To everyone (and everybot) mailing security@kernel.org with multiple public lists on CC at the same time: thank you for your service, I'll never call it mistake and believe it should be the new standard.
Doesn't matter because vulnerabilities in staging drivers aren't considered vulnerabilities now, but the way the Fixes tags are used in these two fixes are going to do wonderful things with the CNA's automation when they issue CVEs anyway: git.kernel.org/pub/scm/linux…
For each commit, the first of the fixes tags points to some irrelevant (but recent) cleanup commit which changed some whitespace or removed some commented-out code, not related at all to bug introduction.
First fix I've seen that'll probably get dropped from stable backports due to not cherry-picking clean from kmalloc_obj churn: git.kernel.org/pub/scm/linux…
We didn't like looking at perfectly good puts so we redid the code with magic scoping rules to make the puts happen invisibly for us so that we could forget...to get: git.kernel.org/pub/scm/linux…