Re: [PATCH] eeprom: at24: Add support for large EEPROMs connected to SMBus adapters

[Guenter Roeck <linux-0h96xk9xTtrk1uMJSBkQmQ@public.gmane.org>]

On Wed, Mar 25, 2015 at 05:15:04PM +0100, Wolfram Sang wrote:
> Guenter,
> 
> thanks for the update
> 
> > Ultimately, the real problem is how i2c-dev accesses a client, not how
> > i2c client drivers (who assume they have exclusive access to a chip)
> > handle multi-command sequences. Forcing extensive locking on all drivers
> > because of i2c-dev just doesn't seem to be the right thing to do.
> 
> I agree. i2c-dev is too much of a special case.
> 
> And since at24 has its own lock (I missed that), your patch might as
> well be good enough to be applied, I'd think.
> 

Ah, sorry, I blindly assumed that you are aware of that. Yes, at24
itself is not the problem, it is parallel access to the chip by i2c-dev.

The same is actually true for all the other drivers I looked at;
usually they have their own lock(s), but such locks do not protect
against interference by i2c-dev.

The bad part is that i2c-dev is heavily used by user space at my
workplace, and that code happily messes with chips which are also
handled by kernel drivers. But as I said, I have no real good idea
how to fix that - neither the user-space code nor how i2c-dev
interfers with (or completely messes up) device access by drivers.

Thanks,
Guenter