Messages in this thread | | | From | Dan Williams <> | Date | Wed, 8 Jul 2020 07:43:48 -0700 | Subject | Re: [PATCH v2] CodingStyle: Inclusive Terminology |
| |
On Wed, Jul 8, 2020 at 1:22 AM Kees Cook <keescook@chromium.org> wrote: > > On Wed, Jul 08, 2020 at 12:23:59AM -0700, Dan Williams wrote: > > Linux maintains a coding-style and its own idiomatic set of terminology. > > Update the style guidelines to recommend replacements for the terms > > master/slave and blacklist/whitelist. > > > > Link: http://lore.kernel.org/r/159389297140.2210796.13590142254668787525.stgit@dwillia2-desk3.amr.corp.intel.com > > Cc: Jonathan Corbet <corbet@lwn.net> > > Acked-by: Randy Dunlap <rdunlap@infradead.org> > > Acked-by: Dave Airlie <airlied@redhat.com> > > Acked-by: Kees Cook <keescook@chromium.org> > > Acked-by: SeongJae Park <sjpark@amazon.de> > > Signed-off-by: Olof Johansson <olof@lixom.net> > > Signed-off-by: Chris Mason <clm@fb.com> > > Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> > > Signed-off-by: Dan Williams <dan.j.williams@intel.com> > > --- > > Changes since v1 [1] > > - Drop inclusive-terminology.rst, it is in the lore archives if the > > arguments are needed for future debates, but otherwise no pressing > > need to carry it in the tree (Linus, James) > > > > - Update the recommended terms to include replacement for 'master' and > > 'whitelist' (Kees, Andy) > > > > - Add 'target' as a replacement (Andy) > > > > - Add 'device' as a replacement (Mark) > > > > - Collect acks and signed-off-bys. Yes, the sign-offs are not reflective > > of a submission chain, but I kept "Signed-off-by" if people offered > > it. > > In that case, I will "upgrade" my Ack. ;) > > Signed-off-by: Kees Cook <keescook@chromium.org> > > :)
Noted.
> > > - Non-change: I did not add explicit language as to what to do with > > existing usages. My personal inclination is to prioritize this > > coding-style cleanup higher than others, but the coding-style document > > has typically not indicated policy on how cleanups are handled by > > subsystems. It will be a case by case effort and consideration. > > While I'd like to have published guidance on fixing existing language > (which is already underway[1]), I agree: let's start here. > > > [...] > > +For symbol names, avoid introducing new usage of 'master/slave' (or > > For symbol names, comments, documentation, and other language, avoid > introducing ...
How about "symbol names and documentation" because I'm struggling to think of an example of where this terminology would leak in outside those broad categories.
> > +'slave' independent of 'master') and 'blacklist/whitelist'. Recommended > > +replacements for 'master/slave' are: 'main/{secondary,subordinate}', > > +'primary/replica', '{initiator,requester}/{target,responder}', > > the main and primary should be merged, IMO: > > '{primary,main}/{secondary,replica,subordinate}'
Ok.
> > > +'host/{device,proxy}', or 'leader/{performer,follower}'. Recommended > > leader/performer does not track for me. Split it out? > > 'leader/follower', 'director/performer'
Sounds good.
> I have also seen: > > 'controller/worker'
Will add.
Thanks Kees.
| |