This reverts commit f17f4864504d754bcbf31e4c89412cdf9946c409.
The reverted commit significantly decreases performance when running
a test where two MT7915 radios have 16 station vdevs each, configured
for AC mode, and transmitting UDP traffic to AP.
Reported-by: Carson Vandegriffe <[email protected]>
Signed-off-by: Ben Greear <[email protected]>
According to the tx-free-done documentation, the DW4 can be repeated,
so have to be more careful about how we test for walking off the
end of the array.
Signed-off-by: Ben Greear <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
If the nic fails to start, it is possible that the
reset_work has already been scheduled. Ensure the
work item is canceled so we do not have use-after-free
crash in case cleanup is called before the work item
is executed.
This fixes crash on my x86_64 apu2 when mt7921k radio
fails to work. Radio still fails, but OS does not
crash.
Signed-off-by: Ben Greear <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
This adds the new mib counters from last patch into ethtool
stats.
Signed-off-by: Ben Greear <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Add register definitions and read & accumulate them in
the mib polling logic. Note that some registers should not
be read since firmware is already reading them. If driver
reads those, they will be cleared-on-read, and so the firmware
stats will be incorrect.
For these 'do-not-read' stats, add them to the registers definition
so that other developers can be aware of these constraints, but do
not actually read them in the driver.
Signed-off-by: Ben Greear <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
These counters are clear-on-read, so we need to accumulate
them in the update_stats poll logic, and read the accumulated
values instead of directly doing register reads when reporting
to debugfs and ethtool stats.
Signed-off-by: Ben Greear <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
The tx status callback is not called for every frame, so
those specific counters under-count, but at least they give
some idea of what is going on.
Signed-off-by: Ben Greear <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Add tx-mode (ofdma, ht, vht, HE) histogram,
tx-ru-idx histogram, and tx-bandwidth histogram.
Also add tx attempts and tx success counters.
All of this is per-station.
Signed-off-by: Ben Greear <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
This exposes some tx-path stats to the ethtool API, so that
ethtool -S wlanX provides some more useful info.
Signed-off-by: Ben Greear <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
Without this change, garbage is seen in the hwmon name and sensors output
for mt7915 is garbled. It appears that the hwmon logic does not make a
copy of the incoming string, but instead just copies a char* and expects
it to never go away.
Fixes: d6938251bb5b ("mt76: mt7915: add thermal sensor device support")
Signed-off-by: Ben Greear <[email protected]>
Signed-off-by: Ryder Lee <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
At 160, this chip can only do 2x2 NSS. Fix the features
accordingly, verified it shows up properly in iw phy foo info
now.
Signed-off-by: Ben Greear <[email protected]>
Signed-off-by: Felix Fietkau <[email protected]>
