drm/i915: Allow user to set cache at BO creation

To comply with the design that buffer objects shall have immutable
cache setting through out their life cycle, {set, get}_caching ioctl's
are no longer supported from MTL onward. With that change caching
policy can only be set at object creation time. The current code
applies a default (platform dependent) cache setting for all objects.
However this is not optimal for performance tuning. The patch extends
the existing gem_create uAPI to let user set PAT index for the object
at creation time.
The new extension is platform independent, so UMD's can switch to using
this extension for older platforms as well, while {set, get}_caching are
still supported on these legacy paltforms for compatibility reason.
However, since PAT index was not clearly defined for platforms prior to
GEN12 (TGL), so we are limiting this externsion to GEN12+ platforms
only. See ext_set_pat() in for the implementation details.

The documentation related to the PAT/MOCS tables is currently available
for Tiger Lake here:
https://www.intel.com/content/www/us/en/docs/graphics-for-linux/developer-reference/1-0/tiger-lake.html

The documentation for other platforms is currently being updated.

BSpec: 45101

Mesa support has been submitted in this merge request:
https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22878

The media driver supprt has bin submitted in this merge request:
https://github.com/intel/media-driver/pull/1680

The IGT test related to this change is
igt@gem_create@create-ext-set-pat

Signed-off-by: Fei Yang <fei.yang@intel.com>
Cc: Chris Wilson <chris.p.wilson@linux.intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Acked-by: Jordan Justen <jordan.l.justen@intel.com>
Tested-by: Jordan Justen <jordan.l.justen@intel.com>
Acked-by: Carl Zhang <carl.zhang@intel.com>
Tested-by: Lihao Gu <lihao.gu@intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Acked-by: Slawomir Milczarek <slawomir.milczarek@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230606100042.482345-2-andi.shyti@linux.intel.com

2 files changed, 46 insertions, 0 deletions

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_create.c b/drivers/gpu/drm/i915/gem/i915_gem_create.c
index bfe1dbda4cb7..d24c0ce8805c 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_create.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_create.c
@@ -245,6 +245,7 @@ struct create_ext {
 	unsigned int n_placements;
 	unsigned int placement_mask;
 	unsigned long flags;
+	unsigned int pat_index;
 };
 
 static void repr_placements(char *buf, size_t size,
@@ -394,11 +395,43 @@ static int ext_set_protected(struct i915_user_extension __user *base, void *data
 	return 0;
 }
 
+static int ext_set_pat(struct i915_user_extension __user *base, void *data)
+{
+	struct create_ext *ext_data = data;
+	struct drm_i915_private *i915 = ext_data->i915;
+	struct drm_i915_gem_create_ext_set_pat ext;
+	unsigned int max_pat_index;
+
+	BUILD_BUG_ON(sizeof(struct drm_i915_gem_create_ext_set_pat) !=
+		     offsetofend(struct drm_i915_gem_create_ext_set_pat, rsvd));
+
+	/* Limiting the extension only to Meteor Lake */
+	if (!IS_METEORLAKE(i915))
+		return -ENODEV;
+
+	if (copy_from_user(&ext, base, sizeof(ext)))
+		return -EFAULT;
+
+	max_pat_index = INTEL_INFO(i915)->max_pat_index;
+
+	if (ext.pat_index > max_pat_index) {
+		drm_dbg(&i915->drm, "PAT index is invalid: %u\n",
+			ext.pat_index);
+		return -EINVAL;
+	}
+
+	ext_data->pat_index = ext.pat_index;
+
+	return 0;
+}
+
 static const i915_user_extension_fn create_extensions[] = {
 	[I915_GEM_CREATE_EXT_MEMORY_REGIONS] = ext_set_placements,
 	[I915_GEM_CREATE_EXT_PROTECTED_CONTENT] = ext_set_protected,
+	[I915_GEM_CREATE_EXT_SET_PAT] = ext_set_pat,
 };
 
+#define PAT_INDEX_NOT_SET	0xffff
 /**
  * i915_gem_create_ext_ioctl - Creates a new mm object and returns a handle to it.
  * @dev: drm device pointer
@@ -418,6 +451,7 @@ i915_gem_create_ext_ioctl(struct drm_device *dev, void *data,
 	if (args->flags & ~I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS)
 		return -EINVAL;
 
+	ext_data.pat_index = PAT_INDEX_NOT_SET;
 	ret = i915_user_extensions(u64_to_user_ptr(args->extensions),
 				   create_extensions,
 				   ARRAY_SIZE(create_extensions),
@@ -454,5 +488,11 @@ i915_gem_create_ext_ioctl(struct drm_device *dev, void *data,
 	if (IS_ERR(obj))
 		return PTR_ERR(obj);
 
+	if (ext_data.pat_index != PAT_INDEX_NOT_SET) {
+		i915_gem_object_set_pat_index(obj, ext_data.pat_index);
+		/* Mark pat_index is set by UMD */
+		obj->pat_set_by_user = true;
+	}
+
 	return i915_gem_publish(obj, file, &args->size, &args->handle);
 }
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c
index 46a19b099ec8..97ac6fb37958 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c
@@ -209,6 +209,12 @@ bool i915_gem_object_can_bypass_llc(struct drm_i915_gem_object *obj)
 		return false;
 
 	/*
+	 * Always flush cache for UMD objects at creation time.
+	 */
+	if (obj->pat_set_by_user)
+		return true;
+
+	/*
 	 * EHL and JSL add the 'Bypass LLC' MOCS entry, which should make it
 	 * possible for userspace to bypass the GTT caching bits set by the
 	 * kernel, as per the given object cache_level. This is troublesome