1. Patchew
  2. linux
  3. [PATCH v3 0/4] clk: spacemit: refactor common ccu driver
  4. View patch

[PATCH v3 4/4] reset: spacemit: fix auxiliary device id

Yixun Lan posted 4 patches 1 month, 1 week ago
There is a newer version of this series
[PATCH v3 4/4] reset: spacemit: fix auxiliary device id
Posted by Yixun Lan 1 month, 1 week ago 
Due to the auxiliary register procedure moved to ccu common module where
the module name changed to spacemit_ccu, then the reset auxiliary device
register id also need to be adjusted in order to prepare for adding new
K3 reset driver, otherwise two reset drivers will claim to support same
"compatible" auxiliary device.

Signed-off-by: Yixun Lan <dlan@gentoo.org>
---
 drivers/reset/reset-spacemit.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/reset/reset-spacemit.c b/drivers/reset/reset-spacemit.c
index e1272aff28f7..cc7fd1f8750d 100644
--- a/drivers/reset/reset-spacemit.c
+++ b/drivers/reset/reset-spacemit.c
@@ -278,7 +278,7 @@ static int spacemit_reset_probe(struct auxiliary_device *adev,
 
 #define K1_AUX_DEV_ID(_unit) \
 	{ \
-		.name = "spacemit_ccu_k1." #_unit "-reset", \
+		.name = "spacemit_ccu.k1-" #_unit "-reset", \
 		.driver_data = (kernel_ulong_t)&k1_ ## _unit ## _reset_data, \
 	}
 

-- 
2.52.0
Re: [PATCH v3 4/4] reset: spacemit: fix auxiliary device id
Posted by Alex Elder 1 month ago 
On 1/3/26 1:26 AM, Yixun Lan wrote:
> Due to the auxiliary register procedure moved to ccu common module where
> the module name changed to spacemit_ccu, then the reset auxiliary device
> register id also need to be adjusted in order to prepare for adding new
> K3 reset driver, otherwise two reset drivers will claim to support same
> "compatible" auxiliary device.
> 
> Signed-off-by: Yixun Lan <dlan@gentoo.org>

This would ideally be merged with the previous patch.  Maybe
Philipp can negotiate with Stephen to have that happen.

Reviewed-by: Alex Elder <elder@riscstar.com>

> ---
>   drivers/reset/reset-spacemit.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/reset/reset-spacemit.c b/drivers/reset/reset-spacemit.c
> index e1272aff28f7..cc7fd1f8750d 100644
> --- a/drivers/reset/reset-spacemit.c
> +++ b/drivers/reset/reset-spacemit.c
> @@ -278,7 +278,7 @@ static int spacemit_reset_probe(struct auxiliary_device *adev,
>   
>   #define K1_AUX_DEV_ID(_unit) \
>   	{ \
> -		.name = "spacemit_ccu_k1." #_unit "-reset", \
> +		.name = "spacemit_ccu.k1-" #_unit "-reset", \
>   		.driver_data = (kernel_ulong_t)&k1_ ## _unit ## _reset_data, \
>   	}
>   
>
Re: [PATCH v3 4/4] reset: spacemit: fix auxiliary device id
Posted by Philipp Zabel 1 month ago 
On Di, 2026-01-06 at 08:43 -0600, Alex Elder wrote:
> On 1/3/26 1:26 AM, Yixun Lan wrote:
> > Due to the auxiliary register procedure moved to ccu common module where
> > the module name changed to spacemit_ccu, then the reset auxiliary device
> > register id also need to be adjusted in order to prepare for adding new
> > K3 reset driver, otherwise two reset drivers will claim to support same
> > "compatible" auxiliary device.
> > 
> > Signed-off-by: Yixun Lan <dlan@gentoo.org>
> 
> This would ideally be merged with the previous patch.  Maybe
> Philipp can negotiate with Stephen to have that happen.
> 
> Reviewed-by: Alex Elder <elder@riscstar.com>
> 
> > ---
> >   drivers/reset/reset-spacemit.c | 2 +-
> >   1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/drivers/reset/reset-spacemit.c b/drivers/reset/reset-spacemit.c
> > index e1272aff28f7..cc7fd1f8750d 100644
> > --- a/drivers/reset/reset-spacemit.c
> > +++ b/drivers/reset/reset-spacemit.c
> > @@ -278,7 +278,7 @@ static int spacemit_reset_probe(struct auxiliary_device *adev,
> >   
> >   #define K1_AUX_DEV_ID(_unit) \
> >   	{ \
> > -		.name = "spacemit_ccu_k1." #_unit "-reset", \
> > +		.name = "spacemit_ccu.k1-" #_unit "-reset", \
> >   		.driver_data = (kernel_ulong_t)&k1_ ## _unit ## _reset_data, \
> >   	}
> >   

Acked-by: Philipp Zabel <p.zabel@pengutronix.de>

to be merged via the clock tree.

Note that the commits for both patches 2 and 3 will fail to probe the
reset driver when bisecting.
If you don't want do merge the corresponding changes to K1_AUX_DEV_ID()
into those two patches, to make this series (runtime) bisectable, you
should add a warning to both their commit messages that the reset
driver will be adapted in the following patch 4.

regards
Philipp
Re: [PATCH v3 4/4] reset: spacemit: fix auxiliary device id
Posted by Yixun Lan 1 month ago 
Hi Philipp,

On 12:07 Thu 08 Jan     , Philipp Zabel wrote:
> On Di, 2026-01-06 at 08:43 -0600, Alex Elder wrote:
> > On 1/3/26 1:26 AM, Yixun Lan wrote:
> > > Due to the auxiliary register procedure moved to ccu common module where
> > > the module name changed to spacemit_ccu, then the reset auxiliary device
> > > register id also need to be adjusted in order to prepare for adding new
> > > K3 reset driver, otherwise two reset drivers will claim to support same
> > > "compatible" auxiliary device.
> > > 
> > > Signed-off-by: Yixun Lan <dlan@gentoo.org>
> > 
> > This would ideally be merged with the previous patch.  Maybe
> > Philipp can negotiate with Stephen to have that happen.
> > 
> > Reviewed-by: Alex Elder <elder@riscstar.com>
> > 
> > > ---
> > >   drivers/reset/reset-spacemit.c | 2 +-
> > >   1 file changed, 1 insertion(+), 1 deletion(-)
> > > 
> > > diff --git a/drivers/reset/reset-spacemit.c b/drivers/reset/reset-spacemit.c
> > > index e1272aff28f7..cc7fd1f8750d 100644
> > > --- a/drivers/reset/reset-spacemit.c
> > > +++ b/drivers/reset/reset-spacemit.c
> > > @@ -278,7 +278,7 @@ static int spacemit_reset_probe(struct auxiliary_device *adev,
> > >   
> > >   #define K1_AUX_DEV_ID(_unit) \
> > >   	{ \
> > > -		.name = "spacemit_ccu_k1." #_unit "-reset", \
> > > +		.name = "spacemit_ccu.k1-" #_unit "-reset", \
> > >   		.driver_data = (kernel_ulong_t)&k1_ ## _unit ## _reset_data, \
> > >   	}
> > >   
> 
> Acked-by: Philipp Zabel <p.zabel@pengutronix.de>
> 
thanks

> to be merged via the clock tree.
> 
ok, got!

and, here is my plan, I will prepare an immutable tag for the patch 4 -
the change to reset driver ID, and merge it with the whole series,
then send them along via clock tree.

the tag would be useful if follow-up K3 reset patch[1] need to be merged, since
both will change the reset driver, and with the tag, patch 4 can be pulled into
both clock and reset subsystem, future reset patches can commit on top of it.
even there is no price to pay if above doesn't happen.


> Note that the commits for both patches 2 and 3 will fail to probe the
> reset driver when bisecting.
> If you don't want do merge the corresponding changes to K1_AUX_DEV_ID()
> into those two patches, to make this series (runtime) bisectable, you
no, I do not want to merge changes K1_AUX_DEV_ID() into patch 2/3

> should add a warning to both their commit messages that the reset
> driver will be adapted in the following patch 4.
ok, I will send a new version and add a warning message

> 
> regards
> Philipp

Link: https://lore.kernel.org/all/20251229-k3-reset-v1-0-eda0747bded3@riscstar.com/ [1]

-- 
Yixun Lan (dlan)

Patchew 2.1-devel-b67e4c2

© 2016 - 2026 Red Hat, Inc.