Content-Type: multipart/mixed; boundary="===============2548014680134449412==" MIME-Version: 1.0 From: James Prestwood Subject: Re: [RFC 4/6] frame-xchg: refactor to work with offchannel module Date: Mon, 22 Jun 2020 13:06:15 -0700 Message-ID: In-Reply-To: <5ed8ec95-5175-c844-750b-1ba8b7940890@gmail.com> List-Id: To: iwd@lists.01.org --===============2548014680134449412== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On Mon, 2020-06-22 at 15:03 -0500, Denis Kenzior wrote: > Hi James, > = > > > > -void frame_xchg_stop(uint64_t wdev_id); > > > > +void frame_xchg_stop(uint64_t wdev_id, uint32_t id); > > > = > > > Do we even need the wdev_id here now? Also, should this be > > > _cancel? > > = > > It just made the lookup easier, but yeah we could iterate each > > queue in > > the phy list. And yeah, I could change that to _cancel. > > = > = > Hmm, how so? I don't even see wdev_id being used now. Ah yeah not in frame-xchg now. I moved that into offchannel.c. So it just makes *that* lookup easier :) But when I think about it we will generally only ever have a couple phy's (or just one), and barely any requests per-phy anyways so a full lookup isnt too expensive. > = > Regards, > -Denis >=20 --===============2548014680134449412==--