Re: [RFC net-next v2 3/4] net: dsa: mt7530: set STP state also on filter ID 1

From: Vladimir Oltean <olteanv@gmail.com>
To: DENG Qingfang <dqfext@gmail.com>
Cc: "Sean Wang" <sean.wang@mediatek.com>,
	"Landen Chao" <Landen.Chao@mediatek.com>,
	"Andrew Lunn" <andrew@lunn.ch>,
	"Vivien Didelot" <vivien.didelot@gmail.com>,
	"Florian Fainelli" <f.fainelli@gmail.com>,
	"David S. Miller" <davem@davemloft.net>,
	"Jakub Kicinski" <kuba@kernel.org>,
	"Matthias Brugger" <matthias.bgg@gmail.com>,
	netdev@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
	linux-mediatek@lists.infradead.org, linux-kernel@vger.kernel.org,
	"Eric Woudstra" <ericwouds@gmail.com>,
	"René van Dorst" <opensource@vdorst.com>,
	"Frank Wunderlich" <frank-w@public-files.de>
Subject: Re: [RFC net-next v2 3/4] net: dsa: mt7530: set STP state also on filter ID 1
Date: Mon, 2 Aug 2021 18:42:26 +0300	[thread overview]
Message-ID: <20210802154226.qggqzkxe6urkx3yf@skbuf> (raw)
In-Reply-To: <20210802153129.1817825-1-dqfext@gmail.com>

On Mon, Aug 02, 2021 at 11:31:29PM +0800, DENG Qingfang wrote:
> On Mon, Aug 02, 2021 at 04:43:36PM +0300, Vladimir Oltean wrote:
> > On Sun, Aug 01, 2021 at 03:10:21AM +0800, DENG Qingfang wrote:
> > > --- a/drivers/net/dsa/mt7530.h
> > > +++ b/drivers/net/dsa/mt7530.h
> > > @@ -181,7 +181,7 @@ enum mt7530_vlan_egress_attr {
> > >
> > >  /* Register for port STP state control */
> > >  #define MT7530_SSP_P(x)			(0x2000 + ((x) * 0x100))
> > > -#define  FID_PST(x)			((x) & 0x3)
> >
> > Shouldn't these macros have _two_ arguments, the FID and the port state?
> >
> > > +#define  FID_PST(x)			(((x) & 0x3) * 0x5)
> >
> > "* 5": explanation?
> >
> > >  #define  FID_PST_MASK			FID_PST(0x3)
> > >
> > >  enum mt7530_stp_state {
> > > --
> > > 2.25.1
> > >
> >
> > I don't exactly understand how this patch works, sorry.
> > Are you altering port state only on bridged ports, or also on standalone
> > ports after this patch? Are standalone ports in the proper STP state
> > (FORWARDING)?
>
> The current code only sets FID 0's STP state. This patch sets both 0's and
> 1's states.
>
> The *5 part is binary magic. [1:0] is FID 0's state, [3:2] is FID 1's state
> and so on. Since 5 == 4'b0101, the value in [1:0] is copied to [3:2] after
> the multiplication.
>
> Perhaps I should only change FID 1's state.

Keep the patches dumb for us mortals please.
If you only change FID 1's state, I am concerned that the driver no
longer initializes FID 0's port state, and might leave that to the
default set by other pre-kernel initialization stage (bootloader?).
So even if you might assume that standalone ports are FORWARDING, they
might not be.