[PATCH v2] wifi: brcmfmac: cyw: set a unique cookie for mgmt tx

[PATCH v2] wifi: brcmfmac: cyw: set a unique cookie for mgmt tx

[Bogdan Nicolae]

Set cookie using an increasing atomic counter. Avoids mismatches of completion
events later in brcmf_notify_mgmt_tx_status, where packet_id != vif->mgmt_tx_id
is checked.

Signed-off-by: Bogdan Nicolae <bogdan.nicolae@gmail.com>
---
v2: split into separate patch and set cookie to atomic counter instead
of packet_id directly.

 drivers/net/wireless/broadcom/brcm80211/brcmfmac/cyw/core.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cyw/core.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cyw/core.c
index 2c59b5e57..ab4ff0cd6 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cyw/core.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cyw/core.c
@@ -23,6 +23,8 @@
 #define MGMT_AUTH_FRAME_DWELL_TIME	4000
 #define MGMT_AUTH_FRAME_WAIT_TIME	(MGMT_AUTH_FRAME_DWELL_TIME + 100)
 
+static atomic_t brcmf_cyw_mgmt_tx_id = ATOMIC_INIT(0);
+
 static int brcmf_cyw_set_sae_pwd(struct brcmf_if *ifp,
 				 struct cfg80211_crypto_settings *crypto)
 {
@@ -123,7 +125,7 @@ int brcmf_cyw_mgmt_tx(struct wiphy *wiphy, struct wireless_dev *wdev,
 	if (!ieee80211_is_auth(mgmt->frame_control))
 		return brcmf_cfg80211_mgmt_tx(wiphy, wdev, params, cookie);
 
-	*cookie = 0;
+	*cookie = atomic_inc_return(&brcmf_cyw_mgmt_tx_id);
 	vif = container_of(wdev, struct brcmf_cfg80211_vif, wdev);
 
 	reinit_completion(&vif->mgmt_tx);
-- 
2.55.0

Re: [PATCH v2] wifi: brcmfmac: cyw: set a unique cookie for mgmt tx

[Arend van Spriel]

On 25/07/2026 03:44, Bogdan Nicolae wrote:
> Set cookie using an increasing atomic counter. Avoids mismatches of completion
> events later in brcmf_notify_mgmt_tx_status, where packet_id != vif->mgmt_tx_id
> is checked.

Thanks for the split. I increased the scope and been looking in the 
cookie usage in the wireless subsystem. So I have a series lined up to 
change the cookie to an input parameter and have cfg80211 hand out the 
cookie value (it already has cfg80211_assign_cookie() function). This
will also cover zero cookie fix here. So I would suggest to drop this 
patch and only let the PMKID patch go in.

Regards,
Arend

Re: [PATCH v2] wifi: brcmfmac: cyw: set a unique cookie for mgmt tx

[Bogdan Nicolae]

Sure, it sounds good. Thanks Arend.

Bogdan

On Sun, Jul 26, 2026 at 2:21 PM Arend van Spriel
<arend.vanspriel@broadcom.com> wrote:
>
> On 25/07/2026 03:44, Bogdan Nicolae wrote:
> > Set cookie using an increasing atomic counter. Avoids mismatches of completion
> > events later in brcmf_notify_mgmt_tx_status, where packet_id != vif->mgmt_tx_id
> > is checked.
>
> Thanks for the split. I increased the scope and been looking in the
> cookie usage in the wireless subsystem. So I have a series lined up to
> change the cookie to an input parameter and have cfg80211 hand out the
> cookie value (it already has cfg80211_assign_cookie() function). This
> will also cover zero cookie fix here. So I would suggest to drop this
> patch and only let the PMKID patch go in.
>
> Regards,
> Arend

Re: [PATCH v2] wifi: brcmfmac: cyw: set a unique cookie for mgmt tx

[Arend van Spriel]

On Fri, 24 Jul 2026 20:44:11 -0500, Bogdan Nicolae wrote:
> Set cookie using an increasing atomic counter. Avoids mismatches of completion
> events later in brcmf_notify_mgmt_tx_status, where packet_id != vif->mgmt_tx_id
> is checked.
>
> Signed-off-by: Bogdan Nicolae <bogdan.nicolae@gmail.com>
> ---
>  drivers/net/wireless/broadcom/brcm80211/brcmfmac/cyw/core.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)

As agreed I will mark this as superseded.

Regards,
Arend