diff options
| author | Sven Eckelmann <sven@narfation.org> | 2019-02-25 20:42:28 +0100 |
|---|---|---|
| committer | Hauke Mehrtens <hauke@hauke-m.de> | 2019-02-26 23:20:04 +0100 |
| commit | ba249bc955419f3d0a943ea3a111ece1624a9291 (patch) | |
| tree | bfc65099ff31b88fae4b5189069c4217c3ad880d /target/linux/gemini | |
| parent | 4beed12d85947f6eb7c935d2f9a90f1999379977 (diff) | |
| download | upstream-ba249bc955419f3d0a943ea3a111ece1624a9291.tar.gz upstream-ba249bc955419f3d0a943ea3a111ece1624a9291.tar.bz2 upstream-ba249bc955419f3d0a943ea3a111ece1624a9291.zip | |
ath10k-ct: fix incorrect multicast/broadcast rate setting
If no mcast_rate is set for the wifi-iface then there is no rate_idx (0)
set for the bss. This can break for example 5GHz meshpoint interfaces
because 0 maps to a CCK rate (11Mbit/s).
It must also be avoided that the ath10k-ct internal state for the rates is
not synced with the mac80211 rates state. Otherwise, the user specified
rate (e.g. a wifi-iface mcast_rate for a meshpoint interface) will only be
set on startup. And a short while after that, ath10k-ct specific code in
ath10k_check_apply_special_rates is missing a valid rate in its own
structures and is then recalculating a new default rate. This default rate
is in most situations not the requested rate.
Fixes: 4df3c71cd4c5 ("ath10k-ct: Update to 2018-12-11 and use version based on 4.19")
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Diffstat (limited to 'target/linux/gemini')
0 files changed, 0 insertions, 0 deletions