On 7/6/20 2:52 PM, Yu, Mingli wrote:
On 7/2/20 12:35 AM, Richard Purdie wrote:
On Wed, 2020-07-01 at 10:05 +0800, Yu, Mingli wrote:
On 7/1/20 4:08 AM, Richard Purdie wrote:
On Mon, 2020-06-29 at 12:31 +0800, Yu, Mingli wrote:
From: Mingli Yu <[email protected]>
Use git repo as the the previous URL only stores
the latest source file and fails to locate the
source tar file if we don't upgrade timely.
Signed-off-by: Mingli Yu <[email protected]>
---
meta/recipes-core/netbase/netbase_6.1.bb | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/meta/recipes-core/netbase/netbase_6.1.bb
b/meta/recipes-core/netbase/netbase_6.1.bb
index bc0049c020..c13685e8ad 100644
--- a/meta/recipes-core/netbase/netbase_6.1.bb
+++ b/meta/recipes-core/netbase/netbase_6.1.bb
@@ -6,10 +6,10 @@ LICENSE = "GPLv2"
LIC_FILES_CHKSUM =
"file://debian/copyright;md5=3dd6192d306f582dee7687da3d8748ab"
PE = "1"
-SRC_URI = "${DEBIAN_MIRROR}/main/n/${BPN}/${BPN}_${PV}.tar.xz"
+SRC_URI = "git://salsa.debian.org/md/netbase.git;protocol=https"
+SRCREV = "0fc1e4ce39328f7388badace0aaf7b7294d5ed61"
-SRC_URI[md5sum] = "e5871a3a5c8390557b8033cf19316a55"
-SRC_URI[sha256sum] =
"084d743bd84d4d9380bac4c71c51e57406dce44f5a69289bb823c903e9b035d8"
+S = "${WORKDIR}/git"
UPSTREAM_CHECK_URI = "${DEBIAN_MIRROR}/main/n/netbase/"
do_install () {
This still breaks all oe-selftest:
https://autobuilder.yoctoproject.org/typhoon/#/builders/79/builds/1095
https://autobuilder.yoctoproject.org/typhoon/#/builders/80/builds/1087
https://autobuilder.yoctoproject.org/typhoon/#/builders/86/builds/1083
https://autobuilder.yoctoproject.org/typhoon/#/builders/87/builds/1098
oe-selftest -r reproducible.ReproducibleTests.test_reproducible_builds
It's weird, I did run reproducible build locally with "oe-selftest -r
reproducible.ReproducibleTests.test_reproducible_builds" and It succeeds
at my side.
BTW, the netbase only introduces 3 files as below:
# rpm -ql netbase
/etc
/etc/protocols
/etc/rpc
/etc/services
And I did the three files' checksum and the checksum turns out to be the
same when with the commit and without the commit.
Anyway, I will investigate the issue again.
This is the diffstat output on the result:
https://autobuilder.yocto.io/pub/repro-fail/oe-reproducible-20200630-_jn1koza/packages/diff-html/
It shows this is a timestamp problem, the timestamps aren't always the
same. It should be using SOURCE_DATE_EPOCH.
Hi RP,
Many thanks your hints!
But I still don't understand why the reproducible test succeeds at my
side and the corresponding SOURCE_DATE_EPOCH in my env as below:
$ cat
/builda/yocto/builds/poky-build6-st/reproducibleA/tmp/work/cortexa57-poky-linux/netbase/1_6.1-r0/source-date-epoch/__source_date_epoch.txt
1581895519
$ cat
/build/yocto/builds/poky-build6-st/reproducibleB/tmp/work/cortexa57-poky-linux/netbase/1_6.1-r0/s-epoch/__source_date_epoch.txt
1581895519
>>> import time
>>> time.ctime(1581895519)
'Sun Feb 16 23:25:19 2020'
That's to say, both reproducibleA and reproducibleB build in my env has
the timestamp 2020-02-16·23:25:19.000000 with your reproducibleB
build as
https://autobuilder.yocto.io/pub/repro-fail/oe-reproducible-20200630-_jn1koza/packages/diff-html/
which you pasted.
Thanks,
Thanks,
Cheers,
Richard
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#140321):
https://lists.openembedded.org/g/openembedded-core/message/140321
Mute This Topic: https://lists.openembedded.org/mt/75185765/21656
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-
