On Mon, 2014-03-10 at 10:03 +0100, Michal Kazior wrote:
> On 10 March 2014 09:32, Eliad Peller <eliad@xxxxxxxxxx> wrote:
> > On Mon, Mar 10, 2014 at 9:37 AM, Luciano Coelho
> > <luciano.coelho@xxxxxxxxx> wrote:
> >> On Sun, 2014-03-09 at 17:13 +0200, Eliad Peller wrote:
> >>> On Sun, Mar 9, 2014 at 4:40 PM, Luciano Coelho <luciano.coelho@xxxxxxxxx> wrote:
> >>> > With single-channel drivers, we need to be able to change a running
> >>> > chanctx if we want to use chanctx reservation. Not all drivers may be
> >>> > able to do this, so add a flag that indicates support for it.
> >>> >
> >>> > Changing a running chanctx can also be used as an optimization in
> >>> > multi-channel drivers when the context needs to be reserved for future
> >>> > usage.
> >>> >
> >>> > Introduce IEEE80211_CHANCTX_RESERVED chanctx mode to mark a channel as
> >>> > reserved so nobody else can use it (since we know it's going to
> >>> > change). In the future, we may allow several vifs to use the same
> >>> > reservation as long as they plan to use the chanctx on the same
> >>> > future channel.
> >>> >
> >>> > Signed-off-by: Luciano Coelho <luciano.coelho@xxxxxxxxx>
> >>> > ---
> >>> [...]
> >>>
> >>> > @@ -177,7 +198,12 @@ ieee80211_find_chanctx(struct ieee80211_local *local,
> >>> > list_for_each_entry(ctx, &local->chanctx_list, list) {
> >>> > const struct cfg80211_chan_def *compat;
> >>> >
> >>> > - if (ctx->mode == IEEE80211_CHANCTX_EXCLUSIVE)
> >>> > + /* We don't support chanctx reservation for multiple
> >>> > + * vifs yet, so don't allow reserved chanctxs to be
> >>> > + * reused.
> >>> > + */
> >>> > + if ((ctx->mode == IEEE80211_CHANCTX_EXCLUSIVE) ||
> >>> > + ieee80211_chanctx_is_reserved(local, ctx))
> >>> > continue;
> >>> >
> >>> i'm not sure that's correct.
> >>> you might want to find a chanctx in order to use it, not only for
> >>> reservation. any reason to reject it in this case?
> >>
> >> At the moment we don't support multi-vifs switching at the same time, so
> >> we don't allow anyone else to use a chanctx while it's reserved.
> >>
> > i'm not talking about multi-vifs switching.
> > consider the following scenario:
> > sta connected on channel 40
> > ap beaconing on channel 60
> >
> > now, you want (for whatever reason) to switch the ap to channel 40.
> > you'll reserve the sta's chanctx, and increase its refcount (to 2).
> >
> > now a new sta vif is added, and tries to connect on channel 40.
> > the check above will skip the existing channel context (because it's
> > also reserved) although there is no reason it won't be used.
>
> Good point. I have this case covered in my multi-vif csa patches. We
> could transplant it into Luca's patch?
Yes, it's a good point. What would happen with this patch is that the
new STA will get a new chanctx, so in theory it would still work (though
wasting resources. as it could be).
If your fix is simple, we could transplant it here. But if it's more
complex, I think we could have it as a new patch on top of my series.
Nobody is using channel reservation yet, so I don't think that would be
a big problem.
--
Luca.
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
