@@ -65,6 +65,7 @@ struct io_uring_sqe {
__u32 xattr_flags;
__u32 msg_ring_flags;
__u32 uring_cmd_flags;
+ __u32 getdents_flags;
};
__u64 user_data; /* data to be passed back at completion time */
/* pack this to avoid bogus arm OABI complaints */
@@ -223,6 +224,7 @@ enum io_uring_op {
IORING_OP_URING_CMD,
IORING_OP_SEND_ZC,
IORING_OP_SENDMSG_ZC,
+ IORING_OP_GETDENTS,
/* this goes last, obviously */
IORING_OP_LAST,
@@ -259,6 +261,11 @@ enum io_uring_op {
*/
#define SPLICE_F_FD_IN_FIXED (1U << 31) /* the last bit of __u32 */
+/*
+ * sqe->getdents_flags
+ */
+#define IORING_GETDENTS_REWIND (1U << 0)
+
/*
* POLL_ADD flags. Note that since sqe->poll_events is the flag space, the
* command flags for POLL_ADD are stored in sqe->len.
@@ -47,6 +47,13 @@ struct io_link {
int flags;
};
+struct io_getdents {
+ struct file *file;
+ struct linux_dirent64 __user *dirent;
+ unsigned int count;
+ int flags;
+};
+
int io_renameat_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
{
struct io_rename *ren = io_kiocb_to_cmd(req, struct io_rename);
@@ -291,3 +298,53 @@ void io_link_cleanup(struct io_kiocb *req)
putname(sl->oldpath);
putname(sl->newpath);
}
+
+int io_getdents_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+ struct io_getdents *gd = io_kiocb_to_cmd(req, struct io_getdents);
+
+ if (READ_ONCE(sqe->off) != 0)
+ return -EINVAL;
+
+ gd->dirent = u64_to_user_ptr(READ_ONCE(sqe->addr));
+ gd->count = READ_ONCE(sqe->len);
+ gd->flags = READ_ONCE(sqe->getdents_flags);
+ if (gd->flags & ~IORING_GETDENTS_REWIND)
+ return -EINVAL;
+ /* rewind cannot be nowait */
+ if (gd->flags & IORING_GETDENTS_REWIND)
+ req->flags |= REQ_F_FORCE_ASYNC;
+
+ return 0;
+}
+
+int io_getdents(struct io_kiocb *req, unsigned int issue_flags)
+{
+ struct io_getdents *gd = io_kiocb_to_cmd(req, struct io_getdents);
+ unsigned long getdents_flags = 0;
+ int ret;
+
+ if (issue_flags & IO_URING_F_NONBLOCK) {
+ if (!(req->file->f_mode & FMODE_NOWAIT))
+ return -EAGAIN;
+
+ getdents_flags = DIR_CONTEXT_F_NOWAIT;
+ }
+ if ((gd->flags & IORING_GETDENTS_REWIND)) {
+ WARN_ON_ONCE(issue_flags & IO_URING_F_NONBLOCK);
+
+ ret = vfs_llseek(req->file, 0, SEEK_SET);
+ if (ret < 0)
+ goto out;
+ }
+
+ ret = vfs_getdents(req->file, gd->dirent, gd->count, getdents_flags);
+out:
+ if (ret == -EAGAIN &&
+ (issue_flags & IO_URING_F_NONBLOCK))
+ return -EAGAIN;
+
+ io_req_set_res(req, ret, 0);
+ return 0;
+}
+
@@ -18,3 +18,6 @@ int io_symlinkat(struct io_kiocb *req, unsigned int issue_flags);
int io_linkat_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe);
int io_linkat(struct io_kiocb *req, unsigned int issue_flags);
void io_link_cleanup(struct io_kiocb *req);
+
+int io_getdents_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe);
+int io_getdents(struct io_kiocb *req, unsigned int issue_flags);
@@ -428,6 +428,11 @@ const struct io_issue_def io_issue_defs[] = {
.prep = io_eopnotsupp_prep,
#endif
},
+ [IORING_OP_GETDENTS] = {
+ .needs_file = 1,
+ .prep = io_getdents_prep,
+ .issue = io_getdents,
+ },
};
@@ -648,6 +653,9 @@ const struct io_cold_def io_cold_defs[] = {
.fail = io_sendrecv_fail,
#endif
},
+ [IORING_OP_GETDENTS] = {
+ .name = "GETDENTS",
+ },
};
const char *io_uring_get_opcode(u8 opcode)
This add support for getdents64 to io_uring, acting exactly like the syscall: the directory is iterated from it's current's position as stored in the file struct, and the file's position is updated exactly as if getdents64 had been called. Additionally, since io_uring has no way of issuing a seek, a flag IORING_GETDENTS_REWIND has been added that will seek to the start of the directory like rewinddir(3) for users that might require such a thing. This will act exactly as if seek then getdents64 have been called sequentially with no atomicity guarantee: if this wasn't clear it is the responsibility of the caller to not use getdents multiple time on a single file in parallel if they want useful results, as is currently the case when using the syscall from multiple threads. For filesystems that support NOWAIT in iterate_shared(), try to use it first; if a user already knows the filesystem they use do not support nowait they can force async through IOSQE_ASYNC in the sqe flags, avoiding the need to bounce back through a useless EAGAIN return. (Note we already do that in prep if rewind is requested) Signed-off-by: Dominique Martinet <asmadeus@codewreck.org> --- include/uapi/linux/io_uring.h | 7 ++++++ io_uring/fs.c | 57 +++++++++++++++++++++++++++++++++++++++++++ io_uring/fs.h | 3 +++ io_uring/opdef.c | 8 ++++++ 4 files changed, 75 insertions(+)