The test script adds a tracing filter: echo '*lock*' > stack_trace_filter
..that takes a few minutes to apply and during that time we see the RCU and/or softlockup messages. This is kind of expected since we're blocked in kernel space updating ftrace points across the kernel which is slow and expensive on the RISC-V unmatched/unleashed H/W. One remedy is to do the following: echo 0 > /proc/sys/kernel/soft_watchdog echo 1 > /sys/module/rcupdate/parameters/rcu_cpu_stall_suppress .. run test echo 0 > /sys/module/rcupdate/parameters/rcu_cpu_stall_suppress echo 1 > /proc/sys/kernel/soft_watchdog The long term fix is to make the ftrace stack_trace_filter interface more yieldy -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux-riscv in Ubuntu. https://bugs.launchpad.net/bugs/1940972 Title: v5.13 kernel ftrace selftest fails on riscv64 unmatched Status in ubuntu-kernel-tests: New Status in linux-riscv package in Ubuntu: New Bug description: $ sudo make -C tools/testing/selftests TARGETS=ftrace run_tests ... # [32] ftrace - Max stack tracer [FAIL] ... Testing using 5.13.0-1002.2 the full ftrace suite takes just over 2 hours on unmatched. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu-kernel-tests/+bug/1940972/+subscriptions -- Mailing list: https://launchpad.net/~kernel-packages Post to : kernel-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~kernel-packages More help : https://help.launchpad.net/ListHelp
