Re: [PATCH 15/17] ppc/pnv: convert pec->stacks[] into pec->phbs[]

[Daniel Henrique Barboza]

On 1/14/22 10:33, Cédric Le Goater wrote:
> > @@ -1520,14 +1520,19 @@ static PnvPhb4PecStack *pnv_phb4_get_stack(PnvChip 
> > *chip, PnvPHB4 *phb,
> >       for (i = 0; i < chip->num_pecs; i++) {
> >           /*
> > -         * For each PEC, check the amount of stacks it supports
> > -         * and see if the given phb4 index matches a stack.
> > +         * For each PEC, check the amount of phbs it supports
> > +         * and see if the given phb4 index matches an index.
> >            */
> >           PnvPhb4PecState *pec = &chip9->pecs[i];
> > -        for (j = 0; j < pec->num_stacks; j++) {
> > +        for (j = 0; j < pec->num_phbs; j++) {
> >               if (index == pnv_phb4_pec_get_phb_id(pec, j)) {
> > -                return &pec->stacks[j];
> > +                pec->phbs[j] = phb;
>
> Why do we need this array ?


Actually we don't. While making  these patches I forgot to assign this pointer 
back
to the array and everything worked. We don't search the PHB back from the PEC at
any point.

This is being kept because I refrain from doing too much design changes at 
once. We
can drop it though - either in this patch or in a follow up.

> > +
> > +                /* Set phb-number now since we already have it */
> > +                object_property_set_int(OBJECT(phb), "phb-number",
> > +                                               j, &error_abort);
>
> that's ugly :/

Not my proudest line of code indeed.

Perhaps we're better of trying to get rid of stack->stack_no altogether before 
even
converting it to phb->stack_no. I'll see how that goes.



Daniel


> C.
>
> > +                return pec;
> >               }
> >           }
> >       }