Petr Vorel
Petr Vorel
Support for reboot in LTP tests would be useful for certain tests. Based on [original proposal from Cyril](https://lists.linux.it/pipermail/ltp/2020-August/018658.html). Adding `reboots_machine` flag to the `tst_test structure`, which will inform the test...
We should TCONF in `tst_net.sh` when `Operation not supported`. Testing `ping02.sh -6`: ``` ping02 1 TINFO: initialize 'lhost' 'ltp_ns_veth2' interface ping02 1 TINFO: add local addr 10.0.0.2/24 ping02 1 TINFO:...
Based on report from Jiri Bohac: ipv6ready/USGv6 testsuite often find bug that IPsec broken by two types of MTU related bugs with wrongly calculated packet size: * **too big packet...
preadv203 fails when storage is fast: ``` preadv203.c:145: TINFO: Number of full_reads 2920600, short reads 0, zero len reads 0, EAGAIN(s) 0 preadv203.c:180: TINFO: Number of writes 3535680 preadv203.c:225: TFAIL:...
Although we have basic `sched_get_priority_{max,min}()` tests, `SCHED_RESET_ON_FORK` deserves it's own test to detect children created by fork(2) do not inherit privileged scheduling policies. https://man7.org/linux/man-pages/man2/sched_setscheduler.2.html https://man7.org/linux/man-pages/man7/sched.7.html https://lwn.net/Articles/335203/ https://lore.kernel.org/ltp/Ye615gN0Yv1%2FWskD@pevik/
Folks from DPDK have very nice patchwork extensions: on [1], they have * Their `Success / Warning / Fail` column contains colored numbers (see S/W/F in [1]) * Patch view...
LTP requires special user and group to be presented (check [IDcheck.sh](https://github.com/linux-test-project/ltp/blob/master/IDcheck.sh) script). Instead of hard-wired them in tests it'd be great to have `tst_user()` and `tst_group()` which would allow via...
There are policy examples in testcases/kernel/security/integrity/ima/datafiles for ima_keys, ima_kexec and ima_policy. It'd be good to automatically load it for tests which don't load any policy to avoid `TCONF`. Things are...
We have bunch of tests in `lib/newlib_tests/` (both C and shell), which we want to run. The best would be to have `make test` target + simple scripts which handles...
Write a regression test once [making unix_bind() undo mknod on failure](https://lore.kernel.org/netdev/[email protected]/) patchset is merged. For more info see workaround https://github.com/linux-test-project/ltp/commit/b003e0f08fa40a4533b234d311f89e32338088e6 for [`bind03`](https://github.com/linux-test-project/ltp/blob/master/testcases/kernel/syscalls/bind/bind03.c) ([original patch](https://patchwork.ozlabs.org/project/ltp/patch/[email protected]/)).