- 24 Oct, 2018 40 commits
-
-
Masami Hiramatsu authored
Strip escape sequences from the stream to the ftracetest summary log file. Note that all test-case results are dumped raw in each file. Signed-off-by:
Masami Hiramatsu <mhiramat@kernel.org> Acked-by:
Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by:
Shuah Khan (Samsung OSG) <shuah@kernel.org>
-
Daniel Díaz authored
If test is being directly executed (with stdout opened on the terminal) and the terminal capabilities indicate enough colors, then use the existing scheme of green, red, and blue to show when tests pass, fail or end in a different way. When running the tests redirecting the stdout, for instance, to a file, then colors are not shown, thus producing a more readable output. Signed-off-by:
Daniel Díaz <daniel.diaz@linaro.org> Acked-by:
-
Peng Hao authored
Makefile contains -D_GNU_SOURCE. remove define "_GNU_SOURCE" in c files. Signed-off-by:
Peng Hao <peng.hao2@zte.com.cn> Signed-off-by:
-
Andrea Parri authored
Fixes the following warnings: dirty_log_test.c: In function ‘help’: dirty_log_test.c:216:9: warning: format ‘%lu’ expects argument of type ‘long unsigned int’, but argument 2 has type ‘int’ [-Wformat=] printf(" -i: specify iteration counts (default: %"PRIu64")\n", ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In file included from include/test_util.h:18:0, from dirty_log_test.c:16: /usr/include/inttypes.h:105:34: note: format string is defined here # define PRIu64 __PRI64_PREFIX "u" dirty_log_test.c:218:9: warning: format ‘%lu’ expects argument of type ‘long unsigned int’, but argument 2 has type ‘int’ [-Wformat=] printf(" -I: specify interval in ms (default: %"PRIu64" ms)\n", ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In file included from include/test_util.h:18:0, from dirty_log_test.c:16: /usr/include/inttypes.h:105:34: note: format string is defined here # define PRIu64 __PRI64_PREFIX "u" Signed-off-by:Andrea Parri <andrea.parri@amarulasolutions.com> Signed-off-by:
-
Steven Rostedt (VMware) authored
Now that ftracetest has over 80 tests, it is difficult to simply scroll up the console window to find the failed tests when it reports just two tests have failed. In order to make this stand out better, have the color of the word "PASS" be green, "FAIL" and "XFAIL" be red, and all other results be blue. This helps tremendously in quickly spotting the failed tests by just scrolling up the console window. Signed-off-by:
-
Colin Ian King authored
Trivial fix to spelling mistake in TEST_ASSERT message text Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
Daniel Díaz authored
When simply running `make' from the selftests top dir, this error shows up: cc -O2 -g -std=gnu99 -Wall -I../../../../usr/include/ -I/usr/include/libmount -I/usr/include/blkid -I/usr/include/uuid gpio-mockup-chardev.c ../../../gpio/gpio-utils.o -lmount -o gpio-mockup-chardev cc: error: ../../../gpio/gpio-utils.o: No such file or directory <builtin>: recipe for target 'gpio-mockup-chardev' failed make[1]: *** [gpio-mockup-chardev] Error 1 because the output directory is set to "selftests/gpio" and all binaries built from ../../../gpio/ end up there. In fact, they appear as, exempli gratia: * gpiogpio-event-mon * gpiogpio-hammer * gpioinclude/ * gpiolsgpio which is wrong, as it's missing a directory separator somewhere. This patch sets straight the output directory when building ../../../gpio/ so that binaries don't cross paths. Signed-off-by:
-
Fathi Boudra authored
This patch cleans up the Makefile by restructuring a couple of things, namely: 1) change explicit paths in targets for variables 2) substitute a variable (BINARIES) for another, part of the selftests build system (TEST_PROGS_EXTENDED) 3) proper cleaning up of the EXTRA objects The resulting Makefile is much more readable and manageable. Signed-off-by:
Fathi Boudra <fathi.boudra@linaro.org> Signed-off-by:
-
Shuah Khan (Samsung OSG) authored
When ioctls for WDIOC_SETOPTIONS (WDIOS_DISABLECARD or WDIOS_ENABLECARD), WDIOC_SETTIMEOUT, and WDIOC_SETPRETIMEOUT fail, the error path continues to handler watchdog timer until user terminates it. When ioctl returns error, it might not be safe to let the watchdog tick. The error could be due an unsupported ioctl command or some other error. Fix it to handle error paths as oneshot to stop the watchdog ticks and exit right away. Signed-off-by: -
Jerry Hoemann authored
Add command line arguments to call ioctl WDIOC_GETTIMEOUT, WDIOC_GETPRETIMEOUT and WDIOC_SETPRETIMEOUT. Signed-off-by:
Jerry Hoemann <jerry.hoemann@hpe.com> Signed-off-by:
-
Jerry Hoemann authored
Printf's say errno but print the string version of error. Make consistent. Signed-off-by:
-
Shuah Khan (Samsung OSG) authored
When /dev/watchdog open fails, watchdog exits with "watchdog not enabled" message. This is incorrect when open fails due to insufficient privilege. Fix message to clearly state the reason when open fails with EACCESS when a non-root user runs it. Signed-off-by: -
Masami Hiramatsu authored
Add a testcase for tracing_cpumask with function tracer. Signed-off-by:
-
Masami Hiramatsu authored
Add a testcase for wakeup_rt tracer. This requires chrt command to test. Signed-off-by:
-
Masami Hiramatsu authored
Add a testcase for wakeup tracer. Signed-off-by:
-
Masami Hiramatsu authored
Add a test case for stacktrace filter command for ftrace. Signed-off-by:
-
Masami Hiramatsu authored
Add a simple testcase for trace_pipe which can consume ringbuffer. Signed-off-by:
-
Masami Hiramatsu authored
Add a testcase for function filter on module. Signed-off-by:
-
Masami Hiramatsu authored
Add a testcase for max stack tracer, which checks basic max stack usage tracing and its filter feature. Signed-off-by:
-
Masami Hiramatsu authored
Add a testcase for function profiling per-cpu statistics interface. There is already func_profile.tc, but that is mainly focusing on the combination of function-profiler and function tracer. This testcase ensures trace_stat per-cpu function statistics is correctly updated. Signed-off-by:
-
Masami Hiramatsu authored
Add a testcase for changing ringbuffer size. This tests not only ringbuffer size but also tests the imbalance per-cpu buffer size change too. Signed-off-by:
-
Masami Hiramatsu authored
Add trace_printk sample module test. This requires to enable trace_printk.ko module for test. Signed-off-by:
-
Masami Hiramatsu authored
Add a testcase for kprobe-event with @symbol argument. Since @symbol needs to refer the kernel data symbol (linux_proc_banner), it requires CONFIG_KALLSYMS_ALL. Signed-off-by:
-
Masami Hiramatsu authored
Add a testcase for testing kprobe_profile interface which provides per-kprobe event hit/misshit counts. Signed-off-by:
-
Masami Hiramatsu authored
Add kprobe-event with $comm argument testcase to ftracetest. This not only checks syntax but also checks log file. Signed-off-by:
-
Masami Hiramatsu authored
Ensure the set_event_pid shows set pid list. Signed-off-by:
-
Masami Hiramatsu authored
Improve the kprobe-event with argument types testcase to test it with various bitsize. kprobe-event argument can be recorded in given types with various bitsize (8, 16, 32, 64), thus the type testcase should test the different bitsize too. Signed-off-by:
-
Masami Hiramatsu authored
Improve kretprobe testcase to check the log data correctness and ensure the event definition is corrctly including argument definition. Signed-off-by:
-
Masami Hiramatsu authored
Improve kprobe testcase to check the log data correctness and ensure the event definition is corrctly including argument definition. Signed-off-by:
-
Masami Hiramatsu authored
Improve kprobe events on module testcase to check module load/unload with disabled/enabled events. This also change the target module to trace_printk.ko, so it depends on CONFIG_SAMPLE_TRACE_PRINTK=m. Signed-off-by:
-
Masami Hiramatsu authored
Use raw loopback address instead of localhost, because "localhost" can depend on nsswitch and in some case we can not resolve the localhost. Signed-off-by:
-
Masami Hiramatsu authored
Fix a test case to make checkbashisms clean. Signed-off-by:
-
Masami Hiramatsu authored
Test $comm in kprobe-event argument syntax testcase only if it is supported on the kernel because $comm has been introduced 4.8 kernel. So on older stable kernel, it should be skipped. Signed-off-by:
-
Masami Hiramatsu authored
Since ftracetest framework calls initialize_ftrace() right before each test and after all tests, we don't need to init/cleanup ftrace for each test case. Just remove such unneeded init/cleanup code because it can increase logfile size. Signed-off-by:
-
Masami Hiramatsu authored
Cleanup ftrace by initialize_ftrace() after running all test cases. This means we also don't need cleanup ftrace on each test case, except for some special options. Signed-off-by:
-
Masami Hiramatsu authored
Add SPDX License Identifier line to template file so that someone who makes new testcase from the template does not forgot it. Signed-off-by:
-
Masami Hiramatsu authored
Clear pid filter, synthetic_events, snapshots, ftrace filter, and trace log in initialize_ftrace(), since those are used in test cases. Signed-off-by:
-
Masami Hiramatsu authored
Add a case number prefix to each logfile. This makes it easier to find which logfile is corresponding to which failure. Signed-off-by:
-
Masami Hiramatsu authored
Add --console hidden option for debug test cases. This option allows to put "sh" or something else when the test case hits a bug. For example, if you find a testcase which doesn't pass, you can insert sh for interactive debug as below ----- #!/bin/sh # description: sample test case good-command suspicious-wrong-command sh # <- add this for interactive debug ----- Signed-off-by:
-
Masami Hiramatsu authored
Add --stop-fail option for debugging the ftracetest. With this option, ftracetest stops right after a testcase fails instead of finish running all testcases. Signed-off-by:
-
