Add bindings, driver and dts to support the Camera Subsystem on the
SM6350 SoC.

These patches were tested on a Fairphone 4 smartphone with WIP sensor
drivers (Sony IMX576 and IMX582), the camera pipeline works properly as
far as I can tell.

Though when stopping the camera stream, the following clock warning
appears in dmesg. But it does not interfere with any functionality,
starting and stopping the stream works and debugcc is showing 426.4 MHz
while the clock is on, and 'off' while it's off.

Any suggestion how to fix this, is appreciated.

[ 5738.590980] ------------[ cut here ]------------
[ 5738.591009] gcc_camera_axi_clk status stuck at 'on'
[ 5738.591049] WARNING: CPU: 0 PID: 6918 at drivers/clk/qcom/clk-branch.c:87 clk_branch_toggle+0x170/0x190
[ 5738.591081] Modules linked in:
[ 5738.591099] CPU: 0 UID: 10000 PID: 6918 Comm: plasma-camera Tainted: G        W           6.17.0-00057-ge6b67db49622 #71 NONE 
[ 5738.591118] Tainted: [W]=WARN
[ 5738.591126] Hardware name: Fairphone 4 (DT)
[ 5738.591136] pstate: 604000c5 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 5738.591150] pc : clk_branch_toggle+0x170/0x190
[ 5738.591164] lr : clk_branch_toggle+0x170/0x190
[ 5738.591177] sp : ffff800086ed3980
[ 5738.591184] x29: ffff800086ed3990 x28: 0000000000000001 x27: ffff800086ed3cd8
[ 5738.591208] x26: 0000000000000000 x25: ffffda14fcfbd250 x24: 0000000000000000
[ 5738.591230] x23: 0000000000000000 x22: ffffda14fc38bce0 x21: 0000000000000000
[ 5738.591252] x20: ffffda14fd33e618 x19: 0000000000000000 x18: 00000000000064c8
[ 5738.591274] x17: 0000000000000000 x16: 00001ae003667e9e x15: ffffda14fd2a07b0
[ 5738.591295] x14: 0000000000000000 x13: 6f27207461206b63 x12: 7574732073757461
[ 5738.591317] x11: 0000000000000058 x10: 0000000000000018 x9 : ffffda14fd2a0838
[ 5738.591338] x8 : 0000000000057fa8 x7 : 0000000000000a16 x6 : ffffda14fd2f8838
[ 5738.591360] x5 : ffff0001f6f59788 x4 : 0000000000000a15 x3 : ffff25ecf9d7e000
[ 5738.591381] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff0000baf5c100
[ 5738.591403] Call trace:
[ 5738.591412]  clk_branch_toggle+0x170/0x190 (P)
[ 5738.591429]  clk_branch2_disable+0x1c/0x30
[ 5738.591445]  clk_core_disable+0x5c/0xb4
[ 5738.591462]  clk_disable+0x38/0x60
[ 5738.591478]  camss_disable_clocks+0x44/0x78
[ 5738.591496]  vfe_put+0x7c/0xc0
[ 5738.591512]  vfe_set_power+0x40/0x50
[ 5738.591528]  pipeline_pm_power_one+0x14c/0x150
[ 5738.591546]  pipeline_pm_power+0x74/0xf4
[ 5738.591561]  v4l2_pipeline_pm_use+0x54/0x9c
[ 5738.591577]  v4l2_pipeline_pm_put+0x14/0x40
[ 5738.591592]  video_unprepare_streaming+0x18/0x24
[ 5738.591608]  __vb2_queue_cancel+0x4c/0x314
[ 5738.591626]  vb2_core_streamoff+0x24/0xc8
[ 5738.591643]  vb2_ioctl_streamoff+0x58/0x98
[ 5738.591657]  v4l_streamoff+0x24/0x30
[ 5738.591672]  __video_do_ioctl+0x430/0x4a8
[ 5738.591689]  video_usercopy+0x2ac/0x680
[ 5738.591705]  video_ioctl2+0x18/0x40
[ 5738.591720]  v4l2_ioctl+0x40/0x60
[ 5738.591734]  __arm64_sys_ioctl+0x90/0xf0
[ 5738.591750]  invoke_syscall.constprop.0+0x40/0xf0
[ 5738.591769]  el0_svc_common.constprop.0+0x38/0xd8
[ 5738.591785]  do_el0_svc+0x1c/0x28
[ 5738.591801]  el0_svc+0x34/0xe8
[ 5738.591820]  el0t_64_sync_handler+0xa0/0xe4
[ 5738.591838]  el0t_64_sync+0x198/0x19c
[ 5738.591854] ---[ end trace 0000000000000000 ]---

Signed-off-by: Luca Weiss <luca.weiss@fairphone.com>
---
Changes in v2:
- Remove prefix from interconnect-names
- Move 'top' power-domain to the top of list
- Update regulator supply names
- Link to v1: https://lore.kernel.org/r/20251024-sm6350-camss-v1-0-63d626638add@fairphone.com

---
Luca Weiss (3):
      dt-bindings: media: camss: Add qcom,sm6350-camss
      media: qcom: camss: Add SM6350 support
      arm64: dts: qcom: sm6350: Add CAMSS node

 .../bindings/media/qcom,sm6350-camss.yaml          | 349 +++++++++++++++++++++
 arch/arm64/boot/dts/qcom/sm6350.dtsi               | 165 ++++++++++
 .../platform/qcom/camss/camss-csiphy-3ph-1-0.c     | 125 ++++++++
 drivers/media/platform/qcom/camss/camss-vfe.c      |   2 +
 drivers/media/platform/qcom/camss/camss.c          | 249 +++++++++++++++
 drivers/media/platform/qcom/camss/camss.h          |   1 +
 6 files changed, 891 insertions(+)
---
base-commit: a92c761bcac3d5042559107fa7679470727a4bcb
change-id: 20251024-sm6350-camss-9c404bf9cfdd

Best regards,
-- 
Luca Weiss <luca.weiss@fairphone.com>

On 14/11/2025 11:15, Luca Weiss wrote:
> Add bindings, driver and dts to support the Camera Subsystem on the
> SM6350 SoC.
> 
> These patches were tested on a Fairphone 4 smartphone with WIP sensor
> drivers (Sony IMX576 and IMX582), the camera pipeline works properly as
> far as I can tell.
> 
> Though when stopping the camera stream, the following clock warning
> appears in dmesg. But it does not interfere with any functionality,
> starting and stopping the stream works and debugcc is showing 426.4 MHz
> while the clock is on, and 'off' while it's off.
> 
> Any suggestion how to fix this, is appreciated.
> 
> [ 5738.590980] ------------[ cut here ]------------
> [ 5738.591009] gcc_camera_axi_clk status stuck at 'on'
> [ 5738.591049] WARNING: CPU: 0 PID: 6918 at drivers/clk/qcom/clk-branch.c:87 clk_branch_toggle+0x170/0x190

Do you have a full and complete kernel tree we could look at here ?

---
bod

On Fri Nov 14, 2025 at 4:51 PM CET, Bryan O'Donoghue wrote:
> On 14/11/2025 11:15, Luca Weiss wrote:
>> Add bindings, driver and dts to support the Camera Subsystem on the
>> SM6350 SoC.
>> 
>> These patches were tested on a Fairphone 4 smartphone with WIP sensor
>> drivers (Sony IMX576 and IMX582), the camera pipeline works properly as
>> far as I can tell.
>> 
>> Though when stopping the camera stream, the following clock warning
>> appears in dmesg. But it does not interfere with any functionality,
>> starting and stopping the stream works and debugcc is showing 426.4 MHz
>> while the clock is on, and 'off' while it's off.
>> 
>> Any suggestion how to fix this, is appreciated.
>> 
>> [ 5738.590980] ------------[ cut here ]------------
>> [ 5738.591009] gcc_camera_axi_clk status stuck at 'on'
>> [ 5738.591049] WARNING: CPU: 0 PID: 6918 at drivers/clk/qcom/clk-branch.c:87 clk_branch_toggle+0x170/0x190
>
> Do you have a full and complete kernel tree we could look at here ?

Sure, this branch has everything in:

https://github.com/sm6350-mainline/linux/tree/sm6350-6.17.y/

For further refence, at least two other people have tested this branch
in postmarketOS, nothing particularly exciting to report from there,
apart from that the sdm-skin-thermal thermal zone (thermistor right next
to SoC) is currently configured with 55 degC as critical trip, which is
quickly achieved when starting a video recording, but that's not really
an issue with camss, but will need some tweaking regardless.

https://gitlab.postmarketos.org/postmarketOS/pmaports/-/merge_requests/7281

Regards
Luca

>
> ---
> bod

On 14/11/2025 15:59, Luca Weiss wrote:
> On Fri Nov 14, 2025 at 4:51 PM CET, Bryan O'Donoghue wrote:
>> On 14/11/2025 11:15, Luca Weiss wrote:
>>> Add bindings, driver and dts to support the Camera Subsystem on the
>>> SM6350 SoC.
>>>
>>> These patches were tested on a Fairphone 4 smartphone with WIP sensor
>>> drivers (Sony IMX576 and IMX582), the camera pipeline works properly as
>>> far as I can tell.
>>>
>>> Though when stopping the camera stream, the following clock warning
>>> appears in dmesg. But it does not interfere with any functionality,
>>> starting and stopping the stream works and debugcc is showing 426.4 MHz
>>> while the clock is on, and 'off' while it's off.
>>>
>>> Any suggestion how to fix this, is appreciated.
>>>
>>> [ 5738.590980] ------------[ cut here ]------------
>>> [ 5738.591009] gcc_camera_axi_clk status stuck at 'on'
>>> [ 5738.591049] WARNING: CPU: 0 PID: 6918 at drivers/clk/qcom/clk-branch.c:87 clk_branch_toggle+0x170/0x190
>>
>> Do you have a full and complete kernel tree we could look at here ?
> 
> Sure, this branch has everything in:
> 
> https://github.com/sm6350-mainline/linux/tree/sm6350-6.17.y/
> 
> For further refence, at least two other people have tested this branch
> in postmarketOS, nothing particularly exciting to report from there,
> apart from that the sdm-skin-thermal thermal zone (thermistor right next
> to SoC) is currently configured with 55 degC as critical trip, which is
> quickly achieved when starting a video recording, but that's not really
> an issue with camss, but will need some tweaking regardless.
> 
> https://gitlab.postmarketos.org/postmarketOS/pmaports/-/merge_requests/7281

diff --git a/drivers/clk/qcom/gcc-sm6350.c b/drivers/clk/qcom/gcc-sm6350.c
index a4d6dff9d0f7f..229629ef82809 100644
--- a/drivers/clk/qcom/gcc-sm6350.c
+++ b/drivers/clk/qcom/gcc-sm6350.c
@@ -908,9 +908,7 @@ static struct clk_branch gcc_camera_ahb_clk = {

  static struct clk_branch gcc_camera_axi_clk = {
         .halt_reg = 0x17018,
-       .halt_check = BRANCH_HALT,
-       .hwcg_reg = 0x17018,
-       .hwcg_bit = 1,
+       .halt_check = BRANCH_VOTED,
         .clkr = {
                 .enable_reg = 0x17018,
                 .enable_mask = BIT(0),

?

---
bod

On 11/16/25 3:30 PM, Bryan O'Donoghue wrote:
> On 14/11/2025 15:59, Luca Weiss wrote:
>> On Fri Nov 14, 2025 at 4:51 PM CET, Bryan O'Donoghue wrote:
>>> On 14/11/2025 11:15, Luca Weiss wrote:
>>>> Add bindings, driver and dts to support the Camera Subsystem on the
>>>> SM6350 SoC.
>>>>
>>>> These patches were tested on a Fairphone 4 smartphone with WIP sensor
>>>> drivers (Sony IMX576 and IMX582), the camera pipeline works properly as
>>>> far as I can tell.
>>>>
>>>> Though when stopping the camera stream, the following clock warning
>>>> appears in dmesg. But it does not interfere with any functionality,
>>>> starting and stopping the stream works and debugcc is showing 426.4 MHz
>>>> while the clock is on, and 'off' while it's off.
>>>>
>>>> Any suggestion how to fix this, is appreciated.
>>>>
>>>> [ 5738.590980] ------------[ cut here ]------------
>>>> [ 5738.591009] gcc_camera_axi_clk status stuck at 'on'
>>>> [ 5738.591049] WARNING: CPU: 0 PID: 6918 at drivers/clk/qcom/clk-branch.c:87 clk_branch_toggle+0x170/0x190
>>>
>>> Do you have a full and complete kernel tree we could look at here ?
>>
>> Sure, this branch has everything in:
>>
>> https://github.com/sm6350-mainline/linux/tree/sm6350-6.17.y/
>>
>> For further refence, at least two other people have tested this branch
>> in postmarketOS, nothing particularly exciting to report from there,
>> apart from that the sdm-skin-thermal thermal zone (thermistor right next
>> to SoC) is currently configured with 55 degC as critical trip, which is
>> quickly achieved when starting a video recording, but that's not really
>> an issue with camss, but will need some tweaking regardless.
>>
>> https://gitlab.postmarketos.org/postmarketOS/pmaports/-/merge_requests/7281
> 
> diff --git a/drivers/clk/qcom/gcc-sm6350.c b/drivers/clk/qcom/gcc-sm6350.c
> index a4d6dff9d0f7f..229629ef82809 100644
> --- a/drivers/clk/qcom/gcc-sm6350.c
> +++ b/drivers/clk/qcom/gcc-sm6350.c
> @@ -908,9 +908,7 @@ static struct clk_branch gcc_camera_ahb_clk = {
> 
>  static struct clk_branch gcc_camera_axi_clk = {
>         .halt_reg = 0x17018,
> -       .halt_check = BRANCH_HALT,
> -       .hwcg_reg = 0x17018,
> -       .hwcg_bit = 1,

No reason to drop the hwcg description

> +       .halt_check = BRANCH_VOTED,

It'd be useful to explain why we should ignore the hw feedback in this case

>         .clkr = {
>                 .enable_reg = 0x17018,
>                 .enable_mask = BIT(0),

Konrad

On 17/11/2025 12:53, Konrad Dybcio wrote:
> On 11/16/25 3:30 PM, Bryan O'Donoghue wrote:
>> On 14/11/2025 15:59, Luca Weiss wrote:
>>> On Fri Nov 14, 2025 at 4:51 PM CET, Bryan O'Donoghue wrote:
>>>> On 14/11/2025 11:15, Luca Weiss wrote:
>>>>> Add bindings, driver and dts to support the Camera Subsystem on the
>>>>> SM6350 SoC.
>>>>>
>>>>> These patches were tested on a Fairphone 4 smartphone with WIP sensor
>>>>> drivers (Sony IMX576 and IMX582), the camera pipeline works properly as
>>>>> far as I can tell.
>>>>>
>>>>> Though when stopping the camera stream, the following clock warning
>>>>> appears in dmesg. But it does not interfere with any functionality,
>>>>> starting and stopping the stream works and debugcc is showing 426.4 MHz
>>>>> while the clock is on, and 'off' while it's off.
>>>>>
>>>>> Any suggestion how to fix this, is appreciated.
>>>>>
>>>>> [ 5738.590980] ------------[ cut here ]------------
>>>>> [ 5738.591009] gcc_camera_axi_clk status stuck at 'on'
>>>>> [ 5738.591049] WARNING: CPU: 0 PID: 6918 at drivers/clk/qcom/clk-branch.c:87 clk_branch_toggle+0x170/0x190
>>>>
>>>> Do you have a full and complete kernel tree we could look at here ?
>>>
>>> Sure, this branch has everything in:
>>>
>>> https://github.com/sm6350-mainline/linux/tree/sm6350-6.17.y/
>>>
>>> For further refence, at least two other people have tested this branch
>>> in postmarketOS, nothing particularly exciting to report from there,
>>> apart from that the sdm-skin-thermal thermal zone (thermistor right next
>>> to SoC) is currently configured with 55 degC as critical trip, which is
>>> quickly achieved when starting a video recording, but that's not really
>>> an issue with camss, but will need some tweaking regardless.
>>>
>>> https://gitlab.postmarketos.org/postmarketOS/pmaports/-/merge_requests/7281
>>
>> diff --git a/drivers/clk/qcom/gcc-sm6350.c b/drivers/clk/qcom/gcc-sm6350.c
>> index a4d6dff9d0f7f..229629ef82809 100644
>> --- a/drivers/clk/qcom/gcc-sm6350.c
>> +++ b/drivers/clk/qcom/gcc-sm6350.c
>> @@ -908,9 +908,7 @@ static struct clk_branch gcc_camera_ahb_clk = {
>>
>>   static struct clk_branch gcc_camera_axi_clk = {
>>          .halt_reg = 0x17018,
>> -       .halt_check = BRANCH_HALT,
>> -       .hwcg_reg = 0x17018,
>> -       .hwcg_bit = 1,
> 
> No reason to drop the hwcg description
> 
>> +       .halt_check = BRANCH_VOTED,
> 
> It'd be useful to explain why we should ignore the hw feedback in this case
> 
>>          .clkr = {
>>                  .enable_reg = 0x17018,
>>                  .enable_mask = BIT(0),
> 
> Konrad

vfe170 is what we have on sdm845

So I'm just asking Luca to try the sdm845 method of waggling this clock 
since what we have doesn't work.

---
bod

On 11/18/25 10:33 AM, Bryan O'Donoghue wrote:
> On 17/11/2025 12:53, Konrad Dybcio wrote:
>> On 11/16/25 3:30 PM, Bryan O'Donoghue wrote:
>>> On 14/11/2025 15:59, Luca Weiss wrote:
>>>> On Fri Nov 14, 2025 at 4:51 PM CET, Bryan O'Donoghue wrote:
>>>>> On 14/11/2025 11:15, Luca Weiss wrote:
>>>>>> Add bindings, driver and dts to support the Camera Subsystem on the
>>>>>> SM6350 SoC.
>>>>>>
>>>>>> These patches were tested on a Fairphone 4 smartphone with WIP sensor
>>>>>> drivers (Sony IMX576 and IMX582), the camera pipeline works properly as
>>>>>> far as I can tell.
>>>>>>
>>>>>> Though when stopping the camera stream, the following clock warning
>>>>>> appears in dmesg. But it does not interfere with any functionality,
>>>>>> starting and stopping the stream works and debugcc is showing 426.4 MHz
>>>>>> while the clock is on, and 'off' while it's off.
>>>>>>
>>>>>> Any suggestion how to fix this, is appreciated.
>>>>>>
>>>>>> [ 5738.590980] ------------[ cut here ]------------
>>>>>> [ 5738.591009] gcc_camera_axi_clk status stuck at 'on'
>>>>>> [ 5738.591049] WARNING: CPU: 0 PID: 6918 at drivers/clk/qcom/clk-branch.c:87 clk_branch_toggle+0x170/0x190
>>>>>
>>>>> Do you have a full and complete kernel tree we could look at here ?
>>>>
>>>> Sure, this branch has everything in:
>>>>
>>>> https://github.com/sm6350-mainline/linux/tree/sm6350-6.17.y/
>>>>
>>>> For further refence, at least two other people have tested this branch
>>>> in postmarketOS, nothing particularly exciting to report from there,
>>>> apart from that the sdm-skin-thermal thermal zone (thermistor right next
>>>> to SoC) is currently configured with 55 degC as critical trip, which is
>>>> quickly achieved when starting a video recording, but that's not really
>>>> an issue with camss, but will need some tweaking regardless.
>>>>
>>>> https://gitlab.postmarketos.org/postmarketOS/pmaports/-/merge_requests/7281
>>>
>>> diff --git a/drivers/clk/qcom/gcc-sm6350.c b/drivers/clk/qcom/gcc-sm6350.c
>>> index a4d6dff9d0f7f..229629ef82809 100644
>>> --- a/drivers/clk/qcom/gcc-sm6350.c
>>> +++ b/drivers/clk/qcom/gcc-sm6350.c
>>> @@ -908,9 +908,7 @@ static struct clk_branch gcc_camera_ahb_clk = {
>>>
>>>   static struct clk_branch gcc_camera_axi_clk = {
>>>          .halt_reg = 0x17018,
>>> -       .halt_check = BRANCH_HALT,
>>> -       .hwcg_reg = 0x17018,
>>> -       .hwcg_bit = 1,
>>
>> No reason to drop the hwcg description
>>
>>> +       .halt_check = BRANCH_VOTED,
>>
>> It'd be useful to explain why we should ignore the hw feedback in this case
>>
>>>          .clkr = {
>>>                  .enable_reg = 0x17018,
>>>                  .enable_mask = BIT(0),
>>
>> Konrad
> 
> vfe170 is what we have on sdm845
> 
> So I'm just asking Luca to try the sdm845 method of waggling this clock since what we have doesn't work.

It's of course going to work because this way you're not calling the
code that throws this error

I was curious whether you know the actual reason why this is being
done in some other GCC drivers

Konrad

On 18/11/2025 10:06, Konrad Dybcio wrote:
>>> Konrad
>> vfe170 is what we have on sdm845
>>
>> So I'm just asking Luca to try the sdm845 method of waggling this clock since what we have doesn't work.
> It's of course going to work because this way you're not calling the
> code that throws this error
> 
> I was curious whether you know the actual reason why this is being
> done in some other GCC drivers
> 
> Konrad

No notion at all, perhaps as a workaround to this very problem.

---
bod

On 11/18/25 12:08 PM, Bryan O'Donoghue wrote:
> On 18/11/2025 10:06, Konrad Dybcio wrote:
>>>> Konrad
>>> vfe170 is what we have on sdm845
>>>
>>> So I'm just asking Luca to try the sdm845 method of waggling this clock since what we have doesn't work.
>> It's of course going to work because this way you're not calling the
>> code that throws this error
>>
>> I was curious whether you know the actual reason why this is being
>> done in some other GCC drivers
>>
>> Konrad
> 
> No notion at all, perhaps as a workaround to this very problem.

I tried digging it up, but only managed to find that there is a signal
between titan and gcc to request it being enabled (perhaps that's a
fancy description of hwcg)

Maybe +Taniya would know? (context: sm6350 gcc_camera_axi_clk stuck at
'on' when disabling)

Konrad

Hi Bryan,

On Tue Nov 18, 2025 at 10:33 AM CET, Bryan O'Donoghue wrote:
> On 17/11/2025 12:53, Konrad Dybcio wrote:
>> On 11/16/25 3:30 PM, Bryan O'Donoghue wrote:
>>> On 14/11/2025 15:59, Luca Weiss wrote:
>>>> On Fri Nov 14, 2025 at 4:51 PM CET, Bryan O'Donoghue wrote:
>>>>> On 14/11/2025 11:15, Luca Weiss wrote:
>>>>>> Add bindings, driver and dts to support the Camera Subsystem on the
>>>>>> SM6350 SoC.
>>>>>>
>>>>>> These patches were tested on a Fairphone 4 smartphone with WIP sensor
>>>>>> drivers (Sony IMX576 and IMX582), the camera pipeline works properly as
>>>>>> far as I can tell.
>>>>>>
>>>>>> Though when stopping the camera stream, the following clock warning
>>>>>> appears in dmesg. But it does not interfere with any functionality,
>>>>>> starting and stopping the stream works and debugcc is showing 426.4 MHz
>>>>>> while the clock is on, and 'off' while it's off.
>>>>>>
>>>>>> Any suggestion how to fix this, is appreciated.
>>>>>>
>>>>>> [ 5738.590980] ------------[ cut here ]------------
>>>>>> [ 5738.591009] gcc_camera_axi_clk status stuck at 'on'
>>>>>> [ 5738.591049] WARNING: CPU: 0 PID: 6918 at drivers/clk/qcom/clk-branch.c:87 clk_branch_toggle+0x170/0x190
>>>>>
>>>>> Do you have a full and complete kernel tree we could look at here ?
>>>>
>>>> Sure, this branch has everything in:
>>>>
>>>> https://github.com/sm6350-mainline/linux/tree/sm6350-6.17.y/
>>>>
>>>> For further refence, at least two other people have tested this branch
>>>> in postmarketOS, nothing particularly exciting to report from there,
>>>> apart from that the sdm-skin-thermal thermal zone (thermistor right next
>>>> to SoC) is currently configured with 55 degC as critical trip, which is
>>>> quickly achieved when starting a video recording, but that's not really
>>>> an issue with camss, but will need some tweaking regardless.
>>>>
>>>> https://gitlab.postmarketos.org/postmarketOS/pmaports/-/merge_requests/7281
>>>
>>> diff --git a/drivers/clk/qcom/gcc-sm6350.c b/drivers/clk/qcom/gcc-sm6350.c
>>> index a4d6dff9d0f7f..229629ef82809 100644
>>> --- a/drivers/clk/qcom/gcc-sm6350.c
>>> +++ b/drivers/clk/qcom/gcc-sm6350.c
>>> @@ -908,9 +908,7 @@ static struct clk_branch gcc_camera_ahb_clk = {
>>>
>>>   static struct clk_branch gcc_camera_axi_clk = {
>>>          .halt_reg = 0x17018,
>>> -       .halt_check = BRANCH_HALT,
>>> -       .hwcg_reg = 0x17018,
>>> -       .hwcg_bit = 1,
>> 
>> No reason to drop the hwcg description
>> 
>>> +       .halt_check = BRANCH_VOTED,
>> 
>> It'd be useful to explain why we should ignore the hw feedback in this case
>> 
>>>          .clkr = {
>>>                  .enable_reg = 0x17018,
>>>                  .enable_mask = BIT(0),
>> 
>> Konrad
>
> vfe170 is what we have on sdm845
>
> So I'm just asking Luca to try the sdm845 method of waggling this clock 
> since what we have doesn't work.

Yes, your change to gcc-sm6350.c does remove the warning and everything
seems to work as expected. Opening and closing the plasma-camera
application, switching between cameras works without any visible issues.

Regards
Luca

>
> ---
> bod