| Message ID | 20250218165757.554178-4-pbonzini@redhat.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | target/riscv: store max SATP mode as a single integer in RISCVCPUConfig | expand |
On Wed, Feb 19, 2025 at 2:58 AM Paolo Bonzini <pbonzini@redhat.com> wrote: > > Check that the argument to set_satp_mode_max_supported is valid for > the MXL value of the CPU. It would be a bug in the CPU definition > if it weren't. > > In fact, there is such a bug in riscv_bare_cpu_init(): not just > SV32 is not a valid VM mode for 64-bit CPUs, SV64 is not a > valid VM mode at all, not yet at least. > > Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Alistair > --- > target/riscv/cpu.c | 6 +++++- > 1 file changed, 5 insertions(+), 1 deletion(-) > > diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c > index cca24b9f1fc..7950b6447f8 100644 > --- a/target/riscv/cpu.c > +++ b/target/riscv/cpu.c > @@ -442,6 +442,8 @@ static void set_satp_mode_max_supported(RISCVCPU *cpu, > cpu->cfg.satp_mode.supported |= (1 << i); > } > } > + > + assert(cpu->cfg.satp_mode.supported & (1 << satp_mode)); > } > > /* Set the satp mode to the max supported */ > @@ -1502,7 +1504,9 @@ static void riscv_bare_cpu_init(Object *obj) > * satp_mode manually (see set_satp_mode_default()). > */ > #ifndef CONFIG_USER_ONLY > - set_satp_mode_max_supported(cpu, VM_1_10_SV64); > + set_satp_mode_max_supported(RISCV_CPU(obj), > + riscv_cpu_mxl(&RISCV_CPU(obj)->env) == MXL_RV32 ? > + VM_1_10_SV32 : VM_1_10_SV57); > #endif > } > > -- > 2.48.1 > >
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index cca24b9f1fc..7950b6447f8 100644 --- a/target/riscv/cpu.c +++ b/target/riscv/cpu.c @@ -442,6 +442,8 @@ static void set_satp_mode_max_supported(RISCVCPU *cpu, cpu->cfg.satp_mode.supported |= (1 << i); } } + + assert(cpu->cfg.satp_mode.supported & (1 << satp_mode)); } /* Set the satp mode to the max supported */ @@ -1502,7 +1504,9 @@ static void riscv_bare_cpu_init(Object *obj) * satp_mode manually (see set_satp_mode_default()). */ #ifndef CONFIG_USER_ONLY - set_satp_mode_max_supported(cpu, VM_1_10_SV64); + set_satp_mode_max_supported(RISCV_CPU(obj), + riscv_cpu_mxl(&RISCV_CPU(obj)->env) == MXL_RV32 ? + VM_1_10_SV32 : VM_1_10_SV57); #endif }
Check that the argument to set_satp_mode_max_supported is valid for the MXL value of the CPU. It would be a bug in the CPU definition if it weren't. In fact, there is such a bug in riscv_bare_cpu_init(): not just SV32 is not a valid VM mode for 64-bit CPUs, SV64 is not a valid VM mode at all, not yet at least. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> --- target/riscv/cpu.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)