From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============9096216517656667241==" MIME-Version: 1.0 From: Denis Kenzior Subject: Re: [PATCH v3 7/7] cdmaphonesim: Add CDMA SMS Support Date: Wed, 05 Jan 2011 12:05:32 -0600 Message-ID: <4D24B2EC.1000907@gmail.com> In-Reply-To: <4D24B25E.2080007@nokia.com> List-Id: To: ofono@ofono.org --===============9096216517656667241== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Hi Lei, >>> diff --git a/plugins/phonesim.conf b/plugins/phonesim.conf >>> index 74bb645..8cd9678 100644 >>> --- a/plugins/phonesim.conf >>> +++ b/plugins/phonesim.conf >>> @@ -12,3 +12,7 @@ >>> #[phonesim] >>> #Address=3D127.0.0.1 >>> #Port=3D12345 >>> + >>> +#[cdmaphonesim] >>> +#Address=3D127.0.0.1 >>> +#Port=3D12345 >> >> If we have both phonesim and cdmaphonesim plugins active, we need to >> make sure they don't interfere with each other. Perhaps adding a >> Type=3Dcdma and having cdmaphonesim look for that Type tag and ignore the >> rest would be a good idea? >> > = > If both phonesim and cdmaphonesim active, they should have to be > configured with running with different port, right? If so,they should > not interfere with each other. And, making sure ports are different is > the job of the end-user, right? That is correct and not what I was referring to. > = > I am a bit lost about the Type attribute. cdmaphonesim is currently > defined as a separate plugin and by the time cdmaphonesim's > parse_config() is invoked, it already knows that it is cdma. Thus, I > don't see why needing another flag indicating this. When you say ignore > the rest, could you pls be more specific about what to ignore and what's > the intention of doing that? Sorry, still learning... :-) > = So correct me if I'm wrong, but there's a race between cdmaphonesim and phonesim plugins: if phonesim starts first it reads phonesim.conf and creates both phonesim and cdmaphonesim modems. When cdmaphonesim starts, it tries to create the modems again, but fails. Similar situation vice-versa. This is why I'm suggesting the Type attribute. Maybe I'm blind and this race does not happen? Regards, -Denis --===============9096216517656667241==--