Re: [PATCH 6/6] perf/x86/intel/pt: Stop tracing across system suspend

[Adrian Hunter <adrian.hunter@intel.com>]

On 02/11/2026 10:59, sashiko-bot@kernel.org wrote:
> Can pt_syscore_resume() prematurely restart an Intel PT event that has been
> throttled by perf_event_throttle(), leading to a duplicate pt_event_start()
> call when the event is unthrottled on the next timer tick?


Yes, the duplicate start can happen

> Also, if pt_syscore_resume() skips calling pt_event_start() when the event is
> throttled, wouldn't clearing pt_suspended_event in pt_syscore_resume() before
> perf_event_unthrottle() later calls pt_event_start() cause a throttled
> snapshot event to read reset PT MSRs in pt_read_offset() and take the
> pt_config_enable() path without reprogramming the output MSRs?

Right, so keep the restart and make the redundant start a no-op
instead.  An event with both PERF_HES_UPTODATE and PERF_HES_STOPPED
clear is already running.

 	if (!(hwc->state & PERF_HES_UPTODATE)) {
+		/* Already running, e.g. restarted by pt_syscore_resume() */
+		if (!(hwc->state & PERF_HES_STOPPED))
+			return;
	hwc->state &= ~PERF_HES_STOPPED;

One side effect remains. A throttled PT event restarted on resume
traces until the next tick while its sibling is still throttled.
That lasts at most one tick and is harmless.