util-linux: split uuid in separate recipe to allow bootstrapping
PATCH 1/1 util-linux: split uuid in separate recipe to allow bootstrapping
From: Luca Boccassi <[email protected]>
In the next release util-linux will gain an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory). Split out util-linux-uuid in a different recipe to break the cycle.
https://github.com/karelzak/util-linux/pull/898
---
util-linux 2.35 is not out yet, but I'd like to get the preparatory work underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
.../util-linux/util-linux-uuid_2.34.bb | 35 +++++++++++++++++++ meta/recipes-core/util-linux/util-linux.inc | 6 ++++ meta/recipes-extended/wget/wget.inc | 2 +- 3 files changed, 42 insertions(+), 1 deletion(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.34.bb
meta/recipes-core/util-linux/util-linux-uuid_2.34.bb new file +35 -0
@@ -0,0 +1,35 @@ +SUMMARY = "Builds libuuid separately from util-linux for bootstrapping" +HOMEPAGE = "http://userweb.kernel.org/~kzak/util-linux/" + +SECTION = "base" + +LICENSE = "GPLv2+ & LGPLv2.1+ & BSD-3-Clause & BSD-4-Clause" + +LIC_FILES_CHKSUM = "file://README.licensing;md5=972a134f1e14b2b060e365df2fab0099 \ + file://COPYING;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ + file://Documentation/licenses/COPYING.GPL-2.0-or-later;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ + file://Documentation/licenses/COPYING.LGPL-2.1-or-later;md5=4fbd65380cdd255951079008b364516c \ + file://Documentation/licenses/COPYING.BSD-3-Clause;md5=58dcd8452651fc8b07d1f65ce07ca8af \ + file://Documentation/licenses/COPYING.BSD-4-Clause-UC;md5=263860f8968d8bafa5392cab74285262 \ + file://libuuid/COPYING;md5=6d2cafc999feb2c2de84d4d24b23290c \ + file://libmount/COPYING;md5=7c7e39fb7d70ffe5d693a643e29987c2 \ + file://libblkid/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04" + +MAJOR_VERSION = "${@'.'.join(d.getVar('PV').split('.')[0:2])}" +SRC_URI = "${KERNELORG_MIRROR}/linux/utils/util-linux/v${MAJOR_VERSION}/util-linux-${MAJOR_VERSION}.tar.xz \ + " + +SRC_URI[md5sum] = "a78cbeaed9c39094b96a48ba8f891d50" +SRC_URI[sha256sum] = "743f9d0c7252b6db246b659c1e1ce0bd45d8d4508b4dfa427bbb4a3e9b9f62b5" + +inherit autotools gettext pkgconfig + +S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +RPROVIDES_libuuid-dev = "util-linux-libuuid-dev" +RPROVIDES_libuuid = "util-linux-libuuid" + +do_install_append() { + rmdir ${D}${bindir} ${D}${sbindir} + rm -rf ${D}${datadir} +}
meta/recipes-core/util-linux/util-linux.inc +6 -0
@@ -288,6 +288,12 @@ ALTERNATIVE_LINK_NAME[utmpdump.1] = "${mandir}/man1/utmpdump.1" ALTERNATIVE_LINK_NAME[uuid.3] = "${mandir}/man3/uuid.3" ALTERNATIVE_LINK_NAME[wall.1] = "${mandir}/man1/wall.1" +# split uuid out to allow in the near future a build-dependency on packages depending on it +RPROVIDES_remove_class-target_${PN}-dev = "util-linux-libuuid-dev" +PACKAGES_remove_class-target = "libuuid1 util-linux-libuuid1" +DEPENDS_append_class-target = " util-linux-uuid" +EXTRA_OECONF_remove_class-target = "--enable-libuuid" +EXTRA_OECONF_append_class-target = " --disable-libuuid" BBCLASSEXTEND = "native nativesdk"
meta/recipes-extended/wget/wget.inc +1 -1
@@ -32,7 +32,7 @@ PACKAGECONFIG[gnutls] = "--with-ssl=gnutls,,gnutls" PACKAGECONFIG[ipv6] = "--enable-ipv6,--disable-ipv6," PACKAGECONFIG[iri] = "--enable-iri,--disable-iri,libidn2" PACKAGECONFIG[libpsl] = "--with-libpsl,--without-libpsl,libpsl" -PACKAGECONFIG[libuuid] = "--with-libuuid,--without-libuuid,util-linux" +PACKAGECONFIG[libuuid] = "--with-libuuid,--without-libuuid,util-linux-uuid" PACKAGECONFIG[openssl] = "--with-ssl=openssl,,openssl" PACKAGECONFIG[pcre] = "--enable-pcre,--disable-pcre,libpcre" PACKAGECONFIG[zlib] = "--with-zlib,--without-zlib,zlib"
Signature · 1 line
2.20.1
Quoted 6 lines · On 05/12/2019 20:13, [email protected] wrote:
From: Luca Boccassi <[email protected]>
In the next release util-linux will gain an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory). Split out util-linux-uuid in a different recipe to break the cycle.
<sigh>
At least please simply make util-linux-uuid just require util-linux and make the configuration changes required (presumably, disabling all tools but uuid).
Ross
PATCH v2 1/1 util-linux: split uuid in separate recipe to allow bootstrapping
In reply to [email protected]From: Luca Boccassi <[email protected]>
In the next release util-linux will gain an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory). Split out util-linux-uuid in a different recipe to break the cycle.
https://github.com/karelzak/util-linux/pull/898
---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
.../util-linux/util-linux-uuid_2.34.bb | 40 +++++++++++++++++++ meta/recipes-core/util-linux/util-linux.inc | 11 +++++ meta/recipes-extended/wget/wget.inc | 2 +- 3 files changed, 52 insertions(+), 1 deletion(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.34.bb
meta/recipes-core/util-linux/util-linux-uuid_2.34.bb new file +40 -0
@@ -0,0 +1,40 @@ +SUMMARY = "Builds libuuid separately from util-linux for bootstrapping" +HOMEPAGE = "http://userweb.kernel.org/~kzak/util-linux/" + +SECTION = "base" + +LICENSE = "GPLv2+ & LGPLv2.1+ & BSD-3-Clause & BSD-4-Clause" + +LIC_FILES_CHKSUM = "file://README.licensing;md5=972a134f1e14b2b060e365df2fab0099 \ + file://COPYING;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ + file://Documentation/licenses/COPYING.GPL-2.0-or-later;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ + file://Documentation/licenses/COPYING.LGPL-2.1-or-later;md5=4fbd65380cdd255951079008b364516c \ + file://Documentation/licenses/COPYING.BSD-3-Clause;md5=58dcd8452651fc8b07d1f65ce07ca8af \ + file://Documentation/licenses/COPYING.BSD-4-Clause-UC;md5=263860f8968d8bafa5392cab74285262 \ + file://libuuid/COPYING;md5=6d2cafc999feb2c2de84d4d24b23290c \ + file://libmount/COPYING;md5=7c7e39fb7d70ffe5d693a643e29987c2 \ + file://libblkid/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04" + +MAJOR_VERSION = "${@'.'.join(d.getVar('PV').split('.')[0:2])}" +SRC_URI = "${KERNELORG_MIRROR}/linux/utils/util-linux/v${MAJOR_VERSION}/util-linux-${MAJOR_VERSION}.tar.xz \ + " + +SRC_URI[md5sum] = "a78cbeaed9c39094b96a48ba8f891d50" +SRC_URI[sha256sum] = "743f9d0c7252b6db246b659c1e1ce0bd45d8d4508b4dfa427bbb4a3e9b9f62b5" + +inherit autotools gettext pkgconfig + +S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +DEBIAN_NOAUTONAME_util-linux-libuuid = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dev = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dbg = "1" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" + +do_install_append() { + rmdir ${D}${bindir} ${D}${sbindir} + rm -rf ${D}${datadir} +}
meta/recipes-core/util-linux/util-linux.inc +11 -0
@@ -288,6 +288,17 @@ ALTERNATIVE_LINK_NAME[utmpdump.1] = "${mandir}/man1/utmpdump.1" ALTERNATIVE_LINK_NAME[uuid.3] = "${mandir}/man3/uuid.3" ALTERNATIVE_LINK_NAME[wall.1] = "${mandir}/man1/wall.1" +# dm-verity support introduces a circular dependency, so util-linux-uuid is split out for target builds +# Need to build libuuid for uuidgen, but then delete it and let the other recipe ship it +RPROVIDES_remove_class-target_${PN}-dev = "util-linux-libuuid-dev" +PACKAGES_remove_class-target = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-dbg" +DEPENDS_append_class-target = " util-linux-uuid" +RDEPENDS_append_class-target_${PN}-dev = " util-linux-libuuid-dev" +RDEPENDS_append_class-target_${PN} = " util-linux-libuuid" +do_install_append_class-target() { + rm -rf ${D}${includedir}/uuid ${D}${libdir}/pkgconfig/uuid.pc ${D}${libdir}/libuuid* ${D}/lib/libuuid* +} + BBCLASSEXTEND = "native nativesdk"
meta/recipes-extended/wget/wget.inc +1 -1
@@ -32,7 +32,7 @@ PACKAGECONFIG[gnutls] = "--with-ssl=gnutls,,gnutls" PACKAGECONFIG[ipv6] = "--enable-ipv6,--disable-ipv6," PACKAGECONFIG[iri] = "--enable-iri,--disable-iri,libidn2" PACKAGECONFIG[libpsl] = "--with-libpsl,--without-libpsl,libpsl" -PACKAGECONFIG[libuuid] = "--with-libuuid,--without-libuuid,util-linux" +PACKAGECONFIG[libuuid] = "--with-libuuid,--without-libuuid,util-linux-uuid" PACKAGECONFIG[openssl] = "--with-ssl=openssl,,openssl" PACKAGECONFIG[pcre] = "--enable-pcre,--disable-pcre,libpcre" PACKAGECONFIG[zlib] = "--with-zlib,--without-zlib,zlib"
Signature · 1 line
2.20.1
PATCH v3 1/1 util-linux: split uuid in separate recipe to allow bootstrapping
From: Luca Boccassi <[email protected]>
In v2.35 util-linux gained an (optional) build
dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory). Split out util-linux-uuid in a different recipe to break the cycle.
Add a packageconfig switch (disabled by default) to allow using the new dependency.
https://github.com/karelzak/util-linux/pull/898
---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
.../nfs-utils/nfs-utils_2.5.1.bb | 2 +- meta/recipes-core/udev/eudev_3.2.9.bb | 2 +- .../util-linux/util-linux-uuid_2.36.bb | 22 +++++++++ meta/recipes-core/util-linux/util-linux.inc | 40 ++++++++++++++++ .../util-linux/util-linux_2.36.bb | 48 ++++--------------- .../btrfs-tools/btrfs-tools_5.7.bb | 2 +- meta/recipes-devtools/e2fsprogs/e2fsprogs.inc | 2 +- meta/recipes-devtools/fdisk/gptfdisk_1.0.5.bb | 2 +- meta/recipes-devtools/mtd/mtd-utils_git.bb | 2 +- meta/recipes-devtools/python/python3_3.9.0.bb | 2 +- .../syslinux/syslinux_6.04-pre2.bb | 2 +- .../tcf-agent/tcf-agent_git.bb | 2 +- meta/recipes-extended/mc/mc_4.8.25.bb | 2 +- meta/recipes-extended/parted/parted_3.3.bb | 2 +- meta/recipes-extended/wget/wget.inc | 2 +- .../fontconfig/fontconfig_2.13.1.bb | 2 +- meta/recipes-graphics/xorg-lib/libxt_1.2.0.bb | 2 +- .../recipes-kernel/lttng/babeltrace2_2.0.3.bb | 2 +- meta/recipes-kernel/lttng/babeltrace_1.5.8.bb | 2 +- .../lttng/lttng-tools_2.12.2.bb | 2 +- meta/recipes-kernel/lttng/lttng-ust_2.12.0.bb | 2 +- .../gstreamer1.0-plugins-base_1.18.0.bb | 2 +- meta/recipes-support/apr/apr_1.7.0.bb | 2 +- meta/recipes-support/serf/serf_1.3.9.bb | 2 +- 24 files changed, 92 insertions(+), 60 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.bb create mode 100644 meta/recipes-core/util-linux/util-linux.inc
meta/recipes-connectivity/nfs-utils/nfs-utils_2.5.1.bb +1 -1
@@ -8,7 +8,7 @@ LICENSE = "MIT & GPLv2+ & BSD" LIC_FILES_CHKSUM = "file://COPYING;md5=95f3a93a5c3c7888de623b46ea085a84" # util-linux for libblkid -DEPENDS = "libcap libevent util-linux sqlite3 libtirpc" +DEPENDS = "libcap libevent util-linux util-linux-uuid sqlite3 libtirpc" RDEPENDS_${PN} = "${PN}-client" RRECOMMENDS_${PN} = "kernel-module-nfsd"
meta/recipes-core/udev/eudev_3.2.9.bb +1 -1
@@ -4,7 +4,7 @@ LICENSE = "GPLv2.0+ & LGPL-2.1+" LICENSE_libudev = "LGPL-2.1+" LIC_FILES_CHKSUM = "file://COPYING;md5=751419260aa954499f7abaabaa882bbe" -DEPENDS = "glib-2.0 glib-2.0-native gperf-native kmod libxslt-native util-linux" +DEPENDS = "glib-2.0 glib-2.0-native gperf-native kmod libxslt-native util-linux util-linux-uuid" PROVIDES = "udev"
meta/recipes-core/util-linux/util-linux-uuid_2.36.bb new file +22 -0
@@ -0,0 +1,22 @@ +# To allow util-linux to optionally build-depend on cryptsetup, libuuid is +# split out of the main recipe, as it's needed by cryptsetup + +require util-linux.inc + +inherit autotools gettext pkgconfig + +S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +DEBIAN_NOAUTONAME_util-linux-libuuid = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dev = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dbg = "1" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" + +do_install_append() { + rm -rf ${D}${datadir} ${D}${bindir} ${D}${base_bindir} ${D}${sbindir} ${D}${base_sbindir} ${D}${exec_prefix}/sbin +} + +BBCLASSEXTEND = "native nativesdk"
meta/recipes-core/util-linux/util-linux.inc new file +40 -0
@@ -0,0 +1,40 @@ +SUMMARY = "A suite of basic system administration utilities" +HOMEPAGE = "https://en.wikipedia.org/wiki/Util-linux" +DESCRIPTION = "Util-linux includes a suite of basic system administration utilities \ +commonly found on most Linux systems. Some of the more important utilities include \ +disk partitioning, kernel message management, filesystem creation, and system login." + +SECTION = "base" + +LICENSE = "GPLv2+ & LGPLv2.1+ & BSD-3-Clause & BSD-4-Clause" +LICENSE_${PN}-libblkid = "LGPLv2.1+" +LICENSE_${PN}-libfdisk = "LGPLv2.1+" +LICENSE_${PN}-libmount = "LGPLv2.1+" +LICENSE_${PN}-libsmartcols = "LGPLv2.1+" +LICENSE_${PN}-libuuid = "BSD-3-Clause" + +LIC_FILES_CHKSUM = "file://README.licensing;md5=0fd5c050c6187d2bf0a4492b7f4e33da \ + file://COPYING;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ + file://Documentation/licenses/COPYING.GPL-2.0-or-later;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ + file://Documentation/licenses/COPYING.LGPL-2.1-or-later;md5=4fbd65380cdd255951079008b364516c \ + file://Documentation/licenses/COPYING.BSD-3-Clause;md5=58dcd8452651fc8b07d1f65ce07ca8af \ + file://Documentation/licenses/COPYING.BSD-4-Clause-UC;md5=263860f8968d8bafa5392cab74285262 \ + file://libuuid/COPYING;md5=6d2cafc999feb2c2de84d4d24b23290c \ + file://libmount/COPYING;md5=7c7e39fb7d70ffe5d693a643e29987c2 \ + file://libblkid/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ + file://libfdisk/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ + file://libsmartcols/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ +" + +MAJOR_VERSION = "${@'.'.join(d.getVar('PV').split('.')[0:2])}" +FILESEXTRAPATHS_prepend := "${THISDIR}/util-linux:" +SRC_URI = "${KERNELORG_MIRROR}/linux/utils/util-linux/v${MAJOR_VERSION}/util-linux-${MAJOR_VERSION}.tar.xz \ + file://configure-sbindir.patch \ + file://runuser.pamd \ + file://runuser-l.pamd \ + file://ptest.patch \ + file://run-ptest \ + file://display_testname_for_subtest.patch \ + file://avoid_parallel_tests.patch \ + " +SRC_URI[sha256sum] = "9e4b1c67eb13b9b67feb32ae1dc0d50e08ce9e5d82e1cccd0ee771ad2fa9e0b1"
meta/recipes-core/util-linux/util-linux_2.36.bb +9 -39
68 lines not on this page · raw message
meta/recipes-devtools/btrfs-tools/btrfs-tools_5.7.bb +1 -1
@@ -10,7 +10,7 @@ HOMEPAGE = "https://btrfs.wiki.kernel.org" LICENSE = "GPLv2" LIC_FILES_CHKSUM = "file://COPYING;md5=fcb02dc552a041dee27e4b85c7396067" SECTION = "base" -DEPENDS = "util-linux attr e2fsprogs lzo acl" +DEPENDS = "util-linux util-linux-uuid attr e2fsprogs lzo acl" DEPENDS_append_class-target = " udev" RDEPENDS_${PN} = "libgcc"
meta/recipes-devtools/e2fsprogs/e2fsprogs.inc +1 -1
@@ -17,7 +17,7 @@ LIC_FILES_CHKSUM = "file://NOTICE;md5=d50be0580c0b0a7fbc7a4830bbe6c12b \ file://lib/et/et_name.c;beginline=1;endline=11;md5=ead236447dac7b980dbc5b4804d8c836 \ file://lib/ss/ss.h;beginline=1;endline=20;md5=6e89ad47da6e75fecd2b5e0e81e1d4a6" SECTION = "base" -DEPENDS = "util-linux attr" +DEPENDS = "util-linux util-linux-uuid attr" SRC_URI = "git://git.kernel.org/pub/scm/fs/ext2/e2fsprogs.git" S = "${WORKDIR}/git"
meta/recipes-devtools/fdisk/gptfdisk_1.0.5.bb +1 -1
@@ -4,7 +4,7 @@ DESCRIPTION = "GPT fdisk is a disk partitioning tool loosely modeled on Linux fd LICENSE = "GPLv2" LIC_FILES_CHKSUM = "file://COPYING;md5=59530bdf33659b29e73d4adb9f9f6552" -DEPENDS = "util-linux" +DEPENDS = "util-linux util-linux-uuid" SRC_URI = "${SOURCEFORGE_MIRROR}/${BPN}/${PV}/${BP}.tar.gz \ file://0001-gptcurses-correctly-include-curses.h.patch \
meta/recipes-devtools/mtd/mtd-utils_git.bb +1 -1
@@ -7,7 +7,7 @@ LIC_FILES_CHKSUM = "file://COPYING;md5=0636e73ff0215e8d672dc4c32c317bb3 \ inherit autotools pkgconfig update-alternatives -DEPENDS = "zlib e2fsprogs util-linux" +DEPENDS = "zlib e2fsprogs util-linux util-linux-uuid" RDEPENDS_mtd-utils-tests += "bash" PV = "2.1.2"
meta/recipes-devtools/python/python3_3.9.0.bb +1 -1
@@ -63,7 +63,7 @@ ALTERNATIVE_LINK_NAME[python3-config] = "${bindir}/python${PYTHON_MAJMIN}-config ALTERNATIVE_TARGET[python3-config] = "${bindir}/python${PYTHON_MAJMIN}-config-${MULTILIB_SUFFIX}" -DEPENDS = "bzip2-replacement-native libffi bzip2 openssl sqlite3 zlib virtual/libintl xz virtual/crypt util-linux libtirpc libnsl2" +DEPENDS = "bzip2-replacement-native libffi bzip2 openssl sqlite3 zlib virtual/libintl xz virtual/crypt util-linux util-linux-uuid libtirpc libnsl2" DEPENDS_append_class-target = " python3-native" DEPENDS_append_class-nativesdk = " python3-native"
meta/recipes-devtools/syslinux/syslinux_6.04-pre2.bb +1 -1
@@ -6,7 +6,7 @@ LIC_FILES_CHKSUM = "file://COPYING;md5=0636e73ff0215e8d672dc4c32c317bb3 \ # If you really want to run syslinux, you need mtools. We just want the # ldlinux.* stuff for now, so skip mtools-native -DEPENDS = "nasm-native util-linux e2fsprogs" +DEPENDS = "nasm-native util-linux util-linux-uuid e2fsprogs" PV = "6.04-pre2" SRC_URI = "https://www.zytor.com/pub/syslinux/Testing/6.04/syslinux-${PV}.tar.xz \
meta/recipes-devtools/tcf-agent/tcf-agent_git.bb +1 -1
@@ -16,7 +16,7 @@ SRC_URI = "git://git.eclipse.org/gitroot/tcf/org.eclipse.tcf.agent \ file://tcf-agent.service \ " -DEPENDS = "util-linux openssl" +DEPENDS = "util-linux util-linux-uuid openssl" RDEPENDS_${PN} = "bash" S = "${WORKDIR}/git/agent"
meta/recipes-extended/mc/mc_4.8.25.bb +1 -1
@@ -3,7 +3,7 @@ HOMEPAGE = "http://www.midnight-commander.org/" LICENSE = "GPLv3" LIC_FILES_CHKSUM = "file://COPYING;md5=270bbafe360e73f9840bd7981621f9c2" SECTION = "console/utils" -DEPENDS = "ncurses glib-2.0 util-linux" +DEPENDS = "ncurses glib-2.0 util-linux util-linux-uuid" RDEPENDS_${PN} = "ncurses-terminfo-base" RRECOMMENDS_${PN} = "ncurses-terminfo"
meta/recipes-extended/parted/parted_3.3.bb +1 -1
@@ -3,7 +3,7 @@ HOMEPAGE = "http://www.gnu.org/software/parted/parted.html" LICENSE = "GPLv3+" LIC_FILES_CHKSUM = "file://COPYING;md5=2f31b266d3440dd7ee50f92cf67d8e6c" SECTION = "console/tools" -DEPENDS = "ncurses readline util-linux virtual/libiconv" +DEPENDS = "ncurses readline util-linux util-linux-uuid virtual/libiconv" SRC_URI = "${GNU_MIRROR}/parted/parted-${PV}.tar.xz \ file://no_check.patch \
meta/recipes-extended/wget/wget.inc +1 -1
@@ -32,7 +32,7 @@ PACKAGECONFIG[gnutls] = "--with-ssl=gnutls,,gnutls" PACKAGECONFIG[ipv6] = "--enable-ipv6,--disable-ipv6," PACKAGECONFIG[iri] = "--enable-iri,--disable-iri,libidn2" PACKAGECONFIG[libpsl] = "--with-libpsl,--without-libpsl,libpsl" -PACKAGECONFIG[libuuid] = "--with-libuuid,--without-libuuid,util-linux" +PACKAGECONFIG[libuuid] = "--with-libuuid,--without-libuuid,util-linux-uuid" PACKAGECONFIG[openssl] = "--with-ssl=openssl,,openssl" PACKAGECONFIG[pcre] = "--enable-pcre,--disable-pcre,libpcre" PACKAGECONFIG[zlib] = "--with-zlib,--without-zlib,zlib"
meta/recipes-graphics/fontconfig/fontconfig_2.13.1.bb +1 -1
@@ -18,7 +18,7 @@ LIC_FILES_CHKSUM = "file://COPYING;md5=7a0449e9bc5370402a94c00204beca3d \ SECTION = "libs" -DEPENDS = "expat freetype zlib gperf-native util-linux" +DEPENDS = "expat freetype zlib gperf-native util-linux util-linux-uuid" SRC_URI = "http://fontconfig.org/release/fontconfig-${PV}.tar.gz \ file://revert-static-pkgconfig.patch \
meta/recipes-graphics/xorg-lib/libxt_1.2.0.bb +1 -1
@@ -15,7 +15,7 @@ require xorg-lib-common.inc LICENSE = "MIT & MIT-style" LIC_FILES_CHKSUM = "file://COPYING;md5=73d55cea4d27ca1a09a5d23378b3ecf8" -DEPENDS += "util-linux libxcb libsm virtual/libx11 xorgproto libxdmcp" +DEPENDS += "util-linux util-linux-uuid libxcb libsm virtual/libx11 xorgproto libxdmcp" PROVIDES = "xt" PE = "1"
meta/recipes-kernel/lttng/babeltrace2_2.0.3.bb +1 -1
@@ -5,7 +5,7 @@ BUGTRACKER = "https://bugs.lttng.org/projects/babeltrace" LICENSE = "MIT & GPLv2 & LGPLv2.1 & BSD-2-Clause" LIC_FILES_CHKSUM = "file://LICENSE;md5=a6a458c13f18385b7bc5069a6d7b176e" -DEPENDS = "glib-2.0 util-linux popt bison-native flex-native" +DEPENDS = "glib-2.0 util-linux util-linux-uuid popt bison-native flex-native" SRC_URI = "git://git.efficios.com/babeltrace.git;branch=stable-2.0 \ file://run-ptest \
meta/recipes-kernel/lttng/babeltrace_1.5.8.bb +1 -1
@@ -5,7 +5,7 @@ BUGTRACKER = "https://bugs.lttng.org/projects/babeltrace" LICENSE = "MIT & GPLv2 & LGPLv2.1" LIC_FILES_CHKSUM = "file://LICENSE;md5=76ba15dd76a248e1dd526bca0e2125fa" -DEPENDS = "glib-2.0 util-linux popt bison-native flex-native" +DEPENDS = "glib-2.0 util-linux util-linux-uuid popt bison-native flex-native" SRC_URI = "git://git.efficios.com/babeltrace.git;branch=stable-1.5 \ file://run-ptest \
meta/recipes-kernel/lttng/lttng-tools_2.12.2.bb +1 -1
@@ -11,7 +11,7 @@ LIC_FILES_CHKSUM = "file://LICENSE;md5=40ef17463fbd6f377db3c47b1cbaded8 \ include lttng-platforms.inc -DEPENDS = "liburcu popt libxml2 util-linux" +DEPENDS = "liburcu popt libxml2 util-linux util-linux-uuid" RDEPENDS_${PN} = "libgcc" RRECOMMENDS_${PN} += "${LTTNGMODULES}" RDEPENDS_${PN}-ptest += "make perl bash gawk babeltrace procps perl-module-overloading coreutils util-linux kmod ${LTTNGMODULES} sed python3-core"
meta/recipes-kernel/lttng/lttng-ust_2.12.0.bb +1 -1
@@ -19,7 +19,7 @@ include lttng-platforms.inc EXTRA_OECONF = "--disable-numa" -DEPENDS = "liburcu util-linux" +DEPENDS = "liburcu util-linux util-linux-uuid" RDEPENDS_${PN}-bin = "python3-core" # For backwards compatibility after rename
meta/recipes-multimedia/gstreamer/gstreamer1.0-plugins-base_1.18.0.bb +1 -1
@@ -14,7 +14,7 @@ SRC_URI[sha256sum] = "762abdd1a950809a1cea62fff7f86b5f7d6bd5f6841e3e585c700b823c S = "${WORKDIR}/gst-plugins-base-${PV}" -DEPENDS += "iso-codes util-linux zlib" +DEPENDS += "iso-codes util-linux util-linux-uuid zlib" inherit gobject-introspection
meta/recipes-support/apr/apr_1.7.0.bb +1 -1
@@ -1,7 +1,7 @@ SUMMARY = "Apache Portable Runtime (APR) library" HOMEPAGE = "http://apr.apache.org/" SECTION = "libs" -DEPENDS = "util-linux" +DEPENDS = "util-linux util-linux-uuid" LICENSE = "Apache-2.0" LIC_FILES_CHKSUM = "file://LICENSE;md5=4dfd4cd216828c8cae5de5a12f3844c8 \
meta/recipes-support/serf/serf_1.3.9.bb +1 -1
@@ -17,7 +17,7 @@ LIC_FILES_CHKSUM = "file://LICENSE;md5=86d3f3a95c324c9479bd8986968f4327" inherit scons -DEPENDS += " openssl apr apr-util util-linux expat" +DEPENDS += " openssl apr apr-util util-linux util-linux-uuid expat" EXTRA_OESCONS = " \ LIBDIR=${libdir} \
Signature · 1 line
2.27.0
Quoted 13 lines · Luca Bocassi
From: Luca Boccassi <[email protected]>
In v2.35 util-linux gained an (optional) build
dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory). Split out util-linux-uuid in a different recipe to break the cycle.Add a packageconfig switch (disabled by default) to allow using the new dependency.
https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>
Quoted 12 lines
---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
I'm afraid this causes do_package_qa errors in basic testing:
https://autobuilder.yoctoproject.org/typhoon/#/builders/83/builds/1668
I am worried about this change as we're starting to see a number of circular dependencies in util-linux (there is a new bug about the pylibmount PACKAGECONFIG option too), maybe we should flag this upstream?
Multiple recipes like this usually turn into a maintenance nightmare unfortunately which is part of my reluctance to go in this direction, not sure we have any choice though.
Cheers,
Richard
Quoted 27 lines · Richard Purdie
Quoted 13 lines · Luca Bocassi
From: Luca Boccassi <[email protected]>
In v2.35 util-linux gained an (optional) build
dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory). Split out util-linux-uuid in a different recipe to break the cycle.Add a packageconfig switch (disabled by default) to allow using the new dependency.
https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>Quoted 12 lines
---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
Quoted 3 lines
I'm afraid this causes
do_package_qaerrors in basic testing:https://autobuilder.yoctoproject.org/typhoon/#/builders/83/builds/1668
Thanks, added RDEPENDS in v4. Strange that those didn't pop up when building poky locally, I usually get QA warnings as expected. Anything in local.conf to enable to get them?
Quoted 8 lines
I am worried about this change as we're starting to see a number of circular dependencies in util-linux (there is a new bug about the pylibmount PACKAGECONFIG option too), maybe we should flag this upstream?
Multiple recipes like this usually turn into a maintenance nightmare unfortunately which is part of my reluctance to go in this direction, not sure we have any choice though.
Well I've added the feature, and both the maintainer and myself were aware of the implications. It's optional, so on distros with multi- stage bootstrapping functionality like Debian/Ubuntu/RHEL/Suse/etc it can be automatically disabled for the first stage build. At runtime it can also be optional via dlopen, if desired (via --configure flag).
Yocto could really use multi stage support - this isn't the first and won't be the last occurrence. Just my 2c...
Signature · 2 lines
Kind regards, Luca Boccassi
Quoted 38 lines · Luca Boccassi
Quoted 37 lines · Richard Purdie
Quoted 35 lines · Luca Bocassi
From: Luca Boccassi <[email protected]>
In v2.35 util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.Add a packageconfig switch (disabled by default) to allow using the
new dependency.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if
there are
alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for
uuidgen build to happen,
as it does not have its own autoconf switch. Delete the
library manualy from
the main recipe after build instead, and add dependency. Might help to break loop python3 -> util-linux -> libselinux
-> python3, as it's
only libuuid that is needed, see
https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
Quoted 8 lines
Quoted 3 lines
I'm afraid this causes
do_package_qaerrors in basic testing:https://autobuilder.yoctoproject.org/typhoon/#/builders/83/builds/1668
Thanks, added RDEPENDS in v4. Strange that those didn't pop up when building poky locally, I usually get QA warnings as expected. Anything
in local.conf to enable to get them?
No, you should see it in a standard build. It did make me wonder how this was tested :/.
Quoted 11 lines
Quoted 10 lines
I am worried about this change as we're starting to see a number of circular dependencies in util-linux (there is a new bug about the pylibmount PACKAGECONFIG option too), maybe we should flag this upstream?
Multiple recipes like this usually turn into a maintenance nightmare
unfortunately which is part of my reluctance to go in this direction,
not sure we have any choice though.
Quoted 6 lines
Well I've added the feature, and both the maintainer and myself were aware of the implications. It's optional, so on distros with multi- stage bootstrapping functionality like Debian/Ubuntu/RHEL/Suse/etc it can be automatically disabled for the first stage build. At runtime it can also be optional via dlopen, if desired (via --configure flag).
I have to ask why libuuid couldn't be done in a separate repository and avoid the need to do a multi-stage build of a component? To me at least, it would seem to make sense to logically split the library code out, then it avoids all the complexity. Yes, that means a different component to release but that isn't unusual.
Quoted 2 lines
Yocto could really use multi stage support - this isn't the first and won't be the last occurrence. Just my 2c...
Well, we can do it as you're proving, its just ugly and hard to maintain. I don't think the other distros will be particularly happy about needing to do it either. Outside of libgcc, we've not really found that we need to do this often at all and the compiler/libc interface is a lot more "special" than uuid.
Thanks for updating the patch. I'll put it back into the queue and test the new version.
Cheers,
Richard
Quoted 43 lines · Richard Purdie
Quoted 42 lines · Luca Boccassi
Quoted 40 lines · Richard Purdie
Quoted 26 lines · Luca Bocassi
From: Luca Boccassi <[email protected]>
In v2.35 util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.Add a packageconfig switch (disabled by default) to allow using the
new dependency.Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if
there are
alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for
uuidgen build to happen,Quoted 9 lines
as it does not have its own autoconf switch. Delete the
library manualy from
the main recipe after build instead, and add dependency. Might help to break loop python3 -> util-linux -> libselinux
-> python3, as it's
only libuuid that is needed, see
https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.yoctoproject.org%2Fg%2Fyocto%2Fmessage%2F47570&data=04%7C01%7CLuca.Boccassi%40microsoft.com%7Cf1cc5d5a36b94793f16708d89d46dfac%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C637432275624402419%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C2000&sdata=poH8DFxARF1Myy%2BRRirbOE3bJzu5uSO%2F%2BLfsalK%2F4ro%3D&reserved=0
v3: rebased and refactored to have a common util-linux.inc fileI'm afraid this causes
do_package_qaerrors in basic testing:
Quoted 7 lines
Quoted 4 lines
Thanks, added RDEPENDS in v4. Strange that those didn't pop up when building poky locally, I usually get QA warnings as expected. Anything
in local.conf to enable to get them?No, you should see it in a standard build. It did make me wonder how this was tested :/.
Very weird. Maybe some packageconfig difference? I did not go through all the possible permutations manually.
We have been using this recipe change in production for a year now, so it has plenty of builds behind.
Quoted 18 lines
Quoted 11 lines
Quoted 10 lines
I am worried about this change as we're starting to see a number of circular dependencies in util-linux (there is a new bug about the pylibmount PACKAGECONFIG option too), maybe we should flag this upstream?
Multiple recipes like this usually turn into a maintenance nightmare
unfortunately which is part of my reluctance to go in this direction,
not sure we have any choice though.Quoted 6 lines
Well I've added the feature, and both the maintainer and myself were aware of the implications. It's optional, so on distros with multi- stage bootstrapping functionality like Debian/Ubuntu/RHEL/Suse/etc it can be automatically disabled for the first stage build. At runtime it can also be optional via dlopen, if desired (via --configure flag).
Quoted 5 lines
I have to ask why libuuid couldn't be done in a separate repository and avoid the need to do a multi-stage build of a component? To me at least, it would seem to make sense to logically split the library code out, then it avoids all the complexity. Yes, that means a different component to release but that isn't unusual.
Because there's no need for the extra complications - again, it's all optional features, so bootstrapping is not an issue when the tooling is there to support it.
I'm not a util-linux maintainer so my opinion on the subject counts for precisely nothing, but as a contributor and user I'd not be very happy if it was stuck to the lowest common denominator.
Quoted 8 lines
Quoted 2 lines
Yocto could really use multi stage support - this isn't the first and won't be the last occurrence. Just my 2c...
Well, we can do it as you're proving, its just ugly and hard to maintain. I don't think the other distros will be particularly happy about needing to do it either. Outside of libgcc, we've not really found that we need to do this often at all and the compiler/libc interface is a lot more "special" than uuid.
But that's what I'm saying: it doesn't have to be ugly, if the infrastructure is there to support it.
On Debian and derivatives, you just mark the dependency with <!stage1>
- and that's it. When bootstrapping you start from stage1 and the resolver skips those. If the package configure/make scripts are done well, by default optional dependencies are skipped if not available and if not explicitly set - and util-linux does that.
In the RPM world, the spec has conditional macros and you set the appropriate one at the build config level (eg: in the lower ring project on OBS).
It's not perfect of course, and requires attention, and there are complications and gotchas, and things do go wrong at times - but such is life in the software world.
Quoted 2 lines
Thanks for updating the patch. I'll put it back into the queue and test the new version.
Thank you - does the approach of adding RDEPENDS look right? The interaction between those variables and the native/nativesdk builds still confuses me a lot, and I get it wrong all the time.
Signature · 2 lines
Kind regards, Luca Boccassi
Quoted 10 lines · Luca Boccassi
Quoted 8 lines · Richard Purdie
Quoted 2 lines · Luca Boccassi
Quoted 1 line · Richard Purdie
On Mon, 2020-11-23 at 13:28 +0000, Luca Bocassi wrote:
I have to ask why libuuid couldn't be done in a separate repository and avoid the need to do a multi-stage build of a component? To me at least, it would seem to make sense to logically split the library code out, then it avoids all the complexity. Yes, that means a different component to release but that isn't unusual.
Quoted 6 lines
Because there's no need for the extra complications - again, it's all optional features, so bootstrapping is not an issue when the tooling is there to support it.
I'm not a util-linux maintainer so my opinion on the subject counts for precisely nothing, but as a contributor and user I'd not be very happy if it was stuck to the lowest common denominator.
If its all optional and not that important I start to wonder why we should bother with it.
My point is there has to be complexity somewhere for this "mutli-stage" approach to work. How much of it you see depends on the system and how it handles it but you'd agree that having a simple more linear dependency tree *is* simpler and easier to work with than something which has multiple stages (and more efficient on build resources too).
Quoted 11 lines
Quoted 8 lines
Quoted 2 lines
Yocto could really use multi stage support - this isn't the first and won't be the last occurrence. Just my 2c...
Well, we can do it as you're proving, its just ugly and hard to maintain. I don't think the other distros will be particularly happy about needing to do it either. Outside of libgcc, we've not really found that we need to do this often at all and the compiler/libc interface is a lot more "special" than uuid.
But that's what I'm saying: it doesn't have to be ugly, if the infrastructure is there to support it.
I'm sure we (OE) could "hide" it but regardless of whether the code is hidden or not, its still ugly, not often used and hard to maintain for someone. We don't want to encourage this though.
Quoted 11 lines
On Debian and derivatives, you just mark the dependency with <!stage1>
- and that's it. When bootstrapping you start from stage1 and the resolver skips those. If the package configure/make scripts are done well, by default optional dependencies are skipped if not available and if not explicitly set - and util-linux does that.
In the RPM world, the spec has conditional macros and you set the appropriate one at the build config level (eg: in the lower ring project on OBS).
It's not perfect of course, and requires attention, and there are complications and gotchas, and things do go wrong at times - but such is life in the software world.
Complexity is fine, where it makes sense and is needed. You're failing to convince me its needed here at all. I believe this does need to be mentioned to the upstream maintainer as they're probably not aware of the issues it causes. Obviously someone else will have to do that though since you believe its "fine".
Quoted 7 lines
Quoted 3 lines
Thanks for updating the patch. I'll put it back into the queue and test the new version.
Thank you - does the approach of adding RDEPENDS look right? The interaction between those variables and the native/nativesdk builds still confuses me a lot, and I get it wrong all the time.
I've just looked and to be honest, no, it doesn't look right at all :(. You're adding dependencies in a recipe where the packages don't exist.
Also, if the recipes are properly structuctred, there should be no need to do this:
PACKAGES_remove = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-dbg"
The more I look at the patch, the more I'm worried :(. As is, its not going to work. I'm not even sure how its being tested or can work.
Cheers,
Richard
Quoted 18 lines · Richard Purdie
Quoted 10 lines · Luca Boccassi
Quoted 8 lines · Richard Purdie
Quoted 2 lines · Luca Boccassi
Quoted 1 line · Richard Purdie
On Mon, 2020-11-23 at 13:28 +0000, Luca Bocassi wrote:
I have to ask why libuuid couldn't be done in a separate repository and avoid the need to do a multi-stage build of a component? To me at least, it would seem to make sense to logically split the library code out, then it avoids all the complexity. Yes, that means a different component to release but that isn't unusual.
Quoted 6 lines
Because there's no need for the extra complications - again, it's all optional features, so bootstrapping is not an issue when the tooling is there to support it.
I'm not a util-linux maintainer so my opinion on the subject counts for precisely nothing, but as a contributor and user I'd not be very happy if it was stuck to the lowest common denominator.
Quoted 8 lines
If its all optional and not that important I start to wonder why we should bother with it.
My point is there has to be complexity somewhere for this "mutli-stage" approach to work. How much of it you see depends on the system and how it handles it but you'd agree that having a simple more linear dependency tree *is* simpler and easier to work with than something which has multiple stages (and more efficient on build resources too).
Optional doesn't always imply unimportant - for some users it's not needed, for others it is, so it's optional. We are in the latter group, hence the work to implement and support it in a multitude of places.
Quoted 28 lines
Quoted 11 lines
Quoted 8 lines
Quoted 2 lines
Yocto could really use multi stage support - this isn't the first and won't be the last occurrence. Just my 2c...
Well, we can do it as you're proving, its just ugly and hard to maintain. I don't think the other distros will be particularly happy about needing to do it either. Outside of libgcc, we've not really found that we need to do this often at all and the compiler/libc interface is a lot more "special" than uuid.
But that's what I'm saying: it doesn't have to be ugly, if the infrastructure is there to support it.
I'm sure we (OE) could "hide" it but regardless of whether the code is hidden or not, its still ugly, not often used and hard to maintain for someone. We don't want to encourage this though.
Quoted 11 lines
On Debian and derivatives, you just mark the dependency with <!stage1>
- and that's it. When bootstrapping you start from stage1 and the resolver skips those. If the package configure/make scripts are done well, by default optional dependencies are skipped if not available and if not explicitly set - and util-linux does that.
In the RPM world, the spec has conditional macros and you set the appropriate one at the build config level (eg: in the lower ring project on OBS).
It's not perfect of course, and requires attention, and there are complications and gotchas, and things do go wrong at times - but such is life in the software world.
Quoted 5 lines
Complexity is fine, where it makes sense and is needed. You're failing to convince me its needed here at all. I believe this does need to be mentioned to the upstream maintainer as they're probably not aware of the issues it causes. Obviously someone else will have to do that though since you believe its "fine".
I'm sure it wasn't intentional, but this passage comes across as quite condescending. Everybody involved is perfectly capable of understanding how this works and what it implies.
Quoted 9 lines
Quoted 6 lines
Quoted 2 lines
Thanks for updating the patch. I'll put it back into the queue and test the new version.
Thank you - does the approach of adding RDEPENDS look right? The interaction between those variables and the native/nativesdk builds still confuses me a lot, and I get it wrong all the time.
I've just looked and to be honest, no, it doesn't look right at all :(. You're adding dependencies in a recipe where the packages don't exist.
Could you please be more specific? Dependencies on packages from other recipes are pretty much normal. In what way is this different? And most importantly, do you have a suggestion on how would you like to see this done instead?
Quoted 4 lines
Also, if the recipes are properly structuctred, there should be no need to do this:
PACKAGES_remove= "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-dbg"
The recipe is what it is - I'm not adding this auto-generation of packages, it was there already, so I have to work with it. If you have a preference for handling it differently please let me know and I'll apply it.
Quoted 2 lines
The more I look at the patch, the more I'm worried :(. As is, its not going to work. I'm not even sure how its being tested or can work.
As already mentioned, the v2 has not only been tested but used in production for a year. Every subsequent revision is recent, so obviously hasn't been used yet, but it is also pretty much a 1:1 application of Yocto maintainers requests. I've already asked for a specific configuration to try, as the basic poky one created by oe- init-build-env didn't show any issue. The CI infrastructure is completely opaque and imperscrutable to a casual passerby, so unfortunately I wasn't able to extract anything usable from it.
Signature · 2 lines
Kind regards, Luca Boccassi
PATCH v4 1/1 util-linux: split uuid in separate recipe to allow bootstrapping
In reply to Luca BocassiFrom: Luca Boccassi <[email protected]>
In the next release util-linux will gain an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory). Split out util-linux-uuid in a different recipe to break the cycle.
https://github.com/karelzak/util-linux/pull/898
---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
.../nfs-utils/nfs-utils_2.5.2.bb | 2 +- meta/recipes-core/udev/eudev_3.2.9.bb | 2 +- .../util-linux/util-linux-uuid_2.36.bb | 22 ++++++++ meta/recipes-core/util-linux/util-linux.inc | 40 +++++++++++++ .../util-linux/util-linux_2.36.bb | 56 ++++++------------- .../btrfs-tools/btrfs-tools_5.9.bb | 2 +- meta/recipes-devtools/e2fsprogs/e2fsprogs.inc | 2 +- meta/recipes-devtools/fdisk/gptfdisk_1.0.5.bb | 2 +- meta/recipes-devtools/mtd/mtd-utils_git.bb | 2 +- meta/recipes-devtools/python/python3_3.9.0.bb | 2 +- .../syslinux/syslinux_6.04-pre2.bb | 2 +- .../tcf-agent/tcf-agent_git.bb | 2 +- meta/recipes-extended/mc/mc_4.8.25.bb | 2 +- meta/recipes-extended/parted/parted_3.3.bb | 2 +- meta/recipes-extended/wget/wget.inc | 2 +- .../fontconfig/fontconfig_2.13.1.bb | 2 +- meta/recipes-graphics/xorg-lib/libxt_1.2.0.bb | 2 +- .../recipes-kernel/lttng/babeltrace2_2.0.3.bb | 2 +- meta/recipes-kernel/lttng/babeltrace_1.5.8.bb | 2 +- .../lttng/lttng-tools_2.12.2.bb | 2 +- meta/recipes-kernel/lttng/lttng-ust_2.12.0.bb | 2 +- .../gstreamer1.0-plugins-base_1.18.1.bb | 2 +- meta/recipes-support/apr/apr_1.7.0.bb | 2 +- meta/recipes-support/serf/serf_1.3.9.bb | 2 +- 24 files changed, 99 insertions(+), 61 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.bb create mode 100644 meta/recipes-core/util-linux/util-linux.inc
meta/recipes-connectivity/nfs-utils/nfs-utils_2.5.2.bb +1 -1
@@ -8,7 +8,7 @@ LICENSE = "MIT & GPLv2+ & BSD" LIC_FILES_CHKSUM = "file://COPYING;md5=95f3a93a5c3c7888de623b46ea085a84" # util-linux for libblkid -DEPENDS = "libcap libevent util-linux sqlite3 libtirpc" +DEPENDS = "libcap libevent util-linux util-linux-uuid sqlite3 libtirpc" RDEPENDS_${PN} = "${PN}-client" RRECOMMENDS_${PN} = "kernel-module-nfsd"
meta/recipes-core/udev/eudev_3.2.9.bb +1 -1
@@ -4,7 +4,7 @@ LICENSE = "GPLv2.0+ & LGPL-2.1+" LICENSE_libudev = "LGPL-2.1+" LIC_FILES_CHKSUM = "file://COPYING;md5=751419260aa954499f7abaabaa882bbe" -DEPENDS = "glib-2.0 glib-2.0-native gperf-native kmod libxslt-native util-linux" +DEPENDS = "glib-2.0 glib-2.0-native gperf-native kmod libxslt-native util-linux util-linux-uuid" PROVIDES = "udev"
meta/recipes-core/util-linux/util-linux-uuid_2.36.bb new file +22 -0
@@ -0,0 +1,22 @@ +# To allow util-linux to optionally build-depend on cryptsetup, libuuid is +# split out of the main recipe, as it's needed by cryptsetup + +require util-linux.inc + +inherit autotools gettext pkgconfig + +S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +DEBIAN_NOAUTONAME_util-linux-libuuid = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dev = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dbg = "1" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" + +do_install_append() { + rm -rf ${D}${datadir} ${D}${bindir} ${D}${base_bindir} ${D}${sbindir} ${D}${base_sbindir} ${D}${exec_prefix}/sbin +} + +BBCLASSEXTEND = "native nativesdk"
meta/recipes-core/util-linux/util-linux.inc new file +40 -0
@@ -0,0 +1,40 @@ +SUMMARY = "A suite of basic system administration utilities" +HOMEPAGE = "https://en.wikipedia.org/wiki/Util-linux" +DESCRIPTION = "Util-linux includes a suite of basic system administration utilities \ +commonly found on most Linux systems. Some of the more important utilities include \ +disk partitioning, kernel message management, filesystem creation, and system login." + +SECTION = "base" + +LICENSE = "GPLv2+ & LGPLv2.1+ & BSD-3-Clause & BSD-4-Clause" +LICENSE_${PN}-libblkid = "LGPLv2.1+" +LICENSE_${PN}-libfdisk = "LGPLv2.1+" +LICENSE_${PN}-libmount = "LGPLv2.1+" +LICENSE_${PN}-libsmartcols = "LGPLv2.1+" +LICENSE_${PN}-libuuid = "BSD-3-Clause" + +LIC_FILES_CHKSUM = "file://README.licensing;md5=0fd5c050c6187d2bf0a4492b7f4e33da \ + file://COPYING;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ + file://Documentation/licenses/COPYING.GPL-2.0-or-later;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ + file://Documentation/licenses/COPYING.LGPL-2.1-or-later;md5=4fbd65380cdd255951079008b364516c \ + file://Documentation/licenses/COPYING.BSD-3-Clause;md5=58dcd8452651fc8b07d1f65ce07ca8af \ + file://Documentation/licenses/COPYING.BSD-4-Clause-UC;md5=263860f8968d8bafa5392cab74285262 \ + file://libuuid/COPYING;md5=6d2cafc999feb2c2de84d4d24b23290c \ + file://libmount/COPYING;md5=7c7e39fb7d70ffe5d693a643e29987c2 \ + file://libblkid/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ + file://libfdisk/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ + file://libsmartcols/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ +" + +MAJOR_VERSION = "${@'.'.join(d.getVar('PV').split('.')[0:2])}" +FILESEXTRAPATHS_prepend := "${THISDIR}/util-linux:" +SRC_URI = "${KERNELORG_MIRROR}/linux/utils/util-linux/v${MAJOR_VERSION}/util-linux-${MAJOR_VERSION}.tar.xz \ + file://configure-sbindir.patch \ + file://runuser.pamd \ + file://runuser-l.pamd \ + file://ptest.patch \ + file://run-ptest \ + file://display_testname_for_subtest.patch \ + file://avoid_parallel_tests.patch \ + " +SRC_URI[sha256sum] = "9e4b1c67eb13b9b67feb32ae1dc0d50e08ce9e5d82e1cccd0ee771ad2fa9e0b1"
meta/recipes-core/util-linux/util-linux_2.36.bb +16 -40
82 lines not on this page · raw message
meta/recipes-devtools/btrfs-tools/btrfs-tools_5.9.bb +1 -1
@@ -10,7 +10,7 @@ HOMEPAGE = "https://btrfs.wiki.kernel.org" LICENSE = "GPLv2" LIC_FILES_CHKSUM = "file://COPYING;md5=fcb02dc552a041dee27e4b85c7396067" SECTION = "base" -DEPENDS = "util-linux attr e2fsprogs lzo acl" +DEPENDS = "util-linux util-linux-uuid attr e2fsprogs lzo acl" DEPENDS_append_class-target = " udev" RDEPENDS_${PN} = "libgcc"
meta/recipes-devtools/e2fsprogs/e2fsprogs.inc +1 -1
@@ -17,7 +17,7 @@ LIC_FILES_CHKSUM = "file://NOTICE;md5=d50be0580c0b0a7fbc7a4830bbe6c12b \ file://lib/et/et_name.c;beginline=1;endline=11;md5=ead236447dac7b980dbc5b4804d8c836 \ file://lib/ss/ss.h;beginline=1;endline=20;md5=6e89ad47da6e75fecd2b5e0e81e1d4a6" SECTION = "base" -DEPENDS = "util-linux attr" +DEPENDS = "util-linux util-linux-uuid attr" SRC_URI = "git://git.kernel.org/pub/scm/fs/ext2/e2fsprogs.git" S = "${WORKDIR}/git"
meta/recipes-devtools/fdisk/gptfdisk_1.0.5.bb +1 -1
@@ -4,7 +4,7 @@ DESCRIPTION = "GPT fdisk is a disk partitioning tool loosely modeled on Linux fd LICENSE = "GPLv2" LIC_FILES_CHKSUM = "file://COPYING;md5=59530bdf33659b29e73d4adb9f9f6552" -DEPENDS = "util-linux" +DEPENDS = "util-linux util-linux-uuid" SRC_URI = "${SOURCEFORGE_MIRROR}/${BPN}/${PV}/${BP}.tar.gz \ file://0001-gptcurses-correctly-include-curses.h.patch \
meta/recipes-devtools/mtd/mtd-utils_git.bb +1 -1
@@ -7,7 +7,7 @@ LIC_FILES_CHKSUM = "file://COPYING;md5=0636e73ff0215e8d672dc4c32c317bb3 \ inherit autotools pkgconfig update-alternatives -DEPENDS = "zlib e2fsprogs util-linux" +DEPENDS = "zlib e2fsprogs util-linux util-linux-uuid" RDEPENDS_mtd-utils-tests += "bash" PV = "2.1.2"
meta/recipes-devtools/python/python3_3.9.0.bb +1 -1
@@ -69,7 +69,7 @@ ALTERNATIVE_LINK_NAME[python3-config] = "${bindir}/python${PYTHON_MAJMIN}-config ALTERNATIVE_TARGET[python3-config] = "${bindir}/python${PYTHON_MAJMIN}-config-${MULTILIB_SUFFIX}" -DEPENDS = "bzip2-replacement-native libffi bzip2 openssl sqlite3 zlib virtual/libintl xz virtual/crypt util-linux libtirpc libnsl2" +DEPENDS = "bzip2-replacement-native libffi bzip2 openssl sqlite3 zlib virtual/libintl xz virtual/crypt util-linux util-linux-uuid libtirpc libnsl2" DEPENDS_append_class-target = " python3-native" DEPENDS_append_class-nativesdk = " python3-native"
meta/recipes-devtools/syslinux/syslinux_6.04-pre2.bb +1 -1
@@ -6,7 +6,7 @@ LIC_FILES_CHKSUM = "file://COPYING;md5=0636e73ff0215e8d672dc4c32c317bb3 \ # If you really want to run syslinux, you need mtools. We just want the # ldlinux.* stuff for now, so skip mtools-native -DEPENDS = "nasm-native util-linux e2fsprogs" +DEPENDS = "nasm-native util-linux util-linux-uuid e2fsprogs" PV = "6.04-pre2" SRC_URI = "https://www.zytor.com/pub/syslinux/Testing/6.04/syslinux-${PV}.tar.xz \
meta/recipes-devtools/tcf-agent/tcf-agent_git.bb +1 -1
@@ -16,7 +16,7 @@ SRC_URI = "git://git.eclipse.org/gitroot/tcf/org.eclipse.tcf.agent \ file://tcf-agent.service \ " -DEPENDS = "util-linux openssl" +DEPENDS = "util-linux util-linux-uuid openssl" RDEPENDS_${PN} = "bash" S = "${WORKDIR}/git/agent"
meta/recipes-extended/mc/mc_4.8.25.bb +1 -1
@@ -3,7 +3,7 @@ HOMEPAGE = "http://www.midnight-commander.org/" LICENSE = "GPLv3" LIC_FILES_CHKSUM = "file://COPYING;md5=270bbafe360e73f9840bd7981621f9c2" SECTION = "console/utils" -DEPENDS = "ncurses glib-2.0 util-linux" +DEPENDS = "ncurses glib-2.0 util-linux util-linux-uuid" RDEPENDS_${PN} = "ncurses-terminfo-base" RRECOMMENDS_${PN} = "ncurses-terminfo"
meta/recipes-extended/parted/parted_3.3.bb +1 -1
@@ -3,7 +3,7 @@ HOMEPAGE = "http://www.gnu.org/software/parted/parted.html" LICENSE = "GPLv3+" LIC_FILES_CHKSUM = "file://COPYING;md5=2f31b266d3440dd7ee50f92cf67d8e6c" SECTION = "console/tools" -DEPENDS = "ncurses readline util-linux virtual/libiconv" +DEPENDS = "ncurses readline util-linux util-linux-uuid virtual/libiconv" SRC_URI = "${GNU_MIRROR}/parted/parted-${PV}.tar.xz \ file://no_check.patch \
meta/recipes-extended/wget/wget.inc +1 -1
@@ -32,7 +32,7 @@ PACKAGECONFIG[gnutls] = "--with-ssl=gnutls,,gnutls" PACKAGECONFIG[ipv6] = "--enable-ipv6,--disable-ipv6," PACKAGECONFIG[iri] = "--enable-iri,--disable-iri,libidn2" PACKAGECONFIG[libpsl] = "--with-libpsl,--without-libpsl,libpsl" -PACKAGECONFIG[libuuid] = "--with-libuuid,--without-libuuid,util-linux" +PACKAGECONFIG[libuuid] = "--with-libuuid,--without-libuuid,util-linux-uuid" PACKAGECONFIG[openssl] = "--with-ssl=openssl,,openssl" PACKAGECONFIG[pcre] = "--enable-pcre,--disable-pcre,libpcre" PACKAGECONFIG[zlib] = "--with-zlib,--without-zlib,zlib"
meta/recipes-graphics/fontconfig/fontconfig_2.13.1.bb +1 -1
@@ -18,7 +18,7 @@ LIC_FILES_CHKSUM = "file://COPYING;md5=7a0449e9bc5370402a94c00204beca3d \ SECTION = "libs" -DEPENDS = "expat freetype zlib gperf-native util-linux" +DEPENDS = "expat freetype zlib gperf-native util-linux util-linux-uuid" SRC_URI = "http://fontconfig.org/release/fontconfig-${PV}.tar.gz \ file://revert-static-pkgconfig.patch \
meta/recipes-graphics/xorg-lib/libxt_1.2.0.bb +1 -1
@@ -15,7 +15,7 @@ require xorg-lib-common.inc LICENSE = "MIT & MIT-style" LIC_FILES_CHKSUM = "file://COPYING;md5=73d55cea4d27ca1a09a5d23378b3ecf8" -DEPENDS += "util-linux libxcb libsm virtual/libx11 xorgproto libxdmcp" +DEPENDS += "util-linux util-linux-uuid libxcb libsm virtual/libx11 xorgproto libxdmcp" PROVIDES = "xt" PE = "1"
meta/recipes-kernel/lttng/babeltrace2_2.0.3.bb +1 -1
@@ -5,7 +5,7 @@ BUGTRACKER = "https://bugs.lttng.org/projects/babeltrace" LICENSE = "MIT & GPLv2 & LGPLv2.1 & BSD-2-Clause" LIC_FILES_CHKSUM = "file://LICENSE;md5=a6a458c13f18385b7bc5069a6d7b176e" -DEPENDS = "glib-2.0 util-linux popt bison-native flex-native" +DEPENDS = "glib-2.0 util-linux util-linux-uuid popt bison-native flex-native" SRC_URI = "git://git.efficios.com/babeltrace.git;branch=stable-2.0 \ file://run-ptest \
meta/recipes-kernel/lttng/babeltrace_1.5.8.bb +1 -1
@@ -5,7 +5,7 @@ BUGTRACKER = "https://bugs.lttng.org/projects/babeltrace" LICENSE = "MIT & GPLv2 & LGPLv2.1" LIC_FILES_CHKSUM = "file://LICENSE;md5=76ba15dd76a248e1dd526bca0e2125fa" -DEPENDS = "glib-2.0 util-linux popt bison-native flex-native" +DEPENDS = "glib-2.0 util-linux util-linux-uuid popt bison-native flex-native" SRC_URI = "git://git.efficios.com/babeltrace.git;branch=stable-1.5 \ file://run-ptest \
meta/recipes-kernel/lttng/lttng-tools_2.12.2.bb +1 -1
@@ -11,7 +11,7 @@ LIC_FILES_CHKSUM = "file://LICENSE;md5=40ef17463fbd6f377db3c47b1cbaded8 \ include lttng-platforms.inc -DEPENDS = "liburcu popt libxml2 util-linux" +DEPENDS = "liburcu popt libxml2 util-linux util-linux-uuid" RDEPENDS_${PN} = "libgcc" RRECOMMENDS_${PN} += "${LTTNGMODULES}" RDEPENDS_${PN}-ptest += "make perl bash gawk babeltrace procps perl-module-overloading coreutils util-linux kmod ${LTTNGMODULES} sed python3-core"
meta/recipes-kernel/lttng/lttng-ust_2.12.0.bb +1 -1
@@ -19,7 +19,7 @@ include lttng-platforms.inc EXTRA_OECONF = "--disable-numa" -DEPENDS = "liburcu util-linux" +DEPENDS = "liburcu util-linux util-linux-uuid" RDEPENDS_${PN}-bin = "python3-core" # For backwards compatibility after rename
meta/recipes-multimedia/gstreamer/gstreamer1.0-plugins-base_1.18.1.bb +1 -1
@@ -13,7 +13,7 @@ SRC_URI[sha256sum] = "1ba654d7de30f7284b4c7071b32f881b609733ce02ab6d9d9ea29386a0 S = "${WORKDIR}/gst-plugins-base-${PV}" -DEPENDS += "iso-codes util-linux zlib" +DEPENDS += "iso-codes util-linux util-linux-uuid zlib" inherit gobject-introspection
meta/recipes-support/apr/apr_1.7.0.bb +1 -1
@@ -1,7 +1,7 @@ SUMMARY = "Apache Portable Runtime (APR) library" HOMEPAGE = "http://apr.apache.org/" SECTION = "libs" -DEPENDS = "util-linux" +DEPENDS = "util-linux util-linux-uuid" LICENSE = "Apache-2.0" LIC_FILES_CHKSUM = "file://LICENSE;md5=4dfd4cd216828c8cae5de5a12f3844c8 \
meta/recipes-support/serf/serf_1.3.9.bb +1 -1
@@ -17,7 +17,7 @@ LIC_FILES_CHKSUM = "file://LICENSE;md5=86d3f3a95c324c9479bd8986968f4327" inherit scons -DEPENDS += " openssl apr apr-util util-linux expat" +DEPENDS += " openssl apr apr-util util-linux util-linux-uuid expat" EXTRA_OESCONS = " \ LIBDIR=${libdir} \
Signature · 1 line
2.29.2
Quoted 10 lines · On Thu, 2020-12-10 at 18:46 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
In the next release util-linux will gain an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory). Split out util-linux-uuid in a different recipe to break the cycle.
https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>
Quoted 12 lines
---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid) and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
This patch got a bit lost before the holidays but I've been reminded we need to resolve it.
I think the one remaining issue here is the need to change the DEPENDS of so many other recipes, likely not just here in this patch but in other layers. I think if util-linux DEPENDS on util-linux-uuid that might remove the need for those changes? That should still allow you to break the circular dependency problem?
I suspect libuuid should really be maintained/built as a separate software project given the dependency problems but that isn't my decision, we just have to deal with it.
I am also worried this is going to break AUH and mean we have to manually handle this recipe but again, I suspect there is little to be done and we just have to deal with it.
Could you resend it with the above DEPENDS change if that will work ok, refreshed against OE-Core since we have util-linux 2.36.1 now and then hopefully we can resolve this.
Cheers,
Richard
Hi Richard
Quoted 5 lines · On Wednesday, 20 January 2021 09:52:41 NZDT Richard Purdie wrote:
I think the one remaining issue here is the need to change the DEPENDS of so many other recipes, likely not just here in this patch but in other layers. I think if util-linux DEPENDS on util-linux-uuid that might remove the need for those changes? That should still allow you to break the circular dependency problem?
I have to admit to a gap in my own knowledge of how our build system handles transitive dependencies. Of course the recipe sysroot should still get everything it needs in it even if the dependency is only indirectly included, in the back of my mind I have the impression that there are expectations that all dependencies are explicitly called out and there are subtle issues if they aren't, but that could be a mistaken impression on my part.
Quoted 3 lines
I suspect libuuid should really be maintained/built as a separate software project given the dependency problems but that isn't my decision, we just have to deal with it.
I agree that it would be better being separate, FWIW.
Quoted 3 lines
I am also worried this is going to break AUH and mean we have to manually handle this recipe but again, I suspect there is little to be done and we just have to deal with it.
Could we perhaps fix the AUH to handle this properly? Do we need some kind of mechanism to get it to always upgrade the two recipes together or is that only part of the issue?
Cheers
Paul
Quoted 7 lines · Paul Eggleton
Quoted 5 lines · On Wednesday, 20 January 2021 09:52:41 NZDT Richard Purdie wrote:
I think the one remaining issue here is the need to change the DEPENDS of so many other recipes, likely not just here in this patch but in other layers. I think if util-linux DEPENDS on util-linux-uuid that might remove the need for those changes? That should still allow you to break the circular dependency problem?
Quoted 6 lines
I have to admit to a gap in my own knowledge of how our build system handles transitive dependencies. Of course the recipe sysroot should still get everything it needs in it even if the dependency is only indirectly included, in the back of my mind I have the impression that there are expectations that all dependencies are explicitly called out and there are subtle issues if they aren't, but that could be a mistaken impression on my part.
I do wonder a little about that as well. As you say, sysroot dependencies should handle this. Anything linking against libuuid should also establish an package level runtime dependency through the linkage so I think this should work.
We definitely don't explicitly list every dependency in every recipe.
If this can work, it makes the migration path for people easier so I think its at least worth investigating/testing.
Just while I'm thinking, the PACKAGES_remove also bothers me a little. Can we rearrange the variables so libuuid is only added in the libuuid recipe variant?
[the idea being that since we control the metadata in oe-core, we shouldn't need to use _remove and can restructure so we don't need to, they're hard to undo. I know we do use it in places sadly even in core]
Quoted 6 lines
Quoted 3 lines
I suspect libuuid should really be maintained/built as a separate software project given the dependency problems but that isn't my decision, we just have to deal with it.
I agree that it would be better being separate, FWIW.
Quoted 7 lines
Quoted 3 lines
I am also worried this is going to break AUH and mean we have to manually handle this recipe but again, I suspect there is little to be done and we just have to deal with it.
Could we perhaps fix the AUH to handle this properly? Do we need some kind of mechanism to get it to always upgrade the two recipes together or is that only part of the issue?
I don't know for sure that AUH won't handle it, I just worry about it. If it doesn't it definitely could be something we could fix there. I just don't know of anyone with the time to spend on what is a marginal corner case if it doesn't work.
Cheers,
Richard
Quoted 21 lines · On Wednesday, 20 January 2021 12:23:34 NZDT Richard Purdie wrote:
Quoted 7 lines · Paul Eggleton
Quoted 5 lines · On Wednesday, 20 January 2021 09:52:41 NZDT Richard Purdie wrote:
I think the one remaining issue here is the need to change the DEPENDS of so many other recipes, likely not just here in this patch but in other layers. I think if util-linux DEPENDS on util-linux-uuid that might remove the need for those changes? That should still allow you to break the circular dependency problem?
Quoted 7 lines
I have to admit to a gap in my own knowledge of how our build system handles transitive dependencies. Of course the recipe sysroot should still get everything it needs in it even if the dependency is only indirectly included, in the back of my mind I have the impression that there are expectations that all dependencies are explicitly called out and there are subtle issues if they aren't, but that could be a mistaken impression on my part.
I do wonder a little about that as well. As you say, sysroot dependencies should handle this. Anything linking against libuuid should also establish an package level runtime dependency through the linkage so I think this should work.
Quoted 4 lines
We definitely don't explicitly list every dependency in every recipe.
If this can work, it makes the migration path for people easier so I think its at least worth investigating/testing.
OK, sure thing.
Quoted 7 lines
Just while I'm thinking, the
PACKAGES_removealso bothers me a little. Can we rearrange the variables so libuuid is only added in the libuuid recipe variant?[the idea being that since we control the metadata in oe-core, we shouldn't need to use
_removeand can restructure so we don't need to, they're hard to undo. I know we do use it in places sadly even in core]
Yes, that is a good point. We should be able to avoid using _remove and I don't like using it either given how heavy a hammer it is.
Quoted 6 lines
Quoted 6 lines
I agree that it would be better being separate, FWIW.
Quoted 3 lines
I am also worried this is going to break AUH and mean we have to manually handle this recipe but again, I suspect there is little to be done and we just have to deal with it.
Quoted 8 lines
Quoted 3 lines
Could we perhaps fix the AUH to handle this properly? Do we need some kind of mechanism to get it to always upgrade the two recipes together or is that only part of the issue?
I don't know for sure that AUH won't handle it, I just worry about it. If it doesn't it definitely could be something we could fix there. I just don't know of anyone with the time to spend on what is a marginal corner case if it doesn't work.
Well, on the other hand if we're "causing" a problem with the AUH with this change it does behoove us to try to resolve that. It is something I'd be prepared to look into at least.
Cheers
Paul
Quoted 55 lines · Paul Eggleton
Quoted 21 lines · On Wednesday, 20 January 2021 12:23:34 NZDT Richard Purdie wrote:
Quoted 7 lines · Paul Eggleton
Quoted 5 lines · On Wednesday, 20 January 2021 09:52:41 NZDT Richard Purdie wrote:
I think the one remaining issue here is the need to change the DEPENDS of so many other recipes, likely not just here in this patch but in other layers. I think if util-linux DEPENDS on util-linux-uuid that might remove the need for those changes? That should still allow you to break the circular dependency problem?
Quoted 7 lines
I have to admit to a gap in my own knowledge of how our build system handles transitive dependencies. Of course the recipe sysroot should still get everything it needs in it even if the dependency is only indirectly included, in the back of my mind I have the impression that there are expectations that all dependencies are explicitly called out and there are subtle issues if they aren't, but that could be a mistaken impression on my part.
I do wonder a little about that as well. As you say, sysroot dependencies should handle this. Anything linking against libuuid should also establish an package level runtime dependency through the linkage so I think this should work.
Quoted 4 lines
We definitely don't explicitly list every dependency in every recipe.
If this can work, it makes the migration path for people easier so I think its at least worth investigating/testing.
OK, sure thing.
Quoted 7 lines
Just while I'm thinking, the
PACKAGES_removealso bothers me a little. Can we rearrange the variables so libuuid is only added in the libuuid recipe variant?[the idea being that since we control the metadata in oe-core, we shouldn't need to use
_removeand can restructure so we don't need to, they're hard to undo. I know we do use it in places sadly even in core]Yes, that is a good point. We should be able to avoid using
_removeand I don't like using it either given how heavy a hammer it is.Quoted 6 lines
Quoted 6 lines
I agree that it would be better being separate, FWIW.
Quoted 3 lines
I am also worried this is going to break AUH and mean we have to manually handle this recipe but again, I suspect there is little to be done and we just have to deal with it.
Quoted 8 lines
Quoted 3 lines
Could we perhaps fix the AUH to handle this properly? Do we need some kind of mechanism to get it to always upgrade the two recipes together or is that only part of the issue?
I don't know for sure that AUH won't handle it, I just worry about it. If it doesn't it definitely could be something we could fix there. I just don't know of anyone with the time to spend on what is a marginal corner case if it doesn't work.
Quoted 6 lines
Well, on the other hand if we're "causing" a problem with the AUH with this change it does behoove us to try to resolve that. It is something I'd be prepared to look into at least.
Cheers
Paul
Thank you both for the reviews and suggestions, sorry it took a while to get back to this. Just sent v5.
Signature · 2 lines
Kind regards, Luca Boccassi
PATCH v5 1/1 util-linux: split uuid in separate recipe to allow bootstrapping
In reply to Luca BocassiFrom: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.
https://github.com/karelzak/util-linux/pull/898
---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
Removed PACKAGES_remove and instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.
.../util-linux/util-linux-uuid_2.36.2.bb | 22 ++++++++ meta/recipes-core/util-linux/util-linux.inc | 41 ++++++++++++++ .../util-linux/util-linux_2.36.2.bb | 53 +++++-------------- 3 files changed, 75 insertions(+), 41 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb create mode 100644 meta/recipes-core/util-linux/util-linux.inc
meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb new file +22 -0
@@ -0,0 +1,22 @@ +# To allow util-linux to optionally build-depend on cryptsetup, libuuid is +# split out of the main recipe, as it's needed by cryptsetup + +require util-linux.inc + +inherit autotools gettext pkgconfig + +S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +DEBIAN_NOAUTONAME_util-linux-libuuid = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dev = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dbg = "1" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" + +do_install_append() { + rm -rf ${D}${datadir} ${D}${bindir} ${D}${base_bindir} ${D}${sbindir} ${D}${base_sbindir} ${D}${exec_prefix}/sbin +} + +BBCLASSEXTEND = "native nativesdk"
meta/recipes-core/util-linux/util-linux.inc new file +41 -0
41 lines not on this page · raw message
meta/recipes-core/util-linux/util-linux_2.36.2.bb +12 -41
84 lines not on this page · raw message
Signature · 1 line
2.29.2
seeing dependency loops
http://sprunge.us/E5c3NC
Quoted 209 lines · Luca Bocassi
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
RemovedPACKAGES_removeand instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2..../util-linux/util-linux-uuid_2.36.2.bb | 22 ++++++++ meta/recipes-core/util-linux/util-linux.inc | 41 ++++++++++++++ .../util-linux/util-linux_2.36.2.bb | 53 +++++-------------- 3 files changed, 75 insertions(+), 41 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb create mode 100644 meta/recipes-core/util-linux/util-linux.incmeta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb new file +22
-0@@ -0,0 +1,22 @@ +# To allow util-linux to optionally build-depend on cryptsetup, libuuid is +# split out of the main recipe, as it's needed by cryptsetup + +require util-linux.inc + +inherit autotools gettext pkgconfig + +S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +DEBIAN_NOAUTONAME_util-linux-libuuid = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dev = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dbg = "1" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" + +do_install_append() { + rm -rf ${D}${datadir} ${D}${bindir} ${D}${base_bindir} ${D}${sbindir} ${D}${base_sbindir} ${D}${exec_prefix}/sbin +} + +BBCLASSEXTEND = "native nativesdk"meta/recipes-core/util-linux/util-linux.inc new file +41
-0@@ -0,0 +1,41 @@ +SUMMARY = "A suite of basic system administration utilities" +HOMEPAGE = "https://en.wikipedia.org/wiki/Util-linux" +DESCRIPTION = "Util-linux includes a suite of basic system administration utilities \ +commonly found on most Linux systems. Some of the more important utilities include \ +disk partitioning, kernel message management, filesystem creation, and system login." + +SECTION = "base" + +LICENSE = "GPLv2+ & LGPLv2.1+ & BSD-3-Clause & BSD-4-Clause" +LICENSE_${PN}-libblkid = "LGPLv2.1+" +LICENSE_${PN}-libfdisk = "LGPLv2.1+" +LICENSE_${PN}-libmount = "LGPLv2.1+" +LICENSE_${PN}-libsmartcols = "LGPLv2.1+" +LICENSE_${PN}-libuuid = "BSD-3-Clause" + +LIC_FILES_CHKSUM = "file://README.licensing;md5=0fd5c050c6187d2bf0a4492b7f4e33da \ + file://COPYING;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ + file://Documentation/licenses/COPYING.GPL-2.0-or-later;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ + file://Documentation/licenses/COPYING.LGPL-2.1-or-later;md5=4fbd65380cdd255951079008b364516c \ + file://Documentation/licenses/COPYING.BSD-3-Clause;md5=58dcd8452651fc8b07d1f65ce07ca8af \ + file://Documentation/licenses/COPYING.BSD-4-Clause-UC;md5=263860f8968d8bafa5392cab74285262 \ + file://libuuid/COPYING;md5=6d2cafc999feb2c2de84d4d24b23290c \ + file://libmount/COPYING;md5=7c7e39fb7d70ffe5d693a643e29987c2 \ + file://libblkid/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ + file://libfdisk/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ + file://libsmartcols/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ +" + +FILESEXTRAPATHS_prepend := "${THISDIR}/util-linux:" +MAJOR_VERSION = "${@'.'.join(d.getVar('PV').split('.')[0:2])}" +BPN = "util-linux" +SRC_URI = "${KERNELORG_MIRROR}/linux/utils/${BPN}/v${MAJOR_VERSION}/${BP}.tar.xz \ + file://configure-sbindir.patch \ + file://runuser.pamd \ + file://runuser-l.pamd \ + file://ptest.patch \ + file://run-ptest \ + file://display_testname_for_subtest.patch \ + file://avoid_parallel_tests.patch \ + " +SRC_URI[sha256sum] = "f7516ba9d8689343594356f0e5e1a5f0da34adfbc89023437735872bb5024c5f"meta/recipes-core/util-linux/util-linux_2.36.2.bb +12
-42@@ -1,46 +1,8 @@ -SUMMARY = "A suite of basic system administration utilities" -HOMEPAGE = "https://en.wikipedia.org/wiki/Util-linux" -DESCRIPTION = "Util-linux includes a suite of basic system administration utilities \ -commonly found on most Linux systems. Some of the more important utilities include \ -disk partitioning, kernel message management, filesystem creation, and system login." - -SECTION = "base" - -LICENSE = "GPLv2+ & LGPLv2.1+ & BSD-3-Clause & BSD-4-Clause" -LICENSE_${PN}-libblkid = "LGPLv2.1+" -LICENSE_${PN}-libfdisk = "LGPLv2.1+" -LICENSE_${PN}-libmount = "LGPLv2.1+" -LICENSE_${PN}-libsmartcols = "LGPLv2.1+" -LICENSE_${PN}-libuuid = "BSD-3-Clause" - -LIC_FILES_CHKSUM = "file://README.licensing;md5=0fd5c050c6187d2bf0a4492b7f4e33da \ - file://COPYING;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ - file://Documentation/licenses/COPYING.GPL-2.0-or-later;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ - file://Documentation/licenses/COPYING.LGPL-2.1-or-later;md5=4fbd65380cdd255951079008b364516c \ - file://Documentation/licenses/COPYING.BSD-3-Clause;md5=58dcd8452651fc8b07d1f65ce07ca8af \ - file://Documentation/licenses/COPYING.BSD-4-Clause-UC;md5=263860f8968d8bafa5392cab74285262 \ - file://libuuid/COPYING;md5=6d2cafc999feb2c2de84d4d24b23290c \ - file://libmount/COPYING;md5=7c7e39fb7d70ffe5d693a643e29987c2 \ - file://libblkid/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ - file://libfdisk/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ - file://libsmartcols/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ -" +require util-linux.inc #gtk-doc is not enabled as it requires xmlto which requires util-linux inherit autotools gettext manpages pkgconfig systemd update-alternatives python3-dir bash-completion ptest -DEPENDS = "libcap-ng ncurses virtual/crypt zlib" - -MAJOR_VERSION = "${@'.'.join(d.getVar('PV').split('.')[0:2])}" -SRC_URI = "${KERNELORG_MIRROR}/linux/utils/${BPN}/v${MAJOR_VERSION}/${BP}.tar.xz \ - file://configure-sbindir.patch \ - file://runuser.pamd \ - file://runuser-l.pamd \ - file://ptest.patch \ - file://run-ptest \ - file://display_testname_for_subtest.patch \ - file://avoid_parallel_tests.patch \ - " -SRC_URI[sha256sum] = "f7516ba9d8689343594356f0e5e1a5f0da34adfbc89023437735872bb5024c5f" +DEPENDS = "libcap-ng ncurses virtual/crypt zlib util-linux-uuid" PACKAGES =+ "${PN}-swaponoff" PACKAGES += "${@bb.utils.contains('PACKAGECONFIG', 'pylibmount', '${PN}-pylibmount', '', d)}"@@ -87,8 +49,9 @@ python util_linux_binpackages () { # we must execute before update-alternatives PACKAGE_PREPROCESS_FUNCS PACKAGE_PREPROCESS_FUNCS =+ "util_linux_binpackages " +# skip libuuid as it will be packaged by the util-linux-uuid recipe python util_linux_libpackages() { - do_split_packages(d, root=d.getVar('UTIL_LINUX_LIBDIR'), file_regex=r'^lib(.*)\.so\..*$', + do_split_packages(d, root=d.getVar('UTIL_LINUX_LIBDIR'), file_regex=r'^lib(?!uuid)(.*)\.so\..*$', output_pattern='${PN}-lib%s', description='${PN} lib%s', extra_depends='', prepend=True, allow_links=True)@@ -141,6 +104,7 @@ PACKAGECONFIG[pylibmount] = "--with-python=3 --enable-pylibmount,--without-pytho PACKAGECONFIG[readline] = "--with-readline,--without-readline,readline" # PCRE support in hardlink PACKAGECONFIG[pcre2] = ",,libpcre2" +PACKAGECONFIG[cryptsetup] = "--with-cryptsetup,--without-cryptsetup,cryptsetup" EXTRA_OEMAKE = "ARCH=${TARGET_ARCH} CPU= CPUOPT= 'OPT=${CFLAGS}'"@@ -165,6 +129,8 @@ RRECOMMENDS_${PN}_class-nativesdk = "" RDEPENDS_${PN}_class-native = "" RDEPENDS_${PN}_class-nativesdk = "" +RDEPENDS_${PN}_append += " util-linux-uuid" + RPROVIDES_${PN}-dev = "${PN}-libblkid-dev ${PN}-libmount-dev ${PN}-libuuid-dev" RDEPENDS_${PN}-bash-completion += "${PN}-lsblk"@@ -316,6 +282,11 @@ ALTERNATIVE_LINK_NAME[utmpdump.1] = "${mandir}/man1/utmpdump.1" ALTERNATIVE_LINK_NAME[uuid.3] = "${mandir}/man3/uuid.3" ALTERNATIVE_LINK_NAME[wall.1] = "${mandir}/man1/wall.1" +# dm-verity support introduces a circular build dependency, so util-linux-uuid is split out for target builds +# Need to build libuuid for uuidgen, but then delete it and let the other recipe ship it +do_install_append() { + rm -rf ${D}${includedir}/uuid ${D}${libdir}/pkgconfig/uuid.pc ${D}${libdir}/libuuid* ${D}/lib/libuuid* +} BBCLASSEXTEND = "native nativesdk" --2.29.2
Quoted 2 lines · Khem Raj
seeing dependency loops
http://sprunge.us/E5c3NC
Thanks for the heads-up - didn't see when building locally with a minimal poky, anything in particular needed in the local.conf to trigger it?
Quoted 209 lines · Luca Bocassi
Quoted 208 lines
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
RemovedPACKAGES_removeand instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2..../util-linux/util-linux-uuid_2.36.2.bb | 22 ++++++++ meta/recipes-core/util-linux/util-linux.inc | 41 ++++++++++++++ .../util-linux/util-linux_2.36.2.bb | 53 +++++-------------- 3 files changed, 75 insertions(+), 41 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb create mode 100644 meta/recipes-core/util-linux/util-linux.incmeta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb new file +22
-0@@ -0,0 +1,22 @@ +# To allow util-linux to optionally build-depend on cryptsetup, libuuid is +# split out of the main recipe, as it's needed by cryptsetup + +require util-linux.inc + +inherit autotools gettext pkgconfig + +S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +DEBIAN_NOAUTONAME_util-linux-libuuid = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dev = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dbg = "1" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" + +do_install_append() { + rm -rf ${D}${datadir} ${D}${bindir} ${D}${base_bindir} ${D}${sbindir} ${D}${base_sbindir} ${D}${exec_prefix}/sbin +} + +BBCLASSEXTEND = "native nativesdk"meta/recipes-core/util-linux/util-linux.inc new file +41
-0@@ -0,0 +1,41 @@ +SUMMARY = "A suite of basic system administration utilities" +HOMEPAGE = "https://en.wikipedia.org/wiki/Util-linux" +DESCRIPTION = "Util-linux includes a suite of basic system administration utilities \ +commonly found on most Linux systems. Some of the more important utilities include \ +disk partitioning, kernel message management, filesystem creation, and system login." + +SECTION = "base" + +LICENSE = "GPLv2+ & LGPLv2.1+ & BSD-3-Clause & BSD-4-Clause" +LICENSE_${PN}-libblkid = "LGPLv2.1+" +LICENSE_${PN}-libfdisk = "LGPLv2.1+" +LICENSE_${PN}-libmount = "LGPLv2.1+" +LICENSE_${PN}-libsmartcols = "LGPLv2.1+" +LICENSE_${PN}-libuuid = "BSD-3-Clause" + +LIC_FILES_CHKSUM = "file://README.licensing;md5=0fd5c050c6187d2bf0a4492b7f4e33da \ + file://COPYING;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ + file://Documentation/licenses/COPYING.GPL-2.0-or-later;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ + file://Documentation/licenses/COPYING.LGPL-2.1-or-later;md5=4fbd65380cdd255951079008b364516c \ + file://Documentation/licenses/COPYING.BSD-3-Clause;md5=58dcd8452651fc8b07d1f65ce07ca8af \ + file://Documentation/licenses/COPYING.BSD-4-Clause-UC;md5=263860f8968d8bafa5392cab74285262 \ + file://libuuid/COPYING;md5=6d2cafc999feb2c2de84d4d24b23290c \ + file://libmount/COPYING;md5=7c7e39fb7d70ffe5d693a643e29987c2 \ + file://libblkid/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ + file://libfdisk/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ + file://libsmartcols/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ +" + +FILESEXTRAPATHS_prepend := "${THISDIR}/util-linux:" +MAJOR_VERSION = "${@'.'.join(d.getVar('PV').split('.')[0:2])}" +BPN = "util-linux" +SRC_URI = "${KERNELORG_MIRROR}/linux/utils/${BPN}/v${MAJOR_VERSION}/${BP}.tar.xz \ + file://configure-sbindir.patch \ + file://runuser.pamd \ + file://runuser-l.pamd \ + file://ptest.patch \ + file://run-ptest \ + file://display_testname_for_subtest.patch \ + file://avoid_parallel_tests.patch \ + " +SRC_URI[sha256sum] = "f7516ba9d8689343594356f0e5e1a5f0da34adfbc89023437735872bb5024c5f"meta/recipes-core/util-linux/util-linux_2.36.2.bb +12
-42@@ -1,46 +1,8 @@ -SUMMARY = "A suite of basic system administration utilities" -HOMEPAGE = "https://en.wikipedia.org/wiki/Util-linux" -DESCRIPTION = "Util-linux includes a suite of basic system administration utilities \ -commonly found on most Linux systems. Some of the more important utilities include \ -disk partitioning, kernel message management, filesystem creation, and system login." - -SECTION = "base" - -LICENSE = "GPLv2+ & LGPLv2.1+ & BSD-3-Clause & BSD-4-Clause" -LICENSE_${PN}-libblkid = "LGPLv2.1+" -LICENSE_${PN}-libfdisk = "LGPLv2.1+" -LICENSE_${PN}-libmount = "LGPLv2.1+" -LICENSE_${PN}-libsmartcols = "LGPLv2.1+" -LICENSE_${PN}-libuuid = "BSD-3-Clause" - -LIC_FILES_CHKSUM = "file://README.licensing;md5=0fd5c050c6187d2bf0a4492b7f4e33da \ - file://COPYING;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ - file://Documentation/licenses/COPYING.GPL-2.0-or-later;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ - file://Documentation/licenses/COPYING.LGPL-2.1-or-later;md5=4fbd65380cdd255951079008b364516c \ - file://Documentation/licenses/COPYING.BSD-3-Clause;md5=58dcd8452651fc8b07d1f65ce07ca8af \ - file://Documentation/licenses/COPYING.BSD-4-Clause-UC;md5=263860f8968d8bafa5392cab74285262 \ - file://libuuid/COPYING;md5=6d2cafc999feb2c2de84d4d24b23290c \ - file://libmount/COPYING;md5=7c7e39fb7d70ffe5d693a643e29987c2 \ - file://libblkid/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ - file://libfdisk/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ - file://libsmartcols/COPYING;md5=693bcbbe16d3a4a4b37bc906bc01cc04 \ -" +require util-linux.inc #gtk-doc is not enabled as it requires xmlto which requires util-linux inherit autotools gettext manpages pkgconfig systemd update-alternatives python3-dir bash-completion ptest -DEPENDS = "libcap-ng ncurses virtual/crypt zlib" - -MAJOR_VERSION = "${@'.'.join(d.getVar('PV').split('.')[0:2])}" -SRC_URI = "${KERNELORG_MIRROR}/linux/utils/${BPN}/v${MAJOR_VERSION}/${BP}.tar.xz \ - file://configure-sbindir.patch \ - file://runuser.pamd \ - file://runuser-l.pamd \ - file://ptest.patch \ - file://run-ptest \ - file://display_testname_for_subtest.patch \ - file://avoid_parallel_tests.patch \ - " -SRC_URI[sha256sum] = "f7516ba9d8689343594356f0e5e1a5f0da34adfbc89023437735872bb5024c5f" +DEPENDS = "libcap-ng ncurses virtual/crypt zlib util-linux-uuid" PACKAGES =+ "${PN}-swaponoff" PACKAGES += "${@bb.utils.contains('PACKAGECONFIG', 'pylibmount', '${PN}-pylibmount', '', d)}"@@ -87,8 +49,9 @@ python util_linux_binpackages () { # we must execute before update-alternatives PACKAGE_PREPROCESS_FUNCS PACKAGE_PREPROCESS_FUNCS =+ "util_linux_binpackages " +# skip libuuid as it will be packaged by the util-linux-uuid recipe python util_linux_libpackages() { - do_split_packages(d, root=d.getVar('UTIL_LINUX_LIBDIR'), file_regex=r'^lib(.*)\.so\..*$', + do_split_packages(d, root=d.getVar('UTIL_LINUX_LIBDIR'), file_regex=r'^lib(?!uuid)(.*)\.so\..*$', output_pattern='${PN}-lib%s', description='${PN} lib%s', extra_depends='', prepend=True, allow_links=True)@@ -141,6 +104,7 @@ PACKAGECONFIG[pylibmount] = "--with-python=3 --enable-pylibmount,--without-pytho PACKAGECONFIG[readline] = "--with-readline,--without-readline,readline" # PCRE support in hardlink PACKAGECONFIG[pcre2] = ",,libpcre2" +PACKAGECONFIG[cryptsetup] = "--with-cryptsetup,--without-cryptsetup,cryptsetup" EXTRA_OEMAKE = "ARCH=${TARGET_ARCH} CPU= CPUOPT= 'OPT=${CFLAGS}'"@@ -165,6 +129,8 @@ RRECOMMENDS_${PN}_class-nativesdk = "" RDEPENDS_${PN}_class-native = "" RDEPENDS_${PN}_class-nativesdk = "" +RDEPENDS_${PN}_append += " util-linux-uuid" + RPROVIDES_${PN}-dev = "${PN}-libblkid-dev ${PN}-libmount-dev ${PN}-libuuid-dev" RDEPENDS_${PN}-bash-completion += "${PN}-lsblk"@@ -316,6 +282,11 @@ ALTERNATIVE_LINK_NAME[utmpdump.1] = "${mandir}/man1/utmpdump.1" ALTERNATIVE_LINK_NAME[uuid.3] = "${mandir}/man3/uuid.3" ALTERNATIVE_LINK_NAME[wall.1] = "${mandir}/man1/wall.1" +# dm-verity support introduces a circular build dependency, so util-linux-uuid is split out for target builds +# Need to build libuuid for uuidgen, but then delete it and let the other recipe ship it +do_install_append() { + rm -rf ${D}${includedir}/uuid ${D}${libdir}/pkgconfig/uuid.pc ${D}${libdir}/libuuid* ${D}/lib/libuuid* +} BBCLASSEXTEND = "native nativesdk" --2.29.2
Hello,
Quoted 27 lines · Luca Bocassi
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
RemovedPACKAGES_removeand instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.
Quoted 7 lines
.../util-linux/util-linux-uuid_2.36.2.bb | 22 ++++++++ meta/recipes-core/util-linux/util-linux.inc | 41 ++++++++++++++ .../util-linux/util-linux_2.36.2.bb | 53 +++++-------------- 3 files changed, 75 insertions(+), 41 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb create mode 100644 meta/recipes-core/util-linux/util-linux.inc
With this patch, there is a very weird failure on the autobuilders, two a-full builds failed with this:
ERROR: musl-1.2.2+gitAUTOINC+e5d2823631-r0 do_package: Error executing a python function in exec_python_func() autogenerated:
The stack trace of python calls that resulted in this exception/failure was:
File: 'exec_python_func() autogenerated', lineno: 2, function: <module>
0001:
*** 0002:extend_recipe_sysroot(d)
0003:File: '/home/pokybuild/yocto-worker/poky-tiny/build/meta/classes/staging.bbclass', lineno: 582, function: extend_recipe_sysroot
0578: if "/bin/" in l or "/sbin/" in l:
0579: # defer /*bin/* files until last in case they need libs
0580: binfiles[l] = (targetdir, dest)
0581: else:
*** 0582: staging_copyfile(l, targetdir, dest, postinsts, seendirs)
0583:
0584: # Handle deferred binfiles
0585: for l in binfiles:
0586: (targetdir, dest) = binfiles[l]
File: '/home/pokybuild/yocto-worker/poky-tiny/build/meta/classes/staging.bbclass', lineno: 157, function: staging_copyfile
0153: os.symlink(linkto, dest)
0154: #bb.warn(c)
0155: else:
0156: try: *** 0157: os.link(c, dest)
0158: except OSError as err:
0159: if err.errno == errno.EXDEV:
0160: bb.utils.copyfile(c, dest)
0161: else:
Exception: FileExistsError: [Errno 17] File exists: '/home/pokybuild/yocto-worker/poky-tiny/build/build/tmp/sysroots-components/core2-32/libgcc/usr/lib/i686-poky-linux-musl/10.2.0/libgcc_eh.a' -> '/home/pokybuild/yocto-worker/poky-tiny/build/build/tmp/work/core2-32-poky-linux-musl/musl/1.2.2+gitAUTOINC+e5d2823631-r0/recipe-sysroot/usr/lib/i686-poky-linux-musl/10.2.0/libgcc_eh.a'
ERROR: Logfile of failure stored in: /home/pokybuild/yocto-worker/poky-tiny/build/build/tmp/work/core2-32-poky-linux-musl/musl/1.2.2+gitAUTOINC+e5d2823631-r0/temp/log.do_package.4051328
NOTE: recipe musl-1.2.2+gitAUTOINC+e5d2823631-r0: task do_package: Failed
ERROR: Task (/home/pokybuild/yocto-worker/poky-tiny/build/meta/recipes-core/musl/musl_git.bb:do_package) failed with exit code '1'
This also happens with glibc-2.33-r0:
ERROR: glibc-2.33-r0 do_package: Error executing a python function in exec_python_func() autogenerated:
[...]
Exception: FileExistsError: [Errno 17] File exists: '/home/pokybuild/yocto-worker/edgerouter/build/build/tmp/sysroots-components/mips64/libgcc/usr/lib/mips64-poky-linux/10.2.0/libgcc.a' -> '/home/pokybuild/yocto-worker/edgerouter/build/build/tmp/work/mips64-poky-linux/glibc/2.33-r0/recipe-sysroot/usr/lib/mips64-poky-linux/10.2.0/libgcc.a'
ERROR: Logfile of failure stored in: /home/pokybuild/yocto-worker/edgerouter/build/build/tmp/work/mips64-poky-linux/glibc/2.33-r0/temp/log.do_package.29993
NOTE: recipe glibc-2.33-r0: task do_package: Failed
ERROR: Task (/home/pokybuild/yocto-worker/edgerouter/build/meta/recipes-core/glibc/glibc_2.33.bb:do_package) failed with exit code '1'
Signature · 3 lines
Alexandre Belloni, co-owner and COO, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
Quoted 75 lines · Alexandre Belloni
Hello,
Quoted 27 lines · Luca Bocassi
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
RemovedPACKAGES_removeand instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.Quoted 7 lines
.../util-linux/util-linux-uuid_2.36.2.bb | 22 ++++++++ meta/recipes-core/util-linux/util-linux.inc | 41 ++++++++++++++ .../util-linux/util-linux_2.36.2.bb | 53 +++++-------------- 3 files changed, 75 insertions(+), 41 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb create mode 100644 meta/recipes-core/util-linux/util-linux.incWith this patch, there is a very weird failure on the autobuilders, two a-full builds failed with this:
ERROR: musl-1.2.2+gitAUTOINC+e5d2823631-r0
do_package: Error executing a python function inexec_python_func()autogenerated:
The stack trace of python calls that resulted in this exception/failure was:
File: 'exec_python_func()autogenerated', lineno: 2, function: <module>0001: *** 0002:extend_recipe_sysroot(d) 0003:File: '
/home/pokybuild/yocto-worker/poky-tiny/build/meta/classes/staging.bbclass', lineno: 582, function:extend_recipe_sysroot
0578: if "/bin/" in l or "/sbin/" in l:
0579: # defer /*bin/* files until last in case they need libs
0580: binfiles[l] = (targetdir, dest)
0581: else:
*** 0582:staging_copyfile(l, targetdir, dest, postinsts, seendirs)
0583:
0584: # Handle deferred binfiles
0585: for l in binfiles:
0586: (targetdir, dest) = binfiles[l]
File: '/home/pokybuild/yocto-worker/poky-tiny/build/meta/classes/staging.bbclass', lineno: 157, function:staging_copyfile0153: os.symlink(linkto, dest) 0154: #bb.warn(c) 0155: else: 0156: try:*** 0157:
os.link(c, dest)
0158: except OSError as err:
0159: if err.errno == errno.EXDEV:
0160:bb.utils.copyfile(c, dest)
0161: else:
Exception: FileExistsError: [Errno 17] File exists: '/home/pokybuild/yocto-worker/poky-tiny/build/build/tmp/sysroots-components/core2-32/libgcc/usr/lib/i686-poky-linux-musl/10.2.0/libgcc_eh.a' -> '/home/pokybuild/yocto-worker/poky-tiny/build/build/tmp/work/core2-32-poky-linux-musl/musl/1.2.2+gitAUTOINC+e5d2823631-r0/recipe-sysroot/usr/lib/i686-poky-linux-musl/10.2.0/libgcc_eh.a'
ERROR: Logfile of failure stored in:/home/pokybuild/yocto-worker/poky-tiny/build/build/tmp/work/core2-32-poky-linux-musl/musl/1.2.2+gitAUTOINC+e5d2823631-r0/temp/log.do_package.4051328
NOTE: recipe musl-1.2.2+gitAUTOINC+e5d2823631-r0: taskdo_package: Failed
ERROR: Task (/home/pokybuild/yocto-worker/poky-tiny/build/meta/recipes-core/musl/musl_git.bb:do_package) failed with exit code '1'This also happens with glibc-2.33-r0:
Quoted 7 lines
ERROR: glibc-2.33-r0
do_package: Error executing a python function inexec_python_func()autogenerated:
[...]
Exception: FileExistsError: [Errno 17] File exists: '/home/pokybuild/yocto-worker/edgerouter/build/build/tmp/sysroots-components/mips64/libgcc/usr/lib/mips64-poky-linux/10.2.0/libgcc.a' -> '/home/pokybuild/yocto-worker/edgerouter/build/build/tmp/work/mips64-poky-linux/glibc/2.33-r0/recipe-sysroot/usr/lib/mips64-poky-linux/10.2.0/libgcc.a'
ERROR: Logfile of failure stored in:/home/pokybuild/yocto-worker/edgerouter/build/build/tmp/work/mips64-poky-linux/glibc/2.33-r0/temp/log.do_package.29993
NOTE: recipe glibc-2.33-r0: taskdo_package: Failed
ERROR: Task (/home/pokybuild/yocto-worker/edgerouter/build/meta/recipes-core/glibc/glibc_2.33.bb:do_package) failed with exit code '1'
Not sure this helps but we also get:
WARNING: nativesdk-util-linux-2.36.2-r0 do_package_qa: QA Issue: nativesdk-util-linux rdepends on nativesdk-util-linux-uuid, but it isn't a build dependency? [build-deps]
Signature · 3 lines
Alexandre Belloni, co-owner and COO, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
Quoted 84 lines · Alexandre Belloni
Quoted 75 lines · Alexandre Belloni
Hello,
Quoted 27 lines · Luca Bocassi
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
RemovedPACKAGES_removeand instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.Quoted 7 lines
.../util-linux/util-linux-uuid_2.36.2.bb | 22 ++++++++ meta/recipes-core/util-linux/util-linux.inc | 41 ++++++++++++++ .../util-linux/util-linux_2.36.2.bb | 53 +++++-------------- 3 files changed, 75 insertions(+), 41 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb create mode 100644 meta/recipes-core/util-linux/util-linux.incWith this patch, there is a very weird failure on the autobuilders, two a-full builds failed with this:
ERROR: musl-1.2.2+gitAUTOINC+e5d2823631-r0
do_package: Error executing a python function inexec_python_func()autogenerated:
The stack trace of python calls that resulted in this exception/failure was:
File: 'exec_python_func()autogenerated', lineno: 2, function: <module>0001: *** 0002:extend_recipe_sysroot(d) 0003:File: '
/home/pokybuild/yocto-worker/poky-tiny/build/meta/classes/staging.bbclass', lineno: 582, function:extend_recipe_sysroot
0578: if "/bin/" in l or "/sbin/" in l:
0579: # defer /*bin/* files until last in case they need libs
0580: binfiles[l] = (targetdir, dest)
0581: else:
*** 0582:staging_copyfile(l, targetdir, dest, postinsts, seendirs)
0583:
0584: # Handle deferred binfiles
0585: for l in binfiles:
0586: (targetdir, dest) = binfiles[l]
File: '/home/pokybuild/yocto-worker/poky-tiny/build/meta/classes/staging.bbclass', lineno: 157, function:staging_copyfile0153: os.symlink(linkto, dest) 0154: #bb.warn(c) 0155: else: 0156: try:*** 0157:
os.link(c, dest)
0158: except OSError as err:
0159: if err.errno == errno.EXDEV:
0160:bb.utils.copyfile(c, dest)
0161: else:
Exception: FileExistsError: [Errno 17] File exists: '/home/pokybuild/yocto-worker/poky-tiny/build/build/tmp/sysroots-components/core2-32/libgcc/usr/lib/i686-poky-linux-musl/10.2.0/libgcc_eh.a' -> '/home/pokybuild/yocto-worker/poky-tiny/build/build/tmp/work/core2-32-poky-linux-musl/musl/1.2.2+gitAUTOINC+e5d2823631-r0/recipe-sysroot/usr/lib/i686-poky-linux-musl/10.2.0/libgcc_eh.a'
ERROR: Logfile of failure stored in:/home/pokybuild/yocto-worker/poky-tiny/build/build/tmp/work/core2-32-poky-linux-musl/musl/1.2.2+gitAUTOINC+e5d2823631-r0/temp/log.do_package.4051328
NOTE: recipe musl-1.2.2+gitAUTOINC+e5d2823631-r0: taskdo_package: Failed
ERROR: Task (/home/pokybuild/yocto-worker/poky-tiny/build/meta/recipes-core/musl/musl_git.bb:do_package) failed with exit code '1'This also happens with glibc-2.33-r0:
Quoted 7 lines
ERROR: glibc-2.33-r0
do_package: Error executing a python function inexec_python_func()autogenerated:
[...]
Exception: FileExistsError: [Errno 17] File exists: '/home/pokybuild/yocto-worker/edgerouter/build/build/tmp/sysroots-components/mips64/libgcc/usr/lib/mips64-poky-linux/10.2.0/libgcc.a' -> '/home/pokybuild/yocto-worker/edgerouter/build/build/tmp/work/mips64-poky-linux/glibc/2.33-r0/recipe-sysroot/usr/lib/mips64-poky-linux/10.2.0/libgcc.a'
ERROR: Logfile of failure stored in:/home/pokybuild/yocto-worker/edgerouter/build/build/tmp/work/mips64-poky-linux/glibc/2.33-r0/temp/log.do_package.29993
NOTE: recipe glibc-2.33-r0: taskdo_package: Failed
ERROR: Task (/home/pokybuild/yocto-worker/edgerouter/build/meta/recipes-core/glibc/glibc_2.33.bb:do_package) failed with exit code '1'
Quoted 5 lines
Not sure this helps but we also get:
WARNING: nativesdk-util-linux-2.36.2-r0
do_package_qa: QA Issue: nativesdk-util-linux rdepends on nativesdk-util-linux-uuid, but it isn't a build dependency? [build-deps]
and I'm adding https://autobuilder.yoctoproject.org/typhoon/#/builders/47/builds/3137/steps/12/logs/stdio :
ERROR: util-linux-uuid-2.36.2-r0 do_package: QA Issue: util-linux-uuid: Files/directories were installed but not shipped in any package:
/usr/lib/libuuid.a
Please set FILES such that these items are packaged. Alternatively if they are unneeded, avoid installing them or delete them within do_install.
Signature · 3 lines
Alexandre Belloni, co-owner and COO, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
Quoted 77 lines · Alexandre Belloni
Quoted 76 lines · Alexandre Belloni
Quoted 75 lines · Alexandre Belloni
Hello,
Quoted 27 lines · Luca Bocassi
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
RemovedPACKAGES_removeand instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.Quoted 7 lines
.../util-linux/util-linux-uuid_2.36.2.bb | 22 ++++++++ meta/recipes-core/util-linux/util-linux.inc | 41 ++++++++++++++ .../util-linux/util-linux_2.36.2.bb | 53 +++++-------------- 3 files changed, 75 insertions(+), 41 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb create mode 100644 meta/recipes-core/util-linux/util-linux.incWith this patch, there is a very weird failure on the autobuilders, two a-full builds failed with this:
ERROR: musl-1.2.2+gitAUTOINC+e5d2823631-r0
do_package: Error executing a python function inexec_python_func()autogenerated:
The stack trace of python calls that resulted in this exception/failure was:
File: 'exec_python_func()autogenerated', lineno: 2, function: <module>0001: *** 0002:extend_recipe_sysroot(d) 0003:File: '
/home/pokybuild/yocto-worker/poky-tiny/build/meta/classes/staging.bbclass', lineno: 582, function:extend_recipe_sysroot
0578: if "/bin/" in l or "/sbin/" in l:
0579: # defer /*bin/* files until last in case they need libs
0580: binfiles[l] = (targetdir, dest)
0581: else:
*** 0582:staging_copyfile(l, targetdir, dest, postinsts, seendirs)
0583:
0584: # Handle deferred binfiles
0585: for l in binfiles:
0586: (targetdir, dest) = binfiles[l]
File: '/home/pokybuild/yocto-worker/poky-tiny/build/meta/classes/staging.bbclass', lineno: 157, function:staging_copyfile0153: os.symlink(linkto, dest) 0154: #bb.warn(c) 0155: else: 0156: try:*** 0157:
os.link(c, dest)
0158: except OSError as err:
0159: if err.errno == errno.EXDEV:
0160:bb.utils.copyfile(c, dest)
0161: else:
Exception: FileExistsError: [Errno 17] File exists: '/home/pokybuild/yocto-worker/poky-tiny/build/build/tmp/sysroots-components/core2-32/libgcc/usr/lib/i686-poky-linux-musl/10.2.0/libgcc_eh.a' -> '/home/pokybuild/yocto-worker/poky-tiny/build/build/tmp/work/core2-32-poky-linux-musl/musl/1.2.2+gitAUTOINC+e5d2823631-r0/recipe-sysroot/usr/lib/i686-poky-linux-musl/10.2.0/libgcc_eh.a'
ERROR: Logfile of failure stored in:/home/pokybuild/yocto-worker/poky-tiny/build/build/tmp/work/core2-32-poky-linux-musl/musl/1.2.2+gitAUTOINC+e5d2823631-r0/temp/log.do_package.4051328
NOTE: recipe musl-1.2.2+gitAUTOINC+e5d2823631-r0: taskdo_package: Failed
ERROR: Task (/home/pokybuild/yocto-worker/poky-tiny/build/meta/recipes-core/musl/musl_git.bb:do_package) failed with exit code '1'This also happens with glibc-2.33-r0:
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
ERROR: glibc-2.33-r0
do_package: Error executing a python function inexec_python_func()autogenerated:
[...]
Exception: FileExistsError: [Errno 17] File exists: '/home/pokybuild/yocto-worker/edgerouter/build/build/tmp/sysroots-components/mips64/libgcc/usr/lib/mips64-poky-linux/10.2.0/libgcc.a' -> '/home/pokybuild/yocto-worker/edgerouter/build/build/tmp/work/mips64-poky-linux/glibc/2.33-r0/recipe-sysroot/usr/lib/mips64-poky-linux/10.2.0/libgcc.a'
ERROR: Logfile of failure stored in:/home/pokybuild/yocto-worker/edgerouter/build/build/tmp/work/mips64-poky-linux/glibc/2.33-r0/temp/log.do_package.29993
NOTE: recipe glibc-2.33-r0: taskdo_package: Failed
ERROR: Task (/home/pokybuild/yocto-worker/edgerouter/build/meta/recipes-core/glibc/glibc_2.33.bb:do_package) failed with exit code '1'
I see those kind of errors without this patch all the time, when a build directory is hosed, and normally it means it's time for a rm -rf and start over. It happens quite often when building in the same directory multiple times. Are you sure it's related to this particular patch?
Quoted 3 lines
Quoted 3 lines
Not sure this helps but we also get:
WARNING: nativesdk-util-linux-2.36.2-r0
do_package_qa: QA Issue: nativesdk-util-linux rdepends on nativesdk-util-linux-uuid, but it isn't a build dependency? [build-deps]
Uhm I don't understand this warning, it _is_ a build dependency? Any idea?
-DEPENDS = "libcap-ng ncurses virtual/crypt zlib" +DEPENDS = "libcap-ng ncurses virtual/crypt zlib util-linux-uuid"
Quoted 5 lines
and I'm adding https://autobuilder.yoctoproject.org/typhoon/#/builders/47/builds/3137/steps/12/logs/stdio :
ERROR: util-linux-uuid-2.36.2-r0
do_package: QA Issue: util-linux-uuid: Files/directories were installed but not shipped in any package:
/usr/lib/libuuid.a
Please set FILES such that these items are packaged. Alternatively if they are unneeded, avoid installing them or delete them withindo_install.
Whops, indeed I forgot the static archive, fixed now.
Signature · 2 lines
Kind regards, Luca Boccassi
Hi,
Quoted 8 lines · Luca Boccassi
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
ERROR: glibc-2.33-r0
do_package: Error executing a python function inexec_python_func()autogenerated:
[...]
Exception: FileExistsError: [Errno 17] File exists: '/home/pokybuild/yocto-worker/edgerouter/build/build/tmp/sysroots-components/mips64/libgcc/usr/lib/mips64-poky-linux/10.2.0/libgcc.a' -> '/home/pokybuild/yocto-worker/edgerouter/build/build/tmp/work/mips64-poky-linux/glibc/2.33-r0/recipe-sysroot/usr/lib/mips64-poky-linux/10.2.0/libgcc.a'
ERROR: Logfile of failure stored in:/home/pokybuild/yocto-worker/edgerouter/build/build/tmp/work/mips64-poky-linux/glibc/2.33-r0/temp/log.do_package.29993
NOTE: recipe glibc-2.33-r0: taskdo_package: Failed
ERROR: Task (/home/pokybuild/yocto-worker/edgerouter/build/meta/recipes-core/glibc/glibc_2.33.bb:do_package) failed with exit code '1'
Quoted 6 lines
I see those kind of errors without this patch all the time, when a build directory is hosed, and normally it means it's time for a rm -rf and start over. It happens quite often when building in the same directory multiple times. Are you sure it's related to this particular patch?
No, not completely sure, however, the issue was gone as soon as the patch has been removed. Richard may have more insight.
Quoted 7 lines
Quoted 3 lines
Quoted 3 lines
Not sure this helps but we also get:
WARNING: nativesdk-util-linux-2.36.2-r0
do_package_qa: QA Issue: nativesdk-util-linux rdepends on nativesdk-util-linux-uuid, but it isn't a build dependency? [build-deps]Uhm I don't understand this warning, it _is_ a build dependency? Any idea?
This was the failing build if you want to look at the config to try to reproduce:
https://autobuilder.yoctoproject.org/typhoon/#/builders/20/builds/3440
Quoted 14 lines
meta/recipes-core/util-linux/util-linux_2.36.2.bb +1
-1-DEPENDS = "libcap-ng ncurses virtual/crypt zlib" +DEPENDS = "libcap-ng ncurses virtual/crypt zlib util-linux-uuid"Quoted 6 lines
and I'm adding https://autobuilder.yoctoproject.org/typhoon/#/builders/47/builds/3137/steps/12/logs/stdio :
ERROR: util-linux-uuid-2.36.2-r0
do_package: QA Issue: util-linux-uuid: Files/directories were installed but not shipped in any package:
/usr/lib/libuuid.a
Please set FILES such that these items are packaged. Alternatively if they are unneeded, avoid installing them or delete them withindo_install.Whops, indeed I forgot the static archive, fixed now.
Signature · 2 lines
Kind regards, Luca Boccassi
Signature · 3 lines
Alexandre Belloni, co-owner and COO, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
Quoted 9 lines · On Thu, 2021-02-25 at 15:30 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb new file +0
-0@@ -0,0 +1,22 @@
[...]
Quoted 10 lines
meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb new file +10
-0+S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +DEBIAN_NOAUTONAME_util-linux-libuuid = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dev = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dbg = "1" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" +
[...]
meta/recipes-core/util-linux/util-linux_2.36.2.bb RRECOMMENDS_${PN}_class-nativesdk = "" · line 133
Quoted 9 lines
meta/recipes-core/util-linux/util-linux_2.36.2.bb +2
-0@@ -1,46 +1,8 @@ +RDEPENDS_${PN}_append += " util-linux-uuid" +
I know this patch has caused a bit of confusion as it blew up in testing. I can't explain why you didn't see that but I did spot why its breaking. I've cut this email to the bare bones of the problem for clarity.
In util-linux-uuid, you set PACKAGES as above with util-linux-libuuid.
In util-linux, you set it to refer to a util-linux-uuid package which does not exist. It needs s/uuid/libuuid/. RDEPENDS refer to PACKAGES namespace so need to be consistent.
Why the tools haven't been clearer about the issue, I don't know but that is no doubt causing many of the issues.
Cheers,
Richard
Quoted 42 lines · Richard Purdie
Quoted 9 lines · On Thu, 2021-02-25 at 15:30 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb new file +0
-0@@ -0,0 +1,22 @@[...]
Quoted 10 lines
meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb new file +10
-0+S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +DEBIAN_NOAUTONAME_util-linux-libuuid = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dev = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dbg = "1" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" +[...]
meta/recipes-core/util-linux/util-linux_2.36.2.bb RRECOMMENDS_${PN}_class-nativesdk = "" · line 133
Quoted 9 lines
meta/recipes-core/util-linux/util-linux_2.36.2.bb +2
-0@@ -1,46 +1,8 @@ +RDEPENDS_${PN}_append += " util-linux-uuid" +I know this patch has caused a bit of confusion as it blew up in testing. I can't explain why you didn't see that but I did spot why its breaking. I've cut this email to the bare bones of the problem for clarity.
In util-linux-uuid, you set PACKAGES as above with util-linux-libuuid.
In util-linux, you set it to refer to a util-linux-uuid package which does not exist. It needs s/uuid/libuuid/. RDEPENDS refer to PACKAGES namespace so need to be consistent.
Quoted 6 lines
Why the tools haven't been clearer about the issue, I don't know but that is no doubt causing many of the issues.
Cheers,
Richard
Thanks for the suggestion, I had completely missed it - I get confused all the time w.r.t. using the source package name or the binary package names in the various fields.
I'll shortly send v6 with the following changes:
- install .a in libuuid-dev
- change rdepends on uuid /uuid/libuuid/
- add rdepends on uuid-dev to libuuid-dev
- remove rprovides of libuuid-dev from uuid-dev
The local.conf I am using to test this adds the following lines on top of the default that's auto-generated:
PACKAGECONFIG_append_pn-util-linux += " cryptsetup" PACKAGECONFIG_remove_pn-cryptsetup += " blkid" PREFERRED_PROVIDER_virtual/kernel = "linux-dummy"
Not sure why the issue wasn't seen, I suspect there's some additional packages that need to be included for it to manifest, although I'm not sure which.
Signature · 2 lines
Kind regards, Luca Boccassi
PATCH v6 1/1 util-linux: split uuid in separate recipe to allow bootstrapping
In reply to Luca BocassiFrom: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.
https://github.com/karelzak/util-linux/pull/898
---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
Removed PACKAGES_remove and instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.
v6: install libuuid.a in libuuid-dev
change rdepends on uuid to use binary package rather than source
add rdepends on uuid-dev to libuuid-dev
remove rprovides of libuuid-dev from uuid-dev
.../util-linux/util-linux-uuid_2.36.2.bb | 22 ++++++++ meta/recipes-core/util-linux/util-linux.inc | 41 ++++++++++++++ .../util-linux/util-linux_2.36.2.bb | 56 +++++-------------- 3 files changed, 77 insertions(+), 42 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb create mode 100644 meta/recipes-core/util-linux/util-linux.inc
meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb new file +22 -0
@@ -0,0 +1,22 @@ +# To allow util-linux to optionally build-depend on cryptsetup, libuuid is +# split out of the main recipe, as it's needed by cryptsetup + +require util-linux.inc + +inherit autotools gettext pkgconfig + +S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +DEBIAN_NOAUTONAME_util-linux-libuuid = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dev = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dbg = "1" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.a ${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" + +do_install_append() { + rm -rf ${D}${datadir} ${D}${bindir} ${D}${base_bindir} ${D}${sbindir} ${D}${base_sbindir} ${D}${exec_prefix}/sbin +} + +BBCLASSEXTEND = "native nativesdk"
meta/recipes-core/util-linux/util-linux.inc new file +41 -0
41 lines not on this page · raw message
meta/recipes-core/util-linux/util-linux_2.36.2.bb +14 -42
87 lines not on this page · raw message
Signature · 1 line
2.29.2
meta/recipes-core/util-linux/util-linux_2.36.2.bb RRECOMMENDS_${PN}_class-nativesdk = "" · line 133
Quoted 4 lines · On Thu, 2021-03-04 at 12:27 +0000, [email protected] wrote:
meta/recipes-core/util-linux/util-linux_2.36.2.bb +2
-1-RPROVIDES_${PN}-dev = "${PN}-libblkid-dev ${PN}-libmount-dev ${PN}-libuuid-dev" +RDEPENDS_${PN}_append += " util-linux-libuuid" +RDEPENDS_${PN}-dev_append += " util-linux-libuuid-dev"
The mysterious libgcov.a issues which appeared in builds when this patch was added are due to the lines above. Changing them to:
RDEPENDS_${PN} += " util-linux-libuuid"
RDEPENDS_${PN}-dev += " util-linux-libuuid-dev"
resolves the issue. To illustrate, have a look at "bitbake glibc -g" in the task-depends.dot file before and after that change.
The problem is the native BBCLASSEXTEND code can't handle the RDEPENDS_${PN}_append correctly and it means in the native case it adds a dependency on util-linux-libuuid rather than a native version.
Why does that cause the gcov error? Its because there are early toolchain util-linux-native dependencies which ends up meaning glibc sees a libgcc dependency. It usually uses libgcc-initial and that has overlapping files with libgcc and libgcov.a is the first in the sorted list.
Again, we should have better errors for a native recipe depending upon a target one and show better errors for this configuration issue. It does at least explain the seemingly strange errors.
Cheers,
Richard
Quoted 19 lines · Richard Purdie
meta/recipes-core/util-linux/util-linux_2.36.2.bb RRECOMMENDS_${PN}_class-nativesdk = "" · line 133
Quoted 3 lines · On Thu, 2021-03-04 at 12:27 +0000, [email protected] wrote:
meta/recipes-core/util-linux/util-linux_2.36.2.bb +2
-1-RPROVIDES_${PN}-dev = "${PN}-libblkid-dev ${PN}-libmount-dev ${PN}-libuuid-dev" +RDEPENDS_${PN}_append += " util-linux-libuuid" +RDEPENDS_${PN}-dev_append += " util-linux-libuuid-dev"The mysterious libgcov.a issues which appeared in builds when this patch was added are due to the lines above. Changing them to:
RDEPENDS_${PN} += " util-linux-libuuid"
RDEPENDS_${PN}-dev += " util-linux-libuuid-dev"resolves the issue. To illustrate, have a look at "bitbake glibc -g" in the task-depends.dot file before and after that change.
The problem is the native BBCLASSEXTEND code can't handle the RDEPENDS_${PN}_append correctly and it means in the native case it adds a dependency on util-linux-libuuid rather than a native version.
Quoted 8 lines
Why does that cause the gcov error? Its because there are early toolchain util-linux-native dependencies which ends up meaning glibc sees a libgcc dependency. It usually uses libgcc-initial and that has overlapping files with libgcc and libgcov.a is the first in the sorted list.
Again, we should have better errors for a native recipe depending upon a target one and show better errors for this configuration issue. It does at least explain the seemingly strange errors.
Thank you! Sent v7 as suggested after a quick rebuild check.
When to use '_append' vs '+=' is one of the things I'm never sure about and invariably I get it wrong most of the times.
Signature · 2 lines
Kind regards, Luca Boccassi
PATCH v7 1/1 util-linux: split uuid in separate recipe to allow bootstrapping
In reply to Luca BocassiFrom: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.
https://github.com/karelzak/util-linux/pull/898
---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
Removed PACKAGES_remove and instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.
v6: install libuuid.a in libuuid-dev
change rdepends on uuid to use binary package rather than source
add rdepends on uuid-dev to libuuid-dev
remove rprovides of libuuid-dev from uuid-dev
v7: do not use '_append' for RDEPENDS, as it is not supported by BBCLASSEXTEND, use
simply '+='
.../util-linux/util-linux-uuid_2.36.2.bb | 22 ++++++++ meta/recipes-core/util-linux/util-linux.inc | 41 ++++++++++++++ .../util-linux/util-linux_2.36.2.bb | 56 +++++-------------- 3 files changed, 77 insertions(+), 42 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb create mode 100644 meta/recipes-core/util-linux/util-linux.inc
meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb new file +22 -0
@@ -0,0 +1,22 @@ +# To allow util-linux to optionally build-depend on cryptsetup, libuuid is +# split out of the main recipe, as it's needed by cryptsetup + +require util-linux.inc + +inherit autotools gettext pkgconfig + +S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +DEBIAN_NOAUTONAME_util-linux-libuuid = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dev = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dbg = "1" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.a ${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" + +do_install_append() { + rm -rf ${D}${datadir} ${D}${bindir} ${D}${base_bindir} ${D}${sbindir} ${D}${base_sbindir} ${D}${exec_prefix}/sbin +} + +BBCLASSEXTEND = "native nativesdk"
meta/recipes-core/util-linux/util-linux.inc new file +41 -0
41 lines not on this page · raw message
meta/recipes-core/util-linux/util-linux_2.36.2.bb +14 -42
87 lines not on this page · raw message
Signature · 1 line
2.29.2
Quoted 24 lines · On Fri, 2021-03-05 at 11:02 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid) and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid. RemovedPACKAGES_removeand instead filter out libuuid via the package generation
Quoted 8 lines
regex.
Rebased on util-linux 2.36.2.
v6: install libuuid.a in libuuid-dev
change rdepends on uuid to use binary package rather than source add rdepends on uuid-dev to libuuid-dev
remove rprovides of libuuid-dev from uuid-dev
v7: do not use '_append' for RDEPENDS, as it is not supported by BBCLASSEXTEND, use simply '+='
This is looking closer but not there yet I'm afraid. Testing showed:
https://autobuilder.yoctoproject.org/typhoon/#/builders/57/builds/3152
https://autobuilder.yoctoproject.org/typhoon/#/builders/47/builds/3188
so far.
Cheers,
Richard
Quoted 36 lines · Richard Purdie
Quoted 31 lines · On Fri, 2021-03-05 at 11:02 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
RemovedPACKAGES_removeand instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.
v6: install libuuid.a in libuuid-dev
change rdepends on uuid to use binary package rather than source
add rdepends on uuid-dev to libuuid-dev
remove rprovides of libuuid-dev from uuid-dev
v7: do not use '_append' for RDEPENDS, as it is not supported by BBCLASSEXTEND, useQuoted 1 line
simply '+='This is looking closer but not there yet I'm afraid. Testing showed:
Quoted 8 lines
https://autobuilder.yoctoproject.org/typhoon/#/builders/57/builds/3152
https://autobuilder.yoctoproject.org/typhoon/#/builders/47/builds/3188so far.
Cheers,
Richard
Thank you, these seems to be easy fixes: -staticdev package was missing (forgot it's separate from -dev) and there's a libx32 directory to take care of. Will do a quick build test and send v8.
Signature · 2 lines
Kind regards, Luca Boccassi
Quoted 34 lines · On Mon, 08 Mar 2021 19:29:20 +0000 "Richard Purdie"…
Quoted 24 lines · On Fri, 2021-03-05 at 11:02 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid) and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency. Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see
https://lists.yoctoproject.org/g/yocto/message/47570 v3: rebased and refactored to have a common util-linux.inc file v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid. RemovedQuoted 8 lines
PACKAGES_removeand instead filter out libuuid via the package generation regex. Rebased on util-linux 2.36.2.
v6: install libuuid.a in libuuid-dev
change rdepends on uuid to use binary package rather than source add rdepends on uuid-dev to libuuid-dev
remove rprovides of libuuid-dev from uuid-dev
v7: do not use '_append' for RDEPENDS, as it is not supported by BBCLASSEXTEND, use simply '+='
Quoted 5 lines
This is looking closer but not there yet I'm afraid. Testing showed:
https://autobuilder.yoctoproject.org/typhoon/#/builders/57/builds/3152
https://autobuilder.yoctoproject.org/typhoon/#/builders/47/builds/3188
It miss also the Maintainer description for linux-util-uuid recipe:
https://autobuilder.yoctoproject.org/typhoon/#/builders/86/builds/1901/steps/14/logs/stdio
Regards,
Köry
Quoted 43 lines · Köry Maincent
Quoted 34 lines · On Mon, 08 Mar 2021 19:29:20 +0000 "Richard Purdie"…
Quoted 24 lines · On Fri, 2021-03-05 at 11:02 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid) and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency. Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see
https://lists.yoctoproject.org/g/yocto/message/47570 v3: rebased and refactored to have a common util-linux.inc file v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid. RemovedQuoted 8 lines
PACKAGES_removeand instead filter out libuuid via the package generation regex. Rebased on util-linux 2.36.2.
v6: install libuuid.a in libuuid-dev
change rdepends on uuid to use binary package rather than source add rdepends on uuid-dev to libuuid-dev
remove rprovides of libuuid-dev from uuid-dev
v7: do not use '_append' for RDEPENDS, as it is not supported by BBCLASSEXTEND, use simply '+='Quoted 5 lines
This is looking closer but not there yet I'm afraid. Testing showed:
https://autobuilder.yoctoproject.org/typhoon/#/builders/57/builds/3152
https://autobuilder.yoctoproject.org/typhoon/#/builders/47/builds/3188
Quoted 6 lines
It miss also the Maintainer description for linux-util-uuid recipe:
https://autobuilder.yoctoproject.org/typhoon/#/builders/86/builds/1901/steps/14/logs/stdioRegards,
Köry
Hi,
The DESCRIPTION field is in the util-linux.inc which is included by the recipe. Or is that something else entirely? What is it, precisely?
Signature · 2 lines
Kind regards, Luca Boccassi
Quoted 4 lines · On Tue, 09 Mar 2021 13:26:55 +0000 Luca Boccassi <[email protected]>…
Quoted 4 lines
It miss also the Maintainer description for linux-util-uuid recipe:
https://autobuilder.yoctoproject.org/typhoon/#/builders/86/builds/1901/steps/14/logs/stdio
Quoted 9 lines
Quoted 3 lines
Regards,
Köry
Hi,
The DESCRIPTION field is in the util-linux.inc which is included by the recipe. Or is that something else entirely? What is it, precisely?
Sorry I was not enough explicit.
When you add a new recipe to oe-core you need to update the file:
meta/conf/distro/include/maintainers.inc
Regards,
Köry
Quoted 13 lines · Köry Maincent
Quoted 10 lines · On Tue, 09 Mar 2021 13:26:55 +0000 Luca Boccassi <[email protected]>…
Quoted 7 lines
It miss also the Maintainer description for linux-util-uuid recipe:
https://autobuilder.yoctoproject.org/typhoon/#/builders/86/builds/1901/steps/14/logs/stdioRegards,
Köry
Hi,
Quoted 7 lines
Quoted 3 lines
The DESCRIPTION field is in the util-linux.inc which is included by the recipe. Or is that something else entirely? What is it, precisely?
Sorry I was not enough explicit.
When you add a new recipe to oe-core you need to update the file:
meta/conf/distro/include/maintainers.inc
The test which failed can be run as:
oe-selftest -r distrodata.Distrodata.test_maintainers
Cheers,
Richard
Quoted 21 lines · Richard Purdie
Quoted 12 lines · Köry Maincent
Quoted 9 lines · On Tue, 09 Mar 2021 13:26:55 +0000 Luca Boccassi <[email protected]>…
Quoted 6 lines
It miss also the Maintainer description for linux-util-uuid recipe:
https://autobuilder.yoctoproject.org/typhoon/#/builders/86/builds/1901/steps/14/logs/stdioRegards,
Köry
Hi,
Quoted 7 lines
Quoted 3 lines
The DESCRIPTION field is in the util-linux.inc which is included by the recipe. Or is that something else entirely? What is it, precisely?
Sorry I was not enough explicit.
When you add a new recipe to oe-core you need to update the file:
meta/conf/distro/include/maintainers.inc
Quoted 7 lines
The test which failed can be run as:
oe-selftest -r distrodata.Distrodata.test_maintainers
Cheers,
Richard
Got it, thanks. Sent v9, set to the same value as util-linux for now, since the sources are shared.
Signature · 2 lines
Kind regards, Luca Boccassi
PATCH v8 1/1 util-linux: split uuid in separate recipe to allow bootstrapping
In reply to Luca BocassiFrom: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.
https://github.com/karelzak/util-linux/pull/898
---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
Removed PACKAGES_remove and instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.
v6: install libuuid.a in libuuid-dev
change rdepends on uuid to use binary package rather than source
add rdepends on uuid-dev to libuuid-dev
remove rprovides of libuuid-dev from uuid-dev
v7: do not use '_append' for RDEPENDS, as it is not supported by BBCLASSEXTEND, use
simply '+='
v8: added missing libuuid-staticdev package
remove libx32/libuuid* too
.../util-linux/util-linux-uuid_2.36.2.bb | 24 ++++++++ meta/recipes-core/util-linux/util-linux.inc | 41 +++++++++++++ .../util-linux/util-linux_2.36.2.bb | 57 +++++-------------- 3 files changed, 80 insertions(+), 42 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb create mode 100644 meta/recipes-core/util-linux/util-linux.inc
meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb new file +24 -0
@@ -0,0 +1,24 @@ +# To allow util-linux to optionally build-depend on cryptsetup, libuuid is +# split out of the main recipe, as it's needed by cryptsetup + +require util-linux.inc + +inherit autotools gettext pkgconfig + +S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +DEBIAN_NOAUTONAME_util-linux-libuuid = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dev = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-staticdev = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dbg = "1" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-staticdev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-staticdev = "${libdir}/libuuid.a" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" + +do_install_append() { + rm -rf ${D}${datadir} ${D}${bindir} ${D}${base_bindir} ${D}${sbindir} ${D}${base_sbindir} ${D}${exec_prefix}/sbin +} + +BBCLASSEXTEND = "native nativesdk"
meta/recipes-core/util-linux/util-linux.inc new file +41 -0
41 lines not on this page · raw message
meta/recipes-core/util-linux/util-linux_2.36.2.bb +15 -42
88 lines not on this page · raw message
Signature · 1 line
2.29.2
PATCH v9 1/1 util-linux: split uuid in separate recipe to allow bootstrapping
In reply to Luca BocassiFrom: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.
https://github.com/karelzak/util-linux/pull/898
---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
Removed PACKAGES_remove and instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.
v6: install libuuid.a in libuuid-dev
change rdepends on uuid to use binary package rather than source
add rdepends on uuid-dev to libuuid-dev
remove rprovides of libuuid-dev from uuid-dev
v7: do not use '_append' for RDEPENDS, as it is not supported by BBCLASSEXTEND, use
simply '+='
v8: added missing libuuid-staticdev package
remove libx32/libuuid* too
v9: set RECIPE_MAINTAINER_pn-util-linux-uuid, same value as RECIPE_MAINTAINER_pn-util-linux
meta/conf/distro/include/maintainers.inc | 1 + .../util-linux/util-linux-uuid_2.36.2.bb | 24 ++++++++ meta/recipes-core/util-linux/util-linux.inc | 41 +++++++++++++ .../util-linux/util-linux_2.36.2.bb | 57 +++++-------------- 4 files changed, 81 insertions(+), 42 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb create mode 100644 meta/recipes-core/util-linux/util-linux.inc
meta/conf/distro/include/maintainers.inc +1 -0
@@ -720,6 +720,7 @@ RECIPE_MAINTAINER_pn-update-rc.d = "Ross Burton <[email protected]>" RECIPE_MAINTAINER_pn-usbinit = "Alexander Kanavin <[email protected]>" RECIPE_MAINTAINER_pn-usbutils = "Alexander Kanavin <[email protected]>" RECIPE_MAINTAINER_pn-util-linux = "Chen Qi <[email protected]>" +RECIPE_MAINTAINER_pn-util-linux-uuid = "Chen Qi <[email protected]>" RECIPE_MAINTAINER_pn-util-macros = "Armin Kuster <[email protected]>" RECIPE_MAINTAINER_pn-v86d = "Alexander Kanavin <[email protected]>" RECIPE_MAINTAINER_pn-vala = "Alexander Kanavin <[email protected]>"
meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb new file +24 -0
@@ -0,0 +1,24 @@ +# To allow util-linux to optionally build-depend on cryptsetup, libuuid is +# split out of the main recipe, as it's needed by cryptsetup + +require util-linux.inc + +inherit autotools gettext pkgconfig + +S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +DEBIAN_NOAUTONAME_util-linux-libuuid = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dev = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-staticdev = "1" +DEBIAN_NOAUTONAME_util-linux-libuuid-dbg = "1" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-staticdev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-staticdev = "${libdir}/libuuid.a" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" + +do_install_append() { + rm -rf ${D}${datadir} ${D}${bindir} ${D}${base_bindir} ${D}${sbindir} ${D}${base_sbindir} ${D}${exec_prefix}/sbin +} + +BBCLASSEXTEND = "native nativesdk"
meta/recipes-core/util-linux/util-linux.inc new file +41 -0
41 lines not on this page · raw message
meta/recipes-core/util-linux/util-linux_2.36.2.bb +15 -42
88 lines not on this page · raw message
Signature · 1 line
2.29.2
Quoted 27 lines · On Tue, 2021-03-09 at 13:56 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid) and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid. RemovedPACKAGES_removeand instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.
v6: install libuuid.a in libuuid-dev
Quoted 8 lines
change rdepends on uuid to use binary package rather than source
add rdepends on uuid-dev to libuuid-dev
remove rprovides of libuuid-dev from uuid-dev
v7: do not use '_append' for RDEPENDS, as it is not supported by BBCLASSEXTEND, use simply '+='
v8: added missing libuuid-staticdev package
remove libx32/libuuid* too
v9: set RECIPE_MAINTAINER_pn-util-linux-uuid, same value as RECIPE_MAINTAINER_pn-util-linux
Sorry, there are still issues:
https://autobuilder.yoctoproject.org/typhoon/#/builders/44/builds/3207
https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1608
https://autobuilder.yoctoproject.org/typhoon/#/builders/52/builds/3159
:(
I really want to get this into the release but we're past feature freeze and running out of opportunity as each broken test run is delaying other patches.
I'm a little unclear if debian renaming was allowed before the split and if so, why it is disabled in the split version. I suspect there is a missing MLPREFIX somewhere causing the multilib issue too.
Cheers,
Richard
Quoted 39 lines · Richard Purdie
Quoted 31 lines · On Tue, 2021-03-09 at 13:56 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
RemovedPACKAGES_removeand instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.
v6: install libuuid.a in libuuid-dev
change rdepends on uuid to use binary package rather than source
add rdepends on uuid-dev to libuuid-dev
remove rprovides of libuuid-dev from uuid-dev
v7: do not use '_append' for RDEPENDS, as it is not supported by BBCLASSEXTEND, useQuoted 4 lines
simply '+='v8: added missing libuuid-staticdev package
remove libx32/libuuid* too
v9: set RECIPE_MAINTAINER_pn-util-linux-uuid, same value as RECIPE_MAINTAINER_pn-util-linuxSorry, there are still issues:
Quoted 8 lines
https://autobuilder.yoctoproject.org/typhoon/#/builders/44/builds/3207
https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1608
https://autobuilder.yoctoproject.org/typhoon/#/builders/52/builds/3159:(
I really want to get this into the release but we're past feature freeze and running out of opportunity as each broken test run is delaying other patches.
No problem at all, it's absolutely fine to postpone, I am in no immediate rush as I have local backports in use. As long as we get things sorted at some point it's all good for me, so please don't delay other patches because of this.
Quoted 3 lines
I'm a little unclear if debian renaming was allowed before the split and if so, why it is disabled in the split version. I suspect there is a missing MLPREFIX somewhere causing the multilib issue too.
I do not remember why the DEBIAN_NOAUTONAME was added at this point. I'll try to send a new version tomorrow.
Signature · 2 lines
Kind regards, Luca Boccassi
Quoted 54 lines · Luca Boccassi
Quoted 39 lines · Richard Purdie
Quoted 31 lines · On Tue, 2021-03-09 at 13:56 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
RemovedPACKAGES_removeand instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.
v6: install libuuid.a in libuuid-dev
change rdepends on uuid to use binary package rather than source
add rdepends on uuid-dev to libuuid-dev
remove rprovides of libuuid-dev from uuid-dev
v7: do not use '_append' for RDEPENDS, as it is not supported by BBCLASSEXTEND, useQuoted 4 lines
simply '+='v8: added missing libuuid-staticdev package
remove libx32/libuuid* too
v9: set RECIPE_MAINTAINER_pn-util-linux-uuid, same value as RECIPE_MAINTAINER_pn-util-linuxSorry, there are still issues:
Quoted 8 lines
https://autobuilder.yoctoproject.org/typhoon/#/builders/44/builds/3207
https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/1608
https://autobuilder.yoctoproject.org/typhoon/#/builders/52/builds/3159:(
I really want to get this into the release but we're past feature freeze and running out of opportunity as each broken test run is delaying other patches.
No problem at all, it's absolutely fine to postpone, I am in no immediate rush as I have local backports in use. As long as we get things sorted at some point it's all good for me, so please don't delay other patches because of this.
Quoted 6 lines
Quoted 3 lines
I'm a little unclear if debian renaming was allowed before the split and if so, why it is disabled in the split version. I suspect there is a missing MLPREFIX somewhere causing the multilib issue too.
I do not remember why the
DEBIAN_NOAUTONAMEwas added at this point. I'll try to send a new version tomorrow.
So without DEBIAN_NOAUTONAME these are the package names:
libuuid1-2.36.2-r0.core2_64.rpm
libuuid-dbg-2.36.2-r0.core2_64.rpm
libuuid-dev-2.36.2-r0.core2_64.rpm
util-linux-uuid-src-2.36.2-r0.core2_64.rpm
that's going to be a problem as they don't match anymore, right? Since they used to be util-linux-libuuid, etc. Although the poky image build completed just fine...
Signature · 2 lines
Kind regards, Luca Boccassi
Quoted 18 lines · Luca Boccassi
Quoted 8 lines · Luca Boccassi
On Tue, 2021-03-09 at 23:43 +0000, Richard Purdie wrote:
No problem at all, it's absolutely fine to postpone, I am in no immediate rush as I have local backports in use. As long as we get things sorted at some point it's all good for me, so please don't delay other patches because of this.
Quoted 6 lines
Quoted 3 lines
I'm a little unclear if debian renaming was allowed before the split and if so, why it is disabled in the split version. I suspect there is a missing MLPREFIX somewhere causing the multilib issue too.
I do not remember why the
DEBIAN_NOAUTONAMEwas added at this point. I'll try to send a new version tomorrow.So without
DEBIAN_NOAUTONAMEthese are the package names:
Quoted 8 lines
libuuid1-2.36.2-r0.core2_64.rpm
libuuid-dbg-2.36.2-r0.core2_64.rpm
libuuid-dev-2.36.2-r0.core2_64.rpm
util-linux-uuid-src-2.36.2-r0.core2_64.rpmthat's going to be a problem as they don't match anymore, right? Since they used to be util-linux-libuuid, etc. Although the poky image build completed just fine...
The above looks fine, they don't have to match. I think this would match what happens to the package before your patch too although I've not checked.
Cheers,
Richard
Quoted 24 lines · Richard Purdie
Quoted 23 lines · Luca Boccassi
Quoted 8 lines · Luca Boccassi
On Tue, 2021-03-09 at 23:43 +0000, Richard Purdie wrote:
No problem at all, it's absolutely fine to postpone, I am in no immediate rush as I have local backports in use. As long as we get things sorted at some point it's all good for me, so please don't delay other patches because of this.
Quoted 6 lines
Quoted 3 lines
I'm a little unclear if debian renaming was allowed before the split and if so, why it is disabled in the split version. I suspect there is a missing MLPREFIX somewhere causing the multilib issue too.
I do not remember why the
DEBIAN_NOAUTONAMEwas added at this point. I'll try to send a new version tomorrow.So without
DEBIAN_NOAUTONAMEthese are the package names:libuuid1-2.36.2-r0.core2_64.rpm
libuuid-dbg-2.36.2-r0.core2_64.rpm
libuuid-dev-2.36.2-r0.core2_64.rpm
util-linux-uuid-src-2.36.2-r0.core2_64.rpm
Quoted 8 lines
Quoted 3 lines
that's going to be a problem as they don't match anymore, right? Since they used to be util-linux-libuuid, etc. Although the poky image build completed just fine...
The above looks fine, they don't have to match. I think this would match what happens to the package before your patch too although I've not checked.
Ok, sent v10.
The other issue was a missing rm -f ${D}/lib64/libuuid* - is there a better way to handle all of this than this line? ${libdir} doesn't seem to match in some cases so I have to manually specify lib/libx32/lib64, is there a better variable to use?
do_install_append () {
rm -rf ${D}${includedir}/uuid ${D}${libdir}/pkgconfig/uuid.pc ${D}${libdir}/libuuid* ${D}/lib/libuuid* ${D}/libx32/libuuid* ${D}/lib64/libuuid*
}Signature · 2 lines
Kind regards, Luca Boccassi
PATCH v10 1/1 util-linux: split uuid in separate recipe to allow bootstrapping
In reply to Luca BocassiFrom: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.
https://github.com/karelzak/util-linux/pull/898
---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
Removed PACKAGES_remove and instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.
v6: install libuuid.a in libuuid-dev
change rdepends on uuid to use binary package rather than source
add rdepends on uuid-dev to libuuid-dev
remove rprovides of libuuid-dev from uuid-dev
v7: do not use '_append' for RDEPENDS, as it is not supported by BBCLASSEXTEND, use
simply '+='
v8: added missing libuuid-staticdev package
remove libx32/libuuid* too
v9: set RECIPE_MAINTAINER_pn-util-linux-uuid, same value as RECIPE_MAINTAINER_pn-util-linux
v10: removed DEBIAN_NOAUTONAME
remove lib64/libuuid* too
meta/conf/distro/include/maintainers.inc | 1 + .../util-linux/util-linux-uuid_2.36.2.bb | 20 +++++++ meta/recipes-core/util-linux/util-linux.inc | 41 +++++++++++++ .../util-linux/util-linux_2.36.2.bb | 57 +++++-------------- 4 files changed, 77 insertions(+), 42 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb create mode 100644 meta/recipes-core/util-linux/util-linux.inc
meta/conf/distro/include/maintainers.inc +1 -0
@@ -720,6 +720,7 @@ RECIPE_MAINTAINER_pn-update-rc.d = "Ross Burton <[email protected]>" RECIPE_MAINTAINER_pn-usbinit = "Alexander Kanavin <[email protected]>" RECIPE_MAINTAINER_pn-usbutils = "Alexander Kanavin <[email protected]>" RECIPE_MAINTAINER_pn-util-linux = "Chen Qi <[email protected]>" +RECIPE_MAINTAINER_pn-util-linux-uuid = "Chen Qi <[email protected]>" RECIPE_MAINTAINER_pn-util-macros = "Armin Kuster <[email protected]>" RECIPE_MAINTAINER_pn-v86d = "Alexander Kanavin <[email protected]>" RECIPE_MAINTAINER_pn-vala = "Alexander Kanavin <[email protected]>"
meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb new file +20 -0
@@ -0,0 +1,20 @@ +# To allow util-linux to optionally build-depend on cryptsetup, libuuid is +# split out of the main recipe, as it's needed by cryptsetup + +require util-linux.inc + +inherit autotools gettext pkgconfig + +S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-staticdev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-staticdev = "${libdir}/libuuid.a" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" + +do_install_append() { + rm -rf ${D}${datadir} ${D}${bindir} ${D}${base_bindir} ${D}${sbindir} ${D}${base_sbindir} ${D}${exec_prefix}/sbin +} + +BBCLASSEXTEND = "native nativesdk"
meta/recipes-core/util-linux/util-linux.inc new file +41 -0
41 lines not on this page · raw message
meta/recipes-core/util-linux/util-linux_2.36.2.bb +15 -42
88 lines not on this page · raw message
Signature · 1 line
2.29.2
meta/recipes-core/util-linux/util-linux_2.36.2.bb do_install_append_class-native () { · line 209
Quoted 6 lines · On Thu, 2021-03-11 at 14:38 +0000, [email protected] wrote:
meta/recipes-core/util-linux/util-linux_2.36.2.bb +6
-0+# dm-verity support introduces a circular build dependency, so util-linux-uuid is split out for target builds +# Need to build libuuid for uuidgen, but then delete it and let the other recipe ship it +do_install_append () { + rm -rf ${D}${includedir}/uuid ${D}${libdir}/pkgconfig/uuid.pc ${D}${libdir}/libuuid* ${D}/lib/libuuid* ${D}/libx32/libuuid* ${D}/lib64/libuuid* +} +
I think you can replace:
${D}/lib/libuuid* ${D}/libx32/libuuid* ${D}/lib64/libuuid*
with
${D}${base_libdir}/libuuid*
?
Cheers,
Richard
Quoted 10 lines · Richard Purdie
meta/recipes-core/util-linux/util-linux_2.36.2.bb do_install_append_class-native () { · line 209
Quoted 6 lines · On Thu, 2021-03-11 at 14:38 +0000, [email protected] wrote:
meta/recipes-core/util-linux/util-linux_2.36.2.bb +6
-0+# dm-verity support introduces a circular build dependency, so util-linux-uuid is split out for target builds +# Need to build libuuid for uuidgen, but then delete it and let the other recipe ship it +do_install_append () { + rm -rf ${D}${includedir}/uuid ${D}${libdir}/pkgconfig/uuid.pc ${D}${libdir}/libuuid* ${D}/lib/libuuid* ${D}/libx32/libuuid* ${D}/lib64/libuuid* +} +I think you can replace:
Quoted 7 lines
${D}/lib/libuuid* ${D}/libx32/libuuid* ${D}/lib64/libuuid*
with
${D}${base_libdir}/libuuid*
?
Thanks, seems to work locally, sent v11
Signature · 2 lines
Kind regards, Luca Boccassi
PATCH v11 1/1 util-linux: split uuid in separate recipe to allow bootstrapping
In reply to Luca BocassiFrom: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.
https://github.com/karelzak/util-linux/pull/898
---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
Removed PACKAGES_remove and instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.
v6: install libuuid.a in libuuid-dev
change rdepends on uuid to use binary package rather than source
add rdepends on uuid-dev to libuuid-dev
remove rprovides of libuuid-dev from uuid-dev
v7: do not use '_append' for RDEPENDS, as it is not supported by BBCLASSEXTEND, use
simply '+='
v8: added missing libuuid-staticdev package
remove libx32/libuuid* too
v9: set RECIPE_MAINTAINER_pn-util-linux-uuid, same value as RECIPE_MAINTAINER_pn-util-linux
v10: removed DEBIAN_NOAUTONAME
remove lib64/libuuid* too
v11: use ${D}${base_libdir}/libuuid* instead of manually specifying /lib|/libx32|/lib64
meta/conf/distro/include/maintainers.inc | 1 + .../util-linux/util-linux-uuid_2.36.2.bb | 20 +++++++ meta/recipes-core/util-linux/util-linux.inc | 41 +++++++++++++ .../util-linux/util-linux_2.36.2.bb | 57 +++++-------------- 4 files changed, 77 insertions(+), 42 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb create mode 100644 meta/recipes-core/util-linux/util-linux.inc
meta/conf/distro/include/maintainers.inc +1 -0
@@ -720,6 +720,7 @@ RECIPE_MAINTAINER_pn-update-rc.d = "Ross Burton <[email protected]>" RECIPE_MAINTAINER_pn-usbinit = "Alexander Kanavin <[email protected]>" RECIPE_MAINTAINER_pn-usbutils = "Alexander Kanavin <[email protected]>" RECIPE_MAINTAINER_pn-util-linux = "Chen Qi <[email protected]>" +RECIPE_MAINTAINER_pn-util-linux-uuid = "Chen Qi <[email protected]>" RECIPE_MAINTAINER_pn-util-macros = "Armin Kuster <[email protected]>" RECIPE_MAINTAINER_pn-v86d = "Alexander Kanavin <[email protected]>" RECIPE_MAINTAINER_pn-vala = "Alexander Kanavin <[email protected]>"
meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb new file +20 -0
@@ -0,0 +1,20 @@ +# To allow util-linux to optionally build-depend on cryptsetup, libuuid is +# split out of the main recipe, as it's needed by cryptsetup + +require util-linux.inc + +inherit autotools gettext pkgconfig + +S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-staticdev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-staticdev = "${libdir}/libuuid.a" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" + +do_install_append() { + rm -rf ${D}${datadir} ${D}${bindir} ${D}${base_bindir} ${D}${sbindir} ${D}${base_sbindir} ${D}${exec_prefix}/sbin +} + +BBCLASSEXTEND = "native nativesdk"
meta/recipes-core/util-linux/util-linux.inc new file +41 -0
41 lines not on this page · raw message
meta/recipes-core/util-linux/util-linux_2.36.2.bb +15 -42
88 lines not on this page · raw message
Signature · 1 line
2.29.2
Quoted 10 lines · On Thu, 2021-03-11 at 15:09 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>
Unfortunately I noticed we had a performance regression in buildtimes in recent changes. The closest I have this narrowed down to so far:
suggests it may be this change. I have more tests queued to confirm that definitively, if so we'll have to figure out why as this shouldn't really happen, its an 8% regression :(.
Cheers,
Richard
Quoted 17 lines · Richard Purdie
Quoted 10 lines · On Thu, 2021-03-11 at 15:09 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>Unfortunately I noticed we had a performance regression in buildtimes in recent changes. The closest I have this narrowed down to so far:
Quoted 7 lines
suggests it may be this change. I have more tests queued to confirm that definitively, if so we'll have to figure out why as this shouldn't really happen, its an 8% regression :(.
Cheers,
Richard
Hi,
Very strange that a single recipe could do that - is there something wrong in the new .bb that I missed and could cause this?
Signature · 2 lines
Kind regards, Luca Boccassi
Quoted 16 lines · Luca Boccassi
Quoted 15 lines · Richard Purdie
Quoted 10 lines · On Thu, 2021-03-11 at 15:09 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>Unfortunately I noticed we had a performance regression in buildtimes in recent changes. The closest I have this narrowed down to so far:
Quoted 8 lines
Quoted 5 lines
suggests it may be this change. I have more tests queued to confirm that definitively, if so we'll have to figure out why as this shouldn't really happen, its an 8% regression :(.
Very strange that a single recipe could do that - is there something wrong in the new .bb that I missed and could cause this?
I'm wondering if it is because we're building util-linux twice now and there is some key choke point in the dependency chain. I have no evidence for that yet, it is just speculation though.
Cheers,
Richard
Quoted 23 lines · Richard Purdie
Quoted 22 lines · Luca Boccassi
Quoted 20 lines · Richard Purdie
Quoted 10 lines · On Thu, 2021-03-11 at 15:09 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>Unfortunately I noticed we had a performance regression in buildtimes in recent changes. The closest I have this narrowed down to so far:
suggests it may be this change. I have more tests queued to confirm that definitively, if so we'll have to figure out why as this shouldn't really happen, its an 8% regression :(.
Quoted 6 lines
Quoted 2 lines
Very strange that a single recipe could do that - is there something wrong in the new .bb that I missed and could cause this?
I'm wondering if it is because we're building util-linux twice now and there is some key choke point in the dependency chain. I have no evidence for that yet, it is just speculation though.
With the autoconf options I've set, on my laptop it takes 32s to do configure + make -j2. Most of that is autoconf - make -j2 takes 8s.
Only 3 libraries are built with this combination: libcommon.a, libtcolors.a, and libuuid.a/so. No executables or anything else is built. It doesn't look like libtcolors is actually needed, I'll see if I can prepare a patch to skip it, but I don't think it will buy more than 1s, it's just two object files.
The good news is that meson support is about to land upstream, which should be significantly faster than autoconf + make:
https://github.com/karelzak/util-linux/commits/topic/meson
Signature · 2 lines
Kind regards, Luca Boccassi
Quoted 40 lines · Luca Boccassi
Quoted 23 lines · Richard Purdie
Quoted 22 lines · Luca Boccassi
Quoted 20 lines · Richard Purdie
Quoted 10 lines · On Thu, 2021-03-11 at 15:09 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>Unfortunately I noticed we had a performance regression in buildtimes in recent changes. The closest I have this narrowed down to so far:
suggests it may be this change. I have more tests queued to confirm that definitively, if so we'll have to figure out why as this shouldn't really happen, its an 8% regression :(.
Quoted 6 lines
Quoted 2 lines
Very strange that a single recipe could do that - is there something wrong in the new .bb that I missed and could cause this?
I'm wondering if it is because we're building util-linux twice now and there is some key choke point in the dependency chain. I have no evidence for that yet, it is just speculation though.
With the autoconf options I've set, on my laptop it takes 32s to do configure + make -j2. Most of that is autoconf - make -j2 takes 8s.
Only 3 libraries are built with this combination: libcommon.a, libtcolors.a, and libuuid.a/so. No executables or anything else is built. It doesn't look like libtcolors is actually needed, I'll see if I can prepare a patch to skip it, but I don't think it will buy more than 1s, it's just two object files.
Quoted 4 lines
The good news is that meson support is about to land upstream, which should be significantly faster than autoconf + make:
Meson definitely improves the speed! I was wondering if it was from configure for example.
I now have more performance test results in (takes time to interleave them with testing of master):
and I think this means it isn't from the util-linux change but one of another three. I'm not entirely convinced those changes could do this but it is what the data says.
I've queued more bisection to narrow it down from there...
Cheers,
Richard
Quoted 58 lines · Richard Purdie
Quoted 40 lines · Luca Boccassi
Quoted 23 lines · Richard Purdie
Quoted 22 lines · Luca Boccassi
Quoted 20 lines · Richard Purdie
Quoted 10 lines · On Thu, 2021-03-11 at 15:09 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>Unfortunately I noticed we had a performance regression in buildtimes in recent changes. The closest I have this narrowed down to so far:
suggests it may be this change. I have more tests queued to confirm that definitively, if so we'll have to figure out why as this shouldn't really happen, its an 8% regression :(.
Quoted 6 lines
Quoted 2 lines
Very strange that a single recipe could do that - is there something wrong in the new .bb that I missed and could cause this?
I'm wondering if it is because we're building util-linux twice now and there is some key choke point in the dependency chain. I have no evidence for that yet, it is just speculation though.
With the autoconf options I've set, on my laptop it takes 32s to do configure + make -j2. Most of that is autoconf - make -j2 takes 8s.
Only 3 libraries are built with this combination: libcommon.a, libtcolors.a, and libuuid.a/so. No executables or anything else is built. It doesn't look like libtcolors is actually needed, I'll see if I can prepare a patch to skip it, but I don't think it will buy more than 1s, it's just two object files.
Quoted 4 lines
The good news is that meson support is about to land upstream, which should be significantly faster than autoconf + make:
Meson definitely improves the speed! I was wondering if it was from configure for example.
I now have more performance test results in (takes time to interleave them with testing of master):
and I think this means it isn't from the util-linux change but one of another three. I'm not entirely convinced those changes could do this but it is what the data says.
Quoted 5 lines
I've queued more bisection to narrow it down from there...
Cheers,
Richard
Ok, that's good news then (at least for me!) - in the meanwhile I had a look at automake anyway, and found out the compile time can be halved since libcommon and libtcolors are not needed by libuuid:
https://github.com/karelzak/util-linux/pull/1262
Not sure if worth backporting, as it's just a dozen object files, the big time chunk is from autoconf anyway.
Signature · 2 lines
Kind regards, Luca Boccassi
Quoted 52 lines · Richard Purdie
Quoted 40 lines · Luca Boccassi
Quoted 23 lines · Richard Purdie
Quoted 22 lines · Luca Boccassi
Quoted 20 lines · Richard Purdie
Quoted 10 lines · On Thu, 2021-03-11 at 15:09 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>Unfortunately I noticed we had a performance regression in buildtimes in recent changes. The closest I have this narrowed down to so far:
suggests it may be this change. I have more tests queued to confirm that definitively, if so we'll have to figure out why as this shouldn't really happen, its an 8% regression :(.
Quoted 6 lines
Quoted 2 lines
Very strange that a single recipe could do that - is there something wrong in the new .bb that I missed and could cause this?
I'm wondering if it is because we're building util-linux twice now and there is some key choke point in the dependency chain. I have no evidence for that yet, it is just speculation though.
With the autoconf options I've set, on my laptop it takes 32s to do configure + make -j2. Most of that is autoconf - make -j2 takes 8s.
Only 3 libraries are built with this combination: libcommon.a, libtcolors.a, and libuuid.a/so. No executables or anything else is built. It doesn't look like libtcolors is actually needed, I'll see if I can prepare a patch to skip it, but I don't think it will buy more than 1s, it's just two object files.
Quoted 4 lines
The good news is that meson support is about to land upstream, which should be significantly faster than autoconf + make:
Meson definitely improves the speed! I was wondering if it was from configure for example.
I now have more performance test results in (takes time to interleave them with testing of master):
Quoted 7 lines
and I think this means it isn't from the util-linux change but one of another three. I'm not entirely convinced those changes could do this but it is what the data says.
I've queued more bisection to narrow it down from there...
BTW: this split also needs manual cleanup in the TMPDIR, right?
Incremental build fails with:
ERROR: util-linux-uuid-2.36.2-r0 do_packagedata: The recipe util-linux-uuid is trying to install files into a shared area when those files already exist. Those files and their manifest location are:
/OE/build/oe-core/tmp-glibc/pkgdata/qemux86-64/shlibs2/util-linux-libuuid.list
(matched in manifest-qemux86_64-util-linux.packagedata)
/OE/build/oe-core/tmp-glibc/pkgdata/qemux86-64/runtime/util-linux-libuuid.packaged
(matched in manifest-qemux86_64-util-linux.packagedata)
/OE/build/oe-core/tmp-glibc/pkgdata/qemux86-64/runtime/util-linux-libuuid
(matched in manifest-qemux86_64-util-linux.packagedata)
Please verify which recipe should provide the above files.
The build has stopped, as continuing in this scenario WILL break things - if not now, possibly in the future (we've seen builds fail several months later). If the system knew how to recover from this automatically it would, however there are several different scenarios which can result in this and we don't know which one this is. It may be you have switched providers of something like virtual/kernel (e.g. from linux-yocto to linux-yocto-dev), in that case you need to execute the clean task for both recipes and it will resolve this error. It may be you changed DISTRO_FEATURES from systemd to udev or vice versa. Cleaning those recipes should again resolve this error, however switching DISTRO_FEATURES on an existing build directory is not supported - you should really clean out tmp and rebuild (reusing sstate should be safe). It could be the overlapping files detected are harmless in which case adding them to SSTATE_DUPWHITELIST may be the correct solution. It could also be your build is including two different conflicting versions of things (e.g. bluez 4 and bluez 5 and the correct solution for that would be to resolve the conflict. If in doubt, please ask on the mailing list, sharing the error and filelist above.
ERROR: util-linux-uuid-2.36.2-r0 do_packagedata: If the above message is too much, the simpler version is you're advised to wipe out tmp and rebuild (reusing sstate is fine). That will likely fix things in most (but not all) cases.
ERROR: Logfile of failure stored in: /OE/build/oe-core/tmp-glibc/work/core2-64-oe-linux/util-linux-uuid/2.36.2-r0/temp/log.do_packagedata.2055738
NOTE: recipe util-linux-uuid-2.36.2-r0: task do_packagedata: Failed
ERROR: Task (/OE/build/oe-core/openembedded-core/meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb:do_packagedata) failed with exit code '1'
as util-linux-uuid doesn't depend on util-linux (and even cannot depend on it), then nothing ensures that older util-linux (including libuuid) is removed from sstate-control before this new separate util-linux-uuid is running packagedata.
Not sure if there is something we can do in cases like this and bumping OELAYOUT_ABI because of this 1 recipe seems a bit overkill.
"bitbake -c clean util-linux" should be enough to unblock util-linux-uuid, but it needs to be executed for every MACHINE (or at least every TUNE_PKGARCH) in given TMPDIR.
Cheers,
Quoted 55 lines · Martin Jansa
Quoted 54 lines · Richard Purdie
Quoted 40 lines · Luca Boccassi
Quoted 23 lines · Richard Purdie
Quoted 22 lines · Luca Boccassi
Quoted 20 lines · Richard Purdie
Quoted 10 lines · On Thu, 2021-03-11 at 15:09 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>Unfortunately I noticed we had a performance regression in buildtimes in recent changes. The closest I have this narrowed down to so far:
suggests it may be this change. I have more tests queued to confirm that definitively, if so we'll have to figure out why as this shouldn't really happen, its an 8% regression :(.
Quoted 6 lines
Quoted 2 lines
Very strange that a single recipe could do that - is there something wrong in the new .bb that I missed and could cause this?
I'm wondering if it is because we're building util-linux twice now and there is some key choke point in the dependency chain. I have no evidence for that yet, it is just speculation though.
With the autoconf options I've set, on my laptop it takes 32s to do configure + make -j2. Most of that is autoconf - make -j2 takes 8s.
Only 3 libraries are built with this combination: libcommon.a, libtcolors.a, and libuuid.a/so. No executables or anything else is built. It doesn't look like libtcolors is actually needed, I'll see if I can prepare a patch to skip it, but I don't think it will buy more than 1s, it's just two object files.
Quoted 4 lines
The good news is that meson support is about to land upstream, which should be significantly faster than autoconf + make:
Meson definitely improves the speed! I was wondering if it was from configure for example.
I now have more performance test results in (takes time to interleave them with testing of master):
Quoted 7 lines
Quoted 5 lines
and I think this means it isn't from the util-linux change but one of another three. I'm not entirely convinced those changes could do this but it is what the data says.
I've queued more bisection to narrow it down from there...
BTW: this split also needs manual cleanup in the TMPDIR, right?
It shouldn't. The system should spot that util-linux has changed and uninstall it from the sysroots as it goes. There is something not working right there :(
Cheers,
Richard
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 60 lines · Richard Purdie via lists.openembedded.org
Quoted 59 lines · Martin Jansa
Quoted 58 lines · Richard Purdie
Quoted 40 lines · Luca Boccassi
Quoted 23 lines · Richard Purdie
Quoted 22 lines · Luca Boccassi
Quoted 20 lines · Richard Purdie
Quoted 10 lines · On Thu, 2021-03-11 at 15:09 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>Unfortunately I noticed we had a performance regression in buildtimes in recent changes. The closest I have this narrowed down to so far:
suggests it may be this change. I have more tests queued to confirm that definitively, if so we'll have to figure out why as this shouldn't really happen, its an 8% regression :(.
Quoted 6 lines
Quoted 2 lines
Very strange that a single recipe could do that - is there something wrong in the new .bb that I missed and could cause this?
I'm wondering if it is because we're building util-linux twice now and there is some key choke point in the dependency chain. I have no evidence for that yet, it is just speculation though.
With the autoconf options I've set, on my laptop it takes 32s to do configure + make -j2. Most of that is autoconf - make -j2 takes 8s.
Only 3 libraries are built with this combination: libcommon.a, libtcolors.a, and libuuid.a/so. No executables or anything else is built. It doesn't look like libtcolors is actually needed, I'll see if I can prepare a patch to skip it, but I don't think it will buy more than 1s, it's just two object files.
Quoted 4 lines
The good news is that meson support is about to land upstream, which should be significantly faster than autoconf + make:
Meson definitely improves the speed! I was wondering if it was from configure for example.
I now have more performance test results in (takes time to interleave them with testing of master):
and I think this means it isn't from the util-linux change but one of another three. I'm not entirely convinced those changes could do this but it is what the data says.
Quoted 6 lines
Quoted 3 lines
Quoted 1 line
I've queued more bisection to narrow it down from there...
BTW: this split also needs manual cleanup in the TMPDIR, right?
It shouldn't. The system should spot that util-linux has changed and uninstall it from the sysroots as it goes. There is something not working right there :(
I was wrong about that, the system doesn't have code for this, it has code for the sysroots but not for other sstate tasks.
I think this is an oversight and we need to simplify things and make this cleanup happen pre-build, much like the "unreachable" tasks cleanup happens today. If we do make this happen, we probably need to add parallelism as the number of stale sstate tasks being cleaned could be substantial.
Cheers,
Richard
I have not debugged yet, but I suspect this change is causing the following failure if I merge the latest poky into our builds:
ERROR: mc:host:util-linux-uuid-2.36.2-r0 do_package: QA Issue: util-linux-uuid: Files/directories were installed but not shipped in any package:
/usr/lib/debug
/usr/lib/debug/usr
/usr/lib/debug/usr/lib
/usr/lib/debug/usr/lib/libuuid.so.1.3.0.debug
Please set FILES such that these items are packaged. Alternatively if they are unneeded, avoid installing them or delete them within do_install.
util-linux-uuid: 4 installed and not shipped files. [installed-vs-shipped]
ERROR: mc:host:util-linux-uuid-2.36.2-r0 do_package: Fatal QA errors found, failing task.
ERROR: Logfile of failure stored in: /hdd/yocto/genx/poky/build/tmp/work/core2-64-poky-linux/util-linux-uuid/2.36.2-r0/temp/log.do_package.31753
ERROR: Task (mc:host:/hdd/yocto/genx/poky/build/../meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb:do_package) failed with exit code '1'
Quoted 90 lines · Richard Purdie
Quoted 60 lines · Richard Purdie via lists.openembedded.org
Quoted 59 lines · Martin Jansa
Quoted 58 lines · Richard Purdie
Quoted 40 lines · Luca Boccassi
Quoted 23 lines · Richard Purdie
Quoted 22 lines · Luca Boccassi
Quoted 20 lines · Richard Purdie
Quoted 10 lines · On Thu, 2021-03-11 at 15:09 +0000, [email protected] wrote:
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>Unfortunately I noticed we had a performance regression in buildtimes in recent changes. The closest I have this narrowed down to so far:
suggests it may be this change. I have more tests queued to confirm that definitively, if so we'll have to figure out why as this shouldn't really happen, its an 8% regression :(.
Quoted 6 lines
Quoted 2 lines
Very strange that a single recipe could do that - is there something wrong in the new .bb that I missed and could cause this?
I'm wondering if it is because we're building util-linux twice now and there is some key choke point in the dependency chain. I have no evidence for that yet, it is just speculation though.
With the autoconf options I've set, on my laptop it takes 32s to do configure + make -j2. Most of that is autoconf - make -j2 takes 8s.
Only 3 libraries are built with this combination: libcommon.a, libtcolors.a, and libuuid.a/so. No executables or anything else is built. It doesn't look like libtcolors is actually needed, I'll see if I can prepare a patch to skip it, but I don't think it will buy more than 1s, it's just two object files.
Quoted 4 lines
The good news is that meson support is about to land upstream, which should be significantly faster than autoconf + make:
Meson definitely improves the speed! I was wondering if it was from configure for example.
I now have more performance test results in (takes time to interleave them with testing of master):
and I think this means it isn't from the util-linux change but one of another three. I'm not entirely convinced those changes could do this but it is what the data says.
Quoted 6 lines
Quoted 3 lines
Quoted 1 line
I've queued more bisection to narrow it down from there...
BTW: this split also needs manual cleanup in the TMPDIR, right?
It shouldn't. The system should spot that util-linux has changed and uninstall it from the sysroots as it goes. There is something not working right there :(
I was wrong about that, the system doesn't have code for this, it has code for the sysroots but not for other sstate tasks.
I think this is an oversight and we need to simplify things and make this cleanup happen pre-build, much like the "unreachable" tasks cleanup happens today. If we do make this happen, we probably need to add parallelism as the number of stale sstate tasks being cleaned could be substantial.
Cheers,
Richard
Quoted 12 lines · Scott Branden
I have not debugged yet, but I suspect this change is causing the following failure if I merge the latest poky into our builds:
ERROR: mc:host:util-linux-uuid-2.36.2-r0
do_package: QA Issue: util-linux-uuid: Files/directories were installed but not shipped in any package:
/usr/lib/debug
/usr/lib/debug/usr
/usr/lib/debug/usr/lib
/usr/lib/debug/usr/lib/libuuid.so.1.3.0.debug
Please set FILES such that these items are packaged. Alternatively if they are unneeded, avoid installing them or delete them withindo_install.
util-linux-uuid: 4 installed and not shipped files. [installed-vs-shipped]
ERROR: mc:host:util-linux-uuid-2.36.2-r0do_package: Fatal QA errors found, failing task.
ERROR: Logfile of failure stored in: /hdd/yocto/genx/poky/build/tmp/work/core2-64-poky-linux/util-linux-uuid/2.36.2-r0/temp/log.do_package.31753
ERROR: Task (mc:host:/hdd/yocto/genx/poky/build/../meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb:do_package) failed with exit code '1'
It most certainly is - is /usr/lib/debug configurable or supported? Currently we have:
FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug"
I'll send a patch to extend it and include /usr/lib/debug too
Quoted 91 lines · Richard Purdie
Quoted 90 lines · Richard Purdie via lists.openembedded.org
Quoted 60 lines · Martin Jansa
Quoted 59 lines · Richard Purdie
Quoted 58 lines · Luca Boccassi
Quoted 40 lines · Richard Purdie
Quoted 23 lines · Luca Boccassi
Quoted 22 lines · Richard Purdie
Quoted 20 lines · On Thu, 2021-03-11 at 15:09 +0000, [email protected] wrote:
Quoted 10 lines
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.https://github.com/karelzak/util-linux/pull/898
Signed-off-by Luca Boccassi <[email protected]>Unfortunately I noticed we had a performance regression in buildtimes in recent changes. The closest I have this narrowed down to so far:
suggests it may be this change. I have more tests queued to confirm that definitively, if so we'll have to figure out why as this shouldn't really happen, its an 8% regression :(.
Quoted 6 lines
Quoted 2 lines
Very strange that a single recipe could do that - is there something wrong in the new .bb that I missed and could cause this?
I'm wondering if it is because we're building util-linux twice now and there is some key choke point in the dependency chain. I have no evidence for that yet, it is just speculation though.
With the autoconf options I've set, on my laptop it takes 32s to do configure + make -j2. Most of that is autoconf - make -j2 takes 8s.
Only 3 libraries are built with this combination: libcommon.a, libtcolors.a, and libuuid.a/so. No executables or anything else is built. It doesn't look like libtcolors is actually needed, I'll see if I can prepare a patch to skip it, but I don't think it will buy more than 1s, it's just two object files.
Quoted 4 lines
The good news is that meson support is about to land upstream, which should be significantly faster than autoconf + make:
Meson definitely improves the speed! I was wondering if it was from configure for example.
I now have more performance test results in (takes time to interleave them with testing of master):
and I think this means it isn't from the util-linux change but one of another three. I'm not entirely convinced those changes could do this but it is what the data says.
Quoted 6 lines
Quoted 3 lines
Quoted 1 line
I've queued more bisection to narrow it down from there...
BTW: this split also needs manual cleanup in the TMPDIR, right?
It shouldn't. The system should spot that util-linux has changed and uninstall it from the sysroots as it goes. There is something not working right there :(
I was wrong about that, the system doesn't have code for this, it has code for the sysroots but not for other sstate tasks.
I think this is an oversight and we need to simplify things and make this cleanup happen pre-build, much like the "unreachable" tasks cleanup happens today. If we do make this happen, we probably need to add parallelism as the number of stale sstate tasks being cleaned could be substantial.
Cheers,
Richard
Quoted 14 lines · Luca Boccassi
Quoted 12 lines · Scott Branden
I have not debugged yet, but I suspect this change is causing the following failure if I merge the latest poky into our builds:
ERROR: mc:host:util-linux-uuid-2.36.2-r0
do_package: QA Issue: util-linux-uuid: Files/directories were installed but not shipped in any package:
/usr/lib/debug
/usr/lib/debug/usr
/usr/lib/debug/usr/lib
/usr/lib/debug/usr/lib/libuuid.so.1.3.0.debug
Please set FILES such that these items are packaged. Alternatively if they are unneeded, avoid installing them or delete them withindo_install.
util-linux-uuid: 4 installed and not shipped files. [installed-vs-shipped]
ERROR: mc:host:util-linux-uuid-2.36.2-r0do_package: Fatal QA errors found, failing task.
ERROR: Logfile of failure stored in: /hdd/yocto/genx/poky/build/tmp/work/core2-64-poky-linux/util-linux-uuid/2.36.2-r0/temp/log.do_package.31753
ERROR: Task (mc:host:/hdd/yocto/genx/poky/build/../meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb:do_package) failed with exit code '1'
Quoted 4 lines
It most certainly is - is
/usr/lib/debugconfigurable or supported? Currently we have:FILES_util-linux-libuuid-dbg = "
/usr/src${libdir}/.debug"
Does it need to be set at all or will the default not cover this?
Cheers,
Richard
Quoted 18 lines · Richard Purdie
Quoted 17 lines · Luca Boccassi
Quoted 12 lines · Scott Branden
I have not debugged yet, but I suspect this change is causing the following failure if I merge the latest poky into our builds:
ERROR: mc:host:util-linux-uuid-2.36.2-r0
do_package: QA Issue: util-linux-uuid: Files/directories were installed but not shipped in any package:
/usr/lib/debug
/usr/lib/debug/usr
/usr/lib/debug/usr/lib
/usr/lib/debug/usr/lib/libuuid.so.1.3.0.debug
Please set FILES such that these items are packaged. Alternatively if they are unneeded, avoid installing them or delete them withindo_install.
util-linux-uuid: 4 installed and not shipped files. [installed-vs-shipped]
ERROR: mc:host:util-linux-uuid-2.36.2-r0do_package: Fatal QA errors found, failing task.
ERROR: Logfile of failure stored in: /hdd/yocto/genx/poky/build/tmp/work/core2-64-poky-linux/util-linux-uuid/2.36.2-r0/temp/log.do_package.31753
ERROR: Task (mc:host:/hdd/yocto/genx/poky/build/../meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb:do_package) failed with exit code '1'It most certainly is - is
/usr/lib/debugconfigurable or supported? Currently we have:
Quoted 7 lines
Quoted 1 line
FILES_util-linux-libuuid-dbg = "
/usr/src${libdir}/.debug"Does it need to be set at all or will the default not cover this?
Cheers,
Richard
IIRC there were QA failures without it, but I cannot remember exactly.
Signature · 2 lines
Kind regards, Luca Boccassi
Could you look into this warning.
WARNING: util-linux-2.36.2-r0 do_package_qa: QA Issue: util-linux-dev rdepends on util-linux-libuuid-dev, but it isn't a build dependency? [build-deps]
https://autobuilder.yoctoproject.org/typhoon/#/builders/61/builds/3226
Regards,
Oleksiy
________________________________________
From: [email protected] <[email protected]> on behalf of Luca Bocassi <[email protected]>
Sent: Thursday, March 11, 2021 17:09
To: [email protected]
Cc: [email protected]; [email protected]
Subject: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrapping
From: Luca Boccassi <[email protected]>
Recently util-linux gained an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid (mandatory).
Split out util-linux-uuid in a different recipe to break the cycle.
https://github.com/karelzak/util-linux/pull/898
---
v1: util-linux 2.35 is not out yet, but I'd like to get the preparatory work
underway as I'm not sure if this is the best approach or if there are alternatives. Suggestions and comments very welcome. Thanks!
v2: changed packages names to reflect old ones (eg: libuuid1 -> util-linux-libuuid)
and leave uuid build enable in main recipe to allow for uuidgen build to happen, as it does not have its own autoconf switch. Delete the library manualy from the main recipe after build instead, and add dependency.
Might help to break loop python3 -> util-linux -> libselinux -> python3, as it's only libuuid that is needed, see https://lists.yoctoproject.org/g/yocto/message/47570
v3: rebased and refactored to have a common util-linux.inc file
v4: added RDEPENDS on util-linux-libuuid on various packages to fix QA warnings
v5: remove RDEPENDS and instead have util-linux RDEPEND on util-linux-uuid.
Removed PACKAGES_remove and instead filter out libuuid via the package generation regex.
Rebased on util-linux 2.36.2.
v6: install libuuid.a in libuuid-dev
change rdepends on uuid to use binary package rather than source
add rdepends on uuid-dev to libuuid-dev
remove rprovides of libuuid-dev from uuid-dev
v7: do not use '_append' for RDEPENDS, as it is not supported by BBCLASSEXTEND, use
simply '+='
v8: added missing libuuid-staticdev package
remove libx32/libuuid* too
v9: set RECIPE_MAINTAINER_pn-util-linux-uuid, same value as RECIPE_MAINTAINER_pn-util-linux
v10: removed DEBIAN_NOAUTONAME
remove lib64/libuuid* too
v11: use ${D}${base_libdir}/libuuid* instead of manually specifying /lib|/libx32|/lib64
meta/conf/distro/include/maintainers.inc | 1 + .../util-linux/util-linux-uuid_2.36.2.bb | 20 +++++++ meta/recipes-core/util-linux/util-linux.inc | 41 +++++++++++++ .../util-linux/util-linux_2.36.2.bb | 57 +++++-------------- 4 files changed, 77 insertions(+), 42 deletions(-) create mode 100644 meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb create mode 100644 meta/recipes-core/util-linux/util-linux.inc
meta/conf/distro/include/maintainers.inc +1 -0
@@ -720,6 +720,7 @@ RECIPE_MAINTAINER_pn-update-rc.d = "Ross Burton <[email protected]>" RECIPE_MAINTAINER_pn-usbinit = "Alexander Kanavin <[email protected]>" RECIPE_MAINTAINER_pn-usbutils = "Alexander Kanavin <[email protected]>" RECIPE_MAINTAINER_pn-util-linux = "Chen Qi <[email protected]>" +RECIPE_MAINTAINER_pn-util-linux-uuid = "Chen Qi <[email protected]>" RECIPE_MAINTAINER_pn-util-macros = "Armin Kuster <[email protected]>" RECIPE_MAINTAINER_pn-v86d = "Alexander Kanavin <[email protected]>" RECIPE_MAINTAINER_pn-vala = "Alexander Kanavin <[email protected]>"
meta/recipes-core/util-linux/util-linux-uuid_2.36.2.bb new file +20 -0
@@ -0,0 +1,20 @@ +# To allow util-linux to optionally build-depend on cryptsetup, libuuid is +# split out of the main recipe, as it's needed by cryptsetup + +require util-linux.inc + +inherit autotools gettext pkgconfig + +S = "${WORKDIR}/util-linux-${PV}" +EXTRA_OECONF += "--disable-all-programs --enable-libuuid" +PACKAGES = "util-linux-libuuid util-linux-libuuid-dev util-linux-libuuid-staticdev util-linux-libuuid-dbg" +FILES_util-linux-libuuid = "${libdir}/libuuid.so.*" +FILES_util-linux-libuuid-dev = "${libdir}/libuuid.so ${includedir} ${libdir}/pkgconfig" +FILES_util-linux-libuuid-staticdev = "${libdir}/libuuid.a" +FILES_util-linux-libuuid-dbg = "/usr/src ${libdir}/.debug" + +do_install_append() { + rm -rf ${D}${datadir} ${D}${bindir} ${D}${base_bindir} ${D}${sbindir} ${D}${base_sbindir} ${D}${exec_prefix}/sbin +} + +BBCLASSEXTEND = "native nativesdk"
meta/recipes-core/util-linux/util-linux.inc new file +41 -0
41 lines not on this page · raw message
meta/recipes-core/util-linux/util-linux_2.36.2.bb +15 -42
88 lines not on this page · raw message
--
2.29.2
Quoted 5 lines · Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco)
Could you look into this warning.
WARNING: util-linux-2.36.2-r0
do_package_qa: QA Issue: util-linux-dev rdepends on util-linux-libuuid-dev, but it isn't a build dependency? [build-deps]https://autobuilder.yoctoproject.org/typhoon/#/builders/61/builds/3226
That failure was my fault when testing some fixes.
I've sent out a patch which renames util-linux-uuid to util-linux-libuuid and sorts out the license issue Peter reported.
I'm optimistic this fixes the various issues people have been having and simplifies the recipe a bit as an added bonus.
The patch has already tested cleanly on the autobuilder (the above warning was from an earlier aborted build before I opted to rename the recipe).
Khem: Thanks for the related meta-oe tweak, I think the rename whilst slightly painful now is the best way forward overall.
Cheers,
Richard
Quoted 18 lines · Richard Purdie
Quoted 5 lines · Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco)
Could you look into this warning.
WARNING: util-linux-2.36.2-r0
do_package_qa: QA Issue: util-linux-dev rdepends on util-linux-libuuid-dev, but it isn't a build dependency? [build-deps]https://autobuilder.yoctoproject.org/typhoon/#/builders/61/builds/3226
That failure was my fault when testing some fixes.
I've sent out a patch which renames util-linux-uuid to util-linux-libuuid and sorts out the license issue Peter reported.
I'm optimistic this fixes the various issues people have been having and simplifies the recipe a bit as an added bonus.
The patch has already tested cleanly on the autobuilder (the above warning was from an earlier aborted build before I opted to rename the recipe).
Quoted 6 lines
Khem: Thanks for the related meta-oe tweak, I think the rename whilst slightly painful now is the best way forward overall.
Cheers,
Richard
Looks good to me, thanks!
Signature · 2 lines
Kind regards, Luca Boccassi
Quoted 19 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 10:34
To: Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca Bocassi <[email protected]>; [email protected]
Cc: [email protected]; Peter Kjellerstedt <[email protected]>; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrappingQuoted 3 lines · Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco)
Could you look into this warning.
WARNING: util-linux-2.36.2-r0
do_package_qa: QA Issue: util-linux-devrdepends on util-linux-libuuid-dev, but it isn't a build dependency? [build-deps]
Quoted 1 line
Quoted 6 lines
That failure was my fault when testing some fixes.
I've sent out a patch which renames util-linux-uuid to util-linux-libuuid and sorts out the license issue Peter reported.
I don't mind the recipe being renamed and cleaned up, but I would prefer to see my entire patch for the license parts being either integrated before this or squashed into it, whichever you prefer. It does not make sense to use the same LIC_FILES_CHKSUM for util-linux-libuuid as for util-linux, and setting the other LICENSE variables in util-linux.inc no longer makes sense as they are only relevant for util-linux.
Quoted 6 lines
I'm optimistic this fixes the various issues people have been having and simplifies the recipe a bit as an added bonus.
The patch has already tested cleanly on the autobuilder (the above warning was from an earlier aborted build before I opted to rename the recipe).
Quoted 6 lines
Khem: Thanks for the related meta-oe tweak, I think the rename whilst slightly painful now is the best way forward overall.
Cheers,
Richard
//Peter
Quoted 26 lines · Peter Kjellerstedt
Quoted 19 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 10:34
To: Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca Bocassi <[email protected]>; [email protected]
Cc: [email protected]; Peter Kjellerstedt <[email protected]>; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrappingQuoted 3 lines · Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco)
Could you look into this warning.
WARNING: util-linux-2.36.2-r0
do_package_qa: QA Issue: util-linux-devrdepends on util-linux-libuuid-dev, but it isn't a build dependency? [build-deps]
Quoted 1 line
Quoted 6 lines
That failure was my fault when testing some fixes.
I've sent out a patch which renames util-linux-uuid to util-linux-libuuid and sorts out the license issue Peter reported.
Quoted 6 lines
I don't mind the recipe being renamed and cleaned up, but I would prefer to see my entire patch for the license parts being either integrated before this or squashed into it, whichever you prefer. It does not make sense to use the same
LIC_FILES_CHKSUMfor util-linux-libuuid as for util-linux, and setting the other LICENSE variables in util-linux.inc no longer makes sense as they are only relevant for util-linux.
I'm torn on that. Code with the other licenses is present, just not used in the final output and I personally suspect that having one LIC_FILES_CHKSUM is going to be easier to maintain in the future rather than two separate ones.
Cheers,
Richard
Quoted 33 lines · Richard Purdie
Quoted 25 lines · Peter Kjellerstedt
Quoted 16 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 10:34
To: Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca Bocassi <[email protected]>; [email protected]
Cc: [email protected]; Peter Kjellerstedt <[email protected]>; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrappingQuoted 3 lines · Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco)
Could you look into this warning.
WARNING: util-linux-2.36.2-r0
do_package_qa: QA Issue: util-linux-devQuoted 8 lines
rdepends on util-linux-libuuid-dev, but it isn't a build dependency? [build-deps]
That failure was my fault when testing some fixes.
I've sent out a patch which renames util-linux-uuid to util-linux-libuuid and sorts out the license issue Peter reported.
Quoted 6 lines
I don't mind the recipe being renamed and cleaned up, but I would prefer to see my entire patch for the license parts being either integrated before this or squashed into it, whichever you prefer. It does not make sense to use the same
LIC_FILES_CHKSUMfor util-linux-libuuid as for util-linux, and setting the other LICENSE variables in util-linux.inc no longer makes sense as they are only relevant for util-linux.
Quoted 3 lines
I'm torn on that. Code with the other licenses is present, just not used in the final output and I personally suspect that having one
LIC_FILES_CHKSUMis going to be easier to maintain in the future rather than two separate ones.
FWIW, in Debian the license definitions always cover the _sources_, not the built binaries. IOW: even if you don't build and distribute a subtool, the license metadata must cover it. I'd think this would be even more important for Yocto since you exclusively distribute sources, not binaries.
Signature · 2 lines
Kind regards, Luca Boccassi
Quoted 39 lines · Luca Boccassi
Quoted 33 lines · Richard Purdie
Quoted 25 lines · Peter Kjellerstedt
Quoted 16 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 10:34
To: Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca Bocassi <[email protected]>; [email protected]
Cc: [email protected]; Peter Kjellerstedt <[email protected]>; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrappingQuoted 3 lines · Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco)
Could you look into this warning.
WARNING: util-linux-2.36.2-r0
do_package_qa: QA Issue: util-linux-devQuoted 8 lines
rdepends on util-linux-libuuid-dev, but it isn't a build dependency? [build-deps]
That failure was my fault when testing some fixes.
I've sent out a patch which renames util-linux-uuid to util-linux-libuuid and sorts out the license issue Peter reported.
Quoted 6 lines
I don't mind the recipe being renamed and cleaned up, but I would prefer to see my entire patch for the license parts being either integrated before this or squashed into it, whichever you prefer. It does not make sense to use the same
LIC_FILES_CHKSUMfor util-linux-libuuid as for util-linux, and setting the other LICENSE variables in util-linux.inc no longer makes sense as they are only relevant for util-linux.Quoted 3 lines
I'm torn on that. Code with the other licenses is present, just not used in the final output and I personally suspect that having one
LIC_FILES_CHKSUMis going to be easier to maintain in the future rather than two separate ones.
Quoted 5 lines
FWIW, in Debian the license definitions always cover the _sources_, not the built binaries. IOW: even if you don't build and distribute a subtool, the license metadata must cover it. I'd think this would be even more important for Yocto since you exclusively distribute sources, not binaries.
ideally, it would be good to have both expressed distinctly since we also have tooling for end users to create manifests for whats strictly on target images. we perhaps
need source license which is perhaps at recipe level and then target license which is at package level.
Quoted 5 lines
--
Kind regards,
Luca Boccassi
Quoted 47 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 15:27
To: Peter Kjellerstedt <[email protected]>; Oleksiy Obitotskyi - X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca Bocassi <[email protected]>; [email protected]
Cc: [email protected]; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrappingQuoted 11 lines · Peter Kjellerstedt
Quoted 11 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 10:34
To: Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca Bocassi <[email protected]>; [email protected]
Cc: [email protected]; Peter Kjellerstedt <[email protected]>; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrappingQuoted 5 lines
Quoted 5 lines
Quoted 3 lines · Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco)
Could you look into this warning.
WARNING: util-linux-2.36.2-r0
do_package_qa: QA Issue: util-linux-dev
Quoted 4 lines
Quoted 4 lines
rdepends on util-linux-libuuid-dev, but it isn't a build dependency? [build-deps]
Quoted 2 lines
https://autobuilder.yoctoproject.org/typhoon/#/builders/61/builds/3226
Quoted 4 lines
Quoted 4 lines
That failure was my fault when testing some fixes.
I've sent out a patch which renames util-linux-uuid to util-linux-
libuuid
Quoted 4 lines
Quoted 1 line
and sorts out the license issue Peter reported.
I don't mind the recipe being renamed and cleaned up, but I would prefer to see my entire patch for the license parts being either integrated
before
Quoted 1 line
this or squashed into it, whichever you prefer. It does not make sense
to
Quoted 2 lines
use the same
LIC_FILES_CHKSUMfor util-linux-libuuid as for util-linux, and setting the other LICENSE variables in util-linux.inc no longer
Quoted 7 lines
makes
Quoted 1 line
sense as they are only relevant for util-linux.
I'm torn on that. Code with the other licenses is present, just not used in the final output and I personally suspect that having one
LIC_FILES_CHKSUMis going to be easier to maintain in the future rather than two separate ones.
I actually checked all the files that go into -dev and -src before suggesting this change, and all files are either marked as public domain or use a BSD-3-Clause license.
Quoted 3 lines
Cheers,
Richard
//Peter
Quoted 50 lines · Peter Kjellerstedt
Quoted 50 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 15:27
To: Peter Kjellerstedt <[email protected]>; Oleksiy Obitotskyi - X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca Bocassi <[email protected]>; [email protected]
Cc: [email protected]; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrappingQuoted 11 lines · Peter Kjellerstedt
Quoted 11 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 10:34
To: Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca Bocassi <[email protected]>; [email protected]
Cc: [email protected]; Peter Kjellerstedt <[email protected]>; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrappingQuoted 5 lines
Quoted 5 lines
Quoted 3 lines · Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco)
Could you look into this warning.
WARNING: util-linux-2.36.2-r0
do_package_qa: QA Issue: util-linux-dev
Quoted 4 lines
Quoted 4 lines
rdepends on util-linux-libuuid-dev, but it isn't a build dependency? [build-deps]
Quoted 2 lines
https://autobuilder.yoctoproject.org/typhoon/#/builders/61/builds/3226
Quoted 4 lines
Quoted 4 lines
That failure was my fault when testing some fixes.
I've sent out a patch which renames util-linux-uuid to util-linux-
libuuid
Quoted 4 lines
Quoted 1 line
and sorts out the license issue Peter reported.
I don't mind the recipe being renamed and cleaned up, but I would prefer to see my entire patch for the license parts being either integrated
before
Quoted 1 line
this or squashed into it, whichever you prefer. It does not make sense
to
Quoted 2 lines
use the same
LIC_FILES_CHKSUMfor util-linux-libuuid as for util-linux, and setting the other LICENSE variables in util-linux.inc no longermakes
Quoted 1 line
sense as they are only relevant for util-linux.
Quoted 8 lines
Quoted 4 lines
I'm torn on that. Code with the other licenses is present, just not used in the final output and I personally suspect that having one
LIC_FILES_CHKSUMis going to be easier to maintain in the future rather than two separate ones.I actually checked all the files that go into -dev and -src before suggesting this change, and all files are either marked as public domain or use a BSD-3-Clause license.
There is a difference between what ends up in ${S} and what ends up in the binary packages. LICENSE clearly governs the latter. Its the scope of LIC_FILES_CHECKSUM which there are differences of opinion on.
Cheers,
Richard
Quoted 78 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 17:52
To: Peter Kjellerstedt <[email protected]>; Oleksiy Obitotskyi - X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca Bocassi <[email protected]>; [email protected]
Cc: [email protected]; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrappingQuoted 4 lines · Peter Kjellerstedt
Quoted 4 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 15:27
To: Peter Kjellerstedt <[email protected]>; OleksiyObitotskyi -
Quoted 1 line
Quoted 1 line
X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca
Bocassi
Quoted 6 lines
Quoted 6 lines
<[email protected]>; [email protected]
Cc: [email protected]; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrappingOn Thu, 2021-03-25 at 14:22 +0000, Peter Kjellerstedt wrote:
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 10:34
To: Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca Bocassi <[email protected]>; [email protected]
Cc: [email protected]; Peter Kjellerstedt <[email protected]>; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid inseparate
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
recipe to allow bootstrapping
On Thu, 2021-03-25 at 09:17 +0000, Oleksiy Obitotskyi -X (oobitots
-
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 3 lines · GLOBALLOGIC INC at Cisco)
Could you look into this warning.
WARNING: util-linux-2.36.2-r0
do_package_qa: QA Issue: util-linux-
Quoted 2 lines
Quoted 2 lines
dev
Quoted 1 line
Quoted 1 line
rdepends on util-linux-libuuid-dev, but it isn't a build
dependency?
Quoted 5 lines
Quoted 5 lines
Quoted 3 lines
Quoted 3 lines
[build-deps]
Quoted 2 lines
https://autobuilder.yoctoproject.org/typhoon/#/builders/61/builds/3226
Quoted 1 line
Quoted 1 line
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
Quoted 3 lines
That failure was my fault when testing some fixes.
I've sent out a patch which renames util-linux-uuid to util-linux-
libuuid
Quoted 3 lines
Quoted 1 line
and sorts out the license issue Peter reported.
I don't mind the recipe being renamed and cleaned up, but I would
prefer
Quoted 3 lines
Quoted 3 lines
Quoted 1 line
to see my entire patch for the license parts being either integrated
before
Quoted 1 line
this or squashed into it, whichever you prefer. It does not make
sense
Quoted 2 lines
Quoted 2 lines
to
Quoted 1 line
use the same
LIC_FILES_CHKSUMfor util-linux-libuuid as for util-linux,
Quoted 5 lines
Quoted 5 lines
Quoted 1 line
and setting the other LICENSE variables in util-linux.inc no longer
makes
Quoted 1 line
sense as they are only relevant for util-linux.
I'm torn on that. Code with the other licenses is present, just not
used
Quoted 1 line
Quoted 1 line
in the final output and I personally suspect that having one
LIC_FILES_CHKSUMQuoted 1 line
Quoted 1 line
is going to be easier to maintain in the future rather than two
separate
Quoted 2 lines
Quoted 1 line
ones.
Quoted 8 lines
Quoted 1 line
I actually checked all the files that go into -dev and -src before
suggesting
Quoted 2 lines
this change, and all files are either marked as public domain or use a BSD-3-Clause license.
There is a difference between what ends up in ${S} and what ends up in the binary packages. LICENSE clearly governs the latter. Its the scope of
LIC_FILES_CHECKSUMwhich there are differences of opinion on.
Well, the latter governs what ends up in ${PN}-lic, so having a lot of unrelated (to the installed packages) license files in LIC_FILES_CHECKSUM does not make sense (to me). If everything that is built and (possibly) installed and thus distributed is covered by BSD-3-Clause licenses, why should ${PN}-lic include a lot of license files for unrelated code?
Quoted 3 lines
Cheers,
Richard
//Peter
Quoted 87 lines · Peter Kjellerstedt
Quoted 78 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 17:52
To: Peter Kjellerstedt <[email protected]>; Oleksiy Obitotskyi - X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca Bocassi <[email protected]>; [email protected]
Cc: [email protected]; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrappingQuoted 4 lines · Peter Kjellerstedt
Quoted 4 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 15:27
To: Peter Kjellerstedt <[email protected]>; OleksiyObitotskyi -
Quoted 1 line
Quoted 1 line
X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca
Bocassi
Quoted 6 lines
Quoted 6 lines
<[email protected]>; [email protected]
Cc: [email protected]; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrappingOn Thu, 2021-03-25 at 14:22 +0000, Peter Kjellerstedt wrote:
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 10:34
To: Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca Bocassi <[email protected]>; [email protected]
Cc: [email protected]; Peter Kjellerstedt <[email protected]>; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid inseparate
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
recipe to allow bootstrapping
On Thu, 2021-03-25 at 09:17 +0000, Oleksiy Obitotskyi -X (oobitots
-
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 3 lines · GLOBALLOGIC INC at Cisco)
Could you look into this warning.
WARNING: util-linux-2.36.2-r0
do_package_qa: QA Issue: util-linux-
Quoted 2 lines
Quoted 2 lines
dev
Quoted 1 line
Quoted 1 line
rdepends on util-linux-libuuid-dev, but it isn't a build
dependency?
Quoted 5 lines
Quoted 5 lines
Quoted 3 lines
Quoted 3 lines
[build-deps]
Quoted 2 lines
https://autobuilder.yoctoproject.org/typhoon/#/builders/61/builds/3226
Quoted 1 line
Quoted 1 line
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
Quoted 3 lines
That failure was my fault when testing some fixes.
I've sent out a patch which renames util-linux-uuid to util-linux-
libuuid
Quoted 3 lines
Quoted 1 line
and sorts out the license issue Peter reported.
I don't mind the recipe being renamed and cleaned up, but I would
prefer
Quoted 3 lines
Quoted 3 lines
Quoted 1 line
to see my entire patch for the license parts being either integrated
before
Quoted 1 line
this or squashed into it, whichever you prefer. It does not make
sense
Quoted 2 lines
Quoted 2 lines
to
Quoted 1 line
use the same
LIC_FILES_CHKSUMfor util-linux-libuuid as for util-linux,
Quoted 5 lines
Quoted 5 lines
Quoted 1 line
and setting the other LICENSE variables in util-linux.inc no longer
makes
Quoted 1 line
sense as they are only relevant for util-linux.
I'm torn on that. Code with the other licenses is present, just not
used
Quoted 1 line
Quoted 1 line
in the final output and I personally suspect that having one
LIC_FILES_CHKSUMQuoted 1 line
Quoted 1 line
is going to be easier to maintain in the future rather than two
separate
Quoted 2 lines
Quoted 1 line
ones.
Quoted 8 lines
Quoted 1 line
I actually checked all the files that go into -dev and -src before
suggesting
Quoted 2 lines
this change, and all files are either marked as public domain or use a BSD-3-Clause license.
There is a difference between what ends up in ${S} and what ends up in the binary packages. LICENSE clearly governs the latter. Its the scope of
LIC_FILES_CHECKSUMwhich there are differences of opinion on.
Quoted 5 lines
Well, the latter governs what ends up in ${PN}-lic, so having a lot of unrelated (to the installed packages) license files in
LIC_FILES_CHECKSUMdoes not make sense (to me). If everything that is built and (possibly) installed and thus distributed is covered by BSD-3-Clause licenses, why should ${PN}-lic include a lot of license files for unrelated code?
I hadn't considered ${PN}-lic :(.
We can't win. If we change LIC_FILES_CHKSUM we'll see complaints from people scanning the code that there are licenses present in WORKDIR that are not in LIC_FILES_CHKSUM. If we don't change it, ${PN}-lic does give more information than necessary. I still think the latter is probably safer and makes recipe upgrades easier.
Licensing in general needs a step back and an overhaul. Sadly people are generally only prepared to do this piecemeal solving their specific issue rather than the general case and big picture.
Cheers,
Richard
Quoted 95 lines · Richard Purdie
Quoted 87 lines · Peter Kjellerstedt
Quoted 78 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 17:52
To: Peter Kjellerstedt <[email protected]>; Oleksiy Obitotskyi - X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca Bocassi <[email protected]>; [email protected]
Cc: [email protected]; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrappingQuoted 4 lines · Peter Kjellerstedt
Quoted 4 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 15:27
To: Peter Kjellerstedt <[email protected]>; OleksiyObitotskyi -
Quoted 1 line
Quoted 1 line
X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca
Bocassi
Quoted 6 lines
Quoted 6 lines
<[email protected]>; [email protected]
Cc: [email protected]; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid in separate recipe to allow bootstrappingOn Thu, 2021-03-25 at 14:22 +0000, Peter Kjellerstedt wrote:
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
-----Original Message-----
From: Richard Purdie <[email protected]>
Sent: den 25 mars 2021 10:34
To: Oleksiy Obitotskyi -X (oobitots - GLOBALLOGIC INC at Cisco) <[email protected]>; Luca Bocassi <[email protected]>; [email protected]
Cc: [email protected]; Peter Kjellerstedt <[email protected]>; Khem Raj <[email protected]>
Subject: Re: [OE-core] [PATCH v11] util-linux: split uuid inseparate
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
recipe to allow bootstrapping
On Thu, 2021-03-25 at 09:17 +0000, Oleksiy Obitotskyi -X (oobitots
-
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 3 lines · GLOBALLOGIC INC at Cisco)
Could you look into this warning.
WARNING: util-linux-2.36.2-r0
do_package_qa: QA Issue: util-linux-
Quoted 2 lines
Quoted 2 lines
dev
Quoted 1 line
Quoted 1 line
rdepends on util-linux-libuuid-dev, but it isn't a build
dependency?
Quoted 5 lines
Quoted 5 lines
Quoted 3 lines
Quoted 3 lines
[build-deps]
Quoted 2 lines
https://autobuilder.yoctoproject.org/typhoon/#/builders/61/builds/3226
Quoted 1 line
Quoted 1 line
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
Quoted 3 lines
That failure was my fault when testing some fixes.
I've sent out a patch which renames util-linux-uuid to util-linux-
libuuid
Quoted 3 lines
Quoted 1 line
and sorts out the license issue Peter reported.
I don't mind the recipe being renamed and cleaned up, but I would
prefer
Quoted 3 lines
Quoted 3 lines
Quoted 1 line
to see my entire patch for the license parts being either integrated
before
Quoted 1 line
this or squashed into it, whichever you prefer. It does not make
sense
Quoted 2 lines
Quoted 2 lines
to
Quoted 1 line
use the same
LIC_FILES_CHKSUMfor util-linux-libuuid as for util-linux,
Quoted 5 lines
Quoted 5 lines
Quoted 1 line
and setting the other LICENSE variables in util-linux.inc no longer
makes
Quoted 1 line
sense as they are only relevant for util-linux.
I'm torn on that. Code with the other licenses is present, just not
used
Quoted 1 line
Quoted 1 line
in the final output and I personally suspect that having one
LIC_FILES_CHKSUMQuoted 1 line
Quoted 1 line
is going to be easier to maintain in the future rather than two
separate
Quoted 2 lines
Quoted 1 line
ones.
Quoted 8 lines
Quoted 1 line
I actually checked all the files that go into -dev and -src before
suggesting
Quoted 2 lines
this change, and all files are either marked as public domain or use a BSD-3-Clause license.
There is a difference between what ends up in ${S} and what ends up in the binary packages. LICENSE clearly governs the latter. Its the scope of
LIC_FILES_CHECKSUMwhich there are differences of opinion on.Quoted 5 lines
Well, the latter governs what ends up in ${PN}-lic, so having a lot of unrelated (to the installed packages) license files in
LIC_FILES_CHECKSUMdoes not make sense (to me). If everything that is built and (possibly) installed and thus distributed is covered by BSD-3-Clause licenses, why should ${PN}-lic include a lot of license files for unrelated code?
Quoted 5 lines
I hadn't considered ${PN}-lic :(.
We can't win. If we change
LIC_FILES_CHKSUMwe'll see complaints from people scanning the code that there are licenses present in WORKDIR that are not inLIC_FILES_CHKSUM.
If there's code in the upstream tar file etc which is not involved at all in the build of the one particular sub component you're interested in then this type of complaint can be solved by removing the unused code from ${S} as part of do_patch.
Quoted 17 lines
If we don't change it, ${PN}-lic does give
more information than necessary. I still think the latter is probably safer and makes recipe upgrades easier.Licensing in general needs a step back and an overhaul. Sadly people are generally only prepared to do this piecemeal solving their specific issue rather than the general case and big picture.
Cheers,
Richard
Quoted 11 lines · On 05/12/2019 20:13, [email protected] wrote:
Quoted 5 lines
From: Luca Boccassi <[email protected]>
In the next release util-linux will gain an (optional) build dependency on libcryptsetup. But libcryptsetup build-depends on util-linux for blkid (optional, can be disabled) and uuid
(mandatory).
Quoted 1 line
Split out util-linux-uuid in a different recipe to break the cycle.
<sigh>
Quoted 7 lines
At least please simply make util-linux-uuid just require util-linux and
make the configuration changes required (presumably, disabling all tools
but uuid).Ross
Hi Ross,
(please CC me, I am not receiving mails to the list. Thanks!)
I tried that at first, but it was so complex that the recipe turned out to be bigger than it is right now. And it wasn't even working, there were tons of conflicts because of who-knows-what it was picking up due to the dynamic generation.
Signature · 2 lines
Kind regards, Luca Boccassi