On Wed, 18 Jun 2025 16:01:11 +0200 Peter Zijlstra <pet...@infradead.org> wrote:
> On Tue, Jun 10, 2025 at 08:54:25PM -0400, Steven Rostedt wrote: > > > +#define UNWIND_MAX_ENTRIES 512 > > The reason this is 512 is so that you end up with a whole page below? Possibly. We could probably even make that configurable. Perhaps just use sysctl_perf_event_max_contexts_per_stack ? Josh, any comments about why you picked this number? -- Steve > > > +int unwind_deferred_trace(struct unwind_stacktrace *trace) > > +{ > > + struct unwind_task_info *info = ¤t->unwind_info; > > + > > + /* Should always be called from faultable context */ > > + might_fault(); > > + > > + if (current->flags & PF_EXITING) > > + return -EINVAL; > > + > > + if (!info->entries) { > > + info->entries = kmalloc_array(UNWIND_MAX_ENTRIES, sizeof(long), > > + GFP_KERNEL); >