It is possible that some performance excursions happened before OS boot
or enable HWP interrupts. So clear MSR_HWP_STATUS bits when we enable
HWP interrupt. In this way a next excursion will results in a HWP
interrupt.
The status bits of MSR_HWP_STATUS must be cleared (0) by software so
that a new status condition change will cause the hardware to set the
bit again and issue the notification.
Fixes: 57577c996d73 ("cpufreq: intel_pstate: Process HWP Guaranteed change notification")
Signed-off-by: Srinivas Pandruvada <[email protected]>
---
drivers/cpufreq/intel_pstate.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c
index 1e6898dc76b6..54cf21896889 100644
--- a/drivers/cpufreq/intel_pstate.c
+++ b/drivers/cpufreq/intel_pstate.c
@@ -1645,6 +1645,7 @@ static void intel_pstate_enable_hwp_interrupt(struct cpudata *cpudata)
/* wrmsrl_on_cpu has to be outside spinlock as this can result in IPC */
wrmsrl_on_cpu(cpudata->cpu, MSR_HWP_INTERRUPT, 0x01);
+ wrmsrl_on_cpu(cpudata->cpu, MSR_HWP_STATUS, 0);
}
}
--
2.31.1
On Thu, Nov 4, 2021 at 11:22 AM Srinivas Pandruvada
<[email protected]> wrote:
>
> It is possible that some performance excursions happened before OS boot
> or enable HWP interrupts. So clear MSR_HWP_STATUS bits when we enable
> HWP interrupt. In this way a next excursion will results in a HWP
> interrupt.
>
> The status bits of MSR_HWP_STATUS must be cleared (0) by software so
> that a new status condition change will cause the hardware to set the
> bit again and issue the notification.
>
> Fixes: 57577c996d73 ("cpufreq: intel_pstate: Process HWP Guaranteed change notification")
> Signed-off-by: Srinivas Pandruvada <[email protected]>
> ---
> drivers/cpufreq/intel_pstate.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c
> index 1e6898dc76b6..54cf21896889 100644
> --- a/drivers/cpufreq/intel_pstate.c
> +++ b/drivers/cpufreq/intel_pstate.c
> @@ -1645,6 +1645,7 @@ static void intel_pstate_enable_hwp_interrupt(struct cpudata *cpudata)
>
> /* wrmsrl_on_cpu has to be outside spinlock as this can result in IPC */
> wrmsrl_on_cpu(cpudata->cpu, MSR_HWP_INTERRUPT, 0x01);
> + wrmsrl_on_cpu(cpudata->cpu, MSR_HWP_STATUS, 0);
> }
> }
>
> --
Applied as 5.16-rc material, thanks!