public inbox for ath11k@lists.infradead.org
 help / color / mirror / Atom feed
* ath11k WiFi bringup on Milos SoC
@ 2025-07-18 13:23 Luca Weiss
  2026-01-16 14:59 ` Luca Weiss
  0 siblings, 1 reply; 3+ messages in thread
From: Luca Weiss @ 2025-07-18 13:23 UTC (permalink / raw)
  To: ath11k, linux-arm-msm

Hi,

I'm trying to bring up WiFi on the Milos SoC (on my device: SM7635)
which is paired with a WCN6755.

There's also the WPSS remoteproc on this SoC, similar to SC7280/Kodiak.

Using the qcom,wcn6750-wifi compatible in dt makes the wpss firmware
crash soon after boot with the following errors:

[ 1263.547381] ath11k 17110040.wifi: wcn6750 hw1.0
[ 1263.577281] remoteproc remoteproc2: powering up wpss
[ 1263.814337] remoteproc remoteproc2: Booting fw image qcom/milos/fairphone/fp6/wpss.mbn, size 8229144
[ 1263.817861] PDM: no support for the platform, userspace daemon might be required.
[ 1264.523796] remoteproc remoteproc2: remote processor wpss is now up
[ 1264.773557] ath11k 17110040.wifi: chip_id 0x1 chip_family 0xb board_id 0xff soc_id 0x40140120
[ 1264.774484] ath11k 17110040.wifi: fw_version 0x3014ffff fw_build_timestamp 2025-04-09 08:57 fw_build_id WLAN.MSL.3.0.1-00328.3-QCAMSLSWPLZ-1.105949.2
[ 1272.362281] ath11k 17110040.wifi: htt event 48 not handled
[ 1273.499832] qcom_q6v5_pas 8a00000.remoteproc: fatal error received: cmnos_assert.c:400:Asserted in whal_wmac_recipes.c:whal_wmac_transition_to_fw_control:1734
[ 1273.500231] remoteproc remoteproc2: crash detected in wpss: type fatal error
[ 1273.501114] remoteproc remoteproc2: handling crash #1 in wpss
[ 1273.501153] remoteproc remoteproc2: recovering wpss

That "htt event 48" should be HTT_T2H_PPDU_ID_FMT_IND, I have briefly
looked in the WiFi driver from the downstream 6.1 kernel and while I
might definitely be wrong here, I'm guessing it's not related to the
crash.

Anyone have an idea how to debug this? Unfortunately my AMSS sources do
not include any sources for WiFi, it's just prebuilts.

Regards
Luca


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: ath11k WiFi bringup on Milos SoC
@ 2026-01-12 13:44 Alexander Koskovich
  0 siblings, 0 replies; 3+ messages in thread
From: Alexander Koskovich @ 2026-01-12 13:44 UTC (permalink / raw)
  To: ath11k@lists.infradead.org, linux-arm-msm@vger.kernel.org; +Cc: Luca Weiss

In-Reply-To: <DBF7OWAWQ94M.FSCP4DPF8ZJY@fairphone.com>

Running into the same firmware assert on another Milos device w/WCN6755 that is running WPSS version WLAN.MSL.3.0.1-00156-QCAMSLSWPLZ-1.122511.2.127363.2.

Would anyone from QCOM be able to provide more context on what situations firmware would assert whal_wmac_transition_to_fw_control, or is it a catch all?

Thanks,
Alex


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: ath11k WiFi bringup on Milos SoC
  2025-07-18 13:23 ath11k WiFi bringup on Milos SoC Luca Weiss
@ 2026-01-16 14:59 ` Luca Weiss
  0 siblings, 0 replies; 3+ messages in thread
From: Luca Weiss @ 2026-01-16 14:59 UTC (permalink / raw)
  To: Luca Weiss, ath11k, linux-arm-msm

On Fri Jul 18, 2025 at 3:23 PM CEST, Luca Weiss wrote:
> Hi,
>
> I'm trying to bring up WiFi on the Milos SoC (on my device: SM7635)
> which is paired with a WCN6755.
>
> There's also the WPSS remoteproc on this SoC, similar to SC7280/Kodiak.
>
> Using the qcom,wcn6750-wifi compatible in dt makes the wpss firmware
> crash soon after boot with the following errors:
>
> [ 1263.547381] ath11k 17110040.wifi: wcn6750 hw1.0
> [ 1263.577281] remoteproc remoteproc2: powering up wpss
> [ 1263.814337] remoteproc remoteproc2: Booting fw image qcom/milos/fairphone/fp6/wpss.mbn, size 8229144
> [ 1263.817861] PDM: no support for the platform, userspace daemon might be required.
> [ 1264.523796] remoteproc remoteproc2: remote processor wpss is now up
> [ 1264.773557] ath11k 17110040.wifi: chip_id 0x1 chip_family 0xb board_id 0xff soc_id 0x40140120
> [ 1264.774484] ath11k 17110040.wifi: fw_version 0x3014ffff fw_build_timestamp 2025-04-09 08:57 fw_build_id WLAN.MSL.3.0.1-00328.3-QCAMSLSWPLZ-1.105949.2
> [ 1272.362281] ath11k 17110040.wifi: htt event 48 not handled
> [ 1273.499832] qcom_q6v5_pas 8a00000.remoteproc: fatal error received: cmnos_assert.c:400:Asserted in whal_wmac_recipes.c:whal_wmac_transition_to_fw_control:1734
> [ 1273.500231] remoteproc remoteproc2: crash detected in wpss: type fatal error
> [ 1273.501114] remoteproc remoteproc2: handling crash #1 in wpss
> [ 1273.501153] remoteproc remoteproc2: recovering wpss

To conclude this, this was missing a pinctrl for gpio5 of pmk8550 which
provides sleep clock to the LF_CLK_IN_GPIO pin of the WCN6755 PMU.

Downstream does define this but not in any way connected to anything
related to the WCN6755.

    &pmk8550_gpios {
    	pinctrl-0 = <&alt_sleep_clk_default>;
    	pinctrl-names = "default";

    	alt_sleep_clk {
    		alt_sleep_clk_default: alt_sleep_clk_default {
    			pins = "gpio5";
    			function = "func1";
    			input-disable;
    			output-enable;
    			bias-disable;
    			power-source = <0>;
    		};
    	};
    };

This is the upstream patch with the pinctrl defined:
https://lore.kernel.org/linux-arm-msm/20260116-milos-fp6-bt-wifi-v1-5-27b4fbb77e9c@fairphone.com/

With this added, the WPSS doesn't crash anymore and WiFi starts working.

Regards
Luca

>
> That "htt event 48" should be HTT_T2H_PPDU_ID_FMT_IND, I have briefly
> looked in the WiFi driver from the downstream 6.1 kernel and while I
> might definitely be wrong here, I'm guessing it's not related to the
> crash.
>
> Anyone have an idea how to debug this? Unfortunately my AMSS sources do
> not include any sources for WiFi, it's just prebuilts.
>
> Regards
> Luca



^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2026-01-16 14:59 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-07-18 13:23 ath11k WiFi bringup on Milos SoC Luca Weiss
2026-01-16 14:59 ` Luca Weiss
  -- strict thread matches above, loose matches on Subject: below --
2026-01-12 13:44 Alexander Koskovich

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox