Allow non-zero vendor_id for libspdm_get_element_from_opaque_data() #2935
Conversation
jyao1
force-pushed
the
opaque
branch
4 times, most recently
from
64851e3 to
8428037
Compare
|
|
||
| opaque_data_ptr = (uint8_t *)&opaque_data; | ||
| opaque_data_size = sizeof(opaque_data); | ||
| status = libspdm_get_element_from_opaque_data(spdm_context, |
There was a problem hiding this comment.
Does libspdm_get_element_from_opaque_data only get the first element_id? What if opaque data contains multiple opaque elements with the same element_id?
There was a problem hiding this comment.
libspdm_get_element_from_opaque_data() will get (element_id, and sm_data_id) pair. Although the name is generic, but it has assumption that element_id is DMTF, because sm_data_id is only under DMTF.
To me, this function is mess, I would not define in this way, if I can redo the work.
I plan to define two functions:
libspdm_get_element_from_opaque_data_with_element_id (element_id, element_index)- that is to get element with ID and index to support multiple elements.libspdm_get_sm_data_element_from_opaque_data (sm_data_id)- that is to get SM_DATA only, with assumption that element_id is DMTF.
The first function can be used for SPDM 1.4 with DMTF-DSP, and support DSP0289 Authorization AODS.
libspdm_get_aods_element_from_opaque_data (aods_id)- that is to get AODS with assumption that element_id is DMTF-DSP.
I will file a new issue for the new function.
There was a problem hiding this comment.
I will file a new issue for the new function.
That will be in a separate pull request?
There was a problem hiding this comment.
yes, separate PR definitely.
Most likely, I will submit after I finish DSP0289 Auth POC. Then we can know how to handle different element ID.
Signed-off-by: Jiewen Yao <[email protected]>
Signed-off-by: Jiewen Yao <[email protected]>
Signed-off-by: Jiewen Yao <[email protected]>
Fix #2934