linux-kernel - Re: [RFC PATCH] [media] rc: filter out not allowed protocols when decoding

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20120904122121.GA13018@pequod.mess.org>
Date:	Tue, 4 Sep 2012 13:21:21 +0100
From:	Sean Young <sean@...s.org>
To:	Changbin Du <changbin.du@...il.com>
Cc:	mchehab@...radead.org, paul.gortmaker@...driver.com,
	sfr@...b.auug.org.au, srinivas.kandagatla@...com,
	linux-media@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH] [media] rc: filter out not allowed protocols when
 decoding

On Tue, Sep 04, 2012 at 11:06:07AM +0800, Changbin Du wrote:
> > >               mutex_lock(&ir_raw_handler_lock);
> > > -             list_for_each_entry(handler, &ir_raw_handler_list, list)
> > > -                     handler->decode(raw->dev, ev);
> > > +             list_for_each_entry(handler, &ir_raw_handler_list, list) {
> > > +                     /* use all protocol by default */
> > > +                     if (raw->dev->allowed_protos == RC_TYPE_UNKNOWN ||
> > > +                         raw->dev->allowed_protos & handler->protocols)
> > > +                             handler->decode(raw->dev, ev);
> > > +             }
> >
> > Each IR protocol decoder already checks whether it is enabled or not;
> > should it not be so that only allowed protocols can be enabled rather
> > than checking both enabled_protocols and allowed_protocols?
> >
> > Just from reading store_protocols it looks like decoders which aren't
> > in allowed_protocols can be enabled, which makes no sense. Also
> > ir_raw_event_register all protocols are enabled rather than the
> > allowed ones.
> >
> >
> > Lastely I don't know why raw ir drivers should dictate which protocols
> > can be enabled. Would it not be better to remove it entirely?
> 
> 
> I agree with you. I just thought that the only thing a decoder should care
> is its decoding logic, but not including decoder management. My idaea is:
>      1) use enabled_protocols to select decoders in ir_raw.c, but not
> placed in decoders to do the judgement.
>      2) remove  allowed_protocols or just use it to set the default
> decoder (also should rename allowed_protocols  to default_protocol).

The default decoder should be the one set by the rc keymap.

> I also have a question:
>      Is there a requirement that one more decoders are enabled for a
> IR device at the same time?

Yes, you want to be able to multiple remotes on the IR device (which
you can do as long as the scancodes don't overlap, I think), and the 
lirc device is implemented as a decoder, so you might want to see the
raw IR as well as have it decoded.

>     And if that will lead to a issue that each decoder may decode a
> same pulse sequence to different evnets since their protocol is
> different?

At the moment, no. David Hardeman has sent a patch for this:

http://patchwork.linuxtv.org/patch/11388/


Sean
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/