| Message ID | 20200831205600.21058-16-thomas@adapt-ip.com (mailing list archive) |
|---|---|
| State | Superseded |
| Delegated to: | Johannes Berg |
| Headers | show |
| Series | add support for S1G association | expand |
On Mon, 2020-08-31 at 13:55 -0700, Thomas Pedersen wrote: > For now just skip the duration calculation for frames > transmitted on the S1G band and avoid a warning. > > Signed-off-by: Thomas Pedersen <thomas@adapt-ip.com> > --- > net/mac80211/tx.c | 4 ++++ > 1 file changed, 4 insertions(+) > > diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c > index d2136007e2eb..bef19616c5f0 100644 > --- a/net/mac80211/tx.c > +++ b/net/mac80211/tx.c > @@ -82,6 +82,10 @@ static __le16 ieee80211_duration(struct ieee80211_tx_data *tx, > > erp = txrate->flags & IEEE80211_RATE_ERP_G; > > + /* TODO */ > + if (sband->band == NL80211_BAND_S1GHZ) > + return 0; I'm not even sure I'd leave a TODO there ... just say /* device is expected to do this */ or so? johannes
Oops, sorry. I just realized I've been replying to v2, while *looking* at v3 in patchwork ... I'll continue on v3, but this seems mostly the same. johannes
On 2020-09-18 03:52, Johannes Berg wrote: > On Mon, 2020-08-31 at 13:55 -0700, Thomas Pedersen wrote: >> For now just skip the duration calculation for frames >> transmitted on the S1G band and avoid a warning. >> >> Signed-off-by: Thomas Pedersen <thomas@adapt-ip.com> >> --- >> net/mac80211/tx.c | 4 ++++ >> 1 file changed, 4 insertions(+) >> >> diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c >> index d2136007e2eb..bef19616c5f0 100644 >> --- a/net/mac80211/tx.c >> +++ b/net/mac80211/tx.c >> @@ -82,6 +82,10 @@ static __le16 ieee80211_duration(struct >> ieee80211_tx_data *tx, >> >> erp = txrate->flags & IEEE80211_RATE_ERP_G; >> >> + /* TODO */ >> + if (sband->band == NL80211_BAND_S1GHZ) >> + return 0; > > > I'm not even sure I'd leave a TODO there ... just say > > /* device is expected to do this */ Eventually it would be nice if mac80211 could calculate duration for S1G? Do you know why it doesn't for HT/VHT?
On Sun, 2020-09-20 at 23:03 -0700, Thomas Pedersen wrote: > > > I'm not even sure I'd leave a TODO there ... just say > > > > /* device is expected to do this */ > > Eventually it would be nice if mac80211 could calculate duration for > S1G? Do you know why it doesn't for HT/VHT? I'm not sure it matters. It almost doesn't even matter for legacy (CCK/OFDM) because devices practically always do it anyway? And for HT/VHT it's just quite a bit more complicated, so we never bothered, since it wasn't necessary. It couldn't even really be done since there are A-MPDUs and other things happening at a lower layer. johannes
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index d2136007e2eb..bef19616c5f0 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -82,6 +82,10 @@ static __le16 ieee80211_duration(struct ieee80211_tx_data *tx, erp = txrate->flags & IEEE80211_RATE_ERP_G; + /* TODO */ + if (sband->band == NL80211_BAND_S1GHZ) + return 0; + /* * data and mgmt (except PS Poll): * - during CFP: 32768
For now just skip the duration calculation for frames transmitted on the S1G band and avoid a warning. Signed-off-by: Thomas Pedersen <thomas@adapt-ip.com> --- net/mac80211/tx.c | 4 ++++ 1 file changed, 4 insertions(+)