[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <D9U8PV79LTVH.38UZBFZD2VB3R@fairphone.com>
Date: Mon, 12 May 2025 15:09:30 +0100
From: "Luca Weiss" <luca.weiss@...rphone.com>
To: Matti Lehtimäki <matti.lehtimaki@...il.com>,
<linux-arm-msm@...r.kernel.org>
Cc: <~postmarketos/upstreaming@...ts.sr.ht>, <phone-devel@...r.kernel.org>,
"Bjorn Andersson" <andersson@...nel.org>, "Mathieu Poirier"
<mathieu.poirier@...aro.org>, "Luca Weiss" <luca@...aweiss.eu>,
<linux-remoteproc@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 1/1] remoteproc: qcom_wcnss: Fix on platforms without
fallback regulators
On Mon May 12, 2025 at 12:40 AM WEST, Matti Lehtimäki wrote:
> Recent change to handle platforms with only single power domain broke
> pronto-v3 which requires power domains and doesn't have fallback voltage
> regulators in case power domains are missing. Add a check to verify
> the number of fallback voltage regulators before using the code which
> handles single power domain situation.
Thanks!
I've tested this on MSM8953/SDM632 Fairphone 3 which is broken on
v6.15-rc5 without this patch! I've also heard of MSM8909 being broken in
v6.15-rc, and msm8976 also uses qcom,pronto-v3-pil.
Hopefully this can still go into the remaining rc's so v6.15 final works.
Tested-by: Luca Weiss <luca.weiss@...rphone.com> # sdm632-fairphone-fp3
Regards
Luca
>
> Fixes: 65991ea8a6d1 ("remoteproc: qcom_wcnss: Handle platforms with only single power domain")
> Signed-off-by: Matti Lehtimäki <matti.lehtimaki@...il.com>
> ---
> drivers/remoteproc/qcom_wcnss.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/remoteproc/qcom_wcnss.c b/drivers/remoteproc/qcom_wcnss.c
> index 775b056d795a..2c7e519a2254 100644
> --- a/drivers/remoteproc/qcom_wcnss.c
> +++ b/drivers/remoteproc/qcom_wcnss.c
> @@ -456,7 +456,8 @@ static int wcnss_init_regulators(struct qcom_wcnss *wcnss,
> if (wcnss->num_pds) {
> info += wcnss->num_pds;
> /* Handle single power domain case */
> - num_vregs += num_pd_vregs - wcnss->num_pds;
> + if (wcnss->num_pds < num_pd_vregs)
> + num_vregs += num_pd_vregs - wcnss->num_pds;
> } else {
> num_vregs += num_pd_vregs;
> }
Powered by blists - more mailing lists