On Fri, 2019-10-04 at 09:25 -0700, James Prestwood wrote:
> 
> > I'm not even entirely sure it _is_ needed - if we've still not
> > created
> > the IBSS but are scanning for it or trying to merge the MAC address
> > won't really matter yet? Probably?
> 
> I guess its just paranoia, rather be safe than sorry. I can take this
> out, but is "Probably?" a good reason? ;)

Fair enough, nobody really cares about IBSS anyway ;-)

I guess I was mostly wondering if you had noticed anything that would
actually be a problem.

> Ok so no switch statement, simply just check that we aren't offchannel
> or scanning. I guess this would then cover the IBSS case too.

Don't think it covers IBSS - that one is really specially accessing some
IBSS data.

> > If you do care about this being more granular then you should check
> > *which* interface is scanning, and then you can still switch the MAC
> > address for *other* interfaces - but I'd still argue it should be
> > independent of interface type.
> 
> I think maybe in the future we might want this, but for now lets not
> worry about it. But just to make sure we are on the same page, your
> talking about e.g. hardware with multiple radios so you could be doing
> offchannel work/scanning/connecting simultaneously without having to
> wait for the current operation to complete?

Not really multiple radios, who cares? Just multiple interfaces would be
sufficient. You're just removing/adding some interface (as far as the
driver is concerned) - doesn't matter if you actually are scanning or
something on another interface right?

> > And, I'm confused, but isn't the polarity of the scanning check
> > wrong?
> 
> Ah yeah, after you pointed that out I realized 'scanning' is a bit
> field. I should be doing:
> 
> test_bit(SCAN_HW_SCANNING, &sdata->local->scanning)

I think checking for all the bits is fine (and necessary, just HW scan
is unlikely to be enough, changing the MAC address would also disrupt a
software scan) - just need to invert the polarity?


> Either way I'll send another patch with these things addressed.

I'll wait, thanks.

johannes

Reply via email to