On Wed, 18 Jun 2025 16:02:47 +0200
Peter Zijlstra [off-list ref] wrote:

On Tue, Jun 10, 2025 at 08:54:25PM -0400, Steven Rostedt wrote:

quoted

+/**
+ * unwind_deferred_trace - Produce a user stacktrace in faultable context
+ * @trace: The descriptor that will store the user stacktrace
+ *
+ * This must be called in a known faultable context (usually when entering
+ * or exiting user space). Depending on the available implementations
+ * the @trace will be loaded with the addresses of the user space stacktrace
+ * if it can be found.  

I am confused -- why would we ever want to call this on exiting
user-space, or rather kernel entry?

I thought the whole point was to request a user trace while in-kernel,
and defer that to return-to-user.

This code was broken out of the unwind deferred trace to be more stand
alone. Actually, it should be renamed to unwind_faultable_trace() or
something to denote that it must be called from a faultable context.

When Josh made perf use the task_work directly, it used this function to do
the trace as it handled the deferring.

Note, a request from the gcc folks is to add a system call that gives the
user space application a backtrace from its current location. This can be
handy for debugging as it would be similar to how we use dump_stack().

This function would be used for that.

-- Steve