Re: [RFC v3 6/7] mac80211: refactor set_channel_type
From: Eliad Peller <hidden>
Date: 2012-06-26 14:04:56
hi Michal, On Tue, Jun 26, 2012 at 3:37 PM, Michal Kazior [off-list ref] wrote:
From: Kazior Michal <redacted> Split functionality for further reuse. Will prevent code duplication when channel context channel_type merging is introduced. Signed-off-by: Michal Kazior <redacted> ---
[...]
+static bool
+ieee80211_channel_types_are_compatible(enum nl80211_channel_type chantype1,
+ enum nl80211_channel_type chantype2,
+ enum nl80211_channel_type *compat)
+{
+ switch (chantype1) {
case NL80211_CHAN_NO_HT:
case NL80211_CHAN_HT20:
/*
* allow any change that doesn't go to no-HT
* (if it already is no-HT no change is needed)
*/
- if (chantype == NL80211_CHAN_NO_HT)
+ if (chantype2 == NL80211_CHAN_NO_HT)
break;
- superchan = chantype;
+ *compat = chantype2;
break;
case NL80211_CHAN_HT40PLUS:
case NL80211_CHAN_HT40MINUS:
+ *compat = chantype1;this should be outside of switch, as it applies to the first cases as well (chantype1=ht20, chantype2=no_ht will return the original *compat value (no_ht) instead of ht20). Eliad.