[prev in list] [next in list] [prev in thread] [next in thread] 

List:       freebsd-wireless
Subject:    Re: 802.11n TX ADDBA request supported?
From:       Adrian Chadd <adrian.chadd () gmail ! com>
Date:       2023-07-11 14:27:50
Message-ID: CAJ-VmomeatKb0kWnfRjvr6FH-BD5BLG090dVzkDjdfBbOtQuPQ () mail ! gmail ! com
[Download RAW message or body]

hi!

So, ampdu_request doesn't send addba itself. It calls ic_send_action
because some NICs want a firmware message to start AMPDU, not sending
frames.

Look at what ath(4) is doing. It registers for ic_send_action and that
sends the actual frames.

mwl(4) (an old 11n marvell chipset, the first with freebsd 11n support)
instead uses a firmware message.


-adrin


On Tue, 4 Jul 2023 at 01:03, =E5=90=B3=E6=81=A9=E7=B7=AF <rickywu0421@gmail=
.com> wrote:

> Hi,
>
> I'm working on 14.0-CURRENT, and I found that it seems net80211(4) lack o=
f
> supporting 802.11n/e TX ADDBA request:
> When we call `ieee80211_ampdu_request()`, then the function ends up
> calling `ic->ic_send_action` (`ieee80211_send_action()`) with category
> `IEEE80211_ACTION_CAT_BA` and action `IEEE80211_ACTION_BA_ADDBA_REQUEST`.
> But then `ieee80211_send_action()` call the dummy function `send_inval()`
> which does nothing. So net80211(4) doesn't send ADDBA requests for us.
>
> The same thing happens when RX an ADDBA request, net80211(4) doesn't repl=
y
> an ADDBA response to the transmitter. The only thing I found is that
> net80211(4) has the `ieee80211_send_bar()` which does send BA requests
> (which is not ADDBA request) for us. I wonder if there is something I'm
> missing or if the "TX/RX ADDBA request/response" is handling most in
> driver/firmware?
>
> Additionally, if I would like to test the 802.11n part in net80211(4),
> which driver/device is suggested to use? Thanks.
>

[Attachment #3 (text/html)]

<div dir="ltr">hi!<div><br></div><div>So, ampdu_request doesn&#39;t send addba \
itself. It calls ic_send_action because some NICs want a firmware message to start \
AMPDU, not sending frames.</div><div><br></div><div>Look at what ath(4) is doing. It \
registers for ic_send_action and that sends the actual \
frames.</div><div><br></div><div>mwl(4) (an old 11n marvell chipset, the first with \
freebsd 11n support) instead uses a firmware \
message.</div><div><br></div><div><br></div><div>-adrin</div><div><br></div></div><br><div \
class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, 4 Jul 2023 at 01:03, \
吳恩緯 &lt;<a href="mailto:rickywu0421@gmail.com">rickywu0421@gmail.com</a>&gt; \
wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px \
0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div \
dir="ltr">Hi,<br><br>I&#39;m working on 14.0-CURRENT, and I found that it seems \
net80211(4) lack of supporting 802.11n/e TX ADDBA request:<br>When we call \
`ieee80211_ampdu_request()`, then the function ends up calling \
`ic-&gt;ic_send_action` (`ieee80211_send_action()`) with category \
`IEEE80211_ACTION_CAT_BA` and action `IEEE80211_ACTION_BA_ADDBA_REQUEST`. But then \
`ieee80211_send_action()` call the dummy function `send_inval()` which does nothing. \
So net80211(4) doesn&#39;t send ADDBA requests for us.<br><br>The same thing happens \
when RX an ADDBA request, net80211(4) doesn&#39;t reply an ADDBA response to the \
transmitter. The only thing I found is that net80211(4) has the \
`ieee80211_send_bar()` which does send BA requests (which is not ADDBA request) for \
us. I wonder if there is something I&#39;m missing or if the &quot;TX/RX ADDBA \
request/response&quot; is handling most in driver/firmware?<br><br>Additionally, if I \
would like to test the 802.11n part in net80211(4), which driver/device is suggested \
to use? Thanks.  <br><div></div></div> </blockquote></div>



[prev in list] [next in list] [prev in thread] [next in thread]