selftests clone3 cap_checkpoint_restore fails on all devices.
Reported-by: Linux Kernel Functional Testing lkft@linaro.org
[ 97.198602] audit: type=1701 audit(1651167820.383:12): auid=4294967295 uid=0 gid=0 ses=4294967295 pid=1732 comm="clone3_cap_chec" exe="/opt/kselftests/default-in-kernel/clone3/clone3_cap_checkpoint_restore" sig=6 res=1
# selftests: clone3: clone3_cap_checkpoint_restore # TAP version 13 # 1..1 # # Starting 1 tests from 1 test cases. # # RUN global.clone3_cap_checkpoint_restore ... # # clone3_cap_checkpoint_restore.c:155:clone3_cap_checkpoint_restore:Child has PID 1733 # # clone3() syscall supported # cap_set_proc: Operation not permitted # # clone3_cap_checkpoint_restore.c:164:clone3_cap_checkpoint_restore:Expected set_capability() (-1) == 0 (0) # # clone3_cap_checkpoint_restore.c:165:clone3_cap_checkpoint_restore:Could not set CAP_CHECKPOINT_RESTORE # # clone3_cap_checkpoint_restore: Test terminated by assertion # # FAIL global.clone3_cap_checkpoint_restore # not ok 1 global.clone3_cap_checkpoint_restore # # FAILED: 0 / 1 tests passed. # # Totals: pass:0 fail:1 xfail:0 xpass:0 skip:0 error:0 not ok 4 selftests: clone3: clone3_cap_checkpoint_restore # exit=1
Test details links, https://lkft.validation.linaro.org/scheduler/job/5812724#L2074
Test results comparison link, https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20221107/tes...
https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20221107/tes...
https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20221107/tes...
metadata: git_ref: master git_repo: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next git_sha: d8e87774068af213ab5b058b1b114dc397b577aa git_describe: next-20221107 kernel_version: 6.1.0-rc3 kernel-config: https://builds.tuxbuild.com/2HChVOSNxUpvfAfLTnaK7zyTdUi/config build-url: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next/-/pipelines/687092361 artifact-location: https://builds.tuxbuild.com/2HChVOSNxUpvfAfLTnaK7zyTdUi toolchain: gcc-11
-- Linaro LKFT https://lkft.linaro.org
On Tue, Nov 08, 2022 at 03:59:56PM +0530, Naresh Kamboju wrote:
selftests clone3 cap_checkpoint_restore fails on all devices.
Reported-by: Linux Kernel Functional Testing lkft@linaro.org
[ 97.198602] audit: type=1701 audit(1651167820.383:12): auid=4294967295 uid=0 gid=0 ses=4294967295 pid=1732 comm="clone3_cap_chec" exe="/opt/kselftests/default-in-kernel/clone3/clone3_cap_checkpoint_restore" sig=6 res=1
# selftests: clone3: clone3_cap_checkpoint_restore # TAP version 13 # 1..1 # # Starting 1 tests from 1 test cases. # # RUN global.clone3_cap_checkpoint_restore ... # # clone3_cap_checkpoint_restore.c:155:clone3_cap_checkpoint_restore:Child has PID 1733 # # clone3() syscall supported # cap_set_proc: Operation not permitted # # clone3_cap_checkpoint_restore.c:164:clone3_cap_checkpoint_restore:Expected set_capability() (-1) == 0 (0) # # clone3_cap_checkpoint_restore.c:165:clone3_cap_checkpoint_restore:Could not set CAP_CHECKPOINT_RESTORE # # clone3_cap_checkpoint_restore: Test terminated by assertion # # FAIL global.clone3_cap_checkpoint_restore # not ok 1 global.clone3_cap_checkpoint_restore # # FAILED: 0 / 1 tests passed. # # Totals: pass:0 fail:1 xfail:0 xpass:0 skip:0 error:0 not ok 4 selftests: clone3: clone3_cap_checkpoint_restore # exit=1
Test details links, https://lkft.validation.linaro.org/scheduler/job/5812724#L2074
Test results comparison link, https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20221107/tes...
https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20221107/tes...
https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20221107/tes...
Similar question to the other report. Is this something that has happened before or are we starting to see this failures just now?
On Tue, 8 Nov 2022 at 16:58, Christian Brauner brauner@kernel.org wrote:
On Tue, Nov 08, 2022 at 03:59:56PM +0530, Naresh Kamboju wrote:
selftests clone3 cap_checkpoint_restore fails on all devices.
Reported-by: Linux Kernel Functional Testing lkft@linaro.org
[ 97.198602] audit: type=1701 audit(1651167820.383:12): auid=4294967295 uid=0 gid=0 ses=4294967295 pid=1732 comm="clone3_cap_chec" exe="/opt/kselftests/default-in-kernel/clone3/clone3_cap_checkpoint_restore" sig=6 res=1
# selftests: clone3: clone3_cap_checkpoint_restore # TAP version 13 # 1..1 # # Starting 1 tests from 1 test cases. # # RUN global.clone3_cap_checkpoint_restore ... # # clone3_cap_checkpoint_restore.c:155:clone3_cap_checkpoint_restore:Child has PID 1733 # # clone3() syscall supported # cap_set_proc: Operation not permitted # # clone3_cap_checkpoint_restore.c:164:clone3_cap_checkpoint_restore:Expected set_capability() (-1) == 0 (0) # # clone3_cap_checkpoint_restore.c:165:clone3_cap_checkpoint_restore:Could not set CAP_CHECKPOINT_RESTORE # # clone3_cap_checkpoint_restore: Test terminated by assertion # # FAIL global.clone3_cap_checkpoint_restore # not ok 1 global.clone3_cap_checkpoint_restore # # FAILED: 0 / 1 tests passed. # # Totals: pass:0 fail:1 xfail:0 xpass:0 skip:0 error:0 not ok 4 selftests: clone3: clone3_cap_checkpoint_restore # exit=1
Test details links, https://lkft.validation.linaro.org/scheduler/job/5812724#L2074
Test results comparison link, https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20221107/tes...
https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20221107/tes...
https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20221107/tes...
Similar question to the other report. Is this something that has happened before or are we starting to see this failures just now?
These failures have been happening for a long time.
FYI, LKFT email is configured to send out email to Linux next mailing list with daily kselftest results / regressions / pass / fail / xfails / skips.
Please refer this link for our reports, https://lore.kernel.org/linux-next/?q=lkft+kselftest+for+
We are improving our process to report failures and regressions on mailing lists.
- Naresh
I presume that test fails because of the difference between definition of struct _cap_struct (struct libcap in test). Commit aca076443591 (Make cap_t operations thread safe.) introduces mutex field at the beginning of the structure. Test uses own definition to set CAP_CHECKPOINT_RESTORE. As data member of these two versions of structures is at the different offsets, CAP_CHECKPOINT_RESTORE is lost and that's why cap_set_proc returns EPERM.
С уважением, Ленар Ханнанов Программист Отдел разработки АО "ИнфоТеКС" в г. Санкт-Петербург 190000, г. Санкт-Петербург, ул. Артиллерийская, д. 1, литер А, БЦ «Европа Хаус» T: +7 495 737-61-92 ( доб. 7196) Ф: +7 495 737-72-78
Lenar.Khannanov@infotecs.ru www.infotecs.ru