hi_mpi_vpc_create_chn
Applicability
|
Product |
Supported |
|---|---|
|
|
√ |
|
|
√ |
|
|
√ |
|
|
√ |
|
|
☓ |
Description
Creates an image processing channel based on the configured channel attributes. The channel ID is specified by the user.
Prototype
hi_s32 hi_mpi_vpc_create_chn (hi_vpc_chn chn, const hi_vpc_chn_attr *attr)
Parameters
|
Parameter |
Input/Output |
Description |
|---|---|---|
|
chn |
Input |
Channel ID for image processing. |
|
attr |
Input |
Pointer to the attributes of an image processing channel. |
Returns
- 0: success
- Other values: failure. For details, see VPC Return Codes.
Restrictions
- The channel ID of a single device must not exceed the maximum channel ID on
Atlas inference products . VDEC involves VPC processing behind the scene. VDEC assigns internal VPC channel IDs within [256, 511]. To avoid channel ID conflicts between VPC and VDEC, you are advised to set the user VPC channel IDs within [0, 255]. - If attr is left empty, the error code HI_ERR_VPC_NULL_PTR is returned.
- Before this API call, ensure that a channel has not been created yet or has been destroyed. Otherwise, a failure is returned.
- For Ascend virtual instances, the requirements on the number of channels are as follows:
Atlas inference products : Number of channels = (Number of allocated VPC hardware units/Number of VPC hardware units) x 256. If the total number of channels is not an integer, round down the value.Atlas A2 training products /Atlas A2 inference products : The maximum number of channels is 256.Atlas 200I/500 A2 inference products : The maximum number of channels is 128.To check the compute resource specifications of the Ascend virtual instances in different scenarios, run the npu-smi info -t template-info command on the server with the Ascend AI Processor installed.