[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAHvy4Aq8G2vLzFCCRRQV5kCD4jp8oYW+c=m_foyHXKoeiCod5A@mail.gmail.com>
Date: Mon, 12 Aug 2024 12:55:21 +0200
From: Pieter <vtpieter@...il.com>
To: Arun.Ramadoss@...rochip.com
Cc: andrew@...n.ch, olteanv@...il.com, davem@...emloft.net,
linux@...linux.org.uk, conor+dt@...nel.org, Woojung.Huh@...rochip.com,
robh@...nel.org, krzk+dt@...nel.org, f.fainelli@...il.com, kuba@...nel.org,
UNGLinuxDriver@...rochip.com, marex@...x.de, edumazet@...gle.com,
pabeni@...hat.com, pieter.van.trappen@...n.ch, devicetree@...r.kernel.org,
netdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next v4 5/5] net: dsa: microchip: apply KSZ87xx family
fixes wrt datasheet
Hi Arun,
> @@ -141,7 +141,7 @@ static struct sk_buff *ksz8795_rcv(struct sk_buff
> > *skb, struct net_device *dev)
> > {
> > u8 *tag = skb_tail_pointer(skb) - KSZ_EGRESS_TAG_LEN;
> >
> > - return ksz_common_rcv(skb, dev, tag[0] & 7,
> > KSZ_EGRESS_TAG_LEN);
> > + return ksz_common_rcv(skb, dev, tag[0] & 3,
> > KSZ_EGRESS_TAG_LEN);
> > }
>
> This change can be separate patch. since it is not related to
> ksz87xx_dev_ops structure. Is it a fix or just good to have one. If it
> is a fix then it should be point to net tree.
>
It's a fix wrt to datasheet but in reality I can see from tests with
a KSZ8794 that bit 2 is always 0 so the bug doesn't manifest itself.
Please advise, keep it in net-next or make a separate patch for the
net tree?
Powered by blists - more mailing lists