[PATCH v4] iommu: simplify list initialization across the iommu subsystem

[PATCH v4] iommu: simplify list initialization across the iommu subsystem

[Can Peng]

Replace separate list head declarations and INIT_LIST_HEAD() calls with
the combined LIST_HEAD() macro throughout the iommu directory.
Using LIST_HEAD() merges declaration and initialization into a single
statement, improving code readability and reducing boilerplate without
altering any functionality.

No functional change intended.

Signed-off-by: Can Peng <pengcan@kylinos.cn>
---
Changes since previous version:
Rebased on top of iommu/next as requested by Vasant Hegde.
Adjusted commit message to be shorter and clearer, following Markus Elfring's suggestion.
Retained the LIST_HEAD() cleanup pattern consistently across the iommu directory.

Notes on prior discussion (from chat history):
The earlier version was reviewed positively:
Link: https://lore.kernel.org/all/4526ca31-71a4-4b3f-b378-32546eb8133d@amd.com/

Markus Elfring suggested using a more concise summary and
recommended the use of Coccinelle for this kind of transformation.

Best regards,
Can Peng

 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 3 +--
 drivers/iommu/arm/arm-smmu/arm-smmu.c       | 3 +--
 drivers/iommu/iommu.c                       | 6 ++----
 drivers/iommu/iommufd/eventq.c              | 3 +--
 4 files changed, 5 insertions(+), 10 deletions(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index fc5a705f6cb3..a772010b5331 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -4659,10 +4659,9 @@ static void __iomem *arm_smmu_ioremap(struct device *dev, resource_size_t start,
 
 static void arm_smmu_rmr_install_bypass_ste(struct arm_smmu_device *smmu)
 {
-	struct list_head rmr_list;
+	LIST_HEAD(rmr_list);
 	struct iommu_resv_region *e;
 
-	INIT_LIST_HEAD(&rmr_list);
 	iort_get_rmr_sids(dev_fwnode(smmu->dev), &rmr_list);
 
 	list_for_each_entry(e, &rmr_list, list) {
diff --git a/drivers/iommu/arm/arm-smmu/arm-smmu.c b/drivers/iommu/arm/arm-smmu/arm-smmu.c
index 5e690cf85ec9..d7e58ac1e14a 100644
--- a/drivers/iommu/arm/arm-smmu/arm-smmu.c
+++ b/drivers/iommu/arm/arm-smmu/arm-smmu.c
@@ -2072,12 +2072,11 @@ static int arm_smmu_device_dt_probe(struct arm_smmu_device *smmu,
 
 static void arm_smmu_rmr_install_bypass_smr(struct arm_smmu_device *smmu)
 {
-	struct list_head rmr_list;
+	LIST_HEAD(rmr_list);
 	struct iommu_resv_region *e;
 	int idx, cnt = 0;
 	u32 reg;
 
-	INIT_LIST_HEAD(&rmr_list);
 	iort_get_rmr_sids(dev_fwnode(smmu->dev), &rmr_list);
 
 	/*
diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index 4926a43118e6..2652ab6f5dde 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -932,7 +932,7 @@ int iommu_get_group_resv_regions(struct iommu_group *group,
 
 	mutex_lock(&group->mutex);
 	for_each_group_device(group, device) {
-		struct list_head dev_resv_regions;
+		LIST_HEAD(dev_resv_regions);
 
 		/*
 		 * Non-API groups still expose reserved_regions in sysfs,
@@ -941,7 +941,6 @@ int iommu_get_group_resv_regions(struct iommu_group *group,
 		if (!dev_has_iommu(device->dev))
 			break;
 
-		INIT_LIST_HEAD(&dev_resv_regions);
 		iommu_get_resv_regions(device->dev, &dev_resv_regions);
 		ret = iommu_insert_device_resv_regions(&dev_resv_regions, head);
 		iommu_put_resv_regions(device->dev, &dev_resv_regions);
@@ -957,10 +956,9 @@ static ssize_t iommu_group_show_resv_regions(struct iommu_group *group,
 					     char *buf)
 {
 	struct iommu_resv_region *region, *next;
-	struct list_head group_resv_regions;
+	LIST_HEAD(group_resv_regions);
 	int offset = 0;
 
-	INIT_LIST_HEAD(&group_resv_regions);
 	iommu_get_group_resv_regions(group, &group_resv_regions);
 
 	list_for_each_entry_safe(region, next, &group_resv_regions, list) {
diff --git a/drivers/iommu/iommufd/eventq.c b/drivers/iommu/iommufd/eventq.c
index e23d9ee4fe38..95a9ee581072 100644
--- a/drivers/iommu/iommufd/eventq.c
+++ b/drivers/iommu/iommufd/eventq.c
@@ -21,12 +21,11 @@ void iommufd_auto_response_faults(struct iommufd_hw_pagetable *hwpt,
 {
 	struct iommufd_fault *fault = hwpt->fault;
 	struct iopf_group *group, *next;
-	struct list_head free_list;
+	LIST_HEAD(free_list);
 	unsigned long index;
 
 	if (!fault || !handle)
 		return;
-	INIT_LIST_HEAD(&free_list);
 
 	mutex_lock(&fault->mutex);
 	spin_lock(&fault->common.lock);
-- 
2.25.1

Re: [PATCH v4] iommu: simplify list initialization across the iommu subsystem

[Markus Elfring]

> Replace separate list head declarations and INIT_LIST_HEAD() calls with
> the combined LIST_HEAD() macro throughout the iommu directory.
> Using LIST_HEAD() merges declaration and initialization into a single
> statement, improving code readability and reducing boilerplate without
> altering any functionality.
…
> Markus Elfring suggested using a more concise summary and
> recommended the use of Coccinelle for this kind of transformation.

Some development tools can help to achieve desirable changes in more convenient ways
and that possible adjustments would become more complete eventually.


…
> +++ b/drivers/iommu/iommu.c
> @@ -932,7 +932,7 @@ int iommu_get_group_resv_regions(struct iommu_group *group,
>  
>  	mutex_lock(&group->mutex);
>  	for_each_group_device(group, device) {
> -		struct list_head dev_resv_regions;
> +		LIST_HEAD(dev_resv_regions);
>  
>  		/*
>  		 * Non-API groups still expose reserved_regions in sysfs,
> @@ -941,7 +941,6 @@ int iommu_get_group_resv_regions(struct iommu_group *group,
>  		if (!dev_has_iommu(device->dev))
>  			break;
>  
> -		INIT_LIST_HEAD(&dev_resv_regions);
>  		iommu_get_resv_regions(device->dev, &dev_resv_regions);
>  		ret = iommu_insert_device_resv_regions(&dev_resv_regions, head);
>  		iommu_put_resv_regions(device->dev, &dev_resv_regions);
…
> +++ b/drivers/iommu/iommufd/eventq.c
> @@ -21,12 +21,11 @@ void iommufd_auto_response_faults(struct iommufd_hw_pagetable *hwpt,
>  {
>  	struct iommufd_fault *fault = hwpt->fault;
>  	struct iopf_group *group, *next;
> -	struct list_head free_list;
> +	LIST_HEAD(free_list);
>  	unsigned long index;
>  
>  	if (!fault || !handle)
>  		return;
> -	INIT_LIST_HEAD(&free_list);
>  
>  	mutex_lock(&fault->mutex);
…

Does anything hinder to apply LIST_HEAD() calls at other source code positions?

May variable scopes be reduced for more use cases?

Regards,
Markus