[PATCH 0/4] iio: convert to use maple tree register cache

[PATCH 0/4] iio: convert to use maple tree register cache

[Chu Guangqing]

The maple tree register cache is based on a much more modern data structure
than the rbtree cache and makes optimisation choices which are probably
more appropriate for modern systems than those made by the rbtree cache.

Chu Guangqing (4):
  iio: adc: ade9000: convert to use maple tree register cache
  iio: light: veml3235: convert to use maple tree register cache
  iio: light: apds9306: convert to use maple tree register cache
  iio: light: apds9960: convert to use maple tree register cache

 drivers/iio/adc/ade9000.c    | 2 +-
 drivers/iio/light/apds9306.c | 2 +-
 drivers/iio/light/apds9960.c | 2 +-
 drivers/iio/light/veml3235.c | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

-- 
2.43.7

[PATCH 1/4] iio: adc: ade9000: convert to use maple tree register cache

[Chu Guangqing]

The maple tree register cache is based on a much more modern data structure
than the rbtree cache and makes optimisation choices which are probably
more appropriate for modern systems than those made by the rbtree cache.

Signed-off-by: Chu Guangqing <chuguangqing@inspur.com>
---
 drivers/iio/adc/ade9000.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/iio/adc/ade9000.c b/drivers/iio/adc/ade9000.c
index 94e05e11abd9..2de8a718d62a 100644
--- a/drivers/iio/adc/ade9000.c
+++ b/drivers/iio/adc/ade9000.c
@@ -1629,7 +1629,7 @@ static const struct regmap_config ade9000_regmap_config = {
 	.val_bits = 32,
 	.max_register = 0x6bc,
 	.zero_flag_mask = true,
-	.cache_type = REGCACHE_RBTREE,
+	.cache_type = REGCACHE_MAPLE,
 	.reg_read = ade9000_spi_read_reg,
 	.reg_write = ade9000_spi_write_reg,
 	.volatile_reg = ade9000_is_volatile_reg,
-- 
2.43.7

[PATCH 2/4] iio: light: veml3235: convert to use maple tree register cache

[Chu Guangqing]

The maple tree register cache is based on a much more modern data structure
than the rbtree cache and makes optimisation choices which are probably
more appropriate for modern systems than those made by the rbtree cache.

Signed-off-by: Chu Guangqing <chuguangqing@inspur.com>
---
 drivers/iio/light/veml3235.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/iio/light/veml3235.c b/drivers/iio/light/veml3235.c
index 77c9ae17ed47..9309ad83ca9e 100644
--- a/drivers/iio/light/veml3235.c
+++ b/drivers/iio/light/veml3235.c
@@ -154,7 +154,7 @@ static const struct regmap_config veml3235_regmap_config = {
 	.rd_table = &veml3235_readable_table,
 	.wr_table = &veml3235_writable_table,
 	.volatile_table = &veml3235_volatile_table,
-	.cache_type = REGCACHE_RBTREE,
+	.cache_type = REGCACHE_MAPLE,
 };
 
 static int veml3235_get_it(struct veml3235_data *data, int *val, int *val2)
-- 
2.43.7

[PATCH 3/4] iio: light: apds9306: convert to use maple tree register cache

[Chu Guangqing]

The maple tree register cache is based on a much more modern data structure
than the rbtree cache and makes optimisation choices which are probably
more appropriate for modern systems than those made by the rbtree cache.

Signed-off-by: Chu Guangqing <chuguangqing@inspur.com>
---
 drivers/iio/light/apds9306.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/iio/light/apds9306.c b/drivers/iio/light/apds9306.c
index 389125675caa..7e68cca0edfa 100644
--- a/drivers/iio/light/apds9306.c
+++ b/drivers/iio/light/apds9306.c
@@ -350,7 +350,7 @@ static const struct regmap_config apds9306_regmap = {
 	.volatile_table = &apds9306_volatile_table,
 	.precious_table = &apds9306_precious_table,
 	.max_register = APDS9306_ALS_THRES_VAR_REG,
-	.cache_type = REGCACHE_RBTREE,
+	.cache_type = REGCACHE_MAPLE,
 };
 
 static const struct reg_field apds9306_rf_sw_reset =
-- 
2.43.7

[PATCH 4/4] iio: light: apds9960: convert to use maple tree register cache

[Chu Guangqing]

The maple tree register cache is based on a much more modern data structure
than the rbtree cache and makes optimisation choices which are probably
more appropriate for modern systems than those made by the rbtree cache.

Signed-off-by: Chu Guangqing <chuguangqing@inspur.com>
---
 drivers/iio/light/apds9960.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/iio/light/apds9960.c b/drivers/iio/light/apds9960.c
index 79b202c59a0f..785c5dbe2d08 100644
--- a/drivers/iio/light/apds9960.c
+++ b/drivers/iio/light/apds9960.c
@@ -234,7 +234,7 @@ static const struct regmap_config apds9960_regmap_config = {
 	.reg_defaults = apds9960_reg_defaults,
 	.num_reg_defaults = ARRAY_SIZE(apds9960_reg_defaults),
 	.max_register = APDS9960_REG_GFIFO_DIR(RIGHT),
-	.cache_type = REGCACHE_RBTREE,
+	.cache_type = REGCACHE_MAPLE,
 };
 
 static const struct iio_event_spec apds9960_pxs_event_spec[] = {
-- 
2.43.7

Re: [PATCH 4/4] iio: light: apds9960: convert to use maple tree register cache

[Andy Shevchenko]

On Fri, Oct 24, 2025 at 03:38:23PM +0800, Chu Guangqing wrote:
> The maple tree register cache is based on a much more modern data structure
> than the rbtree cache and makes optimisation choices which are probably
> more appropriate for modern systems than those made by the rbtree cache.

...

>  	.reg_defaults = apds9960_reg_defaults,
>  	.num_reg_defaults = ARRAY_SIZE(apds9960_reg_defaults),

^^^^ Be careful with such cases, the cache implementations may behave
differently. Have you tested this on the actual HW?

>  	.max_register = APDS9960_REG_GFIFO_DIR(RIGHT),
> -	.cache_type = REGCACHE_RBTREE,
> +	.cache_type = REGCACHE_MAPLE,
>  };

-- 
With Best Regards,
Andy Shevchenko

Re: Re: [PATCH 4/4] iio: light: apds9960: convert to use maple tree register cache

[Andy Shevchenko]

On Sat, Oct 25, 2025 at 11:36:45AM +0800, Gary Chu(楚光庆) wrote:
> >On Fri, Oct 24, 2025 at 03:38:23PM +0800, Chu Guangqing wrote:
> >> The maple tree register cache is based on a much more modern data structure
> >> than the rbtree cache and makes optimisation choices which are probably
> >> more appropriate for modern systems than those made by the rbtree cache.

...

> >>   .reg_defaults = apds9960_reg_defaults,
> >>   .num_reg_defaults = ARRAY_SIZE(apds9960_reg_defaults),
> >
> >^^^^ Be careful with such cases, the cache implementations may behave
> >differently. Have you tested this on the actual HW?
> >
> We have conducted tests on some hardware, and performance improvements were observed,
>  though tests have not been carried out on all hardware models.
> Neither rbtree nor maple tree directly depends on hardware types (such as CPU or peripheral
>  models). Instead, they rely on the address distribution characteristics (discrete/continuous)
>  of hardware registers. The optimal cache type is determined by the hardware layout.
> Red-black trees excel at individual operations on discrete addresses, while Maple Trees are
>  proficient in range operations on contiguous addresses.

It's not about the low-level cache implementation, it's about regmap
abstraction implementation that might differ from cache to cache
implementations. This all in regard how the cold cache is getting filled up.
There is a separate discussion (unrelated to the topic of your series) where
this was brought up. That's why I asked how this was tested.

In any case, up to Jonathan, but I had to rise a potential misbehave, so in my
opinion this kind of corner cases needs to be tested on real HW.

> >>   .max_register = APDS9960_REG_GFIFO_DIR(RIGHT),
> >> - .cache_type = REGCACHE_RBTREE,
> >> + .cache_type = REGCACHE_MAPLE,
> >>  };

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH 3/4] iio: light: apds9306: convert to use maple tree register cache

[Subhajit Ghosh]

On 24/10/25 18:08, Chu Guangqing wrote:
> The maple tree register cache is based on a much more modern data structure
> than the rbtree cache and makes optimisation choices which are probably
> more appropriate for modern systems than those made by the rbtree cache.
> 
> Signed-off-by: Chu Guangqing <chuguangqing@inspur.com>
> ---
>   drivers/iio/light/apds9306.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/iio/light/apds9306.c b/drivers/iio/light/apds9306.c
> index 389125675caa..7e68cca0edfa 100644
> --- a/drivers/iio/light/apds9306.c
> +++ b/drivers/iio/light/apds9306.c
> @@ -350,7 +350,7 @@ static const struct regmap_config apds9306_regmap = {
>   	.volatile_table = &apds9306_volatile_table,
>   	.precious_table = &apds9306_precious_table,
>   	.max_register = APDS9306_ALS_THRES_VAR_REG,
> -	.cache_type = REGCACHE_RBTREE,
> +	.cache_type = REGCACHE_MAPLE,
>   };
>   
>   static const struct reg_field apds9306_rf_sw_reset =
Looks good. I will test it on a real hardware soon. Thank you.
Jonathan will know more about regmap internal implementations.

Acked-by: Subhajit Ghosh <subhajit.ghosh@tweaklogic.com>

Regards,
Subhajit Ghosh

Re: [PATCH 4/4] iio: light: apds9960: convert to use maple tree register cache

[Jonathan Cameron]

On Mon, 27 Oct 2025 10:30:54 +0200
Andy Shevchenko <andriy.shevchenko@intel.com> wrote:

> On Sat, Oct 25, 2025 at 11:36:45AM +0800, Gary Chu(楚光庆) wrote:
> > >On Fri, Oct 24, 2025 at 03:38:23PM +0800, Chu Guangqing wrote:  
> > >> The maple tree register cache is based on a much more modern data structure
> > >> than the rbtree cache and makes optimisation choices which are probably
> > >> more appropriate for modern systems than those made by the rbtree cache.  
> 
> ...
> 
> > >>   .reg_defaults = apds9960_reg_defaults,
> > >>   .num_reg_defaults = ARRAY_SIZE(apds9960_reg_defaults),  
> > >
> > >^^^^ Be careful with such cases, the cache implementations may behave
> > >differently. Have you tested this on the actual HW?
> > >  
> > We have conducted tests on some hardware, and performance improvements were observed,
> >  though tests have not been carried out on all hardware models.
> > Neither rbtree nor maple tree directly depends on hardware types (such as CPU or peripheral
> >  models). Instead, they rely on the address distribution characteristics (discrete/continuous)
> >  of hardware registers. The optimal cache type is determined by the hardware layout.
> > Red-black trees excel at individual operations on discrete addresses, while Maple Trees are
> >  proficient in range operations on contiguous addresses.  
> 
> It's not about the low-level cache implementation, it's about regmap
> abstraction implementation that might differ from cache to cache
> implementations. This all in regard how the cold cache is getting filled up.
> There is a separate discussion (unrelated to the topic of your series) where
> this was brought up.

I appreciate these things can be hard to track down with lots of threads in flight
but any chance of a reference for that? I'd be a little surprised if these uses
are complicated enough to hit corner cases but would like to know more.
I've taken a few similar changes in the past thinking there would be no
practical difference.

Jonathan



> That's why I asked how this was tested.
> 
> In any case, up to Jonathan, but I had to rise a potential misbehave, so in my
> opinion this kind of corner cases needs to be tested on real HW.
> 
> > >>   .max_register = APDS9960_REG_GFIFO_DIR(RIGHT),
> > >> - .cache_type = REGCACHE_RBTREE,
> > >> + .cache_type = REGCACHE_MAPLE,
> > >>  };  
> 

Re: [PATCH 4/4] iio: light: apds9960: convert to use maple tree register cache

[Andy Shevchenko]

On Mon, Oct 27, 2025 at 01:38:06PM +0000, Jonathan Cameron wrote:
> On Mon, 27 Oct 2025 10:30:54 +0200
> Andy Shevchenko <andriy.shevchenko@intel.com> wrote:
> > On Sat, Oct 25, 2025 at 11:36:45AM +0800, Gary Chu(楚光庆) wrote:
> > > >On Fri, Oct 24, 2025 at 03:38:23PM +0800, Chu Guangqing wrote:  
> > > >> The maple tree register cache is based on a much more modern data structure
> > > >> than the rbtree cache and makes optimisation choices which are probably
> > > >> more appropriate for modern systems than those made by the rbtree cache.  

...

> > > >>   .reg_defaults = apds9960_reg_defaults,
> > > >>   .num_reg_defaults = ARRAY_SIZE(apds9960_reg_defaults),  
> > > >
> > > >^^^^ Be careful with such cases, the cache implementations may behave
> > > >differently. Have you tested this on the actual HW?
> > > >  
> > > We have conducted tests on some hardware, and performance improvements were observed,
> > >  though tests have not been carried out on all hardware models.
> > > Neither rbtree nor maple tree directly depends on hardware types (such as CPU or peripheral
> > >  models). Instead, they rely on the address distribution characteristics (discrete/continuous)
> > >  of hardware registers. The optimal cache type is determined by the hardware layout.
> > > Red-black trees excel at individual operations on discrete addresses, while Maple Trees are
> > >  proficient in range operations on contiguous addresses.  
> > 
> > It's not about the low-level cache implementation, it's about regmap
> > abstraction implementation that might differ from cache to cache
> > implementations. This all in regard how the cold cache is getting filled up.
> > There is a separate discussion (unrelated to the topic of your series) where
> > this was brought up.
> 
> I appreciate these things can be hard to track down with lots of threads in flight
> but any chance of a reference for that? I'd be a little surprised if these uses
> are complicated enough to hit corner cases but would like to know more.
> I've taken a few similar changes in the past thinking there would be no
> practical difference.

Sure, it appeared in the discussion of v2 of the following patch:
https://lore.kernel.org/linux-gpio/20251009132651.649099-2-bigunclemax@gmail.com/

> > That's why I asked how this was tested.
> > 
> > In any case, up to Jonathan, but I had to rise a potential misbehave, so in my
> > opinion this kind of corner cases needs to be tested on real HW.
> > 
> > > >>   .max_register = APDS9960_REG_GFIFO_DIR(RIGHT),
> > > >> - .cache_type = REGCACHE_RBTREE,
> > > >> + .cache_type = REGCACHE_MAPLE,
> > > >>  };  

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH 4/4] iio: light: apds9960: convert to use maple tree register cache

[Jonathan Cameron]

On Tue, 28 Oct 2025 10:10:24 +0200
Andy Shevchenko <andriy.shevchenko@intel.com> wrote:

> On Mon, Oct 27, 2025 at 01:38:06PM +0000, Jonathan Cameron wrote:
> > On Mon, 27 Oct 2025 10:30:54 +0200
> > Andy Shevchenko <andriy.shevchenko@intel.com> wrote:  
> > > On Sat, Oct 25, 2025 at 11:36:45AM +0800, Gary Chu(楚光庆) wrote:  
> > > > >On Fri, Oct 24, 2025 at 03:38:23PM +0800, Chu Guangqing wrote:    
> > > > >> The maple tree register cache is based on a much more modern data structure
> > > > >> than the rbtree cache and makes optimisation choices which are probably
> > > > >> more appropriate for modern systems than those made by the rbtree cache.    
> 
> ...
> 
> > > > >>   .reg_defaults = apds9960_reg_defaults,
> > > > >>   .num_reg_defaults = ARRAY_SIZE(apds9960_reg_defaults),    
> > > > >
> > > > >^^^^ Be careful with such cases, the cache implementations may behave
> > > > >differently. Have you tested this on the actual HW?
> > > > >    
> > > > We have conducted tests on some hardware, and performance improvements were observed,
> > > >  though tests have not been carried out on all hardware models.
> > > > Neither rbtree nor maple tree directly depends on hardware types (such as CPU or peripheral
> > > >  models). Instead, they rely on the address distribution characteristics (discrete/continuous)
> > > >  of hardware registers. The optimal cache type is determined by the hardware layout.
> > > > Red-black trees excel at individual operations on discrete addresses, while Maple Trees are
> > > >  proficient in range operations on contiguous addresses.    
> > > 
> > > It's not about the low-level cache implementation, it's about regmap
> > > abstraction implementation that might differ from cache to cache
> > > implementations. This all in regard how the cold cache is getting filled up.
> > > There is a separate discussion (unrelated to the topic of your series) where
> > > this was brought up.  
> > 
> > I appreciate these things can be hard to track down with lots of threads in flight
> > but any chance of a reference for that? I'd be a little surprised if these uses
> > are complicated enough to hit corner cases but would like to know more.
> > I've taken a few similar changes in the past thinking there would be no
> > practical difference.  
> 
> Sure, it appeared in the discussion of v2 of the following patch:
> https://lore.kernel.org/linux-gpio/20251009132651.649099-2-bigunclemax@gmail.com/

Thanks - that's useful info.

Here the uses are all simple and I can't see any likelihood of problems + we've
had a bunch of these changes from RBTREE to MAPLE in the past without issue.

So applied

Thanks,

Jonathan

> 
> > > That's why I asked how this was tested.
> > > 
> > > In any case, up to Jonathan, but I had to rise a potential misbehave, so in my
> > > opinion this kind of corner cases needs to be tested on real HW.
> > >   
> > > > >>   .max_register = APDS9960_REG_GFIFO_DIR(RIGHT),
> > > > >> - .cache_type = REGCACHE_RBTREE,
> > > > >> + .cache_type = REGCACHE_MAPLE,
> > > > >>  };    
>