Re: [PATCH] drivers: misc: ti-st: Use int instead of fuzzy char for callback status
From: Samuel Ortiz
Date: Wed Jul 13 2016 - 07:01:07 EST
Hi Marcel,
On Wed, Jul 13, 2016 at 11:56:02AM +0100, Marcel Holtmann wrote:
> Hi Mauro,
>
> >> On mips and parisc:
> >>
> >> drivers/bluetooth/btwilink.c: In function 'ti_st_open':
> >> drivers/bluetooth/btwilink.c:174:21: warning: overflow in implicit constant conversion [-Woverflow]
> >> hst->reg_status = -EINPROGRESS;
> >>
> >> drivers/nfc/nfcwilink.c: In function 'nfcwilink_open':
> >> drivers/nfc/nfcwilink.c:396:31: warning: overflow in implicit constant conversion [-Woverflow]
> >> drv->st_register_cb_status = -EINPROGRESS;
> >>
> >> There are actually two issues:
> >> 1. Whether "char" is signed or unsigned depends on the architecture.
> >> As the completion callback data is used to pass a (negative) error
> >> code, it should always be signed.
> >> 2. EINPROGRESS is 150 on mips, 245 on parisc.
> >> Hence -EINPROGRESS doesn't fit in a signed 8-bit number.
> >>
> >> Change the callback status from "char" to "int" to fix these.
> >>
> >> Signed-off-by: Geert Uytterhoeven <geert@xxxxxxxxxxxxxx>
> >
> > Patch looks sane to me, but who will apply it?
> >
> > Anyway:
> >
> > Acked-by: Mauro Carvalho Chehab <mchehab@xxxxxxxxxxxxxxxx>
>
> I can take it through bluetooth-next if there is no objection.
>
> Samuel, are you fine with that?
Yes, please go ahead.
Cheers,
Samuel.