Hi, On 28 Nov 2014, at 20:15, Greg KH <[email protected]> wrote:
> On Fri, Nov 28, 2014 at 04:03:34PM +0100, Javier González wrote: >> From: Javier Gonzalez <[email protected]> > > Any reason you sent this twice? My mistake, sorry. > > And your code is full of FIXMEs, which implies that you aren't even > happy with the code, so why would we want to review it as-is now? > The driver is still in development, but I think it is mature enough as to involve more people interested in TrustZone. I have dealt the development repository with parties that I know are want to contribute in this effort; still I think that this is the right place to find both interested people, and good feedback - therefore the RFC. > Please come up with a design, and implementation, that you feel > comfortable with before asking others to do work to review your broken > implementation. My interest lays more in the interface part. I have worked with different TrustZone frameworks and one of the main issues with them is that there is no standard interface from within the kernel. I strongly believe that having a supported driver in the kernel with a defined interface can help getting more people involved in porting more TrustZone drivers, and in general facilitating TrustZone's use. This is of special relevance for mobile devices today, but with ARMv8 down the corner servers and set-top boxes would also profit. Regarding the interface in itself, I am mostly comfortable with the design and implementation, but I am sure it is still improbable. Most of the FIXMEs are in a specific driver I ported. It is aTrustZone open source implementation called Open Virtualization. I still need to work in improving the code, but it performs pretty good at the moment. My intention is to continue working on it, but in the meantime I think it is worth getting feedback from more experience kernel hackers. > >> +static void trustzone_dev_release(struct device *dev) >> +{ >> + /* FIXME: You need to fix all this crap... */ > > Yes you do, why you are asking others to do your work is beyond my > comprehension... > Not the most appropriate comment, I agree. But as I mentioned, it is work in progress, and my intention is to get more people involved and obtain early feedback in a working prototype. Thanks for your comments. I hope that you find the time to look deeper in the code. > greg k-h Best, Javier. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
