linux-stable/io_uring/cancel.c

328 lines
7.4 KiB
C
Raw Normal View History

// 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>
2022-06-18 16:00:50 +00:00
#include <linux/nospec.h>
#include <linux/io_uring.h>
#include <uapi/linux/io_uring.h>
#include "io_uring.h"
#include "tctx.h"
#include "poll.h"
#include "timeout.h"
#include "cancel.h"
struct io_cancel {
struct file *file;
u64 addr;
u32 flags;
s32 fd;
};
#define CANCEL_FLAGS (IORING_ASYNC_CANCEL_ALL | IORING_ASYNC_CANCEL_FD | \
IORING_ASYNC_CANCEL_ANY | IORING_ASYNC_CANCEL_FD_FIXED | \
IORING_ASYNC_CANCEL_USERDATA)
/*
* Returns true if the request matches the criteria outlined by 'cd'.
*/
bool io_cancel_req_match(struct io_kiocb *req, struct io_cancel_data *cd)
{
bool match_user_data = cd->flags & IORING_ASYNC_CANCEL_USERDATA;
if (req->ctx != cd->ctx)
return false;
if (!(cd->flags & (IORING_ASYNC_CANCEL_FD)))
match_user_data = true;
if (cd->flags & IORING_ASYNC_CANCEL_ANY)
goto check_seq;
if (cd->flags & IORING_ASYNC_CANCEL_FD) {
if (req->file != cd->file)
return false;
}
if (match_user_data && req->cqe.user_data != cd->data)
return false;
if (cd->flags & IORING_ASYNC_CANCEL_ALL) {
check_seq:
if (cd->seq == req->work.cancel_seq)
return false;
req->work.cancel_seq = cd->seq;
}
return true;
}
static bool io_cancel_cb(struct io_wq_work *work, void *data)
{
struct io_kiocb *req = container_of(work, struct io_kiocb, work);
struct io_cancel_data *cd = data;
return io_cancel_req_match(req, cd);
}
static int io_async_cancel_one(struct io_uring_task *tctx,
struct io_cancel_data *cd)
{
enum io_wq_cancel cancel_ret;
int ret = 0;
bool all;
if (!tctx || !tctx->io_wq)
return -ENOENT;
all = cd->flags & (IORING_ASYNC_CANCEL_ALL|IORING_ASYNC_CANCEL_ANY);
cancel_ret = io_wq_cancel_cb(tctx->io_wq, io_cancel_cb, cd, all);
switch (cancel_ret) {
case IO_WQ_CANCEL_OK:
ret = 0;
break;
case IO_WQ_CANCEL_RUNNING:
ret = -EALREADY;
break;
case IO_WQ_CANCEL_NOTFOUND:
ret = -ENOENT;
break;
}
return ret;
}
int io_try_cancel(struct io_uring_task *tctx, struct io_cancel_data *cd,
unsigned issue_flags)
{
struct io_ring_ctx *ctx = cd->ctx;
int ret;
WARN_ON_ONCE(!io_wq_current_is_worker() && tctx != current->io_uring);
ret = io_async_cancel_one(tctx, cd);
/*
* Fall-through even for -EALREADY, as we may have poll armed
* that need unarming.
*/
if (!ret)
return 0;
ret = io_poll_cancel(ctx, cd, issue_flags);
if (ret != -ENOENT)
return ret;
spin_lock(&ctx->completion_lock);
if (!(cd->flags & IORING_ASYNC_CANCEL_FD))
ret = io_timeout_cancel(ctx, cd);
spin_unlock(&ctx->completion_lock);
return ret;
}
int io_async_cancel_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
{
struct io_cancel *cancel = io_kiocb_to_cmd(req, struct io_cancel);
if (unlikely(req->flags & REQ_F_BUFFER_SELECT))
return -EINVAL;
if (sqe->off || sqe->len || sqe->splice_fd_in)
return -EINVAL;
cancel->addr = READ_ONCE(sqe->addr);
cancel->flags = READ_ONCE(sqe->cancel_flags);
if (cancel->flags & ~CANCEL_FLAGS)
return -EINVAL;
if (cancel->flags & IORING_ASYNC_CANCEL_FD) {
if (cancel->flags & IORING_ASYNC_CANCEL_ANY)
return -EINVAL;
cancel->fd = READ_ONCE(sqe->fd);
}
return 0;
}
static int __io_async_cancel(struct io_cancel_data *cd,
struct io_uring_task *tctx,
unsigned int issue_flags)
{
bool all = cd->flags & (IORING_ASYNC_CANCEL_ALL|IORING_ASYNC_CANCEL_ANY);
struct io_ring_ctx *ctx = cd->ctx;
struct io_tctx_node *node;
int ret, nr = 0;
do {
ret = io_try_cancel(tctx, cd, issue_flags);
if (ret == -ENOENT)
break;
if (!all)
return ret;
nr++;
} while (1);
/* slow path, try all io-wq's */
io_ring_submit_lock(ctx, issue_flags);
ret = -ENOENT;
list_for_each_entry(node, &ctx->tctx_list, ctx_node) {
struct io_uring_task *tctx = node->task->io_uring;
ret = io_async_cancel_one(tctx, cd);
if (ret != -ENOENT) {
if (!all)
break;
nr++;
}
}
io_ring_submit_unlock(ctx, issue_flags);
return all ? nr : ret;
}
int io_async_cancel(struct io_kiocb *req, unsigned int issue_flags)
{
struct io_cancel *cancel = io_kiocb_to_cmd(req, struct io_cancel);
struct io_cancel_data cd = {
.ctx = req->ctx,
.data = cancel->addr,
.flags = cancel->flags,
.seq = atomic_inc_return(&req->ctx->cancel_seq),
};
struct io_uring_task *tctx = req->task->io_uring;
int ret;
if (cd.flags & IORING_ASYNC_CANCEL_FD) {
if (req->flags & REQ_F_FIXED_FILE ||
cd.flags & IORING_ASYNC_CANCEL_FD_FIXED) {
req->flags |= REQ_F_FIXED_FILE;
req->file = io_file_get_fixed(req, cancel->fd,
issue_flags);
} else {
req->file = io_file_get_normal(req, cancel->fd);
}
if (!req->file) {
ret = -EBADF;
goto done;
}
cd.file = req->file;
}
ret = __io_async_cancel(&cd, tctx, issue_flags);
done:
if (ret < 0)
req_set_fail(req);
io_req_set_res(req, ret, 0);
return IOU_OK;
}
void init_hash_table(struct io_hash_table *table, unsigned size)
{
unsigned int i;
for (i = 0; i < size; i++) {
spin_lock_init(&table->hbs[i].lock);
INIT_HLIST_HEAD(&table->hbs[i].list);
}
}
2022-06-18 16:00:50 +00:00
static int __io_sync_cancel(struct io_uring_task *tctx,
struct io_cancel_data *cd, int fd)
{
struct io_ring_ctx *ctx = cd->ctx;
/* fixed must be grabbed every time since we drop the uring_lock */
if ((cd->flags & IORING_ASYNC_CANCEL_FD) &&
(cd->flags & IORING_ASYNC_CANCEL_FD_FIXED)) {
if (unlikely(fd >= ctx->nr_user_files))
2022-06-18 16:00:50 +00:00
return -EBADF;
fd = array_index_nospec(fd, ctx->nr_user_files);
cd->file = io_file_from_index(&ctx->file_table, fd);
2022-06-18 16:00:50 +00:00
if (!cd->file)
return -EBADF;
}
return __io_async_cancel(cd, tctx, 0);
}
int io_sync_cancel(struct io_ring_ctx *ctx, void __user *arg)
__must_hold(&ctx->uring_lock)
{
struct io_cancel_data cd = {
.ctx = ctx,
.seq = atomic_inc_return(&ctx->cancel_seq),
};
ktime_t timeout = KTIME_MAX;
struct io_uring_sync_cancel_reg sc;
struct fd f = { };
DEFINE_WAIT(wait);
int ret;
if (copy_from_user(&sc, arg, sizeof(sc)))
return -EFAULT;
if (sc.flags & ~CANCEL_FLAGS)
return -EINVAL;
if (sc.pad[0] || sc.pad[1] || sc.pad[2] || sc.pad[3])
return -EINVAL;
cd.data = sc.addr;
cd.flags = sc.flags;
/* we can grab a normal file descriptor upfront */
if ((cd.flags & IORING_ASYNC_CANCEL_FD) &&
!(cd.flags & IORING_ASYNC_CANCEL_FD_FIXED)) {
f = fdget(sc.fd);
if (!f.file)
return -EBADF;
cd.file = f.file;
}
ret = __io_sync_cancel(current->io_uring, &cd, sc.fd);
/* found something, done! */
if (ret != -EALREADY)
goto out;
if (sc.timeout.tv_sec != -1UL || sc.timeout.tv_nsec != -1UL) {
struct timespec64 ts = {
.tv_sec = sc.timeout.tv_sec,
.tv_nsec = sc.timeout.tv_nsec
};
timeout = ktime_add_ns(timespec64_to_ktime(ts), ktime_get_ns());
}
/*
* Keep looking until we get -ENOENT. we'll get woken everytime
* every time a request completes and will retry the cancelation.
*/
do {
cd.seq = atomic_inc_return(&ctx->cancel_seq);
prepare_to_wait(&ctx->cq_wait, &wait, TASK_INTERRUPTIBLE);
ret = __io_sync_cancel(current->io_uring, &cd, sc.fd);
mutex_unlock(&ctx->uring_lock);
2022-06-18 16:00:50 +00:00
if (ret != -EALREADY)
break;
io_uring: add IORING_SETUP_DEFER_TASKRUN Allow deferring async tasks until the user calls io_uring_enter(2) with the IORING_ENTER_GETEVENTS flag. Enable this mode with a flag at io_uring_setup time. This functionality requires that the later io_uring_enter will be called from the same submission task, and therefore restrict this flag to work only when IORING_SETUP_SINGLE_ISSUER is also set. Being able to hand pick when tasks are run prevents the problem where there is current work to be done, however task work runs anyway. For example, a common workload would obtain a batch of CQEs, and process each one. Interrupting this to additional taskwork would add latency but not gain anything. If instead task work is deferred to just before more CQEs are obtained then no additional latency is added. The way this is implemented is by trying to keep task work local to a io_ring_ctx, rather than to the submission task. This is required, as the application will want to wake up only a single io_ring_ctx at a time to process work, and so the lists of work have to be kept separate. This has some other benefits like not having to check the task continually in handle_tw_list (and potentially unlocking/locking those), and reducing locks in the submit & process completions path. There are networking cases where using this option can reduce request latency by 50%. For example a contrived example using [1] where the client sends 2k data and receives the same data back while doing some system calls (to trigger task work) shows this reduction. The reason ends up being that if sending responses is delayed by processing task work, then the client side sits idle. Whereas reordering the sends first means that the client runs it's workload in parallel with the local task work. [1]: Using https://github.com/DylanZA/netbench/tree/defer_run Client: ./netbench --client_only 1 --control_port 10000 --host <host> --tx "epoll --threads 16 --per_thread 1 --size 2048 --resp 2048 --workload 1000" Server: ./netbench --server_only 1 --control_port 10000 --rx "io_uring --defer_taskrun 0 --workload 100" --rx "io_uring --defer_taskrun 1 --workload 100" Signed-off-by: Dylan Yudaken <dylany@fb.com> Link: https://lore.kernel.org/r/20220830125013.570060-5-dylany@fb.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2022-08-30 12:50:10 +00:00
ret = io_run_task_work_sig(ctx);
if (ret < 0)
2022-06-18 16:00:50 +00:00
break;
ret = schedule_hrtimeout(&timeout, HRTIMER_MODE_ABS);
if (!ret) {
ret = -ETIME;
break;
}
mutex_lock(&ctx->uring_lock);
2022-06-18 16:00:50 +00:00
} while (1);
finish_wait(&ctx->cq_wait, &wait);
mutex_lock(&ctx->uring_lock);
2022-06-18 16:00:50 +00:00
if (ret == -ENOENT || ret > 0)
ret = 0;
out:
fdput(f);
return ret;
}