[PATCH libdrm 1/7] addr cs chunk for syncobj timeline

[PATCH libdrm 1/7] addr cs chunk for syncobj timeline

[Chunming Zhou]

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
---
 include/drm/amdgpu_drm.h | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/include/drm/amdgpu_drm.h b/include/drm/amdgpu_drm.h
index d0701ffc..3d0318e6 100644
--- a/include/drm/amdgpu_drm.h
+++ b/include/drm/amdgpu_drm.h
@@ -528,6 +528,8 @@ struct drm_amdgpu_gem_va {
 #define AMDGPU_CHUNK_ID_SYNCOBJ_OUT     0x05
 #define AMDGPU_CHUNK_ID_BO_HANDLES      0x06
 #define AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES	0x07
+#define AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_WAIT    0x08
+#define AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_SIGNAL  0x09
 
 struct drm_amdgpu_cs_chunk {
 	__u32		chunk_id;
@@ -608,6 +610,13 @@ struct drm_amdgpu_cs_chunk_sem {
 	__u32 handle;
 };
 
+struct drm_amdgpu_cs_chunk_syncobj {
+	__u32 handle;
+	__u32 flags;
+	__u64 point;
+};
+
+
 #define AMDGPU_FENCE_TO_HANDLE_GET_SYNCOBJ	0
 #define AMDGPU_FENCE_TO_HANDLE_GET_SYNCOBJ_FD	1
 #define AMDGPU_FENCE_TO_HANDLE_GET_SYNC_FILE_FD	2
-- 
2.17.1

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

[PATCH libdrm 3/7] wrap syncobj timeline query/wait APIs for amdgpu v3

[Chunming Zhou]

v2: symbos are stored in lexical order.
v3: drop export/import and extra query indirection

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
---
 amdgpu/amdgpu-symbol-check |  2 ++
 amdgpu/amdgpu.h            | 39 ++++++++++++++++++++++++++++++++++++++
 amdgpu/amdgpu_cs.c         | 23 ++++++++++++++++++++++
 3 files changed, 64 insertions(+)

diff --git a/amdgpu/amdgpu-symbol-check b/amdgpu/amdgpu-symbol-check
index 4d806922..d3c5bb89 100755
--- a/amdgpu/amdgpu-symbol-check
+++ b/amdgpu/amdgpu-symbol-check
@@ -53,8 +53,10 @@ amdgpu_cs_submit_raw
 amdgpu_cs_submit_raw2
 amdgpu_cs_syncobj_export_sync_file
 amdgpu_cs_syncobj_import_sync_file
+amdgpu_cs_syncobj_query
 amdgpu_cs_syncobj_reset
 amdgpu_cs_syncobj_signal
+amdgpu_cs_syncobj_timeline_wait
 amdgpu_cs_syncobj_wait
 amdgpu_cs_wait_fences
 amdgpu_cs_wait_semaphore
diff --git a/amdgpu/amdgpu.h b/amdgpu/amdgpu.h
index c44a495a..5ebfe1e3 100644
--- a/amdgpu/amdgpu.h
+++ b/amdgpu/amdgpu.h
@@ -1536,6 +1536,45 @@ int amdgpu_cs_syncobj_wait(amdgpu_device_handle dev,
 			   int64_t timeout_nsec, unsigned flags,
 			   uint32_t *first_signaled);
 
+/**
+ *  Wait for one or all sync objects on their points to signal.
+ *
+ * \param   dev	    - \c [in] self-explanatory
+ * \param   handles - \c [in] array of sync object handles
+ * \param   points - \c [in] array of sync points to wait
+ * \param   num_handles - \c [in] self-explanatory
+ * \param   timeout_nsec - \c [in] self-explanatory
+ * \param   flags   - \c [in] a bitmask of DRM_SYNCOBJ_WAIT_FLAGS_*
+ * \param   first_signaled - \c [in] self-explanatory
+ *
+ * \return   0 on success\n
+ *          -ETIME - Timeout
+ *          <0 - Negative POSIX Error code
+ *
+ */
+int amdgpu_cs_syncobj_timeline_wait(amdgpu_device_handle dev,
+				    uint32_t *handles, uint64_t *points,
+				    unsigned num_handles,
+				    int64_t timeout_nsec, unsigned flags,
+				    uint32_t *first_signaled);
+/**
+ *  Query sync objects payloads.
+ *
+ * \param   dev	    - \c [in] self-explanatory
+ * \param   handles - \c [in] array of sync object handles
+ * \param   points - \c [out] array of sync points returned, which presents
+ * syncobj payload.
+ * \param   num_handles - \c [in] self-explanatory
+ *
+ * \return   0 on success\n
+ *          -ETIME - Timeout
+ *          <0 - Negative POSIX Error code
+ *
+ */
+int amdgpu_cs_syncobj_query(amdgpu_device_handle dev,
+			    uint32_t *handles, uint64_t *points,
+			    unsigned num_handles);
+
 /**
  *  Export kernel sync object to shareable fd.
  *
diff --git a/amdgpu/amdgpu_cs.c b/amdgpu/amdgpu_cs.c
index 7c5b9d13..9fcaf2c4 100644
--- a/amdgpu/amdgpu_cs.c
+++ b/amdgpu/amdgpu_cs.c
@@ -686,6 +686,29 @@ drm_public int amdgpu_cs_syncobj_wait(amdgpu_device_handle dev,
 			      flags, first_signaled);
 }
 
+drm_public int amdgpu_cs_syncobj_timeline_wait(amdgpu_device_handle dev,
+					       uint32_t *handles, uint64_t *points,
+					       unsigned num_handles,
+					       int64_t timeout_nsec, unsigned flags,
+					       uint32_t *first_signaled)
+{
+	if (NULL == dev)
+		return -EINVAL;
+
+	return drmSyncobjTimelineWait(dev->fd, handles, points, num_handles,
+				      timeout_nsec, flags, first_signaled);
+}
+
+drm_public int amdgpu_cs_syncobj_query(amdgpu_device_handle dev,
+				       uint32_t *handles, uint64_t *points,
+				       unsigned num_handles)
+{
+	if (NULL == dev)
+		return -EINVAL;
+
+	return drmSyncobjQuery(dev->fd, handles, points, num_handles);
+}
+
 drm_public int amdgpu_cs_export_syncobj(amdgpu_device_handle dev,
 					uint32_t handle,
 					int *shared_fd)
-- 
2.17.1

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

[PATCH libdrm 4/7] add timeline signal/transfer ioctls v2

[Chunming Zhou]

v2: use one transfer ioctl

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
---
 xf86drm.c | 33 +++++++++++++++++++++++++++++++++
 xf86drm.h |  6 ++++++
 2 files changed, 39 insertions(+)

diff --git a/xf86drm.c b/xf86drm.c
index 17e3d880..acd16fab 100644
--- a/xf86drm.c
+++ b/xf86drm.c
@@ -4257,6 +4257,21 @@ drm_public int drmSyncobjSignal(int fd, const uint32_t *handles,
     return ret;
 }
 
+drm_public int drmSyncobjTimelineSignal(int fd, const uint32_t *handles,
+					uint64_t *points, uint32_t handle_count)
+{
+    struct drm_syncobj_timeline_array args;
+    int ret;
+
+    memclear(args);
+    args.handles = (uintptr_t)handles;
+    args.points = (uint64_t)(uintptr_t)points;
+    args.count_handles = handle_count;
+
+    ret = drmIoctl(fd, DRM_IOCTL_SYNCOBJ_TIMELINE_SIGNAL, &args);
+    return ret;
+}
+
 drm_public int drmSyncobjTimelineWait(int fd, uint32_t *handles, uint64_t *points,
 				      unsigned num_handles,
 				      int64_t timeout_nsec, unsigned flags,
@@ -4299,4 +4314,22 @@ drm_public int drmSyncobjQuery(int fd, uint32_t *handles, uint64_t *points,
     return 0;
 }
 
+drm_public int drmSyncobjTransfer(int fd,
+				  uint32_t dst_handle, uint64_t dst_point,
+				  uint32_t src_handle, uint64_t src_point,
+				  uint32_t flags)
+{
+    struct drm_syncobj_transfer args;
+    int ret;
+
+    memclear(args);
+    args.src_handle = src_handle;
+    args.dst_handle = dst_handle;
+    args.src_point = src_point;
+    args.dst_point = dst_point;
+    args.flags = flags;
+
+    ret = drmIoctl(fd, DRM_IOCTL_SYNCOBJ_TRANSFER, &args);
 
+    return ret;
+}
diff --git a/xf86drm.h b/xf86drm.h
index 60c7a84f..3fb1d1ca 100644
--- a/xf86drm.h
+++ b/xf86drm.h
@@ -876,12 +876,18 @@ extern int drmSyncobjWait(int fd, uint32_t *handles, unsigned num_handles,
 			  uint32_t *first_signaled);
 extern int drmSyncobjReset(int fd, const uint32_t *handles, uint32_t handle_count);
 extern int drmSyncobjSignal(int fd, const uint32_t *handles, uint32_t handle_count);
+extern int drmSyncobjTimelineSignal(int fd, const uint32_t *handles,
+				    uint64_t *points, uint32_t handle_count);
 extern int drmSyncobjTimelineWait(int fd, uint32_t *handles, uint64_t *points,
 				  unsigned num_handles,
 				  int64_t timeout_nsec, unsigned flags,
 				  uint32_t *first_signaled);
 extern int drmSyncobjQuery(int fd, uint32_t *handles, uint64_t *points,
 			   uint32_t handle_count);
+extern int drmSyncobjTransfer(int fd,
+			      uint32_t dst_handle, uint64_t dst_point,
+			      uint32_t src_handle, uint64_t src_point,
+			      uint32_t flags);
 
 #if defined(__cplusplus)
 }
-- 
2.17.1

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

Re: [PATCH libdrm 4/7] add timeline signal/transfer ioctls v2

[Lionel Landwerlin]

On 13/05/2019 10:53, Chunming Zhou wrote:
> v2: use one transfer ioctl
>
> Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
> ---
>   xf86drm.c | 33 +++++++++++++++++++++++++++++++++
>   xf86drm.h |  6 ++++++
>   2 files changed, 39 insertions(+)
>
> diff --git a/xf86drm.c b/xf86drm.c
> index 17e3d880..acd16fab 100644
> --- a/xf86drm.c
> +++ b/xf86drm.c
> @@ -4257,6 +4257,21 @@ drm_public int drmSyncobjSignal(int fd, const uint32_t *handles,
>       return ret;
>   }
>   
> +drm_public int drmSyncobjTimelineSignal(int fd, const uint32_t *handles,
> +					uint64_t *points, uint32_t handle_count)
> +{
> +    struct drm_syncobj_timeline_array args;
> +    int ret;
> +
> +    memclear(args);
> +    args.handles = (uintptr_t)handles;
> +    args.points = (uint64_t)(uintptr_t)points;


I think you can drop the uint64_t cast.


> +    args.count_handles = handle_count;
> +
> +    ret = drmIoctl(fd, DRM_IOCTL_SYNCOBJ_TIMELINE_SIGNAL, &args);
> +    return ret;
> +}
> +
>   drm_public int drmSyncobjTimelineWait(int fd, uint32_t *handles, uint64_t *points,
>   				      unsigned num_handles,
>   				      int64_t timeout_nsec, unsigned flags,
> @@ -4299,4 +4314,22 @@ drm_public int drmSyncobjQuery(int fd, uint32_t *handles, uint64_t *points,
>       return 0;
>   }
>   
> +drm_public int drmSyncobjTransfer(int fd,
> +				  uint32_t dst_handle, uint64_t dst_point,
> +				  uint32_t src_handle, uint64_t src_point,
> +				  uint32_t flags)
> +{
> +    struct drm_syncobj_transfer args;
> +    int ret;
> +
> +    memclear(args);
> +    args.src_handle = src_handle;
> +    args.dst_handle = dst_handle;
> +    args.src_point = src_point;
> +    args.dst_point = dst_point;
> +    args.flags = flags;
> +
> +    ret = drmIoctl(fd, DRM_IOCTL_SYNCOBJ_TRANSFER, &args);
>   
> +    return ret;
> +}
> diff --git a/xf86drm.h b/xf86drm.h
> index 60c7a84f..3fb1d1ca 100644
> --- a/xf86drm.h
> +++ b/xf86drm.h
> @@ -876,12 +876,18 @@ extern int drmSyncobjWait(int fd, uint32_t *handles, unsigned num_handles,
>   			  uint32_t *first_signaled);
>   extern int drmSyncobjReset(int fd, const uint32_t *handles, uint32_t handle_count);
>   extern int drmSyncobjSignal(int fd, const uint32_t *handles, uint32_t handle_count);
> +extern int drmSyncobjTimelineSignal(int fd, const uint32_t *handles,
> +				    uint64_t *points, uint32_t handle_count);
>   extern int drmSyncobjTimelineWait(int fd, uint32_t *handles, uint64_t *points,
>   				  unsigned num_handles,
>   				  int64_t timeout_nsec, unsigned flags,
>   				  uint32_t *first_signaled);
>   extern int drmSyncobjQuery(int fd, uint32_t *handles, uint64_t *points,
>   			   uint32_t handle_count);
> +extern int drmSyncobjTransfer(int fd,
> +			      uint32_t dst_handle, uint64_t dst_point,
> +			      uint32_t src_handle, uint64_t src_point,
> +			      uint32_t flags);
>   
>   #if defined(__cplusplus)
>   }


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

[PATCH libdrm 5/7] expose timeline signal/export/import interfaces v2

[Chunming Zhou]

v2: adapt to new one transfer ioctl

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
---
 amdgpu/amdgpu-symbol-check |  3 ++
 amdgpu/amdgpu.h            | 51 ++++++++++++++++++++++++++++
 amdgpu/amdgpu_cs.c         | 68 ++++++++++++++++++++++++++++++++++++++
 3 files changed, 122 insertions(+)

diff --git a/amdgpu/amdgpu-symbol-check b/amdgpu/amdgpu-symbol-check
index d3c5bb89..274b4c6d 100755
--- a/amdgpu/amdgpu-symbol-check
+++ b/amdgpu/amdgpu-symbol-check
@@ -52,10 +52,13 @@ amdgpu_cs_submit
 amdgpu_cs_submit_raw
 amdgpu_cs_submit_raw2
 amdgpu_cs_syncobj_export_sync_file
+amdgpu_cs_syncobj_export_sync_file2
 amdgpu_cs_syncobj_import_sync_file
+amdgpu_cs_syncobj_import_sync_file2
 amdgpu_cs_syncobj_query
 amdgpu_cs_syncobj_reset
 amdgpu_cs_syncobj_signal
+amdgpu_cs_syncobj_timeline_signal
 amdgpu_cs_syncobj_timeline_wait
 amdgpu_cs_syncobj_wait
 amdgpu_cs_wait_fences
diff --git a/amdgpu/amdgpu.h b/amdgpu/amdgpu.h
index 5ebfe1e3..d2480dbe 100644
--- a/amdgpu/amdgpu.h
+++ b/amdgpu/amdgpu.h
@@ -1516,6 +1516,23 @@ int amdgpu_cs_syncobj_reset(amdgpu_device_handle dev,
 int amdgpu_cs_syncobj_signal(amdgpu_device_handle dev,
 			     const uint32_t *syncobjs, uint32_t syncobj_count);
 
+/**
+ * Signal kernel timeline sync objects.
+ *
+ * \param dev           - \c [in] device handle
+ * \param syncobjs      - \c [in] array of sync object handles
+ * \param points	- \c [in] array of timeline points
+ * \param syncobj_count - \c [in] number of handles in syncobjs
+ *
+ * \return   0 on success\n
+ *          <0 - Negative POSIX Error code
+ *
+*/
+int amdgpu_cs_syncobj_timeline_signal(amdgpu_device_handle dev,
+				      const uint32_t *syncobjs,
+				      uint64_t *points,
+				      uint32_t syncobj_count);
+
 /**
  *  Wait for one or all sync objects to signal.
  *
@@ -1633,7 +1650,41 @@ int amdgpu_cs_syncobj_export_sync_file(amdgpu_device_handle dev,
 int amdgpu_cs_syncobj_import_sync_file(amdgpu_device_handle dev,
 				       uint32_t syncobj,
 				       int sync_file_fd);
+/**
+ *  Export kernel timeline sync object to a sync_file.
+ *
+ * \param   dev		- \c [in] device handle
+ * \param   syncobj	- \c [in] sync object handle
+ * \param   point	- \c [in] timeline point
+ * \param   flags	- \c [in] flags
+ * \param   sync_file_fd - \c [out] sync_file file descriptor.
+ *
+ * \return   0 on success\n
+ *          <0 - Negative POSIX Error code
+ *
+ */
+int amdgpu_cs_syncobj_export_sync_file2(amdgpu_device_handle dev,
+					uint32_t syncobj,
+					uint64_t point,
+					uint32_t flags,
+					int *sync_file_fd);
 
+/**
+ *  Import kernel timeline sync object from a sync_file.
+ *
+ * \param   dev		- \c [in] device handle
+ * \param   syncobj	- \c [in] sync object handle
+ * \param   point	- \c [in] timeline point
+ * \param   sync_file_fd - \c [in] sync_file file descriptor.
+ *
+ * \return   0 on success\n
+ *          <0 - Negative POSIX Error code
+ *
+ */
+int amdgpu_cs_syncobj_import_sync_file2(amdgpu_device_handle dev,
+					uint32_t syncobj,
+					uint64_t point,
+					int sync_file_fd);
 /**
  * Export an amdgpu fence as a handle (syncobj or fd).
  *
diff --git a/amdgpu/amdgpu_cs.c b/amdgpu/amdgpu_cs.c
index 9fcaf2c4..daca4421 100644
--- a/amdgpu/amdgpu_cs.c
+++ b/amdgpu/amdgpu_cs.c
@@ -674,6 +674,18 @@ drm_public int amdgpu_cs_syncobj_signal(amdgpu_device_handle dev,
 	return drmSyncobjSignal(dev->fd, syncobjs, syncobj_count);
 }
 
+drm_public int amdgpu_cs_syncobj_timeline_signal(amdgpu_device_handle dev,
+						 const uint32_t *syncobjs,
+						 uint64_t *points,
+						 uint32_t syncobj_count)
+{
+	if (NULL == dev)
+		return -EINVAL;
+
+	return drmSyncobjTimelineSignal(dev->fd, syncobjs,
+					points, syncobj_count);
+}
+
 drm_public int amdgpu_cs_syncobj_wait(amdgpu_device_handle dev,
 				      uint32_t *handles, unsigned num_handles,
 				      int64_t timeout_nsec, unsigned flags,
@@ -749,6 +761,62 @@ drm_public int amdgpu_cs_syncobj_import_sync_file(amdgpu_device_handle dev,
 	return drmSyncobjImportSyncFile(dev->fd, syncobj, sync_file_fd);
 }
 
+drm_public int amdgpu_cs_syncobj_export_sync_file2(amdgpu_device_handle dev,
+						   uint32_t syncobj,
+						   uint64_t point,
+						   uint32_t flags,
+						   int *sync_file_fd)
+{
+	uint32_t binary_handle;
+	int ret;
+
+	if (NULL == dev)
+		return -EINVAL;
+
+	if (!point)
+		return drmSyncobjExportSyncFile(dev->fd, syncobj, sync_file_fd);
+
+	ret = drmSyncobjCreate(dev->fd, 0, &binary_handle);
+	if (ret)
+		return ret;
+
+	ret = drmSyncobjTransfer(dev->fd, binary_handle, 0,
+				 syncobj, point, flags);
+	if (ret)
+		goto out;
+	ret = drmSyncobjExportSyncFile(dev->fd, binary_handle, sync_file_fd);
+out:
+	drmSyncobjDestroy(dev->fd, binary_handle);
+	return ret;
+}
+
+drm_public int amdgpu_cs_syncobj_import_sync_file2(amdgpu_device_handle dev,
+						   uint32_t syncobj,
+						   uint64_t point,
+						   int sync_file_fd)
+{
+	uint32_t binary_handle;
+	int ret;
+
+	if (NULL == dev)
+		return -EINVAL;
+
+	if (!point)
+		return drmSyncobjImportSyncFile(dev->fd, syncobj, sync_file_fd);
+
+	ret = drmSyncobjCreate(dev->fd, 0, &binary_handle);
+	if (ret)
+		return ret;
+	ret = drmSyncobjImportSyncFile(dev->fd, binary_handle, sync_file_fd);
+	if (ret)
+		goto out;
+	ret = drmSyncobjTransfer(dev->fd, syncobj, point,
+				 binary_handle, 0, 0);
+out:
+	drmSyncobjDestroy(dev->fd, binary_handle);
+	return ret;
+}
+
 drm_public int amdgpu_cs_submit_raw(amdgpu_device_handle dev,
 				    amdgpu_context_handle context,
 				    amdgpu_bo_list_handle bo_list_handle,
-- 
2.17.1

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

[PATCH libdrm 6/7] wrap transfer interfaces

[Chunming Zhou]

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
---
 amdgpu/amdgpu.h    | 22 ++++++++++++++++++++++
 amdgpu/amdgpu_cs.c | 16 ++++++++++++++++
 2 files changed, 38 insertions(+)

diff --git a/amdgpu/amdgpu.h b/amdgpu/amdgpu.h
index d2480dbe..9d9b0832 100644
--- a/amdgpu/amdgpu.h
+++ b/amdgpu/amdgpu.h
@@ -1685,6 +1685,28 @@ int amdgpu_cs_syncobj_import_sync_file2(amdgpu_device_handle dev,
 					uint32_t syncobj,
 					uint64_t point,
 					int sync_file_fd);
+
+/**
+ *  transfer between syncbojs.
+ *
+ * \param   dev		- \c [in] device handle
+ * \param   dst_handle	- \c [in] sync object handle
+ * \param   dst_point	- \c [in] timeline point, 0 presents dst is binary
+ * \param   src_handle	- \c [in] sync object handle
+ * \param   src_point	- \c [in] timeline point, 0 presents src is binary
+ * \param   flags	- \c [in] flags
+ *
+ * \return   0 on success\n
+ *          <0 - Negative POSIX Error code
+ *
+ */
+int amdgpu_cs_syncobj_transfer(amdgpu_device_handle dev,
+			       uint32_t dst_handle,
+			       uint64_t dst_point,
+			       uint32_t src_handle,
+			       uint64_t src_point,
+			       uint32_t flags);
+
 /**
  * Export an amdgpu fence as a handle (syncobj or fd).
  *
diff --git a/amdgpu/amdgpu_cs.c b/amdgpu/amdgpu_cs.c
index daca4421..977fa3cf 100644
--- a/amdgpu/amdgpu_cs.c
+++ b/amdgpu/amdgpu_cs.c
@@ -817,6 +817,22 @@ out:
 	return ret;
 }
 
+drm_public int amdgpu_cs_syncobj_transfer(amdgpu_device_handle dev,
+					  uint32_t dst_handle,
+					  uint64_t dst_point,
+					  uint32_t src_handle,
+					  uint64_t src_point,
+					  uint32_t flags)
+{
+	if (NULL == dev)
+		return -EINVAL;
+
+	return drmSyncobjTransfer(dev->fd,
+				  dst_handle, dst_point,
+				  src_handle, src_point,
+				  flags);
+}
+
 drm_public int amdgpu_cs_submit_raw(amdgpu_device_handle dev,
 				    amdgpu_context_handle context,
 				    amdgpu_bo_list_handle bo_list_handle,
-- 
2.17.1

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

[PATCH libdrm 7/7] add syncobj timeline tests v3

[Chunming Zhou]

v2: drop DRM_SYNCOBJ_CREATE_TYPE_TIMELINE, fix timeout calculation,
    fix some warnings
v3: add export/import and cpu signal testing cases

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
---
 tests/amdgpu/Makefile.am     |   3 +-
 tests/amdgpu/amdgpu_test.c   |  11 ++
 tests/amdgpu/amdgpu_test.h   |  21 +++
 tests/amdgpu/meson.build     |   2 +-
 tests/amdgpu/syncobj_tests.c | 290 +++++++++++++++++++++++++++++++++++
 5 files changed, 325 insertions(+), 2 deletions(-)
 create mode 100644 tests/amdgpu/syncobj_tests.c

diff --git a/tests/amdgpu/Makefile.am b/tests/amdgpu/Makefile.am
index 48278848..920882d0 100644
--- a/tests/amdgpu/Makefile.am
+++ b/tests/amdgpu/Makefile.am
@@ -34,4 +34,5 @@ amdgpu_test_SOURCES = \
 	uve_ib.h \
 	deadlock_tests.c \
 	vm_tests.c	\
-	ras_tests.c
+	ras_tests.c \
+	syncobj_tests.c
diff --git a/tests/amdgpu/amdgpu_test.c b/tests/amdgpu/amdgpu_test.c
index 35c8bf6c..73403fb4 100644
--- a/tests/amdgpu/amdgpu_test.c
+++ b/tests/amdgpu/amdgpu_test.c
@@ -57,6 +57,7 @@
 #define DEADLOCK_TESTS_STR "Deadlock Tests"
 #define VM_TESTS_STR "VM Tests"
 #define RAS_TESTS_STR "RAS Tests"
+#define SYNCOBJ_TIMELINE_TESTS_STR "SYNCOBJ TIMELINE Tests"
 
 /**
  *  Open handles for amdgpu devices
@@ -123,6 +124,12 @@ static CU_SuiteInfo suites[] = {
 		.pCleanupFunc = suite_ras_tests_clean,
 		.pTests = ras_tests,
 	},
+	{
+		.pName = SYNCOBJ_TIMELINE_TESTS_STR,
+		.pInitFunc = suite_syncobj_timeline_tests_init,
+		.pCleanupFunc = suite_syncobj_timeline_tests_clean,
+		.pTests = syncobj_timeline_tests,
+	},
 
 	CU_SUITE_INFO_NULL,
 };
@@ -176,6 +183,10 @@ static Suites_Active_Status suites_active_stat[] = {
 			.pName = RAS_TESTS_STR,
 			.pActive = suite_ras_tests_enable,
 		},
+		{
+			.pName = SYNCOBJ_TIMELINE_TESTS_STR,
+			.pActive = suite_syncobj_timeline_tests_enable,
+		},
 };
 
 
diff --git a/tests/amdgpu/amdgpu_test.h b/tests/amdgpu/amdgpu_test.h
index bcd0bc7e..36675ea3 100644
--- a/tests/amdgpu/amdgpu_test.h
+++ b/tests/amdgpu/amdgpu_test.h
@@ -216,6 +216,27 @@ CU_BOOL suite_ras_tests_enable(void);
 extern CU_TestInfo ras_tests[];
 
 
+/**
+ * Initialize syncobj timeline test suite
+ */
+int suite_syncobj_timeline_tests_init();
+
+/**
+ * Deinitialize syncobj timeline test suite
+ */
+int suite_syncobj_timeline_tests_clean();
+
+/**
+ * Decide if the suite is enabled by default or not.
+ */
+CU_BOOL suite_syncobj_timeline_tests_enable(void);
+
+/**
+ * Tests in syncobj timeline test suite
+ */
+extern CU_TestInfo syncobj_timeline_tests[];
+
+
 /**
  * Helper functions
  */
diff --git a/tests/amdgpu/meson.build b/tests/amdgpu/meson.build
index 95ed9305..1726cb43 100644
--- a/tests/amdgpu/meson.build
+++ b/tests/amdgpu/meson.build
@@ -24,7 +24,7 @@ if dep_cunit.found()
     files(
       'amdgpu_test.c', 'basic_tests.c', 'bo_tests.c', 'cs_tests.c',
       'vce_tests.c', 'uvd_enc_tests.c', 'vcn_tests.c', 'deadlock_tests.c',
-      'vm_tests.c', 'ras_tests.c',
+      'vm_tests.c', 'ras_tests.c', 'syncobj_tests.c',
     ),
     dependencies : [dep_cunit, dep_threads],
     include_directories : [inc_root, inc_drm, include_directories('../../amdgpu')],
diff --git a/tests/amdgpu/syncobj_tests.c b/tests/amdgpu/syncobj_tests.c
new file mode 100644
index 00000000..a0c627d7
--- /dev/null
+++ b/tests/amdgpu/syncobj_tests.c
@@ -0,0 +1,290 @@
+/*
+ * Copyright 2017 Advanced Micro Devices, Inc.
+ *
+ * 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 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 COPYRIGHT HOLDER(S) OR AUTHOR(S) 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.
+ *
+*/
+
+#include "CUnit/Basic.h"
+
+#include "amdgpu_test.h"
+#include "amdgpu_drm.h"
+#include "amdgpu_internal.h"
+#include <pthread.h>
+
+static  amdgpu_device_handle device_handle;
+static  uint32_t  major_version;
+static  uint32_t  minor_version;
+
+static void amdgpu_syncobj_timeline_test(void);
+
+CU_BOOL suite_syncobj_timeline_tests_enable(void)
+{
+	return CU_TRUE;
+}
+
+int suite_syncobj_timeline_tests_init(void)
+{
+	int r;
+
+	r = amdgpu_device_initialize(drm_amdgpu[0], &major_version,
+				   &minor_version, &device_handle);
+
+	if (r) {
+		if ((r == -EACCES) && (errno == EACCES))
+			printf("\n\nError:%s. "
+				"Hint:Try to run this test program as root.",
+				strerror(errno));
+		return CUE_SINIT_FAILED;
+	}
+
+	return CUE_SUCCESS;
+}
+
+int suite_syncobj_timeline_tests_clean(void)
+{
+	int r = amdgpu_device_deinitialize(device_handle);
+
+	if (r == 0)
+		return CUE_SUCCESS;
+	else
+		return CUE_SCLEAN_FAILED;
+}
+
+
+CU_TestInfo syncobj_timeline_tests[] = {
+	{ "syncobj timeline test",  amdgpu_syncobj_timeline_test },
+	CU_TEST_INFO_NULL,
+};
+
+#define GFX_COMPUTE_NOP  0xffff1000
+#define SDMA_NOP  0x0
+static int syncobj_command_submission_helper(uint32_t syncobj_handle, bool
+					     wait_or_signal, uint64_t point)
+{
+	amdgpu_context_handle context_handle;
+	amdgpu_bo_handle ib_result_handle;
+	void *ib_result_cpu;
+	uint64_t ib_result_mc_address;
+	struct drm_amdgpu_cs_chunk chunks[2];
+	struct drm_amdgpu_cs_chunk_data chunk_data;
+	struct drm_amdgpu_cs_chunk_syncobj syncobj_data;
+	struct amdgpu_cs_fence fence_status;
+	amdgpu_bo_list_handle bo_list;
+	amdgpu_va_handle va_handle;
+	uint32_t expired, flags;
+	int i, r;
+	uint64_t seq_no;
+	static uint32_t *ptr;
+
+	r = amdgpu_cs_ctx_create(device_handle, &context_handle);
+	CU_ASSERT_EQUAL(r, 0);
+
+	r = amdgpu_bo_alloc_and_map(device_handle, 4096, 4096,
+				    AMDGPU_GEM_DOMAIN_GTT, 0,
+				    &ib_result_handle, &ib_result_cpu,
+				    &ib_result_mc_address, &va_handle);
+	CU_ASSERT_EQUAL(r, 0);
+
+	r = amdgpu_get_bo_list(device_handle, ib_result_handle, NULL,
+			       &bo_list);
+	CU_ASSERT_EQUAL(r, 0);
+
+	ptr = ib_result_cpu;
+
+	for (i = 0; i < 16; ++i)
+		ptr[i] = wait_or_signal ? GFX_COMPUTE_NOP: SDMA_NOP;
+
+	chunks[0].chunk_id = AMDGPU_CHUNK_ID_IB;
+	chunks[0].length_dw = sizeof(struct drm_amdgpu_cs_chunk_ib) / 4;
+	chunks[0].chunk_data = (uint64_t)(uintptr_t)&chunk_data;
+	chunk_data.ib_data._pad = 0;
+	chunk_data.ib_data.va_start = ib_result_mc_address;
+	chunk_data.ib_data.ib_bytes = 16 * 4;
+	chunk_data.ib_data.ip_type = wait_or_signal ? AMDGPU_HW_IP_GFX :
+		AMDGPU_HW_IP_DMA;
+	chunk_data.ib_data.ip_instance = 0;
+	chunk_data.ib_data.ring = 0;
+	chunk_data.ib_data.flags = 0;
+
+	chunks[1].chunk_id = wait_or_signal ?
+		AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_WAIT :
+		AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_SIGNAL;
+	chunks[1].length_dw = sizeof(struct drm_amdgpu_cs_chunk_syncobj) / 4;
+	chunks[1].chunk_data = (uint64_t)(uintptr_t)&syncobj_data;
+	syncobj_data.handle = syncobj_handle;
+	syncobj_data.point = point;
+	syncobj_data.flags = DRM_SYNCOBJ_WAIT_FLAGS_WAIT_FOR_SUBMIT;
+
+	r = amdgpu_cs_submit_raw(device_handle,
+				 context_handle,
+				 bo_list,
+				 2,
+				 chunks,
+				 &seq_no);
+	CU_ASSERT_EQUAL(r, 0);
+
+
+	memset(&fence_status, 0, sizeof(struct amdgpu_cs_fence));
+	fence_status.context = context_handle;
+	fence_status.ip_type = wait_or_signal ? AMDGPU_HW_IP_GFX:
+		AMDGPU_HW_IP_DMA;
+	fence_status.ip_instance = 0;
+	fence_status.ring = 0;
+	fence_status.fence = seq_no;
+
+	r = amdgpu_cs_query_fence_status(&fence_status,
+			AMDGPU_TIMEOUT_INFINITE,0, &expired);
+	CU_ASSERT_EQUAL(r, 0);
+
+	r = amdgpu_bo_list_destroy(bo_list);
+	CU_ASSERT_EQUAL(r, 0);
+
+	r = amdgpu_bo_unmap_and_free(ib_result_handle, va_handle,
+				     ib_result_mc_address, 4096);
+	CU_ASSERT_EQUAL(r, 0);
+
+	r = amdgpu_cs_ctx_free(context_handle);
+	CU_ASSERT_EQUAL(r, 0);
+
+	return r;
+}
+
+struct syncobj_point {
+	uint32_t syncobj_handle;
+	uint64_t point;
+};
+
+static void *syncobj_wait(void *data)
+{
+	struct syncobj_point *sp = (struct syncobj_point *)data;
+	int r;
+
+	r = syncobj_command_submission_helper(sp->syncobj_handle, true,
+					      sp->point);
+	CU_ASSERT_EQUAL(r, 0);
+
+	return (void *)(long)r;
+}
+
+static void *syncobj_signal(void *data)
+{
+	struct syncobj_point *sp = (struct syncobj_point *)data;
+	int r;
+
+	r = syncobj_command_submission_helper(sp->syncobj_handle, false,
+					      sp->point);
+	CU_ASSERT_EQUAL(r, 0);
+
+	return (void *)(long)r;
+}
+
+static void amdgpu_syncobj_timeline_test(void)
+{
+	static pthread_t wait_thread;
+	static pthread_t signal_thread;
+	static pthread_t c_thread;
+	struct syncobj_point sp1, sp2, sp3;
+	uint32_t syncobj_handle;
+	uint64_t payload;
+	uint64_t wait_point, signal_point;
+	uint64_t timeout;
+	struct timespec tp;
+	int r, sync_fd;
+	void *tmp;
+
+	r =  amdgpu_cs_create_syncobj2(device_handle, 0, &syncobj_handle);
+	CU_ASSERT_EQUAL(r, 0);
+
+	// wait on point 5
+	sp1.syncobj_handle = syncobj_handle;
+	sp1.point = 5;
+	r = pthread_create(&wait_thread, NULL, syncobj_wait, &sp1);
+	CU_ASSERT_EQUAL(r, 0);
+
+	// signal on point 10
+	sp2.syncobj_handle = syncobj_handle;
+	sp2.point = 10;
+	r = pthread_create(&signal_thread, NULL, syncobj_signal, &sp2);
+	CU_ASSERT_EQUAL(r, 0);
+
+	r = pthread_join(wait_thread, &tmp);
+	CU_ASSERT_EQUAL(r, 0);
+	CU_ASSERT_EQUAL(tmp, 0);
+
+	r = pthread_join(signal_thread, &tmp);
+	CU_ASSERT_EQUAL(r, 0);
+	CU_ASSERT_EQUAL(tmp, 0);
+
+	//query timeline payload
+	r = amdgpu_cs_syncobj_query(device_handle, &syncobj_handle,
+				    &payload, 1);
+	CU_ASSERT_EQUAL(r, 0);
+	CU_ASSERT_EQUAL(payload, 10);
+
+	//signal on point 16
+	sp3.syncobj_handle = syncobj_handle;
+	sp3.point = 16;
+	r = pthread_create(&c_thread, NULL, syncobj_signal, &sp3);
+	CU_ASSERT_EQUAL(r, 0);
+	//CPU wait on point 16
+	wait_point = 16;
+	timeout = 0;
+	clock_gettime(CLOCK_MONOTONIC, &tp);
+	timeout = tp.tv_sec * 1000000000ULL + tp.tv_nsec;
+	timeout += 0x10000000000; //10s
+	r = amdgpu_cs_syncobj_timeline_wait(device_handle, &syncobj_handle,
+					    &wait_point, 1, timeout,
+					    DRM_SYNCOBJ_WAIT_FLAGS_WAIT_ALL |
+					    DRM_SYNCOBJ_WAIT_FLAGS_WAIT_FOR_SUBMIT,
+					    NULL);
+
+	CU_ASSERT_EQUAL(r, 0);
+	r = pthread_join(c_thread, &tmp);
+	CU_ASSERT_EQUAL(r, 0);
+	CU_ASSERT_EQUAL(tmp, 0);
+
+	// export point 16 and import to point 18
+	r = amdgpu_cs_syncobj_export_sync_file2(device_handle, syncobj_handle,
+						16,
+						DRM_SYNCOBJ_WAIT_FLAGS_WAIT_FOR_SUBMIT,
+						&sync_fd);
+	CU_ASSERT_EQUAL(r, 0);
+	r = amdgpu_cs_syncobj_import_sync_file2(device_handle, syncobj_handle,
+						18, sync_fd);
+	CU_ASSERT_EQUAL(r, 0);
+	r = amdgpu_cs_syncobj_query(device_handle, &syncobj_handle,
+				    &payload, 1);
+	CU_ASSERT_EQUAL(r, 0);
+	CU_ASSERT_EQUAL(payload, 18);
+
+	// CPU signal on point 20
+	signal_point = 20;
+	r = amdgpu_cs_syncobj_timeline_signal(device_handle, &syncobj_handle,
+					      &signal_point, 1);
+	CU_ASSERT_EQUAL(r, 0);
+	r = amdgpu_cs_syncobj_query(device_handle, &syncobj_handle,
+				    &payload, 1);
+	CU_ASSERT_EQUAL(r, 0);
+	CU_ASSERT_EQUAL(payload, 20);
+
+	r = amdgpu_cs_destroy_syncobj(device_handle, syncobj_handle);
+	CU_ASSERT_EQUAL(r, 0);
+
+}
-- 
2.17.1

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

[PATCH libdrm 2/7] add timeline wait/query ioctl v2

[Chunming Zhou]

v2: drop export/import

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
---
 xf86drm.c | 44 ++++++++++++++++++++++++++++++++++++++++++++
 xf86drm.h |  6 ++++++
 2 files changed, 50 insertions(+)

diff --git a/xf86drm.c b/xf86drm.c
index 2c19376b..17e3d880 100644
--- a/xf86drm.c
+++ b/xf86drm.c
@@ -4256,3 +4256,47 @@ drm_public int drmSyncobjSignal(int fd, const uint32_t *handles,
     ret = drmIoctl(fd, DRM_IOCTL_SYNCOBJ_SIGNAL, &args);
     return ret;
 }
+
+drm_public int drmSyncobjTimelineWait(int fd, uint32_t *handles, uint64_t *points,
+				      unsigned num_handles,
+				      int64_t timeout_nsec, unsigned flags,
+				      uint32_t *first_signaled)
+{
+    struct drm_syncobj_timeline_wait args;
+    int ret;
+
+    memclear(args);
+    args.handles = (uintptr_t)handles;
+    args.points = (uint64_t)(uintptr_t)points;
+    args.timeout_nsec = timeout_nsec;
+    args.count_handles = num_handles;
+    args.flags = flags;
+
+    ret = drmIoctl(fd, DRM_IOCTL_SYNCOBJ_TIMELINE_WAIT, &args);
+    if (ret < 0)
+        return -errno;
+
+    if (first_signaled)
+        *first_signaled = args.first_signaled;
+    return ret;
+}
+
+
+drm_public int drmSyncobjQuery(int fd, uint32_t *handles, uint64_t *points,
+			       uint32_t handle_count)
+{
+    struct drm_syncobj_timeline_array args;
+    int ret;
+
+    memclear(args);
+    args.handles = (uintptr_t)handles;
+    args.points = (uint64_t)(uintptr_t)points;
+    args.count_handles = handle_count;
+
+    ret = drmIoctl(fd, DRM_IOCTL_SYNCOBJ_QUERY, &args);
+    if (ret)
+        return ret;
+    return 0;
+}
+
+
diff --git a/xf86drm.h b/xf86drm.h
index 887ecc76..60c7a84f 100644
--- a/xf86drm.h
+++ b/xf86drm.h
@@ -876,6 +876,12 @@ extern int drmSyncobjWait(int fd, uint32_t *handles, unsigned num_handles,
 			  uint32_t *first_signaled);
 extern int drmSyncobjReset(int fd, const uint32_t *handles, uint32_t handle_count);
 extern int drmSyncobjSignal(int fd, const uint32_t *handles, uint32_t handle_count);
+extern int drmSyncobjTimelineWait(int fd, uint32_t *handles, uint64_t *points,
+				  unsigned num_handles,
+				  int64_t timeout_nsec, unsigned flags,
+				  uint32_t *first_signaled);
+extern int drmSyncobjQuery(int fd, uint32_t *handles, uint64_t *points,
+			   uint32_t handle_count);
 
 #if defined(__cplusplus)
 }
-- 
2.17.1

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

Re: [PATCH libdrm 2/7] add timeline wait/query ioctl v2

[Lionel Landwerlin]

On 13/05/2019 10:53, Chunming Zhou wrote:
> v2: drop export/import
>
> Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
> ---
>   xf86drm.c | 44 ++++++++++++++++++++++++++++++++++++++++++++
>   xf86drm.h |  6 ++++++
>   2 files changed, 50 insertions(+)
>
> diff --git a/xf86drm.c b/xf86drm.c
> index 2c19376b..17e3d880 100644
> --- a/xf86drm.c
> +++ b/xf86drm.c
> @@ -4256,3 +4256,47 @@ drm_public int drmSyncobjSignal(int fd, const uint32_t *handles,
>       ret = drmIoctl(fd, DRM_IOCTL_SYNCOBJ_SIGNAL, &args);
>       return ret;
>   }
> +
> +drm_public int drmSyncobjTimelineWait(int fd, uint32_t *handles, uint64_t *points,
> +				      unsigned num_handles,
> +				      int64_t timeout_nsec, unsigned flags,
> +				      uint32_t *first_signaled)
> +{
> +    struct drm_syncobj_timeline_wait args;
> +    int ret;
> +
> +    memclear(args);
> +    args.handles = (uintptr_t)handles;
> +    args.points = (uint64_t)(uintptr_t)points;


I don't think you need those uintptr_t -> uint64_t casts.


> +    args.timeout_nsec = timeout_nsec;
> +    args.count_handles = num_handles;
> +    args.flags = flags;
> +
> +    ret = drmIoctl(fd, DRM_IOCTL_SYNCOBJ_TIMELINE_WAIT, &args);
> +    if (ret < 0)
> +        return -errno;
> +
> +    if (first_signaled)
> +        *first_signaled = args.first_signaled;
> +    return ret;
> +}
> +
> +
> +drm_public int drmSyncobjQuery(int fd, uint32_t *handles, uint64_t *points,
> +			       uint32_t handle_count)
> +{
> +    struct drm_syncobj_timeline_array args;
> +    int ret;
> +
> +    memclear(args);
> +    args.handles = (uintptr_t)handles;
> +    args.points = (uint64_t)(uintptr_t)points;


Same here.


> +    args.count_handles = handle_count;
> +
> +    ret = drmIoctl(fd, DRM_IOCTL_SYNCOBJ_QUERY, &args);
> +    if (ret)
> +        return ret;
> +    return 0;
> +}
> +
> +
> diff --git a/xf86drm.h b/xf86drm.h
> index 887ecc76..60c7a84f 100644
> --- a/xf86drm.h
> +++ b/xf86drm.h
> @@ -876,6 +876,12 @@ extern int drmSyncobjWait(int fd, uint32_t *handles, unsigned num_handles,
>   			  uint32_t *first_signaled);
>   extern int drmSyncobjReset(int fd, const uint32_t *handles, uint32_t handle_count);
>   extern int drmSyncobjSignal(int fd, const uint32_t *handles, uint32_t handle_count);
> +extern int drmSyncobjTimelineWait(int fd, uint32_t *handles, uint64_t *points,
> +				  unsigned num_handles,
> +				  int64_t timeout_nsec, unsigned flags,
> +				  uint32_t *first_signaled);
> +extern int drmSyncobjQuery(int fd, uint32_t *handles, uint64_t *points,
> +			   uint32_t handle_count);
>   
>   #if defined(__cplusplus)
>   }


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

Re: [PATCH libdrm 1/7] addr cs chunk for syncobj timeline

[zhoucm1]

ping... for patch set.


On 2019年05月13日 17:52, Chunming Zhou wrote:
> [CAUTION: External Email]
>
> Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
> ---
>   include/drm/amdgpu_drm.h | 9 +++++++++
>   1 file changed, 9 insertions(+)
>
> diff --git a/include/drm/amdgpu_drm.h b/include/drm/amdgpu_drm.h
> index d0701ffc..3d0318e6 100644
> --- a/include/drm/amdgpu_drm.h
> +++ b/include/drm/amdgpu_drm.h
> @@ -528,6 +528,8 @@ struct drm_amdgpu_gem_va {
>   #define AMDGPU_CHUNK_ID_SYNCOBJ_OUT     0x05
>   #define AMDGPU_CHUNK_ID_BO_HANDLES      0x06
>   #define AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES 0x07
> +#define AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_WAIT    0x08
> +#define AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_SIGNAL  0x09
>
>   struct drm_amdgpu_cs_chunk {
>          __u32           chunk_id;
> @@ -608,6 +610,13 @@ struct drm_amdgpu_cs_chunk_sem {
>          __u32 handle;
>   };
>
> +struct drm_amdgpu_cs_chunk_syncobj {
> +       __u32 handle;
> +       __u32 flags;
> +       __u64 point;
> +};
> +
> +
>   #define AMDGPU_FENCE_TO_HANDLE_GET_SYNCOBJ     0
>   #define AMDGPU_FENCE_TO_HANDLE_GET_SYNCOBJ_FD  1
>   #define AMDGPU_FENCE_TO_HANDLE_GET_SYNC_FILE_FD        2
> --
> 2.17.1
>
> _______________________________________________
> amd-gfx mailing list
> amd-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/amd-gfx

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

Re: [PATCH libdrm 1/7] addr cs chunk for syncobj timeline

[Lionel Landwerlin]

Sorry for the delay, I'll try to review this tomorrow.

-Lionel

On 13/05/2019 11:15, zhoucm1 wrote:
> ping... for patch set.
>
>
> On 2019年05月13日 17:52, Chunming Zhou wrote:
>> [CAUTION: External Email]
>>
>> Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
>> ---
>>   include/drm/amdgpu_drm.h | 9 +++++++++
>>   1 file changed, 9 insertions(+)
>>
>> diff --git a/include/drm/amdgpu_drm.h b/include/drm/amdgpu_drm.h
>> index d0701ffc..3d0318e6 100644
>> --- a/include/drm/amdgpu_drm.h
>> +++ b/include/drm/amdgpu_drm.h
>> @@ -528,6 +528,8 @@ struct drm_amdgpu_gem_va {
>>   #define AMDGPU_CHUNK_ID_SYNCOBJ_OUT     0x05
>>   #define AMDGPU_CHUNK_ID_BO_HANDLES      0x06
>>   #define AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES 0x07
>> +#define AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_WAIT    0x08
>> +#define AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_SIGNAL  0x09
>>
>>   struct drm_amdgpu_cs_chunk {
>>          __u32           chunk_id;
>> @@ -608,6 +610,13 @@ struct drm_amdgpu_cs_chunk_sem {
>>          __u32 handle;
>>   };
>>
>> +struct drm_amdgpu_cs_chunk_syncobj {
>> +       __u32 handle;
>> +       __u32 flags;
>> +       __u64 point;
>> +};
>> +
>> +
>>   #define AMDGPU_FENCE_TO_HANDLE_GET_SYNCOBJ     0
>>   #define AMDGPU_FENCE_TO_HANDLE_GET_SYNCOBJ_FD  1
>>   #define AMDGPU_FENCE_TO_HANDLE_GET_SYNC_FILE_FD        2
>> -- 
>> 2.17.1
>>
>> _______________________________________________
>> amd-gfx mailing list
>> amd-gfx@lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
>
>

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

Re: [PATCH libdrm 1/7] addr cs chunk for syncobj timeline

[Lionel Landwerlin]

With the small nits, patches 2 & 4 are : Reviewed-by: Lionel Landwerlin 
<lionel.g.landwerlin@intel.com>
The other patches are a bit amdgpu specific so maybe you might want 
someone more familiar with amdgpu to review them.
Still I didn't see anything wrong with them so remaining patches are : 
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

I'll send the IGT stuff shortly.

Thanks,

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

Re: [PATCH libdrm 1/7] addr cs chunk for syncobj timeline

[zhoucm1]

Thank you, Lionel.

-David


On 2019年05月14日 17:49, Lionel Landwerlin wrote:
> [CAUTION: External Email]
>
> With the small nits, patches 2 & 4 are : Reviewed-by: Lionel Landwerlin
> <lionel.g.landwerlin@intel.com>
> The other patches are a bit amdgpu specific so maybe you might want
> someone more familiar with amdgpu to review them.
> Still I didn't see anything wrong with them so remaining patches are :
> Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
>
> I'll send the IGT stuff shortly.
>
> Thanks,
>
> -Lionel

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

[PATCH libdrm 1/7] add cs chunk for syncobj timeline

[Chunming Zhou]

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
---
 include/drm/amdgpu_drm.h | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/include/drm/amdgpu_drm.h b/include/drm/amdgpu_drm.h
index d0701ffc..3d0318e6 100644
--- a/include/drm/amdgpu_drm.h
+++ b/include/drm/amdgpu_drm.h
@@ -528,6 +528,8 @@ struct drm_amdgpu_gem_va {
 #define AMDGPU_CHUNK_ID_SYNCOBJ_OUT     0x05
 #define AMDGPU_CHUNK_ID_BO_HANDLES      0x06
 #define AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES	0x07
+#define AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_WAIT    0x08
+#define AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_SIGNAL  0x09
 
 struct drm_amdgpu_cs_chunk {
 	__u32		chunk_id;
@@ -608,6 +610,13 @@ struct drm_amdgpu_cs_chunk_sem {
 	__u32 handle;
 };
 
+struct drm_amdgpu_cs_chunk_syncobj {
+	__u32 handle;
+	__u32 flags;
+	__u64 point;
+};
+
+
 #define AMDGPU_FENCE_TO_HANDLE_GET_SYNCOBJ	0
 #define AMDGPU_FENCE_TO_HANDLE_GET_SYNCOBJ_FD	1
 #define AMDGPU_FENCE_TO_HANDLE_GET_SYNC_FILE_FD	2
-- 
2.17.1

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

[PATCH libdrm 4/7] add timeline signal/transfer ioctls v2

[Chunming Zhou]

v2: use one transfer ioctl

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
---
 xf86drm.c | 33 +++++++++++++++++++++++++++++++++
 xf86drm.h |  6 ++++++
 2 files changed, 39 insertions(+)

diff --git a/xf86drm.c b/xf86drm.c
index 1a717f02..953fc762 100644
--- a/xf86drm.c
+++ b/xf86drm.c
@@ -4257,6 +4257,21 @@ drm_public int drmSyncobjSignal(int fd, const uint32_t *handles,
     return ret;
 }
 
+drm_public int drmSyncobjTimelineSignal(int fd, const uint32_t *handles,
+					uint64_t *points, uint32_t handle_count)
+{
+    struct drm_syncobj_timeline_array args;
+    int ret;
+
+    memclear(args);
+    args.handles = (uintptr_t)handles;
+    args.points = (uintptr_t)points;
+    args.count_handles = handle_count;
+
+    ret = drmIoctl(fd, DRM_IOCTL_SYNCOBJ_TIMELINE_SIGNAL, &args);
+    return ret;
+}
+
 drm_public int drmSyncobjTimelineWait(int fd, uint32_t *handles, uint64_t *points,
 				      unsigned num_handles,
 				      int64_t timeout_nsec, unsigned flags,
@@ -4299,4 +4314,22 @@ drm_public int drmSyncobjQuery(int fd, uint32_t *handles, uint64_t *points,
     return 0;
 }
 
+drm_public int drmSyncobjTransfer(int fd,
+				  uint32_t dst_handle, uint64_t dst_point,
+				  uint32_t src_handle, uint64_t src_point,
+				  uint32_t flags)
+{
+    struct drm_syncobj_transfer args;
+    int ret;
+
+    memclear(args);
+    args.src_handle = src_handle;
+    args.dst_handle = dst_handle;
+    args.src_point = src_point;
+    args.dst_point = dst_point;
+    args.flags = flags;
+
+    ret = drmIoctl(fd, DRM_IOCTL_SYNCOBJ_TRANSFER, &args);
 
+    return ret;
+}
diff --git a/xf86drm.h b/xf86drm.h
index 60c7a84f..3fb1d1ca 100644
--- a/xf86drm.h
+++ b/xf86drm.h
@@ -876,12 +876,18 @@ extern int drmSyncobjWait(int fd, uint32_t *handles, unsigned num_handles,
 			  uint32_t *first_signaled);
 extern int drmSyncobjReset(int fd, const uint32_t *handles, uint32_t handle_count);
 extern int drmSyncobjSignal(int fd, const uint32_t *handles, uint32_t handle_count);
+extern int drmSyncobjTimelineSignal(int fd, const uint32_t *handles,
+				    uint64_t *points, uint32_t handle_count);
 extern int drmSyncobjTimelineWait(int fd, uint32_t *handles, uint64_t *points,
 				  unsigned num_handles,
 				  int64_t timeout_nsec, unsigned flags,
 				  uint32_t *first_signaled);
 extern int drmSyncobjQuery(int fd, uint32_t *handles, uint64_t *points,
 			   uint32_t handle_count);
+extern int drmSyncobjTransfer(int fd,
+			      uint32_t dst_handle, uint64_t dst_point,
+			      uint32_t src_handle, uint64_t src_point,
+			      uint32_t flags);
 
 #if defined(__cplusplus)
 }
-- 
2.17.1

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

[PATCH libdrm 1/7] addr cs chunk for syncobj timeline

[Chunming Zhou]

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
---
 include/drm/amdgpu_drm.h | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/include/drm/amdgpu_drm.h b/include/drm/amdgpu_drm.h
index d0701ffc..3d0318e6 100644
--- a/include/drm/amdgpu_drm.h
+++ b/include/drm/amdgpu_drm.h
@@ -528,6 +528,8 @@ struct drm_amdgpu_gem_va {
 #define AMDGPU_CHUNK_ID_SYNCOBJ_OUT     0x05
 #define AMDGPU_CHUNK_ID_BO_HANDLES      0x06
 #define AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES	0x07
+#define AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_WAIT    0x08
+#define AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_SIGNAL  0x09
 
 struct drm_amdgpu_cs_chunk {
 	__u32		chunk_id;
@@ -608,6 +610,13 @@ struct drm_amdgpu_cs_chunk_sem {
 	__u32 handle;
 };
 
+struct drm_amdgpu_cs_chunk_syncobj {
+	__u32 handle;
+	__u32 flags;
+	__u64 point;
+};
+
+
 #define AMDGPU_FENCE_TO_HANDLE_GET_SYNCOBJ	0
 #define AMDGPU_FENCE_TO_HANDLE_GET_SYNCOBJ_FD	1
 #define AMDGPU_FENCE_TO_HANDLE_GET_SYNC_FILE_FD	2
-- 
2.17.1

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

[PATCH libdrm 4/7] add timeline signal/transfer ioctls v2

[Chunming Zhou]

v2: use one transfer ioctl

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
---
 xf86drm.c | 33 +++++++++++++++++++++++++++++++++
 xf86drm.h |  6 ++++++
 2 files changed, 39 insertions(+)

diff --git a/xf86drm.c b/xf86drm.c
index 17e3d880..acd16fab 100644
--- a/xf86drm.c
+++ b/xf86drm.c
@@ -4257,6 +4257,21 @@ drm_public int drmSyncobjSignal(int fd, const uint32_t *handles,
     return ret;
 }
 
+drm_public int drmSyncobjTimelineSignal(int fd, const uint32_t *handles,
+					uint64_t *points, uint32_t handle_count)
+{
+    struct drm_syncobj_timeline_array args;
+    int ret;
+
+    memclear(args);
+    args.handles = (uintptr_t)handles;
+    args.points = (uint64_t)(uintptr_t)points;
+    args.count_handles = handle_count;
+
+    ret = drmIoctl(fd, DRM_IOCTL_SYNCOBJ_TIMELINE_SIGNAL, &args);
+    return ret;
+}
+
 drm_public int drmSyncobjTimelineWait(int fd, uint32_t *handles, uint64_t *points,
 				      unsigned num_handles,
 				      int64_t timeout_nsec, unsigned flags,
@@ -4299,4 +4314,22 @@ drm_public int drmSyncobjQuery(int fd, uint32_t *handles, uint64_t *points,
     return 0;
 }
 
+drm_public int drmSyncobjTransfer(int fd,
+				  uint32_t dst_handle, uint64_t dst_point,
+				  uint32_t src_handle, uint64_t src_point,
+				  uint32_t flags)
+{
+    struct drm_syncobj_transfer args;
+    int ret;
+
+    memclear(args);
+    args.src_handle = src_handle;
+    args.dst_handle = dst_handle;
+    args.src_point = src_point;
+    args.dst_point = dst_point;
+    args.flags = flags;
+
+    ret = drmIoctl(fd, DRM_IOCTL_SYNCOBJ_TRANSFER, &args);
 
+    return ret;
+}
diff --git a/xf86drm.h b/xf86drm.h
index 60c7a84f..3fb1d1ca 100644
--- a/xf86drm.h
+++ b/xf86drm.h
@@ -876,12 +876,18 @@ extern int drmSyncobjWait(int fd, uint32_t *handles, unsigned num_handles,
 			  uint32_t *first_signaled);
 extern int drmSyncobjReset(int fd, const uint32_t *handles, uint32_t handle_count);
 extern int drmSyncobjSignal(int fd, const uint32_t *handles, uint32_t handle_count);
+extern int drmSyncobjTimelineSignal(int fd, const uint32_t *handles,
+				    uint64_t *points, uint32_t handle_count);
 extern int drmSyncobjTimelineWait(int fd, uint32_t *handles, uint64_t *points,
 				  unsigned num_handles,
 				  int64_t timeout_nsec, unsigned flags,
 				  uint32_t *first_signaled);
 extern int drmSyncobjQuery(int fd, uint32_t *handles, uint64_t *points,
 			   uint32_t handle_count);
+extern int drmSyncobjTransfer(int fd,
+			      uint32_t dst_handle, uint64_t dst_point,
+			      uint32_t src_handle, uint64_t src_point,
+			      uint32_t flags);
 
 #if defined(__cplusplus)
 }
-- 
2.17.1

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