[PATCH 2/2] wifi: ath10k: report WoW wakeups to the PM core

From: Michael Auer

Date: Fri Nov 20 2026 - 15:27:43 EST


ath10k does not report a WoW wakeup to the PM core, so the wakeup
source of the device never counts it, and user space cannot tell a
network wakeup from other resumes. Plasma's powerdevil, for example,
checks the wakeup_count of each wakeup source after resume. It keeps
the display off after a network wakeup and turns it on otherwise.

Call pm_wakeup_event() when the firmware reports a wake reason other
than WOW_REASON_UNSPECIFIED, which it uses when the host resumed for
another reason. iwlwifi does the same since commit 06ae2ad413d2
("iwlwifi: mvm: report wakeup for wowlan").

On a OnePlus 6 the wakeup source of the WLAN device now counts WoW
wakeups, and powerdevil keeps the display off for them.

Tested-on: WCN3990 hw1.0 SNOC WLAN.HL.2.0.c8-00050-QCAHLSWMTPLZ-1

Signed-off-by: Michael Auer <mick@xxxxxxxx>
---
drivers/net/wireless/ath/ath10k/wmi.c | 4 ++++
1 file changed, 4 insertions(+)

diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index eca6bb189577..c171cd96ac6f 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -4675,6 +4675,10 @@ void ath10k_wmi_event_wow_wakeup_host(struct ath10k *ar, struct sk_buff *skb)

ath10k_dbg(ar, ATH10K_DBG_WMI, "wow wakeup host reason %d %s\n",
ev.wake_reason, wow_reason(ev.wake_reason));
+
+ /* UNSPECIFIED: the host resumed for another reason */
+ if (ev.wake_reason != WOW_REASON_UNSPECIFIED)
+ pm_wakeup_event(ar->dev, 0);
}

void ath10k_wmi_event_dcs_interference(struct ath10k *ar, struct sk_buff *skb)
--
2.43.0