Comment 6 for bug 1929035

Hi, so (like mentioned before) IBM usually shares with us a (self-contained and minimal) list of commits (either hashes or - if needed - backports) that constitute a certain function/feature ('logical unit') and that apply cleanly to jammy (22.04) master-next.

Bu anywayt I had a look at the list (that you've referenced) and went down to 2021-10-28 (because the 2021-10-16 are the commits that are target of this ticket anyway) and checked whether these have already landed in jammy master-next or not -- and the majority is already in.
But the following 7 are not:
45c3ff7a9ac1 "net/smc: Clean up local struct sock variables"
af1877b6cad1 "net/smc: Print function name in smcr_link_down tracepoint"
a3a0e81b6fd5 "net/smc: Introduce tracepoint for smcr link down"
aff3083f10bf "net/smc: Introduce tracepoints for tx and rx msg"
482626086820 "net/smc: Introduce tracepoint for fallback"
4c1e34c0dbff "vsock: Enable y2038 safe timeval for timeout"
0daa55d033b0 "octeontx2-af: cn10k: debugfs for dumping LMTST map table"
(the last two make up - with add. two - the "Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net" -- I'm for example not convinced if these are really needed, esp. 0daa55d033b0).

It would be important to know whether these are really needed or can be safely ignored.

At the end a list of commits is needed that can be applied (ideally cherry-picked to) to the current jammy master-next.
(git clone git://git.launchpad.net/~ubuntu-kernel/ubuntu/+source/linux/+git/jammy --branch master-next --single-branch jammy-master-next)

And if starting to pick the three EID commits from LP#1929060 - "smc: Add User-defined EID Support" (which seem to be self-contained) and proceed further with the 10 commits mentioned here (LP#1929035 - "SMC-R v2 Support") [so going bottom up] I face a conflict at 8799e310fb3f "net/smc: add v2 support to the work request layer".
That means even if some of the above 7 are needed, they will not help on solving the conflict with 8799e310fb3f, since the 7 above are all newer.
Hence I assume there is one or even more commits needed (even from before 2021-10-16 ?) that help solving that conflict (...or a backport of 8799e310fb3f).