70 <listitem revision="systemd"> 71 <para>[bdubbs] - Update to dbus-1.14.2. Fixes 72 <ulink url="&lfs-ticket-root;5123">#5123</ulink>.</para> 73 </listitem> 74 <listitem> 75 <para>[bdubbs] - Update to bc-6.0.4. Fixes 76 <ulink url="&lfs-ticket-root;5114">#5114</ulink>.</para> 77 </listitem> 78 <listitem> 79 <para>[bdubbs] - Update to bash-5.2. Fixes 80 <ulink url="&lfs-ticket-root;5122">#5122</ulink>.</para> 81 </listitem> 82 </itemizedlist> 83 </listitem> 84 85 <listitem> 86 <para>2022-09-22</para> 87 <itemizedlist> 88 <listitem> 89 <para>[bdubbs] - Update to expat-2.4.9 (Security Update). Fixes 90 <ulink url="&lfs-ticket-root;5117">#5117</ulink>.</para> 91 </listitem> 92 </itemizedlist> 93 </listitem> 94 95 <listitem> 96 <para>2022-09-20</para> 97 <itemizedlist> 98 <listitem> 99 <para>[bdubbs] - Adapt instructions depending on 100 host setup of /dev/shm when creating virtual filesystems 101 for chroot.</para> 102 </listitem> 103 </itemizedlist> 104 </listitem> 42 105 43 106 <listitem> -
chapter01/how.xml
rf29824e rc093e6b 47 47 work your way through <xref linkend="chapter-cross-tools"/> and beyond.</para> 48 48 49 <para><xref linkend="chapter-cross-tools"/> ,explains the installation of49 <para><xref linkend="chapter-cross-tools"/> explains the installation of 50 50 the initial tool chain, (binutils, gcc, and glibc) using cross compilation 51 51 techniques to isolate the new tools from the host system.</para> … … 62 62 provided in <xref linkend="ch-tools-toolchaintechnotes"/>.</para> 63 63 64 <para>In <xref linkend="chapter-building-system"/> , The64 <para>In <xref linkend="chapter-building-system"/> the 65 65 full LFS system is built. Another advantage provided by the chroot 66 66 environment is that it allows you to continue using the host system -
chapter01/whatsnew.xml
rf29824e rc093e6b 12 12 <title>What's new since the last release</title> 13 13 14 <para>In 11.3 release, <parameter>--enable-default-pie</parameter> 15 and <parameter>--enable-default-ssp</parameter> are enabled for GCC. 16 They can mitigate some type of malicious attacks but they cannot provide 17 a full protection. In case if you are reading a programming textbook, 18 you may need to disable PIE and SSP with GCC options 19 <parameter>-fno-pie -no-pie -fno-stack-protection</parameter> 20 because some textbooks assume they were disabled by default.</para> 21 14 22 <para>Below is a list of package updates made since the previous 15 23 release of the book.</para> … … 39 47 <para>Automake-&automake-version;</para> 40 48 </listitem>--> 41 < !--<listitem>49 <listitem> 42 50 <para>Bash &bash-version;</para> 43 </listitem> -->51 </listitem> 44 52 <listitem> 45 53 <para>Bc &bc-version;</para> … … 63 71 <para>DejaGNU-&dejagnu-version;</para> 64 72 </listitem>--> 65 < !--<listitem revision="systemd">73 <listitem revision="systemd"> 66 74 <para>D-Bus-&dbus-version;</para> 67 </listitem> -->75 </listitem> 68 76 <!--<listitem> 69 77 <para>Diffutils-&diffutils-version;</para> … … 75 83 <para>Eudev-&eudev-version;</para> 76 84 </listitem>--> 77 < !--<listitem>85 <listitem> 78 86 <para>Expat-&expat-version;</para> 79 </listitem> -->87 </listitem> 80 88 <!--<listitem> 81 89 <para>Expect-&expect-version;</para> … … 123 131 <para>Gzip-&gzip-version;</para> 124 132 </listitem>--> 125 < !--<listitem>133 <listitem> 126 134 <para>IANA-Etc-&iana-etc-version;</para> 127 </listitem>-->135 </listitem> 128 136 <!--<listitem> 129 137 <para>Inetutils-&inetutils-version;</para> … … 150 158 <para>LFS-Bootscripts-&lfs-bootscripts-version;</para> 151 159 </listitem>--> 152 < !--<listitem>160 <listitem> 153 161 <para>Libcap-&libcap-version;</para> 154 </listitem> -->162 </listitem> 155 163 <!--<listitem> 156 164 <para>Libelf-&elfutils-version; (from elfutils)</para> 157 165 </listitem>--> 158 < !--<listitem>166 <listitem> 159 167 <para>Libffi-&libffi-version;</para> 160 </listitem> -->168 </listitem> 161 169 <!--<listitem> 162 170 <para>Libpipeline-&libpipeline-version;</para> … … 219 227 <para>Python-&python-version;</para> 220 228 </listitem> 221 < !--<listitem>229 <listitem> 222 230 <para>Readline-&readline-version;</para> 223 </listitem> -->231 </listitem> 224 232 <!--<listitem> 225 233 <para>Sed-&sed-version;</para> … … 246 254 <para>Texinfo-&texinfo-version;</para> 247 255 </listitem>--> 248 < !--<listitem>256 <listitem> 249 257 <para>Tzdata-&tzdata-version;</para> 250 </listitem> -->258 </listitem> 251 259 <!--<listitem> 252 260 <para>Util-Linux-&util-linux-version;</para> -
chapter02/aboutlfs.xml
rf29824e rc093e6b 15 15 throughout the LFS build process. It should be set to the name of the 16 16 directory where you will be building your LFS system - we will use 17 <filename class="directory">/mnt/lfs</filename> as an example, but the18 directory choice is up to you. If you are building LFS on a separate17 <filename class="directory">/mnt/lfs</filename> as an example, but you may 18 choose any directory name you want. If you are building LFS on a separate 19 19 partition, this directory will be the mount point for the partition. 20 20 Choose a directory location and set the variable with the … … 26 26 <command>mkdir -v $LFS/tools</command> can be typed literally. The shell 27 27 will automatically replace <quote>$LFS</quote> with 28 <quote>/mnt/lfs</quote> (or whatever the variable was set to) when it28 <quote>/mnt/lfs</quote> (or whatever value the variable was set to) when it 29 29 processes the command line.</para> 30 30 … … 50 50 enter the export command above. In addition, the shell specified in the 51 51 <filename>/etc/passwd</filename> file for all users that need the 52 <envar>LFS</envar> variable needs tobe bash to ensure that the52 <envar>LFS</envar> variable must be bash to ensure that the 53 53 <filename>/root/.bash_profile</filename> file is incorporated as a part of 54 54 the login process.</para> … … 60 60 the <filename>.bashrc</filename> file for the user and 61 61 <systemitem class="username">root</systemitem>. In addition, 62 some distributions have instructions to not run the<filename>.bashrc</filename>63 instructions in a non-interactive bash invocation. Be sure to addthe64 export command beforethe test for non-interactive use.</para>62 some distributions use an "if" test, and do not run the remaining <filename>.bashrc</filename> 63 instructions for a non-interactive bash invocation. Be sure to place the 64 export command ahead of the test for non-interactive use.</para> 65 65 66 66 </note> -
chapter02/creatingfilesystem.xml
rf29824e rc093e6b 11 11 <title>Creating a File System on the Partition</title> 12 12 13 <para>Now that a blank partition has been set up, the file system can be 14 created. LFS can use any file system recognized by the Linux kernel, but the 15 most common types are ext3 and ext4. The choice of file system can be 16 complex and depends on the characteristics of the files and the size of 13 <para>A partition is just a range of sectors on a disk drive, delimited by 14 boundaries set in a partition table. Before the operating system can use 15 a partition to store any files, the partition must be formatted to contain a file 16 system, typically consisting of a label, directory blocks, data blocks, and 17 an indexing scheme to locate a particular file on demand. The file system 18 also helps the OS keep track of free space on the partition, reserve the 19 needed sectors when a new file is created or an existing file is extended, 20 and recycle the free data segments created when files are deleted. It may 21 also provide support for data redundancy, and for error recovery.</para> 22 23 <para>LFS can use any file system recognized by the Linux kernel, but the 24 most common types are ext3 and ext4. The choice of the right file system can be 25 complex; it depends on the characteristics of the files and the size of 17 26 the partition. For example:</para> 18 27 … … 34 43 <varlistentry> 35 44 <term>ext4</term> 36 <listitem><para>is the latest version of the ext f ile system family of37 partition types. It provides several new capabilities including38 nano-second timestamps, creation and use of very large files (16 TB), and39 speed improvements.</para>45 <listitem><para>is the latest version of the ext family of 46 file systems. It provides several new capabilities including 47 nano-second timestamps, creation and use of very large files 48 (up to 16 TB), and speed improvements.</para> 40 49 </listitem> 41 50 </varlistentry> … … 43 52 44 53 <para>Other file systems, including FAT32, NTFS, ReiserFS, JFS, and XFS are 45 useful for specialized purposes. More information about these file systems46 can be found at <ulink54 useful for specialized purposes. More information about these file systems, 55 and many others, can be found at <ulink 47 56 url="https://en.wikipedia.org/wiki/Comparison_of_file_systems"/>.</para> 48 57 49 <para>LFS assumes that the root file system (/) is of type ext4. 58 <para>LFS assumes that the root file system (/) is of type ext4. To create 50 59 an <systemitem class="filesystem">ext4</systemitem> file system on the LFS 51 partition, run the following:</para>60 partition, issue the following command:</para> 52 61 53 62 <screen role="nodump"><userinput>mkfs -v -t ext4 /dev/<replaceable><xxx></replaceable></userinput></screen> -
chapter02/creatingpartition.xml
rf29824e rc093e6b 95 95 <para>Swapping is never good. For mechanical hard drives you can generally 96 96 tell if a system is swapping by just listening to disk activity and 97 observing how the system reacts to commands. For an SSD driveyou will not98 be able to hear swapping but you can tell how much swap space is being used99 by the <command>top</command> or <command>free</command> programs. Use of100 an SSD drivefor a swap partition should be avoided if possible. The first97 observing how the system reacts to commands. With an SSD you will not 98 be able to hear swapping, but you can tell how much swap space is being used 99 by running the <command>top</command> or <command>free</command> programs. Use of 100 an SSD for a swap partition should be avoided if possible. The first 101 101 reaction to swapping should be to check for an unreasonable command such as 102 102 trying to edit a five gigabyte file. If swapping becomes a normal … … 113 113 loader. This partition will normally be labeled 'BIOS Boot' if using 114 114 <command>fdisk</command> or have a code of <emphasis>EF02</emphasis> if 115 using <command>gdisk</command>.</para>115 using the <command>gdisk</command> command.</para> 116 116 117 117 <note><para>The Grub Bios partition must be on the drive that the BIOS 118 uses to boot the system. This is not necessarily the same drive where the119 LFS root partition is located. Disks on a system may use different120 partition table types. The requirement for this partition depends118 uses to boot the system. This is not necessarily the drive that holds 119 the LFS root partition. The disks on a system may use different 120 partition table types. The necessity of the Grub Bios partition depends 121 121 only on the partition table type of the boot disk.</para></note> 122 122 </sect3> … … 134 134 store kernels and other booting information. To minimize potential boot 135 135 problems with larger disks, make this the first physical partition on 136 your first disk drive. A partition size of 200 megabytes is quite136 your first disk drive. A partition size of 200 megabytes is 137 137 adequate.</para></listitem> 138 138 … … 151 151 <filename class="directory">/lib</filename>, and 152 152 <filename class="directory">/sbin</filename> are symlinks to their 153 counterpart in <filename class="directory">/usr</filename>.154 So <filename class="directory">/usr</filename> contains all binaries153 counterparts in <filename class="directory">/usr</filename>. 154 So <filename class="directory">/usr</filename> contains all the binaries 155 155 needed for the system to run. For LFS a separate partition for 156 156 <filename class="directory">/usr</filename> is normally not needed. 157 If you needit anyway, you should make a partition large enough to158 fit all programs and libraries in the system.The root partition can be157 If you create it anyway, you should make a partition large enough to 158 fit all the programs and libraries in the system. The root partition can be 159 159 very small (maybe just one gigabyte) in this configuration, so it's 160 160 suitable for a thin client or diskless workstation (where 161 161 <filename class="directory">/usr</filename> is mounted from a remote 162 server). However you should take care that an initramfs (not covered by163 LFS) will be needed to boot a system with separate162 server). However, you should be aware that an initramfs (not covered by 163 LFS) will be needed to boot a system with a separate 164 164 <filename class="directory">/usr</filename> partition.</para></listitem> 165 165 166 166 <listitem><para>/opt – This directory is most useful for 167 BLFS where multiple installations of large packages like Gnome or KDEcan167 BLFS, where multiple large packages like KDE or Texlive can 168 168 be installed without embedding the files in the /usr hierarchy. If 169 169 used, 5 to 10 gigabytes is generally adequate.</para> 170 170 </listitem> 171 171 172 <listitem><para>/tmp – A separate /tmp directory is rare, but 173 useful if configuring a thin client. This partition, if used, will 174 usually not need to exceed a couple of gigabytes.</para></listitem> 172 <listitem revision='sysv'><para>/tmp – A separate /tmp directory 173 is rare, but useful if configuring a thin client. This partition, if 174 used, will usually not need to exceed a couple of 175 gigabytes. If you have enough RAM, you can mount a 176 <systemitem class='filesystem'>tmpfs</systemitem> on /tmp to make 177 access to temporary files faster.</para></listitem> 178 179 <listitem revision='systemd'><para>/tmp – By default, systemd 180 mounts a <systemitem class='filesystem'>tmpfs</systemitem> here. 181 If you want to override that behavior, follow 182 <xref linkend='systemd-no-tmpfs'/> when configuring the LFS 183 system.</para></listitem> 175 184 176 185 <listitem><para>/usr/src – This partition is very 177 186 useful for providing a location to store BLFS source files and 178 share them across LFS builds. 179 for building BLFS packages. 180 gigabytes allows plenty of room.</para></listitem>187 share them across LFS builds. It can also be used as a location 188 for building BLFS packages. A reasonably large partition of 30-50 189 gigabytes provides plenty of room.</para></listitem> 181 190 182 191 </itemizedlist> 183 192 184 <para>Any separate partition that you want automatically mounted upon boot185 needs to be specified in the <filename>/etc/fstab</filename>. Details186 about how to specify partitions will be discussed in <xref187 linkend="ch-bootable-fstab"/>. 193 <para>Any separate partition that you want automatically mounted when the 194 system starts must be specified in the <filename>/etc/fstab</filename> file. 195 Details about how to specify partitions will be discussed in <xref 196 linkend="ch-bootable-fstab"/>.</para> 188 197 189 198 </sect3> -
chapter02/hostreqs.xml
rf29824e rc093e6b 37 37 <listitem> 38 38 <para><emphasis role="strong">Bison-2.7</emphasis> (/usr/bin/yacc 39 should be a link to bison or small script that executes bison)</para>39 should be a link to bison or a small script that executes bison)</para> 40 40 </listitem> 41 41 -
chapter02/mounting.xml
rf29824e rc093e6b 11 11 <title>Mounting the New Partition</title> 12 12 13 <para>Now that a file system has been created, the partition needs to 14 be made accessible. In order to do this, the partition needs to be 15 mounted at a chosen mount point. For the purposes of this book, it is 16 assumed that the file system is mounted under the directory specified by the 17 <envar>LFS</envar> environment variable as described in the previous section. 13 <para>Now that a file system has been created, the partition must 14 be mounted so the host system can access it. This book assumes that 15 the file system is mounted at the directory specified by the 16 <envar>LFS</envar> environment variable described in the previous section. 18 17 </para> 18 19 <para>Strictly speaking, one cannot "mount a partition". One mounts the <emphasis>file 20 system</emphasis> embedded in that partition. But since a single partition can't contain 21 more than one file system, people often speak of the partition and the 22 associated file system as if they were one and the same.</para> 19 23 20 <para>Create the mount point and mount the LFS file system by running:</para>24 <para>Create the mount point and mount the LFS file system with these commands:</para> 21 25 22 26 <screen role="nodump"><userinput>mkdir -pv $LFS 23 27 mount -v -t ext4 /dev/<replaceable><xxx></replaceable> $LFS</userinput></screen> 24 28 25 <para>Replace <replaceable><xxx></replaceable> with the designationof the LFS29 <para>Replace <replaceable><xxx></replaceable> with the name of the LFS 26 30 partition.</para> 27 31 28 <para>If using multiple partitions for LFS (e.g., one for <filename29 class="directory">/</filename> and another for <filename30 class="directory">/home</filename>), mount them using:</para>32 <para>If you are using multiple partitions for LFS (e.g., one for 33 <filename class="directory">/</filename> and another for <filename 34 class="directory">/home</filename>), mount them like this:</para> 31 35 32 36 <screen role="nodump"><userinput>mkdir -pv $LFS … … 44 48 without any parameters to see what options are set for the mounted LFS 45 49 partition. If <option>nosuid</option> and/or <option>nodev</option> are set, 46 the partition will need tobe remounted.</para>50 the partition must be remounted.</para> 47 51 48 <warning><para>The above instructions assume that you will not be restarting52 <warning><para>The above instructions assume that you will not restart 49 53 your computer throughout the LFS process. If you shut down your system, 50 54 you will either need to remount the LFS partition each time you restart 51 the build process or modify your host system's /etc/fstab file to automatically 52 remount it upon boot. For example: 55 the build process, or modify the host system's &fstab; file to automatically 56 remount it when you reboot. For example, you might add this line to your 57 &fstab; file: 53 58 54 59 <screen role="nodump">/dev/<replaceable><xxx></replaceable> /mnt/lfs ext4 defaults 1 1</screen> … … 68 73 <systemitem class="filesystem">swap</systemitem> partition.</para> 69 74 70 <para>Now that the re is an established place to work, it is time to75 <para>Now that the new LFS partition is open for business, it's time to 71 76 download the packages.</para> 72 77 -
chapter02/stages.xml
rf29824e rc093e6b 35 35 36 36 <sect2> 37 <title>Chapter 5–6</title>37 <title>Chapters 5–6</title> 38 38 39 39 <itemizedlist> … … 45 45 <para>These two chapters <emphasis>must</emphasis> be done as user 46 46 <systemitem class="username">lfs</systemitem>. 47 A <command>su - lfs</command> needs to be donebefore any task in these48 chapters. Failing todo that, you are at risk of installing packages to the47 A <command>su - lfs</command> command must be issued before any task in these 48 chapters. If you don't do that, you are at risk of installing packages to the 49 49 host, and potentially rendering it unusable.</para> 50 50 </listitem> … … 55 55 doubt about installing a package, ensure any previously expanded 56 56 tarballs are removed, then re-extract the package files, and complete all 57 instructions in that section.</para>57 the instructions in that section.</para> 58 58 </listitem> 59 59 </itemizedlist> … … 61 61 62 62 <sect2> 63 <title>Chapter 7–10</title>63 <title>Chapters 7–10</title> 64 64 65 65 <itemizedlist> … … 70 70 <listitem> 71 71 <para>A few operations, from <quote>Changing Ownership</quote> to 72 <quote>Entering the Chroot Environment</quote> must be done as the72 <quote>Entering the Chroot Environment</quote>, must be done as the 73 73 <systemitem class="username">root</systemitem> user, with the LFS 74 74 environment variable set for the &root; user.</para> … … 78 78 <para> When entering chroot, the LFS environment variable must be set 79 79 for <systemitem class="username">root</systemitem>. The LFS 80 variable is not used after wards.</para>80 variable is not used after entering the chroot environment.</para> 81 81 </listitem> 82 82 -
chapter03/introduction.xml
rf29824e rc093e6b 14 14 order to build a basic Linux system. The listed version numbers correspond to 15 15 versions of the software that are known to work, and this book is based on 16 their use. We highly recommend against using different versions because the build16 their use. We highly recommend against using different versions, because the build 17 17 commands for one version may not work with a different version, unless the 18 different version is specified by a LFS errataor security advisory.18 different version is specified by an LFS erratum or security advisory. 19 19 The newest package versions may also have problems that require 20 20 work-arounds. These work-arounds will be developed and stabilized in the … … 22 22 23 23 <para>For some packages, the release tarball and the (Git or SVN) 24 repository snapshot tarball for th isrelease may be published with25 similar file name . A release tarball contains generated files (for26 example, <command>configure</command> script generated by24 repository snapshot tarball for that release may be published with 25 similar file names. A release tarball contains generated files (for 26 example, a <command>configure</command> script generated by 27 27 <command>autoconf</command>), in addition to the contents of the 28 28 corresponding repository snapshot. The book uses release tarballs … … 70 70 <listitem> 71 71 <para>For stable versions of the book, a tarball of all the needed files 72 can be downloaded from one of the LFS files mirrors listed at72 can be downloaded from one of the mirror sites listed at 73 73 <ulink url="https://www.linuxfromscratch.org/mirrors.html#files"/>.</para> 74 74 </listitem> -
chapter03/packages.xml
rf29824e rc093e6b 16 16 package should be used to avoid security vulnerabilities.</para> 17 17 18 <para>The upstream s may remove old releases, especially when these18 <para>The upstream sources may remove old releases, especially when those 19 19 releases contain a security vulnerability. If one URL below is not 20 20 reachable, you should read the security advisories first to figure out … … 22 22 not, try to download the removed package from a mirror. Although it's 23 23 possible to download an old release from a mirror even if this release 24 has been removed because of a vulnerability, it's not recommendedto25 use a release known to be vulnerable forbuilding your system.</para>24 has been removed because of a vulnerability, it's not a good idea to 25 use a release known to be vulnerable when building your system.</para> 26 26 </note> 27 27 … … 463 463 464 464 <note> 465 <para>The Linux kernel is updated relatively often, many times due to465 <para>The Linux kernel is updated quite frequently, many times due to 466 466 discoveries of security vulnerabilities. The latest available 467 467 <!--&linux-major-version;.&linux-minor-version;.x--> stable kernel -
chapter04/aboutsbus.xml
rf29824e rc093e6b 14 14 it takes to compile and install each package. Because Linux From 15 15 Scratch can be built on many different systems, it is impossible to 16 provide a ccurate time estimates. The biggest package (Glibc) will16 provide absolute time estimates. The biggest package (Glibc) will 17 17 take approximately 20 minutes on the fastest systems, but could take 18 18 up to three days on slower systems! Instead of providing actual times, … … 21 21 22 22 <para>The SBU measure works as follows. The first package to be compiled 23 from this bookis binutils in <xref linkend="chapter-cross-tools"/>. The24 time it takes to compile this package is what w ill be referredto as the25 Standard Build Unit or SBU. All other compile times will be expressed relative26 t o thistime.</para>23 is binutils in <xref linkend="chapter-cross-tools"/>. The 24 time it takes to compile this package is what we will refer to as the 25 Standard Build Unit or SBU. All other compile times will be expressed in 26 terms of this unit of time.</para> 27 27 28 28 <para>For example, consider a package whose compilation time is 4.5 29 SBUs. This means that if asystem took 10 minutes to compile and29 SBUs. This means that if your system took 10 minutes to compile and 30 30 install the first pass of binutils, it will take 31 <emphasis>approximately</emphasis> 45 minutes to build th isexample package.32 Fortunately, most build times are shorter than the one for binutils.</para>31 <emphasis>approximately</emphasis> 45 minutes to build the example package. 32 Fortunately, most build times are shorter than one SBU.</para> 33 33 34 <para> In general,SBUs are not entirely accurate because they depend on many34 <para>SBUs are not entirely accurate because they depend on many 35 35 factors, including the host system's version of GCC. They are provided here 36 36 to give an estimate of how long it might take to install a package, but the … … 46 46 <screen role="nodump"><userinput>export MAKEFLAGS='-j4'</userinput></screen> 47 47 48 <para>or justbuilding with:</para>48 <para>or by building with:</para> 49 49 50 50 <screen role="nodump"><userinput>make -j4</userinput></screen> … … 53 53 book will vary even more than they normally would. In some cases, the make 54 54 step will simply fail. Analyzing the output of the build process will also 55 be more difficult because the lines ofdifferent processes will be56 interleaved. If you run into a problem with a build step, revert backto a55 be more difficult because the lines from different processes will be 56 interleaved. If you run into a problem with a build step, revert to a 57 57 single processor build to properly analyze the error messages.</para> 58 58 </note> -
chapter04/abouttestsuites.xml
rf29824e rc093e6b 28 28 <para>Running the test suites in <xref linkend="chapter-cross-tools"/> 29 29 and <xref linkend="chapter-temporary-tools"/> 30 is impossible, since theprograms are compiled with a cross-compiler,31 so are not supposed to be able torun on the build host.</para>30 is pointless; since the test programs are compiled with a cross-compiler, 31 they probably can't run on the build host.</para> 32 32 </note> 33 33 34 34 <para>A common issue with running the test suites for binutils and GCC 35 is running out of pseudo terminals (PTYs). This can result in a high35 is running out of pseudo terminals (PTYs). This can result in a large 36 36 number of failing tests. This may happen for several reasons, but the 37 37 most likely cause is that the host system does not have the … … 40 40 <ulink url="&lfs-root;lfs/faq.html#no-ptys"/>.</para> 41 41 42 <para>Sometimes package test suites will fail , butfor reasons which the42 <para>Sometimes package test suites will fail for reasons which the 43 43 developers are aware of and have deemed non-critical. Consult the logs located 44 44 at <ulink url="&test-results;"/> to verify whether or not these failures are 45 expected. This site is valid for all test s throughout this book.</para>45 expected. This site is valid for all test suites throughout this book.</para> 46 46 47 47 </sect1> -
chapter04/addinguser.xml
rf29824e rc093e6b 15 15 the packages in the next two chapters are built as an unprivileged user. 16 16 You could use your own user name, but to make it easier to set up a clean 17 working environment, create a new user called <systemitem17 working environment, we will create a new user called <systemitem 18 18 class="username">lfs</systemitem> as a member of a new group (also named 19 <systemitem class="groupname">lfs</systemitem>) and use this userduring19 <systemitem class="groupname">lfs</systemitem>) and run commands as &lfs-user; during 20 20 the installation process. As <systemitem class="username">root</systemitem>, 21 21 issue the following commands to add the new user:</para> … … 25 25 26 26 <variablelist> 27 <title>Th e meaning of the command line options:</title>27 <title>This is what the command line options mean:</title> 28 28 29 29 <varlistentry> … … 55 55 <listitem> 56 56 <para>This parameter prevents possible copying of files from a skeleton 57 directory ( default is <filename class="directory">/etc/skel</filename>)57 directory (the default is <filename class="directory">/etc/skel</filename>) 58 58 by changing the input location to the special null device.</para> 59 59 </listitem> … … 69 69 </variablelist> 70 70 71 <para> To log in as <systemitem class="username">lfs</systemitem> (as opposed72 to switching to user <systemitem class="username">lfs</systemitem> when logged73 in as <systemitem class="username">root</systemitem>, which does not require74 the <systemitem class="username">lfs</systemitem> user to have a password),75 give <systemitem class="username">lfs</systemitem> apassword:</para>71 <para>If you want to log in as &lfs-user; or switch to &lfs-user; from a 72 non-&root; user (as opposed to switching to user &lfs-user; 73 when logged in as &root;, which does not require the &lfs-user; user to 74 have a password), you need to set a password of &lfs-user;. Issue the 75 following command as the &root; user to set the password:</para> 76 76 77 77 <screen role="nodump"><userinput>passwd lfs</userinput></screen> 78 78 79 79 <para>Grant <systemitem class="username">lfs</systemitem> full access to 80 all directories under <filename class="directory">$LFS</filename> by making81 <systemitem class="username">lfs</systemitem> the directoryowner:</para>80 all the directories under <filename class="directory">$LFS</filename> by making 81 <systemitem class="username">lfs</systemitem> the owner:</para> 82 82 83 83 <screen><userinput>chown -v lfs $LFS/{usr{,/*},lib,var,etc,bin,sbin,tools} … … 89 89 <screen arch="ml_all" ><userinput>chown -v lfs $LFS/{lib32,libx32}</userinput></screen> 90 90 91 <note><para>In some host systems, the followingcommand does not complete92 properly and suspends the login tothe &lfs-user; user to the background.91 <note><para>In some host systems, the following <command>su</command> command does not complete 92 properly and suspends the login for the &lfs-user; user to the background. 93 93 If the prompt "lfs:~$" does not appear immediately, entering the 94 94 <command>fg</command> command will fix the issue.</para></note> 95 95 96 <para>Next, login as user <systemitem class="username">lfs</systemitem>.97 This can be done via a virtual console, through a display manager, or with98 the followingsubstitute/switch user command:</para>96 <para>Next, start a shell running as user &lfs-user;. This can be done by 97 logging in as &lfs-user; on a virtual console, or with the following 98 substitute/switch user command:</para> 99 99 100 100 <screen role="nodump"><userinput>su - lfs</userinput></screen> … … 102 102 <para>The <quote><parameter>-</parameter></quote> instructs 103 103 <command>su</command> to start a login shell as opposed to a non-login shell. 104 The difference between these two types of shells can be found in detail in104 The difference between these two types of shells is described in detail in 105 105 <filename>bash(1)</filename> and <command>info bash</command>.</para> 106 106 -
chapter04/creatingminlayout.xml
rf29824e rc093e6b 11 11 <title>Creating a limited directory layout in LFS filesystem</title> 12 12 13 <para>The first task performed in the LFS partition is to create a limited 14 directory hierarchy so that programs compiled in <xref 13 <para>In this section, we begin populating the LFS filesystem with the 14 pieces that will constitute the final Linux system. The first step is to 15 create a limited directory hierarchy, so that the programs compiled in <xref 15 16 linkend="chapter-temporary-tools"/> (as well as glibc and libstdc++ in <xref 16 linkend="chapter-cross-tools"/>) maybe installed in their final17 location. This is needed so that those temporary programs be overwritten18 when rebuilding themin <xref linkend="chapter-building-system"/>.</para>17 linkend="chapter-cross-tools"/>) can be installed in their final 18 location. We do this so those temporary programs will be overwritten when 19 the final versions are built in <xref linkend="chapter-building-system"/>.</para> 19 20 20 <para>Create the required directory layout by running the followingas21 <para>Create the required directory layout by issuing the following commands as 21 22 <systemitem class="username">root</systemitem>:</para> 22 23 … … 39 40 40 41 <para>Programs in <xref linkend="chapter-temporary-tools"/> will be compiled 41 with a cross-compiler (more details in section <xref42 linkend="ch-tools-toolchaintechnotes"/>). In order to separate this43 cross-compiler from the other programs, it will be installed in a special44 directory. Create this directory with:</para>42 with a cross-compiler (more details can be found in section <xref 43 linkend="ch-tools-toolchaintechnotes"/>). This cross-compiler will be installed 44 in a special directory, to separate it from the other programs. Still acting as 45 &root;, create that directory with this command:</para> 45 46 46 47 <screen><userinput>mkdir -pv $LFS/tools</userinput></screen> -
chapter04/introduction.xml
rf29824e rc093e6b 13 13 <para>In this chapter, we will perform a few additional tasks to prepare 14 14 for building the temporary system. We will create a set of directories in 15 <filename class="directory">$LFS</filename> for the installation ofthe16 temporary tools , add an unprivileged user to reduce risk,15 <filename class="directory">$LFS</filename> (in which we will install the 16 temporary tools), add an unprivileged user, 17 17 and create an appropriate build environment for that user. We will also 18 explain the unit of time we use to measure how long LFS packages take to19 build, or <quote>SBUs</quote>, and give some information about package18 explain the units of time (<quote>SBUs</quote>) we use to measure how 19 long it takes to build LFS packages, and provide some information about package 20 20 test suites.</para> 21 21 -
chapter04/settingenviron.xml
rf29824e rc093e6b 20 20 EOF</userinput></screen> 21 21 22 <para>When logged on as user <systemitem class="username">lfs</systemitem>, 23 the initial shell is usually a <emphasis>login</emphasis> shell which reads 22 <para>When logged on as user <systemitem class="username">lfs</systemitem> 23 or switched to the &lfs-user; user using a <command>su</command> command 24 with <quote><parameter>-</parameter></quote> option, 25 the initial shell is a <emphasis>login</emphasis> shell which reads 24 26 the <filename>/etc/profile</filename> of the host (probably containing some 25 27 settings and environment variables) and then <filename>.bash_profile</filename>. … … 33 35 34 36 <para>The new instance of the shell is a <emphasis>non-login</emphasis> 35 shell, which does not read, and execute, the contents of <filename>/etc/profile</filename> or37 shell, which does not read, and execute, the contents of the <filename>/etc/profile</filename> or 36 38 <filename>.bash_profile</filename> files, but rather reads, and executes, the 37 39 <filename>.bashrc</filename> file instead. Create the … … 74 76 <command>bash</command>'s hash function. Hashing is ordinarily a useful 75 77 feature—<command>bash</command> uses a hash table to remember the 76 full path ofexecutable files to avoid searching the <envar>PATH</envar>78 full path to executable files to avoid searching the <envar>PATH</envar> 77 79 time and again to find the same executable. However, the new tools should 78 be used as soon as they are installed. By switching off the hash function,79 the shell will always search the <envar>PATH</envar> whena program is to80 be used as soon as they are installed. Switching off the hash function forces 81 the shell to search the <envar>PATH</envar> whenever a program is to 80 82 be run. As such, the shell will find the newly compiled tools in 81 83 <filename class="directory">$LFS/tools/bin</filename> as soon as they are … … 130 132 <term><parameter>PATH=/usr/bin</parameter></term> 131 133 <listitem> 132 <para>Many modern linux distributions have merged <filename134 <para>Many modern Linux distributions have merged <filename 133 135 class="directory">/bin</filename> and <filename 134 136 class="directory">/usr/bin</filename>. When this is the case, the standard 135 <envar>PATH</envar> variable needs just tobe set to <filename137 <envar>PATH</envar> variable should be set to <filename 136 138 class="directory">/usr/bin/</filename> for the <xref 137 139 linkend="chapter-temporary-tools"/> environment. When this is not the … … 156 158 of <xref linkend="chapter-cross-tools"/> is picked up by the shell 157 159 immediately after its installation. This, combined with turning off hashing, 158 limits the risk that the compiler from the host beused instead of the160 limits the risk that the compiler from the host is used instead of the 159 161 cross-compiler.</para> 160 162 </listitem> … … 210 212 211 213 <para>Finally, to have the environment fully prepared for building the 212 temporary tools, source the just-created user profile:</para> 214 temporary tools, force the <command>bash</command> shell to read 215 the new user profile:</para> 213 216 214 217 <screen><userinput>source ~/.bash_profile</userinput></screen> -
chapter07/cleanup.xml
rf29824e rc093e6b 19 19 <screen><userinput>rm -rf /usr/share/{info,man,doc}/*</userinput></screen> 20 20 21 <para>Second, the libtool .la files are only useful when linking with static22 libraries. They are unneeded and potentially harmful when using dynamic23 shared libraries, especially when using non-autotools build systems.24 While still in chroot, remove those files now:</para>21 <para>Second, on a modern Linux system, the libtool .la files are only 22 useful for libltdl. No libraries in LFS are expected to be loaded by 23 libltdl, and it's known that some .la files can cause BLFS packages 24 fail to build. Remove those files now:</para> 25 25 26 26 <screen><userinput>find /usr/{lib,libexec} -name \*.la -delete</userinput><userinput arch="ml_32"> … … 99 99 <para>Before making a backup, unmount the virtual file systems:</para> 100 100 101 <screen role="nodump"><userinput>umount $LFS/dev/pts 101 <screen role="nodump"><userinput>mountpoint -q $LFS/dev/shm && umount $LFS/dev/shm 102 umount $LFS/dev/pts 102 103 umount $LFS/{sys,proc,run,dev}</userinput></screen> 103 104 -
chapter07/creatingdirs.xml
rf29824e rc093e6b 11 11 <title>Creating Directories</title> 12 12 13 <para>It is time to create the full structure in the LFS file system.</para>13 <para>It is time to create the full directory structure in the LFS file system.</para> 14 14 15 <note><para>Some of the directories mentioned in this section may be16 already created earlier with explicit instructionsor when installing some15 <note><para>Some of the directories mentioned in this section may have 16 already been created earlier with explicit instructions, or when installing some 17 17 packages. They are repeated below for completeness.</para></note> 18 18 … … 46 46 47 47 <para>Directories are, by default, created with permission mode 755, but 48 this is not desirable for all directories. In the commands above, two48 this is not desirable everywhere. In the commands above, two 49 49 changes are made—one to the home directory of user <systemitem 50 50 class="username">root</systemitem>, and another to the directories for … … 52 52 53 53 <para>The first mode change ensures that not just anybody can enter 54 the <filename class="directory">/root</filename> directory— the55 same as a normal user would do with his or herhome directory. The54 the <filename class="directory">/root</filename> directory—just 55 like a normal user would do with his or her own home directory. The 56 56 second mode change makes sure that any user can write to the 57 57 <filename class="directory">/tmp</filename> and <filename … … 63 63 <title>FHS Compliance Note</title> 64 64 65 <para>Th edirectory tree is based on the Filesystem Hierarchy Standard65 <para>This directory tree is based on the Filesystem Hierarchy Standard 66 66 (FHS) (available at <ulink 67 67 url="https://refspecs.linuxfoundation.org/fhs.shtml"/>). The FHS also specifies 68 the optional existence of somedirectories such as <filename68 the optional existence of additional directories such as <filename 69 69 class="directory">/usr/local/games</filename> and <filename 70 class="directory">/usr/share/games</filename>. We create only the71 directories that are needed. However, feel free to create these72 directories . </para>70 class="directory">/usr/share/games</filename>. In LFS, we create only the 71 directories that are really necessary. However, feel free to create more 72 directories, if you wish. </para> 73 73 74 74 </sect2> -
chapter07/introduction.xml
rf29824e rc093e6b 12 12 13 13 <para>This chapter shows how to build the last missing bits of the temporary 14 system: the tools needed by the build machinery ofvarious packages. Now14 system: the tools needed to build the various packages. Now 15 15 that all circular dependencies have been resolved, a <quote>chroot</quote> 16 16 environment, completely isolated from the host operating system (except for … … 18 18 19 19 <para>For proper operation of the isolated environment, some communication 20 with the running kernel must be established. This is done throughthe21 so-called <emphasis>Virtual Kernel File Systems</emphasis>, which mustbe22 mounted when entering the chroot environment. You may want to check23 that they are mounted by issuing <command>findmnt</command>.</para>20 with the running kernel must be established. This is done via the 21 so-called <emphasis>Virtual Kernel File Systems</emphasis>, which will be 22 mounted before entering the chroot environment. You may want to verify 23 that they are mounted by issuing the <command>findmnt</command> command.</para> 24 24 25 25 <para>Until <xref linkend="ch-tools-chroot"/>, the commands must be … … 28 28 are run as &root;, fortunately without access to the OS of the computer 29 29 you built LFS on. Be careful anyway, as it is easy to destroy the whole 30 LFS system with bad ly formedcommands.</para>30 LFS system with bad commands.</para> 31 31 32 32 </sect1> -
chapter07/kernfs.xml
rf29824e rc093e6b 15 15 </indexterm> 16 16 17 <para>Various file systems exported by the kernel are used to communicate to 18 and from the kernel itself. These file systems are virtual in that no disk 17 <para>Applications running in user space utilize various file 18 systems exported by the kernel to communicate 19 with the kernel itself. These file systems are virtual: no disk 19 20 space is used for them. The content of the file systems resides in 20 memory.</para> 21 memory. These file systems must be mounted in the $LFS directory tree 22 so the applications can find them in the chroot environment.</para> 21 23 22 <para>Begin by creating directories on towhich the file systems will be24 <para>Begin by creating directories on which the file systems will be 23 25 mounted:</para> 24 26 … … 28 30 <title>Mounting and Populating /dev</title> 29 31 30 <para>During a normal boot, the kernel automatically mounts the 31 <systemitem class="filesystem">devtmpfs</systemitem> filesystem on the 32 <filename class="directory">/dev</filename> directory, and allow the 33 devices to be created dynamically on that virtual filesystem as they 34 are detected or accessed. Device creation is generally done during the 35 boot process by the kernel and Udev. 36 Since this new system does not yet have Udev and 37 has not yet been booted, it is necessary to mount and populate 38 <filename class="directory">/dev</filename> manually. This is 39 accomplished by bind mounting the host system's 32 <para>During a normal boot of the LFS system, the kernel automatically 33 mounts the <systemitem class="filesystem">devtmpfs</systemitem> 34 filesystem on the 35 <filename class="directory">/dev</filename> directory; the kernel 36 creates device nodes on that virtual filesystem during the boot process 37 or when a device is first detected or accessed. The udev daemon may 38 change the owner or permission of the device nodes created by the 39 kernel, or create new device nodes or symlinks to ease the work of 40 distro maintainers or system administrators. (See 41 <xref linkend='ch-config-udev-device-node-creation'/> for details.) 42 If the host kernel supports &devtmpfs;, we can simply mount a 43 &devtmpfs; at <filename class='directory'>$LFS/dev</filename> and rely 44 on the kernel to populate it (the LFS building process does not need 45 the additional work onto &devtmpfs; by udev daemon).</para> 46 47 <para>But, some host kernels may lack &devtmpfs; support and these 48 host distros maintain the content of 49 <filename class="directory">/dev</filename> with different methods. 50 So the only host-agnostic way for populating 51 <filename class="directory">$LFS/dev</filename> is 52 bind mounting the host system's 40 53 <filename class="directory">/dev</filename> directory. A bind mount is 41 54 a special type of mount that allows you to create a mirror of a 42 directory or mount point tosome other location. Use the following43 command to achievethis:</para>55 directory or mount point at some other location. Use the following 56 command to do this:</para> 44 57 45 58 <screen><userinput>mount -v --bind /dev $LFS/dev</userinput></screen> … … 90 103 directory needs to be created.</para> 91 104 105 <para>In other host systems <filename>/dev/shm</filename> is a mount point 106 for a tmpfs. In that case the mount of /dev above will only create 107 /dev/shm as a directory in the chroot environment. In this situation 108 we must explicitly mount a tmpfs:</para> 109 92 110 <screen><userinput>if [ -h $LFS/dev/shm ]; then 93 111 mkdir -pv $LFS/$(readlink $LFS/dev/shm) 112 else 113 mount -t tmpfs -o nosuid,nodev tmpfs $LFS/dev/shm 94 114 fi</userinput></screen> 95 115 -
chapter08/autoconf.xml
rf29824e rc093e6b 41 41 <sect2 role="installation"> 42 42 <title>Installation of Autoconf</title> 43 <!--44 <para>First, apply a patch fixes several problems that occur with the latest45 perl, libtool, and bash versions.</para>46 43 47 <screen><userinput remap="pre">patch -Np1 -i ../&autoconf-fixes-patch;</userinput></screen> 48 --> 44 <para>First, fix several problems with the tests caused by bash-5.2 and later:</para> 45 46 <screen><userinput remap="pre">sed -e 's/SECONDS|/&SHLVL|/' \ 47 -e '/BASH_ARGV=/a\ /^SHLVL=/ d' \ 48 -i.orig tests/local.at</userinput></screen> 49 49 50 <para>Prepare Autoconf for compilation:</para> 50 51 -
chapter08/binutils.xml
rf29824e rc093e6b 179 179 <screen><userinput remap="test">make -k check</userinput></screen> 180 180 181 <para>Twelve tests fail in the <command>gold</command> testsuite when the181 <para>Twelve tests fail in the <command>gold</command> testsuite when the 182 182 <option>--enable-default-pie</option> and 183 <option>--enable-default-ssp</option> options are passed to GCC. There184 is also a known failure in the <command>as</command> tests.</para>185 <!-- Fixed in 2.39186 https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=01ae03b187 <para>One gold test, <filename>pr17704a_test</filename>, is known to188 fail if <parameter>CONFIG_IA32_EMULATION</parameter> is disabled in the189 k ernel configuration of the host system.</para>190 --> 183 <option>--enable-default-ssp</option> options are passed to GCC. 184 <!-- Caused by egrep deprecation. Note that we don't "patch" temp grep. 185 And it seems unworthy to add a sed into temp grep just for one test 186 failure. (I don't really agree to "patch" grep in the first place, 187 anyway.) --> 188 The test named <filename>ar with versioned solib</filename> is also 189 known to fail.</para> 190 191 191 <para>Install the package:</para> 192 192 -
chapter08/cleanup.xml
rf29824e rc093e6b 17 17 <para>There are also several files installed in the /usr/lib and /usr/libexec 18 18 directories with a file name extension of .la. These are "libtool archive" 19 files. As already said, they are only useful when linking with static20 libraries. They are unneeded, and potentially harmful, when using dynamic21 shared libraries, specially when using also non-autotools build systems.22 To remove them, run:</para>19 files. As already said, on a modern Linux system the libtool .la files are 20 only useful for libltdl. No libraries in LFS are expected to be loaded 21 by libltdl, and it's known that some .la files can cause BLFS packages 22 fail to build. Remove those files now:</para> 23 23 24 24 <screen><userinput>find /usr/lib /usr/libexec -name \*.la -delete</userinput><userinput arch="ml_32,ml_all"> -
chapter08/gawk.xml
rf29824e rc093e6b 47 47 <para>Now fix a programming error identified upstream:</para> 48 48 49 <screen><userinput remap="pre">sed -i -i'241i UPREF(m);' interpret.h</userinput></screen>49 <screen><userinput remap="pre">sed -i '241i UPREF(m);' interpret.h</userinput></screen> 50 50 51 51 <para>Prepare Gawk for compilation:</para> -
chapter08/gcc.xml
rf29824e rc093e6b 133 133 binary programs that can be loaded anywhere in memory. Without PIE, 134 134 the security feature named ASLR (Address Space Layout Randomization) 135 can be applied for the shared libraries, but not the exec tutable135 can be applied for the shared libraries, but not the executable 136 136 itself. Enabling PIE allows ASLR for the executables in addition to 137 137 the shared libraries, and mitigates some attacks based on fixed -
chapter08/libelf.xml
rf29824e rc093e6b 59 59 <screen><userinput remap="test">make check</userinput></screen> 60 60 61 <!-- <para>One test, run-elfclassify.sh, is known to fail.</para>--> 61 <para>One test named <filename>run-low_high_pc.sh</filename> is known to 62 fail on 32-bit x86 system.</para> 62 63 63 64 <para>Install only Libelf:</para> -
chapter08/pkgmgt.xml
rf29824e rc093e6b 12 12 13 13 <para>Package Management is an often requested addition to the LFS Book. A 14 Package Manager allows tracking the installation of files making it easyto14 Package Manager tracks the installation of files, making it easier to 15 15 remove and upgrade packages. As well as the binary and library files, a 16 16 package manager will handle the installation of configuration files. Before … … 18 18 any particular package manager. What it provides is a roundup of the more 19 19 popular techniques and how they work. The perfect package manager for you may 20 be among these techniques ormay be a combination of two or more of these20 be among these techniques, or it may be a combination of two or more of these 21 21 techniques. This section briefly mentions issues that may arise when upgrading 22 22 packages.</para> … … 33 33 <listitem> 34 34 <para>There are multiple solutions for package management, each having 35 its strengths and drawbacks. Including onethat satisfies all audiences35 its strengths and drawbacks. Finding one solution that satisfies all audiences 36 36 is difficult.</para> 37 37 </listitem> … … 40 40 <para>There are some hints written on the topic of package management. Visit 41 41 the <ulink url="&hints-root;">Hints Project</ulink> and see if one of them 42 fits your need .</para>42 fits your needs.</para> 43 43 44 44 <sect2 id='pkgmgmt-upgrade-issues'> … … 52 52 <itemizedlist> 53 53 <listitem> 54 <para>If Linux kernel needs to be upgraded (for example, from55 5.10.17 to 5.10.18 or 5.11.1), nothing else need to be rebuilt.56 The system will keep working fine thanks to the well-defined border57 between kernel and userspace. Specifically, Linux API headers58 need not tobe (and should not be, see the next item) upgraded59 along side the kernel. You'llneed to reboot your system to use the54 <para>If the Linux kernel needs to be upgraded (for example, from 55 5.10.17 to 5.10.18 or 5.11.1), nothing else needs to be rebuilt. 56 The system will keep working fine thanks to the well-defined interface 57 between the kernel and user space. Specifically, Linux API headers 58 need not be (and should not be, see the next item) upgraded 59 along with the kernel. You will merely need to reboot your system to use the 60 60 upgraded kernel.</para> 61 61 </listitem> 62 62 63 63 <listitem> 64 <para>If Linux API headers or Glibc needsto be upgraded to a newer65 version, (e.g. from glibc-2.31 to glibc-2.32), it is safer to64 <para>If Linux API headers or glibc need to be upgraded to a newer 65 version, (e.g., from glibc-2.31 to glibc-2.32), it is safer to 66 66 rebuild LFS. Though you <emphasis>may</emphasis> be able to rebuild 67 67 all the packages in their dependency order, we do not recommend … … 71 71 <listitem> <para>If a package containing a shared library is updated, and 72 72 if the name of the library changes, then any packages dynamically 73 linked to the library need to be recompiled in orderto link against the73 linked to the library must be recompiled, to link against the 74 74 newer library. (Note that there is no correlation between the package 75 75 version and the name of the library.) For example, consider a package 76 foo-1.2.3 that installs a shared library with name <filename77 class='libraryfile'>libfoo.so.1</filename>. Ifyou upgrade the package to78 a newer version foo-1.2.4 that installs a shared library with name76 foo-1.2.3 that installs a shared library with the name <filename 77 class='libraryfile'>libfoo.so.1</filename>. Suppose you upgrade the package to 78 a newer version foo-1.2.4 that installs a shared library with the name 79 79 <filename class='libraryfile'>libfoo.so.2</filename>. In this case, any 80 80 packages that are dynamically linked to <filename 81 81 class='libraryfile'>libfoo.so.1</filename> need to be recompiled to link 82 82 against <filename class='libraryfile'>libfoo.so.2</filename> in order to 83 use the new library version. You should not remove the previous84 libraries un less all the dependent packages arerecompiled.</para>83 use the new library version. You should not remove the old 84 libraries until all the dependent packages have been recompiled.</para> 85 85 </listitem> 86 86 87 87 <listitem> <para>If a package containing a shared library is updated, 88 and the name of library doesn't change, but the version number of the88 and the name of the library doesn't change, but the version number of the 89 89 library <emphasis role="bold">file</emphasis> decreases (for example, 90 the name of the library is keptnamed90 the library is still named 91 91 <filename class='libraryfile'>libfoo.so.1</filename>, 92 but the name of library file is changed from92 but the name of the library file is changed from 93 93 <filename class='libraryfile'>libfoo.so.1.25</filename> to 94 94 <filename class='libraryfile'>libfoo.so.1.24</filename>), 95 95 you should remove the library file from the previously installed version 96 (<filename class='libraryfile'>libfoo.so.1.25</filename> in th ecase).97 O r, a <command>ldconfig</command> run (by yourself using acommand96 (<filename class='libraryfile'>libfoo.so.1.25</filename> in this case). 97 Otherwise, a <command>ldconfig</command> command (invokeed by yourself from the command 98 98 line, or by the installation of some package) will reset the symlink 99 99 <filename class='libraryfile'>libfoo.so.1</filename> to point to 100 the old library file because it seems havinga <quote>newer</quote>101 version , as its version number is larger. This situation may happenif102 you have to downgrade a package, or the package changesthe versioning103 scheme of library files suddenly.</para> </listitem>100 the old library file because it seems to be a <quote>newer</quote> 101 version; its version number is larger. This situation may arise if 102 you have to downgrade a package, or if the authors change the versioning 103 scheme for library files.</para> </listitem> 104 104 105 105 <listitem><para>If a package containing a shared library is updated, 106 and the name of library doesn't change, but a severe issue106 and the name of the library doesn't change, but a severe issue 107 107 (especially, a security vulnerability) is fixed, all running programs 108 108 linked to the shared library should be restarted. The following 109 109 command, run as <systemitem class="username">root</systemitem> after 110 updating, will list what isusing the old versions of those libraries110 the update is cmplete, will list which processes are using the old versions of those libraries 111 111 (replace <replaceable>libfoo</replaceable> with the name of the 112 112 library):</para> … … 116 116 117 117 <para> 118 If <application>OpenSSH</application> is being used for accessing119 the system and it is linked to the updated library, you need to120 restart <command>sshd</command> service, then logout, login again,121 and rerun th at command to confirmnothing is still using the118 If <application>OpenSSH</application> is being used to access 119 the system and it is linked to the updated library, you must 120 restart the <command>sshd</command> service, then logout, login again, 121 and rerun the preceding ps command to confirm that nothing is still using the 122 122 deleted libraries. 123 123 </para> … … 125 125 <para revision='systemd'> 126 126 If the <command>systemd</command> daemon (running as PID 1) is 127 linked to the updated library, you can restart it without reboot 127 linked to the updated library, you can restart it without rebooting 128 128 by running <command>systemctl daemon-reexec</command> as the 129 129 <systemitem class='username'>root</systemitem> user. … … 131 131 132 132 <listitem> 133 <para>If a binaryor a shared library is overwritten, the processes134 using the code or data in th e binaryor library may crash. The135 correct way to update a binaryor a shared library without causing133 <para>If an executable program or a shared library is overwritten, the processes 134 using the code or data in that program or library may crash. The 135 correct way to update a program or a shared library without causing 136 136 the process to crash is to remove it first, then install the new 137 version into position. The <command>install</command> command138 provided by <application> Coreutils</application> has already139 implemented this and most packages use it to install binaries and137 version. The <command>install</command> command 138 provided by <application>coreutils</application> has already 139 implemented this, and most packages use that command to install binary files and 140 140 libraries. This means that you won't be troubled by this issue most of the time. 141 141 However, the install process of some packages (notably Mozilla JS 142 in BLFS) just overwrites the file if it exists and causes a crash, so142 in BLFS) just overwrites the file if it exists; this causes a crash. So 143 143 it's safer to save your work and close unneeded running processes 144 before updating a package.</para> 144 before updating a package.</para> <!-- binary is an adjective, not a noun. --> 145 145 </listitem> 146 146 </itemizedlist> … … 153 153 <para>The following are some common package management techniques. Before 154 154 making a decision on a package manager, do some research on the various 155 techniques, particularly the drawbacks of theparticular scheme.</para>155 techniques, particularly the drawbacks of each particular scheme.</para> 156 156 157 157 <sect3> 158 158 <title>It is All in My Head!</title> 159 159 160 <para>Yes, this is a package management technique. Some folks do not find161 the need fora package manager because they know the packages intimately162 and know wh atfiles are installed by each package. Some users also do not160 <para>Yes, this is a package management technique. Some folks do not 161 need a package manager because they know the packages intimately 162 and know which files are installed by each package. Some users also do not 163 163 need any package management because they plan on rebuilding the entire 164 system when a package is changed.</para>164 system whenever a package is changed.</para> 165 165 166 166 </sect3> … … 169 169 <title>Install in Separate Directories</title> 170 170 171 <para>This is a simplistic package management t hat does not need any extra172 package to manage the installations. Each package is installed in a171 <para>This is a simplistic package management technique that does not need a 172 special program to manage the packages. Each package is installed in a 173 173 separate directory. For example, package foo-1.1 is installed in 174 174 <filename class='directory'>/usr/pkg/foo-1.1</filename> 175 175 and a symlink is made from <filename>/usr/pkg/foo</filename> to 176 <filename class='directory'>/usr/pkg/foo-1.1</filename>. When installing177 a new version foo-1.2 , it is installed in176 <filename class='directory'>/usr/pkg/foo-1.1</filename>. When 177 a new version foo-1.2 comes along, it is installed in 178 178 <filename class='directory'>/usr/pkg/foo-1.2</filename> and the previous 179 179 symlink is replaced by a symlink to the new version.</para> … … 182 182 <envar>LD_LIBRARY_PATH</envar>, <envar>MANPATH</envar>, 183 183 <envar>INFOPATH</envar> and <envar>CPPFLAGS</envar> need to be expanded to 184 include <filename>/usr/pkg/foo</filename>. Formore than a few packages,184 include <filename>/usr/pkg/foo</filename>. If you install more than a few packages, 185 185 this scheme becomes unmanageable.</para> 186 186 … … 191 191 192 192 <para>This is a variation of the previous package management technique. 193 Each package is installed similar tothe previous scheme. But instead of194 making the symlink , each file is symlinked into the193 Each package is installed as in the previous scheme. But instead of 194 making the symlink via a generic package name, each file is symlinked into the 195 195 <filename class='directory'>/usr</filename> hierarchy. This removes the 196 196 need to expand the environment variables. Though the symlinks can be 197 created by the user to automate the creation, many package managers have198 been written using this approach. A few of the popular ones include Stow,197 created by the user, many package managers use this approach, and 198 automate the creation of the symlinks. A few of the popular ones include Stow, 199 199 Epkg, Graft, and Depot.</para> 200 200 201 <para>The installation needs to be faked, so that the package thinks that201 <para>The installation script needs to be fooled, so the package thinks 202 202 it is installed in <filename class="directory">/usr</filename> though in 203 203 reality it is installed in the … … 217 217 instead of <filename class='libraryfile'>/usr/lib/libfoo.so.1</filename> 218 218 as you would expect. The correct approach is to use the 219 <envar>DESTDIR</envar> strategy to fake installation of the package. This219 <envar>DESTDIR</envar> variable to direct the installation. This 220 220 approach works as follows:</para> 221 221 … … 225 225 226 226 <para>Most packages support this approach, but there are some which do not. 227 For the non-compliant packages, you may either need to manuallyinstall the228 package , or you may find that it is easier to install some problematic227 For the non-compliant packages, you may either need to install the 228 package manually, or you may find that it is easier to install some problematic 229 229 packages into <filename class='directory'>/opt</filename>.</para> 230 230 … … 238 238 <command>find</command> command with the appropriate options can generate 239 239 a log of all the files installed after the timestamp file was created. A 240 package manager written withthis approach is install-log.</para>240 package manager that uses this approach is install-log.</para> 241 241 242 242 <para>Though this scheme has the advantage of being simple, it has two 243 243 drawbacks. If, during installation, the files are installed with any 244 244 timestamp other than the current time, those files will not be tracked by 245 the package manager. Also, this scheme can only be used when one package246 is installedat a time. The logs are not reliable if two packages are247 being installed ontwo different consoles.</para>245 the package manager. Also, this scheme can only be used when packages 246 are installed one at a time. The logs are not reliable if two packages are 247 installed simultaneously from two different consoles.</para> 248 248 249 249 </sect3> … … 263 263 executables need to be dynamically linked without the suid or sgid bit. 264 264 Preloading the library may cause some unwanted side-effects during 265 installation. Therefore, it is advised that one performssome tests to266 ensure that the package manager does not break anything andlogs all the265 installation. Therefore, it's a good idea to perform some tests to 266 ensure that the package manager does not break anything, and that it logs all the 267 267 appropriate files.</para> 268 268 269 <para> The secondtechnique is to use <command>strace</command>, which270 logs all system calls made during the execution of the installation269 <para>Another technique is to use <command>strace</command>, which 270 logs all the system calls made during the execution of the installation 271 271 scripts.</para> 272 272 </sect3> … … 276 276 277 277 <para>In this scheme, the package installation is faked into a separate 278 tree as described in the Symlink style package management. After the278 tree as previously described in the symlink style package management section. After the 279 279 installation, a package archive is created using the installed files. 280 This archive is then used to install the package eitheron the local281 machine or can even be used to install the packageon other machines.</para>280 This archive is then used to install the package on the local 281 machine or even on other machines.</para> 282 282 283 283 <para>This approach is used by most of the package managers found in the … … 290 290 url="&hints-root;fakeroot.txt"/>.</para> 291 291 292 <para> Creation of package files that include dependency information is293 complex and isbeyond the scope of LFS.</para>294 295 <para>Slackware uses a <command>tar</command> 292 <para>The creation of package files that include dependency information is 293 complex, and beyond the scope of LFS.</para> 294 295 <para>Slackware uses a <command>tar</command>-based system for package 296 296 archives. This system purposely does not handle package dependencies 297 297 as more complex package managers do. For details of Slackware package … … 323 323 simple as using <command>tar</command> on the LFS partition that contains 324 324 the root directory (about 250MB uncompressed for a base LFS build), copying 325 that file via network transfer or CD-ROM to the new systemand expanding326 it. From that point, a few configuration files will have to be changed.325 that file via network transfer or CD-ROM / USB stick to the new system, and expanding 326 it. After that, a few configuration files will have to be changed. 327 327 Configuration files that may need to be updated include: 328 328 <filename>/etc/hosts</filename>, … … 343 343 </para> 344 344 345 <para>A custom kernel may need to be built for the new systemdepending on345 <para>A custom kernel may be needed for the new system, depending on 346 346 differences in system hardware and the original kernel 347 347 configuration.</para> … … 349 349 <note><para>There have been some reports of issues when copying between 350 350 similar but not identical architectures. For instance, the instruction set 351 for an Intel system is not identical with an AMD processorand later352 versions of some processors may have instructions that are unavailable in351 for an Intel system is not identical with the AMD processor's instructions, and later 352 versions of some processors may provide instructions that are unavailable with 353 353 earlier versions.</para></note> 354 354 355 <para>Finally the new system has to be made bootable via <xref355 <para>Finally, the new system has to be made bootable via <xref 356 356 linkend="ch-bootable-grub"/>.</para> 357 357 -
chapter09/systemd-custom.xml
rf29824e rc093e6b 47 47 </sect2> 48 48 49 <sect2 >49 <sect2 id='systemd-no-tmpfs'> 50 50 <title>Disabling tmpfs for /tmp</title> 51 51 -
chapter09/udev.xml
rf29824e rc093e6b 94 94 </sect3> 95 95 96 <sect3 >96 <sect3 id='ch-config-udev-device-node-creation'> 97 97 <title>Device Node Creation</title> 98 98 -
chapter10/fstab.xml
rf29824e rc093e6b 33 33 tmpfs /run tmpfs defaults 0 0 34 34 devtmpfs /dev devtmpfs mode=0755,nosuid 0 0 35 tmpfs /dev/shm tmpfs nosuid,nodev 0 0 35 36 36 37 # End /etc/fstab</literal> -
chapter10/kernel.xml
rf29824e rc093e6b 118 118 Graphics support ---> 119 119 Frame buffer Devices ---> 120 [*] Support for frame buffer devices ---- 120 <*> Support for frame buffer devices ---> 121 Console display driver support ---> 122 [*] Framebuffer Console support [CONFIG_FRAMEBUFFER_CONSOLE] 121 123 Generic Driver Options ---> 122 124 [ ] Support for uevent helper [CONFIG_UEVENT_HELPER] … … 157 159 Frame buffer Devices ---> 158 160 <*> Support for frame buffer devices ---> 161 Console display driver support ---> 162 [*] Framebuffer Console support [CONFIG_FRAMEBUFFER_CONSOLE] 159 163 File systems ---> 160 164 [*] Inotify support for userspace [CONFIG_INOTIFY_USER] … … 303 307 304 308 <varlistentry> 309 <term><parameter>Framebuffer Console support</parameter></term> 310 <listitem> 311 <para>This is needed to display the Linux console on a frame 312 buffer device. To allow the kernel to print debug messages at an 313 early boot stage, it shouldn't be built as a kernel module 314 unless an initramfs will be used. And, if 315 <option>CONFIG_DRM</option> (Direct Rendering Manager) is enabled, 316 it's likely <option>CONFIG_DRM_FBDEV_EMULATION</option> (Enable 317 legacy fbdev support for your modesetting driver) should be 318 enabled as well.</para> 319 </listitem> 320 </varlistentry> 321 322 <varlistentry> 305 323 <term><parameter>Support x2apic</parameter></term> 306 324 <listitem> … … 350 368 351 369 <caution> 352 <para>If the host system has a separate /boot partition, the files copied 353 below should go there. The easiest way to do that is to bind /boot on the 354 host (outside chroot) to /mnt/lfs/boot before proceeding. As the 355 &root; user in the <emphasis>host system</emphasis>:</para> 356 357 <screen role="nodump"><userinput>mount --bind /boot /mnt/lfs/boot</userinput></screen> 370 <para>If you've decided to use a separate &boot-dir; partition for the 371 LFS system (maybe sharing a &boot-dir; partition with the host 372 distro) , the files copied below should go there. The easiest way to 373 do that is to create the entry for &boot-dir; in &fstab; first (read 374 the previous section for details), then issue the following command 375 as the &root; user in the 376 <emphasis>chroot environment</emphasis>:</para> 377 378 <screen role="nodump"><userinput>mount /boot</userinput></screen> 379 380 <para>The path to the device node is omitted in the command because 381 <command>mount</command> can read it from &fstab;.</para> 358 382 </caution> 359 383 -
chapter11/chapter11.xml
rf29824e rc093e6b 16 16 <xi:include xmlns:xi="http://www.w3.org/2001/XInclude" href="reboot.xml"/> 17 17 <xi:include xmlns:xi="http://www.w3.org/2001/XInclude" href="whatnow.xml"/> 18 <xi:include xmlns:xi="http://www.w3.org/2001/XInclude" href="afterlfs.xml"/> 18 19 19 20 </chapter> -
chapter11/reboot.xml
rf29824e rc093e6b 10 10 11 11 <title>Rebooting the System</title> 12 <para> 13 Now that all of the software has been installed, it is time to reboot 14 your computer. However, there are still a few things to check. 15 Here are some suggestions:</para> 12 16 13 <para>Now that all of the software has been installed, it is time to reboot 14 your computer. However, you should be aware of a few things. The system you 17 18 <!-- 19 The system you 15 20 have created in this book is quite minimal, and most likely will not have 16 21 the functionality you would need to be able to continue forward. By installing … … 18 23 environment, you can leave yourself in a much better position to continue on 19 24 once you reboot into your new LFS installation. Here are some suggestions:</para> 20 25 --> 21 26 <itemizedlist> 22 27 <!-- 23 28 <listitem><para>A text mode browser such as <ulink 24 29 url='&blfs-book;basicnet/lynx.html'>Lynx</ulink> … … 61 66 url='&blfs-book;basicnet/wpa_supplicant.html'>wpa_supplicant</ulink>. 62 67 </para></listitem> 68 --> 69 <listitem> 70 <para> 71 Install any <ulink 72 url='&blfs-book;postlfs/firmware.html'>firmware</ulink> needed if the 73 kernel driver for your hardware requires some firmware files to function 74 properly. 75 </para> 76 </listitem> 63 77 64 <listitem><para>Install <ulink 65 url='&blfs-book;postlfs/firmware.html'>firmwares</ulink> if the kernel 66 driver for your hardware require some firmware to function properly. 67 </para></listitem> 68 69 <listitem><para>Finally, a review of the following configuration files 70 is also appropriate at this point.</para> 78 <listitem> 79 <para> 80 A review of the following configuration files 81 is also appropriate at this point. 82 </para> 71 83 72 84 <itemizedlist> … … 87 99 88 100 <para>Now that we have said that, let's move on to booting our shiny new LFS 89 installation for the first time! First exit from the chroot environment:</para> 101 installation for the first time! <emphasis>First exit from the chroot 102 environment</emphasis>:</para> 90 103 91 104 <screen><userinput>logout</userinput></screen> 92 <!-- We need to show the user the details... 93 <para>Unmount the LFS file system hierarchy:</para> 94 95 <screen><userinput>umount -Rv $LFS</userinput></screen> 96 --> 105 <!-- We need to show the user the details...--> 97 106 98 107 <para>Then unmount the virtual file systems:</para> … … 107 116 partitions before unmounting the main one, like this:</para> 108 117 109 <screen role="nodump"><userinput>umount -v $LFS/usr 110 umount -v $LFS/home 118 <screen role="nodump"><userinput>umount -v $LFS/home 111 119 umount -v $LFS</userinput></screen> 112 120 … … 115 123 <screen role="nodump"><userinput>umount -v $LFS</userinput></screen> 116 124 117 <para>Now, reboot the system with:</para>125 <para>Now, reboot the system.</para> 118 126 119 <screen role="nodump"><userinput>shutdown -r now</userinput></screen> 127 <!-- Commented out because we don't have a host system requirement on 128 its init, and different init system may recommend different commands 129 for reboot. --> 130 <!--<screen role="nodump"><userinput>shutdown -r now</userinput></screen>--> 120 131 121 132 <para>Assuming the GRUB boot loader was set up as outlined earlier, the menu -
chapter11/theend.xml
rf29824e rc093e6b 61 61 EOF</userinput></screen> 62 62 63 <para>Be sure to put some sort of customization for the fields 64 'DISTRIB_CODENAME' and 'VERSION_CODENAME' to make the system uniquely 65 yours.</para> 63 <para>Be sure to customize the fields 'DISTRIB_CODENAME' and 64 'VERSION_CODENAME' to make the system uniquely yours.</para> 66 65 67 66 </sect1> -
chapter11/whatnow.xml
rf29824e rc093e6b 9 9 <?dbhtml filename="whatnow.html"?> 10 10 11 <title> What Now?</title>11 <title>Additional Resources</title> 12 12 13 13 <para>Thank you for reading this LFS book. We hope that you have … … 39 39 40 40 <listitem> 41 <para><ulink url="https://seclists.org/oss-sec/">Open Source Security Mailing List</ulink></para> 41 <para><ulink url="https://seclists.org/oss-sec/">Open Source Security 42 Mailing List</ulink></para> 42 43 43 44 <para>This is a mailing list for discussion of security flaws, … … 47 48 </itemizedlist> 48 49 </listitem> 49 50 <!-- 50 51 <listitem> 51 52 <para>Beyond Linux From Scratch</para> … … 56 57 </para> 57 58 </listitem> 58 59 --> 59 60 <listitem> 60 61 <para>LFS Hints</para> -
general.ent
rf29824e rc093e6b 122 122 <!ENTITY root "<systemitem class='username'>root</systemitem>"> 123 123 <!ENTITY lfs-user "<systemitem class='username'>lfs</systemitem>"> 124 <!ENTITY devtmpfs "<systemitem class='filesystem'>devtmpfs</systemitem>"> 125 <!ENTITY fstab "<filename>/etc/fstab</filename>"> 126 <!ENTITY boot-dir "<filename class='directory'>/boot</filename>"> 127 <!ENTITY ch-final "<xref linkend='chapter-building-system'/>"> 128 <!ENTITY ch-tmp-cross "<xref linkend='chapter-temporary-tools'/>"> 129 <!ENTITY ch-tmp-chroot "<xref linkend='chapter-chroot-temporary-tools'/>"> 124 130 125 131 <!ENTITY % packages-entities SYSTEM "packages.ent"> -
lfs-latest-git.php
rf29824e rc093e6b 35 35 // Isolate the version and put in an array 36 36 $slice = preg_replace( $regex_replace, "$1", $line ); 37 if ( $slice == $line) continue;37 if ( strcmp( $slice, $line ) == 0 ) continue; 38 38 39 39 array_push( $a, $slice ); … … 267 267 return find_max( $lines, "/^\d/", "/^(\d[\d\.]+\d)\/.*$/" ); 268 268 269 if ( $package == "iana-etc" ) 270 return find_max( $lines, "/^\s*20\d\d/", "/^\s+(\d+).*$/" ); 271 272 if ( $package == "meson" ) 273 return find_max( $lines, "/^\s+\d\./", "/^\s+([\d\.]+)$/" ); 274 275 if ( $package == "shadow" ) 276 return find_max( $lines, "/^\s+\d\./", "/^\s+([\d\.]+)$/" ); 277 269 278 if ( $package == "XML-Parser" ) 270 279 { … … 292 301 return str_replace( "_", ".", $max ); 293 302 } 303 304 if ( $package == "libffi" ) 305 return find_max( $lines, "/v\d/", "/^.*v([\d\.]+)$/" ); 294 306 295 307 if ( $package == "procps-ng" ) -
packages.ent
rf29824e rc093e6b 49 49 <!ENTITY automake-fin-sbu "less than 0.1 SBU (about 7.7 SBU with tests)"> 50 50 51 <!ENTITY bash-version "5. 1.16">52 <!ENTITY bash-size "10, 277KB">51 <!ENTITY bash-version "5.2"> 52 <!ENTITY bash-size "10,695 KB"> 53 53 <!ENTITY bash-url "&gnu;bash/bash-&bash-version;.tar.gz"> 54 <!ENTITY bash-md5 "c 17b20a09fc38d67fb303aeb6c130b4e">54 <!ENTITY bash-md5 "cfb4cf795fc239667f187b3d6b3d396f"> 55 55 <!ENTITY bash-home "&gnu-software;bash/"> 56 56 <!ENTITY bash-tmp-du "64 MB"> … … 59 59 <!ENTITY bash-fin-sbu "1.4 SBU"> 60 60 61 <!ENTITY bc-version "6.0. 2">61 <!ENTITY bc-version "6.0.4"> 62 62 <!ENTITY bc-size "442 KB"> 63 63 <!ENTITY bc-url "https://github.com/gavinhoward/bc/releases/download/&bc-version;/bc-&bc-version;.tar.xz"> 64 <!ENTITY bc-md5 "1 01e62dd9c2b90bf18c38d858aa36f0d">64 <!ENTITY bc-md5 "1e1c90de1a11f3499237425de1673ef1"> 65 65 <!ENTITY bc-home "https://git.yzena.com/gavin/bc"> 66 66 <!ENTITY bc-fin-du "7.4 MB"> … … 115 115 <!ENTITY coreutils-fin-sbu "2.8 SBU"> 116 116 117 <!ENTITY dbus-version "1.14. 0">117 <!ENTITY dbus-version "1.14.2"> 118 118 <!ENTITY dbus-size "1,332 KB"> 119 119 <!ENTITY dbus-url "https://dbus.freedesktop.org/releases/dbus/dbus-&dbus-version;.tar.xz"> 120 <!ENTITY dbus-md5 " ddd5570aff05191dbee8e42d751f1b7d">120 <!ENTITY dbus-md5 "2d9a6b441e6f844d41c35a004f0ef50b"> 121 121 <!ENTITY dbus-home "https://www.freedesktop.org/wiki/Software/dbus"> 122 122 <!ENTITY dbus-fin-du "19 MB"> … … 164 164 <!ENTITY eudev-fin-sbu "0.2 SBU"> 165 165 166 <!ENTITY expat-version "2.4. 8">167 <!ENTITY expat-size "44 4KB">166 <!ENTITY expat-version "2.4.9"> 167 <!ENTITY expat-size "449 KB"> 168 168 <!ENTITY expat-url "&sourceforge;expat/expat-&expat-version;.tar.xz"> 169 <!ENTITY expat-md5 " 0584a7318a4c007f7ec94778799d72fe">169 <!ENTITY expat-md5 "8d7fcf7d02d08bf79d9ae5c21cc72c03"> 170 170 <!ENTITY expat-home "https://libexpat.github.io/"> 171 171 <!ENTITY expat-fin-du "12 MB"> … … 318 318 <!ENTITY gzip-fin-sbu "0.3 SBU"> 319 319 320 <!ENTITY iana-etc-version "20220 812">320 <!ENTITY iana-etc-version "20220922"> 321 321 <!ENTITY iana-etc-size "584 KB"> 322 322 <!ENTITY iana-etc-url "https://github.com/Mic92/iana-etc/releases/download/&iana-etc-version;/iana-etc-&iana-etc-version;.tar.gz"> 323 <!ENTITY iana-etc-md5 " 851a53efd53c77d0ad7b3d2b68d8a3fc">323 <!ENTITY iana-etc-md5 "2fdc746cfc1bc10f841760fd6a92618c"> 324 324 <!ENTITY iana-etc-home "https://www.iana.org/protocols"> 325 325 <!ENTITY iana-etc-fin-du "4.8 MB"> … … 392 392 <!ENTITY less-fin-sbu "less than 0.1 SBU"> 393 393 394 <!ENTITY lfs-bootscripts-version "20220 723"> <!-- Scripts depend on this format -->394 <!ENTITY lfs-bootscripts-version "20220920"> <!-- Scripts depend on this format --> 395 395 <!ENTITY lfs-bootscripts-size "BOOTSCRIPTS-SIZE KB"> 396 396 <!ENTITY lfs-bootscripts-url "&downloads-root;lfs-bootscripts-&lfs-bootscripts-version;.tar.xz"> … … 400 400 <!ENTITY lfs-bootscripts-cfg-sbu "less than 0.1 SBU"> 401 401 402 <!ENTITY libcap-version "2.6 5">403 <!ENTITY libcap-size "17 6KB">402 <!ENTITY libcap-version "2.66"> 403 <!ENTITY libcap-size "178 KB"> 404 404 <!ENTITY libcap-url "&kernel;linux/libs/security/linux-privs/libcap2/libcap-&libcap-version;.tar.xz"> 405 <!ENTITY libcap-md5 " 3543e753dd941255c4def6cc67a462bb">405 <!ENTITY libcap-md5 "00afd6e13bc94b2543b1a70770bdb41f"> 406 406 <!ENTITY libcap-home "https://sites.google.com/site/fullycapable/"> 407 407 <!ENTITY libcap-fin-du "2.7 MB"> 408 408 <!ENTITY libcap-fin-sbu "less than 0.1 SBU"> 409 409 410 <!ENTITY libffi-version "3.4. 2">411 <!ENTITY libffi-size "1,32 0KB">410 <!ENTITY libffi-version "3.4.3"> 411 <!ENTITY libffi-size "1,327 KB"> 412 412 <!ENTITY libffi-url "https://github.com/libffi/libffi/releases/download/v&libffi-version;/libffi-&libffi-version;.tar.gz"> 413 <!ENTITY libffi-md5 " 294b921e6cf9ab0fbaea4b639f8fdbe8">413 <!ENTITY libffi-md5 "b57b0ac1d1072681cee9148a417bd2ec"> 414 414 <!ENTITY libffi-home "https://sourceware.org/libffi/"> 415 415 <!ENTITY libffi-fin-du "10 MB"> … … 434 434 <!ENTITY linux-major-version "5"> 435 435 <!ENTITY linux-minor-version "19"> 436 <!ENTITY linux-patch-version " 8">436 <!ENTITY linux-patch-version "12"> 437 437 <!--<!ENTITY linux-version "&linux-major-version;.&linux-minor-version;">--> 438 438 <!ENTITY linux-version "&linux-major-version;.&linux-minor-version;.&linux-patch-version;"> 439 <!ENTITY linux-size "128,5 47KB">439 <!ENTITY linux-size "128,599 KB"> 440 440 <!ENTITY linux-url "&kernel;linux/kernel/v&linux-major-version;.x/linux-&linux-version;.tar.xz"> 441 <!ENTITY linux-md5 " ae08d14f9b7ed3d47c0d22b6d235507a">441 <!ENTITY linux-md5 "6a8c953d04986027b033bc92185745bf"> 442 442 <!ENTITY linux-home "https://www.kernel.org/"> 443 443 <!-- measured for 5.13.4 / gcc-11.1.0 on x86_64 : minimum is … … 612 612 <!ENTITY python-docs-size "7,176 KB"> 613 613 614 <!ENTITY readline-version "8. 1.2">615 <!ENTITY readline-soversion "8. 1"><!-- used for stripping -->616 <!ENTITY readline-size "2,9 23 KB">614 <!ENTITY readline-version "8.2"> 615 <!ENTITY readline-soversion "8.2"><!-- used for stripping --> 616 <!ENTITY readline-size "2,973 KB"> 617 617 <!ENTITY readline-url "&gnu;readline/readline-&readline-version;.tar.gz"> 618 <!ENTITY readline-md5 " 12819fa739a78a6172400f399ab34f81">618 <!ENTITY readline-md5 "4aa1b31be779e6b84f9a96cb66bc50f6"> 619 619 <!ENTITY readline-home "https://tiswww.case.edu/php/chet/readline/rltop.html"> 620 620 <!ENTITY readline-fin-du "15 MB"> … … 704 704 <!ENTITY texinfo-fin-sbu "0.6 SBU"> 705 705 706 <!ENTITY tzdata-version "2022 c">707 <!ENTITY tzdata-size "42 3KB">706 <!ENTITY tzdata-version "2022d"> 707 <!ENTITY tzdata-size "424 KB"> 708 708 <!ENTITY tzdata-url "https://www.iana.org/time-zones/repository/releases/tzdata&tzdata-version;.tar.gz"> 709 <!ENTITY tzdata-md5 " 4e3b2369b68e713ba5d3f7456f20bfdb">709 <!ENTITY tzdata-md5 "e55dbeb2121230a0ae7c58dbb47ae8c8"> 710 710 <!ENTITY tzdata-home "https://www.iana.org/time-zones"> 711 711 -
part3intro/generalinstructions.xml
rf29824e rc093e6b 12 12 <title>General Compilation Instructions</title> 13 13 14 <para>When building packages there are several assumptions made within 15 the instructions:</para> 14 <para>Here are some things you should know about building each package:</para> 16 15 17 16 <itemizedlist> 18 17 19 18 <listitem> 20 <para>Several of thepackages are patched before compilation, but only when19 <para>Several packages are patched before compilation, but only when 21 20 the patch is needed to circumvent a problem. A patch is often needed in 22 both this and the following chapters, but sometimes in only one location. 21 both the current and the following chapters, but sometimes, when the same package 22 is built more than once, the patch is not needed right away. 23 23 Therefore, do not be concerned if instructions for a downloaded patch seem 24 24 to be missing. Warning messages about <emphasis>offset</emphasis> or 25 25 <emphasis>fuzz</emphasis> may also be encountered when applying a patch. Do 26 not worry about these warnings , asthe patch was still successfully26 not worry about these warnings; the patch was still successfully 27 27 applied.</para> 28 28 </listitem> 29 29 30 30 <listitem> 31 <para>During the compilation of most packages, there will be several32 warnings thatscroll by on the screen. These are normal and can safely be33 ignored. These warnings are as they appear—warningsabout31 <para>During the compilation of most packages, some 32 warnings will scroll by on the screen. These are normal and can safely be 33 ignored. These warnings are usually about 34 34 deprecated, but not invalid, use of the C or C++ syntax. C standards change 35 fairly often, and some packages still use the older standard. This is not a36 problem, but does prompt the warning.</para>35 fairly often, and some packages have not yet been updated. This is not a 36 serious problem, but it does cause the warnings to appear.</para> 37 37 </listitem> 38 38 … … 70 70 71 71 <listitem override='bullet'><para><command>/usr/bin/yacc</command> is a 72 symbolic link to <command>bison</command> ora small script that72 symbolic link to <command>bison</command>, or to a small script that 73 73 executes bison.</para></listitem> 74 74 … … 77 77 78 78 <important> 79 <para> To re-emphasize the build process:</para>79 <para>Here is a synopsis of the build process.</para> 80 80 81 81 <orderedlist numeration="arabic" spacing="compact"> 82 82 <listitem> 83 83 <para>Place all the sources and patches in a directory that will be 84 accessible from the chroot environment such as84 accessible from the chroot environment, such as 85 85 <filename class="directory">/mnt/lfs/sources/</filename>.<!-- Do 86 86 <emphasis>not</emphasis> put sources in … … 88 88 </listitem> 89 89 <listitem> 90 <para>Change to the sourcesdirectory.</para>90 <para>Change to the <filename class="directory">/mnt/lfs/sources/</filename> directory.</para> 91 91 </listitem> 92 92 <listitem id='buildinstr' xreflabel='Package build instructions'> … … 98 98 <xref linkend="chapter-temporary-tools"/>, ensure you are 99 99 the <emphasis>lfs</emphasis> user when extracting the package.</para> 100 <para>All methods to get the source code tree being built101 in-position, except extracting the package tarball, are not102 supported. Notably, using <command>cp -R</command>to copy the100 <para>Do not use any method except the <command>tar</command> command 101 to extract the source code. Notably, using the <command>cp -R</command> 102 command to copy the 103 103 source code tree somewhere else can destroy links and 104 timestamps in the sources tree and cause building 105 failure.</para> 104 timestamps in the source tree, and cause the build to fail.</para> 106 105 </listitem> 107 106 <listitem> … … 110 109 </listitem> 111 110 <listitem> 112 <para>Follow the book'sinstructions for building the package.</para>111 <para>Follow the instructions for building the package.</para> 113 112 </listitem> 114 113 <listitem> 115 <para>Change back to the sources directory .</para>114 <para>Change back to the sources directory when the build is complete.</para> 116 115 </listitem> 117 116 <listitem> -
part3intro/introduction.xml
rf29824e rc093e6b 11 11 <title>Introduction</title> 12 12 13 <para>This part is divided into three stages: first building a cross14 compiler and its associated libraries; second, us ethis cross toolchain13 <para>This part is divided into three stages: first, building a cross 14 compiler and its associated libraries; second, using this cross toolchain 15 15 to build several utilities in a way that isolates them from the host 16 distribution; third, enter the chroot environment,which further improves17 host isolation , and buildthe remaining tools needed to build the final16 distribution; and third, entering the chroot environment (which further improves 17 host isolation) and constructing the remaining tools needed to build the final 18 18 system.</para> 19 19 20 <important><para> With this part begins the real work of building a new21 system. It requires much care in ensuring that the instructions are22 followed exactly as the book shows them. You should try to understand23 what they do, and whatever your eagernessto finish your build, you should24 refrain from blindly typ e them as shown, but rather readdocumentation when20 <important><para>This is where the real work of building a new system 21 begins. Be very careful to follow the instructions exactly as the book 22 shows them. You should try to understand what each command does, 23 and no matter how eager you are to finish your build, you should 24 refrain from blindly typing the commands as shown. Read the documentation when 25 25 there is something you do not understand. Also, keep track of your typing 26 and of the output of commands, by sending them to a file, using the27 <command>tee</command> utility. This allows for better diagnosing28 if something g ets wrong.</para></important>26 and of the output of commands, by using the <command>tee</command> utility 27 to send the terminal output to a file. This makes debugging easier 28 if something goes wrong.</para></important> 29 29 30 <para>The next section gives a technical introduction to the build process,31 while the following one contains <emphasis role="strong">very30 <para>The next section is a technical introduction to the build process, 31 while the following one presents <emphasis role="strong">very 32 32 important</emphasis> general instructions.</para> 33 33 -
part3intro/toolchaintechnotes.xml
rf29824e rc093e6b 12 12 13 13 <para>This section explains some of the rationale and technical details 14 behind the overall build method. It is not essentialto immediately14 behind the overall build method. Don't try to immediately 15 15 understand everything in this section. Most of this information will be 16 clearer after performing an actual build. This section can be referred17 to at any time during theprocess.</para>16 clearer after performing an actual build. Come back and re-read this chapter 17 at any time during the build process.</para> 18 18 19 19 <para>The overall goal of <xref linkend="chapter-cross-tools"/> and <xref 20 linkend="chapter-temporary-tools"/> is to produce a temporary area that21 contain s a known-good set of tools that can be isolated from the host system.22 By using <command>chroot</command>, the commands in the remaining chapters23 will be contained within that environment, ensuring a clean, trouble-free20 linkend="chapter-temporary-tools"/> is to produce a temporary area 21 containing a set of tools that are known to be good, and that are isolated from the host system. 22 By using the <command>chroot</command> command, the compilations in the remaining chapters 23 will be isolated within that environment, ensuring a clean, trouble-free 24 24 build of the target LFS system. The build process has been designed to 25 minimize the risks for new readers and to provide the most educational value25 minimize the risks for new readers, and to provide the most educational value 26 26 at the same time.</para> 27 27 28 <para>Th e build process is based on the process of28 <para>This build process is based on 29 29 <emphasis>cross-compilation</emphasis>. Cross-compilation is normally used 30 for building a compiler and itstoolchain for a machine different from31 the one that is used for the build. This is not strictly ne ededfor LFS,30 to build a compiler and its associated toolchain for a machine different from 31 the one that is used for the build. This is not strictly necessary for LFS, 32 32 since the machine where the new system will run is the same as the one 33 used for the build. But cross-compilation has the great advantage that33 used for the build. But cross-compilation has one great advantage: 34 34 anything that is cross-compiled cannot depend on the host environment.</para> 35 35 … … 40 40 <note> 41 41 <para> 42 The LFS book is not , and does not containa general tutorial to43 build a cross (or native) toolchain. Don't use the command in the44 book for a cross toolchain which will be usedfor some purpose other42 The LFS book is not (and does not contain) a general tutorial to 43 build a cross (or native) toolchain. Don't use the commands in the 44 book for a cross toolchain for some purpose other 45 45 than building LFS, unless you really understand what you are doing. 46 46 </para> 47 47 </note> 48 48 49 <para>Cross-compilation involves some concepts that deserve a section o n50 their own. Although this section may be omitted in a first reading,51 coming back to it later will be beneficial to your fullunderstanding of49 <para>Cross-compilation involves some concepts that deserve a section of 50 their own. Although this section may be omitted on a first reading, 51 coming back to it later will help you gain a fuller understanding of 52 52 the process.</para> 53 53 54 <para>Let us first define some terms used in this context :</para>54 <para>Let us first define some terms used in this context.</para> 55 55 56 56 <variablelist> 57 <varlistentry><term> build</term><listitem>57 <varlistentry><term>The build</term><listitem> 58 58 <para>is the machine where we build programs. Note that this machine 59 is referred to as the <quote>host</quote> in other 60 sections.</para></listitem> 59 is also referred to as the <quote>host</quote>.</para></listitem> 61 60 </varlistentry> 62 61 63 <varlistentry><term> host</term><listitem>62 <varlistentry><term>The host</term><listitem> 64 63 <para>is the machine/system where the built programs will run. Note 65 64 that this use of <quote>host</quote> is not the same as in other … … 67 66 </varlistentry> 68 67 69 <varlistentry><term> target</term><listitem>68 <varlistentry><term>The target</term><listitem> 70 69 <para>is only used for compilers. It is the machine the compiler 71 produces code for. It may be different from both build and72 host.</para></listitem>70 produces code for. It may be different from both the build and 71 the host.</para></listitem> 73 72 </varlistentry> 74 73 … … 76 75 77 76 <para>As an example, let us imagine the following scenario (sometimes 78 referred to as <quote>Canadian Cross</quote>): we mayhave a77 referred to as <quote>Canadian Cross</quote>): we have a 79 78 compiler on a slow machine only, let's call it machine A, and the compiler 80 ccA. We may have also a fast machine (B), but with no compiler, and we may81 want to produce code for a nother slow machine (C). Tobuild a82 compiler for machine C , we would have three stages:</para>79 ccA. We also have a fast machine (B), but no compiler for (B), and we 80 want to produce code for a third, slow machine (C). We will build a 81 compiler for machine C in three stages.</para> 83 82 84 83 <informaltable align="center"> … … 96 95 <row> 97 96 <entry>1</entry><entry>A</entry><entry>A</entry><entry>B</entry> 98 <entry> build cross-compiler cc1 using ccA on machine A</entry>97 <entry>Build cross-compiler cc1 using ccA on machine A.</entry> 99 98 </row> 100 99 <row> 101 100 <entry>2</entry><entry>A</entry><entry>B</entry><entry>C</entry> 102 <entry> build cross-compiler cc2 using cc1 on machine A</entry>101 <entry>Build cross-compiler cc2 using cc1 on machine A.</entry> 103 102 </row> 104 103 <row> 105 104 <entry>3</entry><entry>B</entry><entry>C</entry><entry>C</entry> 106 <entry> build compiler ccC using cc2 on machine B</entry>105 <entry>Build compiler ccC using cc2 on machine B.</entry> 107 106 </row> 108 107 </tbody> … … 110 109 </informaltable> 111 110 112 <para>Then, all the otherprograms needed by machine C can be compiled111 <para>Then, all the programs needed by machine C can be compiled 113 112 using cc2 on the fast machine B. Note that unless B can run programs 114 produced for C, there is no way to test the built programs until machine115 C itself is running. For example, for testingccC, we may want to add a113 produced for C, there is no way to test the newly built programs until machine 114 C itself is running. For example, to run a test suite on ccC, we may want to add a 116 115 fourth stage:</para> 117 116 … … 130 129 <row> 131 130 <entry>4</entry><entry>C</entry><entry>C</entry><entry>C</entry> 132 <entry> rebuild and test ccC using itself on machine C</entry>131 <entry>Rebuild and test ccC using ccC on machine C.</entry> 133 132 </row> 134 133 </tbody> … … 147 146 148 147 <note> 149 <para>Almost all the build systems use names of the form 150 cpu-vendor-kernel-os referred to as the machine triplet. An astute 151 reader may wonder why a <quote>triplet</quote> refers to a four component 152 name. The reason is history: initially, three component names were enough 153 to designate a machine unambiguously, but with new machines and systems 154 appearing, that proved insufficient. The word <quote>triplet</quote> 155 remained. A simple way to determine your machine triplet is to run 156 the <command>config.guess</command> 148 <para>All packages involved with cross compilation in the book use an 149 autoconf-based building system. The autoconf-based building system 150 accepts system types in the form cpu-vendor-kernel-os, 151 referred to as the system triplet. Since the vendor field is mostly 152 irrelevant, autoconf allows to omit it. An astute reader may wonder 153 why a <quote>triplet</quote> refers to a four component name. The 154 reason is the kernel field and the os field originiated from one 155 <quote>system</quote> field. Such a three-field form is still valid 156 today for some systems, for example 157 <literal>x86_64-unknown-freebsd</literal>. But for other systems, 158 two systems can share the same kernel but still be too different to 159 use a same triplet for them. For example, an Android running on a 160 mobile phone is completely different from Ubuntu running on an ARM64 161 server, despite they are running on the same type of CPU (ARM64) and 162 using the same kernel (Linux). 163 Without an emulation layer, you cannot run an 164 executable for the server on the mobile phone or vice versa. So the 165 <quote>system</quote> field is separated into kernel and os fields to 166 designate these systems unambiguously. For our example, the Android 167 system is designated <literal>aarch64-unknown-linux-android</literal>, 168 and the Ubuntu system is designated 169 <literal>aarch64-unknown-linux-gnu</literal>. The word 170 <quote>triplet</quote> remained. A simple way to determine your 171 system triplet is to run the <command>config.guess</command> 157 172 script that comes with the source for many packages. Unpack the binutils 158 173 sources and run the script: <userinput>./config.guess</userinput> and note 159 174 the output. For example, for a 32-bit Intel processor the 160 175 output will be <emphasis>i686-pc-linux-gnu</emphasis>. On a 64-bit 161 system it will be <emphasis>x86_64-pc-linux-gnu</emphasis>.</para> 162 163 <para>Also be aware of the name of the platform's dynamic linker, often 176 system it will be <emphasis>x86_64-pc-linux-gnu</emphasis>. On most 177 Linux systems the even simpler <command>gcc -dumpmachine</command> command 178 will give you similar information.</para> 179 180 <para>You should also be aware of the name of the platform's dynamic linker, often 164 181 referred to as the dynamic loader (not to be confused with the standard 165 182 linker <command>ld</command> that is part of binutils). The dynamic linker 166 provided by Glibc finds and loads the shared libraries needed by a183 provided by package glibc finds and loads the shared libraries needed by a 167 184 program, prepares the program to run, and then runs it. The name of the 168 185 dynamic linker for a 32-bit Intel machine is <filename 169 class="libraryfile">ld-linux.so.2</filename> and is <filename170 class="libraryfile">ld-linux-x86-64.so.2</filename> for64-bit systems. A186 class="libraryfile">ld-linux.so.2</filename>; it's <filename 187 class="libraryfile">ld-linux-x86-64.so.2</filename> on 64-bit systems. A 171 188 sure-fire way to determine the name of the dynamic linker is to inspect a 172 189 random binary from the host system by running: <userinput>readelf -l 173 190 <name of binary> | grep interpreter</userinput> and noting the 174 191 output. The authoritative reference covering all platforms is in the 175 <filename>shlib-versions</filename> file in the root of the Glibc source192 <filename>shlib-versions</filename> file in the root of the glibc source 176 193 tree.</para> 177 194 </note> … … 179 196 <para>In order to fake a cross compilation in LFS, the name of the host triplet 180 197 is slightly adjusted by changing the "vendor" field in the 181 <envar>LFS_TGT</envar> variable . We also use the198 <envar>LFS_TGT</envar> variable so it says "lfs". We also use the 182 199 <parameter>--with-sysroot</parameter> option when building the cross linker and 183 200 cross compiler to tell them where to find the needed host files. This 184 201 ensures that none of the other programs built in <xref 185 202 linkend="chapter-temporary-tools"/> can link to libraries on the build 186 machine. Only two stages are mandatory, and one more for tests:</para>203 machine. Only two stages are mandatory, plus one more for tests.</para> 187 204 188 205 <informaltable align="center"> … … 200 217 <row> 201 218 <entry>1</entry><entry>pc</entry><entry>pc</entry><entry>lfs</entry> 202 <entry> build cross-compiler cc1 using cc-pc on pc</entry>219 <entry>Build cross-compiler cc1 using cc-pc on pc.</entry> 203 220 </row> 204 221 <row> 205 222 <entry>2</entry><entry>pc</entry><entry>lfs</entry><entry>lfs</entry> 206 <entry> build compiler cc-lfs using cc1 on pc</entry>223 <entry>Build compiler cc-lfs using cc1 on pc.</entry> 207 224 </row> 208 225 <row> 209 226 <entry>3</entry><entry>lfs</entry><entry>lfs</entry><entry>lfs</entry> 210 <entry> rebuild and test cc-lfs using itself on lfs</entry>227 <entry>Rebuild and test cc-lfs using cc-lfs on lfs.</entry> 211 228 </row> 212 229 </tbody> … … 214 231 </informaltable> 215 232 216 <para>In the abovetable, <quote>on pc</quote> means the commands are run233 <para>In the preceding table, <quote>on pc</quote> means the commands are run 217 234 on a machine using the already installed distribution. <quote>On 218 235 lfs</quote> means the commands are run in a chrooted environment.</para> … … 220 237 <para>Now, there is more about cross-compiling: the C language is not 221 238 just a compiler, but also defines a standard library. In this book, the 222 GNU C library, named glibc, is used . This library must223 be compiled for the lfs machine,that is, using the cross compiler cc1.239 GNU C library, named glibc, is used (there is an alternative, "musl"). This library must 240 be compiled for the LFS machine; that is, using the cross compiler cc1. 224 241 But the compiler itself uses an internal library implementing complex 225 instructions not available in the assembler instruction set. This226 internal library is named libgcc, and must be linked to the glibc242 subroutines for functions not available in the assembler instruction set. This 243 internal library is named libgcc, and it must be linked to the glibc 227 244 library to be fully functional! Furthermore, the standard library for 228 C++ (libstdc++) also needs being linked toglibc. The solution to this229 chicken and egg problem is to first build a degraded cc1based libgcc,230 lacking some functionalities such as threads and exception handling, then231 build glibc using this degraded compiler (glibc itself is not232 degraded), then build libstdc++. But this last library will lackthe233 same functionalities aslibgcc.</para>234 235 <para>This is not the end of the story: the conclusionof the preceding245 C++ (libstdc++) must also be linked with glibc. The solution to this 246 chicken and egg problem is first to build a degraded cc1-based libgcc, 247 lacking some functionalities such as threads and exception handling, and then 248 to build glibc using this degraded compiler (glibc itself is not 249 degraded), and also to build libstdc++. This last library will lack some of the 250 functionality of libgcc.</para> 251 252 <para>This is not the end of the story: the upshot of the preceding 236 253 paragraph is that cc1 is unable to build a fully functional libstdc++, but 237 254 this is the only compiler available for building the C/C++ libraries 238 255 during stage 2! Of course, the compiler built during stage 2, cc-lfs, 239 256 would be able to build those libraries, but (1) the build system of 240 GCC does not know that it is usable on pc, and (2) using it on pc 241 would be at risk of linking to the pc libraries, since cc-lfs is a native 242 compiler. So we have to build libstdc++ later, in chroot.</para> 257 gcc does not know that it is usable on pc, and (2) using it on pc 258 would create a risk of linking to the pc libraries, since cc-lfs is a native 259 compiler. So we have to re-build libstdc++ later as a part of 260 gcc stage 2.</para> 261 262 <para>In &ch-final; (or <quote>stage 3</quote>), all packages needed for 263 the LFS system are built. Even if a package is already installed into 264 the LFS system in a previous chapter, we still rebuild the package 265 unless we are completely sure it's unnecessary. The main reason for 266 rebuilding these packages is to settle them down: if we reinstall a LFS 267 package on a complete LFS system, the installed content of the package 268 should be same as the content of the same package installed in 269 &ch-final;. The temporary packages installed in &ch-tmp-cross; or 270 &ch-tmp-chroot; cannot satisify this expectation because some of them 271 are built without optional dependencies installed, and autoconf cannot 272 perform some feature checks in &ch-tmp-cross; because of cross 273 compilation, causing the temporary packages to lack optional features 274 or use suboptimal code routines. Additionally, a minor reason for 275 rebuilding the packages is allowing to run the testsuite.</para> 243 276 244 277 </sect2> … … 253 286 254 287 <para>Binutils is installed first because the <command>configure</command> 255 runs of both GCC and Glibc perform various feature tests on the assembler288 runs of both gcc and glibc perform various feature tests on the assembler 256 289 and linker to determine which software features to enable or disable. This 257 is more important than one might first realize. An incorrectly configured258 GCC or Glibc can result in a subtly broken toolchain, where the impact of290 is more important than one might realize at first. An incorrectly configured 291 gcc or glibc can result in a subtly broken toolchain, where the impact of 259 292 such breakage might not show up until near the end of the build of an 260 293 entire distribution. A test suite failure will usually highlight this error … … 275 308 will show all the files successfully opened during the linking.</para> 276 309 277 <para>The next package installed is GCC. An example of what can be310 <para>The next package installed is gcc. An example of what can be 278 311 seen during its run of <command>configure</command> is:</para> 279 312 … … 282 315 283 316 <para>This is important for the reasons mentioned above. It also 284 demonstrates that GCC's configure script does not search the PATH317 demonstrates that gcc's configure script does not search the PATH 285 318 directories to find which tools to use. However, during the actual 286 319 operation of <command>gcc</command> itself, the same search paths are not … … 296 329 297 330 <para>Next installed are sanitized Linux API headers. These allow the 298 standard C library ( Glibc) to interface with features that the Linux331 standard C library (glibc) to interface with features that the Linux 299 332 kernel will provide.</para> 300 333 301 <para>The next package installed is Glibc. The most important302 considerations for building Glibc are the compiler, binary tools, and303 kernel headers. The compiler is generally not an issue since Glibc will334 <para>The next package installed is glibc. The most important 335 considerations for building glibc are the compiler, binary tools, and 336 kernel headers. The compiler is generally not an issue since glibc will 304 337 always use the compiler relating to the <parameter>--host</parameter> 305 338 parameter passed to its configure script; e.g. in our case, the compiler … … 314 347 and the use of the <parameter>-nostdinc</parameter> and 315 348 <parameter>-isystem</parameter> flags to control the compiler's include 316 search path. These items highlight an important aspect of the Glibc349 search path. These items highlight an important aspect of the glibc 317 350 package—it is very self-sufficient in terms of its build machinery 318 351 and generally does not rely on toolchain defaults.</para> 319 352 320 <para>As said above, the standard C++ library is compiled next, followed in 321 <xref linkend="chapter-temporary-tools"/> by all the programs that need 322 themselves to be built. The install step of all those packages uses the 323 <envar>DESTDIR</envar> variable to have the 324 programs land into the LFS filesystem.</para> 353 <para>As mentioned above, the standard C++ library is compiled next, followed in 354 <xref linkend="chapter-temporary-tools"/> by other programs that need 355 to be cross compiled for breaking circular dependencies at build time. 356 The install step of all those packages uses the 357 <envar>DESTDIR</envar> variable to force installation 358 in the LFS filesystem.</para> 325 359 326 360 <para>At the end of <xref linkend="chapter-temporary-tools"/> the native 327 lfscompiler is installed. First binutils-pass2 is built,328 with the same <envar>DESTDIR</envar> installas the other programs,329 then the second pass of GCC is constructed, omitting libstdc++330 and other non-important libraries. Due to some weird logic in GCC's361 LFS compiler is installed. First binutils-pass2 is built, 362 in the same <envar>DESTDIR</envar> directory as the other programs, 363 then the second pass of gcc is constructed, omitting some 364 non-critical libraries. Due to some weird logic in gcc's 331 365 configure script, <envar>CC_FOR_TARGET</envar> ends up as 332 <command>cc</command> when the host is the same as the target, but is366 <command>cc</command> when the host is the same as the target, but 333 367 different from the build system. This is why 334 <parameter>CC_FOR_TARGET=$LFS_TGT-gcc</parameter> is put explicitly into335 the configureoptions.</para>368 <parameter>CC_FOR_TARGET=$LFS_TGT-gcc</parameter> is declared explicitly 369 as one of the configuration options.</para> 336 370 337 371 <para>Upon entering the chroot environment in <xref 338 linkend="chapter-chroot-temporary-tools"/>, the first task is to install339 libstdc++. Thentemporary installations of programs needed for the proper372 linkend="chapter-chroot-temporary-tools"/>, 373 the temporary installations of programs needed for the proper 340 374 operation of the toolchain are performed. From this point onwards, the 341 375 core toolchain is self-contained and self-hosted. In
Note:
See TracChangeset
for help on using the changeset viewer.