Since commit 9d5804662ce1 ("ath10k: retrieve MAC address from system firmware if provided"), we support pulling our MAC address from system firmware. Along with this, we should admit that it's possible for system firmware to be screwed up as well, omitting the MAC address there.
Provide a fallback, such that we choose a randomly-generated MAC address, with the local assignment bit set. This is not a preferred mode of operation and really should only serve the purpose of keeping the system chugging along while someone figures out what's wrong with the system. Grepping through other wireless drivers shows very similar fallback logic though, so why not? git grep eth_random_addr drivers/net/wireless/ Signed-off-by: Brian Norris <briannor...@chromium.org> --- drivers/net/wireless/ath/ath10k/mac.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c index 97548f96a2f7..484f9c74ce46 100644 --- a/drivers/net/wireless/ath/ath10k/mac.c +++ b/drivers/net/wireless/ath/ath10k/mac.c @@ -8325,6 +8325,10 @@ int ath10k_mac_register(struct ath10k *ar) void *channels; int ret; + if (!is_valid_ether_addr(ar->mac_addr)) { + ath10k_warn(ar, "invalid MAC address; choosing random\n"); + eth_random_addr(ar->mac_addr); + } SET_IEEE80211_PERM_ADDR(ar->hw, ar->mac_addr); SET_IEEE80211_DEV(ar->hw, ar->dev); -- 2.19.0.rc2.392.g5ba43deb5a-goog _______________________________________________ ath10k mailing list ath10k@lists.infradead.org http://lists.infradead.org/mailman/listinfo/ath10k