Re: [PATCH] net: prestera: acl: Add check for kmemdup

[Joe Perches]

On Fri, 2022-09-30 at 08:44 -0700, Jakub Kicinski wrote:
> On Fri, 30 Sep 2022 08:20:47 -0700 Joe Perches wrote:
> > IMO: If Volodymyr wants to be a maintainer here, he should put
> > his email as an entry in the MAINTAINERS file for the subsystem.
> 
> It's about Fixes tags, unfortunately having everyone of note listed 
> in MAINTAINERS is pretty much impossible. Even tho we are trying.
> 
> > > > Maybe there is a problem of the script that misses one.  
> > 
> > I don't think so.  Maybe you have more evidence...
> 
> I'll CC you when I tell people to CC authors of patches under Fixes
> going forward, I don't have a list going back.
> 
> > > > Anyway, I have already submitted the same patch and added
> > > > "vmytnyk@xxxxxxxxxxx" this time.  
> > > 
> > > Ha! So you do indeed use it in a way I wasn't expecting :S
> > > Thanks for the explanation.
> > > 
> > > Joe, would you be okay to add a "big fat warning" to get_maintainer
> > > when people try to use the -f flag?  
> > 
> > No, not really.  -f isn't required when the file is in git anyway.
> 
> Ah. Yeah. I'd make it error out when run on a source file without -f :S
> 
> > > Maybe we can also change the message
> > > that's displayed when the script is run without arguments to not
> > > mention -f?  
> > 
> > I think that's a poor idea as frequently the script isn't used
> > on patches but simply to identify the maintainers of a particular
> > file or subsystem.
> 
> Identify the maintainers and report a bug, or something else? As a
> maintainer I can tell you that I don't see bug reports as often as I
> see trivial patches from noobs which miss CCs. And I personally don't
> think I ever used get_maintainer on anything else than a patch.
> 
> > > We're getting quite a few fixes which don't CC author, I'm guessing
> > > Jiasheng's approach may be a common one.  
> > 
> > There's no great way to identify "author" or "original submitter"
> > and frequently the "original submitter" isn't a maintainer anyway.
> 
> Confusing sentence. We want for people who s-o-b'd the commit under
> Fixes to be CCed.

If a file or a file modified by a patch is listed in the MAINTAINERS,
git history isn't used unless --git is specified.

For a patch, maybe the author and other SOBs of a commit specified
by a "Fixes:" line SHA-1 in the commit message could be added automatically.