On 5/4/22 07:46, Ben Greear wrote:
On 5/4/22 3:57 AM, Johannes Berg wrote:
On Fri, 2022-02-25 at 15:28 -0800, greearb@xxxxxxxxxxxxxxx wrote:
- if (mvmsta->avg_energy) {
- sinfo->signal_avg = -(s8)mvmsta->avg_energy;
+ if (iwl_mvm_has_new_rx_api(mvm)) { /* rxmq logic */
+ /* Grab chain signal avg, mac80211 cannot do it because
+ * this driver uses RSS. Grab signal_avg here too because firmware
+ * appears go not do DB summing and/or has other bugs. --Ben
+ */
That "--Ben" really seems inappropriate ... please take more care when
sending patches to the list.
I figure if I am submitting a patch that is blaming firmware, and I have no access
to firmware to actually back up my assertion, then I should put my name on it
so that folks looking at the driver don't think it is Intel disparaging it's own firmware.
Hello Johannes,
If there is no reasonable chance to make changes like this to iwlwifi (lockless
signal averaging is the unfixable part I think), let me know
and I'll consider the issue closed.
And, in that case, the mac80211 patch is un-needed since no driver that I know of
would need that code.
Thanks,
Ben
--
Ben Greear <greearb@xxxxxxxxxxxxxxx>
Candela Technologies Inc http://www.candelatech.com
