Ivan Velickovic

The current VMM only supports v0.6 of the hypervisor extension, only `riscv_he_v0.6` is used. I think it's safe to get rid of `riscv_hyp`, should we need to ever go back...

Which test fails in the MCS configuration? I get all tests passing with the following configs: `../init-build.sh -DPLATFORM=qemu-riscv-virt -DRISCV64=1 -DSIMULATION=1 -DQEMU_MEMORY=4095M -DMCS=1` `../init-build.sh -DPLATFORM=qemu-riscv-virt -DRISCV64=1 -DSIMULATION=1 -DQEMU_MEMORY=4095M`

The test `FRAMEEXPORTS0001` requires more than 2GiB of RAM on RISC-V 64-bit and so will currently fail with the following build command: `../init-build.sh -DPLATFORM=qemu-riscv-virt -DRISCV64=1 -DSIMULATION=1`. One way to fix...

I think the tests are failing since `-bios none` is missing from the QEMU command.

> > I think the tests are failing since `-bios none` is missing from the QEMU command. > > You mean `QEMU_CMD` in the cmake change here or should should...

Cool, the `virt` tests pass now.

Are we going to do anything with this? In my opinion there does not seem to be any issues with compiling seL4 on Linux or macOS and so using CMakes...

@kent-mcleod do you plan to update the kernel's manual?

I've successfully run sel4test with the following configurations: ``` ../init-build.sh -DPLATFORM=rpi4 -DAARCH64=1 -DRPI4_MEMORY=4096 -DARM_HYP=1 ../init-build.sh -DPLATFORM=rpi4 -DAARCH64=1 -DRPI4_MEMORY=4096 -DARM_HYP=1 -DRELEASE=1 ``` I would add them to CI but the following...

Actually, it's not the macro options that are being cached, it's (I assume) to do with how `incbin` takes in the same filename and so the object isn't re-compiled. A...