drm/i915: Engine discovery query

Commit Message

Tvrtko Ursulin March 14, 2019, 2:44 p.m. UTC

From: Tvrtko Ursulin <[e-mail]>

Engine discovery query allows userspace to enumerate engines, probe their
configuration features, all without needing to maintain the internal PCI
ID based database.

A new query for the generic i915 query ioctl is added named
DRM_I915_QUERY_ENGINE_INFO, together with accompanying structure
drm_i915_query_engine_info. The address of latter should be passed to the
kernel in the query.data_ptr field, and should be large enough for the
kernel to fill out all known engines as struct drm_i915_engine_info
elements trailing the query.

As with other queries, setting the item query length to zero allows
userspace to query minimum required buffer size.

Enumerated engines have common type mask which can be used to query all
hardware engines, versus engines userspace can submit to using the execbuf
uAPI.

Engines also have capabilities which are per engine class namespace of
bits describing features not present on all engine instances.

v2:
 * Fixed HEVC assignment.
 * Reorder some fields, rename type to flags, increase width. (Lionel)
 * No need to allocate temporary storage if we do it engine by engine.
   (Lionel)

v3:
 * Describe engine flags and mark mbz fields. (Lionel)
 * HEVC only applies to VCS.

v4:
 * Squash SFC flag into main patch.
 * Tidy some comments.

v5:
 * Add uabi_ prefix to engine capabilities. (Chris Wilson)
 * Report exact size of engine info array. (Chris Wilson)
 * Drop the engine flags. (Joonas Lahtinen)
 * Added some more reserved fields.
 * Move flags after class/instance.

v6:
 * Do not check engine info array was zeroed by userspace but zero the
   unused fields for them instead.

v7:
 * Simplify length calculation loop. (Lionel Landwerlin)

v8:
 * Remove MBZ comments where not applicable.
 * Rename ABI flags to match engine class define naming.
 * Rename SFC ABI flag to reflect it applies to VCS and VECS.
 * SFC is wired to even _logical_ engine instances.
 * SFC applies to VCS and VECS.
 * HEVC is present on all instances on Gen11. (Tony)
 * Simplify length calculation even more. (Chris Wilson)
 * Move info_ptr assigment closer to loop for clarity. (Chris Wilson)
 * Use vdbox_sfc_access from runtime info.
 * Rebase for RUNTIME_INFO.
 * Refactor for lower indentation.
 * Rename uAPI class/instance to engine_class/instance to avoid C++
   keyword.

v9:
 * Rebase for s/num_rings/num_engines/ in RUNTIME_INFO.

v10:
 * Use new copy_query_item.

Signed-off-by: Tvrtko Ursulin <[e-mail]>
Cc: Chris Wilson <[e-mail]>
Cc: Jon Bloomfield <[e-mail]>
Cc: Dmitry Rogozhkin <[e-mail]>
Cc: Lionel Landwerlin <[e-mail]>
Cc: Joonas Lahtinen <[e-mail]>
Cc: Tony Ye <[e-mail]>
Reviewed-by: Lionel Landwerlin <[e-mail]> # v7
Reviewed-by: Chris Wilson <[e-mail]> # v7
---
 drivers/gpu/drm/i915/i915_query.c         | 49 +++++++++++++++++++++++
 drivers/gpu/drm/i915/intel_engine_cs.c    | 41 +++++++++++++++++++
 drivers/gpu/drm/i915/intel_engine_types.h |  2 +
 include/uapi/drm/i915_drm.h               | 45 +++++++++++++++++++++
 4 files changed, 137 insertions(+)

Comments

Chris Wilson March 14, 2019, 2:55 p.m. UTC | #1

Quoting Tvrtko Ursulin (2019-03-14 14:44:19)
> From: Tvrtko Ursulin <[e-mail]>
> 
> Engine discovery query allows userspace to enumerate engines, probe their
> configuration features, all without needing to maintain the internal PCI
> ID based database.
> 
> A new query for the generic i915 query ioctl is added named
> DRM_I915_QUERY_ENGINE_INFO, together with accompanying structure
> drm_i915_query_engine_info. The address of latter should be passed to the
> kernel in the query.data_ptr field, and should be large enough for the
> kernel to fill out all known engines as struct drm_i915_engine_info
> elements trailing the query.
> 
> As with other queries, setting the item query length to zero allows
> userspace to query minimum required buffer size.
> 
> Enumerated engines have common type mask which can be used to query all
> hardware engines, versus engines userspace can submit to using the execbuf
> uAPI.

> +       for_each_engine(engine, i915, id) {
> +               info.engine_class = engine->uabi_class;
> +               info.engine_instance = engine->instance;
> +               info.capabilities = engine->uabi_capabilities;

I would love to be able to find out engine->mmio_base,
engine->context_size, and maybe engine->hw_id?

Pretty please.

Or maybe you want to distinguish between uapi info and hw info. Which
would be reasonable.
-Chris

Chris Wilson March 14, 2019, 3:03 p.m. UTC | #2

Quoting Tvrtko Ursulin (2019-03-14 14:44:19)
> From: Tvrtko Ursulin <[e-mail]>
> 
> Engine discovery query allows userspace to enumerate engines, probe their
> configuration features, all without needing to maintain the internal PCI
> ID based database.
> 
> A new query for the generic i915 query ioctl is added named
> DRM_I915_QUERY_ENGINE_INFO, together with accompanying structure

Would ENGINE_UABI_INFO be apt if I were to add my gubbins as
ENGINE_HW_INFO ?
-Chris

Chris Wilson March 14, 2019, 3:49 p.m. UTC | #3

Quoting Tvrtko Ursulin (2019-03-14 14:44:19)
> +struct drm_i915_engine_info {
> +       /** Engine class as in enum drm_i915_gem_engine_class. */
> +       __u16 engine_class;
> +
> +       /** Engine instance number. */
> +       __u16 engine_instance;
> +
> +       /** Reserved field. */
> +       __u32 rsvd0;
> +
> +       /** Engine flags. */
> +       __u64 flags;
> +
> +       /** Capabilities of this engine. */
> +       __u64 capabilities;
> +#define I915_VIDEO_CLASS_CAPABILITY_HEVC               (1 << 0)
> +#define I915_VIDEO_AND_ENHANCE_CLASS_CAPABILITY_SFC    (1 << 1)
> +
> +       /** Reserved fields. */
> +       __u64 rsvd1[4];

Total of 7 __u64. Go on, just add one more.
-Chris

Chris Wilson March 14, 2019, 3:57 p.m. UTC | #4

Quoting Tvrtko Ursulin (2019-03-14 14:44:19)
> +       len = sizeof(struct drm_i915_query_engine_info) +
> +             RUNTIME_INFO(i915)->num_engines *
> +             sizeof(struct drm_i915_engine_info);

Fairly recent macro helps here,
len = struct_size(query_ptr, engines, RUNTIME_INFO(i915)->num_engines);
-Chris

Patch

diff --git a/drivers/gpu/drm/i915/i915_query.c b/drivers/gpu/drm/i915/i915_query.c
index 782183b78f49..ee217ccfbb0f 100644
--- a/drivers/gpu/drm/i915/i915_query.c
+++ b/drivers/gpu/drm/i915/i915_query.c
@@ -96,9 +96,58 @@  static int query_topology_info(struct drm_i915_private *dev_priv,
 	return total_length;
 }
 
+static int
+query_engine_info(struct drm_i915_private *i915,
+		  struct drm_i915_query_item *query_item)
+{
+	struct drm_i915_query_engine_info __user *query_ptr =
+				u64_to_user_ptr(query_item->data_ptr);
+	struct drm_i915_engine_info __user *info_ptr;
+	struct drm_i915_query_engine_info query;
+	struct drm_i915_engine_info info = { };
+	struct intel_engine_cs *engine;
+	enum intel_engine_id id;
+	int len, ret;
+
+	if (query_item->flags)
+		return -EINVAL;
+
+	len = sizeof(struct drm_i915_query_engine_info) +
+	      RUNTIME_INFO(i915)->num_engines *
+	      sizeof(struct drm_i915_engine_info);
+
+	ret = copy_query_item(&query, sizeof(query), len, query_item);
+	if (ret != 0)
+		return ret;
+
+	if (query.num_engines || query.rsvd[0] || query.rsvd[1] ||
+	    query.rsvd[2])
+		return -EINVAL;
+
+	info_ptr = &query_ptr->engines[0];
+
+	for_each_engine(engine, i915, id) {
+		info.engine_class = engine->uabi_class;
+		info.engine_instance = engine->instance;
+		info.capabilities = engine->uabi_capabilities;
+
+		if (__copy_to_user(info_ptr, &info, sizeof(info)))
+			return -EFAULT;
+
+		query.num_engines++;
+		info_ptr++;
+	}
+
+	if (__copy_to_user(query_ptr, &query, sizeof(query)))
+		return -EFAULT;
+
+	return len;
+}
+
 static int (* const i915_query_funcs[])(struct drm_i915_private *dev_priv,
 					struct drm_i915_query_item *query_item) = {
 	query_topology_info,
+	query_engine_info,
 };
 
 int i915_query_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c
index 652c1b3ba190..597bb924e21e 100644
--- a/drivers/gpu/drm/i915/intel_engine_cs.c
+++ b/drivers/gpu/drm/i915/intel_engine_cs.c
@@ -339,6 +339,45 @@  intel_engine_setup(struct drm_i915_private *dev_priv,
 	return 0;
 }
 
+static void __setup_engine_capabilities(struct intel_engine_cs *engine)
+{
+	struct drm_i915_private *i915 = engine->i915;
+
+	if (engine->class == VIDEO_DECODE_CLASS) {
+		/*
+		 * HEVC support is present on first engine instance
+		 * before Gen11 and on all instances afterwards.
+		 */
+		if (INTEL_GEN(i915) >= 11 ||
+		    (INTEL_GEN(i915) >= 9 && engine->instance == 0))
+			engine->uabi_capabilities |=
+				I915_VIDEO_CLASS_CAPABILITY_HEVC;
+
+		/*
+		 * SFC block is present only on even logical engine
+		 * instances.
+		 */
+		if ((INTEL_GEN(i915) >= 11 &&
+		     RUNTIME_INFO(i915)->vdbox_sfc_access & engine->mask) ||
+		    (INTEL_GEN(i915) >= 9 && engine->instance == 0))
+			engine->uabi_capabilities |=
+				I915_VIDEO_AND_ENHANCE_CLASS_CAPABILITY_SFC;
+	} else if (engine->class == VIDEO_ENHANCEMENT_CLASS) {
+		if (INTEL_GEN(i915) >= 9)
+			engine->uabi_capabilities |=
+				I915_VIDEO_AND_ENHANCE_CLASS_CAPABILITY_SFC;
+	}
+}
+
+static void intel_setup_engine_capabilities(struct drm_i915_private *i915)
+{
+	struct intel_engine_cs *engine;
+	enum intel_engine_id id;
+
+	for_each_engine(engine, i915, id)
+		__setup_engine_capabilities(engine);
+}
+
 /**
  * intel_engines_init_mmio() - allocate and prepare the Engine Command Streamers
  * @dev_priv: i915 device private
@@ -389,6 +428,8 @@  int intel_engines_init_mmio(struct drm_i915_private *dev_priv)
 
 	RUNTIME_INFO(dev_priv)->num_engines = hweight32(mask);
 
+	intel_setup_engine_capabilities(dev_priv);
+
 	i915_check_and_clear_faults(dev_priv);
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/intel_engine_types.h b/drivers/gpu/drm/i915/intel_engine_types.h
index b0aa1f0d4e47..cd12e9c96737 100644
--- a/drivers/gpu/drm/i915/intel_engine_types.h
+++ b/drivers/gpu/drm/i915/intel_engine_types.h
@@ -262,6 +262,8 @@  struct intel_engine_cs {
 	u32 context_size;
 	u32 mmio_base;
 
+	u32 uabi_capabilities;
+
 	struct intel_ring *buffer;
 
 	struct i915_timeline timeline;
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index aa2d4c73a97d..8c99925f5111 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -1744,6 +1744,7 @@  struct drm_i915_perf_oa_config {
 struct drm_i915_query_item {
 	__u64 query_id;
 #define DRM_I915_QUERY_TOPOLOGY_INFO    1
+#define DRM_I915_QUERY_ENGINE_INFO	2
 /* Must be kept compact -- no holes and well documented */
 
 	/*
@@ -1842,6 +1843,50 @@  struct drm_i915_query_topology_info {
 	__u8 data[];
 };
 
+/**
+ * struct drm_i915_engine_info
+ *
+ * Describes one engine and it's capabilities as known to the driver.
+ */
+struct drm_i915_engine_info {
+	/** Engine class as in enum drm_i915_gem_engine_class. */
+	__u16 engine_class;
+
+	/** Engine instance number. */
+	__u16 engine_instance;
+
+	/** Reserved field. */
+	__u32 rsvd0;
+
+	/** Engine flags. */
+	__u64 flags;
+
+	/** Capabilities of this engine. */
+	__u64 capabilities;
+#define I915_VIDEO_CLASS_CAPABILITY_HEVC		(1 << 0)
+#define I915_VIDEO_AND_ENHANCE_CLASS_CAPABILITY_SFC	(1 << 1)
+
+	/** Reserved fields. */
+	__u64 rsvd1[4];
+};
+
+/**
+ * struct drm_i915_query_engine_info
+ *
+ * Engine info query enumerates all engines known to the driver by filling in
+ * an array of struct drm_i915_engine_info structures.
+ */
+struct drm_i915_query_engine_info {
+	/** Number of struct drm_i915_engine_info structs following. */
+	__u32 num_engines;
+
+	/** MBZ */
+	__u32 rsvd[3];
+
+	/** Marker for drm_i915_engine_info structures. */
+	struct drm_i915_engine_info engines[];
+};
+
 #if defined(__cplusplus)
 }
 #endif