Re: [PATCH] mmc: core: Fix error return code in sd_read_ext_regs()

[Ulf Hansson]

On Wed, 30 Nov 2022 at 12:45, Leizhen (ThunderTown)
<thunder.leizhen@xxxxxxxxxx> wrote:
>
>
>
> On 2022/11/30 19:33, Ulf Hansson wrote:
> > On Wed, 30 Nov 2022 at 12:18, Leizhen (ThunderTown)
> > <thunder.leizhen@xxxxxxxxxx> wrote:
> >>
> >>
> >>
> >> On 2022/11/30 18:04, Ulf Hansson wrote:
> >>> On Wed, 30 Nov 2022 at 10:29, Zhen Lei <thunder.leizhen@xxxxxxxxxx> wrote:
> >>>>
> >>>> Fix to return a negative error code from the error handling
> >>>> case instead of 0, as done elsewhere in this function.
> >>>>
> >>>> Fixes: c784f92769ae ("mmc: core: Read the SD function extension registers for power management")
> >>>> Signed-off-by: Zhen Lei <thunder.leizhen@xxxxxxxxxx>
> >>>> ---
> >>>>  drivers/mmc/core/sd.c | 1 +
> >>>>  1 file changed, 1 insertion(+)
> >>>>
> >>>> diff --git a/drivers/mmc/core/sd.c b/drivers/mmc/core/sd.c
> >>>> index 3662bf5320ce56d..7b64f76f0179ca8 100644
> >>>> --- a/drivers/mmc/core/sd.c
> >>>> +++ b/drivers/mmc/core/sd.c
> >>>> @@ -1277,6 +1277,7 @@ static int sd_read_ext_regs(struct mmc_card *card)
> >>>>         if (rev != 0 || len > 512) {
> >>>>                 pr_warn("%s: non-supported SD ext reg layout\n",
> >>>>                         mmc_hostname(card->host));
> >>>> +               err = -EOPNOTSUPP;
> >>>
> >>> The original intent was to not return an error code. Simply, because
> >>> the card remains functional and all but the new features from the SD
> >>> function extensions registers can still be used.
> >>
> >> OK, I got it.
> >>
> >>>
> >>> Perhaps, we should update the comment a few lines above to better
> >>> reflect that this is in-fact what we intend here.
> >>
> >> How about also add 'warning' to the output? The other two outputs
> >> contain the 'error' keyword.
> >
> > There is already a pr_warn(), or is there something you want to add to that?
>
> No, I don't have anything else to add.
>
> I just saw the other two error outputs are also use pr_warn(). Maybe they
> need to be changed to pr_err().

Aha, that's a good point! We should change those to pr_err().

Kind regards
Uffe