| Message ID | 20230215171832.635085-1-Martin.Jansa@gmail.com |
|---|---|
| State | New |
| Headers | show |
| Series | [RFC] libssh2: Update patch to apply and fix malformed Upstream-Status | expand |
On Wed, 2023-02-15 at 18:18 +0100, Martin Jansa wrote: > * I've noticed this, because patchreview.py reports Malformed > Upstream-Status in this one now, but the QA check in insane.bbclass > wasn't reporting it before > > * the reason why insane.bbclass doesn't report it is because: > SRC_URI:append:ptest = " file://0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch" > doesn't work and probably never did as "ptest" is not an override > and patch-status-core checks only the .patch files which are really > in SRC_URI. > > * I've added it directly to SRC_URI and it also didn't apply: > http://errors.yoctoproject.org/Errors/Details/689955/ > as the "endif" at the end was removed in 1.10.0 version: > https://github.com/libssh2/libssh2/commit/ecd6a74e44562797a1e92186ad4a402c5641720e#diff-32103f666ff2fb42b025a47ccf1b959bbcc6db89f217e5943b1de73c81a4f9db > > after updating the patch to apply I got expected patch-status-core failure: > http://errors.yoctoproject.org/Errors/Details/689956/ > ERROR: QA Issue: Malformed Upstream-Status in patch > TOPDIR/openembedded-core/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > Please correct according to https://www.openembedded.org/wiki/Commit_Patch_Message_Guidelines#Patch_Header_Recommendations:_Upstream-Status : > Upstream-Status: Inappropriate[oe specific] [patch-status-core] > > * so the mystery is solved, but what should be do with this never-used > patch, I think we should just delete it together with > SRC_URI:append:ptest, but please Changqing Li confirm it's not needed > > * it was originally added by Changqing Li in meta-oe with: > https://git.openembedded.org/meta-openembedded/commit/?id=d7aa7173405c3b36235af736cd31dbe110708787 > then imported to oe-core by Randy MacLeod with: > https://git.openembedded.org/openembedded-core/commit/?id=57df134b1be56a688f41851e5ff014dd859c0bc3 > > Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> > --- > ...nviroment-to-decide-if-a-test-is-bui.patch | 22 ++++++------------- > 1 file changed, 7 insertions(+), 15 deletions(-) > > diff --git a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > index b1204e49eb..e0000b2658 100644 > --- a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > +++ b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > @@ -1,24 +1,23 @@ > -From f6abce5ba41a412a247250dcd80e387e53474466 Mon Sep 17 00:00:00 2001 > -From: Your Name <you@example.com> > +From 7ffa1773be8d262bed0f5c8bdcb4dd8f906de095 Mon Sep 17 00:00:00 2001 > +From: Changqing Li <changqing.li@windriver.com> > Date: Mon, 28 Dec 2020 02:08:03 +0000 > Subject: [PATCH] Don't let host enviroment to decide if a test is build > > test ssh2.sh need sshd, for cross compile, we need it on target, so > don't use SSHD on host to decide weither to build a test > > -Upstream-Status: Inappropriate[oe specific] > +Upstream-Status: Inappropriate [oe specific] > > Signed-off-by: Changqing Li <changqing.li@windriver.com> The patch is to fix ssh2.sh as a test. The test was disabled here: https://git.yoctoproject.org/poky/commit/meta/recipes-support/libssh2?id=af067d427fc1000c2dcce84c5578ea26f9048208 referencing: https://github.com/libssh2/libssh2/issues/630 which is now fixed. The autobuilder doesn't currently run the test: https://autobuilder.yocto.io/pub/non-release/20230215-12/testresults/qemux86-64-ptest/libssh2.log I don't know if the test still needs the patch to fix it. Cheers, Richard
On Wed, Feb 15, 2023 at 9:18 AM Martin Jansa <Martin.Jansa@gmail.com> wrote: > > * I've noticed this, because patchreview.py reports Malformed > Upstream-Status in this one now, but the QA check in insane.bbclass > wasn't reporting it before > > * the reason why insane.bbclass doesn't report it is because: > SRC_URI:append:ptest = " file://0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch" > doesn't work and probably never did as "ptest" is not an override > and patch-status-core checks only the .patch files which are really > in SRC_URI. > > * I've added it directly to SRC_URI and it also didn't apply: > http://errors.yoctoproject.org/Errors/Details/689955/ > as the "endif" at the end was removed in 1.10.0 version: > https://github.com/libssh2/libssh2/commit/ecd6a74e44562797a1e92186ad4a402c5641720e#diff-32103f666ff2fb42b025a47ccf1b959bbcc6db89f217e5943b1de73c81a4f9db > > after updating the patch to apply I got expected patch-status-core failure: > http://errors.yoctoproject.org/Errors/Details/689956/ > ERROR: QA Issue: Malformed Upstream-Status in patch > TOPDIR/openembedded-core/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > Please correct according to https://www.openembedded.org/wiki/Commit_Patch_Message_Guidelines#Patch_Header_Recommendations:_Upstream-Status : > Upstream-Status: Inappropriate[oe specific] [patch-status-core] > > * so the mystery is solved, but what should be do with this never-used > patch, I think we should just delete it together with > SRC_URI:append:ptest, but please Changqing Li confirm it's not needed > > * it was originally added by Changqing Li in meta-oe with: > https://git.openembedded.org/meta-openembedded/commit/?id=d7aa7173405c3b36235af736cd31dbe110708787 > then imported to oe-core by Randy MacLeod with: > https://git.openembedded.org/openembedded-core/commit/?id=57df134b1be56a688f41851e5ff014dd859c0bc3 > > Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> > --- > ...nviroment-to-decide-if-a-test-is-bui.patch | 22 ++++++------------- > 1 file changed, 7 insertions(+), 15 deletions(-) > > diff --git a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > index b1204e49eb..e0000b2658 100644 > --- a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > +++ b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > @@ -1,24 +1,23 @@ > -From f6abce5ba41a412a247250dcd80e387e53474466 Mon Sep 17 00:00:00 2001 > -From: Your Name <you@example.com> > +From 7ffa1773be8d262bed0f5c8bdcb4dd8f906de095 Mon Sep 17 00:00:00 2001 > +From: Changqing Li <changqing.li@windriver.com> > Date: Mon, 28 Dec 2020 02:08:03 +0000 > Subject: [PATCH] Don't let host enviroment to decide if a test is build > > test ssh2.sh need sshd, for cross compile, we need it on target, so > don't use SSHD on host to decide weither to build a test > > -Upstream-Status: Inappropriate[oe specific] > +Upstream-Status: Inappropriate [oe specific] Since we are already using a special token '[' to separate reason from status, I always wondered if having this space before '[' is something that could be ignored. > > Signed-off-by: Changqing Li <changqing.li@windriver.com> > - > --- > - tests/Makefile.am | 6 +----- > - 1 file changed, 1 insertion(+), 5 deletions(-) > + tests/Makefile.am | 4 ---- > + 1 file changed, 4 deletions(-) > > diff --git a/tests/Makefile.am b/tests/Makefile.am > -index dc0922f..6cbc35d 100644 > +index 27ddc2d..13de8ab 100644 > --- a/tests/Makefile.am > +++ b/tests/Makefile.am > -@@ -1,16 +1,12 @@ > +@@ -3,16 +3,12 @@ SUBDIRS = ossfuzz > AM_CPPFLAGS = -I$(top_srcdir)/src -I$(top_srcdir)/include -I$(top_builddir)/src > LDADD = ../src/libssh2.la > > @@ -35,10 +34,3 @@ index dc0922f..6cbc35d 100644 > check_PROGRAMS = $(ctests) > > TESTS_ENVIRONMENT = SSHD=$(SSHD) EXEEXT=$(EXEEXT) > -@@ -38,4 +34,4 @@ if OPENSSL > - # EXTRA_DIST += test_public_key_auth_succeeds_with_correct_encrypted_ed25519_key.c > - # EXTRA_DIST += test_public_key_auth_succeeds_with_correct_ed25519_key_from_mem.c > - EXTRA_DIST += test_public_key_auth_succeeds_with_correct_rsa_openssh_key.c > --endif > -\ No newline at end of file > -+endif > -- > 2.39.2 > > > -=-=-=-=-=-=-=-=-=-=-=- > Links: You receive all messages sent to this group. > View/Reply Online (#177204): https://lists.openembedded.org/g/openembedded-core/message/177204 > Mute This Topic: https://lists.openembedded.org/mt/96987532/1997914 > Group Owner: openembedded-core+owner@lists.openembedded.org > Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [raj.khem@gmail.com] > -=-=-=-=-=-=-=-=-=-=-=- >
Please no. Much less readable without a space. Alex On Wed 15. Feb 2023 at 20.27, Khem Raj <raj.khem@gmail.com> wrote: > On Wed, Feb 15, 2023 at 9:18 AM Martin Jansa <Martin.Jansa@gmail.com> > wrote: > > > > * I've noticed this, because patchreview.py reports Malformed > > Upstream-Status in this one now, but the QA check in insane.bbclass > > wasn't reporting it before > > > > * the reason why insane.bbclass doesn't report it is because: > > SRC_URI:append:ptest = " > file://0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch" > > doesn't work and probably never did as "ptest" is not an override > > and patch-status-core checks only the .patch files which are really > > in SRC_URI. > > > > * I've added it directly to SRC_URI and it also didn't apply: > > http://errors.yoctoproject.org/Errors/Details/689955/ > > as the "endif" at the end was removed in 1.10.0 version: > > > https://github.com/libssh2/libssh2/commit/ecd6a74e44562797a1e92186ad4a402c5641720e#diff-32103f666ff2fb42b025a47ccf1b959bbcc6db89f217e5943b1de73c81a4f9db > > > > after updating the patch to apply I got expected patch-status-core > failure: > > http://errors.yoctoproject.org/Errors/Details/689956/ > > ERROR: QA Issue: Malformed Upstream-Status in patch > > > TOPDIR/openembedded-core/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > Please correct according to > https://www.openembedded.org/wiki/Commit_Patch_Message_Guidelines#Patch_Header_Recommendations:_Upstream-Status > : > > Upstream-Status: Inappropriate[oe specific] [patch-status-core] > > > > * so the mystery is solved, but what should be do with this never-used > > patch, I think we should just delete it together with > > SRC_URI:append:ptest, but please Changqing Li confirm it's not needed > > > > * it was originally added by Changqing Li in meta-oe with: > > > https://git.openembedded.org/meta-openembedded/commit/?id=d7aa7173405c3b36235af736cd31dbe110708787 > > then imported to oe-core by Randy MacLeod with: > > > https://git.openembedded.org/openembedded-core/commit/?id=57df134b1be56a688f41851e5ff014dd859c0bc3 > > > > Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> > > --- > > ...nviroment-to-decide-if-a-test-is-bui.patch | 22 ++++++------------- > > 1 file changed, 7 insertions(+), 15 deletions(-) > > > > diff --git > a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > index b1204e49eb..e0000b2658 100644 > > --- > a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > +++ > b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > @@ -1,24 +1,23 @@ > > -From f6abce5ba41a412a247250dcd80e387e53474466 Mon Sep 17 00:00:00 2001 > > -From: Your Name <you@example.com> > > +From 7ffa1773be8d262bed0f5c8bdcb4dd8f906de095 Mon Sep 17 00:00:00 2001 > > +From: Changqing Li <changqing.li@windriver.com> > > Date: Mon, 28 Dec 2020 02:08:03 +0000 > > Subject: [PATCH] Don't let host enviroment to decide if a test is build > > > > test ssh2.sh need sshd, for cross compile, we need it on target, so > > don't use SSHD on host to decide weither to build a test > > > > -Upstream-Status: Inappropriate[oe specific] > > +Upstream-Status: Inappropriate [oe specific] > > Since we are already using a special token '[' to separate reason from > status, I always wondered if having this space before '[' is something > that could be ignored. > > > > > Signed-off-by: Changqing Li <changqing.li@windriver.com> > > - > > --- > > - tests/Makefile.am | 6 +----- > > - 1 file changed, 1 insertion(+), 5 deletions(-) > > + tests/Makefile.am | 4 ---- > > + 1 file changed, 4 deletions(-) > > > > diff --git a/tests/Makefile.am b/tests/Makefile.am > > -index dc0922f..6cbc35d 100644 > > +index 27ddc2d..13de8ab 100644 > > --- a/tests/Makefile.am > > +++ b/tests/Makefile.am > > -@@ -1,16 +1,12 @@ > > +@@ -3,16 +3,12 @@ SUBDIRS = ossfuzz > > AM_CPPFLAGS = -I$(top_srcdir)/src -I$(top_srcdir)/include > -I$(top_builddir)/src > > LDADD = ../src/libssh2.la > > > > @@ -35,10 +34,3 @@ index dc0922f..6cbc35d 100644 > > check_PROGRAMS = $(ctests) > > > > TESTS_ENVIRONMENT = SSHD=$(SSHD) EXEEXT=$(EXEEXT) > > -@@ -38,4 +34,4 @@ if OPENSSL > > - # EXTRA_DIST += > test_public_key_auth_succeeds_with_correct_encrypted_ed25519_key.c > > - # EXTRA_DIST += > test_public_key_auth_succeeds_with_correct_ed25519_key_from_mem.c > > - EXTRA_DIST += > test_public_key_auth_succeeds_with_correct_rsa_openssh_key.c > > --endif > > -\ No newline at end of file > > -+endif > > -- > > 2.39.2 > > > > > > > > > > -=-=-=-=-=-=-=-=-=-=-=- > Links: You receive all messages sent to this group. > View/Reply Online (#177209): > https://lists.openembedded.org/g/openembedded-core/message/177209 > Mute This Topic: https://lists.openembedded.org/mt/96987532/1686489 > Group Owner: openembedded-core+owner@lists.openembedded.org > Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [ > alex.kanavin@gmail.com] > -=-=-=-=-=-=-=-=-=-=-=- > >
On Wed, Feb 15, 2023 at 11:26 AM Khem Raj <raj.khem@gmail.com> wrote: > > On Wed, Feb 15, 2023 at 9:18 AM Martin Jansa <Martin.Jansa@gmail.com> wrote: > > > > * I've noticed this, because patchreview.py reports Malformed > > Upstream-Status in this one now, but the QA check in insane.bbclass > > wasn't reporting it before > > > > * the reason why insane.bbclass doesn't report it is because: > > SRC_URI:append:ptest = " file://0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch" > > doesn't work and probably never did as "ptest" is not an override > > and patch-status-core checks only the .patch files which are really > > in SRC_URI. > > > > * I've added it directly to SRC_URI and it also didn't apply: > > http://errors.yoctoproject.org/Errors/Details/689955/ > > as the "endif" at the end was removed in 1.10.0 version: > > https://github.com/libssh2/libssh2/commit/ecd6a74e44562797a1e92186ad4a402c5641720e#diff-32103f666ff2fb42b025a47ccf1b959bbcc6db89f217e5943b1de73c81a4f9db > > > > after updating the patch to apply I got expected patch-status-core failure: > > http://errors.yoctoproject.org/Errors/Details/689956/ > > ERROR: QA Issue: Malformed Upstream-Status in patch > > TOPDIR/openembedded-core/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > Please correct according to https://www.openembedded.org/wiki/Commit_Patch_Message_Guidelines#Patch_Header_Recommendations:_Upstream-Status : > > Upstream-Status: Inappropriate[oe specific] [patch-status-core] > > > > * so the mystery is solved, but what should be do with this never-used > > patch, I think we should just delete it together with > > SRC_URI:append:ptest, but please Changqing Li confirm it's not needed > > > > * it was originally added by Changqing Li in meta-oe with: > > https://git.openembedded.org/meta-openembedded/commit/?id=d7aa7173405c3b36235af736cd31dbe110708787 > > then imported to oe-core by Randy MacLeod with: > > https://git.openembedded.org/openembedded-core/commit/?id=57df134b1be56a688f41851e5ff014dd859c0bc3 > > > > Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> > > --- > > ...nviroment-to-decide-if-a-test-is-bui.patch | 22 ++++++------------- > > 1 file changed, 7 insertions(+), 15 deletions(-) > > > > diff --git a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > index b1204e49eb..e0000b2658 100644 > > --- a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > +++ b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > @@ -1,24 +1,23 @@ > > -From f6abce5ba41a412a247250dcd80e387e53474466 Mon Sep 17 00:00:00 2001 > > -From: Your Name <you@example.com> > > +From 7ffa1773be8d262bed0f5c8bdcb4dd8f906de095 Mon Sep 17 00:00:00 2001 > > +From: Changqing Li <changqing.li@windriver.com> > > Date: Mon, 28 Dec 2020 02:08:03 +0000 > > Subject: [PATCH] Don't let host enviroment to decide if a test is build > > > > test ssh2.sh need sshd, for cross compile, we need it on target, so > > don't use SSHD on host to decide weither to build a test > > > > -Upstream-Status: Inappropriate[oe specific] > > +Upstream-Status: Inappropriate [oe specific] > > Since we are already using a special token '[' to separate reason from > status, I always wondered if having this space before '[' is something > that could be ignored. > I see that scripts/contrib/patchreview.py does not care for space, that answers my question. Maybe it would be good to tabulate reason as well perhaps. > > > > Signed-off-by: Changqing Li <changqing.li@windriver.com> > > - > > --- > > - tests/Makefile.am | 6 +----- > > - 1 file changed, 1 insertion(+), 5 deletions(-) > > + tests/Makefile.am | 4 ---- > > + 1 file changed, 4 deletions(-) > > > > diff --git a/tests/Makefile.am b/tests/Makefile.am > > -index dc0922f..6cbc35d 100644 > > +index 27ddc2d..13de8ab 100644 > > --- a/tests/Makefile.am > > +++ b/tests/Makefile.am > > -@@ -1,16 +1,12 @@ > > +@@ -3,16 +3,12 @@ SUBDIRS = ossfuzz > > AM_CPPFLAGS = -I$(top_srcdir)/src -I$(top_srcdir)/include -I$(top_builddir)/src > > LDADD = ../src/libssh2.la > > > > @@ -35,10 +34,3 @@ index dc0922f..6cbc35d 100644 > > check_PROGRAMS = $(ctests) > > > > TESTS_ENVIRONMENT = SSHD=$(SSHD) EXEEXT=$(EXEEXT) > > -@@ -38,4 +34,4 @@ if OPENSSL > > - # EXTRA_DIST += test_public_key_auth_succeeds_with_correct_encrypted_ed25519_key.c > > - # EXTRA_DIST += test_public_key_auth_succeeds_with_correct_ed25519_key_from_mem.c > > - EXTRA_DIST += test_public_key_auth_succeeds_with_correct_rsa_openssh_key.c > > --endif > > -\ No newline at end of file > > -+endif > > -- > > 2.39.2 > > > > > > -=-=-=-=-=-=-=-=-=-=-=- > > Links: You receive all messages sent to this group. > > View/Reply Online (#177204): https://lists.openembedded.org/g/openembedded-core/message/177204 > > Mute This Topic: https://lists.openembedded.org/mt/96987532/1997914 > > Group Owner: openembedded-core+owner@lists.openembedded.org > > Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [raj.khem@gmail.com] > > -=-=-=-=-=-=-=-=-=-=-=- > >
patchreview.py will care about the space (like insane.bbclass does) if https://lists.openembedded.org/g/openembedded-core/message/177207 is merged On Wed, Feb 15, 2023 at 9:35 PM Khem Raj <raj.khem@gmail.com> wrote: > On Wed, Feb 15, 2023 at 11:26 AM Khem Raj <raj.khem@gmail.com> wrote: > > > > On Wed, Feb 15, 2023 at 9:18 AM Martin Jansa <Martin.Jansa@gmail.com> > wrote: > > > > > > * I've noticed this, because patchreview.py reports Malformed > > > Upstream-Status in this one now, but the QA check in insane.bbclass > > > wasn't reporting it before > > > > > > * the reason why insane.bbclass doesn't report it is because: > > > SRC_URI:append:ptest = " > file://0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch" > > > doesn't work and probably never did as "ptest" is not an override > > > and patch-status-core checks only the .patch files which are really > > > in SRC_URI. > > > > > > * I've added it directly to SRC_URI and it also didn't apply: > > > http://errors.yoctoproject.org/Errors/Details/689955/ > > > as the "endif" at the end was removed in 1.10.0 version: > > > > https://github.com/libssh2/libssh2/commit/ecd6a74e44562797a1e92186ad4a402c5641720e#diff-32103f666ff2fb42b025a47ccf1b959bbcc6db89f217e5943b1de73c81a4f9db > > > > > > after updating the patch to apply I got expected patch-status-core > failure: > > > http://errors.yoctoproject.org/Errors/Details/689956/ > > > ERROR: QA Issue: Malformed Upstream-Status in patch > > > > TOPDIR/openembedded-core/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > > Please correct according to > https://www.openembedded.org/wiki/Commit_Patch_Message_Guidelines#Patch_Header_Recommendations:_Upstream-Status > : > > > Upstream-Status: Inappropriate[oe specific] [patch-status-core] > > > > > > * so the mystery is solved, but what should be do with this never-used > > > patch, I think we should just delete it together with > > > SRC_URI:append:ptest, but please Changqing Li confirm it's not needed > > > > > > * it was originally added by Changqing Li in meta-oe with: > > > > https://git.openembedded.org/meta-openembedded/commit/?id=d7aa7173405c3b36235af736cd31dbe110708787 > > > then imported to oe-core by Randy MacLeod with: > > > > https://git.openembedded.org/openembedded-core/commit/?id=57df134b1be56a688f41851e5ff014dd859c0bc3 > > > > > > Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> > > > --- > > > ...nviroment-to-decide-if-a-test-is-bui.patch | 22 ++++++------------- > > > 1 file changed, 7 insertions(+), 15 deletions(-) > > > > > > diff --git > a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > > index b1204e49eb..e0000b2658 100644 > > > --- > a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > > +++ > b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > > @@ -1,24 +1,23 @@ > > > -From f6abce5ba41a412a247250dcd80e387e53474466 Mon Sep 17 00:00:00 2001 > > > -From: Your Name <you@example.com> > > > +From 7ffa1773be8d262bed0f5c8bdcb4dd8f906de095 Mon Sep 17 00:00:00 2001 > > > +From: Changqing Li <changqing.li@windriver.com> > > > Date: Mon, 28 Dec 2020 02:08:03 +0000 > > > Subject: [PATCH] Don't let host enviroment to decide if a test is > build > > > > > > test ssh2.sh need sshd, for cross compile, we need it on target, so > > > don't use SSHD on host to decide weither to build a test > > > > > > -Upstream-Status: Inappropriate[oe specific] > > > +Upstream-Status: Inappropriate [oe specific] > > > > Since we are already using a special token '[' to separate reason from > > status, I always wondered if having this space before '[' is something > > that could be ignored. > > > > I see that scripts/contrib/patchreview.py does not care for space, > that answers my question. Maybe it would be > good to tabulate reason as well perhaps. > > > > > > > Signed-off-by: Changqing Li <changqing.li@windriver.com> > > > - > > > --- > > > - tests/Makefile.am | 6 +----- > > > - 1 file changed, 1 insertion(+), 5 deletions(-) > > > + tests/Makefile.am | 4 ---- > > > + 1 file changed, 4 deletions(-) > > > > > > diff --git a/tests/Makefile.am b/tests/Makefile.am > > > -index dc0922f..6cbc35d 100644 > > > +index 27ddc2d..13de8ab 100644 > > > --- a/tests/Makefile.am > > > +++ b/tests/Makefile.am > > > -@@ -1,16 +1,12 @@ > > > +@@ -3,16 +3,12 @@ SUBDIRS = ossfuzz > > > AM_CPPFLAGS = -I$(top_srcdir)/src -I$(top_srcdir)/include > -I$(top_builddir)/src > > > LDADD = ../src/libssh2.la > > > > > > @@ -35,10 +34,3 @@ index dc0922f..6cbc35d 100644 > > > check_PROGRAMS = $(ctests) > > > > > > TESTS_ENVIRONMENT = SSHD=$(SSHD) EXEEXT=$(EXEEXT) > > > -@@ -38,4 +34,4 @@ if OPENSSL > > > - # EXTRA_DIST += > test_public_key_auth_succeeds_with_correct_encrypted_ed25519_key.c > > > - # EXTRA_DIST += > test_public_key_auth_succeeds_with_correct_ed25519_key_from_mem.c > > > - EXTRA_DIST += > test_public_key_auth_succeeds_with_correct_rsa_openssh_key.c > > > --endif > > > -\ No newline at end of file > > > -+endif > > > -- > > > 2.39.2 > > > > > > > > > -=-=-=-=-=-=-=-=-=-=-=- > > > Links: You receive all messages sent to this group. > > > View/Reply Online (#177204): > https://lists.openembedded.org/g/openembedded-core/message/177204 > > > Mute This Topic: https://lists.openembedded.org/mt/96987532/1997914 > > > Group Owner: openembedded-core+owner@lists.openembedded.org > > > Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub > [raj.khem@gmail.com] > > > -=-=-=-=-=-=-=-=-=-=-=- > > > >
On Wed, Feb 15, 2023 at 12:46 PM Martin Jansa <martin.jansa@gmail.com> wrote: > > patchreview.py will care about the space (like insane.bbclass does) if https://lists.openembedded.org/g/openembedded-core/message/177207 is merged > I see. thanks. I think this should be something every developer should then be able to run on the patchset they are about to post for review. Otherwise, it just add on to upstreaming friction that developer experiences and it won't scale. > On Wed, Feb 15, 2023 at 9:35 PM Khem Raj <raj.khem@gmail.com> wrote: >> >> On Wed, Feb 15, 2023 at 11:26 AM Khem Raj <raj.khem@gmail.com> wrote: >> > >> > On Wed, Feb 15, 2023 at 9:18 AM Martin Jansa <Martin.Jansa@gmail.com> wrote: >> > > >> > > * I've noticed this, because patchreview.py reports Malformed >> > > Upstream-Status in this one now, but the QA check in insane.bbclass >> > > wasn't reporting it before >> > > >> > > * the reason why insane.bbclass doesn't report it is because: >> > > SRC_URI:append:ptest = " file://0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch" >> > > doesn't work and probably never did as "ptest" is not an override >> > > and patch-status-core checks only the .patch files which are really >> > > in SRC_URI. >> > > >> > > * I've added it directly to SRC_URI and it also didn't apply: >> > > http://errors.yoctoproject.org/Errors/Details/689955/ >> > > as the "endif" at the end was removed in 1.10.0 version: >> > > https://github.com/libssh2/libssh2/commit/ecd6a74e44562797a1e92186ad4a402c5641720e#diff-32103f666ff2fb42b025a47ccf1b959bbcc6db89f217e5943b1de73c81a4f9db >> > > >> > > after updating the patch to apply I got expected patch-status-core failure: >> > > http://errors.yoctoproject.org/Errors/Details/689956/ >> > > ERROR: QA Issue: Malformed Upstream-Status in patch >> > > TOPDIR/openembedded-core/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch >> > > Please correct according to https://www.openembedded.org/wiki/Commit_Patch_Message_Guidelines#Patch_Header_Recommendations:_Upstream-Status : >> > > Upstream-Status: Inappropriate[oe specific] [patch-status-core] >> > > >> > > * so the mystery is solved, but what should be do with this never-used >> > > patch, I think we should just delete it together with >> > > SRC_URI:append:ptest, but please Changqing Li confirm it's not needed >> > > >> > > * it was originally added by Changqing Li in meta-oe with: >> > > https://git.openembedded.org/meta-openembedded/commit/?id=d7aa7173405c3b36235af736cd31dbe110708787 >> > > then imported to oe-core by Randy MacLeod with: >> > > https://git.openembedded.org/openembedded-core/commit/?id=57df134b1be56a688f41851e5ff014dd859c0bc3 >> > > >> > > Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> >> > > --- >> > > ...nviroment-to-decide-if-a-test-is-bui.patch | 22 ++++++------------- >> > > 1 file changed, 7 insertions(+), 15 deletions(-) >> > > >> > > diff --git a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch >> > > index b1204e49eb..e0000b2658 100644 >> > > --- a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch >> > > +++ b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch >> > > @@ -1,24 +1,23 @@ >> > > -From f6abce5ba41a412a247250dcd80e387e53474466 Mon Sep 17 00:00:00 2001 >> > > -From: Your Name <you@example.com> >> > > +From 7ffa1773be8d262bed0f5c8bdcb4dd8f906de095 Mon Sep 17 00:00:00 2001 >> > > +From: Changqing Li <changqing.li@windriver.com> >> > > Date: Mon, 28 Dec 2020 02:08:03 +0000 >> > > Subject: [PATCH] Don't let host enviroment to decide if a test is build >> > > >> > > test ssh2.sh need sshd, for cross compile, we need it on target, so >> > > don't use SSHD on host to decide weither to build a test >> > > >> > > -Upstream-Status: Inappropriate[oe specific] >> > > +Upstream-Status: Inappropriate [oe specific] >> > >> > Since we are already using a special token '[' to separate reason from >> > status, I always wondered if having this space before '[' is something >> > that could be ignored. >> > >> >> I see that scripts/contrib/patchreview.py does not care for space, >> that answers my question. Maybe it would be >> good to tabulate reason as well perhaps. >> >> > > >> > > Signed-off-by: Changqing Li <changqing.li@windriver.com> >> > > - >> > > --- >> > > - tests/Makefile.am | 6 +----- >> > > - 1 file changed, 1 insertion(+), 5 deletions(-) >> > > + tests/Makefile.am | 4 ---- >> > > + 1 file changed, 4 deletions(-) >> > > >> > > diff --git a/tests/Makefile.am b/tests/Makefile.am >> > > -index dc0922f..6cbc35d 100644 >> > > +index 27ddc2d..13de8ab 100644 >> > > --- a/tests/Makefile.am >> > > +++ b/tests/Makefile.am >> > > -@@ -1,16 +1,12 @@ >> > > +@@ -3,16 +3,12 @@ SUBDIRS = ossfuzz >> > > AM_CPPFLAGS = -I$(top_srcdir)/src -I$(top_srcdir)/include -I$(top_builddir)/src >> > > LDADD = ../src/libssh2.la >> > > >> > > @@ -35,10 +34,3 @@ index dc0922f..6cbc35d 100644 >> > > check_PROGRAMS = $(ctests) >> > > >> > > TESTS_ENVIRONMENT = SSHD=$(SSHD) EXEEXT=$(EXEEXT) >> > > -@@ -38,4 +34,4 @@ if OPENSSL >> > > - # EXTRA_DIST += test_public_key_auth_succeeds_with_correct_encrypted_ed25519_key.c >> > > - # EXTRA_DIST += test_public_key_auth_succeeds_with_correct_ed25519_key_from_mem.c >> > > - EXTRA_DIST += test_public_key_auth_succeeds_with_correct_rsa_openssh_key.c >> > > --endif >> > > -\ No newline at end of file >> > > -+endif >> > > -- >> > > 2.39.2 >> > > >> > > >> > > -=-=-=-=-=-=-=-=-=-=-=- >> > > Links: You receive all messages sent to this group. >> > > View/Reply Online (#177204): https://lists.openembedded.org/g/openembedded-core/message/177204 >> > > Mute This Topic: https://lists.openembedded.org/mt/96987532/1997914 >> > > Group Owner: openembedded-core+owner@lists.openembedded.org >> > > Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [raj.khem@gmail.com] >> > > -=-=-=-=-=-=-=-=-=-=-=- >> > >
On Wed, 2023-02-15 at 17:29 +0000, Richard Purdie via lists.openembedded.org wrote: > On Wed, 2023-02-15 at 18:18 +0100, Martin Jansa wrote: > > * I've noticed this, because patchreview.py reports Malformed > > Upstream-Status in this one now, but the QA check in insane.bbclass > > wasn't reporting it before > > > > * the reason why insane.bbclass doesn't report it is because: > > SRC_URI:append:ptest = " file://0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch" > > doesn't work and probably never did as "ptest" is not an override > > and patch-status-core checks only the .patch files which are really > > in SRC_URI. > > > > * I've added it directly to SRC_URI and it also didn't apply: > > http://errors.yoctoproject.org/Errors/Details/689955/ > > as the "endif" at the end was removed in 1.10.0 version: > > https://github.com/libssh2/libssh2/commit/ecd6a74e44562797a1e92186ad4a402c5641720e#diff-32103f666ff2fb42b025a47ccf1b959bbcc6db89f217e5943b1de73c81a4f9db > > > > after updating the patch to apply I got expected patch-status-core failure: > > http://errors.yoctoproject.org/Errors/Details/689956/ > > ERROR: QA Issue: Malformed Upstream-Status in patch > > TOPDIR/openembedded-core/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > Please correct according to https://www.openembedded.org/wiki/Commit_Patch_Message_Guidelines#Patch_Header_Recommendations:_Upstream-Status : > > Upstream-Status: Inappropriate[oe specific] [patch-status-core] > > > > * so the mystery is solved, but what should be do with this never-used > > patch, I think we should just delete it together with > > SRC_URI:append:ptest, but please Changqing Li confirm it's not needed > > > > * it was originally added by Changqing Li in meta-oe with: > > https://git.openembedded.org/meta-openembedded/commit/?id=d7aa7173405c3b36235af736cd31dbe110708787 > > then imported to oe-core by Randy MacLeod with: > > https://git.openembedded.org/openembedded-core/commit/?id=57df134b1be56a688f41851e5ff014dd859c0bc3 > > > > Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> > > --- > > ...nviroment-to-decide-if-a-test-is-bui.patch | 22 ++++++------------- > > 1 file changed, 7 insertions(+), 15 deletions(-) > > > > diff --git a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > index b1204e49eb..e0000b2658 100644 > > --- a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > +++ b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch > > @@ -1,24 +1,23 @@ > > -From f6abce5ba41a412a247250dcd80e387e53474466 Mon Sep 17 00:00:00 2001 > > -From: Your Name <you@example.com> > > +From 7ffa1773be8d262bed0f5c8bdcb4dd8f906de095 Mon Sep 17 00:00:00 2001 > > +From: Changqing Li <changqing.li@windriver.com> > > Date: Mon, 28 Dec 2020 02:08:03 +0000 > > Subject: [PATCH] Don't let host enviroment to decide if a test is build > > > > test ssh2.sh need sshd, for cross compile, we need it on target, so > > don't use SSHD on host to decide weither to build a test > > > > -Upstream-Status: Inappropriate[oe specific] > > +Upstream-Status: Inappropriate [oe specific] > > > > Signed-off-by: Changqing Li <changqing.li@windriver.com> > > The patch is to fix ssh2.sh as a test. The test was disabled here: > > https://git.yoctoproject.org/poky/commit/meta/recipes-support/libssh2?id=af067d427fc1000c2dcce84c5578ea26f9048208 > > referencing: > > https://github.com/libssh2/libssh2/issues/630 > > which is now fixed. > > The autobuilder doesn't currently run the test: > > https://autobuilder.yocto.io/pub/non-release/20230215-12/testresults/qemux86-64-ptest/libssh2.log > > I don't know if the test still needs the patch to fix it. I did some tests and there is a patch to fix issue 630 which then does allow ssh2.sh to pass without this patch. We can therefore remove it. We should also add a patch to sort issue 630 and enable the test. Cheers, Richard
diff --git a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch index b1204e49eb..e0000b2658 100644 --- a/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch +++ b/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch @@ -1,24 +1,23 @@ -From f6abce5ba41a412a247250dcd80e387e53474466 Mon Sep 17 00:00:00 2001 -From: Your Name <you@example.com> +From 7ffa1773be8d262bed0f5c8bdcb4dd8f906de095 Mon Sep 17 00:00:00 2001 +From: Changqing Li <changqing.li@windriver.com> Date: Mon, 28 Dec 2020 02:08:03 +0000 Subject: [PATCH] Don't let host enviroment to decide if a test is build test ssh2.sh need sshd, for cross compile, we need it on target, so don't use SSHD on host to decide weither to build a test -Upstream-Status: Inappropriate[oe specific] +Upstream-Status: Inappropriate [oe specific] Signed-off-by: Changqing Li <changqing.li@windriver.com> - --- - tests/Makefile.am | 6 +----- - 1 file changed, 1 insertion(+), 5 deletions(-) + tests/Makefile.am | 4 ---- + 1 file changed, 4 deletions(-) diff --git a/tests/Makefile.am b/tests/Makefile.am -index dc0922f..6cbc35d 100644 +index 27ddc2d..13de8ab 100644 --- a/tests/Makefile.am +++ b/tests/Makefile.am -@@ -1,16 +1,12 @@ +@@ -3,16 +3,12 @@ SUBDIRS = ossfuzz AM_CPPFLAGS = -I$(top_srcdir)/src -I$(top_srcdir)/include -I$(top_builddir)/src LDADD = ../src/libssh2.la @@ -35,10 +34,3 @@ index dc0922f..6cbc35d 100644 check_PROGRAMS = $(ctests) TESTS_ENVIRONMENT = SSHD=$(SSHD) EXEEXT=$(EXEEXT) -@@ -38,4 +34,4 @@ if OPENSSL - # EXTRA_DIST += test_public_key_auth_succeeds_with_correct_encrypted_ed25519_key.c - # EXTRA_DIST += test_public_key_auth_succeeds_with_correct_ed25519_key_from_mem.c - EXTRA_DIST += test_public_key_auth_succeeds_with_correct_rsa_openssh_key.c --endif -\ No newline at end of file -+endif
* I've noticed this, because patchreview.py reports Malformed Upstream-Status in this one now, but the QA check in insane.bbclass wasn't reporting it before * the reason why insane.bbclass doesn't report it is because: SRC_URI:append:ptest = " file://0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch" doesn't work and probably never did as "ptest" is not an override and patch-status-core checks only the .patch files which are really in SRC_URI. * I've added it directly to SRC_URI and it also didn't apply: http://errors.yoctoproject.org/Errors/Details/689955/ as the "endif" at the end was removed in 1.10.0 version: https://github.com/libssh2/libssh2/commit/ecd6a74e44562797a1e92186ad4a402c5641720e#diff-32103f666ff2fb42b025a47ccf1b959bbcc6db89f217e5943b1de73c81a4f9db after updating the patch to apply I got expected patch-status-core failure: http://errors.yoctoproject.org/Errors/Details/689956/ ERROR: QA Issue: Malformed Upstream-Status in patch TOPDIR/openembedded-core/meta/recipes-support/libssh2/files/0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch Please correct according to https://www.openembedded.org/wiki/Commit_Patch_Message_Guidelines#Patch_Header_Recommendations:_Upstream-Status : Upstream-Status: Inappropriate[oe specific] [patch-status-core] * so the mystery is solved, but what should be do with this never-used patch, I think we should just delete it together with SRC_URI:append:ptest, but please Changqing Li confirm it's not needed * it was originally added by Changqing Li in meta-oe with: https://git.openembedded.org/meta-openembedded/commit/?id=d7aa7173405c3b36235af736cd31dbe110708787 then imported to oe-core by Randy MacLeod with: https://git.openembedded.org/openembedded-core/commit/?id=57df134b1be56a688f41851e5ff014dd859c0bc3 Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> --- ...nviroment-to-decide-if-a-test-is-bui.patch | 22 ++++++------------- 1 file changed, 7 insertions(+), 15 deletions(-)