[PATCH net-next 2/2] net: stmmac: disable TX checksum offload for jumbo frames on bugged cores

[Jisheng Zhang]

On Wed, Oct 21, 2026 at 12:33:37PM +0800, Jisheng Zhang wrote:
> On Tue, Oct 20, 2026 at 02:28:26PM +0200, Andrew Lunn wrote:
> > > > Here I want to query a generic rule from netdev maintainers:
> > > > can XDP frame enable HW checksum on dwmac IP? Per my understanding, the
> > > > dwmac can only do HW checksum for normal particular packets layouts, such as
> > > > TCP, UDP. What about if XDP transmits a non-IP frame?
> > > 
> > > or a generic XDP guide from netdev maintainers:
> > > Since XDP can tranmist arbitrary Eth frames, so how does the eth driver cope
> > > with HW checksum if the HW checksum is intended for supported IP packet
> > > layouts rather than arbitrary layer2 frames?
> > > 
> > > choices:
> > > c1. Simply disabled HW checksum by default. Then XDP prog must provide
> > > checksums themselves, lost the HW offload cap for IP packets.
> > > 
> > > c2. detect the frame in eth driver's xdp code path, then enable/disable
> > > HW checksum accordingly. Then how? Any example in exisiting eth drivers
> > > is appreciated.
> > 
> > Why is this specific to XDP? Cannot the network stack transmit
> > arbitrary layer2 frames? Isn't this a generic problem?
> > 
> 
> network stack normal tx supplies the skb->ip_summed metadata, and
> dwmac driver specifically can decide to enable/disable HW csum
> by checking the protocol which is extraced from skb, example
> stmmac_has_ip_ethertype()
> But XDP path doesn't have the necessary metadata, and I'm not sure
> whether stmmac_has_ip_ethertype xdb version can be implemented or
> not. That's the reason I request generic XDP guidence from netdev
> maintainers.
> 

Hi,

Given above, I think the dwmac hw csum needs to be disabled by default
to make XDP always behave properly. Since this patch2/2 has been
merged, I will update my patch:
https://lore.kernel.org/netdev/20260929121025.20821-1-jszhang@kernel.org/

1.rebase on the lastest net-next or wait for new rc1

2.update the commit msg of my patch to include above reply.

Kindly let me know if this way can be accepted or not

Thanks