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.
v2:
Make sure to initialize ->specific_ctx_id when opening, without
relying on _pin_notify hook, in case ctx already pinned.
v3:
Revert back to pinning ctx upfront when opening stream, removing
need to hook in to pinning and to update OACONTROL on the fly.
Signed-off-by: Robert Bragg
Signed-off-by: Zhenyu Wang
Cc: Chris Wilson
Reviewed-by: Matthew Auld
---
drivers/gpu/drm/i915/i915_drv.h | 66 ++-
drivers/gpu/drm/i915/i915_perf.c | 1036 +-
drivers/gpu/drm/i915/i915_reg.h | 338 +
include/uapi/drm/i915_drm.h | 71 ++-
4 files changed, 1482 insertions(+), 29 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index bdebb66..8003120 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1785,6 +1785,11 @@ struct intel_wm_config {
bool sprites_scaled;
};
+struct i915_oa_format {
+ u32 format;
+ int size;
+};
+
struct i915_oa_reg {
i915_reg_t addr;
u32 value;
@@ -1805,11 +1810,6 @@ struct i915_perf_stream_ops {
*/
void (*disable)(struct i915_perf_stream *stream);
- /* Return: true if any i915 perf records are ready to read()
-* for this stream.
-*/
- bool (*can_read)(struct i915_perf_stream *stream);
-
/* Call poll_wait, passing a wait queue that will be woken
* once there is something ready to read() for the stream
*/
@@ -1819,9 +1819,7 @@ struct i915_perf_stream_ops {
/* For handling a blocking read, wait until there is something
* to ready to read() for the stream. E.g. wait on the same
-* wait queue that would be passed to poll_wait() until
-* ->can_read() returns true (if its safe to call ->can_read()
-* without the i915 perf lock held).
+* wait queue that would be passed to poll_wait().
*/
int (*wait_unlocked)(struct i915_perf_stream *stream);
@@ -1861,11 +1859,28 @@ struct i915_perf_stream {
struct list_head link;
u32 sample_flags;
+ int sample_size;
struct i915_gem_context *ctx;
bool enabled;
- struct i915_perf_stream_ops *ops;
+ const struct i915_perf_stream_ops *ops;
+};
+
+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,
+ char __user *buf,
+ size_t count,
+ size_t *offset);
+ bool (*oa_buffer_is_empty)(struct drm_i915_private *dev_priv);
};
struct drm_i915_private {
@@ -2171,16 +2186,47 @@ struct drm_i915_private {
struct {
bool initialized;
+
struct mutex lock;
struct list_head streams;
+ spinlock_t hook_lock;
+
struct {
- u32 metrics_set;
+ struct i915_perf_stream *exclusive_stream;
+
+ u32 specific_ctx_id;
+ struct i915_vma *pinned_rcs_vma;
+
+ struct hrtimer poll_check_timer;
+ wait_queue_head_t poll_wq;
+ bool pollin;
+
+ bool periodic;
+ int period_exponent;
+ int timestamp_frequency;
+
+ int tail_margin;
+
+ int 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 i915_vma *vma;
+ u8 *vaddr;
+ int format;
+ int format_size;
+ } oa_buffer;
+
+ u32 gen7_latched_oastatus1;
+
+ struct i915_oa_ops ops;
+ const struct i915_oa_format *oa_formats;
+ int n_builtin_sets;
} oa;
} perf;
diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index