On Fri, Sep 28, 2012 at 4:06 AM, Johannes Berg
<johannes@xxxxxxxxxxxxxxxx> wrote:
> On Thu, 2012-09-27 at 11:59 -0700, Bing Zhao wrote:
>
>> + if (associated && !tx_empty) {
>> + if (unlikely(local->scan_req->flags &
>> + CFG80211_SCAN_FLAG_LOW_PRIORITY))
>
> I don't really see value in the "unlikely()" here, that just clutters
> the code and probably has very little effect on the runtime behaviour,
> this is very infrequently executed.
>
>
>> + case SCAN_ABORT:
>> + aborted = true;
>> + goto out_complete;
>
> Maybe we should have different flags though ... I mean, this is the
> first implementation that I hear of that interprets a background scan as
> "ok to abort at any time"? It seems very unlikely that other
> implementations would do that. I *think* (like I said before, I don't
> really know) that ours (Intel's) will just shorten the dwell time, or
> similar instead.
Well previous implementations I've done have used this technique for
many years and you can find them in various products (assuming they
haven't been totally rewritten) :-)
-Sam
--
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
