On 03/24/2020 10:16 PM, Brian Norris wrote:
On Tue, Mar 24, 2020 at 7:55 PM Tony Chuang <yhchuang@xxxxxxxxxxx> wrote:
Brian Norris <briannorris@xxxxxxxxxxxx> writes:
We want to measure the TX power, and the equipment just cannot
detect the signal on some rates, unless we "fix" the rate exactly.
I think we all understand this now.
So NL80211_CMD_SET_TX_BITRATE_MASK is not so useful for us
sometimes.
I'm not sure if you have directly explained why this is the case. See
your comment earlier:
"This command just mask out some of rates that are not allowed."
Sure, but if you mask out all but 1 bitrate...voila! A fixed rate!
So, see this thread from a while back. Has anyone even *tried* to use
this API you are proposing?
http://lists.infradead.org/pipermail/ath10k/2017-October/010291.html
Here is the fix I have had in my tree for a few years to let ath10k actually
set a single rate:
[greearb@ben-dt4 linux-5.4.dev.y]$ git show cccf04cc3440ddee0760249da51026bf2532f926
commit cccf04cc3440ddee0760249da51026bf2532f926
Author: Ben Greear <greearb@xxxxxxxxxxxxxxx>
Date: Tue Oct 10 13:56:29 2017 -0700
mac80211: Revert some of e8e4f5, fixes setting single rate in ath10k.
This lets us successfully set a single rate in ath10k again.
Signed-off-by: Ben Greear <greearb@xxxxxxxxxxxxxxx>
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 78cf453cda2c..3f248ad70805 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -2886,8 +2886,10 @@ static int ieee80211_set_bitrate_mask(struct wiphy *wiphy,
u32 basic_rates = sdata->vif.bss_conf.basic_rates;
enum nl80211_band band = sdata->vif.bss_conf.chandef.chan->band;
- if (!(mask->control[band].legacy & basic_rates))
- return -EINVAL;
+ if (!(mask->control[band].legacy & basic_rates)) {
+ pr_err("%s: WARNING: no legacy rates for band[%d] in set-bitrate-mask.\n",
+ sdata->dev->name, band);
+ }
}
if (ieee80211_hw_check(&local->hw, HAS_RATE_CONTROL)) {
Thanks,
Ben
And this:
"But actually the firmware has its own rate
adaptive mechanism, so mask out the other rates does not mean the rate
left will be chosen."
That's entirely your fault, not the fault of the API. If your firmware
doesn't listen to your driver, then that's either your firmware or
your driver's fault. If you set a mask that has exactly 1 bitrate in
it... well, that's exactly what you should tell your firmware to do.
Not, "use this 1 bitrate, but try something else if you feel like it."
Now, there are other problems, like the others that Ben mentioned: the
rest of the mac80211 framework doesn't like it too much if you really
disable all but 1 rate (arguably a mac80211 bug -- but not a nl80211
bug); or maybe you want to differentiate management frames and data
frames (and that's not what the current
NL80211_CMD_SET_TX_BITRATE_MASK allows for).
I'm still not (personally) expecting that you *must* do this all via
the existing CMD_SET_TX_BITRATE_MASK, but to satisfy everyone involved
here, you at least need to be clear about why you aren't.
Regards,
Brian
--
Ben Greear <greearb@xxxxxxxxxxxxxxx>
Candela Technologies Inc http://www.candelatech.com
