Hi Adrian, thank you for taking a look. I have a monitor device next to the target which is in promiscuous mode from which I can see that the device injecting CTS frames (another device) is functioning. It's just that the target's behaviour is puzzling as it doesn't necessarily respect the CTS, or doesn't receive it.
One theory I have is that the power saving mode is preventing it from receiving the frame, so it unwittingly wakes up to send its frame when the channel _should_ be kept clear. Unfortunately CSMA doesn't help here as I'm trying to clear the channel for another protocol. Do you think it might be possible to get some visibility into the current value of the NAV timer? That would certainly be informative for MAC research. Hugh On 17 January 2013 15:37, Adrian Chadd <adr...@freebsd.org> wrote: > On 16 January 2013 18:45, Hugh O'Brien <hugh.obr...@ul.ie> wrote: > > Hi there, > > > > I'm doing some research on CTS based co-existence with wifi for other > 2.4GHz > > protocols. I've been using packet injection to get the CTS frames on the > > air, which I can verify, but I'm not sure that certain stations are > > receiving them, as their radio occupancy seems unaffected. > > > > Is there a way to see how many CTS packets the ath5k driver has seen? I'm > > unsure if NAV properties make it up to the driver level, would anyone > know > > if the chipset could be interrogated for those values? > > Hm, it doesn't _look_ like the chips have a CTS received counter. > > There's AR_RTS_OK / AR_RTS_FAIL but I don't know if that's a MAC TX > RTS counter, or a MAC RX RTS counter. > And that doesn't help for CTS-to-self, which doesn't have an RTS exchange. > > What about just enabling promisc mode and counting the RTS/CTS/ACK > exchanges? > > > > adrian >
_______________________________________________ ath5k-devel mailing list ath5k-devel@lists.ath5k.org https://lists.ath5k.org/mailman/listinfo/ath5k-devel
