[v5] dax/hmem, cxl: Coordinate Soft Reserved handling with CXL and HMEM

[PATCH v5 7/7] dax/hmem: Reintroduce Soft Reserved ranges back into the iomem tree

Posted by Smita Koralahalli 2 weeks, 4 days ago

Reworked from a patch by Alison Schofield <alison.schofield@intel.com>

Reintroduce Soft Reserved range into the iomem_resource tree for HMEM
to consume.

This restores visibility in /proc/iomem for ranges actively in use, while
avoiding the early-boot conflicts that occurred when Soft Reserved was
published into iomem before CXL window and region discovery.

Link: https://lore.kernel.org/linux-cxl/29312c0765224ae76862d59a17748c8188fb95f1.1692638817.git.alison.schofield@intel.com/
Co-developed-by: Alison Schofield <alison.schofield@intel.com>
Signed-off-by: Alison Schofield <alison.schofield@intel.com>
Co-developed-by: Zhijian Li <lizhijian@fujitsu.com>
Signed-off-by: Zhijian Li <lizhijian@fujitsu.com>
Signed-off-by: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dax/hmem/hmem.c | 32 +++++++++++++++++++++++++++++++-
 1 file changed, 31 insertions(+), 1 deletion(-)

diff --git a/drivers/dax/hmem/hmem.c b/drivers/dax/hmem/hmem.c
index bcb57d8678d7..f3ef4faf158f 100644
--- a/drivers/dax/hmem/hmem.c
+++ b/drivers/dax/hmem/hmem.c
@@ -64,6 +64,34 @@ struct dax_defer_work {
 	struct work_struct work;
 };
 
+static void remove_soft_reserved(void *r)
+{
+	remove_resource(r);
+	kfree(r);
+}
+
+static int add_soft_reserve_into_iomem(struct device *host,
+				       const struct resource *res)
+{
+	struct resource *soft __free(kfree) =
+		kzalloc(sizeof(*soft), GFP_KERNEL);
+	int rc;
+
+	if (!soft)
+		return -ENOMEM;
+
+	*soft = DEFINE_RES_NAMED_DESC(res->start, (res->end - res->start + 1),
+				      "Soft Reserved", IORESOURCE_MEM,
+				      IORES_DESC_SOFT_RESERVED);
+
+	rc = insert_resource(&iomem_resource, soft);
+	if (rc)
+		return rc;
+
+	return devm_add_action_or_reset(host, remove_soft_reserved,
+					no_free_ptr(soft));
+}
+
 static int hmem_register_device(struct device *host, int target_nid,
 				const struct resource *res)
 {
@@ -94,7 +122,9 @@ static int hmem_register_device(struct device *host, int target_nid,
 	if (rc != REGION_INTERSECTS)
 		return 0;
 
-	/* TODO: Add Soft-Reserved memory back to iomem */
+	rc = add_soft_reserve_into_iomem(host, res);
+	if (rc)
+		return rc;
 
 	id = memregion_alloc(GFP_KERNEL);
 	if (id < 0) {
-- 
2.17.1

Re: [PATCH v5 7/7] dax/hmem: Reintroduce Soft Reserved ranges back into the iomem tree

Posted by Jonathan Cameron 2 weeks, 4 days ago

On Thu, 22 Jan 2026 04:55:43 +0000
Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com> wrote:

> Reworked from a patch by Alison Schofield <alison.schofield@intel.com>
> 
> Reintroduce Soft Reserved range into the iomem_resource tree for HMEM
> to consume.
> 
> This restores visibility in /proc/iomem for ranges actively in use, while
> avoiding the early-boot conflicts that occurred when Soft Reserved was
> published into iomem before CXL window and region discovery.
> 
> Link: https://lore.kernel.org/linux-cxl/29312c0765224ae76862d59a17748c8188fb95f1.1692638817.git.alison.schofield@intel.com/
> Co-developed-by: Alison Schofield <alison.schofield@intel.com>
> Signed-off-by: Alison Schofield <alison.schofield@intel.com>
> Co-developed-by: Zhijian Li <lizhijian@fujitsu.com>
> Signed-off-by: Zhijian Li <lizhijian@fujitsu.com>
> Signed-off-by: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com>
> Reviewed-by: Dave Jiang <dave.jiang@intel.com>
> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com>
> Reviewed-by: Dan Williams <dan.j.williams@intel.com>
A few minor things from a fresh read.
> ---
>  drivers/dax/hmem/hmem.c | 32 +++++++++++++++++++++++++++++++-
>  1 file changed, 31 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/dax/hmem/hmem.c b/drivers/dax/hmem/hmem.c
> index bcb57d8678d7..f3ef4faf158f 100644
> --- a/drivers/dax/hmem/hmem.c
> +++ b/drivers/dax/hmem/hmem.c
> @@ -64,6 +64,34 @@ struct dax_defer_work {
>  	struct work_struct work;
>  };
>  
> +static void remove_soft_reserved(void *r)
> +{
> +	remove_resource(r);
> +	kfree(r);
> +}
> +
> +static int add_soft_reserve_into_iomem(struct device *host,
> +				       const struct resource *res)
> +{
> +	struct resource *soft __free(kfree) =
> +		kzalloc(sizeof(*soft), GFP_KERNEL);

On one line. I think it's exactly 80 chars.

> +	int rc;
> +
The declaration and check should be together. For __free stuff inline declarations
are preferred.

You fully assign it so kmalloc is all that's needed.
	
	
	struct resource *soft __free(kfree) = kzalloc(sizeof(*soft), GFP_KERNEL);
	if (!soft)

If not, just switch the two declarations.

> +	if (!soft)
> +		return -ENOMEM;
> +
> +	*soft = DEFINE_RES_NAMED_DESC(res->start, (res->end - res->start + 1),
> +				      "Soft Reserved", IORESOURCE_MEM,
> +				      IORES_DESC_SOFT_RESERVED);
> +
> +	rc = insert_resource(&iomem_resource, soft);
> +	if (rc)
> +		return rc;
> +
> +	return devm_add_action_or_reset(host, remove_soft_reserved,
> +					no_free_ptr(soft));
> +}
> +
>  static int hmem_register_device(struct device *host, int target_nid,
>  				const struct resource *res)
>  {
> @@ -94,7 +122,9 @@ static int hmem_register_device(struct device *host, int target_nid,
>  	if (rc != REGION_INTERSECTS)
>  		return 0;
>  
> -	/* TODO: Add Soft-Reserved memory back to iomem */
> +	rc = add_soft_reserve_into_iomem(host, res);
> +	if (rc)
> +		return rc;
>  
>  	id = memregion_alloc(GFP_KERNEL);
>  	if (id < 0) {

Re: [PATCH v5 7/7] dax/hmem: Reintroduce Soft Reserved ranges back into the iomem tree

Posted by Koralahalli Channabasappa, Smita 1 week, 5 days ago

On 1/22/2026 8:39 AM, Jonathan Cameron wrote:
> On Thu, 22 Jan 2026 04:55:43 +0000
> Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com> wrote:
> 
>> Reworked from a patch by Alison Schofield <alison.schofield@intel.com>
>>
>> Reintroduce Soft Reserved range into the iomem_resource tree for HMEM
>> to consume.
>>
>> This restores visibility in /proc/iomem for ranges actively in use, while
>> avoiding the early-boot conflicts that occurred when Soft Reserved was
>> published into iomem before CXL window and region discovery.
>>
>> Link: https://lore.kernel.org/linux-cxl/29312c0765224ae76862d59a17748c8188fb95f1.1692638817.git.alison.schofield@intel.com/
>> Co-developed-by: Alison Schofield <alison.schofield@intel.com>
>> Signed-off-by: Alison Schofield <alison.schofield@intel.com>
>> Co-developed-by: Zhijian Li <lizhijian@fujitsu.com>
>> Signed-off-by: Zhijian Li <lizhijian@fujitsu.com>
>> Signed-off-by: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com>
>> Reviewed-by: Dave Jiang <dave.jiang@intel.com>
>> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com>
>> Reviewed-by: Dan Williams <dan.j.williams@intel.com>
> A few minor things from a fresh read.
>> ---
>>   drivers/dax/hmem/hmem.c | 32 +++++++++++++++++++++++++++++++-
>>   1 file changed, 31 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/dax/hmem/hmem.c b/drivers/dax/hmem/hmem.c
>> index bcb57d8678d7..f3ef4faf158f 100644
>> --- a/drivers/dax/hmem/hmem.c
>> +++ b/drivers/dax/hmem/hmem.c
>> @@ -64,6 +64,34 @@ struct dax_defer_work {
>>   	struct work_struct work;
>>   };
>>   
>> +static void remove_soft_reserved(void *r)
>> +{
>> +	remove_resource(r);
>> +	kfree(r);
>> +}
>> +
>> +static int add_soft_reserve_into_iomem(struct device *host,
>> +				       const struct resource *res)
>> +{
>> +	struct resource *soft __free(kfree) =
>> +		kzalloc(sizeof(*soft), GFP_KERNEL);
> 
> On one line. I think it's exactly 80 chars.
> 
>> +	int rc;
>> +
> The declaration and check should be together. For __free stuff inline declarations
> are preferred.
> 
> You fully assign it so kmalloc is all that's needed.
> 	
> 	
> 	struct resource *soft __free(kfree) = kzalloc(sizeof(*soft), GFP_KERNEL);
> 	if (!soft)
> 
> If not, just switch the two declarations.

Sure, I will make the changes.

Thanks
Smita
> 
>> +	if (!soft)
>> +		return -ENOMEM;
>> +
>> +	*soft = DEFINE_RES_NAMED_DESC(res->start, (res->end - res->start + 1),
>> +				      "Soft Reserved", IORESOURCE_MEM,
>> +				      IORES_DESC_SOFT_RESERVED);
>> +
>> +	rc = insert_resource(&iomem_resource, soft);
>> +	if (rc)
>> +		return rc;
>> +
>> +	return devm_add_action_or_reset(host, remove_soft_reserved,
>> +					no_free_ptr(soft));
>> +}
>> +
>>   static int hmem_register_device(struct device *host, int target_nid,
>>   				const struct resource *res)
>>   {
>> @@ -94,7 +122,9 @@ static int hmem_register_device(struct device *host, int target_nid,
>>   	if (rc != REGION_INTERSECTS)
>>   		return 0;
>>   
>> -	/* TODO: Add Soft-Reserved memory back to iomem */
>> +	rc = add_soft_reserve_into_iomem(host, res);
>> +	if (rc)
>> +		return rc;
>>   
>>   	id = memregion_alloc(GFP_KERNEL);
>>   	if (id < 0) {
>

[PATCH v5 1/7] dax/hmem: Request cxl_acpi and cxl_pci before walking Soft Reserved ranges
[PATCH v5 2/7] dax/hmem: Gate Soft Reserved deferral on DEV_DAX_CXL
[PATCH v5 3/7] cxl/region: Skip decoder reset on detach for autodiscovered regions
[PATCH v5 4/7] cxl/region: Add helper to check Soft Reserved containment by CXL regions
[PATCH v5 5/7] dax: Introduce dax_cxl_mode for CXL coordination
[PATCH v5 6/7] dax/hmem, cxl: Defer and resolve ownership of Soft Reserved memory ranges
[PATCH v5 7/7] dax/hmem: Reintroduce Soft Reserved ranges back into the iomem tree