diff mbox

[igt,v9] igt: Add gem_ctx_freq to exercise requesting freq on a ctx

Message ID 20180413122620.26227-1-chris@chris-wilson.co.uk (mailing list archive)
State New, archived
Headers show

Commit Message

Chris Wilson April 13, 2018, 12:26 p.m. UTC
Exercise some new API that allows applications to request that
individual contexts are executed within a desired frequency range.

v2: Split single/continuous set_freq subtests
v3: Do an up/down ramp for individual freq request, check nothing
changes after each invalid request
v4: Check the frequencies reported by the kernel across the entire
range.
v5: Rewrite sandwich to create a sandwich between multiple concurrent
engines.
v6: Exercise sysfs overrides.
v7: Reset min/max of default context after independent(); don't ask
about failure
v8: Check transition beyond randomly chosen frequencies as well as
up/down ramps.
v9: Check inheritance and sprinkle comments around the place.

With meson.build provided by Arkadiusz Hiler.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Praveen Paneri <praveen.paneri@intel.com>
Cc: Sagar A Kamble <sagar.a.kamble@intel.com>
Cc: Antonio Argenziano <antonio.argenziano@intel.com>
Reviewed-by: Antonio Argenziano <antonio.argenziano@intel.com> #v5
---
 lib/i915/gem_scheduler.c |  15 +
 lib/i915/gem_scheduler.h |   2 +
 tests/Makefile.am        |   1 +
 tests/Makefile.sources   |   1 +
 tests/gem_ctx_freq.c     | 953 +++++++++++++++++++++++++++++++++++++++
 tests/meson.build        |  21 +-
 6 files changed, 987 insertions(+), 6 deletions(-)
 create mode 100644 tests/gem_ctx_freq.c
diff mbox

Patch

diff --git a/lib/i915/gem_scheduler.c b/lib/i915/gem_scheduler.c
index ad156306f..fd0f2e84f 100644
--- a/lib/i915/gem_scheduler.c
+++ b/lib/i915/gem_scheduler.c
@@ -104,6 +104,19 @@  bool gem_scheduler_has_preemption(int fd)
 	       LOCAL_I915_SCHEDULER_CAP_PREEMPTION;
 }
 
+/**
+ * gem_scheduler_has_frequency_control:
+ * @fd: open i915 drm file descriptor
+ *
+ * Feature test macro to query whether the driver supports controlling the
+ * GPU frequency for individual contexts and requests.
+ */
+bool gem_scheduler_has_frequency_control(int fd)
+{
+	return gem_scheduler_capability(fd) &
+	       LOCAL_I915_SCHEDULER_CAP_FREQUENCY;
+}
+
 /**
  * gem_scheduler_print_capability:
  * @fd: open i915 drm file descriptor
@@ -122,4 +135,6 @@  void gem_scheduler_print_capability(int fd)
 		igt_info(" - With priority sorting\n");
 	if (caps & LOCAL_I915_SCHEDULER_CAP_PREEMPTION)
 		igt_info(" - With preemption enabled\n");
+	if (caps & LOCAL_I915_SCHEDULER_CAP_FREQUENCY)
+		igt_info(" - With frequency control\n");
 }
diff --git a/lib/i915/gem_scheduler.h b/lib/i915/gem_scheduler.h
index 9fcb02665..6cc8444f6 100644
--- a/lib/i915/gem_scheduler.h
+++ b/lib/i915/gem_scheduler.h
@@ -27,11 +27,13 @@ 
 #define LOCAL_I915_SCHEDULER_CAP_ENABLED	(1 << 0)
 #define LOCAL_I915_SCHEDULER_CAP_PRIORITY	(1 << 1)
 #define LOCAL_I915_SCHEDULER_CAP_PREEMPTION	(1 << 2)
+#define LOCAL_I915_SCHEDULER_CAP_FREQUENCY	(1 << 3)
 
 unsigned gem_scheduler_capability(int fd);
 bool gem_scheduler_enabled(int fd);
 bool gem_scheduler_has_ctx_priority(int fd);
 bool gem_scheduler_has_preemption(int fd);
+bool gem_scheduler_has_frequency_control(int fd);
 void gem_scheduler_print_capability(int fd);
 
 #endif /* GEM_SCHEDULER_H */
diff --git a/tests/Makefile.am b/tests/Makefile.am
index f41ad5096..408c8b423 100644
--- a/tests/Makefile.am
+++ b/tests/Makefile.am
@@ -104,6 +104,7 @@  drm_import_export_CFLAGS = $(AM_CFLAGS) $(THREAD_CFLAGS)
 drm_import_export_LDADD = $(LDADD) -lpthread
 gem_close_race_CFLAGS = $(AM_CFLAGS) $(THREAD_CFLAGS)
 gem_close_race_LDADD = $(LDADD) -lpthread
+gem_ctx_freq_LDADD = $(LDADD) $(top_builddir)/lib/libigt_perf.la
 gem_ctx_thrash_CFLAGS = $(AM_CFLAGS) $(THREAD_CFLAGS)
 gem_ctx_thrash_LDADD = $(LDADD) -lpthread
 gem_exec_parallel_CFLAGS = $(AM_CFLAGS) $(THREAD_CFLAGS)
diff --git a/tests/Makefile.sources b/tests/Makefile.sources
index 791e4f83d..5ce345963 100644
--- a/tests/Makefile.sources
+++ b/tests/Makefile.sources
@@ -58,6 +58,7 @@  TESTS_progs = \
 	gem_ctx_bad_exec \
 	gem_ctx_create \
 	gem_ctx_exec \
+	gem_ctx_freq \
 	gem_ctx_isolation \
 	gem_ctx_param \
 	gem_ctx_switch \
diff --git a/tests/gem_ctx_freq.c b/tests/gem_ctx_freq.c
new file mode 100644
index 000000000..236c2a667
--- /dev/null
+++ b/tests/gem_ctx_freq.c
@@ -0,0 +1,953 @@ 
+/*
+ * Copyright © 2018 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ */
+
+#include <errno.h>
+#include <fcntl.h>
+#include <sched.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <unistd.h>
+
+#include "igt.h"
+#include "igt_perf.h"
+#include "igt_sysfs.h"
+#include "sw_sync.h"
+
+#define LOCAL_CONTEXT_PARAM_FREQUENCY 8
+
+#define SAMPLE_PERIOD (USEC_PER_SEC / 10)
+#define PMU_TOLERANCE 100
+
+static int sysfs = -1;
+
+static int __set_freq(int fd, uint32_t ctx, uint32_t min, uint32_t max)
+{
+	struct drm_i915_gem_context_param param = {
+		.ctx_id = ctx,
+		.param = LOCAL_CONTEXT_PARAM_FREQUENCY,
+		.value = (uint64_t)max << 32 | min,
+	};
+
+	return __gem_context_set_param(fd, &param);
+}
+
+static void set_freq(int fd, uint32_t ctx, uint32_t min, uint32_t max)
+{
+	igt_assert_eq(__set_freq(fd, ctx, min, max), 0);
+}
+
+static void get_freq(int fd, uint32_t ctx, uint32_t *min, uint32_t *max)
+{
+	struct drm_i915_gem_context_param param = {
+		.ctx_id = ctx,
+		.param = LOCAL_CONTEXT_PARAM_FREQUENCY,
+	};
+
+	gem_context_get_param(fd, &param);
+
+	*min = param.value & 0xffffffff;
+	*max = param.value >> 32;
+}
+
+static uint32_t set_single_freq(int fd, uint32_t ctx, uint32_t freq)
+{
+	uint32_t discard;
+
+	set_freq(fd, ctx, freq, freq);
+	get_freq(fd, ctx, &freq, &discard);
+	igt_assert_eq(freq, discard);
+
+	return freq;
+}
+
+static void kick_rps_worker(void)
+{
+	sched_yield();
+	usleep(SAMPLE_PERIOD);
+}
+
+static double measure_frequency(int pmu, int period_us)
+{
+	uint64_t data[2];
+	uint64_t d_t, d_v;
+
+	kick_rps_worker(); /* let the kthreads (intel_rps_work) run */
+
+	igt_assert_eq(read(pmu, data, sizeof(data)), sizeof(data));
+	d_v = -data[0];
+	d_t = -data[1];
+
+	usleep(period_us);
+
+	igt_assert_eq(read(pmu, data, sizeof(data)), sizeof(data));
+	d_v += data[0];
+	d_t += data[1];
+
+	return d_v * 1e9 / d_t;
+}
+
+static bool __pmu_within_tolerance(double actual, double target)
+{
+	return (actual > target - PMU_TOLERANCE &&
+		actual < target + PMU_TOLERANCE);
+}
+
+static void pmu_assert(double actual, double target)
+{
+	igt_assert_f(__pmu_within_tolerance(actual, target),
+		     "Measured frequency %.2fMHz, is beyond target %.0f±%dMhz\n",
+		     actual, target, PMU_TOLERANCE);
+}
+
+static void busy_wait_until_idle(int fd, igt_spin_t *spin)
+{
+	igt_spin_batch_end(spin);
+	do {
+		usleep(10000);
+	} while (gem_bo_busy(fd, spin->handle));
+}
+
+static void __igt_spin_batch_free_idle(int fd, igt_spin_t *spin)
+{
+	busy_wait_until_idle(fd, spin);
+
+	igt_spin_batch_free(fd, spin);
+}
+
+#define TRIANGLE_SIZE(x) (2 * (x) + 1)
+static void triangle_fill(uint32_t *t, unsigned int nstep,
+			  uint32_t min, uint32_t max)
+{
+	for (unsigned int step = 0; step <= 2*nstep; step++) {
+		int frac = step > nstep ? 2*nstep - step : step;
+		t[step] = min + (max - min) * frac / nstep;
+	}
+}
+
+static void single(int fd, const struct intel_execution_engine *e)
+{
+#define N_STEPS 10
+	const unsigned int engine = e->exec_id | e->flags;
+	uint32_t ctx = gem_context_create(fd);
+	uint32_t frequencies[TRIANGLE_SIZE(N_STEPS)];
+	uint32_t min, max;
+	double measured;
+	igt_spin_t *spin;
+	int pmu;
+
+	/* Check that the requests runs at our context frequency. */
+
+	get_freq(fd, ctx, &min, &max);
+	igt_info("Min freq: %dMHz; Max freq: %dMHz\n", min, max);
+	triangle_fill(frequencies, N_STEPS, min, max);
+
+	pmu = perf_i915_open(I915_PMU_REQUESTED_FREQUENCY);
+	igt_require(pmu >= 0);
+
+	for (int pass = 0; pass < 3; pass++) {
+		for (int i = 0; i < ARRAY_SIZE(frequencies); i++) {
+			uint32_t freq = frequencies[i];
+
+			freq = set_single_freq(fd, ctx, freq);
+
+			gem_quiescent_gpu(fd);
+			spin = __igt_spin_batch_new(fd, ctx, engine, 0);
+
+			measured = measure_frequency(pmu, SAMPLE_PERIOD);
+			igt_debugfs_dump(fd, "i915_rps_boost_info");
+
+			__igt_spin_batch_free_idle(fd, spin);
+
+			igt_info("%s(%s): Measured %.1fMHz, expected %dMhz\n",
+				 e->name, __func__,  measured, freq);
+			pmu_assert(measured, freq);
+		}
+
+		igt_permute_array(frequencies,
+				  ARRAY_SIZE(frequencies),
+				  igt_exchange_int);
+	}
+	gem_quiescent_gpu(fd);
+
+	close(pmu);
+	gem_context_destroy(fd, ctx);
+
+#undef N_STEPS
+}
+
+static void continuous(int fd, const struct intel_execution_engine *e)
+{
+#define N_STEPS 10
+	const unsigned int engine = e->exec_id | e->flags;
+	uint32_t ctx = gem_context_create(fd);
+	uint32_t frequencies[TRIANGLE_SIZE(N_STEPS)];
+	uint32_t min, max;
+	double measured;
+	igt_spin_t *spin;
+	int pmu;
+
+	/*
+	 * Check that we can adjust the frequency range of new requests
+	 * on the fly, even if the context is already executing.
+	 */
+
+	get_freq(fd, ctx, &min, &max);
+	igt_info("Min freq: %dMHz; Max freq: %dMHz\n", min, max);
+	triangle_fill(frequencies, N_STEPS, min, max);
+
+	pmu = perf_i915_open(I915_PMU_REQUESTED_FREQUENCY);
+	igt_require(pmu >= 0);
+
+	gem_quiescent_gpu(fd);
+	spin = __igt_spin_batch_new(fd, ctx, engine, 0);
+	for (int pass = 0; pass < 3; pass++) {
+		for (int i = 0; i < ARRAY_SIZE(frequencies); i++) {
+			uint32_t freq;
+			igt_spin_t *kick;
+
+			freq = set_single_freq(fd, ctx, frequencies[i]);
+
+			/*
+			 * When requesting a new frequency on the currently
+			 * executing context, it does not take effect until the
+			 * next context switch. In this case, we trigger a lite
+			 * restore.
+			 */
+			kick = __igt_spin_batch_new(fd, ctx, engine, 0);
+			igt_spin_batch_free(fd, spin);
+			spin = kick;
+
+			measured = measure_frequency(pmu, SAMPLE_PERIOD);
+			igt_debugfs_dump(fd, "i915_rps_boost_info");
+
+			igt_info("%s(continuous): Measured %.1fMHz, expected %dMhz\n",
+				 e->name, measured, freq);
+			pmu_assert(measured, freq);
+		}
+
+		igt_permute_array(frequencies,
+				  ARRAY_SIZE(frequencies),
+				  igt_exchange_int);
+	}
+	igt_spin_batch_free(fd, spin);
+	gem_quiescent_gpu(fd);
+
+	close(pmu);
+	gem_context_destroy(fd, ctx);
+#undef N_STEPS
+}
+
+static void set_sysfs_freq(uint32_t min, uint32_t max)
+{
+	igt_sysfs_printf(sysfs, "gt_min_freq_mhz", "%u", min);
+	igt_sysfs_printf(sysfs, "gt_max_freq_mhz", "%u", max);
+}
+
+static void get_sysfs_freq(uint32_t *min, uint32_t *max)
+{
+	igt_sysfs_scanf(sysfs, "gt_min_freq_mhz", "%u", min);
+	igt_sysfs_scanf(sysfs, "gt_max_freq_mhz", "%u", max);
+}
+
+static void sysfs_clamp(int fd, const struct intel_execution_engine *e)
+{
+#define N_STEPS 10
+	const unsigned int engine = e->exec_id | e->flags;
+	uint32_t ctx = gem_context_create(fd);
+	uint32_t frequencies[TRIANGLE_SIZE(N_STEPS)];
+	uint32_t sys_min, sys_max;
+	uint32_t min, max;
+	double measured;
+	igt_spin_t *spin;
+	int pmu;
+
+	/*
+	 * The sysfs interface sets the global limits and overrides the
+	 * user's request. So we can to check that if the user requests
+	 * a range outside of the sysfs, the requests are only run at the
+	 * constriained sysfs range.
+	 */
+
+	get_sysfs_freq(&sys_min, &sys_max);
+	igt_info("System min freq: %dMHz; max freq: %dMHz\n", sys_min, sys_max);
+
+	get_freq(fd, ctx, &min, &max);
+	igt_info("Context min freq: %dMHz; max freq: %dMHz\n", min, max);
+	triangle_fill(frequencies, N_STEPS, min, max);
+
+	pmu = perf_i915_open(I915_PMU_REQUESTED_FREQUENCY);
+	igt_require(pmu >= 0);
+
+	for (int outer = 0; outer <= 2*N_STEPS; outer++) {
+		uint32_t sys_freq = frequencies[outer];
+
+		for (int inner = 0; inner <= 2*N_STEPS; inner++) {
+			uint32_t ctx_freq, cur, discard;
+
+			ctx_freq = set_single_freq(fd, ctx, frequencies[inner]);
+
+			gem_quiescent_gpu(fd);
+			spin = __igt_spin_batch_new(fd, ctx, engine, 0);
+			usleep(10000);
+
+			set_sysfs_freq(sys_freq, sys_freq);
+			get_sysfs_freq(&cur, &discard);
+
+			measured = measure_frequency(pmu, SAMPLE_PERIOD);
+			igt_debugfs_dump(fd, "i915_rps_boost_info");
+
+			set_sysfs_freq(sys_min, sys_max);
+			__igt_spin_batch_free_idle(fd, spin);
+
+			igt_info("%s(sysfs): Measured %.1fMHz, context %dMhz, expected %dMhz\n",
+				 e->name, measured, ctx_freq, cur);
+			pmu_assert(measured, cur);
+		}
+	}
+	gem_quiescent_gpu(fd);
+
+	close(pmu);
+	gem_context_destroy(fd, ctx);
+
+#undef N_STEPS
+}
+
+static void sandwich_engine(int fd, unsigned int engine, int timeout)
+{
+	uint32_t ctx = gem_context_create(fd);
+	uint32_t min, max;
+	int pmu;
+
+	pmu = perf_i915_open(I915_PMU_REQUESTED_FREQUENCY);
+	igt_require(pmu >= 0);
+
+	get_freq(fd, ctx, &min, &max);
+
+	igt_until_timeout(timeout) {
+		uint32_t range[2];
+		igt_spin_t *spin;
+		double measured;
+
+		/* make sure we keep an overlap between all engines */
+		range[0] = min + (rand() % (max - min) / 2);
+		range[1] = max - (rand() % (max - min) / 2);
+
+		set_freq(fd, ctx, range[0], range[1]);
+		get_freq(fd, ctx, &range[0], &range[1]);
+
+		spin = __igt_spin_batch_new(fd, ctx, engine, 0);
+
+		measured = measure_frequency(pmu, SAMPLE_PERIOD);
+		igt_spin_batch_free(fd, spin);
+
+		igt_assert(measured >= range[0] - PMU_TOLERANCE &&
+			   measured <= range[1] + PMU_TOLERANCE);
+	}
+
+	gem_context_destroy(fd, ctx);
+	close(pmu);
+}
+
+static void sandwich(int fd, int timeout)
+{
+	unsigned int engine;
+
+	for_each_physical_engine(fd, engine) {
+		igt_fork(child, 1)
+			sandwich_engine(fd, engine, timeout);
+	}
+
+	igt_waitchildren();
+	gem_quiescent_gpu(fd);
+}
+
+static void inheritance(int fd)
+{
+	const uint32_t bbe = MI_BATCH_BUFFER_END;
+	struct drm_i915_gem_exec_object2 obj[2] = {
+		{ .flags = EXEC_OBJECT_WRITE, },
+		{ .handle = gem_create(fd, 4096) }
+	};
+	uint32_t ctx_lo, ctx_hi;
+	uint32_t min, max;
+	unsigned int engine;
+	IGT_CORK_FENCE(cork);
+	int pmu;
+
+	/*
+	 * We support "priority inheritance" across requests. So if a later
+	 * higher priority request is submitted depending on some earlier
+	 * requests, those earlier requests are bumped to the new priority
+	 * and rescheduled. They also inherit the extended frequency range
+	 * of the higher priority request.
+	 */
+
+	igt_require_sw_sync();
+	get_freq(fd, 0, &min, &max);
+
+	ctx_lo = gem_context_create(fd);
+	gem_context_set_priority(fd, ctx_lo, I915_CONTEXT_MIN_USER_PRIORITY);
+	set_freq(fd, ctx_lo, min, min);
+
+	ctx_hi = gem_context_create(fd);
+	gem_context_set_priority(fd, ctx_hi, I915_CONTEXT_MAX_USER_PRIORITY);
+	set_freq(fd, ctx_hi, max, max);
+
+	gem_write(fd, obj[1].handle, 0, &bbe, sizeof(bbe));
+
+	pmu = perf_i915_open(I915_PMU_REQUESTED_FREQUENCY);
+	igt_require(pmu >= 0);
+
+	gem_quiescent_gpu(fd);
+	for_each_physical_engine(fd, engine) {
+		struct drm_i915_gem_execbuffer2 eb = {
+			.buffer_count = 2,
+			.buffers_ptr = to_user_pointer(obj),
+			.flags = engine | I915_EXEC_FENCE_IN,
+			.rsvd1 = ctx_hi,
+			.rsvd2 = igt_cork_plug(&cork, fd),
+		};
+		igt_spin_t *spin = igt_spin_batch_new(fd, ctx_lo, engine, 0);
+		double measured;
+
+		measured = measure_frequency(pmu, SAMPLE_PERIOD);
+
+		igt_info("%s/lo: Measured %.1fMHz, expected %dMhz\n",
+			 e__->name, measured, min);
+		pmu_assert(measured, min);
+
+		/* Add a dependency from hi to lo. */
+		obj[0].handle = spin->handle;
+
+		/* Submit high priority work, but don't let it run */
+		gem_execbuf(fd, &eb);
+
+		measured = measure_frequency(pmu, SAMPLE_PERIOD);
+
+		igt_info("%s/hi: Measured %.1fMHz, expected %dMhz\n",
+			 e__->name, measured, max);
+		igt_debugfs_dump(fd, "i915_engine_info");
+		igt_debugfs_dump(fd, "i915_rps_boost_info");
+
+		igt_spin_batch_free(fd, spin);
+		igt_cork_unplug(&cork);
+
+		pmu_assert(measured, max);
+		gem_quiescent_gpu(fd);
+	}
+
+	/* Also works between engines */
+	for_each_physical_engine(fd, engine) {
+		struct drm_i915_gem_execbuffer2 eb = {
+			.buffer_count = 2,
+			.buffers_ptr = to_user_pointer(obj),
+			.flags = I915_EXEC_DEFAULT | I915_EXEC_FENCE_IN,
+			.rsvd1 = ctx_hi,
+			.rsvd2 = igt_cork_plug(&cork, fd),
+		};
+		igt_spin_t *spin = igt_spin_batch_new(fd, ctx_lo, engine, 0);
+		double measured;
+
+		obj[0].handle = spin->handle;
+		gem_execbuf(fd, &eb);
+
+		measured = measure_frequency(pmu, SAMPLE_PERIOD);
+
+		igt_info("%s-default: Measured %.1fMHz, expected %dMhz\n",
+			 e__->name, measured, max);
+		igt_debugfs_dump(fd, "i915_engine_info");
+		igt_debugfs_dump(fd, "i915_rps_boost_info");
+
+		igt_spin_batch_free(fd, spin);
+		igt_cork_unplug(&cork);
+
+		pmu_assert(measured, max);
+		gem_quiescent_gpu(fd);
+	}
+
+	gem_close(fd, obj[1].handle);
+	close(pmu);
+}
+
+static void pwm(int fd, unsigned int *engines, unsigned int nengine, int link)
+{
+	uint32_t ctx[nengine];
+
+	fcntl(link, F_SETFL, fcntl(link, F_GETFL) | O_NONBLOCK);
+
+	for (unsigned int n = 0; n < nengine; n++)
+		ctx[n] = gem_context_create(fd);
+
+	do {
+		igt_spin_t *spin;
+		struct {
+			uint32_t engine;
+			uint32_t min;
+			uint32_t max;
+		} req;
+
+		while (read(link, &req, sizeof(req)) > 0) {
+			if ((req.engine | req.min | req.max) == 0)
+				goto out;
+
+			igt_assert(req.engine < nengine);
+			set_freq(fd, ctx[req.engine], req.min, req.max);
+		}
+
+		/* Create a ~20% load on this engine  using busy spinners */
+		spin = __igt_spin_batch_new(fd, ctx[0], engines[0], 0);
+		for (unsigned int n = 1; n < nengine; n++) {
+			struct drm_i915_gem_exec_object2 obj = {
+				.handle = spin->handle,
+			};
+			struct drm_i915_gem_execbuffer2 eb = {
+				.buffer_count = 1,
+				.buffers_ptr = to_user_pointer(&obj),
+				.flags = engines[n],
+				.rsvd1 = ctx[n],
+			};
+			gem_execbuf(fd, &eb);
+		}
+		usleep(100);
+
+		__igt_spin_batch_free_idle(fd, spin);
+
+		usleep(400);
+	} while (1);
+
+out:
+	for (unsigned int n = 0; n < nengine; n++)
+		gem_context_destroy(fd, ctx[n]);
+}
+
+static void smoketest(int fd, int timeout)
+{
+	unsigned int engines[16];
+	unsigned int nengine;
+	unsigned int engine;
+	uint32_t min[16], max[16];
+	int pmu, link[2];
+
+	/*
+	 * Basic stress testing (smoketesting).
+	 *
+	 * Apply a random workload that lets the frequency oscillate, giving
+	 * plenty of opportunity for RPS to perform min/max evaluations
+	 * searching for an instance where it dies.
+	 */
+
+	get_freq(fd, 0, &min[0], &max[0]);
+
+	nengine = 0;
+	for_each_physical_engine(fd, engine) {
+		if (nengine == ARRAY_SIZE(engines) - 1)
+			break;
+
+		min[nengine] = min[0];
+		max[nengine] = max[0];
+		engines[nengine] = engine;
+		nengine++;
+	}
+	igt_require(nengine);
+
+	igt_assert(pipe(link) == 0);
+	igt_fork(child, 1)
+		pwm(fd, engines, nengine, link[0]);
+	close(link[0]);
+
+	pmu = perf_i915_open(I915_PMU_REQUESTED_FREQUENCY);
+	igt_require(pmu >= 0);
+
+	igt_until_timeout(timeout) {
+		struct {
+			uint32_t engine;
+			uint32_t min;
+			uint32_t max;
+		} req;
+		double measured;
+		uint32_t ctx;
+
+		req.engine = rand() % nengine;
+
+		/*
+		 * Since we apply a different range to each engine, we
+		 * want to ensure that all ranges are conjoint (overlap).
+		 * This allows us to apply the simple rule that in order
+		 * to satisfy the independent contexts running on each
+		 * engine, the frequency must be within the intersection
+		 * of all ranges. If that intersection is empty, we leave
+		 * it up to the kernel to decide (and do not want to imply
+		 * any contract to the user, so ignore it).
+		 */
+		ctx = gem_context_create(fd);
+		get_freq(fd, ctx, &req.min, &req.max);
+		req.min = rand() % (req.max - req.min) + req.min;
+		req.max = rand() % (req.max - req.min) + req.min;
+		set_freq(fd, ctx, req.min, req.max);
+		get_freq(fd, ctx, &req.min, &req.max);
+
+		igt_debug("Replacing (%d, %d) on engine %x with (%d, %d)\n",
+			  min[req.engine], max[req.engine], req.engine,
+			  req.min, req.max);
+		igt_assert(write(link[1], &req, sizeof(req)) == sizeof(req));
+		gem_context_destroy(fd, ctx);
+
+		min[req.engine] = req.min;
+		max[req.engine] = req.max;
+
+		for (unsigned int n = 0; n < nengine; n++) {
+			igt_debug("[%d]: [%d, %d]\n", n, min[n], max[n]);
+			if (min[n] < req.min)
+				req.min = min[n];
+			if (max[n] > req.max)
+				req.max = max[n];
+		}
+		igt_assert(req.max >= req.min);
+
+		usleep(50000);
+		measured = measure_frequency(pmu, SAMPLE_PERIOD);
+
+		if (measured <= req.min - PMU_TOLERANCE ||
+		    measured >= req.max + PMU_TOLERANCE)
+			igt_debugfs_dump(fd, "i915_rps_boost_info");
+		igt_info("Measured %.1fMHz, expected [%d, %d]Mhz\n",
+			 measured, req.min, req.max);
+		igt_assert(measured > req.min - PMU_TOLERANCE &&
+			   measured < req.max + PMU_TOLERANCE);
+	}
+
+	do {
+		struct {
+			uint32_t engine;
+			uint32_t min;
+			uint32_t max;
+		} req = {};
+
+		write(link[1], &req, sizeof(req));
+		close(link[1]);
+	} while (0);
+	igt_waitchildren();
+	gem_quiescent_gpu(fd);
+
+	close(pmu);
+}
+
+static void invalid_context(int fd, uint32_t ctx, uint32_t min, uint32_t max)
+{
+	const struct test {
+		uint32_t min, max;
+	} tests[] = {
+		{ min - 50, max - 50 },
+		{ min - 50, max },
+		{ min - 50, max + 50 },
+		{ min, max + 50 },
+		{ min + 50, max + 50 },
+
+		{ min - 50, min - 50 },
+
+		{ min - 50, min },
+		{ min + 50, min },
+		{ min, min - 50 },
+
+		{ max + 50, max },
+		{ max, max + 50 },
+		{ max, max - 50 },
+
+		{ max + 50, max + 50 },
+
+		{}
+	};
+
+	for (const struct test *t = tests; t->min | t->max; t++) {
+		uint32_t cur_min, cur_max;
+
+		igt_assert_f(__set_freq(fd, ctx, t->min, t->max) == -EINVAL,
+			     "Failed to reject invalid [%d, %d] (valid range [%d, %d]) on context %d\n",
+			     t->min, t->max, min, max, ctx);
+
+		get_freq(fd, 0, &cur_min, &cur_max);
+		igt_assert_eq(cur_min, min);
+		igt_assert_eq(cur_max, max);
+	}
+}
+
+static void uabi_invalid(int fd)
+{
+	uint32_t min, max, ctx;
+
+	/* Minimal negative testing for rejection of invalid parameters */
+
+	get_freq(fd, 0, &min, &max);
+
+	invalid_context(fd, 0, min, max);
+
+	ctx = gem_context_create(fd);
+	invalid_context(fd, ctx, min, max);
+	gem_context_destroy(fd, ctx);
+}
+
+static void idempotent_context(int fd, uint32_t ctx)
+{
+	uint32_t min, max;
+	uint32_t cur_min, cur_max;
+	uint32_t *all, count;
+
+	get_freq(fd, ctx, &min, &max);
+
+	set_freq(fd, ctx, max, max);
+	get_freq(fd, ctx, &cur_min, &cur_max);
+	igt_assert_eq(cur_min, max);
+	igt_assert_eq(cur_max, max);
+
+	set_freq(fd, ctx, min, min);
+	get_freq(fd, ctx, &cur_min, &cur_max);
+	igt_assert_eq(cur_min, min);
+	igt_assert_eq(cur_max, min);
+
+	/* Bin each frequency in range and check that each is reported back */
+	count = max - min + 1;
+	all = malloc(sizeof(uint32_t) * count);
+	igt_assert(all);
+	for (uint32_t n = 0; n < count; n++) {
+		set_freq(fd, ctx, n + min, n + min);
+		get_freq(fd, ctx, &cur_min, &cur_max);
+		all[n] = cur_min;
+	}
+	igt_permute_array(all, count, igt_exchange_int);
+	for (uint32_t n = 0; n < count; n++) {
+		set_freq(fd, ctx, min, all[n]);
+		get_freq(fd, ctx, &cur_min, &cur_max);
+		igt_assert_eq(cur_min, min);
+		igt_assert_eq(cur_max, all[n]);
+
+		set_freq(fd, ctx, all[n], max);
+		get_freq(fd, ctx, &cur_min, &cur_max);
+		igt_assert_eq(cur_min, all[n]);
+		igt_assert_eq(cur_max, max);
+
+		set_freq(fd, ctx, all[n], all[n]);
+		get_freq(fd, ctx, &cur_min, &cur_max);
+		igt_assert_eq(cur_min, all[n]);
+		igt_assert_eq(cur_max, all[n]);
+	}
+
+	set_freq(fd, ctx, min, max);
+	get_freq(fd, ctx, &cur_min, &cur_max);
+	igt_assert_eq(cur_min, min);
+	igt_assert_eq(cur_max, max);
+}
+
+static void uabi_idempotent(int fd)
+{
+	uint32_t ctx;
+
+	/*
+	 * A general principle for an API of least-surprise is that the
+	 * user receives the same value as they put in. We apply this
+	 * principle for I915_CONTEXT_PARAM_FREQUENCY, modulo rounding
+	 * into the nearest bin.
+	 */
+
+	idempotent_context(fd, 0);
+
+	ctx = gem_context_create(fd);
+	idempotent_context(fd, ctx);
+	gem_context_destroy(fd, ctx);
+}
+
+static void range_context(int fd, uint32_t ctx)
+{
+	uint32_t min, max;
+	uint32_t cur_min, cur_max;
+
+	get_freq(fd, ctx, &min, &max);
+
+	for (uint32_t freq = min; freq <= max; freq++) {
+		set_freq(fd, ctx, freq, freq);
+		get_freq(fd, ctx, &cur_min, &cur_max);
+
+		igt_assert(cur_min >= min);
+		igt_assert(cur_max <= max);
+	}
+
+	set_freq(fd, ctx, min, max);
+	get_freq(fd, ctx, &cur_min, &cur_max);
+	igt_assert_eq(cur_min, min);
+	igt_assert_eq(cur_max, max);
+}
+
+static void uabi_range(int fd)
+{
+	uint32_t ctx;
+
+	/* Check that all reqported frequencies are within the min/max */
+
+	range_context(fd, 0);
+
+	ctx = gem_context_create(fd);
+	range_context(fd, ctx);
+	gem_context_destroy(fd, ctx);
+}
+
+static void uabi_independent(int fd)
+{
+	uint32_t min, max;
+	uint32_t cur_min, cur_max;
+	uint32_t ctx[2];
+
+	/*
+	 * Newly created contexts should not derive their limits from older
+	 * contexts, but be always created with the same full range.
+	 */
+
+	get_freq(fd, 0, &min, &max);
+
+	set_freq(fd, 0, max, max);
+	ctx[0] = gem_context_create(fd);
+	get_freq(fd, ctx[0], &cur_min, &cur_max);
+	igt_assert_eq(cur_min, min);
+	igt_assert_eq(cur_max, max);
+
+	set_freq(fd, 0, min, min);
+	get_freq(fd, ctx[0], &cur_min, &cur_max);
+	igt_assert_eq(cur_min, min);
+	igt_assert_eq(cur_max, max);
+
+	ctx[1] = gem_context_create(fd);
+	get_freq(fd, ctx[1], &cur_min, &cur_max);
+	igt_assert_eq(cur_min, min);
+	igt_assert_eq(cur_max, max);
+
+	set_freq(fd, ctx[1], max, max);
+	get_freq(fd, ctx[0], &cur_min, &cur_max);
+	igt_assert_eq(cur_min, min);
+	igt_assert_eq(cur_max, max);
+
+	get_freq(fd, 0, &cur_min, &cur_max);
+	igt_assert_eq(cur_min, min);
+	igt_assert_eq(cur_max, min);
+
+	get_freq(fd, ctx[1], &cur_min, &cur_max);
+	igt_assert_eq(cur_min, max);
+	igt_assert_eq(cur_max, max);
+	gem_context_destroy(fd, ctx[1]);
+
+	get_freq(fd, ctx[0], &cur_min, &cur_max);
+	igt_assert_eq(cur_min, min);
+	igt_assert_eq(cur_max, max);
+	gem_context_destroy(fd, ctx[0]);
+
+	set_freq(fd, 0, min, max);
+	get_freq(fd, 0, &cur_min, &cur_max);
+	igt_assert_eq(cur_min, min);
+	igt_assert_eq(cur_max, max);
+}
+
+static void restore_sysfs_freq(int sig)
+{
+	char buf[256];
+
+	if (igt_sysfs_read(sysfs, "gt_RPn_freq_mhz", buf, sizeof(buf)) > 0) {
+		igt_sysfs_set(sysfs, "gt_idle_freq_mhz", buf);
+		igt_sysfs_set(sysfs, "gt_min_freq_mhz", buf);
+	}
+
+	if (igt_sysfs_read(sysfs, "gt_RP0_freq_mhz", buf, sizeof(buf)) > 0) {
+		igt_sysfs_set(sysfs, "gt_max_freq_mhz", buf);
+		igt_sysfs_set(sysfs, "gt_boost_freq_mhz", buf);
+	}
+}
+
+static void disable_boost(int fd)
+{
+	char *value;
+
+	value = igt_sysfs_get(fd, "gt_RPn_freq_mhz");
+	igt_sysfs_set(fd, "gt_min_freq_mhz", value);
+	igt_sysfs_set(fd, "gt_boost_freq_mhz", value);
+	free(value);
+
+	value = igt_sysfs_get(fd, "gt_RP0_freq_mhz");
+	igt_sysfs_set(fd, "gt_max_freq_mhz", value);
+	free(value);
+}
+
+igt_main
+{
+	const struct intel_execution_engine *e;
+	int fd = -1;
+
+	igt_fixture {
+		fd = drm_open_driver(DRIVER_INTEL);
+		igt_require_gem(fd);
+		igt_require(gem_scheduler_has_frequency_control(fd));
+
+		sysfs = igt_sysfs_open(fd, NULL);
+		igt_assert(sysfs != -1);
+		igt_install_exit_handler(restore_sysfs_freq);
+
+		disable_boost(sysfs);
+	}
+
+	igt_subtest("uabi-invalid")
+		uabi_invalid(fd);
+
+	igt_subtest("uabi-range")
+		uabi_range(fd);
+
+	igt_subtest("uabi-idempotent")
+		uabi_idempotent(fd);
+
+	igt_subtest("uabi-independent")
+		uabi_independent(fd);
+
+	igt_skip_on_simulation();
+
+	for (e = intel_execution_engines; e->name; e++) {
+		igt_subtest_group {
+			igt_fixture {
+				gem_require_ring(fd, e->exec_id | e->flags);
+			}
+
+			igt_subtest_f("%s-single", e->name)
+				single(fd, e);
+			igt_subtest_f("%s-continuous", e->name)
+				continuous(fd, e);
+			igt_subtest_f("%s-sysfs", e->name)
+				sysfs_clamp(fd, e);
+		}
+	}
+
+	igt_subtest("sandwich")
+		sandwich(fd, 20);
+
+	igt_subtest("inheritance")
+		inheritance(fd);
+
+	igt_subtest("smoketest") {
+		igt_fork_hang_detector(fd);
+		smoketest(fd, 20);
+		igt_stop_hang_detector();
+	}
+}
diff --git a/tests/meson.build b/tests/meson.build
index 4720dfe21..0052b3640 100644
--- a/tests/meson.build
+++ b/tests/meson.build
@@ -1,3 +1,9 @@ 
+# those will be linked with lib_igt_perf
+perf_test_progs = [
+	'gem_ctx_freq',
+	'perf_pmu',
+]
+
 test_progs = [
 	'core_auth',
 	'core_get_client_auth',
@@ -35,6 +41,7 @@  test_progs = [
 	'gem_ctx_bad_exec',
 	'gem_ctx_create',
 	'gem_ctx_exec',
+	'gem_ctx_freq',
 	'gem_ctx_isolation',
 	'gem_ctx_param',
 	'gem_ctx_switch',
@@ -295,12 +302,14 @@  test_executables += executable('gem_eio', 'gem_eio.c',
 	   install : true)
 test_progs += 'gem_eio'
 
-test_executables += executable('perf_pmu', 'perf_pmu.c',
-	   dependencies : test_deps + [ lib_igt_perf ],
-	   install_dir : libexecdir,
-	   install_rpath : rpathdir,
-	   install : true)
-test_progs += 'perf_pmu'
+foreach prog : perf_test_progs
+	test_executables += executable(prog, prog + '.c',
+		dependencies : test_deps + [ lib_igt_perf ],
+		install_dir : libexecdir,
+		install_rpath : rpathdir,
+		install : true)
+	test_progs += prog
+endforeach
 
 executable('testdisplay', ['testdisplay.c', 'testdisplay_hotplug.c'],
 	   dependencies : test_deps,