2019-02-13 14:06:46

by Adrian Bunk

[permalink] [raw]
Subject: [PATCH net] mwifiex: Fix NL80211_TX_POWER_LIMITED

NL80211_TX_POWER_LIMITED was treated as NL80211_TX_POWER_AUTOMATIC,
which is the opposite of what should happen and can cause nasty
regulatory problems.

if/else converted to a switch without default to make gcc warn
on unhandled enum values.

Signed-off-by: Adrian Bunk <[email protected]>

---

I tried to follow the conventions used inside this driver for
variable types, dbm_min is u16 like dbm.
---
drivers/net/wireless/marvell/mwifiex/cfg80211.c | 13 +++++++++++--
drivers/net/wireless/marvell/mwifiex/ioctl.h | 1 +
drivers/net/wireless/marvell/mwifiex/sta_ioctl.c | 11 +++++++----
3 files changed, 19 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
index 1467af22e394..9ef0fc108641 100644
--- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c
+++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
@@ -376,11 +376,20 @@ mwifiex_cfg80211_set_tx_power(struct wiphy *wiphy,
struct mwifiex_power_cfg power_cfg;
int dbm = MBM_TO_DBM(mbm);

- if (type == NL80211_TX_POWER_FIXED) {
+ switch (type) {
+ case NL80211_TX_POWER_FIXED:
power_cfg.is_power_auto = 0;
+ power_cfg.is_power_fixed = 1;
power_cfg.power_level = dbm;
- } else {
+ break;
+ case NL80211_TX_POWER_LIMITED:
+ power_cfg.is_power_auto = 0;
+ power_cfg.is_power_fixed = 0;
+ power_cfg.power_level = dbm;
+ break;
+ case NL80211_TX_POWER_AUTOMATIC:
power_cfg.is_power_auto = 1;
+ break;
}

priv = mwifiex_get_priv(adapter, MWIFIEX_BSS_ROLE_ANY);
diff --git a/drivers/net/wireless/marvell/mwifiex/ioctl.h b/drivers/net/wireless/marvell/mwifiex/ioctl.h
index 48e154e1865d..0dd592ea6e83 100644
--- a/drivers/net/wireless/marvell/mwifiex/ioctl.h
+++ b/drivers/net/wireless/marvell/mwifiex/ioctl.h
@@ -267,6 +267,7 @@ struct mwifiex_ds_encrypt_key {

struct mwifiex_power_cfg {
u32 is_power_auto;
+ u32 is_power_fixed;
u32 power_level;
};

diff --git a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
index b454b5f85503..ebc0e41e5d3b 100644
--- a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
+++ b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
@@ -688,6 +688,9 @@ int mwifiex_set_tx_power(struct mwifiex_private *priv,
txp_cfg = (struct host_cmd_ds_txpwr_cfg *) buf;
txp_cfg->action = cpu_to_le16(HostCmd_ACT_GEN_SET);
if (!power_cfg->is_power_auto) {
+ u16 dbm_min = power_cfg->is_power_fixed ?
+ dbm : priv->min_tx_power_level;
+
txp_cfg->mode = cpu_to_le32(1);
pg_tlv = (struct mwifiex_types_power_group *)
(buf + sizeof(struct host_cmd_ds_txpwr_cfg));
@@ -702,7 +705,7 @@ int mwifiex_set_tx_power(struct mwifiex_private *priv,
pg->last_rate_code = 0x03;
pg->modulation_class = MOD_CLASS_HR_DSSS;
pg->power_step = 0;
- pg->power_min = (s8) dbm;
+ pg->power_min = (s8) dbm_min;
pg->power_max = (s8) dbm;
pg++;
/* Power group for modulation class OFDM */
@@ -710,7 +713,7 @@ int mwifiex_set_tx_power(struct mwifiex_private *priv,
pg->last_rate_code = 0x07;
pg->modulation_class = MOD_CLASS_OFDM;
pg->power_step = 0;
- pg->power_min = (s8) dbm;
+ pg->power_min = (s8) dbm_min;
pg->power_max = (s8) dbm;
pg++;
/* Power group for modulation class HTBW20 */
@@ -718,7 +721,7 @@ int mwifiex_set_tx_power(struct mwifiex_private *priv,
pg->last_rate_code = 0x20;
pg->modulation_class = MOD_CLASS_HT;
pg->power_step = 0;
- pg->power_min = (s8) dbm;
+ pg->power_min = (s8) dbm_min;
pg->power_max = (s8) dbm;
pg->ht_bandwidth = HT_BW_20;
pg++;
@@ -727,7 +730,7 @@ int mwifiex_set_tx_power(struct mwifiex_private *priv,
pg->last_rate_code = 0x20;
pg->modulation_class = MOD_CLASS_HT;
pg->power_step = 0;
- pg->power_min = (s8) dbm;
+ pg->power_min = (s8) dbm_min;
pg->power_max = (s8) dbm;
pg->ht_bandwidth = HT_BW_40;
}
--
2.11.0



2019-02-19 15:15:39

by Kalle Valo

[permalink] [raw]
Subject: Re: [PATCH net] mwifiex: Fix NL80211_TX_POWER_LIMITED

Adrian Bunk <[email protected]> wrote:

> NL80211_TX_POWER_LIMITED was treated as NL80211_TX_POWER_AUTOMATIC,
> which is the opposite of what should happen and can cause nasty
> regulatory problems.
>
> if/else converted to a switch without default to make gcc warn
> on unhandled enum values.
>
> Signed-off-by: Adrian Bunk <[email protected]>

Patch applied to wireless-drivers-next.git, thanks.

65a576e27309 mwifiex: Fix NL80211_TX_POWER_LIMITED

--
https://patchwork.kernel.org/patch/10810011/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches