Re: [PATCH] net: ipa: Fix valid_state leak in ipa_smp2p_init()

[Wentao Liang]

Hi Greg, Alex,

Sorry for the very late reply to both of you, and for the extra work this
caused.

Alex, you are right that 96ca1e658ae45 ("net: ipa: fix SMEM state handle
leaks in SMP2P init") already covers this.  It releases both SMEM state
handles on the init error paths and in ipa_smp2p_exit(), and it has been
in mainline since v7.2-rc2.  My patch duplicates it, was incomplete on
its own, and was generated from an old snapshot (v7.1-rc4 based), which
is why its context does not match v7.2.  There is nothing left to fix, so
please disregard it; the patchwork entry already shows as not applicable.

Greg, I owe you an answer on the Assisted-by tag, and I am sorry for
leaving it hanging.  The candidate came from our static analysis tooling,
and DeepSeek and GPT-6.1 Sol were used to audit the finding and help
write the change; I reviewed the result and I am responsible for what I
sent.  I should have declared that in the changelog itself, and I did
not.  I have re-read Documentation/process/generated-content.rst on this
and will state the tools used in the changelog text from now on.

Since all three changes here are already upstream, there is no v2 to
send.  On the volume: you are right, I sent far more patches than I could
follow up on, and the review time that cost is on me.  This has been an
important lesson.  I am now going through all of the outstanding feedback
on every patch I sent, by hand, thread by thread, and I will answer each
one.  I will not post patches in batches larger than I can personally
review and follow up on.

Thanks,
Wentao