Re: [PATCH bpf-next v2 3/4] selftests/bpf: Update qemu config for riscv64 in vmtest.sh

[Pu Lehui]

On 2026/9/30 15:19, bot+bpf-ci@xxxxxxxxxx wrote:
> > selftests/bpf: Update qemu config for riscv64 in vmtest.sh
>
> TL;DR: The new '-cpu max' riscv64 option in vmtest.sh needs qemu 8.2.0+, so
> cross-runs on qemu 7.2 to 8.1 may fail; the "v7.2.0+" comment and commit
> message may be stale.
>
> > diff --git a/tools/testing/selftests/bpf/vmtest.sh b/tools/testing/selftests/bpf/vmtest.sh
> > --- a/tools/testing/selftests/bpf/vmtest.sh
> > +++ b/tools/testing/selftests/bpf/vmtest.sh
>
> [ ... ]
>
> > @@ -36,13 +36,13 @@ aarch64)
> >   	;;
> >   riscv64)
> >   	# required qemu version v7.2.0+
> >   	QEMU_BINARY=qemu-system-riscv64
> >   	QEMU_CONSOLE="ttyS0,115200"
> >   	HOST_FLAGS=(-M virt -cpu host -enable-kvm -smp 8)
> > -	CROSS_FLAGS=(-M virt -cpu rv64,sscofpmf=true -smp 8)
> > +	CROSS_FLAGS=(-M virt -cpu max,sscofpmf=false -smp 8)
> >   	BZIMAGE="arch/riscv/boot/Image"
> >   	ARCH="riscv"
> >   	;;
>
> Is the "required qemu version v7.2.0+" comment still accurate after this
> change?
>
> The riscv 'max' CPU model isn't available in qemu before v8.2.0.  In
> target/riscv/cpu.c, v7.2.0 and v8.1.0 have no riscv_max_cpu_init() and no
> TYPE_RISCV_CPU_MAX, and v8.2.0 is the release that adds:
>
>      DEFINE_DYNAMIC_CPU(TYPE_RISCV_CPU_MAX, riscv_max_cpu_init)
>
> The previous '-cpu rv64,sscofpmf=true' worked on qemu 7.2 because the
> sscofpmf property already exists there.
>
> On a non-riscv host, run-qemu takes the CROSS_FLAGS path and runs
> ${QEMU_BINARY} with "${QEMU_FLAGS[@]}".  With qemu 7.2 through 8.1 (Debian 12
> ships 7.2, for example), qemu-system-riscv64 would fail with "unable to find
> CPU model 'max'" and the script would abort under 'set -e'.
>
> The commit message doesn't mention the raised minimum qemu version either.
> Should the comment be updated to v8.2.0+, or should the commit message state
> the new requirement?

make sense

> [ ... ]
>
>
> ---
> AI reviewed your patch. Please fix the bug or email reply why it's not a bug.
> See: https://github.com/kernel-patches/vmtest/blob/master/ci/claude/README.md
>
> CI run summary: https://github.com/kernel-patches/bpf/actions/runs/36681333169