From: Greg Kroah-Hartman gregkh@linuxfoundation.org
... e70af8d040d2 has a Fixes tag. Not sure why it's not automatically
backported.
Because "Fixes:" is not the flag that we are sure to trigger off of. Please read: https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html for how to do this properly.
Thanks, I just read this again to refresh my memory :-) I remember Sasha has an AI algorithm to pick up patches into the stable tree and a "Fixes" tag should be a strong indicator.
I tired to manually cherry-pick e70af8d040d2 to 5.15.y and got some merge conflicts. Probably that's why Sasha's script didn't automatically do the backport. @Carl, @Jeffrey, may I know if you have some cycles to help backport e70af8d040d2?
That being said, because some subsystem maintainers do NOT put cc: stable in their patches, we do sometimes sweep the tree and try to pick up things with only "Fixes:" but we don't always catch everything.
So if you want to be sure a patch is applied, please always add a cc: stable in the patch.
greg k-h
Hi Greg, since you're here, I'd like to ask a question:
If I add the cc: stable line in a patch and use git-send-email to post the patch, git-send-email also posts the patch to the stable list -- is this acceptable? Sometimes a patch may have to undergo multiple revisions, meaning all the discussion emails go to the stable list unnecessarily, and I guess this is not good?
It looks like there is no git-send-email option to exclude an email.
Thanks, Dexuan