Now when we use FS_VE_MOUNT for nfs instead of FS_USERNS_MOUNT we
already have a check that nfs can be mounted only from initial userns.

https://jira.sw.ru/browse/PSBM-131949
Fixes: f1c852869b95 ("ve/fs/nfs: NFS containerization")
Signed-off-by: Pavel Tikhomirov <[email protected]>
---
 fs/nfs/fs_context.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/fs/nfs/fs_context.c b/fs/nfs/fs_context.c
index 9eed176870ee..68efb1e9b143 100644
--- a/fs/nfs/fs_context.c
+++ b/fs/nfs/fs_context.c
@@ -1364,8 +1364,6 @@ static int nfs_init_fs_context(struct fs_context *fc)
 
        if (!(get_exec_env()->features & VE_FEATURE_NFS))
                return -ENODEV;
-       if (!current_user_ns_initial())
-               return -EPERM;
 
        ctx = kzalloc(sizeof(struct nfs_fs_context), GFP_KERNEL);
        if (unlikely(!ctx))
-- 
2.31.1

_______________________________________________
Devel mailing list
[email protected]
https://lists.openvz.org/mailman/listinfo/devel

Reply via email to