From: Thomas Pedersen <thomas@adapt-ip.com>
To: Johannes Berg <johannes@sipsolutions.net>
Cc: linux-wireless <linux-wireless@vger.kernel.org>,
Thomas Pedersen <thomas@adapt-ip.com>
Subject: [PATCH v2 03/22] cfg80211: regulatory: handle S1G channels
Date: Mon, 31 Aug 2020 13:55:41 -0700 [thread overview]
Message-ID: <20200831205600.21058-4-thomas@adapt-ip.com> (raw)
In-Reply-To: <20200831205600.21058-1-thomas@adapt-ip.com>
S1G channels have a minimum bandwidth of 1Mhz, and there
is a 1:1 mapping of allowed bandwidth to channel number.
Signed-off-by: Thomas Pedersen <thomas@adapt-ip.com>
---
v2:
- drop the freq_reg_info() interface changes and move the
check for S1G band inside. Fixes a driver compile
error.
- fix iterating past bws[] in __freq_reg_info() by
setting initial element to 0.
Reported-by: kernel test robot <lkp@intel.com>
---
net/wireless/reg.c | 70 ++++++++++++++++++++++++++++++++++++++--------
1 file changed, 58 insertions(+), 12 deletions(-)
diff --git a/net/wireless/reg.c b/net/wireless/reg.c
index 0ab7808fcec8..be6f54b70ad3 100644
--- a/net/wireless/reg.c
+++ b/net/wireless/reg.c
@@ -1617,9 +1617,11 @@ __freq_reg_info(struct wiphy *wiphy, u32 center_freq, u32 min_bw)
{
const struct ieee80211_regdomain *regd = reg_get_regdomain(wiphy);
const struct ieee80211_reg_rule *reg_rule = NULL;
+ const u32 bws[] = {0, 1, 2, 4, 5, 8, 10, 16, 20};
+ int i = sizeof(bws) / sizeof(u32) - 1;
u32 bw;
- for (bw = MHZ_TO_KHZ(20); bw >= min_bw; bw = bw / 2) {
+ for (bw = MHZ_TO_KHZ(bws[i]); bw >= min_bw; bw = MHZ_TO_KHZ(bws[i--])) {
reg_rule = freq_reg_info_regd(center_freq, regd, bw);
if (!IS_ERR(reg_rule))
return reg_rule;
@@ -1631,7 +1633,9 @@ __freq_reg_info(struct wiphy *wiphy, u32 center_freq, u32 min_bw)
const struct ieee80211_reg_rule *freq_reg_info(struct wiphy *wiphy,
u32 center_freq)
{
- return __freq_reg_info(wiphy, center_freq, MHZ_TO_KHZ(20));
+ u32 min_bw = center_freq < MHZ_TO_KHZ(1000) ? 1 : 20;
+
+ return __freq_reg_info(wiphy, center_freq, MHZ_TO_KHZ(min_bw));
}
EXPORT_SYMBOL(freq_reg_info);
@@ -1659,6 +1663,7 @@ static uint32_t reg_rule_to_chan_bw_flags(const struct ieee80211_regdomain *regd
{
const struct ieee80211_freq_range *freq_range = NULL;
u32 max_bandwidth_khz, center_freq_khz, bw_flags = 0;
+ bool is_s1g = chan->band == NL80211_BAND_S1GHZ;
freq_range = ®_rule->freq_range;
@@ -1678,16 +1683,57 @@ static uint32_t reg_rule_to_chan_bw_flags(const struct ieee80211_regdomain *regd
MHZ_TO_KHZ(20)))
bw_flags |= IEEE80211_CHAN_NO_20MHZ;
- if (max_bandwidth_khz < MHZ_TO_KHZ(10))
- bw_flags |= IEEE80211_CHAN_NO_10MHZ;
- if (max_bandwidth_khz < MHZ_TO_KHZ(20))
- bw_flags |= IEEE80211_CHAN_NO_20MHZ;
- if (max_bandwidth_khz < MHZ_TO_KHZ(40))
- bw_flags |= IEEE80211_CHAN_NO_HT40;
- if (max_bandwidth_khz < MHZ_TO_KHZ(80))
- bw_flags |= IEEE80211_CHAN_NO_80MHZ;
- if (max_bandwidth_khz < MHZ_TO_KHZ(160))
- bw_flags |= IEEE80211_CHAN_NO_160MHZ;
+ if (is_s1g) {
+ /* S1G is strict about non overlapping channels. We can
+ * calculate which bandwidth is allowed per channel by finding
+ * the largest bandwidth which cleanly divides the freq_range.
+ */
+ int edge_offset;
+ int ch_bw = max_bandwidth_khz;
+
+ while (ch_bw) {
+ edge_offset = (center_freq_khz - ch_bw / 2) -
+ freq_range->start_freq_khz;
+ if (edge_offset % ch_bw == 0) {
+ switch (KHZ_TO_MHZ(ch_bw)) {
+ case 1:
+ bw_flags |= IEEE80211_CHAN_1MHZ;
+ break;
+ case 2:
+ bw_flags |= IEEE80211_CHAN_2MHZ;
+ break;
+ case 4:
+ bw_flags |= IEEE80211_CHAN_4MHZ;
+ break;
+ case 8:
+ bw_flags |= IEEE80211_CHAN_8MHZ;
+ break;
+ case 16:
+ bw_flags |= IEEE80211_CHAN_16MHZ;
+ break;
+ default:
+ /* If we got here, no bandwidths fit on
+ * this frequency, ie. band edge.
+ */
+ bw_flags |= IEEE80211_CHAN_DISABLED;
+ break;
+ }
+ break;
+ }
+ ch_bw /= 2;
+ }
+ } else {
+ if (max_bandwidth_khz < MHZ_TO_KHZ(10))
+ bw_flags |= IEEE80211_CHAN_NO_10MHZ;
+ if (max_bandwidth_khz < MHZ_TO_KHZ(20))
+ bw_flags |= IEEE80211_CHAN_NO_20MHZ;
+ if (max_bandwidth_khz < MHZ_TO_KHZ(40))
+ bw_flags |= IEEE80211_CHAN_NO_HT40;
+ if (max_bandwidth_khz < MHZ_TO_KHZ(80))
+ bw_flags |= IEEE80211_CHAN_NO_80MHZ;
+ if (max_bandwidth_khz < MHZ_TO_KHZ(160))
+ bw_flags |= IEEE80211_CHAN_NO_160MHZ;
+ }
return bw_flags;
}
--
2.20.1
next prev parent reply other threads:[~2020-08-31 20:56 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-08-31 20:55 [PATCH v2 00/22] add support for S1G association Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 01/22] ieee80211: redefine S1G bits with GENMASK Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 02/22] nl80211: advertise supported channel width in S1G Thomas Pedersen
2020-08-31 20:55 ` Thomas Pedersen [this message]
2020-09-05 18:38 ` [PATCH v2 03/22] cfg80211: regulatory: handle S1G channels Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 04/22] nl80211: correctly validate S1G beacon head Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 05/22] nl80211: support setting S1G channels Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 06/22] {cfg,mac}80211: get correct default channel width for S1G Thomas Pedersen
2020-09-18 10:38 ` Johannes Berg
2020-09-21 4:59 ` Thomas Pedersen
2020-09-21 7:01 ` Johannes Berg
2020-09-21 16:26 ` Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 07/22] mac80211: s1g: choose scanning width based on frequency Thomas Pedersen
2020-09-18 10:40 ` Johannes Berg
2020-09-21 5:06 ` Thomas Pedersen
2020-09-21 6:58 ` Johannes Berg
2020-08-31 20:55 ` [PATCH v2 08/22] nl80211: support S1G capabilities Thomas Pedersen
2020-09-18 10:41 ` Johannes Berg
2020-09-18 10:47 ` Johannes Berg
2020-09-21 4:34 ` Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 09/22] mac80211: support S1G STA capabilities Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 10/22] cfg80211: convert S1G beacon to scan results Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 11/22] cfg80211: parse S1G Operation element for BSS channel Thomas Pedersen
2020-09-18 10:45 ` Johannes Berg
2020-09-21 5:12 ` Thomas Pedersen
2020-09-21 6:54 ` Johannes Berg
2020-08-31 20:55 ` [PATCH v2 12/22] mac80211: convert S1G beacon to scan results Thomas Pedersen
2020-09-18 10:48 ` Johannes Berg
2020-09-21 5:45 ` Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 13/22] cfg80211: handle Association Response from S1G STA Thomas Pedersen
2020-09-18 10:50 ` Johannes Berg
2020-09-21 5:52 ` Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 14/22] mac80211: encode listen interval for S1G Thomas Pedersen
2020-09-18 10:51 ` Johannes Berg
2020-09-21 5:56 ` Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 15/22] mac80211: don't calculate duration " Thomas Pedersen
2020-09-18 10:52 ` Johannes Berg
2020-09-18 10:55 ` Johannes Berg
2020-09-21 6:03 ` Thomas Pedersen
2020-09-21 6:51 ` Johannes Berg
2020-08-31 20:55 ` [PATCH v2 16/22] mac80211: handle S1G low rates Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 17/22] mac80211: avoid rate init for S1G band Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 18/22] mac80211: receive and process S1G beacons Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 19/22] mac80211: support S1G association Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 20/22] nl80211: include frequency offset in survey info Thomas Pedersen
2020-08-31 20:55 ` [PATCH v2 21/22] mac80211_hwsim: indicate support for S1G Thomas Pedersen
2020-09-06 8:49 ` [mac80211_hwsim] 8cafe19852: hwsim.fst_ap_config_default.fail kernel test robot
2020-09-08 18:00 ` Thomas Pedersen
2020-09-08 18:41 ` Johannes Berg
2020-09-08 18:53 ` Thomas Pedersen
2020-08-31 20:56 ` [PATCH v2 22/22] mac80211_hwsim: fix TSF timestamp write to S1G beacon Thomas Pedersen
2020-09-06 16:04 ` [PATCH v2 00/22] add support for S1G association Johannes Berg
2020-09-08 18:30 ` Thomas Pedersen
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200831205600.21058-4-thomas@adapt-ip.com \
--to=thomas@adapt-ip.com \
--cc=johannes@sipsolutions.net \
--cc=linux-wireless@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).