| Message ID | 20190725071955.19117-1-qiangqing.zhang@nxp.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [V2] perf: imx8_ddr_perf: add AXI ID filter support | expand |
> -----Original Message----- > From: Joakim Zhang > Sent: Thursday, July 25, 2019 2:22 AM > To: robin.murphy@arm.com; will@kernel.org; mark.rutland@arm.com; Frank Li > <frank.li@nxp.com> > Cc: linux-arm-kernel@lists.infradead.org; dl-linux-imx <linux-imx@nxp.com>; > kernel@pengutronix.de; Joakim Zhang <qiangqing.zhang@nxp.com> > Subject: [PATCH V2] perf: imx8_ddr_perf: add AXI ID filter support > > AXI filtering is used by CSV modes 0x41 and 0x42 to count reads or writes with > an ARID or AXID matching filter setting. Granularity is at subsystem level. > Implementation does not allow filtring between masters within a subsystem. > Filter is defined with 2 configuration registers. > > --AXI_ID defines AxID matching value > --AXI_MASKING defines which bits of AxID are meaningful for the matching > > When non-masked bits are matching corresponding AXI_ID bits then counter is > incremented. This filter allows counting read or write access from a subsystem > or multiple subsystems. > > Perf counter is incremented if AxID && AXI_MASKING == AXI_ID && > AXI_MASKING > > AXI_ID and AXI_MASKING are mapped on DPCR1 register in performance > counter. > > Read and write AXI ID filter can't be specified at the same time as this filter is > shared between counters. > > e.g. > perf stat -a -e imx8_ddr0/axi-id-read,axi_id=0xMMMMDDDD/ cmd > MMMM: AXI_MASKING > DDDD: AXI_ID > > ChangeLog: > V1 -> V2: > * add error log if user specifies read/write AXI ID filter at > the same time. > * of_device_get_match_data() instead of of_match_device(), and > remove the check of return value. > > Signed-off-by: Joakim Zhang <qiangqing.zhang@nxp.com> > --- > drivers/perf/fsl_imx8_ddr_perf.c | 48 ++++++++++++++++++++++++++++++-- > 1 file changed, 46 insertions(+), 2 deletions(-) > > diff --git a/drivers/perf/fsl_imx8_ddr_perf.c b/drivers/perf/fsl_imx8_ddr_perf.c > index 63fe21600072..da851da74cbd 100644 > --- a/drivers/perf/fsl_imx8_ddr_perf.c > +++ b/drivers/perf/fsl_imx8_ddr_perf.c > @@ -42,9 +42,22 @@ > > static DEFINE_IDA(ddr_ida); > > +/* DDR Perf hardware feature */ > +#define DDR_CAP_AXI_ID_FILTER 0x1 /* support AXI ID filter > */ > + > +struct fsl_ddr_devtype_data { > + unsigned int quirks; /* quirks needed for different DDR Perf core */ > +}; > + > +static const struct fsl_ddr_devtype_data imx8_devtype_data; > + > +static const struct fsl_ddr_devtype_data imx8m_devtype_data = { > + .quirks = DDR_CAP_AXI_ID_FILTER, > +}; > + > static const struct of_device_id imx_ddr_pmu_dt_ids[] = { > - { .compatible = "fsl,imx8-ddr-pmu",}, > - { .compatible = "fsl,imx8m-ddr-pmu",}, > + { .compatible = "fsl,imx8-ddr-pmu", .data = &imx8_devtype_data}, > + { .compatible = "fsl,imx8m-ddr-pmu", .data = &imx8m_devtype_data}, > { /* sentinel */ } > }; > > @@ -57,6 +70,8 @@ struct ddr_pmu { > struct perf_event *events[NUM_COUNTERS]; > int active_events; > enum cpuhp_state cpuhp_state; > + const struct fsl_ddr_devtype_data *devtype_data; > + bool axi_id_read, axi_id_write; > int irq; > int id; > }; > @@ -128,6 +143,8 @@ static struct attribute *ddr_perf_events_attrs[] = { > IMX8_DDR_PMU_EVENT_ATTR(refresh, 0x37), > IMX8_DDR_PMU_EVENT_ATTR(write, 0x38), > IMX8_DDR_PMU_EVENT_ATTR(raw-hazard, 0x39), > + IMX8_DDR_PMU_EVENT_ATTR(axi-id-read, 0x41), > + IMX8_DDR_PMU_EVENT_ATTR(axi-id-write, 0x42), > NULL, > }; > > @@ -137,9 +154,11 @@ static struct attribute_group > ddr_perf_events_attr_group = { }; > > PMU_FORMAT_ATTR(event, "config:0-7"); > +PMU_FORMAT_ATTR(axi_id, "config1:0-31"); > > static struct attribute *ddr_perf_format_attrs[] = { > &format_attr_event.attr, > + &format_attr_axi_id.attr, > NULL, > }; > > @@ -274,6 +293,22 @@ static void ddr_perf_event_start(struct perf_event > *event, int flags) > struct hw_perf_event *hwc = &event->hw; > int counter = hwc->idx; > > + if (pmu->devtype_data->quirks & DDR_CAP_AXI_ID_FILTER) { > + if (event->attr.config == 0x41) > + pmu->axi_id_read = true; > + > + if (event->attr.config == 0x42) > + pmu->axi_id_write = true; > + > + if (pmu->axi_id_read && pmu->axi_id_write) { > + dev_err(pmu->dev, "axi-id-read/write event can't be > specified at the same time\n"); axi_id can be the same value for 0x41 and 0x42. It should be checked at event_add. So it provide user early error information. Best regards Frank Li > + } else { > + int val = event->attr.config1; > + > + writel(val, pmu->base + COUNTER_DPCR1); > + } > + } > + > local64_set(&hwc->prev_count, 0); > > ddr_perf_counter_enable(pmu, event->attr.config, counter, true); @@ - > 316,6 +351,11 @@ static void ddr_perf_event_stop(struct perf_event *event, > int flags) > ddr_perf_event_update(event); > > hwc->state |= PERF_HES_STOPPED; > + > + if (pmu->devtype_data->quirks & DDR_CAP_AXI_ID_FILTER) { > + pmu->axi_id_read = false; > + pmu->axi_id_write = false; > + } > } > > static void ddr_perf_event_del(struct perf_event *event, int flags) @@ -445,6 > +485,7 @@ static int ddr_perf_offline_cpu(unsigned int cpu, struct hlist_node > *node) > > static int ddr_perf_probe(struct platform_device *pdev) { > + const struct fsl_ddr_devtype_data *data; > struct ddr_pmu *pmu; > struct device_node *np; > void __iomem *base; > @@ -472,6 +513,9 @@ static int ddr_perf_probe(struct platform_device *pdev) > if (!name) > return -ENOMEM; > > + data = of_device_get_match_data(&pdev->dev); > + pmu->devtype_data = data; > + > pmu->cpu = raw_smp_processor_id(); > ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN, > DDR_CPUHP_CB_NAME, > -- > 2.17.1
diff --git a/drivers/perf/fsl_imx8_ddr_perf.c b/drivers/perf/fsl_imx8_ddr_perf.c index 63fe21600072..da851da74cbd 100644 --- a/drivers/perf/fsl_imx8_ddr_perf.c +++ b/drivers/perf/fsl_imx8_ddr_perf.c @@ -42,9 +42,22 @@ static DEFINE_IDA(ddr_ida); +/* DDR Perf hardware feature */ +#define DDR_CAP_AXI_ID_FILTER 0x1 /* support AXI ID filter */ + +struct fsl_ddr_devtype_data { + unsigned int quirks; /* quirks needed for different DDR Perf core */ +}; + +static const struct fsl_ddr_devtype_data imx8_devtype_data; + +static const struct fsl_ddr_devtype_data imx8m_devtype_data = { + .quirks = DDR_CAP_AXI_ID_FILTER, +}; + static const struct of_device_id imx_ddr_pmu_dt_ids[] = { - { .compatible = "fsl,imx8-ddr-pmu",}, - { .compatible = "fsl,imx8m-ddr-pmu",}, + { .compatible = "fsl,imx8-ddr-pmu", .data = &imx8_devtype_data}, + { .compatible = "fsl,imx8m-ddr-pmu", .data = &imx8m_devtype_data}, { /* sentinel */ } }; @@ -57,6 +70,8 @@ struct ddr_pmu { struct perf_event *events[NUM_COUNTERS]; int active_events; enum cpuhp_state cpuhp_state; + const struct fsl_ddr_devtype_data *devtype_data; + bool axi_id_read, axi_id_write; int irq; int id; }; @@ -128,6 +143,8 @@ static struct attribute *ddr_perf_events_attrs[] = { IMX8_DDR_PMU_EVENT_ATTR(refresh, 0x37), IMX8_DDR_PMU_EVENT_ATTR(write, 0x38), IMX8_DDR_PMU_EVENT_ATTR(raw-hazard, 0x39), + IMX8_DDR_PMU_EVENT_ATTR(axi-id-read, 0x41), + IMX8_DDR_PMU_EVENT_ATTR(axi-id-write, 0x42), NULL, }; @@ -137,9 +154,11 @@ static struct attribute_group ddr_perf_events_attr_group = { }; PMU_FORMAT_ATTR(event, "config:0-7"); +PMU_FORMAT_ATTR(axi_id, "config1:0-31"); static struct attribute *ddr_perf_format_attrs[] = { &format_attr_event.attr, + &format_attr_axi_id.attr, NULL, }; @@ -274,6 +293,22 @@ static void ddr_perf_event_start(struct perf_event *event, int flags) struct hw_perf_event *hwc = &event->hw; int counter = hwc->idx; + if (pmu->devtype_data->quirks & DDR_CAP_AXI_ID_FILTER) { + if (event->attr.config == 0x41) + pmu->axi_id_read = true; + + if (event->attr.config == 0x42) + pmu->axi_id_write = true; + + if (pmu->axi_id_read && pmu->axi_id_write) { + dev_err(pmu->dev, "axi-id-read/write event can't be specified at the same time\n"); + } else { + int val = event->attr.config1; + + writel(val, pmu->base + COUNTER_DPCR1); + } + } + local64_set(&hwc->prev_count, 0); ddr_perf_counter_enable(pmu, event->attr.config, counter, true); @@ -316,6 +351,11 @@ static void ddr_perf_event_stop(struct perf_event *event, int flags) ddr_perf_event_update(event); hwc->state |= PERF_HES_STOPPED; + + if (pmu->devtype_data->quirks & DDR_CAP_AXI_ID_FILTER) { + pmu->axi_id_read = false; + pmu->axi_id_write = false; + } } static void ddr_perf_event_del(struct perf_event *event, int flags) @@ -445,6 +485,7 @@ static int ddr_perf_offline_cpu(unsigned int cpu, struct hlist_node *node) static int ddr_perf_probe(struct platform_device *pdev) { + const struct fsl_ddr_devtype_data *data; struct ddr_pmu *pmu; struct device_node *np; void __iomem *base; @@ -472,6 +513,9 @@ static int ddr_perf_probe(struct platform_device *pdev) if (!name) return -ENOMEM; + data = of_device_get_match_data(&pdev->dev); + pmu->devtype_data = data; + pmu->cpu = raw_smp_processor_id(); ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN, DDR_CPUHP_CB_NAME,
AXI filtering is used by CSV modes 0x41 and 0x42 to count reads or writes with an ARID or AXID matching filter setting. Granularity is at subsystem level. Implementation does not allow filtring between masters within a subsystem. Filter is defined with 2 configuration registers. --AXI_ID defines AxID matching value --AXI_MASKING defines which bits of AxID are meaningful for the matching When non-masked bits are matching corresponding AXI_ID bits then counter is incremented. This filter allows counting read or write access from a subsystem or multiple subsystems. Perf counter is incremented if AxID && AXI_MASKING == AXI_ID && AXI_MASKING AXI_ID and AXI_MASKING are mapped on DPCR1 register in performance counter. Read and write AXI ID filter can't be specified at the same time as this filter is shared between counters. e.g. perf stat -a -e imx8_ddr0/axi-id-read,axi_id=0xMMMMDDDD/ cmd MMMM: AXI_MASKING DDDD: AXI_ID ChangeLog: V1 -> V2: * add error log if user specifies read/write AXI ID filter at the same time. * of_device_get_match_data() instead of of_match_device(), and remove the check of return value. Signed-off-by: Joakim Zhang <qiangqing.zhang@nxp.com> --- drivers/perf/fsl_imx8_ddr_perf.c | 48 ++++++++++++++++++++++++++++++-- 1 file changed, 46 insertions(+), 2 deletions(-)