[PATCH v9 2/2] drm/xe/pmu: Enable PMU interface

Umesh Nerlige Ramappa umesh.nerlige.ramappa at intel.com
Fri Jun 28 16:52:36 UTC 2024


On Fri, Jun 28, 2024 at 10:55:06AM -0500, Lucas De Marchi wrote:
>On Thu, Jun 20, 2024 at 12:52:05PM GMT, Umesh Nerlige Ramappa wrote:
>>On Thu, Jun 13, 2024 at 03:34:11PM +0530, Riana Tauro wrote:
>>>From: Aravind Iddamsetty <aravind.iddamsetty at linux.intel.com>
>>>
>>>There are a set of engine group busyness counters provided by HW which are
>>>perfect fit to be exposed via PMU perf events.
>>>
>>>BSPEC: 46559, 46560, 46722, 46729, 52071, 71028
>>>
>>>events can be listed using:
>>>perf list
>>>xe_0000_03_00.0/any-engine-group-busy-gt0/         [Kernel PMU event]
>>>xe_0000_03_00.0/copy-group-busy-gt0/               [Kernel PMU event]
>>>xe_0000_03_00.0/media-group-busy-gt0/              [Kernel PMU event]
>>>xe_0000_03_00.0/render-group-busy-gt0/             [Kernel PMU event]
>>>
>>>and can be read using:
>>>
>>>perf stat -e "xe_0000_8c_00.0/render-group-busy-gt0/" -I 1000
>>>         time        counts unit events
>>>   1.001139062            0 ns  xe_0000_8c_00.0/render-group-busy-gt0/
>>>   2.003294678            0 ns  xe_0000_8c_00.0/render-group-busy-gt0/
>>>   3.005199582            0 ns  xe_0000_8c_00.0/render-group-busy-gt0/
>>>   4.007076497            0 ns  xe_0000_8c_00.0/render-group-busy-gt0/
>>>   5.008553068            0 ns  xe_0000_8c_00.0/render-group-busy-gt0/
>>>   6.010531563        43520 ns  xe_0000_8c_00.0/render-group-busy-gt0/
>>>   7.012468029        44800 ns  xe_0000_8c_00.0/render-group-busy-gt0/
>>>   8.013463515            0 ns  xe_0000_8c_00.0/render-group-busy-gt0/
>>>   9.015300183            0 ns  xe_0000_8c_00.0/render-group-busy-gt0/
>>>  10.017233010            0 ns  xe_0000_8c_00.0/render-group-busy-gt0/
>>>  10.971934120            0 ns  xe_0000_8c_00.0/render-group-busy-gt0/
>>>
>>>The pmu base implementation is taken from i915.
>>>
>>>v2:
>>>Store last known value when device is awake return that while the GT is
>>>suspended and then update the driver copy when read during awake.
>>>
>>>v3:
>>>1. drop init_samples, as storing counters before going to suspend should
>>>be sufficient.
>>>2. ported the "drm/i915/pmu: Make PMU sample array two-dimensional" and
>>>dropped helpers to store and read samples.
>>>3. use xe_device_mem_access_get_if_ongoing to check if device is active
>>>before reading the OA registers.
>>>4. dropped format attr as no longer needed
>>>5. introduce xe_pmu_suspend to call engine_group_busyness_store
>>>6. few other nits.
>>>
>>>v4: minor nits.
>>>
>>>v5: take forcewake when accessing the OAG registers
>>>
>>>v6:
>>>1. drop engine_busyness_sample_type
>>>2. update UAPI documentation
>>>
>>>v7:
>>>1. update UAPI documentation
>>>2. drop MEDIA_GT specific change for media busyness counter.
>>>
>>>v8:
>>>1. rebase
>>>2. replace mem_access_if_ongoing with xe_pm_runtime_get_if_active
>>>3. remove interrupts pmu event
>>>
>>>v9: replace drmm_add_action_or_reset with devm (Matthew Auld)
>>>
>>>Co-developed-by: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
>>>Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
>>>Co-developed-by: Bommu Krishnaiah <krishnaiah.bommu at intel.com>
>>>Signed-off-by: Bommu Krishnaiah <krishnaiah.bommu at intel.com>
>>>Signed-off-by: Aravind Iddamsetty <aravind.iddamsetty at linux.intel.com>
>>>Reviewed-by: Ashutosh Dixit <ashutosh.dixit at intel.com>
>>>Signed-off-by: Riana Tauro <riana.tauro at intel.com>
>>>---
>>>drivers/gpu/drm/xe/Makefile          |   2 +
>>>drivers/gpu/drm/xe/regs/xe_gt_regs.h |   5 +
>>>drivers/gpu/drm/xe/xe_device.c       |   2 +
>>>drivers/gpu/drm/xe/xe_device_types.h |   4 +
>>>drivers/gpu/drm/xe/xe_gt.c           |   2 +
>>>drivers/gpu/drm/xe/xe_module.c       |   5 +
>>>drivers/gpu/drm/xe/xe_pmu.c          | 631 +++++++++++++++++++++++++++
>>>drivers/gpu/drm/xe/xe_pmu.h          |  26 ++
>>>drivers/gpu/drm/xe/xe_pmu_types.h    |  67 +++
>>>include/uapi/drm/xe_drm.h            |  39 ++
>>>10 files changed, 783 insertions(+)
>>>create mode 100644 drivers/gpu/drm/xe/xe_pmu.c
>>>create mode 100644 drivers/gpu/drm/xe/xe_pmu.h
>>>create mode 100644 drivers/gpu/drm/xe/xe_pmu_types.h
>>>
>>>diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
>>>index cbf961b90237..83bf1e07669b 100644
>>>--- a/drivers/gpu/drm/xe/Makefile
>>>+++ b/drivers/gpu/drm/xe/Makefile
>>>@@ -278,6 +278,8 @@ xe-$(CONFIG_DRM_XE_DISPLAY) += \
>>>	i915-display/skl_universal_plane.o \
>>>	i915-display/skl_watermark.o
>>>
>>>+xe-$(CONFIG_PERF_EVENTS) += xe_pmu.o
>>>+
>>>ifeq ($(CONFIG_ACPI),y)
>>>	xe-$(CONFIG_DRM_XE_DISPLAY) += \
>>>		i915-display/intel_acpi.o \
>>>diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
>>>index 47c26c37608d..22821dcd4e1b 100644
>>>--- a/drivers/gpu/drm/xe/regs/xe_gt_regs.h
>>>+++ b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
>>>@@ -390,6 +390,11 @@
>>>#define   INVALIDATION_BROADCAST_MODE_DIS	REG_BIT(12)
>>>#define   GLOBAL_INVALIDATION_MODE		REG_BIT(2)
>>>
>>>+#define XE_OAG_RC0_ANY_ENGINE_BUSY_FREE		XE_REG(0xdb80)
>>>+#define XE_OAG_ANY_MEDIA_FF_BUSY_FREE		XE_REG(0xdba0)
>>>+#define XE_OAG_BLT_BUSY_FREE			XE_REG(0xdbbc)
>>>+#define XE_OAG_RENDER_BUSY_FREE			XE_REG(0xdbdc)
>>>+
>>>#define HALF_SLICE_CHICKEN5			XE_REG_MCR(0xe188, XE_REG_OPTION_MASKED)
>>>#define   DISABLE_SAMPLE_G_PERFORMANCE		REG_BIT(0)
>>>
>>>diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
>>>index 64691a56d59c..bb00c8c9ec9b 100644
>>>--- a/drivers/gpu/drm/xe/xe_device.c
>>>+++ b/drivers/gpu/drm/xe/xe_device.c
>>>@@ -668,6 +668,8 @@ int xe_device_probe(struct xe_device *xe)
>>>
>>>	xe_hwmon_register(xe);
>>>
>>>+	xe_pmu_register(&xe->pmu);
>>>+
>>>	return devm_add_action_or_reset(xe->drm.dev, xe_device_sanitize, xe);
>>>
>>>err_fini_display:
>>>diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
>>>index 52bc461171d5..a5dba7325cf1 100644
>>>--- a/drivers/gpu/drm/xe/xe_device_types.h
>>>+++ b/drivers/gpu/drm/xe/xe_device_types.h
>>>@@ -18,6 +18,7 @@
>>>#include "xe_lmtt_types.h"
>>>#include "xe_memirq_types.h"
>>>#include "xe_platform_types.h"
>>>+#include "xe_pmu.h"
>>>#include "xe_pt_types.h"
>>>#include "xe_sriov_types.h"
>>>#include "xe_step_types.h"
>>>@@ -473,6 +474,9 @@ struct xe_device {
>>>		int mode;
>>>	} wedged;
>>>
>>>+	/** @pmu: performance monitoring unit */
>>>+	struct xe_pmu pmu;
>>>+
>>>	/* private: */
>>>
>>>#if IS_ENABLED(CONFIG_DRM_XE_DISPLAY)
>>>diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
>>>index 57d84751e160..477d0ae5f230 100644
>>>--- a/drivers/gpu/drm/xe/xe_gt.c
>>>+++ b/drivers/gpu/drm/xe/xe_gt.c
>>>@@ -782,6 +782,8 @@ int xe_gt_suspend(struct xe_gt *gt)
>>>	if (err)
>>>		goto err_msg;
>>>
>>>+	xe_pmu_suspend(gt);
>>>+
>>>	err = xe_uc_suspend(&gt->uc);
>>>	if (err)
>>>		goto err_force_wake;
>>>diff --git a/drivers/gpu/drm/xe/xe_module.c b/drivers/gpu/drm/xe/xe_module.c
>>>index 3edeb30d5ccb..26f814f97fc2 100644
>>>--- a/drivers/gpu/drm/xe/xe_module.c
>>>+++ b/drivers/gpu/drm/xe/xe_module.c
>>>@@ -11,6 +11,7 @@
>>>#include "xe_drv.h"
>>>#include "xe_hw_fence.h"
>>>#include "xe_pci.h"
>>>+#include "xe_pmu.h"
>>>#include "xe_sched_job.h"
>>>
>>>struct xe_modparam xe_modparam = {
>>>@@ -74,6 +75,10 @@ static const struct init_funcs init_funcs[] = {
>>>		.init = xe_sched_job_module_init,
>>>		.exit = xe_sched_job_module_exit,
>>>	},
>>>+	{
>>>+		.init = xe_pmu_init,
>>>+		.exit = xe_pmu_exit,
>>>+	},
>>>	{
>>>		.init = xe_register_pci_driver,
>>>		.exit = xe_unregister_pci_driver,
>>>diff --git a/drivers/gpu/drm/xe/xe_pmu.c b/drivers/gpu/drm/xe/xe_pmu.c
>>>new file mode 100644
>>>index 000000000000..64960a358af2
>>>--- /dev/null
>>>+++ b/drivers/gpu/drm/xe/xe_pmu.c
>>>@@ -0,0 +1,631 @@
>>>+// SPDX-License-Identifier: MIT
>>>+/*
>>>+ * Copyright © 2024 Intel Corporation
>>>+ */
>>>+
>>>+#include <drm/drm_drv.h>
>>>+#include <drm/drm_managed.h>
>>>+#include <drm/xe_drm.h>
>>>+
>>>+#include "regs/xe_gt_regs.h"
>>>+#include "xe_device.h"
>>>+#include "xe_force_wake.h"
>>>+#include "xe_gt_clock.h"
>>>+#include "xe_mmio.h"
>>>+#include "xe_macros.h"
>>>+#include "xe_pm.h"
>>>+
>>>+static cpumask_t xe_pmu_cpumask;
>>>+static unsigned int xe_pmu_target_cpu = -1;
>>>+
>>>+static unsigned int config_gt_id(const u64 config)
>>>+{
>>>+	return config >> __XE_PMU_GT_SHIFT;
>>>+}
>>>+
>>>+static u64 config_counter(const u64 config)
>>>+{
>>>+	return config & ~(~0ULL << __XE_PMU_GT_SHIFT);
>>>+}
>>>+
>>>+static void xe_pmu_event_destroy(struct perf_event *event)
>>>+{
>>>+	struct xe_device *xe =
>>>+		container_of(event->pmu, typeof(*xe), pmu.base);
>>>+
>>>+	drm_WARN_ON(&xe->drm, event->parent);
>>>+
>>>+	drm_dev_put(&xe->drm);
>>>+}
>>>+
>>>+static u64 __engine_group_busyness_read(struct xe_gt *gt, int sample_type)
>>>+{
>>>+	u64 val;
>>>+
>>>+	switch (sample_type) {
>>>+	case __XE_SAMPLE_RENDER_GROUP_BUSY:
>>>+		val = xe_mmio_read32(gt, XE_OAG_RENDER_BUSY_FREE);
>>>+		break;
>>>+	case __XE_SAMPLE_COPY_GROUP_BUSY:
>>>+		val = xe_mmio_read32(gt, XE_OAG_BLT_BUSY_FREE);
>>>+		break;
>>>+	case __XE_SAMPLE_MEDIA_GROUP_BUSY:
>>>+		val = xe_mmio_read32(gt, XE_OAG_ANY_MEDIA_FF_BUSY_FREE);
>>>+		break;
>>>+	case __XE_SAMPLE_ANY_ENGINE_GROUP_BUSY:
>>>+		val = xe_mmio_read32(gt, XE_OAG_RC0_ANY_ENGINE_BUSY_FREE);
>>>+		break;
>>>+	default:
>>>+		drm_warn(&gt->tile->xe->drm, "unknown pmu event\n");
>>>+	}
>>>+
>>>+	return xe_gt_clock_cycles_to_ns(gt, val * 16);
>>>+}
>>>+
>>>+static u64 engine_group_busyness_read(struct xe_gt *gt, u64 config)
>>>+{
>>>+	int sample_type = config_counter(config);
>>>+	const unsigned int gt_id = gt->info.id;
>>>+	struct xe_device *xe = gt->tile->xe;
>>>+	struct xe_pmu *pmu = &xe->pmu;
>>>+	unsigned long flags;
>>>+	bool device_awake;
>>>+	u64 val;
>>>+
>>>+	device_awake = xe_pm_runtime_get_if_active(xe);
>>>+	if (device_awake) {
>>>+		XE_WARN_ON(xe_force_wake_get(gt_to_fw(gt), XE_FW_GT));
>>>+		val = __engine_group_busyness_read(gt, sample_type);
>>>+		XE_WARN_ON(xe_force_wake_put(gt_to_fw(gt), XE_FW_GT));
>>>+		xe_pm_runtime_put(xe);
>>>+	}
>>>+
>>>+	spin_lock_irqsave(&pmu->lock, flags);
>>>+
>>>+	if (device_awake)
>>>+		pmu->sample[gt_id][sample_type] = val;
>>>+	else
>>>+		val = pmu->sample[gt_id][sample_type];
>>>+
>>>+	spin_unlock_irqrestore(&pmu->lock, flags);
>>>+
>>>+	return val;
>>>+}
>>>+
>>>+static void engine_group_busyness_store(struct xe_gt *gt)
>>>+{
>>>+	struct xe_pmu *pmu = &gt->tile->xe->pmu;
>>>+	unsigned int gt_id = gt->info.id;
>>>+	unsigned long flags;
>>>+	int i;
>>>+
>>>+	spin_lock_irqsave(&pmu->lock, flags);
>>>+
>>>+	for (i = __XE_SAMPLE_RENDER_GROUP_BUSY; i <= __XE_SAMPLE_ANY_ENGINE_GROUP_BUSY; i++)
>>>+		pmu->sample[gt_id][i] = __engine_group_busyness_read(gt, i);
>>>+
>>>+	spin_unlock_irqrestore(&pmu->lock, flags);
>>>+}
>>>+
>>>+static int
>>>+config_status(struct xe_device *xe, u64 config)
>>>+{
>>>+	unsigned int gt_id = config_gt_id(config);
>>>+	struct xe_gt *gt = xe_device_get_gt(xe, gt_id);
>>>+
>>>+	if (gt_id >= XE_PMU_MAX_GT)
>>>+		return -ENOENT;
>>>+
>>>+	switch (config_counter(config)) {
>>>+	case XE_PMU_RENDER_GROUP_BUSY(0):
>>>+	case XE_PMU_COPY_GROUP_BUSY(0):
>>>+	case XE_PMU_ANY_ENGINE_GROUP_BUSY(0):
>>>+		if (gt->info.type == XE_GT_TYPE_MEDIA)
>>>+			return -ENOENT;
>>>+		break;
>>>+	case XE_PMU_MEDIA_GROUP_BUSY(0):
>>>+		if (!(gt->info.engine_mask & (BIT(XE_HW_ENGINE_VCS0) | BIT(XE_HW_ENGINE_VECS0))))
>>>+			return -ENOENT;
>>>+		break;
>>>+	default:
>>>+		return -ENOENT;
>>>+	}
>>>+
>>>+	return 0;
>>>+}
>>>+
>>>+static int xe_pmu_event_init(struct perf_event *event)
>>>+{
>>>+	struct xe_device *xe =
>>>+		container_of(event->pmu, typeof(*xe), pmu.base);
>>>+	struct xe_pmu *pmu = &xe->pmu;
>>>+	int ret;
>>>+
>>>+	if (pmu->closed)
>>>+		return -ENODEV;
>>>+
>>>+	if (event->attr.type != event->pmu->type)
>>>+		return -ENOENT;
>>>+
>>>+	/* unsupported modes and filters */
>>>+	if (event->attr.sample_period) /* no sampling */
>>>+		return -EINVAL;
>>>+
>>>+	if (has_branch_stack(event))
>>>+		return -EOPNOTSUPP;
>>>+
>>>+	if (event->cpu < 0)
>>>+		return -EINVAL;
>>>+
>>>+	/* only allow running on one cpu at a time */
>>>+	if (!cpumask_test_cpu(event->cpu, &xe_pmu_cpumask))
>>>+		return -EINVAL;
>>>+
>>>+	ret = config_status(xe, event->attr.config);
>>>+	if (ret)
>>>+		return ret;
>>>+
>>>+	if (!event->parent) {
>>>+		drm_dev_get(&xe->drm);
>>>+		event->destroy = xe_pmu_event_destroy;
>>>+	}
>>>+
>>>+	return 0;
>>>+}
>>>+
>>>+static u64 __xe_pmu_event_read(struct perf_event *event)
>>>+{
>>>+	struct xe_device *xe =
>>>+		container_of(event->pmu, typeof(*xe), pmu.base);
>>>+	const unsigned int gt_id = config_gt_id(event->attr.config);
>>>+	const u64 config = event->attr.config;
>>>+	struct xe_gt *gt = xe_device_get_gt(xe, gt_id);
>>>+	u64 val;
>>>+
>>>+	switch (config_counter(config)) {
>>>+	case XE_PMU_RENDER_GROUP_BUSY(0):
>>>+	case XE_PMU_COPY_GROUP_BUSY(0):
>>>+	case XE_PMU_ANY_ENGINE_GROUP_BUSY(0):
>>>+	case XE_PMU_MEDIA_GROUP_BUSY(0):
>>>+		val = engine_group_busyness_read(gt, config);
>>>+		break;
>>>+	default:
>>>+		drm_warn(&gt->tile->xe->drm, "unknown pmu event\n");
>>>+	}
>>>+
>>>+	return val;
>>>+}
>>>+
>>>+static void xe_pmu_event_read(struct perf_event *event)
>>>+{
>>>+	struct xe_device *xe =
>>>+		container_of(event->pmu, typeof(*xe), pmu.base);
>>>+	struct hw_perf_event *hwc = &event->hw;
>>>+	struct xe_pmu *pmu = &xe->pmu;
>>>+	u64 prev, new;
>>>+
>>>+	if (pmu->closed) {
>>>+		event->hw.state = PERF_HES_STOPPED;
>>>+		return;
>>>+	}
>>>+again:
>>>+	prev = local64_read(&hwc->prev_count);
>>>+	new = __xe_pmu_event_read(event);
>>>+
>>>+	if (local64_cmpxchg(&hwc->prev_count, prev, new) != prev)
>>>+		goto again;
>>>+
>>>+	local64_add(new - prev, &event->count);
>>>+}
>>>+
>>>+static void xe_pmu_enable(struct perf_event *event)
>>>+{
>>>+	/*
>>>+	 * Store the current counter value so we can report the correct delta
>>>+	 * for all listeners. Even when the event was already enabled and has
>>>+	 * an existing non-zero value.
>>>+	 */
>>>+	local64_set(&event->hw.prev_count, __xe_pmu_event_read(event));
>>>+}
>>>+
>>>+static void xe_pmu_event_start(struct perf_event *event, int flags)
>>>+{
>>>+	struct xe_device *xe =
>>>+		container_of(event->pmu, typeof(*xe), pmu.base);
>>>+	struct xe_pmu *pmu = &xe->pmu;
>>>+
>>>+	if (pmu->closed)
>>>+		return;
>>>+
>>>+	xe_pmu_enable(event);
>>>+	event->hw.state = 0;
>>>+}
>>>+
>>>+static void xe_pmu_event_stop(struct perf_event *event, int flags)
>>>+{
>>>+	if (flags & PERF_EF_UPDATE)
>>>+		xe_pmu_event_read(event);
>>>+
>>>+	event->hw.state = PERF_HES_STOPPED;
>>>+}
>>>+
>>>+static int xe_pmu_event_add(struct perf_event *event, int flags)
>>>+{
>>>+	struct xe_device *xe =
>>>+		container_of(event->pmu, typeof(*xe), pmu.base);
>>>+	struct xe_pmu *pmu = &xe->pmu;
>>>+
>>>+	if (pmu->closed)
>>>+		return -ENODEV;
>>>+
>>>+	if (flags & PERF_EF_START)
>>>+		xe_pmu_event_start(event, flags);
>>>+
>>>+	return 0;
>>>+}
>>>+
>>>+static void xe_pmu_event_del(struct perf_event *event, int flags)
>>>+{
>>>+	xe_pmu_event_stop(event, PERF_EF_UPDATE);
>>>+}
>>>+
>>>+static int xe_pmu_event_event_idx(struct perf_event *event)
>>>+{
>>>+	return 0;
>>>+}
>>>+
>>>+struct xe_ext_attribute {
>>>+	struct device_attribute attr;
>>>+	unsigned long val;
>>>+};
>>>+
>>>+static ssize_t xe_pmu_event_show(struct device *dev,
>>>+				 struct device_attribute *attr, char *buf)
>>>+{
>>>+	struct xe_ext_attribute *eattr;
>>>+
>>>+	eattr = container_of(attr, struct xe_ext_attribute, attr);
>>>+	return sprintf(buf, "config=0x%lx\n", eattr->val);
>>>+}
>>>+
>>>+static ssize_t cpumask_show(struct device *dev,
>>>+			    struct device_attribute *attr, char *buf)
>>>+{
>>>+	return cpumap_print_to_pagebuf(true, buf, &xe_pmu_cpumask);
>>>+}
>>>+
>>>+static DEVICE_ATTR_RO(cpumask);
>>>+
>>>+static struct attribute *xe_cpumask_attrs[] = {
>>>+	&dev_attr_cpumask.attr,
>>>+	NULL,
>>>+};
>>>+
>>>+static const struct attribute_group xe_pmu_cpumask_attr_group = {
>>>+	.attrs = xe_cpumask_attrs,
>>>+};
>>>+
>>>+#define __event(__counter, __name, __unit) \
>>>+{ \
>>>+	.counter = (__counter), \
>>>+	.name = (__name), \
>>>+	.unit = (__unit), \
>>>+}
>>>+
>>>+static struct xe_ext_attribute *
>>>+add_xe_attr(struct xe_ext_attribute *attr, const char *name, u64 config)
>>>+{
>>>+	sysfs_attr_init(&attr->attr.attr);
>>>+	attr->attr.attr.name = name;
>>>+	attr->attr.attr.mode = 0444;
>>>+	attr->attr.show = xe_pmu_event_show;
>>>+	attr->val = config;
>>>+
>>>+	return ++attr;
>>>+}
>>>+
>>>+static struct perf_pmu_events_attr *
>>>+add_pmu_attr(struct perf_pmu_events_attr *attr, const char *name,
>>>+	     const char *str)
>>>+{
>>>+	sysfs_attr_init(&attr->attr.attr);
>>>+	attr->attr.attr.name = name;
>>>+	attr->attr.attr.mode = 0444;
>>>+	attr->attr.show = perf_event_sysfs_show;
>>>+	attr->event_str = str;
>>>+
>>>+	return ++attr;
>>>+}
>>>+
>>>+static struct attribute **
>>>+create_event_attributes(struct xe_pmu *pmu)
>>>+{
>>>+	struct xe_device *xe = container_of(pmu, typeof(*xe), pmu);
>>>+	static const struct {
>>>+		unsigned int counter;
>>>+		const char *name;
>>>+		const char *unit;
>>>+	} events[] = {
>>>+		__event(0, "render-group-busy", "ns"),
>>>+		__event(1, "copy-group-busy", "ns"),
>>>+		__event(2, "media-group-busy", "ns"),
>>>+		__event(3, "any-engine-group-busy", "ns"),
>>>+	};
>>>+
>>>+	struct perf_pmu_events_attr *pmu_attr = NULL, *pmu_iter;
>>>+	struct xe_ext_attribute *xe_attr = NULL, *xe_iter;
>>>+	struct attribute **attr = NULL, **attr_iter;
>>>+	unsigned int count = 0;
>>>+	unsigned int i, j;
>>>+	struct xe_gt *gt;
>>>+
>>>+	/* Count how many counters we will be exposing. */
>>>+	for_each_gt(gt, xe, j) {
>>>+		for (i = 0; i < ARRAY_SIZE(events); i++) {
>>>+			u64 config = ___XE_PMU_OTHER(j, events[i].counter);
>>>+
>>>+			if (!config_status(xe, config))
>>>+				count++;
>>>+		}
>>>+	}
>>>+
>>>+	/* Allocate attribute objects and table. */
>>>+	xe_attr = kcalloc(count, sizeof(*xe_attr), GFP_KERNEL);
>>>+	if (!xe_attr)
>>>+		goto err_alloc;
>>>+
>>>+	pmu_attr = kcalloc(count, sizeof(*pmu_attr), GFP_KERNEL);
>>>+	if (!pmu_attr)
>>>+		goto err_alloc;
>>>+
>>>+	/* Max one pointer of each attribute type plus a termination entry. */
>>>+	attr = kcalloc(count * 2 + 1, sizeof(*attr), GFP_KERNEL);
>>>+	if (!attr)
>>>+		goto err_alloc;
>>>+
>>>+	xe_iter = xe_attr;
>>>+	pmu_iter = pmu_attr;
>>>+	attr_iter = attr;
>>>+
>>>+	for_each_gt(gt, xe, j) {
>>>+		for (i = 0; i < ARRAY_SIZE(events); i++) {
>>>+			u64 config = ___XE_PMU_OTHER(j, events[i].counter);
>>>+			char *str;
>>>+
>>>+			if (config_status(xe, config))
>>>+				continue;
>>>+
>>>+			str = kasprintf(GFP_KERNEL, "%s-gt%u",
>>>+					events[i].name, j);
>>>+			if (!str)
>>>+				goto err;
>>>+
>>>+			*attr_iter++ = &xe_iter->attr.attr;
>>>+			xe_iter = add_xe_attr(xe_iter, str, config);
>>>+
>>>+			if (events[i].unit) {
>>>+				str = kasprintf(GFP_KERNEL, "%s-gt%u.unit",
>>>+						events[i].name, j);
>>>+				if (!str)
>>>+					goto err;
>>>+
>>>+				*attr_iter++ = &pmu_iter->attr.attr;
>>>+				pmu_iter = add_pmu_attr(pmu_iter, str,
>>>+							events[i].unit);
>>>+			}
>>>+		}
>>>+	}
>>>+
>>>+	pmu->xe_attr = xe_attr;
>>>+	pmu->pmu_attr = pmu_attr;
>>>+
>>>+	return attr;
>>>+
>>>+err:
>>>+	for (attr_iter = attr; *attr_iter; attr_iter++)
>>>+		kfree((*attr_iter)->name);
>>>+
>>>+err_alloc:
>>>+	kfree(attr);
>>>+	kfree(xe_attr);
>>>+	kfree(pmu_attr);
>>>+
>>>+	return NULL;
>>>+}
>>>+
>>>+static void free_event_attributes(struct xe_pmu *pmu)
>>>+{
>>>+	struct attribute **attr_iter = pmu->events_attr_group.attrs;
>>>+
>>>+	for (; *attr_iter; attr_iter++)
>>>+		kfree((*attr_iter)->name);
>>>+
>>>+	kfree(pmu->events_attr_group.attrs);
>>>+	kfree(pmu->xe_attr);
>>>+	kfree(pmu->pmu_attr);
>>>+
>>>+	pmu->events_attr_group.attrs = NULL;
>>>+	pmu->xe_attr = NULL;
>>>+	pmu->pmu_attr = NULL;
>>>+}
>>>+
>>>+static int xe_pmu_cpu_online(unsigned int cpu, struct hlist_node *node)
>>>+{
>>>+	struct xe_pmu *pmu = hlist_entry_safe(node, typeof(*pmu), cpuhp.node);
>>>+
>>>+	/* Select the first online CPU as a designated reader. */
>>>+	if (cpumask_empty(&xe_pmu_cpumask))
>>>+		cpumask_set_cpu(cpu, &xe_pmu_cpumask);
>>>+
>>>+	return 0;
>>>+}
>>>+
>>>+static int xe_pmu_cpu_offline(unsigned int cpu, struct hlist_node *node)
>>>+{
>>>+	struct xe_pmu *pmu = hlist_entry_safe(node, typeof(*pmu), cpuhp.node);
>>>+	unsigned int target = xe_pmu_target_cpu;
>>>+
>>>+	/*
>>>+	 * Unregistering an instance generates a CPU offline event which we must
>>>+	 * ignore to avoid incorrectly modifying the shared xe_pmu_cpumask.
>>>+	 */
>>>+	if (pmu->closed)
>>>+		return 0;
>>>+
>>>+	if (cpumask_test_and_clear_cpu(cpu, &xe_pmu_cpumask)) {
>>>+		target = cpumask_any_but(topology_sibling_cpumask(cpu), cpu);
>>>+
>>>+		/* Migrate events if there is a valid target */
>>>+		if (target < nr_cpu_ids) {
>>>+			cpumask_set_cpu(target, &xe_pmu_cpumask);
>>>+			xe_pmu_target_cpu = target;
>>>+		}
>>>+	}
>>>+
>>>+	if (target < nr_cpu_ids && target != pmu->cpuhp.cpu) {
>>>+		perf_pmu_migrate_context(&pmu->base, cpu, target);
>>>+		pmu->cpuhp.cpu = target;
>>>+	}
>>>+
>>>+	return 0;
>>>+}
>>>+
>>>+static enum cpuhp_state cpuhp_slot = CPUHP_INVALID;
>>>+
>>>+int xe_pmu_init(void)
>>>+{
>>>+	int ret;
>>>+
>>>+	ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN,
>>>+				      "perf/x86/intel/xe:online",
>>>+				      xe_pmu_cpu_online,
>>>+				      xe_pmu_cpu_offline);
>>>+	if (ret < 0)
>>>+		pr_notice("Failed to setup cpuhp state for xe PMU! (%d)\n",
>>>+			  ret);
>>>+	else
>>>+		cpuhp_slot = ret;
>>>+
>>>+	return 0;
>>>+}
>>>+
>>>+void xe_pmu_exit(void)
>>>+{
>>>+	if (cpuhp_slot != CPUHP_INVALID)
>>>+		cpuhp_remove_multi_state(cpuhp_slot);
>>>+}
>>>+
>>>+static int xe_pmu_register_cpuhp_state(struct xe_pmu *pmu)
>>>+{
>>>+	if (cpuhp_slot == CPUHP_INVALID)
>>>+		return -EINVAL;
>>>+
>>>+	return cpuhp_state_add_instance(cpuhp_slot, &pmu->cpuhp.node);
>>>+}
>>>+
>>>+static void xe_pmu_unregister_cpuhp_state(struct xe_pmu *pmu)
>>>+{
>>>+	cpuhp_state_remove_instance(cpuhp_slot, &pmu->cpuhp.node);
>>>+}
>>>+
>>>+void xe_pmu_suspend(struct xe_gt *gt)
>>>+{
>>>+	engine_group_busyness_store(gt);
>>>+}
>>>+
>>>+static void xe_pmu_unregister(void *arg)
>>>+{
>>>+	struct xe_pmu *pmu = arg;
>>>+
>>>+	if (!pmu->base.event_init)
>>>+		return;
>>>+
>>>+	/*
>>>+	 * "Disconnect" the PMU callbacks - since all are atomic synchronize_rcu
>>>+	 * ensures all currently executing ones will have exited before we
>>>+	 * proceed with unregistration.
>>>+	 */
>>>+	pmu->closed = true;
>>>+	synchronize_rcu();
>>>+
>>>+	xe_pmu_unregister_cpuhp_state(pmu);
>>>+
>>>+	perf_pmu_unregister(&pmu->base);
>>>+	pmu->base.event_init = NULL;
>>>+	kfree(pmu->base.attr_groups);
>>>+	kfree(pmu->name);
>>>+	free_event_attributes(pmu);
>>>+}
>>>+
>>>+void xe_pmu_register(struct xe_pmu *pmu)
>>>+{
>>>+	struct xe_device *xe = container_of(pmu, typeof(*xe), pmu);
>>>+	const struct attribute_group *attr_groups[] = {
>>>+		&pmu->events_attr_group,
>>>+		&xe_pmu_cpumask_attr_group,
>>>+		NULL
>>>+	};
>>>+
>>>+	int ret = -ENOMEM;
>>>+
>>>+	spin_lock_init(&pmu->lock);
>>>+	pmu->cpuhp.cpu = -1;
>>>+
>>>+	pmu->name = kasprintf(GFP_KERNEL,
>>>+			      "xe_%s",
>>>+			      dev_name(xe->drm.dev));
>>>+	if (pmu->name)
>>>+		/* tools/perf reserves colons as special. */
>>>+		strreplace((char *)pmu->name, ':', '_');
>>>+
>>>+	if (!pmu->name)
>>>+		goto err;
>>>+
>>>+	pmu->events_attr_group.name = "events";
>>>+	pmu->events_attr_group.attrs = create_event_attributes(pmu);
>>>+	if (!pmu->events_attr_group.attrs)
>>>+		goto err_name;
>>>+
>>>+	pmu->base.attr_groups = kmemdup(attr_groups, sizeof(attr_groups),
>>>+					GFP_KERNEL);
>>>+	if (!pmu->base.attr_groups)
>>>+		goto err_attr;
>>>+
>>>+	pmu->base.module	= THIS_MODULE;
>>>+	pmu->base.task_ctx_nr	= perf_invalid_context;
>>>+	pmu->base.event_init	= xe_pmu_event_init;
>>>+	pmu->base.add		= xe_pmu_event_add;
>>>+	pmu->base.del		= xe_pmu_event_del;
>>>+	pmu->base.start		= xe_pmu_event_start;
>>>+	pmu->base.stop		= xe_pmu_event_stop;
>>>+	pmu->base.read		= xe_pmu_event_read;
>>>+	pmu->base.event_idx	= xe_pmu_event_event_idx;
>>>+
>>>+	ret = perf_pmu_register(&pmu->base, pmu->name, -1);
>>>+	if (ret)
>>>+		goto err_groups;
>>>+
>>>+	ret = xe_pmu_register_cpuhp_state(pmu);
>>>+	if (ret)
>>>+		goto err_unreg;
>>>+
>>>+	ret = devm_add_action_or_reset(xe->drm.dev, xe_pmu_unregister, pmu);
>>>+	if (ret)
>>>+		goto err_cpuhp;
>>>+
>>>+	return;
>>>+
>>>+err_cpuhp:
>>>+	xe_pmu_unregister_cpuhp_state(pmu);
>>>+err_unreg:
>>>+	perf_pmu_unregister(&pmu->base);
>>>+err_groups:
>>>+	kfree(pmu->base.attr_groups);
>>>+err_attr:
>>>+	pmu->base.event_init = NULL;
>>>+	free_event_attributes(pmu);
>>>+err_name:
>>>+	kfree(pmu->name);
>>>+err:
>>>+	drm_notice(&xe->drm, "Failed to register PMU!\n");
>>>+}
>>>diff --git a/drivers/gpu/drm/xe/xe_pmu.h b/drivers/gpu/drm/xe/xe_pmu.h
>>>new file mode 100644
>>>index 000000000000..8afa256f9dac
>>>--- /dev/null
>>>+++ b/drivers/gpu/drm/xe/xe_pmu.h
>>>@@ -0,0 +1,26 @@
>>>+/* SPDX-License-Identifier: MIT */
>>>+/*
>>>+ * Copyright © 2024 Intel Corporation
>>>+ */
>>>+
>>>+#ifndef _XE_PMU_H_
>>>+#define _XE_PMU_H_
>>>+
>>>+#include "xe_pmu_types.h"
>>>+
>>>+struct xe_gt;
>>>+
>>>+#if IS_ENABLED(CONFIG_PERF_EVENTS)
>>>+int xe_pmu_init(void);
>>>+void xe_pmu_exit(void);
>>>+void xe_pmu_register(struct xe_pmu *pmu);
>>>+void xe_pmu_suspend(struct xe_gt *gt);
>>>+#else
>>>+static inline int xe_pmu_init(void) { return 0; }
>>>+static inline void xe_pmu_exit(void) {}
>>>+static inline void xe_pmu_register(struct xe_pmu *pmu) {}
>>>+static inline void xe_pmu_suspend(struct xe_gt *gt) {}
>>>+#endif
>>>+
>>>+#endif
>>>+
>>>diff --git a/drivers/gpu/drm/xe/xe_pmu_types.h b/drivers/gpu/drm/xe/xe_pmu_types.h
>>>new file mode 100644
>>>index 000000000000..e86e8d7e0356
>>>--- /dev/null
>>>+++ b/drivers/gpu/drm/xe/xe_pmu_types.h
>>>@@ -0,0 +1,67 @@
>>>+/* SPDX-License-Identifier: MIT */
>>>+/*
>>>+ * Copyright © 2024 Intel Corporation
>>>+ */
>>>+
>>>+#ifndef _XE_PMU_TYPES_H_
>>>+#define _XE_PMU_TYPES_H_
>>>+
>>>+#include <linux/perf_event.h>
>>>+#include <linux/spinlock_types.h>
>>>+#include <uapi/drm/xe_drm.h>
>>>+
>>>+enum {
>>>+	__XE_SAMPLE_RENDER_GROUP_BUSY,
>>>+	__XE_SAMPLE_COPY_GROUP_BUSY,
>>>+	__XE_SAMPLE_MEDIA_GROUP_BUSY,
>>>+	__XE_SAMPLE_ANY_ENGINE_GROUP_BUSY,
>>>+	__XE_NUM_PMU_SAMPLERS
>>>+};
>>>+
>>>+#define XE_PMU_MAX_GT 2
>>>+
>>>+struct xe_pmu {
>>>+	/**
>>>+	 * @cpuhp: Struct used for CPU hotplug handling.
>>>+	 */
>>>+	struct {
>>>+		struct hlist_node node;
>>>+		unsigned int cpu;
>>>+	} cpuhp;
>>>+	/**
>>>+	 * @base: PMU base.
>>>+	 */
>>>+	struct pmu base;
>>>+	/**
>>>+	 * @closed: xe is unregistering.
>>>+	 */
>>>+	bool closed;
>>>+	/**
>>>+	 * @name: Name as registered with perf core.
>>>+	 */
>>>+	const char *name;
>>>+	/**
>>>+	 * @lock: Lock protecting enable mask and ref count handling.
>>>+	 */
>>>+	spinlock_t lock;
>>>+	/**
>>>+	 * @sample: Current and previous (raw) counters.
>>>+	 *
>>>+	 * These counters are updated when the device is awake.
>>>+	 */
>>>+	u64 sample[XE_PMU_MAX_GT][__XE_NUM_PMU_SAMPLERS];
>>>+	/**
>>>+	 * @events_attr_group: Device events attribute group.
>>>+	 */
>>>+	struct attribute_group events_attr_group;
>>>+	/**
>>>+	 * @xe_attr: Memory block holding device attributes.
>>>+	 */
>>>+	void *xe_attr;
>>>+	/**
>>>+	 * @pmu_attr: Memory block holding device attributes.
>>>+	 */
>>>+	void *pmu_attr;
>>>+};
>>>+
>>>+#endif
>>>diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h
>>>index d7b0903c22b2..07ca545354f7 100644
>>>--- a/include/uapi/drm/xe_drm.h
>>>+++ b/include/uapi/drm/xe_drm.h
>>>@@ -1370,6 +1370,45 @@ struct drm_xe_wait_user_fence {
>>>	__u64 reserved[2];
>>>};
>>>
>>>+/**
>>>+ * DOC: XE PMU event config IDs
>>>+ *
>>>+ * Check 'man perf_event_open' to use the ID's XE_PMU_XXXX listed in xe_drm.h
>>>+ * in 'struct perf_event_attr' as part of perf_event_open syscall to read a
>>>+ * particular event.
>>>+ *
>>>+ * For example to open the XE_PMU_RENDER_GROUP_BUSY(0):
>>>+ *
>>>+ * .. code-block:: C
>>>+ *
>>>+ *	struct perf_event_attr attr;
>>>+ *	long long count;
>>>+ *	int cpu = 0;
>>>+ *	int fd;
>>>+ *
>>>+ *	memset(&attr, 0, sizeof(struct perf_event_attr));
>>>+ *	attr.type = type; // eg: /sys/bus/event_source/devices/xe_0000_56_00.0/type
>>>+ *	attr.read_format = PERF_FORMAT_TOTAL_TIME_ENABLED;
>>>+ *	attr.use_clockid = 1;
>>>+ *	attr.clockid = CLOCK_MONOTONIC;
>>>+ *	attr.config = XE_PMU_RENDER_GROUP_BUSY(0);
>>>+ *
>>>+ *	fd = syscall(__NR_perf_event_open, &attr, -1, cpu, -1, 0);
>>>+ */
>>>+
>>>+/*
>>>+ * Top bits of every counter are GT id.
>>>+ */
>>>+#define __XE_PMU_GT_SHIFT (56)
>>>+
>>>+#define ___XE_PMU_OTHER(gt, x) \
>>>+	(((__u64)(x)) | ((__u64)(gt) << __XE_PMU_GT_SHIFT))
>>>+
>>>+#define XE_PMU_RENDER_GROUP_BUSY(gt)		___XE_PMU_OTHER(gt, 0)
>>>+#define XE_PMU_COPY_GROUP_BUSY(gt)		___XE_PMU_OTHER(gt, 1)
>>>+#define XE_PMU_MEDIA_GROUP_BUSY(gt)		___XE_PMU_OTHER(gt, 2)
>>>+#define XE_PMU_ANY_ENGINE_GROUP_BUSY(gt)	___XE_PMU_OTHER(gt, 3)
>>
>>+ Lucas for inputs
>>
>>We should align this to the interface planned for other PMU busyness 
>>counters as well as how we do PCEU. i.e.
>>
>>1) counters are in ticks
>>2) total time in ticks is also exported to the user.
>>
>>For 1), I would just append TICKS to the counter names and drop the
>
>this uses perf and as such I believe we should use the terms used by
>perf.
>
>$ sudo perf stat sleep 1
>
> Performance counter stats for 'sleep 1':
>
>              0.91 msec task-clock                       #    0.001 CPUs utilized
>                 1      context-switches                 #    1.096 K/sec
>                 0      cpu-migrations                   #    0.000 /sec
>                72      page-faults                      #   78.924 K/sec
>------>  2,033,156      cycles                           #    2.229 GHz
>         1,560,992      instructions                     #    0.77  insn per cycle
>           290,814      branches                         #  318.779 M/sec
>            10,449      branch-misses                    #    3.59% of all branches
>
>       1.001580466 seconds time elapsed
>
>       0.000000000 seconds user
>       0.001545000 seconds sys
>
>so... s/ticks/cycles/
>
>I think I said that before, but what's up with all these "group" in the
>names? It's confusing since apparently group and engine class are mixed.

These are counters defined in the HW and indicate busyness of a group of 
engines (spanning multiple classes) rather than a single engine. The 
free running counters are directly read from HW.

Single engine busyness is a different API and wip.

Regards,
Umesh
>
>We are also missing proper kernel-doc in xe_pmu.c
>
>Lucas De Marchi
>
>>conversion to _ns in __engine_group_busyness_read(). Also, drop the 
>>patch that adds this conversion helper.
>>
>>For 2) define a new counter - total active ticks that would return 
>>the 'CPU' timestamp converted to gpu ticks. The reason I am 
>>insisting on CPU timestamp here is because we want to have a time 
>>base that is ticking even when the GPU is idle.
>>
>>Regards,
>>Umesh
>>
>>>+
>>>#if defined(__cplusplus)
>>>}
>>>#endif
>>>-- 
>>>2.40.0
>>>


More information about the Intel-xe mailing list