diff options
| author | Steven Rostedt <rostedt@goodmis.org> | 2026-09-02 09:55:01 -0400 |
|---|---|---|
| committer | Steven Rostedt <rostedt@goodmis.org> | 2026-09-02 11:08:21 -0400 |
| commit | 4617721c502b2ddaa4e324e86da4997edf738fa5 (patch) | |
| tree | 350aa83b47da0b574a814795392b20b0bed83cbc /tools/tracing/rtla/README.txt | |
| download | linux-stable-4617721c502b2ddaa4e324e86da4997edf738fa5.tar.gz linux-stable-4617721c502b2ddaa4e324e86da4997edf738fa5.zip | |
ftrace: Synchronize the initialization of ftrace_opsgrafted
There's some internal state that ftrace_ops needs to have set, but since
it can be declared outside of the ftrace.c code, it calls
ftrace_ops_init() on the ops in every global function. The issue is that
if two tasks call it on the same ops at the same time it is possible to
have the initialization of one corrupt the initialization of the other
call.
Create a ops_mutex to use to synchronize every initialization of the
ftrace_ops. The mutex is taken within checking the ftrace_ops flag that
states it was initializied but the flag is checked again after the mutex
has been taken. Checking first outside the mutex allows it to shortcut
having to take the mutex. But then the check needs to be done again after
the mute is taken in case of races.
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20260902095501.6b59af20@gandalf.local.home
Fixes: f04f24fb7e48d ("ftrace, kprobes: Fix a deadlock on ftrace_regex_lock")
Reported-by: sashiko-bot@kernel.org
Close: https://lore.kernel.org/all/20260829025528.49A831F000E9@smtp.kernel.org/
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Diffstat (limited to 'tools/tracing/rtla/README.txt')
| -rw-r--r-- | tools/tracing/rtla/README.txt | 75 |
1 files changed, 75 insertions, 0 deletions
diff --git a/tools/tracing/rtla/README.txt b/tools/tracing/rtla/README.txt new file mode 100644 index 000000000..13b4a798b --- /dev/null +++ b/tools/tracing/rtla/README.txt @@ -0,0 +1,75 @@ +RTLA: Real-Time Linux Analysis tools + +The rtla meta-tool includes a set of commands that aims to analyze +the real-time properties of Linux. Instead of testing Linux as a black box, +rtla leverages kernel tracing capabilities to provide precise information +about the properties and root causes of unexpected results. + +Installing RTLA + +RTLA depends on the following libraries and tools: + + - libtracefs + - libtraceevent + - libcpupower (optional, for --deepest-idle-state) + - libcheck (optional, for unit tests) + +For BPF sample collection support, the following extra dependencies are +required: + + - libbpf 1.0.0 or later + - bpftool with skeleton support + - clang with BPF CO-RE support + +It also depends on python3-docutils to compile man pages. + +For development, we suggest the following steps for compiling rtla: + + $ git clone git://git.kernel.org/pub/scm/libs/libtrace/libtraceevent.git + $ cd libtraceevent/ + $ make + $ sudo make install + $ cd .. + $ git clone git://git.kernel.org/pub/scm/libs/libtrace/libtracefs.git + $ cd libtracefs/ + $ make + $ sudo make install + $ cd .. + $ cd $libcpupower_src + $ make + $ sudo make install + $ cd $rtla_src + $ make + $ sudo make install + +Running tests + +RTLA has two test suites: a runtime test suite and a unit test suite. + +The runtime test suite is available as "make check" (root required) and has +the following dependencies, in addition to RTLA build dependencies: + +- Perl +- Test::Harness (libtest-harness-perl on Debian/Ubuntu, perl-Test-Harness on Fedora/RHEL) +- bash +- coreutils +- ldd +- util-linux +- procps(-ng) +- bpftool (if rtla is built against libbpf) + +as well as the following required system configuration: + +- CONFIG_OSNOISE_TRACER=y +- CONFIG_TIMERLAT_TRACER=y +- tracefs mounted and readable at /sys/kernel/tracing + +The unit test suite is available as "make unit-tests" and has the following +dependencies: + +- libcheck + +Unlike the runtime test suite, root is not required to run unit tests, nor is +a tracefs/osnoise/timerlat-capable kernel required. + +For further information, please refer to the rtla man page. |
