Re: [PATCH v1 03/11] sched: Add sched tracepoints for RV task model
From: Peter Zijlstra <peterz@infradead.org>
Date: 2025-02-11 14:37:11
Also in:
lkml
From: Peter Zijlstra <peterz@infradead.org>
Date: 2025-02-11 14:37:11
Also in:
lkml
On Tue, Feb 11, 2025 at 01:54:44PM +0100, Gabriele Monaco wrote:
quoted
quoted
+void do_trace_set_current_state(int state_value) +{ + trace_sched_set_state_tp(current, current->__state, state_value);Should this be: __do_trace_sched_set_state_tp() ?Mmh, you mean avoiding the static_branch_unlikely in the helper function, since it's supposed to be used before calling it?
Yep, seems pointless to repeat that.
quoted
quoted
+} +EXPORT_SYMBOL(do_trace_set_current_state);_GPLI'm absolutely not against this change but, out of curiosity, would that imply non-GPL modules are not going to be able to sleep going forward? At least not using this pattern.
Bah, you're right. Killing non-GPL modules seems like a worthy goal though, but perhaps another day ;-)