[PATCH v4 1/2] perf/imx_ddr: speed up overflow frequency of cycle

Frank Li frank.li at nxp.com
Thu Aug 10 11:43:03 PDT 2023



> -----Original Message-----
> From: Xu Yang <xu.yang_2 at nxp.com>
> Sent: Thursday, August 10, 2023 6:00 AM
> To: Frank Li <frank.li at nxp.com>
> Cc: will at kernel.org; mark.rutland at arm.com; shawnguo at kernel.org;
> s.hauer at pengutronix.de; kernel at pengutronix.de; dl-linux-imx <linux-
> imx at nxp.com>; linux-arm-kernel at lists.infradead.org
> Subject: [PATCH v4 1/2] perf/imx_ddr: speed up overflow frequency of cycle
> 
> For i.MX8MP, we cannot ensure that cycle counter overflow occurs at least
> 4 times as often as other events. Due to byte counters will count for any
> event configured, it will overflow more often. And if byte counters
> overflow that related counters would stop since they share the
> COUNTER_CNTL. We can speed up cycle counter overflow frequency by
> setting
> counter parameter (CP) field of cycle counter. In this way, we can avoid
> stop counting byte counters when interrupt didn't come and the byte
> counters can be fetched or updated from each cycle counter overflow
> interrupt.
> 
> Because we initialize CP filed to shorten counter0 overflow time, the cycle
> counter will start couting from a fixed/base value each time. We need to
> remove the base from the result too. Therefore, we could get precise result
> from cycle counter.
> 
> Signed-off-by: Xu Yang <xu.yang_2 at nxp.com>

Minor nits below, then

Reviewed-by: Frank Li <Frank.Li at nxp.com> 

> 
> ---
> Changes in v4:
>  - change comments bit error
>  - add CYCLES_COUNTER_MASK
> Changes in v3:
>  - modify the comments as suggested from Mark
>  - use mask to remove bias value
>  - merge two patches to one
> Changes in v2:
>  - improve if condition
> ---
>  drivers/perf/fsl_imx8_ddr_perf.c | 23 +++++++++++++++++++++++
>  1 file changed, 23 insertions(+)
> 
> diff --git a/drivers/perf/fsl_imx8_ddr_perf.c
> b/drivers/perf/fsl_imx8_ddr_perf.c
> index 5222ba1e79d0..e59e4fc6378d 100644
> --- a/drivers/perf/fsl_imx8_ddr_perf.c
> +++ b/drivers/perf/fsl_imx8_ddr_perf.c
> @@ -28,6 +28,8 @@
>  #define CNTL_CLEAR_MASK		0xFFFFFFFD
>  #define CNTL_OVER_MASK		0xFFFFFFFE
> 
> +#define CNTL_CP_SHIFT		16
> +#define CNTL_CP_MASK		(0xFF << CNTL_CP_SHIFT)
>  #define CNTL_CSV_SHIFT		24
>  #define CNTL_CSV_MASK		(0xFFU << CNTL_CSV_SHIFT)
> 
> @@ -35,6 +37,8 @@
>  #define EVENT_CYCLES_COUNTER	0
>  #define NUM_COUNTERS		4
> 
> +/* For removing bias if cycle counter CNTL.CP is set to 0xf0 */
> +#define CYCLES_COUNTER_MASK	0x0FFFFFFF
>  #define AXI_MASKING_REVERT	0xffff0000	/*
> AXI_MASKING(MSB 16bits) + AXI_ID(LSB 16bits) */
> 
>  #define to_ddr_pmu(p)		container_of(p, struct ddr_pmu, pmu)
> @@ -427,6 +431,17 @@ static void ddr_perf_counter_enable(struct
> ddr_pmu *pmu, int config,
>  		writel(0, pmu->base + reg);
>  		val = CNTL_EN | CNTL_CLEAR;
>  		val |= FIELD_PREP(CNTL_CSV_MASK, config);
> +
> +		/*
> +		 * On i.MX8MP we need to bias the cycle counter to overflow
> more often.
> +		 * We do this by initializing bits [23:16] of the counter value
> via the
> +		 * COUNTER_CTRL Counter Parameter (CP) field.
> +		 */
> +		if (pmu->devtype_data->quirks &
> DDR_CAP_AXI_ID_FILTER_ENHANCED) {
> +			if (counter == EVENT_CYCLES_COUNTER)
> +				val |= FIELD_PREP(CNTL_CP_MASK, 0xf0);
> +		}
> +
>  		writel(val, pmu->base + reg);
>  	} else {
>  		/* Disable counter */
> @@ -466,6 +481,14 @@ static void ddr_perf_event_update(struct
> perf_event *event)
>  	int ret;
> 
>  	new_raw_count = ddr_perf_read_counter(pmu, counter);
> +	/*
> +	 * Remove the bias applied in ddr_perf_counter_enable().
> +	 */

One line comments should be
/* Remove the bias applied in ddr_perf_counter_enable(). */

> +	if (pmu->devtype_data->quirks &
> DDR_CAP_AXI_ID_FILTER_ENHANCED) {
> +		if (counter == EVENT_CYCLES_COUNTER)
> +			new_raw_count &= CYCLES_COUNTER_MASK;
> +	}
> +
>  	local64_add(new_raw_count, &event->count);
> 
>  	/*
> --
> 2.34.1




More information about the linux-arm-kernel mailing list