Core Control Interface is not being used to drive cv32e20 primary inputs
MikeOpenHWGroup opened this issue · comments
While looking for something else I noticed that the core_control interface is reporting erroneous inputs to the core:
UVM_INFO /scratch/mike/GitHubRepos/openhwgroup/core-v-verif/cv32e20/dev/cv32e20/tb/uvmt/uvmt_cv32e20_tb_ifs.sv(188) @ 0: reporter [CORE_CNTRL_IF] Quasi-static CORE control inputs:
pulp_clock_en = 0
scan_cg_en = 0
boot_addr = 00000080
mtvec_addr = 00000000
dm_halt_addr = 1a110800
dm_exception_addr = 1a111000
hart_id = 00000000
Several of these are just plain wrong and at least one (pulp_clock_en
) is a hold-over from the CV32E40P. A quick visual inspection reveals that the core control interface is not being used as intended.
There are two paths to resolve this:
- Depreciate the use of the core control interface entirely.
- Update the core control interface and use it to drive the core control primary inputs to the core.
I favour option 2, but we can discuss at our next verification team meeting.