Skip to contentSkip to conversation

oeqa/runtime: fix regression in minidebuginfo test

PATCH Superseded [email protected] 3 participants 12 messages May 29, 2024 – Jun 5, 2024 2 messages not in the archive

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 oeqa/runtime: fix regression in minidebuginfo test

From: Etienne Cordonnier <[email protected]>

The builtin command "sleep" was getting executed instead of the busybox binary. Hence "pidof sleep" was empty and the test was failing.

---
 meta/lib/oeqa/runtime/cases/systemd.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
meta/lib/oeqa/runtime/cases/systemd.py +1 -1
index 5481e1d840..a2229848e7 100644
@@ -145,7 +145,7 @@ class SystemdServiceTests(SystemdTest):
         Verify that call-stacks generated by systemd-coredump contain symbolicated call-stacks,
         extracted from the minidebuginfo metadata (.gnu_debugdata elf section).
         """
-        t_thread = threading.Thread(target=self.target.run, args=("ulimit -c unlimited && sleep 1000",))
+        t_thread = threading.Thread(target=self.target.run, args=("ulimit -c unlimited && /usr/bin/sleep 1000",))
         t_thread.start()
         time.sleep(1)
 
Signature · 1 line
2.36.1.vfs.0.0
#1 lore raw
Quoted 16 lines · Etienne Cordonnier via lists.openembedded.org

From: Etienne Cordonnier <[email protected]>

The builtin command "sleep" was getting executed instead of the busybox binary. Hence "pidof sleep" was empty and the test was failing.

---
 meta/lib/oeqa/runtime/cases/systemd.py | 2 +-

 1 file changed, 1 insertion(+), 1 deletion(-)
meta/lib/oeqa/runtime/cases/systemd.py +0 -0
index 5481e1d840..a2229848e7 100644
@@ -145,7 +145,7 @@ class SystemdServiceTests(SystemdTest):
          Verify that call-stacks generated by systemd-coredump contain symbolicated call-stacks,
Quoted 6 lines

         extracted from the minidebuginfo metadata (.gnu_debugdata elf section).
         """
-        t_thread = threading.Thread(target=self.target.run, args=("ulimit -c unlimited && sleep 1000",)) +        t_thread = threading.Thread(target=self.target.run, args=("ulimit -c unlimited && /usr/bin/sleep 1000",)) t_thread.start()
         time.sleep(1)

I get a bit nervous about hardcoding paths like "/usr/bin", particularly in the context of some distros liking to move things around. We may want to put a shutil.which("sleep") call in instead (and an import shutil if needed)?

Cheers,

Richard

#2 lore raw

Hi Richard,
I thought it was OK because systemd requires usrmerge, but I've sent a new version of the patch using "env sleep" instead of "/usr/bin/sleep" in order to be safe. I hope this also fixes the test failure which was happening on CI.
One reason this could be failing in CI is that this functionality currently requires user-namespaces enabled in the kernel (e.g. core-image-sato does have CONFIG_USER_NS, thus it works with this image). This is not really by design, and I'm working on changing this in the upstream systemd:
https://github.com/systemd/systemd/pull/33043 . I can then backport this on scarthgap if the change is merged upstream.

Étienne

Quoted 41 lines · Richard Purdie
Quoted 3 lines · Etienne Cordonnier via lists.openembedded.org

From: Etienne Cordonnier <[email protected]>

The builtin command "sleep" was getting executed instead of the busybox

binary.

Review

Quoted 8 lines

Hence "pidof sleep" was empty and the test was failing.

---
 meta/lib/oeqa/runtime/cases/systemd.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
(quoted diff) +0 -0

b/meta/lib/oeqa/runtime/cases/systemd.py

meta/lib/oeqa/runtime/cases/systemd.py class SystemdServiceTests(SystemdTest): · line 145

Quoted 5 lines

index 5481e1d840..a2229848e7 100644

meta/lib/oeqa/runtime/cases/systemd.py +0 -0
@@ -145,7 +145,7 @@ class SystemdServiceTests(SystemdTest):
         Verify that call-stacks generated by systemd-coredump contain

symbolicated call-stacks,

meta/lib/oeqa/runtime/cases/systemd.py class SystemdServiceTests(SystemdTest): · line 146

Quoted 1 line
meta/lib/oeqa/runtime/cases/systemd.py +0 -0
         extracted from the minidebuginfo metadata (.gnu_debugdata elf

section).

meta/lib/oeqa/runtime/cases/systemd.py class SystemdServiceTests(SystemdTest): · line 148

Quoted 2 lines
meta/lib/oeqa/runtime/cases/systemd.py +0 -1
         """
-        t_thread = threading.Thread(target=self.target.run,

args=("ulimit -c unlimited && sleep 1000",))

meta/lib/oeqa/runtime/cases/systemd.py class SystemdServiceTests(SystemdTest): · line 148

Quoted 1 line
meta/lib/oeqa/runtime/cases/systemd.py +1 -0
+        t_thread = threading.Thread(target=self.target.run,

args=("ulimit -c unlimited && /usr/bin/sleep 1000",))

meta/lib/oeqa/runtime/cases/systemd.py class SystemdServiceTests(SystemdTest): · line 150

Quoted 2 lines
meta/lib/oeqa/runtime/cases/systemd.py +0 -0
         t_thread.start()
         time.sleep(1)

I get a bit nervous about hardcoding paths like "/usr/bin", particularly in the context of some distros liking to move things around. We may want to put a shutil.which("sleep") call in instead (and an import shutil if needed)?

Cheers,

Richard

A message that is not in the archive (sent off-list or not yet received): [email protected]

#4 lore raw

Hi Richard,
thanks for merging the v2 of this patch. Is this test running in CI currently? I'm thinking I could send a PR to enable minidebuginfo in poky-altcfg, but is there a way I can verify that this test works in yocto's CI? (it works locally, but as I mentioned it would fail on a kernel with CONFIG_USER_NS disabled.)

Étienne

Quoted 73 lines · Etienne Cordonnier via lists.openembedded.org

Hi Richard,
I thought it was OK because systemd requires usrmerge, but I've sent a new version of the patch using "env sleep" instead of "/usr/bin/sleep" in order to be safe. I hope this also fixes the test failure which was happening on CI.
One reason this could be failing in CI is that this functionality currently requires user-namespaces enabled in the kernel (e.g. core-image-sato does have CONFIG_USER_NS, thus it works with this image). This is not really by design, and I'm working on changing this in the upstream systemd: https://github.com/systemd/systemd/pull/33043 . I can then backport this on scarthgap if the change is merged upstream.

Étienne

Quoted 33 lines · Richard Purdie

On Wed, 2024-05-29 at 12:08 +0200, Etienne Cordonnier via lists.openembedded.org
<https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.openembedded.org&d=DwMFaQ&c=ncDTmphkJTvjIDPh0hpF_4vCHvabgGkICC2epckfdiw&r=AhkbNonVuMIGRfPx_Qj9TsRih1DULJTKUkSGa66m67E&m=qu0Injv6pHxfa5BgL3x29y11Al_3EpMdBOTkj8cWHLaFlxRoh9vy7WMU3VU2o6c-&s=jZOAoPx0ETGk3PoW9Wmzdl9wuDwTZtC_WxYX7O3EPFY&e=>

Quoted 3 lines · wrote:

From: Etienne Cordonnier <[email protected]>

The builtin command "sleep" was getting executed instead of the busybox

binary.

Review

Quoted 8 lines

Hence "pidof sleep" was empty and the test was failing.

---
 meta/lib/oeqa/runtime/cases/systemd.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
(quoted diff) +0 -0

b/meta/lib/oeqa/runtime/cases/systemd.py

meta/lib/oeqa/runtime/cases/systemd.py class SystemdServiceTests(SystemdTest): · line 145

Quoted 5 lines

index 5481e1d840..a2229848e7 100644

meta/lib/oeqa/runtime/cases/systemd.py +0 -0
@@ -145,7 +145,7 @@ class SystemdServiceTests(SystemdTest):
         Verify that call-stacks generated by systemd-coredump contain

symbolicated call-stacks,

meta/lib/oeqa/runtime/cases/systemd.py class SystemdServiceTests(SystemdTest): · line 146

Quoted 1 line
meta/lib/oeqa/runtime/cases/systemd.py +0 -0
         extracted from the minidebuginfo metadata (.gnu_debugdata elf

section).

meta/lib/oeqa/runtime/cases/systemd.py class SystemdServiceTests(SystemdTest): · line 148

Quoted 2 lines
meta/lib/oeqa/runtime/cases/systemd.py +0 -1
         """
-        t_thread = threading.Thread(target=self.target.run,

args=("ulimit -c unlimited && sleep 1000",))

meta/lib/oeqa/runtime/cases/systemd.py class SystemdServiceTests(SystemdTest): · line 148

Quoted 1 line
meta/lib/oeqa/runtime/cases/systemd.py +1 -0
+        t_thread = threading.Thread(target=self.target.run,

args=("ulimit -c unlimited && /usr/bin/sleep 1000",))

meta/lib/oeqa/runtime/cases/systemd.py class SystemdServiceTests(SystemdTest): · line 150

Quoted 2 lines
meta/lib/oeqa/runtime/cases/systemd.py +0 -0
         t_thread.start()
         time.sleep(1)
Quoted 10 lines

I get a bit nervous about hardcoding paths like "/usr/bin", particularly in the context of some distros liking to move things around. We may want to put a shutil.which("sleep") call in instead (and an import shutil if needed)?

Cheers,

Richard

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group. View/Reply Online (#200026):
https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.openembedded.org_g_openembedded-2Dcore_message_200026&d=DwIFaQ&c=ncDTmphkJTvjIDPh0hpF_4vCHvabgGkICC2epckfdiw&r=AhkbNonVuMIGRfPx_Qj9TsRih1DULJTKUkSGa66m67E&m=qu0Injv6pHxfa5BgL3x29y11Al_3EpMdBOTkj8cWHLaFlxRoh9vy7WMU3VU2o6c-&s=EhlpO7VCVBhijcCMY2T86qar5UZZTaUidFKXQTXXBXA&e=
Mute This Topic:
https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.openembedded.org_mt_106365956_7048771&d=DwIFaQ&c=ncDTmphkJTvjIDPh0hpF_4vCHvabgGkICC2epckfdiw&r=AhkbNonVuMIGRfPx_Qj9TsRih1DULJTKUkSGa66m67E&m=qu0Injv6pHxfa5BgL3x29y11Al_3EpMdBOTkj8cWHLaFlxRoh9vy7WMU3VU2o6c-&s=v89AFdBy-BUex49TuXqDiEFkidJxP27psNJVdVXrgCM&e=
Group Owner: [email protected] Unsubscribe:
https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.openembedded.org_g_openembedded-2Dcore_unsub&d=DwIFaQ&c=ncDTmphkJTvjIDPh0hpF_4vCHvabgGkICC2epckfdiw&r=AhkbNonVuMIGRfPx_Qj9TsRih1DULJTKUkSGa66m67E&m=qu0Injv6pHxfa5BgL3x29y11Al_3EpMdBOTkj8cWHLaFlxRoh9vy7WMU3VU2o6c-&s=vCbKyJLrNTiyVgVTio-Tl9XdO6erBWJiypaEgcrkqX8&e=
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

#5 lore raw

It's not typically expected for contributors to know the ins and outs of the AB test matrix; just send the needed patches, and any fails will be reported back to you.

Alex

Quoted 62 lines · Etienne Cordonnier via lists.openembedded.org

Hi Richard,
thanks for merging the v2 of this patch. Is this test running in CI currently? I'm thinking I could send a PR to enable minidebuginfo in poky-altcfg, but is there a way I can verify that this test works in yocto's CI? (it works locally, but as I mentioned it would fail on a kernel with CONFIG_USER_NS disabled.)

Étienne

Quoted 45 lines · Etienne Cordonnier via lists.openembedded.org

Hi Richard,
I thought it was OK because systemd requires usrmerge, but I've sent a new version of the patch using "env sleep" instead of "/usr/bin/sleep" in order to be safe. I hope this also fixes the test failure which was happening on CI.
One reason this could be failing in CI is that this functionality currently requires user-namespaces enabled in the kernel (e.g. core-image-sato does have CONFIG_USER_NS, thus it works with this image). This is not really by design, and I'm working on changing this in the upstream systemd: https://github.com/systemd/systemd/pull/33043 . I can then backport this on scarthgap if the change is merged upstream.

Étienne

Quoted 34 lines · Richard Purdie
Quoted 15 lines · Etienne Cordonnier via lists.openembedded.org

From: Etienne Cordonnier <[email protected]>

The builtin command "sleep" was getting executed instead of the busybox binary. Hence "pidof sleep" was empty and the test was failing.

---
 meta/lib/oeqa/runtime/cases/systemd.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
meta/lib/oeqa/runtime/cases/systemd.py +0 -0
index 5481e1d840..a2229848e7 100644

meta/lib/oeqa/runtime/cases/systemd.py class SystemdServiceTests(SystemdTest): · line 150

Quoted 7 lines
meta/lib/oeqa/runtime/cases/systemd.py +1 -1
@@ -145,7 +145,7 @@ class SystemdServiceTests(SystemdTest):
         Verify that call-stacks generated by systemd-coredump contain symbolicated call-stacks,
         extracted from the minidebuginfo metadata (.gnu_debugdata elf section).
         """
-        t_thread = threading.Thread(target=self.target.run, args=("ulimit -c unlimited && sleep 1000",))
+        t_thread = threading.Thread(target=self.target.run, args=("ulimit -c unlimited && /usr/bin/sleep 1000",))
         t_thread.start()
         time.sleep(1)

I get a bit nervous about hardcoding paths like "/usr/bin", particularly in the context of some distros liking to move things around. We may want to put a shutil.which("sleep") call in instead (and an import shutil if needed)?

Cheers,

Richard

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#200246): https://lists.openembedded.org/g/openembedded-core/message/200246
Mute This Topic: https://lists.openembedded.org/mt/106365956/1686489
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

#6 lore raw
In reply to Etienne Cordonnier
Quoted 6 lines · Etienne Cordonnier

Hi Richard,
thanks for merging the v2 of this patch. Is this test running in CI currently? I'm thinking I could send a PR to enable minidebuginfo in poky-altcfg, but is there a way I can verify that this test works in yocto's CI? (it works locally, but as I mentioned it would fail on a kernel with CONFIG_USER_NS disabled.)

I'd have expected it to be tested in CI, yes.

You can see results here for example:

https://git.yoctoproject.org/yocto-testresults/tree/runtime/poky-altcfg/qemux86-64/core-image-sato-sdk/testresults.json

Cheers,

Richard

A message that is not in the archive (sent off-list or not yet received): [email protected]

#8 lore raw
Quoted 11 lines · Richard Purdie via lists.openembedded.org
Quoted 9 lines · Etienne Cordonnier

Hi Richard,
thanks for merging the v2 of this patch. Is this test running in CI currently? I'm thinking I could send a PR to enable minidebuginfo in
poky-altcfg, but is there a way I can verify that this test works in
yocto's CI? (it works locally, but as I mentioned it would fail on a
kernel with CONFIG_USER_NS disabled.)

Quoted 5 lines

I'd have expected it to be tested in CI, yes.

You can see results here for example:

https://git.yoctoproject.org/yocto-testresults/tree/runtime/poky-altcfg/qemux86-64/core-image-sato-sdk/testresults.json

Being more specific, there are minidebuginfo tests here:

https://git.yoctoproject.org/poky/tree/meta/lib/oeqa/selftest/cases/minidebuginfo.py

with test results here:

https://git.yoctoproject.org/yocto-testresults/tree/oeselftest/testresults.json

but that is a larger json file.

Cheers,

Richard

#9 lore raw

Thanks Richard, this is helpful.
This shows that test_systemd_coredump_minidebuginfo is skipped / not executed, so I'm quite sure the test is still failing in CI as you had reported in December. I've sent a patch to fix what I think is causing the CI issue, and another patch to enable the feature in poky-altcfg once the CI failure is fixed.

Étienne

Quoted 37 lines · Richard Purdie
Quoted 11 lines · Richard Purdie via lists.openembedded.org
Quoted 9 lines · Etienne Cordonnier

Hi Richard,
thanks for merging the v2 of this patch. Is this test running in CI currently? I'm thinking I could send a PR to enable minidebuginfo in
poky-altcfg, but is there a way I can verify that this test works in
yocto's CI? (it works locally, but as I mentioned it would fail on a
kernel with CONFIG_USER_NS disabled.)

Quoted 5 lines

I'd have expected it to be tested in CI, yes.

You can see results here for example:

https://urldefense.proofpoint.com/v2/url?u=https-3A__git.yoctoproject.org_yocto-2Dtestresults_tree_runtime_poky-2Daltcfg_qemux86-2D64_core-2Dimage-2Dsato-2Dsdk_testresults.json&d=DwIFaQ&c=ncDTmphkJTvjIDPh0hpF_4vCHvabgGkICC2epckfdiw&r=AhkbNonVuMIGRfPx_Qj9TsRih1DULJTKUkSGa66m67E&m=VT5mIgXAiw0SxyBc46dQyS_ECUHBzZ6-efmyplXXwC7bFU6wHrv6BRxih0bgXf-O&s=fxG471cE71Up_h9_5zn5AMCspSvX9vZwR-_3lGUEQxc&e=

Being more specific, there are minidebuginfo tests here:

https://urldefense.proofpoint.com/v2/url?u=https-3A__git.yoctoproject.org_poky_tree_meta_lib_oeqa_selftest_cases_minidebuginfo.py&d=DwIFaQ&c=ncDTmphkJTvjIDPh0hpF_4vCHvabgGkICC2epckfdiw&r=AhkbNonVuMIGRfPx_Qj9TsRih1DULJTKUkSGa66m67E&m=VT5mIgXAiw0SxyBc46dQyS_ECUHBzZ6-efmyplXXwC7bFU6wHrv6BRxih0bgXf-O&s=rBKJ_kiFzQzdclkgzStD5LO90KUAh_xjZmZzYCmy4Mc&e=

with test results here:

https://urldefense.proofpoint.com/v2/url?u=https-3A__git.yoctoproject.org_yocto-2Dtestresults_tree_oeselftest_testresults.json&d=DwIFaQ&c=ncDTmphkJTvjIDPh0hpF_4vCHvabgGkICC2epckfdiw&r=AhkbNonVuMIGRfPx_Qj9TsRih1DULJTKUkSGa66m67E&m=VT5mIgXAiw0SxyBc46dQyS_ECUHBzZ6-efmyplXXwC7bFU6wHrv6BRxih0bgXf-O&s=kvjEIBBwJ6pnlJEHyyAg_Fzemajwma8z2SVR_mHja0g&e=

but that is a larger json file.

Cheers,

Richard

#10 lore raw
Quoted 6 lines · Etienne Cordonnier

Thanks Richard, this is helpful.
This shows that test_systemd_coredump_minidebuginfo is skipped / not executed, so I'm quite sure the test is still failing in CI as you had reported in December. I've sent a patch to fix what I think is causing the CI issue, and another patch to enable the feature in poky-altcfg once the CI failure is fixed.

I think it is being tested. There is this test:

https://git.yoctoproject.org/poky/tree/meta/lib/oeqa/selftest/cases/minidebuginfo.py

which is run with oe-selftest -r minidebuginfo

which generates results here:

https://git.yoctoproject.org/yocto-testresults/tree/oeselftest/testresults.json

e.g.:

            "minidebuginfo.Minidebuginfo.test_minidebuginfo": {
                "duration": 1073.8962650299072,
                "status": "PASSED"
            },

            "minidebuginfo.Minidebuginfo.test_minidebuginfo": {
                "duration": 764.4718866348267,
                "status": "PASSED"
            },

so I think it is being tested in each oe-selftest run (hence the multiple results)?

Cheers,

Richard

#11 lore raw

I'm talking specifically about test_systemd_coredump_minidebuginfo (the minidebuginfo test which I wrote in meta/lib/oeqa/runtime/cases/systemd.py, not the tests in meta/lib/oeqa/selftest/cases/minidebuginfo.py).

Étienne

Quoted 41 lines · Richard Purdie
Quoted 6 lines · Etienne Cordonnier

Thanks Richard, this is helpful.
This shows that test_systemd_coredump_minidebuginfo is skipped / not executed, so I'm quite sure the test is still failing in CI as you had reported in December. I've sent a patch to fix what I think is causing the CI issue, and another patch to enable the feature in poky-altcfg once the CI failure is fixed.

I think it is being tested. There is this test:

https://urldefense.proofpoint.com/v2/url?u=https-3A__git.yoctoproject.org_poky_tree_meta_lib_oeqa_selftest_cases_minidebuginfo.py&d=DwIFaQ&c=ncDTmphkJTvjIDPh0hpF_4vCHvabgGkICC2epckfdiw&r=AhkbNonVuMIGRfPx_Qj9TsRih1DULJTKUkSGa66m67E&m=3LHd-W_iTYnP8P40BAL8-r4JdqCery51l_QUwjUomCUCwtluWLU6LhJf_b7DoH6Y&s=i7GSqs0NbXHgfofD0OVBT7Vn7ofzIzlRvW91KAVDDBI&e=

which is run with oe-selftest -r minidebuginfo

which generates results here:

https://urldefense.proofpoint.com/v2/url?u=https-3A__git.yoctoproject.org_yocto-2Dtestresults_tree_oeselftest_testresults.json&d=DwIFaQ&c=ncDTmphkJTvjIDPh0hpF_4vCHvabgGkICC2epckfdiw&r=AhkbNonVuMIGRfPx_Qj9TsRih1DULJTKUkSGa66m67E&m=3LHd-W_iTYnP8P40BAL8-r4JdqCery51l_QUwjUomCUCwtluWLU6LhJf_b7DoH6Y&s=wGnnzbHed5xpJ4j1BjBdN0lpjhd5E3d1MrDS59kZVT0&e=

e.g.:

            "minidebuginfo.Minidebuginfo.test_minidebuginfo": {
                "duration": 1073.8962650299072,
                "status": "PASSED"
            },

            "minidebuginfo.Minidebuginfo.test_minidebuginfo": {
                "duration": 764.4718866348267,
                "status": "PASSED"
            },

so I think it is being tested in each oe-selftest run (hence the multiple results)?

Cheers,

Richard

#12 lore raw
Quoted 4 lines · Etienne Cordonnier

I'm talking specifically about test_systemd_coredump_minidebuginfo (the minidebuginfo test which I wrote
in meta/lib/oeqa/runtime/cases/systemd.py, not the tests in meta/lib/oeqa/selftest/cases/minidebuginfo.py).

Ok, sorry, I'm struggling to remember all the right context!

I get nervous about blanket distro changes as whilst it adds test coverage to minidebuginfo, it also removes test coverage from other things (the non-mini systemd case).

How are we all runtime coverage to the selftest in question, so it boots and runs the systemd selftests?

Specifically, we'd add config along the lines of:

DISTRO_FEATURES:append = " systemd"
TEST_SUITES = "ping ssh systemd"

and then add a:

bitbake('-c testimage core-image-minimal')

to that testcase in selftest, which would then ensure we do run the right tests?

If we want to test both the systemd and non-systemd cases we could add a second test to selftest instead which covers this?

Cheers,

Richard

#13 lore raw

Hi Richard,
I'm assuming you meant "How about we add runtime coverage to the selftest" (it got autocorrected to "How are we all runtime coverage to the selftest").

I'll have a look at adding this to selftest to ensure that the test is executed, however for context you had suggested another solution in
https://lists.openembedded.org/g/openembedded-core/topic/102891486#msg191495
, which is why I hadn't done it initially (it is not a problem at all, I can change the implementation, I'm just making sure you still have the context since the previous conversation was 6 months ago):

"I was looking at your patches and thinking we might want to enable minidebuginfo in poky or perhaps OE-Core by default too."

"We could assume we'll end up enabling this by default and if we do that, we could add a new runtime tests if it is configured knowing it will be by default in some of our builds."

"and include running things under runqemu to test target pieces but a runtime test may be enough for this feature."

Quoted 35 lines · Richard Purdie
Quoted 4 lines · Etienne Cordonnier

I'm talking specifically about test_systemd_coredump_minidebuginfo (the minidebuginfo test which I wrote
in meta/lib/oeqa/runtime/cases/systemd.py, not the tests in meta/lib/oeqa/selftest/cases/minidebuginfo.py).

Ok, sorry, I'm struggling to remember all the right context!

I get nervous about blanket distro changes as whilst it adds test coverage to minidebuginfo, it also removes test coverage from other things (the non-mini systemd case).

How are we all runtime coverage to the selftest in question, so it boots and runs the systemd selftests?

Specifically, we'd add config along the lines of:

DISTRO_FEATURES:append = " systemd"
TEST_SUITES = "ping ssh systemd"

and then add a:

bitbake('-c testimage core-image-minimal')

to that testcase in selftest, which would then ensure we do run the right tests?

If we want to test both the systemd and non-systemd cases we could add a second test to selftest instead which covers this?

Cheers,

Richard