On Wed, Apr 03, 2024 at 10:13:52AM -0400, Steven Rostedt wrote:
> On Fri, 29 Mar 2024 14:40:55 -0400
> Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:
>
> > > +static vm_fault_t tracing_buffers_mmap_fault(struct vm_fault *vmf)
> > > +{
> > > + return VM_FAULT_SIGBUS;
> > > +}
> >
> > If this is all it does, I don't believe it's needed.
> >
> > > +
> > > +#ifdef CONFIG_TRACER_MAX_TRACE
> > > +static int get_snapshot_map(struct trace_array *tr)
> > > +{
> > > + int err = 0;
> > > +
> > > + /*
> > > + * Called with mmap_lock held. lockdep would be unhappy if we would now
> > > + * take trace_types_lock. Instead use the specific
> > > + * snapshot_trigger_lock.
> > > + */
> > > + spin_lock(&tr->snapshot_trigger_lock);
> > > +
> > > + if (tr->snapshot || tr->mapped == UINT_MAX)
> > > + err = -EBUSY;
> > > + else
> > > + tr->mapped++;
> > > +
> > > + spin_unlock(&tr->snapshot_trigger_lock);
> > > +
> > > + /* Wait for update_max_tr() to observe iter->tr->mapped */
> > > + if (tr->mapped == 1)
> > > + synchronize_rcu();
> > > +
> > > + return err;
> > > +
> > > +}
> > > +static void put_snapshot_map(struct trace_array *tr)
> > > +{
> > > + spin_lock(&tr->snapshot_trigger_lock);
> > > + if (!WARN_ON(!tr->mapped))
> > > + tr->mapped--;
> > > + spin_unlock(&tr->snapshot_trigger_lock);
> > > +}
> > > +#else
> > > +static inline int get_snapshot_map(struct trace_array *tr) { return 0; }
> > > +static inline void put_snapshot_map(struct trace_array *tr) { }
> > > +#endif
> > > +
> > > +static void tracing_buffers_mmap_close(struct vm_area_struct *vma)
> > > +{
> > > + struct ftrace_buffer_info *info = vma->vm_file->private_data;
> > > + struct trace_iterator *iter = &info->iter;
> > > +
> > > + WARN_ON(ring_buffer_unmap(iter->array_buffer->buffer, iter->cpu_file));
> > > + put_snapshot_map(iter->tr);
> > > +}
> > > +
> > > +static void tracing_buffers_mmap_open(struct vm_area_struct *vma) { }
> >
> > Same for the open.
> >
> >
> > > +
> > > +static const struct vm_operations_struct tracing_buffers_vmops = {
> > > + .open = tracing_buffers_mmap_open,
> > > + .close = tracing_buffers_mmap_close,
> > > + .fault = tracing_buffers_mmap_fault,
> > > +};
> >
> > I replaced this with:
> >
> > static const struct vm_operations_struct tracing_buffers_vmops = {
> > .close = tracing_buffers_mmap_close,
> > };
> >
> > And it appears to work just fine. The mm code handles the NULL cases for
> > .open and .fault.
> >
> > Is there any reason to do something different than the mm defaults?
No other reason here than my own ignorance. I will remove.
>
> Hi Vincent,
>
> Do you plan on sending out a v20 series?
Of course, let me spin that this week! Got also few typos to fix in the doc and
I believe an include missing for riscv.
>
> -- Steve
