base-passwd: add clock group
PATCH v3 1/2 base-passwd: add clock group
From: Chen Qi <[email protected]>
New systemd version (v258) introduces a new group, clock, to "enable applications like linuxptp to open clocks without root privileges".[1]
This results in warning at do_rootfs time:
WARNING: Group clock has never been defined
Add group clock with gid 81 to fix this issue. The wheel group's gid is 80, so 81 is chosen. And as the patch to add wheel group was not accepted by upstream, directly mark this one's upstream status to be 'Inappropriate'.
[1] https://github.com/systemd/systemd/commit/af96ccfc24bc4803078a46b4ef2cdeb5decdfbcd
--- .../base-passwd/0001-Add-clock-group.patch | 34 +++++++++++++++++++ .../base-passwd/base-passwd_3.6.7.bb | 1 + 2 files changed, 35 insertions(+) create mode 100644 meta/recipes-core/base-passwd/base-passwd/0001-Add-clock-group.patch
meta/recipes-core/base-passwd/base-passwd/0001-Add-clock-group.patch new file +34 -0
@@ -0,0 +1,34 @@ +From 4b3a8f347179d7e2a1062eeed13a6474deb28728 Mon Sep 17 00:00:00 2001 +From: Chen Qi <[email protected]> +Date: Thu, 16 Oct 2025 08:48:49 +0800 +Subject: [PATCH] Add clock group + +The clock group is needed to "to enable applications like linuxptp +to open clocks without root privileges". See the link below: +https://github.com/systemd/systemd/commit/af96ccfc24bc4803078a46b4ef2cdeb5decdfbcd + +Note that this group is needed as long as people use udev from +systemd because 50-udev-default.rules needs it. See the above +commit. + +Upstream-Status: Inappropriate [enable feature] + +Signed-off-by: Chen Qi <[email protected]> +--- + group.master | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/group.master b/group.master +index e54fd1d..6f84539 100644 +--- a/group.master ++++ b/group.master +@@ -39,5 +39,6 @@ staff:*:50: + games:*:60: + shutdown:*:70: + wheel:*:80: ++clock:*:81: + users:*:100: + nogroup:*:65534: +-- +2.34.1 +
meta/recipes-core/base-passwd/base-passwd_3.6.7.bb +1 -0
@@ -13,6 +13,7 @@ SRC_URI = "https://launchpad.net/debian/+archive/primary/+files/${BPN}_${PV}.tar file://0005-Add-kvm-group.patch \ file://0007-Add-wheel-group.patch \ file://0001-base-passwd-Add-the-sgx-group.patch \ + file://0001-Add-clock-group.patch \ " SRC_URI[sha256sum] = "cf869870fed7862b57bfa9e99cd5cd6f365e2349705a1b65af7fc182629bd1ab"
Signature · 1 line
2.34.1
PATCH v3 2/2 systemd: upgrade from 257.8 to 258
From: Chen Qi <[email protected]>
258 Release Changes:
https://github.com/systemd/systemd/releases/tag/v258
The 'runlevel', 'initctl', 'telinit' related contents are removed because they are offically removed in v258.
Disable tests explicitly. This is because we're not installing any tests. There's no point building them out. And this puts extra burden on making things compile for musl.
gcrypt and gnutls PACKAGECONFIGs are removed as "OpenSSL is now the only supported cryptography backend for systemd-resolved and systemd-importd."
musl changes:
1. This version has big changes in musl patches. The goal is to help maintenance. I want patches to be more easily applied to new versions. The count of files modified by musl specific patches decreased from 113 (v257) to 34 (v258).
2. All patches are re-evaluated.
This version of systemd introduces big changes in header files and other places. So the previous musl patches are re-evaluated one by one. Those that are dropped are considered not needed. A few new ones are added. There are some notable ones. 0012-do-not-disable-buffer-in-writing-files.patch is dropped because there is not runtime error. 0016-Fix-the-segfault-for-glob-related-codes-and-define-d.patch is added because this version of systemd uses strv_free to free gl_pathv instead of relying on globfree provided by libc.
---
...ve_257.8.bb => systemd-boot-native_258.bb} | 0
...temd-boot_257.8.bb => systemd-boot_258.bb} | 0
...7.8.bb => systemd-systemctl-native_258.bb} | 0
meta/recipes-core/systemd/systemd.inc | 4 +-
...tall-dependency-links-at-install-tim.patch | 10 +-
...implment-systemd-sysv-install-for-OE.patch | 13 +-
...> 0003-Do-not-create-var-log-README.patch} | 9 +-
...3-missing_type.h-add-comparison_fn_t.patch | 61 --
...ntroduce-header-file-and-add-__THROW.patch | 61 ++
...-parse_printf_format-implementation.patch} | 79 +--
...OB_BRACE-and-GLOB_ALTDIRFUNC-is-not-.patch | 156 ------
...info-related-contents-glibc-specific.patch | 117 ++++
...006-add-missing-FTW_-macros-for-musl.patch | 44 --
...07-Use-uintmax_t-for-handling-rlim_t.patch | 106 ----
...override-sys-prctl.h-to-avoid-redefi.patch | 43 ++
...patible-basename-for-non-glibc-syste.patch | 34 --
...ompliant-strerror_r-from-GNU-specif.patch} | 20 +-
...uffering-when-writing-to-oom_score_a.patch | 41 --
...til-Make-STRERROR-portable-for-musl.patch} | 17 +-
...-util.h-define-RLIM_FMT-to-fit-musl-.patch | 40 ++
...definition-of-prctl_mm_map-structure.patch | 32 --
...ride-malloc.h-define-dummy-malloc_tr.patch | 41 ++
...-not-disable-buffer-in-writing-files.patch | 530 ------------------
...tion.c-avoid-using-glibc-ConditionVe.patch | 47 ++
.../0013-Handle-__cpu_mask-usage.patch | 60 --
...c-avoid-boot-time-segfault-for-musl.patch} | 8 +-
...014-Handle-missing-gshadow-for-musl.patch} | 77 +--
.../0015-Avoid-sequence-point-error.patch | 35 ++
...l.h-Define-MIPS-ABI-defines-for-musl.patch | 42 --
...-for-glob-related-codes-and-define-d.patch | 409 ++++++++++++++
...ass-correct-parameters-to-getdents64.patch | 37 --
.../0017-Adjust-for-musl-headers.patch | 526 -----------------
...ays-include-netinet-if_ether.h-first.patch | 40 ++
...trerror-is-assumed-to-be-GNU-specifi.patch | 52 --
...ake-malloc_trim-conditional-on-glibc.patch | 39 --
...hared-Do-not-use-malloc_info-on-musl.patch | 57 --
...22-avoid-missing-LOCK_EX-declaration.patch | 56 --
...gnal.h-to-avoid-the-undeclared-error.patch | 27 -
...references-using-stdin-as-a-struct-m.patch | 48 --
...clusion-order-to-avoid-redeclaration.patch | 288 ----------
.../{systemd_257.8.bb => systemd_258.bb} | 67 +--
41 files changed, 966 insertions(+), 2407 deletions(-)
rename meta/recipes-core/systemd/{systemd-boot-native_257.8.bb => systemd-boot-native_258.bb} (100%)
rename meta/recipes-core/systemd/{systemd-boot_257.8.bb => systemd-boot_258.bb} (100%)
rename meta/recipes-core/systemd/{systemd-systemctl-native_257.8.bb => systemd-systemctl-native_258.bb} (100%)
rename meta/recipes-core/systemd/systemd/{0001-Do-not-create-var-log-README.patch => 0003-Do-not-create-var-log-README.patch} (84%)
delete mode 100644 meta/recipes-core/systemd/systemd/0003-missing_type.h-add-comparison_fn_t.patch
create mode 100644 meta/recipes-core/systemd/systemd/0004-musl.h-introduce-header-file-and-add-__THROW.patch
rename meta/recipes-core/systemd/systemd/{0004-add-fallback-parse_printf_format-implementation.patch => 0005-add-fallback-parse_printf_format-implementation.patch} (88%)
delete mode 100644 meta/recipes-core/systemd/systemd/0005-don-t-fail-if-GLOB_BRACE-and-GLOB_ALTDIRFUNC-is-not-.patch
create mode 100644 meta/recipes-core/systemd/systemd/0006-Make-mallinfo-related-contents-glibc-specific.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0006-add-missing-FTW_-macros-for-musl.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0007-Use-uintmax_t-for-handling-rlim_t.patch
create mode 100644 meta/recipes-core/systemd/systemd/0007-add-src-include-override-sys-prctl.h-to-avoid-redefi.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0008-Define-glibc-compatible-basename-for-non-glibc-syste.patch
rename meta/recipes-core/systemd/systemd/{0010-distinguish-XSI-compliant-strerror_r-from-GNU-specif.patch => 0008-distinguish-XSI-compliant-strerror_r-from-GNU-specif.patch} (77%)
delete mode 100644 meta/recipes-core/systemd/systemd/0009-Do-not-disable-buffering-when-writing-to-oom_score_a.patch
rename meta/recipes-core/systemd/systemd/{0019-errno-util-Make-STRERROR-portable-for-musl.patch => 0009-errno-util-Make-STRERROR-portable-for-musl.patch} (72%)
create mode 100644 meta/recipes-core/systemd/systemd/0010-src-basic-format-util.h-define-RLIM_FMT-to-fit-musl-.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0011-avoid-redefinition-of-prctl_mm_map-structure.patch
create mode 100644 meta/recipes-core/systemd/systemd/0011-src-include-override-malloc.h-define-dummy-malloc_tr.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0012-do-not-disable-buffer-in-writing-files.patch
create mode 100644 meta/recipes-core/systemd/systemd/0012-src-shared-condition.c-avoid-using-glibc-ConditionVe.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0013-Handle-__cpu_mask-usage.patch
rename meta/recipes-core/systemd/systemd/{0026-build-path.c-avoid-boot-time-segfault-for-musl.patch => 0013-build-path.c-avoid-boot-time-segfault-for-musl.patch} (77%)
rename meta/recipes-core/systemd/systemd/{0014-Handle-missing-gshadow.patch => 0014-Handle-missing-gshadow-for-musl.patch} (71%)
create mode 100644 meta/recipes-core/systemd/systemd/0015-Avoid-sequence-point-error.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0015-missing_syscall.h-Define-MIPS-ABI-defines-for-musl.patch
create mode 100644 meta/recipes-core/systemd/systemd/0016-Fix-the-segfault-for-glob-related-codes-and-define-d.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0016-pass-correct-parameters-to-getdents64.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0017-Adjust-for-musl-headers.patch
create mode 100644 meta/recipes-core/systemd/systemd/0017-Always-include-netinet-if_ether.h-first.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0018-test-bus-error-strerror-is-assumed-to-be-GNU-specifi.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0020-sd-event-Make-malloc_trim-conditional-on-glibc.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0021-shared-Do-not-use-malloc_info-on-musl.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0022-avoid-missing-LOCK_EX-declaration.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0023-include-signal.h-to-avoid-the-undeclared-error.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0024-undef-stdin-for-references-using-stdin-as-a-struct-m.patch
delete mode 100644 meta/recipes-core/systemd/systemd/0025-adjust-header-inclusion-order-to-avoid-redeclaration.patch
rename meta/recipes-core/systemd/{systemd_257.8.bb => systemd_258.bb} (92%)meta/recipes-core/systemd/systemd-boot-native_258.bb renamed from meta/recipes-core/systemd/systemd-boot-native_257.8.bb +0 -0
meta/recipes-core/systemd/systemd-boot_258.bb renamed from meta/recipes-core/systemd/systemd-boot_257.8.bb +0 -0
meta/recipes-core/systemd/systemd-systemctl-native_258.bb renamed from meta/recipes-core/systemd/systemd-systemctl-native_257.8.bb +0 -0
meta/recipes-core/systemd/systemd.inc +2 -2
@@ -15,8 +15,8 @@ LICENSE:libsystemd = "LGPL-2.1-or-later" LIC_FILES_CHKSUM = "file://LICENSE.GPL2;md5=751419260aa954499f7abaabaa882bbe \ file://LICENSE.LGPL2.1;md5=4fbd65380cdd255951079008b364516c" -SRCREV = "5e38d199a623563698ab4a69acbbe3afa9135198" -SRCBRANCH = "v257-stable" +SRCREV = "781d9d0789379d1ea1f2ecefb804d41e9c8b6c38" +SRCBRANCH = "v258-stable" SRC_URI = "git://github.com/systemd/systemd.git;protocol=https;branch=${SRCBRANCH};tag=v${PV}" CVE_PRODUCT = "systemd"
meta/recipes-core/systemd/systemd/0001-binfmt-Don-t-install-dependency-links-at-install-tim.patch +5 -5
@@ -1,7 +1,7 @@ -From e5fd143f215f072404c544f694cb026a4231503e Mon Sep 17 00:00:00 2001 +From b1ab58d3b4a8164e5978409706d6769f5cfde404 Mon Sep 17 00:00:00 2001 From: Chen Qi <[email protected]> Date: Thu, 21 Feb 2019 16:23:24 +0800 -Subject: [PATCH 01/26] binfmt: Don't install dependency links at install time +Subject: [PATCH 01/16] binfmt: Don't install dependency links at install time for the binfmt services use [Install] blocks so that they get created when the service is enabled
@@ -25,10 +25,10 @@ Signed-off-by: Scott Murray <[email protected]> 3 files changed, 7 insertions(+), 2 deletions(-) diff --git a/units/meson.build b/units/meson.build -index 96f4852741..0a3a4fee67 100644 +index 4f47a3b2bd..237870025b 100644 --- a/units/meson.build +++ b/units/meson.build -@@ -156,7 +156,6 @@ units = [ +@@ -154,7 +154,6 @@ units = [ { 'file' : 'proc-sys-fs-binfmt_misc.automount', 'conditions' : ['ENABLE_BINFMT'],
@@ -36,7 +36,7 @@ index 96f4852741..0a3a4fee67 100644 }, { 'file' : 'proc-sys-fs-binfmt_misc.mount', -@@ -258,7 +257,6 @@ units = [ +@@ -263,7 +262,6 @@ units = [ { 'file' : 'systemd-binfmt.service.in', 'conditions' : ['ENABLE_BINFMT'],
meta/recipes-core/systemd/systemd/0002-implment-systemd-sysv-install-for-OE.patch +8 -5
@@ -1,7 +1,7 @@ -From 4a5602ede9881fd8e578a3c8bc40dd5df7c4d802 Mon Sep 17 00:00:00 2001 +From 08ab9f1eaa86ca2c6a0bb45219a0153495708acb Mon Sep 17 00:00:00 2001 From: Khem Raj <[email protected]> Date: Sat, 5 Sep 2015 06:31:47 +0000 -Subject: [PATCH] implement systemd-sysv-install for OE +Subject: [PATCH 02/16] implment systemd-sysv-install for OE Use update-rc.d for enabling/disabling and status command to check the status of the sysv service
@@ -14,7 +14,7 @@ Signed-off-by: Khem Raj <[email protected]> 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/src/systemctl/systemd-sysv-install.SKELETON b/src/systemctl/systemd-sysv-install.SKELETON -index cb58d8243b..eff3f5f579 100755 +index cb58d8243b..000bdf6165 100755 --- a/src/systemctl/systemd-sysv-install.SKELETON +++ b/src/systemctl/systemd-sysv-install.SKELETON @@ -34,17 +34,17 @@ case "$1" in
@@ -22,13 +22,13 @@ index cb58d8243b..eff3f5f579 100755 # call the command to enable SysV init script $NAME here # (consider optional $ROOT) - echo "IMPLEMENT ME: enabling SysV init.d script $NAME" -+ update-rc.d ${ROOT:+-r $ROOT} -f $NAME defaults ++ update-rc.d -f $NAME defaults ;; disable) # call the command to disable SysV init script $NAME here # (consider optional $ROOT) - echo "IMPLEMENT ME: disabling SysV init.d script $NAME" -+ update-rc.d ${ROOT:+-r $ROOT} -f $NAME remove ++ update-rc.d -f $NAME remove ;; is-enabled) # exit with 0 if $NAME is enabled, non-zero if it is disabled
@@ -38,3 +38,6 @@ index cb58d8243b..eff3f5f579 100755 ;; *) usage ;; +-- +2.34.1 +
meta/recipes-core/systemd/systemd/0003-Do-not-create-var-log-README.patch renamed from meta/recipes-core/systemd/systemd/0001-Do-not-create-var-log-README.patch +6 -3
@@ -1,7 +1,7 @@ -From 425ad51e727058b48dd4580fd6afe7e51e96a28a Mon Sep 17 00:00:00 2001 +From a7f6a296707642d05463aec22ea3dfce7d06c989 Mon Sep 17 00:00:00 2001 From: Peter Kjellerstedt <[email protected]> Date: Tue, 21 Jan 2025 05:02:00 +0100 -Subject: [PATCH] Do not create /var/log/README +Subject: [PATCH 03/16] Do not create /var/log/README /var/log/README is a link to /usr/share/doc/systemd/README.logs. The latter is packaged in systemd-doc and likely not installed, which leaves
@@ -15,7 +15,7 @@ Signed-off-by: Peter Kjellerstedt <[email protected]> 1 file changed, 3 deletions(-) diff --git a/tmpfiles.d/legacy.conf.in b/tmpfiles.d/legacy.conf.in -index b475500e58..650c91a8da 100644 +index cdef21fa9b..03798c953e 100644 --- a/tmpfiles.d/legacy.conf.in +++ b/tmpfiles.d/legacy.conf.in @@ -13,9 +13,6 @@
@@ -28,3 +28,6 @@ index b475500e58..650c91a8da 100644 {% if HAVE_SYSV_COMPAT %} # /run/lock/subsys is used for serializing SysV service execution, and +-- +2.34.1 +
meta/recipes-core/systemd/systemd/0003-missing_type.h-add-comparison_fn_t.patch deleted +0 -61
61 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0004-musl.h-introduce-header-file-and-add-__THROW.patch new file +61 -0
61 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0005-add-fallback-parse_printf_format-implementation.patch renamed from meta/recipes-core/systemd/systemd/0004-add-fallback-parse_printf_format-implementation.patch +28 -51
138 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0005-don-t-fail-if-GLOB_BRACE-and-GLOB_ALTDIRFUNC-is-not-.patch deleted +0 -156
156 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0006-Make-mallinfo-related-contents-glibc-specific.patch new file +117 -0
117 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0006-add-missing-FTW_-macros-for-musl.patch deleted +0 -44
44 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0007-Use-uintmax_t-for-handling-rlim_t.patch deleted +0 -106
106 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0007-add-src-include-override-sys-prctl.h-to-avoid-redefi.patch new file +43 -0
43 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0008-Define-glibc-compatible-basename-for-non-glibc-syste.patch deleted +0 -34
@@ -1,34 +0,0 @@ -From f4cd939c7cc1ce0a59bab2693768f2c95d9ced00 Mon Sep 17 00:00:00 2001 -From: Khem Raj <[email protected]> -Date: Sun, 27 May 2018 08:36:44 -0700 -Subject: [PATCH 08/26] Define glibc compatible basename() for non-glibc - systems - -Fixes builds with musl, even though systemd is adamant about -using non-posix basename implementation, we have a way out - -Upstream-Status: Inappropriate [musl specific] - -Signed-off-by: Khem Raj <[email protected]> ---- - src/basic/string-util.h | 4 ++++ - 1 file changed, 4 insertions(+) - -diff --git a/src/basic/string-util.h b/src/basic/string-util.h -index cc6aa183c0..0b035125cd 100644 ---- a/src/basic/string-util.h -+++ b/src/basic/string-util.h -@@ -27,6 +27,10 @@ - #define URI_UNRESERVED ALPHANUMERICAL "-._~" /* [RFC3986] */ - #define URI_VALID URI_RESERVED URI_UNRESERVED /* [RFC3986] */ - -+#if !defined(__GLIBC__) -+#define basename(src) (strrchr(src,'/') ? strrchr(src,'/')+1 : src) -+#endif -+ - static inline char* strstr_ptr(const char *haystack, const char *needle) { - if (!haystack || !needle) - return NULL; --- -2.34.1 -
meta/recipes-core/systemd/systemd/0008-distinguish-XSI-compliant-strerror_r-from-GNU-specif.patch renamed from meta/recipes-core/systemd/systemd/0010-distinguish-XSI-compliant-strerror_r-from-GNU-specif.patch +10 -10
@@ -1,10 +1,10 @@ -From b7f6c245b4ae72999f23eecc2bbb6d6fb8db667c Mon Sep 17 00:00:00 2001 +From bbe0096367618026fe5ebb81c266fd50084f113a Mon Sep 17 00:00:00 2001 From: Chen Qi <[email protected]> Date: Tue, 10 Jul 2018 15:40:17 +0800 -Subject: [PATCH 10/26] distinguish XSI-compliant strerror_r from GNU-specifi +Subject: [PATCH 08/16] distinguish XSI-compliant strerror_r from GNU-specific strerror_r -XSI-compliant strerror_r and GNU-specifi strerror_r are different. +XSI-compliant strerror_r and GNU-specific strerror_r are different. int strerror_r(int errnum, char *buf, size_t buflen); /* XSI-compliant */
@@ -24,10 +24,10 @@ Signed-off-by: Chen Qi <[email protected]> 2 files changed, 15 insertions(+), 1 deletion(-) diff --git a/src/libsystemd/sd-bus/bus-error.c b/src/libsystemd/sd-bus/bus-error.c -index 58c24d25c0..69a0d09d42 100644 +index bc0164c1b4..e69ffe4f75 100644 --- a/src/libsystemd/sd-bus/bus-error.c +++ b/src/libsystemd/sd-bus/bus-error.c -@@ -405,7 +405,12 @@ static void bus_error_strerror(sd_bus_error *e, int error) { +@@ -400,7 +400,12 @@ static void bus_error_strerror(sd_bus_error *e, int error) { return; errno = 0;
@@ -40,25 +40,25 @@ index 58c24d25c0..69a0d09d42 100644 if (errno == ERANGE || strlen(x) >= k - 1) { free(m); k *= 2; -@@ -590,8 +595,12 @@ const char* _bus_error_message(const sd_bus_error *e, int error, char buf[static +@@ -585,8 +590,12 @@ const char* _bus_error_message(const sd_bus_error *e, int error, char buf[static if (e && e->message) return e->message; - +#ifndef __GLIBC__ -+ strerror_r(abs(error), buf, ERRNO_BUF_LEN); ++ strerror_r(ABS(error), buf, ERRNO_BUF_LEN); + return buf; +#else - return strerror_r(abs(error), buf, ERRNO_BUF_LEN); + return strerror_r(ABS(error), buf, ERRNO_BUF_LEN); +#endif } static bool map_ok(const sd_bus_error_map *map) { diff --git a/src/libsystemd/sd-journal/journal-send.c b/src/libsystemd/sd-journal/journal-send.c -index 75e8e08add..41e5c7c2b8 100644 +index e838e7e0d7..28b6e014c0 100644 --- a/src/libsystemd/sd-journal/journal-send.c +++ b/src/libsystemd/sd-journal/journal-send.c -@@ -361,7 +361,12 @@ static int fill_iovec_perror_and_send(const char *message, int skip, struct iove +@@ -338,7 +338,12 @@ static int fill_iovec_perror_and_send(const char *message, int skip, struct iove char* j; errno = 0;
meta/recipes-core/systemd/systemd/0009-Do-not-disable-buffering-when-writing-to-oom_score_a.patch deleted +0 -41
41 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0009-errno-util-Make-STRERROR-portable-for-musl.patch renamed from meta/recipes-core/systemd/systemd/0019-errno-util-Make-STRERROR-portable-for-musl.patch +10 -7
@@ -1,7 +1,7 @@ -From 28fa1d5f56c6ddee9e336e6f2051c55e9f2f98b4 Mon Sep 17 00:00:00 2001 +From 7ce68a17deeda43c31a5de5956bad4c92481f1ca Mon Sep 17 00:00:00 2001 From: Khem Raj <[email protected]> Date: Mon, 23 Jan 2023 23:39:46 -0800 -Subject: [PATCH 19/26] errno-util: Make STRERROR portable for musl +Subject: [PATCH 09/16] errno-util: Make STRERROR portable for musl Sadly, systemd has decided to use yet another GNU extention in a macro lets make this such that we can use XSI compliant strerror_r() for
@@ -10,27 +10,30 @@ non-glibc hosts Upstream-Status: Inappropriate [musl specific] Signed-off-by: Khem Raj <[email protected]> + +Rebased for v258 +Signed-off-by: Chen Qi <[email protected]> --- src/basic/errno-util.h | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/src/basic/errno-util.h b/src/basic/errno-util.h -index 48b76e4bf7..6e7653e2d9 100644 +index 5fe337cfb1..dee34167cb 100644 --- a/src/basic/errno-util.h +++ b/src/basic/errno-util.h -@@ -15,8 +15,16 @@ +@@ -13,8 +13,16 @@ * https://stackoverflow.com/questions/34880638/compound-literal-lifetime-and-if-blocks * * Note that we use the GNU variant of strerror_r() here. */ --#define STRERROR(errnum) strerror_r(abs(errnum), (char[ERRNO_BUF_LEN]){}, ERRNO_BUF_LEN) +-#define STRERROR(errnum) strerror_r(ABS(errnum), (char[ERRNO_BUF_LEN]){}, ERRNO_BUF_LEN) +static inline const char * STRERROR(int errnum); +static inline const char * STRERROR(int errnum) { +#ifdef __GLIBC__ -+ return strerror_r(abs(errnum), (char[ERRNO_BUF_LEN]){}, ERRNO_BUF_LEN); ++ return strerror_r(ABS(errnum), (char[ERRNO_BUF_LEN]){}, ERRNO_BUF_LEN); +#else + static __thread char buf[ERRNO_BUF_LEN]; -+ return strerror_r(abs(errnum), buf, ERRNO_BUF_LEN) ? "unknown error" : buf; ++ return strerror_r(ABS(errnum), buf, ERRNO_BUF_LEN) ? "unknown error" : buf; +#endif +} /* A helper to print an error message or message for functions that return 0 on EOF.
meta/recipes-core/systemd/systemd/0010-src-basic-format-util.h-define-RLIM_FMT-to-fit-musl-.patch new file +40 -0
@@ -0,0 +1,40 @@ +From adf7bd6cf6305f5454cfb1ad4f1755880e831f7e Mon Sep 17 00:00:00 2001 +From: Chen Qi <[email protected]> +Date: Mon, 29 Sep 2025 14:41:52 +0800 +Subject: [PATCH 10/16] src/basic/format-util.h: define RLIM_FMT to fit musl's + rlim_t definition + +musl always defines rlim_t to be "unsigned long long". + +Upstream-Status: Inappropriate [musl specific] + +Signed-off-by: Chen Qi <[email protected]> +--- + src/basic/format-util.h | 4 ++++ + 1 file changed, 4 insertions(+) + +diff --git a/src/basic/format-util.h b/src/basic/format-util.h +index e42f788ce6..9275f9615d 100644 +--- a/src/basic/format-util.h ++++ b/src/basic/format-util.h +@@ -39,6 +39,7 @@ assert_cc(sizeof(gid_t) == sizeof(uint32_t)); + # error Unknown timex member size + #endif + ++#ifdef __GLIBC__ + #if SIZEOF_RLIM_T == 8 + # define RLIM_FMT "%" PRIu64 + #elif SIZEOF_RLIM_T == 4 +@@ -46,6 +47,9 @@ assert_cc(sizeof(gid_t) == sizeof(uint32_t)); + #else + # error Unknown rlim_t size + #endif ++#else ++#define RLIM_FMT "%llu" ++#endif + + #if SIZEOF_DEV_T == 8 + # define DEV_FMT "%" PRIu64 +-- +2.34.1 +
meta/recipes-core/systemd/systemd/0011-avoid-redefinition-of-prctl_mm_map-structure.patch deleted +0 -32
@@ -1,32 +0,0 @@ -From 43b0269e850a2fbcb6ca615258aa8f8a9b4f6a9d Mon Sep 17 00:00:00 2001 -From: Chen Qi <[email protected]> -Date: Mon, 25 Feb 2019 15:44:54 +0800 -Subject: [PATCH 11/26] avoid redefinition of prctl_mm_map structure - -Fix the following compile failure: -error: redefinition of 'struct prctl_mm_map' - -Upstream-Status: Inappropriate [musl specific] - -Signed-off-by: Chen Qi <[email protected]> ---- - src/basic/missing_prctl.h | 2 ++ - 1 file changed, 2 insertions(+) - -diff --git a/src/basic/missing_prctl.h b/src/basic/missing_prctl.h -index 2c9f9f6c50..65a984b564 100644 ---- a/src/basic/missing_prctl.h -+++ b/src/basic/missing_prctl.h -@@ -1,7 +1,9 @@ - /* SPDX-License-Identifier: LGPL-2.1-or-later */ - #pragma once - -+#ifdef __GLIBC__ - #include <linux/prctl.h> -+#endif - - #include "macro.h" - --- -2.34.1 -
meta/recipes-core/systemd/systemd/0011-src-include-override-malloc.h-define-dummy-malloc_tr.patch new file +41 -0
41 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0012-do-not-disable-buffer-in-writing-files.patch deleted +0 -530
530 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0012-src-shared-condition.c-avoid-using-glibc-ConditionVe.patch new file +47 -0
47 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0013-Handle-__cpu_mask-usage.patch deleted +0 -60
60 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0013-build-path.c-avoid-boot-time-segfault-for-musl.patch renamed from meta/recipes-core/systemd/systemd/0026-build-path.c-avoid-boot-time-segfault-for-musl.patch +4 -4
@@ -1,7 +1,7 @@ -From f2a7cf1d2a2bc2516a180809efd85c828cd9c7f4 Mon Sep 17 00:00:00 2001 +From a4b921b8d408e0f1c9d035a3ec93c3da3fe42352 Mon Sep 17 00:00:00 2001 From: Chen Qi <[email protected]> Date: Wed, 3 Jul 2024 07:18:42 -0700 -Subject: [PATCH 26/26] build-path.c: avoid boot time segfault for musl +Subject: [PATCH 13/16] build-path.c: avoid boot time segfault for musl This function, at runtime, should return -ENOEXEC. For musl, it somehow segfaults. I think it's related to getauxval, but it's
@@ -15,10 +15,10 @@ Signed-off-by: Chen Qi <[email protected]> 1 file changed, 1 insertion(+) diff --git a/src/basic/build-path.c b/src/basic/build-path.c -index b5972658df..4ef551034e 100644 +index 577ff72bce..3a7876798c 100644 --- a/src/basic/build-path.c +++ b/src/basic/build-path.c -@@ -151,6 +151,7 @@ int get_build_exec_dir(char **ret) { +@@ -152,6 +152,7 @@ int get_build_exec_dir(char **ret) { */ static int runpath_cached = -ERRNO_MAX-1;
meta/recipes-core/systemd/systemd/0014-Handle-missing-gshadow-for-musl.patch renamed from meta/recipes-core/systemd/systemd/0014-Handle-missing-gshadow.patch +40 -37
162 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0015-Avoid-sequence-point-error.patch new file +35 -0
@@ -0,0 +1,35 @@ +From 5dbeaee2085d3edcf20983fce4208eb174e46422 Mon Sep 17 00:00:00 2001 +From: Chen Qi <[email protected]> +Date: Thu, 9 Oct 2025 15:19:09 +0800 +Subject: [PATCH 15/16] Avoid sequence-point error + +See https://www.openwall.com/lists/musl/2025/09/30/4 + +Upstream-Status: Inappropriate [musl specific] + +Signed-off-by: Chen Qi <[email protected]> +--- + src/shared/cpu-set-util.c | 6 ++++-- + 1 file changed, 4 insertions(+), 2 deletions(-) + +diff --git a/src/shared/cpu-set-util.c b/src/shared/cpu-set-util.c +index 7a155c12d4..a9c57fc31c 100644 +--- a/src/shared/cpu-set-util.c ++++ b/src/shared/cpu-set-util.c +@@ -97,9 +97,11 @@ char* cpu_set_to_mask_string(const CPUSet *c) { + for (size_t i = c->allocated * 8; i > 0; ) { + uint32_t m = 0; + +- for (int j = (i % 32 ?: 32) - 1; j >= 0; j--) +- if (CPU_ISSET_S(--i, c->allocated, c->set)) ++ for (int j = (i % 32 ?: 32) - 1; j >= 0; j--) { ++ --i; ++ if (CPU_ISSET_S(i, c->allocated, c->set)) + SET_BIT(m, j); ++ } + + if (!found_nonzero) { + if (m == 0) +-- +2.34.1 +
meta/recipes-core/systemd/systemd/0015-missing_syscall.h-Define-MIPS-ABI-defines-for-musl.patch deleted +0 -42
42 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0016-Fix-the-segfault-for-glob-related-codes-and-define-d.patch new file +409 -0
409 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0016-pass-correct-parameters-to-getdents64.patch deleted +0 -37
@@ -1,37 +0,0 @@ -From 9079b158779a9c395c24f882f72a1c734795045d Mon Sep 17 00:00:00 2001 -From: Khem Raj <[email protected]> -Date: Fri, 21 Jan 2022 15:15:11 -0800 -Subject: [PATCH 16/26] pass correct parameters to getdents64 - -Fixes -../git/src/basic/recurse-dir.c:57:40: error: incompatible pointer types passing 'uint8_t *' (aka 'unsigned char *') to parameter of type 'struct dirent *' [-Werror,-Wincompatible-pointer-types] - n = getdents64(dir_fd, (uint8_t*) de->buffer + de->buffer_size, bs - de->buffer_size); - ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - -../git/src/basic/stat-util.c:102:28: error: incompatible pointer types passing 'union (unnamed union at ../git/src/basic/stat-util.c:78:9) *' to parameter of type 'struct dirent *' [-Werror,-Wincompatible-pointer-types] - n = getdents64(fd, &buffer, sizeof(buffer)); - ^~~~~~~ - -Upstream-Status: Inappropriate [musl specific] -Signed-off-by: Khem Raj <[email protected]> -Signed-off-by: Jiaqing Zhao <[email protected]> ---- - src/basic/recurse-dir.c | 2 +- - 1 file changed, 1 insertion(+), 1 deletion(-) - -diff --git a/src/basic/recurse-dir.c b/src/basic/recurse-dir.c -index 378fd92b06..5b567b457d 100644 ---- a/src/basic/recurse-dir.c -+++ b/src/basic/recurse-dir.c -@@ -56,7 +56,7 @@ int readdir_all(int dir_fd, - bs = MIN(MALLOC_SIZEOF_SAFE(de) - offsetof(DirectoryEntries, buffer), (size_t) SSIZE_MAX); - assert(bs > de->buffer_size); - -- n = getdents64(dir_fd, (uint8_t*) de->buffer + de->buffer_size, bs - de->buffer_size); -+ n = getdents64(dir_fd, (struct dirent*)((uint8_t*) de->buffer + de->buffer_size), bs - de->buffer_size); - if (n < 0) - return -errno; - if (n == 0) --- -2.34.1 -
meta/recipes-core/systemd/systemd/0017-Adjust-for-musl-headers.patch deleted +0 -526
526 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0017-Always-include-netinet-if_ether.h-first.patch new file +40 -0
@@ -0,0 +1,40 @@ +From 675460e529f7373d042db581da3fdc3c9c25bf63 Mon Sep 17 00:00:00 2001 +From: Chen Qi <[email protected]> +Date: Wed, 15 Oct 2025 15:48:09 +0800 +Subject: [PATCH 17/17] Always include netinet/if_ether.h first + +When compiling systemd with musl, we get many ethhdr redinition +errors. This is because netinet/if_ether.h in musl does not check +__UAPI_DEF_ETHHDR and expects programs to include it before other +kernel headers. +See https://www.openwall.com/lists/musl/2025/10/14/2 + +To avoid this redefinition error, and to avoid resolving conflicts +every time we upgrade systemd, use '-include netinet/if_ether.h' to +make sure this file header is included first. + +The major consideration behind such method is maintenance. + +Upstream-Status: Inappropriate [musl specific]
Lines of this file are not shown. View the raw message.
meta/recipes-core/systemd/systemd/0018-test-bus-error-strerror-is-assumed-to-be-GNU-specifi.patch deleted +0 -52
52 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0020-sd-event-Make-malloc_trim-conditional-on-glibc.patch deleted +0 -39
Lines of this file are not shown. View the raw message.
meta/recipes-core/systemd/systemd/0021-shared-Do-not-use-malloc_info-on-musl.patch deleted +0 -57
57 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0022-avoid-missing-LOCK_EX-declaration.patch deleted +0 -56
56 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0023-include-signal.h-to-avoid-the-undeclared-error.patch deleted +0 -27
Lines of this file are not shown. View the raw message.
meta/recipes-core/systemd/systemd/0024-undef-stdin-for-references-using-stdin-as-a-struct-m.patch deleted +0 -48
48 lines not on this page · raw message
meta/recipes-core/systemd/systemd/0025-adjust-header-inclusion-order-to-avoid-redeclaration.patch deleted +0 -288
288 lines not on this page · raw message
meta/recipes-core/systemd/systemd_258.bb renamed from meta/recipes-core/systemd/systemd_257.8.bb +20 -47
67 lines not on this page · raw message
The rest of this diff is not shown. View the raw message.
Signature · 1 line
2.34.1
Part of the diff is not shown here. View the raw message.
Re: [OE-core][PATCH V3 2/2] systemd: upgrade from 257.8 to 258
Quoted 31 lines · Chen Qi via lists.openembedded.org
From: Chen Qi <[email protected]>
258 Release Changes:
https://github.com/systemd/systemd/releases/tag/v258The 'runlevel', 'initctl', 'telinit' related contents are removed because they are offically removed in v258.
Disable tests explicitly. This is because we're not installing any tests. There's no point building them out. And this puts extra burden on making things compile for musl.
gcrypt and gnutls PACKAGECONFIGs are removed as "OpenSSL is now the only supported cryptography backend for systemd-resolved and systemd-importd."
musl changes:
1. This version has big changes in musl patches. The goal is to help maintenance. I want patches to be more easily applied to new versions. The count of files modified by musl specific patches decreased from 113 (v257) to 34 (v258).
2. All patches are re-evaluated.
This version of systemd introduces big changes in header files and other places. So the previous musl patches are re-evaluated one by one. Those that are dropped are considered not needed. A few new ones are added. There are some notable ones. 0012-do-not-disable-buffer-in-writing-files.patch is dropped because there is not runtime error. 0016-Fix-the-segfault-for-glob-related-codes-and-define-d.patch is added because this version of systemd usesstrv_freeto freegl_pathvinstead of relying on globfree provided by libc.
Quoted 2 lines
Signed-off-by Chen Qi <[email protected]>---
Hi Chen,
Thanks for the new version. Still, we get some errors with post install scripts:
ERROR: core-image-sato-1.0-r0 do_rootfs: Postinstall scriptlets of ['bluez5', 'ofono', 'connman', 'xserver-nodm-init', 'avahi-daemon', 'rpcbind', 'run-postinsts'] have failed. If the intention is to defer them to first boot, then please place them into pkg_postinst_ontarget:${PN} ().
Deferring to first boot via 'exit 1' is no longer supported.
Details of the failure are in /srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.
ERROR: Logfile of failure stored in: /srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.2882514
NOTE: recipe core-image-sato-1.0-r0: task do_rootfs: Failed
ERROR: Task (/srv/pokybuild/yocto-worker/qa-extras2/build/layers/openembedded-core/meta/recipes-sato/images/core-image-sato.bb:do_rootfs) failed with exit code '1'
https://autobuilder.yoctoproject.org/valkyrie/#/builders/26/builds/2590
Can you have a look at these, please?
Thanks,
Mathieu
Signature · 3 lines
Mathieu Dubois-Briand, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
Re: [OE-core][PATCH V3 2/2] systemd: upgrade from 257.8 to 258
Quoted 48 lines · Mathieu Dubois-Briand
Quoted 31 lines · Chen Qi via lists.openembedded.org
From: Chen Qi <[email protected]>
258 Release Changes:
https://github.com/systemd/systemd/releases/tag/v258The 'runlevel', 'initctl', 'telinit' related contents are removed because they are offically removed in v258.
Disable tests explicitly. This is because we're not installing any tests. There's no point building them out. And this puts extra burden on making things compile for musl.
gcrypt and gnutls PACKAGECONFIGs are removed as "OpenSSL is now the only supported cryptography backend for systemd-resolved and systemd-importd."
musl changes:
1. This version has big changes in musl patches. The goal is to help maintenance. I want patches to be more easily applied to new versions. The count of files modified by musl specific patches decreased from 113 (v257) to 34 (v258).
2. All patches are re-evaluated.
This version of systemd introduces big changes in header files and other places. So the previous musl patches are re-evaluated one by one. Those that are dropped are considered not needed. A few new ones are added. There are some notable ones. 0012-do-not-disable-buffer-in-writing-files.patch is dropped because there is not runtime error. 0016-Fix-the-segfault-for-glob-related-codes-and-define-d.patch is added because this version of systemd usesstrv_freeto freegl_pathvinstead of relying on globfree provided by libc.Quoted 2 lines
Signed-off-by Chen Qi <[email protected]>---
Hi Chen,
Thanks for the new version. Still, we get some errors with post install scripts:
ERROR: core-image-sato-1.0-r0
do_rootfs: Postinstall scriptlets of ['bluez5', 'ofono', 'connman', 'xserver-nodm-init', 'avahi-daemon', 'rpcbind', 'run-postinsts'] have failed. If the intention is to defer them to first boot, then please place them into pkg_postinst_ontarget:${PN} ().
Deferring to first boot via 'exit 1' is no longer supported.
Details of the failure are in/srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.
ERROR: Logfile of failure stored in:/srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.2882514
NOTE: recipe core-image-sato-1.0-r0: taskdo_rootfs: Failed
ERROR: Task (/srv/pokybuild/yocto-worker/qa-extras2/build/layers/openembedded-core/meta/recipes-sato/images/core-image-sato.bb:do_rootfs) failed with exit code '1'
Quoted 6 lines
https://autobuilder.yoctoproject.org/valkyrie/#/builders/26/builds/2590
Can you have a look at these, please?
Thanks,
Mathieu
Sorry, another issue that I've just identified, during image tests:
Traceback (most recent call last):
File "/srv/pokybuild/yocto-worker/qemuarm-alt/build/layers/openembedded-core/meta/lib/oeqa/core/decorator/__init__.py", line 35, in wrapped_f
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/srv/pokybuild/yocto-worker/qemuarm-alt/build/layers/openembedded-core/meta/lib/oeqa/runtime/cases/parselogs.py", line 185, in test_parselogs
self.assertEqual(errcount, 0, msg=self.msg)
AssertionError: 1 != 0 : Log: /srv/pokybuild/yocto-worker/qemuarm-alt/build/build/tmp/work/qemuarm-poky-linux-gnueabi/core-image-sato/1.0/target_logs/Xorg.0.log
-----------------------
Central error: [ 27.402] failed to find screen to remove
***********************
[ 26.897] (**) Option "Device" "/dev/input/event1" [ 26.955] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 26.956] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard [ 26.959] (II) event1 - QEMU QEMU USB Keyboard: device removed [ 26.991] (**) Option "config_info" "udev:/sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:02.0/usb1/1-2/1-2:1.0/0003:0627:0001.0002/input/input1/event1" [ 26.991] (II) XINPUT: Adding extended input device "QEMU QEMU USB Keyboard" (type: KEYBOARD, id 7) [ 27.049] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 27.051] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard [ 27.401] (II) config/udev: removing GPU device /sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:03.0/drm/card0 /dev/dri/card0 [ 27.402] xf86: remove device 0 /sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:03.0/drm/card0 [ 27.402] failed to find screen to remove [ 27.428] (II) config/udev: removing device QEMU QEMU USB Keyboard [ 27.431] (II) event1 - QEMU QEMU USB Keyboard: device removed [ 27.472] (II) UnloadModule: "libinput" [ 27.508] (II) config/udev: Adding input device QEMU QEMU USB Keyboard (/dev/input/event1) [ 27.509] (**) QEMU QEMU USB Keyboard: Applying InputClass "libinput keyboard catchall" [ 27.509] (II) Using input driver 'libinput' for 'QEMU QEMU USB Keyboard' [ 27.509] (**) QEMU QEMU USB Keyboard: always reports core events [ 27.509] (**) Option "Device" "/dev/input/event1" [ 27.567] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 27.568] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard
***********************
1 errors found in logs.
https://autobuilder.yoctoproject.org/valkyrie/#/builders/74/builds/2588
Thanks,
Mathieu
Signature · 3 lines
Mathieu Dubois-Briand, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
Re: [OE-core][PATCH V3 2/2] systemd: upgrade from 257.8 to 258
Quoted 88 lines · Mathieu Dubois-Briand via lists.openembedded.org
Quoted 47 lines · Mathieu Dubois-Briand
Quoted 31 lines · Chen Qi via lists.openembedded.org
From: Chen Qi <[email protected]>
258 Release Changes:
https://github.com/systemd/systemd/releases/tag/v258The 'runlevel', 'initctl', 'telinit' related contents are removed because they are offically removed in v258.
Disable tests explicitly. This is because we're not installing any tests. There's no point building them out. And this puts extra burden on making things compile for musl.
gcrypt and gnutls PACKAGECONFIGs are removed as "OpenSSL is now the only supported cryptography backend for systemd-resolved and systemd-importd."
musl changes:
1. This version has big changes in musl patches. The goal is to help maintenance. I want patches to be more easily applied to new versions. The count of files modified by musl specific patches decreased from 113 (v257) to 34 (v258).
2. All patches are re-evaluated.
This version of systemd introduces big changes in header files and other places. So the previous musl patches are re-evaluated one by one. Those that are dropped are considered not needed. A few new ones are added. There are some notable ones. 0012-do-not-disable-buffer-in-writing-files.patch is dropped because there is not runtime error. 0016-Fix-the-segfault-for-glob-related-codes-and-define-d.patch is added because this version of systemd usesstrv_freeto freegl_pathvinstead of relying on globfree provided by libc.Quoted 2 lines
Signed-off-by Chen Qi <[email protected]>---
Hi Chen,
Thanks for the new version. Still, we get some errors with post install scripts:
ERROR: core-image-sato-1.0-r0
do_rootfs: Postinstall scriptlets of ['bluez5', 'ofono', 'connman', 'xserver-nodm-init', 'avahi-daemon', 'rpcbind', 'run-postinsts'] have failed. If the intention is to defer them to first boot, then please place them into pkg_postinst_ontarget:${PN} ().
Deferring to first boot via 'exit 1' is no longer supported.
Details of the failure are in/srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.
ERROR: Logfile of failure stored in:/srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.2882514
NOTE: recipe core-image-sato-1.0-r0: taskdo_rootfs: Failed
ERROR: Task (/srv/pokybuild/yocto-worker/qa-extras2/build/layers/openembedded-core/meta/recipes-sato/images/core-image-sato.bb:do_rootfs) failed with exit code '1'Quoted 6 lines
https://autobuilder.yoctoproject.org/valkyrie/#/builders/26/builds/2590
Can you have a look at these, please?
Thanks,
MathieuSorry, another issue that I've just identified, during image tests:
Traceback (most recent call last):
File "/srv/pokybuild/yocto-worker/qemuarm-alt/build/layers/openembedded-core/meta/lib/oeqa/core/decorator/__init__.py", line 35, inwrapped_f
returnfunc(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/srv/pokybuild/yocto-worker/qemuarm-alt/build/layers/openembedded-core/meta/lib/oeqa/runtime/cases/parselogs.py", line 185, intest_parselogs
self.assertEqual(errcount, 0, msg=self.msg)
AssertionError: 1 != 0 : Log:/srv/pokybuild/yocto-worker/qemuarm-alt/build/build/tmp/work/qemuarm-poky-linux-gnueabi/core-image-sato/1.0/target_logs/Xorg.0.log
-----------------------
Central error: [ 27.402] failed to find screen to remove
***********************[ 26.897] (**) Option "Device" "/dev/input/event1" [ 26.955] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 26.956] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard [ 26.959] (II) event1 - QEMU QEMU USB Keyboard: device removed [ 26.991] (**) Option "config_info" "udev:/sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:02.0/usb1/1-2/1-2:1.0/0003:0627:0001.0002/input/input1/event1" [ 26.991] (II) XINPUT: Adding extended input device "QEMU QEMU USB Keyboard" (type: KEYBOARD, id 7) [ 27.049] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 27.051] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard [ 27.401] (II) config/udev: removing GPU device /sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:03.0/drm/card0 /dev/dri/card0 [ 27.402] xf86: remove device 0 /sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:03.0/drm/card0 [ 27.402] failed to find screen to remove [ 27.428] (II) config/udev: removing device QEMU QEMU USB Keyboard [ 27.431] (II) event1 - QEMU QEMU USB Keyboard: device removed [ 27.472] (II) UnloadModule: "libinput" [ 27.508] (II) config/udev: Adding input device QEMU QEMU USB Keyboard (/dev/input/event1) [ 27.509] (**) QEMU QEMU USB Keyboard: Applying InputClass "libinput keyboard catchall" [ 27.509] (II) Using input driver 'libinput' for 'QEMU QEMU USB Keyboard' [ 27.509] (**) QEMU QEMU USB Keyboard: always reports core events [ 27.509] (**) Option "Device" "/dev/input/event1" [ 27.567] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 27.568] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard
Quoted 7 lines
***********************
1 errors found in logs.https://autobuilder.yoctoproject.org/valkyrie/#/builders/74/builds/2588
Thanks,
Mathieu
Got it. I'll fix these two errors in V4.
Regards,
Qi
Quoted 9 lines
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#225075): https://lists.openembedded.org/g/openembedded-core/message/225075
Mute This Topic: https://lists.openembedded.org/mt/115785954/3618072
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [OE-core][PATCH V3 2/2] systemd: upgrade from 257.8 to 258
In reply to Mathieu Dubois-BriandQuoted 88 lines · Mathieu Dubois-Briand via lists.openembedded.org
Quoted 47 lines · Mathieu Dubois-Briand
Quoted 31 lines · Chen Qi via lists.openembedded.org
From: Chen Qi <[email protected]>
258 Release Changes:
https://github.com/systemd/systemd/releases/tag/v258The 'runlevel', 'initctl', 'telinit' related contents are removed because they are offically removed in v258.
Disable tests explicitly. This is because we're not installing any tests. There's no point building them out. And this puts extra burden on making things compile for musl.
gcrypt and gnutls PACKAGECONFIGs are removed as "OpenSSL is now the only supported cryptography backend for systemd-resolved and systemd-importd."
musl changes:
1. This version has big changes in musl patches. The goal is to help maintenance. I want patches to be more easily applied to new versions. The count of files modified by musl specific patches decreased from 113 (v257) to 34 (v258).
2. All patches are re-evaluated.
This version of systemd introduces big changes in header files and other places. So the previous musl patches are re-evaluated one by one. Those that are dropped are considered not needed. A few new ones are added. There are some notable ones. 0012-do-not-disable-buffer-in-writing-files.patch is dropped because there is not runtime error. 0016-Fix-the-segfault-for-glob-related-codes-and-define-d.patch is added because this version of systemd usesstrv_freeto freegl_pathvinstead of relying on globfree provided by libc.Quoted 2 lines
Signed-off-by Chen Qi <[email protected]>---
Hi Chen,
Thanks for the new version. Still, we get some errors with post install scripts:
ERROR: core-image-sato-1.0-r0
do_rootfs: Postinstall scriptlets of ['bluez5', 'ofono', 'connman', 'xserver-nodm-init', 'avahi-daemon', 'rpcbind', 'run-postinsts'] have failed. If the intention is to defer them to first boot, then please place them into pkg_postinst_ontarget:${PN} ().
Deferring to first boot via 'exit 1' is no longer supported.
Details of the failure are in/srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.
ERROR: Logfile of failure stored in:/srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.2882514
NOTE: recipe core-image-sato-1.0-r0: taskdo_rootfs: Failed
ERROR: Task (/srv/pokybuild/yocto-worker/qa-extras2/build/layers/openembedded-core/meta/recipes-sato/images/core-image-sato.bb:do_rootfs) failed with exit code '1'Quoted 6 lines
https://autobuilder.yoctoproject.org/valkyrie/#/builders/26/builds/2590
Can you have a look at these, please?
Thanks,
MathieuSorry, another issue that I've just identified, during image tests:
Traceback (most recent call last):
File "/srv/pokybuild/yocto-worker/qemuarm-alt/build/layers/openembedded-core/meta/lib/oeqa/core/decorator/__init__.py", line 35, inwrapped_f
returnfunc(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/srv/pokybuild/yocto-worker/qemuarm-alt/build/layers/openembedded-core/meta/lib/oeqa/runtime/cases/parselogs.py", line 185, intest_parselogs
self.assertEqual(errcount, 0, msg=self.msg)
AssertionError: 1 != 0 : Log:/srv/pokybuild/yocto-worker/qemuarm-alt/build/build/tmp/work/qemuarm-poky-linux-gnueabi/core-image-sato/1.0/target_logs/Xorg.0.log
-----------------------
Central error: [ 27.402] failed to find screen to remove
***********************[ 26.897] (**) Option "Device" "/dev/input/event1" [ 26.955] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 26.956] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard [ 26.959] (II) event1 - QEMU QEMU USB Keyboard: device removed [ 26.991] (**) Option "config_info" "udev:/sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:02.0/usb1/1-2/1-2:1.0/0003:0627:0001.0002/input/input1/event1" [ 26.991] (II) XINPUT: Adding extended input device "QEMU QEMU USB Keyboard" (type: KEYBOARD, id 7) [ 27.049] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 27.051] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard [ 27.401] (II) config/udev: removing GPU device /sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:03.0/drm/card0 /dev/dri/card0 [ 27.402] xf86: remove device 0 /sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:03.0/drm/card0 [ 27.402] failed to find screen to remove [ 27.428] (II) config/udev: removing device QEMU QEMU USB Keyboard [ 27.431] (II) event1 - QEMU QEMU USB Keyboard: device removed [ 27.472] (II) UnloadModule: "libinput" [ 27.508] (II) config/udev: Adding input device QEMU QEMU USB Keyboard (/dev/input/event1) [ 27.509] (**) QEMU QEMU USB Keyboard: Applying InputClass "libinput keyboard catchall" [ 27.509] (II) Using input driver 'libinput' for 'QEMU QEMU USB Keyboard' [ 27.509] (**) QEMU QEMU USB Keyboard: always reports core events [ 27.509] (**) Option "Device" "/dev/input/event1" [ 27.567] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 27.568] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard
Quoted 7 lines
***********************
1 errors found in logs.https://autobuilder.yoctoproject.org/valkyrie/#/builders/74/builds/2588
Thanks,
Mathieu
Hi Mathieu,
I reproduced the postinstall error you mentioned in another email.
But I cannot reproduce this one. I used 'poky-altfcfg + qemuarm' as in the above link. Is there any other config I need to use? I referred to the 'Write config' step.
Regards,
Qi
Quoted 9 lines
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#225075): https://lists.openembedded.org/g/openembedded-core/message/225075
Mute This Topic: https://lists.openembedded.org/mt/115785954/3618072
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-
Re: [OE-core][PATCH V3 2/2] systemd: upgrade from 257.8 to 258
Quoted 99 lines · Chen Qi via lists.openembedded.org
Quoted 88 lines · Mathieu Dubois-Briand via lists.openembedded.org
Quoted 47 lines · Mathieu Dubois-Briand
Quoted 31 lines · Chen Qi via lists.openembedded.org
From: Chen Qi <[email protected]>
258 Release Changes:
https://github.com/systemd/systemd/releases/tag/v258The 'runlevel', 'initctl', 'telinit' related contents are removed because they are offically removed in v258.
Disable tests explicitly. This is because we're not installing any tests. There's no point building them out. And this puts extra burden on making things compile for musl.
gcrypt and gnutls PACKAGECONFIGs are removed as "OpenSSL is now the only supported cryptography backend for systemd-resolved and systemd-importd."
musl changes:
1. This version has big changes in musl patches. The goal is to help maintenance. I want patches to be more easily applied to new versions. The count of files modified by musl specific patches decreased from 113 (v257) to 34 (v258).
2. All patches are re-evaluated.
This version of systemd introduces big changes in header files and other places. So the previous musl patches are re-evaluated one by one. Those that are dropped are considered not needed. A few new ones are added. There are some notable ones. 0012-do-not-disable-buffer-in-writing-files.patch is dropped because there is not runtime error. 0016-Fix-the-segfault-for-glob-related-codes-and-define-d.patch is added because this version of systemd usesstrv_freeto freegl_pathvinstead of relying on globfree provided by libc.Quoted 2 lines
Signed-off-by Chen Qi <[email protected]>---
Hi Chen,
Thanks for the new version. Still, we get some errors with post install scripts:
ERROR: core-image-sato-1.0-r0
do_rootfs: Postinstall scriptlets of ['bluez5', 'ofono', 'connman', 'xserver-nodm-init', 'avahi-daemon', 'rpcbind', 'run-postinsts'] have failed. If the intention is to defer them to first boot, then please place them into pkg_postinst_ontarget:${PN} ().
Deferring to first boot via 'exit 1' is no longer supported.
Details of the failure are in/srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.
ERROR: Logfile of failure stored in:/srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.2882514
NOTE: recipe core-image-sato-1.0-r0: taskdo_rootfs: Failed
ERROR: Task (/srv/pokybuild/yocto-worker/qa-extras2/build/layers/openembedded-core/meta/recipes-sato/images/core-image-sato.bb:do_rootfs) failed with exit code '1'Quoted 6 lines
https://autobuilder.yoctoproject.org/valkyrie/#/builders/26/builds/2590
Can you have a look at these, please?
Thanks,
MathieuSorry, another issue that I've just identified, during image tests:
Traceback (most recent call last):
File "/srv/pokybuild/yocto-worker/qemuarm-alt/build/layers/openembedded-core/meta/lib/oeqa/core/decorator/__init__.py", line 35, inwrapped_f
returnfunc(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/srv/pokybuild/yocto-worker/qemuarm-alt/build/layers/openembedded-core/meta/lib/oeqa/runtime/cases/parselogs.py", line 185, intest_parselogs
self.assertEqual(errcount, 0, msg=self.msg)
AssertionError: 1 != 0 : Log:/srv/pokybuild/yocto-worker/qemuarm-alt/build/build/tmp/work/qemuarm-poky-linux-gnueabi/core-image-sato/1.0/target_logs/Xorg.0.log
-----------------------
Central error: [ 27.402] failed to find screen to remove
***********************[ 26.897] (**) Option "Device" "/dev/input/event1" [ 26.955] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 26.956] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard [ 26.959] (II) event1 - QEMU QEMU USB Keyboard: device removed [ 26.991] (**) Option "config_info" "udev:/sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:02.0/usb1/1-2/1-2:1.0/0003:0627:0001.0002/input/input1/event1" [ 26.991] (II) XINPUT: Adding extended input device "QEMU QEMU USB Keyboard" (type: KEYBOARD, id 7) [ 27.049] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 27.051] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard [ 27.401] (II) config/udev: removing GPU device /sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:03.0/drm/card0 /dev/dri/card0 [ 27.402] xf86: remove device 0 /sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:03.0/drm/card0 [ 27.402] failed to find screen to remove [ 27.428] (II) config/udev: removing device QEMU QEMU USB Keyboard [ 27.431] (II) event1 - QEMU QEMU USB Keyboard: device removed [ 27.472] (II) UnloadModule: "libinput" [ 27.508] (II) config/udev: Adding input device QEMU QEMU USB Keyboard (/dev/input/event1) [ 27.509] (**) QEMU QEMU USB Keyboard: Applying InputClass "libinput keyboard catchall" [ 27.509] (II) Using input driver 'libinput' for 'QEMU QEMU USB Keyboard' [ 27.509] (**) QEMU QEMU USB Keyboard: always reports core events [ 27.509] (**) Option "Device" "/dev/input/event1" [ 27.567] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 27.568] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboardQuoted 7 lines
***********************
1 errors found in logs.https://autobuilder.yoctoproject.org/valkyrie/#/builders/74/builds/2588
Thanks,
MathieuHi Mathieu,
Quoted 12 lines
I reproduced the postinstall error you mentioned in another email.
But I cannot reproduce this one. I used 'poky-altfcfg + qemuarm' as in the above link. Is there any other config I need to use? I referred to the 'Write config' step.
Regards,
QiQuoted 3 lines
Yes, that should be all you need. In the write config step there is a bit more than that, but it should not impact the test result.
I will try to play a bit with it locally, to see if I manage to reproduce this.
Thanks,
Mathieu
Signature · 3 lines
Mathieu Dubois-Briand, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
Re: [OE-core][PATCH V3 2/2] systemd: upgrade from 257.8 to 258
In reply to ChenQiQuoted 99 lines · Chen Qi via lists.openembedded.org
Quoted 88 lines · Mathieu Dubois-Briand via lists.openembedded.org
Quoted 47 lines · Mathieu Dubois-Briand
Quoted 31 lines · Chen Qi via lists.openembedded.org
From: Chen Qi <[email protected]>
258 Release Changes:
https://github.com/systemd/systemd/releases/tag/v258The 'runlevel', 'initctl', 'telinit' related contents are removed because they are offically removed in v258.
Disable tests explicitly. This is because we're not installing any tests. There's no point building them out. And this puts extra burden on making things compile for musl.
gcrypt and gnutls PACKAGECONFIGs are removed as "OpenSSL is now the only supported cryptography backend for systemd-resolved and systemd-importd."
musl changes:
1. This version has big changes in musl patches. The goal is to help maintenance. I want patches to be more easily applied to new versions. The count of files modified by musl specific patches decreased from 113 (v257) to 34 (v258).
2. All patches are re-evaluated.
This version of systemd introduces big changes in header files and other places. So the previous musl patches are re-evaluated one by one. Those that are dropped are considered not needed. A few new ones are added. There are some notable ones. 0012-do-not-disable-buffer-in-writing-files.patch is dropped because there is not runtime error. 0016-Fix-the-segfault-for-glob-related-codes-and-define-d.patch is added because this version of systemd usesstrv_freeto freegl_pathvinstead of relying on globfree provided by libc.Quoted 2 lines
Signed-off-by Chen Qi <[email protected]>---
Hi Chen,
Thanks for the new version. Still, we get some errors with post install scripts:
ERROR: core-image-sato-1.0-r0
do_rootfs: Postinstall scriptlets of ['bluez5', 'ofono', 'connman', 'xserver-nodm-init', 'avahi-daemon', 'rpcbind', 'run-postinsts'] have failed. If the intention is to defer them to first boot, then please place them into pkg_postinst_ontarget:${PN} ().
Deferring to first boot via 'exit 1' is no longer supported.
Details of the failure are in/srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.
ERROR: Logfile of failure stored in:/srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.2882514
NOTE: recipe core-image-sato-1.0-r0: taskdo_rootfs: Failed
ERROR: Task (/srv/pokybuild/yocto-worker/qa-extras2/build/layers/openembedded-core/meta/recipes-sato/images/core-image-sato.bb:do_rootfs) failed with exit code '1'Quoted 6 lines
https://autobuilder.yoctoproject.org/valkyrie/#/builders/26/builds/2590
Can you have a look at these, please?
Thanks,
MathieuSorry, another issue that I've just identified, during image tests:
Traceback (most recent call last):
File "/srv/pokybuild/yocto-worker/qemuarm-alt/build/layers/openembedded-core/meta/lib/oeqa/core/decorator/__init__.py", line 35, inwrapped_f
returnfunc(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/srv/pokybuild/yocto-worker/qemuarm-alt/build/layers/openembedded-core/meta/lib/oeqa/runtime/cases/parselogs.py", line 185, intest_parselogs
self.assertEqual(errcount, 0, msg=self.msg)
AssertionError: 1 != 0 : Log:/srv/pokybuild/yocto-worker/qemuarm-alt/build/build/tmp/work/qemuarm-poky-linux-gnueabi/core-image-sato/1.0/target_logs/Xorg.0.log
-----------------------
Central error: [ 27.402] failed to find screen to remove
***********************[ 26.897] (**) Option "Device" "/dev/input/event1" [ 26.955] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 26.956] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard [ 26.959] (II) event1 - QEMU QEMU USB Keyboard: device removed [ 26.991] (**) Option "config_info" "udev:/sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:02.0/usb1/1-2/1-2:1.0/0003:0627:0001.0002/input/input1/event1" [ 26.991] (II) XINPUT: Adding extended input device "QEMU QEMU USB Keyboard" (type: KEYBOARD, id 7) [ 27.049] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 27.051] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard [ 27.401] (II) config/udev: removing GPU device /sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:03.0/drm/card0 /dev/dri/card0 [ 27.402] xf86: remove device 0 /sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:03.0/drm/card0 [ 27.402] failed to find screen to remove [ 27.428] (II) config/udev: removing device QEMU QEMU USB Keyboard [ 27.431] (II) event1 - QEMU QEMU USB Keyboard: device removed [ 27.472] (II) UnloadModule: "libinput" [ 27.508] (II) config/udev: Adding input device QEMU QEMU USB Keyboard (/dev/input/event1) [ 27.509] (**) QEMU QEMU USB Keyboard: Applying InputClass "libinput keyboard catchall" [ 27.509] (II) Using input driver 'libinput' for 'QEMU QEMU USB Keyboard' [ 27.509] (**) QEMU QEMU USB Keyboard: always reports core events [ 27.509] (**) Option "Device" "/dev/input/event1" [ 27.567] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 27.568] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboardQuoted 7 lines
***********************
1 errors found in logs.https://autobuilder.yoctoproject.org/valkyrie/#/builders/74/builds/2588
Thanks,
MathieuHi Mathieu,
Quoted 12 lines
I reproduced the postinstall error you mentioned in another email.
But I cannot reproduce this one. I used 'poky-altfcfg + qemuarm' as in the above link. Is there any other config I need to use? I referred to the 'Write config' step.
Regards,
QiQuoted 3 lines
Hi Qi,
I've been able to reproduce the test failure with this branch:
https://git.openembedded.org/openembedded-core-contrib/log/?h=mathieu/systemd-upgrade-tests
Following config in my local.conf:
---
SDKMACHINE = "x86_64"OE_FRAGMENTS += "machine/qemuarm distro/poky-altcfg" EXTRA_IMAGE_FEATURES ?= 'allow-empty-password empty-root-password allow-root-login' IMAGE_CLASSES += "testimage"
---
And then just:
bitbake core-image-sato && bitbake core-image-sato:do_testimage
Hope this can help you!
Thanks,
Mathieu
Signature · 3 lines
Mathieu Dubois-Briand, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
Re: [OE-core][PATCH V3 2/2] systemd: upgrade from 257.8 to 258
Quoted 124 lines · Mathieu Dubois-Briand
Quoted 98 lines · Chen Qi via lists.openembedded.org
Quoted 88 lines · Mathieu Dubois-Briand via lists.openembedded.org
Quoted 47 lines · Mathieu Dubois-Briand
Quoted 31 lines · Chen Qi via lists.openembedded.org
From: Chen Qi <[email protected]>
258 Release Changes:
https://github.com/systemd/systemd/releases/tag/v258The 'runlevel', 'initctl', 'telinit' related contents are removed because they are offically removed in v258.
Disable tests explicitly. This is because we're not installing any tests. There's no point building them out. And this puts extra burden on making things compile for musl.
gcrypt and gnutls PACKAGECONFIGs are removed as "OpenSSL is now the only supported cryptography backend for systemd-resolved and systemd-importd."
musl changes:
1. This version has big changes in musl patches. The goal is to help maintenance. I want patches to be more easily applied to new versions. The count of files modified by musl specific patches decreased from 113 (v257) to 34 (v258).
2. All patches are re-evaluated.
This version of systemd introduces big changes in header files and other places. So the previous musl patches are re-evaluated one by one. Those that are dropped are considered not needed. A few new ones are added. There are some notable ones. 0012-do-not-disable-buffer-in-writing-files.patch is dropped because there is not runtime error. 0016-Fix-the-segfault-for-glob-related-codes-and-define-d.patch is added because this version of systemd usesstrv_freeto freegl_pathvinstead of relying on globfree provided by libc.Quoted 2 lines
Signed-off-by Chen Qi <[email protected]>---
Hi Chen,
Thanks for the new version. Still, we get some errors with post install scripts:
ERROR: core-image-sato-1.0-r0
do_rootfs: Postinstall scriptlets of ['bluez5', 'ofono', 'connman', 'xserver-nodm-init', 'avahi-daemon', 'rpcbind', 'run-postinsts'] have failed. If the intention is to defer them to first boot, then please place them into pkg_postinst_ontarget:${PN} ().
Deferring to first boot via 'exit 1' is no longer supported.
Details of the failure are in/srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.
ERROR: Logfile of failure stored in:/srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.2882514
NOTE: recipe core-image-sato-1.0-r0: taskdo_rootfs: Failed
ERROR: Task (/srv/pokybuild/yocto-worker/qa-extras2/build/layers/openembedded-core/meta/recipes-sato/images/core-image-sato.bb:do_rootfs) failed with exit code '1'Quoted 6 lines
https://autobuilder.yoctoproject.org/valkyrie/#/builders/26/builds/2590
Can you have a look at these, please?
Thanks,
MathieuSorry, another issue that I've just identified, during image tests:
Traceback (most recent call last):
File "/srv/pokybuild/yocto-worker/qemuarm-alt/build/layers/openembedded-core/meta/lib/oeqa/core/decorator/__init__.py", line 35, inwrapped_f
returnfunc(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/srv/pokybuild/yocto-worker/qemuarm-alt/build/layers/openembedded-core/meta/lib/oeqa/runtime/cases/parselogs.py", line 185, intest_parselogs
self.assertEqual(errcount, 0, msg=self.msg)
AssertionError: 1 != 0 : Log:/srv/pokybuild/yocto-worker/qemuarm-alt/build/build/tmp/work/qemuarm-poky-linux-gnueabi/core-image-sato/1.0/target_logs/Xorg.0.log
-----------------------
Central error: [ 27.402] failed to find screen to remove
***********************[ 26.897] (**) Option "Device" "/dev/input/event1" [ 26.955] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 26.956] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard [ 26.959] (II) event1 - QEMU QEMU USB Keyboard: device removed [ 26.991] (**) Option "config_info" "udev:/sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:02.0/usb1/1-2/1-2:1.0/0003:0627:0001.0002/input/input1/event1" [ 26.991] (II) XINPUT: Adding extended input device "QEMU QEMU USB Keyboard" (type: KEYBOARD, id 7) [ 27.049] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 27.051] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard [ 27.401] (II) config/udev: removing GPU device /sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:03.0/drm/card0 /dev/dri/card0 [ 27.402] xf86: remove device 0 /sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:03.0/drm/card0 [ 27.402] failed to find screen to remove [ 27.428] (II) config/udev: removing device QEMU QEMU USB Keyboard [ 27.431] (II) event1 - QEMU QEMU USB Keyboard: device removed [ 27.472] (II) UnloadModule: "libinput" [ 27.508] (II) config/udev: Adding input device QEMU QEMU USB Keyboard (/dev/input/event1) [ 27.509] (**) QEMU QEMU USB Keyboard: Applying InputClass "libinput keyboard catchall" [ 27.509] (II) Using input driver 'libinput' for 'QEMU QEMU USB Keyboard' [ 27.509] (**) QEMU QEMU USB Keyboard: always reports core events [ 27.509] (**) Option "Device" "/dev/input/event1" [ 27.567] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 27.568] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboardQuoted 7 lines
***********************
1 errors found in logs.https://autobuilder.yoctoproject.org/valkyrie/#/builders/74/builds/2588
Thanks,
MathieuHi Mathieu,
Quoted 11 lines
I reproduced the postinstall error you mentioned in another email.
But I cannot reproduce this one. I used 'poky-altfcfg + qemuarm' as in the above link. Is there any other config I need to use? I referred to the 'Write config' step.
Regards,
QiQuoted 2 lines
Hi Qi,
I've been able to reproduce the test failure with this branch:
https://git.openembedded.org/openembedded-core-contrib/log/?h=mathieu/systemd-upgrade-testsFollowing config in my local.conf:
---
SDKMACHINE = "x86_64"OE_FRAGMENTS+= "machine/qemuarm distro/poky-altcfg"EXTRA_IMAGE_FEATURES?= 'allow-empty-password empty-root-password allow-root-login'IMAGE_CLASSES+= "testimage"
---
Quoted 8 lines
And then just:
bitbake core-image-sato && bitbake core-image-sato:do_testimageHope this can help you!
Thanks,
Mathieu
Got it! Thanks a lot!
Regards,
Qi
Re: [OE-core][PATCH V3 2/2] systemd: upgrade from 257.8 to 258
In reply to Mathieu Dubois-BriandQuoted 124 lines · Mathieu Dubois-Briand
Quoted 98 lines · Chen Qi via lists.openembedded.org
Quoted 88 lines · Mathieu Dubois-Briand via lists.openembedded.org
Quoted 47 lines · Mathieu Dubois-Briand
Quoted 31 lines · Chen Qi via lists.openembedded.org
From: Chen Qi <[email protected]>
258 Release Changes:
https://github.com/systemd/systemd/releases/tag/v258The 'runlevel', 'initctl', 'telinit' related contents are removed because they are offically removed in v258.
Disable tests explicitly. This is because we're not installing any tests. There's no point building them out. And this puts extra burden on making things compile for musl.
gcrypt and gnutls PACKAGECONFIGs are removed as "OpenSSL is now the only supported cryptography backend for systemd-resolved and systemd-importd."
musl changes:
1. This version has big changes in musl patches. The goal is to help maintenance. I want patches to be more easily applied to new versions. The count of files modified by musl specific patches decreased from 113 (v257) to 34 (v258).
2. All patches are re-evaluated.
This version of systemd introduces big changes in header files and other places. So the previous musl patches are re-evaluated one by one. Those that are dropped are considered not needed. A few new ones are added. There are some notable ones. 0012-do-not-disable-buffer-in-writing-files.patch is dropped because there is not runtime error. 0016-Fix-the-segfault-for-glob-related-codes-and-define-d.patch is added because this version of systemd usesstrv_freeto freegl_pathvinstead of relying on globfree provided by libc.Quoted 2 lines
Signed-off-by Chen Qi <[email protected]>---
Hi Chen,
Thanks for the new version. Still, we get some errors with post install scripts:
ERROR: core-image-sato-1.0-r0
do_rootfs: Postinstall scriptlets of ['bluez5', 'ofono', 'connman', 'xserver-nodm-init', 'avahi-daemon', 'rpcbind', 'run-postinsts'] have failed. If the intention is to defer them to first boot, then please place them into pkg_postinst_ontarget:${PN} ().
Deferring to first boot via 'exit 1' is no longer supported.
Details of the failure are in/srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.
ERROR: Logfile of failure stored in:/srv/pokybuild/yocto-worker/qa-extras2/build/build/tmp/work/qemux86_64-poky-linux/core-image-sato/1.0/temp/log.do_rootfs.2882514
NOTE: recipe core-image-sato-1.0-r0: taskdo_rootfs: Failed
ERROR: Task (/srv/pokybuild/yocto-worker/qa-extras2/build/layers/openembedded-core/meta/recipes-sato/images/core-image-sato.bb:do_rootfs) failed with exit code '1'Quoted 6 lines
https://autobuilder.yoctoproject.org/valkyrie/#/builders/26/builds/2590
Can you have a look at these, please?
Thanks,
MathieuSorry, another issue that I've just identified, during image tests:
Traceback (most recent call last):
File "/srv/pokybuild/yocto-worker/qemuarm-alt/build/layers/openembedded-core/meta/lib/oeqa/core/decorator/__init__.py", line 35, inwrapped_f
returnfunc(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/srv/pokybuild/yocto-worker/qemuarm-alt/build/layers/openembedded-core/meta/lib/oeqa/runtime/cases/parselogs.py", line 185, intest_parselogs
self.assertEqual(errcount, 0, msg=self.msg)
AssertionError: 1 != 0 : Log:/srv/pokybuild/yocto-worker/qemuarm-alt/build/build/tmp/work/qemuarm-poky-linux-gnueabi/core-image-sato/1.0/target_logs/Xorg.0.log
-----------------------
Central error: [ 27.402] failed to find screen to remove
***********************[ 26.897] (**) Option "Device" "/dev/input/event1" [ 26.955] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 26.956] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard [ 26.959] (II) event1 - QEMU QEMU USB Keyboard: device removed [ 26.991] (**) Option "config_info" "udev:/sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:02.0/usb1/1-2/1-2:1.0/0003:0627:0001.0002/input/input1/event1" [ 26.991] (II) XINPUT: Adding extended input device "QEMU QEMU USB Keyboard" (type: KEYBOARD, id 7) [ 27.049] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 27.051] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboard [ 27.401] (II) config/udev: removing GPU device /sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:03.0/drm/card0 /dev/dri/card0 [ 27.402] xf86: remove device 0 /sys/devices/platform/3f000000.pcie/pci0000:00/0000:00:03.0/drm/card0 [ 27.402] failed to find screen to remove [ 27.428] (II) config/udev: removing device QEMU QEMU USB Keyboard [ 27.431] (II) event1 - QEMU QEMU USB Keyboard: device removed [ 27.472] (II) UnloadModule: "libinput" [ 27.508] (II) config/udev: Adding input device QEMU QEMU USB Keyboard (/dev/input/event1) [ 27.509] (**) QEMU QEMU USB Keyboard: Applying InputClass "libinput keyboard catchall" [ 27.509] (II) Using input driver 'libinput' for 'QEMU QEMU USB Keyboard' [ 27.509] (**) QEMU QEMU USB Keyboard: always reports core events [ 27.509] (**) Option "Device" "/dev/input/event1" [ 27.567] (II) event1 - QEMU QEMU USB Keyboard: is tagged by udev as: Keyboard [ 27.568] (II) event1 - QEMU QEMU USB Keyboard: device is a keyboardQuoted 7 lines
***********************
1 errors found in logs.https://autobuilder.yoctoproject.org/valkyrie/#/builders/74/builds/2588
Thanks,
MathieuHi Mathieu,
Quoted 11 lines
I reproduced the postinstall error you mentioned in another email.
But I cannot reproduce this one. I used 'poky-altfcfg + qemuarm' as in the above link. Is there any other config I need to use? I referred to the 'Write config' step.
Regards,
QiQuoted 2 lines
Hi Qi,
I've been able to reproduce the test failure with this branch:
https://git.openembedded.org/openembedded-core-contrib/log/?h=mathieu/systemd-upgrade-testsFollowing config in my local.conf:
---
SDKMACHINE = "x86_64"OE_FRAGMENTS+= "machine/qemuarm distro/poky-altcfg"EXTRA_IMAGE_FEATURES?= 'allow-empty-password empty-root-password allow-root-login'IMAGE_CLASSES+= "testimage"
---
Quoted 8 lines
And then just:
bitbake core-image-sato && bitbake core-image-sato:do_testimageHope this can help you!
Thanks,
Mathieu
Hi Mathieu,
I finally reproduced the error! It has to use OE_FRAGMENTS to reproduce the issue. I'll fix this issue ASAP. Other issues have been fixed. Thanks a lot for your help.
Regards,
Qi
Quoted 55 lines
-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Chen Qi via lists.openembedded.org
Sent: den 16 oktober 2025 10:47
To: [email protected]
Subject: [OE-core][PATCH V3 1/2] base-passwd: add clock groupFrom: Chen Qi <[email protected]>
New systemd version (v258) introduces a new group, clock, to "enable applications like linuxptp to open clocks without root privileges".[1]
This results in warning at
do_rootfstime:
WARNING: Group clock has never been definedAdd group clock with gid 81 to fix this issue. The wheel group's gid is 80, so 81 is chosen. And as the patch to add wheel group was not accepted by upstream, directly mark this one's upstream status to be 'Inappropriate'.
[1] https://github.com/systemd/systemd/commit/af96ccfc24bc4803078a46b4ef2cdeb5decdfbcd
Signed-off-by Chen Qi <[email protected]>--- .../base-passwd/0001-Add-clock-group.patch | 34 +++++++++++++++++++ .../base-passwd/base-passwd_3.6.7.bb | 1 + 2 files changed, 35 insertions(+) create mode 100644 meta/recipes-core/base-passwd/base-passwd/0001-Add-clock-group.patchmeta/recipes-core/base-passwd/base-passwd/0001-Add-clock-group.patch new file +8
-0@@ -0,0 +1,34 @@ +From 4b3a8f347179d7e2a1062eeed13a6474deb28728 Mon Sep 17 00:00:00 2001 +From: Chen Qi <[email protected]> +Date: Thu, 16 Oct 2025 08:48:49 +0800 +Subject: [PATCH] Add clock group + +The clock group is needed to "to enable applications like linuxptp +to open clocks without root privileges". See the link below: +https://github.com/systemd/systemd/commit/af96ccfc24bc4803078a46b4ef2cdeb
5decdfbcdmeta/recipes-core/base-passwd/base-passwd/0001-Add-clock-group.patch new file +11
-0+ +Note that this group is needed as long as people use udev from +systemd because 50-udev-default.rules needs it. See the above +commit. + +Upstream-Status: Inappropriate [enable feature] + +Signed-off-by: Chen Qi <[email protected]> +--- + group.master | 1 + + 1 file changed, 1 insertion(+)
meta/recipes-core/base-passwd/base-passwd/0001-Add-clock-group.patch line 29
Quoted 10 lines
meta/recipes-core/base-passwd/base-passwd/0001-Add-clock-group.patch new file +10
-0+ +diff --git a/group.master b/group.master +index e54fd1d..6f84539 100644 +--- a/group.master ++++ b/group.master +@@ -39,5 +39,6 @@ staff:*:50: + games:*:60: + shutdown:*:70: + wheel:*:80: ++clock:*:81:
Why does this have to go here. Why can't you just update the GROUPADD_PARAM:${PN} in the systemd recipe to create the clock group:
GROUPADD_PARAM:${PN} = "-r systemd-journal; -r clock;"
That way it is only created if systemd is actually used.
Quoted 17 lines
+ users:*:100:
+ nogroup:*:65534:
+--
+2.34.1meta/recipes-core/base-passwd/base-passwd/0001-Add-clock-group.patch new file +1
-0+meta/recipes-core/base-passwd/base-passwd_3.6.7.bb +0
-0@@ -13,6 +13,7 @@ SRC_URI ="https://launchpad.net/debian/+archive/primary/+files/${BPN}_${PV}.tar
file://0005-Add-kvm-group.patch \ file://0007-Add-wheel-group.patch \ file://0001-base-passwd-Add-the-sgx-group.patch \ + file://0001-Add-clock-group.patch \ "
Quoted 3 lines
SRC_URI[sha256sum] = "
cf869870fed7862b57bfa9e99cd5cd6f365e2349705a1b65af7fc182629bd1ab"
--
2.34.1
//Peter
Quoted 68 lines · Peter Kjellerstedt
Quoted 55 lines
-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Chen Qi via lists.openembedded.org
Sent: den 16 oktober 2025 10:47
To: [email protected]
Subject: [OE-core][PATCH V3 1/2] base-passwd: add clock groupFrom: Chen Qi <[email protected]>
New systemd version (v258) introduces a new group, clock, to "enable applications like linuxptp to open clocks without root privileges".[1]
This results in warning at
do_rootfstime:
WARNING: Group clock has never been definedAdd group clock with gid 81 to fix this issue. The wheel group's gid is 80, so 81 is chosen. And as the patch to add wheel group was not accepted by upstream, directly mark this one's upstream status to be 'Inappropriate'.
[1] https://github.com/systemd/systemd/commit/af96ccfc24bc4803078a46b4ef2cdeb5decdfbcd
Signed-off-by Chen Qi <[email protected]>--- .../base-passwd/0001-Add-clock-group.patch | 34 +++++++++++++++++++ .../base-passwd/base-passwd_3.6.7.bb | 1 + 2 files changed, 35 insertions(+)create mode 100644
meta/recipes-core/base-passwd/base-passwd/0001-Add-clock-group.patchmeta/recipes-core/base-passwd/base-passwd/0001-Add-clock-group.patch new file +8
-0@@ -0,0 +1,34 @@ +From 4b3a8f347179d7e2a1062eeed13a6474deb28728 Mon Sep 17 00:00:00 2001 +From: Chen Qi <[email protected]> +Date: Thu, 16 Oct 2025 08:48:49 +0800 +Subject: [PATCH] Add clock group + +The clock group is needed to "to enable applications like linuxptp +to open clocks without root privileges". See the link below: +https://github.com/systemd/systemd/commit/af96ccfc24bc4803078a46b4ef2cdeb
5decdfbcdmeta/recipes-core/base-passwd/base-passwd/0001-Add-clock-group.patch new file +11
-0+ +Note that this group is needed as long as people use udev from +systemd because 50-udev-default.rules needs it. See the above +commit. + +Upstream-Status: Inappropriate [enable feature] + +Signed-off-by: Chen Qi <[email protected]> +--- + group.master | 1 + + 1 file changed, 1 insertion(+)meta/recipes-core/base-passwd/base-passwd/0001-Add-clock-group.patch line 29
Quoted 10 lines
meta/recipes-core/base-passwd/base-passwd/0001-Add-clock-group.patch new file +10
-0+ +diff --git a/group.master b/group.master +index e54fd1d..6f84539 100644 +--- a/group.master ++++ b/group.master +@@ -39,5 +39,6 @@ staff:*:50: + games:*:60: + shutdown:*:70: + wheel:*:80: ++clock:*:81:Why does this have to go here. Why can't you just update the GROUPADD_PARAM:${PN} in the systemd recipe to create the clock group:
Quoted 3 lines
GROUPADD_PARAM:${PN} = "-r systemd-journal; -r clock;"
That way it is only created if systemd is actually used.
I consider this clock similar to the 'wheel' and 'kvm' groups. 'wheel' and 'kvm' groups are added here, so I put 'clock' group here.
From my understanding, such groups are likely to be used by other applications (recipes). You can see this from the original commit: https://github.com/systemd/systemd/commit/af96ccfc24bc4803078a46b4ef2cdeb
It's possible that applications such as linuxptp will use this clock group. What systemd (udev) does is to change the device's group to 'clock'. This means applications using those devices could be put in 'clock' group to be granted access.
Regards,
Qi
Quoted 23 lines
Quoted 18 lines
+ users:*:100:
+ nogroup:*:65534:
+--
+2.34.1meta/recipes-core/base-passwd/base-passwd/0001-Add-clock-group.patch new file +1
-0+meta/recipes-core/base-passwd/base-passwd_3.6.7.bb +0
-0@@ -13,6 +13,7 @@ SRC_URI ="https://launchpad.net/debian/+archive/primary/+files/${BPN}_${PV}.tar
file://0005-Add-kvm-group.patch \ file://0007-Add-wheel-group.patch \ file://0001-base-passwd-Add-the-sgx-group.patch \ + file://0001-Add-clock-group.patch \ "Quoted 3 lines
SRC_URI[sha256sum] = "
cf869870fed7862b57bfa9e99cd5cd6f365e2349705a1b65af7fc182629bd1ab"
--
2.34.1//Peter
Quoted 10 lines · Chen Qi via lists.openembedded.org
Quoted 6 lines
Why does this have to go here. Why can't you just update the GROUPADD_PARAM:${PN} in the systemd recipe to create the clock group:
GROUPADD_PARAM:${PN} = "-r systemd-journal; -r clock;"
That way it is only created if systemd is actually used.
I consider this clock similar to the 'wheel' and 'kvm' groups. 'wheel' and 'kvm' groups are added here, so I put 'clock' group here.
Quoted 8 lines
From my understanding, such groups are likely to be used by other
applications (recipes). You can see this from the original commit:
https://github.com/systemd/systemd/commit/af96ccfc24bc4803078a46b4ef2cdebIt's possible that applications such as linuxptp will use this clock group. What systemd (udev) does is to change the device's group to 'clock'. This means applications using those devices could be put in 'clock' group to be granted access.
But how do classic binary distributions handle this? Can you research where e.g. Fedora or Debian create this group? We could probably follow that, and avoid having to carry an Inappropriate patch. Maybe base-passwd upstream already has a fix?
Alex
Quoted 19 lines · Alexander Kanavin
Quoted 9 lines · Chen Qi via lists.openembedded.org
Quoted 6 lines
Why does this have to go here. Why can't you just update the GROUPADD_PARAM:${PN} in the systemd recipe to create the clock group:
GROUPADD_PARAM:${PN} = "-r systemd-journal; -r clock;"
That way it is only created if systemd is actually used.
I consider this clock similar to the 'wheel' and 'kvm' groups. 'wheel' and 'kvm' groups are added here, so I put 'clock' group here.
Quoted 8 lines
From my understanding, such groups are likely to be used by other
applications (recipes). You can see this from the original commit:
https://github.com/systemd/systemd/commit/af96ccfc24bc4803078a46b4ef2cdebIt's possible that applications such as linuxptp will use this clock group. What systemd (udev) does is to change the device's group to 'clock'. This means applications using those devices could be put in 'clock' group to be granted access.
Quoted 6 lines
But how do classic binary distributions handle this? Can you research where e.g. Fedora or Debian create this group? We could probably follow that, and avoid having to carry an Inappropriate patch. Maybe base-passwd upstream already has a fix?
Alex
Hi Alex,
I just checked the fedora:latest container. It also puts 'clock' group in a basic setup package, which, according to its description, is basically 'base-passwd + base-files' in Yocto.
See commands and results below:
"""
[root@607fbfbf12bb /]# grep clock /etc/group
clock:x:103:
[root@607fbfbf12bb /]# rpm -qf /etc/group setup-2.15.0-13.fc42.noarch
[root@607fbfbf12bb /]# rpm -qi setup | grep -A 4 Summary Summary : A set of system configuration and setup files Description :
The setup package contains a set of important system configuration and setup files, such as passwd, group, and profile.
"""
I also checked base-passwd upstream and debian:latest container. base-passwd upstream does not have clock group: https://salsa.debian.org/debian/base-passwd/-/blob/master/group.master?ref_type=heads. And debian:latest is using systemd 257.8 so it does not have this 'clock' group issue now.
I'll use gid 103 in patch V4 to sync with fedora (though not necessary).
Regards,
Qi
Quoted 7 lines · Chen Qi via lists.openembedded.org
I also checked base-passwd upstream and debian:latest container. base-passwd upstream does not have clock group:
https://salsa.debian.org/debian/base-passwd/-/blob/master/group.master?ref_type=heads.
And debian:latest is using systemd 257.8 so it does not have this 'clock' group issue now.I'll use gid 103 in patch V4 to sync with fedora (though not necessary).
Thanks for the research. I think you can simply send the patch upstream to Debian via a pull request?
https://salsa.debian.org/debian/base-passwd/-/merge_requests
Alex
Quoted 7 lines
-----Original Message-----
From: Alexander Kanavin <[email protected]>
Sent: den 20 oktober 2025 13:59
To: [email protected]
Cc: Peter Kjellerstedt <[email protected]>; [email protected]
Subject: Re: [OE-core][PATCH V3 1/2] base-passwd: add clock group
Quoted 8 lines
Quoted 7 lines · Chen Qi via lists.openembedded.org
I also checked base-passwd upstream and debian:latest container. base-passwd upstream does not have clock group:
https://salsa.debian.org/debian/base-passwd/-/blob/master/group.master?ref_type=heads.
And debian:latest is using systemd 257.8 so it does not have this 'clock' group issue now.I'll use gid 103 in patch V4 to sync with fedora (though not necessary).
Since most of the other groups provided by base-passwd aren't synced with Fedora, can we stick with using IDs <= 100 for base-passwd? It would at least make my life easier since we have already used a huge number of static IDs > 100.
Quoted 6 lines
Thanks for the research. I think you can simply send the patch upstream to Debian via a pull request?
https://salsa.debian.org/debian/base-passwd/-/merge_requestsAlex
//Peter