On Wed, Jun 26, 2019 at 11:24:31AM +0200, Sven Eckelmann wrote:
> Btw. thanks for fixing all the ieee80211 typos. Am not 100% sure where
> following changes came from:
>
> @@ -798,8 +799,9 @@ int hostapd_start_dfs_cac(struct hostapd_iface *iface,
> struct hostapd_data *hapd = iface->bss[0];
> struct hostapd_freq_params data;
> int res;
> + struct hostapd_hw_modes *cmode = iface->current_mode;
>
> - if (!hapd->driver || !hapd->driver->start_dfs_cac)
> + if (!hapd->driver || !hapd->driver->start_dfs_cac || !cmode)
That !cmode part is just for completeness since iface->current_mode is
not guaranteed to be non-NULL in general case (it might be here in
practice, though).
> @@ -812,8 +814,8 @@ int hostapd_start_dfs_cac(struct hostapd_iface *iface,
> vht_enabled, he_enabled, sec_channel_offset,
> oper_chwidth, center_segment0,
> center_segment1,
> - iface->current_mode->vht_capab,
> - &iface->current_mode->he_capab[IEEE80211_MODE_AP])) {
> + cmode->vht_capab,
> + &cmode->he_capab[IEEE80211_MODE_AP])) {
But the main reason for introducing that local cmode variable was to
keep this part reasonable.. There was no way of making the new version
fit in 80 characters without very ugly line wrapping without it.
> Seems like this was important to you to get this changed. Just find it odd
> that I only knew about this because I noticed that my patch was silently
> modified. And then I find something (the "if") in it which is not self
> explanatory and might be an actual bug fix for a different problem which is
> now hidden in this larger patch.
It was a coding style change and not a bug fix for any known issue.
--
Jouni Malinen PGP id EFC895FA
_______________________________________________
Hostap mailing list
Hostap@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/hostap
