Re: [OE-core] [PATCH v5 2/3] mesa: add support for RustiCL under PACKAGECONFIG "opencl"
On Fri, 12 Jan 2024 at 12:21, Böszörményi Zoltán wrote:
>
> 2024. 01. 02. 23:57 keltezéssel, Dmitry Baryshkov írta:
> > On Tue, 2 Jan 2024 at 17:30, Richard Purdie
> > wrote:
> >> On Tue, 2024-01-02 at 17:25 +0200, Dmitry Baryshkov wrote:
> >>> On Tue, 2 Jan 2024 at 13:40, Richard Purdie
> >>> wrote:
> On Tue, 2024-01-02 at 12:32 +0200, Dmitry Baryshkov wrote:
> > On Sat, 30 Dec 2023 at 23:02, Richard Purdie
> > wrote:
> >> On Sat, 2023-12-30 at 21:14 +0200, Dmitry Baryshkov wrote:
> >>> From: Zoltán Böszörményi
> >>>
> >>> Mesa now has two OpenCL implementations: Clover (older one) and
> >>> RustiCL
> >>> (newer one). Extend the "opencl" package config to include both
> >>> implementations into the libopencl-mesa (-dev) packages. As the Clover
> >>> implementation is in maintenance mode and is expected to be removed at
> >>> some point, it doesn't make sense to split the packages or to provide
> >>> two config options.
> >>>
> >>> Signed-off-by: Zoltán Böszörményi
> >>> [DB: reworked rust class inheritance]
> >>> [DB: merged "rusticl" and "clover" back to "opencl"]
> >>> Signed-off-by: Dmitry Baryshkov
> >>> ---
> >>> meta/recipes-graphics/mesa/mesa.inc | 21 -
> >>> 1 file changed, 16 insertions(+), 5 deletions(-)
> >>>
> >>> diff --git a/meta/recipes-graphics/mesa/mesa.inc
> >>> b/meta/recipes-graphics/mesa/mesa.inc
> >>> index efbce5caade5..a192562a7935 100644
> >>> --- a/meta/recipes-graphics/mesa/mesa.inc
> >>> +++ b/meta/recipes-graphics/mesa/mesa.inc
> >>> @@ -48,6 +48,12 @@ PROVIDES = " \
> >>>
> >>> inherit meson pkgconfig python3native gettext features_check
> >>>
> >>> +# We can not include rust conditionally, otherwise it can not be
> >>> overriden by
> >>> +# the bbappends. Instead always include it, but remove the
> >>> dependencies by
> >>> +# default.
> >>> +inherit rust
> >>> +INHIBIT_DEFAULT_RUST_DEPS = "${@bb.utils.contains('PACKAGECONFIG',
> >>> 'opencl', '', '1', d)}"
> >>> +
> >> I'd really prefer not to have the rust inherit unconditionally. I have
> >> some ideas about how we could fix this in bitbake so we may need to
> >> wait for that if we can't avoid the PACKAGECONFIG issue from appends...
> > Well, it's not just appends. Extending PACKAGECONFIG from local.conf
> > also doesn't work. In fact even adding "opencl" to the default
> > PACKAGECONFIG doesn't work unless I move `inherit
> > @{bb.utils.contains` after the PACKAGECONFIG definition. Using
> > `INHIBIT_DEFAULT_RUST_DEPS` works around that issue, but yes, it has a
> > drawback of including rust unconditionally.
> PACKAGECONFIG:append:pn-mesa = " opencl"
> >>> Yes, I tried that. I got an error about rust tools not being defined.
> >>> I think the :append: and overrides are handled after the inherit
> >>> argument is handled. So it is either PACKAGECONFIG being fully defined
> >>> before inherit (literally before), or inherit ends up with an empty
> >>> argument.
> >> The append and overrides would be expanded at the time the expression
> >> in the inherit statement is evaluated so if that didn't work there is
> >> something else at play...
> > I did not attempt debugging the bitbake itself. However I can see that
> > with this line in local.conf, PACKAGECONFIG is updated, but the RUSTC
> > is not defined (which means that rust.bbclass was not inherited). Then
> > I tried adding opencl to the default PACKAGECONFIG in mesa.inc. If
> > inherit comes before PACKAGECONFIG being set, it is not honoured. If
> > inherit comes after PACKAGECONFIG, it works as expected (and RUSTC
> > gets defined).
>
> Something like this should work:
>
> inherit ${@bb.utils.contains('PACKAGECONFIG', 'opencl', 'rust', '', d)}
>
> E.g. the libxml2 recipe does it, although for a different bbclass.
This way enabling opencl in .bbappend is not honoured.
--
With best wishes
Dmitry
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193600):
https://lists.openembedded.org/g/openembedded-core/message/193600
Mute This Topic: https://lists.openembedded.org/mt/103436170/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH] mdadm: Disable 10ddf-fail-spare and 10ddf-fail-stop-readd testcases
This always fails in my CI setup. There are some others which fail
intermittently however, this one is deterministic and fails always e.g.
AssertionError:- [1/1]
Failed ptests:
{'mdadm':
['/usr/lib/mdadm/ptest/tests/10ddf-fail-spareFAILED_-_see_/usr/lib/mdadm/ptest/logs/10ddf-fail-spare.log_and_/usr/lib/mdadm/ptest/lo
Mark them broken for now.
Signed-off-by: Khem Raj
---
meta/recipes-extended/mdadm/mdadm_4.2.bb | 2 ++
1 file changed, 2 insertions(+)
diff --git a/meta/recipes-extended/mdadm/mdadm_4.2.bb
b/meta/recipes-extended/mdadm/mdadm_4.2.bb
index fcfb4e1d502..3d09bc7d6b4 100644
--- a/meta/recipes-extended/mdadm/mdadm_4.2.bb
+++ b/meta/recipes-extended/mdadm/mdadm_4.2.bb
@@ -111,6 +111,8 @@ do_install_ptest() {
echo "intermittent failure on autobuilder" >
${D}${PTEST_PATH}/tests/19raid6check.broken
echo "intermittent failure on autobuilder" >
${D}${PTEST_PATH}/tests/20raid5journal.broken
echo "intermittent failure on autobuilder" >
${D}${PTEST_PATH}/tests/21raid5cache.broken
+ echo "intermittent failure on autobuilder" >
${D}${PTEST_PATH}/tests/10ddf-fail-spare.broken
+ echo "intermittent failure on autobuilder" >
${D}${PTEST_PATH}/tests/10ddf-fail-stop-readd.broken
}
RDEPENDS:${PN} += "bash"
--
2.43.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193599):
https://lists.openembedded.org/g/openembedded-core/message/193599
Mute This Topic: https://lists.openembedded.org/mt/103692859/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH] webkitgtk: Workaround for clang compiler segfault
This is a compiler crash which is under works upstream, until its fixed
lets disable musttail calls on clang+arm
Signed-off-by: Khem Raj
---
.../webkit/webkitgtk/no-musttail-arm.patch| 21 +++
meta/recipes-sato/webkit/webkitgtk_2.42.2.bb | 1 +
2 files changed, 22 insertions(+)
create mode 100644 meta/recipes-sato/webkit/webkitgtk/no-musttail-arm.patch
diff --git a/meta/recipes-sato/webkit/webkitgtk/no-musttail-arm.patch
b/meta/recipes-sato/webkit/webkitgtk/no-musttail-arm.patch
new file mode 100644
index 000..a6d67c1d703
--- /dev/null
+++ b/meta/recipes-sato/webkit/webkitgtk/no-musttail-arm.patch
@@ -0,0 +1,21 @@
+clang/arm: Do not use MUST_TAIL_CALL
+
+This causes clang-17 to crash see [1]
+this code is new in webkit 2.42[2] thats why we do not see the crash in older
webkit
+
+[1] https://github.com/llvm/llvm-project/issues/67767
+[2]
https://github.com/WebKit/WebKit/commit/4d816460b765acd8aef90ab474615850b91ecc35
+
+Upstream-Status: Inappropriate [work around to avoid clang compiler crash]
+Signed-off-by: Khem Raj
+--- a/Source/WTF/wtf/Compiler.h
b/Source/WTF/wtf/Compiler.h
+@@ -284,7 +284,7 @@
+ /* MUST_TAIL_CALL */
+
+ #if !defined(MUST_TAIL_CALL) && defined(__cplusplus) &&
defined(__has_cpp_attribute)
+-#if __has_cpp_attribute(clang::musttail)
++#if __has_cpp_attribute(clang::musttail) && !defined(__arm__)
+ #define MUST_TAIL_CALL [[clang::musttail]]
+ #endif
+ #endif
diff --git a/meta/recipes-sato/webkit/webkitgtk_2.42.2.bb
b/meta/recipes-sato/webkit/webkitgtk_2.42.2.bb
index 2ba60eb65ba..997aa5575b0 100644
--- a/meta/recipes-sato/webkit/webkitgtk_2.42.2.bb
+++ b/meta/recipes-sato/webkit/webkitgtk_2.42.2.bb
@@ -14,6 +14,7 @@ SRC_URI =
"https://www.webkitgtk.org/releases/${BPN}-${PV}.tar.xz \
file://reproducibility.patch \
file://0d3344e17d258106617b0e6d783d073b188a2548.patch \
file://0001-CMake-Add-a-variable-to-control-macro-__PAS_ALWAYS_I.patch \
+ file://no-musttail-arm.patch \
"
SRC_URI[sha256sum] =
"5720aa3e8627f1b9f63252187d4df0f8233ae71d697b1796ebfbe5ca750bd118"
--
2.43.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193598):
https://lists.openembedded.org/g/openembedded-core/message/193598
Mute This Topic: https://lists.openembedded.org/mt/103687376/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] Patchtest results for [PATCH v2 3/3] insane.bbclass: Python code cleanup in check_32bit_symbols
Thank you for your submission. Patchtest identified one or more issues with the patch. Please see the log below for more information: --- Testing patch /home/patchtest/share/mboxes/v2-3-3-insane.bbclass-Python-code-cleanup-in-check_32bit_symbols.patch FAIL: test commit message presence: Please include a commit message on your patch explaining the change (test_mbox.TestMbox.test_commit_message_presence) PASS: test Signed-off-by presence (test_mbox.TestMbox.test_signed_off_by_presence) PASS: test author valid (test_mbox.TestMbox.test_author_valid) PASS: test max line length (test_metadata.TestMetadata.test_max_line_length) PASS: test mbox format (test_mbox.TestMbox.test_mbox_format) PASS: test non-AUH upgrade (test_mbox.TestMbox.test_non_auh_upgrade) PASS: test shortlog format (test_mbox.TestMbox.test_shortlog_format) PASS: test shortlog length (test_mbox.TestMbox.test_shortlog_length) SKIP: pretest pylint: No python related patches, skipping test (test_python_pylint.PyLint.pretest_pylint) SKIP: pretest src uri left files: No modified recipes, skipping pretest (test_metadata.TestMetadata.pretest_src_uri_left_files) SKIP: test CVE check ignore: No modified recipes, skipping test (test_metadata.TestMetadata.test_cve_check_ignore) SKIP: test CVE tag format: No new CVE patches introduced (test_patch.TestPatch.test_cve_tag_format) SKIP: test Signed-off-by presence: No new CVE patches introduced (test_patch.TestPatch.test_signed_off_by_presence) SKIP: test Upstream-Status presence: No new CVE patches introduced (test_patch.TestPatch.test_upstream_status_presence_format) SKIP: test bugzilla entry format: No bug ID found (test_mbox.TestMbox.test_bugzilla_entry_format) SKIP: test lic files chksum modified not mentioned: No modified recipes, skipping test (test_metadata.TestMetadata.test_lic_files_chksum_modified_not_mentioned) SKIP: test lic files chksum presence: No added recipes, skipping test (test_metadata.TestMetadata.test_lic_files_chksum_presence) SKIP: test license presence: No added recipes, skipping test (test_metadata.TestMetadata.test_license_presence) SKIP: test pylint: No python related patches, skipping test (test_python_pylint.PyLint.test_pylint) SKIP: test series merge on head: Merge test is disabled for now (test_mbox.TestMbox.test_series_merge_on_head) SKIP: test src uri left files: No modified recipes, skipping pretest (test_metadata.TestMetadata.test_src_uri_left_files) SKIP: test summary presence: No added recipes, skipping test (test_metadata.TestMetadata.test_summary_presence) SKIP: test target mailing list: Series merged, no reason to check other mailing lists (test_mbox.TestMbox.test_target_mailing_list) --- Please address the issues identified and submit a new revision of the patch, or alternatively, reply to this email with an explanation of why the patch should be accepted. If you believe these results are due to an error in patchtest, please submit a bug at https://bugzilla.yoctoproject.org/ (use the 'Patchtest' category under 'Yocto Project Subprojects'). For more information on specific failures, see: https://wiki.yoctoproject.org/wiki/Patchtest. Thank you! -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#193597): https://lists.openembedded.org/g/openembedded-core/message/193597 Mute This Topic: https://lists.openembedded.org/mt/103684920/21656 Group Owner: openembedded-core+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH] siteinfo.bbclass: add support for darwin19 and darwin21
From: Etienne Cordonnier
Ported from the meta-darwin layer which supports darwin19 (honister version)
and darwin21 (kirkstone version).
Signed-off-by: Dominik Schnitzer
Signed-off-by: Etienne Cordonnier
---
meta/classes-recipe/siteinfo.bbclass | 4
1 file changed, 4 insertions(+)
diff --git a/meta/classes-recipe/siteinfo.bbclass
b/meta/classes-recipe/siteinfo.bbclass
index 05253ae90d..68aefb8eda 100644
--- a/meta/classes-recipe/siteinfo.bbclass
+++ b/meta/classes-recipe/siteinfo.bbclass
@@ -73,6 +73,8 @@ def siteinfo_data_for_machine(arch, os, d):
osinfo = {
"darwin": "common-darwin",
"darwin9": "common-darwin",
+"darwin19": "common-darwin",
+"darwin21": "common-darwin",
"linux": "common-linux common-glibc",
"linux-gnu": "common-linux common-glibc",
"linux-gnu_ilp32": "common-linux common-glibc",
@@ -130,6 +132,8 @@ def siteinfo_data_for_machine(arch, os, d):
"x86_64-cygwin": "bit-64",
"x86_64-darwin": "bit-64",
"x86_64-darwin9": "bit-64",
+"x86_64-darwin19": "bit-64",
+"x86_64-darwin21": "bit-64",
"x86_64-linux": "bit-64",
"x86_64-linux-musl": "x86_64-linux bit-64",
"x86_64-linux-muslx32": "bit-32 ix86-common x32-linux",
--
2.36.1.vfs.0.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193596):
https://lists.openembedded.org/g/openembedded-core/message/193596
Mute This Topic: https://lists.openembedded.org/mt/103684823/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH] chrpath.bbclass: fix Darwin support
From: Etienne Cordonnier
Ported from the meta-darwin layer:
The call to out.split("\n") expects a string, thus the parameter text=True is
needed (otherwise Popen returns a bytes object).
Note that "text" is just a more readable alias for universal_newlines.
Signed-off-by: Dominik Schnitzer
Signed-off-by: Etienne Cordonnier
---
meta/classes/chrpath.bbclass | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta/classes/chrpath.bbclass b/meta/classes/chrpath.bbclass
index 1aecb4df66..16729dcf61 100644
--- a/meta/classes/chrpath.bbclass
+++ b/meta/classes/chrpath.bbclass
@@ -62,7 +62,7 @@ def process_file_linux(cmd, fpath, rootdir, baseprefix,
tmpdir, d, break_hardlin
def process_file_darwin(cmd, fpath, rootdir, baseprefix, tmpdir, d,
break_hardlinks = False):
import subprocess as sub
-p = sub.Popen([d.expand("${HOST_PREFIX}otool"), '-L',
fpath],stdout=sub.PIPE,stderr=sub.PIPE)
+p = sub.Popen([d.expand("${HOST_PREFIX}otool"), '-L',
fpath],stdout=sub.PIPE,stderr=sub.PIPE, text=True)
out, err = p.communicate()
# If returned successfully, process stdout for results
if p.returncode != 0:
--
2.36.1.vfs.0.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193595):
https://lists.openembedded.org/g/openembedded-core/message/193595
Mute This Topic: https://lists.openembedded.org/mt/103684811/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] Patchtest results for [PATCH 3/3] insane.bbclass: Python code cleanup in check_32bit_symbols
Thank you for your submission. Patchtest identified one or more issues with the patch. Please see the log below for more information: --- Testing patch /home/patchtest/share/mboxes/3-3-insane.bbclass-Python-code-cleanup-in-check_32bit_symbols.patch FAIL: test Signed-off-by presence: Mbox is missing Signed-off-by. Add it manually or with "git commit --amend -s" (test_mbox.TestMbox.test_signed_off_by_presence) FAIL: test commit message presence: Please include a commit message on your patch explaining the change (test_mbox.TestMbox.test_commit_message_presence) PASS: test author valid (test_mbox.TestMbox.test_author_valid) PASS: test max line length (test_metadata.TestMetadata.test_max_line_length) PASS: test mbox format (test_mbox.TestMbox.test_mbox_format) PASS: test non-AUH upgrade (test_mbox.TestMbox.test_non_auh_upgrade) PASS: test shortlog format (test_mbox.TestMbox.test_shortlog_format) PASS: test shortlog length (test_mbox.TestMbox.test_shortlog_length) SKIP: pretest pylint: No python related patches, skipping test (test_python_pylint.PyLint.pretest_pylint) SKIP: pretest src uri left files: No modified recipes, skipping pretest (test_metadata.TestMetadata.pretest_src_uri_left_files) SKIP: test CVE check ignore: No modified recipes, skipping test (test_metadata.TestMetadata.test_cve_check_ignore) SKIP: test CVE tag format: No new CVE patches introduced (test_patch.TestPatch.test_cve_tag_format) SKIP: test Signed-off-by presence: No new CVE patches introduced (test_patch.TestPatch.test_signed_off_by_presence) SKIP: test Upstream-Status presence: No new CVE patches introduced (test_patch.TestPatch.test_upstream_status_presence_format) SKIP: test bugzilla entry format: No bug ID found (test_mbox.TestMbox.test_bugzilla_entry_format) SKIP: test lic files chksum modified not mentioned: No modified recipes, skipping test (test_metadata.TestMetadata.test_lic_files_chksum_modified_not_mentioned) SKIP: test lic files chksum presence: No added recipes, skipping test (test_metadata.TestMetadata.test_lic_files_chksum_presence) SKIP: test license presence: No added recipes, skipping test (test_metadata.TestMetadata.test_license_presence) SKIP: test pylint: No python related patches, skipping test (test_python_pylint.PyLint.test_pylint) SKIP: test series merge on head: Merge test is disabled for now (test_mbox.TestMbox.test_series_merge_on_head) SKIP: test src uri left files: No modified recipes, skipping pretest (test_metadata.TestMetadata.test_src_uri_left_files) SKIP: test summary presence: No added recipes, skipping test (test_metadata.TestMetadata.test_summary_presence) SKIP: test target mailing list: Series merged, no reason to check other mailing lists (test_mbox.TestMbox.test_target_mailing_list) --- Please address the issues identified and submit a new revision of the patch, or alternatively, reply to this email with an explanation of why the patch should be accepted. If you believe these results are due to an error in patchtest, please submit a bug at https://bugzilla.yoctoproject.org/ (use the 'Patchtest' category under 'Yocto Project Subprojects'). For more information on specific failures, see: https://wiki.yoctoproject.org/wiki/Patchtest. Thank you! -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#193594): https://lists.openembedded.org/g/openembedded-core/message/193594 Mute This Topic: https://lists.openembedded.org/mt/103684612/21656 Group Owner: openembedded-core+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[OE-core][PATCH v2] package.py: fix Darwin support
From: Etienne Cordonnier
- 'subprocess.Popen([d.expand("${HOST_PREFIX}otool)' requires text-mode (a more
readable alias for the universal_newlines parameter), since otool produces
text and the code 'out.split("\n")' expects a string, not a bytes object.
otool is used on MacOS only, so this error isn't triggered on Linux.
- use 'startswith("darwin")' in order to support all darwin versions and not
just specific versions (meta-darwin supports darwin21 at the moment).
Signed-off-by: Dominik Schnitzer
Signed-off-by: Etienne Cordonnier
---
meta/lib/oe/package.py | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/meta/lib/oe/package.py b/meta/lib/oe/package.py
index 9a465eaa09..702d8403be 100644
--- a/meta/lib/oe/package.py
+++ b/meta/lib/oe/package.py
@@ -1615,7 +1615,7 @@ def process_shlibs(pkgfiles, d):
sonames.add(prov)
if file.endswith('.dylib') or file.endswith('.so'):
rpath = []
-p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"), '-l',
file], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"), '-l',
file], stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
out, err = p.communicate()
# If returned successfully, process stdout for results
if p.returncode == 0:
@@ -1624,7 +1624,7 @@ def process_shlibs(pkgfiles, d):
if l.startswith('path '):
rpath.append(l.split()[1])
-p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"), '-L', file],
stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"), '-L', file],
stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
out, err = p.communicate()
# If returned successfully, process stdout for results
if p.returncode == 0:
@@ -1686,7 +1686,7 @@ def process_shlibs(pkgfiles, d):
soname = None
if cpath.islink(file):
continue
-if hostos == "darwin" or hostos == "darwin8":
+if hostos.startswith("darwin"):
darwin_so(file, needed, sonames, renames, pkgver)
elif hostos.startswith("mingw"):
mingw_dll(file, needed, sonames, renames, pkgver)
--
2.36.1.vfs.0.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193593):
https://lists.openembedded.org/g/openembedded-core/message/193593
Mute This Topic: https://lists.openembedded.org/mt/103684561/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [OE-core] [PATCH 3/3] insane.bbclass: Python code cleanup in check_32bit_symbols
Forgot the signed-off footer, v2 sent
/Ola
On Fri, Jan 12 2024, Ola x Nilsson wrote:
> ---
> meta/classes-global/insane.bbclass | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/meta/classes-global/insane.bbclass
> b/meta/classes-global/insane.bbclass
> index 1ff6a319c4..26457764ce 100644
> --- a/meta/classes-global/insane.bbclass
> +++ b/meta/classes-global/insane.bbclass
> @@ -514,9 +514,9 @@ def check_32bit_symbols(path, packagename, d, elf,
> messages):
> """
> Check that ELF files do not use any 32 bit time APIs from glibc.
> """
> -thirtytwo_bit_time_archs =
> set(('arm','armeb','mipsarcho32','powerpc','x86'))
> +thirtytwo_bit_time_archs = {'arm','armeb','mipsarcho32','powerpc','x86'}
> overrides = set(d.getVar('OVERRIDES').split(':'))
> -if not(thirtytwo_bit_time_archs & overrides):
> +if not (thirtytwo_bit_time_archs & overrides):
> return
>
> import re
> @@ -1606,7 +1606,7 @@ do_unpack[postfuncs] += "do_qa_unpack"
>
> python () {
> import re
> -
> +
> tests = d.getVar('ALL_QA').split()
> if "desktop" in tests:
> d.appendVar("PACKAGE_DEPENDS", " desktop-file-utils-native")
--
Ola x Nilsson
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193592):
https://lists.openembedded.org/g/openembedded-core/message/193592
Mute This Topic: https://lists.openembedded.org/mt/103684136/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH v2 3/3] insane.bbclass: Python code cleanup in check_32bit_symbols
Signed-off-by: Ola x Nilsson
---
meta/classes-global/insane.bbclass | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/meta/classes-global/insane.bbclass
b/meta/classes-global/insane.bbclass
index 1ff6a319c4..26457764ce 100644
--- a/meta/classes-global/insane.bbclass
+++ b/meta/classes-global/insane.bbclass
@@ -514,9 +514,9 @@ def check_32bit_symbols(path, packagename, d, elf,
messages):
"""
Check that ELF files do not use any 32 bit time APIs from glibc.
"""
-thirtytwo_bit_time_archs =
set(('arm','armeb','mipsarcho32','powerpc','x86'))
+thirtytwo_bit_time_archs = {'arm','armeb','mipsarcho32','powerpc','x86'}
overrides = set(d.getVar('OVERRIDES').split(':'))
-if not(thirtytwo_bit_time_archs & overrides):
+if not (thirtytwo_bit_time_archs & overrides):
return
import re
@@ -1606,7 +1606,7 @@ do_unpack[postfuncs] += "do_qa_unpack"
python () {
import re
-
+
tests = d.getVar('ALL_QA').split()
if "desktop" in tests:
d.appendVar("PACKAGE_DEPENDS", " desktop-file-utils-native")
--
2.39.2
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193590):
https://lists.openembedded.org/g/openembedded-core/message/193590
Mute This Topic: https://lists.openembedded.org/mt/103684365/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH v2 1/3] insane.bbclass: Check for adjtime in check_32_bit_symbols
adjtime was overlooked in the original commit. Signed-off-by: Ola x Nilsson --- meta/classes-global/insane.bbclass | 1 + 1 file changed, 1 insertion(+) diff --git a/meta/classes-global/insane.bbclass b/meta/classes-global/insane.bbclass index 6f3cd3026d..69741a6a79 100644 --- a/meta/classes-global/insane.bbclass +++ b/meta/classes-global/insane.bbclass @@ -534,6 +534,7 @@ def check_32bit_symbols(path, packagename, d, elf, messages): # /usr/include/signal.h "sigtimedwait", # /usr/include/sys/time.h +"adjtime", "futimes", "futimesat", "getitimer", "gettimeofday", "lutimes", "setitimer", "settimeofday", "utimes", # /usr/include/sys/timex.h -- 2.39.2 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#193591): https://lists.openembedded.org/g/openembedded-core/message/193591 Mute This Topic: https://lists.openembedded.org/mt/103684366/21656 Group Owner: openembedded-core+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH v2 2/3] insane.bbclass: Make check_32bit_symbols check for file functions too
Since _TIME_BITS=64 forces the use of _FILE_OFFSET_BITS=64, also check for functions redirected only based on _FILE_OFFSET_BITS and __USE_FILE_OFFSET64. Signed-off-by: Ola x Nilsson --- meta/classes-global/insane.bbclass | 31 ++ 1 file changed, 31 insertions(+) diff --git a/meta/classes-global/insane.bbclass b/meta/classes-global/insane.bbclass index 69741a6a79..1ff6a319c4 100644 --- a/meta/classes-global/insane.bbclass +++ b/meta/classes-global/insane.bbclass @@ -598,6 +598,37 @@ def check_32bit_symbols(path, packagename, d, elf, messages): "fts_set", # /usr/include/netdb.h "gai_suspend", + +# Since _TIME_BITS=64 forces the use of _FILE_OFFSET_BITS=64, +# also check for functions redirected only based on _FILE_OFFSET_BITS +# and __USE_FILE_OFFSET64 +# /usr/include/bits/fcntl-linux.h +"fallocate", +# /usr/include/bits/resource.h +"prlimit", +# /usr/include/sys/statfs.h +"statfs", "fstatfs", "statvfs", "fstatvfs", +# /usr/include/sys/sendfile.h +"sendfile", +# /usr/include/sys/resource.h +"getrlimit", "setrlimit", +# /usr/include/sys/uio.h +"preadv", "pwritev", "preadv2", "pwritev2", +# /usr/include/sys/mman.h +"mmap", +# /usr/include/stdlib.h +"mkstemp", "mkstemps", "mkostemp", "mkostemps", +# /usr/include/stdio.h +"fopen", "tmpfile", "freopen", "fseeko", "ftello", "fgetpos", +"fsetpos", +# /usr/include/dirent.h +"readdir", +"readdir_r", "scandir", "scandirat", "alphasort", "getdirentries", +"versionsort", +# /usr/include/unistd.h +"lseek", "pread", "pwrite", "truncate", "ftruncate", "lockf", +# /usr/include/fcntl.h +"open", "openat", "creat", "posix_fadvise", "posix_fallocate", } ptrn = re.compile( -- 2.39.2 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#193589): https://lists.openembedded.org/g/openembedded-core/message/193589 Mute This Topic: https://lists.openembedded.org/mt/103684364/21656 Group Owner: openembedded-core+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 3/3] insane.bbclass: Python code cleanup in check_32bit_symbols
---
meta/classes-global/insane.bbclass | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/meta/classes-global/insane.bbclass
b/meta/classes-global/insane.bbclass
index 1ff6a319c4..26457764ce 100644
--- a/meta/classes-global/insane.bbclass
+++ b/meta/classes-global/insane.bbclass
@@ -514,9 +514,9 @@ def check_32bit_symbols(path, packagename, d, elf,
messages):
"""
Check that ELF files do not use any 32 bit time APIs from glibc.
"""
-thirtytwo_bit_time_archs =
set(('arm','armeb','mipsarcho32','powerpc','x86'))
+thirtytwo_bit_time_archs = {'arm','armeb','mipsarcho32','powerpc','x86'}
overrides = set(d.getVar('OVERRIDES').split(':'))
-if not(thirtytwo_bit_time_archs & overrides):
+if not (thirtytwo_bit_time_archs & overrides):
return
import re
@@ -1606,7 +1606,7 @@ do_unpack[postfuncs] += "do_qa_unpack"
python () {
import re
-
+
tests = d.getVar('ALL_QA').split()
if "desktop" in tests:
d.appendVar("PACKAGE_DEPENDS", " desktop-file-utils-native")
--
2.39.2
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193588):
https://lists.openembedded.org/g/openembedded-core/message/193588
Mute This Topic: https://lists.openembedded.org/mt/103684136/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 1/3] insane.bbclass: Check for adjtime in check_32_bit_symbols
adjtime was overlooked in the original commit. Signed-off-by: Ola x Nilsson --- meta/classes-global/insane.bbclass | 1 + 1 file changed, 1 insertion(+) diff --git a/meta/classes-global/insane.bbclass b/meta/classes-global/insane.bbclass index 6f3cd3026d..69741a6a79 100644 --- a/meta/classes-global/insane.bbclass +++ b/meta/classes-global/insane.bbclass @@ -534,6 +534,7 @@ def check_32bit_symbols(path, packagename, d, elf, messages): # /usr/include/signal.h "sigtimedwait", # /usr/include/sys/time.h +"adjtime", "futimes", "futimesat", "getitimer", "gettimeofday", "lutimes", "setitimer", "settimeofday", "utimes", # /usr/include/sys/timex.h -- 2.39.2 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#193586): https://lists.openembedded.org/g/openembedded-core/message/193586 Mute This Topic: https://lists.openembedded.org/mt/103684131/21656 Group Owner: openembedded-core+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 2/3] insane.bbclass: Make check_32bit_symbols check for file functions too
Since _TIME_BITS=64 forces the use of _FILE_OFFSET_BITS=64, also check for functions redirected only based on _FILE_OFFSET_BITS and __USE_FILE_OFFSET64. Signed-off-by: Ola x Nilsson --- meta/classes-global/insane.bbclass | 31 ++ 1 file changed, 31 insertions(+) diff --git a/meta/classes-global/insane.bbclass b/meta/classes-global/insane.bbclass index 69741a6a79..1ff6a319c4 100644 --- a/meta/classes-global/insane.bbclass +++ b/meta/classes-global/insane.bbclass @@ -598,6 +598,37 @@ def check_32bit_symbols(path, packagename, d, elf, messages): "fts_set", # /usr/include/netdb.h "gai_suspend", + +# Since _TIME_BITS=64 forces the use of _FILE_OFFSET_BITS=64, +# also check for functions redirected only based on _FILE_OFFSET_BITS +# and __USE_FILE_OFFSET64 +# /usr/include/bits/fcntl-linux.h +"fallocate", +# /usr/include/bits/resource.h +"prlimit", +# /usr/include/sys/statfs.h +"statfs", "fstatfs", "statvfs", "fstatvfs", +# /usr/include/sys/sendfile.h +"sendfile", +# /usr/include/sys/resource.h +"getrlimit", "setrlimit", +# /usr/include/sys/uio.h +"preadv", "pwritev", "preadv2", "pwritev2", +# /usr/include/sys/mman.h +"mmap", +# /usr/include/stdlib.h +"mkstemp", "mkstemps", "mkostemp", "mkostemps", +# /usr/include/stdio.h +"fopen", "tmpfile", "freopen", "fseeko", "ftello", "fgetpos", +"fsetpos", +# /usr/include/dirent.h +"readdir", +"readdir_r", "scandir", "scandirat", "alphasort", "getdirentries", +"versionsort", +# /usr/include/unistd.h +"lseek", "pread", "pwrite", "truncate", "ftruncate", "lockf", +# /usr/include/fcntl.h +"open", "openat", "creat", "posix_fadvise", "posix_fallocate", } ptrn = re.compile( -- 2.39.2 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#193587): https://lists.openembedded.org/g/openembedded-core/message/193587 Mute This Topic: https://lists.openembedded.org/mt/103684132/21656 Group Owner: openembedded-core+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [OE-core] [PATCH] package.py: fix Darwin support
hi Alex,
as far as I know, universal_newlines=True is always required when executing
commands which produce text. I'll send an updated commit, but it's already
used a lot in the code-base, so I didn't think it needed an explanation.
See this example:
>>> p = subprocess.Popen(['ls', '-l'], stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
>>> stdout, stderr = p.communicate()
>>> print(stdout)
b'total 32\ndrwxrwxr-x 2 ecordonnier ecordonnier 4096 Sep 19 13:43
classes\ndrwxrwxr-x 3 ecordonnier ecordonnier 4096 Aug 31 17:32
conf\n-rw-rw-r-- 1 ecordonnier ecordonnier 1023 Jun 23 2022
COPYING.MIT\n-rw-rw-r-- 1 ecordonnier ecordonnier 1946 Sep 19 13:43
README\ndrwxrwxr-x 5 ecordonnier ecordonnier 4096 Jun 23 2022
recipes-core\ndrwxrwxr-x 13 ecordonnier ecordonnier 4096 Aug 31 17:32
recipes-devtools\ndrwxrwxr-x 3 ecordonnier ecordonnier 4096 Aug 31 17:32
recipes-extended\ndrwxrwxr-x 5 ecordonnier ecordonnier 4096 Aug 31 17:32
recipes-support\n'
>>> p = subprocess.Popen(['ls', '-l'], stdout=subprocess.PIPE,
stderr=subprocess.PIPE, universal_newlines=True)
>>> stdout, stderr = p.communicate()
>>> print(stdout)
total 32
drwxrwxr-x 2 ecordonnier ecordonnier 4096 Sep 19 13:43 classes
drwxrwxr-x 3 ecordonnier ecordonnier 4096 Aug 31 17:32 conf
-rw-rw-r-- 1 ecordonnier ecordonnier 1023 Jun 23 2022 COPYING.MIT
-rw-rw-r-- 1 ecordonnier ecordonnier 1946 Sep 19 13:43 README
drwxrwxr-x 5 ecordonnier ecordonnier 4096 Jun 23 2022 recipes-core
drwxrwxr-x 13 ecordonnier ecordonnier 4096 Aug 31 17:32 recipes-devtools
drwxrwxr-x 3 ecordonnier ecordonnier 4096 Aug 31 17:32 recipes-extended
drwxrwxr-x 5 ecordonnier ecordonnier 4096 Aug 31 17:32 recipes-support
On Fri, Jan 12, 2024 at 3:06 PM Alexander Kanavin
wrote:
> The universal_newlines=True changes need to be explained in the commit
> message. Why are they added?
>
> The need for hostos.startswith("darwin") is easier to guess, but that
> too, needs an explanation.
>
> Alex
>
> On Fri, 12 Jan 2024 at 14:53, Etienne Cordonnier via
> lists.openembedded.org
> wrote:
> >
> > From: Etienne Cordonnier
> >
> > Signed-off-by: Dominik Schnitzer
> > Signed-off-by: Etienne Cordonnier
> > ---
> > meta/lib/oe/package.py | 6 +++---
> > 1 file changed, 3 insertions(+), 3 deletions(-)
> >
> > diff --git a/meta/lib/oe/package.py b/meta/lib/oe/package.py
> > index 9a465eaa09..635efc9763 100644
> > --- a/meta/lib/oe/package.py
> > +++ b/meta/lib/oe/package.py
> > @@ -1615,7 +1615,7 @@ def process_shlibs(pkgfiles, d):
> > sonames.add(prov)
> > if file.endswith('.dylib') or file.endswith('.so'):
> > rpath = []
> > -p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"),
> '-l', file], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
> > +p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"),
> '-l', file], stdout=subprocess.PIPE, stderr=subprocess.PIPE,
> universal_newlines=True)
> > out, err = p.communicate()
> > # If returned successfully, process stdout for results
> > if p.returncode == 0:
> > @@ -1624,7 +1624,7 @@ def process_shlibs(pkgfiles, d):
> > if l.startswith('path '):
> > rpath.append(l.split()[1])
> >
> > -p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"), '-L',
> file], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
> > +p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"), '-L',
> file], stdout=subprocess.PIPE, stderr=subprocess.PIPE,
> universal_newlines=True)
> > out, err = p.communicate()
> > # If returned successfully, process stdout for results
> > if p.returncode == 0:
> > @@ -1686,7 +1686,7 @@ def process_shlibs(pkgfiles, d):
> > soname = None
> > if cpath.islink(file):
> > continue
> > -if hostos == "darwin" or hostos == "darwin8":
> > +if hostos.startswith("darwin"):
> > darwin_so(file, needed, sonames, renames, pkgver)
> > elif hostos.startswith("mingw"):
> > mingw_dll(file, needed, sonames, renames, pkgver)
> > --
> > 2.36.1.vfs.0.0
> >
> >
> >
> >
>
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193585):
https://lists.openembedded.org/g/openembedded-core/message/193585
Mute This Topic: https://lists.openembedded.org/mt/103682827/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] base-files ships /usr/src as 2755 in rpm (honister, kirkstone)
Hi, I have a very strange issue, base-files and *-src rpm packages have permission conflicts for /usr/src. As expected after reading the recipe and all the bbappends from the layers I use, it should use 0755 and according to devshell, it does: $ bitbake -c install base-files $ bitbake -c devshell base-files $ ls -ld image/usr/src drwxr-xr-x 2 root root 6 Jan 12 12:54 image/usr/src $ exit But for some reason, the base-files rpm package ends up with 2755 for this directory: $ bitbake base-files $ rpm -qp --dump deploy/glibc/rpm/intel_corei7_64/base-files-3.0.14-r0.intel_corei7_64.rpm 2>/dev/null | grep usr/src /usr/src 0 1302044400 042755 root root 0 0 0 X When I install a dbg and src subpackages for a package that I need to debug on the target machine, dnf and rpm complain about the conflicting permissions. I have tested it in honister and kirkstone where it's an issue. In mickledore and nanbield it's not, *-src subpackages also use 2755. What may have fixed this? I would like to backport it in my local trees. Thanks in advance, Zoltán Böszörményi -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#193584): https://lists.openembedded.org/g/openembedded-core/message/193584 Mute This Topic: https://lists.openembedded.org/mt/103683808/21656 Group Owner: openembedded-core+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH] bmaptool: add 3 fixes
I found a couple issues with bmaptool, including a race condition, which I
have fixed and submitted upstream. This patch adds these fixes to the project
now while waiting for feedback and a new release:
BmapCopy.py: fix error message
CLI.py: fix block device udev race condition
BmapCopy.py: tweak suggested udev rule
Signed-off-by: Trevor Woerner
---
.../bmap-tools/bmap-tools_git.bb | 7 +-
.../0001-BmapCopy.py-fix-error-message.patch | 36
...fix-block-device-udev-race-condition.patch | 83 +++
...mapCopy.py-tweak-suggested-udev-rule.patch | 43 ++
4 files changed, 168 insertions(+), 1 deletion(-)
create mode 100644
meta/recipes-support/bmap-tools/files/0001-BmapCopy.py-fix-error-message.patch
create mode 100644
meta/recipes-support/bmap-tools/files/0002-CLI.py-fix-block-device-udev-race-condition.patch
create mode 100644
meta/recipes-support/bmap-tools/files/0003-BmapCopy.py-tweak-suggested-udev-rule.patch
diff --git a/meta/recipes-support/bmap-tools/bmap-tools_git.bb
b/meta/recipes-support/bmap-tools/bmap-tools_git.bb
index effba2ecad68..9bbd7d51c831 100644
--- a/meta/recipes-support/bmap-tools/bmap-tools_git.bb
+++ b/meta/recipes-support/bmap-tools/bmap-tools_git.bb
@@ -9,7 +9,12 @@ SECTION = "console/utils"
LICENSE = "GPL-2.0-only"
LIC_FILES_CHKSUM = "file://LICENSE;md5=b234ee4d69f5fce4486a80fdaf4a4263"
-SRC_URI = "git://github.com/intel/${BPN};branch=main;protocol=https"
+FILESEXTRAPATHS:prepend := "${THISDIR}/files:"
+SRC_URI = "git://github.com/intel/${BPN};branch=main;protocol=https \
+ file://0001-BmapCopy.py-fix-error-message.patch \
+ file://0002-CLI.py-fix-block-device-udev-race-condition.patch \
+ file://0003-BmapCopy.py-tweak-suggested-udev-rule.patch \
+ "
SRCREV = "d84a6fd202fe246a0bc19ed2082e41bcdd75fb13"
S = "${WORKDIR}/git"
diff --git
a/meta/recipes-support/bmap-tools/files/0001-BmapCopy.py-fix-error-message.patch
b/meta/recipes-support/bmap-tools/files/0001-BmapCopy.py-fix-error-message.patch
new file mode 100644
index ..ddac322e9135
--- /dev/null
+++
b/meta/recipes-support/bmap-tools/files/0001-BmapCopy.py-fix-error-message.patch
@@ -0,0 +1,36 @@
+From ad0b0513a46c7d238d0fdabee0267c7084b75e84 Mon Sep 17 00:00:00 2001
+From: Trevor Woerner
+Date: Thu, 11 Jan 2024 22:04:23 -0500
+Subject: [PATCH 1/3] BmapCopy.py: fix error message
+
+The wrong variable was being used when attempting to print out an informative
+message to the user. Leading to nonsense messages such as:
+
+ bmaptool: info: failed to enable I/O optimization, expect suboptimal
speed (reason: cannot switch to the 1 I/O scheduler: 1 in use. None)
+
+Upstream-Status: Submitted [https://github.com/intel/bmap-tools/pull/129]
+Signed-off-by: Trevor Woerner
+---
+ bmaptools/BmapCopy.py | 6 +++---
+ 1 file changed, 3 insertions(+), 3 deletions(-)
+
+diff --git a/bmaptools/BmapCopy.py b/bmaptools/BmapCopy.py
+index 9de7ef434233..b1e8e0fcbdb7 100644
+--- a/bmaptools/BmapCopy.py
b/bmaptools/BmapCopy.py
+@@ -892,9 +892,9 @@ class BmapBdevCopy(BmapCopy):
+ _log.info(
+ "failed to enable I/O optimization, expect "
+ "suboptimal speed (reason: cannot switch to the "
+-f"{max_ratio_chg.temp_value} I/O scheduler: "
+-f"{max_ratio_chg.old_value or 'unknown scheduler'} in
use. "
+-f"{max_ratio_chg.error})"
++f"'{scheduler_chg.temp_value}' I/O scheduler: "
++f"'{scheduler_chg.old_value or 'unknown scheduler'}' in
use. "
++f"{scheduler_chg.error})"
+ )
+ if max_ratio_chg.error or scheduler_chg.error:
+ _log.info(
+--
+2.43.0.76.g1a87c842ece3
+
diff --git
a/meta/recipes-support/bmap-tools/files/0002-CLI.py-fix-block-device-udev-race-condition.patch
b/meta/recipes-support/bmap-tools/files/0002-CLI.py-fix-block-device-udev-race-condition.patch
new file mode 100644
index ..ea2749a26432
--- /dev/null
+++
b/meta/recipes-support/bmap-tools/files/0002-CLI.py-fix-block-device-udev-race-condition.patch
@@ -0,0 +1,83 @@
+From 34f4321dfce28697f830639260076e60d765698b Mon Sep 17 00:00:00 2001
+From: Trevor Woerner
+Date: Fri, 12 Jan 2024 01:16:19 -0500
+Subject: [PATCH 2/3] CLI.py: fix block device udev race condition
+
+We are encouraged to add a udev rule to change a block device's
+bdi/max_ratio to '1' and queue/scheduler to 'none', which I did.
+So I was surprised when, about 50% of the time, I kept seeing:
+
+ ...
+ bmaptool: info: failed to enable I/O optimization, expect suboptimal
speed (reason: cannot switch to the 'none' I/O scheduler: 'bfq' in use. [Errno
13] Permission denied: '/sys/dev/block/8:160/queue/scheduler')
+ bmaptool: info: You may want to set these I/O optimizations through a
udev rule like this:
+ ...
+
+The strange part is
Re: [OE-core] [PATCH] package.py: fix Darwin support
The universal_newlines=True changes need to be explained in the commit
message. Why are they added?
The need for hostos.startswith("darwin") is easier to guess, but that
too, needs an explanation.
Alex
On Fri, 12 Jan 2024 at 14:53, Etienne Cordonnier via
lists.openembedded.org
wrote:
>
> From: Etienne Cordonnier
>
> Signed-off-by: Dominik Schnitzer
> Signed-off-by: Etienne Cordonnier
> ---
> meta/lib/oe/package.py | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/meta/lib/oe/package.py b/meta/lib/oe/package.py
> index 9a465eaa09..635efc9763 100644
> --- a/meta/lib/oe/package.py
> +++ b/meta/lib/oe/package.py
> @@ -1615,7 +1615,7 @@ def process_shlibs(pkgfiles, d):
> sonames.add(prov)
> if file.endswith('.dylib') or file.endswith('.so'):
> rpath = []
> -p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"), '-l',
> file], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
> +p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"), '-l',
> file], stdout=subprocess.PIPE, stderr=subprocess.PIPE,
> universal_newlines=True)
> out, err = p.communicate()
> # If returned successfully, process stdout for results
> if p.returncode == 0:
> @@ -1624,7 +1624,7 @@ def process_shlibs(pkgfiles, d):
> if l.startswith('path '):
> rpath.append(l.split()[1])
>
> -p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"), '-L', file],
> stdout=subprocess.PIPE, stderr=subprocess.PIPE)
> +p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"), '-L', file],
> stdout=subprocess.PIPE, stderr=subprocess.PIPE, universal_newlines=True)
> out, err = p.communicate()
> # If returned successfully, process stdout for results
> if p.returncode == 0:
> @@ -1686,7 +1686,7 @@ def process_shlibs(pkgfiles, d):
> soname = None
> if cpath.islink(file):
> continue
> -if hostos == "darwin" or hostos == "darwin8":
> +if hostos.startswith("darwin"):
> darwin_so(file, needed, sonames, renames, pkgver)
> elif hostos.startswith("mingw"):
> mingw_dll(file, needed, sonames, renames, pkgver)
> --
> 2.36.1.vfs.0.0
>
>
>
>
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193582):
https://lists.openembedded.org/g/openembedded-core/message/193582
Mute This Topic: https://lists.openembedded.org/mt/103682827/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH] package.py: fix Darwin support
From: Etienne Cordonnier
Signed-off-by: Dominik Schnitzer
Signed-off-by: Etienne Cordonnier
---
meta/lib/oe/package.py | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/meta/lib/oe/package.py b/meta/lib/oe/package.py
index 9a465eaa09..635efc9763 100644
--- a/meta/lib/oe/package.py
+++ b/meta/lib/oe/package.py
@@ -1615,7 +1615,7 @@ def process_shlibs(pkgfiles, d):
sonames.add(prov)
if file.endswith('.dylib') or file.endswith('.so'):
rpath = []
-p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"), '-l',
file], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"), '-l',
file], stdout=subprocess.PIPE, stderr=subprocess.PIPE, universal_newlines=True)
out, err = p.communicate()
# If returned successfully, process stdout for results
if p.returncode == 0:
@@ -1624,7 +1624,7 @@ def process_shlibs(pkgfiles, d):
if l.startswith('path '):
rpath.append(l.split()[1])
-p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"), '-L', file],
stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+p = subprocess.Popen([d.expand("${HOST_PREFIX}otool"), '-L', file],
stdout=subprocess.PIPE, stderr=subprocess.PIPE, universal_newlines=True)
out, err = p.communicate()
# If returned successfully, process stdout for results
if p.returncode == 0:
@@ -1686,7 +1686,7 @@ def process_shlibs(pkgfiles, d):
soname = None
if cpath.islink(file):
continue
-if hostos == "darwin" or hostos == "darwin8":
+if hostos.startswith("darwin"):
darwin_so(file, needed, sonames, renames, pkgver)
elif hostos.startswith("mingw"):
mingw_dll(file, needed, sonames, renames, pkgver)
--
2.36.1.vfs.0.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193581):
https://lists.openembedded.org/g/openembedded-core/message/193581
Mute This Topic: https://lists.openembedded.org/mt/103682827/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH] cmake.bbclass: add Darwin support
From: Etienne Cordonnier
This was initially added by
https://git.yoctoproject.org/meta-darwin/commit/?id=97a84e083d0c2812e44231d48fb478e9a9cfac79
to the meta-darwin repository, however having it directory in openembedded-core
will make the maintenance of meta-darwin easier.
Signed-off-by: Dominik Schnitzer
Signed-off-by: Etienne Cordonnier
---
meta/classes-recipe/cmake.bbclass | 2 ++
1 file changed, 2 insertions(+)
diff --git a/meta/classes-recipe/cmake.bbclass
b/meta/classes-recipe/cmake.bbclass
index d978b88944..1e353f660f 100644
--- a/meta/classes-recipe/cmake.bbclass
+++ b/meta/classes-recipe/cmake.bbclass
@@ -72,6 +72,8 @@ OECMAKE_TARGET_COMPILE ?= "all"
OECMAKE_TARGET_INSTALL ?= "install"
def map_host_os_to_system_name(host_os):
+if host_os.startswith('darwin'):
+return 'Darwin'
if host_os.startswith('mingw'):
return 'Windows'
if host_os.startswith('linux'):
--
2.36.1.vfs.0.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193580):
https://lists.openembedded.org/g/openembedded-core/message/193580
Mute This Topic: https://lists.openembedded.org/mt/103682815/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 2/2] python: update 3.11.5 -> 3.12.1
Drop distutils and smtpd modules from packaging, as both are gone in 3.12.
Rebase:
0020-configure.ac-setup.py-do-not-add-a-curses-include-pa.patch
(drop setup.py chunk as the file is gone)
Drop patches:
0001-Do-not-add-usr-lib-termcap-to-linker-flags-to-avoid-.patch
(setup.py gone, lib/termcap not mentioned anywhere else)
0001-Don-t-search-system-for-headers-libraries.patch
(setup.py gone, usr/lib64 not mentioned anywhere else)
0001-Makefile-do-not-compile-.pyc-in-parallel.patch
(replaced with COMPILEALL_OPTS= in EXTRA_OEMAKE)
0001-setup.py-Do-not-detect-multiarch-paths-when-cross-co.patch
(setup.py gone, add_multiarch_paths not mentioned anywhere else)
0017-setup.py-do-not-report-missing-dependencies-for-disa.patch
(has been superseded by Setup.local tweak in do_configure:prepend)
12-distutils-prefix-is-inside-staging-area.patch
(distutils has been removed upstream, so this old, unplesant hack can be
finally dropped)
avoid_warning_about_tkinter.patch
(setup.py gone, tkinter detection logic performed in configure.ac)
Signed-off-by: Alexander Kanavin
---
meta/classes-recipe/python3-dir.bbclass | 2 +-
...shebang-overflow-on-python-config.py.patch | 8 +-
...ib-termcap-to-linker-flags-to-avoid-.patch | 25 -
...-search-system-for-headers-libraries.patch | 27 -
...e-stdin-I-O-errors-same-way-as-maste.patch | 6 +-
...-use-prefix-value-from-build-configu.patch | 30 +++---
...file-do-not-compile-.pyc-in-parallel.patch | 65
...-qemu-wrapper-when-gathering-profile.patch | 16 +--
...sts-due-to-load-variability-on-YP-AB.patch | 24 ++---
...est_sysconfig-for-posix_user-purelib.patch | 10 +-
...asename-to-replace-CC-for-checking-c.patch | 44
...detect-multiarch-paths-when-cross-co.patch | 42
..._fileno-test-due-to-load-variability.patch | 10 +-
...g.py-use-platlibdir-also-for-purelib.patch | 6 +-
...pes.test_find-skip-without-tools-sdk.patch | 10 +-
...le.py-correct-the-test-output-format.patch | 6 +-
...orlines-skip-due-to-load-variability.patch | 10 +-
...report-missing-dependencies-for-disa.patch | 38 ---
...up.py-do-not-add-a-curses-include-pa.patch | 37 ++-
...tutils-prefix-is-inside-staging-area.patch | 58 --
.../python3/avoid_warning_about_tkinter.patch | 30 --
.../python/python3/cgi_py.patch | 4 +-
.../python/python3/crosspythonpath.patch | 6 +-
.../python3/deterministic_imports.patch | 18 +++-
.../python/python3/makerace.patch | 10 +-
.../python/python3/python3-manifest.json | 100 +++---
.../{python3_3.11.5.bb => python3_3.12.1.bb} | 27 +++--
27 files changed, 154 insertions(+), 515 deletions(-)
delete mode 100644
meta/recipes-devtools/python/python3/0001-Do-not-add-usr-lib-termcap-to-linker-flags-to-avoid-.patch
delete mode 100644
meta/recipes-devtools/python/python3/0001-Don-t-search-system-for-headers-libraries.patch
delete mode 100644
meta/recipes-devtools/python/python3/0001-Makefile-do-not-compile-.pyc-in-parallel.patch
delete mode 100644
meta/recipes-devtools/python/python3/0001-setup.py-Do-not-detect-multiarch-paths-when-cross-co.patch
delete mode 100644
meta/recipes-devtools/python/python3/0017-setup.py-do-not-report-missing-dependencies-for-disa.patch
delete mode 100644
meta/recipes-devtools/python/python3/12-distutils-prefix-is-inside-staging-area.patch
delete mode 100644
meta/recipes-devtools/python/python3/avoid_warning_about_tkinter.patch
rename meta/recipes-devtools/python/{python3_3.11.5.bb => python3_3.12.1.bb}
(96%)
diff --git a/meta/classes-recipe/python3-dir.bbclass
b/meta/classes-recipe/python3-dir.bbclass
index d93d337f76c..3d07de99b88 100644
--- a/meta/classes-recipe/python3-dir.bbclass
+++ b/meta/classes-recipe/python3-dir.bbclass
@@ -4,7 +4,7 @@
# SPDX-License-Identifier: MIT
#
-PYTHON_BASEVERSION = "3.11"
+PYTHON_BASEVERSION = "3.12"
PYTHON_ABI = ""
PYTHON_DIR = "python${PYTHON_BASEVERSION}"
PYTHON_PN = "python3"
diff --git
a/meta/recipes-devtools/python/python3/0001-Avoid-shebang-overflow-on-python-config.py.patch
b/meta/recipes-devtools/python/python3/0001-Avoid-shebang-overflow-on-python-config.py.patch
index db084350045..03ecda98c2c 100644
---
a/meta/recipes-devtools/python/python3/0001-Avoid-shebang-overflow-on-python-config.py.patch
+++
b/meta/recipes-devtools/python/python3/0001-Avoid-shebang-overflow-on-python-config.py.patch
@@ -1,4 +1,4 @@
-From 78f482b91d94b44a02e02c4580166757119061ea Mon Sep 17 00:00:00 2001
+From 33bf96f31c4873032ace2cc9afacc3785997527a Mon Sep 17 00:00:00 2001
From: Paulo Neves
Date: Tue, 7 Jun 2022 16:16:41 +0200
Subject: [PATCH] Avoid shebang overflow on python-config.py
@@ -17,13 +17,13 @@ Upstream-Status: Denied [distribution]
1 file changed, 2 insertions(+)
diff --git a/Makefile.pre.in b/Makefile.pre.in
-index 55c7c46..1f6500a 100644
+index fb0b525..75d2e73 100644
--- a/Makefile.pre.in
+++ b/Makefile.pre.in
-@@ -2115,6
[OE-core] [PATCH 1/2] glib-2.0: pull in python3-dbusmock unconditionally
With python 3.12 some of the glib ptests started failing. Inspection
revealed that they fail because all tests in some glib's test suites are
skipped;
python 3.12's unittest module deems "no tests in a test suite were executed"
a failure (and I tend to agree).
Why are all the tests skipped? Because python dbus module is missing from
the image, and it's missing because it's pulled in by dbusmock, which
in turn is missing because it is pulled in subject to GI_DATA_ENABLED, and
that variable is not global but defined in a g-i class that glib does
not include and never did. So it was simply always excluded regardless
of g-i settings.
Back when this condition was added python3-dbusmock wouldn't build without
introspection data enabled; now it does, so we can again pull it
into ptest dependencies without conditions.
Signed-off-by: Alexander Kanavin
---
meta/recipes-core/glib-2.0/glib.inc | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta/recipes-core/glib-2.0/glib.inc
b/meta/recipes-core/glib-2.0/glib.inc
index ae6fc519922..9562bc23ebb 100644
--- a/meta/recipes-core/glib-2.0/glib.inc
+++ b/meta/recipes-core/glib-2.0/glib.inc
@@ -175,7 +175,7 @@ RDEPENDS:${PN}-ptest += "\
${PN}-locale-th \
python3-core \
python3-modules \
-${@bb.utils.contains('GI_DATA_ENABLED', 'True',
'python3-dbusmock', '', d)} \
+python3-dbusmock \
${PN}-codegen \
locale-base-de-de \
locale-base-es-es \
--
2.39.2
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193578):
https://lists.openembedded.org/g/openembedded-core/message/193578
Mute This Topic: https://lists.openembedded.org/mt/103681543/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [OE-core] [PATCH 13/21] alsa-utils-scripts: merge into alsa-utils
On Fri, 12 Jan 2024 at 10:14, Marko, Peter wrote: > I'm upgrading my layer from kirkstone to scarthgap and observed that my image > failed due to gpl3 license. > > I want to conveniently install whole alsa-utils except for parts where > license forbids me to do it. > After your path I would need to list all alsa-utils subpackages except the > scripts. > And on top of that during every core layer update I'd have to check if there > is not a new one. > > So from this point of view, your statement > "There is no reason to keep the recipe separate: bash dependency is not a > problem until the alsa-utils-scripts package is explicitly installed into a > target image." > Is not true, it's a problem for maintenance. > > Would something like this be accepted when I'd submit it? > This would return alsa-utils package installation to status before your > commit, while still keeping alsa-utils single recipe. Does a .bbappend with ALSA_UTILS_PKGS:remove = "alsa-utils-scripts" work for you? I would not want to exclude things that upstream installs from default choices in oe-core, if they logically belong to a set. GPLv3 is OSI approved, and Siemens not being able to ship it in a product is a special case. Alex -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#193577): https://lists.openembedded.org/g/openembedded-core/message/193577 Mute This Topic: https://lists.openembedded.org/mt/91575161/21656 Group Owner: openembedded-core+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [RFC][oe-core][PATCHv2] webp: switch recipe to cmake buildsystem
On Fri, 12 Jan 2024 at 11:26, Alexander Kanavin via lists.openembedded.org wrote: > > Aside from the fact that I can't send suggestions to upstream at google > > anyway, because I'm not willing to accept their requirement for a google > > account with a gmail address. Also: this is not correct. Google account can be created with a non-google email address (there is that option when creating it), and if so it's not any different from a github or whatever account. Alex -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#193576): https://lists.openembedded.org/g/openembedded-core/message/193576 Mute This Topic: https://lists.openembedded.org/mt/103623686/21656 Group Owner: openembedded-core+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [RFC][oe-core][PATCHv2] webp: switch recipe to cmake buildsystem
On Fri, 12 Jan 2024 at 09:56, Markus Volk wrote: > > Why is the patch marked inappropriate? > > > > Please send upstream first, and only after upstream has expressed > > their approval/disaporoval/request for changes we can take some form > > of it in oe-core. Maybe they would simply want to turn these into hard > > dependencies instead of adding even more options. > I guess upstream simply wants to leave everything as it is, because the > disadvantage of options would definitely be that everybody would have to > adapt their build systems. And that for a problem that is rather > yocto-specific. I'm not sure I wanted to suggest such a change upstream. > Aside from the fact that I can't send suggestions to upstream at google > anyway, because I'm not willing to accept their requirement for a google > account with a gmail address. In that case, please abstain from making patches that you are unable to send upstream. It's a basic expectation in oe-core. Also please don't make assumptions about what upstream wants; floating dependencies is a real problem that they might be interested in solving, and if not, we would still have a reference to their position. It comes across as you trying to find reasons to avoid interacting with them. For this specific recipe, you can still rewrite it to use cmake, without adding patches, if you follow my suggestion elsewhere (drop the existing packageconfigs, and make a new set that strictly follows the options that upstream offers right now, pulling in all the optional libraries for each of them). It doesn't solve the floating dependencies problem completely (it could silently regress to that on version updates), but it would avoid it at this moment. Alex -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#193575): https://lists.openembedded.org/g/openembedded-core/message/193575 Mute This Topic: https://lists.openembedded.org/mt/103623686/21656 Group Owner: openembedded-core+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [OE-core] [PATCH v5 2/3] mesa: add support for RustiCL under PACKAGECONFIG "opencl"
2024. 01. 02. 23:57 keltezéssel, Dmitry Baryshkov írta:
On Tue, 2 Jan 2024 at 17:30, Richard Purdie
wrote:
On Tue, 2024-01-02 at 17:25 +0200, Dmitry Baryshkov wrote:
On Tue, 2 Jan 2024 at 13:40, Richard Purdie
wrote:
On Tue, 2024-01-02 at 12:32 +0200, Dmitry Baryshkov wrote:
On Sat, 30 Dec 2023 at 23:02, Richard Purdie
wrote:
On Sat, 2023-12-30 at 21:14 +0200, Dmitry Baryshkov wrote:
From: Zoltán Böszörményi
Mesa now has two OpenCL implementations: Clover (older one) and RustiCL
(newer one). Extend the "opencl" package config to include both
implementations into the libopencl-mesa (-dev) packages. As the Clover
implementation is in maintenance mode and is expected to be removed at
some point, it doesn't make sense to split the packages or to provide
two config options.
Signed-off-by: Zoltán Böszörményi
[DB: reworked rust class inheritance]
[DB: merged "rusticl" and "clover" back to "opencl"]
Signed-off-by: Dmitry Baryshkov
---
meta/recipes-graphics/mesa/mesa.inc | 21 -
1 file changed, 16 insertions(+), 5 deletions(-)
diff --git a/meta/recipes-graphics/mesa/mesa.inc
b/meta/recipes-graphics/mesa/mesa.inc
index efbce5caade5..a192562a7935 100644
--- a/meta/recipes-graphics/mesa/mesa.inc
+++ b/meta/recipes-graphics/mesa/mesa.inc
@@ -48,6 +48,12 @@ PROVIDES = " \
inherit meson pkgconfig python3native gettext features_check
+# We can not include rust conditionally, otherwise it can not be overriden by
+# the bbappends. Instead always include it, but remove the dependencies by
+# default.
+inherit rust
+INHIBIT_DEFAULT_RUST_DEPS = "${@bb.utils.contains('PACKAGECONFIG', 'opencl', '',
'1', d)}"
+
I'd really prefer not to have the rust inherit unconditionally. I have
some ideas about how we could fix this in bitbake so we may need to
wait for that if we can't avoid the PACKAGECONFIG issue from appends...
Well, it's not just appends. Extending PACKAGECONFIG from local.conf
also doesn't work. In fact even adding "opencl" to the default
PACKAGECONFIG doesn't work unless I move `inherit
@{bb.utils.contains` after the PACKAGECONFIG definition. Using
`INHIBIT_DEFAULT_RUST_DEPS` works around that issue, but yes, it has a
drawback of including rust unconditionally.
PACKAGECONFIG:append:pn-mesa = " opencl"
Yes, I tried that. I got an error about rust tools not being defined.
I think the :append: and overrides are handled after the inherit
argument is handled. So it is either PACKAGECONFIG being fully defined
before inherit (literally before), or inherit ends up with an empty
argument.
The append and overrides would be expanded at the time the expression
in the inherit statement is evaluated so if that didn't work there is
something else at play...
I did not attempt debugging the bitbake itself. However I can see that
with this line in local.conf, PACKAGECONFIG is updated, but the RUSTC
is not defined (which means that rust.bbclass was not inherited). Then
I tried adding opencl to the default PACKAGECONFIG in mesa.inc. If
inherit comes before PACKAGECONFIG being set, it is not honoured. If
inherit comes after PACKAGECONFIG, it works as expected (and RUSTC
gets defined).
Something like this should work:
inherit ${@bb.utils.contains('PACKAGECONFIG', 'opencl', 'rust', '', d)}
E.g. the libxml2 recipe does it, although for a different bbclass.
Best regards,
Zoltán Böszörményi
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193574):
https://lists.openembedded.org/g/openembedded-core/message/193574
Mute This Topic: https://lists.openembedded.org/mt/103436170/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [OE-core] [PATCH 13/21] alsa-utils-scripts: merge into alsa-utils
Hi Alex,
I'm upgrading my layer from kirkstone to scarthgap and observed that my image
failed due to gpl3 license.
I want to conveniently install whole alsa-utils except for parts where license
forbids me to do it.
After your path I would need to list all alsa-utils subpackages except the
scripts.
And on top of that during every core layer update I'd have to check if there is
not a new one.
So from this point of view, your statement
"There is no reason to keep the recipe separate: bash dependency is not a
problem until the alsa-utils-scripts package is explicitly installed into a
target image."
Is not true, it's a problem for maintenance.
Would something like this be accepted when I'd submit it?
This would return alsa-utils package installation to status before your commit,
while still keeping alsa-utils single recipe.
diff --git a/meta/recipes-multimedia/alsa/alsa-utils_1.2.10.bb
b/meta/recipes-multimedia/alsa/alsa-utils_1.2.10.bb
index 4e5ed8dfa0..5f4048938f 100644
--- a/meta/recipes-multimedia/alsa/alsa-utils_1.2.10.bb
+++ b/meta/recipes-multimedia/alsa/alsa-utils_1.2.10.bb
@@ -56,10 +56,10 @@ ALSA_UTILS_PKGS = "\
alsa-utils-alsactl \
alsa-utils-alsaloop \
alsa-utils-alsaucm \
- alsa-utils-scripts \
alsa-utils-nhltdmicinfo \
"
+PACKAGES += "alsa-utils-scripts"
PACKAGES += "${ALSA_UTILS_PKGS}"
RDEPENDS:${PN} += "${ALSA_UTILS_PKGS}"
Thanks,
Peter
-Original Message-
From: openembedded-core@lists.openembedded.org
On Behalf Of Alexander Kanavin via
lists.openembedded.org
Sent: Monday, June 6, 2022 14:01
To: openembedded-core@lists.openembedded.org
Cc: Kanavin, Alexander (EXT) (Linutronix GmbH)
Subject: [OE-core] [PATCH 13/21] alsa-utils-scripts: merge into alsa-utils
> There is no reason to keep the recipe separate: bash dependency is not a
> problem until the alsa-utils-scripts package is explicitly installed into a
> target image.
>
> Signed-off-by: Alexander Kanavin
> ---
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#193573):
https://lists.openembedded.org/g/openembedded-core/message/193573
Mute This Topic: https://lists.openembedded.org/mt/91575161/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
