Re: [PATCH v2 05/26] xen/riscv: introduce guest riscv,isa string

[Oleksii Kurochko <oleksii.kurochko@xxxxxxxxx>]

On 5/19/26 5:56 PM, Jan Beulich wrote:
> On 19.05.2026 17:17, Oleksii Kurochko wrote:
> > On 5/19/26 4:53 PM, Jan Beulich wrote:
> > > On 19.05.2026 16:49, Oleksii Kurochko wrote:
> > > > int init_guest_isa(struct domain *d)
> > > > {
> > > >        int len;
> > > >
> > > >        bitmap_andnot(d->arch.isa, riscv_isa, guest_unsupp,
> > > >                      RISCV_ISA_EXT_MAX);
> > > >
> > > >        len = build_guest_isa_str(NULL, 0, d->arch.isa);
> > > >        if ( len < 0 )
> > > >            return len;
> > > >
> > > >        d->arch.isa_str = xmalloc_array(char, len + 1);
> > > >        if ( !d->arch.isa_str )
> > > >            return -ENOMEM;
> > > >
> > > >        build_guest_isa_str(d->arch.isa_str, len + 1, d->arch.isa);
> > > At least ASSERT() the success of this?
> > I will add:
> >
> > ASSERT(build_guest_isa_str(d->arch.isa_str, len + 1, d->arch.isa) == len);
> Ehem. Please check how ASSERT() works (and the difference to BUG_ON()).
Condition itself looks correct for ASSERT(). If build_guest_isa_str() 
returns value equal to len then assert_failed() shouldn't be called.
Maybe do you mean that it will never fire in release build then yes it 
should be changed to BUG_ON() and the condition should be inverted:
 BUG_ON(build_guest_isa_str(d->arch.isa_str, len + 1, d->arch.isa) != len);

~ Oleksii