> + if (env->priv_ver >= PRIV_VERSION_1_10_0) {
> + if (get_field(env->satp, SATP_MODE) == VM_1_09_MBARE) {
> + mode = PRV_M;
> + }
> + } else {
> + if (get_field(env->mstatus, MSTATUS_VM) == VM_1_10_MBARE) {
> + mode = PRV_M;
> + }
> + }
This mixes up VM_1_09_MBARE and VM_1_10_MBARE, but they evaluate to
the same value anyway.
And as Richard said just rely on the mmu_idx from cpu_mmu_index. I
actually already did the change to remove it in a patch for a new riscv
CSR I developed and can thus confirm it works fine.
> + case CSR_SPTBR:
This should use CSR_SATP. In fact even if you want to keep 1.9.1 support
I would highly recommend to remove the CSR_SPTBR define and only use
CSR_SATP in code, with sptbr limited to comments to avoid confusion.
