drm sync objects (vn+1)

drm sync objects (vn+1)

[Dave Airlie]

Okay another day, another sync object patchset.

In my drm-syncobj branch.

This round should:
a) address the amdgpu CS ERESTARTSYS failure case, I've reworked
the sync_file/syncobj/amdgpu bits to have a lookup and just replace
in separate phases and just do the wait lookup and sync, then later
replace the waits with NULL, once the CS is submitted.

b) Block polling on the semaphore sync files, and cease worrying
about the polling state and callback when doing the replace fence
dance, which can only be called on semaphore sync files.

c) Add IGT tests - in my fd.o intel-gpu-tools repo syncobj branch,
are a basic sets of tests to do bad things and poke some corner
cases, and I've fixed up most of these in here hopefully. Checks
that padding is 0, legal fds, poll doesn't work etc.

d) bump SYNCOBJ CAP to 0x13 as 0x12 is already in use in drm-next.

Dave.

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 2/8] sync_file: export some interfaces needed by drm sync objects.

[Dave Airlie]

From: Dave Airlie <airlied@redhat.com>

These are just alloc and fdget interfaces needed by the drm sync
objects code.

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>
---
 drivers/dma-buf/sync_file.c | 21 +++++++++++++++++++--
 include/linux/sync_file.h   |  3 ++-
 2 files changed, 21 insertions(+), 3 deletions(-)

diff --git a/drivers/dma-buf/sync_file.c b/drivers/dma-buf/sync_file.c
index 4342d92..8667b48 100644
--- a/drivers/dma-buf/sync_file.c
+++ b/drivers/dma-buf/sync_file.c
@@ -46,7 +46,16 @@ int sync_file_validate_type_flags(uint32_t type, uint32_t flags)
 }
 EXPORT_SYMBOL(sync_file_validate_type_flags);
 
-static struct sync_file *sync_file_alloc(uint32_t type, uint32_t flags)
+/**
+ * sync_file_alloc() - allocate an unfenced sync file
+ * @type: type of sync file object
+ * @flags: creation flags for sync file object
+ *
+ * Creates a sync_file.
+ * The sync_file can be released with fput(sync_file->file).
+ * Returns the sync_file or NULL in case of error.
+ */
+struct sync_file *sync_file_alloc(uint32_t type, uint32_t flags)
 {
 	struct sync_file *sync_file;
 	int ret;
@@ -78,6 +87,7 @@ static struct sync_file *sync_file_alloc(uint32_t type, uint32_t flags)
 	kfree(sync_file);
 	return NULL;
 }
+EXPORT_SYMBOL(sync_file_alloc);
 
 static void fence_check_cb_func(struct dma_fence *f, struct dma_fence_cb *cb)
 {
@@ -120,7 +130,13 @@ struct sync_file *sync_file_create(struct dma_fence *fence,
 }
 EXPORT_SYMBOL(sync_file_create);
 
-static struct sync_file *sync_file_fdget(int fd)
+/**
+ * sync_file_fdget - get a reference to the file and return sync_file.
+ * @fd: file descriptor pointing at a sync_file.
+ *
+ * Returns the corresponding sync_file object.
+ */
+struct sync_file *sync_file_fdget(int fd)
 {
 	struct file *file = fget(fd);
 
@@ -136,6 +152,7 @@ static struct sync_file *sync_file_fdget(int fd)
 	fput(file);
 	return NULL;
 }
+EXPORT_SYMBOL(sync_file_fdget);
 
 /**
  * sync_file_get_fence - get the fence related to the sync_file fd
diff --git a/include/linux/sync_file.h b/include/linux/sync_file.h
index ede4182..e683dd1 100644
--- a/include/linux/sync_file.h
+++ b/include/linux/sync_file.h
@@ -54,7 +54,8 @@ struct sync_file {
 #define POLL_ENABLED DMA_FENCE_FLAG_USER_BITS
 
 int sync_file_validate_type_flags(uint32_t type, uint32_t flags);
+struct sync_file *sync_file_alloc(uint32_t type, uint32_t flags);
 struct sync_file *sync_file_create(struct dma_fence *fence, uint32_t type, uint32_t flags);
 struct dma_fence *sync_file_get_fence(int fd);
-
+struct sync_file *sync_file_fdget(int fd);
 #endif /* _LINUX_SYNC_H */
-- 
2.9.3

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[PATCH 5/8] sync_file: add support for a semaphore object (v2)

[Dave Airlie]

From: Dave Airlie <airlied@redhat.com>

This object can be used to implement the Vulkan semaphores.

The object behaviour differs from fence, in that you can
replace the underlying fence, and you cannot merge semaphores.

v2: change replace fence API to have a return value,
don't allow polling on semaphore objects,
don't worry about polling when replacing fences (since
we don't allow polling anymore).
switch APIs to one to retreive a fence from a sync_file
to make AMDGPU implementation more robust.

Signed-off-by: Dave Airlie <airlied@redhat.com>
---
 drivers/dma-buf/sync_file.c    | 73 ++++++++++++++++++++++++++++++++++++------
 include/linux/sync_file.h      |  4 +++
 include/uapi/linux/sync_file.h | 14 ++++++++
 3 files changed, 81 insertions(+), 10 deletions(-)

diff --git a/drivers/dma-buf/sync_file.c b/drivers/dma-buf/sync_file.c
index 57ea7c7..01c12e7 100644
--- a/drivers/dma-buf/sync_file.c
+++ b/drivers/dma-buf/sync_file.c
@@ -44,7 +44,7 @@ int sync_file_validate_type_flags(uint32_t type, uint32_t flags)
 {
 	if (flags)
 		return -EINVAL;
-	if (type != SYNC_FILE_TYPE_FENCE)
+	if (type != SYNC_FILE_TYPE_FENCE && type != SYNC_FILE_TYPE_SEMAPHORE)
 		return -EINVAL;
 	return 0;
 }
@@ -165,6 +165,28 @@ struct sync_file *sync_file_fdget(int fd)
 EXPORT_SYMBOL(sync_file_fdget);
 
 /**
+ * sync_file_get_fence_from_sync_file - get the fence related to the sync_file
+ * @sync_file:		sync_file fd to get the fence from
+ *
+ * Returns a fence that represents all fence in the sync_file. On error NULL is returned.
+ */
+struct dma_fence *sync_file_get_fence_from_sync_file(struct sync_file *sync_file)
+{
+	struct dma_fence *fence;
+
+	if (!rcu_access_pointer(sync_file->fence)) {
+		fput(sync_file->file);
+		return NULL;
+	}
+
+	rcu_read_lock();
+	fence = dma_fence_get_rcu_safe(&sync_file->fence);
+	rcu_read_unlock();
+	return fence;
+}
+EXPORT_SYMBOL(sync_file_get_fence_from_sync_file);
+
+/**
  * sync_file_get_fence - get the fence related to the sync_file fd
  * @fd:		sync_file fd to get the fence from
  *
@@ -180,14 +202,7 @@ struct dma_fence *sync_file_get_fence(int fd)
 	if (!sync_file)
 		return NULL;
 
-	if (!rcu_access_pointer(sync_file->fence)) {
-		fput(sync_file->file);
-		return NULL;
-	}
-
-	rcu_read_lock();
-	fence = dma_fence_get_rcu_safe(&sync_file->fence);
-	rcu_read_unlock();
+	fence = sync_file_get_fence_from_sync_file(sync_file);
 
 	fput(sync_file->file);
 
@@ -202,6 +217,39 @@ sync_file_get_fence_locked(struct sync_file *sync_file)
 					 sync_file_held(sync_file));
 }
 
+/**
+ * sync_file_replace_fence - replace the fence related to the sync_file
+ * @sync_file:	 sync file to replace fence in
+ * @fence: fence to replace with (or NULL for no fence).
+ * Returns previous fence.
+ */
+int sync_file_replace_fence(struct sync_file *sync_file,
+			    struct dma_fence *fence,
+			    struct dma_fence **old_fence)
+{
+	struct dma_fence *ret_fence = NULL;
+
+	/* don't allow replace on fence sync files */
+	if (sync_file->type != SYNC_FILE_TYPE_SEMAPHORE)
+		return -EINVAL;
+
+	if (fence == sync_file->fence)
+		return -EINVAL;
+
+	if (fence)
+		dma_fence_get(fence);
+
+	mutex_lock(&sync_file->lock);
+
+	ret_fence = sync_file_get_fence_locked(sync_file);
+	RCU_INIT_POINTER(sync_file->fence, fence);
+
+	mutex_unlock(&sync_file->lock);
+	*old_fence = ret_fence;
+	return 0;
+}
+EXPORT_SYMBOL(sync_file_replace_fence);
+
 static int sync_file_set_fence(struct sync_file *sync_file,
 			       struct dma_fence **fences, int num_fences)
 {
@@ -406,6 +454,10 @@ static unsigned int sync_file_poll(struct file *file, poll_table *wait)
 	unsigned int ret_val = 0;
 	struct dma_fence *fence;
 
+	/* ban polls on the sync file semaphores */
+	if (!sync_file || sync_file->type != SYNC_FILE_TYPE_FENCE)
+		return POLLERR | POLLNVAL;
+
 	poll_wait(file, &sync_file->wq, wait);
 
 	mutex_lock(&sync_file->lock);
@@ -453,7 +505,8 @@ static long sync_file_ioctl_merge(struct sync_file *sync_file,
 
 	data.name[sizeof(data.name) - 1] = '\0';
 
-	if (sync_file->type != fence2->type) {
+	if (sync_file->type != fence2->type ||
+	    sync_file->type != SYNC_FILE_TYPE_FENCE) {
 		err = -EINVAL;
 		goto err_put_fd;
 	}
diff --git a/include/linux/sync_file.h b/include/linux/sync_file.h
index 4bf661b..14dff25 100644
--- a/include/linux/sync_file.h
+++ b/include/linux/sync_file.h
@@ -61,5 +61,9 @@ int sync_file_validate_type_flags(uint32_t type, uint32_t flags);
 struct sync_file *sync_file_alloc(uint32_t type, uint32_t flags);
 struct sync_file *sync_file_create(struct dma_fence *fence, uint32_t type, uint32_t flags);
 struct dma_fence *sync_file_get_fence(int fd);
+struct dma_fence *sync_file_get_fence_from_sync_file(struct sync_file *sync_file);
 struct sync_file *sync_file_fdget(int fd);
+int sync_file_replace_fence(struct sync_file *sync_file,
+			    struct dma_fence *fence,
+			    struct dma_fence **old_fence);
 #endif /* _LINUX_SYNC_H */
diff --git a/include/uapi/linux/sync_file.h b/include/uapi/linux/sync_file.h
index f439cda..5f266e0 100644
--- a/include/uapi/linux/sync_file.h
+++ b/include/uapi/linux/sync_file.h
@@ -80,6 +80,20 @@ struct sync_file_info {
 #define SYNC_FILE_TYPE_FENCE 0
 
 /**
+ * DOC: SYNC_FILE_TYPE_SEMAPHORE - semaphore sync file object
+ *
+ * This is a sync file that operates like a Vulkan semaphore.
+ * The object should just be imported/exported but not use the
+ * sync file ioctls (except info).
+ * This object can have it's backing fence replaced multiple times.
+ * Each signal operation assigns a backing fence.
+ * Each wait operation waits on the current fence, and removes it.
+ * These operations should happen via driver command submission interfaces.
+ * This is useful for shared vulkan semaphores.
+ */
+#define SYNC_FILE_TYPE_SEMAPHORE 1
+
+/**
  * struct sync_file_type - data returned from sync file type ioctl
  * @type:	sync_file type
  * @flags:	sync_file creation flags
-- 
2.9.3

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[PATCH 1/8] sync_file: add type/flags to sync file object creation.

[Dave Airlie]

From: Dave Airlie <airlied@redhat.com>

This allows us to create sync files with different semantics,
and clearly define the interoperation between them it also
provides flags to allow for tweaks on those semantics.

This provides a validation interface for drivers that accept
types from userspace so they can return EINVAL instead of ENOMEM.

This provides an ioctl for userspace to retrieve the type/flags
of an object it may recieve from somewhere else.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
---
 Documentation/driver-api/dma-buf.rst |  6 ++++
 drivers/dma-buf/sw_sync.c            |  2 +-
 drivers/dma-buf/sync_file.c          | 67 +++++++++++++++++++++++++++++++++---
 drivers/gpu/drm/drm_atomic.c         |  2 +-
 include/linux/sync_file.h            |  9 ++++-
 include/uapi/linux/sync_file.h       | 27 +++++++++++++++
 6 files changed, 106 insertions(+), 7 deletions(-)

diff --git a/Documentation/driver-api/dma-buf.rst b/Documentation/driver-api/dma-buf.rst
index 31671b4..bf2f7d5 100644
--- a/Documentation/driver-api/dma-buf.rst
+++ b/Documentation/driver-api/dma-buf.rst
@@ -163,3 +163,9 @@ DMA Fence uABI/Sync File
 .. kernel-doc:: include/linux/sync_file.h
    :internal:
 
+
+Sync File IOCTL Definitions
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. kernel-doc:: include/uapi/linux/sync_file.h
+   :internal:
diff --git a/drivers/dma-buf/sw_sync.c b/drivers/dma-buf/sw_sync.c
index 69c5ff3..1c47de6 100644
--- a/drivers/dma-buf/sw_sync.c
+++ b/drivers/dma-buf/sw_sync.c
@@ -315,7 +315,7 @@ static long sw_sync_ioctl_create_fence(struct sync_timeline *obj,
 		goto err;
 	}
 
-	sync_file = sync_file_create(&pt->base);
+	sync_file = sync_file_create(&pt->base, SYNC_FILE_TYPE_FENCE, 0);
 	dma_fence_put(&pt->base);
 	if (!sync_file) {
 		err = -ENOMEM;
diff --git a/drivers/dma-buf/sync_file.c b/drivers/dma-buf/sync_file.c
index 2321035..4342d92 100644
--- a/drivers/dma-buf/sync_file.c
+++ b/drivers/dma-buf/sync_file.c
@@ -28,9 +28,32 @@
 
 static const struct file_operations sync_file_fops;
 
-static struct sync_file *sync_file_alloc(void)
+/**
+ * sync_file_validate_type_flags - validate type/flags for support
+ * @type: type of sync file object
+ * @flags: flags to sync object.
+ *
+ * Validates the flags are correct so userspace can get a more
+ * detailed error type.
+ */
+int sync_file_validate_type_flags(uint32_t type, uint32_t flags)
+{
+	if (flags)
+		return -EINVAL;
+	if (type != SYNC_FILE_TYPE_FENCE)
+		return -EINVAL;
+	return 0;
+}
+EXPORT_SYMBOL(sync_file_validate_type_flags);
+
+static struct sync_file *sync_file_alloc(uint32_t type, uint32_t flags)
 {
 	struct sync_file *sync_file;
+	int ret;
+
+	ret = sync_file_validate_type_flags(type, flags);
+	if (ret)
+		return NULL;
 
 	sync_file = kzalloc(sizeof(*sync_file), GFP_KERNEL);
 	if (!sync_file)
@@ -47,6 +70,8 @@ static struct sync_file *sync_file_alloc(void)
 
 	INIT_LIST_HEAD(&sync_file->cb.node);
 
+	sync_file->type = type;
+	sync_file->flags = flags;
 	return sync_file;
 
 err:
@@ -66,17 +91,21 @@ static void fence_check_cb_func(struct dma_fence *f, struct dma_fence_cb *cb)
 /**
  * sync_file_create() - creates a sync file
  * @fence:	fence to add to the sync_fence
+ * @type: type of sync file to create
+ * @flags: flags to create sync file with.
  *
  * Creates a sync_file containg @fence. This function acquires and additional
  * reference of @fence for the newly-created &sync_file, if it succeeds. The
  * sync_file can be released with fput(sync_file->file). Returns the
  * sync_file or NULL in case of error.
  */
-struct sync_file *sync_file_create(struct dma_fence *fence)
+struct sync_file *sync_file_create(struct dma_fence *fence,
+				   uint32_t type,
+				   uint32_t flags)
 {
 	struct sync_file *sync_file;
 
-	sync_file = sync_file_alloc();
+	sync_file = sync_file_alloc(type, flags);
 	if (!sync_file)
 		return NULL;
 
@@ -200,7 +229,7 @@ static struct sync_file *sync_file_merge(const char *name, struct sync_file *a,
 	struct dma_fence **fences, **nfences, **a_fences, **b_fences;
 	int i, i_a, i_b, num_fences, a_num_fences, b_num_fences;
 
-	sync_file = sync_file_alloc();
+	sync_file = sync_file_alloc(a->type, a->flags);
 	if (!sync_file)
 		return NULL;
 
@@ -339,6 +368,12 @@ static long sync_file_ioctl_merge(struct sync_file *sync_file,
 	}
 
 	data.name[sizeof(data.name) - 1] = '\0';
+
+	if (sync_file->type != fence2->type) {
+		err = -EINVAL;
+		goto err_put_fd;
+	}
+
 	fence3 = sync_file_merge(data.name, sync_file, fence2);
 	if (!fence3) {
 		err = -ENOMEM;
@@ -437,6 +472,27 @@ static long sync_file_ioctl_fence_info(struct sync_file *sync_file,
 	return ret;
 }
 
+static long sync_file_ioctl_type(struct sync_file *sync_file,
+				 unsigned long arg)
+{
+	struct sync_file_type type;
+	int ret;
+	if (copy_from_user(&type, (void __user *)arg, sizeof(type)))
+		return -EFAULT;
+
+	if (type.flags || type.type)
+		return -EINVAL;
+
+	type.type = sync_file->type;
+	type.flags = sync_file->flags;
+
+	if (copy_to_user((void __user *)arg, &type, sizeof(type)))
+		ret = -EFAULT;
+	else
+		ret = 0;
+	return ret;
+}
+
 static long sync_file_ioctl(struct file *file, unsigned int cmd,
 			    unsigned long arg)
 {
@@ -449,6 +505,9 @@ static long sync_file_ioctl(struct file *file, unsigned int cmd,
 	case SYNC_IOC_FILE_INFO:
 		return sync_file_ioctl_fence_info(sync_file, arg);
 
+	case SYNC_IOC_TYPE:
+		return sync_file_ioctl_type(sync_file, arg);
+
 	default:
 		return -ENOTTY;
 	}
diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
index a567310..bb5a740 100644
--- a/drivers/gpu/drm/drm_atomic.c
+++ b/drivers/gpu/drm/drm_atomic.c
@@ -1917,7 +1917,7 @@ static int setup_out_fence(struct drm_out_fence_state *fence_state,
 	if (put_user(fence_state->fd, fence_state->out_fence_ptr))
 		return -EFAULT;
 
-	fence_state->sync_file = sync_file_create(fence);
+	fence_state->sync_file = sync_file_create(fence, SYNC_FILE_TYPE_FENCE, 0);
 	if (!fence_state->sync_file)
 		return -ENOMEM;
 
diff --git a/include/linux/sync_file.h b/include/linux/sync_file.h
index 3e3ab84..ede4182 100644
--- a/include/linux/sync_file.h
+++ b/include/linux/sync_file.h
@@ -20,6 +20,8 @@
 #include <linux/spinlock.h>
 #include <linux/dma-fence.h>
 #include <linux/dma-fence-array.h>
+#include <uapi/linux/sync_file.h>
+
 
 /**
  * struct sync_file - sync file to export to the userspace
@@ -30,6 +32,8 @@
  * @wq:			wait queue for fence signaling
  * @fence:		fence with the fences in the sync_file
  * @cb:			fence callback information
+ * @type:               sync file type
+ * @flags:              flags used to create sync file
  */
 struct sync_file {
 	struct file		*file;
@@ -43,11 +47,14 @@ struct sync_file {
 
 	struct dma_fence	*fence;
 	struct dma_fence_cb cb;
+	uint32_t type;
+	uint32_t flags;
 };
 
 #define POLL_ENABLED DMA_FENCE_FLAG_USER_BITS
 
-struct sync_file *sync_file_create(struct dma_fence *fence);
+int sync_file_validate_type_flags(uint32_t type, uint32_t flags);
+struct sync_file *sync_file_create(struct dma_fence *fence, uint32_t type, uint32_t flags);
 struct dma_fence *sync_file_get_fence(int fd);
 
 #endif /* _LINUX_SYNC_H */
diff --git a/include/uapi/linux/sync_file.h b/include/uapi/linux/sync_file.h
index 5b287d6..f439cda 100644
--- a/include/uapi/linux/sync_file.h
+++ b/include/uapi/linux/sync_file.h
@@ -69,6 +69,26 @@ struct sync_file_info {
 #define SYNC_IOC_MAGIC		'>'
 
 /**
+ * DOC: SYNC_FILE_TYPE_FENCE - fence sync file object
+ *
+ * This sync file is a wrapper around a dma fence or a dma fence array.
+ * It can be merged with another fence sync file object to create a new
+ * merged object.
+ * The fence backing this object cannot be replaced.
+ * This is useful for shared fences.
+ */
+#define SYNC_FILE_TYPE_FENCE 0
+
+/**
+ * struct sync_file_type - data returned from sync file type ioctl
+ * @type:	sync_file type
+ * @flags:	sync_file creation flags
+ */
+struct sync_file_type {
+	__u32	type;
+	__u32	flags;
+};
+/**
  * Opcodes  0, 1 and 2 were burned during a API change to avoid users of the
  * old API to get weird errors when trying to handling sync_files. The API
  * change happened during the de-stage of the Sync Framework when there was
@@ -94,4 +114,11 @@ struct sync_file_info {
  */
 #define SYNC_IOC_FILE_INFO	_IOWR(SYNC_IOC_MAGIC, 4, struct sync_file_info)
 
+/**
+ * DOC: SYNC_IOC_TYPE - get creation type and flags of sync_file.
+ *
+ * Takes a struct sync_file_type. Returns the created values of type and flags.
+ */
+#define SYNC_IOC_TYPE		_IOWR(SYNC_IOC_MAGIC, 5, struct sync_file_type)
+
 #endif /* _UAPI_LINUX_SYNC_H */
-- 
2.9.3

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 3/8] drm: introduce sync objects as sync file objects with no fd (v2.1)

[Dave Airlie]

From: Dave Airlie <airlied@redhat.com>

Sync objects are new toplevel drm object, that have the same
semantics as sync_file objects, but without requiring an fd
to be consumed to support them.

This patch just enables the DRM interface to create these
objects, it doesn't actually provide any semaphore objects
or new features.

v2: do get/put on the sync file file to avoid race (Daniel)
v2.1: fix reference counting bug on import (Jason)
cleanup destroy (Jason)
fix a bunch of bugs for the igt test.
restrict semaphore objects for create/and fd conversions.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
---
 Documentation/gpu/drm-internals.rst |   5 +
 Documentation/gpu/drm-mm.rst        |   6 +
 drivers/gpu/drm/Makefile            |   2 +-
 drivers/gpu/drm/drm_fops.c          |   8 +
 drivers/gpu/drm/drm_internal.h      |  15 ++
 drivers/gpu/drm/drm_ioctl.c         |  14 ++
 drivers/gpu/drm/drm_syncobj.c       | 287 ++++++++++++++++++++++++++++++++++++
 include/drm/drmP.h                  |   5 +
 include/drm/drm_drv.h               |   1 +
 include/uapi/drm/drm.h              |  27 ++++
 10 files changed, 369 insertions(+), 1 deletion(-)
 create mode 100644 drivers/gpu/drm/drm_syncobj.c

diff --git a/Documentation/gpu/drm-internals.rst b/Documentation/gpu/drm-internals.rst
index e35920d..743b751 100644
--- a/Documentation/gpu/drm-internals.rst
+++ b/Documentation/gpu/drm-internals.rst
@@ -98,6 +98,11 @@ DRIVER_ATOMIC
     implement appropriate obj->atomic_get_property() vfuncs for any
     modeset objects with driver specific properties.
 
+DRIVER_SYNCOBJ
+    Driver support sync objects. These are just sync files that don't
+    use a file descriptor. They can be used to implement Vulkan shared
+    semaphores.
+
 Major, Minor and Patchlevel
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
diff --git a/Documentation/gpu/drm-mm.rst b/Documentation/gpu/drm-mm.rst
index f5760b1..28aebe8 100644
--- a/Documentation/gpu/drm-mm.rst
+++ b/Documentation/gpu/drm-mm.rst
@@ -483,3 +483,9 @@ DRM Cache Handling
 
 .. kernel-doc:: drivers/gpu/drm/drm_cache.c
    :export:
+
+DRM Sync Objects
+===========================
+
+.. kernel-doc:: drivers/gpu/drm/drm_syncobj.c
+   :export:
diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index 3ee9579..b5e565c 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -16,7 +16,7 @@ drm-y       :=	drm_auth.o drm_bufs.o drm_cache.o \
 		drm_framebuffer.o drm_connector.o drm_blend.o \
 		drm_encoder.o drm_mode_object.o drm_property.o \
 		drm_plane.o drm_color_mgmt.o drm_print.o \
-		drm_dumb_buffers.o drm_mode_config.o
+		drm_dumb_buffers.o drm_mode_config.o drm_syncobj.o
 
 drm-$(CONFIG_DRM_LIB_RANDOM) += lib/drm_random.o
 drm-$(CONFIG_DRM_VM) += drm_vm.o
diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c
index afdf5b1..9a61df2 100644
--- a/drivers/gpu/drm/drm_fops.c
+++ b/drivers/gpu/drm/drm_fops.c
@@ -219,6 +219,9 @@ static int drm_open_helper(struct file *filp, struct drm_minor *minor)
 	if (drm_core_check_feature(dev, DRIVER_GEM))
 		drm_gem_open(dev, priv);
 
+	if (drm_core_check_feature(dev, DRIVER_SYNCOBJ))
+		drm_syncobj_open(priv);
+
 	if (drm_core_check_feature(dev, DRIVER_PRIME))
 		drm_prime_init_file_private(&priv->prime);
 
@@ -266,6 +269,8 @@ static int drm_open_helper(struct file *filp, struct drm_minor *minor)
 out_prime_destroy:
 	if (drm_core_check_feature(dev, DRIVER_PRIME))
 		drm_prime_destroy_file_private(&priv->prime);
+	if (drm_core_check_feature(dev, DRIVER_SYNCOBJ))
+		drm_syncobj_release(priv);
 	if (drm_core_check_feature(dev, DRIVER_GEM))
 		drm_gem_release(dev, priv);
 	put_pid(priv->pid);
@@ -400,6 +405,9 @@ int drm_release(struct inode *inode, struct file *filp)
 		drm_property_destroy_user_blobs(dev, file_priv);
 	}
 
+	if (drm_core_check_feature(dev, DRIVER_SYNCOBJ))
+		drm_syncobj_release(file_priv);
+
 	if (drm_core_check_feature(dev, DRIVER_GEM))
 		drm_gem_release(dev, file_priv);
 
diff --git a/drivers/gpu/drm/drm_internal.h b/drivers/gpu/drm/drm_internal.h
index f37388c..70c27a0 100644
--- a/drivers/gpu/drm/drm_internal.h
+++ b/drivers/gpu/drm/drm_internal.h
@@ -142,4 +142,19 @@ static inline int drm_debugfs_crtc_crc_add(struct drm_crtc *crtc)
 {
 	return 0;
 }
+
 #endif
+
+/* drm_syncobj.c */
+void drm_syncobj_open(struct drm_file *file_private);
+void drm_syncobj_release(struct drm_file *file_private);
+int drm_syncobj_create_ioctl(struct drm_device *dev, void *data,
+			     struct drm_file *file_private);
+int drm_syncobj_destroy_ioctl(struct drm_device *dev, void *data,
+			      struct drm_file *file_private);
+int drm_syncobj_handle_to_fd_ioctl(struct drm_device *dev, void *data,
+				   struct drm_file *file_private);
+int drm_syncobj_fd_to_handle_ioctl(struct drm_device *dev, void *data,
+				   struct drm_file *file_private);
+int drm_syncobj_info_ioctl(struct drm_device *dev, void *data,
+			   struct drm_file *file_private);
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index a7c61c2..4a8ed66 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -240,6 +240,9 @@ static int drm_getcap(struct drm_device *dev, void *data, struct drm_file *file_
 		req->value |= dev->driver->prime_fd_to_handle ? DRM_PRIME_CAP_IMPORT : 0;
 		req->value |= dev->driver->prime_handle_to_fd ? DRM_PRIME_CAP_EXPORT : 0;
 		return 0;
+	case DRM_CAP_SYNCOBJ:
+		req->value = drm_core_check_feature(dev, DRIVER_SYNCOBJ);
+		return 0;
 	}
 
 	/* Other caps only work with KMS drivers */
@@ -641,6 +644,17 @@ static const struct drm_ioctl_desc drm_ioctls[] = {
 	DRM_IOCTL_DEF(DRM_IOCTL_MODE_ATOMIC, drm_mode_atomic_ioctl, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED),
 	DRM_IOCTL_DEF(DRM_IOCTL_MODE_CREATEPROPBLOB, drm_mode_createblob_ioctl, DRM_CONTROL_ALLOW|DRM_UNLOCKED),
 	DRM_IOCTL_DEF(DRM_IOCTL_MODE_DESTROYPROPBLOB, drm_mode_destroyblob_ioctl, DRM_CONTROL_ALLOW|DRM_UNLOCKED),
+
+	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_CREATE, drm_syncobj_create_ioctl,
+		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_DESTROY, drm_syncobj_destroy_ioctl,
+		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_HANDLE_TO_FD, drm_syncobj_handle_to_fd_ioctl,
+		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_FD_TO_HANDLE, drm_syncobj_fd_to_handle_ioctl,
+		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF(DRM_IOCTL_SYNCOBJ_INFO, drm_syncobj_info_ioctl,
+		      DRM_UNLOCKED|DRM_RENDER_ALLOW),
 };
 
 #define DRM_CORE_IOCTL_COUNT	ARRAY_SIZE( drm_ioctls )
diff --git a/drivers/gpu/drm/drm_syncobj.c b/drivers/gpu/drm/drm_syncobj.c
new file mode 100644
index 0000000..4827f90
--- /dev/null
+++ b/drivers/gpu/drm/drm_syncobj.c
@@ -0,0 +1,287 @@
+/*
+ * Copyright © 2017 Red Hat
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Authors:
+ *
+ */
+
+/**
+ * DOC: Overview
+ *
+ * DRM synchronisation objects (syncobj) are wrappers for sync_file objects,
+ * that don't use fd space, and can be converted to/from sync_file fds.
+ *
+ * Depending on the sync object type, they can expose different sync_file
+ * semantics and restrictions.
+ *
+ * Their primary use-case is to implement Vulkan shared semaphores.
+ */
+
+#include <drm/drmP.h>
+#include <linux/sync_file.h>
+#include "drm_internal.h"
+
+static struct sync_file *drm_syncobj_get(struct drm_file *file_private,
+					 u32 handle)
+{
+	struct sync_file *sync_file;
+
+	spin_lock(&file_private->syncobj_table_lock);
+
+	/* Check if we currently have a reference on the object */
+	sync_file = idr_find(&file_private->syncobj_idr, handle);
+	if (sync_file)
+		get_file(sync_file->file);
+
+	spin_unlock(&file_private->syncobj_table_lock);
+
+	return sync_file;
+}
+
+static int drm_syncobj_create(struct drm_file *file_private,
+			      unsigned type,
+			      unsigned flags, u32 *handle)
+{
+	struct sync_file *sync_file;
+	int ret;
+
+	ret = sync_file_validate_type_flags(type, flags);
+	if (ret)
+		return ret;
+
+	sync_file = sync_file_alloc(type, flags);
+	if (!sync_file)
+		return -ENOMEM;
+
+	idr_preload(GFP_KERNEL);
+	spin_lock(&file_private->syncobj_table_lock);
+	ret = idr_alloc(&file_private->syncobj_idr, sync_file, 1, 0, GFP_NOWAIT);
+	spin_unlock(&file_private->syncobj_table_lock);
+
+	idr_preload_end();
+
+	if (ret < 0) {
+		fput(sync_file->file);
+		return ret;
+	}
+
+	*handle = ret;
+	return 0;
+}
+
+static int drm_syncobj_destroy(struct drm_file *file_private,
+			       u32 handle)
+{
+	struct sync_file *sync_file;
+
+	spin_lock(&file_private->syncobj_table_lock);
+	sync_file = idr_remove(&file_private->syncobj_idr, handle);
+	spin_unlock(&file_private->syncobj_table_lock);
+
+	if (!sync_file)
+		return -EINVAL;
+
+	fput(sync_file->file);
+	return 0;
+}
+
+static int drm_syncobj_handle_to_fd(struct drm_file *file_private,
+				    u32 handle, int *fd)
+{
+	struct sync_file *sync_file = drm_syncobj_get(file_private, handle);
+	int ret;
+	if (!sync_file)
+		return -EINVAL;
+
+	if (sync_file->type != SYNC_FILE_TYPE_SEMAPHORE) {
+		fput(sync_file->file);
+		return -EINVAL;
+	}
+
+	ret = get_unused_fd_flags(O_CLOEXEC);
+	if (ret < 0) {
+		fput(sync_file->file);
+		return ret;
+	}
+	fd_install(ret, sync_file->file);
+	*fd = ret;
+	/* leave reference with fd */
+	return 0;
+}
+
+static int drm_syncobj_fd_to_handle(struct drm_file *file_private,
+				    int fd, u32 *handle)
+{
+	struct sync_file *sync_file = sync_file_fdget(fd);
+	int ret;
+	if (!sync_file)
+		return -EINVAL;
+
+	if (sync_file->type != SYNC_FILE_TYPE_SEMAPHORE) {
+		fput(sync_file->file);
+		return -EINVAL;
+	}
+
+	idr_preload(GFP_KERNEL);
+	spin_lock(&file_private->syncobj_table_lock);
+	ret = idr_alloc(&file_private->syncobj_idr, sync_file, 1, 0, GFP_NOWAIT);
+	spin_unlock(&file_private->syncobj_table_lock);
+	idr_preload_end();
+
+	if (ret < 0) {
+		fput(sync_file->file);
+		return ret;
+	}
+	*handle = ret;
+	return 0;
+}
+
+static int drm_syncobj_info(struct drm_file *file_private,
+			    u32 handle, u32 *type, u32 *flags)
+{
+	struct sync_file *sync_file = drm_syncobj_get(file_private, handle);
+
+	if (!sync_file)
+		return -EINVAL;
+	*type = sync_file->type;
+	*flags = sync_file->flags;
+	fput(sync_file->file);
+	return 0;
+}
+
+/**
+ * drm_syncobj_open - initalizes syncobj file-private structures at devnode open time
+ * @dev: drm_device which is being opened by userspace
+ * @file_private: drm file-private structure to set up
+ *
+ * Called at device open time, sets up the structure for handling refcounting
+ * of sync objects.
+ */
+void
+drm_syncobj_open(struct drm_file *file_private)
+{
+	idr_init(&file_private->syncobj_idr);
+	spin_lock_init(&file_private->syncobj_table_lock);
+}
+
+static int
+drm_syncobj_release_handle(int id, void *ptr, void *data)
+{
+	struct sync_file *sync_file = ptr;
+
+	fput(sync_file->file);
+	return 0;
+}
+
+/**
+ * drm_syncobj_release - release file-private sync object resources
+ * @dev: drm_device which is being closed by userspace
+ * @file_private: drm file-private structure to clean up
+ *
+ * Called at close time when the filp is going away.
+ *
+ * Releases any remaining references on objects by this filp.
+ */
+void
+drm_syncobj_release(struct drm_file *file_private)
+{
+	idr_for_each(&file_private->syncobj_idr,
+		     &drm_syncobj_release_handle, file_private);
+	idr_destroy(&file_private->syncobj_idr);
+}
+
+int
+drm_syncobj_create_ioctl(struct drm_device *dev, void *data,
+			 struct drm_file *file_private)
+{
+	struct drm_syncobj_create_info *args = data;
+
+	if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ))
+		return -ENODEV;
+
+	/* make sure padding is empty */
+	if (args->pad)
+		return -EINVAL;
+
+	/* only allow sync objects for now */
+	if (args->type != SYNC_FILE_TYPE_SEMAPHORE)
+		return -EINVAL;
+	return drm_syncobj_create(file_private, args->type,
+				 args->flags, &args->handle);
+}
+
+int
+drm_syncobj_destroy_ioctl(struct drm_device *dev, void *data,
+			  struct drm_file *file_private)
+{
+	struct drm_syncobj_destroy *args = data;
+
+	if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ))
+		return -ENODEV;
+
+	/* make sure padding is empty */
+	if (args->pad)
+		return -EINVAL;
+	return drm_syncobj_destroy(file_private, args->handle);
+}
+
+int
+drm_syncobj_handle_to_fd_ioctl(struct drm_device *dev, void *data,
+				   struct drm_file *file_private)
+{
+	struct drm_syncobj_handle *args = data;
+
+	if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ))
+		return -ENODEV;
+
+	return drm_syncobj_handle_to_fd(file_private, args->handle,
+					&args->fd);
+}
+
+int
+drm_syncobj_fd_to_handle_ioctl(struct drm_device *dev, void *data,
+				   struct drm_file *file_private)
+{
+	struct drm_syncobj_handle *args = data;
+
+	if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ))
+		return -ENODEV;
+
+	return drm_syncobj_fd_to_handle(file_private, args->fd,
+					&args->handle);
+}
+
+int
+drm_syncobj_info_ioctl(struct drm_device *dev, void *data,
+		       struct drm_file *file_private)
+{
+	struct drm_syncobj_create_info *args = data;
+
+	if (!drm_core_check_feature(dev, DRIVER_SYNCOBJ))
+		return -ENODEV;
+
+	/* make sure padding is empty */
+	if (args->pad)
+		return -EINVAL;
+	return drm_syncobj_info(file_private, args->handle,
+				&args->type, &args->flags);
+}
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index 6105c05..1d48f6f 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -405,6 +405,11 @@ struct drm_file {
 	/** Lock for synchronization of access to object_idr. */
 	spinlock_t table_lock;
 
+	/** Mapping of sync object handles to object pointers. */
+	struct idr syncobj_idr;
+	/** Lock for synchronization of access to syncobj_idr. */
+	spinlock_t syncobj_table_lock;
+
 	struct file *filp;
 	void *driver_priv;
 
diff --git a/include/drm/drm_drv.h b/include/drm/drm_drv.h
index 5699f42..48ff06b 100644
--- a/include/drm/drm_drv.h
+++ b/include/drm/drm_drv.h
@@ -53,6 +53,7 @@ struct drm_mode_create_dumb;
 #define DRIVER_RENDER			0x8000
 #define DRIVER_ATOMIC			0x10000
 #define DRIVER_KMS_LEGACY_CONTEXT	0x20000
+#define DRIVER_SYNCOBJ                  0x40000
 
 /**
  * struct drm_driver - DRM driver structure
diff --git a/include/uapi/drm/drm.h b/include/uapi/drm/drm.h
index b2c5284..f2d3f06 100644
--- a/include/uapi/drm/drm.h
+++ b/include/uapi/drm/drm.h
@@ -647,6 +647,7 @@ struct drm_gem_open {
 #define DRM_CAP_CURSOR_HEIGHT		0x9
 #define DRM_CAP_ADDFB2_MODIFIERS	0x10
 #define DRM_CAP_PAGE_FLIP_TARGET	0x11
+#define DRM_CAP_SYNCOBJ		0x13
 
 /** DRM_IOCTL_GET_CAP ioctl argument type */
 struct drm_get_cap {
@@ -696,6 +697,26 @@ struct drm_prime_handle {
 	__s32 fd;
 };
 
+struct drm_syncobj_create_info {
+	__u32 handle;
+	__u32 type;
+	__u32 flags;
+	__u32 pad;
+};
+
+struct drm_syncobj_destroy {
+	__u32 handle;
+	__u32 pad;
+};
+
+struct drm_syncobj_handle {
+	__u32 handle;
+	/** Flags.. only applicable for handle->fd */
+	__u32 flags;
+
+	__s32 fd;
+};
+
 #if defined(__cplusplus)
 }
 #endif
@@ -814,6 +835,12 @@ extern "C" {
 #define DRM_IOCTL_MODE_CREATEPROPBLOB	DRM_IOWR(0xBD, struct drm_mode_create_blob)
 #define DRM_IOCTL_MODE_DESTROYPROPBLOB	DRM_IOWR(0xBE, struct drm_mode_destroy_blob)
 
+#define DRM_IOCTL_SYNCOBJ_CREATE	DRM_IOWR(0xBF, struct drm_syncobj_create_info)
+#define DRM_IOCTL_SYNCOBJ_DESTROY	DRM_IOWR(0xC0, struct drm_syncobj_destroy)
+#define DRM_IOCTL_SYNCOBJ_HANDLE_TO_FD	DRM_IOWR(0xC1, struct drm_syncobj_handle)
+#define DRM_IOCTL_SYNCOBJ_FD_TO_HANDLE	DRM_IOWR(0xC2, struct drm_syncobj_handle)
+#define DRM_IOCTL_SYNCOBJ_INFO		DRM_IOWR(0xC3, struct drm_syncobj_create_info)
+
 /**
  * Device specific ioctls should only be in their respective headers
  * The device specific ioctl range is from 0x40 to 0x9f.
-- 
2.9.3

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

Re: [PATCH 3/8] drm: introduce sync objects as sync file objects with no fd (v2.1)

[Emil Velikov]

Hi Dave,

On 12 April 2017 at 05:57, Dave Airlie <airlied@gmail.com> wrote:

> +struct drm_syncobj_handle {
> +       __u32 handle;
> +       /** Flags.. only applicable for handle->fd */
> +       __u32 flags;
> +
> +       __s32 fd;
> +};
I think this struct need a __u32 pad as well.

Thanks
Emil
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[PATCH 4/8] sync_file: add a mutex to protect fence and callback members. (v4)

[Dave Airlie]

From: Dave Airlie <airlied@redhat.com>

This patch allows the underlying fence in a sync_file to be changed
or set to NULL. This isn't currently required but for Vulkan
semaphores we need to be able to swap and reset the fence.

In order to faciliate this, it uses rcu to protect the fence,
along with a new mutex. The mutex also protects the callback.
It also checks for NULL when retrieving the rcu protected
fence in case it has been reset.

v1.1: fix the locking (Julia Lawall).
v2: use rcu try one
v3: fix poll to use proper rcu, fixup merge/fill ioctls
to not crash on NULL fence cases.
v4: use rcu in even more places, add missing fput

Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>
---
 drivers/dma-buf/sync_file.c | 135 +++++++++++++++++++++++++++++++++++---------
 include/linux/sync_file.h   |   6 +-
 2 files changed, 114 insertions(+), 27 deletions(-)

diff --git a/drivers/dma-buf/sync_file.c b/drivers/dma-buf/sync_file.c
index 8667b48..57ea7c7 100644
--- a/drivers/dma-buf/sync_file.c
+++ b/drivers/dma-buf/sync_file.c
@@ -28,6 +28,10 @@
 
 static const struct file_operations sync_file_fops;
 
+#define sync_file_held(obj) lockdep_is_held(&(obj)->lock)
+#define sync_file_assert_held(obj) \
+	lockdep_assert_held(&(obj)->lock)
+
 /**
  * sync_file_validate_type_flags - validate type/flags for support
  * @type: type of sync file object
@@ -81,6 +85,10 @@ struct sync_file *sync_file_alloc(uint32_t type, uint32_t flags)
 
 	sync_file->type = type;
 	sync_file->flags = flags;
+
+	RCU_INIT_POINTER(sync_file->fence, NULL);
+
+	mutex_init(&sync_file->lock);
 	return sync_file;
 
 err:
@@ -119,7 +127,9 @@ struct sync_file *sync_file_create(struct dma_fence *fence,
 	if (!sync_file)
 		return NULL;
 
-	sync_file->fence = dma_fence_get(fence);
+	dma_fence_get(fence);
+
+	RCU_INIT_POINTER(sync_file->fence, fence);
 
 	snprintf(sync_file->name, sizeof(sync_file->name), "%s-%s%llu-%d",
 		 fence->ops->get_driver_name(fence),
@@ -170,13 +180,28 @@ struct dma_fence *sync_file_get_fence(int fd)
 	if (!sync_file)
 		return NULL;
 
-	fence = dma_fence_get(sync_file->fence);
+	if (!rcu_access_pointer(sync_file->fence)) {
+		fput(sync_file->file);
+		return NULL;
+	}
+
+	rcu_read_lock();
+	fence = dma_fence_get_rcu_safe(&sync_file->fence);
+	rcu_read_unlock();
+
 	fput(sync_file->file);
 
 	return fence;
 }
 EXPORT_SYMBOL(sync_file_get_fence);
 
+static inline struct dma_fence *
+sync_file_get_fence_locked(struct sync_file *sync_file)
+{
+	return rcu_dereference_protected(sync_file->fence,
+					 sync_file_held(sync_file));
+}
+
 static int sync_file_set_fence(struct sync_file *sync_file,
 			       struct dma_fence **fences, int num_fences)
 {
@@ -189,7 +214,7 @@ static int sync_file_set_fence(struct sync_file *sync_file,
 	 * we own the reference of the dma_fence_array creation.
 	 */
 	if (num_fences == 1) {
-		sync_file->fence = fences[0];
+		RCU_INIT_POINTER(sync_file->fence, fences[0]);
 		kfree(fences);
 	} else {
 		array = dma_fence_array_create(num_fences, fences,
@@ -198,24 +223,30 @@ static int sync_file_set_fence(struct sync_file *sync_file,
 		if (!array)
 			return -ENOMEM;
 
-		sync_file->fence = &array->base;
+		RCU_INIT_POINTER(sync_file->fence, &array->base);
 	}
 
 	return 0;
 }
 
-static struct dma_fence **get_fences(struct sync_file *sync_file,
+/* must be called with rcu read lock taken */
+static struct dma_fence **get_fences(struct dma_fence **fence,
 				     int *num_fences)
 {
-	if (dma_fence_is_array(sync_file->fence)) {
-		struct dma_fence_array *array = to_dma_fence_array(sync_file->fence);
+	if (!*fence) {
+		*num_fences = 0;
+		return NULL;
+	}
+
+	if (dma_fence_is_array(*fence)) {
+		struct dma_fence_array *array = to_dma_fence_array(*fence);
 
 		*num_fences = array->num_fences;
 		return array->fences;
 	}
 
 	*num_fences = 1;
-	return &sync_file->fence;
+	return fence;
 }
 
 static void add_fence(struct dma_fence **fences,
@@ -245,16 +276,33 @@ static struct sync_file *sync_file_merge(const char *name, struct sync_file *a,
 	struct sync_file *sync_file;
 	struct dma_fence **fences, **nfences, **a_fences, **b_fences;
 	int i, i_a, i_b, num_fences, a_num_fences, b_num_fences;
+	struct dma_fence *a_fence, *b_fence;
 
 	sync_file = sync_file_alloc(a->type, a->flags);
 	if (!sync_file)
 		return NULL;
 
-	a_fences = get_fences(a, &a_num_fences);
-	b_fences = get_fences(b, &b_num_fences);
-	if (a_num_fences > INT_MAX - b_num_fences)
+	if (!rcu_access_pointer(a->fence) ||
+	    !rcu_access_pointer(b->fence))
 		return NULL;
 
+	rcu_read_lock();
+	a_fence = dma_fence_get_rcu_safe(&a->fence);
+	b_fence = dma_fence_get_rcu_safe(&b->fence);
+	rcu_read_unlock();
+
+	a_fences = get_fences(&a_fence, &a_num_fences);
+	b_fences = get_fences(&b_fence, &b_num_fences);
+	if (!a_num_fences || !b_num_fences)
+		goto put_src_fences;
+
+	if (a_num_fences > INT_MAX - b_num_fences)
+		goto put_src_fences;
+
+	sync_file = sync_file_alloc(a->type, a->flags);
+	if (!sync_file)
+		goto put_src_fences;
+
 	num_fences = a_num_fences + b_num_fences;
 
 	fences = kcalloc(num_fences, sizeof(*fences), GFP_KERNEL);
@@ -314,11 +362,16 @@ static struct sync_file *sync_file_merge(const char *name, struct sync_file *a,
 		goto err;
 	}
 
+	dma_fence_put(a_fence);
+	dma_fence_put(b_fence);
 	strlcpy(sync_file->name, name, sizeof(sync_file->name));
 	return sync_file;
 
 err:
 	fput(sync_file->file);
+put_src_fences:
+	dma_fence_put(a_fence);
+	dma_fence_put(b_fence);
 	return NULL;
 
 }
@@ -327,10 +380,15 @@ static void sync_file_free(struct kref *kref)
 {
 	struct sync_file *sync_file = container_of(kref, struct sync_file,
 						     kref);
+	struct dma_fence *fence;
+
+	fence = rcu_dereference_protected(sync_file->fence, 1);
+	if (fence) {
+		if (test_bit(POLL_ENABLED, &fence->flags))
+			dma_fence_remove_callback(fence, &sync_file->cb);
+		dma_fence_put(fence);
+	}
 
-	if (test_bit(POLL_ENABLED, &sync_file->fence->flags))
-		dma_fence_remove_callback(sync_file->fence, &sync_file->cb);
-	dma_fence_put(sync_file->fence);
 	kfree(sync_file);
 }
 
@@ -345,16 +403,25 @@ static int sync_file_release(struct inode *inode, struct file *file)
 static unsigned int sync_file_poll(struct file *file, poll_table *wait)
 {
 	struct sync_file *sync_file = file->private_data;
+	unsigned int ret_val = 0;
+	struct dma_fence *fence;
 
 	poll_wait(file, &sync_file->wq, wait);
 
-	if (!test_and_set_bit(POLL_ENABLED, &sync_file->fence->flags)) {
-		if (dma_fence_add_callback(sync_file->fence, &sync_file->cb,
-					   fence_check_cb_func) < 0)
-			wake_up_all(&sync_file->wq);
+	mutex_lock(&sync_file->lock);
+
+	fence = sync_file_get_fence_locked(sync_file);
+	if (fence) {
+		if (!test_and_set_bit(POLL_ENABLED, &fence->flags)) {
+			if (dma_fence_add_callback(fence, &sync_file->cb,
+						   fence_check_cb_func) < 0)
+				wake_up_all(&sync_file->wq);
+		}
+		ret_val = dma_fence_is_signaled(fence) ? POLLIN : 0;
 	}
+	mutex_unlock(&sync_file->lock);
 
-	return dma_fence_is_signaled(sync_file->fence) ? POLLIN : 0;
+	return ret_val;
 }
 
 static long sync_file_ioctl_merge(struct sync_file *sync_file,
@@ -436,6 +503,7 @@ static long sync_file_ioctl_fence_info(struct sync_file *sync_file,
 	struct sync_file_info info;
 	struct sync_fence_info *fence_info = NULL;
 	struct dma_fence **fences;
+	struct dma_fence *fence;
 	__u32 size;
 	int num_fences, ret, i;
 
@@ -445,7 +513,15 @@ static long sync_file_ioctl_fence_info(struct sync_file *sync_file,
 	if (info.flags || info.pad)
 		return -EINVAL;
 
-	fences = get_fences(sync_file, &num_fences);
+	rcu_read_lock();
+	fence = dma_fence_get_rcu_safe(&sync_file->fence);
+	rcu_read_unlock();
+
+	fences = get_fences(&fence, &num_fences);
+
+	/* if there are no fences in the sync_file just return */
+	if (!num_fences)
+		goto no_fences;
 
 	/*
 	 * Passing num_fences = 0 means that userspace doesn't want to
@@ -456,13 +532,17 @@ static long sync_file_ioctl_fence_info(struct sync_file *sync_file,
 	if (!info.num_fences)
 		goto no_fences;
 
-	if (info.num_fences < num_fences)
-		return -EINVAL;
+	if (info.num_fences < num_fences) {
+		ret = -EINVAL;
+		goto out;
+	}
 
 	size = num_fences * sizeof(*fence_info);
 	fence_info = kzalloc(size, GFP_KERNEL);
-	if (!fence_info)
-		return -ENOMEM;
+	if (!fence_info) {
+		ret = -ENOMEM;
+		goto out;
+	}
 
 	for (i = 0; i < num_fences; i++)
 		sync_fill_fence_info(fences[i], &fence_info[i]);
@@ -475,7 +555,10 @@ static long sync_file_ioctl_fence_info(struct sync_file *sync_file,
 
 no_fences:
 	strlcpy(info.name, sync_file->name, sizeof(info.name));
-	info.status = dma_fence_is_signaled(sync_file->fence);
+	if (num_fences)
+		info.status = dma_fence_is_signaled(sync_file->fence);
+	else
+		info.status = -ENOENT;
 	info.num_fences = num_fences;
 
 	if (copy_to_user((void __user *)arg, &info, sizeof(info)))
@@ -485,7 +568,7 @@ static long sync_file_ioctl_fence_info(struct sync_file *sync_file,
 
 out:
 	kfree(fence_info);
-
+	dma_fence_put(fence);
 	return ret;
 }
 
diff --git a/include/linux/sync_file.h b/include/linux/sync_file.h
index e683dd1..4bf661b 100644
--- a/include/linux/sync_file.h
+++ b/include/linux/sync_file.h
@@ -34,6 +34,7 @@
  * @cb:			fence callback information
  * @type:               sync file type
  * @flags:              flags used to create sync file
+ * @lock:               mutex to protect fence/cb - used for semaphores
  */
 struct sync_file {
 	struct file		*file;
@@ -45,10 +46,13 @@ struct sync_file {
 
 	wait_queue_head_t	wq;
 
-	struct dma_fence	*fence;
+	struct dma_fence __rcu	*fence;
 	struct dma_fence_cb cb;
 	uint32_t type;
 	uint32_t flags;
+
+	/* protects the fence pointer and cb */
+	struct mutex lock;
 };
 
 #define POLL_ENABLED DMA_FENCE_FLAG_USER_BITS
-- 
2.9.3

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 6/8] drm/syncobj: add semaphore support helpers. (v2)

[Dave Airlie]

From: Dave Airlie <airlied@redhat.com>

This just adds two helper interfaces to bridge the gap from
drivers to sync_file for the semaphore objects.

These will be used by the amdgpu driver.

v2: drop one of the APIs and replace with a fence
lookup to make the amdgpu api more robust.

Signed-off-by: Dave Airlie <airlied@redhat.com>
---
 drivers/gpu/drm/drm_syncobj.c | 60 +++++++++++++++++++++++++++++++++++++++++++
 include/drm/drm_syncobj.h     | 36 ++++++++++++++++++++++++++
 2 files changed, 96 insertions(+)
 create mode 100644 include/drm/drm_syncobj.h

diff --git a/drivers/gpu/drm/drm_syncobj.c b/drivers/gpu/drm/drm_syncobj.c
index 4827f90..fb135ae 100644
--- a/drivers/gpu/drm/drm_syncobj.c
+++ b/drivers/gpu/drm/drm_syncobj.c
@@ -39,6 +39,7 @@
 #include <drm/drmP.h>
 #include <linux/sync_file.h>
 #include "drm_internal.h"
+#include <drm/drm_syncobj.h>
 
 static struct sync_file *drm_syncobj_get(struct drm_file *file_private,
 					 u32 handle)
@@ -57,6 +58,65 @@ static struct sync_file *drm_syncobj_get(struct drm_file *file_private,
 	return sync_file;
 }
 
+static int drm_syncobj_swap_fences_nocheck(struct drm_file *file_private,
+					   uint32_t handle,
+					   struct dma_fence *fence,
+					   struct dma_fence **old_fence)
+{
+	struct sync_file *sync_file = drm_syncobj_get(file_private, handle);
+	int ret;
+
+	if (!sync_file)
+		return -EINVAL;
+
+	ret = sync_file_replace_fence(sync_file, fence, old_fence);
+	fput(sync_file->file);
+	return ret;
+}
+
+/**
+ * drm_syncobj_replace_fence - lookup and replace fence in a sync object.
+ * @file_private - drm file private pointer.
+ * @handle - syncobj handle to lookup
+ * @fence - fence to install in sync file.
+ * Returns:
+ * 0 on success, or -EINVAL when the handle doesn't point at a valid
+ * sync_file with %SYNC_FILE_TYPE_SEMAPHORE.
+ *
+ * This looks up a sync object and replaces the fence on it, freeing
+ * the old one.
+ */
+int drm_syncobj_replace_fence(struct drm_file *file_private,
+			      u32 handle,
+			      struct dma_fence *fence)
+{
+	struct dma_fence *old_fence;
+	int r;
+
+	r = drm_syncobj_swap_fences_nocheck(file_private, handle, fence, &old_fence);
+	if (r)
+		return r;
+	dma_fence_put(old_fence);
+	return 0;
+}
+EXPORT_SYMBOL(drm_syncobj_replace_fence);
+
+int drm_syncobj_fence_get(struct drm_file *file_private,
+			  u32 handle,
+			  struct dma_fence **fence)
+{
+	struct sync_file *sync_file = drm_syncobj_get(file_private, handle);
+
+	if (!sync_file)
+		return -EINVAL;
+
+	*fence = sync_file_get_fence_from_sync_file(sync_file);
+	if (!*fence)
+		return -EINVAL;
+	return 0;
+}
+EXPORT_SYMBOL(drm_syncobj_fence_get);
+
 static int drm_syncobj_create(struct drm_file *file_private,
 			      unsigned type,
 			      unsigned flags, u32 *handle)
diff --git a/include/drm/drm_syncobj.h b/include/drm/drm_syncobj.h
new file mode 100644
index 0000000..714b658
--- /dev/null
+++ b/include/drm/drm_syncobj.h
@@ -0,0 +1,36 @@
+/*
+ * Copyright © 2017 Red Hat
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Authors:
+ *
+ */
+#ifndef __DRM_SYNCOBJ_H__
+#define __DRM_SYNCOBJ_H__
+
+int drm_syncobj_fence_get(struct drm_file *file_private,
+			  u32 handle,
+			  struct dma_fence **fence);
+int drm_syncobj_replace_fence(struct drm_file *file_private,
+			      u32 handle,
+			      struct dma_fence *fence);
+
+#endif
-- 
2.9.3

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 7/8] amdgpu/cs: split out fence dependency checking

[Dave Airlie]

From: Dave Airlie <airlied@redhat.com>

This just splits out the fence depenency checking into it's
own function to make it easier to add semaphore dependencies.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 85 +++++++++++++++++++---------------
 1 file changed, 47 insertions(+), 38 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 99424cb..df25b32 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -963,56 +963,65 @@ static int amdgpu_cs_ib_fill(struct amdgpu_device *adev,
 	return 0;
 }
 
-static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
-				  struct amdgpu_cs_parser *p)
+static int amdgpu_process_fence_dep(struct amdgpu_cs_parser *p,
+				    struct amdgpu_cs_chunk *chunk)
 {
 	struct amdgpu_fpriv *fpriv = p->filp->driver_priv;
-	int i, j, r;
-
-	for (i = 0; i < p->nchunks; ++i) {
-		struct drm_amdgpu_cs_chunk_dep *deps;
-		struct amdgpu_cs_chunk *chunk;
-		unsigned num_deps;
+	unsigned num_deps;
+	int i, r;
+	struct drm_amdgpu_cs_chunk_dep *deps;
 
-		chunk = &p->chunks[i];
+	deps = (struct drm_amdgpu_cs_chunk_dep *)chunk->kdata;
+	num_deps = chunk->length_dw * 4 /
+		sizeof(struct drm_amdgpu_cs_chunk_dep);
 
-		if (chunk->chunk_id != AMDGPU_CHUNK_ID_DEPENDENCIES)
-			continue;
+	for (i = 0; i < num_deps; ++i) {
+		struct amdgpu_ring *ring;
+		struct amdgpu_ctx *ctx;
+		struct dma_fence *fence;
 
-		deps = (struct drm_amdgpu_cs_chunk_dep *)chunk->kdata;
-		num_deps = chunk->length_dw * 4 /
-			sizeof(struct drm_amdgpu_cs_chunk_dep);
+		r = amdgpu_cs_get_ring(p->adev, deps[i].ip_type,
+				       deps[i].ip_instance,
+				       deps[i].ring, &ring);
+		if (r)
+			return r;
 
-		for (j = 0; j < num_deps; ++j) {
-			struct amdgpu_ring *ring;
-			struct amdgpu_ctx *ctx;
-			struct dma_fence *fence;
+		ctx = amdgpu_ctx_get(fpriv, deps[i].ctx_id);
+		if (ctx == NULL)
+			return -EINVAL;
 
-			r = amdgpu_cs_get_ring(adev, deps[j].ip_type,
-					       deps[j].ip_instance,
-					       deps[j].ring, &ring);
+		fence = amdgpu_ctx_get_fence(ctx, ring,
+					     deps[i].handle);
+		if (IS_ERR(fence)) {
+			r = PTR_ERR(fence);
+			amdgpu_ctx_put(ctx);
+			return r;
+		} else if (fence) {
+			r = amdgpu_sync_fence(p->adev, &p->job->sync,
+					      fence);
+			dma_fence_put(fence);
+			amdgpu_ctx_put(ctx);
 			if (r)
 				return r;
+		}
+	}
+	return 0;
+}
 
-			ctx = amdgpu_ctx_get(fpriv, deps[j].ctx_id);
-			if (ctx == NULL)
-				return -EINVAL;
+static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
+				  struct amdgpu_cs_parser *p)
+{
+	int i, r;
 
-			fence = amdgpu_ctx_get_fence(ctx, ring,
-						     deps[j].handle);
-			if (IS_ERR(fence)) {
-				r = PTR_ERR(fence);
-				amdgpu_ctx_put(ctx);
-				return r;
+	for (i = 0; i < p->nchunks; ++i) {
+		struct amdgpu_cs_chunk *chunk;
 
-			} else if (fence) {
-				r = amdgpu_sync_fence(adev, &p->job->sync,
-						      fence);
-				dma_fence_put(fence);
-				amdgpu_ctx_put(ctx);
-				if (r)
-					return r;
-			}
+		chunk = &p->chunks[i];
+
+		if (chunk->chunk_id == AMDGPU_CHUNK_ID_DEPENDENCIES) {
+			r = amdgpu_process_fence_dep(p, chunk);
+			if (r)
+				return r;
 		}
 	}
 
-- 
2.9.3

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH 8/8] amdgpu: use sync file for shared semaphores (v3)

[Dave Airlie]

From: Dave Airlie <airlied@redhat.com>

This creates a new command submission chunk for amdgpu
to add wait and signal sync objects around the submission.

Sync objects are managed via the drm syncobj ioctls.

The command submission interface is enhanced with two new
chunks, one for semaphore waiting, one for semaphore signalling
and just takes a list of handles for each.

This is based on work originally done by David Zhou at AMD,
with input from Christian Konig on what things should look like.

NOTE: this interface addition needs a version bump to expose
it to userspace.

v1.1: keep file reference on import.
v2: move to using syncobjs
v2.1: change some APIs to just use p pointer.
v3: make more robust against CS failures, we now add the
wait sems but only remove them once the CS job has been
submitted.

Signed-off-by: Dave Airlie <airlied@redhat.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c  | 120 +++++++++++++++++++++++++++++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c |   2 +-
 include/uapi/drm/amdgpu_drm.h           |   6 ++
 3 files changed, 126 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index df25b32..e9b5796 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -27,6 +27,7 @@
 #include <linux/pagemap.h>
 #include <drm/drmP.h>
 #include <drm/amdgpu_drm.h>
+#include <drm/drm_syncobj.h>
 #include "amdgpu.h"
 #include "amdgpu_trace.h"
 
@@ -217,6 +218,8 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
 			break;
 
 		case AMDGPU_CHUNK_ID_DEPENDENCIES:
+		case AMDGPU_CHUNK_ID_SEM_WAIT:
+		case AMDGPU_CHUNK_ID_SEM_SIGNAL:
 			break;
 
 		default:
@@ -1008,6 +1011,73 @@ static int amdgpu_process_fence_dep(struct amdgpu_cs_parser *p,
 	return 0;
 }
 
+static int amdgpu_sem_lookup_and_add_to_sync(struct amdgpu_cs_parser *p,
+					     uint32_t handle)
+{
+	int r;
+	struct dma_fence *fence;
+	r = drm_syncobj_fence_get(p->filp, handle, &fence);
+	if (r)
+		return r;
+
+	r = amdgpu_sync_fence(p->adev, &p->job->sync, fence);
+	dma_fence_put(fence);
+
+	return r;
+}
+
+static int amdgpu_sem_lookup_and_remove(struct amdgpu_cs_parser *p,
+				      uint32_t handle)
+{
+	int r;
+	struct dma_fence *old_fence;
+
+	r = drm_syncobj_replace_fence(p->filp, handle, NULL);
+	if (r)
+		return r;
+	dma_fence_put(old_fence);
+
+	return r;
+}
+
+static int amdgpu_process_sem_wait_dep(struct amdgpu_cs_parser *p,
+				       struct amdgpu_cs_chunk *chunk)
+{
+	unsigned num_deps;
+	int i, r;
+	struct drm_amdgpu_cs_chunk_sem *deps;
+
+	deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
+	num_deps = chunk->length_dw * 4 /
+		sizeof(struct drm_amdgpu_cs_chunk_sem);
+
+	for (i = 0; i < num_deps; ++i) {
+		r = amdgpu_sem_lookup_and_add_to_sync(p, deps[i].handle);
+		if (r)
+			return r;
+	}
+	return 0;
+}
+
+static int amdgpu_process_sem_wait_post(struct amdgpu_cs_parser *p,
+					struct amdgpu_cs_chunk *chunk)
+{
+	unsigned num_deps;
+	int i, r;
+	struct drm_amdgpu_cs_chunk_sem *deps;
+
+	deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
+	num_deps = chunk->length_dw * 4 /
+		sizeof(struct drm_amdgpu_cs_chunk_sem);
+
+	for (i = 0; i < num_deps; ++i) {
+		r = amdgpu_sem_lookup_and_remove(p, deps[i].handle);
+		if (r)
+			return r;
+	}
+	return 0;
+}
+
 static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
 				  struct amdgpu_cs_parser *p)
 {
@@ -1022,12 +1092,60 @@ static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
 			r = amdgpu_process_fence_dep(p, chunk);
 			if (r)
 				return r;
+		} else if (chunk->chunk_id == AMDGPU_CHUNK_ID_SEM_WAIT) {
+			r = amdgpu_process_sem_wait_dep(p, chunk);
+			if (r)
+				return r;
 		}
 	}
 
 	return 0;
 }
 
+static int amdgpu_process_sem_signal_dep(struct amdgpu_cs_parser *p,
+					 struct amdgpu_cs_chunk *chunk)
+{
+	unsigned num_deps;
+	int i, r;
+	struct drm_amdgpu_cs_chunk_sem *deps;
+
+	deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
+	num_deps = chunk->length_dw * 4 /
+		sizeof(struct drm_amdgpu_cs_chunk_sem);
+
+	for (i = 0; i < num_deps; ++i) {
+		r = drm_syncobj_replace_fence(p->filp, deps[i].handle,
+					      p->fence);
+		if (r)
+			return r;
+	}
+	return 0;
+}
+
+static int amdgpu_cs_post_dependencies(struct amdgpu_cs_parser *p)
+{
+	int i, r;
+
+	for (i = 0; i < p->nchunks; ++i) {
+		struct amdgpu_cs_chunk *chunk;
+
+		chunk = &p->chunks[i];
+
+		if (chunk->chunk_id == AMDGPU_CHUNK_ID_SEM_WAIT) {
+			r = amdgpu_process_sem_wait_post(p, chunk);
+			if (r)
+				return r;
+		}
+
+		if (chunk->chunk_id == AMDGPU_CHUNK_ID_SEM_SIGNAL) {
+			r = amdgpu_process_sem_signal_dep(p, chunk);
+			if (r)
+				return r;
+		}
+	}
+	return 0;
+}
+
 static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
 			    union drm_amdgpu_cs *cs)
 {
@@ -1055,7 +1173,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
 	trace_amdgpu_cs_ioctl(job);
 	amd_sched_entity_push_job(&job->base);
 
-	return 0;
+	return amdgpu_cs_post_dependencies(p);
 }
 
 int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index b76cd69..e95951e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -683,7 +683,7 @@ static struct drm_driver kms_driver = {
 	.driver_features =
 	    DRIVER_USE_AGP |
 	    DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | DRIVER_GEM |
-	    DRIVER_PRIME | DRIVER_RENDER | DRIVER_MODESET,
+	    DRIVER_PRIME | DRIVER_RENDER | DRIVER_MODESET | DRIVER_SYNCOBJ,
 	.load = amdgpu_driver_load_kms,
 	.open = amdgpu_driver_open_kms,
 	.preclose = amdgpu_driver_preclose_kms,
diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index 5797283..647c520 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -390,6 +390,8 @@ struct drm_amdgpu_gem_va {
 #define AMDGPU_CHUNK_ID_IB		0x01
 #define AMDGPU_CHUNK_ID_FENCE		0x02
 #define AMDGPU_CHUNK_ID_DEPENDENCIES	0x03
+#define AMDGPU_CHUNK_ID_SEM_WAIT        0x04
+#define AMDGPU_CHUNK_ID_SEM_SIGNAL      0x05
 
 struct drm_amdgpu_cs_chunk {
 	__u32		chunk_id;
@@ -454,6 +456,10 @@ struct drm_amdgpu_cs_chunk_fence {
 	__u32 offset;
 };
 
+struct drm_amdgpu_cs_chunk_sem {
+	__u32 handle;
+};
+
 struct drm_amdgpu_cs_chunk_data {
 	union {
 		struct drm_amdgpu_cs_chunk_ib		ib_data;
-- 
2.9.3

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

Re: [PATCH 8/8] amdgpu: use sync file for shared semaphores (v3)

[Christian König]

Am 12.04.2017 um 06:57 schrieb Dave Airlie:
> From: Dave Airlie <airlied@redhat.com>
>
> This creates a new command submission chunk for amdgpu
> to add wait and signal sync objects around the submission.
>
> Sync objects are managed via the drm syncobj ioctls.
>
> The command submission interface is enhanced with two new
> chunks, one for semaphore waiting, one for semaphore signalling
> and just takes a list of handles for each.
>
> This is based on work originally done by David Zhou at AMD,
> with input from Christian Konig on what things should look like.
>
> NOTE: this interface addition needs a version bump to expose
> it to userspace.
>
> v1.1: keep file reference on import.
> v2: move to using syncobjs
> v2.1: change some APIs to just use p pointer.
> v3: make more robust against CS failures, we now add the
> wait sems but only remove them once the CS job has been
> submitted.
>
> Signed-off-by: Dave Airlie <airlied@redhat.com>
> ---
>   drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c  | 120 +++++++++++++++++++++++++++++++-
>   drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c |   2 +-
>   include/uapi/drm/amdgpu_drm.h           |   6 ++
>   3 files changed, 126 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> index df25b32..e9b5796 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> @@ -27,6 +27,7 @@
>   #include <linux/pagemap.h>
>   #include <drm/drmP.h>
>   #include <drm/amdgpu_drm.h>
> +#include <drm/drm_syncobj.h>
>   #include "amdgpu.h"
>   #include "amdgpu_trace.h"
>   
> @@ -217,6 +218,8 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
>   			break;
>   
>   		case AMDGPU_CHUNK_ID_DEPENDENCIES:
> +		case AMDGPU_CHUNK_ID_SEM_WAIT:
> +		case AMDGPU_CHUNK_ID_SEM_SIGNAL:
>   			break;
>   
>   		default:
> @@ -1008,6 +1011,73 @@ static int amdgpu_process_fence_dep(struct amdgpu_cs_parser *p,
>   	return 0;
>   }
>   
> +static int amdgpu_sem_lookup_and_add_to_sync(struct amdgpu_cs_parser *p,
> +					     uint32_t handle)
> +{
> +	int r;
> +	struct dma_fence *fence;

Missing new line between declaration and code.

> +	r = drm_syncobj_fence_get(p->filp, handle, &fence);
> +	if (r)
> +		return r;
> +
> +	r = amdgpu_sync_fence(p->adev, &p->job->sync, fence);
> +	dma_fence_put(fence);
> +
> +	return r;
> +}
> +
> +static int amdgpu_sem_lookup_and_remove(struct amdgpu_cs_parser *p,
> +				      uint32_t handle)
> +{
> +	int r;
> +	struct dma_fence *old_fence;
> +
> +	r = drm_syncobj_replace_fence(p->filp, handle, NULL);
> +	if (r)
> +		return r;
> +	dma_fence_put(old_fence);

Am I wrong or is old_fence not initialized here?

Additional to that what happens when the fence in the sync object was 
changed while we do the CS? Or even worse the handle got assigned to a 
new sync object.

Regards,
Christian.

> +
> +	return r;
> +}
> +
> +static int amdgpu_process_sem_wait_dep(struct amdgpu_cs_parser *p,
> +				       struct amdgpu_cs_chunk *chunk)
> +{
> +	unsigned num_deps;
> +	int i, r;
> +	struct drm_amdgpu_cs_chunk_sem *deps;
> +
> +	deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
> +	num_deps = chunk->length_dw * 4 /
> +		sizeof(struct drm_amdgpu_cs_chunk_sem);
> +
> +	for (i = 0; i < num_deps; ++i) {
> +		r = amdgpu_sem_lookup_and_add_to_sync(p, deps[i].handle);
> +		if (r)
> +			return r;
> +	}
> +	return 0;
> +}
> +
> +static int amdgpu_process_sem_wait_post(struct amdgpu_cs_parser *p,
> +					struct amdgpu_cs_chunk *chunk)
> +{
> +	unsigned num_deps;
> +	int i, r;
> +	struct drm_amdgpu_cs_chunk_sem *deps;
> +
> +	deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
> +	num_deps = chunk->length_dw * 4 /
> +		sizeof(struct drm_amdgpu_cs_chunk_sem);
> +
> +	for (i = 0; i < num_deps; ++i) {
> +		r = amdgpu_sem_lookup_and_remove(p, deps[i].handle);
> +		if (r)
> +			return r;
> +	}
> +	return 0;
> +}
> +
>   static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
>   				  struct amdgpu_cs_parser *p)
>   {
> @@ -1022,12 +1092,60 @@ static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
>   			r = amdgpu_process_fence_dep(p, chunk);
>   			if (r)
>   				return r;
> +		} else if (chunk->chunk_id == AMDGPU_CHUNK_ID_SEM_WAIT) {
> +			r = amdgpu_process_sem_wait_dep(p, chunk);
> +			if (r)
> +				return r;
>   		}
>   	}
>   
>   	return 0;
>   }
>   
> +static int amdgpu_process_sem_signal_dep(struct amdgpu_cs_parser *p,
> +					 struct amdgpu_cs_chunk *chunk)
> +{
> +	unsigned num_deps;
> +	int i, r;
> +	struct drm_amdgpu_cs_chunk_sem *deps;
> +
> +	deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
> +	num_deps = chunk->length_dw * 4 /
> +		sizeof(struct drm_amdgpu_cs_chunk_sem);
> +
> +	for (i = 0; i < num_deps; ++i) {
> +		r = drm_syncobj_replace_fence(p->filp, deps[i].handle,
> +					      p->fence);
> +		if (r)
> +			return r;
> +	}
> +	return 0;
> +}
> +
> +static int amdgpu_cs_post_dependencies(struct amdgpu_cs_parser *p)
> +{
> +	int i, r;
> +
> +	for (i = 0; i < p->nchunks; ++i) {
> +		struct amdgpu_cs_chunk *chunk;
> +
> +		chunk = &p->chunks[i];
> +
> +		if (chunk->chunk_id == AMDGPU_CHUNK_ID_SEM_WAIT) {
> +			r = amdgpu_process_sem_wait_post(p, chunk);
> +			if (r)
> +				return r;
> +		}
> +
> +		if (chunk->chunk_id == AMDGPU_CHUNK_ID_SEM_SIGNAL) {
> +			r = amdgpu_process_sem_signal_dep(p, chunk);
> +			if (r)
> +				return r;
> +		}
> +	}
> +	return 0;
> +}
> +
>   static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
>   			    union drm_amdgpu_cs *cs)
>   {
> @@ -1055,7 +1173,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
>   	trace_amdgpu_cs_ioctl(job);
>   	amd_sched_entity_push_job(&job->base);
>   
> -	return 0;
> +	return amdgpu_cs_post_dependencies(p);
>   }
>   
>   int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
> index b76cd69..e95951e 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
> @@ -683,7 +683,7 @@ static struct drm_driver kms_driver = {
>   	.driver_features =
>   	    DRIVER_USE_AGP |
>   	    DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | DRIVER_GEM |
> -	    DRIVER_PRIME | DRIVER_RENDER | DRIVER_MODESET,
> +	    DRIVER_PRIME | DRIVER_RENDER | DRIVER_MODESET | DRIVER_SYNCOBJ,
>   	.load = amdgpu_driver_load_kms,
>   	.open = amdgpu_driver_open_kms,
>   	.preclose = amdgpu_driver_preclose_kms,
> diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
> index 5797283..647c520 100644
> --- a/include/uapi/drm/amdgpu_drm.h
> +++ b/include/uapi/drm/amdgpu_drm.h
> @@ -390,6 +390,8 @@ struct drm_amdgpu_gem_va {
>   #define AMDGPU_CHUNK_ID_IB		0x01
>   #define AMDGPU_CHUNK_ID_FENCE		0x02
>   #define AMDGPU_CHUNK_ID_DEPENDENCIES	0x03
> +#define AMDGPU_CHUNK_ID_SEM_WAIT        0x04
> +#define AMDGPU_CHUNK_ID_SEM_SIGNAL      0x05
>   
>   struct drm_amdgpu_cs_chunk {
>   	__u32		chunk_id;
> @@ -454,6 +456,10 @@ struct drm_amdgpu_cs_chunk_fence {
>   	__u32 offset;
>   };
>   
> +struct drm_amdgpu_cs_chunk_sem {
> +	__u32 handle;
> +};
> +
>   struct drm_amdgpu_cs_chunk_data {
>   	union {
>   		struct drm_amdgpu_cs_chunk_ib		ib_data;


_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

Re: [PATCH 8/8] amdgpu: use sync file for shared semaphores (v3)

[Chris Wilson]

On Wed, Apr 12, 2017 at 11:55:24AM +0200, Christian König wrote:
> Am 12.04.2017 um 06:57 schrieb Dave Airlie:
> >+static int amdgpu_sem_lookup_and_remove(struct amdgpu_cs_parser *p,
> >+				      uint32_t handle)
> >+{
> >+	int r;
> >+	struct dma_fence *old_fence;
> >+
> >+	r = drm_syncobj_replace_fence(p->filp, handle, NULL);
> >+	if (r)
> >+		return r;
> >+	dma_fence_put(old_fence);
> 
> Am I wrong or is old_fence not initialized here?
> 
> Additional to that what happens when the fence in the sync object
> was changed while we do the CS? Or even worse the handle got
> assigned to a new sync object.

We either ww_mutex the lot, or regard that as a userspace race where the
order between the two concurrent CS emits is undefined and who gets the
in-semaphore is happenstance.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

Re: [PATCH 8/8] amdgpu: use sync file for shared semaphores (v3)

[Christian König]

Am 12.04.2017 um 12:18 schrieb Chris Wilson:
> On Wed, Apr 12, 2017 at 11:55:24AM +0200, Christian König wrote:
>> Am 12.04.2017 um 06:57 schrieb Dave Airlie:
>>> +static int amdgpu_sem_lookup_and_remove(struct amdgpu_cs_parser *p,
>>> +				      uint32_t handle)
>>> +{
>>> +	int r;
>>> +	struct dma_fence *old_fence;
>>> +
>>> +	r = drm_syncobj_replace_fence(p->filp, handle, NULL);
>>> +	if (r)
>>> +		return r;
>>> +	dma_fence_put(old_fence);
>> Am I wrong or is old_fence not initialized here?
>>
>> Additional to that what happens when the fence in the sync object
>> was changed while we do the CS? Or even worse the handle got
>> assigned to a new sync object.
> We either ww_mutex the lot, or regard that as a userspace race where the
> order between the two concurrent CS emits is undefined and who gets the
> in-semaphore is happenstance.

My thinking to solve this was rather a) resolve the handle to an object 
only once and b) replace the old fence with NULL only if it didn't changed.

Letting userspace race here is also an option, but I would prefer it 
another solution.

Christian.

> -Chris
>

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

Re: [PATCH 8/8] amdgpu: use sync file for shared semaphores (v3)

[Alex Deucher]

On Wed, Apr 12, 2017 at 12:57 AM, Dave Airlie <airlied@gmail.com> wrote:
> From: Dave Airlie <airlied@redhat.com>
>
> This creates a new command submission chunk for amdgpu
> to add wait and signal sync objects around the submission.
>
> Sync objects are managed via the drm syncobj ioctls.
>
> The command submission interface is enhanced with two new
> chunks, one for semaphore waiting, one for semaphore signalling
> and just takes a list of handles for each.
>
> This is based on work originally done by David Zhou at AMD,
> with input from Christian Konig on what things should look like.
>
> NOTE: this interface addition needs a version bump to expose
> it to userspace.
>
> v1.1: keep file reference on import.
> v2: move to using syncobjs
> v2.1: change some APIs to just use p pointer.
> v3: make more robust against CS failures, we now add the
> wait sems but only remove them once the CS job has been
> submitted.
>
> Signed-off-by: Dave Airlie <airlied@redhat.com>
> ---
>  drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c  | 120 +++++++++++++++++++++++++++++++-
>  drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c |   2 +-
>  include/uapi/drm/amdgpu_drm.h           |   6 ++
>  3 files changed, 126 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> index df25b32..e9b5796 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> @@ -27,6 +27,7 @@
>  #include <linux/pagemap.h>
>  #include <drm/drmP.h>
>  #include <drm/amdgpu_drm.h>
> +#include <drm/drm_syncobj.h>
>  #include "amdgpu.h"
>  #include "amdgpu_trace.h"
>
> @@ -217,6 +218,8 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
>                         break;
>
>                 case AMDGPU_CHUNK_ID_DEPENDENCIES:
> +               case AMDGPU_CHUNK_ID_SEM_WAIT:
> +               case AMDGPU_CHUNK_ID_SEM_SIGNAL:
>                         break;
>
>                 default:
> @@ -1008,6 +1011,73 @@ static int amdgpu_process_fence_dep(struct amdgpu_cs_parser *p,
>         return 0;
>  }
>
> +static int amdgpu_sem_lookup_and_add_to_sync(struct amdgpu_cs_parser *p,
> +                                            uint32_t handle)
> +{
> +       int r;
> +       struct dma_fence *fence;
> +       r = drm_syncobj_fence_get(p->filp, handle, &fence);
> +       if (r)
> +               return r;
> +
> +       r = amdgpu_sync_fence(p->adev, &p->job->sync, fence);
> +       dma_fence_put(fence);
> +
> +       return r;
> +}
> +
> +static int amdgpu_sem_lookup_and_remove(struct amdgpu_cs_parser *p,
> +                                     uint32_t handle)
> +{
> +       int r;
> +       struct dma_fence *old_fence;
> +
> +       r = drm_syncobj_replace_fence(p->filp, handle, NULL);
> +       if (r)
> +               return r;
> +       dma_fence_put(old_fence);
> +
> +       return r;
> +}
> +
> +static int amdgpu_process_sem_wait_dep(struct amdgpu_cs_parser *p,
> +                                      struct amdgpu_cs_chunk *chunk)
> +{
> +       unsigned num_deps;
> +       int i, r;
> +       struct drm_amdgpu_cs_chunk_sem *deps;
> +
> +       deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
> +       num_deps = chunk->length_dw * 4 /
> +               sizeof(struct drm_amdgpu_cs_chunk_sem);
> +
> +       for (i = 0; i < num_deps; ++i) {
> +               r = amdgpu_sem_lookup_and_add_to_sync(p, deps[i].handle);
> +               if (r)
> +                       return r;
> +       }
> +       return 0;
> +}
> +
> +static int amdgpu_process_sem_wait_post(struct amdgpu_cs_parser *p,
> +                                       struct amdgpu_cs_chunk *chunk)
> +{
> +       unsigned num_deps;
> +       int i, r;
> +       struct drm_amdgpu_cs_chunk_sem *deps;
> +
> +       deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
> +       num_deps = chunk->length_dw * 4 /
> +               sizeof(struct drm_amdgpu_cs_chunk_sem);
> +
> +       for (i = 0; i < num_deps; ++i) {
> +               r = amdgpu_sem_lookup_and_remove(p, deps[i].handle);
> +               if (r)
> +                       return r;
> +       }
> +       return 0;
> +}
> +
>  static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
>                                   struct amdgpu_cs_parser *p)
>  {
> @@ -1022,12 +1092,60 @@ static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
>                         r = amdgpu_process_fence_dep(p, chunk);
>                         if (r)
>                                 return r;
> +               } else if (chunk->chunk_id == AMDGPU_CHUNK_ID_SEM_WAIT) {
> +                       r = amdgpu_process_sem_wait_dep(p, chunk);
> +                       if (r)
> +                               return r;
>                 }
>         }
>
>         return 0;
>  }
>
> +static int amdgpu_process_sem_signal_dep(struct amdgpu_cs_parser *p,
> +                                        struct amdgpu_cs_chunk *chunk)
> +{
> +       unsigned num_deps;
> +       int i, r;
> +       struct drm_amdgpu_cs_chunk_sem *deps;
> +
> +       deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
> +       num_deps = chunk->length_dw * 4 /
> +               sizeof(struct drm_amdgpu_cs_chunk_sem);
> +
> +       for (i = 0; i < num_deps; ++i) {
> +               r = drm_syncobj_replace_fence(p->filp, deps[i].handle,
> +                                             p->fence);
> +               if (r)
> +                       return r;
> +       }
> +       return 0;
> +}
> +
> +static int amdgpu_cs_post_dependencies(struct amdgpu_cs_parser *p)
> +{
> +       int i, r;
> +
> +       for (i = 0; i < p->nchunks; ++i) {
> +               struct amdgpu_cs_chunk *chunk;
> +
> +               chunk = &p->chunks[i];
> +
> +               if (chunk->chunk_id == AMDGPU_CHUNK_ID_SEM_WAIT) {
> +                       r = amdgpu_process_sem_wait_post(p, chunk);
> +                       if (r)
> +                               return r;
> +               }
> +
> +               if (chunk->chunk_id == AMDGPU_CHUNK_ID_SEM_SIGNAL) {
> +                       r = amdgpu_process_sem_signal_dep(p, chunk);
> +                       if (r)
> +                               return r;
> +               }
> +       }
> +       return 0;
> +}
> +
>  static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
>                             union drm_amdgpu_cs *cs)
>  {
> @@ -1055,7 +1173,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
>         trace_amdgpu_cs_ioctl(job);
>         amd_sched_entity_push_job(&job->base);
>
> -       return 0;
> +       return amdgpu_cs_post_dependencies(p);
>  }
>
>  int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
> index b76cd69..e95951e 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
> @@ -683,7 +683,7 @@ static struct drm_driver kms_driver = {
>         .driver_features =
>             DRIVER_USE_AGP |
>             DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | DRIVER_GEM |
> -           DRIVER_PRIME | DRIVER_RENDER | DRIVER_MODESET,
> +           DRIVER_PRIME | DRIVER_RENDER | DRIVER_MODESET | DRIVER_SYNCOBJ,
>         .load = amdgpu_driver_load_kms,
>         .open = amdgpu_driver_open_kms,
>         .preclose = amdgpu_driver_preclose_kms,
> diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
> index 5797283..647c520 100644
> --- a/include/uapi/drm/amdgpu_drm.h
> +++ b/include/uapi/drm/amdgpu_drm.h
> @@ -390,6 +390,8 @@ struct drm_amdgpu_gem_va {
>  #define AMDGPU_CHUNK_ID_IB             0x01
>  #define AMDGPU_CHUNK_ID_FENCE          0x02
>  #define AMDGPU_CHUNK_ID_DEPENDENCIES   0x03
> +#define AMDGPU_CHUNK_ID_SEM_WAIT        0x04
> +#define AMDGPU_CHUNK_ID_SEM_SIGNAL      0x05
>
>  struct drm_amdgpu_cs_chunk {
>         __u32           chunk_id;
> @@ -454,6 +456,10 @@ struct drm_amdgpu_cs_chunk_fence {
>         __u32 offset;
>  };
>
> +struct drm_amdgpu_cs_chunk_sem {
> +       __u32 handle;


Do we need a pad here?

Alex

> +};
> +
>  struct drm_amdgpu_cs_chunk_data {
>         union {
>                 struct drm_amdgpu_cs_chunk_ib           ib_data;
> --
> 2.9.3
>
> _______________________________________________
> dri-devel mailing list
> dri-devel@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/dri-devel
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx