__group__ ticket summary component version milestone type owner status created _changetime _description _reporter Active Tickets 58 Libtool BOOK CLFS Sysroot GIT CLFS Sysroot 1.0.0 defect clfs-commits@… new 2006-07-16T17:04:05-07:00 2011-01-09T07:50:01-08:00 Libtool is completely broken. It's retaining all of the information about the cross-compiler. Which is reasonable because I can't figure out a way to make it look anywhere else. Libtool is a complete mess to the point where anything that may use it after you boot into the final system wont work. Joe Ciccone Active Tickets 1171 Ncurses 6.0 and Hosts with mawk and gawk BOOK CLFS Standard GIT CLFS Standard 3.1.0 defect clfs-commits@… new 2017-06-20T06:24:59-07:00 2017-06-20T06:24:59-07:00 "It may be that older versions of mawk are keeping ncurses 6.0 from building. Anyone with a host system install that has mawk and no gawk, or mawk and gawk may encounter issues where mawk is found first before gawk because of configure system. Add the following before configure: {{{ AWK=gawk }}} Another option is to add a '''sed''' command: {{{ sed -i 's/.:space:./ \t/g' ncurses/base/MKlib_gen.sh }}} or {{{ sed -i s/mawk// configure }}}" William Harrington Active Tickets 1173 Bootscripts mountfs script needs updating when using current util-linux mount BOOK CLFS Standard SYSVINIT CLFS Sysvinit 3.1.0 defect clfs-commits@… new 2017-06-22T17:24:29-07:00 2017-06-22T17:24:29-07:00 "With latest util-linux when fake mounting proc sys dev run then the bootscript will output a message about the filesystem not being found in /etc/fstab. Tested with an old build using util-linux 2.24.2 a fake mount didn't complain about it not being in /etc/fstab. Using latest util linux it will output that the entry isn't in /etc/fstab. For our mountfs script have this: {{{ mount -f /proc proc || failed=1 mount -f /sys proc || failed=1 mount -f /run proc || failed=1 mount -f /dev proc || failed=1 }}} It does matter when using a fake mount that device and dir are specified." William Harrington Active Tickets 923 "Integrate Joe's ""simp"" changes from main book" BOOK CLFS Standard GIT CLFS Standard 3.1.0 enhancement Andrew Bradford assigned 2013-06-18T19:00:46-07:00 2014-11-02T22:05:50-08:00 "Joe had worked on the 'simp' branch of the main book back in 2011. The goal was to provide a clfs specific schema that would present writing the book in a way that could have common and not-common sections all in one file. For instance, most of the instructions for building GCC are such that there's a ton of text that's common to all archs. Only some text is specific to which ever arch is being read. Right now we end up with a common file and one file per arch with lots of references back to the common file. This makes reading either the common file or each of the arch's files difficult as seeing how to get from the files to the output is difficult. With Joe's 'simp' changes everything goes in one file and the section that's specific to any arch or another goes inline with indication for which arch it's for. Multiple archs can go together in a block, such as if just one arch is different. Migration to this new 'simp' style can be done seamlessly based on the last tests I remember Joe doing. As things are converted over they automatically start using the new schemas and validation. Until they get converted, the old just docbook is used." Andrew Bradford Active Tickets 1011 Add ARM Arch to the book BOOK CLFS Standard GIT CLFS Standard 3.1.0 enhancement clfs-commits@… new 2014-10-30T19:32:44-07:00 2016-07-15T07:22:46-07:00 "ARM needs to be added as an architecture for our build system. Adding the book is not much different from a normal x86 build except that these few things need to be adjusted: Kernel Headers GCC configure options Kernel headers, either install the arm headers or download a kernel specific for any platform ( aka Raspberry pi) GCC, adjust options for cpu, float, and vfpu, and possibly something else. We need a table for various ARM cpu's with the options to use with GCC during configure regarding float, vfp, and target triplet. We'll have a table for targets like with mips. Any input will be great for starting a table with the proper options and can update GCC with some variables to use during configure from the table regarding target triplet, cpu, vfp, and float info." William Harrington Active Tickets 989 Eudev 3.2.2 Version Upgrade BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2014-10-24T17:15:11-07:00 2017-06-03T12:56:04-07:00 "Download: [http://dev.gentoo.org/~blueness/eudev/eudev-3.2.2.tar.gz] Size: 1.822 KB MD5: http://dev.gentoo.org/~blueness/eudev/eudev-3.2.2.tar.gz" William Harrington Active Tickets 1135 Add Gperf 3.0.4 to Ch 7 BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2017-04-14T14:45:26-07:00 2017-08-02T11:12:18-07:00 "The new Eudev [[ticket:989]] requires gperf. We need to add this to Ch 7 when the user wants to use the boot method." William Harrington Active Tickets 1185 GLIBC 2.26 GLOB_TILDE exploit BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2018-01-05T07:44:30-08:00 2018-01-05T11:06:48-08:00 "'''Major vulnerability''' Refer to the following URL's (CVE-2017-15670, CVE-2017-15671, CVE-2017-15804): * [https://sourceware.org/bugzilla/show_bug.cgi?id=22320] * [https://sourceware.org/bugzilla/show_bug.cgi?id=22325] * [https://sourceware.org/bugzilla/show_bug.cgi?id=22332] Attached a patch for GLIBC 2.26 until 2.27 is released. With the patch, not running a patched glibc on the system: * FAIL: posix/tst-glob-tilde * FAIL: posix/tst-glob-tilde-mem {{{ posix/tst-glob-tilde.out Didn't expect signal from child: got `Aborted' }}} {{{ posix/tst-glob-tilde-mem.out Memory not freed: ----------------- Address Size Caller 0x00007f6d0f2a4010 0x7a131 at 0x402c8d 0x00007f6d10ccd8e0 0x1869a at 0x7f6d0edf4ab5 0x00007f6d10cfbc90 0x18694 at 0x7f6d0edf4ab5 0x00007f6d10d14330 0x1869b at 0x7f6d0edf4ab5 0x00007f6d10d39180 0x18695 at 0x7f6d0edf4ab5 0x00007f6d10d69ec0 0x18696 at 0x7f6d0edf4ab5 0x00007f6d10d82560 0x186a0 at 0x7f6d0ed8e6f7 0x00007f6d10d9ac10 0x1869c at 0x7f6d0edf4ab5 }}}" William Harrington Active Tickets 915 Book mirroring BOOK task clfs-commits@… new 2012-09-23T21:48:28-07:00 2012-09-23T21:48:28-07:00 "Hello. I would like to mirror the cross-lfs book and patches. I think this is what the majority of users want when they come to cross-lfs.org. This is an example of what I'm thinking: [http://www.secondfloor.ca/cross-lfs.org/files/] I asked the server admin to install the xml packages needed to render the dev book daily from git. I'm also thinking to patch/sed the mirrored books to use my mirrored patches instead of patches.cross-lfs.org. Are you all cool with this? Is there anything else I can do?" Robert Connolly Active Tickets 983 CC and LDFLAGS in early final-system packages BOOK task clfs-commits@… new 2014-08-29T22:21:04-07:00 2014-09-26T19:50:56-07:00 "Several packages in the final-system have lines similar to this: {{{ CC=""gcc -isystem /usr/include"" \ CXX=""g++ -isystem /usr/include"" \ LDFLAGS=""-Wl,-rpath-link,/usr/lib:/lib"" \ ./configure [...] }}} This is found in most packages between Glibc and GCC in the final-system. There's no explanation given for these variables. I guess they're supposed to be to ensure that those packages look for headers and libs in the right place, though in my experience they build just fine without those. Also, even if the -isystem is needed, I believe that could just be added to GCC's specsfile during the toolchain adjustment, as LFS does. And of course, if this stuff is needed/useful, it should also be on the rest of the packages between Glibc and GCC, even if only for consistency. Looking for thoughts from anyone who knows more than I do about how the toolchain all works together..." chris@… Active Tickets 1040 Use /usr/libexec in accordance with FHS 3.0 BOOK task clfs-commits@… new 2015-06-13T20:34:08-07:00 2015-10-01T02:38:51-07:00 The recently-released FHS 3.0 recommends the usage of /usr/libexec for packages' internal executables, rather than a subdirectory in /usr/lib. chris@… Active Tickets 1097 Expect may not need --with-tcl options BOOK task clfs-commits@… new 2015-12-23T05:30:49-08:00 2015-12-28T05:59:48-08:00 "Expect seems to be able to find Tcl on its own, without needing the --with-tcl or --with-tclinclude options. I've tested this on x86 Pure64, please verify for other archs. {{{ root:~/build/expect5.45# ./configure --prefix=/tools checking for correct TEA configuration... ok (TEA 3.9) configure: configuring expect 5.45 checking for Tcl configuration... found /tools/lib/tclConfig.sh checking for existence of /tools/lib/tclConfig.sh... loading configure: --exec-prefix defaulting to TCL_EXEC_PREFIX /tools ... checking for Tcl public headers... /tools/include checking for Tcl private include files... private headers found with public headers ... checking for tclsh... /tools/bin/tclsh8.6 ... checking for Tcl configuration... found /tools/lib/tclConfig.sh checking for existence of /tools/lib/tclConfig.sh... loading ... }}} " chris@… Active Tickets 1111 Reorganization of final-system Glibc page BOOK task clfs-commits@… new 2016-03-07T03:56:49-08:00 2016-03-07T04:03:24-08:00 "I have a patch where I've edited and rearranged a number of elements of the final-system Glibc page. A summary of the changes: * I don't see any need for the Note about iconv(). I think it's one of those things where anyone who cares will probably already know and won't need to be told about it by CLFS. * I think the paragraph about how Glibc will build fine despite the toolchain pointing to /tools should be made a Note, and should be at the top. * I put together the commands for nscd config file and systemd support files into one block. * I think the sentence describing installation of locales could be slightly reworded. The current phrasing, specifying that the locales were not installed by the ""make install"", seems to imply that the user should expect them to be, and I don't believe a new builder would. * I believe timezone config should be in the i18n section instead of just ""Configuring"". It does relate to the user's localization settings so I think it fits better there. Any further thoughts on these modifications?" chris@… Active Tickets 853 Intel x86 Build BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2011-07-13T10:35:00-07:00 2014-06-13T20:02:37-07:00 Complete a build for the x86 architecture without errors. Jonathan Active Tickets 854 PowerPC 32 Build BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2011-07-13T10:35:57-07:00 2014-07-01T05:27:57-07:00 "Complete a build for the PowerPC 32 architecture without errors. " Jonathan Active Tickets 855 Mips 32 Build BOOK CLFS Standard GIT CLFS Standard 3.1.0 task Jonathan new 2011-07-13T10:36:49-07:00 2013-10-10T11:28:45-07:00 "Complete a build for the Mips 32 architecture without errors. Either Little or Big Enden " Jonathan Active Tickets 856 SPARC V8 Build BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2011-07-13T10:37:29-07:00 2013-10-10T11:28:54-07:00 "Complete a build for the SPARC v8 architecture without errors. " Jonathan Active Tickets 858 Mips pure 64 build BOOK CLFS Standard GIT CLFS Standard 3.1.0 task Jonathan new 2011-07-13T10:39:02-07:00 2013-10-10T11:29:03-07:00 "Complete a build for the Mips 64 architecture without errors. " Jonathan Active Tickets 859 UltraSPARC pure 64 build BOOK CLFS Standard GIT CLFS Standard 3.1.0 task Jonathan new 2011-07-13T10:39:49-07:00 2013-10-17T05:37:06-07:00 "Complete a build for the UltraSPARC pure 64 architecture without errors. " Jonathan Active Tickets 860 x86-64 Multilib build BOOK CLFS Standard GIT CLFS Standard 3.1.0 task Jonathan new 2011-07-13T10:40:39-07:00 2014-06-28T17:33:02-07:00 "Complete a build for the X86_64 Multilib architecture without errors. " Jonathan Active Tickets 861 Mips Multilib Build BOOK CLFS Standard GIT CLFS Standard 3.1.0 task Jonathan new 2011-07-13T10:41:17-07:00 2013-10-10T11:29:26-07:00 "Complete a build for the Mips Multilib architecture without errors. " Jonathan Active Tickets 862 UltraSPARC Multilib Build BOOK CLFS Standard GIT CLFS Standard 3.1.0 task Jonathan new 2011-07-13T10:42:04-07:00 2013-10-10T11:29:34-07:00 "Complete a build for the UltraSPARC Multilib architecture without errors. " Jonathan Active Tickets 916 x86-64 pure 64 build BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2013-01-19T13:36:11-08:00 2014-06-15T11:58:31-07:00 Complete a build without errors for the x86_64 pure 64 architecture without errors. William Harrington Active Tickets 932 PowerPC Pure64 build BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2013-10-10T11:32:55-07:00 2013-10-10T11:35:12-07:00 Complete a build for the PowerPC pure64 architecture without errors. William Harrington Active Tickets 933 PowerPC 64 Multilib Build BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2013-10-10T11:33:36-07:00 2013-10-10T11:35:23-07:00 Complete a build for the PowerPC 64 multilib architecture without errors. William Harrington Active Tickets 1061 Bash Testsuite BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2015-08-22T15:45:58-07:00 2019-03-01T07:36:24-08:00 "I notice LFS is changing ownership to the user '''nobody''' during the tests. {{{ chown -Rv nobody . }}} {{{ su nobody -s /bin/bash -c ""PATH=$PATH make tests"" }}} I need to find out why it was changed from just: {{{ make tests }}}" William Harrington Active Tickets 1063 GPERF Test Suite Failures BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2015-08-22T22:02:14-07:00 2015-08-22T22:02:14-07:00 We need to add '''-j 1''' to Gperf's test suite. Failures occur if the user use multiple jobs. William Harrington Active Tickets 1070 Grub2 2.02-beta2 update BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2015-08-27T17:36:38-07:00 2017-06-02T03:32:04-07:00 "Update to grub2.02. Also a crud of upstream patches. [http://alpha.gnu.org/gnu/grub/grub-2.02~beta2.tar.xz] Grub deals with the deprecated gets() fine now. Can remove: {{{ sed -i -e '/gets is a/d' grub-core/gnulib/stdio.in.h }}}" William Harrington Active Tickets 1071 SPARC64 and GCCTARGET variable BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2015-09-04T06:41:02-07:00 2015-09-04T06:41:02-07:00 "Right now, the book has the GCCTARGET in variables, which this is not needed for Sparc 32 bit or Sparc64 pure 64bit. It is required for Sparc64 multilib for cross-tools gcc and temp-system gcc-static. Review Jim Gifford's last edit for GCCTARGET regarding sparc64 somewhere in the history of final-preps variables.xml and cross-tools sparc64 gcc-static and temp-system sparc64 gcc. GCCTARGET is used during make all with CFLAGS_FOR_TARGET Then in 2009 Joe came back and made it used with only sparc and took it out of sparc64 multilib cross-tools gcc-static. So when do we need GCCTARGET when building gcc in cross-tools and temp-system?" William Harrington Active Tickets 1126 Boot-scripts init.d script remove command BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2017-04-14T13:53:27-07:00 2017-04-14T13:53:27-07:00 "Newer kernels should not need this command anymore: {{{ echo """" > /sys/kernel/uevent_helper }}} We need to double check and if found true, remove this from our init.d/udev bootscript." William Harrington Active Tickets 1146 Ncurses 6.0 Ch 10 audit BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2017-04-14T17:49:42-07:00 2017-04-14T17:49:42-07:00 "== Rollup patch level == [ftp://invisible-island.net/ncurses/6.0/ncurses-6.0-20170408.patch.gz] Weekly rollups can be found: [ftp://invisible-island.net/ncurses/6.0/] == Ch 10 == * Check installed libraries * Install docs {{{ mkdir -v /usr/share/doc/ncurses-6.0 cp -v -R doc/* /usr/share/doc/ncurses-6.0 }}} " William Harrington Active Tickets 1161 GCC configure options for checking and libstdcxxtime BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2017-05-12T14:10:09-07:00 2017-05-12T14:10:09-07:00 "Do we want --enable-checking=release and --enable-libstdcxx-time? {{{ --enable-libstdcxx-time=OPTION }}} Enables link-type checks for the availability of the clock_gettime clocks, used in the implementation of [time.clock], and of the nanosleep and sched_yield functions, used in the implementation of [thread.thread.this] of the 2011 ISO C++ standard. The choice OPTION=yes checks for the availability of the facilities in libc and libposix4. In case it's needed the latter is also linked to libstdc++ as part of the build process. OPTION=rt also searches (and, if needed, links) librt. Note that the latter is not always desirable because, in glibc, for example, in turn it triggers the linking of libpthread too, which activates locking, a large overhead for single-thread programs. OPTION=no skips the tests completely. The default is OPTION=auto, which skips the checks and enables the features only for targets known to support them. {{{ --enable-checking --enable-checking=list }}} When you specify this option, the compiler is built to perform internal consistency checks of the requested complexity. This does not change the generated code, but adds error checking within the compiler. This will slow down the compiler and may only work properly if you are building the compiler with GCC. This is ‘yes,extra’ by default when building from SVN or snapshots, but ‘release’ for releases. The default for building the stage1 compiler is ‘yes’. More control over the checks may be had by specifying list. The categories of checks available are ‘yes’ (most common checks ‘assert,misc,tree,gc,rtlflag,runtime’), ‘no’ (no checks at all), ‘all’ (all but ‘valgrind’), ‘release’ (cheapest checks ‘assert,runtime’) or ‘none’ (same as ‘no’). Individual checks can be enabled with these flags ‘assert’, ‘df’, ‘fold’, ‘gc’, ‘gcac’, ‘misc’, ‘rtl’, ‘rtlflag’, ‘runtime’, ‘tree’, ‘extra’ and ‘valgrind’. ‘extra’ adds for ‘misc’ checking extra checks that might affect code generation and should therefore not differ between stage1 and later stages. The ‘valgrind’ check requires the external valgrind simulator, available from http://valgrind.org/. The ‘df’, ‘rtl’, ‘gcac’ and ‘valgrind’ checks are very expensive. To disable all checking, ‘--disable-checking’ or ‘--enable-checking=none’ must be explicitly requested. Disabling assertions will make the compiler and runtime slightly faster but increase the risk of undetected internal errors causing wrong code to be generated. --disable-stage1-checking --enable-stage1-checking --enable-stage1-checking=list If no --enable-checking option is specified the stage1 compiler will be built with ‘yes’ checking enabled, otherwise the stage1 checking flags are the same as specified by --enable-checking. To build the stage1 compiler with different checking options use --enable-stage1-checking. The list of checking options is the same as for --enable-checking. If your system is too slow or too small to bootstrap a released compiler with checking for stage1 enabled, you can use ‘--disable-stage1-checking’ to disable checking for the stage1 compiler." William Harrington Active Tickets 1164 IPutils s20161105 BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2017-06-02T21:19:48-07:00 2017-08-25T12:24:51-07:00 "Some added USE variables: {{{ make USE_IDN=no USE_NETTLE=no USE_CRYPTO=no TARGETS=""clockdiff ping rdisc tracepath traceroute6"" }}} Could create a patch like our current version. Nothing from the previous IPutils version needs to apply. Generate man-pages and create patch. Add attr, acl, libcap to sysvinit book." William Harrington Active Tickets 1165 Systemd timesyncd commands removal BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2017-06-03T12:40:00-07:00 2017-06-03T12:40:00-07:00 "We do not need to create the user and group for timesyncd. The service is also already enabled. If anything, it'd need to be disabled if not wanted." William Harrington Active Tickets 1166 Util-linux 2.30 Upgrade BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2017-06-03T13:56:05-07:00 2017-08-02T11:49:01-07:00 "Upgrade to Util-linux 2.30 ChangeLog: [https://www.kernel.org/pub/linux/utils/util-linux/v2.30/v2.30-ChangeLog] Download: [https://www.kernel.org/pub/linux/utils/util-linux/v2.30/util-linux-2.30.tar.xz] MD5: eaa3429150268027908a1b8ae6ee9a62 Size: 4.478 KB" William Harrington Active Tickets 1176 Upgrade Automake to 1.15.1 BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2017-08-02T11:18:41-07:00 2017-08-02T11:18:41-07:00 "URL: [http://ftp.gnu.org/gnu/automake/automake-1.15.1.tar.xz] MD5: 24cd3501b6ad8cd4d7e2546f07e8b4d4 Size: 1,509 KB Release: [https://lists.gnu.org/archive/html/info-gnu/2017-06/msg00007.html]" William Harrington Active Tickets 1177 Upgrade Binutils to 2.28.1 BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2017-08-02T11:23:49-07:00 2017-08-02T11:23:49-07:00 "URL: [http://ftp.gnu.org/gnu/binutils/binutils-2.28.1.tar.xz] MD5: a3bf359889e4b299fce1f4cb919dc7b6 Size: 19,440 KB" William Harrington Active Tickets 1178 Upgrade GREP to 3.1 BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2017-08-02T11:27:06-07:00 2017-08-02T11:27:06-07:00 "URL: [http://ftp.gnu.org/gnu/grep/grep-3.1.tar.xz] MD5: feca7b3e7c7f4aab2b42ecbfc513b070 Size: 1,371 KB Release: [https://lists.gnu.org/archive/html/info-gnu/2017-07/msg00000.html]" William Harrington Active Tickets 1179 Upgrade Linux Kernel Sublevel patch to 40 BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2017-08-02T11:32:52-07:00 2017-08-02T11:32:52-07:00 "URL: [https://www.kernel.org/pub/linux/kernel/v4.x/patch-4.9.40.xz] MD5: a03f01580773606fb6e8669c8f2e0088 Size: 856 KB" William Harrington Active Tickets 1181 Upgrade PSMISC to 23.1 BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2017-08-02T11:43:25-07:00 2017-08-02T11:43:25-07:00 "URL: [https://sourceforge.net/projects/psmisc/files/psmisc/psmisc-23.1.tar.xz] MD5: bbba1f701c02fb50d59540d1ff90d8d1 Size: 296 KB Changes in 23.1 * killall: Remove debug output * docs: distribute pslog.1 * docs: distribute README.md Changes in 23.0 * killall: Fix process age race condition !11 * killall: only use scope of current namespace by default * killall: better parsing of command names * pstree: add cgroup namespaces !10 * peekfd: Help give long options too !5 * killall: correctly report when 32+ procs match !8 * prtat: Remove duplicate options in man page !3 * pslog: New program to check what logs are open merge !1 * fuser: increase path length for octal escaping * Make usage of linked lists of devices found in /proc/self/mountinfo optional * Make timeout() in timeout.c work with shared mmap to reduce the load due write/read the stat buffers * Add list_sort() to lists.h to be able to sort lists * fuser: Fixed typo for -M flag. * pstree: by default doesn't show threadnames, use -t to show as it disables compaction. * pstree: Removed need for PATH_MAX * pstree: ignores disappeared processes. * killall: -o and -y work with -r flags. * m4/gettext.m4: Upgrade to gettext-0.19.4. * m4/iconv.m4: Upgrade to gettext-0.19.4. * m4/lib-ld.m4: Upgrade to gettext-0.19.4. * m4/lib-link.m4: Upgrade to gettext-0.19.4. * m4/lib-prefix.m4: Upgrade to gettext-0.19.4. * m4/nls.m4: Upgrade to gettext-0.19.4. * m4/po.m4: Upgrade to gettext-0.19.4. * m4/progtest.m4: Upgrade to gettext-0.19.4. * configure.ac (AM_GNU_GETTEXT_VERSION): Bump to 0.19.4." William Harrington Active Tickets 1182 Upgrade Texinfo to 6.4 BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2017-08-02T11:45:34-07:00 2017-08-02T11:45:34-07:00 "URL: [http://ftp.gnu.org/gnu/texinfo/texinfo-6.4.tar.xz] MD5: 2a676c8339efe6ddea0f1cb52e626d15 Size: 4,498 KB" William Harrington Active Tickets 101 Glibc locale installation text needs to be rewritten BOOK CLFS Sysroot GIT CLFS Sysroot 1.0.0 task clfs-commits@… new 2007-01-04T20:52:56-08:00 2011-01-09T07:50:08-08:00 "I am not sure exactly how much needs to be changed for the locale installation in the Glibc build, but at minimum the descriptive text is wrong. It mentions that the minimal list of locales is what's needed for various testsuites to pass, but this certainly does not apply to a sysroot build. I don't really know exactly what else would have to be changed there and what (if any) locales would still be required (beyond just ""install the one you use"")." chris@… Active Tickets 102 Build order/hard-coded binary pathname issues... BOOK CLFS Sysroot GIT CLFS Sysroot 1.0.0 task clfs-commits@… new 2007-01-26T23:27:01-08:00 2011-01-09T07:50:13-08:00 "Many packages end up with hard-coded names to various binaries, which has usually been solved by either switching the build order around when needed to ensure that certain programs were in specific places when they were found by other packages, or creating symlinks to some binaries. However, this is essentially a completely different issue for the sysroot book since it will be looking at program locations on the host system. In a way it simplifies things because the build order is less important, but it can also cause problems when programs on the host system that packages hard-code paths to are not in the same place on the host system as on the final CLFS system. The most obvious way to solve this is to create symlinks on the host to any such programs, but we do want to avoid touching the host whenever possible. Does anyone else have ideas for any other potential solution?" chris@… Active Tickets 219 Update Package Contents in Sysroot BOOK CLFS Sysroot GIT CLFS Sysroot 1.0.0 task Joe Ciccone assigned 2009-07-11T13:25:50-07:00 2011-01-09T07:50:21-08:00 Copy the package contents from the main branch into clfs-sysroot. Joe Ciccone Active Tickets 221 Fix rpath for GMP/MPFR/PPL/CLooG-PPL BOOK CLFS Sysroot GIT CLFS Sysroot 1.0.0 task Joe Ciccone assigned 2009-07-12T08:36:32-07:00 2011-01-09T07:50:31-08:00 Its leaving $CLFS in the path after the sysroot build is complete. Need to clean up. Joe Ciccone Active Tickets 1016 Eudev 3.2.1 Version Upgrade BOOK CLFS Standard SYSVINIT CLFS Sysvinit 3.1.0 task clfs-commits@… new 2014-11-05T12:18:02-08:00 2017-04-14T12:33:35-07:00 "http://dev.gentoo.org/~blueness/eudev/eudev-3.2.1.tar.gz There may be need for fixing a test script: {{{ sed -r -i 's|/usr(/bin/test)|\1|' test/udev-test.pl }}} We need to check the sysvinit book for proper test binary location, as well as mentioned in [[ticket:989|Eudev 3.1.5 Version Upgrade for SYSTEMD Book]]" William Harrington Active Tickets 1116 Add Gperf 3.0.4 BOOK CLFS Standard SYSVINIT CLFS Sysvinit 3.1.0 task clfs-commits@… new 2016-07-15T21:45:18-07:00 2017-08-02T11:11:49-07:00 "Latest Eudev needs GPerf. FTP:[ftp://ftp.gnu.org/gnu/gperf/gperf-3.0.4.tar.gz] MD5SUM: c1f1db32fb6598d6a93e6e88796a8632" William Harrington Active Tickets 1180 Upgrade Libpipeline to 1.4.2 BOOK CLFS Standard SYSVINIT CLFS Sysvinit 3.1.0 task clfs-commits@… new 2017-08-02T11:36:32-07:00 2017-08-02T11:36:32-07:00 "URL: [http://download.savannah.gnu.org/releases/libpipeline/libpipeline-1.4.2.tar.gz] MD5: d5c80387eb9c9e5d089da2a06e8a6b12 Size: 827 KB" William Harrington Active Tickets 956 Do away with /tools and /cross-tools symlinks BOOK task clfs-commits@… new 2014-05-07T18:41:23-07:00 2015-12-16T04:32:34-08:00 "I believe there's no real reason to keep the /tools and /cross-tools symlinks, so they can just be disposed of. We already use --with-sysroot in the cross-toolchain, so we may as well take full advantage of it. This would also allow for building the whole temp-system without the need for any root access, or testing multiple builds at once with jhalfs or similar tools. Plus, I believe in general that reducing host system modifications is a good thing. I've managed to get almost a complete temp system this way, so far. Of course I modified most references to ""/tools"" and ""/cross-tools"" in the cross-tools, temp-system and boot sections by prefixing them with ""${CLFS}"", and added ""DESTDIR=${CLFS}"" to the installation commands for temp-system/boot packages. Aside from those, the only changes I've had to make was adding ""--with-sysroot"" to temp-system MPFR and MPC, due to libtool problems (not finding libgmp.la), and modifying Util-linux instructions so that it links to libpthread, as it does not do so automatically. I've hit similar problems with boot/e2fsprogs, as apparently dependent libraries are not being automatically found. If something links to librt, it will be located in ${CLFS}/tools, but then libpthread, which librt links to, will not. However, it works fine if both are explicitly specified. In other words, if I create a dummy.c with ""main(){}""... {{{ ${CC} dummy.c -lrt }}} fails due to not finding libpthread, but {{{ ${CC} dummy.c -lrt -libpthread }}} works just fine. I've been trying to find the cause for this, with assistance from William Harrington in the chat room, who suspects there might be a problem with -rpath or something similar. Compiling Glibc with LDFLAGS=""-Wl,rpath,/tools/lib:${CLFS}/tools/lib"" does not seem to help, though. I have a patch with the changes I've made so far on my local copy - http://www.clfs.org/~chris/no-tools.patch" chris@… Active Tickets 1110 Obsolete rpc headers not needed in /tools BOOK task clfs-commits@… new 2016-02-26T14:23:58-08:00 2016-02-26T14:23:58-08:00 "For Glibc, we currently use --enable-obsolete-rpc in both Cross Tools and Final System, to install several rpc header for some packages that might need them. I'm testing a build now without this option, and it appears it is not needed, at least in /tools and possibly not in the final-system as well. There might be some packages that I don't use that still need those headers, but I haven't seen any problems from leaving them out. In any case, --enable-obsolete-rpc does not appear to be needed for the Cross Tools build of Glibc. It's possible that it might not be needed in the Final System either, but I think it could be left there for now." chris@… Active Tickets 1058 Bzip2 Install Docs BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2015-08-22T07:53:15-07:00 2015-08-22T07:53:15-07:00 "LFS installs bzip2 docs with the patch. We may want to include that with our builds. [http://www.linuxfromscratch.org/patches/lfs/systemd/bzip2-1.0.6-install_docs-1.patch] The install performs the following: {{{ DOCDIR=share/doc/$(DISTNAME) if ( test ! -d $(PREFIX)/$(DOCDIR) ) ; then mkdir -p $(PREFIX)/$(DOCDIR); fi cp -f manual.html $(PREFIX)/$(DOCDIR) cp -f manual.pdf $(PREFIX)/$(DOCDIR) cp -f manual.ps $(PREFIX)/$(DOCDIR) cp -f bzip2.txt $(PREFIX)/$(DOCDIR) chmod a+r $(PREFIX)/$(DOCDIR)/manual.html chmod a+r $(PREFIX)/$(DOCDIR)/manual.pdf chmod a+r $(PREFIX)/$(DOCDIR)/manual.ps chmod a+r $(PREFIX)/$(DOCDIR)/bzip2.txt }}}" William Harrington Active Tickets 1060 Readline 7.0 Install Documentation BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2015-08-22T08:56:26-07:00 2017-04-14T12:45:09-07:00 "Our current instruction installs: CHANGES INSTALL README history.html readline.html rluseman.html The following installs the rest of the documentation: {{{ install -v -m644 doc/*.{ps,pdf,html,dvi} /usr/share/doc/readline-7.0 }}} This will install the following documentation: {{{ history.dvi history.ps readline.html readline_3.ps rluserman.pdf history.html history_3.ps readline.pdf rluserman.dvi rluserman.ps history.pdf readline.dvi readline.ps rluserman.html }}}" William Harrington Active Tickets 1062 Findutils 4.6.0 find location BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2015-08-22T18:10:48-07:00 2017-04-14T19:13:45-07:00 "Do we need to move find to /bin? LFS states that some packages in BLFS and beyond expect find to be in /bin. Has anyone ran across any packages which require that? {{{ mv -v /usr/bin/find /bin sed -i 's|find:=${BINDIR}|find:=/bin|' /usr/bin/updatedb }}}" William Harrington Active Tickets 1078 M4 test-update-copyright.sh test failure BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2015-09-25T17:55:14-07:00 2015-11-22T20:33:39-08:00 "[http://lists.linuxfromscratch.org/pipermail/lfs-dev/2015-June/070225.html] Perl 5.22 issue." William Harrington Active Tickets 1079 Final System Bison 3 tests fail BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2015-09-25T18:32:15-07:00 2015-12-25T09:36:14-08:00 "Our description states 3 tests will fail. Which 3 tests? If more than 3 tests fail how will someone know if the 3 tests are with it? Here is the list of tests which fail: {{{ Executing test test-bison-yylloc Test test-bison-yylloc FAILED. See test-bison-yylloc/OUTPUT for details. Executing test test-bison-yylval Test test-bison-yylval FAILED. See test-bison-yylval/OUTPUT for details. Executing test test-bison-nr Test test-bison-nr FAILED. See test-bison-nr/OUTPUT for details. }}}" William Harrington Active Tickets 1147 Move to Pkg-config from Pkg-config-lite BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2017-04-14T17:59:19-07:00 2017-04-14T17:59:19-07:00 "Do we want to move to Pkg-config? We only used pkg-config-lite because of the minimal install for our base system. It has been known for a while that regular pkg-config can build without any deps from glib." William Harrington Active Tickets 66 Gettext finds Java on my system BOOK CLFS Sysroot GIT CLFS Sysroot 1.0.0 task clfs-commits@… new 2006-07-21T21:43:15-07:00 2011-01-09T07:50:37-08:00 In the past, LFS tried hard to ensure that Chapter 5 Gettext doesn't find Java or C# compilers on the host. In CLFS-2.0, Gettext finds Java on my host. alexander@… Active Tickets 1092 Current status for /etc/mtab BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2015-11-30T12:30:09-08:00 2016-02-21T10:44:56-08:00 "'''''/etc/mtab''''' is no longer being referenced in '''Util-linux-2.27.1''', '''Systemd-228''', and future '''GLIBC''' releases. So far it has been appropriate to have /etc/mtab as a link to /proc/self/mounts. We should be able to remove the creation of mtab altogether in the future. Refer to the following for more information: [https://github.com/systemd/systemd/issues/1495] [https://sourceware.org/bugzilla/show_bug.cgi?id=19108] [https://sourceware.org/ml/libc-alpha/2015-11/msg00418.html] [https://github.com/karelzak/util-linux/commit/02501746] [https://github.com/systemd/systemd/blob/v228/NEWS] [https://github.com/systemd/systemd/blob/v228/README]" William Harrington Active Tickets 1095 Host System Requirements Check BOOK CLFS Standard GIT CLFS Standard 3.1.0 task clfs-commits@… new 2015-12-18T10:56:52-08:00 2015-12-18T16:18:04-08:00 "Chris mentioned the notion to check our Host System Requirements section. I have been testing with the updated lfs livecd which is based on 6.3 with my own updates: [http://clfs.org/~kb0iic/livecdupd/] So far GCC needs to be upgraded to at least 4.2.3 because of newer Binutils and GOLD Linker requirements and the heavy use of templates. A 4.1.2 compiler can only work if using the 4.1.3 prerelease fixes (4.1.3 was never released). This is why I say GCC should at least use 4.2.3 as of now. Same issue with 4.2.0, 4.2.1, and 4.2.2. 4.2.3 is fine. ISL 0.15 has a header issue at ctx.h line 83 which causes issues with older compilers regarding C99 and a comma existing at the end of an enumeration. During ISL cross-tools install: {{{ cp -v include/isl/ctx.h{,.orig} sed '83/s/.$//' include/isl/ctx.h.orig > include/isl/ctx.h }}} https://groups.google.com/d/msg/isl-development/olvv6MU_3gE/VMf8qzZLCQAJ Editing the header in ISL for cross-tools is best no matter which host compiler version is used. Chris found some issues, but I'm still at gcc-static right now." William Harrington Active Tickets 1098 Remove extra /lib64, /libx32 references from ldconfig for Pure64 BOOK CLFS Standard 3.1.0 task clfs-commits@… new 2015-12-27T14:35:25-08:00 2017-06-02T19:46:50-07:00 "ldconfig still has references to other dynamic linker locations in /lib64 and /libx32. These can be removed with this command: {{{ sed -e 's/lib64/lib/' -e '/ld-linux\(-x32\|\.so\)/d' -i sysdeps/unix/sysv/linux/x86_64/ldconfig.h }}} Also it add those dirs to its default search locations, with these lines: {{{ add_system_dir (SLIBDIR); if (strcmp (SLIBDIR, LIBDIR)) add_system_dir (LIBDIR); }}} Apparently LIBDIR and/or SLIBDIR is set elsewhere, but I couldn't find where. I'd prefer if we can find where it's getting those from, but if not, those lines can just be removed: {{{ sed -e '/LIBDIR/d' -i elf/ldconfig.c }}}" chris@… Active Tickets 1082 GMP Cross-tools note for ABI=32 BOOK CLFS Sysvinit 3.1.0 task clfs-commits@… new 2015-10-16T18:18:45-07:00 2015-10-16T18:18:45-07:00 "GMP has matured and the configure system configures the ABI properly. If anyone has any issues put it in this ticket. The note will be removed as it doesn't seem to be required anymore. I haven't had this problem in many builds. This note was added in a very early version of GMP. I haven't witnessed an issue sinced GMP version 5 or 6. It may have been a problem in 3 and 4. Of course, the ABI variable still does the job. If the note is no longer needed we should remove it." William Harrington