linux-next/io_uring/advise.c
Jens Axboe 3b87184f7e io_uring/advise: support 64-bit lengths
The existing fadvise/madvise support only supports 32-bit lengths. Add
support for 64-bit lengths, enabled by the application setting sqe->off
rather than sqe->len for the length. If sqe->len is set, then that is
used as the 32-bit length. If sqe->len is zero, then sqe->off is read
for full 64-bit support.

Older kernels will return -EINVAL if 64-bit support isn't available.

Fixes: 4840e418c2 ("io_uring: add IORING_OP_FADVISE")
Fixes: c1ca757bd6 ("io_uring: add IORING_OP_MADVISE")
Reported-by: Stefan <source@s.muenzel.net>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2024-06-16 14:54:55 -06:00

109 lines
2.4 KiB
C

// SPDX-License-Identifier: GPL-2.0
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/fs.h>
#include <linux/file.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/namei.h>
#include <linux/io_uring.h>
#include <uapi/linux/fadvise.h>
#include <uapi/linux/io_uring.h>
#include "io_uring.h"
#include "advise.h"
struct io_fadvise {
struct file *file;
u64 offset;
u64 len;
u32 advice;
};
struct io_madvise {
struct file *file;
u64 addr;
u64 len;
u32 advice;
};
int io_madvise_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
{
#if defined(CONFIG_ADVISE_SYSCALLS) && defined(CONFIG_MMU)
struct io_madvise *ma = io_kiocb_to_cmd(req, struct io_madvise);
if (sqe->buf_index || sqe->splice_fd_in)
return -EINVAL;
ma->addr = READ_ONCE(sqe->addr);
ma->len = READ_ONCE(sqe->off);
if (!ma->len)
ma->len = READ_ONCE(sqe->len);
ma->advice = READ_ONCE(sqe->fadvise_advice);
req->flags |= REQ_F_FORCE_ASYNC;
return 0;
#else
return -EOPNOTSUPP;
#endif
}
int io_madvise(struct io_kiocb *req, unsigned int issue_flags)
{
#if defined(CONFIG_ADVISE_SYSCALLS) && defined(CONFIG_MMU)
struct io_madvise *ma = io_kiocb_to_cmd(req, struct io_madvise);
int ret;
WARN_ON_ONCE(issue_flags & IO_URING_F_NONBLOCK);
ret = do_madvise(current->mm, ma->addr, ma->len, ma->advice);
io_req_set_res(req, ret, 0);
return IOU_OK;
#else
return -EOPNOTSUPP;
#endif
}
static bool io_fadvise_force_async(struct io_fadvise *fa)
{
switch (fa->advice) {
case POSIX_FADV_NORMAL:
case POSIX_FADV_RANDOM:
case POSIX_FADV_SEQUENTIAL:
return false;
default:
return true;
}
}
int io_fadvise_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
{
struct io_fadvise *fa = io_kiocb_to_cmd(req, struct io_fadvise);
if (sqe->buf_index || sqe->splice_fd_in)
return -EINVAL;
fa->offset = READ_ONCE(sqe->off);
fa->len = READ_ONCE(sqe->addr);
if (!fa->len)
fa->len = READ_ONCE(sqe->len);
fa->advice = READ_ONCE(sqe->fadvise_advice);
if (io_fadvise_force_async(fa))
req->flags |= REQ_F_FORCE_ASYNC;
return 0;
}
int io_fadvise(struct io_kiocb *req, unsigned int issue_flags)
{
struct io_fadvise *fa = io_kiocb_to_cmd(req, struct io_fadvise);
int ret;
WARN_ON_ONCE(issue_flags & IO_URING_F_NONBLOCK && io_fadvise_force_async(fa));
ret = vfs_fadvise(req->file, fa->offset, fa->len, fa->advice);
if (ret < 0)
req_set_fail(req);
io_req_set_res(req, ret, 0);
return IOU_OK;
}