Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1528162

Re: [PATCH] Add support for disabling Intel PT trace in ftrace

Show all headers | View raw

Andi Kleen <andi@firstfloor.org> writes:

> +/*
> + * Disable the PT trace for debugging purposes.
> + */
> +void pt_disable(void)
> +{
> +	u64 val;
> +
> +	if (!boot_cpu_has(X86_FEATURE_INTEL_PT))
> +		return;
> +
> +	rdmsrl_safe(MSR_IA32_RTIT_CTL, &val);
> +	val &= ~RTIT_CTL_TRACEEN;
> +	wrmsrl_safe(MSR_IA32_RTIT_CTL, val);
> +}
> +EXPORT_SYMBOL(pt_disable);

This will create unexplainable gaps in the trace, at least we should
output RECORD_AUX when this happens, maybe add a flag for "had to stop
the trace for reasons external to perf".

Also, I can't tell if this is called from an atomic context.

But I'd suggest something more generic like perf_pmu_off($pmu):
 - we already have the code to stop the output;
 - this won't be a driver-specific api then;
 - this will be reflected in the event hw state;
 - it will also go through the driver's callbacks, so its internal
 states will actually match the reality;
 - will work equally well for intel_bts or the ARM/Coresight tracers.

Regards,
--
Alex

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH] Add support for disabling Intel PT trace in ftrace Andi Kleen <andi@firstfloor.org> - 2016-11-18 18:00 +0100
  Re: [PATCH] Add support for disabling Intel PT trace in ftrace Steven Rostedt <rostedt@goodmis.org> - 2016-11-22 23:40 +0100
    Re: [PATCH] Add support for disabling Intel PT trace in ftrace Thomas Gleixner <tglx@linutronix.de> - 2016-11-23 11:30 +0100
  Re: [PATCH] Add support for disabling Intel PT trace in ftrace Alexander Shishkin <alexander.shishkin@intel.com> - 2016-11-23 08:50 +0100
    Re: [PATCH] Add support for disabling Intel PT trace in ftrace Andi Kleen <andi@firstfloor.org> - 2016-11-23 18:00 +0100

csiph-web