[PATCH v6 0/5] io_uring: add xattr support

[PATCH v6 0/5] io_uring: add xattr support

[Stefan Roesch]

This adds the xattr support to io_uring. The intent is to have a more
complete support for file operations in io_uring.

This change adds support for the following functions to io_uring:
- fgetxattr
- fsetxattr
- getxattr
- setxattr

Patch 1: fs: split off do_user_path_at_empty from user_path_at_empty()
  This splits off a new function do_user_path_at_empty from
  user_path_at_empty that is based on filename and not on a
  user-specified string.

Patch 2: fs: split off setxattr_setup function from setxattr
  Split off the setup part of the setxattr function.

Patch 3: fs: split off do_getxattr from getxattr
  Split of the do_getxattr part from getxattr. This will
  allow it to be invoked it from io_uring.

Patch 4: io_uring: add fsetxattr and setxattr support
  This adds new functions to support the fsetxattr and setxattr
  functions.

Patch 5: io_uring: add fgetxattr and getxattr support
  This adds new functions to support the fgetxattr and getxattr
  functions.


There are two additional patches:
  liburing: Add support for xattr api's.
            This also includes the tests for the new code.
  xfstests: Add support for io_uring xattr support.


V6: - reverted addition of kname array to xattr_ctx structure
      Adding the kname array increases the io_kiocb beyond 64 bytes
      (increases it to 224 bytes). We try hard to limit it to 64 bytes.
      Keeping the original interface also is a bit more efficient.
    - rebased on for-5.17/io_uring-getdents64
V5: - add kname array to xattr_ctx structure
V4: - rebased patch series
V3: - remove req->file checks in prep functions
    - change size parameter in do_xattr
V2: - split off function do_user_path_empty instead of changing
      the function signature of user_path_at
    - Fix datatype size problem in do_getxattr



Stefan Roesch (5):
  fs: split off do_user_path_at_empty from user_path_at_empty()
  fs: split off setxattr_setup function from setxattr
  fs: split off do_getxattr from getxattr
  io_uring: add fsetxattr and setxattr support
  io_uring: add fgetxattr and getxattr support

 fs/internal.h                 |  23 +++
 fs/io_uring.c                 | 318 ++++++++++++++++++++++++++++++++++
 fs/namei.c                    |  10 +-
 fs/xattr.c                    | 107 ++++++++----
 include/linux/namei.h         |   2 +
 include/uapi/linux/io_uring.h |   8 +-
 6 files changed, 428 insertions(+), 40 deletions(-)


base-commit: b4518682080d3a1cdd6ea45a54ff6772b8b2797a
-- 
2.30.2

[PATCH v6 1/5] fs: split off do_user_path_at_empty from user_path_at_empty()

[Stefan Roesch]

This splits off a do_user_path_at_empty function from the
user_path_at_empty_function. This is required so it can be
called from io_uring.

Signed-off-by: Stefan Roesch <shr@fb.com>
---
 fs/namei.c            | 10 ++++++++--
 include/linux/namei.h |  2 ++
 2 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/fs/namei.c b/fs/namei.c
index 1f9d2187c765..d988e241b32c 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -2794,12 +2794,18 @@ int path_pts(struct path *path)
 }
 #endif
 
+int do_user_path_at_empty(int dfd, struct filename *filename, unsigned int flags,
+		       struct path *path)
+{
+	return filename_lookup(dfd, filename, flags, path, NULL);
+}
+
 int user_path_at_empty(int dfd, const char __user *name, unsigned flags,
-		 struct path *path, int *empty)
+		struct path *path, int *empty)
 {
 	struct filename *filename = getname_flags(name, flags, empty);
-	int ret = filename_lookup(dfd, filename, flags, path, NULL);
 
+	int ret = do_user_path_at_empty(dfd, filename, flags, path);
 	putname(filename);
 	return ret;
 }
diff --git a/include/linux/namei.h b/include/linux/namei.h
index e89329bb3134..8f3ef38c057b 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -49,6 +49,8 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT};
 
 extern int path_pts(struct path *path);
 
+extern int do_user_path_at_empty(int dfd, struct filename *filename,
+				unsigned int flags, struct path *path);
 extern int user_path_at_empty(int, const char __user *, unsigned, struct path *, int *empty);
 
 static inline int user_path_at(int dfd, const char __user *name, unsigned flags,
-- 
2.30.2

[PATCH v6 2/5] fs: split off setxattr_setup function from setxattr

[Stefan Roesch]

This splits of the setup part of the function
setxattr in its own dedicated function called
setxattr_setup.

This makes it possible to call this function
from io_uring in the pre-processing of an
xattr request.

Signed-off-by: Stefan Roesch <shr@fb.com>
---
 fs/internal.h | 17 ++++++++++++
 fs/xattr.c    | 75 ++++++++++++++++++++++++++++++++++-----------------
 2 files changed, 67 insertions(+), 25 deletions(-)

diff --git a/fs/internal.h b/fs/internal.h
index 432ea3ce76ec..e7d5b4a9fb43 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -202,3 +202,20 @@ struct linux_dirent64;
 
 int vfs_getdents(struct file *file, struct linux_dirent64 __user *dirent,
 		 unsigned int count, loff_t *pos);
+
+ /*
+  * fs/xattr.c:
+  */
+struct xattr_ctx {
+	/* Value of attribute */
+	const void __user *value;
+	size_t size;
+	/* Attribute name */
+	char *kname;
+	int kname_sz;
+	unsigned int flags;
+};
+
+void *setxattr_setup(struct user_namespace *mnt_userns,
+		     const char __user *name,
+		     struct xattr_ctx *ctx);
diff --git a/fs/xattr.c b/fs/xattr.c
index 5c8c5175b385..79afea64d7ba 100644
--- a/fs/xattr.c
+++ b/fs/xattr.c
@@ -25,6 +25,8 @@
 
 #include <linux/uaccess.h>
 
+#include "internal.h"
+
 static const char *
 strcmp_prefix(const char *a, const char *a_prefix)
 {
@@ -539,43 +541,66 @@ EXPORT_SYMBOL_GPL(vfs_removexattr);
 /*
  * Extended attribute SET operations
  */
-static long
-setxattr(struct user_namespace *mnt_userns, struct dentry *d,
-	 const char __user *name, const void __user *value, size_t size,
-	 int flags)
+
+void *setxattr_setup(struct user_namespace *mnt_userns, const char __user *name,
+		struct xattr_ctx *ctx)
 {
-	int error;
 	void *kvalue = NULL;
-	char kname[XATTR_NAME_MAX + 1];
+	int error;
 
-	if (flags & ~(XATTR_CREATE|XATTR_REPLACE))
-		return -EINVAL;
+	if (ctx->flags & ~(XATTR_CREATE|XATTR_REPLACE))
+		return ERR_PTR(-EINVAL);
 
-	error = strncpy_from_user(kname, name, sizeof(kname));
-	if (error == 0 || error == sizeof(kname))
-		error = -ERANGE;
+	error = strncpy_from_user(ctx->kname, name, ctx->kname_sz);
+	if (error == 0 || error == ctx->kname_sz)
+		return  ERR_PTR(-ERANGE);
 	if (error < 0)
-		return error;
+		return ERR_PTR(error);
 
-	if (size) {
-		if (size > XATTR_SIZE_MAX)
-			return -E2BIG;
-		kvalue = kvmalloc(size, GFP_KERNEL);
+	if (ctx->size) {
+		if (ctx->size > XATTR_SIZE_MAX)
+			return ERR_PTR(-E2BIG);
+
+		kvalue = kvmalloc(ctx->size, GFP_KERNEL);
 		if (!kvalue)
-			return -ENOMEM;
-		if (copy_from_user(kvalue, value, size)) {
-			error = -EFAULT;
-			goto out;
+			return ERR_PTR(-ENOMEM);
+
+		if (copy_from_user(kvalue, ctx->value, ctx->size)) {
+			kvfree(kvalue);
+			return ERR_PTR(-EFAULT);
 		}
-		if ((strcmp(kname, XATTR_NAME_POSIX_ACL_ACCESS) == 0) ||
-		    (strcmp(kname, XATTR_NAME_POSIX_ACL_DEFAULT) == 0))
-			posix_acl_fix_xattr_from_user(mnt_userns, kvalue, size);
+
+		if ((strcmp(ctx->kname, XATTR_NAME_POSIX_ACL_ACCESS) == 0) ||
+		    (strcmp(ctx->kname, XATTR_NAME_POSIX_ACL_DEFAULT) == 0))
+			posix_acl_fix_xattr_from_user(mnt_userns, kvalue, ctx->size);
 	}
 
+	return kvalue;
+}
+
+static long
+setxattr(struct user_namespace *mnt_userns, struct dentry *d,
+	const char __user *name, const void __user *value, size_t size,
+	int flags)
+{
+	char kname[XATTR_NAME_MAX + 1];
+	struct xattr_ctx ctx = {
+		.value    = value,
+		.size     = size,
+		.kname    = kname,
+		.kname_sz = sizeof(kname),
+		.flags    = flags,
+	};
+	void *kvalue;
+	int error;
+
+	kvalue = setxattr_setup(mnt_userns, name, &ctx);
+	if (IS_ERR(kvalue))
+		return PTR_ERR(kvalue);
+
 	error = vfs_setxattr(mnt_userns, d, kname, kvalue, size, flags);
-out:
-	kvfree(kvalue);
 
+	kvfree(kvalue);
 	return error;
 }
 
-- 
2.30.2

[PATCH v6 3/5] fs: split off do_getxattr from getxattr

[Stefan Roesch]

This splits off do_getxattr function from the getxattr
function. This will allow io_uring to call it from its
io worker.

Signed-off-by: Stefan Roesch <shr@fb.com>
---
 fs/internal.h |  6 ++++++
 fs/xattr.c    | 32 ++++++++++++++++++++------------
 2 files changed, 26 insertions(+), 12 deletions(-)

diff --git a/fs/internal.h b/fs/internal.h
index e7d5b4a9fb43..ea0433799dbc 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -216,6 +216,12 @@ struct xattr_ctx {
 	unsigned int flags;
 };
 
+ssize_t do_getxattr(struct user_namespace *mnt_userns,
+		    struct dentry *d,
+		    const char *kname,
+		    void __user *value,
+		    size_t size);
+
 void *setxattr_setup(struct user_namespace *mnt_userns,
 		     const char __user *name,
 		     struct xattr_ctx *ctx);
diff --git a/fs/xattr.c b/fs/xattr.c
index 79afea64d7ba..a675c7f0ea0c 100644
--- a/fs/xattr.c
+++ b/fs/xattr.c
@@ -666,19 +666,12 @@ SYSCALL_DEFINE5(fsetxattr, int, fd, const char __user *, name,
 /*
  * Extended attribute GET operations
  */
-static ssize_t
-getxattr(struct user_namespace *mnt_userns, struct dentry *d,
-	 const char __user *name, void __user *value, size_t size)
+ssize_t
+do_getxattr(struct user_namespace *mnt_userns, struct dentry *d,
+	const char *kname, void __user *value, size_t size)
 {
-	ssize_t error;
 	void *kvalue = NULL;
-	char kname[XATTR_NAME_MAX + 1];
-
-	error = strncpy_from_user(kname, name, sizeof(kname));
-	if (error == 0 || error == sizeof(kname))
-		error = -ERANGE;
-	if (error < 0)
-		return error;
+	ssize_t error;
 
 	if (size) {
 		if (size > XATTR_SIZE_MAX)
@@ -702,10 +695,25 @@ getxattr(struct user_namespace *mnt_userns, struct dentry *d,
 	}
 
 	kvfree(kvalue);
-
 	return error;
 }
 
+static ssize_t
+getxattr(struct user_namespace *mnt_userns, struct dentry *d,
+	 const char __user *name, void __user *value, size_t size)
+{
+	ssize_t error;
+	char kname[XATTR_NAME_MAX + 1];
+
+	error = strncpy_from_user(kname, name, sizeof(kname));
+	if (error == 0 || error == sizeof(kname))
+		error = -ERANGE;
+	if (error < 0)
+		return error;
+
+	return do_getxattr(mnt_userns, d, kname, value, size);
+}
+
 static ssize_t path_getxattr(const char __user *pathname,
 			     const char __user *name, void __user *value,
 			     size_t size, unsigned int lookup_flags)
-- 
2.30.2

[PATCH v6 4/5] io_uring: add fsetxattr and setxattr support

[Stefan Roesch]

This adds support to io_uring for the fsetxattr and setxattr API.

Signed-off-by: Stefan Roesch <shr@fb.com>
---
 fs/io_uring.c                 | 170 ++++++++++++++++++++++++++++++++++
 include/uapi/linux/io_uring.h |   6 +-
 2 files changed, 175 insertions(+), 1 deletion(-)

diff --git a/fs/io_uring.c b/fs/io_uring.c
index c8258c784116..8b6c70d6cacc 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -82,6 +82,7 @@
 #include <linux/audit.h>
 #include <linux/security.h>
 #include <linux/atomic-ref.h>
+#include <linux/xattr.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/io_uring.h>
@@ -726,6 +727,13 @@ struct io_async_rw {
 	struct wait_page_queue		wpq;
 };
 
+struct io_xattr {
+	struct file			*file;
+	struct xattr_ctx		ctx;
+	void				*value;
+	struct filename			*filename;
+};
+
 enum {
 	REQ_F_FIXED_FILE_BIT	= IOSQE_FIXED_FILE_BIT,
 	REQ_F_IO_DRAIN_BIT	= IOSQE_IO_DRAIN_BIT,
@@ -866,6 +874,7 @@ struct io_kiocb {
 		struct io_symlink	symlink;
 		struct io_hardlink	hardlink;
 		struct io_getdents	getdents;
+		struct io_xattr		xattr;
 	};
 
 	u8				opcode;
@@ -1118,6 +1127,10 @@ static const struct io_op_def io_op_defs[] = {
 	[IORING_OP_GETDENTS] = {
 		.needs_file		= 1,
 	},
+	[IORING_OP_FSETXATTR] = {
+		.needs_file = 1
+	},
+	[IORING_OP_SETXATTR] = {},
 };
 
 /* requests with any of those set should undergo io_disarm_next() */
@@ -3887,6 +3900,144 @@ static int io_renameat(struct io_kiocb *req, unsigned int issue_flags)
 	return 0;
 }
 
+static int __io_setxattr_prep(struct io_kiocb *req,
+			const struct io_uring_sqe *sqe,
+			struct user_namespace *user_ns)
+{
+	struct io_xattr *ix = &req->xattr;
+	const char __user *name;
+	void *ret;
+
+	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
+		return -EINVAL;
+	if (unlikely(sqe->ioprio))
+		return -EINVAL;
+	if (unlikely(req->flags & REQ_F_FIXED_FILE))
+		return -EBADF;
+
+	ix->filename = NULL;
+	name = u64_to_user_ptr(READ_ONCE(sqe->addr));
+	ix->ctx.value = u64_to_user_ptr(READ_ONCE(sqe->addr2));
+	ix->ctx.size = READ_ONCE(sqe->len);
+	ix->ctx.flags = READ_ONCE(sqe->xattr_flags);
+
+	ix->ctx.kname = kmalloc(XATTR_NAME_MAX + 1, GFP_KERNEL);
+	if (!ix->ctx.kname)
+		return -ENOMEM;
+	ix->ctx.kname_sz = XATTR_NAME_MAX + 1;
+
+	ret = setxattr_setup(user_ns, name, &ix->ctx);
+	if (IS_ERR(ret)) {
+		kfree(ix->ctx.kname);
+		return PTR_ERR(ret);
+	}
+
+	ix->value = ret;
+	req->flags |= REQ_F_NEED_CLEANUP;
+	return 0;
+}
+
+static int io_setxattr_prep(struct io_kiocb *req,
+			const struct io_uring_sqe *sqe)
+{
+	struct io_xattr *ix = &req->xattr;
+	const char __user *path;
+	int ret;
+
+	ret = __io_setxattr_prep(req, sqe, current_user_ns());
+	if (ret)
+		return ret;
+
+	path = u64_to_user_ptr(READ_ONCE(sqe->addr3));
+
+	ix->filename = getname_flags(path, LOOKUP_FOLLOW, NULL);
+	if (IS_ERR(ix->filename)) {
+		ret = PTR_ERR(ix->filename);
+		ix->filename = NULL;
+	}
+
+	return ret;
+}
+
+static int io_fsetxattr_prep(struct io_kiocb *req,
+			const struct io_uring_sqe *sqe)
+{
+	return __io_setxattr_prep(req, sqe, file_mnt_user_ns(req->file));
+}
+
+static int __io_setxattr(struct io_kiocb *req, unsigned int issue_flags,
+			struct path *path)
+{
+	struct io_xattr *ix = &req->xattr;
+	int ret;
+
+	ret = mnt_want_write(path->mnt);
+	if (!ret) {
+		ret = vfs_setxattr(mnt_user_ns(path->mnt), path->dentry,
+				ix->ctx.kname, ix->value, ix->ctx.size,
+				ix->ctx.flags);
+		mnt_drop_write(path->mnt);
+	}
+
+	return ret;
+}
+
+static int io_fsetxattr(struct io_kiocb *req, unsigned int issue_flags)
+{
+	struct io_xattr *ix = &req->xattr;
+	int ret;
+
+	if (issue_flags & IO_URING_F_NONBLOCK)
+		return -EAGAIN;
+
+	ret = __io_setxattr(req, issue_flags, &req->file->f_path);
+
+	req->flags &= ~REQ_F_NEED_CLEANUP;
+	kfree(ix->ctx.kname);
+
+	if (ix->value)
+		kvfree(ix->value);
+	if (ret < 0)
+		req_set_fail(req);
+
+	io_req_complete(req, ret);
+	return 0;
+}
+
+static int io_setxattr(struct io_kiocb *req, unsigned int issue_flags)
+{
+	struct io_xattr *ix = &req->xattr;
+	unsigned int lookup_flags = LOOKUP_FOLLOW;
+	struct path path;
+	int ret;
+
+	if (issue_flags & IO_URING_F_NONBLOCK)
+		return -EAGAIN;
+
+retry:
+	ret = do_user_path_at_empty(AT_FDCWD, ix->filename, lookup_flags, &path);
+	if (!ret) {
+		ret = __io_setxattr(req, issue_flags, &path);
+		path_put(&path);
+		if (retry_estale(ret, lookup_flags)) {
+			lookup_flags |= LOOKUP_REVAL;
+			goto retry;
+		}
+	}
+	putname(ix->filename);
+
+	req->flags &= ~REQ_F_NEED_CLEANUP;
+	kfree(ix->ctx.kname);
+
+	if (ix->value)
+		kvfree(ix->value);
+	if (ret < 0)
+		req_set_fail(req);
+
+	io_req_complete(req, ret);
+	return 0;
+}
+
 static int io_unlinkat_prep(struct io_kiocb *req,
 			    const struct io_uring_sqe *sqe)
 {
@@ -6623,6 +6774,10 @@ static int io_req_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 		return io_linkat_prep(req, sqe);
 	case IORING_OP_GETDENTS:
 		return io_getdents_prep(req, sqe);
+	case IORING_OP_FSETXATTR:
+		return io_fsetxattr_prep(req, sqe);
+	case IORING_OP_SETXATTR:
+		return io_setxattr_prep(req, sqe);
 	}
 
 	printk_once(KERN_WARNING "io_uring: unhandled opcode %d\n",
@@ -6764,6 +6919,14 @@ static void io_clean_op(struct io_kiocb *req)
 			putname(req->hardlink.oldpath);
 			putname(req->hardlink.newpath);
 			break;
+		case IORING_OP_SETXATTR:
+			if (req->xattr.filename)
+				putname(req->xattr.filename);
+			fallthrough;
+		case IORING_OP_FSETXATTR:
+			kfree(req->xattr.ctx.kname);
+			kvfree(req->xattr.value);
+			break;
 		}
 	}
 	if ((req->flags & REQ_F_POLLED) && req->apoll) {
@@ -6909,6 +7072,12 @@ static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags)
 	case IORING_OP_GETDENTS:
 		ret = io_getdents(req, issue_flags);
 		break;
+	case IORING_OP_FSETXATTR:
+		ret = io_fsetxattr(req, issue_flags);
+		break;
+	case IORING_OP_SETXATTR:
+		ret = io_setxattr(req, issue_flags);
+		break;
 	default:
 		ret = -EINVAL;
 		break;
@@ -11277,6 +11446,7 @@ static int __init io_uring_init(void)
 	BUILD_BUG_SQE_ELEM(42, __u16,  personality);
 	BUILD_BUG_SQE_ELEM(44, __s32,  splice_fd_in);
 	BUILD_BUG_SQE_ELEM(44, __u32,  file_index);
+	BUILD_BUG_SQE_ELEM(48, __u64,  addr3);
 
 	BUILD_BUG_ON(sizeof(struct io_uring_files_update) !=
 		     sizeof(struct io_uring_rsrc_update));
diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h
index 57dc88db5793..c62a8bec8cd4 100644
--- a/include/uapi/linux/io_uring.h
+++ b/include/uapi/linux/io_uring.h
@@ -45,6 +45,7 @@ struct io_uring_sqe {
 		__u32		rename_flags;
 		__u32		unlink_flags;
 		__u32		hardlink_flags;
+		__u32		xattr_flags;
 	};
 	__u64	user_data;	/* data to be passed back at completion time */
 	/* pack this to avoid bogus arm OABI complaints */
@@ -60,7 +61,8 @@ struct io_uring_sqe {
 		__s32	splice_fd_in;
 		__u32	file_index;
 	};
-	__u64	__pad2[2];
+	__u64	addr3;
+	__u64	__pad2[1];
 };
 
 enum {
@@ -144,6 +146,8 @@ enum {
 	IORING_OP_SYMLINKAT,
 	IORING_OP_LINKAT,
 	IORING_OP_GETDENTS,
+	IORING_OP_FSETXATTR,
+	IORING_OP_SETXATTR,
 
 	/* this goes last, obviously */
 	IORING_OP_LAST,
-- 
2.30.2

[PATCH v6 5/5] io_uring: add fgetxattr and getxattr support

[Stefan Roesch]

This adds support to io_uring for the fgetxattr and getxattr API.

Signed-off-by: Stefan Roesch <shr@fb.com>
---
 fs/io_uring.c                 | 148 ++++++++++++++++++++++++++++++++++
 include/uapi/linux/io_uring.h |   2 +
 2 files changed, 150 insertions(+)

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 8b6c70d6cacc..4d8c99370f14 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -1131,6 +1131,10 @@ static const struct io_op_def io_op_defs[] = {
 		.needs_file = 1
 	},
 	[IORING_OP_SETXATTR] = {},
+	[IORING_OP_FGETXATTR] = {
+		.needs_file = 1
+	},
+	[IORING_OP_GETXATTR] = {},
 };
 
 /* requests with any of those set should undergo io_disarm_next() */
@@ -3900,6 +3904,133 @@ static int io_renameat(struct io_kiocb *req, unsigned int issue_flags)
 	return 0;
 }
 
+static int __io_getxattr_prep(struct io_kiocb *req,
+			      const struct io_uring_sqe *sqe)
+{
+	struct io_xattr *ix = &req->xattr;
+	const char __user *name;
+	int ret;
+
+	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
+		return -EINVAL;
+	if (unlikely(sqe->ioprio))
+		return -EINVAL;
+	if (unlikely(req->flags & REQ_F_FIXED_FILE))
+		return -EBADF;
+
+	ix->filename = NULL;
+	ix->value = NULL;
+	name = u64_to_user_ptr(READ_ONCE(sqe->addr));
+	ix->ctx.value = u64_to_user_ptr(READ_ONCE(sqe->addr2));
+	ix->ctx.size = READ_ONCE(sqe->len);
+	ix->ctx.flags = READ_ONCE(sqe->xattr_flags);
+
+	if (ix->ctx.flags)
+		return -EINVAL;
+
+	ix->ctx.kname = kmalloc(XATTR_NAME_MAX + 1, GFP_KERNEL);
+	if (!ix->ctx.kname)
+		return -ENOMEM;
+
+	ret = strncpy_from_user(ix->ctx.kname, name, XATTR_NAME_MAX + 1);
+	if (!ret || ret == XATTR_NAME_MAX + 1)
+		ret = -ERANGE;
+	if (ret < 0) {
+		kfree(ix->ctx.kname);
+		return ret;
+	}
+
+	req->flags |= REQ_F_NEED_CLEANUP;
+	return 0;
+}
+
+static int io_fgetxattr_prep(struct io_kiocb *req,
+			     const struct io_uring_sqe *sqe)
+{
+	return __io_getxattr_prep(req, sqe);
+}
+
+static int io_getxattr_prep(struct io_kiocb *req,
+			    const struct io_uring_sqe *sqe)
+{
+	struct io_xattr *ix = &req->xattr;
+	const char __user *path;
+	int ret;
+
+	ret = __io_getxattr_prep(req, sqe);
+	if (ret)
+		return ret;
+
+	path = u64_to_user_ptr(READ_ONCE(sqe->addr3));
+
+	ix->filename = getname_flags(path, LOOKUP_FOLLOW, NULL);
+	if (IS_ERR(ix->filename)) {
+		ret = PTR_ERR(ix->filename);
+		ix->filename = NULL;
+	}
+
+	return ret;
+}
+
+static int io_fgetxattr(struct io_kiocb *req, unsigned int issue_flags)
+{
+	struct io_xattr *ix = &req->xattr;
+	int ret;
+
+	if (issue_flags & IO_URING_F_NONBLOCK)
+		return -EAGAIN;
+
+	ret = do_getxattr(mnt_user_ns(req->file->f_path.mnt),
+			req->file->f_path.dentry,
+			ix->ctx.kname,
+			(void __user *)ix->ctx.value,
+			ix->ctx.size);
+
+	req->flags &= ~REQ_F_NEED_CLEANUP;
+	kfree(ix->ctx.kname);
+	if (ret < 0)
+		req_set_fail(req);
+
+	io_req_complete(req, ret);
+	return 0;
+}
+
+static int io_getxattr(struct io_kiocb *req, unsigned int issue_flags)
+{
+	struct io_xattr *ix = &req->xattr;
+	unsigned int lookup_flags = LOOKUP_FOLLOW;
+	struct path path;
+	int ret;
+
+	if (issue_flags & IO_URING_F_NONBLOCK)
+		return -EAGAIN;
+
+retry:
+	ret = do_user_path_at_empty(AT_FDCWD, ix->filename, lookup_flags, &path);
+	if (!ret) {
+		ret = do_getxattr(mnt_user_ns(path.mnt),
+				path.dentry,
+				ix->ctx.kname,
+				(void __user *)ix->ctx.value,
+				ix->ctx.size);
+
+		path_put(&path);
+		if (retry_estale(ret, lookup_flags)) {
+			lookup_flags |= LOOKUP_REVAL;
+			goto retry;
+		}
+	}
+	putname(ix->filename);
+
+	req->flags &= ~REQ_F_NEED_CLEANUP;
+	kfree(ix->ctx.kname);
+	if (ret < 0)
+		req_set_fail(req);
+
+	io_req_complete(req, ret);
+	return 0;
+}
+
 static int __io_setxattr_prep(struct io_kiocb *req,
 			const struct io_uring_sqe *sqe,
 			struct user_namespace *user_ns)
@@ -6778,6 +6909,10 @@ static int io_req_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 		return io_fsetxattr_prep(req, sqe);
 	case IORING_OP_SETXATTR:
 		return io_setxattr_prep(req, sqe);
+	case IORING_OP_FGETXATTR:
+		return io_fgetxattr_prep(req, sqe);
+	case IORING_OP_GETXATTR:
+		return io_getxattr_prep(req, sqe);
 	}
 
 	printk_once(KERN_WARNING "io_uring: unhandled opcode %d\n",
@@ -6927,6 +7062,13 @@ static void io_clean_op(struct io_kiocb *req)
 			kfree(req->xattr.ctx.kname);
 			kvfree(req->xattr.value);
 			break;
+		case IORING_OP_GETXATTR:
+			if (req->xattr.filename)
+				putname(req->xattr.filename);
+			fallthrough;
+		case IORING_OP_FGETXATTR:
+			kfree(req->xattr.ctx.kname);
+			break;
 		}
 	}
 	if ((req->flags & REQ_F_POLLED) && req->apoll) {
@@ -7078,6 +7220,12 @@ static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags)
 	case IORING_OP_SETXATTR:
 		ret = io_setxattr(req, issue_flags);
 		break;
+	case IORING_OP_FGETXATTR:
+		ret = io_fgetxattr(req, issue_flags);
+		break;
+	case IORING_OP_GETXATTR:
+		ret = io_getxattr(req, issue_flags);
+		break;
 	default:
 		ret = -EINVAL;
 		break;
diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h
index c62a8bec8cd4..efc7ac9b3a6b 100644
--- a/include/uapi/linux/io_uring.h
+++ b/include/uapi/linux/io_uring.h
@@ -148,6 +148,8 @@ enum {
 	IORING_OP_GETDENTS,
 	IORING_OP_FSETXATTR,
 	IORING_OP_SETXATTR,
+	IORING_OP_FGETXATTR,
+	IORING_OP_GETXATTR,
 
 	/* this goes last, obviously */
 	IORING_OP_LAST,
-- 
2.30.2

Re: [PATCH v6 2/5] fs: split off setxattr_setup function from setxattr

[Christian Brauner]

On Wed, Dec 22, 2021 at 01:01:24PM -0800, Stefan Roesch wrote:
> This splits of the setup part of the function
> setxattr in its own dedicated function called
> setxattr_setup.
> 
> This makes it possible to call this function
> from io_uring in the pre-processing of an
> xattr request.
> 
> Signed-off-by: Stefan Roesch <shr@fb.com>
> ---

I like the introduction of struct xattr_ctx.
But I would prefer if we called this setxattr_prepare() to mirror
setattr_prepare() and change the signature to:

int setxattr_setup(struct user_namespace *mnt_userns,
                   const char __user *name,
		   struct xattr_ctx *ctx,
		   void **xattr_val);

Since NULL is a success condition I think it makes more sense to have an
error returned and the value be a return argument. So sm like
(uncompiled and untested):

int setxattr_prepare(struct user_namespace *mnt_userns, const char __user *name,
		     struct xattr_ctx *ctx, void **xattr_val)
{
	void *kvalue = NULL;
	int error;

	if (ctx->flags & ~(XATTR_CREATE | XATTR_REPLACE))
		return -EINVAL;

	error = strncpy_from_user(ctx->kname, name, ctx->kname_sz);
	if (error == 0 || error == ctx->kname_sz)
		return -ERANGE;
	if (error < 0)
		return error;

	if (ctx->size) {
		if (ctx->size > XATTR_SIZE_MAX)
			return -E2BIG;

		kvalue = kvmalloc(ctx->size, GFP_KERNEL);
		if (!kvalue)
			return -ENOMEM;

		if (copy_from_user(kvalue, ctx->value, ctx->size)) {
			kvfree(kvalue);
			return -EFAULT;
		}

		if ((strcmp(ctx->kname, XATTR_NAME_POSIX_ACL_ACCESS) == 0) ||
		    (strcmp(ctx->kname, XATTR_NAME_POSIX_ACL_DEFAULT) == 0))
			posix_acl_fix_xattr_from_user(mnt_userns, kvalue, ctx->size);
	}

	*xattr_val = kvalue;
	return 0;
}

Re: [PATCH v6 1/5] fs: split off do_user_path_at_empty from user_path_at_empty()

[Christian Brauner]

On Wed, Dec 22, 2021 at 01:01:23PM -0800, Stefan Roesch wrote:
> This splits off a do_user_path_at_empty function from the
> user_path_at_empty_function. This is required so it can be
> called from io_uring.
> 
> Signed-off-by: Stefan Roesch <shr@fb.com>
> ---

(Not excited we're continuing exposing two variants one with char *name
and struct filename *filename but we've done it for the mkdir/mknod etc
series already. But we should earmark this for something that we might
look into doing better in the near future.)

Acked-by: Christian Brauner <christian.brauner@ubuntu.com>

Re: [PATCH v6 3/5] fs: split off do_getxattr from getxattr

[Christian Brauner]

On Wed, Dec 22, 2021 at 01:01:25PM -0800, Stefan Roesch wrote:
> This splits off do_getxattr function from the getxattr
> function. This will allow io_uring to call it from its
> io worker.
> 
> Signed-off-by: Stefan Roesch <shr@fb.com>
> ---

Looks good.
Acked-by: Christian Brauner <christian.brauner@ubuntu.com>

Re: [PATCH v6 0/5] io_uring: add xattr support

[Christian Brauner]

On Wed, Dec 22, 2021 at 01:01:22PM -0800, Stefan Roesch wrote:
> This adds the xattr support to io_uring. The intent is to have a more
> complete support for file operations in io_uring.
> 
> This change adds support for the following functions to io_uring:
> - fgetxattr
> - fsetxattr
> - getxattr
> - setxattr
> 
> Patch 1: fs: split off do_user_path_at_empty from user_path_at_empty()
>   This splits off a new function do_user_path_at_empty from
>   user_path_at_empty that is based on filename and not on a
>   user-specified string.
> 
> Patch 2: fs: split off setxattr_setup function from setxattr
>   Split off the setup part of the setxattr function.
> 
> Patch 3: fs: split off do_getxattr from getxattr
>   Split of the do_getxattr part from getxattr. This will
>   allow it to be invoked it from io_uring.
> 
> Patch 4: io_uring: add fsetxattr and setxattr support
>   This adds new functions to support the fsetxattr and setxattr
>   functions.
> 
> Patch 5: io_uring: add fgetxattr and getxattr support
>   This adds new functions to support the fgetxattr and getxattr
>   functions.
> 
> 
> There are two additional patches:
>   liburing: Add support for xattr api's.
>             This also includes the tests for the new code.
>   xfstests: Add support for io_uring xattr support.
> 
> 
> V6: - reverted addition of kname array to xattr_ctx structure
>       Adding the kname array increases the io_kiocb beyond 64 bytes
>       (increases it to 224 bytes). We try hard to limit it to 64 bytes.
>       Keeping the original interface also is a bit more efficient.
>     - rebased on for-5.17/io_uring-getdents64
> V5: - add kname array to xattr_ctx structure
> V4: - rebased patch series
> V3: - remove req->file checks in prep functions
>     - change size parameter in do_xattr
> V2: - split off function do_user_path_empty instead of changing
>       the function signature of user_path_at
>     - Fix datatype size problem in do_getxattr
> 
> 
> 
> Stefan Roesch (5):
>   fs: split off do_user_path_at_empty from user_path_at_empty()
>   fs: split off setxattr_setup function from setxattr
>   fs: split off do_getxattr from getxattr
>   io_uring: add fsetxattr and setxattr support
>   io_uring: add fgetxattr and getxattr support
> 
>  fs/internal.h                 |  23 +++
>  fs/io_uring.c                 | 318 ++++++++++++++++++++++++++++++++++
>  fs/namei.c                    |  10 +-
>  fs/xattr.c                    | 107 ++++++++----
>  include/linux/namei.h         |   2 +
>  include/uapi/linux/io_uring.h |   8 +-
>  6 files changed, 428 insertions(+), 40 deletions(-)
> 
> 
> base-commit: b4518682080d3a1cdd6ea45a54ff6772b8b2797a

Jens, please keep me in the loop once this series lands.
I maintain a large vfs testsuite for idmapped mounts (It's actually a
generic testsuite which also tests idmapped mounts.) and it currently
already has tests for io_uring:

https://git.kernel.org/pub/scm/fs/xfs/xfstests-dev.git/tree/src/idmapped-mounts/idmapped-mounts.c#n6942

Once this lands we need to expand it to test xattr support for io_uring
as well (It should probably also include mkdir/link/mknod that we added
last cycle.).

Christian

Re: [PATCH v6 0/5] io_uring: add xattr support

[Jens Axboe]

On 12/23/21 4:04 AM, Christian Brauner wrote:
> On Wed, Dec 22, 2021 at 01:01:22PM -0800, Stefan Roesch wrote:
>> This adds the xattr support to io_uring. The intent is to have a more
>> complete support for file operations in io_uring.
>>
>> This change adds support for the following functions to io_uring:
>> - fgetxattr
>> - fsetxattr
>> - getxattr
>> - setxattr
>>
>> Patch 1: fs: split off do_user_path_at_empty from user_path_at_empty()
>>   This splits off a new function do_user_path_at_empty from
>>   user_path_at_empty that is based on filename and not on a
>>   user-specified string.
>>
>> Patch 2: fs: split off setxattr_setup function from setxattr
>>   Split off the setup part of the setxattr function.
>>
>> Patch 3: fs: split off do_getxattr from getxattr
>>   Split of the do_getxattr part from getxattr. This will
>>   allow it to be invoked it from io_uring.
>>
>> Patch 4: io_uring: add fsetxattr and setxattr support
>>   This adds new functions to support the fsetxattr and setxattr
>>   functions.
>>
>> Patch 5: io_uring: add fgetxattr and getxattr support
>>   This adds new functions to support the fgetxattr and getxattr
>>   functions.
>>
>>
>> There are two additional patches:
>>   liburing: Add support for xattr api's.
>>             This also includes the tests for the new code.
>>   xfstests: Add support for io_uring xattr support.
>>
>>
>> V6: - reverted addition of kname array to xattr_ctx structure
>>       Adding the kname array increases the io_kiocb beyond 64 bytes
>>       (increases it to 224 bytes). We try hard to limit it to 64 bytes.
>>       Keeping the original interface also is a bit more efficient.
>>     - rebased on for-5.17/io_uring-getdents64
>> V5: - add kname array to xattr_ctx structure
>> V4: - rebased patch series
>> V3: - remove req->file checks in prep functions
>>     - change size parameter in do_xattr
>> V2: - split off function do_user_path_empty instead of changing
>>       the function signature of user_path_at
>>     - Fix datatype size problem in do_getxattr
>>
>>
>>
>> Stefan Roesch (5):
>>   fs: split off do_user_path_at_empty from user_path_at_empty()
>>   fs: split off setxattr_setup function from setxattr
>>   fs: split off do_getxattr from getxattr
>>   io_uring: add fsetxattr and setxattr support
>>   io_uring: add fgetxattr and getxattr support
>>
>>  fs/internal.h                 |  23 +++
>>  fs/io_uring.c                 | 318 ++++++++++++++++++++++++++++++++++
>>  fs/namei.c                    |  10 +-
>>  fs/xattr.c                    | 107 ++++++++----
>>  include/linux/namei.h         |   2 +
>>  include/uapi/linux/io_uring.h |   8 +-
>>  6 files changed, 428 insertions(+), 40 deletions(-)
>>
>>
>> base-commit: b4518682080d3a1cdd6ea45a54ff6772b8b2797a
> 
> Jens, please keep me in the loop once this series lands.

You bet, and thanks for the reviews!

> I maintain a large vfs testsuite for idmapped mounts (It's actually a
> generic testsuite which also tests idmapped mounts.) and it currently
> already has tests for io_uring:
> 
> https://git.kernel.org/pub/scm/fs/xfs/xfstests-dev.git/tree/src/idmapped-mounts/idmapped-mounts.c#n6942
> 
> Once this lands we need to expand it to test xattr support for io_uring
> as well (It should probably also include mkdir/link/mknod that we added
> last cycle.).

There are a few basic tests here:

https://git.kernel.dk/cgit/liburing/log/?h=xattr

as well, but more of a sanity kind of checking, would be great if tests
were added to the VFS suite as well.

With the last few kinks ironed out, I hope to queue the next version
posted for 5.17.

-- 
Jens Axboe

Re: [PATCH v6 4/5] io_uring: add fsetxattr and setxattr support

[Christian Brauner]

On Wed, Dec 22, 2021 at 01:01:26PM -0800, Stefan Roesch wrote:
> This adds support to io_uring for the fsetxattr and setxattr API.
> 
> Signed-off-by: Stefan Roesch <shr@fb.com>
> ---
>  fs/io_uring.c                 | 170 ++++++++++++++++++++++++++++++++++
>  include/uapi/linux/io_uring.h |   6 +-
>  2 files changed, 175 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/io_uring.c b/fs/io_uring.c
> index c8258c784116..8b6c70d6cacc 100644
> --- a/fs/io_uring.c
> +++ b/fs/io_uring.c
> @@ -82,6 +82,7 @@
>  #include <linux/audit.h>
>  #include <linux/security.h>
>  #include <linux/atomic-ref.h>
> +#include <linux/xattr.h>
>  
>  #define CREATE_TRACE_POINTS
>  #include <trace/events/io_uring.h>
> @@ -726,6 +727,13 @@ struct io_async_rw {
>  	struct wait_page_queue		wpq;
>  };
>  
> +struct io_xattr {
> +	struct file			*file;
> +	struct xattr_ctx		ctx;
> +	void				*value;
> +	struct filename			*filename;
> +};
> +
>  enum {
>  	REQ_F_FIXED_FILE_BIT	= IOSQE_FIXED_FILE_BIT,
>  	REQ_F_IO_DRAIN_BIT	= IOSQE_IO_DRAIN_BIT,
> @@ -866,6 +874,7 @@ struct io_kiocb {
>  		struct io_symlink	symlink;
>  		struct io_hardlink	hardlink;
>  		struct io_getdents	getdents;
> +		struct io_xattr		xattr;
>  	};
>  
>  	u8				opcode;
> @@ -1118,6 +1127,10 @@ static const struct io_op_def io_op_defs[] = {
>  	[IORING_OP_GETDENTS] = {
>  		.needs_file		= 1,
>  	},
> +	[IORING_OP_FSETXATTR] = {
> +		.needs_file = 1
> +	},
> +	[IORING_OP_SETXATTR] = {},
>  };
>  
>  /* requests with any of those set should undergo io_disarm_next() */
> @@ -3887,6 +3900,144 @@ static int io_renameat(struct io_kiocb *req, unsigned int issue_flags)
>  	return 0;
>  }
>  
> +static int __io_setxattr_prep(struct io_kiocb *req,
> +			const struct io_uring_sqe *sqe,
> +			struct user_namespace *user_ns)
> +{
> +	struct io_xattr *ix = &req->xattr;
> +	const char __user *name;
> +	void *ret;
> +
> +	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
> +		return -EINVAL;
> +	if (unlikely(sqe->ioprio))
> +		return -EINVAL;
> +	if (unlikely(req->flags & REQ_F_FIXED_FILE))
> +		return -EBADF;
> +
> +	ix->filename = NULL;
> +	name = u64_to_user_ptr(READ_ONCE(sqe->addr));
> +	ix->ctx.value = u64_to_user_ptr(READ_ONCE(sqe->addr2));
> +	ix->ctx.size = READ_ONCE(sqe->len);
> +	ix->ctx.flags = READ_ONCE(sqe->xattr_flags);
> +
> +	ix->ctx.kname = kmalloc(XATTR_NAME_MAX + 1, GFP_KERNEL);
> +	if (!ix->ctx.kname)
> +		return -ENOMEM;
> +	ix->ctx.kname_sz = XATTR_NAME_MAX + 1;
> +
> +	ret = setxattr_setup(user_ns, name, &ix->ctx);

Looking at this a bit closer, the setxattr_setup() function converts the
vfs caps prior to vfs_setxattr(). That shouldn't be done there though.
The conversion should be done when mnt_want_write() is held in
__io_setxattr() exactly how we do for setxattr()-based calls in
fs/xattr.c. This will guard against changes of relevant mount properties
(current or future). It will also allow you to simplify your
setxattr_setup() function a bit and you don't need to retrieve the
mount's idmapping until __io_setxattr().

Right now you're splitting updating the xattrs over the prep and commit
stage and I worry that in fully async contexts this is easy to miss. So
I'd rather do it in one place. Since we can't move it all into
vfs_setxattr() similar to what we did for fscaps because it's used in a
bunch of contexts where the conversion isn't wanted we should simply
expose do_setxattr() similar to do_getxattr() you're adding.

So on top of your current patchset I'd suggest you do something like the
following (completely untested):

From 6bcd3efc3293bb91599ee73272262ac596ab4608 Mon Sep 17 00:00:00 2001
From: Christian Brauner <christian.brauner@ubuntu.com>
Date: Thu, 23 Dec 2021 15:23:14 +0100
Subject: [PATCH] UNTESTED

---
 fs/internal.h |  8 +++++---
 fs/io_uring.c | 21 +++++++++-----------
 fs/xattr.c    | 55 ++++++++++++++++++++++++++++++++++-----------------
 3 files changed, 51 insertions(+), 33 deletions(-)

diff --git a/fs/internal.h b/fs/internal.h
index ea0433799dbc..08259fa98b2e 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -222,6 +222,8 @@ ssize_t do_getxattr(struct user_namespace *mnt_userns,
 		    void __user *value,
 		    size_t size);
 
-void *setxattr_setup(struct user_namespace *mnt_userns,
-		     const char __user *name,
-		     struct xattr_ctx *ctx);
+int do_setxattr(struct user_namespace *mnt_userns, struct dentry *dentry,
+		struct xattr_ctx *ctx, void *xattr_val);
+
+int setxattr_copy(const char __user *name, struct xattr_ctx *ctx,
+		  void **xattr_val);
diff --git a/fs/io_uring.c b/fs/io_uring.c
index 5dd01f19d915..c910c29e1632 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -4040,12 +4040,11 @@ static int io_getxattr(struct io_kiocb *req, unsigned int issue_flags)
 }
 
 static int __io_setxattr_prep(struct io_kiocb *req,
-			const struct io_uring_sqe *sqe,
-			struct user_namespace *user_ns)
+			const struct io_uring_sqe *sqe)
 {
 	struct io_xattr *ix = &req->xattr;
 	const char __user *name;
-	void *ret;
+	int ret;
 
 	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
 		return -EINVAL;
@@ -4065,13 +4064,12 @@ static int __io_setxattr_prep(struct io_kiocb *req,
 		return -ENOMEM;
 	ix->ctx.kname_sz = XATTR_NAME_MAX + 1;
 
-	ret = setxattr_setup(user_ns, name, &ix->ctx);
-	if (IS_ERR(ret)) {
+	ret = setxattr_copy(name, &ix->ctx, &ix->value);
+	if (ret) {
 		kfree(ix->ctx.kname);
-		return PTR_ERR(ret);
+		return ret;
 	}
 
-	ix->value = ret;
 	req->flags |= REQ_F_NEED_CLEANUP;
 	return 0;
 }
@@ -4083,7 +4081,7 @@ static int io_setxattr_prep(struct io_kiocb *req,
 	const char __user *path;
 	int ret;
 
-	ret = __io_setxattr_prep(req, sqe, current_user_ns());
+	ret = __io_setxattr_prep(req, sqe);
 	if (ret)
 		return ret;
 
@@ -4101,7 +4099,7 @@ static int io_setxattr_prep(struct io_kiocb *req,
 static int io_fsetxattr_prep(struct io_kiocb *req,
 			const struct io_uring_sqe *sqe)
 {
-	return __io_setxattr_prep(req, sqe, file_mnt_user_ns(req->file));
+	return __io_setxattr_prep(req, sqe);
 }
 
 static int __io_setxattr(struct io_kiocb *req, unsigned int issue_flags,
@@ -4112,9 +4110,8 @@ static int __io_setxattr(struct io_kiocb *req, unsigned int issue_flags,
 
 	ret = mnt_want_write(path->mnt);
 	if (!ret) {
-		ret = vfs_setxattr(mnt_user_ns(path->mnt), path->dentry,
-				ix->ctx.kname, ix->value, ix->ctx.size,
-				ix->ctx.flags);
+		ret = do_setxattr(mnt_user_ns(path->mnt), path->dentry,
+				  &ix->ctx, ix->value);
 		mnt_drop_write(path->mnt);
 	}
 
diff --git a/fs/xattr.c b/fs/xattr.c
index a675c7f0ea0c..03a44c5895d1 100644
--- a/fs/xattr.c
+++ b/fs/xattr.c
@@ -542,40 +542,59 @@ EXPORT_SYMBOL_GPL(vfs_removexattr);
  * Extended attribute SET operations
  */
 
-void *setxattr_setup(struct user_namespace *mnt_userns, const char __user *name,
-		struct xattr_ctx *ctx)
+int setxattr_copy(const char __user *name, struct xattr_ctx *ctx,
+		  void **xattr_val)
 {
 	void *kvalue = NULL;
 	int error;
 
 	if (ctx->flags & ~(XATTR_CREATE|XATTR_REPLACE))
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 
 	error = strncpy_from_user(ctx->kname, name, ctx->kname_sz);
 	if (error == 0 || error == ctx->kname_sz)
-		return  ERR_PTR(-ERANGE);
+		return  -ERANGE;
 	if (error < 0)
-		return ERR_PTR(error);
+		return error;
 
 	if (ctx->size) {
 		if (ctx->size > XATTR_SIZE_MAX)
-			return ERR_PTR(-E2BIG);
+			return -E2BIG;
 
 		kvalue = kvmalloc(ctx->size, GFP_KERNEL);
 		if (!kvalue)
-			return ERR_PTR(-ENOMEM);
+			return -ENOMEM;
 
 		if (copy_from_user(kvalue, ctx->value, ctx->size)) {
 			kvfree(kvalue);
-			return ERR_PTR(-EFAULT);
+			return -EFAULT;
 		}
-
-		if ((strcmp(ctx->kname, XATTR_NAME_POSIX_ACL_ACCESS) == 0) ||
-		    (strcmp(ctx->kname, XATTR_NAME_POSIX_ACL_DEFAULT) == 0))
-			posix_acl_fix_xattr_from_user(mnt_userns, kvalue, ctx->size);
 	}
 
-	return kvalue;
+	*xattr_val = kvalue;
+	return 0;
+}
+
+static void setxattr_convert(struct user_namespace *mnt_userns,
+			     struct xattr_ctx *ctx, void *kvalue)
+{
+	if (ctx->size &&
+	    ((strcmp(ctx->kname, XATTR_NAME_POSIX_ACL_ACCESS) == 0) ||
+	     (strcmp(ctx->kname, XATTR_NAME_POSIX_ACL_DEFAULT) == 0)))
+		posix_acl_fix_xattr_from_user(mnt_userns, kvalue, ctx->size);
+}
+
+int do_setxattr(struct user_namespace *mnt_userns, struct dentry *dentry,
+		struct xattr_ctx *ctx, void *xattr_val)
+{
+	void *kvalue = NULL;
+	int error;
+
+	setxattr_convert(mnt_userns, ctx, kvalue);
+	error = vfs_setxattr(mnt_userns, dentry, ctx->kname,
+			     kvalue, ctx->size, ctx->flags);
+	kvfree(kvalue);
+	return error;
 }
 
 static long
@@ -591,14 +610,14 @@ setxattr(struct user_namespace *mnt_userns, struct dentry *d,
 		.kname_sz = sizeof(kname),
 		.flags    = flags,
 	};
-	void *kvalue;
+	void *kvalue = NULL;
 	int error;
 
-	kvalue = setxattr_setup(mnt_userns, name, &ctx);
-	if (IS_ERR(kvalue))
-		return PTR_ERR(kvalue);
+	error = setxattr_copy(name, &ctx, &kvalue);
+	if (error)
+		return error;
 
-	error = vfs_setxattr(mnt_userns, d, kname, kvalue, size, flags);
+	error = do_setxattr(mnt_userns, d, &ctx, kvalue);
 
 	kvfree(kvalue);
 	return error;
-- 
2.30.2

Re: [PATCH v6 4/5] io_uring: add fsetxattr and setxattr support

[Stefan Roesch]

On 12/23/21 6:52 AM, Christian Brauner wrote:
> On Wed, Dec 22, 2021 at 01:01:26PM -0800, Stefan Roesch wrote:
>> This adds support to io_uring for the fsetxattr and setxattr API.
>>
>> Signed-off-by: Stefan Roesch <shr@fb.com>
>> ---
>>  fs/io_uring.c                 | 170 ++++++++++++++++++++++++++++++++++
>>  include/uapi/linux/io_uring.h |   6 +-
>>  2 files changed, 175 insertions(+), 1 deletion(-)
>>
>> diff --git a/fs/io_uring.c b/fs/io_uring.c
>> index c8258c784116..8b6c70d6cacc 100644
>> --- a/fs/io_uring.c
>> +++ b/fs/io_uring.c
>> @@ -82,6 +82,7 @@
>>  #include <linux/audit.h>
>>  #include <linux/security.h>
>>  #include <linux/atomic-ref.h>
>> +#include <linux/xattr.h>
>>  
>>  #define CREATE_TRACE_POINTS
>>  #include <trace/events/io_uring.h>
>> @@ -726,6 +727,13 @@ struct io_async_rw {
>>  	struct wait_page_queue		wpq;
>>  };
>>  
>> +struct io_xattr {
>> +	struct file			*file;
>> +	struct xattr_ctx		ctx;
>> +	void				*value;
>> +	struct filename			*filename;
>> +};
>> +
>>  enum {
>>  	REQ_F_FIXED_FILE_BIT	= IOSQE_FIXED_FILE_BIT,
>>  	REQ_F_IO_DRAIN_BIT	= IOSQE_IO_DRAIN_BIT,
>> @@ -866,6 +874,7 @@ struct io_kiocb {
>>  		struct io_symlink	symlink;
>>  		struct io_hardlink	hardlink;
>>  		struct io_getdents	getdents;
>> +		struct io_xattr		xattr;
>>  	};
>>  
>>  	u8				opcode;
>> @@ -1118,6 +1127,10 @@ static const struct io_op_def io_op_defs[] = {
>>  	[IORING_OP_GETDENTS] = {
>>  		.needs_file		= 1,
>>  	},
>> +	[IORING_OP_FSETXATTR] = {
>> +		.needs_file = 1
>> +	},
>> +	[IORING_OP_SETXATTR] = {},
>>  };
>>  
>>  /* requests with any of those set should undergo io_disarm_next() */
>> @@ -3887,6 +3900,144 @@ static int io_renameat(struct io_kiocb *req, unsigned int issue_flags)
>>  	return 0;
>>  }
>>  
>> +static int __io_setxattr_prep(struct io_kiocb *req,
>> +			const struct io_uring_sqe *sqe,
>> +			struct user_namespace *user_ns)
>> +{
>> +	struct io_xattr *ix = &req->xattr;
>> +	const char __user *name;
>> +	void *ret;
>> +
>> +	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
>> +		return -EINVAL;
>> +	if (unlikely(sqe->ioprio))
>> +		return -EINVAL;
>> +	if (unlikely(req->flags & REQ_F_FIXED_FILE))
>> +		return -EBADF;
>> +
>> +	ix->filename = NULL;
>> +	name = u64_to_user_ptr(READ_ONCE(sqe->addr));
>> +	ix->ctx.value = u64_to_user_ptr(READ_ONCE(sqe->addr2));
>> +	ix->ctx.size = READ_ONCE(sqe->len);
>> +	ix->ctx.flags = READ_ONCE(sqe->xattr_flags);
>> +
>> +	ix->ctx.kname = kmalloc(XATTR_NAME_MAX + 1, GFP_KERNEL);
>> +	if (!ix->ctx.kname)
>> +		return -ENOMEM;
>> +	ix->ctx.kname_sz = XATTR_NAME_MAX + 1;
>> +
>> +	ret = setxattr_setup(user_ns, name, &ix->ctx);
> 
> Looking at this a bit closer, the setxattr_setup() function converts the
> vfs caps prior to vfs_setxattr(). That shouldn't be done there though.
> The conversion should be done when mnt_want_write() is held in
> __io_setxattr() exactly how we do for setxattr()-based calls in
> fs/xattr.c. This will guard against changes of relevant mount properties
> (current or future). It will also allow you to simplify your
> setxattr_setup() function a bit and you don't need to retrieve the
> mount's idmapping until __io_setxattr().
> 
> Right now you're splitting updating the xattrs over the prep and commit
> stage and I worry that in fully async contexts this is easy to miss. So
> I'd rather do it in one place. Since we can't move it all into
> vfs_setxattr() similar to what we did for fscaps because it's used in a
> bunch of contexts where the conversion isn't wanted we should simply
> expose do_setxattr() similar to do_getxattr() you're adding.
> 
> So on top of your current patchset I'd suggest you do something like the
> following (completely untested):
> 


Thanks for your review and the code. I only changed the below code that the
do_setxattr does not use a kvalue, I assume you wanted to use xattr_val.


> From 6bcd3efc3293bb91599ee73272262ac596ab4608 Mon Sep 17 00:00:00 2001
> From: Christian Brauner <christian.brauner@ubuntu.com>
> Date: Thu, 23 Dec 2021 15:23:14 +0100
> Subject: [PATCH] UNTESTED
> 
> ---
>  fs/internal.h |  8 +++++---
>  fs/io_uring.c | 21 +++++++++-----------
>  fs/xattr.c    | 55 ++++++++++++++++++++++++++++++++++-----------------
>  3 files changed, 51 insertions(+), 33 deletions(-)
> 
> diff --git a/fs/internal.h b/fs/internal.h
> index ea0433799dbc..08259fa98b2e 100644
> --- a/fs/internal.h
> +++ b/fs/internal.h
> @@ -222,6 +222,8 @@ ssize_t do_getxattr(struct user_namespace *mnt_userns,
>  		    void __user *value,
>  		    size_t size);
>  
> -void *setxattr_setup(struct user_namespace *mnt_userns,
> -		     const char __user *name,
> -		     struct xattr_ctx *ctx);
> +int do_setxattr(struct user_namespace *mnt_userns, struct dentry *dentry,
> +		struct xattr_ctx *ctx, void *xattr_val);
> +
> +int setxattr_copy(const char __user *name, struct xattr_ctx *ctx,
> +		  void **xattr_val);
> diff --git a/fs/io_uring.c b/fs/io_uring.c
> index 5dd01f19d915..c910c29e1632 100644
> --- a/fs/io_uring.c
> +++ b/fs/io_uring.c
> @@ -4040,12 +4040,11 @@ static int io_getxattr(struct io_kiocb *req, unsigned int issue_flags)
>  }
>  
>  static int __io_setxattr_prep(struct io_kiocb *req,
> -			const struct io_uring_sqe *sqe,
> -			struct user_namespace *user_ns)
> +			const struct io_uring_sqe *sqe)
>  {
>  	struct io_xattr *ix = &req->xattr;
>  	const char __user *name;
> -	void *ret;
> +	int ret;
>  
>  	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
>  		return -EINVAL;
> @@ -4065,13 +4064,12 @@ static int __io_setxattr_prep(struct io_kiocb *req,
>  		return -ENOMEM;
>  	ix->ctx.kname_sz = XATTR_NAME_MAX + 1;
>  
> -	ret = setxattr_setup(user_ns, name, &ix->ctx);
> -	if (IS_ERR(ret)) {
> +	ret = setxattr_copy(name, &ix->ctx, &ix->value);
> +	if (ret) {
>  		kfree(ix->ctx.kname);
> -		return PTR_ERR(ret);
> +		return ret;
>  	}
>  
> -	ix->value = ret;
>  	req->flags |= REQ_F_NEED_CLEANUP;
>  	return 0;
>  }
> @@ -4083,7 +4081,7 @@ static int io_setxattr_prep(struct io_kiocb *req,
>  	const char __user *path;
>  	int ret;
>  
> -	ret = __io_setxattr_prep(req, sqe, current_user_ns());
> +	ret = __io_setxattr_prep(req, sqe);
>  	if (ret)
>  		return ret;
>  
> @@ -4101,7 +4099,7 @@ static int io_setxattr_prep(struct io_kiocb *req,
>  static int io_fsetxattr_prep(struct io_kiocb *req,
>  			const struct io_uring_sqe *sqe)
>  {
> -	return __io_setxattr_prep(req, sqe, file_mnt_user_ns(req->file));
> +	return __io_setxattr_prep(req, sqe);
>  }
>  
>  static int __io_setxattr(struct io_kiocb *req, unsigned int issue_flags,
> @@ -4112,9 +4110,8 @@ static int __io_setxattr(struct io_kiocb *req, unsigned int issue_flags,
>  
>  	ret = mnt_want_write(path->mnt);
>  	if (!ret) {
> -		ret = vfs_setxattr(mnt_user_ns(path->mnt), path->dentry,
> -				ix->ctx.kname, ix->value, ix->ctx.size,
> -				ix->ctx.flags);
> +		ret = do_setxattr(mnt_user_ns(path->mnt), path->dentry,
> +				  &ix->ctx, ix->value);
>  		mnt_drop_write(path->mnt);
>  	}
>  
> diff --git a/fs/xattr.c b/fs/xattr.c
> index a675c7f0ea0c..03a44c5895d1 100644
> --- a/fs/xattr.c
> +++ b/fs/xattr.c
> @@ -542,40 +542,59 @@ EXPORT_SYMBOL_GPL(vfs_removexattr);
>   * Extended attribute SET operations
>   */
>  
> -void *setxattr_setup(struct user_namespace *mnt_userns, const char __user *name,
> -		struct xattr_ctx *ctx)
> +int setxattr_copy(const char __user *name, struct xattr_ctx *ctx,
> +		  void **xattr_val)
>  {
>  	void *kvalue = NULL;
>  	int error;
>  
>  	if (ctx->flags & ~(XATTR_CREATE|XATTR_REPLACE))
> -		return ERR_PTR(-EINVAL);
> +		return -EINVAL;
>  
>  	error = strncpy_from_user(ctx->kname, name, ctx->kname_sz);
>  	if (error == 0 || error == ctx->kname_sz)
> -		return  ERR_PTR(-ERANGE);
> +		return  -ERANGE;
>  	if (error < 0)
> -		return ERR_PTR(error);
> +		return error;
>  
>  	if (ctx->size) {
>  		if (ctx->size > XATTR_SIZE_MAX)
> -			return ERR_PTR(-E2BIG);
> +			return -E2BIG;
>  
>  		kvalue = kvmalloc(ctx->size, GFP_KERNEL);
>  		if (!kvalue)
> -			return ERR_PTR(-ENOMEM);
> +			return -ENOMEM;
>  
>  		if (copy_from_user(kvalue, ctx->value, ctx->size)) {
>  			kvfree(kvalue);
> -			return ERR_PTR(-EFAULT);
> +			return -EFAULT;
>  		}
> -
> -		if ((strcmp(ctx->kname, XATTR_NAME_POSIX_ACL_ACCESS) == 0) ||
> -		    (strcmp(ctx->kname, XATTR_NAME_POSIX_ACL_DEFAULT) == 0))
> -			posix_acl_fix_xattr_from_user(mnt_userns, kvalue, ctx->size);
>  	}
>  
> -	return kvalue;
> +	*xattr_val = kvalue;
> +	return 0;
> +}
> +
> +static void setxattr_convert(struct user_namespace *mnt_userns,
> +			     struct xattr_ctx *ctx, void *kvalue)
> +{
> +	if (ctx->size &&
> +	    ((strcmp(ctx->kname, XATTR_NAME_POSIX_ACL_ACCESS) == 0) ||
> +	     (strcmp(ctx->kname, XATTR_NAME_POSIX_ACL_DEFAULT) == 0)))
> +		posix_acl_fix_xattr_from_user(mnt_userns, kvalue, ctx->size);
> +}
> +
> +int do_setxattr(struct user_namespace *mnt_userns, struct dentry *dentry,
> +		struct xattr_ctx *ctx, void *xattr_val)
> +{
> +	void *kvalue = NULL;
> +	int error;
> +
> +	setxattr_convert(mnt_userns, ctx, kvalue);
> +	error = vfs_setxattr(mnt_userns, dentry, ctx->kname,
> +			     kvalue, ctx->size, ctx->flags);
> +	kvfree(kvalue);
> +	return error;
>  }
>  
>  static long
> @@ -591,14 +610,14 @@ setxattr(struct user_namespace *mnt_userns, struct dentry *d,
>  		.kname_sz = sizeof(kname),
>  		.flags    = flags,
>  	};
> -	void *kvalue;
> +	void *kvalue = NULL;
>  	int error;
>  
> -	kvalue = setxattr_setup(mnt_userns, name, &ctx);
> -	if (IS_ERR(kvalue))
> -		return PTR_ERR(kvalue);
> +	error = setxattr_copy(name, &ctx, &kvalue);
> +	if (error)
> +		return error;
>  
> -	error = vfs_setxattr(mnt_userns, d, kname, kvalue, size, flags);
> +	error = do_setxattr(mnt_userns, d, &ctx, kvalue);
>  
>  	kvfree(kvalue);
>  	return error;
>