On 1/19/24 20:19, Johannes Berg wrote:
I was trying to let you come up with a patch for learning, but I think
at this point just making one illustrates better what I'm thinking ...
Here's one, but I haven't even compiled this:
Hmm... this:
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -701,7 +701,9 @@ ieee80211_tx_h_rate_ctrl(struct ieee80211_tx_data *tx)
txrc.bss_conf = &tx->sdata->vif.bss_conf;
txrc.skb = tx->skb;
txrc.reported_rate.idx = -1;
- txrc.rate_idx_mask = tx->sdata->rc_rateidx_mask[info->band];
+
+ if (!(info->control.flags & IEEE80211_TX_CTRL_SCAN_TX))
+ txrc.rate_idx_mask = tx->sdata->rc_rateidx_mask[info->band];
if (tx->sdata->rc_has_mcs_mask[info->band])
txrc.rate_idx_mcs_mask =
leaves 'rate_idx_mask' unset (i.e. zero) even for band 0 and then
causes the same WARN_ONCE() even earlier (before switching to band 1).
Dmitry
