On 8/27/19 12:15 PM, Jan Klare wrote:
Hi,
Thanks for all the initial help and the quick followup review. I
successfully build systemd for the patched SRCREV for all the platforms
you mentioned, should I post something of the output or some proof
somewhere?
No, that usually would be fine.
We trust you and the automated builders will verify that things work as
well. :)
But...
In this case I just realized that my oneline build-all was wrong;
you need:
$ for i in `ls ../../oe-core.git/meta/conf/machine | \
grep qemu | sed -e 's/\.conf//'`; \
do \
MACHINE=$i bitbake systemd && \
touch sd-okay-$i || \
touch sd-fail-$i; \
done
I had left off the setting of the MACHINE variable. Sorry about that.
Can you re-run the test after removing the sd-* files?
../Randy
Cheers,
Jan
On 27. Aug 2019, at 17:14, Randy MacLeod <[email protected]
<mailto:[email protected]>> wrote:
On 8/27/19 10:52 AM, Jan Klare wrote:
Currently systemd 241 does break for kernels 5.2+ with the error
described here:
* https://github.com/systemd/systemd/issues/12784
The issue has been fixed in master and will be fixed in the release
243. The
necessary patches have been backported to systemd/systemd-stable in
the branch
v241-stable, but currently in warrior an old version of that branch
is pulled
in.
This patch updates the SRCREV to the latest commit from that branch and
therefore pulls in the needed fix to run systemd 241 on 5.2+ kernels.
Signed-off-by: Jan Klare <[email protected] <mailto:[email protected]>>
---
meta/recipes-core/systemd/systemd.inc | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta/recipes-core/systemd/systemd.inc
b/meta/recipes-core/systemd/systemd.inc
index 5bd88ed6ed..2b9c291959 100644
--- a/meta/recipes-core/systemd/systemd.inc
+++ b/meta/recipes-core/systemd/systemd.inc
@@ -14,7 +14,7 @@ LICENSE = "GPLv2 & LGPLv2.1"
LIC_FILES_CHKSUM =
"file://LICENSE.GPL2;md5=751419260aa954499f7abaabaa882bbe \
file://LICENSE.LGPL2.1;md5=4fbd65380cdd255951079008b364516c"
-SRCREV = "c1f8ff8d0de7e303b8004b02a0a47d4cc103a7f8"
+SRCREV = "511646b8ac5c82f210b16920044465756913d238"
SRCBRANCH = "v241-stable"
SRC_URI =
"git://github.com/systemd/systemd-stable.git;protocol=git;branch=${SRCBRANCH}"
That's a better log since it explains your motivation.
We don't require it but if you send something like this commit again,
it would be good to analyze the commit logs, like this (but likely
without the tutorial flavour):
$ cd .../systemd-stable.git
$ git checkout v241-stable
$ git pull
$ git log --oneline c1f8ff8d0de7..511646b8ac5c | wc -l
157
The patches are almost all (1) are cherry-picked back from master to
fix bugs. The four commits (2) that are not cherry-picks,
are:
A revert of something that does not belong in stable,
A customized fix for stable,
A bug fix
A merge commit
I was talking with Jan on IRC and he said that he had built
and tested this update for his target (agema-ag7648 switch).
I've asked Jan to build for all qemus so hopefully that goes well.
Looks good to me.
../Randy
[1]
$ git log --oneline \
--grep="cherry picked from commit" c1f8ff8d0de7..511646b8ac5c | wc -l
153
$ git log --oneline c1f8ff8d0de7..511646b8ac5c | wc -l
157
[2]
$ git log --oneline --grep="cherry picked from commit" \
--invert-grep c1f8ff8d0de7..511646b8ac5c
099c5e4f5f Revert "meson: stop creating enablement symlinks
in /etc during installation"
dc903ec516 Enable RestrictSUIDSGID=yes for our services
with DynamicUser=yes
ce52d60450 fstab-generator: use DefaultDependencies=no for
/sysroot mounts
35387ea7e1 Merge pull request #48 from
yuwata/v241-stable-dbus-and-timedated
--
# Randy MacLeod
# Wind River Linux
--
_______________________________________________
Openembedded-core mailing list
[email protected]
<mailto:[email protected]>
http://lists.openembedded.org/mailman/listinfo/openembedded-core
--
# Randy MacLeod
# Wind River Linux
--
_______________________________________________
Openembedded-core mailing list
[email protected]
http://lists.openembedded.org/mailman/listinfo/openembedded-core
