* [PATCH V4 1/4] dt-bindings: perf: imx8-ddr: add imx8qxp ddr performance monitor
@ 2019-02-15 18:03 ` Frank Li
0 siblings, 0 replies; 43+ messages in thread
From: Frank Li @ 2019-02-15 18:03 UTC (permalink / raw)
To: mark.rutland, will.deacon, shawnguo, s.hauer, kernel, festevam,
dl-linux-imx, robh+dt, Aisheng Dong, devicetree, lznuaa,
linux-arm-kernel
Cc: Frank Li
Added binding doc for imx8qxp ddr performance monitor
Signed-off-by: Frank Li <Frank.Li@nxp.com>
---
Change from v4 to v4
* remove "standard xxx"
Change from v2 to v3
* ddr_pmu0 -> ddr-pmu
.../devicetree/bindings/perf/fsl-imx-ddr.txt | 22 ++++++++++++++++++++++
1 file changed, 22 insertions(+)
create mode 100644 Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
diff --git a/Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt b/Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
new file mode 100644
index 0000000..c667b87
--- /dev/null
+++ b/Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
@@ -0,0 +1,22 @@
+* Freescale(NXP) IMX8 DDR performance monitor
+
+Required properties:
+
+- compatible: should be one of:
+ "fsl,imx8-ddr-pmu"
+ "fsl,imx8m-ddr-pmu"
+
+- reg: physical address and size
+
+- interrupts: single interrupt
+ generated by the control block
+
+Example:
+
+ ddr-pmu@5c020000 {
+ compatible = "fsl,imx8-ddr-pmu";
+ reg = <0x0 0x5c020000 0x0 0x10000>;
+ interrupt-parent = <&gic>;
+ interrupts = <GIC_SPI 131 IRQ_TYPE_LEVEL_HIGH>;
+ };
+
--
2.5.2
^ permalink raw reply related [flat|nested] 43+ messages in thread
* [PATCH V4 1/4] dt-bindings: perf: imx8-ddr: add imx8qxp ddr performance monitor
@ 2019-02-15 18:03 ` Frank Li
0 siblings, 0 replies; 43+ messages in thread
From: Frank Li @ 2019-02-15 18:03 UTC (permalink / raw)
To: mark.rutland, will.deacon, shawnguo, s.hauer, kernel, festevam,
dl-linux-imx, robh+dt, Aisheng Dong, devicetree, lznuaa,
linux-arm-kernel
Cc: Frank Li
Added binding doc for imx8qxp ddr performance monitor
Signed-off-by: Frank Li <Frank.Li@nxp.com>
---
Change from v4 to v4
* remove "standard xxx"
Change from v2 to v3
* ddr_pmu0 -> ddr-pmu
.../devicetree/bindings/perf/fsl-imx-ddr.txt | 22 ++++++++++++++++++++++
1 file changed, 22 insertions(+)
create mode 100644 Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
diff --git a/Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt b/Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
new file mode 100644
index 0000000..c667b87
--- /dev/null
+++ b/Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
@@ -0,0 +1,22 @@
+* Freescale(NXP) IMX8 DDR performance monitor
+
+Required properties:
+
+- compatible: should be one of:
+ "fsl,imx8-ddr-pmu"
+ "fsl,imx8m-ddr-pmu"
+
+- reg: physical address and size
+
+- interrupts: single interrupt
+ generated by the control block
+
+Example:
+
+ ddr-pmu@5c020000 {
+ compatible = "fsl,imx8-ddr-pmu";
+ reg = <0x0 0x5c020000 0x0 0x10000>;
+ interrupt-parent = <&gic>;
+ interrupts = <GIC_SPI 131 IRQ_TYPE_LEVEL_HIGH>;
+ };
+
--
2.5.2
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 43+ messages in thread
* [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
2019-02-15 18:03 ` Frank Li
@ 2019-02-15 18:03 ` Frank Li
-1 siblings, 0 replies; 43+ messages in thread
From: Frank Li @ 2019-02-15 18:03 UTC (permalink / raw)
To: mark.rutland, will.deacon, shawnguo, s.hauer, kernel, festevam,
dl-linux-imx, robh+dt, Aisheng Dong, devicetree, lznuaa,
linux-arm-kernel
Cc: Frank Li
Add ddr performance monitor support for iMX8QXP
There are 4 counters for ddr perfomance events.
counter 0 is dedicated for cycles.
you choose any up to 3 no cycles events.
for example:
perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
Support below events.
ddr0/activate/ [Kernel PMU event]
ddr0/axid-read/ [Kernel PMU event]
ddr0/axid-write/ [Kernel PMU event]
ddr0/cycles/ [Kernel PMU event]
ddr0/hp-read-credit-cnt/ [Kernel PMU event]
ddr0/hp-read/ [Kernel PMU event]
ddr0/hp-req-nodcredit/ [Kernel PMU event]
ddr0/hp-xact-credit/ [Kernel PMU event]
ddr0/load-mode/ [Kernel PMU event]
ddr0/lp-read-credit-cnt/ [Kernel PMU event]
ddr0/lp-req-nocredit/ [Kernel PMU event]
ddr0/lp-xact-credit/ [Kernel PMU event]
ddr0/mwr/ [Kernel PMU event]
ddr0/precharge/ [Kernel PMU event]
ddr0/raw-hazard/ [Kernel PMU event]
ddr0/read-access/ [Kernel PMU event]
ddr0/read-activate/ [Kernel PMU event]
ddr0/read-command/ [Kernel PMU event]
ddr0/read-cycles/ [Kernel PMU event]
ddr0/read-modify-write-command/ [Kernel PMU event]
ddr0/read-queue-depth/ [Kernel PMU event]
ddr0/read-write-transition/ [Kernel PMU event]
ddr0/read/ [Kernel PMU event]
ddr0/refresh/ [Kernel PMU event]
ddr0/selfresh/ [Kernel PMU event]
ddr0/wr-xact-credit/ [Kernel PMU event]
ddr0/write-access/ [Kernel PMU event]
ddr0/write-command/ [Kernel PMU event]
ddr0/write-credit-cnt/ [Kernel PMU event]
ddr0/write-cycles/ [Kernel PMU event]
ddr0/write-queue-depth/ [Kernel PMU event]
ddr0/write/
Signed-off-by: Frank Li <Frank.Li@nxp.com>
---
Change from v3 to v4
* Change FSL_IMX8_DDR_PERF to FSL_IMX8_DDR_PMU
* sort include
* remove struct fsl_ddr_devtype_data
* Added comment need disable control first
* Added comment about must enable cycle counter
* Added macro for EVENT_AXI_READ, remove hardcode 0x41 and 0x42
* Added comment about cycle counter is fastest one
Change from v2 to v3
* remove kfree
Change from V1 to V2
* update Kconfig by use i.MX8 instead of i.MX8 QXP
* remove gpl statememnt since SPDX tag
* use dev_kzalloc
* use dev_err
* commit message show axi_read 0x41\axi_write 0x42
* commit message show cycles must be enabled
* Irq only issue at cycles overflow
* use NUM_COUNTER
* use devm_request_irq
* add hotplug callback to handle context migration
drivers/perf/Kconfig | 7 +
drivers/perf/Makefile | 1 +
drivers/perf/fsl_imx8_ddr_perf.c | 573 +++++++++++++++++++++++++++++++++++++++
3 files changed, 581 insertions(+)
create mode 100644 drivers/perf/fsl_imx8_ddr_perf.c
diff --git a/drivers/perf/Kconfig b/drivers/perf/Kconfig
index af9bc17..e567437 100644
--- a/drivers/perf/Kconfig
+++ b/drivers/perf/Kconfig
@@ -61,6 +61,13 @@ config ARM_DSU_PMU
system, control logic. The PMU allows counting various events related
to DSU.
+config FSL_IMX8_DDR_PMU
+ tristate "Freescale i.MX8 DDR perf monitor"
+ depends on ARCH_MXC
+ help
+ Provides support for ddr perfomance monitor in i.MX8. Provide memory
+ througput information.
+
config HISI_PMU
bool "HiSilicon SoC PMU"
depends on ARM64 && ACPI
diff --git a/drivers/perf/Makefile b/drivers/perf/Makefile
index 909f27f..0bc066b 100644
--- a/drivers/perf/Makefile
+++ b/drivers/perf/Makefile
@@ -4,6 +4,7 @@ obj-$(CONFIG_ARM_CCN) += arm-ccn.o
obj-$(CONFIG_ARM_DSU_PMU) += arm_dsu_pmu.o
obj-$(CONFIG_ARM_PMU) += arm_pmu.o arm_pmu_platform.o
obj-$(CONFIG_ARM_PMU_ACPI) += arm_pmu_acpi.o
+obj-$(CONFIG_FSL_IMX8_DDR_PMU) += fsl_imx8_ddr_perf.o
obj-$(CONFIG_HISI_PMU) += hisilicon/
obj-$(CONFIG_QCOM_L2_PMU) += qcom_l2_pmu.o
obj-$(CONFIG_QCOM_L3_PMU) += qcom_l3_pmu.o
diff --git a/drivers/perf/fsl_imx8_ddr_perf.c b/drivers/perf/fsl_imx8_ddr_perf.c
new file mode 100644
index 0000000..3d02daf
--- /dev/null
+++ b/drivers/perf/fsl_imx8_ddr_perf.c
@@ -0,0 +1,573 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2017 NXP
+ * Copyright 2016 Freescale Semiconductor, Inc.
+ */
+
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_device.h>
+#include <linux/of_irq.h>
+#include <linux/perf_event.h>
+#include <linux/slab.h>
+
+#define COUNTER_CNTL 0x0
+#define COUNTER_READ 0x20
+
+#define COUNTER_DPCR1 0x30
+
+#define CNTL_OVER 0x1
+#define CNTL_CLEAR 0x2
+#define CNTL_EN 0x4
+#define CNTL_EN_MASK 0xFFFFFFFB
+#define CNTL_CLEAR_MASK 0xFFFFFFFD
+#define CNTL_OVER_MASK 0xFFFFFFFE
+
+#define CNTL_CSV_SHIFT 24
+#define CNTL_CSV_MASK (0xFF << CNTL_CSV_SHIFT)
+
+#define EVENT_CYCLES_ID 0
+#define EVENT_CYCLES_COUNTER 0
+#define NUM_COUNTER 4
+#define MAX_EVENT 3
+#define EVENT_AXI_READ 0x41
+#define EVENT_AXI_WRITE 0x42
+
+#define to_ddr_pmu(p) container_of(p, struct ddr_pmu, pmu)
+
+#define DDR_PERF_DEV_NAME "ddr_perf"
+
+static DEFINE_IDA(ddr_ida);
+
+PMU_EVENT_ATTR_STRING(cycles, ddr_perf_cycles, "event=0x00");
+PMU_EVENT_ATTR_STRING(selfresh, ddr_perf_selfresh, "event=0x01");
+PMU_EVENT_ATTR_STRING(read-access, ddr_perf_read_accesses, "event=0x04");
+PMU_EVENT_ATTR_STRING(write-access, ddr_perf_write_accesses, "event=0x05");
+PMU_EVENT_ATTR_STRING(read-queue-depth, ddr_perf_read_queue_depth,
+ "event=0x08");
+PMU_EVENT_ATTR_STRING(write-queue-depth, ddr_perf_write_queue_depth,
+ "event=0x09");
+PMU_EVENT_ATTR_STRING(lp-read-credit-cnt, ddr_perf_lp_read_credit_cnt,
+ "event=0x10");
+PMU_EVENT_ATTR_STRING(hp-read-credit-cnt, ddr_perf_hp_read_credit_cnt,
+ "event=0x11");
+PMU_EVENT_ATTR_STRING(write-credit-cnt, ddr_perf_write_credit_cnt,
+ "event=0x12");
+PMU_EVENT_ATTR_STRING(read-command, ddr_perf_read_command, "event=0x20");
+PMU_EVENT_ATTR_STRING(write-command, ddr_perf_write_command, "event=0x21");
+PMU_EVENT_ATTR_STRING(read-modify-write-command,
+ ddr_perf_read_modify_write_command, "event=0x22");
+PMU_EVENT_ATTR_STRING(hp-read, ddr_perf_hp_read, "event=0x23");
+PMU_EVENT_ATTR_STRING(hp-req-nodcredit, ddr_perf_hp_req_nocredit, "event=0x24");
+PMU_EVENT_ATTR_STRING(hp-xact-credit, ddr_perf_hp_xact_credit, "event=0x25");
+PMU_EVENT_ATTR_STRING(lp-req-nocredit, ddr_perf_lp_req_nocredit, "event=0x26");
+PMU_EVENT_ATTR_STRING(lp-xact-credit, ddr_perf_lp_xact_credit, "event=0x27");
+PMU_EVENT_ATTR_STRING(wr-xact-credit, ddr_perf_wr_xact_credit, "event=0x29");
+PMU_EVENT_ATTR_STRING(read-cycles, ddr_perf_read_cycles, "event=0x2a");
+PMU_EVENT_ATTR_STRING(write-cycles, ddr_perf_write_cycles, "event=0x2b");
+PMU_EVENT_ATTR_STRING(read-write-transition, ddr_perf_read_write_transition,
+ "event=0x30");
+PMU_EVENT_ATTR_STRING(precharge, ddr_perf_precharge, "event=0x31");
+PMU_EVENT_ATTR_STRING(activate, ddr_perf_activate, "event=0x32");
+PMU_EVENT_ATTR_STRING(load-mode, ddr_perf_load_mode, "event=0x33");
+PMU_EVENT_ATTR_STRING(mwr, ddr_perf_mwr, "event=0x34");
+PMU_EVENT_ATTR_STRING(read, ddr_perf_read, "event=0x35");
+PMU_EVENT_ATTR_STRING(read-activate, ddr_perf_read_activate, "event=0x36");
+PMU_EVENT_ATTR_STRING(refresh, ddr_perf_refresh, "event=0x37");
+PMU_EVENT_ATTR_STRING(write, ddr_perf_write, "event=0x38");
+PMU_EVENT_ATTR_STRING(raw-hazard, ddr_perf_raw_hazard, "event=0x39");
+
+PMU_EVENT_ATTR_STRING(axid-read, ddr_perf_axid_read,
+ "event="__stringify(EVENT_AXI_READ));
+PMU_EVENT_ATTR_STRING(axid-write, ddr_perf_axid_write,
+ "event="__stringify(EVENT_AXI_WRITE));
+
+#define DDR_CAP_AXI_ID 0x1
+
+static const struct of_device_id imx_ddr_pmu_dt_ids[] = {
+ { .compatible = "fsl,imx8-ddr-pmu",},
+ { .compatible = "fsl,imx8m-ddr-pmu", .data = (void *)DDR_CAP_AXI_ID},
+ { /* sentinel */ }
+};
+
+struct ddr_pmu {
+ struct pmu pmu;
+ void __iomem *base;
+ cpumask_t cpu;
+ struct hlist_node node;
+ struct device *dev;
+ struct perf_event *active_events[NUM_COUNTER];
+ int total_events;
+ enum cpuhp_state cpuhp_state;
+ bool cycles_active;
+ uintptr_t flags;
+};
+
+static ssize_t ddr_perf_cpumask_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct ddr_pmu *pmu = dev_get_drvdata(dev);
+
+ return cpumap_print_to_pagebuf(true, buf, &pmu->cpu);
+}
+
+static struct device_attribute ddr_perf_cpumask_attr =
+ __ATTR(cpumask, 0444, ddr_perf_cpumask_show, NULL);
+
+static struct attribute *ddr_perf_cpumask_attrs[] = {
+ &ddr_perf_cpumask_attr.attr,
+ NULL,
+};
+
+static struct attribute_group ddr_perf_cpumask_attr_group = {
+ .attrs = ddr_perf_cpumask_attrs,
+};
+
+static struct attribute *ddr_perf_events_attrs[] = {
+ &ddr_perf_cycles.attr.attr,
+ &ddr_perf_selfresh.attr.attr,
+ &ddr_perf_read_accesses.attr.attr,
+ &ddr_perf_write_accesses.attr.attr,
+ &ddr_perf_read_queue_depth.attr.attr,
+ &ddr_perf_write_queue_depth.attr.attr,
+ &ddr_perf_lp_read_credit_cnt.attr.attr,
+ &ddr_perf_hp_read_credit_cnt.attr.attr,
+ &ddr_perf_write_credit_cnt.attr.attr,
+ &ddr_perf_read_command.attr.attr,
+ &ddr_perf_write_command.attr.attr,
+ &ddr_perf_read_modify_write_command.attr.attr,
+ &ddr_perf_hp_read.attr.attr,
+ &ddr_perf_hp_req_nocredit.attr.attr,
+ &ddr_perf_hp_xact_credit.attr.attr,
+ &ddr_perf_lp_req_nocredit.attr.attr,
+ &ddr_perf_lp_xact_credit.attr.attr,
+ &ddr_perf_wr_xact_credit.attr.attr,
+ &ddr_perf_read_cycles.attr.attr,
+ &ddr_perf_write_cycles.attr.attr,
+ &ddr_perf_read_write_transition.attr.attr,
+ &ddr_perf_precharge.attr.attr,
+ &ddr_perf_activate.attr.attr,
+ &ddr_perf_load_mode.attr.attr,
+ &ddr_perf_mwr.attr.attr,
+ &ddr_perf_read.attr.attr,
+ &ddr_perf_read_activate.attr.attr,
+ &ddr_perf_refresh.attr.attr,
+ &ddr_perf_write.attr.attr,
+ &ddr_perf_raw_hazard.attr.attr,
+ &ddr_perf_axid_read.attr.attr,
+ &ddr_perf_axid_write.attr.attr,
+ NULL,
+};
+
+static struct attribute_group ddr_perf_events_attr_group = {
+ .name = "events",
+ .attrs = ddr_perf_events_attrs,
+};
+
+PMU_FORMAT_ATTR(event, "config:0-63");
+PMU_FORMAT_ATTR(axi_id, "config1:0-63");
+
+static struct attribute *ddr_perf_format_attrs[] = {
+ &format_attr_event.attr,
+ &format_attr_axi_id.attr,
+ NULL,
+};
+
+static struct attribute_group ddr_perf_format_attr_group = {
+ .name = "format",
+ .attrs = ddr_perf_format_attrs,
+};
+
+static const struct attribute_group *attr_groups[] = {
+ &ddr_perf_events_attr_group,
+ &ddr_perf_format_attr_group,
+ &ddr_perf_cpumask_attr_group,
+ NULL,
+};
+
+static u32 ddr_perf_alloc_counter(struct ddr_pmu *pmu, int event)
+{
+ int i;
+
+ /* Always map cycle event to counter 0
+ Cycles counter is dedicated for cycle event
+ can't used for the other events
+ */
+ if (event == EVENT_CYCLES_ID)
+ return EVENT_CYCLES_COUNTER;
+
+ for (i = 1; i < NUM_COUNTER; i++)
+ if (pmu->active_events[i] == NULL)
+ return i;
+
+ return -ENOENT;
+}
+
+static u32 ddr_perf_free_counter(struct ddr_pmu *pmu, int counter)
+{
+ if (counter < 0 || counter >= NUM_COUNTER)
+ return -ENOENT;
+
+ pmu->active_events[counter] = NULL;
+
+ return 0;
+}
+
+static u32 ddr_perf_read_counter(struct ddr_pmu *pmu, int counter)
+{
+ return readl(pmu->base + COUNTER_READ + counter * 4);
+}
+
+static int ddr_perf_event_init(struct perf_event *event)
+{
+ struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
+ struct hw_perf_event *hwc = &event->hw;
+
+ if (event->attr.type != event->pmu->type)
+ return -ENOENT;
+
+ if (is_sampling_event(event) || event->attach_state & PERF_ATTACH_TASK)
+ return -EOPNOTSUPP;
+
+ if (event->cpu < 0) {
+ dev_warn(pmu->dev, "Can't provide per-task data!\n");
+ return -EOPNOTSUPP;
+ }
+
+ if (event->attr.exclude_user ||
+ event->attr.exclude_kernel ||
+ event->attr.exclude_hv ||
+ event->attr.exclude_idle ||
+ event->attr.exclude_host ||
+ event->attr.exclude_guest ||
+ event->attr.sample_period)
+ return -EINVAL;
+
+ event->cpu = cpumask_first(&pmu->cpu);
+ hwc->idx = -1;
+
+ return 0;
+}
+
+
+static void ddr_perf_event_update(struct perf_event *event)
+{
+ struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
+ struct hw_perf_event *hwc = &event->hw;
+ u64 delta, prev_raw_count, new_raw_count;
+ int counter = hwc->idx;
+
+ do {
+ prev_raw_count = local64_read(&hwc->prev_count);
+ new_raw_count = ddr_perf_read_counter(pmu, counter);
+ } while (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
+ new_raw_count) != prev_raw_count);
+
+ delta = (new_raw_count - prev_raw_count) & 0xFFFFFFFF;
+
+ local64_add(delta, &event->count);
+}
+
+static void ddr_perf_event_enable(struct ddr_pmu *pmu, int config,
+ int counter, bool enable)
+{
+ u8 reg = counter * 4 + COUNTER_CNTL;
+ int val;
+
+ if (enable) {
+ /* must disable first, then enable again
+ * otherwise, cycle counter will not work
+ * if previous state is enabled.
+ */
+ writel(0, pmu->base + reg);
+ val = CNTL_EN | CNTL_CLEAR;
+ val |= (config << CNTL_CSV_SHIFT) & CNTL_CSV_MASK;
+ } else {
+ /* Disable counter */
+ val = readl(pmu->base + reg) & CNTL_EN_MASK;
+ }
+
+ writel(val, pmu->base + reg);
+
+ if (config == EVENT_CYCLES_ID)
+ pmu->cycles_active = enable;
+}
+
+static void ddr_perf_event_start(struct perf_event *event, int flags)
+{
+ struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
+ struct hw_perf_event *hwc = &event->hw;
+ int counter = hwc->idx;
+
+ if (pmu->flags & DDR_CAP_AXI_ID) {
+ if (event->attr.config == EVENT_AXI_READ ||
+ event->attr.config == EVENT_AXI_WRITE) {
+ int val = event->attr.config1;
+
+ writel(val, pmu->base + COUNTER_DPCR1);
+ }
+ }
+
+ local64_set(&hwc->prev_count, 0);
+
+ ddr_perf_event_enable(pmu, event->attr.config, counter, true);
+ /*
+ * If the cycles counter wasn't explicitly selected,
+ * we will enable it now.
+ * cycles counter must be enabled otherwise other counters will
+ * stopped.
+ */
+ if (counter > 0 && !pmu->cycles_active)
+ ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
+ EVENT_CYCLES_COUNTER, true);
+}
+
+static int ddr_perf_event_add(struct perf_event *event, int flags)
+{
+ struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
+ struct hw_perf_event *hwc = &event->hw;
+ int counter;
+ int cfg = event->attr.config;
+
+ counter = ddr_perf_alloc_counter(pmu, cfg);
+ if (counter < 0) {
+ dev_dbg(pmu->dev, "There are not enough counters\n");
+ return -EOPNOTSUPP;
+ }
+
+ pmu->active_events[counter] = event;
+ pmu->total_events++;
+ hwc->idx = counter;
+
+ if (flags & PERF_EF_START)
+ ddr_perf_event_start(event, flags);
+
+ local64_set(&hwc->prev_count, ddr_perf_read_counter(pmu, counter));
+
+ return 0;
+}
+
+static void ddr_perf_event_stop(struct perf_event *event, int flags)
+{
+ struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
+ struct hw_perf_event *hwc = &event->hw;
+ int counter = hwc->idx;
+
+ ddr_perf_event_enable(pmu, event->attr.config, counter, false);
+ ddr_perf_event_update(event);
+}
+
+static void ddr_perf_event_del(struct perf_event *event, int flags)
+{
+ struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
+ struct hw_perf_event *hwc = &event->hw;
+ int counter = hwc->idx;
+
+ ddr_perf_event_stop(event, PERF_EF_UPDATE);
+
+ ddr_perf_free_counter(pmu, counter);
+ pmu->total_events--;
+ hwc->idx = -1;
+
+ /* If all events have stopped, stop the cycles counter as well */
+ if ((pmu->total_events == 0) && pmu->cycles_active)
+ ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
+ EVENT_CYCLES_COUNTER, false);
+}
+
+static int ddr_perf_init(struct ddr_pmu *pmu, void __iomem *base,
+ struct device *dev)
+{
+ *pmu = (struct ddr_pmu) {
+ .pmu = (struct pmu) {
+ .task_ctx_nr = perf_invalid_context,
+ .attr_groups = attr_groups,
+ .event_init = ddr_perf_event_init,
+ .add = ddr_perf_event_add,
+ .del = ddr_perf_event_del,
+ .start = ddr_perf_event_start,
+ .stop = ddr_perf_event_stop,
+ .read = ddr_perf_event_update,
+ },
+ .base = base,
+ .dev = dev,
+ };
+
+ return ida_simple_get(&ddr_ida, 0, 0, GFP_KERNEL);
+}
+
+static irqreturn_t ddr_perf_irq_handler(int irq, void *p)
+{
+ int i;
+ u8 reg;
+ int val;
+ int counter;
+ struct ddr_pmu *pmu = (struct ddr_pmu *) p;
+ struct perf_event *event;
+
+ /* Only cycles counter overflowed can issue irq. all counters will
+ * be stopped when cycles counter overflow. but other counter don't stop
+ * when overflow happen. Update all of the local counter values,
+ * then reset the cycles counter, so the others can continue
+ * counting. cycles counter is fastest counter in all events. at last
+ * 4 times than other counters.
+ */
+ for (i = 0; i < NUM_COUNTER; i++) {
+
+ if (!pmu->active_events[i])
+ continue;
+
+ event = pmu->active_events[i];
+ counter = event->hw.idx;
+ reg = counter * 4 + COUNTER_CNTL;
+ val = readl(pmu->base + reg);
+
+ ddr_perf_event_update(event);
+
+ if (counter == EVENT_CYCLES_COUNTER) {
+ ddr_perf_event_enable(pmu,
+ EVENT_CYCLES_ID,
+ EVENT_CYCLES_COUNTER,
+ true);
+ ddr_perf_event_update(event);
+ }
+ }
+
+ return IRQ_HANDLED;
+}
+
+static int ddr_perf_offline_cpu(unsigned int cpu, struct hlist_node *node)
+{
+ struct ddr_pmu *pmu = hlist_entry_safe(node, struct ddr_pmu, node);
+ int target;
+
+ if (!cpumask_test_and_clear_cpu(cpu, &pmu->cpu))
+ return 0;
+
+ target = cpumask_any_but(cpu_online_mask, cpu);
+ if (target >= nr_cpu_ids)
+ return 0;
+
+ perf_pmu_migrate_context(&pmu->pmu, cpu, target);
+ cpumask_set_cpu(target, &pmu->cpu);
+
+ return 0;
+}
+
+static int ddr_perf_probe(struct platform_device *pdev)
+{
+ struct ddr_pmu *pmu;
+ struct device_node *np;
+ void __iomem *base;
+ struct resource *iomem;
+ char *name;
+ char *hpname;
+ int num;
+ int ret;
+ u32 irq;
+ const struct of_device_id *of_id =
+ of_match_device(imx_ddr_pmu_dt_ids, &pdev->dev);
+
+ iomem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+ base = devm_ioremap_resource(&pdev->dev, iomem);
+ if (IS_ERR(base))
+ return PTR_ERR(base);
+
+ np = pdev->dev.of_node;
+
+ pmu = devm_kzalloc(&pdev->dev, sizeof(*pmu), GFP_KERNEL);
+ if (!pmu)
+ return -ENOMEM;
+
+ num = ddr_perf_init(pmu, base, &pdev->dev);
+
+ name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "ddr%d", num);
+ if (!name)
+ return -ENOMEM;
+
+ hpname = devm_kasprintf(&pdev->dev, GFP_KERNEL,
+ "perf/imx/ddr%d:online", num);
+ if (!hpname)
+ return -ENOMEM;
+
+ pmu->flags = (uintptr_t) of_id->data;
+
+ cpumask_set_cpu(raw_smp_processor_id(), &pmu->cpu);
+ ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN, hpname, NULL,
+ ddr_perf_offline_cpu);
+
+ if (ret < 0) {
+ dev_err(&pdev->dev, "cpuhp_setup_state_multi failed\n");
+ goto ddr_perf_err;
+ }
+
+ pmu->cpuhp_state = ret;
+
+ /* Register the pmu instance for cpu hotplug */
+ cpuhp_state_add_instance_nocalls(pmu->cpuhp_state, &pmu->node);
+
+ ret = perf_pmu_register(&(pmu->pmu), name, -1);
+ if (ret)
+ goto ddr_perf_err;
+
+ /* Request irq */
+ irq = of_irq_get(np, 0);
+ if (irq < 0) {
+ dev_err(&pdev->dev, "Failed to get irq: %d", irq);
+ ret = irq;
+ goto ddr_perf_irq_err;
+ }
+
+ ret = devm_request_irq(&pdev->dev, irq,
+ ddr_perf_irq_handler,
+ IRQF_TRIGGER_RISING | IRQF_ONESHOT,
+ DDR_PERF_DEV_NAME,
+ pmu);
+ if (ret < 0) {
+ dev_err(&pdev->dev, "Request irq failed: %d", ret);
+ goto ddr_perf_irq_err;
+ }
+
+ return 0;
+
+ddr_perf_irq_err:
+ perf_pmu_unregister(&(pmu->pmu));
+
+ddr_perf_err:
+ if (pmu->cpuhp_state)
+ cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
+
+ dev_warn(&pdev->dev, "i.MX8 DDR Perf PMU failed (%d), disabled\n", ret);
+ return ret;
+}
+
+static int ddr_perf_remove(struct platform_device *pdev)
+{
+ struct ddr_pmu *pmu = platform_get_drvdata(pdev);
+
+ cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
+ perf_pmu_unregister(&pmu->pmu);
+
+ return 0;
+}
+
+static struct platform_driver imx_ddr_pmu_driver = {
+ .driver = {
+ .name = "imx-ddr-pmu",
+ .of_match_table = imx_ddr_pmu_dt_ids,
+ },
+ .probe = ddr_perf_probe,
+ .remove = ddr_perf_remove,
+};
+
+static int __init imx_ddr_pmu_init(void)
+{
+ return platform_driver_register(&imx_ddr_pmu_driver);
+}
+
+module_init(imx_ddr_pmu_init);
+
--
2.5.2
^ permalink raw reply related [flat|nested] 43+ messages in thread
* [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
@ 2019-02-15 18:03 ` Frank Li
0 siblings, 0 replies; 43+ messages in thread
From: Frank Li @ 2019-02-15 18:03 UTC (permalink / raw)
To: mark.rutland, will.deacon, shawnguo, s.hauer, kernel, festevam,
dl-linux-imx, robh+dt, Aisheng Dong, devicetree, lznuaa,
linux-arm-kernel
Cc: Frank Li
Add ddr performance monitor support for iMX8QXP
There are 4 counters for ddr perfomance events.
counter 0 is dedicated for cycles.
you choose any up to 3 no cycles events.
for example:
perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
Support below events.
ddr0/activate/ [Kernel PMU event]
ddr0/axid-read/ [Kernel PMU event]
ddr0/axid-write/ [Kernel PMU event]
ddr0/cycles/ [Kernel PMU event]
ddr0/hp-read-credit-cnt/ [Kernel PMU event]
ddr0/hp-read/ [Kernel PMU event]
ddr0/hp-req-nodcredit/ [Kernel PMU event]
ddr0/hp-xact-credit/ [Kernel PMU event]
ddr0/load-mode/ [Kernel PMU event]
ddr0/lp-read-credit-cnt/ [Kernel PMU event]
ddr0/lp-req-nocredit/ [Kernel PMU event]
ddr0/lp-xact-credit/ [Kernel PMU event]
ddr0/mwr/ [Kernel PMU event]
ddr0/precharge/ [Kernel PMU event]
ddr0/raw-hazard/ [Kernel PMU event]
ddr0/read-access/ [Kernel PMU event]
ddr0/read-activate/ [Kernel PMU event]
ddr0/read-command/ [Kernel PMU event]
ddr0/read-cycles/ [Kernel PMU event]
ddr0/read-modify-write-command/ [Kernel PMU event]
ddr0/read-queue-depth/ [Kernel PMU event]
ddr0/read-write-transition/ [Kernel PMU event]
ddr0/read/ [Kernel PMU event]
ddr0/refresh/ [Kernel PMU event]
ddr0/selfresh/ [Kernel PMU event]
ddr0/wr-xact-credit/ [Kernel PMU event]
ddr0/write-access/ [Kernel PMU event]
ddr0/write-command/ [Kernel PMU event]
ddr0/write-credit-cnt/ [Kernel PMU event]
ddr0/write-cycles/ [Kernel PMU event]
ddr0/write-queue-depth/ [Kernel PMU event]
ddr0/write/
Signed-off-by: Frank Li <Frank.Li@nxp.com>
---
Change from v3 to v4
* Change FSL_IMX8_DDR_PERF to FSL_IMX8_DDR_PMU
* sort include
* remove struct fsl_ddr_devtype_data
* Added comment need disable control first
* Added comment about must enable cycle counter
* Added macro for EVENT_AXI_READ, remove hardcode 0x41 and 0x42
* Added comment about cycle counter is fastest one
Change from v2 to v3
* remove kfree
Change from V1 to V2
* update Kconfig by use i.MX8 instead of i.MX8 QXP
* remove gpl statememnt since SPDX tag
* use dev_kzalloc
* use dev_err
* commit message show axi_read 0x41\axi_write 0x42
* commit message show cycles must be enabled
* Irq only issue at cycles overflow
* use NUM_COUNTER
* use devm_request_irq
* add hotplug callback to handle context migration
drivers/perf/Kconfig | 7 +
drivers/perf/Makefile | 1 +
drivers/perf/fsl_imx8_ddr_perf.c | 573 +++++++++++++++++++++++++++++++++++++++
3 files changed, 581 insertions(+)
create mode 100644 drivers/perf/fsl_imx8_ddr_perf.c
diff --git a/drivers/perf/Kconfig b/drivers/perf/Kconfig
index af9bc17..e567437 100644
--- a/drivers/perf/Kconfig
+++ b/drivers/perf/Kconfig
@@ -61,6 +61,13 @@ config ARM_DSU_PMU
system, control logic. The PMU allows counting various events related
to DSU.
+config FSL_IMX8_DDR_PMU
+ tristate "Freescale i.MX8 DDR perf monitor"
+ depends on ARCH_MXC
+ help
+ Provides support for ddr perfomance monitor in i.MX8. Provide memory
+ througput information.
+
config HISI_PMU
bool "HiSilicon SoC PMU"
depends on ARM64 && ACPI
diff --git a/drivers/perf/Makefile b/drivers/perf/Makefile
index 909f27f..0bc066b 100644
--- a/drivers/perf/Makefile
+++ b/drivers/perf/Makefile
@@ -4,6 +4,7 @@ obj-$(CONFIG_ARM_CCN) += arm-ccn.o
obj-$(CONFIG_ARM_DSU_PMU) += arm_dsu_pmu.o
obj-$(CONFIG_ARM_PMU) += arm_pmu.o arm_pmu_platform.o
obj-$(CONFIG_ARM_PMU_ACPI) += arm_pmu_acpi.o
+obj-$(CONFIG_FSL_IMX8_DDR_PMU) += fsl_imx8_ddr_perf.o
obj-$(CONFIG_HISI_PMU) += hisilicon/
obj-$(CONFIG_QCOM_L2_PMU) += qcom_l2_pmu.o
obj-$(CONFIG_QCOM_L3_PMU) += qcom_l3_pmu.o
diff --git a/drivers/perf/fsl_imx8_ddr_perf.c b/drivers/perf/fsl_imx8_ddr_perf.c
new file mode 100644
index 0000000..3d02daf
--- /dev/null
+++ b/drivers/perf/fsl_imx8_ddr_perf.c
@@ -0,0 +1,573 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2017 NXP
+ * Copyright 2016 Freescale Semiconductor, Inc.
+ */
+
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_device.h>
+#include <linux/of_irq.h>
+#include <linux/perf_event.h>
+#include <linux/slab.h>
+
+#define COUNTER_CNTL 0x0
+#define COUNTER_READ 0x20
+
+#define COUNTER_DPCR1 0x30
+
+#define CNTL_OVER 0x1
+#define CNTL_CLEAR 0x2
+#define CNTL_EN 0x4
+#define CNTL_EN_MASK 0xFFFFFFFB
+#define CNTL_CLEAR_MASK 0xFFFFFFFD
+#define CNTL_OVER_MASK 0xFFFFFFFE
+
+#define CNTL_CSV_SHIFT 24
+#define CNTL_CSV_MASK (0xFF << CNTL_CSV_SHIFT)
+
+#define EVENT_CYCLES_ID 0
+#define EVENT_CYCLES_COUNTER 0
+#define NUM_COUNTER 4
+#define MAX_EVENT 3
+#define EVENT_AXI_READ 0x41
+#define EVENT_AXI_WRITE 0x42
+
+#define to_ddr_pmu(p) container_of(p, struct ddr_pmu, pmu)
+
+#define DDR_PERF_DEV_NAME "ddr_perf"
+
+static DEFINE_IDA(ddr_ida);
+
+PMU_EVENT_ATTR_STRING(cycles, ddr_perf_cycles, "event=0x00");
+PMU_EVENT_ATTR_STRING(selfresh, ddr_perf_selfresh, "event=0x01");
+PMU_EVENT_ATTR_STRING(read-access, ddr_perf_read_accesses, "event=0x04");
+PMU_EVENT_ATTR_STRING(write-access, ddr_perf_write_accesses, "event=0x05");
+PMU_EVENT_ATTR_STRING(read-queue-depth, ddr_perf_read_queue_depth,
+ "event=0x08");
+PMU_EVENT_ATTR_STRING(write-queue-depth, ddr_perf_write_queue_depth,
+ "event=0x09");
+PMU_EVENT_ATTR_STRING(lp-read-credit-cnt, ddr_perf_lp_read_credit_cnt,
+ "event=0x10");
+PMU_EVENT_ATTR_STRING(hp-read-credit-cnt, ddr_perf_hp_read_credit_cnt,
+ "event=0x11");
+PMU_EVENT_ATTR_STRING(write-credit-cnt, ddr_perf_write_credit_cnt,
+ "event=0x12");
+PMU_EVENT_ATTR_STRING(read-command, ddr_perf_read_command, "event=0x20");
+PMU_EVENT_ATTR_STRING(write-command, ddr_perf_write_command, "event=0x21");
+PMU_EVENT_ATTR_STRING(read-modify-write-command,
+ ddr_perf_read_modify_write_command, "event=0x22");
+PMU_EVENT_ATTR_STRING(hp-read, ddr_perf_hp_read, "event=0x23");
+PMU_EVENT_ATTR_STRING(hp-req-nodcredit, ddr_perf_hp_req_nocredit, "event=0x24");
+PMU_EVENT_ATTR_STRING(hp-xact-credit, ddr_perf_hp_xact_credit, "event=0x25");
+PMU_EVENT_ATTR_STRING(lp-req-nocredit, ddr_perf_lp_req_nocredit, "event=0x26");
+PMU_EVENT_ATTR_STRING(lp-xact-credit, ddr_perf_lp_xact_credit, "event=0x27");
+PMU_EVENT_ATTR_STRING(wr-xact-credit, ddr_perf_wr_xact_credit, "event=0x29");
+PMU_EVENT_ATTR_STRING(read-cycles, ddr_perf_read_cycles, "event=0x2a");
+PMU_EVENT_ATTR_STRING(write-cycles, ddr_perf_write_cycles, "event=0x2b");
+PMU_EVENT_ATTR_STRING(read-write-transition, ddr_perf_read_write_transition,
+ "event=0x30");
+PMU_EVENT_ATTR_STRING(precharge, ddr_perf_precharge, "event=0x31");
+PMU_EVENT_ATTR_STRING(activate, ddr_perf_activate, "event=0x32");
+PMU_EVENT_ATTR_STRING(load-mode, ddr_perf_load_mode, "event=0x33");
+PMU_EVENT_ATTR_STRING(mwr, ddr_perf_mwr, "event=0x34");
+PMU_EVENT_ATTR_STRING(read, ddr_perf_read, "event=0x35");
+PMU_EVENT_ATTR_STRING(read-activate, ddr_perf_read_activate, "event=0x36");
+PMU_EVENT_ATTR_STRING(refresh, ddr_perf_refresh, "event=0x37");
+PMU_EVENT_ATTR_STRING(write, ddr_perf_write, "event=0x38");
+PMU_EVENT_ATTR_STRING(raw-hazard, ddr_perf_raw_hazard, "event=0x39");
+
+PMU_EVENT_ATTR_STRING(axid-read, ddr_perf_axid_read,
+ "event="__stringify(EVENT_AXI_READ));
+PMU_EVENT_ATTR_STRING(axid-write, ddr_perf_axid_write,
+ "event="__stringify(EVENT_AXI_WRITE));
+
+#define DDR_CAP_AXI_ID 0x1
+
+static const struct of_device_id imx_ddr_pmu_dt_ids[] = {
+ { .compatible = "fsl,imx8-ddr-pmu",},
+ { .compatible = "fsl,imx8m-ddr-pmu", .data = (void *)DDR_CAP_AXI_ID},
+ { /* sentinel */ }
+};
+
+struct ddr_pmu {
+ struct pmu pmu;
+ void __iomem *base;
+ cpumask_t cpu;
+ struct hlist_node node;
+ struct device *dev;
+ struct perf_event *active_events[NUM_COUNTER];
+ int total_events;
+ enum cpuhp_state cpuhp_state;
+ bool cycles_active;
+ uintptr_t flags;
+};
+
+static ssize_t ddr_perf_cpumask_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct ddr_pmu *pmu = dev_get_drvdata(dev);
+
+ return cpumap_print_to_pagebuf(true, buf, &pmu->cpu);
+}
+
+static struct device_attribute ddr_perf_cpumask_attr =
+ __ATTR(cpumask, 0444, ddr_perf_cpumask_show, NULL);
+
+static struct attribute *ddr_perf_cpumask_attrs[] = {
+ &ddr_perf_cpumask_attr.attr,
+ NULL,
+};
+
+static struct attribute_group ddr_perf_cpumask_attr_group = {
+ .attrs = ddr_perf_cpumask_attrs,
+};
+
+static struct attribute *ddr_perf_events_attrs[] = {
+ &ddr_perf_cycles.attr.attr,
+ &ddr_perf_selfresh.attr.attr,
+ &ddr_perf_read_accesses.attr.attr,
+ &ddr_perf_write_accesses.attr.attr,
+ &ddr_perf_read_queue_depth.attr.attr,
+ &ddr_perf_write_queue_depth.attr.attr,
+ &ddr_perf_lp_read_credit_cnt.attr.attr,
+ &ddr_perf_hp_read_credit_cnt.attr.attr,
+ &ddr_perf_write_credit_cnt.attr.attr,
+ &ddr_perf_read_command.attr.attr,
+ &ddr_perf_write_command.attr.attr,
+ &ddr_perf_read_modify_write_command.attr.attr,
+ &ddr_perf_hp_read.attr.attr,
+ &ddr_perf_hp_req_nocredit.attr.attr,
+ &ddr_perf_hp_xact_credit.attr.attr,
+ &ddr_perf_lp_req_nocredit.attr.attr,
+ &ddr_perf_lp_xact_credit.attr.attr,
+ &ddr_perf_wr_xact_credit.attr.attr,
+ &ddr_perf_read_cycles.attr.attr,
+ &ddr_perf_write_cycles.attr.attr,
+ &ddr_perf_read_write_transition.attr.attr,
+ &ddr_perf_precharge.attr.attr,
+ &ddr_perf_activate.attr.attr,
+ &ddr_perf_load_mode.attr.attr,
+ &ddr_perf_mwr.attr.attr,
+ &ddr_perf_read.attr.attr,
+ &ddr_perf_read_activate.attr.attr,
+ &ddr_perf_refresh.attr.attr,
+ &ddr_perf_write.attr.attr,
+ &ddr_perf_raw_hazard.attr.attr,
+ &ddr_perf_axid_read.attr.attr,
+ &ddr_perf_axid_write.attr.attr,
+ NULL,
+};
+
+static struct attribute_group ddr_perf_events_attr_group = {
+ .name = "events",
+ .attrs = ddr_perf_events_attrs,
+};
+
+PMU_FORMAT_ATTR(event, "config:0-63");
+PMU_FORMAT_ATTR(axi_id, "config1:0-63");
+
+static struct attribute *ddr_perf_format_attrs[] = {
+ &format_attr_event.attr,
+ &format_attr_axi_id.attr,
+ NULL,
+};
+
+static struct attribute_group ddr_perf_format_attr_group = {
+ .name = "format",
+ .attrs = ddr_perf_format_attrs,
+};
+
+static const struct attribute_group *attr_groups[] = {
+ &ddr_perf_events_attr_group,
+ &ddr_perf_format_attr_group,
+ &ddr_perf_cpumask_attr_group,
+ NULL,
+};
+
+static u32 ddr_perf_alloc_counter(struct ddr_pmu *pmu, int event)
+{
+ int i;
+
+ /* Always map cycle event to counter 0
+ Cycles counter is dedicated for cycle event
+ can't used for the other events
+ */
+ if (event == EVENT_CYCLES_ID)
+ return EVENT_CYCLES_COUNTER;
+
+ for (i = 1; i < NUM_COUNTER; i++)
+ if (pmu->active_events[i] == NULL)
+ return i;
+
+ return -ENOENT;
+}
+
+static u32 ddr_perf_free_counter(struct ddr_pmu *pmu, int counter)
+{
+ if (counter < 0 || counter >= NUM_COUNTER)
+ return -ENOENT;
+
+ pmu->active_events[counter] = NULL;
+
+ return 0;
+}
+
+static u32 ddr_perf_read_counter(struct ddr_pmu *pmu, int counter)
+{
+ return readl(pmu->base + COUNTER_READ + counter * 4);
+}
+
+static int ddr_perf_event_init(struct perf_event *event)
+{
+ struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
+ struct hw_perf_event *hwc = &event->hw;
+
+ if (event->attr.type != event->pmu->type)
+ return -ENOENT;
+
+ if (is_sampling_event(event) || event->attach_state & PERF_ATTACH_TASK)
+ return -EOPNOTSUPP;
+
+ if (event->cpu < 0) {
+ dev_warn(pmu->dev, "Can't provide per-task data!\n");
+ return -EOPNOTSUPP;
+ }
+
+ if (event->attr.exclude_user ||
+ event->attr.exclude_kernel ||
+ event->attr.exclude_hv ||
+ event->attr.exclude_idle ||
+ event->attr.exclude_host ||
+ event->attr.exclude_guest ||
+ event->attr.sample_period)
+ return -EINVAL;
+
+ event->cpu = cpumask_first(&pmu->cpu);
+ hwc->idx = -1;
+
+ return 0;
+}
+
+
+static void ddr_perf_event_update(struct perf_event *event)
+{
+ struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
+ struct hw_perf_event *hwc = &event->hw;
+ u64 delta, prev_raw_count, new_raw_count;
+ int counter = hwc->idx;
+
+ do {
+ prev_raw_count = local64_read(&hwc->prev_count);
+ new_raw_count = ddr_perf_read_counter(pmu, counter);
+ } while (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
+ new_raw_count) != prev_raw_count);
+
+ delta = (new_raw_count - prev_raw_count) & 0xFFFFFFFF;
+
+ local64_add(delta, &event->count);
+}
+
+static void ddr_perf_event_enable(struct ddr_pmu *pmu, int config,
+ int counter, bool enable)
+{
+ u8 reg = counter * 4 + COUNTER_CNTL;
+ int val;
+
+ if (enable) {
+ /* must disable first, then enable again
+ * otherwise, cycle counter will not work
+ * if previous state is enabled.
+ */
+ writel(0, pmu->base + reg);
+ val = CNTL_EN | CNTL_CLEAR;
+ val |= (config << CNTL_CSV_SHIFT) & CNTL_CSV_MASK;
+ } else {
+ /* Disable counter */
+ val = readl(pmu->base + reg) & CNTL_EN_MASK;
+ }
+
+ writel(val, pmu->base + reg);
+
+ if (config == EVENT_CYCLES_ID)
+ pmu->cycles_active = enable;
+}
+
+static void ddr_perf_event_start(struct perf_event *event, int flags)
+{
+ struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
+ struct hw_perf_event *hwc = &event->hw;
+ int counter = hwc->idx;
+
+ if (pmu->flags & DDR_CAP_AXI_ID) {
+ if (event->attr.config == EVENT_AXI_READ ||
+ event->attr.config == EVENT_AXI_WRITE) {
+ int val = event->attr.config1;
+
+ writel(val, pmu->base + COUNTER_DPCR1);
+ }
+ }
+
+ local64_set(&hwc->prev_count, 0);
+
+ ddr_perf_event_enable(pmu, event->attr.config, counter, true);
+ /*
+ * If the cycles counter wasn't explicitly selected,
+ * we will enable it now.
+ * cycles counter must be enabled otherwise other counters will
+ * stopped.
+ */
+ if (counter > 0 && !pmu->cycles_active)
+ ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
+ EVENT_CYCLES_COUNTER, true);
+}
+
+static int ddr_perf_event_add(struct perf_event *event, int flags)
+{
+ struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
+ struct hw_perf_event *hwc = &event->hw;
+ int counter;
+ int cfg = event->attr.config;
+
+ counter = ddr_perf_alloc_counter(pmu, cfg);
+ if (counter < 0) {
+ dev_dbg(pmu->dev, "There are not enough counters\n");
+ return -EOPNOTSUPP;
+ }
+
+ pmu->active_events[counter] = event;
+ pmu->total_events++;
+ hwc->idx = counter;
+
+ if (flags & PERF_EF_START)
+ ddr_perf_event_start(event, flags);
+
+ local64_set(&hwc->prev_count, ddr_perf_read_counter(pmu, counter));
+
+ return 0;
+}
+
+static void ddr_perf_event_stop(struct perf_event *event, int flags)
+{
+ struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
+ struct hw_perf_event *hwc = &event->hw;
+ int counter = hwc->idx;
+
+ ddr_perf_event_enable(pmu, event->attr.config, counter, false);
+ ddr_perf_event_update(event);
+}
+
+static void ddr_perf_event_del(struct perf_event *event, int flags)
+{
+ struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
+ struct hw_perf_event *hwc = &event->hw;
+ int counter = hwc->idx;
+
+ ddr_perf_event_stop(event, PERF_EF_UPDATE);
+
+ ddr_perf_free_counter(pmu, counter);
+ pmu->total_events--;
+ hwc->idx = -1;
+
+ /* If all events have stopped, stop the cycles counter as well */
+ if ((pmu->total_events == 0) && pmu->cycles_active)
+ ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
+ EVENT_CYCLES_COUNTER, false);
+}
+
+static int ddr_perf_init(struct ddr_pmu *pmu, void __iomem *base,
+ struct device *dev)
+{
+ *pmu = (struct ddr_pmu) {
+ .pmu = (struct pmu) {
+ .task_ctx_nr = perf_invalid_context,
+ .attr_groups = attr_groups,
+ .event_init = ddr_perf_event_init,
+ .add = ddr_perf_event_add,
+ .del = ddr_perf_event_del,
+ .start = ddr_perf_event_start,
+ .stop = ddr_perf_event_stop,
+ .read = ddr_perf_event_update,
+ },
+ .base = base,
+ .dev = dev,
+ };
+
+ return ida_simple_get(&ddr_ida, 0, 0, GFP_KERNEL);
+}
+
+static irqreturn_t ddr_perf_irq_handler(int irq, void *p)
+{
+ int i;
+ u8 reg;
+ int val;
+ int counter;
+ struct ddr_pmu *pmu = (struct ddr_pmu *) p;
+ struct perf_event *event;
+
+ /* Only cycles counter overflowed can issue irq. all counters will
+ * be stopped when cycles counter overflow. but other counter don't stop
+ * when overflow happen. Update all of the local counter values,
+ * then reset the cycles counter, so the others can continue
+ * counting. cycles counter is fastest counter in all events. at last
+ * 4 times than other counters.
+ */
+ for (i = 0; i < NUM_COUNTER; i++) {
+
+ if (!pmu->active_events[i])
+ continue;
+
+ event = pmu->active_events[i];
+ counter = event->hw.idx;
+ reg = counter * 4 + COUNTER_CNTL;
+ val = readl(pmu->base + reg);
+
+ ddr_perf_event_update(event);
+
+ if (counter == EVENT_CYCLES_COUNTER) {
+ ddr_perf_event_enable(pmu,
+ EVENT_CYCLES_ID,
+ EVENT_CYCLES_COUNTER,
+ true);
+ ddr_perf_event_update(event);
+ }
+ }
+
+ return IRQ_HANDLED;
+}
+
+static int ddr_perf_offline_cpu(unsigned int cpu, struct hlist_node *node)
+{
+ struct ddr_pmu *pmu = hlist_entry_safe(node, struct ddr_pmu, node);
+ int target;
+
+ if (!cpumask_test_and_clear_cpu(cpu, &pmu->cpu))
+ return 0;
+
+ target = cpumask_any_but(cpu_online_mask, cpu);
+ if (target >= nr_cpu_ids)
+ return 0;
+
+ perf_pmu_migrate_context(&pmu->pmu, cpu, target);
+ cpumask_set_cpu(target, &pmu->cpu);
+
+ return 0;
+}
+
+static int ddr_perf_probe(struct platform_device *pdev)
+{
+ struct ddr_pmu *pmu;
+ struct device_node *np;
+ void __iomem *base;
+ struct resource *iomem;
+ char *name;
+ char *hpname;
+ int num;
+ int ret;
+ u32 irq;
+ const struct of_device_id *of_id =
+ of_match_device(imx_ddr_pmu_dt_ids, &pdev->dev);
+
+ iomem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+ base = devm_ioremap_resource(&pdev->dev, iomem);
+ if (IS_ERR(base))
+ return PTR_ERR(base);
+
+ np = pdev->dev.of_node;
+
+ pmu = devm_kzalloc(&pdev->dev, sizeof(*pmu), GFP_KERNEL);
+ if (!pmu)
+ return -ENOMEM;
+
+ num = ddr_perf_init(pmu, base, &pdev->dev);
+
+ name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "ddr%d", num);
+ if (!name)
+ return -ENOMEM;
+
+ hpname = devm_kasprintf(&pdev->dev, GFP_KERNEL,
+ "perf/imx/ddr%d:online", num);
+ if (!hpname)
+ return -ENOMEM;
+
+ pmu->flags = (uintptr_t) of_id->data;
+
+ cpumask_set_cpu(raw_smp_processor_id(), &pmu->cpu);
+ ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN, hpname, NULL,
+ ddr_perf_offline_cpu);
+
+ if (ret < 0) {
+ dev_err(&pdev->dev, "cpuhp_setup_state_multi failed\n");
+ goto ddr_perf_err;
+ }
+
+ pmu->cpuhp_state = ret;
+
+ /* Register the pmu instance for cpu hotplug */
+ cpuhp_state_add_instance_nocalls(pmu->cpuhp_state, &pmu->node);
+
+ ret = perf_pmu_register(&(pmu->pmu), name, -1);
+ if (ret)
+ goto ddr_perf_err;
+
+ /* Request irq */
+ irq = of_irq_get(np, 0);
+ if (irq < 0) {
+ dev_err(&pdev->dev, "Failed to get irq: %d", irq);
+ ret = irq;
+ goto ddr_perf_irq_err;
+ }
+
+ ret = devm_request_irq(&pdev->dev, irq,
+ ddr_perf_irq_handler,
+ IRQF_TRIGGER_RISING | IRQF_ONESHOT,
+ DDR_PERF_DEV_NAME,
+ pmu);
+ if (ret < 0) {
+ dev_err(&pdev->dev, "Request irq failed: %d", ret);
+ goto ddr_perf_irq_err;
+ }
+
+ return 0;
+
+ddr_perf_irq_err:
+ perf_pmu_unregister(&(pmu->pmu));
+
+ddr_perf_err:
+ if (pmu->cpuhp_state)
+ cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
+
+ dev_warn(&pdev->dev, "i.MX8 DDR Perf PMU failed (%d), disabled\n", ret);
+ return ret;
+}
+
+static int ddr_perf_remove(struct platform_device *pdev)
+{
+ struct ddr_pmu *pmu = platform_get_drvdata(pdev);
+
+ cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
+ perf_pmu_unregister(&pmu->pmu);
+
+ return 0;
+}
+
+static struct platform_driver imx_ddr_pmu_driver = {
+ .driver = {
+ .name = "imx-ddr-pmu",
+ .of_match_table = imx_ddr_pmu_dt_ids,
+ },
+ .probe = ddr_perf_probe,
+ .remove = ddr_perf_remove,
+};
+
+static int __init imx_ddr_pmu_init(void)
+{
+ return platform_driver_register(&imx_ddr_pmu_driver);
+}
+
+module_init(imx_ddr_pmu_init);
+
--
2.5.2
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 43+ messages in thread
* [PATCH V4 3/4] arm64: dts: imx8qxp: added ddr performance monitor nodes
2019-02-15 18:03 ` Frank Li
@ 2019-02-15 18:03 ` Frank Li
-1 siblings, 0 replies; 43+ messages in thread
From: Frank Li @ 2019-02-15 18:03 UTC (permalink / raw)
To: mark.rutland, will.deacon, shawnguo, s.hauer, kernel, festevam,
dl-linux-imx, robh+dt, Aisheng Dong, devicetree, lznuaa,
linux-arm-kernel
Cc: Frank Li
Add ddr performance monitor
Signed-off-by: Frank Li <Frank.Li@nxp.com>
---
Change from v3 to v4
* none
Change from v2 to v3
* ddr_pmu0 -> ddr-pmu
arch/arm64/boot/dts/freescale/imx8qxp.dtsi | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/arch/arm64/boot/dts/freescale/imx8qxp.dtsi b/arch/arm64/boot/dts/freescale/imx8qxp.dtsi
index 4c3dd95..ac0691e 100644
--- a/arch/arm64/boot/dts/freescale/imx8qxp.dtsi
+++ b/arch/arm64/boot/dts/freescale/imx8qxp.dtsi
@@ -79,6 +79,13 @@
interrupts = <GIC_PPI 7 IRQ_TYPE_LEVEL_HIGH>;
};
+ ddr-pmu@5c020000 {
+ compatible = "fsl,imx8-ddr-pmu";
+ reg = <0x0 0x5c020000 0x0 0x10000>;
+ interrupt-parent = <&gic>;
+ interrupts = <GIC_SPI 131 IRQ_TYPE_LEVEL_HIGH>;
+ };
+
psci {
compatible = "arm,psci-1.0";
method = "smc";
--
2.5.2
^ permalink raw reply related [flat|nested] 43+ messages in thread
* [PATCH V4 3/4] arm64: dts: imx8qxp: added ddr performance monitor nodes
@ 2019-02-15 18:03 ` Frank Li
0 siblings, 0 replies; 43+ messages in thread
From: Frank Li @ 2019-02-15 18:03 UTC (permalink / raw)
To: mark.rutland, will.deacon, shawnguo, s.hauer, kernel, festevam,
dl-linux-imx, robh+dt, Aisheng Dong, devicetree, lznuaa,
linux-arm-kernel
Cc: Frank Li
Add ddr performance monitor
Signed-off-by: Frank Li <Frank.Li@nxp.com>
---
Change from v3 to v4
* none
Change from v2 to v3
* ddr_pmu0 -> ddr-pmu
arch/arm64/boot/dts/freescale/imx8qxp.dtsi | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/arch/arm64/boot/dts/freescale/imx8qxp.dtsi b/arch/arm64/boot/dts/freescale/imx8qxp.dtsi
index 4c3dd95..ac0691e 100644
--- a/arch/arm64/boot/dts/freescale/imx8qxp.dtsi
+++ b/arch/arm64/boot/dts/freescale/imx8qxp.dtsi
@@ -79,6 +79,13 @@
interrupts = <GIC_PPI 7 IRQ_TYPE_LEVEL_HIGH>;
};
+ ddr-pmu@5c020000 {
+ compatible = "fsl,imx8-ddr-pmu";
+ reg = <0x0 0x5c020000 0x0 0x10000>;
+ interrupt-parent = <&gic>;
+ interrupts = <GIC_SPI 131 IRQ_TYPE_LEVEL_HIGH>;
+ };
+
psci {
compatible = "arm,psci-1.0";
method = "smc";
--
2.5.2
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 43+ messages in thread
* [PATCH V4 4/4] MAINTAINERS: Added imx DDR performonitor driver maintainer information
2019-02-15 18:03 ` Frank Li
@ 2019-02-15 18:03 ` Frank Li
-1 siblings, 0 replies; 43+ messages in thread
From: Frank Li @ 2019-02-15 18:03 UTC (permalink / raw)
To: mark.rutland, will.deacon, shawnguo, s.hauer, kernel, festevam,
dl-linux-imx, robh+dt, Aisheng Dong, devicetree, lznuaa,
linux-arm-kernel
Cc: Frank Li
Add DDR perf counter driver maintainer information
Signed-off-by: Frank Li <Frank.Li@nxp.com>
---
No change from v1 to v4
MAINTAINERS | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 6ba068e..687be0f 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -6163,6 +6163,13 @@ L: linux-i2c@vger.kernel.org
S: Maintained
F: drivers/i2c/busses/i2c-cpm.c
+FREESCALE IMX DDR Performance Monitor DRIVER
+M: Frank Li <Frank.li@nxp.com>
+L: linux-arm-kernel@lists.infradead.org
+S: Maintained
+F: drivers/perf/fsl_imx8_ddr_perf.c
+F: Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
+
FREESCALE IMX LPI2C DRIVER
M: Dong Aisheng <aisheng.dong@nxp.com>
L: linux-i2c@vger.kernel.org
--
2.5.2
^ permalink raw reply related [flat|nested] 43+ messages in thread
* [PATCH V4 4/4] MAINTAINERS: Added imx DDR performonitor driver maintainer information
@ 2019-02-15 18:03 ` Frank Li
0 siblings, 0 replies; 43+ messages in thread
From: Frank Li @ 2019-02-15 18:03 UTC (permalink / raw)
To: mark.rutland, will.deacon, shawnguo, s.hauer, kernel, festevam,
dl-linux-imx, robh+dt, Aisheng Dong, devicetree, lznuaa,
linux-arm-kernel
Cc: Frank Li
Add DDR perf counter driver maintainer information
Signed-off-by: Frank Li <Frank.Li@nxp.com>
---
No change from v1 to v4
MAINTAINERS | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 6ba068e..687be0f 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -6163,6 +6163,13 @@ L: linux-i2c@vger.kernel.org
S: Maintained
F: drivers/i2c/busses/i2c-cpm.c
+FREESCALE IMX DDR Performance Monitor DRIVER
+M: Frank Li <Frank.li@nxp.com>
+L: linux-arm-kernel@lists.infradead.org
+S: Maintained
+F: drivers/perf/fsl_imx8_ddr_perf.c
+F: Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
+
FREESCALE IMX LPI2C DRIVER
M: Dong Aisheng <aisheng.dong@nxp.com>
L: linux-i2c@vger.kernel.org
--
2.5.2
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 43+ messages in thread
* Re: [PATCH V4 4/4] MAINTAINERS: Added imx DDR performonitor driver maintainer information
2019-02-15 18:03 ` Frank Li
@ 2019-02-15 20:01 ` Uwe Kleine-König
-1 siblings, 0 replies; 43+ messages in thread
From: Uwe Kleine-König @ 2019-02-15 20:01 UTC (permalink / raw)
To: Frank Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
will.deacon, robh+dt, dl-linux-imx, kernel, lznuaa, shawnguo,
linux-arm-kernel
On Fri, Feb 15, 2019 at 06:03:16PM +0000, Frank Li wrote:
> Add DDR perf counter driver maintainer information
>
> Signed-off-by: Frank Li <Frank.Li@nxp.com>
> ---
> No change from v1 to v4
>
> MAINTAINERS | 7 +++++++
> 1 file changed, 7 insertions(+)
>
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 6ba068e..687be0f 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -6163,6 +6163,13 @@ L: linux-i2c@vger.kernel.org
> S: Maintained
> F: drivers/i2c/busses/i2c-cpm.c
>
> +FREESCALE IMX DDR Performance Monitor DRIVER
> +M: Frank Li <Frank.li@nxp.com>
> +L: linux-arm-kernel@lists.infradead.org
> +S: Maintained
> +F: drivers/perf/fsl_imx8_ddr_perf.c
> +F: Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
> +
With da8b7f0fb02b ("MAINTAINERS: add all files matching "imx" and "mxs"
to the IMX entry") that sits currently in next the "ARM/FREESCALE IMX /
MXC ARM ARCHITECTURE" also matches for these two files.
Is this good here or not?
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-König |
Industrial Linux Solutions | http://www.pengutronix.de/ |
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 4/4] MAINTAINERS: Added imx DDR performonitor driver maintainer information
@ 2019-02-15 20:01 ` Uwe Kleine-König
0 siblings, 0 replies; 43+ messages in thread
From: Uwe Kleine-König @ 2019-02-15 20:01 UTC (permalink / raw)
To: Frank Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
will.deacon, robh+dt, dl-linux-imx, kernel, lznuaa, shawnguo,
linux-arm-kernel
On Fri, Feb 15, 2019 at 06:03:16PM +0000, Frank Li wrote:
> Add DDR perf counter driver maintainer information
>
> Signed-off-by: Frank Li <Frank.Li@nxp.com>
> ---
> No change from v1 to v4
>
> MAINTAINERS | 7 +++++++
> 1 file changed, 7 insertions(+)
>
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 6ba068e..687be0f 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -6163,6 +6163,13 @@ L: linux-i2c@vger.kernel.org
> S: Maintained
> F: drivers/i2c/busses/i2c-cpm.c
>
> +FREESCALE IMX DDR Performance Monitor DRIVER
> +M: Frank Li <Frank.li@nxp.com>
> +L: linux-arm-kernel@lists.infradead.org
> +S: Maintained
> +F: drivers/perf/fsl_imx8_ddr_perf.c
> +F: Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
> +
With da8b7f0fb02b ("MAINTAINERS: add all files matching "imx" and "mxs"
to the IMX entry") that sits currently in next the "ARM/FREESCALE IMX /
MXC ARM ARCHITECTURE" also matches for these two files.
Is this good here or not?
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-König |
Industrial Linux Solutions | http://www.pengutronix.de/ |
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 4/4] MAINTAINERS: Added imx DDR performonitor driver maintainer information
2019-02-15 20:01 ` Uwe Kleine-König
(?)
@ 2019-02-15 21:13 ` Zhi Li
2019-02-15 21:44 ` Uwe Kleine-König
-1 siblings, 1 reply; 43+ messages in thread
From: Zhi Li @ 2019-02-15 21:13 UTC (permalink / raw)
To: Uwe Kleine-König
Cc: mark.rutland, Aisheng Dong, festevam, devicetree, Frank Li,
s.hauer, will.deacon, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Feb 15, 2019 at 2:01 PM Uwe Kleine-König
<u.kleine-koenig@pengutronix.de> wrote:
>
> On Fri, Feb 15, 2019 at 06:03:16PM +0000, Frank Li wrote:
> > Add DDR perf counter driver maintainer information
> >
> > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > ---
> > No change from v1 to v4
> >
> > MAINTAINERS | 7 +++++++
> > 1 file changed, 7 insertions(+)
> >
> > diff --git a/MAINTAINERS b/MAINTAINERS
> > index 6ba068e..687be0f 100644
> > --- a/MAINTAINERS
> > +++ b/MAINTAINERS
> > @@ -6163,6 +6163,13 @@ L: linux-i2c@vger.kernel.org
> > S: Maintained
> > F: drivers/i2c/busses/i2c-cpm.c
> >
> > +FREESCALE IMX DDR Performance Monitor DRIVER
> > +M: Frank Li <Frank.li@nxp.com>
> > +L: linux-arm-kernel@lists.infradead.org
> > +S: Maintained
> > +F: drivers/perf/fsl_imx8_ddr_perf.c
> > +F: Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
> > +
>
> With da8b7f0fb02b ("MAINTAINERS: add all files matching "imx" and "mxs"
> to the IMX entry") that sits currently in next the "ARM/FREESCALE IMX /
> MXC ARM ARCHITECTURE" also matches for these two files.
>
> Is this good here or not?
Yes! what's do you means?
Needn't these entry because da8b7f0fb02b?
>
> Best regards
> Uwe
>
> --
> Pengutronix e.K. | Uwe Kleine-König |
> Industrial Linux Solutions | http://www.pengutronix.de/ |
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 4/4] MAINTAINERS: Added imx DDR performonitor driver maintainer information
2019-02-15 21:13 ` Zhi Li
@ 2019-02-15 21:44 ` Uwe Kleine-König
0 siblings, 0 replies; 43+ messages in thread
From: Uwe Kleine-König @ 2019-02-15 21:44 UTC (permalink / raw)
To: Zhi Li
Cc: mark.rutland, Aisheng Dong, festevam, devicetree, Frank Li,
s.hauer, will.deacon, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Feb 15, 2019 at 03:13:00PM -0600, Zhi Li wrote:
> On Fri, Feb 15, 2019 at 2:01 PM Uwe Kleine-König
> <u.kleine-koenig@pengutronix.de> wrote:
> >
> > On Fri, Feb 15, 2019 at 06:03:16PM +0000, Frank Li wrote:
> > > Add DDR perf counter driver maintainer information
> > >
> > > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > > ---
> > > No change from v1 to v4
> > >
> > > MAINTAINERS | 7 +++++++
> > > 1 file changed, 7 insertions(+)
> > >
> > > diff --git a/MAINTAINERS b/MAINTAINERS
> > > index 6ba068e..687be0f 100644
> > > --- a/MAINTAINERS
> > > +++ b/MAINTAINERS
> > > @@ -6163,6 +6163,13 @@ L: linux-i2c@vger.kernel.org
> > > S: Maintained
> > > F: drivers/i2c/busses/i2c-cpm.c
> > >
> > > +FREESCALE IMX DDR Performance Monitor DRIVER
> > > +M: Frank Li <Frank.li@nxp.com>
> > > +L: linux-arm-kernel@lists.infradead.org
> > > +S: Maintained
> > > +F: drivers/perf/fsl_imx8_ddr_perf.c
> > > +F: Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
> > > +
> >
> > With da8b7f0fb02b ("MAINTAINERS: add all files matching "imx" and "mxs"
> > to the IMX entry") that sits currently in next the "ARM/FREESCALE IMX /
> > MXC ARM ARCHITECTURE" also matches for these two files.
> >
> > Is this good here or not?
>
> Yes! what's do you means?
> Needn't these entry because da8b7f0fb02b?
It's just that we then will have two entries that will match. In a tree
where I have both, this series and da8b7f0fb02b I see:
$ scripts/get_maintainer.pl -f drivers/perf/fsl_imx8_ddr_perf.c
Frank Li <Frank.li@nxp.com> (maintainer:FREESCALE IMX DDR Performance Monitor DRIVER)
Will Deacon <will.deacon@arm.com> (maintainer:ARM PMU PROFILING AND DEBUGGING)
Mark Rutland <mark.rutland@arm.com> (maintainer:ARM PMU PROFILING AND DEBUGGING)
Shawn Guo <shawnguo@kernel.org> (maintainer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
Sascha Hauer <s.hauer@pengutronix.de> (maintainer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
Pengutronix Kernel Team <kernel@pengutronix.de> (reviewer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
Fabio Estevam <fabio.estevam@nxp.com> (reviewer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
NXP Linux Team <linux-imx@nxp.com> (reviewer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
linux-arm-kernel@lists.infradead.org (open list:FREESCALE IMX DDR Performance Monitor DRIVER)
linux-kernel@vger.kernel.org (open list)
That's not necessarily a bad thing, but something I want people to make
aware of.
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-König |
Industrial Linux Solutions | http://www.pengutronix.de/ |
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 4/4] MAINTAINERS: Added imx DDR performonitor driver maintainer information
@ 2019-02-15 21:44 ` Uwe Kleine-König
0 siblings, 0 replies; 43+ messages in thread
From: Uwe Kleine-König @ 2019-02-15 21:44 UTC (permalink / raw)
To: Zhi Li
Cc: mark.rutland, Aisheng Dong, festevam, devicetree, Frank Li,
s.hauer, will.deacon, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Feb 15, 2019 at 03:13:00PM -0600, Zhi Li wrote:
> On Fri, Feb 15, 2019 at 2:01 PM Uwe Kleine-König
> <u.kleine-koenig@pengutronix.de> wrote:
> >
> > On Fri, Feb 15, 2019 at 06:03:16PM +0000, Frank Li wrote:
> > > Add DDR perf counter driver maintainer information
> > >
> > > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > > ---
> > > No change from v1 to v4
> > >
> > > MAINTAINERS | 7 +++++++
> > > 1 file changed, 7 insertions(+)
> > >
> > > diff --git a/MAINTAINERS b/MAINTAINERS
> > > index 6ba068e..687be0f 100644
> > > --- a/MAINTAINERS
> > > +++ b/MAINTAINERS
> > > @@ -6163,6 +6163,13 @@ L: linux-i2c@vger.kernel.org
> > > S: Maintained
> > > F: drivers/i2c/busses/i2c-cpm.c
> > >
> > > +FREESCALE IMX DDR Performance Monitor DRIVER
> > > +M: Frank Li <Frank.li@nxp.com>
> > > +L: linux-arm-kernel@lists.infradead.org
> > > +S: Maintained
> > > +F: drivers/perf/fsl_imx8_ddr_perf.c
> > > +F: Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
> > > +
> >
> > With da8b7f0fb02b ("MAINTAINERS: add all files matching "imx" and "mxs"
> > to the IMX entry") that sits currently in next the "ARM/FREESCALE IMX /
> > MXC ARM ARCHITECTURE" also matches for these two files.
> >
> > Is this good here or not?
>
> Yes! what's do you means?
> Needn't these entry because da8b7f0fb02b?
It's just that we then will have two entries that will match. In a tree
where I have both, this series and da8b7f0fb02b I see:
$ scripts/get_maintainer.pl -f drivers/perf/fsl_imx8_ddr_perf.c
Frank Li <Frank.li@nxp.com> (maintainer:FREESCALE IMX DDR Performance Monitor DRIVER)
Will Deacon <will.deacon@arm.com> (maintainer:ARM PMU PROFILING AND DEBUGGING)
Mark Rutland <mark.rutland@arm.com> (maintainer:ARM PMU PROFILING AND DEBUGGING)
Shawn Guo <shawnguo@kernel.org> (maintainer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
Sascha Hauer <s.hauer@pengutronix.de> (maintainer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
Pengutronix Kernel Team <kernel@pengutronix.de> (reviewer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
Fabio Estevam <fabio.estevam@nxp.com> (reviewer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
NXP Linux Team <linux-imx@nxp.com> (reviewer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
linux-arm-kernel@lists.infradead.org (open list:FREESCALE IMX DDR Performance Monitor DRIVER)
linux-kernel@vger.kernel.org (open list)
That's not necessarily a bad thing, but something I want people to make
aware of.
Best regards
Uwe
--
Pengutronix e.K. | Uwe Kleine-König |
Industrial Linux Solutions | http://www.pengutronix.de/ |
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 4/4] MAINTAINERS: Added imx DDR performonitor driver maintainer information
2019-02-15 21:44 ` Uwe Kleine-König
@ 2019-02-18 1:22 ` Shawn Guo
-1 siblings, 0 replies; 43+ messages in thread
From: Shawn Guo @ 2019-02-18 1:22 UTC (permalink / raw)
To: Uwe Kleine-König
Cc: mark.rutland, Aisheng Dong, devicetree, will.deacon, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, Zhi Li, festevam,
linux-arm-kernel
On Fri, Feb 15, 2019 at 10:44:44PM +0100, Uwe Kleine-König wrote:
> On Fri, Feb 15, 2019 at 03:13:00PM -0600, Zhi Li wrote:
> > On Fri, Feb 15, 2019 at 2:01 PM Uwe Kleine-König
> > <u.kleine-koenig@pengutronix.de> wrote:
> > >
> > > On Fri, Feb 15, 2019 at 06:03:16PM +0000, Frank Li wrote:
> > > > Add DDR perf counter driver maintainer information
> > > >
> > > > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > > > ---
> > > > No change from v1 to v4
> > > >
> > > > MAINTAINERS | 7 +++++++
> > > > 1 file changed, 7 insertions(+)
> > > >
> > > > diff --git a/MAINTAINERS b/MAINTAINERS
> > > > index 6ba068e..687be0f 100644
> > > > --- a/MAINTAINERS
> > > > +++ b/MAINTAINERS
> > > > @@ -6163,6 +6163,13 @@ L: linux-i2c@vger.kernel.org
> > > > S: Maintained
> > > > F: drivers/i2c/busses/i2c-cpm.c
> > > >
> > > > +FREESCALE IMX DDR Performance Monitor DRIVER
> > > > +M: Frank Li <Frank.li@nxp.com>
> > > > +L: linux-arm-kernel@lists.infradead.org
> > > > +S: Maintained
> > > > +F: drivers/perf/fsl_imx8_ddr_perf.c
> > > > +F: Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
> > > > +
> > >
> > > With da8b7f0fb02b ("MAINTAINERS: add all files matching "imx" and "mxs"
> > > to the IMX entry") that sits currently in next the "ARM/FREESCALE IMX /
> > > MXC ARM ARCHITECTURE" also matches for these two files.
> > >
> > > Is this good here or not?
> >
> > Yes! what's do you means?
> > Needn't these entry because da8b7f0fb02b?
>
> It's just that we then will have two entries that will match. In a tree
> where I have both, this series and da8b7f0fb02b I see:
>
> $ scripts/get_maintainer.pl -f drivers/perf/fsl_imx8_ddr_perf.c
> Frank Li <Frank.li@nxp.com> (maintainer:FREESCALE IMX DDR Performance Monitor DRIVER)
> Will Deacon <will.deacon@arm.com> (maintainer:ARM PMU PROFILING AND DEBUGGING)
> Mark Rutland <mark.rutland@arm.com> (maintainer:ARM PMU PROFILING AND DEBUGGING)
> Shawn Guo <shawnguo@kernel.org> (maintainer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
> Sascha Hauer <s.hauer@pengutronix.de> (maintainer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
> Pengutronix Kernel Team <kernel@pengutronix.de> (reviewer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
> Fabio Estevam <fabio.estevam@nxp.com> (reviewer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
> NXP Linux Team <linux-imx@nxp.com> (reviewer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
> linux-arm-kernel@lists.infradead.org (open list:FREESCALE IMX DDR Performance Monitor DRIVER)
> linux-kernel@vger.kernel.org (open list)
>
> That's not necessarily a bad thing, but something I want people to make
> aware of.
For me, that's actually a good thing. We, i.MX platform folks, get
copied on i.MX related driver changes in the right order - after driver
maintainers.
Shawn
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 4/4] MAINTAINERS: Added imx DDR performonitor driver maintainer information
@ 2019-02-18 1:22 ` Shawn Guo
0 siblings, 0 replies; 43+ messages in thread
From: Shawn Guo @ 2019-02-18 1:22 UTC (permalink / raw)
To: Uwe Kleine-König
Cc: mark.rutland, Aisheng Dong, devicetree, will.deacon, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, Zhi Li, festevam,
linux-arm-kernel
On Fri, Feb 15, 2019 at 10:44:44PM +0100, Uwe Kleine-König wrote:
> On Fri, Feb 15, 2019 at 03:13:00PM -0600, Zhi Li wrote:
> > On Fri, Feb 15, 2019 at 2:01 PM Uwe Kleine-König
> > <u.kleine-koenig@pengutronix.de> wrote:
> > >
> > > On Fri, Feb 15, 2019 at 06:03:16PM +0000, Frank Li wrote:
> > > > Add DDR perf counter driver maintainer information
> > > >
> > > > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > > > ---
> > > > No change from v1 to v4
> > > >
> > > > MAINTAINERS | 7 +++++++
> > > > 1 file changed, 7 insertions(+)
> > > >
> > > > diff --git a/MAINTAINERS b/MAINTAINERS
> > > > index 6ba068e..687be0f 100644
> > > > --- a/MAINTAINERS
> > > > +++ b/MAINTAINERS
> > > > @@ -6163,6 +6163,13 @@ L: linux-i2c@vger.kernel.org
> > > > S: Maintained
> > > > F: drivers/i2c/busses/i2c-cpm.c
> > > >
> > > > +FREESCALE IMX DDR Performance Monitor DRIVER
> > > > +M: Frank Li <Frank.li@nxp.com>
> > > > +L: linux-arm-kernel@lists.infradead.org
> > > > +S: Maintained
> > > > +F: drivers/perf/fsl_imx8_ddr_perf.c
> > > > +F: Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
> > > > +
> > >
> > > With da8b7f0fb02b ("MAINTAINERS: add all files matching "imx" and "mxs"
> > > to the IMX entry") that sits currently in next the "ARM/FREESCALE IMX /
> > > MXC ARM ARCHITECTURE" also matches for these two files.
> > >
> > > Is this good here or not?
> >
> > Yes! what's do you means?
> > Needn't these entry because da8b7f0fb02b?
>
> It's just that we then will have two entries that will match. In a tree
> where I have both, this series and da8b7f0fb02b I see:
>
> $ scripts/get_maintainer.pl -f drivers/perf/fsl_imx8_ddr_perf.c
> Frank Li <Frank.li@nxp.com> (maintainer:FREESCALE IMX DDR Performance Monitor DRIVER)
> Will Deacon <will.deacon@arm.com> (maintainer:ARM PMU PROFILING AND DEBUGGING)
> Mark Rutland <mark.rutland@arm.com> (maintainer:ARM PMU PROFILING AND DEBUGGING)
> Shawn Guo <shawnguo@kernel.org> (maintainer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
> Sascha Hauer <s.hauer@pengutronix.de> (maintainer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
> Pengutronix Kernel Team <kernel@pengutronix.de> (reviewer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
> Fabio Estevam <fabio.estevam@nxp.com> (reviewer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
> NXP Linux Team <linux-imx@nxp.com> (reviewer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
> linux-arm-kernel@lists.infradead.org (open list:FREESCALE IMX DDR Performance Monitor DRIVER)
> linux-kernel@vger.kernel.org (open list)
>
> That's not necessarily a bad thing, but something I want people to make
> aware of.
For me, that's actually a good thing. We, i.MX platform folks, get
copied on i.MX related driver changes in the right order - after driver
maintainers.
Shawn
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 1/4] dt-bindings: perf: imx8-ddr: add imx8qxp ddr performance monitor
2019-02-15 18:03 ` Frank Li
@ 2019-02-18 16:56 ` Rob Herring
-1 siblings, 0 replies; 43+ messages in thread
From: Rob Herring @ 2019-02-18 16:56 UTC (permalink / raw)
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
will.deacon, robh+dt, dl-linux-imx, kernel, lznuaa, shawnguo,
Frank Li, linux-arm-kernel
On Fri, 15 Feb 2019 18:03:08 +0000, Frank Li wrote:
> Added binding doc for imx8qxp ddr performance monitor
>
> Signed-off-by: Frank Li <Frank.Li@nxp.com>
> ---
> Change from v4 to v4
> * remove "standard xxx"
>
> Change from v2 to v3
> * ddr_pmu0 -> ddr-pmu
>
> .../devicetree/bindings/perf/fsl-imx-ddr.txt | 22 ++++++++++++++++++++++
> 1 file changed, 22 insertions(+)
> create mode 100644 Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
>
Reviewed-by: Rob Herring <robh@kernel.org>
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 1/4] dt-bindings: perf: imx8-ddr: add imx8qxp ddr performance monitor
@ 2019-02-18 16:56 ` Rob Herring
0 siblings, 0 replies; 43+ messages in thread
From: Rob Herring @ 2019-02-18 16:56 UTC (permalink / raw)
To: Frank Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
will.deacon, robh+dt, dl-linux-imx, kernel, lznuaa, shawnguo,
Frank Li, linux-arm-kernel
On Fri, 15 Feb 2019 18:03:08 +0000, Frank Li wrote:
> Added binding doc for imx8qxp ddr performance monitor
>
> Signed-off-by: Frank Li <Frank.Li@nxp.com>
> ---
> Change from v4 to v4
> * remove "standard xxx"
>
> Change from v2 to v3
> * ddr_pmu0 -> ddr-pmu
>
> .../devicetree/bindings/perf/fsl-imx-ddr.txt | 22 ++++++++++++++++++++++
> 1 file changed, 22 insertions(+)
> create mode 100644 Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
>
Reviewed-by: Rob Herring <robh@kernel.org>
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 3/4] arm64: dts: imx8qxp: added ddr performance monitor nodes
2019-02-15 18:03 ` Frank Li
@ 2019-02-18 17:38 ` Fabio Estevam
-1 siblings, 0 replies; 43+ messages in thread
From: Fabio Estevam @ 2019-02-18 17:38 UTC (permalink / raw)
To: Frank Li
Cc: mark.rutland, Aisheng Dong, devicetree, s.hauer, will.deacon,
robh+dt, dl-linux-imx, kernel, lznuaa, shawnguo,
linux-arm-kernel
On Fri, Feb 15, 2019 at 4:03 PM Frank Li <frank.li@nxp.com> wrote:
>
> Add ddr performance monitor
>
> Signed-off-by: Frank Li <Frank.Li@nxp.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 3/4] arm64: dts: imx8qxp: added ddr performance monitor nodes
@ 2019-02-18 17:38 ` Fabio Estevam
0 siblings, 0 replies; 43+ messages in thread
From: Fabio Estevam @ 2019-02-18 17:38 UTC (permalink / raw)
To: Frank Li
Cc: mark.rutland, Aisheng Dong, devicetree, s.hauer, will.deacon,
robh+dt, dl-linux-imx, kernel, lznuaa, shawnguo,
linux-arm-kernel
On Fri, Feb 15, 2019 at 4:03 PM Frank Li <frank.li@nxp.com> wrote:
>
> Add ddr performance monitor
>
> Signed-off-by: Frank Li <Frank.Li@nxp.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* RE: [PATCH V4 4/4] MAINTAINERS: Added imx DDR performonitor driver maintainer information
2019-02-15 21:44 ` Uwe Kleine-König
@ 2019-02-20 6:28 ` Aisheng Dong
-1 siblings, 0 replies; 43+ messages in thread
From: Aisheng Dong @ 2019-02-20 6:28 UTC (permalink / raw)
To: Uwe Kleine-König, Zhi Li
Cc: mark.rutland, devicetree, festevam, Frank Li, s.hauer,
will.deacon, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
> From: Uwe Kleine-König [mailto:u.kleine-koenig@pengutronix.de]
> Sent: Saturday, February 16, 2019 5:45 AM
>
> On Fri, Feb 15, 2019 at 03:13:00PM -0600, Zhi Li wrote:
> > On Fri, Feb 15, 2019 at 2:01 PM Uwe Kleine-König
> > <u.kleine-koenig@pengutronix.de> wrote:
> > >
> > > On Fri, Feb 15, 2019 at 06:03:16PM +0000, Frank Li wrote:
> > > > Add DDR perf counter driver maintainer information
> > > >
> > > > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > > > ---
> > > > No change from v1 to v4
> > > >
> > > > MAINTAINERS | 7 +++++++
> > > > 1 file changed, 7 insertions(+)
> > > >
> > > > diff --git a/MAINTAINERS b/MAINTAINERS index 6ba068e..687be0f
> > > > 100644
> > > > --- a/MAINTAINERS
> > > > +++ b/MAINTAINERS
> > > > @@ -6163,6 +6163,13 @@ L: linux-i2c@vger.kernel.org
> > > > S: Maintained
> > > > F: drivers/i2c/busses/i2c-cpm.c
> > > >
> > > > +FREESCALE IMX DDR Performance Monitor DRIVER
> > > > +M: Frank Li <Frank.li@nxp.com>
> > > > +L: linux-arm-kernel@lists.infradead.org
> > > > +S: Maintained
> > > > +F: drivers/perf/fsl_imx8_ddr_perf.c
> > > > +F: Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
> > > > +
> > >
> > > With da8b7f0fb02b ("MAINTAINERS: add all files matching "imx" and "mxs"
> > > to the IMX entry") that sits currently in next the "ARM/FREESCALE
> > > IMX / MXC ARM ARCHITECTURE" also matches for these two files.
> > >
> > > Is this good here or not?
> >
> > Yes! what's do you means?
> > Needn't these entry because da8b7f0fb02b?
>
> It's just that we then will have two entries that will match. In a tree where I
> have both, this series and da8b7f0fb02b I see:
>
> $ scripts/get_maintainer.pl -f drivers/perf/fsl_imx8_ddr_perf.c
> Frank Li <Frank.li@nxp.com> (maintainer:FREESCALE IMX DDR
> Performance Monitor DRIVER)
> Will Deacon <will.deacon@arm.com> (maintainer:ARM PMU PROFILING
> AND DEBUGGING)
> Mark Rutland <mark.rutland@arm.com> (maintainer:ARM PMU
> PROFILING AND DEBUGGING)
> Shawn Guo <shawnguo@kernel.org> (maintainer:ARM/FREESCALE IMX /
> MXC ARM ARCHITECTURE)
> Sascha Hauer <s.hauer@pengutronix.de> (maintainer:ARM/FREESCALE
> IMX / MXC ARM ARCHITECTURE)
> Pengutronix Kernel Team <kernel@pengutronix.de>
> (reviewer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
> Fabio Estevam <fabio.estevam@nxp.com> (reviewer:ARM/FREESCALE IMX
> / MXC ARM ARCHITECTURE)
> NXP Linux Team <linux-imx@nxp.com> (reviewer:ARM/FREESCALE IMX /
> MXC ARM ARCHITECTURE)
> linux-arm-kernel@lists.infradead.org (open list:FREESCALE IMX DDR
> Performance Monitor DRIVER)
> linux-kernel@vger.kernel.org (open list)
>
> That's not necessarily a bad thing, but something I want people to make aware
> of.
>
IMHO it's better to have a more specific maintainer in the MAINTAINER file
for a specific module.
Don't reply on the general pattern to filter all IMX patches to send ARCH
Maintainers to help review.
Regards
Dong Aisheng
> Best regards
> Uwe
>
> --
> Pengutronix e.K. | Uwe Kleine-König
> |
> Industrial Linux Solutions |
> https://emea01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.
> pengutronix.de%2F&data=02%7C01%7Caisheng.dong%40nxp.com%7C88
> f88bc0bd7e4812d00008d6938ed1af%7C686ea1d3bc2b4c6fa92cd99c5c3016
> 35%7C0%7C0%7C636858638948593375&sdata=KwEG5SXUsLCgUU9pg8
> hnZ9JQq4%2FPGHA6EFhmF%2Bcmlo8%3D&reserved=0 |
^ permalink raw reply [flat|nested] 43+ messages in thread
* RE: [PATCH V4 4/4] MAINTAINERS: Added imx DDR performonitor driver maintainer information
@ 2019-02-20 6:28 ` Aisheng Dong
0 siblings, 0 replies; 43+ messages in thread
From: Aisheng Dong @ 2019-02-20 6:28 UTC (permalink / raw)
To: Uwe Kleine-König, Zhi Li
Cc: mark.rutland, devicetree, festevam, Frank Li, s.hauer,
will.deacon, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
> From: Uwe Kleine-König [mailto:u.kleine-koenig@pengutronix.de]
> Sent: Saturday, February 16, 2019 5:45 AM
>
> On Fri, Feb 15, 2019 at 03:13:00PM -0600, Zhi Li wrote:
> > On Fri, Feb 15, 2019 at 2:01 PM Uwe Kleine-König
> > <u.kleine-koenig@pengutronix.de> wrote:
> > >
> > > On Fri, Feb 15, 2019 at 06:03:16PM +0000, Frank Li wrote:
> > > > Add DDR perf counter driver maintainer information
> > > >
> > > > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > > > ---
> > > > No change from v1 to v4
> > > >
> > > > MAINTAINERS | 7 +++++++
> > > > 1 file changed, 7 insertions(+)
> > > >
> > > > diff --git a/MAINTAINERS b/MAINTAINERS index 6ba068e..687be0f
> > > > 100644
> > > > --- a/MAINTAINERS
> > > > +++ b/MAINTAINERS
> > > > @@ -6163,6 +6163,13 @@ L: linux-i2c@vger.kernel.org
> > > > S: Maintained
> > > > F: drivers/i2c/busses/i2c-cpm.c
> > > >
> > > > +FREESCALE IMX DDR Performance Monitor DRIVER
> > > > +M: Frank Li <Frank.li@nxp.com>
> > > > +L: linux-arm-kernel@lists.infradead.org
> > > > +S: Maintained
> > > > +F: drivers/perf/fsl_imx8_ddr_perf.c
> > > > +F: Documentation/devicetree/bindings/perf/fsl-imx-ddr.txt
> > > > +
> > >
> > > With da8b7f0fb02b ("MAINTAINERS: add all files matching "imx" and "mxs"
> > > to the IMX entry") that sits currently in next the "ARM/FREESCALE
> > > IMX / MXC ARM ARCHITECTURE" also matches for these two files.
> > >
> > > Is this good here or not?
> >
> > Yes! what's do you means?
> > Needn't these entry because da8b7f0fb02b?
>
> It's just that we then will have two entries that will match. In a tree where I
> have both, this series and da8b7f0fb02b I see:
>
> $ scripts/get_maintainer.pl -f drivers/perf/fsl_imx8_ddr_perf.c
> Frank Li <Frank.li@nxp.com> (maintainer:FREESCALE IMX DDR
> Performance Monitor DRIVER)
> Will Deacon <will.deacon@arm.com> (maintainer:ARM PMU PROFILING
> AND DEBUGGING)
> Mark Rutland <mark.rutland@arm.com> (maintainer:ARM PMU
> PROFILING AND DEBUGGING)
> Shawn Guo <shawnguo@kernel.org> (maintainer:ARM/FREESCALE IMX /
> MXC ARM ARCHITECTURE)
> Sascha Hauer <s.hauer@pengutronix.de> (maintainer:ARM/FREESCALE
> IMX / MXC ARM ARCHITECTURE)
> Pengutronix Kernel Team <kernel@pengutronix.de>
> (reviewer:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE)
> Fabio Estevam <fabio.estevam@nxp.com> (reviewer:ARM/FREESCALE IMX
> / MXC ARM ARCHITECTURE)
> NXP Linux Team <linux-imx@nxp.com> (reviewer:ARM/FREESCALE IMX /
> MXC ARM ARCHITECTURE)
> linux-arm-kernel@lists.infradead.org (open list:FREESCALE IMX DDR
> Performance Monitor DRIVER)
> linux-kernel@vger.kernel.org (open list)
>
> That's not necessarily a bad thing, but something I want people to make aware
> of.
>
IMHO it's better to have a more specific maintainer in the MAINTAINER file
for a specific module.
Don't reply on the general pattern to filter all IMX patches to send ARCH
Maintainers to help review.
Regards
Dong Aisheng
> Best regards
> Uwe
>
> --
> Pengutronix e.K. | Uwe Kleine-König
> |
> Industrial Linux Solutions |
> https://emea01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.
> pengutronix.de%2F&data=02%7C01%7Caisheng.dong%40nxp.com%7C88
> f88bc0bd7e4812d00008d6938ed1af%7C686ea1d3bc2b4c6fa92cd99c5c3016
> 35%7C0%7C0%7C636858638948593375&sdata=KwEG5SXUsLCgUU9pg8
> hnZ9JQq4%2FPGHA6EFhmF%2Bcmlo8%3D&reserved=0 |
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
2019-02-15 18:03 ` Frank Li
@ 2019-02-22 19:45 ` Zhi Li
-1 siblings, 0 replies; 43+ messages in thread
From: Zhi Li @ 2019-02-22 19:45 UTC (permalink / raw)
To: Frank Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
will.deacon, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Feb 15, 2019 at 12:03 PM Frank Li <frank.li@nxp.com> wrote:
>
> Add ddr performance monitor support for iMX8QXP
>
> There are 4 counters for ddr perfomance events.
> counter 0 is dedicated for cycles.
> you choose any up to 3 no cycles events.
Any additional comments for these patches?
best regards
Frank Li
>
> for example:
>
> perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
>
> Support below events.
>
> ddr0/activate/ [Kernel PMU event]
> ddr0/axid-read/ [Kernel PMU event]
> ddr0/axid-write/ [Kernel PMU event]
> ddr0/cycles/ [Kernel PMU event]
> ddr0/hp-read-credit-cnt/ [Kernel PMU event]
> ddr0/hp-read/ [Kernel PMU event]
> ddr0/hp-req-nodcredit/ [Kernel PMU event]
> ddr0/hp-xact-credit/ [Kernel PMU event]
> ddr0/load-mode/ [Kernel PMU event]
> ddr0/lp-read-credit-cnt/ [Kernel PMU event]
> ddr0/lp-req-nocredit/ [Kernel PMU event]
> ddr0/lp-xact-credit/ [Kernel PMU event]
> ddr0/mwr/ [Kernel PMU event]
> ddr0/precharge/ [Kernel PMU event]
> ddr0/raw-hazard/ [Kernel PMU event]
> ddr0/read-access/ [Kernel PMU event]
> ddr0/read-activate/ [Kernel PMU event]
> ddr0/read-command/ [Kernel PMU event]
> ddr0/read-cycles/ [Kernel PMU event]
> ddr0/read-modify-write-command/ [Kernel PMU event]
> ddr0/read-queue-depth/ [Kernel PMU event]
> ddr0/read-write-transition/ [Kernel PMU event]
> ddr0/read/ [Kernel PMU event]
> ddr0/refresh/ [Kernel PMU event]
> ddr0/selfresh/ [Kernel PMU event]
> ddr0/wr-xact-credit/ [Kernel PMU event]
> ddr0/write-access/ [Kernel PMU event]
> ddr0/write-command/ [Kernel PMU event]
> ddr0/write-credit-cnt/ [Kernel PMU event]
> ddr0/write-cycles/ [Kernel PMU event]
> ddr0/write-queue-depth/ [Kernel PMU event]
> ddr0/write/
>
> Signed-off-by: Frank Li <Frank.Li@nxp.com>
> ---
> Change from v3 to v4
> * Change FSL_IMX8_DDR_PERF to FSL_IMX8_DDR_PMU
> * sort include
> * remove struct fsl_ddr_devtype_data
> * Added comment need disable control first
> * Added comment about must enable cycle counter
> * Added macro for EVENT_AXI_READ, remove hardcode 0x41 and 0x42
> * Added comment about cycle counter is fastest one
>
> Change from v2 to v3
> * remove kfree
>
> Change from V1 to V2
> * update Kconfig by use i.MX8 instead of i.MX8 QXP
> * remove gpl statememnt since SPDX tag
> * use dev_kzalloc
> * use dev_err
> * commit message show axi_read 0x41\axi_write 0x42
> * commit message show cycles must be enabled
> * Irq only issue at cycles overflow
> * use NUM_COUNTER
> * use devm_request_irq
> * add hotplug callback to handle context migration
>
> drivers/perf/Kconfig | 7 +
> drivers/perf/Makefile | 1 +
> drivers/perf/fsl_imx8_ddr_perf.c | 573 +++++++++++++++++++++++++++++++++++++++
> 3 files changed, 581 insertions(+)
> create mode 100644 drivers/perf/fsl_imx8_ddr_perf.c
>
> diff --git a/drivers/perf/Kconfig b/drivers/perf/Kconfig
> index af9bc17..e567437 100644
> --- a/drivers/perf/Kconfig
> +++ b/drivers/perf/Kconfig
> @@ -61,6 +61,13 @@ config ARM_DSU_PMU
> system, control logic. The PMU allows counting various events related
> to DSU.
>
> +config FSL_IMX8_DDR_PMU
> + tristate "Freescale i.MX8 DDR perf monitor"
> + depends on ARCH_MXC
> + help
> + Provides support for ddr perfomance monitor in i.MX8. Provide memory
> + througput information.
> +
> config HISI_PMU
> bool "HiSilicon SoC PMU"
> depends on ARM64 && ACPI
> diff --git a/drivers/perf/Makefile b/drivers/perf/Makefile
> index 909f27f..0bc066b 100644
> --- a/drivers/perf/Makefile
> +++ b/drivers/perf/Makefile
> @@ -4,6 +4,7 @@ obj-$(CONFIG_ARM_CCN) += arm-ccn.o
> obj-$(CONFIG_ARM_DSU_PMU) += arm_dsu_pmu.o
> obj-$(CONFIG_ARM_PMU) += arm_pmu.o arm_pmu_platform.o
> obj-$(CONFIG_ARM_PMU_ACPI) += arm_pmu_acpi.o
> +obj-$(CONFIG_FSL_IMX8_DDR_PMU) += fsl_imx8_ddr_perf.o
> obj-$(CONFIG_HISI_PMU) += hisilicon/
> obj-$(CONFIG_QCOM_L2_PMU) += qcom_l2_pmu.o
> obj-$(CONFIG_QCOM_L3_PMU) += qcom_l3_pmu.o
> diff --git a/drivers/perf/fsl_imx8_ddr_perf.c b/drivers/perf/fsl_imx8_ddr_perf.c
> new file mode 100644
> index 0000000..3d02daf
> --- /dev/null
> +++ b/drivers/perf/fsl_imx8_ddr_perf.c
> @@ -0,0 +1,573 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Copyright 2017 NXP
> + * Copyright 2016 Freescale Semiconductor, Inc.
> + */
> +
> +#include <linux/init.h>
> +#include <linux/interrupt.h>
> +#include <linux/io.h>
> +#include <linux/module.h>
> +#include <linux/of.h>
> +#include <linux/of_address.h>
> +#include <linux/of_device.h>
> +#include <linux/of_irq.h>
> +#include <linux/perf_event.h>
> +#include <linux/slab.h>
> +
> +#define COUNTER_CNTL 0x0
> +#define COUNTER_READ 0x20
> +
> +#define COUNTER_DPCR1 0x30
> +
> +#define CNTL_OVER 0x1
> +#define CNTL_CLEAR 0x2
> +#define CNTL_EN 0x4
> +#define CNTL_EN_MASK 0xFFFFFFFB
> +#define CNTL_CLEAR_MASK 0xFFFFFFFD
> +#define CNTL_OVER_MASK 0xFFFFFFFE
> +
> +#define CNTL_CSV_SHIFT 24
> +#define CNTL_CSV_MASK (0xFF << CNTL_CSV_SHIFT)
> +
> +#define EVENT_CYCLES_ID 0
> +#define EVENT_CYCLES_COUNTER 0
> +#define NUM_COUNTER 4
> +#define MAX_EVENT 3
> +#define EVENT_AXI_READ 0x41
> +#define EVENT_AXI_WRITE 0x42
> +
> +#define to_ddr_pmu(p) container_of(p, struct ddr_pmu, pmu)
> +
> +#define DDR_PERF_DEV_NAME "ddr_perf"
> +
> +static DEFINE_IDA(ddr_ida);
> +
> +PMU_EVENT_ATTR_STRING(cycles, ddr_perf_cycles, "event=0x00");
> +PMU_EVENT_ATTR_STRING(selfresh, ddr_perf_selfresh, "event=0x01");
> +PMU_EVENT_ATTR_STRING(read-access, ddr_perf_read_accesses, "event=0x04");
> +PMU_EVENT_ATTR_STRING(write-access, ddr_perf_write_accesses, "event=0x05");
> +PMU_EVENT_ATTR_STRING(read-queue-depth, ddr_perf_read_queue_depth,
> + "event=0x08");
> +PMU_EVENT_ATTR_STRING(write-queue-depth, ddr_perf_write_queue_depth,
> + "event=0x09");
> +PMU_EVENT_ATTR_STRING(lp-read-credit-cnt, ddr_perf_lp_read_credit_cnt,
> + "event=0x10");
> +PMU_EVENT_ATTR_STRING(hp-read-credit-cnt, ddr_perf_hp_read_credit_cnt,
> + "event=0x11");
> +PMU_EVENT_ATTR_STRING(write-credit-cnt, ddr_perf_write_credit_cnt,
> + "event=0x12");
> +PMU_EVENT_ATTR_STRING(read-command, ddr_perf_read_command, "event=0x20");
> +PMU_EVENT_ATTR_STRING(write-command, ddr_perf_write_command, "event=0x21");
> +PMU_EVENT_ATTR_STRING(read-modify-write-command,
> + ddr_perf_read_modify_write_command, "event=0x22");
> +PMU_EVENT_ATTR_STRING(hp-read, ddr_perf_hp_read, "event=0x23");
> +PMU_EVENT_ATTR_STRING(hp-req-nodcredit, ddr_perf_hp_req_nocredit, "event=0x24");
> +PMU_EVENT_ATTR_STRING(hp-xact-credit, ddr_perf_hp_xact_credit, "event=0x25");
> +PMU_EVENT_ATTR_STRING(lp-req-nocredit, ddr_perf_lp_req_nocredit, "event=0x26");
> +PMU_EVENT_ATTR_STRING(lp-xact-credit, ddr_perf_lp_xact_credit, "event=0x27");
> +PMU_EVENT_ATTR_STRING(wr-xact-credit, ddr_perf_wr_xact_credit, "event=0x29");
> +PMU_EVENT_ATTR_STRING(read-cycles, ddr_perf_read_cycles, "event=0x2a");
> +PMU_EVENT_ATTR_STRING(write-cycles, ddr_perf_write_cycles, "event=0x2b");
> +PMU_EVENT_ATTR_STRING(read-write-transition, ddr_perf_read_write_transition,
> + "event=0x30");
> +PMU_EVENT_ATTR_STRING(precharge, ddr_perf_precharge, "event=0x31");
> +PMU_EVENT_ATTR_STRING(activate, ddr_perf_activate, "event=0x32");
> +PMU_EVENT_ATTR_STRING(load-mode, ddr_perf_load_mode, "event=0x33");
> +PMU_EVENT_ATTR_STRING(mwr, ddr_perf_mwr, "event=0x34");
> +PMU_EVENT_ATTR_STRING(read, ddr_perf_read, "event=0x35");
> +PMU_EVENT_ATTR_STRING(read-activate, ddr_perf_read_activate, "event=0x36");
> +PMU_EVENT_ATTR_STRING(refresh, ddr_perf_refresh, "event=0x37");
> +PMU_EVENT_ATTR_STRING(write, ddr_perf_write, "event=0x38");
> +PMU_EVENT_ATTR_STRING(raw-hazard, ddr_perf_raw_hazard, "event=0x39");
> +
> +PMU_EVENT_ATTR_STRING(axid-read, ddr_perf_axid_read,
> + "event="__stringify(EVENT_AXI_READ));
> +PMU_EVENT_ATTR_STRING(axid-write, ddr_perf_axid_write,
> + "event="__stringify(EVENT_AXI_WRITE));
> +
> +#define DDR_CAP_AXI_ID 0x1
> +
> +static const struct of_device_id imx_ddr_pmu_dt_ids[] = {
> + { .compatible = "fsl,imx8-ddr-pmu",},
> + { .compatible = "fsl,imx8m-ddr-pmu", .data = (void *)DDR_CAP_AXI_ID},
> + { /* sentinel */ }
> +};
> +
> +struct ddr_pmu {
> + struct pmu pmu;
> + void __iomem *base;
> + cpumask_t cpu;
> + struct hlist_node node;
> + struct device *dev;
> + struct perf_event *active_events[NUM_COUNTER];
> + int total_events;
> + enum cpuhp_state cpuhp_state;
> + bool cycles_active;
> + uintptr_t flags;
> +};
> +
> +static ssize_t ddr_perf_cpumask_show(struct device *dev,
> + struct device_attribute *attr, char *buf)
> +{
> + struct ddr_pmu *pmu = dev_get_drvdata(dev);
> +
> + return cpumap_print_to_pagebuf(true, buf, &pmu->cpu);
> +}
> +
> +static struct device_attribute ddr_perf_cpumask_attr =
> + __ATTR(cpumask, 0444, ddr_perf_cpumask_show, NULL);
> +
> +static struct attribute *ddr_perf_cpumask_attrs[] = {
> + &ddr_perf_cpumask_attr.attr,
> + NULL,
> +};
> +
> +static struct attribute_group ddr_perf_cpumask_attr_group = {
> + .attrs = ddr_perf_cpumask_attrs,
> +};
> +
> +static struct attribute *ddr_perf_events_attrs[] = {
> + &ddr_perf_cycles.attr.attr,
> + &ddr_perf_selfresh.attr.attr,
> + &ddr_perf_read_accesses.attr.attr,
> + &ddr_perf_write_accesses.attr.attr,
> + &ddr_perf_read_queue_depth.attr.attr,
> + &ddr_perf_write_queue_depth.attr.attr,
> + &ddr_perf_lp_read_credit_cnt.attr.attr,
> + &ddr_perf_hp_read_credit_cnt.attr.attr,
> + &ddr_perf_write_credit_cnt.attr.attr,
> + &ddr_perf_read_command.attr.attr,
> + &ddr_perf_write_command.attr.attr,
> + &ddr_perf_read_modify_write_command.attr.attr,
> + &ddr_perf_hp_read.attr.attr,
> + &ddr_perf_hp_req_nocredit.attr.attr,
> + &ddr_perf_hp_xact_credit.attr.attr,
> + &ddr_perf_lp_req_nocredit.attr.attr,
> + &ddr_perf_lp_xact_credit.attr.attr,
> + &ddr_perf_wr_xact_credit.attr.attr,
> + &ddr_perf_read_cycles.attr.attr,
> + &ddr_perf_write_cycles.attr.attr,
> + &ddr_perf_read_write_transition.attr.attr,
> + &ddr_perf_precharge.attr.attr,
> + &ddr_perf_activate.attr.attr,
> + &ddr_perf_load_mode.attr.attr,
> + &ddr_perf_mwr.attr.attr,
> + &ddr_perf_read.attr.attr,
> + &ddr_perf_read_activate.attr.attr,
> + &ddr_perf_refresh.attr.attr,
> + &ddr_perf_write.attr.attr,
> + &ddr_perf_raw_hazard.attr.attr,
> + &ddr_perf_axid_read.attr.attr,
> + &ddr_perf_axid_write.attr.attr,
> + NULL,
> +};
> +
> +static struct attribute_group ddr_perf_events_attr_group = {
> + .name = "events",
> + .attrs = ddr_perf_events_attrs,
> +};
> +
> +PMU_FORMAT_ATTR(event, "config:0-63");
> +PMU_FORMAT_ATTR(axi_id, "config1:0-63");
> +
> +static struct attribute *ddr_perf_format_attrs[] = {
> + &format_attr_event.attr,
> + &format_attr_axi_id.attr,
> + NULL,
> +};
> +
> +static struct attribute_group ddr_perf_format_attr_group = {
> + .name = "format",
> + .attrs = ddr_perf_format_attrs,
> +};
> +
> +static const struct attribute_group *attr_groups[] = {
> + &ddr_perf_events_attr_group,
> + &ddr_perf_format_attr_group,
> + &ddr_perf_cpumask_attr_group,
> + NULL,
> +};
> +
> +static u32 ddr_perf_alloc_counter(struct ddr_pmu *pmu, int event)
> +{
> + int i;
> +
> + /* Always map cycle event to counter 0
> + Cycles counter is dedicated for cycle event
> + can't used for the other events
> + */
> + if (event == EVENT_CYCLES_ID)
> + return EVENT_CYCLES_COUNTER;
> +
> + for (i = 1; i < NUM_COUNTER; i++)
> + if (pmu->active_events[i] == NULL)
> + return i;
> +
> + return -ENOENT;
> +}
> +
> +static u32 ddr_perf_free_counter(struct ddr_pmu *pmu, int counter)
> +{
> + if (counter < 0 || counter >= NUM_COUNTER)
> + return -ENOENT;
> +
> + pmu->active_events[counter] = NULL;
> +
> + return 0;
> +}
> +
> +static u32 ddr_perf_read_counter(struct ddr_pmu *pmu, int counter)
> +{
> + return readl(pmu->base + COUNTER_READ + counter * 4);
> +}
> +
> +static int ddr_perf_event_init(struct perf_event *event)
> +{
> + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> + struct hw_perf_event *hwc = &event->hw;
> +
> + if (event->attr.type != event->pmu->type)
> + return -ENOENT;
> +
> + if (is_sampling_event(event) || event->attach_state & PERF_ATTACH_TASK)
> + return -EOPNOTSUPP;
> +
> + if (event->cpu < 0) {
> + dev_warn(pmu->dev, "Can't provide per-task data!\n");
> + return -EOPNOTSUPP;
> + }
> +
> + if (event->attr.exclude_user ||
> + event->attr.exclude_kernel ||
> + event->attr.exclude_hv ||
> + event->attr.exclude_idle ||
> + event->attr.exclude_host ||
> + event->attr.exclude_guest ||
> + event->attr.sample_period)
> + return -EINVAL;
> +
> + event->cpu = cpumask_first(&pmu->cpu);
> + hwc->idx = -1;
> +
> + return 0;
> +}
> +
> +
> +static void ddr_perf_event_update(struct perf_event *event)
> +{
> + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> + struct hw_perf_event *hwc = &event->hw;
> + u64 delta, prev_raw_count, new_raw_count;
> + int counter = hwc->idx;
> +
> + do {
> + prev_raw_count = local64_read(&hwc->prev_count);
> + new_raw_count = ddr_perf_read_counter(pmu, counter);
> + } while (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
> + new_raw_count) != prev_raw_count);
> +
> + delta = (new_raw_count - prev_raw_count) & 0xFFFFFFFF;
> +
> + local64_add(delta, &event->count);
> +}
> +
> +static void ddr_perf_event_enable(struct ddr_pmu *pmu, int config,
> + int counter, bool enable)
> +{
> + u8 reg = counter * 4 + COUNTER_CNTL;
> + int val;
> +
> + if (enable) {
> + /* must disable first, then enable again
> + * otherwise, cycle counter will not work
> + * if previous state is enabled.
> + */
> + writel(0, pmu->base + reg);
> + val = CNTL_EN | CNTL_CLEAR;
> + val |= (config << CNTL_CSV_SHIFT) & CNTL_CSV_MASK;
> + } else {
> + /* Disable counter */
> + val = readl(pmu->base + reg) & CNTL_EN_MASK;
> + }
> +
> + writel(val, pmu->base + reg);
> +
> + if (config == EVENT_CYCLES_ID)
> + pmu->cycles_active = enable;
> +}
> +
> +static void ddr_perf_event_start(struct perf_event *event, int flags)
> +{
> + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> + struct hw_perf_event *hwc = &event->hw;
> + int counter = hwc->idx;
> +
> + if (pmu->flags & DDR_CAP_AXI_ID) {
> + if (event->attr.config == EVENT_AXI_READ ||
> + event->attr.config == EVENT_AXI_WRITE) {
> + int val = event->attr.config1;
> +
> + writel(val, pmu->base + COUNTER_DPCR1);
> + }
> + }
> +
> + local64_set(&hwc->prev_count, 0);
> +
> + ddr_perf_event_enable(pmu, event->attr.config, counter, true);
> + /*
> + * If the cycles counter wasn't explicitly selected,
> + * we will enable it now.
> + * cycles counter must be enabled otherwise other counters will
> + * stopped.
> + */
> + if (counter > 0 && !pmu->cycles_active)
> + ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
> + EVENT_CYCLES_COUNTER, true);
> +}
> +
> +static int ddr_perf_event_add(struct perf_event *event, int flags)
> +{
> + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> + struct hw_perf_event *hwc = &event->hw;
> + int counter;
> + int cfg = event->attr.config;
> +
> + counter = ddr_perf_alloc_counter(pmu, cfg);
> + if (counter < 0) {
> + dev_dbg(pmu->dev, "There are not enough counters\n");
> + return -EOPNOTSUPP;
> + }
> +
> + pmu->active_events[counter] = event;
> + pmu->total_events++;
> + hwc->idx = counter;
> +
> + if (flags & PERF_EF_START)
> + ddr_perf_event_start(event, flags);
> +
> + local64_set(&hwc->prev_count, ddr_perf_read_counter(pmu, counter));
> +
> + return 0;
> +}
> +
> +static void ddr_perf_event_stop(struct perf_event *event, int flags)
> +{
> + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> + struct hw_perf_event *hwc = &event->hw;
> + int counter = hwc->idx;
> +
> + ddr_perf_event_enable(pmu, event->attr.config, counter, false);
> + ddr_perf_event_update(event);
> +}
> +
> +static void ddr_perf_event_del(struct perf_event *event, int flags)
> +{
> + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> + struct hw_perf_event *hwc = &event->hw;
> + int counter = hwc->idx;
> +
> + ddr_perf_event_stop(event, PERF_EF_UPDATE);
> +
> + ddr_perf_free_counter(pmu, counter);
> + pmu->total_events--;
> + hwc->idx = -1;
> +
> + /* If all events have stopped, stop the cycles counter as well */
> + if ((pmu->total_events == 0) && pmu->cycles_active)
> + ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
> + EVENT_CYCLES_COUNTER, false);
> +}
> +
> +static int ddr_perf_init(struct ddr_pmu *pmu, void __iomem *base,
> + struct device *dev)
> +{
> + *pmu = (struct ddr_pmu) {
> + .pmu = (struct pmu) {
> + .task_ctx_nr = perf_invalid_context,
> + .attr_groups = attr_groups,
> + .event_init = ddr_perf_event_init,
> + .add = ddr_perf_event_add,
> + .del = ddr_perf_event_del,
> + .start = ddr_perf_event_start,
> + .stop = ddr_perf_event_stop,
> + .read = ddr_perf_event_update,
> + },
> + .base = base,
> + .dev = dev,
> + };
> +
> + return ida_simple_get(&ddr_ida, 0, 0, GFP_KERNEL);
> +}
> +
> +static irqreturn_t ddr_perf_irq_handler(int irq, void *p)
> +{
> + int i;
> + u8 reg;
> + int val;
> + int counter;
> + struct ddr_pmu *pmu = (struct ddr_pmu *) p;
> + struct perf_event *event;
> +
> + /* Only cycles counter overflowed can issue irq. all counters will
> + * be stopped when cycles counter overflow. but other counter don't stop
> + * when overflow happen. Update all of the local counter values,
> + * then reset the cycles counter, so the others can continue
> + * counting. cycles counter is fastest counter in all events. at last
> + * 4 times than other counters.
> + */
> + for (i = 0; i < NUM_COUNTER; i++) {
> +
> + if (!pmu->active_events[i])
> + continue;
> +
> + event = pmu->active_events[i];
> + counter = event->hw.idx;
> + reg = counter * 4 + COUNTER_CNTL;
> + val = readl(pmu->base + reg);
> +
> + ddr_perf_event_update(event);
> +
> + if (counter == EVENT_CYCLES_COUNTER) {
> + ddr_perf_event_enable(pmu,
> + EVENT_CYCLES_ID,
> + EVENT_CYCLES_COUNTER,
> + true);
> + ddr_perf_event_update(event);
> + }
> + }
> +
> + return IRQ_HANDLED;
> +}
> +
> +static int ddr_perf_offline_cpu(unsigned int cpu, struct hlist_node *node)
> +{
> + struct ddr_pmu *pmu = hlist_entry_safe(node, struct ddr_pmu, node);
> + int target;
> +
> + if (!cpumask_test_and_clear_cpu(cpu, &pmu->cpu))
> + return 0;
> +
> + target = cpumask_any_but(cpu_online_mask, cpu);
> + if (target >= nr_cpu_ids)
> + return 0;
> +
> + perf_pmu_migrate_context(&pmu->pmu, cpu, target);
> + cpumask_set_cpu(target, &pmu->cpu);
> +
> + return 0;
> +}
> +
> +static int ddr_perf_probe(struct platform_device *pdev)
> +{
> + struct ddr_pmu *pmu;
> + struct device_node *np;
> + void __iomem *base;
> + struct resource *iomem;
> + char *name;
> + char *hpname;
> + int num;
> + int ret;
> + u32 irq;
> + const struct of_device_id *of_id =
> + of_match_device(imx_ddr_pmu_dt_ids, &pdev->dev);
> +
> + iomem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> + base = devm_ioremap_resource(&pdev->dev, iomem);
> + if (IS_ERR(base))
> + return PTR_ERR(base);
> +
> + np = pdev->dev.of_node;
> +
> + pmu = devm_kzalloc(&pdev->dev, sizeof(*pmu), GFP_KERNEL);
> + if (!pmu)
> + return -ENOMEM;
> +
> + num = ddr_perf_init(pmu, base, &pdev->dev);
> +
> + name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "ddr%d", num);
> + if (!name)
> + return -ENOMEM;
> +
> + hpname = devm_kasprintf(&pdev->dev, GFP_KERNEL,
> + "perf/imx/ddr%d:online", num);
> + if (!hpname)
> + return -ENOMEM;
> +
> + pmu->flags = (uintptr_t) of_id->data;
> +
> + cpumask_set_cpu(raw_smp_processor_id(), &pmu->cpu);
> + ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN, hpname, NULL,
> + ddr_perf_offline_cpu);
> +
> + if (ret < 0) {
> + dev_err(&pdev->dev, "cpuhp_setup_state_multi failed\n");
> + goto ddr_perf_err;
> + }
> +
> + pmu->cpuhp_state = ret;
> +
> + /* Register the pmu instance for cpu hotplug */
> + cpuhp_state_add_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> +
> + ret = perf_pmu_register(&(pmu->pmu), name, -1);
> + if (ret)
> + goto ddr_perf_err;
> +
> + /* Request irq */
> + irq = of_irq_get(np, 0);
> + if (irq < 0) {
> + dev_err(&pdev->dev, "Failed to get irq: %d", irq);
> + ret = irq;
> + goto ddr_perf_irq_err;
> + }
> +
> + ret = devm_request_irq(&pdev->dev, irq,
> + ddr_perf_irq_handler,
> + IRQF_TRIGGER_RISING | IRQF_ONESHOT,
> + DDR_PERF_DEV_NAME,
> + pmu);
> + if (ret < 0) {
> + dev_err(&pdev->dev, "Request irq failed: %d", ret);
> + goto ddr_perf_irq_err;
> + }
> +
> + return 0;
> +
> +ddr_perf_irq_err:
> + perf_pmu_unregister(&(pmu->pmu));
> +
> +ddr_perf_err:
> + if (pmu->cpuhp_state)
> + cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> +
> + dev_warn(&pdev->dev, "i.MX8 DDR Perf PMU failed (%d), disabled\n", ret);
> + return ret;
> +}
> +
> +static int ddr_perf_remove(struct platform_device *pdev)
> +{
> + struct ddr_pmu *pmu = platform_get_drvdata(pdev);
> +
> + cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> + perf_pmu_unregister(&pmu->pmu);
> +
> + return 0;
> +}
> +
> +static struct platform_driver imx_ddr_pmu_driver = {
> + .driver = {
> + .name = "imx-ddr-pmu",
> + .of_match_table = imx_ddr_pmu_dt_ids,
> + },
> + .probe = ddr_perf_probe,
> + .remove = ddr_perf_remove,
> +};
> +
> +static int __init imx_ddr_pmu_init(void)
> +{
> + return platform_driver_register(&imx_ddr_pmu_driver);
> +}
> +
> +module_init(imx_ddr_pmu_init);
> +
> --
> 2.5.2
>
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
@ 2019-02-22 19:45 ` Zhi Li
0 siblings, 0 replies; 43+ messages in thread
From: Zhi Li @ 2019-02-22 19:45 UTC (permalink / raw)
To: Frank Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
will.deacon, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Feb 15, 2019 at 12:03 PM Frank Li <frank.li@nxp.com> wrote:
>
> Add ddr performance monitor support for iMX8QXP
>
> There are 4 counters for ddr perfomance events.
> counter 0 is dedicated for cycles.
> you choose any up to 3 no cycles events.
Any additional comments for these patches?
best regards
Frank Li
>
> for example:
>
> perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
>
> Support below events.
>
> ddr0/activate/ [Kernel PMU event]
> ddr0/axid-read/ [Kernel PMU event]
> ddr0/axid-write/ [Kernel PMU event]
> ddr0/cycles/ [Kernel PMU event]
> ddr0/hp-read-credit-cnt/ [Kernel PMU event]
> ddr0/hp-read/ [Kernel PMU event]
> ddr0/hp-req-nodcredit/ [Kernel PMU event]
> ddr0/hp-xact-credit/ [Kernel PMU event]
> ddr0/load-mode/ [Kernel PMU event]
> ddr0/lp-read-credit-cnt/ [Kernel PMU event]
> ddr0/lp-req-nocredit/ [Kernel PMU event]
> ddr0/lp-xact-credit/ [Kernel PMU event]
> ddr0/mwr/ [Kernel PMU event]
> ddr0/precharge/ [Kernel PMU event]
> ddr0/raw-hazard/ [Kernel PMU event]
> ddr0/read-access/ [Kernel PMU event]
> ddr0/read-activate/ [Kernel PMU event]
> ddr0/read-command/ [Kernel PMU event]
> ddr0/read-cycles/ [Kernel PMU event]
> ddr0/read-modify-write-command/ [Kernel PMU event]
> ddr0/read-queue-depth/ [Kernel PMU event]
> ddr0/read-write-transition/ [Kernel PMU event]
> ddr0/read/ [Kernel PMU event]
> ddr0/refresh/ [Kernel PMU event]
> ddr0/selfresh/ [Kernel PMU event]
> ddr0/wr-xact-credit/ [Kernel PMU event]
> ddr0/write-access/ [Kernel PMU event]
> ddr0/write-command/ [Kernel PMU event]
> ddr0/write-credit-cnt/ [Kernel PMU event]
> ddr0/write-cycles/ [Kernel PMU event]
> ddr0/write-queue-depth/ [Kernel PMU event]
> ddr0/write/
>
> Signed-off-by: Frank Li <Frank.Li@nxp.com>
> ---
> Change from v3 to v4
> * Change FSL_IMX8_DDR_PERF to FSL_IMX8_DDR_PMU
> * sort include
> * remove struct fsl_ddr_devtype_data
> * Added comment need disable control first
> * Added comment about must enable cycle counter
> * Added macro for EVENT_AXI_READ, remove hardcode 0x41 and 0x42
> * Added comment about cycle counter is fastest one
>
> Change from v2 to v3
> * remove kfree
>
> Change from V1 to V2
> * update Kconfig by use i.MX8 instead of i.MX8 QXP
> * remove gpl statememnt since SPDX tag
> * use dev_kzalloc
> * use dev_err
> * commit message show axi_read 0x41\axi_write 0x42
> * commit message show cycles must be enabled
> * Irq only issue at cycles overflow
> * use NUM_COUNTER
> * use devm_request_irq
> * add hotplug callback to handle context migration
>
> drivers/perf/Kconfig | 7 +
> drivers/perf/Makefile | 1 +
> drivers/perf/fsl_imx8_ddr_perf.c | 573 +++++++++++++++++++++++++++++++++++++++
> 3 files changed, 581 insertions(+)
> create mode 100644 drivers/perf/fsl_imx8_ddr_perf.c
>
> diff --git a/drivers/perf/Kconfig b/drivers/perf/Kconfig
> index af9bc17..e567437 100644
> --- a/drivers/perf/Kconfig
> +++ b/drivers/perf/Kconfig
> @@ -61,6 +61,13 @@ config ARM_DSU_PMU
> system, control logic. The PMU allows counting various events related
> to DSU.
>
> +config FSL_IMX8_DDR_PMU
> + tristate "Freescale i.MX8 DDR perf monitor"
> + depends on ARCH_MXC
> + help
> + Provides support for ddr perfomance monitor in i.MX8. Provide memory
> + througput information.
> +
> config HISI_PMU
> bool "HiSilicon SoC PMU"
> depends on ARM64 && ACPI
> diff --git a/drivers/perf/Makefile b/drivers/perf/Makefile
> index 909f27f..0bc066b 100644
> --- a/drivers/perf/Makefile
> +++ b/drivers/perf/Makefile
> @@ -4,6 +4,7 @@ obj-$(CONFIG_ARM_CCN) += arm-ccn.o
> obj-$(CONFIG_ARM_DSU_PMU) += arm_dsu_pmu.o
> obj-$(CONFIG_ARM_PMU) += arm_pmu.o arm_pmu_platform.o
> obj-$(CONFIG_ARM_PMU_ACPI) += arm_pmu_acpi.o
> +obj-$(CONFIG_FSL_IMX8_DDR_PMU) += fsl_imx8_ddr_perf.o
> obj-$(CONFIG_HISI_PMU) += hisilicon/
> obj-$(CONFIG_QCOM_L2_PMU) += qcom_l2_pmu.o
> obj-$(CONFIG_QCOM_L3_PMU) += qcom_l3_pmu.o
> diff --git a/drivers/perf/fsl_imx8_ddr_perf.c b/drivers/perf/fsl_imx8_ddr_perf.c
> new file mode 100644
> index 0000000..3d02daf
> --- /dev/null
> +++ b/drivers/perf/fsl_imx8_ddr_perf.c
> @@ -0,0 +1,573 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Copyright 2017 NXP
> + * Copyright 2016 Freescale Semiconductor, Inc.
> + */
> +
> +#include <linux/init.h>
> +#include <linux/interrupt.h>
> +#include <linux/io.h>
> +#include <linux/module.h>
> +#include <linux/of.h>
> +#include <linux/of_address.h>
> +#include <linux/of_device.h>
> +#include <linux/of_irq.h>
> +#include <linux/perf_event.h>
> +#include <linux/slab.h>
> +
> +#define COUNTER_CNTL 0x0
> +#define COUNTER_READ 0x20
> +
> +#define COUNTER_DPCR1 0x30
> +
> +#define CNTL_OVER 0x1
> +#define CNTL_CLEAR 0x2
> +#define CNTL_EN 0x4
> +#define CNTL_EN_MASK 0xFFFFFFFB
> +#define CNTL_CLEAR_MASK 0xFFFFFFFD
> +#define CNTL_OVER_MASK 0xFFFFFFFE
> +
> +#define CNTL_CSV_SHIFT 24
> +#define CNTL_CSV_MASK (0xFF << CNTL_CSV_SHIFT)
> +
> +#define EVENT_CYCLES_ID 0
> +#define EVENT_CYCLES_COUNTER 0
> +#define NUM_COUNTER 4
> +#define MAX_EVENT 3
> +#define EVENT_AXI_READ 0x41
> +#define EVENT_AXI_WRITE 0x42
> +
> +#define to_ddr_pmu(p) container_of(p, struct ddr_pmu, pmu)
> +
> +#define DDR_PERF_DEV_NAME "ddr_perf"
> +
> +static DEFINE_IDA(ddr_ida);
> +
> +PMU_EVENT_ATTR_STRING(cycles, ddr_perf_cycles, "event=0x00");
> +PMU_EVENT_ATTR_STRING(selfresh, ddr_perf_selfresh, "event=0x01");
> +PMU_EVENT_ATTR_STRING(read-access, ddr_perf_read_accesses, "event=0x04");
> +PMU_EVENT_ATTR_STRING(write-access, ddr_perf_write_accesses, "event=0x05");
> +PMU_EVENT_ATTR_STRING(read-queue-depth, ddr_perf_read_queue_depth,
> + "event=0x08");
> +PMU_EVENT_ATTR_STRING(write-queue-depth, ddr_perf_write_queue_depth,
> + "event=0x09");
> +PMU_EVENT_ATTR_STRING(lp-read-credit-cnt, ddr_perf_lp_read_credit_cnt,
> + "event=0x10");
> +PMU_EVENT_ATTR_STRING(hp-read-credit-cnt, ddr_perf_hp_read_credit_cnt,
> + "event=0x11");
> +PMU_EVENT_ATTR_STRING(write-credit-cnt, ddr_perf_write_credit_cnt,
> + "event=0x12");
> +PMU_EVENT_ATTR_STRING(read-command, ddr_perf_read_command, "event=0x20");
> +PMU_EVENT_ATTR_STRING(write-command, ddr_perf_write_command, "event=0x21");
> +PMU_EVENT_ATTR_STRING(read-modify-write-command,
> + ddr_perf_read_modify_write_command, "event=0x22");
> +PMU_EVENT_ATTR_STRING(hp-read, ddr_perf_hp_read, "event=0x23");
> +PMU_EVENT_ATTR_STRING(hp-req-nodcredit, ddr_perf_hp_req_nocredit, "event=0x24");
> +PMU_EVENT_ATTR_STRING(hp-xact-credit, ddr_perf_hp_xact_credit, "event=0x25");
> +PMU_EVENT_ATTR_STRING(lp-req-nocredit, ddr_perf_lp_req_nocredit, "event=0x26");
> +PMU_EVENT_ATTR_STRING(lp-xact-credit, ddr_perf_lp_xact_credit, "event=0x27");
> +PMU_EVENT_ATTR_STRING(wr-xact-credit, ddr_perf_wr_xact_credit, "event=0x29");
> +PMU_EVENT_ATTR_STRING(read-cycles, ddr_perf_read_cycles, "event=0x2a");
> +PMU_EVENT_ATTR_STRING(write-cycles, ddr_perf_write_cycles, "event=0x2b");
> +PMU_EVENT_ATTR_STRING(read-write-transition, ddr_perf_read_write_transition,
> + "event=0x30");
> +PMU_EVENT_ATTR_STRING(precharge, ddr_perf_precharge, "event=0x31");
> +PMU_EVENT_ATTR_STRING(activate, ddr_perf_activate, "event=0x32");
> +PMU_EVENT_ATTR_STRING(load-mode, ddr_perf_load_mode, "event=0x33");
> +PMU_EVENT_ATTR_STRING(mwr, ddr_perf_mwr, "event=0x34");
> +PMU_EVENT_ATTR_STRING(read, ddr_perf_read, "event=0x35");
> +PMU_EVENT_ATTR_STRING(read-activate, ddr_perf_read_activate, "event=0x36");
> +PMU_EVENT_ATTR_STRING(refresh, ddr_perf_refresh, "event=0x37");
> +PMU_EVENT_ATTR_STRING(write, ddr_perf_write, "event=0x38");
> +PMU_EVENT_ATTR_STRING(raw-hazard, ddr_perf_raw_hazard, "event=0x39");
> +
> +PMU_EVENT_ATTR_STRING(axid-read, ddr_perf_axid_read,
> + "event="__stringify(EVENT_AXI_READ));
> +PMU_EVENT_ATTR_STRING(axid-write, ddr_perf_axid_write,
> + "event="__stringify(EVENT_AXI_WRITE));
> +
> +#define DDR_CAP_AXI_ID 0x1
> +
> +static const struct of_device_id imx_ddr_pmu_dt_ids[] = {
> + { .compatible = "fsl,imx8-ddr-pmu",},
> + { .compatible = "fsl,imx8m-ddr-pmu", .data = (void *)DDR_CAP_AXI_ID},
> + { /* sentinel */ }
> +};
> +
> +struct ddr_pmu {
> + struct pmu pmu;
> + void __iomem *base;
> + cpumask_t cpu;
> + struct hlist_node node;
> + struct device *dev;
> + struct perf_event *active_events[NUM_COUNTER];
> + int total_events;
> + enum cpuhp_state cpuhp_state;
> + bool cycles_active;
> + uintptr_t flags;
> +};
> +
> +static ssize_t ddr_perf_cpumask_show(struct device *dev,
> + struct device_attribute *attr, char *buf)
> +{
> + struct ddr_pmu *pmu = dev_get_drvdata(dev);
> +
> + return cpumap_print_to_pagebuf(true, buf, &pmu->cpu);
> +}
> +
> +static struct device_attribute ddr_perf_cpumask_attr =
> + __ATTR(cpumask, 0444, ddr_perf_cpumask_show, NULL);
> +
> +static struct attribute *ddr_perf_cpumask_attrs[] = {
> + &ddr_perf_cpumask_attr.attr,
> + NULL,
> +};
> +
> +static struct attribute_group ddr_perf_cpumask_attr_group = {
> + .attrs = ddr_perf_cpumask_attrs,
> +};
> +
> +static struct attribute *ddr_perf_events_attrs[] = {
> + &ddr_perf_cycles.attr.attr,
> + &ddr_perf_selfresh.attr.attr,
> + &ddr_perf_read_accesses.attr.attr,
> + &ddr_perf_write_accesses.attr.attr,
> + &ddr_perf_read_queue_depth.attr.attr,
> + &ddr_perf_write_queue_depth.attr.attr,
> + &ddr_perf_lp_read_credit_cnt.attr.attr,
> + &ddr_perf_hp_read_credit_cnt.attr.attr,
> + &ddr_perf_write_credit_cnt.attr.attr,
> + &ddr_perf_read_command.attr.attr,
> + &ddr_perf_write_command.attr.attr,
> + &ddr_perf_read_modify_write_command.attr.attr,
> + &ddr_perf_hp_read.attr.attr,
> + &ddr_perf_hp_req_nocredit.attr.attr,
> + &ddr_perf_hp_xact_credit.attr.attr,
> + &ddr_perf_lp_req_nocredit.attr.attr,
> + &ddr_perf_lp_xact_credit.attr.attr,
> + &ddr_perf_wr_xact_credit.attr.attr,
> + &ddr_perf_read_cycles.attr.attr,
> + &ddr_perf_write_cycles.attr.attr,
> + &ddr_perf_read_write_transition.attr.attr,
> + &ddr_perf_precharge.attr.attr,
> + &ddr_perf_activate.attr.attr,
> + &ddr_perf_load_mode.attr.attr,
> + &ddr_perf_mwr.attr.attr,
> + &ddr_perf_read.attr.attr,
> + &ddr_perf_read_activate.attr.attr,
> + &ddr_perf_refresh.attr.attr,
> + &ddr_perf_write.attr.attr,
> + &ddr_perf_raw_hazard.attr.attr,
> + &ddr_perf_axid_read.attr.attr,
> + &ddr_perf_axid_write.attr.attr,
> + NULL,
> +};
> +
> +static struct attribute_group ddr_perf_events_attr_group = {
> + .name = "events",
> + .attrs = ddr_perf_events_attrs,
> +};
> +
> +PMU_FORMAT_ATTR(event, "config:0-63");
> +PMU_FORMAT_ATTR(axi_id, "config1:0-63");
> +
> +static struct attribute *ddr_perf_format_attrs[] = {
> + &format_attr_event.attr,
> + &format_attr_axi_id.attr,
> + NULL,
> +};
> +
> +static struct attribute_group ddr_perf_format_attr_group = {
> + .name = "format",
> + .attrs = ddr_perf_format_attrs,
> +};
> +
> +static const struct attribute_group *attr_groups[] = {
> + &ddr_perf_events_attr_group,
> + &ddr_perf_format_attr_group,
> + &ddr_perf_cpumask_attr_group,
> + NULL,
> +};
> +
> +static u32 ddr_perf_alloc_counter(struct ddr_pmu *pmu, int event)
> +{
> + int i;
> +
> + /* Always map cycle event to counter 0
> + Cycles counter is dedicated for cycle event
> + can't used for the other events
> + */
> + if (event == EVENT_CYCLES_ID)
> + return EVENT_CYCLES_COUNTER;
> +
> + for (i = 1; i < NUM_COUNTER; i++)
> + if (pmu->active_events[i] == NULL)
> + return i;
> +
> + return -ENOENT;
> +}
> +
> +static u32 ddr_perf_free_counter(struct ddr_pmu *pmu, int counter)
> +{
> + if (counter < 0 || counter >= NUM_COUNTER)
> + return -ENOENT;
> +
> + pmu->active_events[counter] = NULL;
> +
> + return 0;
> +}
> +
> +static u32 ddr_perf_read_counter(struct ddr_pmu *pmu, int counter)
> +{
> + return readl(pmu->base + COUNTER_READ + counter * 4);
> +}
> +
> +static int ddr_perf_event_init(struct perf_event *event)
> +{
> + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> + struct hw_perf_event *hwc = &event->hw;
> +
> + if (event->attr.type != event->pmu->type)
> + return -ENOENT;
> +
> + if (is_sampling_event(event) || event->attach_state & PERF_ATTACH_TASK)
> + return -EOPNOTSUPP;
> +
> + if (event->cpu < 0) {
> + dev_warn(pmu->dev, "Can't provide per-task data!\n");
> + return -EOPNOTSUPP;
> + }
> +
> + if (event->attr.exclude_user ||
> + event->attr.exclude_kernel ||
> + event->attr.exclude_hv ||
> + event->attr.exclude_idle ||
> + event->attr.exclude_host ||
> + event->attr.exclude_guest ||
> + event->attr.sample_period)
> + return -EINVAL;
> +
> + event->cpu = cpumask_first(&pmu->cpu);
> + hwc->idx = -1;
> +
> + return 0;
> +}
> +
> +
> +static void ddr_perf_event_update(struct perf_event *event)
> +{
> + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> + struct hw_perf_event *hwc = &event->hw;
> + u64 delta, prev_raw_count, new_raw_count;
> + int counter = hwc->idx;
> +
> + do {
> + prev_raw_count = local64_read(&hwc->prev_count);
> + new_raw_count = ddr_perf_read_counter(pmu, counter);
> + } while (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
> + new_raw_count) != prev_raw_count);
> +
> + delta = (new_raw_count - prev_raw_count) & 0xFFFFFFFF;
> +
> + local64_add(delta, &event->count);
> +}
> +
> +static void ddr_perf_event_enable(struct ddr_pmu *pmu, int config,
> + int counter, bool enable)
> +{
> + u8 reg = counter * 4 + COUNTER_CNTL;
> + int val;
> +
> + if (enable) {
> + /* must disable first, then enable again
> + * otherwise, cycle counter will not work
> + * if previous state is enabled.
> + */
> + writel(0, pmu->base + reg);
> + val = CNTL_EN | CNTL_CLEAR;
> + val |= (config << CNTL_CSV_SHIFT) & CNTL_CSV_MASK;
> + } else {
> + /* Disable counter */
> + val = readl(pmu->base + reg) & CNTL_EN_MASK;
> + }
> +
> + writel(val, pmu->base + reg);
> +
> + if (config == EVENT_CYCLES_ID)
> + pmu->cycles_active = enable;
> +}
> +
> +static void ddr_perf_event_start(struct perf_event *event, int flags)
> +{
> + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> + struct hw_perf_event *hwc = &event->hw;
> + int counter = hwc->idx;
> +
> + if (pmu->flags & DDR_CAP_AXI_ID) {
> + if (event->attr.config == EVENT_AXI_READ ||
> + event->attr.config == EVENT_AXI_WRITE) {
> + int val = event->attr.config1;
> +
> + writel(val, pmu->base + COUNTER_DPCR1);
> + }
> + }
> +
> + local64_set(&hwc->prev_count, 0);
> +
> + ddr_perf_event_enable(pmu, event->attr.config, counter, true);
> + /*
> + * If the cycles counter wasn't explicitly selected,
> + * we will enable it now.
> + * cycles counter must be enabled otherwise other counters will
> + * stopped.
> + */
> + if (counter > 0 && !pmu->cycles_active)
> + ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
> + EVENT_CYCLES_COUNTER, true);
> +}
> +
> +static int ddr_perf_event_add(struct perf_event *event, int flags)
> +{
> + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> + struct hw_perf_event *hwc = &event->hw;
> + int counter;
> + int cfg = event->attr.config;
> +
> + counter = ddr_perf_alloc_counter(pmu, cfg);
> + if (counter < 0) {
> + dev_dbg(pmu->dev, "There are not enough counters\n");
> + return -EOPNOTSUPP;
> + }
> +
> + pmu->active_events[counter] = event;
> + pmu->total_events++;
> + hwc->idx = counter;
> +
> + if (flags & PERF_EF_START)
> + ddr_perf_event_start(event, flags);
> +
> + local64_set(&hwc->prev_count, ddr_perf_read_counter(pmu, counter));
> +
> + return 0;
> +}
> +
> +static void ddr_perf_event_stop(struct perf_event *event, int flags)
> +{
> + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> + struct hw_perf_event *hwc = &event->hw;
> + int counter = hwc->idx;
> +
> + ddr_perf_event_enable(pmu, event->attr.config, counter, false);
> + ddr_perf_event_update(event);
> +}
> +
> +static void ddr_perf_event_del(struct perf_event *event, int flags)
> +{
> + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> + struct hw_perf_event *hwc = &event->hw;
> + int counter = hwc->idx;
> +
> + ddr_perf_event_stop(event, PERF_EF_UPDATE);
> +
> + ddr_perf_free_counter(pmu, counter);
> + pmu->total_events--;
> + hwc->idx = -1;
> +
> + /* If all events have stopped, stop the cycles counter as well */
> + if ((pmu->total_events == 0) && pmu->cycles_active)
> + ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
> + EVENT_CYCLES_COUNTER, false);
> +}
> +
> +static int ddr_perf_init(struct ddr_pmu *pmu, void __iomem *base,
> + struct device *dev)
> +{
> + *pmu = (struct ddr_pmu) {
> + .pmu = (struct pmu) {
> + .task_ctx_nr = perf_invalid_context,
> + .attr_groups = attr_groups,
> + .event_init = ddr_perf_event_init,
> + .add = ddr_perf_event_add,
> + .del = ddr_perf_event_del,
> + .start = ddr_perf_event_start,
> + .stop = ddr_perf_event_stop,
> + .read = ddr_perf_event_update,
> + },
> + .base = base,
> + .dev = dev,
> + };
> +
> + return ida_simple_get(&ddr_ida, 0, 0, GFP_KERNEL);
> +}
> +
> +static irqreturn_t ddr_perf_irq_handler(int irq, void *p)
> +{
> + int i;
> + u8 reg;
> + int val;
> + int counter;
> + struct ddr_pmu *pmu = (struct ddr_pmu *) p;
> + struct perf_event *event;
> +
> + /* Only cycles counter overflowed can issue irq. all counters will
> + * be stopped when cycles counter overflow. but other counter don't stop
> + * when overflow happen. Update all of the local counter values,
> + * then reset the cycles counter, so the others can continue
> + * counting. cycles counter is fastest counter in all events. at last
> + * 4 times than other counters.
> + */
> + for (i = 0; i < NUM_COUNTER; i++) {
> +
> + if (!pmu->active_events[i])
> + continue;
> +
> + event = pmu->active_events[i];
> + counter = event->hw.idx;
> + reg = counter * 4 + COUNTER_CNTL;
> + val = readl(pmu->base + reg);
> +
> + ddr_perf_event_update(event);
> +
> + if (counter == EVENT_CYCLES_COUNTER) {
> + ddr_perf_event_enable(pmu,
> + EVENT_CYCLES_ID,
> + EVENT_CYCLES_COUNTER,
> + true);
> + ddr_perf_event_update(event);
> + }
> + }
> +
> + return IRQ_HANDLED;
> +}
> +
> +static int ddr_perf_offline_cpu(unsigned int cpu, struct hlist_node *node)
> +{
> + struct ddr_pmu *pmu = hlist_entry_safe(node, struct ddr_pmu, node);
> + int target;
> +
> + if (!cpumask_test_and_clear_cpu(cpu, &pmu->cpu))
> + return 0;
> +
> + target = cpumask_any_but(cpu_online_mask, cpu);
> + if (target >= nr_cpu_ids)
> + return 0;
> +
> + perf_pmu_migrate_context(&pmu->pmu, cpu, target);
> + cpumask_set_cpu(target, &pmu->cpu);
> +
> + return 0;
> +}
> +
> +static int ddr_perf_probe(struct platform_device *pdev)
> +{
> + struct ddr_pmu *pmu;
> + struct device_node *np;
> + void __iomem *base;
> + struct resource *iomem;
> + char *name;
> + char *hpname;
> + int num;
> + int ret;
> + u32 irq;
> + const struct of_device_id *of_id =
> + of_match_device(imx_ddr_pmu_dt_ids, &pdev->dev);
> +
> + iomem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> + base = devm_ioremap_resource(&pdev->dev, iomem);
> + if (IS_ERR(base))
> + return PTR_ERR(base);
> +
> + np = pdev->dev.of_node;
> +
> + pmu = devm_kzalloc(&pdev->dev, sizeof(*pmu), GFP_KERNEL);
> + if (!pmu)
> + return -ENOMEM;
> +
> + num = ddr_perf_init(pmu, base, &pdev->dev);
> +
> + name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "ddr%d", num);
> + if (!name)
> + return -ENOMEM;
> +
> + hpname = devm_kasprintf(&pdev->dev, GFP_KERNEL,
> + "perf/imx/ddr%d:online", num);
> + if (!hpname)
> + return -ENOMEM;
> +
> + pmu->flags = (uintptr_t) of_id->data;
> +
> + cpumask_set_cpu(raw_smp_processor_id(), &pmu->cpu);
> + ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN, hpname, NULL,
> + ddr_perf_offline_cpu);
> +
> + if (ret < 0) {
> + dev_err(&pdev->dev, "cpuhp_setup_state_multi failed\n");
> + goto ddr_perf_err;
> + }
> +
> + pmu->cpuhp_state = ret;
> +
> + /* Register the pmu instance for cpu hotplug */
> + cpuhp_state_add_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> +
> + ret = perf_pmu_register(&(pmu->pmu), name, -1);
> + if (ret)
> + goto ddr_perf_err;
> +
> + /* Request irq */
> + irq = of_irq_get(np, 0);
> + if (irq < 0) {
> + dev_err(&pdev->dev, "Failed to get irq: %d", irq);
> + ret = irq;
> + goto ddr_perf_irq_err;
> + }
> +
> + ret = devm_request_irq(&pdev->dev, irq,
> + ddr_perf_irq_handler,
> + IRQF_TRIGGER_RISING | IRQF_ONESHOT,
> + DDR_PERF_DEV_NAME,
> + pmu);
> + if (ret < 0) {
> + dev_err(&pdev->dev, "Request irq failed: %d", ret);
> + goto ddr_perf_irq_err;
> + }
> +
> + return 0;
> +
> +ddr_perf_irq_err:
> + perf_pmu_unregister(&(pmu->pmu));
> +
> +ddr_perf_err:
> + if (pmu->cpuhp_state)
> + cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> +
> + dev_warn(&pdev->dev, "i.MX8 DDR Perf PMU failed (%d), disabled\n", ret);
> + return ret;
> +}
> +
> +static int ddr_perf_remove(struct platform_device *pdev)
> +{
> + struct ddr_pmu *pmu = platform_get_drvdata(pdev);
> +
> + cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> + perf_pmu_unregister(&pmu->pmu);
> +
> + return 0;
> +}
> +
> +static struct platform_driver imx_ddr_pmu_driver = {
> + .driver = {
> + .name = "imx-ddr-pmu",
> + .of_match_table = imx_ddr_pmu_dt_ids,
> + },
> + .probe = ddr_perf_probe,
> + .remove = ddr_perf_remove,
> +};
> +
> +static int __init imx_ddr_pmu_init(void)
> +{
> + return platform_driver_register(&imx_ddr_pmu_driver);
> +}
> +
> +module_init(imx_ddr_pmu_init);
> +
> --
> 2.5.2
>
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
2019-02-22 19:45 ` Zhi Li
@ 2019-02-28 21:39 ` Zhi Li
-1 siblings, 0 replies; 43+ messages in thread
From: Zhi Li @ 2019-02-28 21:39 UTC (permalink / raw)
To: Frank Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
will.deacon, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Feb 22, 2019 at 1:45 PM Zhi Li <lznuaa@gmail.com> wrote:
>
> On Fri, Feb 15, 2019 at 12:03 PM Frank Li <frank.li@nxp.com> wrote:
> >
> > Add ddr performance monitor support for iMX8QXP
> >
> > There are 4 counters for ddr perfomance events.
> > counter 0 is dedicated for cycles.
> > you choose any up to 3 no cycles events.
>
> Any additional comments for these patches?
Mark:
There are no comments for two weeks.
Did you plan pick up these patches?
best regards
Frank Li
>
> best regards
> Frank Li
>
> >
> > for example:
> >
> > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> >
> > Support below events.
> >
> > ddr0/activate/ [Kernel PMU event]
> > ddr0/axid-read/ [Kernel PMU event]
> > ddr0/axid-write/ [Kernel PMU event]
> > ddr0/cycles/ [Kernel PMU event]
> > ddr0/hp-read-credit-cnt/ [Kernel PMU event]
> > ddr0/hp-read/ [Kernel PMU event]
> > ddr0/hp-req-nodcredit/ [Kernel PMU event]
> > ddr0/hp-xact-credit/ [Kernel PMU event]
> > ddr0/load-mode/ [Kernel PMU event]
> > ddr0/lp-read-credit-cnt/ [Kernel PMU event]
> > ddr0/lp-req-nocredit/ [Kernel PMU event]
> > ddr0/lp-xact-credit/ [Kernel PMU event]
> > ddr0/mwr/ [Kernel PMU event]
> > ddr0/precharge/ [Kernel PMU event]
> > ddr0/raw-hazard/ [Kernel PMU event]
> > ddr0/read-access/ [Kernel PMU event]
> > ddr0/read-activate/ [Kernel PMU event]
> > ddr0/read-command/ [Kernel PMU event]
> > ddr0/read-cycles/ [Kernel PMU event]
> > ddr0/read-modify-write-command/ [Kernel PMU event]
> > ddr0/read-queue-depth/ [Kernel PMU event]
> > ddr0/read-write-transition/ [Kernel PMU event]
> > ddr0/read/ [Kernel PMU event]
> > ddr0/refresh/ [Kernel PMU event]
> > ddr0/selfresh/ [Kernel PMU event]
> > ddr0/wr-xact-credit/ [Kernel PMU event]
> > ddr0/write-access/ [Kernel PMU event]
> > ddr0/write-command/ [Kernel PMU event]
> > ddr0/write-credit-cnt/ [Kernel PMU event]
> > ddr0/write-cycles/ [Kernel PMU event]
> > ddr0/write-queue-depth/ [Kernel PMU event]
> > ddr0/write/
> >
> > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > ---
> > Change from v3 to v4
> > * Change FSL_IMX8_DDR_PERF to FSL_IMX8_DDR_PMU
> > * sort include
> > * remove struct fsl_ddr_devtype_data
> > * Added comment need disable control first
> > * Added comment about must enable cycle counter
> > * Added macro for EVENT_AXI_READ, remove hardcode 0x41 and 0x42
> > * Added comment about cycle counter is fastest one
> >
> > Change from v2 to v3
> > * remove kfree
> >
> > Change from V1 to V2
> > * update Kconfig by use i.MX8 instead of i.MX8 QXP
> > * remove gpl statememnt since SPDX tag
> > * use dev_kzalloc
> > * use dev_err
> > * commit message show axi_read 0x41\axi_write 0x42
> > * commit message show cycles must be enabled
> > * Irq only issue at cycles overflow
> > * use NUM_COUNTER
> > * use devm_request_irq
> > * add hotplug callback to handle context migration
> >
> > drivers/perf/Kconfig | 7 +
> > drivers/perf/Makefile | 1 +
> > drivers/perf/fsl_imx8_ddr_perf.c | 573 +++++++++++++++++++++++++++++++++++++++
> > 3 files changed, 581 insertions(+)
> > create mode 100644 drivers/perf/fsl_imx8_ddr_perf.c
> >
> > diff --git a/drivers/perf/Kconfig b/drivers/perf/Kconfig
> > index af9bc17..e567437 100644
> > --- a/drivers/perf/Kconfig
> > +++ b/drivers/perf/Kconfig
> > @@ -61,6 +61,13 @@ config ARM_DSU_PMU
> > system, control logic. The PMU allows counting various events related
> > to DSU.
> >
> > +config FSL_IMX8_DDR_PMU
> > + tristate "Freescale i.MX8 DDR perf monitor"
> > + depends on ARCH_MXC
> > + help
> > + Provides support for ddr perfomance monitor in i.MX8. Provide memory
> > + througput information.
> > +
> > config HISI_PMU
> > bool "HiSilicon SoC PMU"
> > depends on ARM64 && ACPI
> > diff --git a/drivers/perf/Makefile b/drivers/perf/Makefile
> > index 909f27f..0bc066b 100644
> > --- a/drivers/perf/Makefile
> > +++ b/drivers/perf/Makefile
> > @@ -4,6 +4,7 @@ obj-$(CONFIG_ARM_CCN) += arm-ccn.o
> > obj-$(CONFIG_ARM_DSU_PMU) += arm_dsu_pmu.o
> > obj-$(CONFIG_ARM_PMU) += arm_pmu.o arm_pmu_platform.o
> > obj-$(CONFIG_ARM_PMU_ACPI) += arm_pmu_acpi.o
> > +obj-$(CONFIG_FSL_IMX8_DDR_PMU) += fsl_imx8_ddr_perf.o
> > obj-$(CONFIG_HISI_PMU) += hisilicon/
> > obj-$(CONFIG_QCOM_L2_PMU) += qcom_l2_pmu.o
> > obj-$(CONFIG_QCOM_L3_PMU) += qcom_l3_pmu.o
> > diff --git a/drivers/perf/fsl_imx8_ddr_perf.c b/drivers/perf/fsl_imx8_ddr_perf.c
> > new file mode 100644
> > index 0000000..3d02daf
> > --- /dev/null
> > +++ b/drivers/perf/fsl_imx8_ddr_perf.c
> > @@ -0,0 +1,573 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * Copyright 2017 NXP
> > + * Copyright 2016 Freescale Semiconductor, Inc.
> > + */
> > +
> > +#include <linux/init.h>
> > +#include <linux/interrupt.h>
> > +#include <linux/io.h>
> > +#include <linux/module.h>
> > +#include <linux/of.h>
> > +#include <linux/of_address.h>
> > +#include <linux/of_device.h>
> > +#include <linux/of_irq.h>
> > +#include <linux/perf_event.h>
> > +#include <linux/slab.h>
> > +
> > +#define COUNTER_CNTL 0x0
> > +#define COUNTER_READ 0x20
> > +
> > +#define COUNTER_DPCR1 0x30
> > +
> > +#define CNTL_OVER 0x1
> > +#define CNTL_CLEAR 0x2
> > +#define CNTL_EN 0x4
> > +#define CNTL_EN_MASK 0xFFFFFFFB
> > +#define CNTL_CLEAR_MASK 0xFFFFFFFD
> > +#define CNTL_OVER_MASK 0xFFFFFFFE
> > +
> > +#define CNTL_CSV_SHIFT 24
> > +#define CNTL_CSV_MASK (0xFF << CNTL_CSV_SHIFT)
> > +
> > +#define EVENT_CYCLES_ID 0
> > +#define EVENT_CYCLES_COUNTER 0
> > +#define NUM_COUNTER 4
> > +#define MAX_EVENT 3
> > +#define EVENT_AXI_READ 0x41
> > +#define EVENT_AXI_WRITE 0x42
> > +
> > +#define to_ddr_pmu(p) container_of(p, struct ddr_pmu, pmu)
> > +
> > +#define DDR_PERF_DEV_NAME "ddr_perf"
> > +
> > +static DEFINE_IDA(ddr_ida);
> > +
> > +PMU_EVENT_ATTR_STRING(cycles, ddr_perf_cycles, "event=0x00");
> > +PMU_EVENT_ATTR_STRING(selfresh, ddr_perf_selfresh, "event=0x01");
> > +PMU_EVENT_ATTR_STRING(read-access, ddr_perf_read_accesses, "event=0x04");
> > +PMU_EVENT_ATTR_STRING(write-access, ddr_perf_write_accesses, "event=0x05");
> > +PMU_EVENT_ATTR_STRING(read-queue-depth, ddr_perf_read_queue_depth,
> > + "event=0x08");
> > +PMU_EVENT_ATTR_STRING(write-queue-depth, ddr_perf_write_queue_depth,
> > + "event=0x09");
> > +PMU_EVENT_ATTR_STRING(lp-read-credit-cnt, ddr_perf_lp_read_credit_cnt,
> > + "event=0x10");
> > +PMU_EVENT_ATTR_STRING(hp-read-credit-cnt, ddr_perf_hp_read_credit_cnt,
> > + "event=0x11");
> > +PMU_EVENT_ATTR_STRING(write-credit-cnt, ddr_perf_write_credit_cnt,
> > + "event=0x12");
> > +PMU_EVENT_ATTR_STRING(read-command, ddr_perf_read_command, "event=0x20");
> > +PMU_EVENT_ATTR_STRING(write-command, ddr_perf_write_command, "event=0x21");
> > +PMU_EVENT_ATTR_STRING(read-modify-write-command,
> > + ddr_perf_read_modify_write_command, "event=0x22");
> > +PMU_EVENT_ATTR_STRING(hp-read, ddr_perf_hp_read, "event=0x23");
> > +PMU_EVENT_ATTR_STRING(hp-req-nodcredit, ddr_perf_hp_req_nocredit, "event=0x24");
> > +PMU_EVENT_ATTR_STRING(hp-xact-credit, ddr_perf_hp_xact_credit, "event=0x25");
> > +PMU_EVENT_ATTR_STRING(lp-req-nocredit, ddr_perf_lp_req_nocredit, "event=0x26");
> > +PMU_EVENT_ATTR_STRING(lp-xact-credit, ddr_perf_lp_xact_credit, "event=0x27");
> > +PMU_EVENT_ATTR_STRING(wr-xact-credit, ddr_perf_wr_xact_credit, "event=0x29");
> > +PMU_EVENT_ATTR_STRING(read-cycles, ddr_perf_read_cycles, "event=0x2a");
> > +PMU_EVENT_ATTR_STRING(write-cycles, ddr_perf_write_cycles, "event=0x2b");
> > +PMU_EVENT_ATTR_STRING(read-write-transition, ddr_perf_read_write_transition,
> > + "event=0x30");
> > +PMU_EVENT_ATTR_STRING(precharge, ddr_perf_precharge, "event=0x31");
> > +PMU_EVENT_ATTR_STRING(activate, ddr_perf_activate, "event=0x32");
> > +PMU_EVENT_ATTR_STRING(load-mode, ddr_perf_load_mode, "event=0x33");
> > +PMU_EVENT_ATTR_STRING(mwr, ddr_perf_mwr, "event=0x34");
> > +PMU_EVENT_ATTR_STRING(read, ddr_perf_read, "event=0x35");
> > +PMU_EVENT_ATTR_STRING(read-activate, ddr_perf_read_activate, "event=0x36");
> > +PMU_EVENT_ATTR_STRING(refresh, ddr_perf_refresh, "event=0x37");
> > +PMU_EVENT_ATTR_STRING(write, ddr_perf_write, "event=0x38");
> > +PMU_EVENT_ATTR_STRING(raw-hazard, ddr_perf_raw_hazard, "event=0x39");
> > +
> > +PMU_EVENT_ATTR_STRING(axid-read, ddr_perf_axid_read,
> > + "event="__stringify(EVENT_AXI_READ));
> > +PMU_EVENT_ATTR_STRING(axid-write, ddr_perf_axid_write,
> > + "event="__stringify(EVENT_AXI_WRITE));
> > +
> > +#define DDR_CAP_AXI_ID 0x1
> > +
> > +static const struct of_device_id imx_ddr_pmu_dt_ids[] = {
> > + { .compatible = "fsl,imx8-ddr-pmu",},
> > + { .compatible = "fsl,imx8m-ddr-pmu", .data = (void *)DDR_CAP_AXI_ID},
> > + { /* sentinel */ }
> > +};
> > +
> > +struct ddr_pmu {
> > + struct pmu pmu;
> > + void __iomem *base;
> > + cpumask_t cpu;
> > + struct hlist_node node;
> > + struct device *dev;
> > + struct perf_event *active_events[NUM_COUNTER];
> > + int total_events;
> > + enum cpuhp_state cpuhp_state;
> > + bool cycles_active;
> > + uintptr_t flags;
> > +};
> > +
> > +static ssize_t ddr_perf_cpumask_show(struct device *dev,
> > + struct device_attribute *attr, char *buf)
> > +{
> > + struct ddr_pmu *pmu = dev_get_drvdata(dev);
> > +
> > + return cpumap_print_to_pagebuf(true, buf, &pmu->cpu);
> > +}
> > +
> > +static struct device_attribute ddr_perf_cpumask_attr =
> > + __ATTR(cpumask, 0444, ddr_perf_cpumask_show, NULL);
> > +
> > +static struct attribute *ddr_perf_cpumask_attrs[] = {
> > + &ddr_perf_cpumask_attr.attr,
> > + NULL,
> > +};
> > +
> > +static struct attribute_group ddr_perf_cpumask_attr_group = {
> > + .attrs = ddr_perf_cpumask_attrs,
> > +};
> > +
> > +static struct attribute *ddr_perf_events_attrs[] = {
> > + &ddr_perf_cycles.attr.attr,
> > + &ddr_perf_selfresh.attr.attr,
> > + &ddr_perf_read_accesses.attr.attr,
> > + &ddr_perf_write_accesses.attr.attr,
> > + &ddr_perf_read_queue_depth.attr.attr,
> > + &ddr_perf_write_queue_depth.attr.attr,
> > + &ddr_perf_lp_read_credit_cnt.attr.attr,
> > + &ddr_perf_hp_read_credit_cnt.attr.attr,
> > + &ddr_perf_write_credit_cnt.attr.attr,
> > + &ddr_perf_read_command.attr.attr,
> > + &ddr_perf_write_command.attr.attr,
> > + &ddr_perf_read_modify_write_command.attr.attr,
> > + &ddr_perf_hp_read.attr.attr,
> > + &ddr_perf_hp_req_nocredit.attr.attr,
> > + &ddr_perf_hp_xact_credit.attr.attr,
> > + &ddr_perf_lp_req_nocredit.attr.attr,
> > + &ddr_perf_lp_xact_credit.attr.attr,
> > + &ddr_perf_wr_xact_credit.attr.attr,
> > + &ddr_perf_read_cycles.attr.attr,
> > + &ddr_perf_write_cycles.attr.attr,
> > + &ddr_perf_read_write_transition.attr.attr,
> > + &ddr_perf_precharge.attr.attr,
> > + &ddr_perf_activate.attr.attr,
> > + &ddr_perf_load_mode.attr.attr,
> > + &ddr_perf_mwr.attr.attr,
> > + &ddr_perf_read.attr.attr,
> > + &ddr_perf_read_activate.attr.attr,
> > + &ddr_perf_refresh.attr.attr,
> > + &ddr_perf_write.attr.attr,
> > + &ddr_perf_raw_hazard.attr.attr,
> > + &ddr_perf_axid_read.attr.attr,
> > + &ddr_perf_axid_write.attr.attr,
> > + NULL,
> > +};
> > +
> > +static struct attribute_group ddr_perf_events_attr_group = {
> > + .name = "events",
> > + .attrs = ddr_perf_events_attrs,
> > +};
> > +
> > +PMU_FORMAT_ATTR(event, "config:0-63");
> > +PMU_FORMAT_ATTR(axi_id, "config1:0-63");
> > +
> > +static struct attribute *ddr_perf_format_attrs[] = {
> > + &format_attr_event.attr,
> > + &format_attr_axi_id.attr,
> > + NULL,
> > +};
> > +
> > +static struct attribute_group ddr_perf_format_attr_group = {
> > + .name = "format",
> > + .attrs = ddr_perf_format_attrs,
> > +};
> > +
> > +static const struct attribute_group *attr_groups[] = {
> > + &ddr_perf_events_attr_group,
> > + &ddr_perf_format_attr_group,
> > + &ddr_perf_cpumask_attr_group,
> > + NULL,
> > +};
> > +
> > +static u32 ddr_perf_alloc_counter(struct ddr_pmu *pmu, int event)
> > +{
> > + int i;
> > +
> > + /* Always map cycle event to counter 0
> > + Cycles counter is dedicated for cycle event
> > + can't used for the other events
> > + */
> > + if (event == EVENT_CYCLES_ID)
> > + return EVENT_CYCLES_COUNTER;
> > +
> > + for (i = 1; i < NUM_COUNTER; i++)
> > + if (pmu->active_events[i] == NULL)
> > + return i;
> > +
> > + return -ENOENT;
> > +}
> > +
> > +static u32 ddr_perf_free_counter(struct ddr_pmu *pmu, int counter)
> > +{
> > + if (counter < 0 || counter >= NUM_COUNTER)
> > + return -ENOENT;
> > +
> > + pmu->active_events[counter] = NULL;
> > +
> > + return 0;
> > +}
> > +
> > +static u32 ddr_perf_read_counter(struct ddr_pmu *pmu, int counter)
> > +{
> > + return readl(pmu->base + COUNTER_READ + counter * 4);
> > +}
> > +
> > +static int ddr_perf_event_init(struct perf_event *event)
> > +{
> > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > + struct hw_perf_event *hwc = &event->hw;
> > +
> > + if (event->attr.type != event->pmu->type)
> > + return -ENOENT;
> > +
> > + if (is_sampling_event(event) || event->attach_state & PERF_ATTACH_TASK)
> > + return -EOPNOTSUPP;
> > +
> > + if (event->cpu < 0) {
> > + dev_warn(pmu->dev, "Can't provide per-task data!\n");
> > + return -EOPNOTSUPP;
> > + }
> > +
> > + if (event->attr.exclude_user ||
> > + event->attr.exclude_kernel ||
> > + event->attr.exclude_hv ||
> > + event->attr.exclude_idle ||
> > + event->attr.exclude_host ||
> > + event->attr.exclude_guest ||
> > + event->attr.sample_period)
> > + return -EINVAL;
> > +
> > + event->cpu = cpumask_first(&pmu->cpu);
> > + hwc->idx = -1;
> > +
> > + return 0;
> > +}
> > +
> > +
> > +static void ddr_perf_event_update(struct perf_event *event)
> > +{
> > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > + struct hw_perf_event *hwc = &event->hw;
> > + u64 delta, prev_raw_count, new_raw_count;
> > + int counter = hwc->idx;
> > +
> > + do {
> > + prev_raw_count = local64_read(&hwc->prev_count);
> > + new_raw_count = ddr_perf_read_counter(pmu, counter);
> > + } while (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
> > + new_raw_count) != prev_raw_count);
> > +
> > + delta = (new_raw_count - prev_raw_count) & 0xFFFFFFFF;
> > +
> > + local64_add(delta, &event->count);
> > +}
> > +
> > +static void ddr_perf_event_enable(struct ddr_pmu *pmu, int config,
> > + int counter, bool enable)
> > +{
> > + u8 reg = counter * 4 + COUNTER_CNTL;
> > + int val;
> > +
> > + if (enable) {
> > + /* must disable first, then enable again
> > + * otherwise, cycle counter will not work
> > + * if previous state is enabled.
> > + */
> > + writel(0, pmu->base + reg);
> > + val = CNTL_EN | CNTL_CLEAR;
> > + val |= (config << CNTL_CSV_SHIFT) & CNTL_CSV_MASK;
> > + } else {
> > + /* Disable counter */
> > + val = readl(pmu->base + reg) & CNTL_EN_MASK;
> > + }
> > +
> > + writel(val, pmu->base + reg);
> > +
> > + if (config == EVENT_CYCLES_ID)
> > + pmu->cycles_active = enable;
> > +}
> > +
> > +static void ddr_perf_event_start(struct perf_event *event, int flags)
> > +{
> > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > + struct hw_perf_event *hwc = &event->hw;
> > + int counter = hwc->idx;
> > +
> > + if (pmu->flags & DDR_CAP_AXI_ID) {
> > + if (event->attr.config == EVENT_AXI_READ ||
> > + event->attr.config == EVENT_AXI_WRITE) {
> > + int val = event->attr.config1;
> > +
> > + writel(val, pmu->base + COUNTER_DPCR1);
> > + }
> > + }
> > +
> > + local64_set(&hwc->prev_count, 0);
> > +
> > + ddr_perf_event_enable(pmu, event->attr.config, counter, true);
> > + /*
> > + * If the cycles counter wasn't explicitly selected,
> > + * we will enable it now.
> > + * cycles counter must be enabled otherwise other counters will
> > + * stopped.
> > + */
> > + if (counter > 0 && !pmu->cycles_active)
> > + ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
> > + EVENT_CYCLES_COUNTER, true);
> > +}
> > +
> > +static int ddr_perf_event_add(struct perf_event *event, int flags)
> > +{
> > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > + struct hw_perf_event *hwc = &event->hw;
> > + int counter;
> > + int cfg = event->attr.config;
> > +
> > + counter = ddr_perf_alloc_counter(pmu, cfg);
> > + if (counter < 0) {
> > + dev_dbg(pmu->dev, "There are not enough counters\n");
> > + return -EOPNOTSUPP;
> > + }
> > +
> > + pmu->active_events[counter] = event;
> > + pmu->total_events++;
> > + hwc->idx = counter;
> > +
> > + if (flags & PERF_EF_START)
> > + ddr_perf_event_start(event, flags);
> > +
> > + local64_set(&hwc->prev_count, ddr_perf_read_counter(pmu, counter));
> > +
> > + return 0;
> > +}
> > +
> > +static void ddr_perf_event_stop(struct perf_event *event, int flags)
> > +{
> > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > + struct hw_perf_event *hwc = &event->hw;
> > + int counter = hwc->idx;
> > +
> > + ddr_perf_event_enable(pmu, event->attr.config, counter, false);
> > + ddr_perf_event_update(event);
> > +}
> > +
> > +static void ddr_perf_event_del(struct perf_event *event, int flags)
> > +{
> > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > + struct hw_perf_event *hwc = &event->hw;
> > + int counter = hwc->idx;
> > +
> > + ddr_perf_event_stop(event, PERF_EF_UPDATE);
> > +
> > + ddr_perf_free_counter(pmu, counter);
> > + pmu->total_events--;
> > + hwc->idx = -1;
> > +
> > + /* If all events have stopped, stop the cycles counter as well */
> > + if ((pmu->total_events == 0) && pmu->cycles_active)
> > + ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
> > + EVENT_CYCLES_COUNTER, false);
> > +}
> > +
> > +static int ddr_perf_init(struct ddr_pmu *pmu, void __iomem *base,
> > + struct device *dev)
> > +{
> > + *pmu = (struct ddr_pmu) {
> > + .pmu = (struct pmu) {
> > + .task_ctx_nr = perf_invalid_context,
> > + .attr_groups = attr_groups,
> > + .event_init = ddr_perf_event_init,
> > + .add = ddr_perf_event_add,
> > + .del = ddr_perf_event_del,
> > + .start = ddr_perf_event_start,
> > + .stop = ddr_perf_event_stop,
> > + .read = ddr_perf_event_update,
> > + },
> > + .base = base,
> > + .dev = dev,
> > + };
> > +
> > + return ida_simple_get(&ddr_ida, 0, 0, GFP_KERNEL);
> > +}
> > +
> > +static irqreturn_t ddr_perf_irq_handler(int irq, void *p)
> > +{
> > + int i;
> > + u8 reg;
> > + int val;
> > + int counter;
> > + struct ddr_pmu *pmu = (struct ddr_pmu *) p;
> > + struct perf_event *event;
> > +
> > + /* Only cycles counter overflowed can issue irq. all counters will
> > + * be stopped when cycles counter overflow. but other counter don't stop
> > + * when overflow happen. Update all of the local counter values,
> > + * then reset the cycles counter, so the others can continue
> > + * counting. cycles counter is fastest counter in all events. at last
> > + * 4 times than other counters.
> > + */
> > + for (i = 0; i < NUM_COUNTER; i++) {
> > +
> > + if (!pmu->active_events[i])
> > + continue;
> > +
> > + event = pmu->active_events[i];
> > + counter = event->hw.idx;
> > + reg = counter * 4 + COUNTER_CNTL;
> > + val = readl(pmu->base + reg);
> > +
> > + ddr_perf_event_update(event);
> > +
> > + if (counter == EVENT_CYCLES_COUNTER) {
> > + ddr_perf_event_enable(pmu,
> > + EVENT_CYCLES_ID,
> > + EVENT_CYCLES_COUNTER,
> > + true);
> > + ddr_perf_event_update(event);
> > + }
> > + }
> > +
> > + return IRQ_HANDLED;
> > +}
> > +
> > +static int ddr_perf_offline_cpu(unsigned int cpu, struct hlist_node *node)
> > +{
> > + struct ddr_pmu *pmu = hlist_entry_safe(node, struct ddr_pmu, node);
> > + int target;
> > +
> > + if (!cpumask_test_and_clear_cpu(cpu, &pmu->cpu))
> > + return 0;
> > +
> > + target = cpumask_any_but(cpu_online_mask, cpu);
> > + if (target >= nr_cpu_ids)
> > + return 0;
> > +
> > + perf_pmu_migrate_context(&pmu->pmu, cpu, target);
> > + cpumask_set_cpu(target, &pmu->cpu);
> > +
> > + return 0;
> > +}
> > +
> > +static int ddr_perf_probe(struct platform_device *pdev)
> > +{
> > + struct ddr_pmu *pmu;
> > + struct device_node *np;
> > + void __iomem *base;
> > + struct resource *iomem;
> > + char *name;
> > + char *hpname;
> > + int num;
> > + int ret;
> > + u32 irq;
> > + const struct of_device_id *of_id =
> > + of_match_device(imx_ddr_pmu_dt_ids, &pdev->dev);
> > +
> > + iomem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> > + base = devm_ioremap_resource(&pdev->dev, iomem);
> > + if (IS_ERR(base))
> > + return PTR_ERR(base);
> > +
> > + np = pdev->dev.of_node;
> > +
> > + pmu = devm_kzalloc(&pdev->dev, sizeof(*pmu), GFP_KERNEL);
> > + if (!pmu)
> > + return -ENOMEM;
> > +
> > + num = ddr_perf_init(pmu, base, &pdev->dev);
> > +
> > + name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "ddr%d", num);
> > + if (!name)
> > + return -ENOMEM;
> > +
> > + hpname = devm_kasprintf(&pdev->dev, GFP_KERNEL,
> > + "perf/imx/ddr%d:online", num);
> > + if (!hpname)
> > + return -ENOMEM;
> > +
> > + pmu->flags = (uintptr_t) of_id->data;
> > +
> > + cpumask_set_cpu(raw_smp_processor_id(), &pmu->cpu);
> > + ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN, hpname, NULL,
> > + ddr_perf_offline_cpu);
> > +
> > + if (ret < 0) {
> > + dev_err(&pdev->dev, "cpuhp_setup_state_multi failed\n");
> > + goto ddr_perf_err;
> > + }
> > +
> > + pmu->cpuhp_state = ret;
> > +
> > + /* Register the pmu instance for cpu hotplug */
> > + cpuhp_state_add_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> > +
> > + ret = perf_pmu_register(&(pmu->pmu), name, -1);
> > + if (ret)
> > + goto ddr_perf_err;
> > +
> > + /* Request irq */
> > + irq = of_irq_get(np, 0);
> > + if (irq < 0) {
> > + dev_err(&pdev->dev, "Failed to get irq: %d", irq);
> > + ret = irq;
> > + goto ddr_perf_irq_err;
> > + }
> > +
> > + ret = devm_request_irq(&pdev->dev, irq,
> > + ddr_perf_irq_handler,
> > + IRQF_TRIGGER_RISING | IRQF_ONESHOT,
> > + DDR_PERF_DEV_NAME,
> > + pmu);
> > + if (ret < 0) {
> > + dev_err(&pdev->dev, "Request irq failed: %d", ret);
> > + goto ddr_perf_irq_err;
> > + }
> > +
> > + return 0;
> > +
> > +ddr_perf_irq_err:
> > + perf_pmu_unregister(&(pmu->pmu));
> > +
> > +ddr_perf_err:
> > + if (pmu->cpuhp_state)
> > + cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> > +
> > + dev_warn(&pdev->dev, "i.MX8 DDR Perf PMU failed (%d), disabled\n", ret);
> > + return ret;
> > +}
> > +
> > +static int ddr_perf_remove(struct platform_device *pdev)
> > +{
> > + struct ddr_pmu *pmu = platform_get_drvdata(pdev);
> > +
> > + cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> > + perf_pmu_unregister(&pmu->pmu);
> > +
> > + return 0;
> > +}
> > +
> > +static struct platform_driver imx_ddr_pmu_driver = {
> > + .driver = {
> > + .name = "imx-ddr-pmu",
> > + .of_match_table = imx_ddr_pmu_dt_ids,
> > + },
> > + .probe = ddr_perf_probe,
> > + .remove = ddr_perf_remove,
> > +};
> > +
> > +static int __init imx_ddr_pmu_init(void)
> > +{
> > + return platform_driver_register(&imx_ddr_pmu_driver);
> > +}
> > +
> > +module_init(imx_ddr_pmu_init);
> > +
> > --
> > 2.5.2
> >
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
@ 2019-02-28 21:39 ` Zhi Li
0 siblings, 0 replies; 43+ messages in thread
From: Zhi Li @ 2019-02-28 21:39 UTC (permalink / raw)
To: Frank Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
will.deacon, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Feb 22, 2019 at 1:45 PM Zhi Li <lznuaa@gmail.com> wrote:
>
> On Fri, Feb 15, 2019 at 12:03 PM Frank Li <frank.li@nxp.com> wrote:
> >
> > Add ddr performance monitor support for iMX8QXP
> >
> > There are 4 counters for ddr perfomance events.
> > counter 0 is dedicated for cycles.
> > you choose any up to 3 no cycles events.
>
> Any additional comments for these patches?
Mark:
There are no comments for two weeks.
Did you plan pick up these patches?
best regards
Frank Li
>
> best regards
> Frank Li
>
> >
> > for example:
> >
> > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> >
> > Support below events.
> >
> > ddr0/activate/ [Kernel PMU event]
> > ddr0/axid-read/ [Kernel PMU event]
> > ddr0/axid-write/ [Kernel PMU event]
> > ddr0/cycles/ [Kernel PMU event]
> > ddr0/hp-read-credit-cnt/ [Kernel PMU event]
> > ddr0/hp-read/ [Kernel PMU event]
> > ddr0/hp-req-nodcredit/ [Kernel PMU event]
> > ddr0/hp-xact-credit/ [Kernel PMU event]
> > ddr0/load-mode/ [Kernel PMU event]
> > ddr0/lp-read-credit-cnt/ [Kernel PMU event]
> > ddr0/lp-req-nocredit/ [Kernel PMU event]
> > ddr0/lp-xact-credit/ [Kernel PMU event]
> > ddr0/mwr/ [Kernel PMU event]
> > ddr0/precharge/ [Kernel PMU event]
> > ddr0/raw-hazard/ [Kernel PMU event]
> > ddr0/read-access/ [Kernel PMU event]
> > ddr0/read-activate/ [Kernel PMU event]
> > ddr0/read-command/ [Kernel PMU event]
> > ddr0/read-cycles/ [Kernel PMU event]
> > ddr0/read-modify-write-command/ [Kernel PMU event]
> > ddr0/read-queue-depth/ [Kernel PMU event]
> > ddr0/read-write-transition/ [Kernel PMU event]
> > ddr0/read/ [Kernel PMU event]
> > ddr0/refresh/ [Kernel PMU event]
> > ddr0/selfresh/ [Kernel PMU event]
> > ddr0/wr-xact-credit/ [Kernel PMU event]
> > ddr0/write-access/ [Kernel PMU event]
> > ddr0/write-command/ [Kernel PMU event]
> > ddr0/write-credit-cnt/ [Kernel PMU event]
> > ddr0/write-cycles/ [Kernel PMU event]
> > ddr0/write-queue-depth/ [Kernel PMU event]
> > ddr0/write/
> >
> > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > ---
> > Change from v3 to v4
> > * Change FSL_IMX8_DDR_PERF to FSL_IMX8_DDR_PMU
> > * sort include
> > * remove struct fsl_ddr_devtype_data
> > * Added comment need disable control first
> > * Added comment about must enable cycle counter
> > * Added macro for EVENT_AXI_READ, remove hardcode 0x41 and 0x42
> > * Added comment about cycle counter is fastest one
> >
> > Change from v2 to v3
> > * remove kfree
> >
> > Change from V1 to V2
> > * update Kconfig by use i.MX8 instead of i.MX8 QXP
> > * remove gpl statememnt since SPDX tag
> > * use dev_kzalloc
> > * use dev_err
> > * commit message show axi_read 0x41\axi_write 0x42
> > * commit message show cycles must be enabled
> > * Irq only issue at cycles overflow
> > * use NUM_COUNTER
> > * use devm_request_irq
> > * add hotplug callback to handle context migration
> >
> > drivers/perf/Kconfig | 7 +
> > drivers/perf/Makefile | 1 +
> > drivers/perf/fsl_imx8_ddr_perf.c | 573 +++++++++++++++++++++++++++++++++++++++
> > 3 files changed, 581 insertions(+)
> > create mode 100644 drivers/perf/fsl_imx8_ddr_perf.c
> >
> > diff --git a/drivers/perf/Kconfig b/drivers/perf/Kconfig
> > index af9bc17..e567437 100644
> > --- a/drivers/perf/Kconfig
> > +++ b/drivers/perf/Kconfig
> > @@ -61,6 +61,13 @@ config ARM_DSU_PMU
> > system, control logic. The PMU allows counting various events related
> > to DSU.
> >
> > +config FSL_IMX8_DDR_PMU
> > + tristate "Freescale i.MX8 DDR perf monitor"
> > + depends on ARCH_MXC
> > + help
> > + Provides support for ddr perfomance monitor in i.MX8. Provide memory
> > + througput information.
> > +
> > config HISI_PMU
> > bool "HiSilicon SoC PMU"
> > depends on ARM64 && ACPI
> > diff --git a/drivers/perf/Makefile b/drivers/perf/Makefile
> > index 909f27f..0bc066b 100644
> > --- a/drivers/perf/Makefile
> > +++ b/drivers/perf/Makefile
> > @@ -4,6 +4,7 @@ obj-$(CONFIG_ARM_CCN) += arm-ccn.o
> > obj-$(CONFIG_ARM_DSU_PMU) += arm_dsu_pmu.o
> > obj-$(CONFIG_ARM_PMU) += arm_pmu.o arm_pmu_platform.o
> > obj-$(CONFIG_ARM_PMU_ACPI) += arm_pmu_acpi.o
> > +obj-$(CONFIG_FSL_IMX8_DDR_PMU) += fsl_imx8_ddr_perf.o
> > obj-$(CONFIG_HISI_PMU) += hisilicon/
> > obj-$(CONFIG_QCOM_L2_PMU) += qcom_l2_pmu.o
> > obj-$(CONFIG_QCOM_L3_PMU) += qcom_l3_pmu.o
> > diff --git a/drivers/perf/fsl_imx8_ddr_perf.c b/drivers/perf/fsl_imx8_ddr_perf.c
> > new file mode 100644
> > index 0000000..3d02daf
> > --- /dev/null
> > +++ b/drivers/perf/fsl_imx8_ddr_perf.c
> > @@ -0,0 +1,573 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * Copyright 2017 NXP
> > + * Copyright 2016 Freescale Semiconductor, Inc.
> > + */
> > +
> > +#include <linux/init.h>
> > +#include <linux/interrupt.h>
> > +#include <linux/io.h>
> > +#include <linux/module.h>
> > +#include <linux/of.h>
> > +#include <linux/of_address.h>
> > +#include <linux/of_device.h>
> > +#include <linux/of_irq.h>
> > +#include <linux/perf_event.h>
> > +#include <linux/slab.h>
> > +
> > +#define COUNTER_CNTL 0x0
> > +#define COUNTER_READ 0x20
> > +
> > +#define COUNTER_DPCR1 0x30
> > +
> > +#define CNTL_OVER 0x1
> > +#define CNTL_CLEAR 0x2
> > +#define CNTL_EN 0x4
> > +#define CNTL_EN_MASK 0xFFFFFFFB
> > +#define CNTL_CLEAR_MASK 0xFFFFFFFD
> > +#define CNTL_OVER_MASK 0xFFFFFFFE
> > +
> > +#define CNTL_CSV_SHIFT 24
> > +#define CNTL_CSV_MASK (0xFF << CNTL_CSV_SHIFT)
> > +
> > +#define EVENT_CYCLES_ID 0
> > +#define EVENT_CYCLES_COUNTER 0
> > +#define NUM_COUNTER 4
> > +#define MAX_EVENT 3
> > +#define EVENT_AXI_READ 0x41
> > +#define EVENT_AXI_WRITE 0x42
> > +
> > +#define to_ddr_pmu(p) container_of(p, struct ddr_pmu, pmu)
> > +
> > +#define DDR_PERF_DEV_NAME "ddr_perf"
> > +
> > +static DEFINE_IDA(ddr_ida);
> > +
> > +PMU_EVENT_ATTR_STRING(cycles, ddr_perf_cycles, "event=0x00");
> > +PMU_EVENT_ATTR_STRING(selfresh, ddr_perf_selfresh, "event=0x01");
> > +PMU_EVENT_ATTR_STRING(read-access, ddr_perf_read_accesses, "event=0x04");
> > +PMU_EVENT_ATTR_STRING(write-access, ddr_perf_write_accesses, "event=0x05");
> > +PMU_EVENT_ATTR_STRING(read-queue-depth, ddr_perf_read_queue_depth,
> > + "event=0x08");
> > +PMU_EVENT_ATTR_STRING(write-queue-depth, ddr_perf_write_queue_depth,
> > + "event=0x09");
> > +PMU_EVENT_ATTR_STRING(lp-read-credit-cnt, ddr_perf_lp_read_credit_cnt,
> > + "event=0x10");
> > +PMU_EVENT_ATTR_STRING(hp-read-credit-cnt, ddr_perf_hp_read_credit_cnt,
> > + "event=0x11");
> > +PMU_EVENT_ATTR_STRING(write-credit-cnt, ddr_perf_write_credit_cnt,
> > + "event=0x12");
> > +PMU_EVENT_ATTR_STRING(read-command, ddr_perf_read_command, "event=0x20");
> > +PMU_EVENT_ATTR_STRING(write-command, ddr_perf_write_command, "event=0x21");
> > +PMU_EVENT_ATTR_STRING(read-modify-write-command,
> > + ddr_perf_read_modify_write_command, "event=0x22");
> > +PMU_EVENT_ATTR_STRING(hp-read, ddr_perf_hp_read, "event=0x23");
> > +PMU_EVENT_ATTR_STRING(hp-req-nodcredit, ddr_perf_hp_req_nocredit, "event=0x24");
> > +PMU_EVENT_ATTR_STRING(hp-xact-credit, ddr_perf_hp_xact_credit, "event=0x25");
> > +PMU_EVENT_ATTR_STRING(lp-req-nocredit, ddr_perf_lp_req_nocredit, "event=0x26");
> > +PMU_EVENT_ATTR_STRING(lp-xact-credit, ddr_perf_lp_xact_credit, "event=0x27");
> > +PMU_EVENT_ATTR_STRING(wr-xact-credit, ddr_perf_wr_xact_credit, "event=0x29");
> > +PMU_EVENT_ATTR_STRING(read-cycles, ddr_perf_read_cycles, "event=0x2a");
> > +PMU_EVENT_ATTR_STRING(write-cycles, ddr_perf_write_cycles, "event=0x2b");
> > +PMU_EVENT_ATTR_STRING(read-write-transition, ddr_perf_read_write_transition,
> > + "event=0x30");
> > +PMU_EVENT_ATTR_STRING(precharge, ddr_perf_precharge, "event=0x31");
> > +PMU_EVENT_ATTR_STRING(activate, ddr_perf_activate, "event=0x32");
> > +PMU_EVENT_ATTR_STRING(load-mode, ddr_perf_load_mode, "event=0x33");
> > +PMU_EVENT_ATTR_STRING(mwr, ddr_perf_mwr, "event=0x34");
> > +PMU_EVENT_ATTR_STRING(read, ddr_perf_read, "event=0x35");
> > +PMU_EVENT_ATTR_STRING(read-activate, ddr_perf_read_activate, "event=0x36");
> > +PMU_EVENT_ATTR_STRING(refresh, ddr_perf_refresh, "event=0x37");
> > +PMU_EVENT_ATTR_STRING(write, ddr_perf_write, "event=0x38");
> > +PMU_EVENT_ATTR_STRING(raw-hazard, ddr_perf_raw_hazard, "event=0x39");
> > +
> > +PMU_EVENT_ATTR_STRING(axid-read, ddr_perf_axid_read,
> > + "event="__stringify(EVENT_AXI_READ));
> > +PMU_EVENT_ATTR_STRING(axid-write, ddr_perf_axid_write,
> > + "event="__stringify(EVENT_AXI_WRITE));
> > +
> > +#define DDR_CAP_AXI_ID 0x1
> > +
> > +static const struct of_device_id imx_ddr_pmu_dt_ids[] = {
> > + { .compatible = "fsl,imx8-ddr-pmu",},
> > + { .compatible = "fsl,imx8m-ddr-pmu", .data = (void *)DDR_CAP_AXI_ID},
> > + { /* sentinel */ }
> > +};
> > +
> > +struct ddr_pmu {
> > + struct pmu pmu;
> > + void __iomem *base;
> > + cpumask_t cpu;
> > + struct hlist_node node;
> > + struct device *dev;
> > + struct perf_event *active_events[NUM_COUNTER];
> > + int total_events;
> > + enum cpuhp_state cpuhp_state;
> > + bool cycles_active;
> > + uintptr_t flags;
> > +};
> > +
> > +static ssize_t ddr_perf_cpumask_show(struct device *dev,
> > + struct device_attribute *attr, char *buf)
> > +{
> > + struct ddr_pmu *pmu = dev_get_drvdata(dev);
> > +
> > + return cpumap_print_to_pagebuf(true, buf, &pmu->cpu);
> > +}
> > +
> > +static struct device_attribute ddr_perf_cpumask_attr =
> > + __ATTR(cpumask, 0444, ddr_perf_cpumask_show, NULL);
> > +
> > +static struct attribute *ddr_perf_cpumask_attrs[] = {
> > + &ddr_perf_cpumask_attr.attr,
> > + NULL,
> > +};
> > +
> > +static struct attribute_group ddr_perf_cpumask_attr_group = {
> > + .attrs = ddr_perf_cpumask_attrs,
> > +};
> > +
> > +static struct attribute *ddr_perf_events_attrs[] = {
> > + &ddr_perf_cycles.attr.attr,
> > + &ddr_perf_selfresh.attr.attr,
> > + &ddr_perf_read_accesses.attr.attr,
> > + &ddr_perf_write_accesses.attr.attr,
> > + &ddr_perf_read_queue_depth.attr.attr,
> > + &ddr_perf_write_queue_depth.attr.attr,
> > + &ddr_perf_lp_read_credit_cnt.attr.attr,
> > + &ddr_perf_hp_read_credit_cnt.attr.attr,
> > + &ddr_perf_write_credit_cnt.attr.attr,
> > + &ddr_perf_read_command.attr.attr,
> > + &ddr_perf_write_command.attr.attr,
> > + &ddr_perf_read_modify_write_command.attr.attr,
> > + &ddr_perf_hp_read.attr.attr,
> > + &ddr_perf_hp_req_nocredit.attr.attr,
> > + &ddr_perf_hp_xact_credit.attr.attr,
> > + &ddr_perf_lp_req_nocredit.attr.attr,
> > + &ddr_perf_lp_xact_credit.attr.attr,
> > + &ddr_perf_wr_xact_credit.attr.attr,
> > + &ddr_perf_read_cycles.attr.attr,
> > + &ddr_perf_write_cycles.attr.attr,
> > + &ddr_perf_read_write_transition.attr.attr,
> > + &ddr_perf_precharge.attr.attr,
> > + &ddr_perf_activate.attr.attr,
> > + &ddr_perf_load_mode.attr.attr,
> > + &ddr_perf_mwr.attr.attr,
> > + &ddr_perf_read.attr.attr,
> > + &ddr_perf_read_activate.attr.attr,
> > + &ddr_perf_refresh.attr.attr,
> > + &ddr_perf_write.attr.attr,
> > + &ddr_perf_raw_hazard.attr.attr,
> > + &ddr_perf_axid_read.attr.attr,
> > + &ddr_perf_axid_write.attr.attr,
> > + NULL,
> > +};
> > +
> > +static struct attribute_group ddr_perf_events_attr_group = {
> > + .name = "events",
> > + .attrs = ddr_perf_events_attrs,
> > +};
> > +
> > +PMU_FORMAT_ATTR(event, "config:0-63");
> > +PMU_FORMAT_ATTR(axi_id, "config1:0-63");
> > +
> > +static struct attribute *ddr_perf_format_attrs[] = {
> > + &format_attr_event.attr,
> > + &format_attr_axi_id.attr,
> > + NULL,
> > +};
> > +
> > +static struct attribute_group ddr_perf_format_attr_group = {
> > + .name = "format",
> > + .attrs = ddr_perf_format_attrs,
> > +};
> > +
> > +static const struct attribute_group *attr_groups[] = {
> > + &ddr_perf_events_attr_group,
> > + &ddr_perf_format_attr_group,
> > + &ddr_perf_cpumask_attr_group,
> > + NULL,
> > +};
> > +
> > +static u32 ddr_perf_alloc_counter(struct ddr_pmu *pmu, int event)
> > +{
> > + int i;
> > +
> > + /* Always map cycle event to counter 0
> > + Cycles counter is dedicated for cycle event
> > + can't used for the other events
> > + */
> > + if (event == EVENT_CYCLES_ID)
> > + return EVENT_CYCLES_COUNTER;
> > +
> > + for (i = 1; i < NUM_COUNTER; i++)
> > + if (pmu->active_events[i] == NULL)
> > + return i;
> > +
> > + return -ENOENT;
> > +}
> > +
> > +static u32 ddr_perf_free_counter(struct ddr_pmu *pmu, int counter)
> > +{
> > + if (counter < 0 || counter >= NUM_COUNTER)
> > + return -ENOENT;
> > +
> > + pmu->active_events[counter] = NULL;
> > +
> > + return 0;
> > +}
> > +
> > +static u32 ddr_perf_read_counter(struct ddr_pmu *pmu, int counter)
> > +{
> > + return readl(pmu->base + COUNTER_READ + counter * 4);
> > +}
> > +
> > +static int ddr_perf_event_init(struct perf_event *event)
> > +{
> > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > + struct hw_perf_event *hwc = &event->hw;
> > +
> > + if (event->attr.type != event->pmu->type)
> > + return -ENOENT;
> > +
> > + if (is_sampling_event(event) || event->attach_state & PERF_ATTACH_TASK)
> > + return -EOPNOTSUPP;
> > +
> > + if (event->cpu < 0) {
> > + dev_warn(pmu->dev, "Can't provide per-task data!\n");
> > + return -EOPNOTSUPP;
> > + }
> > +
> > + if (event->attr.exclude_user ||
> > + event->attr.exclude_kernel ||
> > + event->attr.exclude_hv ||
> > + event->attr.exclude_idle ||
> > + event->attr.exclude_host ||
> > + event->attr.exclude_guest ||
> > + event->attr.sample_period)
> > + return -EINVAL;
> > +
> > + event->cpu = cpumask_first(&pmu->cpu);
> > + hwc->idx = -1;
> > +
> > + return 0;
> > +}
> > +
> > +
> > +static void ddr_perf_event_update(struct perf_event *event)
> > +{
> > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > + struct hw_perf_event *hwc = &event->hw;
> > + u64 delta, prev_raw_count, new_raw_count;
> > + int counter = hwc->idx;
> > +
> > + do {
> > + prev_raw_count = local64_read(&hwc->prev_count);
> > + new_raw_count = ddr_perf_read_counter(pmu, counter);
> > + } while (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
> > + new_raw_count) != prev_raw_count);
> > +
> > + delta = (new_raw_count - prev_raw_count) & 0xFFFFFFFF;
> > +
> > + local64_add(delta, &event->count);
> > +}
> > +
> > +static void ddr_perf_event_enable(struct ddr_pmu *pmu, int config,
> > + int counter, bool enable)
> > +{
> > + u8 reg = counter * 4 + COUNTER_CNTL;
> > + int val;
> > +
> > + if (enable) {
> > + /* must disable first, then enable again
> > + * otherwise, cycle counter will not work
> > + * if previous state is enabled.
> > + */
> > + writel(0, pmu->base + reg);
> > + val = CNTL_EN | CNTL_CLEAR;
> > + val |= (config << CNTL_CSV_SHIFT) & CNTL_CSV_MASK;
> > + } else {
> > + /* Disable counter */
> > + val = readl(pmu->base + reg) & CNTL_EN_MASK;
> > + }
> > +
> > + writel(val, pmu->base + reg);
> > +
> > + if (config == EVENT_CYCLES_ID)
> > + pmu->cycles_active = enable;
> > +}
> > +
> > +static void ddr_perf_event_start(struct perf_event *event, int flags)
> > +{
> > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > + struct hw_perf_event *hwc = &event->hw;
> > + int counter = hwc->idx;
> > +
> > + if (pmu->flags & DDR_CAP_AXI_ID) {
> > + if (event->attr.config == EVENT_AXI_READ ||
> > + event->attr.config == EVENT_AXI_WRITE) {
> > + int val = event->attr.config1;
> > +
> > + writel(val, pmu->base + COUNTER_DPCR1);
> > + }
> > + }
> > +
> > + local64_set(&hwc->prev_count, 0);
> > +
> > + ddr_perf_event_enable(pmu, event->attr.config, counter, true);
> > + /*
> > + * If the cycles counter wasn't explicitly selected,
> > + * we will enable it now.
> > + * cycles counter must be enabled otherwise other counters will
> > + * stopped.
> > + */
> > + if (counter > 0 && !pmu->cycles_active)
> > + ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
> > + EVENT_CYCLES_COUNTER, true);
> > +}
> > +
> > +static int ddr_perf_event_add(struct perf_event *event, int flags)
> > +{
> > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > + struct hw_perf_event *hwc = &event->hw;
> > + int counter;
> > + int cfg = event->attr.config;
> > +
> > + counter = ddr_perf_alloc_counter(pmu, cfg);
> > + if (counter < 0) {
> > + dev_dbg(pmu->dev, "There are not enough counters\n");
> > + return -EOPNOTSUPP;
> > + }
> > +
> > + pmu->active_events[counter] = event;
> > + pmu->total_events++;
> > + hwc->idx = counter;
> > +
> > + if (flags & PERF_EF_START)
> > + ddr_perf_event_start(event, flags);
> > +
> > + local64_set(&hwc->prev_count, ddr_perf_read_counter(pmu, counter));
> > +
> > + return 0;
> > +}
> > +
> > +static void ddr_perf_event_stop(struct perf_event *event, int flags)
> > +{
> > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > + struct hw_perf_event *hwc = &event->hw;
> > + int counter = hwc->idx;
> > +
> > + ddr_perf_event_enable(pmu, event->attr.config, counter, false);
> > + ddr_perf_event_update(event);
> > +}
> > +
> > +static void ddr_perf_event_del(struct perf_event *event, int flags)
> > +{
> > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > + struct hw_perf_event *hwc = &event->hw;
> > + int counter = hwc->idx;
> > +
> > + ddr_perf_event_stop(event, PERF_EF_UPDATE);
> > +
> > + ddr_perf_free_counter(pmu, counter);
> > + pmu->total_events--;
> > + hwc->idx = -1;
> > +
> > + /* If all events have stopped, stop the cycles counter as well */
> > + if ((pmu->total_events == 0) && pmu->cycles_active)
> > + ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
> > + EVENT_CYCLES_COUNTER, false);
> > +}
> > +
> > +static int ddr_perf_init(struct ddr_pmu *pmu, void __iomem *base,
> > + struct device *dev)
> > +{
> > + *pmu = (struct ddr_pmu) {
> > + .pmu = (struct pmu) {
> > + .task_ctx_nr = perf_invalid_context,
> > + .attr_groups = attr_groups,
> > + .event_init = ddr_perf_event_init,
> > + .add = ddr_perf_event_add,
> > + .del = ddr_perf_event_del,
> > + .start = ddr_perf_event_start,
> > + .stop = ddr_perf_event_stop,
> > + .read = ddr_perf_event_update,
> > + },
> > + .base = base,
> > + .dev = dev,
> > + };
> > +
> > + return ida_simple_get(&ddr_ida, 0, 0, GFP_KERNEL);
> > +}
> > +
> > +static irqreturn_t ddr_perf_irq_handler(int irq, void *p)
> > +{
> > + int i;
> > + u8 reg;
> > + int val;
> > + int counter;
> > + struct ddr_pmu *pmu = (struct ddr_pmu *) p;
> > + struct perf_event *event;
> > +
> > + /* Only cycles counter overflowed can issue irq. all counters will
> > + * be stopped when cycles counter overflow. but other counter don't stop
> > + * when overflow happen. Update all of the local counter values,
> > + * then reset the cycles counter, so the others can continue
> > + * counting. cycles counter is fastest counter in all events. at last
> > + * 4 times than other counters.
> > + */
> > + for (i = 0; i < NUM_COUNTER; i++) {
> > +
> > + if (!pmu->active_events[i])
> > + continue;
> > +
> > + event = pmu->active_events[i];
> > + counter = event->hw.idx;
> > + reg = counter * 4 + COUNTER_CNTL;
> > + val = readl(pmu->base + reg);
> > +
> > + ddr_perf_event_update(event);
> > +
> > + if (counter == EVENT_CYCLES_COUNTER) {
> > + ddr_perf_event_enable(pmu,
> > + EVENT_CYCLES_ID,
> > + EVENT_CYCLES_COUNTER,
> > + true);
> > + ddr_perf_event_update(event);
> > + }
> > + }
> > +
> > + return IRQ_HANDLED;
> > +}
> > +
> > +static int ddr_perf_offline_cpu(unsigned int cpu, struct hlist_node *node)
> > +{
> > + struct ddr_pmu *pmu = hlist_entry_safe(node, struct ddr_pmu, node);
> > + int target;
> > +
> > + if (!cpumask_test_and_clear_cpu(cpu, &pmu->cpu))
> > + return 0;
> > +
> > + target = cpumask_any_but(cpu_online_mask, cpu);
> > + if (target >= nr_cpu_ids)
> > + return 0;
> > +
> > + perf_pmu_migrate_context(&pmu->pmu, cpu, target);
> > + cpumask_set_cpu(target, &pmu->cpu);
> > +
> > + return 0;
> > +}
> > +
> > +static int ddr_perf_probe(struct platform_device *pdev)
> > +{
> > + struct ddr_pmu *pmu;
> > + struct device_node *np;
> > + void __iomem *base;
> > + struct resource *iomem;
> > + char *name;
> > + char *hpname;
> > + int num;
> > + int ret;
> > + u32 irq;
> > + const struct of_device_id *of_id =
> > + of_match_device(imx_ddr_pmu_dt_ids, &pdev->dev);
> > +
> > + iomem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> > + base = devm_ioremap_resource(&pdev->dev, iomem);
> > + if (IS_ERR(base))
> > + return PTR_ERR(base);
> > +
> > + np = pdev->dev.of_node;
> > +
> > + pmu = devm_kzalloc(&pdev->dev, sizeof(*pmu), GFP_KERNEL);
> > + if (!pmu)
> > + return -ENOMEM;
> > +
> > + num = ddr_perf_init(pmu, base, &pdev->dev);
> > +
> > + name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "ddr%d", num);
> > + if (!name)
> > + return -ENOMEM;
> > +
> > + hpname = devm_kasprintf(&pdev->dev, GFP_KERNEL,
> > + "perf/imx/ddr%d:online", num);
> > + if (!hpname)
> > + return -ENOMEM;
> > +
> > + pmu->flags = (uintptr_t) of_id->data;
> > +
> > + cpumask_set_cpu(raw_smp_processor_id(), &pmu->cpu);
> > + ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN, hpname, NULL,
> > + ddr_perf_offline_cpu);
> > +
> > + if (ret < 0) {
> > + dev_err(&pdev->dev, "cpuhp_setup_state_multi failed\n");
> > + goto ddr_perf_err;
> > + }
> > +
> > + pmu->cpuhp_state = ret;
> > +
> > + /* Register the pmu instance for cpu hotplug */
> > + cpuhp_state_add_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> > +
> > + ret = perf_pmu_register(&(pmu->pmu), name, -1);
> > + if (ret)
> > + goto ddr_perf_err;
> > +
> > + /* Request irq */
> > + irq = of_irq_get(np, 0);
> > + if (irq < 0) {
> > + dev_err(&pdev->dev, "Failed to get irq: %d", irq);
> > + ret = irq;
> > + goto ddr_perf_irq_err;
> > + }
> > +
> > + ret = devm_request_irq(&pdev->dev, irq,
> > + ddr_perf_irq_handler,
> > + IRQF_TRIGGER_RISING | IRQF_ONESHOT,
> > + DDR_PERF_DEV_NAME,
> > + pmu);
> > + if (ret < 0) {
> > + dev_err(&pdev->dev, "Request irq failed: %d", ret);
> > + goto ddr_perf_irq_err;
> > + }
> > +
> > + return 0;
> > +
> > +ddr_perf_irq_err:
> > + perf_pmu_unregister(&(pmu->pmu));
> > +
> > +ddr_perf_err:
> > + if (pmu->cpuhp_state)
> > + cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> > +
> > + dev_warn(&pdev->dev, "i.MX8 DDR Perf PMU failed (%d), disabled\n", ret);
> > + return ret;
> > +}
> > +
> > +static int ddr_perf_remove(struct platform_device *pdev)
> > +{
> > + struct ddr_pmu *pmu = platform_get_drvdata(pdev);
> > +
> > + cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> > + perf_pmu_unregister(&pmu->pmu);
> > +
> > + return 0;
> > +}
> > +
> > +static struct platform_driver imx_ddr_pmu_driver = {
> > + .driver = {
> > + .name = "imx-ddr-pmu",
> > + .of_match_table = imx_ddr_pmu_dt_ids,
> > + },
> > + .probe = ddr_perf_probe,
> > + .remove = ddr_perf_remove,
> > +};
> > +
> > +static int __init imx_ddr_pmu_init(void)
> > +{
> > + return platform_driver_register(&imx_ddr_pmu_driver);
> > +}
> > +
> > +module_init(imx_ddr_pmu_init);
> > +
> > --
> > 2.5.2
> >
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
2019-02-28 21:39 ` Zhi Li
@ 2019-03-12 21:49 ` Zhi Li
-1 siblings, 0 replies; 43+ messages in thread
From: Zhi Li @ 2019-03-12 21:49 UTC (permalink / raw)
To: Frank Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
will.deacon, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Thu, Feb 28, 2019 at 3:39 PM Zhi Li <lznuaa@gmail.com> wrote:
>
> On Fri, Feb 22, 2019 at 1:45 PM Zhi Li <lznuaa@gmail.com> wrote:
> >
> > On Fri, Feb 15, 2019 at 12:03 PM Frank Li <frank.li@nxp.com> wrote:
> > >
> > > Add ddr performance monitor support for iMX8QXP
> > >
> > > There are 4 counters for ddr perfomance events.
> > > counter 0 is dedicated for cycles.
> > > you choose any up to 3 no cycles events.
> >
> > Any additional comments for these patches?
>
> Mark:
>
> There are no comments for two weeks.
> Did you plan pick up these patches?
ping
>
> best regards
> Frank Li
>
>
> >
> > best regards
> > Frank Li
> >
> > >
> > > for example:
> > >
> > > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> > >
> > > Support below events.
> > >
> > > ddr0/activate/ [Kernel PMU event]
> > > ddr0/axid-read/ [Kernel PMU event]
> > > ddr0/axid-write/ [Kernel PMU event]
> > > ddr0/cycles/ [Kernel PMU event]
> > > ddr0/hp-read-credit-cnt/ [Kernel PMU event]
> > > ddr0/hp-read/ [Kernel PMU event]
> > > ddr0/hp-req-nodcredit/ [Kernel PMU event]
> > > ddr0/hp-xact-credit/ [Kernel PMU event]
> > > ddr0/load-mode/ [Kernel PMU event]
> > > ddr0/lp-read-credit-cnt/ [Kernel PMU event]
> > > ddr0/lp-req-nocredit/ [Kernel PMU event]
> > > ddr0/lp-xact-credit/ [Kernel PMU event]
> > > ddr0/mwr/ [Kernel PMU event]
> > > ddr0/precharge/ [Kernel PMU event]
> > > ddr0/raw-hazard/ [Kernel PMU event]
> > > ddr0/read-access/ [Kernel PMU event]
> > > ddr0/read-activate/ [Kernel PMU event]
> > > ddr0/read-command/ [Kernel PMU event]
> > > ddr0/read-cycles/ [Kernel PMU event]
> > > ddr0/read-modify-write-command/ [Kernel PMU event]
> > > ddr0/read-queue-depth/ [Kernel PMU event]
> > > ddr0/read-write-transition/ [Kernel PMU event]
> > > ddr0/read/ [Kernel PMU event]
> > > ddr0/refresh/ [Kernel PMU event]
> > > ddr0/selfresh/ [Kernel PMU event]
> > > ddr0/wr-xact-credit/ [Kernel PMU event]
> > > ddr0/write-access/ [Kernel PMU event]
> > > ddr0/write-command/ [Kernel PMU event]
> > > ddr0/write-credit-cnt/ [Kernel PMU event]
> > > ddr0/write-cycles/ [Kernel PMU event]
> > > ddr0/write-queue-depth/ [Kernel PMU event]
> > > ddr0/write/
> > >
> > > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > > ---
> > > Change from v3 to v4
> > > * Change FSL_IMX8_DDR_PERF to FSL_IMX8_DDR_PMU
> > > * sort include
> > > * remove struct fsl_ddr_devtype_data
> > > * Added comment need disable control first
> > > * Added comment about must enable cycle counter
> > > * Added macro for EVENT_AXI_READ, remove hardcode 0x41 and 0x42
> > > * Added comment about cycle counter is fastest one
> > >
> > > Change from v2 to v3
> > > * remove kfree
> > >
> > > Change from V1 to V2
> > > * update Kconfig by use i.MX8 instead of i.MX8 QXP
> > > * remove gpl statememnt since SPDX tag
> > > * use dev_kzalloc
> > > * use dev_err
> > > * commit message show axi_read 0x41\axi_write 0x42
> > > * commit message show cycles must be enabled
> > > * Irq only issue at cycles overflow
> > > * use NUM_COUNTER
> > > * use devm_request_irq
> > > * add hotplug callback to handle context migration
> > >
> > > drivers/perf/Kconfig | 7 +
> > > drivers/perf/Makefile | 1 +
> > > drivers/perf/fsl_imx8_ddr_perf.c | 573 +++++++++++++++++++++++++++++++++++++++
> > > 3 files changed, 581 insertions(+)
> > > create mode 100644 drivers/perf/fsl_imx8_ddr_perf.c
> > >
> > > diff --git a/drivers/perf/Kconfig b/drivers/perf/Kconfig
> > > index af9bc17..e567437 100644
> > > --- a/drivers/perf/Kconfig
> > > +++ b/drivers/perf/Kconfig
> > > @@ -61,6 +61,13 @@ config ARM_DSU_PMU
> > > system, control logic. The PMU allows counting various events related
> > > to DSU.
> > >
> > > +config FSL_IMX8_DDR_PMU
> > > + tristate "Freescale i.MX8 DDR perf monitor"
> > > + depends on ARCH_MXC
> > > + help
> > > + Provides support for ddr perfomance monitor in i.MX8. Provide memory
> > > + througput information.
> > > +
> > > config HISI_PMU
> > > bool "HiSilicon SoC PMU"
> > > depends on ARM64 && ACPI
> > > diff --git a/drivers/perf/Makefile b/drivers/perf/Makefile
> > > index 909f27f..0bc066b 100644
> > > --- a/drivers/perf/Makefile
> > > +++ b/drivers/perf/Makefile
> > > @@ -4,6 +4,7 @@ obj-$(CONFIG_ARM_CCN) += arm-ccn.o
> > > obj-$(CONFIG_ARM_DSU_PMU) += arm_dsu_pmu.o
> > > obj-$(CONFIG_ARM_PMU) += arm_pmu.o arm_pmu_platform.o
> > > obj-$(CONFIG_ARM_PMU_ACPI) += arm_pmu_acpi.o
> > > +obj-$(CONFIG_FSL_IMX8_DDR_PMU) += fsl_imx8_ddr_perf.o
> > > obj-$(CONFIG_HISI_PMU) += hisilicon/
> > > obj-$(CONFIG_QCOM_L2_PMU) += qcom_l2_pmu.o
> > > obj-$(CONFIG_QCOM_L3_PMU) += qcom_l3_pmu.o
> > > diff --git a/drivers/perf/fsl_imx8_ddr_perf.c b/drivers/perf/fsl_imx8_ddr_perf.c
> > > new file mode 100644
> > > index 0000000..3d02daf
> > > --- /dev/null
> > > +++ b/drivers/perf/fsl_imx8_ddr_perf.c
> > > @@ -0,0 +1,573 @@
> > > +// SPDX-License-Identifier: GPL-2.0
> > > +/*
> > > + * Copyright 2017 NXP
> > > + * Copyright 2016 Freescale Semiconductor, Inc.
> > > + */
> > > +
> > > +#include <linux/init.h>
> > > +#include <linux/interrupt.h>
> > > +#include <linux/io.h>
> > > +#include <linux/module.h>
> > > +#include <linux/of.h>
> > > +#include <linux/of_address.h>
> > > +#include <linux/of_device.h>
> > > +#include <linux/of_irq.h>
> > > +#include <linux/perf_event.h>
> > > +#include <linux/slab.h>
> > > +
> > > +#define COUNTER_CNTL 0x0
> > > +#define COUNTER_READ 0x20
> > > +
> > > +#define COUNTER_DPCR1 0x30
> > > +
> > > +#define CNTL_OVER 0x1
> > > +#define CNTL_CLEAR 0x2
> > > +#define CNTL_EN 0x4
> > > +#define CNTL_EN_MASK 0xFFFFFFFB
> > > +#define CNTL_CLEAR_MASK 0xFFFFFFFD
> > > +#define CNTL_OVER_MASK 0xFFFFFFFE
> > > +
> > > +#define CNTL_CSV_SHIFT 24
> > > +#define CNTL_CSV_MASK (0xFF << CNTL_CSV_SHIFT)
> > > +
> > > +#define EVENT_CYCLES_ID 0
> > > +#define EVENT_CYCLES_COUNTER 0
> > > +#define NUM_COUNTER 4
> > > +#define MAX_EVENT 3
> > > +#define EVENT_AXI_READ 0x41
> > > +#define EVENT_AXI_WRITE 0x42
> > > +
> > > +#define to_ddr_pmu(p) container_of(p, struct ddr_pmu, pmu)
> > > +
> > > +#define DDR_PERF_DEV_NAME "ddr_perf"
> > > +
> > > +static DEFINE_IDA(ddr_ida);
> > > +
> > > +PMU_EVENT_ATTR_STRING(cycles, ddr_perf_cycles, "event=0x00");
> > > +PMU_EVENT_ATTR_STRING(selfresh, ddr_perf_selfresh, "event=0x01");
> > > +PMU_EVENT_ATTR_STRING(read-access, ddr_perf_read_accesses, "event=0x04");
> > > +PMU_EVENT_ATTR_STRING(write-access, ddr_perf_write_accesses, "event=0x05");
> > > +PMU_EVENT_ATTR_STRING(read-queue-depth, ddr_perf_read_queue_depth,
> > > + "event=0x08");
> > > +PMU_EVENT_ATTR_STRING(write-queue-depth, ddr_perf_write_queue_depth,
> > > + "event=0x09");
> > > +PMU_EVENT_ATTR_STRING(lp-read-credit-cnt, ddr_perf_lp_read_credit_cnt,
> > > + "event=0x10");
> > > +PMU_EVENT_ATTR_STRING(hp-read-credit-cnt, ddr_perf_hp_read_credit_cnt,
> > > + "event=0x11");
> > > +PMU_EVENT_ATTR_STRING(write-credit-cnt, ddr_perf_write_credit_cnt,
> > > + "event=0x12");
> > > +PMU_EVENT_ATTR_STRING(read-command, ddr_perf_read_command, "event=0x20");
> > > +PMU_EVENT_ATTR_STRING(write-command, ddr_perf_write_command, "event=0x21");
> > > +PMU_EVENT_ATTR_STRING(read-modify-write-command,
> > > + ddr_perf_read_modify_write_command, "event=0x22");
> > > +PMU_EVENT_ATTR_STRING(hp-read, ddr_perf_hp_read, "event=0x23");
> > > +PMU_EVENT_ATTR_STRING(hp-req-nodcredit, ddr_perf_hp_req_nocredit, "event=0x24");
> > > +PMU_EVENT_ATTR_STRING(hp-xact-credit, ddr_perf_hp_xact_credit, "event=0x25");
> > > +PMU_EVENT_ATTR_STRING(lp-req-nocredit, ddr_perf_lp_req_nocredit, "event=0x26");
> > > +PMU_EVENT_ATTR_STRING(lp-xact-credit, ddr_perf_lp_xact_credit, "event=0x27");
> > > +PMU_EVENT_ATTR_STRING(wr-xact-credit, ddr_perf_wr_xact_credit, "event=0x29");
> > > +PMU_EVENT_ATTR_STRING(read-cycles, ddr_perf_read_cycles, "event=0x2a");
> > > +PMU_EVENT_ATTR_STRING(write-cycles, ddr_perf_write_cycles, "event=0x2b");
> > > +PMU_EVENT_ATTR_STRING(read-write-transition, ddr_perf_read_write_transition,
> > > + "event=0x30");
> > > +PMU_EVENT_ATTR_STRING(precharge, ddr_perf_precharge, "event=0x31");
> > > +PMU_EVENT_ATTR_STRING(activate, ddr_perf_activate, "event=0x32");
> > > +PMU_EVENT_ATTR_STRING(load-mode, ddr_perf_load_mode, "event=0x33");
> > > +PMU_EVENT_ATTR_STRING(mwr, ddr_perf_mwr, "event=0x34");
> > > +PMU_EVENT_ATTR_STRING(read, ddr_perf_read, "event=0x35");
> > > +PMU_EVENT_ATTR_STRING(read-activate, ddr_perf_read_activate, "event=0x36");
> > > +PMU_EVENT_ATTR_STRING(refresh, ddr_perf_refresh, "event=0x37");
> > > +PMU_EVENT_ATTR_STRING(write, ddr_perf_write, "event=0x38");
> > > +PMU_EVENT_ATTR_STRING(raw-hazard, ddr_perf_raw_hazard, "event=0x39");
> > > +
> > > +PMU_EVENT_ATTR_STRING(axid-read, ddr_perf_axid_read,
> > > + "event="__stringify(EVENT_AXI_READ));
> > > +PMU_EVENT_ATTR_STRING(axid-write, ddr_perf_axid_write,
> > > + "event="__stringify(EVENT_AXI_WRITE));
> > > +
> > > +#define DDR_CAP_AXI_ID 0x1
> > > +
> > > +static const struct of_device_id imx_ddr_pmu_dt_ids[] = {
> > > + { .compatible = "fsl,imx8-ddr-pmu",},
> > > + { .compatible = "fsl,imx8m-ddr-pmu", .data = (void *)DDR_CAP_AXI_ID},
> > > + { /* sentinel */ }
> > > +};
> > > +
> > > +struct ddr_pmu {
> > > + struct pmu pmu;
> > > + void __iomem *base;
> > > + cpumask_t cpu;
> > > + struct hlist_node node;
> > > + struct device *dev;
> > > + struct perf_event *active_events[NUM_COUNTER];
> > > + int total_events;
> > > + enum cpuhp_state cpuhp_state;
> > > + bool cycles_active;
> > > + uintptr_t flags;
> > > +};
> > > +
> > > +static ssize_t ddr_perf_cpumask_show(struct device *dev,
> > > + struct device_attribute *attr, char *buf)
> > > +{
> > > + struct ddr_pmu *pmu = dev_get_drvdata(dev);
> > > +
> > > + return cpumap_print_to_pagebuf(true, buf, &pmu->cpu);
> > > +}
> > > +
> > > +static struct device_attribute ddr_perf_cpumask_attr =
> > > + __ATTR(cpumask, 0444, ddr_perf_cpumask_show, NULL);
> > > +
> > > +static struct attribute *ddr_perf_cpumask_attrs[] = {
> > > + &ddr_perf_cpumask_attr.attr,
> > > + NULL,
> > > +};
> > > +
> > > +static struct attribute_group ddr_perf_cpumask_attr_group = {
> > > + .attrs = ddr_perf_cpumask_attrs,
> > > +};
> > > +
> > > +static struct attribute *ddr_perf_events_attrs[] = {
> > > + &ddr_perf_cycles.attr.attr,
> > > + &ddr_perf_selfresh.attr.attr,
> > > + &ddr_perf_read_accesses.attr.attr,
> > > + &ddr_perf_write_accesses.attr.attr,
> > > + &ddr_perf_read_queue_depth.attr.attr,
> > > + &ddr_perf_write_queue_depth.attr.attr,
> > > + &ddr_perf_lp_read_credit_cnt.attr.attr,
> > > + &ddr_perf_hp_read_credit_cnt.attr.attr,
> > > + &ddr_perf_write_credit_cnt.attr.attr,
> > > + &ddr_perf_read_command.attr.attr,
> > > + &ddr_perf_write_command.attr.attr,
> > > + &ddr_perf_read_modify_write_command.attr.attr,
> > > + &ddr_perf_hp_read.attr.attr,
> > > + &ddr_perf_hp_req_nocredit.attr.attr,
> > > + &ddr_perf_hp_xact_credit.attr.attr,
> > > + &ddr_perf_lp_req_nocredit.attr.attr,
> > > + &ddr_perf_lp_xact_credit.attr.attr,
> > > + &ddr_perf_wr_xact_credit.attr.attr,
> > > + &ddr_perf_read_cycles.attr.attr,
> > > + &ddr_perf_write_cycles.attr.attr,
> > > + &ddr_perf_read_write_transition.attr.attr,
> > > + &ddr_perf_precharge.attr.attr,
> > > + &ddr_perf_activate.attr.attr,
> > > + &ddr_perf_load_mode.attr.attr,
> > > + &ddr_perf_mwr.attr.attr,
> > > + &ddr_perf_read.attr.attr,
> > > + &ddr_perf_read_activate.attr.attr,
> > > + &ddr_perf_refresh.attr.attr,
> > > + &ddr_perf_write.attr.attr,
> > > + &ddr_perf_raw_hazard.attr.attr,
> > > + &ddr_perf_axid_read.attr.attr,
> > > + &ddr_perf_axid_write.attr.attr,
> > > + NULL,
> > > +};
> > > +
> > > +static struct attribute_group ddr_perf_events_attr_group = {
> > > + .name = "events",
> > > + .attrs = ddr_perf_events_attrs,
> > > +};
> > > +
> > > +PMU_FORMAT_ATTR(event, "config:0-63");
> > > +PMU_FORMAT_ATTR(axi_id, "config1:0-63");
> > > +
> > > +static struct attribute *ddr_perf_format_attrs[] = {
> > > + &format_attr_event.attr,
> > > + &format_attr_axi_id.attr,
> > > + NULL,
> > > +};
> > > +
> > > +static struct attribute_group ddr_perf_format_attr_group = {
> > > + .name = "format",
> > > + .attrs = ddr_perf_format_attrs,
> > > +};
> > > +
> > > +static const struct attribute_group *attr_groups[] = {
> > > + &ddr_perf_events_attr_group,
> > > + &ddr_perf_format_attr_group,
> > > + &ddr_perf_cpumask_attr_group,
> > > + NULL,
> > > +};
> > > +
> > > +static u32 ddr_perf_alloc_counter(struct ddr_pmu *pmu, int event)
> > > +{
> > > + int i;
> > > +
> > > + /* Always map cycle event to counter 0
> > > + Cycles counter is dedicated for cycle event
> > > + can't used for the other events
> > > + */
> > > + if (event == EVENT_CYCLES_ID)
> > > + return EVENT_CYCLES_COUNTER;
> > > +
> > > + for (i = 1; i < NUM_COUNTER; i++)
> > > + if (pmu->active_events[i] == NULL)
> > > + return i;
> > > +
> > > + return -ENOENT;
> > > +}
> > > +
> > > +static u32 ddr_perf_free_counter(struct ddr_pmu *pmu, int counter)
> > > +{
> > > + if (counter < 0 || counter >= NUM_COUNTER)
> > > + return -ENOENT;
> > > +
> > > + pmu->active_events[counter] = NULL;
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +static u32 ddr_perf_read_counter(struct ddr_pmu *pmu, int counter)
> > > +{
> > > + return readl(pmu->base + COUNTER_READ + counter * 4);
> > > +}
> > > +
> > > +static int ddr_perf_event_init(struct perf_event *event)
> > > +{
> > > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > > + struct hw_perf_event *hwc = &event->hw;
> > > +
> > > + if (event->attr.type != event->pmu->type)
> > > + return -ENOENT;
> > > +
> > > + if (is_sampling_event(event) || event->attach_state & PERF_ATTACH_TASK)
> > > + return -EOPNOTSUPP;
> > > +
> > > + if (event->cpu < 0) {
> > > + dev_warn(pmu->dev, "Can't provide per-task data!\n");
> > > + return -EOPNOTSUPP;
> > > + }
> > > +
> > > + if (event->attr.exclude_user ||
> > > + event->attr.exclude_kernel ||
> > > + event->attr.exclude_hv ||
> > > + event->attr.exclude_idle ||
> > > + event->attr.exclude_host ||
> > > + event->attr.exclude_guest ||
> > > + event->attr.sample_period)
> > > + return -EINVAL;
> > > +
> > > + event->cpu = cpumask_first(&pmu->cpu);
> > > + hwc->idx = -1;
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +
> > > +static void ddr_perf_event_update(struct perf_event *event)
> > > +{
> > > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > > + struct hw_perf_event *hwc = &event->hw;
> > > + u64 delta, prev_raw_count, new_raw_count;
> > > + int counter = hwc->idx;
> > > +
> > > + do {
> > > + prev_raw_count = local64_read(&hwc->prev_count);
> > > + new_raw_count = ddr_perf_read_counter(pmu, counter);
> > > + } while (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
> > > + new_raw_count) != prev_raw_count);
> > > +
> > > + delta = (new_raw_count - prev_raw_count) & 0xFFFFFFFF;
> > > +
> > > + local64_add(delta, &event->count);
> > > +}
> > > +
> > > +static void ddr_perf_event_enable(struct ddr_pmu *pmu, int config,
> > > + int counter, bool enable)
> > > +{
> > > + u8 reg = counter * 4 + COUNTER_CNTL;
> > > + int val;
> > > +
> > > + if (enable) {
> > > + /* must disable first, then enable again
> > > + * otherwise, cycle counter will not work
> > > + * if previous state is enabled.
> > > + */
> > > + writel(0, pmu->base + reg);
> > > + val = CNTL_EN | CNTL_CLEAR;
> > > + val |= (config << CNTL_CSV_SHIFT) & CNTL_CSV_MASK;
> > > + } else {
> > > + /* Disable counter */
> > > + val = readl(pmu->base + reg) & CNTL_EN_MASK;
> > > + }
> > > +
> > > + writel(val, pmu->base + reg);
> > > +
> > > + if (config == EVENT_CYCLES_ID)
> > > + pmu->cycles_active = enable;
> > > +}
> > > +
> > > +static void ddr_perf_event_start(struct perf_event *event, int flags)
> > > +{
> > > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > > + struct hw_perf_event *hwc = &event->hw;
> > > + int counter = hwc->idx;
> > > +
> > > + if (pmu->flags & DDR_CAP_AXI_ID) {
> > > + if (event->attr.config == EVENT_AXI_READ ||
> > > + event->attr.config == EVENT_AXI_WRITE) {
> > > + int val = event->attr.config1;
> > > +
> > > + writel(val, pmu->base + COUNTER_DPCR1);
> > > + }
> > > + }
> > > +
> > > + local64_set(&hwc->prev_count, 0);
> > > +
> > > + ddr_perf_event_enable(pmu, event->attr.config, counter, true);
> > > + /*
> > > + * If the cycles counter wasn't explicitly selected,
> > > + * we will enable it now.
> > > + * cycles counter must be enabled otherwise other counters will
> > > + * stopped.
> > > + */
> > > + if (counter > 0 && !pmu->cycles_active)
> > > + ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
> > > + EVENT_CYCLES_COUNTER, true);
> > > +}
> > > +
> > > +static int ddr_perf_event_add(struct perf_event *event, int flags)
> > > +{
> > > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > > + struct hw_perf_event *hwc = &event->hw;
> > > + int counter;
> > > + int cfg = event->attr.config;
> > > +
> > > + counter = ddr_perf_alloc_counter(pmu, cfg);
> > > + if (counter < 0) {
> > > + dev_dbg(pmu->dev, "There are not enough counters\n");
> > > + return -EOPNOTSUPP;
> > > + }
> > > +
> > > + pmu->active_events[counter] = event;
> > > + pmu->total_events++;
> > > + hwc->idx = counter;
> > > +
> > > + if (flags & PERF_EF_START)
> > > + ddr_perf_event_start(event, flags);
> > > +
> > > + local64_set(&hwc->prev_count, ddr_perf_read_counter(pmu, counter));
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +static void ddr_perf_event_stop(struct perf_event *event, int flags)
> > > +{
> > > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > > + struct hw_perf_event *hwc = &event->hw;
> > > + int counter = hwc->idx;
> > > +
> > > + ddr_perf_event_enable(pmu, event->attr.config, counter, false);
> > > + ddr_perf_event_update(event);
> > > +}
> > > +
> > > +static void ddr_perf_event_del(struct perf_event *event, int flags)
> > > +{
> > > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > > + struct hw_perf_event *hwc = &event->hw;
> > > + int counter = hwc->idx;
> > > +
> > > + ddr_perf_event_stop(event, PERF_EF_UPDATE);
> > > +
> > > + ddr_perf_free_counter(pmu, counter);
> > > + pmu->total_events--;
> > > + hwc->idx = -1;
> > > +
> > > + /* If all events have stopped, stop the cycles counter as well */
> > > + if ((pmu->total_events == 0) && pmu->cycles_active)
> > > + ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
> > > + EVENT_CYCLES_COUNTER, false);
> > > +}
> > > +
> > > +static int ddr_perf_init(struct ddr_pmu *pmu, void __iomem *base,
> > > + struct device *dev)
> > > +{
> > > + *pmu = (struct ddr_pmu) {
> > > + .pmu = (struct pmu) {
> > > + .task_ctx_nr = perf_invalid_context,
> > > + .attr_groups = attr_groups,
> > > + .event_init = ddr_perf_event_init,
> > > + .add = ddr_perf_event_add,
> > > + .del = ddr_perf_event_del,
> > > + .start = ddr_perf_event_start,
> > > + .stop = ddr_perf_event_stop,
> > > + .read = ddr_perf_event_update,
> > > + },
> > > + .base = base,
> > > + .dev = dev,
> > > + };
> > > +
> > > + return ida_simple_get(&ddr_ida, 0, 0, GFP_KERNEL);
> > > +}
> > > +
> > > +static irqreturn_t ddr_perf_irq_handler(int irq, void *p)
> > > +{
> > > + int i;
> > > + u8 reg;
> > > + int val;
> > > + int counter;
> > > + struct ddr_pmu *pmu = (struct ddr_pmu *) p;
> > > + struct perf_event *event;
> > > +
> > > + /* Only cycles counter overflowed can issue irq. all counters will
> > > + * be stopped when cycles counter overflow. but other counter don't stop
> > > + * when overflow happen. Update all of the local counter values,
> > > + * then reset the cycles counter, so the others can continue
> > > + * counting. cycles counter is fastest counter in all events. at last
> > > + * 4 times than other counters.
> > > + */
> > > + for (i = 0; i < NUM_COUNTER; i++) {
> > > +
> > > + if (!pmu->active_events[i])
> > > + continue;
> > > +
> > > + event = pmu->active_events[i];
> > > + counter = event->hw.idx;
> > > + reg = counter * 4 + COUNTER_CNTL;
> > > + val = readl(pmu->base + reg);
> > > +
> > > + ddr_perf_event_update(event);
> > > +
> > > + if (counter == EVENT_CYCLES_COUNTER) {
> > > + ddr_perf_event_enable(pmu,
> > > + EVENT_CYCLES_ID,
> > > + EVENT_CYCLES_COUNTER,
> > > + true);
> > > + ddr_perf_event_update(event);
> > > + }
> > > + }
> > > +
> > > + return IRQ_HANDLED;
> > > +}
> > > +
> > > +static int ddr_perf_offline_cpu(unsigned int cpu, struct hlist_node *node)
> > > +{
> > > + struct ddr_pmu *pmu = hlist_entry_safe(node, struct ddr_pmu, node);
> > > + int target;
> > > +
> > > + if (!cpumask_test_and_clear_cpu(cpu, &pmu->cpu))
> > > + return 0;
> > > +
> > > + target = cpumask_any_but(cpu_online_mask, cpu);
> > > + if (target >= nr_cpu_ids)
> > > + return 0;
> > > +
> > > + perf_pmu_migrate_context(&pmu->pmu, cpu, target);
> > > + cpumask_set_cpu(target, &pmu->cpu);
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +static int ddr_perf_probe(struct platform_device *pdev)
> > > +{
> > > + struct ddr_pmu *pmu;
> > > + struct device_node *np;
> > > + void __iomem *base;
> > > + struct resource *iomem;
> > > + char *name;
> > > + char *hpname;
> > > + int num;
> > > + int ret;
> > > + u32 irq;
> > > + const struct of_device_id *of_id =
> > > + of_match_device(imx_ddr_pmu_dt_ids, &pdev->dev);
> > > +
> > > + iomem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> > > + base = devm_ioremap_resource(&pdev->dev, iomem);
> > > + if (IS_ERR(base))
> > > + return PTR_ERR(base);
> > > +
> > > + np = pdev->dev.of_node;
> > > +
> > > + pmu = devm_kzalloc(&pdev->dev, sizeof(*pmu), GFP_KERNEL);
> > > + if (!pmu)
> > > + return -ENOMEM;
> > > +
> > > + num = ddr_perf_init(pmu, base, &pdev->dev);
> > > +
> > > + name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "ddr%d", num);
> > > + if (!name)
> > > + return -ENOMEM;
> > > +
> > > + hpname = devm_kasprintf(&pdev->dev, GFP_KERNEL,
> > > + "perf/imx/ddr%d:online", num);
> > > + if (!hpname)
> > > + return -ENOMEM;
> > > +
> > > + pmu->flags = (uintptr_t) of_id->data;
> > > +
> > > + cpumask_set_cpu(raw_smp_processor_id(), &pmu->cpu);
> > > + ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN, hpname, NULL,
> > > + ddr_perf_offline_cpu);
> > > +
> > > + if (ret < 0) {
> > > + dev_err(&pdev->dev, "cpuhp_setup_state_multi failed\n");
> > > + goto ddr_perf_err;
> > > + }
> > > +
> > > + pmu->cpuhp_state = ret;
> > > +
> > > + /* Register the pmu instance for cpu hotplug */
> > > + cpuhp_state_add_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> > > +
> > > + ret = perf_pmu_register(&(pmu->pmu), name, -1);
> > > + if (ret)
> > > + goto ddr_perf_err;
> > > +
> > > + /* Request irq */
> > > + irq = of_irq_get(np, 0);
> > > + if (irq < 0) {
> > > + dev_err(&pdev->dev, "Failed to get irq: %d", irq);
> > > + ret = irq;
> > > + goto ddr_perf_irq_err;
> > > + }
> > > +
> > > + ret = devm_request_irq(&pdev->dev, irq,
> > > + ddr_perf_irq_handler,
> > > + IRQF_TRIGGER_RISING | IRQF_ONESHOT,
> > > + DDR_PERF_DEV_NAME,
> > > + pmu);
> > > + if (ret < 0) {
> > > + dev_err(&pdev->dev, "Request irq failed: %d", ret);
> > > + goto ddr_perf_irq_err;
> > > + }
> > > +
> > > + return 0;
> > > +
> > > +ddr_perf_irq_err:
> > > + perf_pmu_unregister(&(pmu->pmu));
> > > +
> > > +ddr_perf_err:
> > > + if (pmu->cpuhp_state)
> > > + cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> > > +
> > > + dev_warn(&pdev->dev, "i.MX8 DDR Perf PMU failed (%d), disabled\n", ret);
> > > + return ret;
> > > +}
> > > +
> > > +static int ddr_perf_remove(struct platform_device *pdev)
> > > +{
> > > + struct ddr_pmu *pmu = platform_get_drvdata(pdev);
> > > +
> > > + cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> > > + perf_pmu_unregister(&pmu->pmu);
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +static struct platform_driver imx_ddr_pmu_driver = {
> > > + .driver = {
> > > + .name = "imx-ddr-pmu",
> > > + .of_match_table = imx_ddr_pmu_dt_ids,
> > > + },
> > > + .probe = ddr_perf_probe,
> > > + .remove = ddr_perf_remove,
> > > +};
> > > +
> > > +static int __init imx_ddr_pmu_init(void)
> > > +{
> > > + return platform_driver_register(&imx_ddr_pmu_driver);
> > > +}
> > > +
> > > +module_init(imx_ddr_pmu_init);
> > > +
> > > --
> > > 2.5.2
> > >
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
@ 2019-03-12 21:49 ` Zhi Li
0 siblings, 0 replies; 43+ messages in thread
From: Zhi Li @ 2019-03-12 21:49 UTC (permalink / raw)
To: Frank Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
will.deacon, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Thu, Feb 28, 2019 at 3:39 PM Zhi Li <lznuaa@gmail.com> wrote:
>
> On Fri, Feb 22, 2019 at 1:45 PM Zhi Li <lznuaa@gmail.com> wrote:
> >
> > On Fri, Feb 15, 2019 at 12:03 PM Frank Li <frank.li@nxp.com> wrote:
> > >
> > > Add ddr performance monitor support for iMX8QXP
> > >
> > > There are 4 counters for ddr perfomance events.
> > > counter 0 is dedicated for cycles.
> > > you choose any up to 3 no cycles events.
> >
> > Any additional comments for these patches?
>
> Mark:
>
> There are no comments for two weeks.
> Did you plan pick up these patches?
ping
>
> best regards
> Frank Li
>
>
> >
> > best regards
> > Frank Li
> >
> > >
> > > for example:
> > >
> > > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> > >
> > > Support below events.
> > >
> > > ddr0/activate/ [Kernel PMU event]
> > > ddr0/axid-read/ [Kernel PMU event]
> > > ddr0/axid-write/ [Kernel PMU event]
> > > ddr0/cycles/ [Kernel PMU event]
> > > ddr0/hp-read-credit-cnt/ [Kernel PMU event]
> > > ddr0/hp-read/ [Kernel PMU event]
> > > ddr0/hp-req-nodcredit/ [Kernel PMU event]
> > > ddr0/hp-xact-credit/ [Kernel PMU event]
> > > ddr0/load-mode/ [Kernel PMU event]
> > > ddr0/lp-read-credit-cnt/ [Kernel PMU event]
> > > ddr0/lp-req-nocredit/ [Kernel PMU event]
> > > ddr0/lp-xact-credit/ [Kernel PMU event]
> > > ddr0/mwr/ [Kernel PMU event]
> > > ddr0/precharge/ [Kernel PMU event]
> > > ddr0/raw-hazard/ [Kernel PMU event]
> > > ddr0/read-access/ [Kernel PMU event]
> > > ddr0/read-activate/ [Kernel PMU event]
> > > ddr0/read-command/ [Kernel PMU event]
> > > ddr0/read-cycles/ [Kernel PMU event]
> > > ddr0/read-modify-write-command/ [Kernel PMU event]
> > > ddr0/read-queue-depth/ [Kernel PMU event]
> > > ddr0/read-write-transition/ [Kernel PMU event]
> > > ddr0/read/ [Kernel PMU event]
> > > ddr0/refresh/ [Kernel PMU event]
> > > ddr0/selfresh/ [Kernel PMU event]
> > > ddr0/wr-xact-credit/ [Kernel PMU event]
> > > ddr0/write-access/ [Kernel PMU event]
> > > ddr0/write-command/ [Kernel PMU event]
> > > ddr0/write-credit-cnt/ [Kernel PMU event]
> > > ddr0/write-cycles/ [Kernel PMU event]
> > > ddr0/write-queue-depth/ [Kernel PMU event]
> > > ddr0/write/
> > >
> > > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > > ---
> > > Change from v3 to v4
> > > * Change FSL_IMX8_DDR_PERF to FSL_IMX8_DDR_PMU
> > > * sort include
> > > * remove struct fsl_ddr_devtype_data
> > > * Added comment need disable control first
> > > * Added comment about must enable cycle counter
> > > * Added macro for EVENT_AXI_READ, remove hardcode 0x41 and 0x42
> > > * Added comment about cycle counter is fastest one
> > >
> > > Change from v2 to v3
> > > * remove kfree
> > >
> > > Change from V1 to V2
> > > * update Kconfig by use i.MX8 instead of i.MX8 QXP
> > > * remove gpl statememnt since SPDX tag
> > > * use dev_kzalloc
> > > * use dev_err
> > > * commit message show axi_read 0x41\axi_write 0x42
> > > * commit message show cycles must be enabled
> > > * Irq only issue at cycles overflow
> > > * use NUM_COUNTER
> > > * use devm_request_irq
> > > * add hotplug callback to handle context migration
> > >
> > > drivers/perf/Kconfig | 7 +
> > > drivers/perf/Makefile | 1 +
> > > drivers/perf/fsl_imx8_ddr_perf.c | 573 +++++++++++++++++++++++++++++++++++++++
> > > 3 files changed, 581 insertions(+)
> > > create mode 100644 drivers/perf/fsl_imx8_ddr_perf.c
> > >
> > > diff --git a/drivers/perf/Kconfig b/drivers/perf/Kconfig
> > > index af9bc17..e567437 100644
> > > --- a/drivers/perf/Kconfig
> > > +++ b/drivers/perf/Kconfig
> > > @@ -61,6 +61,13 @@ config ARM_DSU_PMU
> > > system, control logic. The PMU allows counting various events related
> > > to DSU.
> > >
> > > +config FSL_IMX8_DDR_PMU
> > > + tristate "Freescale i.MX8 DDR perf monitor"
> > > + depends on ARCH_MXC
> > > + help
> > > + Provides support for ddr perfomance monitor in i.MX8. Provide memory
> > > + througput information.
> > > +
> > > config HISI_PMU
> > > bool "HiSilicon SoC PMU"
> > > depends on ARM64 && ACPI
> > > diff --git a/drivers/perf/Makefile b/drivers/perf/Makefile
> > > index 909f27f..0bc066b 100644
> > > --- a/drivers/perf/Makefile
> > > +++ b/drivers/perf/Makefile
> > > @@ -4,6 +4,7 @@ obj-$(CONFIG_ARM_CCN) += arm-ccn.o
> > > obj-$(CONFIG_ARM_DSU_PMU) += arm_dsu_pmu.o
> > > obj-$(CONFIG_ARM_PMU) += arm_pmu.o arm_pmu_platform.o
> > > obj-$(CONFIG_ARM_PMU_ACPI) += arm_pmu_acpi.o
> > > +obj-$(CONFIG_FSL_IMX8_DDR_PMU) += fsl_imx8_ddr_perf.o
> > > obj-$(CONFIG_HISI_PMU) += hisilicon/
> > > obj-$(CONFIG_QCOM_L2_PMU) += qcom_l2_pmu.o
> > > obj-$(CONFIG_QCOM_L3_PMU) += qcom_l3_pmu.o
> > > diff --git a/drivers/perf/fsl_imx8_ddr_perf.c b/drivers/perf/fsl_imx8_ddr_perf.c
> > > new file mode 100644
> > > index 0000000..3d02daf
> > > --- /dev/null
> > > +++ b/drivers/perf/fsl_imx8_ddr_perf.c
> > > @@ -0,0 +1,573 @@
> > > +// SPDX-License-Identifier: GPL-2.0
> > > +/*
> > > + * Copyright 2017 NXP
> > > + * Copyright 2016 Freescale Semiconductor, Inc.
> > > + */
> > > +
> > > +#include <linux/init.h>
> > > +#include <linux/interrupt.h>
> > > +#include <linux/io.h>
> > > +#include <linux/module.h>
> > > +#include <linux/of.h>
> > > +#include <linux/of_address.h>
> > > +#include <linux/of_device.h>
> > > +#include <linux/of_irq.h>
> > > +#include <linux/perf_event.h>
> > > +#include <linux/slab.h>
> > > +
> > > +#define COUNTER_CNTL 0x0
> > > +#define COUNTER_READ 0x20
> > > +
> > > +#define COUNTER_DPCR1 0x30
> > > +
> > > +#define CNTL_OVER 0x1
> > > +#define CNTL_CLEAR 0x2
> > > +#define CNTL_EN 0x4
> > > +#define CNTL_EN_MASK 0xFFFFFFFB
> > > +#define CNTL_CLEAR_MASK 0xFFFFFFFD
> > > +#define CNTL_OVER_MASK 0xFFFFFFFE
> > > +
> > > +#define CNTL_CSV_SHIFT 24
> > > +#define CNTL_CSV_MASK (0xFF << CNTL_CSV_SHIFT)
> > > +
> > > +#define EVENT_CYCLES_ID 0
> > > +#define EVENT_CYCLES_COUNTER 0
> > > +#define NUM_COUNTER 4
> > > +#define MAX_EVENT 3
> > > +#define EVENT_AXI_READ 0x41
> > > +#define EVENT_AXI_WRITE 0x42
> > > +
> > > +#define to_ddr_pmu(p) container_of(p, struct ddr_pmu, pmu)
> > > +
> > > +#define DDR_PERF_DEV_NAME "ddr_perf"
> > > +
> > > +static DEFINE_IDA(ddr_ida);
> > > +
> > > +PMU_EVENT_ATTR_STRING(cycles, ddr_perf_cycles, "event=0x00");
> > > +PMU_EVENT_ATTR_STRING(selfresh, ddr_perf_selfresh, "event=0x01");
> > > +PMU_EVENT_ATTR_STRING(read-access, ddr_perf_read_accesses, "event=0x04");
> > > +PMU_EVENT_ATTR_STRING(write-access, ddr_perf_write_accesses, "event=0x05");
> > > +PMU_EVENT_ATTR_STRING(read-queue-depth, ddr_perf_read_queue_depth,
> > > + "event=0x08");
> > > +PMU_EVENT_ATTR_STRING(write-queue-depth, ddr_perf_write_queue_depth,
> > > + "event=0x09");
> > > +PMU_EVENT_ATTR_STRING(lp-read-credit-cnt, ddr_perf_lp_read_credit_cnt,
> > > + "event=0x10");
> > > +PMU_EVENT_ATTR_STRING(hp-read-credit-cnt, ddr_perf_hp_read_credit_cnt,
> > > + "event=0x11");
> > > +PMU_EVENT_ATTR_STRING(write-credit-cnt, ddr_perf_write_credit_cnt,
> > > + "event=0x12");
> > > +PMU_EVENT_ATTR_STRING(read-command, ddr_perf_read_command, "event=0x20");
> > > +PMU_EVENT_ATTR_STRING(write-command, ddr_perf_write_command, "event=0x21");
> > > +PMU_EVENT_ATTR_STRING(read-modify-write-command,
> > > + ddr_perf_read_modify_write_command, "event=0x22");
> > > +PMU_EVENT_ATTR_STRING(hp-read, ddr_perf_hp_read, "event=0x23");
> > > +PMU_EVENT_ATTR_STRING(hp-req-nodcredit, ddr_perf_hp_req_nocredit, "event=0x24");
> > > +PMU_EVENT_ATTR_STRING(hp-xact-credit, ddr_perf_hp_xact_credit, "event=0x25");
> > > +PMU_EVENT_ATTR_STRING(lp-req-nocredit, ddr_perf_lp_req_nocredit, "event=0x26");
> > > +PMU_EVENT_ATTR_STRING(lp-xact-credit, ddr_perf_lp_xact_credit, "event=0x27");
> > > +PMU_EVENT_ATTR_STRING(wr-xact-credit, ddr_perf_wr_xact_credit, "event=0x29");
> > > +PMU_EVENT_ATTR_STRING(read-cycles, ddr_perf_read_cycles, "event=0x2a");
> > > +PMU_EVENT_ATTR_STRING(write-cycles, ddr_perf_write_cycles, "event=0x2b");
> > > +PMU_EVENT_ATTR_STRING(read-write-transition, ddr_perf_read_write_transition,
> > > + "event=0x30");
> > > +PMU_EVENT_ATTR_STRING(precharge, ddr_perf_precharge, "event=0x31");
> > > +PMU_EVENT_ATTR_STRING(activate, ddr_perf_activate, "event=0x32");
> > > +PMU_EVENT_ATTR_STRING(load-mode, ddr_perf_load_mode, "event=0x33");
> > > +PMU_EVENT_ATTR_STRING(mwr, ddr_perf_mwr, "event=0x34");
> > > +PMU_EVENT_ATTR_STRING(read, ddr_perf_read, "event=0x35");
> > > +PMU_EVENT_ATTR_STRING(read-activate, ddr_perf_read_activate, "event=0x36");
> > > +PMU_EVENT_ATTR_STRING(refresh, ddr_perf_refresh, "event=0x37");
> > > +PMU_EVENT_ATTR_STRING(write, ddr_perf_write, "event=0x38");
> > > +PMU_EVENT_ATTR_STRING(raw-hazard, ddr_perf_raw_hazard, "event=0x39");
> > > +
> > > +PMU_EVENT_ATTR_STRING(axid-read, ddr_perf_axid_read,
> > > + "event="__stringify(EVENT_AXI_READ));
> > > +PMU_EVENT_ATTR_STRING(axid-write, ddr_perf_axid_write,
> > > + "event="__stringify(EVENT_AXI_WRITE));
> > > +
> > > +#define DDR_CAP_AXI_ID 0x1
> > > +
> > > +static const struct of_device_id imx_ddr_pmu_dt_ids[] = {
> > > + { .compatible = "fsl,imx8-ddr-pmu",},
> > > + { .compatible = "fsl,imx8m-ddr-pmu", .data = (void *)DDR_CAP_AXI_ID},
> > > + { /* sentinel */ }
> > > +};
> > > +
> > > +struct ddr_pmu {
> > > + struct pmu pmu;
> > > + void __iomem *base;
> > > + cpumask_t cpu;
> > > + struct hlist_node node;
> > > + struct device *dev;
> > > + struct perf_event *active_events[NUM_COUNTER];
> > > + int total_events;
> > > + enum cpuhp_state cpuhp_state;
> > > + bool cycles_active;
> > > + uintptr_t flags;
> > > +};
> > > +
> > > +static ssize_t ddr_perf_cpumask_show(struct device *dev,
> > > + struct device_attribute *attr, char *buf)
> > > +{
> > > + struct ddr_pmu *pmu = dev_get_drvdata(dev);
> > > +
> > > + return cpumap_print_to_pagebuf(true, buf, &pmu->cpu);
> > > +}
> > > +
> > > +static struct device_attribute ddr_perf_cpumask_attr =
> > > + __ATTR(cpumask, 0444, ddr_perf_cpumask_show, NULL);
> > > +
> > > +static struct attribute *ddr_perf_cpumask_attrs[] = {
> > > + &ddr_perf_cpumask_attr.attr,
> > > + NULL,
> > > +};
> > > +
> > > +static struct attribute_group ddr_perf_cpumask_attr_group = {
> > > + .attrs = ddr_perf_cpumask_attrs,
> > > +};
> > > +
> > > +static struct attribute *ddr_perf_events_attrs[] = {
> > > + &ddr_perf_cycles.attr.attr,
> > > + &ddr_perf_selfresh.attr.attr,
> > > + &ddr_perf_read_accesses.attr.attr,
> > > + &ddr_perf_write_accesses.attr.attr,
> > > + &ddr_perf_read_queue_depth.attr.attr,
> > > + &ddr_perf_write_queue_depth.attr.attr,
> > > + &ddr_perf_lp_read_credit_cnt.attr.attr,
> > > + &ddr_perf_hp_read_credit_cnt.attr.attr,
> > > + &ddr_perf_write_credit_cnt.attr.attr,
> > > + &ddr_perf_read_command.attr.attr,
> > > + &ddr_perf_write_command.attr.attr,
> > > + &ddr_perf_read_modify_write_command.attr.attr,
> > > + &ddr_perf_hp_read.attr.attr,
> > > + &ddr_perf_hp_req_nocredit.attr.attr,
> > > + &ddr_perf_hp_xact_credit.attr.attr,
> > > + &ddr_perf_lp_req_nocredit.attr.attr,
> > > + &ddr_perf_lp_xact_credit.attr.attr,
> > > + &ddr_perf_wr_xact_credit.attr.attr,
> > > + &ddr_perf_read_cycles.attr.attr,
> > > + &ddr_perf_write_cycles.attr.attr,
> > > + &ddr_perf_read_write_transition.attr.attr,
> > > + &ddr_perf_precharge.attr.attr,
> > > + &ddr_perf_activate.attr.attr,
> > > + &ddr_perf_load_mode.attr.attr,
> > > + &ddr_perf_mwr.attr.attr,
> > > + &ddr_perf_read.attr.attr,
> > > + &ddr_perf_read_activate.attr.attr,
> > > + &ddr_perf_refresh.attr.attr,
> > > + &ddr_perf_write.attr.attr,
> > > + &ddr_perf_raw_hazard.attr.attr,
> > > + &ddr_perf_axid_read.attr.attr,
> > > + &ddr_perf_axid_write.attr.attr,
> > > + NULL,
> > > +};
> > > +
> > > +static struct attribute_group ddr_perf_events_attr_group = {
> > > + .name = "events",
> > > + .attrs = ddr_perf_events_attrs,
> > > +};
> > > +
> > > +PMU_FORMAT_ATTR(event, "config:0-63");
> > > +PMU_FORMAT_ATTR(axi_id, "config1:0-63");
> > > +
> > > +static struct attribute *ddr_perf_format_attrs[] = {
> > > + &format_attr_event.attr,
> > > + &format_attr_axi_id.attr,
> > > + NULL,
> > > +};
> > > +
> > > +static struct attribute_group ddr_perf_format_attr_group = {
> > > + .name = "format",
> > > + .attrs = ddr_perf_format_attrs,
> > > +};
> > > +
> > > +static const struct attribute_group *attr_groups[] = {
> > > + &ddr_perf_events_attr_group,
> > > + &ddr_perf_format_attr_group,
> > > + &ddr_perf_cpumask_attr_group,
> > > + NULL,
> > > +};
> > > +
> > > +static u32 ddr_perf_alloc_counter(struct ddr_pmu *pmu, int event)
> > > +{
> > > + int i;
> > > +
> > > + /* Always map cycle event to counter 0
> > > + Cycles counter is dedicated for cycle event
> > > + can't used for the other events
> > > + */
> > > + if (event == EVENT_CYCLES_ID)
> > > + return EVENT_CYCLES_COUNTER;
> > > +
> > > + for (i = 1; i < NUM_COUNTER; i++)
> > > + if (pmu->active_events[i] == NULL)
> > > + return i;
> > > +
> > > + return -ENOENT;
> > > +}
> > > +
> > > +static u32 ddr_perf_free_counter(struct ddr_pmu *pmu, int counter)
> > > +{
> > > + if (counter < 0 || counter >= NUM_COUNTER)
> > > + return -ENOENT;
> > > +
> > > + pmu->active_events[counter] = NULL;
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +static u32 ddr_perf_read_counter(struct ddr_pmu *pmu, int counter)
> > > +{
> > > + return readl(pmu->base + COUNTER_READ + counter * 4);
> > > +}
> > > +
> > > +static int ddr_perf_event_init(struct perf_event *event)
> > > +{
> > > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > > + struct hw_perf_event *hwc = &event->hw;
> > > +
> > > + if (event->attr.type != event->pmu->type)
> > > + return -ENOENT;
> > > +
> > > + if (is_sampling_event(event) || event->attach_state & PERF_ATTACH_TASK)
> > > + return -EOPNOTSUPP;
> > > +
> > > + if (event->cpu < 0) {
> > > + dev_warn(pmu->dev, "Can't provide per-task data!\n");
> > > + return -EOPNOTSUPP;
> > > + }
> > > +
> > > + if (event->attr.exclude_user ||
> > > + event->attr.exclude_kernel ||
> > > + event->attr.exclude_hv ||
> > > + event->attr.exclude_idle ||
> > > + event->attr.exclude_host ||
> > > + event->attr.exclude_guest ||
> > > + event->attr.sample_period)
> > > + return -EINVAL;
> > > +
> > > + event->cpu = cpumask_first(&pmu->cpu);
> > > + hwc->idx = -1;
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +
> > > +static void ddr_perf_event_update(struct perf_event *event)
> > > +{
> > > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > > + struct hw_perf_event *hwc = &event->hw;
> > > + u64 delta, prev_raw_count, new_raw_count;
> > > + int counter = hwc->idx;
> > > +
> > > + do {
> > > + prev_raw_count = local64_read(&hwc->prev_count);
> > > + new_raw_count = ddr_perf_read_counter(pmu, counter);
> > > + } while (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
> > > + new_raw_count) != prev_raw_count);
> > > +
> > > + delta = (new_raw_count - prev_raw_count) & 0xFFFFFFFF;
> > > +
> > > + local64_add(delta, &event->count);
> > > +}
> > > +
> > > +static void ddr_perf_event_enable(struct ddr_pmu *pmu, int config,
> > > + int counter, bool enable)
> > > +{
> > > + u8 reg = counter * 4 + COUNTER_CNTL;
> > > + int val;
> > > +
> > > + if (enable) {
> > > + /* must disable first, then enable again
> > > + * otherwise, cycle counter will not work
> > > + * if previous state is enabled.
> > > + */
> > > + writel(0, pmu->base + reg);
> > > + val = CNTL_EN | CNTL_CLEAR;
> > > + val |= (config << CNTL_CSV_SHIFT) & CNTL_CSV_MASK;
> > > + } else {
> > > + /* Disable counter */
> > > + val = readl(pmu->base + reg) & CNTL_EN_MASK;
> > > + }
> > > +
> > > + writel(val, pmu->base + reg);
> > > +
> > > + if (config == EVENT_CYCLES_ID)
> > > + pmu->cycles_active = enable;
> > > +}
> > > +
> > > +static void ddr_perf_event_start(struct perf_event *event, int flags)
> > > +{
> > > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > > + struct hw_perf_event *hwc = &event->hw;
> > > + int counter = hwc->idx;
> > > +
> > > + if (pmu->flags & DDR_CAP_AXI_ID) {
> > > + if (event->attr.config == EVENT_AXI_READ ||
> > > + event->attr.config == EVENT_AXI_WRITE) {
> > > + int val = event->attr.config1;
> > > +
> > > + writel(val, pmu->base + COUNTER_DPCR1);
> > > + }
> > > + }
> > > +
> > > + local64_set(&hwc->prev_count, 0);
> > > +
> > > + ddr_perf_event_enable(pmu, event->attr.config, counter, true);
> > > + /*
> > > + * If the cycles counter wasn't explicitly selected,
> > > + * we will enable it now.
> > > + * cycles counter must be enabled otherwise other counters will
> > > + * stopped.
> > > + */
> > > + if (counter > 0 && !pmu->cycles_active)
> > > + ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
> > > + EVENT_CYCLES_COUNTER, true);
> > > +}
> > > +
> > > +static int ddr_perf_event_add(struct perf_event *event, int flags)
> > > +{
> > > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > > + struct hw_perf_event *hwc = &event->hw;
> > > + int counter;
> > > + int cfg = event->attr.config;
> > > +
> > > + counter = ddr_perf_alloc_counter(pmu, cfg);
> > > + if (counter < 0) {
> > > + dev_dbg(pmu->dev, "There are not enough counters\n");
> > > + return -EOPNOTSUPP;
> > > + }
> > > +
> > > + pmu->active_events[counter] = event;
> > > + pmu->total_events++;
> > > + hwc->idx = counter;
> > > +
> > > + if (flags & PERF_EF_START)
> > > + ddr_perf_event_start(event, flags);
> > > +
> > > + local64_set(&hwc->prev_count, ddr_perf_read_counter(pmu, counter));
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +static void ddr_perf_event_stop(struct perf_event *event, int flags)
> > > +{
> > > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > > + struct hw_perf_event *hwc = &event->hw;
> > > + int counter = hwc->idx;
> > > +
> > > + ddr_perf_event_enable(pmu, event->attr.config, counter, false);
> > > + ddr_perf_event_update(event);
> > > +}
> > > +
> > > +static void ddr_perf_event_del(struct perf_event *event, int flags)
> > > +{
> > > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > > + struct hw_perf_event *hwc = &event->hw;
> > > + int counter = hwc->idx;
> > > +
> > > + ddr_perf_event_stop(event, PERF_EF_UPDATE);
> > > +
> > > + ddr_perf_free_counter(pmu, counter);
> > > + pmu->total_events--;
> > > + hwc->idx = -1;
> > > +
> > > + /* If all events have stopped, stop the cycles counter as well */
> > > + if ((pmu->total_events == 0) && pmu->cycles_active)
> > > + ddr_perf_event_enable(pmu, EVENT_CYCLES_ID,
> > > + EVENT_CYCLES_COUNTER, false);
> > > +}
> > > +
> > > +static int ddr_perf_init(struct ddr_pmu *pmu, void __iomem *base,
> > > + struct device *dev)
> > > +{
> > > + *pmu = (struct ddr_pmu) {
> > > + .pmu = (struct pmu) {
> > > + .task_ctx_nr = perf_invalid_context,
> > > + .attr_groups = attr_groups,
> > > + .event_init = ddr_perf_event_init,
> > > + .add = ddr_perf_event_add,
> > > + .del = ddr_perf_event_del,
> > > + .start = ddr_perf_event_start,
> > > + .stop = ddr_perf_event_stop,
> > > + .read = ddr_perf_event_update,
> > > + },
> > > + .base = base,
> > > + .dev = dev,
> > > + };
> > > +
> > > + return ida_simple_get(&ddr_ida, 0, 0, GFP_KERNEL);
> > > +}
> > > +
> > > +static irqreturn_t ddr_perf_irq_handler(int irq, void *p)
> > > +{
> > > + int i;
> > > + u8 reg;
> > > + int val;
> > > + int counter;
> > > + struct ddr_pmu *pmu = (struct ddr_pmu *) p;
> > > + struct perf_event *event;
> > > +
> > > + /* Only cycles counter overflowed can issue irq. all counters will
> > > + * be stopped when cycles counter overflow. but other counter don't stop
> > > + * when overflow happen. Update all of the local counter values,
> > > + * then reset the cycles counter, so the others can continue
> > > + * counting. cycles counter is fastest counter in all events. at last
> > > + * 4 times than other counters.
> > > + */
> > > + for (i = 0; i < NUM_COUNTER; i++) {
> > > +
> > > + if (!pmu->active_events[i])
> > > + continue;
> > > +
> > > + event = pmu->active_events[i];
> > > + counter = event->hw.idx;
> > > + reg = counter * 4 + COUNTER_CNTL;
> > > + val = readl(pmu->base + reg);
> > > +
> > > + ddr_perf_event_update(event);
> > > +
> > > + if (counter == EVENT_CYCLES_COUNTER) {
> > > + ddr_perf_event_enable(pmu,
> > > + EVENT_CYCLES_ID,
> > > + EVENT_CYCLES_COUNTER,
> > > + true);
> > > + ddr_perf_event_update(event);
> > > + }
> > > + }
> > > +
> > > + return IRQ_HANDLED;
> > > +}
> > > +
> > > +static int ddr_perf_offline_cpu(unsigned int cpu, struct hlist_node *node)
> > > +{
> > > + struct ddr_pmu *pmu = hlist_entry_safe(node, struct ddr_pmu, node);
> > > + int target;
> > > +
> > > + if (!cpumask_test_and_clear_cpu(cpu, &pmu->cpu))
> > > + return 0;
> > > +
> > > + target = cpumask_any_but(cpu_online_mask, cpu);
> > > + if (target >= nr_cpu_ids)
> > > + return 0;
> > > +
> > > + perf_pmu_migrate_context(&pmu->pmu, cpu, target);
> > > + cpumask_set_cpu(target, &pmu->cpu);
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +static int ddr_perf_probe(struct platform_device *pdev)
> > > +{
> > > + struct ddr_pmu *pmu;
> > > + struct device_node *np;
> > > + void __iomem *base;
> > > + struct resource *iomem;
> > > + char *name;
> > > + char *hpname;
> > > + int num;
> > > + int ret;
> > > + u32 irq;
> > > + const struct of_device_id *of_id =
> > > + of_match_device(imx_ddr_pmu_dt_ids, &pdev->dev);
> > > +
> > > + iomem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> > > + base = devm_ioremap_resource(&pdev->dev, iomem);
> > > + if (IS_ERR(base))
> > > + return PTR_ERR(base);
> > > +
> > > + np = pdev->dev.of_node;
> > > +
> > > + pmu = devm_kzalloc(&pdev->dev, sizeof(*pmu), GFP_KERNEL);
> > > + if (!pmu)
> > > + return -ENOMEM;
> > > +
> > > + num = ddr_perf_init(pmu, base, &pdev->dev);
> > > +
> > > + name = devm_kasprintf(&pdev->dev, GFP_KERNEL, "ddr%d", num);
> > > + if (!name)
> > > + return -ENOMEM;
> > > +
> > > + hpname = devm_kasprintf(&pdev->dev, GFP_KERNEL,
> > > + "perf/imx/ddr%d:online", num);
> > > + if (!hpname)
> > > + return -ENOMEM;
> > > +
> > > + pmu->flags = (uintptr_t) of_id->data;
> > > +
> > > + cpumask_set_cpu(raw_smp_processor_id(), &pmu->cpu);
> > > + ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN, hpname, NULL,
> > > + ddr_perf_offline_cpu);
> > > +
> > > + if (ret < 0) {
> > > + dev_err(&pdev->dev, "cpuhp_setup_state_multi failed\n");
> > > + goto ddr_perf_err;
> > > + }
> > > +
> > > + pmu->cpuhp_state = ret;
> > > +
> > > + /* Register the pmu instance for cpu hotplug */
> > > + cpuhp_state_add_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> > > +
> > > + ret = perf_pmu_register(&(pmu->pmu), name, -1);
> > > + if (ret)
> > > + goto ddr_perf_err;
> > > +
> > > + /* Request irq */
> > > + irq = of_irq_get(np, 0);
> > > + if (irq < 0) {
> > > + dev_err(&pdev->dev, "Failed to get irq: %d", irq);
> > > + ret = irq;
> > > + goto ddr_perf_irq_err;
> > > + }
> > > +
> > > + ret = devm_request_irq(&pdev->dev, irq,
> > > + ddr_perf_irq_handler,
> > > + IRQF_TRIGGER_RISING | IRQF_ONESHOT,
> > > + DDR_PERF_DEV_NAME,
> > > + pmu);
> > > + if (ret < 0) {
> > > + dev_err(&pdev->dev, "Request irq failed: %d", ret);
> > > + goto ddr_perf_irq_err;
> > > + }
> > > +
> > > + return 0;
> > > +
> > > +ddr_perf_irq_err:
> > > + perf_pmu_unregister(&(pmu->pmu));
> > > +
> > > +ddr_perf_err:
> > > + if (pmu->cpuhp_state)
> > > + cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> > > +
> > > + dev_warn(&pdev->dev, "i.MX8 DDR Perf PMU failed (%d), disabled\n", ret);
> > > + return ret;
> > > +}
> > > +
> > > +static int ddr_perf_remove(struct platform_device *pdev)
> > > +{
> > > + struct ddr_pmu *pmu = platform_get_drvdata(pdev);
> > > +
> > > + cpuhp_state_remove_instance_nocalls(pmu->cpuhp_state, &pmu->node);
> > > + perf_pmu_unregister(&pmu->pmu);
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +static struct platform_driver imx_ddr_pmu_driver = {
> > > + .driver = {
> > > + .name = "imx-ddr-pmu",
> > > + .of_match_table = imx_ddr_pmu_dt_ids,
> > > + },
> > > + .probe = ddr_perf_probe,
> > > + .remove = ddr_perf_remove,
> > > +};
> > > +
> > > +static int __init imx_ddr_pmu_init(void)
> > > +{
> > > + return platform_driver_register(&imx_ddr_pmu_driver);
> > > +}
> > > +
> > > +module_init(imx_ddr_pmu_init);
> > > +
> > > --
> > > 2.5.2
> > >
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
2019-02-15 18:03 ` Frank Li
@ 2019-04-04 11:17 ` Will Deacon
-1 siblings, 0 replies; 43+ messages in thread
From: Will Deacon @ 2019-04-04 11:17 UTC (permalink / raw)
To: Frank Li
Cc: mark.rutland, Aisheng Dong, devicetree, shawnguo, s.hauer,
robh+dt, dl-linux-imx, kernel, lznuaa, festevam,
linux-arm-kernel
Hi Frank,
On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> Add ddr performance monitor support for iMX8QXP
>
> There are 4 counters for ddr perfomance events.
> counter 0 is dedicated for cycles.
> you choose any up to 3 no cycles events.
>
> for example:
>
> perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
does the COUNTER_DPCR1 register work and what happens if I specify two
simultaneous events with different values in config1? I'm a little wary
of taking raw user input and throwing it straight into a device register!
Will
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
@ 2019-04-04 11:17 ` Will Deacon
0 siblings, 0 replies; 43+ messages in thread
From: Will Deacon @ 2019-04-04 11:17 UTC (permalink / raw)
To: Frank Li
Cc: mark.rutland, Aisheng Dong, devicetree, shawnguo, s.hauer,
robh+dt, dl-linux-imx, kernel, lznuaa, festevam,
linux-arm-kernel
Hi Frank,
On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> Add ddr performance monitor support for iMX8QXP
>
> There are 4 counters for ddr perfomance events.
> counter 0 is dedicated for cycles.
> you choose any up to 3 no cycles events.
>
> for example:
>
> perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
does the COUNTER_DPCR1 register work and what happens if I specify two
simultaneous events with different values in config1? I'm a little wary
of taking raw user input and throwing it straight into a device register!
Will
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
2019-04-04 11:17 ` Will Deacon
@ 2019-04-05 14:34 ` Zhi Li
-1 siblings, 0 replies; 43+ messages in thread
From: Zhi Li @ 2019-04-05 14:34 UTC (permalink / raw)
To: Will Deacon
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Thu, Apr 4, 2019 at 6:17 AM Will Deacon <will.deacon@arm.com> wrote:
>
> Hi Frank,
>
> On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> > Add ddr performance monitor support for iMX8QXP
> >
> > There are 4 counters for ddr perfomance events.
> > counter 0 is dedicated for cycles.
> > you choose any up to 3 no cycles events.
> >
> > for example:
> >
> > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
>
> Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
> does the COUNTER_DPCR1 register work and what happens if I specify two
> simultaneous events with different values in config1? I'm a little wary
> of taking raw user input and throwing it straight into a device register!
I have not find better method to do that.
bit 0-15: AXIID
bit 16-31: AXID_MASK
Check condition AxID && AXI_MASKING == AXI_ID && AXI_MASKING
best regards
Frank Li
>
> Will
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
@ 2019-04-05 14:34 ` Zhi Li
0 siblings, 0 replies; 43+ messages in thread
From: Zhi Li @ 2019-04-05 14:34 UTC (permalink / raw)
To: Will Deacon
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Thu, Apr 4, 2019 at 6:17 AM Will Deacon <will.deacon@arm.com> wrote:
>
> Hi Frank,
>
> On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> > Add ddr performance monitor support for iMX8QXP
> >
> > There are 4 counters for ddr perfomance events.
> > counter 0 is dedicated for cycles.
> > you choose any up to 3 no cycles events.
> >
> > for example:
> >
> > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
>
> Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
> does the COUNTER_DPCR1 register work and what happens if I specify two
> simultaneous events with different values in config1? I'm a little wary
> of taking raw user input and throwing it straight into a device register!
I have not find better method to do that.
bit 0-15: AXIID
bit 16-31: AXID_MASK
Check condition AxID && AXI_MASKING == AXI_ID && AXI_MASKING
best regards
Frank Li
>
> Will
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
2019-04-05 14:34 ` Zhi Li
@ 2019-04-05 14:38 ` Will Deacon
-1 siblings, 0 replies; 43+ messages in thread
From: Will Deacon @ 2019-04-05 14:38 UTC (permalink / raw)
To: Zhi Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Apr 05, 2019 at 09:34:38AM -0500, Zhi Li wrote:
> On Thu, Apr 4, 2019 at 6:17 AM Will Deacon <will.deacon@arm.com> wrote:
> >
> > Hi Frank,
> >
> > On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> > > Add ddr performance monitor support for iMX8QXP
> > >
> > > There are 4 counters for ddr perfomance events.
> > > counter 0 is dedicated for cycles.
> > > you choose any up to 3 no cycles events.
> > >
> > > for example:
> > >
> > > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> >
> > Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
> > does the COUNTER_DPCR1 register work and what happens if I specify two
> > simultaneous events with different values in config1? I'm a little wary
> > of taking raw user input and throwing it straight into a device register!
>
> I have not find better method to do that.
> bit 0-15: AXIID
> bit 16-31: AXID_MASK
>
> Check condition AxID && AXI_MASKING == AXI_ID && AXI_MASKING
Hmm, that doesn't really answer my second question :(
What happens if I specify two simultaneous events with different values
in config1?
Will
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
@ 2019-04-05 14:38 ` Will Deacon
0 siblings, 0 replies; 43+ messages in thread
From: Will Deacon @ 2019-04-05 14:38 UTC (permalink / raw)
To: Zhi Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Apr 05, 2019 at 09:34:38AM -0500, Zhi Li wrote:
> On Thu, Apr 4, 2019 at 6:17 AM Will Deacon <will.deacon@arm.com> wrote:
> >
> > Hi Frank,
> >
> > On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> > > Add ddr performance monitor support for iMX8QXP
> > >
> > > There are 4 counters for ddr perfomance events.
> > > counter 0 is dedicated for cycles.
> > > you choose any up to 3 no cycles events.
> > >
> > > for example:
> > >
> > > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> >
> > Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
> > does the COUNTER_DPCR1 register work and what happens if I specify two
> > simultaneous events with different values in config1? I'm a little wary
> > of taking raw user input and throwing it straight into a device register!
>
> I have not find better method to do that.
> bit 0-15: AXIID
> bit 16-31: AXID_MASK
>
> Check condition AxID && AXI_MASKING == AXI_ID && AXI_MASKING
Hmm, that doesn't really answer my second question :(
What happens if I specify two simultaneous events with different values
in config1?
Will
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
2019-04-05 14:38 ` Will Deacon
@ 2019-04-05 14:58 ` Zhi Li
-1 siblings, 0 replies; 43+ messages in thread
From: Zhi Li @ 2019-04-05 14:58 UTC (permalink / raw)
To: Will Deacon
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Apr 5, 2019 at 9:38 AM Will Deacon <will.deacon@arm.com> wrote:
>
> On Fri, Apr 05, 2019 at 09:34:38AM -0500, Zhi Li wrote:
> > On Thu, Apr 4, 2019 at 6:17 AM Will Deacon <will.deacon@arm.com> wrote:
> > >
> > > Hi Frank,
> > >
> > > On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> > > > Add ddr performance monitor support for iMX8QXP
> > > >
> > > > There are 4 counters for ddr perfomance events.
> > > > counter 0 is dedicated for cycles.
> > > > you choose any up to 3 no cycles events.
> > > >
> > > > for example:
> > > >
> > > > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > > > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> > >
> > > Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
Only imx845 have AXID filter capability now.
> > > does the COUNTER_DPCR1 register work and what happens if I specify two
> > > simultaneous events with different values in config1? I'm a little wary
There are difference match register for each event.
1. Read event with config 1 A
2. Read event with config 1 B
1 will show read count with filter A
2 will show read count with filter B.
> > > of taking raw user input and throwing it straight into a device register!
> >
> > I have not find better method to do that.
> > bit 0-15: AXIID
> > bit 16-31: AXID_MASK
> >
> > Check condition AxID && AXI_MASKING == AXI_ID && AXI_MASKING
>
> Hmm, that doesn't really answer my second question :(
>
> What happens if I specify two simultaneous events with different values
> in config1?
Sorry, see above answer.
>
> Will
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
@ 2019-04-05 14:58 ` Zhi Li
0 siblings, 0 replies; 43+ messages in thread
From: Zhi Li @ 2019-04-05 14:58 UTC (permalink / raw)
To: Will Deacon
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Apr 5, 2019 at 9:38 AM Will Deacon <will.deacon@arm.com> wrote:
>
> On Fri, Apr 05, 2019 at 09:34:38AM -0500, Zhi Li wrote:
> > On Thu, Apr 4, 2019 at 6:17 AM Will Deacon <will.deacon@arm.com> wrote:
> > >
> > > Hi Frank,
> > >
> > > On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> > > > Add ddr performance monitor support for iMX8QXP
> > > >
> > > > There are 4 counters for ddr perfomance events.
> > > > counter 0 is dedicated for cycles.
> > > > you choose any up to 3 no cycles events.
> > > >
> > > > for example:
> > > >
> > > > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > > > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> > >
> > > Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
Only imx845 have AXID filter capability now.
> > > does the COUNTER_DPCR1 register work and what happens if I specify two
> > > simultaneous events with different values in config1? I'm a little wary
There are difference match register for each event.
1. Read event with config 1 A
2. Read event with config 1 B
1 will show read count with filter A
2 will show read count with filter B.
> > > of taking raw user input and throwing it straight into a device register!
> >
> > I have not find better method to do that.
> > bit 0-15: AXIID
> > bit 16-31: AXID_MASK
> >
> > Check condition AxID && AXI_MASKING == AXI_ID && AXI_MASKING
>
> Hmm, that doesn't really answer my second question :(
>
> What happens if I specify two simultaneous events with different values
> in config1?
Sorry, see above answer.
>
> Will
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
2019-04-05 14:58 ` Zhi Li
@ 2019-04-05 15:03 ` Will Deacon
-1 siblings, 0 replies; 43+ messages in thread
From: Will Deacon @ 2019-04-05 15:03 UTC (permalink / raw)
To: Zhi Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Apr 05, 2019 at 09:58:38AM -0500, Zhi Li wrote:
> On Fri, Apr 5, 2019 at 9:38 AM Will Deacon <will.deacon@arm.com> wrote:
> > On Fri, Apr 05, 2019 at 09:34:38AM -0500, Zhi Li wrote:
> > > On Thu, Apr 4, 2019 at 6:17 AM Will Deacon <will.deacon@arm.com> wrote:
> > > > On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> > > > > Add ddr performance monitor support for iMX8QXP
> > > > >
> > > > > There are 4 counters for ddr perfomance events.
> > > > > counter 0 is dedicated for cycles.
> > > > > you choose any up to 3 no cycles events.
> > > > >
> > > > > for example:
> > > > >
> > > > > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > > > > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> > > >
> > > > Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
>
> Only imx845 have AXID filter capability now.
>
> > > > does the COUNTER_DPCR1 register work and what happens if I specify two
> > > > simultaneous events with different values in config1? I'm a little wary
>
> There are difference match register for each event.
> 1. Read event with config 1 A
> 2. Read event with config 1 B
>
> 1 will show read count with filter A
> 2 will show read count with filter B.
Thanks, that makes sense, but I can't see how that corresponds to the code
in the patch:
+static void ddr_perf_event_start(struct perf_event *event, int flags)
+{
+ struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
+ struct hw_perf_event *hwc = &event->hw;
+ int counter = hwc->idx;
+
+ if (pmu->flags & DDR_CAP_AXI_ID) {
+ if (event->attr.config == EVENT_AXI_READ ||
+ event->attr.config == EVENT_AXI_WRITE) {
+ int val = event->attr.config1;
+
+ writel(val, pmu->base + COUNTER_DPCR1);
+ }
+ }
+
+ local64_set(&hwc->prev_count, 0);
+
+ ddr_perf_event_enable(pmu, event->attr.config, counter, true);
ddr_perf_event_enable() does what you'd expect, and uses hwc->idx to
allocate a counter, but the code before it just always writes to DPCR1.
What am I missing?
Will
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
@ 2019-04-05 15:03 ` Will Deacon
0 siblings, 0 replies; 43+ messages in thread
From: Will Deacon @ 2019-04-05 15:03 UTC (permalink / raw)
To: Zhi Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Apr 05, 2019 at 09:58:38AM -0500, Zhi Li wrote:
> On Fri, Apr 5, 2019 at 9:38 AM Will Deacon <will.deacon@arm.com> wrote:
> > On Fri, Apr 05, 2019 at 09:34:38AM -0500, Zhi Li wrote:
> > > On Thu, Apr 4, 2019 at 6:17 AM Will Deacon <will.deacon@arm.com> wrote:
> > > > On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> > > > > Add ddr performance monitor support for iMX8QXP
> > > > >
> > > > > There are 4 counters for ddr perfomance events.
> > > > > counter 0 is dedicated for cycles.
> > > > > you choose any up to 3 no cycles events.
> > > > >
> > > > > for example:
> > > > >
> > > > > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > > > > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> > > >
> > > > Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
>
> Only imx845 have AXID filter capability now.
>
> > > > does the COUNTER_DPCR1 register work and what happens if I specify two
> > > > simultaneous events with different values in config1? I'm a little wary
>
> There are difference match register for each event.
> 1. Read event with config 1 A
> 2. Read event with config 1 B
>
> 1 will show read count with filter A
> 2 will show read count with filter B.
Thanks, that makes sense, but I can't see how that corresponds to the code
in the patch:
+static void ddr_perf_event_start(struct perf_event *event, int flags)
+{
+ struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
+ struct hw_perf_event *hwc = &event->hw;
+ int counter = hwc->idx;
+
+ if (pmu->flags & DDR_CAP_AXI_ID) {
+ if (event->attr.config == EVENT_AXI_READ ||
+ event->attr.config == EVENT_AXI_WRITE) {
+ int val = event->attr.config1;
+
+ writel(val, pmu->base + COUNTER_DPCR1);
+ }
+ }
+
+ local64_set(&hwc->prev_count, 0);
+
+ ddr_perf_event_enable(pmu, event->attr.config, counter, true);
ddr_perf_event_enable() does what you'd expect, and uses hwc->idx to
allocate a counter, but the code before it just always writes to DPCR1.
What am I missing?
Will
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
2019-04-05 15:03 ` Will Deacon
@ 2019-04-08 19:10 ` Zhi Li
-1 siblings, 0 replies; 43+ messages in thread
From: Zhi Li @ 2019-04-08 19:10 UTC (permalink / raw)
To: Will Deacon
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Apr 5, 2019 at 10:03 AM Will Deacon <will.deacon@arm.com> wrote:
>
> On Fri, Apr 05, 2019 at 09:58:38AM -0500, Zhi Li wrote:
> > On Fri, Apr 5, 2019 at 9:38 AM Will Deacon <will.deacon@arm.com> wrote:
> > > On Fri, Apr 05, 2019 at 09:34:38AM -0500, Zhi Li wrote:
> > > > On Thu, Apr 4, 2019 at 6:17 AM Will Deacon <will.deacon@arm.com> wrote:
> > > > > On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> > > > > > Add ddr performance monitor support for iMX8QXP
> > > > > >
> > > > > > There are 4 counters for ddr perfomance events.
> > > > > > counter 0 is dedicated for cycles.
> > > > > > you choose any up to 3 no cycles events.
> > > > > >
> > > > > > for example:
> > > > > >
> > > > > > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > > > > > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> > > > >
> > > > > Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
> >
> > Only imx845 have AXID filter capability now.
> >
> > > > > does the COUNTER_DPCR1 register work and what happens if I specify two
> > > > > simultaneous events with different values in config1? I'm a little wary
> >
> > There are difference match register for each event.
> > 1. Read event with config 1 A
> > 2. Read event with config 1 B
> >
> > 1 will show read count with filter A
> > 2 will show read count with filter B.
>
> Thanks, that makes sense, but I can't see how that corresponds to the code
> in the patch:
>
> +static void ddr_perf_event_start(struct perf_event *event, int flags)
> +{
> + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> + struct hw_perf_event *hwc = &event->hw;
> + int counter = hwc->idx;
> +
> + if (pmu->flags & DDR_CAP_AXI_ID) {
> + if (event->attr.config == EVENT_AXI_READ ||
> + event->attr.config == EVENT_AXI_WRITE) {
> + int val = event->attr.config1;
> +
> + writel(val, pmu->base + COUNTER_DPCR1);
> + }
> + }
> +
> + local64_set(&hwc->prev_count, 0);
> +
> + ddr_perf_event_enable(pmu, event->attr.config, counter, true);
>
> ddr_perf_event_enable() does what you'd expect, and uses hwc->idx to
> allocate a counter, but the code before it just always writes to DPCR1.
>
> What am I missing?
I double check with our IC design team. My previous answer is wrong.
Only one DPCR1. It will impact all event EVENT_AXI_READ and EVENT_AXI_WRITE.
Thank means just choose one config1 for event EVENT_AXI_READ or EVENT_AXI_WRITE
best regards
Frank Li
>
> Will
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
@ 2019-04-08 19:10 ` Zhi Li
0 siblings, 0 replies; 43+ messages in thread
From: Zhi Li @ 2019-04-08 19:10 UTC (permalink / raw)
To: Will Deacon
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Fri, Apr 5, 2019 at 10:03 AM Will Deacon <will.deacon@arm.com> wrote:
>
> On Fri, Apr 05, 2019 at 09:58:38AM -0500, Zhi Li wrote:
> > On Fri, Apr 5, 2019 at 9:38 AM Will Deacon <will.deacon@arm.com> wrote:
> > > On Fri, Apr 05, 2019 at 09:34:38AM -0500, Zhi Li wrote:
> > > > On Thu, Apr 4, 2019 at 6:17 AM Will Deacon <will.deacon@arm.com> wrote:
> > > > > On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> > > > > > Add ddr performance monitor support for iMX8QXP
> > > > > >
> > > > > > There are 4 counters for ddr perfomance events.
> > > > > > counter 0 is dedicated for cycles.
> > > > > > you choose any up to 3 no cycles events.
> > > > > >
> > > > > > for example:
> > > > > >
> > > > > > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > > > > > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> > > > >
> > > > > Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
> >
> > Only imx845 have AXID filter capability now.
> >
> > > > > does the COUNTER_DPCR1 register work and what happens if I specify two
> > > > > simultaneous events with different values in config1? I'm a little wary
> >
> > There are difference match register for each event.
> > 1. Read event with config 1 A
> > 2. Read event with config 1 B
> >
> > 1 will show read count with filter A
> > 2 will show read count with filter B.
>
> Thanks, that makes sense, but I can't see how that corresponds to the code
> in the patch:
>
> +static void ddr_perf_event_start(struct perf_event *event, int flags)
> +{
> + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> + struct hw_perf_event *hwc = &event->hw;
> + int counter = hwc->idx;
> +
> + if (pmu->flags & DDR_CAP_AXI_ID) {
> + if (event->attr.config == EVENT_AXI_READ ||
> + event->attr.config == EVENT_AXI_WRITE) {
> + int val = event->attr.config1;
> +
> + writel(val, pmu->base + COUNTER_DPCR1);
> + }
> + }
> +
> + local64_set(&hwc->prev_count, 0);
> +
> + ddr_perf_event_enable(pmu, event->attr.config, counter, true);
>
> ddr_perf_event_enable() does what you'd expect, and uses hwc->idx to
> allocate a counter, but the code before it just always writes to DPCR1.
>
> What am I missing?
I double check with our IC design team. My previous answer is wrong.
Only one DPCR1. It will impact all event EVENT_AXI_READ and EVENT_AXI_WRITE.
Thank means just choose one config1 for event EVENT_AXI_READ or EVENT_AXI_WRITE
best regards
Frank Li
>
> Will
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
2019-04-08 19:10 ` Zhi Li
@ 2019-04-16 13:51 ` Will Deacon
-1 siblings, 0 replies; 43+ messages in thread
From: Will Deacon @ 2019-04-16 13:51 UTC (permalink / raw)
To: Zhi Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Mon, Apr 08, 2019 at 02:10:28PM -0500, Zhi Li wrote:
> On Fri, Apr 5, 2019 at 10:03 AM Will Deacon <will.deacon@arm.com> wrote:
> >
> > On Fri, Apr 05, 2019 at 09:58:38AM -0500, Zhi Li wrote:
> > > On Fri, Apr 5, 2019 at 9:38 AM Will Deacon <will.deacon@arm.com> wrote:
> > > > On Fri, Apr 05, 2019 at 09:34:38AM -0500, Zhi Li wrote:
> > > > > On Thu, Apr 4, 2019 at 6:17 AM Will Deacon <will.deacon@arm.com> wrote:
> > > > > > On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> > > > > > > Add ddr performance monitor support for iMX8QXP
> > > > > > >
> > > > > > > There are 4 counters for ddr perfomance events.
> > > > > > > counter 0 is dedicated for cycles.
> > > > > > > you choose any up to 3 no cycles events.
> > > > > > >
> > > > > > > for example:
> > > > > > >
> > > > > > > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > > > > > > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> > > > > >
> > > > > > Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
> > >
> > > Only imx845 have AXID filter capability now.
> > >
> > > > > > does the COUNTER_DPCR1 register work and what happens if I specify two
> > > > > > simultaneous events with different values in config1? I'm a little wary
> > >
> > > There are difference match register for each event.
> > > 1. Read event with config 1 A
> > > 2. Read event with config 1 B
> > >
> > > 1 will show read count with filter A
> > > 2 will show read count with filter B.
> >
> > Thanks, that makes sense, but I can't see how that corresponds to the code
> > in the patch:
> >
> > +static void ddr_perf_event_start(struct perf_event *event, int flags)
> > +{
> > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > + struct hw_perf_event *hwc = &event->hw;
> > + int counter = hwc->idx;
> > +
> > + if (pmu->flags & DDR_CAP_AXI_ID) {
> > + if (event->attr.config == EVENT_AXI_READ ||
> > + event->attr.config == EVENT_AXI_WRITE) {
> > + int val = event->attr.config1;
> > +
> > + writel(val, pmu->base + COUNTER_DPCR1);
> > + }
> > + }
> > +
> > + local64_set(&hwc->prev_count, 0);
> > +
> > + ddr_perf_event_enable(pmu, event->attr.config, counter, true);
> >
> > ddr_perf_event_enable() does what you'd expect, and uses hwc->idx to
> > allocate a counter, but the code before it just always writes to DPCR1.
> >
> > What am I missing?
>
> I double check with our IC design team. My previous answer is wrong.
> Only one DPCR1. It will impact all event EVENT_AXI_READ and EVENT_AXI_WRITE.
>
> Thank means just choose one config1 for event EVENT_AXI_READ or EVENT_AXI_WRITE
I think you need either to enforce that in the driver, or drop this
filtering capability for now. We shouldn't just tell userspace "oh yeah,
don't do that."
Given that your patch 3 (which updates a .dts file) doesn't appear to
support the filtering, I suggest just dropping that support initially so
that we can queue the rest of this driver for 5.2.
Will
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
@ 2019-04-16 13:51 ` Will Deacon
0 siblings, 0 replies; 43+ messages in thread
From: Will Deacon @ 2019-04-16 13:51 UTC (permalink / raw)
To: Zhi Li
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Mon, Apr 08, 2019 at 02:10:28PM -0500, Zhi Li wrote:
> On Fri, Apr 5, 2019 at 10:03 AM Will Deacon <will.deacon@arm.com> wrote:
> >
> > On Fri, Apr 05, 2019 at 09:58:38AM -0500, Zhi Li wrote:
> > > On Fri, Apr 5, 2019 at 9:38 AM Will Deacon <will.deacon@arm.com> wrote:
> > > > On Fri, Apr 05, 2019 at 09:34:38AM -0500, Zhi Li wrote:
> > > > > On Thu, Apr 4, 2019 at 6:17 AM Will Deacon <will.deacon@arm.com> wrote:
> > > > > > On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> > > > > > > Add ddr performance monitor support for iMX8QXP
> > > > > > >
> > > > > > > There are 4 counters for ddr perfomance events.
> > > > > > > counter 0 is dedicated for cycles.
> > > > > > > you choose any up to 3 no cycles events.
> > > > > > >
> > > > > > > for example:
> > > > > > >
> > > > > > > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > > > > > > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> > > > > >
> > > > > > Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
> > >
> > > Only imx845 have AXID filter capability now.
> > >
> > > > > > does the COUNTER_DPCR1 register work and what happens if I specify two
> > > > > > simultaneous events with different values in config1? I'm a little wary
> > >
> > > There are difference match register for each event.
> > > 1. Read event with config 1 A
> > > 2. Read event with config 1 B
> > >
> > > 1 will show read count with filter A
> > > 2 will show read count with filter B.
> >
> > Thanks, that makes sense, but I can't see how that corresponds to the code
> > in the patch:
> >
> > +static void ddr_perf_event_start(struct perf_event *event, int flags)
> > +{
> > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > + struct hw_perf_event *hwc = &event->hw;
> > + int counter = hwc->idx;
> > +
> > + if (pmu->flags & DDR_CAP_AXI_ID) {
> > + if (event->attr.config == EVENT_AXI_READ ||
> > + event->attr.config == EVENT_AXI_WRITE) {
> > + int val = event->attr.config1;
> > +
> > + writel(val, pmu->base + COUNTER_DPCR1);
> > + }
> > + }
> > +
> > + local64_set(&hwc->prev_count, 0);
> > +
> > + ddr_perf_event_enable(pmu, event->attr.config, counter, true);
> >
> > ddr_perf_event_enable() does what you'd expect, and uses hwc->idx to
> > allocate a counter, but the code before it just always writes to DPCR1.
> >
> > What am I missing?
>
> I double check with our IC design team. My previous answer is wrong.
> Only one DPCR1. It will impact all event EVENT_AXI_READ and EVENT_AXI_WRITE.
>
> Thank means just choose one config1 for event EVENT_AXI_READ or EVENT_AXI_WRITE
I think you need either to enforce that in the driver, or drop this
filtering capability for now. We shouldn't just tell userspace "oh yeah,
don't do that."
Given that your patch 3 (which updates a .dts file) doesn't appear to
support the filtering, I suggest just dropping that support initially so
that we can queue the rest of this driver for 5.2.
Will
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
2019-04-16 13:51 ` Will Deacon
@ 2019-04-16 20:41 ` Zhi Li
-1 siblings, 0 replies; 43+ messages in thread
From: Zhi Li @ 2019-04-16 20:41 UTC (permalink / raw)
To: Will Deacon
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Tue, Apr 16, 2019 at 8:51 AM Will Deacon <will.deacon@arm.com> wrote:
>
> On Mon, Apr 08, 2019 at 02:10:28PM -0500, Zhi Li wrote:
> > On Fri, Apr 5, 2019 at 10:03 AM Will Deacon <will.deacon@arm.com> wrote:
> > >
> > > On Fri, Apr 05, 2019 at 09:58:38AM -0500, Zhi Li wrote:
> > > > On Fri, Apr 5, 2019 at 9:38 AM Will Deacon <will.deacon@arm.com> wrote:
> > > > > On Fri, Apr 05, 2019 at 09:34:38AM -0500, Zhi Li wrote:
> > > > > > On Thu, Apr 4, 2019 at 6:17 AM Will Deacon <will.deacon@arm.com> wrote:
> > > > > > > On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> > > > > > > > Add ddr performance monitor support for iMX8QXP
> > > > > > > >
> > > > > > > > There are 4 counters for ddr perfomance events.
> > > > > > > > counter 0 is dedicated for cycles.
> > > > > > > > you choose any up to 3 no cycles events.
> > > > > > > >
> > > > > > > > for example:
> > > > > > > >
> > > > > > > > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > > > > > > > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> > > > > > >
> > > > > > > Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
> > > >
> > > > Only imx845 have AXID filter capability now.
> > > >
> > > > > > > does the COUNTER_DPCR1 register work and what happens if I specify two
> > > > > > > simultaneous events with different values in config1? I'm a little wary
> > > >
> > > > There are difference match register for each event.
> > > > 1. Read event with config 1 A
> > > > 2. Read event with config 1 B
> > > >
> > > > 1 will show read count with filter A
> > > > 2 will show read count with filter B.
> > >
> > > Thanks, that makes sense, but I can't see how that corresponds to the code
> > > in the patch:
> > >
> > > +static void ddr_perf_event_start(struct perf_event *event, int flags)
> > > +{
> > > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > > + struct hw_perf_event *hwc = &event->hw;
> > > + int counter = hwc->idx;
> > > +
> > > + if (pmu->flags & DDR_CAP_AXI_ID) {
> > > + if (event->attr.config == EVENT_AXI_READ ||
> > > + event->attr.config == EVENT_AXI_WRITE) {
> > > + int val = event->attr.config1;
> > > +
> > > + writel(val, pmu->base + COUNTER_DPCR1);
> > > + }
> > > + }
> > > +
> > > + local64_set(&hwc->prev_count, 0);
> > > +
> > > + ddr_perf_event_enable(pmu, event->attr.config, counter, true);
> > >
> > > ddr_perf_event_enable() does what you'd expect, and uses hwc->idx to
> > > allocate a counter, but the code before it just always writes to DPCR1.
> > >
> > > What am I missing?
> >
> > I double check with our IC design team. My previous answer is wrong.
> > Only one DPCR1. It will impact all event EVENT_AXI_READ and EVENT_AXI_WRITE.
> >
> > Thank means just choose one config1 for event EVENT_AXI_READ or EVENT_AXI_WRITE
>
> I think you need either to enforce that in the driver, or drop this
> filtering capability for now. We shouldn't just tell userspace "oh yeah,
> don't do that."
>
> Given that your patch 3 (which updates a .dts file) doesn't appear to
> support the filtering, I suggest just dropping that support initially so
> that we can queue the rest of this driver for 5.2.
V5(remove AXI ID) sent out.
best regards
Frank Li
>
> Will
^ permalink raw reply [flat|nested] 43+ messages in thread
* Re: [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support
@ 2019-04-16 20:41 ` Zhi Li
0 siblings, 0 replies; 43+ messages in thread
From: Zhi Li @ 2019-04-16 20:41 UTC (permalink / raw)
To: Will Deacon
Cc: mark.rutland, Aisheng Dong, devicetree, festevam, s.hauer,
Frank Li, robh+dt, dl-linux-imx, kernel, shawnguo,
linux-arm-kernel
On Tue, Apr 16, 2019 at 8:51 AM Will Deacon <will.deacon@arm.com> wrote:
>
> On Mon, Apr 08, 2019 at 02:10:28PM -0500, Zhi Li wrote:
> > On Fri, Apr 5, 2019 at 10:03 AM Will Deacon <will.deacon@arm.com> wrote:
> > >
> > > On Fri, Apr 05, 2019 at 09:58:38AM -0500, Zhi Li wrote:
> > > > On Fri, Apr 5, 2019 at 9:38 AM Will Deacon <will.deacon@arm.com> wrote:
> > > > > On Fri, Apr 05, 2019 at 09:34:38AM -0500, Zhi Li wrote:
> > > > > > On Thu, Apr 4, 2019 at 6:17 AM Will Deacon <will.deacon@arm.com> wrote:
> > > > > > > On Fri, Feb 15, 2019 at 06:03:11PM +0000, Frank Li wrote:
> > > > > > > > Add ddr performance monitor support for iMX8QXP
> > > > > > > >
> > > > > > > > There are 4 counters for ddr perfomance events.
> > > > > > > > counter 0 is dedicated for cycles.
> > > > > > > > you choose any up to 3 no cycles events.
> > > > > > > >
> > > > > > > > for example:
> > > > > > > >
> > > > > > > > perf stat -a -e ddr0/read-access/,ddr0/write-access/,ddr0/precharge/ ls
> > > > > > > > perf stat -a -e ddr0/cycles/,ddr0/read-access/,ddr0/write-access/ ls
> > > > > > >
> > > > > > > Could you elaborate a bit on DDR_CAP_AXI_ID, please? Specifically, how
> > > >
> > > > Only imx845 have AXID filter capability now.
> > > >
> > > > > > > does the COUNTER_DPCR1 register work and what happens if I specify two
> > > > > > > simultaneous events with different values in config1? I'm a little wary
> > > >
> > > > There are difference match register for each event.
> > > > 1. Read event with config 1 A
> > > > 2. Read event with config 1 B
> > > >
> > > > 1 will show read count with filter A
> > > > 2 will show read count with filter B.
> > >
> > > Thanks, that makes sense, but I can't see how that corresponds to the code
> > > in the patch:
> > >
> > > +static void ddr_perf_event_start(struct perf_event *event, int flags)
> > > +{
> > > + struct ddr_pmu *pmu = to_ddr_pmu(event->pmu);
> > > + struct hw_perf_event *hwc = &event->hw;
> > > + int counter = hwc->idx;
> > > +
> > > + if (pmu->flags & DDR_CAP_AXI_ID) {
> > > + if (event->attr.config == EVENT_AXI_READ ||
> > > + event->attr.config == EVENT_AXI_WRITE) {
> > > + int val = event->attr.config1;
> > > +
> > > + writel(val, pmu->base + COUNTER_DPCR1);
> > > + }
> > > + }
> > > +
> > > + local64_set(&hwc->prev_count, 0);
> > > +
> > > + ddr_perf_event_enable(pmu, event->attr.config, counter, true);
> > >
> > > ddr_perf_event_enable() does what you'd expect, and uses hwc->idx to
> > > allocate a counter, but the code before it just always writes to DPCR1.
> > >
> > > What am I missing?
> >
> > I double check with our IC design team. My previous answer is wrong.
> > Only one DPCR1. It will impact all event EVENT_AXI_READ and EVENT_AXI_WRITE.
> >
> > Thank means just choose one config1 for event EVENT_AXI_READ or EVENT_AXI_WRITE
>
> I think you need either to enforce that in the driver, or drop this
> filtering capability for now. We shouldn't just tell userspace "oh yeah,
> don't do that."
>
> Given that your patch 3 (which updates a .dts file) doesn't appear to
> support the filtering, I suggest just dropping that support initially so
> that we can queue the rest of this driver for 5.2.
V5(remove AXI ID) sent out.
best regards
Frank Li
>
> Will
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 43+ messages in thread
end of thread, other threads:[~2019-04-16 20:41 UTC | newest]
Thread overview: 43+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-02-15 18:03 [PATCH V4 1/4] dt-bindings: perf: imx8-ddr: add imx8qxp ddr performance monitor Frank Li
2019-02-15 18:03 ` Frank Li
2019-02-15 18:03 ` [PATCH V4 2/4] drivers/perf: imx_ddr: Add ddr performance counter support Frank Li
2019-02-15 18:03 ` Frank Li
2019-02-22 19:45 ` Zhi Li
2019-02-22 19:45 ` Zhi Li
2019-02-28 21:39 ` Zhi Li
2019-02-28 21:39 ` Zhi Li
2019-03-12 21:49 ` Zhi Li
2019-03-12 21:49 ` Zhi Li
2019-04-04 11:17 ` Will Deacon
2019-04-04 11:17 ` Will Deacon
2019-04-05 14:34 ` Zhi Li
2019-04-05 14:34 ` Zhi Li
2019-04-05 14:38 ` Will Deacon
2019-04-05 14:38 ` Will Deacon
2019-04-05 14:58 ` Zhi Li
2019-04-05 14:58 ` Zhi Li
2019-04-05 15:03 ` Will Deacon
2019-04-05 15:03 ` Will Deacon
2019-04-08 19:10 ` Zhi Li
2019-04-08 19:10 ` Zhi Li
2019-04-16 13:51 ` Will Deacon
2019-04-16 13:51 ` Will Deacon
2019-04-16 20:41 ` Zhi Li
2019-04-16 20:41 ` Zhi Li
2019-02-15 18:03 ` [PATCH V4 3/4] arm64: dts: imx8qxp: added ddr performance monitor nodes Frank Li
2019-02-15 18:03 ` Frank Li
2019-02-18 17:38 ` Fabio Estevam
2019-02-18 17:38 ` Fabio Estevam
2019-02-15 18:03 ` [PATCH V4 4/4] MAINTAINERS: Added imx DDR performonitor driver maintainer information Frank Li
2019-02-15 18:03 ` Frank Li
2019-02-15 20:01 ` Uwe Kleine-König
2019-02-15 20:01 ` Uwe Kleine-König
2019-02-15 21:13 ` Zhi Li
2019-02-15 21:44 ` Uwe Kleine-König
2019-02-15 21:44 ` Uwe Kleine-König
2019-02-18 1:22 ` Shawn Guo
2019-02-18 1:22 ` Shawn Guo
2019-02-20 6:28 ` Aisheng Dong
2019-02-20 6:28 ` Aisheng Dong
2019-02-18 16:56 ` [PATCH V4 1/4] dt-bindings: perf: imx8-ddr: add imx8qxp ddr performance monitor Rob Herring
2019-02-18 16:56 ` Rob Herring
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.