Re: [PATCH v2 2/6] target/arm: Fix nregs computation in do_ld_zpa

[Richard Henderson]

On 2/7/24 00:46, Peter Maydell wrote:
> > @@ -4600,7 +4601,7 @@ static void do_ld_zpa(DisasContext *s, int zt, int pg,
> >        * accessible via the instruction encoding.
> >        */
> >       assert(fn != NULL);
> > -    do_mem_zpa(s, zt, pg, addr, dtype, nreg, false, fn);
> > +    do_mem_zpa(s, zt, pg, addr, dtype, nreg + 1, false, fn);
> >   }
> >
> >   static bool trans_LD_zprr(DisasContext *s, arg_rprr_load *a)
>
> What about do_st_zpa() ? It's not obvious what the 'nreg'
> encoding is in the a->nreg field in arg_rprr_store, but
> it's definitely confusing that do_st_zpa() calls
> do_mem_zpa() passing "nreg" whereas do_ld_zpa() now
> passes it "nreg + 1". Can we make it so the handling
> in these two functions lines up?

Yes, I think there may be a bug in store as well.
Comparing the two is complicated by the cut outs for LDFF1, LDNF1, LD1R and PRF.


r~