Re: [PATCH v8 1/2] wifi: mwifiex: add host mlme for client mode

[Francesco Dolcini <francesco@dolcini.it>]

On Fri, Dec 22, 2023 at 11:21:22AM +0800, David Lin wrote:
> Add host based MLME to enable WPA3 functionalities in client mode.
> This feature required a firmware with the corresponding V2 Key API
> support. The feature (WPA3) is currently enabled and verified only
> on IW416. Also, verified no regression with change when host MLME
> is disabled.
>
> Signed-off-by: David Lin <yu-hao.lin@nxp.com>

Currently this do not apply cleanly on wireless-next/main git tree, it
should be rebased to that branch as v9 (the reason is your addition
of fw_ready_extra_delay from a previous, now merged, patch).

In general the patch looks good to me, however I am no expert on
wireless driver or the related linux subsystem. I just have a couple of
small comments that I would suggest address in v9 (given that you need
to do it as a minimum to rebase your code).


> ---
>  .../net/wireless/marvell/mwifiex/cfg80211.c   | 315 ++++++++++++++++++
>  drivers/net/wireless/marvell/mwifiex/cmdevt.c |  25 ++
>  drivers/net/wireless/marvell/mwifiex/decl.h   |  22 ++
>  drivers/net/wireless/marvell/mwifiex/fw.h     |  33 ++
>  drivers/net/wireless/marvell/mwifiex/init.c   |   6 +
>  drivers/net/wireless/marvell/mwifiex/join.c   |  66 +++-
>  drivers/net/wireless/marvell/mwifiex/main.c   |  54 +++
>  drivers/net/wireless/marvell/mwifiex/main.h   |  17 +
>  drivers/net/wireless/marvell/mwifiex/scan.c   |   6 +
>  drivers/net/wireless/marvell/mwifiex/sdio.c   |  13 +
>  drivers/net/wireless/marvell/mwifiex/sdio.h   |   2 +
>  .../net/wireless/marvell/mwifiex/sta_event.c  |  36 +-
>  .../net/wireless/marvell/mwifiex/sta_ioctl.c  |   3 +-
>  drivers/net/wireless/marvell/mwifiex/sta_tx.c |   9 +-
>  drivers/net/wireless/marvell/mwifiex/util.c   |  80 +++++
>  15 files changed, 673 insertions(+), 14 deletions(-)
> 
> diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
> index 7a15ea8072e6..3cee1b58465e 100644
> --- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c
> +++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
> @@ -4202,6 +4208,302 @@ mwifiex_cfg80211_change_station(struct wiphy *wiphy, struct net_device *dev,

...

> +static int
> +mwifiex_cfg80211_probe_client(struct wiphy *wiphy,
> +			      struct net_device *dev, const u8 *peer,
> +			      u64 *cookie)
> +{
> +	return -1;

See my following comment on this

> +}
> +
>  /* station cfg80211 operations */
>  static struct cfg80211_ops mwifiex_cfg80211_ops = {
>  	.add_virtual_intf = mwifiex_add_virtual_intf,
> @@ -4347,6 +4649,16 @@ int mwifiex_register_cfg80211(struct mwifiex_adapter *adapter)
>  			    "%s: creating new wiphy\n", __func__);
>  		return -ENOMEM;
>  	}
> +	if (adapter->host_mlme_enabled) {
> +		mwifiex_cfg80211_ops.auth = mwifiex_cfg80211_authenticate;
> +		mwifiex_cfg80211_ops.assoc = mwifiex_cfg80211_associate;
> +		mwifiex_cfg80211_ops.deauth = mwifiex_cfg80211_deauthenticate;
> +		mwifiex_cfg80211_ops.disassoc = mwifiex_cfg80211_disassociate;
> +		mwifiex_cfg80211_ops.disconnect = NULL;
> +		mwifiex_cfg80211_ops.connect = NULL;
> +		mwifiex_cfg80211_ops.probe_client =
> +			mwifiex_cfg80211_probe_client;

Can you omit this one? You should get `-EOPNOTSUPP` for free with probe_client
set to NULL. Am I wrong?

> +	}
>  	wiphy->max_scan_ssids = MWIFIEX_MAX_SSID_LIST_LENGTH;
>  	wiphy->max_scan_ie_len = MWIFIEX_MAX_VSIE_LEN;
>  	wiphy->mgmt_stypes = mwifiex_mgmt_stypes;

...

> diff --git a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
> index a2ad2b53f016..23639aacf092 100644
> --- a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
> +++ b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c
> @@ -136,6 +136,7 @@ int mwifiex_fill_new_bss_desc(struct mwifiex_private *priv,
>  	const struct cfg80211_bss_ies *ies;
>  
>  	rcu_read_lock();
> +	bss_desc->bss = bss;
what is this change for? I was not able to understand it, nor to find
any user of this bss parameter. This looks like an unrelated fix, but no
code seems to be affected.


With these 2 comments addressed, please feel free to add to v9

Reviewed-by: Francesco Dolcini <francesco.dolcini@toradex.com>


And thanks for this work!
Francesco