[PATCH v3 0/5] regcache: Split out ->populate() and use it

[PATCH v3 0/5] regcache: Split out ->populate() and use it

[Andy Shevchenko]

This is a refactoring series to decouple cache initialisation and population.
On its own it has no functional impact but will be used in the further
development. Besides that I found this split useful on its own (from the design
perspective). That's why I decided to send it out as is separately from a bigger
(and ongoing) work.

Changelog v3:
- added missing space (Mark)
- collected tags (Sander)

Changelog v2:
v2: https://lore.kernel.org/r/20251030173915.3886882-1-andriy.shevchenko@linux.intel.com
- rebased on top of the recent regcache changes (Mark)
- a new clean up patch induced by the previous

v1: https://lore.kernel.org/r/20251029073131.3004660-1-andriy.shevchenko@linux.intel.com

Andy Shevchenko (5):
  regcache: Add ->populate() callback to separate from ->init()
  regcache: rbtree: Split ->populate() from ->init()
  regcache: flat: Remove unneeded check and error message for -ENOMEM
  regcache: flat: Split ->populate() from ->init()
  regcache: maple: Split ->populate() from ->init()

 drivers/base/regmap/internal.h        |  1 +
 drivers/base/regmap/regcache-flat.c   | 37 +++++++++++----------
 drivers/base/regmap/regcache-maple.c  | 47 ++++++++++++---------------
 drivers/base/regmap/regcache-rbtree.c | 31 ++++++++++--------
 drivers/base/regmap/regcache.c        | 16 +++++++++
 5 files changed, 74 insertions(+), 58 deletions(-)

-- 
2.50.1

[PATCH v3 1/5] regcache: Add ->populate() callback to separate from ->init()

[Andy Shevchenko]

In the future changes we would like to change the flow of the cache handling.
Add ->populate() callback in order to prepare for that.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 drivers/base/regmap/internal.h |  1 +
 drivers/base/regmap/regcache.c | 16 ++++++++++++++++
 2 files changed, 17 insertions(+)

diff --git a/drivers/base/regmap/internal.h b/drivers/base/regmap/internal.h
index 8d19a1414d5b..1477329410ec 100644
--- a/drivers/base/regmap/internal.h
+++ b/drivers/base/regmap/internal.h
@@ -186,6 +186,7 @@ struct regcache_ops {
 	enum regcache_type type;
 	int (*init)(struct regmap *map);
 	int (*exit)(struct regmap *map);
+	int (*populate)(struct regmap *map);
 #ifdef CONFIG_DEBUG_FS
 	void (*debugfs_init)(struct regmap *map);
 #endif
diff --git a/drivers/base/regmap/regcache.c b/drivers/base/regmap/regcache.c
index 0392f5525cf3..319c342bf5a0 100644
--- a/drivers/base/regmap/regcache.c
+++ b/drivers/base/regmap/regcache.c
@@ -222,8 +222,24 @@ int regcache_init(struct regmap *map, const struct regmap_config *config)
 		if (ret)
 			goto err_free;
 	}
+
+	if (map->num_reg_defaults && map->cache_ops->populate) {
+		dev_dbg(map->dev, "Populating %s cache\n", map->cache_ops->name);
+		map->lock(map->lock_arg);
+		ret = map->cache_ops->populate(map);
+		map->unlock(map->lock_arg);
+		if (ret)
+			goto err_exit;
+	}
 	return 0;
 
+err_exit:
+	if (map->cache_ops->exit) {
+		dev_dbg(map->dev, "Destroying %s cache\n", map->cache_ops->name);
+		map->lock(map->lock_arg);
+		ret = map->cache_ops->exit(map);
+		map->unlock(map->lock_arg);
+	}
 err_free:
 	kfree(map->reg_defaults);
 	if (map->cache_free)
-- 
2.50.1

Re: [PATCH v3 1/5] regcache: Add ->populate() callback to separate from ->init()

[Charles Keepax]

On Fri, Oct 31, 2025 at 09:03:16AM +0100, Andy Shevchenko wrote:
> In the future changes we would like to change the flow of the cache handling.
> Add ->populate() callback in order to prepare for that.
> 
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---

Reviewed-by: Charles Keepax <ckeepax@opensource.cirrus.com>

Thanks,
Charles

Re: [PATCH v3 1/5] regcache: Add ->populate() callback to separate from ->init()

[Andy Shevchenko]

On Fri, Oct 31, 2025 at 10:48:37AM +0000, Charles Keepax wrote:
> On Fri, Oct 31, 2025 at 09:03:16AM +0100, Andy Shevchenko wrote:
> > In the future changes we would like to change the flow of the cache handling.
> > Add ->populate() callback in order to prepare for that.
> 
> Reviewed-by: Charles Keepax <ckeepax@opensource.cirrus.com>

Thank you!

-- 
With Best Regards,
Andy Shevchenko

[PATCH v3 2/5] regcache: rbtree: Split ->populate() from ->init()

[Andy Shevchenko]

Split ->populate() implementation from ->init() code.
This decoupling will help for the further changes.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 drivers/base/regmap/regcache-rbtree.c | 31 +++++++++++++++------------
 1 file changed, 17 insertions(+), 14 deletions(-)

diff --git a/drivers/base/regmap/regcache-rbtree.c b/drivers/base/regmap/regcache-rbtree.c
index a9d17f316e55..3344b82c3799 100644
--- a/drivers/base/regmap/regcache-rbtree.c
+++ b/drivers/base/regmap/regcache-rbtree.c
@@ -184,8 +184,6 @@ static void rbtree_debugfs_init(struct regmap *map)
 static int regcache_rbtree_init(struct regmap *map)
 {
 	struct regcache_rbtree_ctx *rbtree_ctx;
-	int i;
-	int ret;
 
 	map->cache = kmalloc(sizeof *rbtree_ctx, map->alloc_flags);
 	if (!map->cache)
@@ -195,19 +193,7 @@ static int regcache_rbtree_init(struct regmap *map)
 	rbtree_ctx->root = RB_ROOT;
 	rbtree_ctx->cached_rbnode = NULL;
 
-	for (i = 0; i < map->num_reg_defaults; i++) {
-		ret = regcache_rbtree_write(map,
-					    map->reg_defaults[i].reg,
-					    map->reg_defaults[i].def);
-		if (ret)
-			goto err;
-	}
-
 	return 0;
-
-err:
-	regcache_rbtree_exit(map);
-	return ret;
 }
 
 static int regcache_rbtree_exit(struct regmap *map)
@@ -239,6 +225,22 @@ static int regcache_rbtree_exit(struct regmap *map)
 	return 0;
 }
 
+static int regcache_rbtree_populate(struct regmap *map)
+{
+	unsigned int i;
+	int ret;
+
+	for (i = 0; i < map->num_reg_defaults; i++) {
+		ret = regcache_rbtree_write(map,
+					    map->reg_defaults[i].reg,
+					    map->reg_defaults[i].def);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
 static int regcache_rbtree_read(struct regmap *map,
 				unsigned int reg, unsigned int *value)
 {
@@ -546,6 +548,7 @@ struct regcache_ops regcache_rbtree_ops = {
 	.name = "rbtree",
 	.init = regcache_rbtree_init,
 	.exit = regcache_rbtree_exit,
+	.populate = regcache_rbtree_populate,
 #ifdef CONFIG_DEBUG_FS
 	.debugfs_init = rbtree_debugfs_init,
 #endif
-- 
2.50.1

Re: [PATCH v3 2/5] regcache: rbtree: Split ->populate() from ->init()

[Charles Keepax]

On Fri, Oct 31, 2025 at 09:03:17AM +0100, Andy Shevchenko wrote:
> Split ->populate() implementation from ->init() code.
> This decoupling will help for the further changes.
> 
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---

Reviewed-by: Charles Keepax <ckeepax@opensource.cirrus.com>

Thanks,
Charles

[PATCH v3 3/5] regcache: flat: Remove unneeded check and error message for -ENOMEM

[Andy Shevchenko]

There is a convention in the kernel to avoid error messages
in the cases of -ENOMEM errors. Besides that, the idea behind
using struct_size() and other macros from overflow.h is
to saturate the size that the following allocation call will
definitely fail, hence the check and the error messaging added
in regcache_flat_init() are redundant. Remove them.

Acked-by: Sander Vanheule <sander@svanheule.net>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 drivers/base/regmap/regcache-flat.c | 10 +---------
 1 file changed, 1 insertion(+), 9 deletions(-)

diff --git a/drivers/base/regmap/regcache-flat.c b/drivers/base/regmap/regcache-flat.c
index 3b9235bb8313..bacb7137092f 100644
--- a/drivers/base/regmap/regcache-flat.c
+++ b/drivers/base/regmap/regcache-flat.c
@@ -30,7 +30,6 @@ struct regcache_flat_data {
 static int regcache_flat_init(struct regmap *map)
 {
 	int i;
-	size_t cache_data_size;
 	unsigned int cache_size;
 	struct regcache_flat_data *cache;
 
@@ -38,14 +37,7 @@ static int regcache_flat_init(struct regmap *map)
 		return -EINVAL;
 
 	cache_size = regcache_flat_get_index(map, map->max_register) + 1;
-	cache_data_size = struct_size(cache, data, cache_size);
-
-	if (cache_data_size == SIZE_MAX) {
-		dev_err(map->dev, "cannot allocate regmap cache");
-		return -ENOMEM;
-	}
-
-	cache = kzalloc(cache_data_size, map->alloc_flags);
+	cache = kzalloc(struct_size(cache, data, cache_size), map->alloc_flags);
 	if (!cache)
 		return -ENOMEM;
 
-- 
2.50.1

Re: [PATCH v3 3/5] regcache: flat: Remove unneeded check and error message for -ENOMEM

[Charles Keepax]

On Fri, Oct 31, 2025 at 09:03:18AM +0100, Andy Shevchenko wrote:
> There is a convention in the kernel to avoid error messages
> in the cases of -ENOMEM errors. Besides that, the idea behind
> using struct_size() and other macros from overflow.h is
> to saturate the size that the following allocation call will
> definitely fail, hence the check and the error messaging added
> in regcache_flat_init() are redundant. Remove them.
> 
> Acked-by: Sander Vanheule <sander@svanheule.net>
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---

Reviewed-by: Charles Keepax <ckeepax@opensource.cirrus.com>

Thanks,
Charles

[PATCH v3 4/5] regcache: flat: Split ->populate() from ->init()

[Andy Shevchenko]

Split ->populate() implementation from ->init() code.
This decoupling will help for the further changes.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 drivers/base/regmap/regcache-flat.c | 27 ++++++++++++++++++---------
 1 file changed, 18 insertions(+), 9 deletions(-)

diff --git a/drivers/base/regmap/regcache-flat.c b/drivers/base/regmap/regcache-flat.c
index bacb7137092f..53cc59c84e2f 100644
--- a/drivers/base/regmap/regcache-flat.c
+++ b/drivers/base/regmap/regcache-flat.c
@@ -29,7 +29,6 @@ struct regcache_flat_data {
 
 static int regcache_flat_init(struct regmap *map)
 {
-	int i;
 	unsigned int cache_size;
 	struct regcache_flat_data *cache;
 
@@ -47,14 +46,6 @@ static int regcache_flat_init(struct regmap *map)
 
 	map->cache = cache;
 
-	for (i = 0; i < map->num_reg_defaults; i++) {
-		unsigned int reg = map->reg_defaults[i].reg;
-		unsigned int index = regcache_flat_get_index(map, reg);
-
-		cache->data[index] = map->reg_defaults[i].def;
-		__set_bit(index, cache->valid);
-	}
-
 	return 0;
 
 err_free:
@@ -75,6 +66,22 @@ static int regcache_flat_exit(struct regmap *map)
 	return 0;
 }
 
+static int regcache_flat_populate(struct regmap *map)
+{
+	struct regcache_flat_data *cache = map->cache;
+	unsigned int i;
+
+	for (i = 0; i < map->num_reg_defaults; i++) {
+		unsigned int reg = map->reg_defaults[i].reg;
+		unsigned int index = regcache_flat_get_index(map, reg);
+
+		cache->data[index] = map->reg_defaults[i].def;
+		__set_bit(index, cache->valid);
+	}
+
+	return 0;
+}
+
 static int regcache_flat_read(struct regmap *map,
 			      unsigned int reg, unsigned int *value)
 {
@@ -134,6 +141,7 @@ struct regcache_ops regcache_flat_ops = {
 	.name = "flat",
 	.init = regcache_flat_init,
 	.exit = regcache_flat_exit,
+	.populate = regcache_flat_populate,
 	.read = regcache_flat_read,
 	.write = regcache_flat_write,
 };
@@ -143,6 +151,7 @@ struct regcache_ops regcache_flat_sparse_ops = {
 	.name = "flat-sparse",
 	.init = regcache_flat_init,
 	.exit = regcache_flat_exit,
+	.populate = regcache_flat_populate,
 	.read = regcache_flat_sparse_read,
 	.write = regcache_flat_write,
 	.drop = regcache_flat_drop,
-- 
2.50.1

Re: [PATCH v3 4/5] regcache: flat: Split ->populate() from ->init()

[Charles Keepax]

On Fri, Oct 31, 2025 at 09:03:19AM +0100, Andy Shevchenko wrote:
> Split ->populate() implementation from ->init() code.
> This decoupling will help for the further changes.
> 
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---

Reviewed-by: Charles Keepax <ckeepax@opensource.cirrus.com>

Thanks,
Charles

[PATCH v3 5/5] regcache: maple: Split ->populate() from ->init()

[Andy Shevchenko]

Split ->populate() implementation from ->init() code.
This decoupling will help for the further changes.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 drivers/base/regmap/regcache-maple.c | 47 +++++++++++++---------------
 1 file changed, 21 insertions(+), 26 deletions(-)

diff --git a/drivers/base/regmap/regcache-maple.c b/drivers/base/regmap/regcache-maple.c
index 2319c30283a6..ca1c72b68f31 100644
--- a/drivers/base/regmap/regcache-maple.c
+++ b/drivers/base/regmap/regcache-maple.c
@@ -289,6 +289,23 @@ static int regcache_maple_sync(struct regmap *map, unsigned int min,
 	return ret;
 }
 
+static int regcache_maple_init(struct regmap *map)
+{
+	struct maple_tree *mt;
+
+	mt = kmalloc(sizeof(*mt), map->alloc_flags);
+	if (!mt)
+		return -ENOMEM;
+	map->cache = mt;
+
+	mt_init(mt);
+
+	if (!mt_external_lock(mt) && map->lock_key)
+		lockdep_set_class_and_subclass(&mt->ma_lock, map->lock_key, 1);
+
+	return 0;
+}
+
 static int regcache_maple_exit(struct regmap *map)
 {
 	struct maple_tree *mt = map->cache;
@@ -340,26 +357,12 @@ static int regcache_maple_insert_block(struct regmap *map, int first,
 	return ret;
 }
 
-static int regcache_maple_init(struct regmap *map)
+static int regcache_maple_populate(struct regmap *map)
 {
-	struct maple_tree *mt;
 	int i;
 	int ret;
 	int range_start;
 
-	mt = kmalloc(sizeof(*mt), map->alloc_flags);
-	if (!mt)
-		return -ENOMEM;
-	map->cache = mt;
-
-	mt_init(mt);
-
-	if (!mt_external_lock(mt) && map->lock_key)
-		lockdep_set_class_and_subclass(&mt->ma_lock, map->lock_key, 1);
-
-	if (!map->num_reg_defaults)
-		return 0;
-
 	range_start = 0;
 
 	/* Scan for ranges of contiguous registers */
@@ -369,23 +372,14 @@ static int regcache_maple_init(struct regmap *map)
 			ret = regcache_maple_insert_block(map, range_start,
 							  i - 1);
 			if (ret != 0)
-				goto err;
+				return ret;
 
 			range_start = i;
 		}
 	}
 
 	/* Add the last block */
-	ret = regcache_maple_insert_block(map, range_start,
-					  map->num_reg_defaults - 1);
-	if (ret != 0)
-		goto err;
-
-	return 0;
-
-err:
-	regcache_maple_exit(map);
-	return ret;
+	return regcache_maple_insert_block(map, range_start, map->num_reg_defaults - 1);
 }
 
 struct regcache_ops regcache_maple_ops = {
@@ -393,6 +387,7 @@ struct regcache_ops regcache_maple_ops = {
 	.name = "maple",
 	.init = regcache_maple_init,
 	.exit = regcache_maple_exit,
+	.populate = regcache_maple_populate,
 	.read = regcache_maple_read,
 	.write = regcache_maple_write,
 	.drop = regcache_maple_drop,
-- 
2.50.1

Re: [PATCH v3 5/5] regcache: maple: Split ->populate() from ->init()

[Charles Keepax]

On Fri, Oct 31, 2025 at 09:03:20AM +0100, Andy Shevchenko wrote:
> Split ->populate() implementation from ->init() code.
> This decoupling will help for the further changes.
> 
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---

Reviewed-by: Charles Keepax <ckeepax@opensource.cirrus.com>

Thanks,
Charles

Re: [PATCH v3 0/5] regcache: Split out ->populate() and use it

[Mark Brown]

On Fri, 31 Oct 2025 09:03:15 +0100, Andy Shevchenko wrote:
> This is a refactoring series to decouple cache initialisation and population.
> On its own it has no functional impact but will be used in the further
> development. Besides that I found this split useful on its own (from the design
> perspective). That's why I decided to send it out as is separately from a bigger
> (and ongoing) work.
> 
> Changelog v3:
> - added missing space (Mark)
> - collected tags (Sander)
> 
> [...]

Applied to

   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap.git for-next

Thanks!

[1/5] regcache: Add ->populate() callback to separate from ->init()
      commit: 94a3a95f03154d8d4c6206950a7f6ef9a30baec6
[2/5] regcache: rbtree: Split ->populate() from ->init()
      commit: bda6f8749c8e0b10f083dc7a1edf169f349fb776
[3/5] regcache: flat: Remove unneeded check and error message for -ENOMEM
      commit: 27fef3048fe95934f6f2f87341eb33ef6581a075
[4/5] regcache: flat: Split ->populate() from ->init()
      commit: 44c1a444b030647803d900e60f5a8af31a782f0e
[5/5] regcache: maple: Split ->populate() from ->init()
      commit: ed5d499b5c9cc11dd3edae1a7a55db7dfa4f1bdc

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark