Backport relevant changes from Zeus
cover Backport relevant changes from Zeus
This patch set contains backports of our changes from Zeus that are relevant for Warrior.
//Peter
The following changes since commit 79a850a10a4b88a6d20d607b322542f947874323:
conf/poky: add Fedora 30 and Opensuse Leap 15.1 to supported distributions (2019-10-14 09:49:27 +0100)
are available in the Git repository at:
git://push.yoctoproject.org/poky-contrib pkj/warrior_sync
Niclas Svensson (1):
devtool: finish: Keep patches ordered when updating bbappend
Peter Kjellerstedt (7):
meson.bbclass: Remove the MESON_*_ARGS variables
nativesdk-meson: Remove some unused variables
devtool: Avoid failure for recipes with S == WORKDIR and no local
files
package_rpm.bbclass: Remove a misleading bb.note()
tzdata: Correct the packaging of /etc/localtime and /etc/timezone
devtool: finish: Add suppport for the --no-clean option lib/oe/lsb: Make sure the distro ID is always lowercased
meta/classes/meson.bbclass | 15 +++++---------- meta/classes/package_rpm.bbclass | 5 ++--- meta/lib/oe/lsb.py | 8 ++++---- .../meson/nativesdk-meson_0.49.2.bb | 5 ----- meta/recipes-extended/timezone/tzdata.bb | 5 +++-- scripts/lib/devtool/standard.py | 17 +++++++++-------- 6 files changed, 23 insertions(+), 32 deletions(-)
Signature · 1 line
2.21.0
PATCH 1/8 meson.bbclass: Remove the MESON_*_ARGS variables
From: Peter Kjellerstedt <[email protected]>
The options in ${HOST_CC_ARCH}${TOOLCHAIN_OPTIONS} are already passed via ${CC}/${CXX} and there is no reason to pass them a second time. Thus we can remove MESON_TOOLCHAIN_ARGS. And when it is removed, the other MESON_*_ARGS variables revert to the standard CFLAGS, CXXFLAGS and LDFLAGS, so just use them directly instead.
Apart from the obvious improvement with not passing a lot of options twice, this also solves a problem where -pie would be passed on the command line in a way that it would prevent building any dynamic libraries using meson if using a toolchain that is not built with --enable-default-pie and if security_flags.inc is used.
--- meta/classes/meson.bbclass | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-)
meta/classes/meson.bbclass +5 -10
@@ -30,11 +30,6 @@ MESONOPTS = " --prefix ${prefix} \ -Dcpp_args='${BUILD_CPPFLAGS} ${BUILD_CXXFLAGS}' \ -Dcpp_link_args='${BUILD_LDFLAGS}'" -MESON_TOOLCHAIN_ARGS = "${HOST_CC_ARCH}${TOOLCHAIN_OPTIONS}" -MESON_C_ARGS = "${MESON_TOOLCHAIN_ARGS} ${CFLAGS}" -MESON_CPP_ARGS = "${MESON_TOOLCHAIN_ARGS} ${CXXFLAGS}" -MESON_LINK_ARGS = "${MESON_TOOLCHAIN_ARGS} ${LDFLAGS}" - EXTRA_OEMESON_append = " ${PACKAGECONFIG_CONFARGS}" MESON_CROSS_FILE = ""
@@ -76,7 +71,7 @@ def meson_endian(prefix, d): bb.fatal("Cannot determine endianism for %s-%s" % (arch, os)) addtask write_config before do_configure -do_write_config[vardeps] += "MESON_C_ARGS MESON_CPP_ARGS MESON_LINK_ARGS CC CXX LD AR NM STRIP READELF" +do_write_config[vardeps] += "CC CXX LD AR NM STRIP READELF CFLAGS CXXFLAGS LDFLAGS" do_write_config() { # This needs to be Py to split the args into single-element lists cat >${WORKDIR}/meson.cross <<EOF
@@ -93,10 +88,10 @@ llvm-config = 'llvm-config8.0.0' [properties] needs_exe_wrapper = true -c_args = ${@meson_array('MESON_C_ARGS', d)} -c_link_args = ${@meson_array('MESON_LINK_ARGS', d)} -cpp_args = ${@meson_array('MESON_CPP_ARGS', d)} -cpp_link_args = ${@meson_array('MESON_LINK_ARGS', d)} +c_args = ${@meson_array('CFLAGS', d)} +c_link_args = ${@meson_array('LDFLAGS', d)} +cpp_args = ${@meson_array('CXXFLAGS', d)} +cpp_link_args = ${@meson_array('LDFLAGS', d)} gtkdoc_exe_wrapper = '${B}/gtkdoc-qemuwrapper' [host_machine]
Signature · 1 line
2.21.0
Re: [warrior][PATCH 1/8] meson.bbclass: Remove the MESON_*_ARGS variables
Quoted 7 lines · Peter Kjellerstedt
From: Peter Kjellerstedt <[email protected]>
The options in ${HOST_CC_ARCH}${TOOLCHAIN_OPTIONS} are already passed via ${CC}/${CXX} and there is no reason to pass them a second time. Thus we can remove
MESON_TOOLCHAIN_ARGS. And when it is removed, the other MESON_*_ARGS variables revert to the standard CFLAGS, CXXFLAGS and LDFLAGS, so just use them directly instead.
the above is not a reason to backport to warrior
Quoted 5 lines
Apart from the obvious improvement with not passing a lot of options twice, this also solves a problem where -pie would be passed on the command line in a way that it would prevent building any dynamic libraries using meson if using a toolchain that is not built with --enable-default-pie and if security_flags.inc is used.
but this is.
- armin
Quoted 47 lines
Signed-off-by Peter Kjellerstedt <[email protected]>Signed-off-by Richard Purdie <[email protected]>--- meta/classes/meson.bbclass | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-)meta/classes/meson.bbclass +5
-10@@ -30,11 +30,6 @@ MESONOPTS = " --prefix ${prefix} \ -Dcpp_args='${BUILD_CPPFLAGS} ${BUILD_CXXFLAGS}' \ -Dcpp_link_args='${BUILD_LDFLAGS}'" -MESON_TOOLCHAIN_ARGS = "${HOST_CC_ARCH}${TOOLCHAIN_OPTIONS}" -MESON_C_ARGS = "${MESON_TOOLCHAIN_ARGS} ${CFLAGS}" -MESON_CPP_ARGS = "${MESON_TOOLCHAIN_ARGS} ${CXXFLAGS}" -MESON_LINK_ARGS = "${MESON_TOOLCHAIN_ARGS} ${LDFLAGS}" - EXTRA_OEMESON_append = " ${PACKAGECONFIG_CONFARGS}" MESON_CROSS_FILE = ""@@ -76,7 +71,7 @@ def meson_endian(prefix, d): bb.fatal("Cannot determine endianism for %s-%s" % (arch, os)) addtask write_config before do_configure -do_write_config[vardeps] += "MESON_C_ARGS MESON_CPP_ARGS MESON_LINK_ARGS CC CXX LD AR NM STRIP READELF" +do_write_config[vardeps] += "CC CXX LD AR NM STRIP READELF CFLAGS CXXFLAGS LDFLAGS" do_write_config() { # This needs to be Py to split the args into single-element lists cat >${WORKDIR}/meson.cross <<EOF@@ -93,10 +88,10 @@ llvm-config = 'llvm-config8.0.0' [properties] needs_exe_wrapper = true -c_args = ${@meson_array('MESON_C_ARGS', d)} -c_link_args = ${@meson_array('MESON_LINK_ARGS', d)} -cpp_args = ${@meson_array('MESON_CPP_ARGS', d)} -cpp_link_args = ${@meson_array('MESON_LINK_ARGS', d)} +c_args = ${@meson_array('CFLAGS', d)} +c_link_args = ${@meson_array('LDFLAGS', d)} +cpp_args = ${@meson_array('CXXFLAGS', d)} +cpp_link_args = ${@meson_array('LDFLAGS', d)} gtkdoc_exe_wrapper = '${B}/gtkdoc-qemuwrapper' [host_machine]
PATCH 2/8 nativesdk-meson: Remove some unused variables
In reply to Peter KjellerstedtFrom: Peter Kjellerstedt <[email protected]>
--- meta/recipes-devtools/meson/nativesdk-meson_0.49.2.bb | 5 ----- 1 file changed, 5 deletions(-)
meta/recipes-devtools/meson/nativesdk-meson_0.49.2.bb +0 -5
@@ -16,11 +16,6 @@ def meson_endian(prefix, d): else: bb.fatal("Cannot determine endianism for %s-%s" % (arch, os)) -MESON_TOOLCHAIN_ARGS = "${BUILDSDK_CC_ARCH}${TOOLCHAIN_OPTIONS}" -MESON_C_ARGS = "${MESON_TOOLCHAIN_ARGS} ${BUILDSDK_CFLAGS}" -MESON_CPP_ARGS = "${MESON_TOOLCHAIN_ARGS} ${BUILDSDK_CXXFLAGS}" -MESON_LINK_ARGS = "${MESON_TOOLCHAIN_ARGS} ${BUILDSDK_LDFLAGS}" - # The cross file logic is similar but not identical to that in meson.bbclass, # since it's generating for an SDK rather than a cross-compile. Important # differences are:
Signature · 1 line
2.21.0
PATCH 3/8 devtool: Avoid failure for recipes with S == WORKDIR and no local files
In reply to Peter KjellerstedtFrom: Peter Kjellerstedt <[email protected]>
When extracting the sources for a recipe that has S == WORKDIR and no local files in the SRC_URI (which, e.g., can happen for a recipe with a URI that has the unpack=false attribute), the extraction fails with the following backtrace:
Traceback (most recent call last):
File ".../scripts/devtool", line 344, in <module>
ret = main()
File ".../scripts/devtool", line 331, in main
ret = args.func(args, config, basepath, workspace)
File ".../poky/scripts/lib/devtool/standard.py", line 762, in modify
initial_rev, _ = _extract_source(srctree, args.keep_temp, args.branch, False, config, basepath, workspace, args.fixed_setup, rd, tinfoil, no_overrides=args.no_overrides)
File ".../poky/scripts/lib/devtool/standard.py", line 647, in _extract_source
bb.process.run('git %s commit -a -m "Committing local file symlinks\n\n%s"' % (' '.join(useroptions), oe.patch.GitApplyTree.ignore_commit_prefix), cwd=srctree)
File ".../poky/bitbake/lib/bb/process.py", line 178, in run
raise ExecutionError(cmd, pipe.returncode, stdout, stderr)
bb.process.ExecutionError: Execution of 'git commit -a -m "Committing local file symlinks
%% ignore"' failed with exit code 1:
On branch devtool
nothing to commit, working tree clean
This is because no files were found in the oe-local-files directory and consequently no symbolic links were added using git add, but the git commit command was still executed.
--- scripts/lib/devtool/standard.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
scripts/lib/devtool/standard.py +3 -3
@@ -637,9 +637,9 @@ def _extract_source(srctree, keep_temp, devbranch, sync, config, basepath, works addfiles.append(os.path.join(relpth, fn)) if addfiles: bb.process.run('git add %s' % ' '.join(addfiles), cwd=srctree) - useroptions = [] - oe.patch.GitApplyTree.gitCommandUserOptions(useroptions, d=d) - bb.process.run('git %s commit -a -m "Committing local file symlinks\n\n%s"' % (' '.join(useroptions), oe.patch.GitApplyTree.ignore_commit_prefix), cwd=srctree) + useroptions = [] + oe.patch.GitApplyTree.gitCommandUserOptions(useroptions, d=d) + bb.process.run('git %s commit -m "Committing local file symlinks\n\n%s"' % (' '.join(useroptions), oe.patch.GitApplyTree.ignore_commit_prefix), cwd=srctree) if is_kernel_yocto: logger.info('Copying kernel config to srctree')
Signature · 1 line
2.21.0
PATCH 4/8 package_rpm.bbclass: Remove a misleading bb.note()
In reply to Peter KjellerstedtFrom: Peter Kjellerstedt <[email protected]>
It should have been removed in 3db9d865 (classes/package_rpm.bbclass:
Enhance diagnostic messages) when it was split in two new notes.
Also change the casing of two other notes to align them with the other notes.
--- meta/classes/package_rpm.bbclass | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
meta/classes/package_rpm.bbclass +2 -3
@@ -409,7 +409,6 @@ python write_specfile () { if not file_list and localdata.getVar('ALLOW_EMPTY', False) != "1": bb.note("Not creating empty RPM package for %s" % splitname) else: - bb.note("Creating RPM package for %s" % splitname) spec_files_top.append('%files') if extra_pkgdata: package_rpm_extra_pkgdata(splitname, spec_files_top, localdata)
@@ -418,7 +417,7 @@ python write_specfile () { bb.note("Creating RPM package for %s" % splitname) spec_files_top.extend(file_list) else: - bb.note("Creating EMPTY RPM Package for %s" % splitname) + bb.note("Creating empty RPM package for %s" % splitname) spec_files_top.append('') continue
@@ -510,7 +509,7 @@ python write_specfile () { bb.note("Creating RPM package for %s" % splitname) spec_files_bottom.extend(file_list) else: - bb.note("Creating EMPTY RPM Package for %s" % splitname) + bb.note("Creating empty RPM package for %s" % splitname) spec_files_bottom.append('') del localdata
Signature · 1 line
2.21.0
PATCH 5/8 tzdata: Correct the packaging of /etc/localtime and /etc/timezone
In reply to Peter KjellerstedtFrom: Peter Kjellerstedt <[email protected]>
During restructuring of the packaging in 2af4d6eb (tzdata: Install everything by default), these two files remained in the tzdata package, which is supposed to be empty. Move them to tzdata-core where they belong.
Also simplify the definition of CONFFILES_tzdata-core. As its value only takes effect for files that actually exist, there is no need to complicate its definition by checking if a file is created before adding it to the list of configuration files.
--- meta/recipes-extended/timezone/tzdata.bb | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
meta/recipes-extended/timezone/tzdata.bb +3 -2
@@ -147,6 +147,8 @@ FILES_tzdata-misc += "${datadir}/zoneinfo/Cuba \ RPROVIDES_tzdata-misc = "tzdata-misc" FILES_tzdata-core += " \ + ${sysconfdir}/localtime \ + ${sysconfdir}/timezone \ ${datadir}/zoneinfo/Pacific/Honolulu \ ${datadir}/zoneinfo/America/Anchorage \ ${datadir}/zoneinfo/America/Los_Angeles \
@@ -202,8 +204,7 @@ FILES_tzdata-core += " \ ${datadir}/zoneinfo/iso3166.tab \ ${datadir}/zoneinfo/Etc/*" -CONFFILES_tzdata-core += "${@ "${sysconfdir}/timezone" if bb.utils.to_boolean(d.getVar('INSTALL_TIMEZONE_FILE')) else "" }" -CONFFILES_tzdata-core += "${sysconfdir}/localtime" +CONFFILES_tzdata-core = "${sysconfdir}/localtime ${sysconfdir}/timezone" ALLOW_EMPTY_${PN} = "1" RDEPENDS_${PN} = "${TZ_PACKAGES}"
Signature · 1 line
2.21.0
PATCH 6/8 devtool: finish: Keep patches ordered when updating bbappend
In reply to Peter KjellerstedtFrom: Niclas Svensson <[email protected]>
The _get_patchset_revs() function returns the patches in an OrderedDict to keep them ordered. However, this information was lost when the patches were added to the bbappend file.
--- scripts/lib/devtool/standard.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
scripts/lib/devtool/standard.py +4 -4
@@ -1520,17 +1520,17 @@ def _update_recipe_patch(recipename, workspace, srctree, rd, appendlayerdir, wil patches_dir, changed_revs) logger.debug('Pre-filtering: update: %s, new: %s' % (dict(upd_p), dict(new_p))) if filter_patches: - new_p = {} - upd_p = {k:v for k,v in upd_p.items() if k in filter_patches} + new_p = OrderedDict() + upd_p = OrderedDict((k,v) for k,v in upd_p.items() if k in filter_patches) remove_files = [f for f in remove_files if f in filter_patches] updatefiles = False updaterecipe = False destpath = None srcuri = (rd.getVar('SRC_URI', False) or '').split() if appendlayerdir: - files = dict((os.path.join(local_files_dir, key), val) for + files = OrderedDict((os.path.join(local_files_dir, key), val) for key, val in list(upd_f.items()) + list(new_f.items())) - files.update(dict((os.path.join(patches_dir, key), val) for + files.update(OrderedDict((os.path.join(patches_dir, key), val) for key, val in list(upd_p.items()) + list(new_p.items()))) if files or remove_files: removevalues = None
Signature · 1 line
2.21.0
PATCH 7/8 devtool: finish: Add suppport for the --no-clean option
In reply to Peter KjellerstedtFrom: Peter Kjellerstedt <[email protected]>
This works just like the already existing --no-clean option to the devtool reset command.
--- scripts/lib/devtool/standard.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
scripts/lib/devtool/standard.py +2 -1
@@ -1895,7 +1895,7 @@ def finish(args, config, basepath, workspace): else: raise DevtoolError('Source tree is not clean:\n\n%s\nEnsure you have committed your changes or use -f/--force if you are sure there\'s nothing that needs to be committed' % dirty) - no_clean = False + no_clean = args.no_clean tinfoil = setup_tinfoil(basepath=basepath, tracking=True) try: rd = parse_recipe(config, tinfoil, args.recipename, True)
@@ -2169,6 +2169,7 @@ def register_commands(subparsers, context): parser_finish.add_argument('--mode', '-m', choices=['patch', 'srcrev', 'auto'], default='auto', help='Update mode (where %(metavar)s is %(choices)s; default is %(default)s)', metavar='MODE') parser_finish.add_argument('--initial-rev', help='Override starting revision for patches') parser_finish.add_argument('--force', '-f', action="store_true", help='Force continuing even if there are uncommitted changes in the source tree repository') + parser_finish.add_argument('--no-clean', '-n', action="store_true", help='Don\'t clean the sysroot to remove recipe output') parser_finish.add_argument('--no-overrides', '-O', action="store_true", help='Do not handle other override branches (if they exist)') parser_finish.add_argument('--dry-run', '-N', action="store_true", help='Dry-run (just report changes instead of writing them)') parser_finish.add_argument('--force-patch-refresh', action="store_true", help='Update patches in the layer even if they have not been modified (useful for refreshing patch context)')
Signature · 1 line
2.21.0
Re: [warrior][PATCH 7/8] devtool: finish: Add suppport for the --no-clean option
Quoted 4 lines · Peter Kjellerstedt
From: Peter Kjellerstedt <[email protected]>
This works just like the already existing --no-clean option to the
devtool resetcommand.
what problem is this solving in warrior?
-a rmin
Quoted 28 lines
Signed-off-by Peter Kjellerstedt <[email protected]>Signed-off-by Richard Purdie <[email protected]>--- scripts/lib/devtool/standard.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)scripts/lib/devtool/standard.py +2
-1@@ -1895,7 +1895,7 @@ def finish(args, config, basepath, workspace): else: raise DevtoolError('Source tree is not clean:\n\n%s\nEnsure you have committed your changes or use -f/--force if you are sure there\'s nothing that needs to be committed' % dirty) - no_clean = False + no_clean = args.no_clean tinfoil = setup_tinfoil(basepath=basepath, tracking=True) try: rd = parse_recipe(config, tinfoil, args.recipename, True)@@ -2169,6 +2169,7 @@ def register_commands(subparsers, context): parser_finish.add_argument('--mode', '-m', choices=['patch', 'srcrev', 'auto'], default='auto', help='Update mode (where %(metavar)s is %(choices)s; default is %(default)s)', metavar='MODE') parser_finish.add_argument('--initial-rev', help='Override starting revision for patches') parser_finish.add_argument('--force', '-f', action="store_true", help='Force continuing even if there are uncommitted changes in the source tree repository') + parser_finish.add_argument('--no-clean', '-n', action="store_true", help='Don\'t clean the sysroot to remove recipe output') parser_finish.add_argument('--no-overrides', '-O', action="store_true", help='Do not handle other override branches (if they exist)') parser_finish.add_argument('--dry-run', '-N', action="store_true", help='Dry-run (just report changes instead of writing them)') parser_finish.add_argument('--force-patch-refresh', action="store_true", help='Update patches in the layer even if they have not been modified (useful for refreshing patch context)')
Re: [warrior][PATCH 7/8] devtool: finish: Add suppport for the --no-clean option
Quoted 8 lines
-----Original Message-----
From: akuster808 <[email protected]>
Sent: den 16 oktober 2019 22:25
To: Peter Kjellerstedt <[email protected]>; openembedded- [email protected]
Subject: Re: [OE-core] [warrior][PATCH 7/8] devtool: finish: Add suppport for the --no-clean option
Quoted 7 lines
Quoted 4 lines · Peter Kjellerstedt
From: Peter Kjellerstedt <[email protected]>
This works just like the already existing --no-clean option to the
devtool resetcommand.what problem is this solving in warrior?
-a rmin
Keeping the development tools in sync with master so one can use the same options when working regardless of if one is working with Warrior or Zeus. Since it does not affect what is built and goes into the rootfs, it should not be a problem that it technically is a new feature.
//Peter
Quoted 44 lines
Review
Quoted 8 lines
Signed-off-by Peter Kjellerstedt <[email protected]>Signed-off-by Richard Purdie <[email protected]>--- scripts/lib/devtool/standard.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)(quoted diff) +0
-0
b/scripts/lib/devtool/standard.pyscripts/lib/devtool/standard.py def finish(args, config, basepath, workspace): · line 1896
Quoted 6 lines
index
b944ec3966..aeb9452a8a100644scripts/lib/devtool/standard.py +0
-0@@ -1895,7 +1895,7 @@ def finish(args, config, basepath, workspace): else: raise DevtoolError('Source tree is notclean:\n\n%s\nEnsure you have committed your changes or use -f/--force if you are sure there\'s nothing that needs to be committed' % dirty)
scripts/lib/devtool/standard.py def register_commands(subparsers, context): · line 2169
Quoted 8 lines
scripts/lib/devtool/standard.py +1
-1- no_clean = False + no_clean = args.no_clean tinfoil = setup_tinfoil(basepath=basepath, tracking=True) try: rd = parse_recipe(config, tinfoil, args.recipename, True)@@ -2169,6 +2169,7 @@ def register_commands(subparsers, context): parser_finish.add_argument('--mode', '-m', choices=['patch','srcrev', 'auto'], default='auto', help='Update mode (where %(metavar)s is %(choices)s; default is %(default)s)', metavar='MODE')
scripts/lib/devtool/standard.py def register_commands(subparsers, context): · line 2170
Quoted 1 line
scripts/lib/devtool/standard.py +0
-0parser_finish.add_argument('--initial-rev', help='Overridestarting revision for patches')
scripts/lib/devtool/standard.py def register_commands(subparsers, context): · line 2171
Quoted 1 line
scripts/lib/devtool/standard.py +0
-0parser_finish.add_argument('--force', '-f', action="store_true",help='Force continuing even if there are uncommitted changes in the source tree repository')
scripts/lib/devtool/standard.py def register_commands(subparsers, context): · line 2172
Quoted 1 line
scripts/lib/devtool/standard.py +1
-0+ parser_finish.add_argument('--no-clean', '-n',action="store_true", help='Don\'t clean the sysroot to remove recipe output')
scripts/lib/devtool/standard.py def register_commands(subparsers, context): · line 2173
Quoted 1 line
scripts/lib/devtool/standard.py +0
-0parser_finish.add_argument('--no-overrides', '-O',action="store_true", help='Do not handle other override branches (if they exist)')
scripts/lib/devtool/standard.py def register_commands(subparsers, context): · line 2174
Quoted 1 line
scripts/lib/devtool/standard.py +0
-0parser_finish.add_argument('--dry-run', '-N',action="store_true", help='Dry-run (just report changes instead of writing them)')
scripts/lib/devtool/standard.py def register_commands(subparsers, context): · line 2175
Quoted 1 line
scripts/lib/devtool/standard.py +0
-0parser_finish.add_argument('--force-patch-refresh',action="store_true", help='Update patches in the layer even if they have not been modified (useful for refreshing patch context)')
PATCH 8/8 lib/oe/lsb: Make sure the distro ID is always lowercased
In reply to Peter KjellerstedtFrom: Peter Kjellerstedt <[email protected]>
In commit 8689e561 (lib/oe/lsb: attempt to ensure consistent distro id regardless of source), the distro ID returned by oe.lsb.distro_identifier() was lowercased, but only if a release version is also present.
This changes the code to always lowercase the distro ID, including the default distro ID "unknown", which is used if no other ID can be identified.
--- meta/lib/oe/lsb.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
meta/lib/oe/lsb.py +4 -4
@@ -110,12 +110,12 @@ def distro_identifier(adjust_hook=None): if adjust_hook: distro_id, release = adjust_hook(distro_id, release) if not distro_id: - return "Unknown" - # Filter out any non-alphanumerics - distro_id = re.sub(r'\W', '', distro_id) + return "unknown" + # Filter out any non-alphanumerics and convert to lowercase + distro_id = re.sub(r'\W', '', distro_id).lower() if release: - id_str = '{0}-{1}'.format(distro_id.lower(), release) + id_str = '{0}-{1}'.format(distro_id, release) else: id_str = distro_id return id_str.replace(' ','-').replace('/','-')
Signature · 1 line
2.21.0
✗ patchtest: failure for Backport relevant changes from Zeus
In reply to Peter Kjellerstedt== Series Details ==
Series: Backport relevant changes from Zeus
Revision: 1
URL : https://patchwork.openembedded.org/series/20503/
State : failure
== Summary ==
Thank you for submitting this patch series to OpenEmbedded Core. This is an automated response. Several tests have been executed on the proposed series by patchtest resulting in the following failures:
* Issue Errors in your Python code were encountered [test_pylint]
Suggested fix Correct the lines introduced by your patch
Output Please, fix the listed issues:
scripts/lib/devtool/standard.py does not exist
If you believe any of these test results are incorrect, please reply to the mailing list ([email protected]) raising your concerns. Otherwise we would appreciate you correcting the issues and submitting a new version of the patchset if applicable. Please ensure you add/increment the version number when sending the new version (i.e. [PATCH] -> [PATCH v2] -> [PATCH v3] -> ...).
---
Guidelines: https://www.openembedded.org/wiki/Commit_Patch_Message_Guidelines Test framework: http://git.yoctoproject.org/cgit/cgit.cgi/patchtest Test suite: http://git.yoctoproject.org/cgit/cgit.cgi/patchtest-oe
Peter,
Quoted 5 lines · Peter Kjellerstedt
This patch set contains backports of our changes from Zeus that are relevant for Warrior.
//Peter
Quoted 7 lines
The following changes since commit
79a850a10a4b88a6d20d607b322542f947874323:conf/poky: add Fedora 30 and Opensuse Leap 15.1 to supported distributions (2019-10-14 09:49:27 +0100)
are available in the Git repository at:
git://push.yoctoproject.org/poky-contribpkj/warrior_sync
The series is causing build failures.
https://errors.yoctoproject.org/Errors/Details/274748/
https://errors.yoctoproject.org/Errors/Details/274757/
I have not had the time to figure out which ones giving I have been working on a different build failure. So if you could double check them, that would be great.
regards,
Armin
Quoted 22 lines
Niclas Svensson (1):
devtool: finish: Keep patches ordered when updating bbappendPeter Kjellerstedt (7):
meson.bbclass: Remove the MESON_*_ARGS variables
nativesdk-meson: Remove some unused variables
devtool: Avoid failure for recipes with S == WORKDIR and no local
files
package_rpm.bbclass: Remove a misleadingbb.note()
tzdata: Correct the packaging of/etc/localtimeand/etc/timezone
devtool: finish: Add suppport for the --no-clean optionlib/oe/lsb: Make sure the distro ID is always lowercasedmeta/classes/meson.bbclass | 15 +++++---------- meta/classes/package_rpm.bbclass | 5 ++--- meta/lib/oe/lsb.py | 8 ++++---- .../meson/nativesdk-meson_0.49.2.bb | 5 ----- meta/recipes-extended/timezone/tzdata.bb | 5 +++-- scripts/lib/devtool/standard.py | 17 +++++++++-------- 6 files changed, 23 insertions(+), 32 deletions(-)
Quoted 33 lines
-----Original Message-----
From: akuster808 <[email protected]>
Sent: den 27 oktober 2019 09:20
To: Peter Kjellerstedt <[email protected]>; openembedded- [email protected]
Subject: Re: [OE-core] [warrior][PATCH 0/8] Backport relevant changes from ZeusPeter,
Quoted 6 lines · Peter Kjellerstedt
This patch set contains backports of our changes from Zeus that are relevant for Warrior.
//Peter
The following changes since commit
79a850a10a4b88a6d20d607b322542f947874323:Quoted 2 lines
conf/poky: add Fedora 30 and Opensuse Leap 15.1 to supported
distributions (2019-10-14 09:49:27 +0100)
Quoted 4 lines
are available in the Git repository at:
git://push.yoctoproject.org/poky-contribpkj/warrior_syncThe series is causing build failures.
Quoted 7 lines
I have not had the time to figure out which ones giving I have been working on a different build failure. So if you could double check them,
that would be great.regards,
Armin
The problems are caused by the "meson.bbclass: Remove the MESON_*_ARGS variables" change. I have not had time to look into why it breaks yet, but I can reproduce it so it should hopefully not be too hard to figure it out once I have a little time.
//Peter
Quoted 23 lines
Quoted 13 lines
Niclas Svensson (1):
devtool: finish: Keep patches ordered when updating bbappendPeter Kjellerstedt (7):
meson.bbclass: Remove the MESON_*_ARGS variables
nativesdk-meson: Remove some unused variables
devtool: Avoid failure for recipes with S == WORKDIR and no local
files
package_rpm.bbclass: Remove a misleadingbb.note()
tzdata: Correct the packaging of/etc/localtimeand/etc/timezone
devtool: finish: Add suppport for the --no-clean optionlib/oe/lsb: Make sure the distro ID is always lowercasedQuoted 6 lines
meta/classes/meson.bbclass | 15 +++++---------- meta/classes/package_rpm.bbclass | 5 ++--- meta/lib/oe/lsb.py | 8 ++++---- .../meson/nativesdk-meson_0.49.2.bb | 5 ----- meta/recipes-extended/timezone/tzdata.bb | 5 +++-- scripts/lib/devtool/standard.py | 17 +++++++++--------
Quoted 2 lines
6 files changed, 23 insertions(+), 32 deletions(-)
Quoted 50 lines
-----Original Message-----
From: [email protected] <openembedded-core- [email protected]> On Behalf Of Peter Kjellerstedt
Sent: den 3 november 2019 16:18
To: akuster808 <[email protected]>; openembedded- [email protected]
Subject: Re: [OE-core] [warrior][PATCH 0/8] Backport relevant changes from ZeusQuoted 33 lines
-----Original Message-----
From: akuster808 <[email protected]>
Sent: den 27 oktober 2019 09:20
To: Peter Kjellerstedt <[email protected]>; openembedded- [email protected]
Subject: Re: [OE-core] [warrior][PATCH 0/8] Backport relevant changes from ZeusPeter,
Quoted 6 lines · Peter Kjellerstedt
This patch set contains backports of our changes from Zeus that are relevant for Warrior.
//Peter
The following changes since commit
79a850a10a4b88a6d20d607b322542f947874323:Quoted 2 lines
conf/poky: add Fedora 30 and Opensuse Leap 15.1 to supported
distributions (2019-10-14 09:49:27 +0100)
Quoted 4 lines
are available in the Git repository at:
git://push.yoctoproject.org/poky-contribpkj/warrior_syncThe series is causing build failures.
Quoted 7 lines
I have not had the time to figure out which ones giving I have been working on a different build failure. So if you could double check them,
that would be great.regards,
Armin
Quoted 6 lines
The problems are caused by the "meson.bbclass: Remove the MESON_*_ARGS variables" change. I have not had time to look into why it breaks yet, but I can reproduce it so it should hopefully not be too hard to figure it out once I have a little time.
//Peter
Ok, I have figured it out now. It turns out that the change to meson.bbclass uncovered a misuse of the meson.cross file. Over a year ago, Ross changed meson_array() to return a string instead of an array if only one argument is specified. This was for the benefit of cross tools specified in the meson.cross file, however it also affected properties. Up until now there has always been more than one argument for all properties so this did not change anything. However, my change to meson.bbclass to remove some unnecessary arguments caused the c_link_args for nativesdk to be left with only one argument.
Support for specifying a property as a string instead of a list was added to meson in version 0.50, which is why there never was a problem with Zeus due to the change to meson.bbclass. I have now backported the relevant part from meson 0.50 and included it in my patch series.
//Peter
Quoted 21 lines
Quoted 21 lines
Quoted 13 lines
Niclas Svensson (1):
devtool: finish: Keep patches ordered when updating bbappendPeter Kjellerstedt (7):
meson.bbclass: Remove the MESON_*_ARGS variables
nativesdk-meson: Remove some unused variables
devtool: Avoid failure for recipes with S == WORKDIR and no local
files
package_rpm.bbclass: Remove a misleadingbb.note()
tzdata: Correct the packaging of/etc/localtimeand/etc/timezone
devtool: finish: Add suppport for the --no-clean optionlib/oe/lsb: Make sure the distro ID is always lowercasedQuoted 6 lines
meta/classes/meson.bbclass | 15 +++++---------- meta/classes/package_rpm.bbclass | 5 ++--- meta/lib/oe/lsb.py | 8 ++++---- .../meson/nativesdk-meson_0.49.2.bb | 5 ----- meta/recipes-extended/timezone/tzdata.bb | 5 +++-- scripts/lib/devtool/standard.py | 17 +++++++++--------
Quoted 1 line
6 files changed, 23 insertions(+), 32 deletions(-)