Shayne Chen <shayne.chen@xxxxxxxxxxxx> wrote:
>
> Hi Ping-Ke,
>
> On Mon, 2025-03-31 at 04:40 +0000, Ping-Ke Shih wrote:
> >
> > External email : Please do not click links or open attachments until
> > you have verified the sender or the content.
> >
> >
> > Shayne Chen <shayne.chen@xxxxxxxxxxxx> wrote:
> > > --- a/drivers/net/wireless/mediatek/mt76/mt76_connac.h
> > > +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac.h
> > > @@ -232,9 +232,14 @@ static inline bool is_mt7992(struct mt76_dev
> > > *dev)
> > > return mt76_chip(dev) == 0x7992;
> > > }
> > >
> > > +static inline bool is_mt7990(struct mt76_dev *dev)
> > > +{
> > > + return mt76_chip(dev) == 0x7993;
> >
> > It seems like patch 01/10 want to avoid this kind of confusion.
> >
> This helper is added in mt76_connac.h, which can be used by mt76 common
> part and mt7996 per-chip folder.
>
> > Also, there are two IDs for MT7990:
> >
> > #define MT7990_DEVICE_ID 0x7993
> > #define MT7990_DEVICE_ID_2 0x799b
> >
> These two are only used by mt7996 per-chip folder.
>
I was not aware this hardware has two PCI ID. One is 0x7993 and the other
is 0x799b for HIF2. I guess suffix '_2' means 'HIF2', which would be
clearer?
Another thing is usage of return value of mt76_chip(). Here, you compare
with raw hexadecimal 0x7993, but rarely in mt7996_eeprom_name()
MT7992_DEVICE_ID is used.
