Nebula is a collaborative project aiming to be a Linux-capable CPU supporting the RV64IMAFDCV_Zicsr_Zifencei
and virtal memory. It will support 32-bit or 64-bit, optionally multiply/divide (M), atomics (A), float (FD), compressed instructions (C) and vector instructions (V). Nebula will be superscalar in-order and may later become out-of-order.
This work ("Nebula RISC-V") is licensed under CERN-OHL-P v2
If you create an FPGA bitstream, a compiled simulation, a tapeout or other compiled design with a Nebula-RISC-V core, the marchid
CSR is all the necessary attribution. For this, the value of marchid
must not be changed and it must be readable by software running in M-mode. It is recommended that any third-party releases of a modified Nebula-RISC-V CPU set the highest bit of mimpid
to 1 to distinguish from official releases.