Alice Mikityanska wrote:

On Sun, 8 Mar 2026 at 01:23, Willem de Bruijn
[off-list ref] wrote:

quoted

Alice Mikityanska wrote:

quoted

Thanks for reviewing my series!

On Fri, 6 Mar 2026 at 22:55, Willem de Bruijn
[off-list ref] wrote:

quoted

Alice Mikityanska wrote:

quoted

From: Alice Mikityanska <redacted>

Taking further the idea of commit b10b446ce7ad ("udp: gso: Use single
MSS length in UDP header for GSO_PARTIAL"), simplify the implementation
and fix the checksum (apparently ignored by hardware anyway).

The mentioned commit started using msslen for uh->len, but still uses
newlen to adjust uh->check. If the formula for check is fixed, newlen is
assigned but never used before the loop, and newlen is overwritten after
the loop. This makes msslen not really necessary, as we can reuse
newlen, if we don't adjust mss before.

That's a big if. Why is the udp length now set to the segment length,
and not to the GSO packet length, as before?

Just so that we are on the same page: the behavior of the UDP length
field changed in Gal's commit b10b446ce7ad, not in mine. You reviewed
that commit and approved it:

https://lore.kernel.org/netdev/willemdebruijn.kernel.218d53621fba7@gmail.com/ (local)

I'm just refactoring to simplify the code a little bit + fixing the
checksum field that went out of sync with length after Gal's change.
Hope that clarifies it.

Oh right. When respinning can you add this brief summary?

It's already in the first sentence of the commit message... I can
rephrase it to add something about "refactoring without changing
behavior" if "simplify the implementation" reads ambiguous.

I totally read over that. As is is great. Thanks.