diff mbox

[4/8] drm/i915: Add i915 perf event for Haswell OA unit

Message ID 1454448655-11178-5-git-send-email-robert@sixbynine.org (mailing list archive)
State New, archived
Headers show

Commit Message

Robert Bragg Feb. 2, 2016, 9:30 p.m. UTC
Gen graphics hardware can be set up to periodically write snapshots of
performance counters into a circular buffer via its Observation
Architecture and this patch exposes that capability to userspace via the
i915 perf interface.

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Robert Bragg <robert@sixbynine.org>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
---
 drivers/gpu/drm/i915/i915_drv.h         |  49 ++
 drivers/gpu/drm/i915/i915_gem_context.c |  23 +-
 drivers/gpu/drm/i915/i915_perf.c        | 783 +++++++++++++++++++++++++++++++-
 drivers/gpu/drm/i915/i915_reg.h         | 338 ++++++++++++++
 include/uapi/drm/i915_drm.h             |  56 ++-
 5 files changed, 1233 insertions(+), 16 deletions(-)

Comments

kernel test robot Feb. 2, 2016, 10:18 p.m. UTC | #1
Hi Robert,

[auto build test WARNING on v4.4-rc8]
[cannot apply to drm-intel/for-linux-next v4.5-rc2 v4.5-rc1 next-20160202]
[if your patch is applied to the wrong git tree, please drop us a note to help improving the system]

url:    https://github.com/0day-ci/linux/commits/Robert-Bragg/Enable-Gen-7-Observation-Architecture/20160203-053423
config: i386-defconfig (attached as .config)
reproduce:
        # save the attached .config to linux build tree
        make ARCH=i386 

All warnings (new ones prefixed by >>):

   drivers/gpu/drm/i915/i915_perf.c: In function 'gen7_oa_read':
>> drivers/gpu/drm/i915/i915_perf.c:264:2: warning: #warning "WIP: check the correct way to clear OASTATUS1 bits" [-Wcpp]
    #warning "WIP: check the correct way to clear OASTATUS1 bits"
     ^

vim +264 drivers/gpu/drm/i915/i915_perf.c

   248						       DRM_I915_PERF_RECORD_OA_REPORT_LOST);
   249				if (n_records && ret == -EFAULT)
   250					ret = 0;
   251				else if (ret < 0)
   252					return ret;
   253				else if (ret > 0) {
   254					n_records += ret;
   255					oastatus1 &= ~GEN7_OASTATUS1_REPORT_LOST;
   256				}
   257			}
   258	
   259			/* XXX: This register contains the OA tail pointer
   260			 * which may be updated asynchronously while periodic
   261			 * sampling is enabled so it's racy to
   262			 * read-modify-write like we are here (esp after the
   263			 * append/copy_to_user work above.) */
 > 264	#warning "WIP: check the correct way to clear OASTATUS1 bits"
   265			I915_WRITE(GEN7_OASTATUS1, oastatus1);
   266		}
   267	
   268		/* If there is still buffer space */
   269		if (ret) {
   270			ret = gen7_append_oa_reports(stream, read_state, &head, tail);
   271			if (n_records && ret == -EFAULT)
   272				ret = 0;

---
0-DAY kernel test infrastructure                Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all                   Intel Corporation
diff mbox

Patch

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index ebd556f..aaf1a5b 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1698,6 +1698,11 @@  struct i915_execbuffer_params {
 	struct drm_i915_gem_request     *request;
 };
 
+struct i915_oa_format {
+	u32 format;
+	int size;
+};
+
 struct i915_oa_reg {
 	u32 addr;
 	u32 value;
@@ -1715,6 +1720,7 @@  struct i915_perf_stream {
 	struct list_head link;
 
 	u32 sample_flags;
+	int sample_size;
 
 	struct intel_context *ctx;
 	bool enabled;
@@ -1778,6 +1784,20 @@  struct i915_perf_stream {
 	void (*destroy)(struct i915_perf_stream *stream);
 };
 
+struct i915_oa_ops {
+	void (*init_oa_buffer)(struct drm_i915_private *dev_priv);
+	int (*enable_metric_set)(struct drm_i915_private *dev_priv);
+	void (*disable_metric_set)(struct drm_i915_private *dev_priv);
+	void (*oa_enable)(struct drm_i915_private *dev_priv);
+	void (*oa_disable)(struct drm_i915_private *dev_priv);
+	void (*update_oacontrol)(struct drm_i915_private *dev_priv);
+	void (*update_hw_ctx_id_locked)(struct drm_i915_private *dev_priv,
+					u32 ctx_id);
+	int (*read)(struct i915_perf_stream *stream,
+		    struct i915_perf_read_state *read_state);
+	bool (*oa_buffer_is_empty)(struct drm_i915_private *dev_priv);
+};
+
 struct drm_i915_private {
 	struct drm_device *dev;
 	struct kmem_cache *objects;
@@ -2023,16 +2043,43 @@  struct drm_i915_private {
 
 	struct {
 		bool initialized;
+
 		struct mutex lock;
 		struct list_head streams;
 
+		spinlock_t hook_lock;
+
 		struct {
+			struct i915_perf_stream *exclusive_stream;
+
+			u32 specific_ctx_id;
+
+			struct hrtimer poll_check_timer;
+			wait_queue_head_t poll_wq;
+
+			bool periodic;
+			u32 period_exponent;
+
 			u32 metrics_set;
 
 			const struct i915_oa_reg *mux_regs;
 			int mux_regs_len;
 			const struct i915_oa_reg *b_counter_regs;
 			int b_counter_regs_len;
+
+			struct {
+				struct drm_i915_gem_object *obj;
+				u32 gtt_offset;
+				u8 *addr;
+				u32 head;
+				u32 tail;
+				int format;
+				int format_size;
+			} oa_buffer;
+
+			struct i915_oa_ops ops;
+			const struct i915_oa_format *oa_formats;
+			int n_builtin_sets;
 		} oa;
 	} perf;
 
@@ -3272,6 +3319,8 @@  int i915_gem_context_setparam_ioctl(struct drm_device *dev, void *data,
 
 int i915_perf_open_ioctl(struct drm_device *dev, void *data,
 			 struct drm_file *file);
+void i915_oa_context_pin_notify(struct drm_i915_private *dev_priv,
+				struct intel_context *context);
 
 /* i915_gem_evict.c */
 int __must_check i915_gem_evict_something(struct drm_device *dev,
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 02ceb7a..b8d2d41 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -133,6 +133,23 @@  static int get_context_size(struct drm_device *dev)
 	return ret;
 }
 
+static int i915_gem_context_pin_state(struct drm_device *dev,
+				      struct intel_context *ctx)
+{
+	int ret;
+
+	BUG_ON(!mutex_is_locked(&dev->struct_mutex));
+
+	ret = i915_gem_obj_ggtt_pin(ctx->legacy_hw_ctx.rcs_state,
+				    get_context_alignment(dev), 0);
+	if (ret)
+		return ret;
+
+	i915_oa_context_pin_notify(dev->dev_private, ctx);
+
+	return 0;
+}
+
 static void i915_gem_context_clean(struct intel_context *ctx)
 {
 	struct i915_hw_ppgtt *ppgtt = ctx->ppgtt;
@@ -280,8 +297,7 @@  i915_gem_create_context(struct drm_device *dev,
 		 * be available. To avoid this we always pin the default
 		 * context.
 		 */
-		ret = i915_gem_obj_ggtt_pin(ctx->legacy_hw_ctx.rcs_state,
-					    get_context_alignment(dev), 0);
+		ret = i915_gem_context_pin_state(dev, ctx);
 		if (ret) {
 			DRM_DEBUG_DRIVER("Couldn't pin %d\n", ret);
 			goto err_destroy;
@@ -663,8 +679,7 @@  static int do_switch(struct drm_i915_gem_request *req)
 
 	/* Trying to pin first makes error handling easier. */
 	if (ring == &dev_priv->ring[RCS]) {
-		ret = i915_gem_obj_ggtt_pin(to->legacy_hw_ctx.rcs_state,
-					    get_context_alignment(ring->dev), 0);
+		ret = i915_gem_context_pin_state(ring->dev, to);
 		if (ret)
 			return ret;
 	}
diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index 9af9e52..b54f719 100644
--- a/drivers/gpu/drm/i915/i915_perf.c
+++ b/drivers/gpu/drm/i915/i915_perf.c
@@ -25,14 +25,692 @@ 
 #include <linux/sizes.h>
 
 #include "i915_drv.h"
+#include "intel_ringbuffer.h"
+#include "intel_lrc.h"
+#include "i915_oa_hsw.h"
+
+/* Must be a power of two */
+#define OA_BUFFER_SIZE	     SZ_16M
+#define OA_TAKEN(tail, head) ((tail - head) & (OA_BUFFER_SIZE - 1))
+
+/* frequency for forwarding samples from OA to perf buffer */
+#define POLL_FREQUENCY 200
+#define POLL_PERIOD max_t(u64, 10000, NSEC_PER_SEC / POLL_FREQUENCY)
+
+#define OA_EXPONENT_MAX 0x3f
+
+static struct i915_oa_format hsw_oa_formats[I915_OA_FORMAT_MAX] = {
+	[I915_OA_FORMAT_A13]	    = { 0, 64 },
+	[I915_OA_FORMAT_A29]	    = { 1, 128 },
+	[I915_OA_FORMAT_A13_B8_C8]  = { 2, 128 },
+	/* A29_B8_C8 Disallowed as 192 bytes doesn't factor into buffer size */
+	[I915_OA_FORMAT_B4_C8]	    = { 4, 64 },
+	[I915_OA_FORMAT_A45_B8_C8]  = { 5, 256 },
+	[I915_OA_FORMAT_B4_C8_A16]  = { 6, 128 },
+	[I915_OA_FORMAT_C4_B8]	    = { 7, 64 },
+};
+
+#define SAMPLE_OA_REPORT      (1<<0)
 
 struct perf_open_properties {
 	u32 sample_flags;
 
 	u64 single_context:1;
 	u64 ctx_handle;
+
+	/* OA sampling state */
+	int metrics_set;
+	int oa_format;
+	bool oa_periodic;
+	u32 oa_period_exponent;
 };
 
+static bool gen7_oa_buffer_is_empty(struct drm_i915_private *dev_priv)
+{
+	u32 oastatus2 = I915_READ(GEN7_OASTATUS2);
+	u32 oastatus1 = I915_READ(GEN7_OASTATUS1);
+	u32 head = oastatus2 & GEN7_OASTATUS2_HEAD_MASK;
+	u32 tail = oastatus1 & GEN7_OASTATUS1_TAIL_MASK;
+
+	return OA_TAKEN(tail, head) == 0;
+}
+
+/**
+ * Appends a status record to a userspace read() buffer.
+ *
+ * Returns 0 if no room left in userspace read() buffer, 1 if status
+ * record successfully appended or -EFAULT for copy_to_user failure.
+ */
+static int append_oa_status(struct i915_perf_stream *stream,
+			    struct i915_perf_read_state *read_state,
+			    enum drm_i915_perf_record_type type)
+{
+	struct drm_i915_perf_record_header header = { type, 0, sizeof(header) };
+
+	if ((read_state->count - read_state->read) < header.size)
+		return 0;
+
+	if (copy_to_user(read_state->buf, &header, sizeof(header)))
+		return -EFAULT;
+
+	read_state->buf += sizeof(header);
+	read_state->read += header.size;
+
+	return 1;
+}
+
+/**
+ * Copies single OA report into userspace read() buffer.
+ *
+ * Returns 0 if no room left in userspace read() buffer, 1 if sample
+ * record successfully appended or -EFAULT for copy_to_user failure.
+ */
+static int append_oa_sample(struct i915_perf_stream *stream,
+			    struct i915_perf_read_state *read_state,
+			    const u8 *report)
+{
+	struct drm_i915_private *dev_priv = stream->dev_priv;
+	int report_size = dev_priv->perf.oa.oa_buffer.format_size;
+	struct drm_i915_perf_record_header header;
+	u32 sample_flags = stream->sample_flags;
+	char __user *buf = read_state->buf;
+
+	header.type = DRM_I915_PERF_RECORD_SAMPLE;
+	header.pad = 0;
+	header.size = stream->sample_size;
+
+	if ((read_state->count - read_state->read) < header.size)
+		return 0;
+
+	if (copy_to_user(buf, &header, sizeof(header)))
+		return -EFAULT;
+	buf += sizeof(header);
+
+	if (sample_flags & SAMPLE_OA_REPORT) {
+		if (copy_to_user(buf, report, report_size))
+			return -EFAULT;
+		buf += report_size;
+	}
+
+	read_state->buf = buf;
+	read_state->read += header.size;
+
+	return 1;
+}
+
+/**
+ * Copies all buffered OA reports into userspace read() buffer.
+ * @head_ptr: (inout): the head pointer before and after appending
+ *
+ * Returns the number of records appended (may be zero if the read()
+ * buffer is already full), or -EFAULT for failures while copying.
+ *
+ * Consistent with i915_perf_stream::read(), this function should only
+ * return an -EFAULT from copying if no records have been copied and
+ * otherwise reports a short read.
+ */
+static int gen7_append_oa_reports(struct i915_perf_stream *stream,
+				  struct i915_perf_read_state *read_state,
+				  u32 *head_ptr,
+				  u32 tail)
+{
+	struct drm_i915_private *dev_priv = stream->dev_priv;
+	int report_size = dev_priv->perf.oa.oa_buffer.format_size;
+	u8 *oa_buf_base = dev_priv->perf.oa.oa_buffer.addr;
+	u32 mask = (OA_BUFFER_SIZE - 1);
+	u32 head;
+	u8 *report;
+	u32 taken;
+	int ret = 0;
+	int n_records = 0;
+
+	head = *head_ptr - dev_priv->perf.oa.oa_buffer.gtt_offset;
+	tail -= dev_priv->perf.oa.oa_buffer.gtt_offset;
+
+	/* Note: the gpu doesn't wrap the tail according to the OA buffer size
+	 * so when we need to make sure our head/tail values are in-bounds we
+	 * use the above mask.
+	 */
+
+	while ((taken = OA_TAKEN(tail, head))) {
+		/* The tail increases in 64 byte increments, not in
+		 * format_size steps.
+		 */
+		if (taken < report_size)
+			break;
+
+		report = oa_buf_base + (head & mask);
+
+		if (dev_priv->perf.oa.exclusive_stream->enabled) {
+			ret = append_oa_sample(stream, read_state, report);
+			if (n_records && ret == -EFAULT)
+				ret = 0;
+			if (ret > 0)
+				n_records += ret;
+			else
+				break;
+		}
+
+		/* We don't want to update head if there was a problem
+		 * appending the sample.
+		 */
+		head += report_size;
+	}
+
+	*head_ptr = dev_priv->perf.oa.oa_buffer.gtt_offset + head;
+
+	return n_records;
+}
+
+/**
+ * Check OA status registers, appending status records if necessary
+ * and copy as many buffered OA reports to userspace as possible.
+ *
+ * NB: this function should only return an -EFAULT from copying if no
+ * records have been copied and otherwise report a short read.
+ *
+ * Returns the number of records appended (may be zero if no room in
+ * read() buffer), or -EFAULT (the only errno we expect to see here)
+ */
+static int gen7_oa_read(struct i915_perf_stream *stream,
+			struct i915_perf_read_state *read_state)
+{
+	struct drm_i915_private *dev_priv = stream->dev_priv;
+	u32 oastatus2;
+	u32 oastatus1;
+	u32 head;
+	u32 tail;
+	int ret = 1; /* > 0 implies a prior successful append */
+	int n_records = 0;
+
+	WARN_ON(!dev_priv->perf.oa.oa_buffer.addr);
+
+	oastatus2 = I915_READ(GEN7_OASTATUS2);
+	oastatus1 = I915_READ(GEN7_OASTATUS1);
+
+	head = oastatus2 & GEN7_OASTATUS2_HEAD_MASK;
+	tail = oastatus1 & GEN7_OASTATUS1_TAIL_MASK;
+
+	if (unlikely(oastatus1 & (GEN7_OASTATUS1_OABUFFER_OVERFLOW |
+				  GEN7_OASTATUS1_REPORT_LOST))) {
+
+		if (oastatus1 & GEN7_OASTATUS1_OABUFFER_OVERFLOW) {
+			ret = append_oa_status(stream, read_state,
+					       DRM_I915_PERF_RECORD_OA_BUFFER_OVERFLOW);
+			if (ret <= 0)
+				return ret;
+			n_records += ret;
+			oastatus1 &= ~GEN7_OASTATUS1_OABUFFER_OVERFLOW;
+		}
+
+		if (oastatus1 & GEN7_OASTATUS1_REPORT_LOST) {
+			ret = append_oa_status(stream, read_state,
+					       DRM_I915_PERF_RECORD_OA_REPORT_LOST);
+			if (n_records && ret == -EFAULT)
+				ret = 0;
+			else if (ret < 0)
+				return ret;
+			else if (ret > 0) {
+				n_records += ret;
+				oastatus1 &= ~GEN7_OASTATUS1_REPORT_LOST;
+			}
+		}
+
+		/* XXX: This register contains the OA tail pointer
+		 * which may be updated asynchronously while periodic
+		 * sampling is enabled so it's racy to
+		 * read-modify-write like we are here (esp after the
+		 * append/copy_to_user work above.) */
+#warning "WIP: check the correct way to clear OASTATUS1 bits"
+		I915_WRITE(GEN7_OASTATUS1, oastatus1);
+	}
+
+	/* If there is still buffer space */
+	if (ret) {
+		ret = gen7_append_oa_reports(stream, read_state, &head, tail);
+		if (n_records && ret == -EFAULT)
+			ret = 0;
+		else if (ret < 0)
+			return ret;
+		else if (ret > 0) {
+			n_records += ret;
+
+			I915_WRITE(GEN7_OASTATUS2,
+				   ((head & GEN7_OASTATUS2_HEAD_MASK) |
+				    OA_MEM_SELECT_GGTT));
+		}
+	}
+
+	return n_records;
+}
+
+static bool i915_oa_can_read(struct i915_perf_stream *stream)
+{
+	struct drm_i915_private *dev_priv = stream->dev_priv;
+
+	return !dev_priv->perf.oa.ops.oa_buffer_is_empty(dev_priv);
+}
+
+static int i915_oa_wait_unlocked(struct i915_perf_stream *stream)
+{
+	struct drm_i915_private *dev_priv = stream->dev_priv;
+
+	/* Note: the oa_buffer_is_empty() condition is ok to run unlocked as it
+	 * just performs mmio reads of the OA buffer head + tail pointers and
+	 * it's assumed we're handling some operation that implies the stream
+	 * can't be destroyed until completion (such as a read()) that ensures
+	 * the device + OA buffer can't disappear
+	 */
+	return wait_event_interruptible(dev_priv->perf.oa.poll_wq,
+					!dev_priv->perf.oa.ops.oa_buffer_is_empty(dev_priv));
+}
+
+static void i915_oa_poll_wait(struct i915_perf_stream *stream,
+			      struct file *file,
+			      poll_table *wait)
+{
+	struct drm_i915_private *dev_priv = stream->dev_priv;
+
+	poll_wait(file, &dev_priv->perf.oa.poll_wq, wait);
+}
+
+static int i915_oa_read(struct i915_perf_stream *stream,
+			struct i915_perf_read_state *read_state)
+{
+	struct drm_i915_private *dev_priv = stream->dev_priv;
+
+	return dev_priv->perf.oa.ops.read(stream, read_state);
+}
+
+static void
+free_oa_buffer(struct drm_i915_private *i915)
+{
+	mutex_lock(&i915->dev->struct_mutex);
+
+	vunmap(i915->perf.oa.oa_buffer.addr);
+	i915_gem_object_ggtt_unpin(i915->perf.oa.oa_buffer.obj);
+	drm_gem_object_unreference(&i915->perf.oa.oa_buffer.obj->base);
+
+	i915->perf.oa.oa_buffer.obj = NULL;
+	i915->perf.oa.oa_buffer.gtt_offset = 0;
+	i915->perf.oa.oa_buffer.addr = NULL;
+
+	mutex_unlock(&i915->dev->struct_mutex);
+}
+
+static void i915_oa_stream_destroy(struct i915_perf_stream *stream)
+{
+	struct drm_i915_private *dev_priv = stream->dev_priv;
+
+	BUG_ON(stream != dev_priv->perf.oa.exclusive_stream);
+
+	dev_priv->perf.oa.ops.disable_metric_set(dev_priv);
+
+	free_oa_buffer(dev_priv);
+
+	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+	intel_runtime_pm_put(dev_priv);
+
+	dev_priv->perf.oa.exclusive_stream = NULL;
+}
+
+static void *vmap_oa_buffer(struct drm_i915_gem_object *obj)
+{
+	int i;
+	void *addr = NULL;
+	struct sg_page_iter sg_iter;
+	struct page **pages;
+
+	pages = drm_malloc_ab(obj->base.size >> PAGE_SHIFT, sizeof(*pages));
+	if (pages == NULL) {
+		DRM_DEBUG_DRIVER("Failed to get space for pages\n");
+		goto finish;
+	}
+
+	i = 0;
+	for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, 0) {
+		pages[i] = sg_page_iter_page(&sg_iter);
+		i++;
+	}
+
+	addr = vmap(pages, i, 0, PAGE_KERNEL);
+	if (addr == NULL) {
+		DRM_DEBUG_DRIVER("Failed to vmap pages\n");
+		goto finish;
+	}
+
+finish:
+	if (pages)
+		drm_free_large(pages);
+	return addr;
+}
+
+static void gen7_init_oa_buffer(struct drm_i915_private *dev_priv)
+{
+	/* Pre-DevBDW: OABUFFER must be set with counters off,
+	 * before OASTATUS1, but after OASTATUS2
+	 */
+	I915_WRITE(GEN7_OASTATUS2, dev_priv->perf.oa.oa_buffer.gtt_offset |
+		   OA_MEM_SELECT_GGTT); /* head */
+	I915_WRITE(GEN7_OABUFFER, dev_priv->perf.oa.oa_buffer.gtt_offset);
+	I915_WRITE(GEN7_OASTATUS1, dev_priv->perf.oa.oa_buffer.gtt_offset |
+		   OABUFFER_SIZE_16M); /* tail */
+}
+
+static int alloc_oa_buffer(struct drm_i915_private *dev_priv)
+{
+	struct drm_i915_gem_object *bo;
+	int ret;
+
+	BUG_ON(dev_priv->perf.oa.oa_buffer.obj);
+
+	ret = i915_mutex_lock_interruptible(dev_priv->dev);
+	if (ret)
+		return ret;
+
+	bo = i915_gem_alloc_object(dev_priv->dev, OA_BUFFER_SIZE);
+	if (bo == NULL) {
+		DRM_ERROR("Failed to allocate OA buffer\n");
+		ret = -ENOMEM;
+		goto unlock;
+	}
+	dev_priv->perf.oa.oa_buffer.obj = bo;
+
+	ret = i915_gem_object_set_cache_level(bo, I915_CACHE_LLC);
+	if (ret)
+		goto err_unref;
+
+	/* PreHSW required 512K alignment, HSW requires 16M */
+	ret = i915_gem_obj_ggtt_pin(bo, SZ_16M, 0);
+	if (ret)
+		goto err_unref;
+
+	dev_priv->perf.oa.oa_buffer.gtt_offset = i915_gem_obj_ggtt_offset(bo);
+	dev_priv->perf.oa.oa_buffer.addr = vmap_oa_buffer(bo);
+
+	dev_priv->perf.oa.ops.init_oa_buffer(dev_priv);
+
+	DRM_DEBUG_DRIVER("OA Buffer initialized, gtt offset = 0x%x, vaddr = %p",
+			 dev_priv->perf.oa.oa_buffer.gtt_offset,
+			 dev_priv->perf.oa.oa_buffer.addr);
+
+	goto unlock;
+
+err_unref:
+	drm_gem_object_unreference(&bo->base);
+
+unlock:
+	mutex_unlock(&dev_priv->dev->struct_mutex);
+	return ret;
+}
+
+static void config_oa_regs(struct drm_i915_private *dev_priv,
+			   const struct i915_oa_reg *regs,
+			   int n_regs)
+{
+	int i;
+
+	for (i = 0; i < n_regs; i++) {
+		const struct i915_oa_reg *reg = regs + i;
+
+		I915_WRITE(reg->addr, reg->value);
+	}
+}
+
+static int hsw_enable_metric_set(struct drm_i915_private *dev_priv)
+{
+	int ret = i915_oa_select_metric_set_hsw(dev_priv);
+
+	if (ret)
+		return ret;
+
+	I915_WRITE(GDT_CHICKEN_BITS, GT_NOA_ENABLE);
+
+	/* PRM:
+	 *
+	 * OA unit is using “crclk” for its functionality. When trunk
+	 * level clock gating takes place, OA clock would be gated,
+	 * unable to count the events from non-render clock domain.
+	 * Render clock gating must be disabled when OA is enabled to
+	 * count the events from non-render domain. Unit level clock
+	 * gating for RCS should also be disabled.
+	 */
+	I915_WRITE(GEN7_MISCCPCTL, (I915_READ(GEN7_MISCCPCTL) &
+				    ~GEN7_DOP_CLOCK_GATE_ENABLE));
+	I915_WRITE(GEN6_UCGCTL1, (I915_READ(GEN6_UCGCTL1) |
+				  GEN6_CSUNIT_CLOCK_GATE_DISABLE));
+
+	config_oa_regs(dev_priv, dev_priv->perf.oa.mux_regs,
+		       dev_priv->perf.oa.mux_regs_len);
+	config_oa_regs(dev_priv, dev_priv->perf.oa.b_counter_regs,
+		       dev_priv->perf.oa.b_counter_regs_len);
+
+	return 0;
+}
+
+static void hsw_disable_metric_set(struct drm_i915_private *dev_priv)
+{
+	I915_WRITE(GEN6_UCGCTL1, (I915_READ(GEN6_UCGCTL1) &
+				  ~GEN6_CSUNIT_CLOCK_GATE_DISABLE));
+	I915_WRITE(GEN7_MISCCPCTL, (I915_READ(GEN7_MISCCPCTL) |
+				    GEN7_DOP_CLOCK_GATE_ENABLE));
+
+	I915_WRITE(GDT_CHICKEN_BITS, (I915_READ(GDT_CHICKEN_BITS) &
+				      ~GT_NOA_ENABLE));
+}
+
+static void gen7_update_oacontrol_locked(struct drm_i915_private *dev_priv)
+{
+	assert_spin_locked(&dev_priv->perf.hook_lock);
+
+	if (dev_priv->perf.oa.exclusive_stream->enabled) {
+		unsigned long ctx_id = 0;
+		bool pinning_ok = false;
+
+		if (dev_priv->perf.oa.exclusive_stream->ctx &&
+		    dev_priv->perf.oa.specific_ctx_id) {
+			ctx_id = dev_priv->perf.oa.specific_ctx_id;
+			pinning_ok = true;
+		}
+
+		if (dev_priv->perf.oa.exclusive_stream->ctx == NULL ||
+		    pinning_ok) {
+			bool periodic = dev_priv->perf.oa.periodic;
+			u32 period_exponent = dev_priv->perf.oa.period_exponent;
+			u32 report_format = dev_priv->perf.oa.oa_buffer.format;
+
+			I915_WRITE(GEN7_OACONTROL,
+				   (ctx_id & GEN7_OACONTROL_CTX_MASK) |
+				   (period_exponent <<
+				    GEN7_OACONTROL_TIMER_PERIOD_SHIFT) |
+				   (periodic ?
+				    GEN7_OACONTROL_TIMER_ENABLE : 0) |
+				   (report_format <<
+				    GEN7_OACONTROL_FORMAT_SHIFT) |
+				   (ctx_id ?
+				    GEN7_OACONTROL_PER_CTX_ENABLE : 0) |
+				   GEN7_OACONTROL_ENABLE);
+			return;
+		}
+	}
+
+	I915_WRITE(GEN7_OACONTROL, 0);
+}
+
+static void gen7_oa_enable(struct drm_i915_private *dev_priv)
+{
+	unsigned long flags;
+	u32 oastatus1, tail;
+
+	spin_lock_irqsave(&dev_priv->perf.hook_lock, flags);
+	gen7_update_oacontrol_locked(dev_priv);
+	spin_unlock_irqrestore(&dev_priv->perf.hook_lock, flags);
+
+	/* Reset the head ptr so we don't forward reports from before now. */
+	oastatus1 = I915_READ(GEN7_OASTATUS1);
+	tail = oastatus1 & GEN7_OASTATUS1_TAIL_MASK;
+	I915_WRITE(GEN7_OASTATUS2, (tail & GEN7_OASTATUS2_HEAD_MASK) |
+				    OA_MEM_SELECT_GGTT);
+}
+
+static void i915_oa_stream_enable(struct i915_perf_stream *stream)
+{
+	struct drm_i915_private *dev_priv = stream->dev_priv;
+
+	dev_priv->perf.oa.ops.oa_enable(dev_priv);
+
+	if (dev_priv->perf.oa.periodic)
+		hrtimer_start(&dev_priv->perf.oa.poll_check_timer,
+			      ns_to_ktime(POLL_PERIOD),
+			      HRTIMER_MODE_REL_PINNED);
+}
+
+static void gen7_oa_disable(struct drm_i915_private *dev_priv)
+{
+	I915_WRITE(GEN7_OACONTROL, 0);
+}
+
+static void i915_oa_stream_disable(struct i915_perf_stream *stream)
+{
+	struct drm_i915_private *dev_priv = stream->dev_priv;
+
+	dev_priv->perf.oa.ops.oa_disable(dev_priv);
+
+	if (dev_priv->perf.oa.periodic)
+		hrtimer_cancel(&dev_priv->perf.oa.poll_check_timer);
+}
+
+static int i915_oa_stream_init(struct i915_perf_stream *stream,
+			       struct drm_i915_perf_open_param *param,
+			       struct perf_open_properties *props)
+{
+	struct drm_i915_private *dev_priv = stream->dev_priv;
+	int format_size;
+	int ret;
+
+	if (!(props->sample_flags & SAMPLE_OA_REPORT)) {
+		DRM_ERROR("Only OA report sampling supported\n");
+		return -EINVAL;
+	}
+
+	if (!dev_priv->perf.oa.ops.init_oa_buffer) {
+		DRM_ERROR("OA unit not supported\n");
+		return -ENODEV;
+	}
+
+	/* To avoid the complexity of having to accurately filter
+	 * counter reports and marshal to the appropriate client
+	 * we currently only allow exclusive access
+	 */
+	if (dev_priv->perf.oa.exclusive_stream) {
+		DRM_ERROR("OA unit already in use\n");
+		return -EBUSY;
+	}
+
+	if (!props->metrics_set) {
+		DRM_ERROR("OA metric set not specified\n");
+		return -EINVAL;
+	}
+
+	if (!props->oa_format) {
+		DRM_ERROR("OA report format not specified\n");
+		return -EINVAL;
+	}
+
+	stream->sample_size = sizeof(struct drm_i915_perf_record_header);
+
+	format_size = dev_priv->perf.oa.oa_formats[props->oa_format].size;
+
+	stream->sample_flags |= SAMPLE_OA_REPORT;
+	stream->sample_size += format_size;
+
+	dev_priv->perf.oa.oa_buffer.format_size = format_size;
+	BUG_ON(dev_priv->perf.oa.oa_buffer.format_size == 0);
+
+	dev_priv->perf.oa.oa_buffer.format =
+		dev_priv->perf.oa.oa_formats[props->oa_format].format;
+
+	dev_priv->perf.oa.metrics_set = props->metrics_set;
+
+	dev_priv->perf.oa.periodic = props->oa_periodic;
+	if (dev_priv->perf.oa.periodic)
+		dev_priv->perf.oa.period_exponent = props->oa_period_exponent;
+
+	ret = alloc_oa_buffer(dev_priv);
+	if (ret)
+		return ret;
+
+	dev_priv->perf.oa.exclusive_stream = stream;
+
+	/* PRM - observability performance counters:
+	 *
+	 *   OACONTROL, performance counter enable, note:
+	 *
+	 *   "When this bit is set, in order to have coherent counts,
+	 *   RC6 power state and trunk clock gating must be disabled.
+	 *   This can be achieved by programming MMIO registers as
+	 *   0xA094=0 and 0xA090[31]=1"
+	 *
+	 *   In our case we are expecting that taking pm + FORCEWAKE
+	 *   references will effectively disable RC6.
+	 */
+	intel_runtime_pm_get(dev_priv);
+	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+
+	dev_priv->perf.oa.ops.enable_metric_set(dev_priv);
+
+	stream->destroy = i915_oa_stream_destroy;
+	stream->enable = i915_oa_stream_enable;
+	stream->disable = i915_oa_stream_disable;
+	stream->can_read = i915_oa_can_read;
+	stream->wait_unlocked = i915_oa_wait_unlocked;
+	stream->poll_wait = i915_oa_poll_wait;
+	stream->read = i915_oa_read;
+
+	return 0;
+}
+
+static void gen7_update_hw_ctx_id_locked(struct drm_i915_private *dev_priv,
+					 u32 ctx_id)
+{
+	assert_spin_locked(&dev_priv->perf.hook_lock);
+
+	dev_priv->perf.oa.specific_ctx_id = ctx_id;
+	gen7_update_oacontrol_locked(dev_priv);
+}
+
+static void i915_oa_context_pin_notify_locked(struct drm_i915_private *dev_priv,
+					      struct intel_context *context)
+{
+	assert_spin_locked(&dev_priv->perf.hook_lock);
+
+	if (i915.enable_execlists ||
+	    dev_priv->perf.oa.ops.update_hw_ctx_id_locked == NULL)
+		return;
+
+	if (dev_priv->perf.oa.exclusive_stream &&
+	    dev_priv->perf.oa.exclusive_stream->ctx == context) {
+		struct drm_i915_gem_object *obj =
+			context->legacy_hw_ctx.rcs_state;
+		u32 ctx_id = i915_gem_obj_ggtt_offset(obj);
+
+		dev_priv->perf.oa.ops.update_hw_ctx_id_locked(dev_priv, ctx_id);
+	}
+}
+
+void i915_oa_context_pin_notify(struct drm_i915_private *dev_priv,
+				struct intel_context *context)
+{
+	unsigned long flags;
+
+	if (!dev_priv->perf.initialized)
+		return;
+
+	spin_lock_irqsave(&dev_priv->perf.hook_lock, flags);
+	i915_oa_context_pin_notify_locked(dev_priv, context);
+	spin_unlock_irqrestore(&dev_priv->perf.hook_lock, flags);
+}
+
 static ssize_t i915_perf_read_locked(struct i915_perf_stream *stream,
 				     struct file *file,
 				     char __user *buf,
@@ -86,6 +764,20 @@  static ssize_t i915_perf_read(struct file *file,
 	return ret;
 }
 
+static enum hrtimer_restart poll_check_timer_cb(struct hrtimer *hrtimer)
+{
+	struct drm_i915_private *dev_priv =
+		container_of(hrtimer, typeof(*dev_priv),
+			     perf.oa.poll_check_timer);
+
+	if (!dev_priv->perf.oa.ops.oa_buffer_is_empty(dev_priv))
+		wake_up(&dev_priv->perf.oa.poll_wq);
+
+	hrtimer_forward_now(hrtimer, ns_to_ktime(POLL_PERIOD));
+
+	return HRTIMER_RESTART;
+}
+
 static unsigned int i915_perf_poll_locked(struct i915_perf_stream *stream,
 					  struct file *file,
 					  poll_table *wait)
@@ -276,18 +968,18 @@  int i915_perf_open_ioctl_locked(struct drm_device *dev,
 		goto err_ctx;
 	}
 
-	stream->sample_flags = props->sample_flags;
 	stream->dev_priv = dev_priv;
 	stream->ctx = specific_ctx;
 
-	/*
-	 * TODO: support sampling something
-	 *
-	 * For now this is as far as we can go.
+	ret = i915_oa_stream_init(stream, param, props);
+	if (ret)
+		goto err_alloc;
+
+	/* we avoid simply assigning stream->sample_flags = props->sample_flags
+	 * to have _stream_init check the combination of sample flags more
+	 * thoroughly, but still this is the expected result at this point.
 	 */
-	DRM_ERROR("Unsupported i915 perf stream configuration\n");
-	ret = -EINVAL;
-	goto err_alloc;
+	BUG_ON(stream->sample_flags != props->sample_flags);
 
 	list_add(&stream->link, &dev_priv->perf.streams);
 
@@ -372,7 +1064,53 @@  static int read_properties_unlocked(struct drm_i915_private *dev_priv,
 			props->single_context = 1;
 			props->ctx_handle = value;
 			break;
-
+		case DRM_I915_PERF_SAMPLE_OA_PROP:
+			props->sample_flags |= SAMPLE_OA_REPORT;
+			break;
+		case DRM_I915_PERF_OA_METRICS_SET_PROP:
+			if (value == 0 ||
+			    value > dev_priv->perf.oa.n_builtin_sets) {
+				DRM_ERROR("Unknown OA metric set ID");
+				return -EINVAL;
+			}
+			props->metrics_set = value;
+			break;
+		case DRM_I915_PERF_OA_FORMAT_PROP:
+			if (value == 0 || value >= I915_OA_FORMAT_MAX) {
+				DRM_ERROR("Invalid OA report format\n");
+				return -EINVAL;
+			}
+			if (!dev_priv->perf.oa.oa_formats[value].size) {
+				DRM_ERROR("Invalid OA report format\n");
+				return -EINVAL;
+			}
+			props->oa_format = value;
+			break;
+		case DRM_I915_PERF_OA_EXPONENT_PROP:
+			if (value > OA_EXPONENT_MAX)
+				return -EINVAL;
+
+			/* NB: The exponent represents a period as follows:
+			 *
+			 *   80ns * 2^(period_exponent + 1)
+			 *
+			 * Theoretically we can program the OA unit to sample
+			 * every 160ns but don't allow that by default unless
+			 * root.
+			 *
+			 * Referring to perf's
+			 * kernel.perf_event_max_sample_rate for a precedent
+			 * (100000 by default); with an OA exponent of 6 we get
+			 * a period of 10.240 microseconds -just under 100000Hz
+			 */
+			if (value < 6 && !capable(CAP_SYS_ADMIN)) {
+				DRM_ERROR("Sampling period too high without root privileges\n");
+				return -EACCES;
+			}
+
+			props->oa_periodic = true;
+			props->oa_period_exponent = value;
+			break;
 		case DRM_I915_PERF_PROP_MAX:
 			BUG();
 		}
@@ -423,8 +1161,33 @@  void i915_perf_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
 
+	if (!IS_HASWELL(dev))
+		return;
+
+	hrtimer_init(&dev_priv->perf.oa.poll_check_timer,
+		     CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	dev_priv->perf.oa.poll_check_timer.function = poll_check_timer_cb;
+	init_waitqueue_head(&dev_priv->perf.oa.poll_wq);
+
 	INIT_LIST_HEAD(&dev_priv->perf.streams);
 	mutex_init(&dev_priv->perf.lock);
+	spin_lock_init(&dev_priv->perf.hook_lock);
+
+	dev_priv->perf.oa.ops.init_oa_buffer = gen7_init_oa_buffer;
+	dev_priv->perf.oa.ops.enable_metric_set = hsw_enable_metric_set;
+	dev_priv->perf.oa.ops.disable_metric_set = hsw_disable_metric_set;
+	dev_priv->perf.oa.ops.oa_enable = gen7_oa_enable;
+	dev_priv->perf.oa.ops.oa_disable = gen7_oa_disable;
+	dev_priv->perf.oa.ops.update_hw_ctx_id_locked = gen7_update_hw_ctx_id_locked;
+	dev_priv->perf.oa.ops.read = gen7_oa_read;
+	dev_priv->perf.oa.ops.oa_buffer_is_empty = gen7_oa_buffer_is_empty;
+
+	dev_priv->perf.oa.oa_formats = hsw_oa_formats;
+
+	dev_priv->perf.oa.n_builtin_sets =
+		i915_oa_n_builtin_metric_sets_hsw;
+
+	dev_priv->perf.oa.oa_formats = hsw_oa_formats;
 
 	dev_priv->perf.initialized = true;
 }
@@ -436,7 +1199,7 @@  void i915_perf_fini(struct drm_device *dev)
 	if (!dev_priv->perf.initialized)
 		return;
 
-	/* Currently nothing to clean up */
+	dev_priv->perf.oa.ops.init_oa_buffer = NULL;
 
 	dev_priv->perf.initialized = false;
 }
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index e64c866..3d9841c 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -541,6 +541,343 @@ 
 #define GEN7_GPGPU_DISPATCHDIMZ         0x2508
 
 #define GEN7_OACONTROL 0x2360
+#define  GEN7_OACONTROL_CTX_MASK	    0xFFFFF000
+#define  GEN7_OACONTROL_TIMER_PERIOD_MASK   0x3F
+#define  GEN7_OACONTROL_TIMER_PERIOD_SHIFT  6
+#define  GEN7_OACONTROL_TIMER_ENABLE	    (1<<5)
+#define  GEN7_OACONTROL_FORMAT_A13	    (0<<2)
+#define  GEN7_OACONTROL_FORMAT_A29	    (1<<2)
+#define  GEN7_OACONTROL_FORMAT_A13_B8_C8    (2<<2)
+#define  GEN7_OACONTROL_FORMAT_A29_B8_C8    (3<<2)
+#define  GEN7_OACONTROL_FORMAT_B4_C8	    (4<<2)
+#define  GEN7_OACONTROL_FORMAT_A45_B8_C8    (5<<2)
+#define  GEN7_OACONTROL_FORMAT_B4_C8_A16    (6<<2)
+#define  GEN7_OACONTROL_FORMAT_C4_B8	    (7<<2)
+#define  GEN7_OACONTROL_FORMAT_SHIFT	    2
+#define  GEN7_OACONTROL_PER_CTX_ENABLE	    (1<<1)
+#define  GEN7_OACONTROL_ENABLE		    (1<<0)
+
+#define GEN8_OACTXID 0x2364
+
+#define GEN8_OACONTROL 0x2B00
+#define  GEN8_OA_REPORT_FORMAT_A12	    (0<<2)
+#define  GEN8_OA_REPORT_FORMAT_A12_B8_C8    (2<<2)
+#define  GEN8_OA_REPORT_FORMAT_A36_B8_C8    (5<<2)
+#define  GEN8_OA_REPORT_FORMAT_C4_B8	    (7<<2)
+#define  GEN8_OA_REPORT_FORMAT_SHIFT	    2
+#define  GEN8_OA_SPECIFIC_CONTEXT_ENABLE    (1<<1)
+#define  GEN8_OA_COUNTER_ENABLE             (1<<0)
+
+#define GEN8_OACTXCONTROL 0x2360
+#define  GEN8_OA_TIMER_PERIOD_MASK	    0x3F
+#define  GEN8_OA_TIMER_PERIOD_SHIFT	    2
+#define  GEN8_OA_TIMER_ENABLE		    (1<<1)
+#define  GEN8_OA_COUNTER_RESUME		    (1<<0)
+
+#define GEN7_OABUFFER 0x23B0 /* R/W */
+#define  GEN7_OABUFFER_OVERRUN_DISABLE	    (1<<3)
+#define  GEN7_OABUFFER_EDGE_TRIGGER	    (1<<2)
+#define  GEN7_OABUFFER_STOP_RESUME_ENABLE   (1<<1)
+#define  GEN7_OABUFFER_RESUME		    (1<<0)
+
+#define GEN8_OABUFFER 0x2b14
+
+#define GEN7_OASTATUS1 0x2364
+#define  GEN7_OASTATUS1_TAIL_MASK	    0xffffffc0
+#define  GEN7_OASTATUS1_COUNTER_OVERFLOW    (1<<2)
+#define  GEN7_OASTATUS1_OABUFFER_OVERFLOW   (1<<1)
+#define  GEN7_OASTATUS1_REPORT_LOST	    (1<<0)
+
+#define GEN7_OASTATUS2 0x2368
+#define GEN7_OASTATUS2_HEAD_MASK    0xffffffc0
+
+#define GEN8_OASTATUS 0x2b08
+#define  GEN8_OASTATUS_OVERRUN_STATUS	    (1<<3)
+#define  GEN8_OASTATUS_COUNTER_OVERFLOW     (1<<2)
+#define  GEN8_OASTATUS_OABUFFER_OVERFLOW    (1<<1)
+#define  GEN8_OASTATUS_REPORT_LOST	    (1<<0)
+
+#define GEN8_OAHEADPTR 0x2B0C
+#define GEN8_OATAILPTR 0x2B10
+
+#define OABUFFER_SIZE_128K  (0<<3)
+#define OABUFFER_SIZE_256K  (1<<3)
+#define OABUFFER_SIZE_512K  (2<<3)
+#define OABUFFER_SIZE_1M    (3<<3)
+#define OABUFFER_SIZE_2M    (4<<3)
+#define OABUFFER_SIZE_4M    (5<<3)
+#define OABUFFER_SIZE_8M    (6<<3)
+#define OABUFFER_SIZE_16M   (7<<3)
+
+#define OA_MEM_SELECT_GGTT  (1<<0)
+
+#define EU_PERF_CNTL0	    0xe458
+
+#define GDT_CHICKEN_BITS    0x9840
+#define GT_NOA_ENABLE	    0x00000080
+
+/*
+ * OA Boolean state
+ */
+
+#define OAREPORTTRIG1 0x2740
+#define OAREPORTTRIG1_THRESHOLD_MASK 0xffff
+#define OAREPORTTRIG1_EDGE_LEVEL_TRIGER_SELECT_MASK 0xffff0000 /* 0=level */
+
+#define OAREPORTTRIG2 0x2744
+#define OAREPORTTRIG2_INVERT_A_0  (1<<0)
+#define OAREPORTTRIG2_INVERT_A_1  (1<<1)
+#define OAREPORTTRIG2_INVERT_A_2  (1<<2)
+#define OAREPORTTRIG2_INVERT_A_3  (1<<3)
+#define OAREPORTTRIG2_INVERT_A_4  (1<<4)
+#define OAREPORTTRIG2_INVERT_A_5  (1<<5)
+#define OAREPORTTRIG2_INVERT_A_6  (1<<6)
+#define OAREPORTTRIG2_INVERT_A_7  (1<<7)
+#define OAREPORTTRIG2_INVERT_A_8  (1<<8)
+#define OAREPORTTRIG2_INVERT_A_9  (1<<9)
+#define OAREPORTTRIG2_INVERT_A_10 (1<<10)
+#define OAREPORTTRIG2_INVERT_A_11 (1<<11)
+#define OAREPORTTRIG2_INVERT_A_12 (1<<12)
+#define OAREPORTTRIG2_INVERT_A_13 (1<<13)
+#define OAREPORTTRIG2_INVERT_A_14 (1<<14)
+#define OAREPORTTRIG2_INVERT_A_15 (1<<15)
+#define OAREPORTTRIG2_INVERT_B_0  (1<<16)
+#define OAREPORTTRIG2_INVERT_B_1  (1<<17)
+#define OAREPORTTRIG2_INVERT_B_2  (1<<18)
+#define OAREPORTTRIG2_INVERT_B_3  (1<<19)
+#define OAREPORTTRIG2_INVERT_C_0  (1<<20)
+#define OAREPORTTRIG2_INVERT_C_1  (1<<21)
+#define OAREPORTTRIG2_INVERT_D_0  (1<<22)
+#define OAREPORTTRIG2_THRESHOLD_ENABLE	    (1<<23)
+#define OAREPORTTRIG2_REPORT_TRIGGER_ENABLE (1<<31)
+
+#define OAREPORTTRIG3 0x2748
+#define OAREPORTTRIG3_NOA_SELECT_MASK	    0xf
+#define OAREPORTTRIG3_NOA_SELECT_8_SHIFT    0
+#define OAREPORTTRIG3_NOA_SELECT_9_SHIFT    4
+#define OAREPORTTRIG3_NOA_SELECT_10_SHIFT   8
+#define OAREPORTTRIG3_NOA_SELECT_11_SHIFT   12
+#define OAREPORTTRIG3_NOA_SELECT_12_SHIFT   16
+#define OAREPORTTRIG3_NOA_SELECT_13_SHIFT   20
+#define OAREPORTTRIG3_NOA_SELECT_14_SHIFT   24
+#define OAREPORTTRIG3_NOA_SELECT_15_SHIFT   28
+
+#define OAREPORTTRIG4 0x274c
+#define OAREPORTTRIG4_NOA_SELECT_MASK	    0xf
+#define OAREPORTTRIG4_NOA_SELECT_0_SHIFT    0
+#define OAREPORTTRIG4_NOA_SELECT_1_SHIFT    4
+#define OAREPORTTRIG4_NOA_SELECT_2_SHIFT    8
+#define OAREPORTTRIG4_NOA_SELECT_3_SHIFT    12
+#define OAREPORTTRIG4_NOA_SELECT_4_SHIFT    16
+#define OAREPORTTRIG4_NOA_SELECT_5_SHIFT    20
+#define OAREPORTTRIG4_NOA_SELECT_6_SHIFT    24
+#define OAREPORTTRIG4_NOA_SELECT_7_SHIFT    28
+
+#define OAREPORTTRIG5 0x2750
+#define OAREPORTTRIG5_THRESHOLD_MASK 0xffff
+#define OAREPORTTRIG5_EDGE_LEVEL_TRIGER_SELECT_MASK 0xffff0000 /* 0=level */
+
+#define OAREPORTTRIG6 0x2754
+#define OAREPORTTRIG6_INVERT_A_0  (1<<0)
+#define OAREPORTTRIG6_INVERT_A_1  (1<<1)
+#define OAREPORTTRIG6_INVERT_A_2  (1<<2)
+#define OAREPORTTRIG6_INVERT_A_3  (1<<3)
+#define OAREPORTTRIG6_INVERT_A_4  (1<<4)
+#define OAREPORTTRIG6_INVERT_A_5  (1<<5)
+#define OAREPORTTRIG6_INVERT_A_6  (1<<6)
+#define OAREPORTTRIG6_INVERT_A_7  (1<<7)
+#define OAREPORTTRIG6_INVERT_A_8  (1<<8)
+#define OAREPORTTRIG6_INVERT_A_9  (1<<9)
+#define OAREPORTTRIG6_INVERT_A_10 (1<<10)
+#define OAREPORTTRIG6_INVERT_A_11 (1<<11)
+#define OAREPORTTRIG6_INVERT_A_12 (1<<12)
+#define OAREPORTTRIG6_INVERT_A_13 (1<<13)
+#define OAREPORTTRIG6_INVERT_A_14 (1<<14)
+#define OAREPORTTRIG6_INVERT_A_15 (1<<15)
+#define OAREPORTTRIG6_INVERT_B_0  (1<<16)
+#define OAREPORTTRIG6_INVERT_B_1  (1<<17)
+#define OAREPORTTRIG6_INVERT_B_2  (1<<18)
+#define OAREPORTTRIG6_INVERT_B_3  (1<<19)
+#define OAREPORTTRIG6_INVERT_C_0  (1<<20)
+#define OAREPORTTRIG6_INVERT_C_1  (1<<21)
+#define OAREPORTTRIG6_INVERT_D_0  (1<<22)
+#define OAREPORTTRIG6_THRESHOLD_ENABLE	    (1<<23)
+#define OAREPORTTRIG6_REPORT_TRIGGER_ENABLE (1<<31)
+
+#define OAREPORTTRIG7 0x2758
+#define OAREPORTTRIG7_NOA_SELECT_MASK	    0xf
+#define OAREPORTTRIG7_NOA_SELECT_8_SHIFT    0
+#define OAREPORTTRIG7_NOA_SELECT_9_SHIFT    4
+#define OAREPORTTRIG7_NOA_SELECT_10_SHIFT   8
+#define OAREPORTTRIG7_NOA_SELECT_11_SHIFT   12
+#define OAREPORTTRIG7_NOA_SELECT_12_SHIFT   16
+#define OAREPORTTRIG7_NOA_SELECT_13_SHIFT   20
+#define OAREPORTTRIG7_NOA_SELECT_14_SHIFT   24
+#define OAREPORTTRIG7_NOA_SELECT_15_SHIFT   28
+
+#define OAREPORTTRIG8 0x275c
+#define OAREPORTTRIG8_NOA_SELECT_MASK	    0xf
+#define OAREPORTTRIG8_NOA_SELECT_0_SHIFT    0
+#define OAREPORTTRIG8_NOA_SELECT_1_SHIFT    4
+#define OAREPORTTRIG8_NOA_SELECT_2_SHIFT    8
+#define OAREPORTTRIG8_NOA_SELECT_3_SHIFT    12
+#define OAREPORTTRIG8_NOA_SELECT_4_SHIFT    16
+#define OAREPORTTRIG8_NOA_SELECT_5_SHIFT    20
+#define OAREPORTTRIG8_NOA_SELECT_6_SHIFT    24
+#define OAREPORTTRIG8_NOA_SELECT_7_SHIFT    28
+
+#define OASTARTTRIG1 0x2710
+#define OASTARTTRIG1_THRESHOLD_COUNT_MASK_MBZ 0xffff0000
+#define OASTARTTRIG1_THRESHOLD_MASK	      0xffff
+
+#define OASTARTTRIG2 0x2714
+#define OASTARTTRIG2_INVERT_A_0 (1<<0)
+#define OASTARTTRIG2_INVERT_A_1 (1<<1)
+#define OASTARTTRIG2_INVERT_A_2 (1<<2)
+#define OASTARTTRIG2_INVERT_A_3 (1<<3)
+#define OASTARTTRIG2_INVERT_A_4 (1<<4)
+#define OASTARTTRIG2_INVERT_A_5 (1<<5)
+#define OASTARTTRIG2_INVERT_A_6 (1<<6)
+#define OASTARTTRIG2_INVERT_A_7 (1<<7)
+#define OASTARTTRIG2_INVERT_A_8 (1<<8)
+#define OASTARTTRIG2_INVERT_A_9 (1<<9)
+#define OASTARTTRIG2_INVERT_A_10 (1<<10)
+#define OASTARTTRIG2_INVERT_A_11 (1<<11)
+#define OASTARTTRIG2_INVERT_A_12 (1<<12)
+#define OASTARTTRIG2_INVERT_A_13 (1<<13)
+#define OASTARTTRIG2_INVERT_A_14 (1<<14)
+#define OASTARTTRIG2_INVERT_A_15 (1<<15)
+#define OASTARTTRIG2_INVERT_B_0 (1<<16)
+#define OASTARTTRIG2_INVERT_B_1 (1<<17)
+#define OASTARTTRIG2_INVERT_B_2 (1<<18)
+#define OASTARTTRIG2_INVERT_B_3 (1<<19)
+#define OASTARTTRIG2_INVERT_C_0 (1<<20)
+#define OASTARTTRIG2_INVERT_C_1 (1<<21)
+#define OASTARTTRIG2_INVERT_D_0 (1<<22)
+#define OASTARTTRIG2_THRESHOLD_ENABLE	    (1<<23)
+#define OASTARTTRIG2_START_TRIG_FLAG_MBZ    (1<<24)
+#define OASTARTTRIG2_EVENT_SELECT_0  (1<<28)
+#define OASTARTTRIG2_EVENT_SELECT_1  (1<<29)
+#define OASTARTTRIG2_EVENT_SELECT_2  (1<<30)
+#define OASTARTTRIG2_EVENT_SELECT_3  (1<<31)
+
+#define OASTARTTRIG3 0x2718
+#define OASTARTTRIG3_NOA_SELECT_MASK	   0xf
+#define OASTARTTRIG3_NOA_SELECT_8_SHIFT    0
+#define OASTARTTRIG3_NOA_SELECT_9_SHIFT    4
+#define OASTARTTRIG3_NOA_SELECT_10_SHIFT   8
+#define OASTARTTRIG3_NOA_SELECT_11_SHIFT   12
+#define OASTARTTRIG3_NOA_SELECT_12_SHIFT   16
+#define OASTARTTRIG3_NOA_SELECT_13_SHIFT   20
+#define OASTARTTRIG3_NOA_SELECT_14_SHIFT   24
+#define OASTARTTRIG3_NOA_SELECT_15_SHIFT   28
+
+#define OASTARTTRIG4 0x271c
+#define OASTARTTRIG4_NOA_SELECT_MASK	    0xf
+#define OASTARTTRIG4_NOA_SELECT_0_SHIFT    0
+#define OASTARTTRIG4_NOA_SELECT_1_SHIFT    4
+#define OASTARTTRIG4_NOA_SELECT_2_SHIFT    8
+#define OASTARTTRIG4_NOA_SELECT_3_SHIFT    12
+#define OASTARTTRIG4_NOA_SELECT_4_SHIFT    16
+#define OASTARTTRIG4_NOA_SELECT_5_SHIFT    20
+#define OASTARTTRIG4_NOA_SELECT_6_SHIFT    24
+#define OASTARTTRIG4_NOA_SELECT_7_SHIFT    28
+
+#define OASTARTTRIG5 0x2720
+#define OASTARTTRIG5_THRESHOLD_COUNT_MASK_MBZ 0xffff0000
+#define OASTARTTRIG5_THRESHOLD_MASK	      0xffff
+
+#define OASTARTTRIG6 0x2724
+#define OASTARTTRIG6_INVERT_A_0 (1<<0)
+#define OASTARTTRIG6_INVERT_A_1 (1<<1)
+#define OASTARTTRIG6_INVERT_A_2 (1<<2)
+#define OASTARTTRIG6_INVERT_A_3 (1<<3)
+#define OASTARTTRIG6_INVERT_A_4 (1<<4)
+#define OASTARTTRIG6_INVERT_A_5 (1<<5)
+#define OASTARTTRIG6_INVERT_A_6 (1<<6)
+#define OASTARTTRIG6_INVERT_A_7 (1<<7)
+#define OASTARTTRIG6_INVERT_A_8 (1<<8)
+#define OASTARTTRIG6_INVERT_A_9 (1<<9)
+#define OASTARTTRIG6_INVERT_A_10 (1<<10)
+#define OASTARTTRIG6_INVERT_A_11 (1<<11)
+#define OASTARTTRIG6_INVERT_A_12 (1<<12)
+#define OASTARTTRIG6_INVERT_A_13 (1<<13)
+#define OASTARTTRIG6_INVERT_A_14 (1<<14)
+#define OASTARTTRIG6_INVERT_A_15 (1<<15)
+#define OASTARTTRIG6_INVERT_B_0 (1<<16)
+#define OASTARTTRIG6_INVERT_B_1 (1<<17)
+#define OASTARTTRIG6_INVERT_B_2 (1<<18)
+#define OASTARTTRIG6_INVERT_B_3 (1<<19)
+#define OASTARTTRIG6_INVERT_C_0 (1<<20)
+#define OASTARTTRIG6_INVERT_C_1 (1<<21)
+#define OASTARTTRIG6_INVERT_D_0 (1<<22)
+#define OASTARTTRIG6_THRESHOLD_ENABLE	    (1<<23)
+#define OASTARTTRIG6_START_TRIG_FLAG_MBZ    (1<<24)
+#define OASTARTTRIG6_EVENT_SELECT_4  (1<<28)
+#define OASTARTTRIG6_EVENT_SELECT_5  (1<<29)
+#define OASTARTTRIG6_EVENT_SELECT_6  (1<<30)
+#define OASTARTTRIG6_EVENT_SELECT_7  (1<<31)
+
+#define OASTARTTRIG7 0x2728
+#define OASTARTTRIG7_NOA_SELECT_MASK	   0xf
+#define OASTARTTRIG7_NOA_SELECT_8_SHIFT    0
+#define OASTARTTRIG7_NOA_SELECT_9_SHIFT    4
+#define OASTARTTRIG7_NOA_SELECT_10_SHIFT   8
+#define OASTARTTRIG7_NOA_SELECT_11_SHIFT   12
+#define OASTARTTRIG7_NOA_SELECT_12_SHIFT   16
+#define OASTARTTRIG7_NOA_SELECT_13_SHIFT   20
+#define OASTARTTRIG7_NOA_SELECT_14_SHIFT   24
+#define OASTARTTRIG7_NOA_SELECT_15_SHIFT   28
+
+#define OASTARTTRIG8 0x272c
+#define OASTARTTRIG8_NOA_SELECT_MASK	   0xf
+#define OASTARTTRIG8_NOA_SELECT_0_SHIFT    0
+#define OASTARTTRIG8_NOA_SELECT_1_SHIFT    4
+#define OASTARTTRIG8_NOA_SELECT_2_SHIFT    8
+#define OASTARTTRIG8_NOA_SELECT_3_SHIFT    12
+#define OASTARTTRIG8_NOA_SELECT_4_SHIFT    16
+#define OASTARTTRIG8_NOA_SELECT_5_SHIFT    20
+#define OASTARTTRIG8_NOA_SELECT_6_SHIFT    24
+#define OASTARTTRIG8_NOA_SELECT_7_SHIFT    28
+
+/* CECX_0 */
+#define OACEC_COMPARE_LESS_OR_EQUAL	6
+#define OACEC_COMPARE_NOT_EQUAL		5
+#define OACEC_COMPARE_LESS_THAN		4
+#define OACEC_COMPARE_GREATER_OR_EQUAL	3
+#define OACEC_COMPARE_EQUAL		2
+#define OACEC_COMPARE_GREATER_THAN	1
+#define OACEC_COMPARE_ANY_EQUAL		0
+
+#define OACEC_COMPARE_VALUE_MASK    0xffff
+#define OACEC_COMPARE_VALUE_SHIFT   3
+
+#define OACEC_SELECT_NOA	(0<<19)
+#define OACEC_SELECT_PREV	(1<<19)
+#define OACEC_SELECT_BOOLEAN	(2<<19)
+
+/* CECX_1 */
+#define OACEC_MASK_MASK		    0xffff
+#define OACEC_CONSIDERATIONS_MASK   0xffff
+#define OACEC_CONSIDERATIONS_SHIFT  16
+
+#define OACEC0_0 0x2770
+#define OACEC0_1 0x2774
+#define OACEC1_0 0x2778
+#define OACEC1_1 0x277c
+#define OACEC2_0 0x2780
+#define OACEC2_1 0x2784
+#define OACEC3_0 0x2788
+#define OACEC3_1 0x278c
+#define OACEC4_0 0x2790
+#define OACEC4_1 0x2794
+#define OACEC5_0 0x2798
+#define OACEC5_1 0x279c
+#define OACEC6_0 0x27a0
+#define OACEC6_1 0x27a4
+#define OACEC7_0 0x27a8
+#define OACEC7_1 0x27ac
+
 
 #define _GEN7_PIPEA_DE_LOAD_SL	0x70068
 #define _GEN7_PIPEB_DE_LOAD_SL	0x71068
@@ -6750,6 +7087,7 @@  enum skl_disp_power_wells {
 # define GEN6_RCCUNIT_CLOCK_GATE_DISABLE		(1 << 11)
 
 #define GEN6_UCGCTL3				0x9408
+# define GEN6_OACSUNIT_CLOCK_GATE_DISABLE		(1 << 20)
 
 #define GEN7_UCGCTL4				0x940c
 #define  GEN7_L3BANK2X_CLOCK_GATE_DISABLE	(1<<25)
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index ff4507e..e747702 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -1132,6 +1132,18 @@  struct drm_i915_gem_context_param {
 	__u64 value;
 };
 
+enum drm_i915_oa_format {
+	I915_OA_FORMAT_A13 = 1,
+	I915_OA_FORMAT_A29,
+	I915_OA_FORMAT_A13_B8_C8,
+	I915_OA_FORMAT_B4_C8,
+	I915_OA_FORMAT_A45_B8_C8,
+	I915_OA_FORMAT_B4_C8_A16,
+	I915_OA_FORMAT_C4_B8,
+
+	I915_OA_FORMAT_MAX	    /* non-ABI */
+};
+
 #define I915_PERF_FLAG_FD_CLOEXEC	(1<<0)
 #define I915_PERF_FLAG_FD_NONBLOCK	(1<<1)
 #define I915_PERF_FLAG_DISABLED		(1<<2)
@@ -1144,6 +1156,32 @@  enum drm_i915_perf_property_id {
 	 */
 	DRM_I915_PERF_CTX_HANDLE_PROP = 1,
 
+	/**
+	 * A value of 1 requests the inclusion of raw OA unit reports as
+	 * part of stream samples.
+	 */
+	DRM_I915_PERF_SAMPLE_OA_PROP,
+
+	/**
+	 * The value specifies which set of OA unit metrics should be
+	 * be configured, defining the contents of any OA unit reports.
+	 */
+	DRM_I915_PERF_OA_METRICS_SET_PROP,
+
+	/**
+	 * The value specifies the size and layout of OA unit reports.
+	 */
+	DRM_I915_PERF_OA_FORMAT_PROP,
+
+	/**
+	 * Specifying this property implicitly requests periodic OA unit
+	 * sampling and (at least on Haswell) the sampling frequency is derived
+	 * from this exponent as follows:
+	 *
+	 *   80ns * 2^(period_exponent + 1)
+	 */
+	DRM_I915_PERF_OA_EXPONENT_PROP,
+
 	DRM_I915_PERF_PROP_MAX /* non-ABI */
 };
 
@@ -1185,17 +1223,31 @@  enum drm_i915_perf_record_type {
 	 * every sample.
 	 *
 	 * The order of these sample properties given by userspace has no
-	 * affect on the ordering of data within a sample. The order will be
+	 * affect on the ordering of data within a sample. The order is
 	 * documented here.
 	 *
 	 * struct {
 	 *     struct drm_i915_perf_record_header header;
 	 *
-	 *     TODO: itemize extensible sample data here
+	 *     { u32 oa_report[]; } && DRM_I915_PERF_SAMPLE_OA_PROP
 	 * };
 	 */
 	DRM_I915_PERF_RECORD_SAMPLE = 1,
 
+	/*
+	 * Indicates that one or more OA reports was not written
+	 * by the hardware.
+	 */
+	DRM_I915_PERF_RECORD_OA_REPORT_LOST = 2,
+
+	/*
+	 * Indicates that the internal circular buffer that Gen
+	 * graphics writes OA reports into has filled, which may
+	 * either mean that old reports could be overwritten or
+	 * subsequent reports lost until the buffer is cleared.
+	 */
+	DRM_I915_PERF_RECORD_OA_BUFFER_OVERFLOW = 3,
+
 	DRM_I915_PERF_RECORD_MAX /* non-ABI */
 };