drm/i915: The return of i915_gpu_info to debugfs

Once upon a time before we had automated GPU state capture upon hangs,
we had intel_gpu_dump. Now we come almost full circle and reinstate that
view of the current GPU queues and registers by using the error capture
facility to snapshot the GPU state when debugfs/.../i915_gpu_info is
opened - which should provided useful debugging to both the error
capture routines (without having to cause a hang and avoid the error
state being eaten by igt) and generally.

v2: Rename drm_i915_error_state to i915_gpu_state to alleviate some name
collisions between the error state dump and inspecting the gpu state.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170214164611.11381-1-chris@chris-wilson.co.uk

drivers/gpu/drm/i915/i915_debugfs.c

@@ -988,89 +988,93 @@ static int i915_gem_fence_regs_info(struct seq_file *m, void *data)
 }
 
 #if IS_ENABLED(CONFIG_DRM_I915_CAPTURE_ERROR)
-
-static ssize_t
-i915_error_state_write(struct file *filp,
-		       const char __user *ubuf,
-		       size_t cnt,
-		       loff_t *ppos)
+static ssize_t gpu_state_read(struct file *file, char __user *ubuf,
+			      size_t count, loff_t *pos)
 {
-	struct i915_error_state_file_priv *error_priv = filp->private_data;
+	struct i915_gpu_state *error = file->private_data;
+	struct drm_i915_error_state_buf str;
+	ssize_t ret;
+	loff_t tmp;
 
-	DRM_DEBUG_DRIVER("Resetting error state\n");
-	i915_destroy_error_state(error_priv->i915);
-
-	return cnt;
-}
-
-static int i915_error_state_open(struct inode *inode, struct file *file)
-{
-	struct drm_i915_private *dev_priv = inode->i_private;
-	struct i915_error_state_file_priv *error_priv;
+	if (!error)
+		return 0;
 
-	error_priv = kzalloc(sizeof(*error_priv), GFP_KERNEL);
-	if (!error_priv)
-		return -ENOMEM;
+	ret = i915_error_state_buf_init(&str, error->i915, count, *pos);
+	if (ret)
+		return ret;
 
-	error_priv->i915 = dev_priv;
+	ret = i915_error_state_to_str(&str, error);
+	if (ret)
+		goto out;
 
-	i915_error_state_get(&dev_priv->drm, error_priv);
+	tmp = 0;
+	ret = simple_read_from_buffer(ubuf, count, &tmp, str.buf, str.bytes);
+	if (ret < 0)
+		goto out;
 
-	file->private_data = error_priv;
+	*pos = str.start + ret;
+out:
+	i915_error_state_buf_release(&str);
+	return ret;
+}
 
+static int gpu_state_release(struct inode *inode, struct file *file)
+{
+	i915_gpu_state_put(file->private_data);
 	return 0;
 }
 
-static int i915_error_state_release(struct inode *inode, struct file *file)
+static int i915_gpu_info_open(struct inode *inode, struct file *file)
 {
-	struct i915_error_state_file_priv *error_priv = file->private_data;
+	struct i915_gpu_state *gpu;
 
-	i915_error_state_put(error_priv);
-	kfree(error_priv);
+	gpu = i915_capture_gpu_state(inode->i_private);
+	if (!gpu)
+		return -ENOMEM;
 
+	file->private_data = gpu;
 	return 0;
 }
 
-static ssize_t i915_error_state_read(struct file *file, char __user *userbuf,
-				     size_t count, loff_t *pos)
+static const struct file_operations i915_gpu_info_fops = {
+	.owner = THIS_MODULE,
+	.open = i915_gpu_info_open,
+	.read = gpu_state_read,
+	.llseek = default_llseek,
+	.release = gpu_state_release,
+};
+
+static ssize_t
+i915_error_state_write(struct file *filp,
+		       const char __user *ubuf,
+		       size_t cnt,
+		       loff_t *ppos)
 {
-	struct i915_error_state_file_priv *error_priv = file->private_data;
-	struct drm_i915_error_state_buf error_str;
-	loff_t tmp_pos = 0;
-	ssize_t ret_count = 0;
-	int ret;
+	struct i915_gpu_state *error = filp->private_data;
 
-	ret = i915_error_state_buf_init(&error_str, error_priv->i915,
-					count, *pos);
-	if (ret)
-		return ret;
+	if (!error)
+		return 0;
 
-	ret = i915_error_state_to_str(&error_str, error_priv);
-	if (ret)
-		goto out;
+	DRM_DEBUG_DRIVER("Resetting error state\n");
+	i915_reset_error_state(error->i915);
 
-	ret_count = simple_read_from_buffer(userbuf, count, &tmp_pos,
-					    error_str.buf,
-					    error_str.bytes);
+	return cnt;
+}
 
-	if (ret_count < 0)
-		ret = ret_count;
-	else
-		*pos = error_str.start + ret_count;
-out:
-	i915_error_state_buf_release(&error_str);
-	return ret ?: ret_count;
+static int i915_error_state_open(struct inode *inode, struct file *file)
+{
+	file->private_data = i915_first_error_state(inode->i_private);
+	return 0;
 }
 
 static const struct file_operations i915_error_state_fops = {
 	.owner = THIS_MODULE,
 	.open = i915_error_state_open,
-	.read = i915_error_state_read,
+	.read = gpu_state_read,
 	.write = i915_error_state_write,
 	.llseek = default_llseek,
-	.release = i915_error_state_release,
+	.release = gpu_state_release,
 };
-
 #endif
 
 static int
@@ -4811,6 +4815,7 @@ static const struct i915_debugfs_files {
 	{"i915_gem_drop_caches", &i915_drop_caches_fops},
 #if IS_ENABLED(CONFIG_DRM_I915_CAPTURE_ERROR)
 	{"i915_error_state", &i915_error_state_fops},
+	{"i915_gpu_info", &i915_gpu_info_fops},
 #endif
 	{"i915_next_seqno", &i915_next_seqno_fops},
 	{"i915_display_crc_ctl", &i915_display_crc_ctl_fops},

drivers/gpu/drm/i915/i915_drv.c

@@ -1370,7 +1370,7 @@ void i915_driver_unload(struct drm_device *dev)
 
 	/* Free error state after interrupts are fully disabled. */
 	cancel_delayed_work_sync(&dev_priv->gpu_error.hangcheck_work);
-	i915_destroy_error_state(dev_priv);
+	i915_reset_error_state(dev_priv);
 
 	/* Flush any outstanding unpin_work. */
 	drain_workqueue(dev_priv->wq);

drivers/gpu/drm/i915/i915_drv.h

@@ -897,7 +897,7 @@ struct intel_device_info {
 
 struct intel_display_error_state;
 
-struct drm_i915_error_state {
+struct i915_gpu_state {
 	struct kref ref;
 	struct timeval time;
 	struct timeval boottime;
@@ -917,7 +917,7 @@ struct drm_i915_error_state {
 	u32 eir;
 	u32 pgtbl_er;
 	u32 ier;
-	u32 gtier[4];
+	u32 gtier[4], ngtier;
 	u32 ccid;
 	u32 derrmr;
 	u32 forcewake;
@@ -931,6 +931,7 @@ struct drm_i915_error_state {
 	u32 gab_ctl;
 	u32 gfx_mode;
 
+	u32 nfence;
 	u64 fence[I915_MAX_NUM_FENCES];
 	struct intel_overlay_error_state *overlay;
 	struct intel_display_error_state *display;
@@ -1512,11 +1513,6 @@ struct drm_i915_error_state_buf {
 	loff_t pos;
 };
 
-struct i915_error_state_file_priv {
-	struct drm_i915_private *i915;
-	struct drm_i915_error_state *error;
-};
-
 #define I915_RESET_TIMEOUT (10 * HZ) /* 10s */
 #define I915_FENCE_TIMEOUT (10 * HZ) /* 10s */
 
@@ -1533,7 +1529,7 @@ struct i915_gpu_error {
 	/* For reset and error_state handling. */
 	spinlock_t lock;
 	/* Protected by the above dev->gpu_error.lock. */
-	struct drm_i915_error_state *first_error;
+	struct i915_gpu_state *first_error;
 
 	unsigned long missed_irq_rings;
 
@@ -3574,7 +3570,7 @@ static inline void intel_display_crc_init(struct drm_i915_private *dev_priv) {}
 __printf(2, 3)
 void i915_error_printf(struct drm_i915_error_state_buf *e, const char *f, ...);
 int i915_error_state_to_str(struct drm_i915_error_state_buf *estr,
-			    const struct i915_error_state_file_priv *error);
+			    const struct i915_gpu_state *gpu);
 int i915_error_state_buf_init(struct drm_i915_error_state_buf *eb,
 			      struct drm_i915_private *i915,
 			      size_t count, loff_t pos);
@@ -3583,13 +3579,28 @@ static inline void i915_error_state_buf_release(
 {
 	kfree(eb->buf);
 }
+
+struct i915_gpu_state *i915_capture_gpu_state(struct drm_i915_private *i915);
 void i915_capture_error_state(struct drm_i915_private *dev_priv,
 			      u32 engine_mask,
 			      const char *error_msg);
-void i915_error_state_get(struct drm_device *dev,
-			  struct i915_error_state_file_priv *error_priv);
-void i915_error_state_put(struct i915_error_state_file_priv *error_priv);
-void i915_destroy_error_state(struct drm_i915_private *dev_priv);
+
+static inline struct i915_gpu_state *
+i915_gpu_state_get(struct i915_gpu_state *gpu)
+{
+	kref_get(&gpu->ref);
+	return gpu;
+}
+
+void __i915_gpu_state_free(struct kref *kref);
+static inline void i915_gpu_state_put(struct i915_gpu_state *gpu)
+{
+	if (gpu)
+		kref_put(&gpu->ref, __i915_gpu_state_free);
+}
+
+struct i915_gpu_state *i915_first_error_state(struct drm_i915_private *i915);
+void i915_reset_error_state(struct drm_i915_private *i915);
 
 #else
 
@@ -3599,7 +3610,13 @@ static inline void i915_capture_error_state(struct drm_i915_private *dev_priv,
 {
 }
 
-static inline void i915_destroy_error_state(struct drm_i915_private *dev_priv)
+static inline struct i915_gpu_state *
+i915_first_error_state(struct drm_i915_private *i915)
+{
+	return NULL;
+}
+
+static inline void i915_reset_error_state(struct drm_i915_private *i915)
 {
 }
 
@@ -3747,7 +3764,6 @@ extern void intel_overlay_print_error_state(struct drm_i915_error_state_buf *e,
 extern struct intel_display_error_state *
 intel_display_capture_error_state(struct drm_i915_private *dev_priv);
 extern void intel_display_print_error_state(struct drm_i915_error_state_buf *e,
-					    struct drm_i915_private *dev_priv,
 					    struct intel_display_error_state *error);
 
 int sandybridge_pcode_read(struct drm_i915_private *dev_priv, u32 mbox, u32 *val);

drivers/gpu/drm/i915/i915_sysfs.c

@@ -522,33 +522,27 @@ static ssize_t error_state_read(struct file *filp, struct kobject *kobj,
 
 	struct device *kdev = kobj_to_dev(kobj);
 	struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev);
-	struct drm_device *dev = &dev_priv->drm;
-	struct i915_error_state_file_priv error_priv;
 	struct drm_i915_error_state_buf error_str;
-	ssize_t ret_count = 0;
-	int ret;
-
-	memset(&error_priv, 0, sizeof(error_priv));
+	struct i915_gpu_state *gpu;
+	ssize_t ret;
 
-	ret = i915_error_state_buf_init(&error_str, to_i915(dev), count, off);
+	ret = i915_error_state_buf_init(&error_str, dev_priv, count, off);
 	if (ret)
 		return ret;
 
-	error_priv.i915 = dev_priv;
-	i915_error_state_get(dev, &error_priv);
-
-	ret = i915_error_state_to_str(&error_str, &error_priv);
+	gpu = i915_first_error_state(dev_priv);
+	ret = i915_error_state_to_str(&error_str, gpu);
 	if (ret)
 		goto out;
 
-	ret_count = count < error_str.bytes ? count : error_str.bytes;
+	ret = count < error_str.bytes ? count : error_str.bytes;
+	memcpy(buf, error_str.buf, ret);
 
-	memcpy(buf, error_str.buf, ret_count);
 out:
-	i915_error_state_put(&error_priv);
+	i915_gpu_state_put(gpu);
 	i915_error_state_buf_release(&error_str);
 
-	return ret ?: ret_count;
+	return ret;
 }
 
 static ssize_t error_state_write(struct file *file, struct kobject *kobj,
@@ -559,7 +553,7 @@ static ssize_t error_state_write(struct file *file, struct kobject *kobj,
 	struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev);
 
 	DRM_DEBUG_DRIVER("Resetting error state\n");
-	i915_destroy_error_state(dev_priv);
+	i915_reset_error_state(dev_priv);
 
 	return count;
 }

drivers/gpu/drm/i915/intel_display.c

@@ -15838,9 +15838,9 @@ intel_display_capture_error_state(struct drm_i915_private *dev_priv)
 
 void
 intel_display_print_error_state(struct drm_i915_error_state_buf *m,
-				struct drm_i915_private *dev_priv,
 				struct intel_display_error_state *error)
 {
+	struct drm_i915_private *dev_priv = m->i915;
 	int i;
 
 	if (!error)