Re: [PATCH V3 04/20] nvdimm/label: Update mutex_lock() with guard(mutex)()

[Neeraj Kumar]

On 17/09/25 03:46PM, Jonathan Cameron wrote:

On Wed, 17 Sep 2025 18:59:24 +0530
Neeraj Kumar <s.nee...@samsung.com> wrote:

Updated mutex_lock() with guard(mutex)()

Say why.

Sure, I will update it in next patch-set.

Signed-off-by: Neeraj Kumar <s.nee...@samsung.com>
---
 drivers/nvdimm/label.c | 36 +++++++++++++++++-------------------
 1 file changed, 17 insertions(+), 19 deletions(-)

diff --git a/drivers/nvdimm/label.c b/drivers/nvdimm/label.c
index 668e1e146229..3235562d0e1c 100644
--- a/drivers/nvdimm/label.c
+++ b/drivers/nvdimm/label.c
@@ -948,7 +948,7 @@ static int __pmem_label_update(struct nd_region *nd_region,
                return rc;
+       list_for_each_entry(label_ent, &nd_mapping->labels, list)
+               if (!label_ent->label) {
+                       label_ent->label = nd_label;
+                       nd_label = NULL;
+                       break;

Perhaps it will change in later patches, but you could have done
                if (!label_ent->label) {
                        label_ent->label = nd_label;
                        return;
                }
as nothing else happens if we find a match.

Yes, I have updated it in later patch. I will update it here itself.

+               }
@@ -998,9 +998,8 @@ static int init_labels(struct nd_mapping *nd_mapping, int 
num_labels)
                label_ent = kzalloc(sizeof(*label_ent), GFP_KERNEL);
                if (!label_ent)
                        return -ENOMEM;
-               mutex_lock(&nd_mapping->lock);
+               guard(mutex)(&nd_mapping->lock);
                list_add_tail(&label_ent->list, &nd_mapping->labels);
-               mutex_unlock(&nd_mapping->lock);

Not sure I'd bother with cases like this but harmless.

        }

-       mutex_lock(&nd_mapping->lock);
+       guard(mutex)(&nd_mapping->lock);
        list_for_each_entry_safe(label_ent, e, &nd_mapping->labels, list) {
                struct nd_namespace_label *nd_label = label_ent->label;

@@ -1061,7 +1060,6 @@ static int del_labels(struct nd_mapping *nd_mapping, 
uuid_t *uuid)
                nd_mapping_free_labels(nd_mapping);
                dev_dbg(ndd->dev, "no more active labels\n");
        }
-       mutex_unlock(&nd_mapping->lock);

This is a potential functional change as the lock is held for longer than 
before.
nd_label_write_index is not trivial so reviewing if that is safe is not trivial.

The benefit is small so far (maybe that changes in later patches) so I would not
make the change.

Sure, I will revert it back in next patch-set

Regards,
Neeraj

        return nd_label_write_index(ndd, ndd->ns_next,
                        nd_inc_seq(__le32_to_cpu(nsindex->seq)), 0);