[PATCH] wifi: mt76: mt7915: split mcu chan_mib array up
AngeloGioacchino Del Regno
angelogioacchino.delregno at collabora.com
Mon Dec 5 06:29:01 PST 2022
Il 03/12/22 08:37, Ryder Lee ha scritto:
> The current flow confuses coverity check that leads to false reporting,
> so split the offs[] into two pieces according to chipset revision to
> silence coverity tool.
>
> Signed-off-by: Ryder Lee <ryder.lee at mediatek.com>
> ---
> .../net/wireless/mediatek/mt76/mt7915/mcu.c | 50 ++++++++++---------
> 1 file changed, 27 insertions(+), 23 deletions(-)
>
> diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
> index b2652de082ba..ca315af3905b 100644
> --- a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
> +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
> @@ -2974,38 +2974,42 @@ int mt7915_mcu_apply_tx_dpd(struct mt7915_phy *phy)
>
> int mt7915_mcu_get_chan_mib_info(struct mt7915_phy *phy, bool chan_switch)
> {
> - /* strict order */
> - static const u32 offs[] = {
> - MIB_NON_WIFI_TIME,
> - MIB_TX_TIME,
> - MIB_RX_TIME,
> - MIB_OBSS_AIRTIME,
> - MIB_TXOP_INIT_COUNT,
> - /* v2 */
> - MIB_NON_WIFI_TIME_V2,
> - MIB_TX_TIME_V2,
> - MIB_RX_TIME_V2,
> - MIB_OBSS_AIRTIME_V2
> - };
> struct mt76_channel_state *state = phy->mt76->chan_state;
> struct mt76_channel_state *state_ts = &phy->state_ts;
> struct mt7915_dev *dev = phy->dev;
> struct mt7915_mcu_mib *res, req[5];
> struct sk_buff *skb;
> - int i, ret, start = 0, ofs = 20;
> + static const u32 *offs;
> + int i, ret, len, offs_cc;
> u64 cc_tx;
>
A more readable option would be
/* strict order */
static const u32 chip_offs_v1[] = {
MIB_NON_WIFI_TIME,
MIB_TX_TIME,
MIB_RX_TIME,
MIB_OBSS_AIRTIME,
MIB_TXOP_INIT_COUNT,
};
static const u32 chip_offs_v2[] = {
MIB_NON_WIFI_TIME_V2,
MIB_TX_TIME_V2,
MIB_RX_TIME_V2,
MIB_OBSS_AIRTIME_V2
};
if (is_mt7915(&dev->mt76)) {
len = ARRAY_SIZE(chip_offs_v1);
offs = chip_offs_v1;
offs_cc = 20;
} else {
len = ARRAY_SIZE(chip_offs_v2);
offs = chip_offs_v2;
offs_cc = 20;
}
Regards,
Angelo
More information about the Linux-mediatek
mailing list