[PATCH v2 02/13] storage-by-alias: drop fake cdev
Sascha Hauer
s.hauer at pengutronix.de
Tue Dec 9 04:51:21 PST 2025
storage-by-alias creates its own cdev, but that is never used. The only
purpose is to create a cdev with its own device node which otherwise
links to the original cdev. This means we can drop all the cdev
operations and use the newly created devfs_create_link_node() with the
same effect.
Signed-off-by: Sascha Hauer <s.hauer at pengutronix.de>
---
drivers/misc/storage-by-alias.c | 120 +++-------------------------------------
1 file changed, 7 insertions(+), 113 deletions(-)
diff --git a/drivers/misc/storage-by-alias.c b/drivers/misc/storage-by-alias.c
index 2c498e0d0483ca71093612382c78372df399c437..6c759867ccbb7a7b4bc922152699eb346d42d271 100644
--- a/drivers/misc/storage-by-alias.c
+++ b/drivers/misc/storage-by-alias.c
@@ -24,102 +24,8 @@ static LIST_HEAD(sba_list);
struct sba {
char *alias;
struct device *dev;
- struct cdev *rcdev;
- struct cdev cdev;
struct list_head list;
-};
-
-static ssize_t sba_read(struct cdev *cdev, void *buf, size_t count, loff_t offset, ulong flags)
-{
- struct sba *sba = cdev->priv;
-
- return cdev_read(sba->rcdev, buf, count, offset, flags);
-}
-
-static ssize_t sba_write(struct cdev *cdev, const void *buf, size_t count, loff_t offset, ulong flags)
-{
- struct sba *sba = cdev->priv;
-
- return cdev_write(sba->rcdev, buf, count, offset, flags);
-}
-
-static int sba_ioctl(struct cdev *cdev, unsigned int request, void *buf)
-{
- struct sba *sba = cdev->priv;
-
- return cdev_ioctl(sba->rcdev, request, buf);
-}
-
-static int sba_open(struct cdev *cdev, unsigned long flags)
-{
- struct sba *sba = cdev->priv;
-
- return cdev_open(sba->rcdev, flags);
-}
-
-static int sba_close(struct cdev *cdev)
-{
- struct sba *sba = cdev->priv;
-
- cdev_close(sba->rcdev);
-
- return 0;
-}
-
-static int sba_flush(struct cdev *cdev)
-{
- struct sba *sba = cdev->priv;
-
- return cdev_flush(sba->rcdev);
-}
-
-static int sba_erase(struct cdev *cdev, loff_t count, loff_t offset)
-{
- struct sba *sba = cdev->priv;
-
- return cdev_erase(sba->rcdev, count, offset);
-}
-
-static int sba_protect(struct cdev *cdev, size_t count, loff_t offset, int prot)
-{
- struct sba *sba = cdev->priv;
-
- return cdev_protect(sba->rcdev, count, offset, prot);
-}
-
-static int sba_discard_range(struct cdev *cdev, loff_t count, loff_t offset)
-{
- struct sba *sba = cdev->priv;
-
- return cdev_discard_range(sba->rcdev, count, offset);
-}
-
-static int sba_memmap(struct cdev *cdev, void **map, int flags)
-{
- struct sba *sba = cdev->priv;
-
- return cdev_memmap(sba->rcdev, map, flags);
-}
-
-static int sba_truncate(struct cdev *cdev, size_t size)
-{
- struct sba *sba = cdev->priv;
-
- return cdev_truncate(sba->rcdev, size);
-}
-
-static struct cdev_operations sba_ops = {
- .read = sba_read,
- .write = sba_write,
- .ioctl = sba_ioctl,
- .open = sba_open,
- .close = sba_close,
- .flush = sba_flush,
- .erase = sba_erase,
- .protect = sba_protect,
- .discard_range = sba_discard_range,
- .memmap = sba_memmap,
- .truncate = sba_truncate,
+ bool populated;
};
static int sba_add_partitions(struct cdev *rcdev, void *data)
@@ -127,29 +33,17 @@ static int sba_add_partitions(struct cdev *rcdev, void *data)
struct sba *sba = data;
int ret;
- dev_dbg(sba->dev, "Adding %s -> %s\n", sba->alias, rcdev->name);
-
- if (sba->rcdev)
+ if (sba->populated)
return 0;
- sba->rcdev = rcdev;
- sba->cdev.link = rcdev;
- sba->cdev.name = sba->alias;
- sba->cdev.size = rcdev->size;
- sba->cdev.ops = &sba_ops;
- sba->cdev.dev = sba->dev;
- sba->cdev.priv = sba;
+ dev_info(sba->dev, "Adding %s -> %s\n", sba->alias, rcdev->name);
- if (rcdev->flags & DEVFS_WRITE_AUTOERASE)
- sba->cdev.flags |= DEVFS_WRITE_AUTOERASE;
+ ret = devfs_create_link_node(rcdev, sba->alias, sba->dev->device_node);
+ if (ret)
+ return ret;
- ret = devfs_create(&sba->cdev);
- if (ret) {
- dev_err(sba->dev, "Failed to create cdev: %pe\n", ERR_PTR(ret));
- return 0;
- }
+ sba->populated = true;
- of_parse_partitions(&sba->cdev, sba->dev->of_node);
return 0;
}
--
2.47.3
More information about the barebox
mailing list