#define IORING_MAX_CQ_ENTRIES (2 * IORING_MAX_ENTRIES)
#define IORING_SQPOLL_CAP_ENTRIES_VALUE 8
-/* 512 entries per page on 64-bit archs, 64 pages max */
+/* only define max */
#define IORING_MAX_FIXED_FILES (1U << 15)
#define IORING_MAX_RESTRICTIONS (IORING_RESTRICTION_LAST + \
IORING_REGISTER_LAST + IORING_OP_LAST)
-#define IO_RSRC_TAG_TABLE_SHIFT 9
+#define IO_RSRC_TAG_TABLE_SHIFT (PAGE_SHIFT - 3)
#define IO_RSRC_TAG_TABLE_MAX (1U << IO_RSRC_TAG_TABLE_SHIFT)
#define IO_RSRC_TAG_TABLE_MASK (IO_RSRC_TAG_TABLE_MAX - 1)
struct io_submit_state submit_state;
struct list_head timeout_list;
+ struct list_head ltimeout_list;
struct list_head cq_overflow_list;
struct xarray io_buffers;
struct xarray personalities;
struct hrtimer timer;
struct timespec64 ts;
enum hrtimer_mode mode;
+ u32 flags;
};
struct io_accept {
struct sockaddr __user *addr;
int __user *addr_len;
int flags;
+ u32 file_slot;
unsigned long nofile;
};
/* timeout update */
struct timespec64 ts;
u32 flags;
+ bool ltimeout;
};
struct io_rw {
struct io_open {
struct file *file;
int dfd;
+ u32 file_slot;
struct filename *filename;
struct open_how how;
unsigned long nofile;
struct io_poll_iocb *double_poll;
};
-typedef void (*io_req_tw_func_t)(struct io_kiocb *req);
+typedef void (*io_req_tw_func_t)(struct io_kiocb *req, bool *locked);
struct io_task_work {
union {
},
[IORING_OP_WRITE] = {
.needs_file = 1,
+ .hash_reg_file = 1,
.unbound_nonreg_file = 1,
.pollout = 1,
.plug = 1,
static void io_submit_flush_completions(struct io_ring_ctx *ctx);
static int io_req_prep_async(struct io_kiocb *req);
+static int io_install_fixed_file(struct io_kiocb *req, struct file *file,
+ unsigned int issue_flags, u32 slot_index);
+static enum hrtimer_restart io_link_timeout_fn(struct hrtimer *timer);
+
static struct kmem_cache *req_cachep;
static const struct file_operations io_uring_fops;
}
EXPORT_SYMBOL(io_uring_get_socket);
+static inline void io_tw_lock(struct io_ring_ctx *ctx, bool *locked)
+{
+ if (!*locked) {
+ mutex_lock(&ctx->uring_lock);
+ *locked = true;
+ }
+}
+
#define io_for_each_link(pos, head) \
for (pos = (head); pos; pos = pos->link)
req->flags |= REQ_F_FAIL;
}
+static inline void req_fail_link_node(struct io_kiocb *req, int res)
+{
+ req_set_fail(req);
+ req->result = res;
+}
+
static void io_ring_ctx_ref_free(struct percpu_ref *ref)
{
struct io_ring_ctx *ctx = container_of(ref, struct io_ring_ctx, refs);
fallback_work.work);
struct llist_node *node = llist_del_all(&ctx->fallback_llist);
struct io_kiocb *req, *tmp;
+ bool locked = false;
percpu_ref_get(&ctx->refs);
llist_for_each_entry_safe(req, tmp, node, io_task_work.fallback_node)
- req->io_task_work.func(req);
+ req->io_task_work.func(req, &locked);
+
+ if (locked) {
+ if (ctx->submit_state.compl_nr)
+ io_submit_flush_completions(ctx);
+ mutex_unlock(&ctx->uring_lock);
+ }
percpu_ref_put(&ctx->refs);
+
}
static struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p)
INIT_LIST_HEAD(&ctx->iopoll_list);
INIT_LIST_HEAD(&ctx->defer_list);
INIT_LIST_HEAD(&ctx->timeout_list);
+ INIT_LIST_HEAD(&ctx->ltimeout_list);
spin_lock_init(&ctx->rsrc_ref_lock);
INIT_LIST_HEAD(&ctx->rsrc_ref_list);
INIT_DELAYED_WORK(&ctx->rsrc_put_work, io_rsrc_put_work);
}
}
-static void io_queue_async_work(struct io_kiocb *req)
+static void io_queue_async_work(struct io_kiocb *req, bool *locked)
{
struct io_ring_ctx *ctx = req->ctx;
struct io_kiocb *link = io_prep_linked_timeout(req);
struct io_uring_task *tctx = req->task->io_uring;
+ /* must not take the lock, NULL it as a precaution */
+ locked = NULL;
+
BUG_ON(!tctx);
BUG_ON(!tctx->io_wq);
return !ctx->eventfd_async || io_wq_current_is_worker();
}
+/*
+ * This should only get called when at least one event has been posted.
+ * Some applications rely on the eventfd notification count only changing
+ * IFF a new CQE has been added to the CQ ring. There's no depedency on
+ * 1:1 relationship between how many times this function is called (and
+ * hence the eventfd count) and number of CQEs posted to the CQ ring.
+ */
static void io_cqring_ev_posted(struct io_ring_ctx *ctx)
{
/*
{
struct io_uring_task *tctx = task->io_uring;
- percpu_counter_sub(&tctx->inflight, nr);
- if (unlikely(atomic_read(&tctx->in_idle)))
- wake_up(&tctx->wait);
- put_task_struct_many(task, nr);
+ if (likely(task == current)) {
+ tctx->cached_refs += nr;
+ } else {
+ percpu_counter_sub(&tctx->inflight, nr);
+ if (unlikely(atomic_read(&tctx->in_idle)))
+ wake_up(&tctx->wait);
+ put_task_struct_many(task, nr);
+ }
+}
+
+static void io_task_refs_refill(struct io_uring_task *tctx)
+{
+ unsigned int refill = -tctx->cached_refs + IO_TCTX_REFS_CACHE_NR;
+
+ percpu_counter_add(&tctx->inflight, refill);
+ refcount_add(refill, ¤t->usage);
+ tctx->cached_refs += refill;
+}
+
+static inline void io_get_task_refs(int nr)
+{
+ struct io_uring_task *tctx = current->io_uring;
+
+ tctx->cached_refs -= nr;
+ if (unlikely(tctx->cached_refs < 0))
+ io_task_refs_refill(tctx);
}
static bool io_cqring_event_overflow(struct io_ring_ctx *ctx, u64 user_data,
io_req_complete_post(req, res, 0);
}
+static void io_req_complete_fail_submit(struct io_kiocb *req)
+{
+ /*
+ * We don't submit, fail them all, for that replace hardlinks with
+ * normal links. Extra REQ_F_LINK is tolerated.
+ */
+ req->flags &= ~REQ_F_HARDLINK;
+ req->flags |= REQ_F_LINK;
+ io_req_complete_failed(req, req->result);
+}
+
/*
* Don't initialise the fields below on every allocation, but do that in
* advance and keep them valid across allocations.
io_remove_next_linked(req);
link->timeout.head = NULL;
if (hrtimer_try_to_cancel(&io->timer) != -1) {
+ list_del(&link->timeout.list);
io_cqring_fill_event(link->ctx, link->user_data,
-ECANCELED, 0);
io_put_req_deferred(link);
req->link = NULL;
while (link) {
+ long res = -ECANCELED;
+
+ if (link->flags & REQ_F_FAIL)
+ res = link->result;
+
nxt = link->link;
link->link = NULL;
trace_io_uring_fail_link(req, link);
- io_cqring_fill_event(link->ctx, link->user_data, -ECANCELED, 0);
+ io_cqring_fill_event(link->ctx, link->user_data, res, 0);
io_put_req_deferred(link);
link = nxt;
}
return __io_req_find_next(req);
}
-static void ctx_flush_and_put(struct io_ring_ctx *ctx)
+static void ctx_flush_and_put(struct io_ring_ctx *ctx, bool *locked)
{
if (!ctx)
return;
- if (ctx->submit_state.compl_nr) {
- mutex_lock(&ctx->uring_lock);
- io_submit_flush_completions(ctx);
+ if (*locked) {
+ if (ctx->submit_state.compl_nr)
+ io_submit_flush_completions(ctx);
mutex_unlock(&ctx->uring_lock);
+ *locked = false;
}
percpu_ref_put(&ctx->refs);
}
static void tctx_task_work(struct callback_head *cb)
{
+ bool locked = false;
struct io_ring_ctx *ctx = NULL;
struct io_uring_task *tctx = container_of(cb, struct io_uring_task,
task_work);
io_task_work.node);
if (req->ctx != ctx) {
- ctx_flush_and_put(ctx);
+ ctx_flush_and_put(ctx, &locked);
ctx = req->ctx;
+ /* if not contended, grab and improve batching */
+ locked = mutex_trylock(&ctx->uring_lock);
percpu_ref_get(&ctx->refs);
}
- req->io_task_work.func(req);
+ req->io_task_work.func(req, &locked);
node = next;
} while (node);
cond_resched();
}
- ctx_flush_and_put(ctx);
+ ctx_flush_and_put(ctx, &locked);
}
static void io_req_task_work_add(struct io_kiocb *req)
}
}
-static void io_req_task_cancel(struct io_kiocb *req)
+static void io_req_task_cancel(struct io_kiocb *req, bool *locked)
{
struct io_ring_ctx *ctx = req->ctx;
- /* ctx is guaranteed to stay alive while we hold uring_lock */
- mutex_lock(&ctx->uring_lock);
+ /* not needed for normal modes, but SQPOLL depends on it */
+ io_tw_lock(ctx, locked);
io_req_complete_failed(req, req->result);
- mutex_unlock(&ctx->uring_lock);
}
-static void io_req_task_submit(struct io_kiocb *req)
+static void io_req_task_submit(struct io_kiocb *req, bool *locked)
{
struct io_ring_ctx *ctx = req->ctx;
- /* ctx stays valid until unlock, even if we drop all ours ctx->refs */
- mutex_lock(&ctx->uring_lock);
+ io_tw_lock(ctx, locked);
+ /* req->task == current here, checking PF_EXITING is safe */
if (likely(!(req->task->flags & PF_EXITING)))
__io_queue_sqe(req);
else
io_req_complete_failed(req, -EFAULT);
- mutex_unlock(&ctx->uring_lock);
}
static void io_req_task_queue_fail(struct io_kiocb *req, int ret)
__io_free_req(req);
}
+static void io_free_req_work(struct io_kiocb *req, bool *locked)
+{
+ io_free_req(req);
+}
+
struct req_batch {
struct task_struct *task;
int task_refs;
{
if (rb->ctx_refs)
percpu_ref_put_many(&ctx->refs, rb->ctx_refs);
- if (rb->task == current)
- current->io_uring->cached_refs += rb->task_refs;
- else if (rb->task)
+ if (rb->task)
io_put_task(rb->task, rb->task_refs);
}
static inline void io_put_req_deferred(struct io_kiocb *req)
{
if (req_ref_put_and_test(req)) {
- req->io_task_work.func = io_free_req;
+ req->io_task_work.func = io_free_req_work;
io_req_task_work_add(req);
}
}
return false;
}
-static void io_req_task_complete(struct io_kiocb *req)
+static void io_req_task_complete(struct io_kiocb *req, bool *locked)
{
- __io_req_complete(req, 0, req->result, io_put_rw_kbuf(req));
+ unsigned int cflags = io_put_rw_kbuf(req);
+ long res = req->result;
+
+ if (*locked) {
+ struct io_ring_ctx *ctx = req->ctx;
+ struct io_submit_state *state = &ctx->submit_state;
+
+ io_req_complete_state(req, res, cflags);
+ state->compl_reqs[state->compl_nr++] = req;
+ if (state->compl_nr == ARRAY_SIZE(state->compl_reqs))
+ io_submit_flush_completions(ctx);
+ } else {
+ io_req_complete_post(req, res, cflags);
+ }
}
static void __io_complete_rw(struct io_kiocb *req, long res, long res2,
{
if (__io_complete_rw_common(req, res))
return;
- io_req_task_complete(req);
+ __io_req_complete(req, 0, req->result, io_put_rw_kbuf(req));
}
static void io_complete_rw(struct kiocb *kiocb, long res, long res2)
return -EINVAL;
}
+static bool need_read_all(struct io_kiocb *req)
+{
+ return req->flags & REQ_F_ISREG ||
+ S_ISBLK(file_inode(req->file)->i_mode);
+}
+
static int io_read(struct io_kiocb *req, unsigned int issue_flags)
{
struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs;
} else if (ret == -EIOCBQUEUED) {
goto out_free;
} else if (ret <= 0 || ret == io_size || !force_nonblock ||
- (req->flags & REQ_F_NOWAIT) || !(req->flags & REQ_F_ISREG)) {
+ (req->flags & REQ_F_NOWAIT) || !need_read_all(req)) {
/* read all, failed, already did sync or don't want to retry */
goto done;
}
if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
return -EINVAL;
- if (sqe->ioprio || sqe->buf_index)
+ if (sqe->ioprio || sqe->buf_index || sqe->splice_fd_in)
return -EINVAL;
if (unlikely(req->flags & REQ_F_FIXED_FILE))
return -EBADF;
if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
return -EINVAL;
- if (sqe->ioprio || sqe->off || sqe->len || sqe->buf_index)
+ if (sqe->ioprio || sqe->off || sqe->len || sqe->buf_index ||
+ sqe->splice_fd_in)
return -EINVAL;
if (unlikely(req->flags & REQ_F_FIXED_FILE))
return -EBADF;
#if defined(CONFIG_NET)
if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
return -EINVAL;
- if (sqe->ioprio || sqe->off || sqe->addr || sqe->rw_flags ||
- sqe->buf_index)
+ if (unlikely(sqe->ioprio || sqe->off || sqe->addr || sqe->rw_flags ||
+ sqe->buf_index || sqe->splice_fd_in))
return -EINVAL;
req->shutdown.how = READ_ONCE(sqe->len);
if (unlikely(ctx->flags & IORING_SETUP_IOPOLL))
return -EINVAL;
- if (unlikely(sqe->addr || sqe->ioprio || sqe->buf_index))
+ if (unlikely(sqe->addr || sqe->ioprio || sqe->buf_index ||
+ sqe->splice_fd_in))
return -EINVAL;
req->sync.flags = READ_ONCE(sqe->fsync_flags);
static int io_fallocate_prep(struct io_kiocb *req,
const struct io_uring_sqe *sqe)
{
- if (sqe->ioprio || sqe->buf_index || sqe->rw_flags)
+ if (sqe->ioprio || sqe->buf_index || sqe->rw_flags ||
+ sqe->splice_fd_in)
return -EINVAL;
if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
return -EINVAL;
req->open.filename = NULL;
return ret;
}
+
+ req->open.file_slot = READ_ONCE(sqe->file_index);
+ if (req->open.file_slot && (req->open.how.flags & O_CLOEXEC))
+ return -EINVAL;
+
req->open.nofile = rlimit(RLIMIT_NOFILE);
req->flags |= REQ_F_NEED_CLEANUP;
return 0;
{
struct open_flags op;
struct file *file;
- bool nonblock_set;
- bool resolve_nonblock;
+ bool resolve_nonblock, nonblock_set;
+ bool fixed = !!req->open.file_slot;
int ret;
ret = build_open_flags(&req->open.how, &op);
op.open_flag |= O_NONBLOCK;
}
- ret = __get_unused_fd_flags(req->open.how.flags, req->open.nofile);
- if (ret < 0)
- goto err;
+ if (!fixed) {
+ ret = __get_unused_fd_flags(req->open.how.flags, req->open.nofile);
+ if (ret < 0)
+ goto err;
+ }
file = do_filp_open(req->open.dfd, req->open.filename, &op);
if (IS_ERR(file)) {
* marginal gain for something that is now known to be a slower
* path. So just put it, and we'll get a new one when we retry.
*/
- put_unused_fd(ret);
+ if (!fixed)
+ put_unused_fd(ret);
ret = PTR_ERR(file);
/* only retry if RESOLVE_CACHED wasn't already set by application */
if ((issue_flags & IO_URING_F_NONBLOCK) && !nonblock_set)
file->f_flags &= ~O_NONBLOCK;
fsnotify_open(file);
- fd_install(ret, file);
+
+ if (!fixed)
+ fd_install(ret, file);
+ else
+ ret = io_install_fixed_file(req, file, issue_flags,
+ req->open.file_slot - 1);
err:
putname(req->open.filename);
req->flags &= ~REQ_F_NEED_CLEANUP;
struct io_provide_buf *p = &req->pbuf;
u64 tmp;
- if (sqe->ioprio || sqe->rw_flags || sqe->addr || sqe->len || sqe->off)
+ if (sqe->ioprio || sqe->rw_flags || sqe->addr || sqe->len || sqe->off ||
+ sqe->splice_fd_in)
return -EINVAL;
tmp = READ_ONCE(sqe->fd);
struct io_provide_buf *p = &req->pbuf;
u64 tmp;
- if (sqe->ioprio || sqe->rw_flags)
+ if (sqe->ioprio || sqe->rw_flags || sqe->splice_fd_in)
return -EINVAL;
tmp = READ_ONCE(sqe->fd);
const struct io_uring_sqe *sqe)
{
#if defined(CONFIG_EPOLL)
- if (sqe->ioprio || sqe->buf_index)
+ if (sqe->ioprio || sqe->buf_index || sqe->splice_fd_in)
return -EINVAL;
if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
return -EINVAL;
static int io_madvise_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
{
#if defined(CONFIG_ADVISE_SYSCALLS) && defined(CONFIG_MMU)
- if (sqe->ioprio || sqe->buf_index || sqe->off)
+ if (sqe->ioprio || sqe->buf_index || sqe->off || sqe->splice_fd_in)
return -EINVAL;
if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
return -EINVAL;
static int io_fadvise_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
{
- if (sqe->ioprio || sqe->buf_index || sqe->addr)
+ if (sqe->ioprio || sqe->buf_index || sqe->addr || sqe->splice_fd_in)
return -EINVAL;
if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
return -EINVAL;
{
if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
return -EINVAL;
- if (sqe->ioprio || sqe->buf_index)
+ if (sqe->ioprio || sqe->buf_index || sqe->splice_fd_in)
return -EINVAL;
if (req->flags & REQ_F_FIXED_FILE)
return -EBADF;
if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
return -EINVAL;
if (sqe->ioprio || sqe->off || sqe->addr || sqe->len ||
- sqe->rw_flags || sqe->buf_index)
+ sqe->rw_flags || sqe->buf_index || sqe->splice_fd_in)
return -EINVAL;
if (req->flags & REQ_F_FIXED_FILE)
return -EBADF;
if (unlikely(ctx->flags & IORING_SETUP_IOPOLL))
return -EINVAL;
- if (unlikely(sqe->addr || sqe->ioprio || sqe->buf_index))
+ if (unlikely(sqe->addr || sqe->ioprio || sqe->buf_index ||
+ sqe->splice_fd_in))
return -EINVAL;
req->sync.off = READ_ONCE(sqe->off);
accept->addr_len = u64_to_user_ptr(READ_ONCE(sqe->addr2));
accept->flags = READ_ONCE(sqe->accept_flags);
accept->nofile = rlimit(RLIMIT_NOFILE);
+
+ accept->file_slot = READ_ONCE(sqe->file_index);
+ if (accept->file_slot && ((req->open.how.flags & O_CLOEXEC) ||
+ (accept->flags & SOCK_CLOEXEC)))
+ return -EINVAL;
+ if (accept->flags & ~(SOCK_CLOEXEC | SOCK_NONBLOCK))
+ return -EINVAL;
+ if (SOCK_NONBLOCK != O_NONBLOCK && (accept->flags & SOCK_NONBLOCK))
+ accept->flags = (accept->flags & ~SOCK_NONBLOCK) | O_NONBLOCK;
return 0;
}
struct io_accept *accept = &req->accept;
bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK;
unsigned int file_flags = force_nonblock ? O_NONBLOCK : 0;
- int ret;
+ bool fixed = !!accept->file_slot;
+ struct file *file;
+ int ret, fd;
if (req->file->f_flags & O_NONBLOCK)
req->flags |= REQ_F_NOWAIT;
- ret = __sys_accept4_file(req->file, file_flags, accept->addr,
- accept->addr_len, accept->flags,
- accept->nofile);
- if (ret == -EAGAIN && force_nonblock)
- return -EAGAIN;
- if (ret < 0) {
+ if (!fixed) {
+ fd = __get_unused_fd_flags(accept->flags, accept->nofile);
+ if (unlikely(fd < 0))
+ return fd;
+ }
+ file = do_accept(req->file, file_flags, accept->addr, accept->addr_len,
+ accept->flags);
+ if (IS_ERR(file)) {
+ if (!fixed)
+ put_unused_fd(fd);
+ ret = PTR_ERR(file);
+ if (ret == -EAGAIN && force_nonblock)
+ return -EAGAIN;
if (ret == -ERESTARTSYS)
ret = -EINTR;
req_set_fail(req);
+ } else if (!fixed) {
+ fd_install(fd, file);
+ ret = fd;
+ } else {
+ ret = io_install_fixed_file(req, file, issue_flags,
+ accept->file_slot - 1);
}
__io_req_complete(req, issue_flags, ret, 0);
return 0;
if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
return -EINVAL;
- if (sqe->ioprio || sqe->len || sqe->buf_index || sqe->rw_flags)
+ if (sqe->ioprio || sqe->len || sqe->buf_index || sqe->rw_flags ||
+ sqe->splice_fd_in)
return -EINVAL;
conn->addr = u64_to_user_ptr(READ_ONCE(sqe->addr));
{
struct io_ring_ctx *ctx = req->ctx;
+ /* req->task == current here, checking PF_EXITING is safe */
if (unlikely(req->task->flags & PF_EXITING))
WRITE_ONCE(poll->canceled, true);
return !(flags & IORING_CQE_F_MORE);
}
-static void io_poll_task_func(struct io_kiocb *req)
+static void io_poll_task_func(struct io_kiocb *req, bool *locked)
{
struct io_ring_ctx *ctx = req->ctx;
struct io_kiocb *nxt;
if (done) {
nxt = io_put_req_find_next(req);
if (nxt)
- io_req_task_submit(nxt);
+ io_req_task_submit(nxt, locked);
}
}
}
__io_queue_proc(&apoll->poll, pt, head, &apoll->double_poll);
}
-static void io_async_task_func(struct io_kiocb *req)
+static void io_async_task_func(struct io_kiocb *req, bool *locked)
{
struct async_poll *apoll = req->apoll;
struct io_ring_ctx *ctx = req->ctx;
spin_unlock(&ctx->completion_lock);
if (!READ_ONCE(apoll->poll.canceled))
- io_req_task_submit(req);
+ io_req_task_submit(req, locked);
else
io_req_complete_failed(req, -ECANCELED);
}
if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
return -EINVAL;
- if (sqe->ioprio || sqe->buf_index)
+ if (sqe->ioprio || sqe->buf_index || sqe->splice_fd_in)
return -EINVAL;
flags = READ_ONCE(sqe->len);
if (flags & ~(IORING_POLL_UPDATE_EVENTS | IORING_POLL_UPDATE_USER_DATA |
return 0;
}
-static void io_req_task_timeout(struct io_kiocb *req)
+static void io_req_task_timeout(struct io_kiocb *req, bool *locked)
{
req_set_fail(req);
io_req_complete_post(req, -ETIME, 0);
}
static int io_timeout_cancel(struct io_ring_ctx *ctx, __u64 user_data)
+ __must_hold(&ctx->completion_lock)
__must_hold(&ctx->timeout_lock)
{
struct io_kiocb *req = io_timeout_extract(ctx, user_data);
return 0;
}
+static clockid_t io_timeout_get_clock(struct io_timeout_data *data)
+{
+ switch (data->flags & IORING_TIMEOUT_CLOCK_MASK) {
+ case IORING_TIMEOUT_BOOTTIME:
+ return CLOCK_BOOTTIME;
+ case IORING_TIMEOUT_REALTIME:
+ return CLOCK_REALTIME;
+ default:
+ /* can't happen, vetted at prep time */
+ WARN_ON_ONCE(1);
+ fallthrough;
+ case 0:
+ return CLOCK_MONOTONIC;
+ }
+}
+
+static int io_linked_timeout_update(struct io_ring_ctx *ctx, __u64 user_data,
+ struct timespec64 *ts, enum hrtimer_mode mode)
+ __must_hold(&ctx->timeout_lock)
+{
+ struct io_timeout_data *io;
+ struct io_kiocb *req;
+ bool found = false;
+
+ list_for_each_entry(req, &ctx->ltimeout_list, timeout.list) {
+ found = user_data == req->user_data;
+ if (found)
+ break;
+ }
+ if (!found)
+ return -ENOENT;
+
+ io = req->async_data;
+ if (hrtimer_try_to_cancel(&io->timer) == -1)
+ return -EALREADY;
+ hrtimer_init(&io->timer, io_timeout_get_clock(io), mode);
+ io->timer.function = io_link_timeout_fn;
+ hrtimer_start(&io->timer, timespec64_to_ktime(*ts), mode);
+ return 0;
+}
+
static int io_timeout_update(struct io_ring_ctx *ctx, __u64 user_data,
struct timespec64 *ts, enum hrtimer_mode mode)
__must_hold(&ctx->timeout_lock)
req->timeout.off = 0; /* noseq */
data = req->async_data;
list_add_tail(&req->timeout.list, &ctx->timeout_list);
- hrtimer_init(&data->timer, CLOCK_MONOTONIC, mode);
+ hrtimer_init(&data->timer, io_timeout_get_clock(data), mode);
data->timer.function = io_timeout_fn;
hrtimer_start(&data->timer, timespec64_to_ktime(*ts), mode);
return 0;
return -EINVAL;
if (unlikely(req->flags & (REQ_F_FIXED_FILE | REQ_F_BUFFER_SELECT)))
return -EINVAL;
- if (sqe->ioprio || sqe->buf_index || sqe->len)
+ if (sqe->ioprio || sqe->buf_index || sqe->len || sqe->splice_fd_in)
return -EINVAL;
+ tr->ltimeout = false;
tr->addr = READ_ONCE(sqe->addr);
tr->flags = READ_ONCE(sqe->timeout_flags);
- if (tr->flags & IORING_TIMEOUT_UPDATE) {
- if (tr->flags & ~(IORING_TIMEOUT_UPDATE|IORING_TIMEOUT_ABS))
+ if (tr->flags & IORING_TIMEOUT_UPDATE_MASK) {
+ if (hweight32(tr->flags & IORING_TIMEOUT_CLOCK_MASK) > 1)
+ return -EINVAL;
+ if (tr->flags & IORING_LINK_TIMEOUT_UPDATE)
+ tr->ltimeout = true;
+ if (tr->flags & ~(IORING_TIMEOUT_UPDATE_MASK|IORING_TIMEOUT_ABS))
return -EINVAL;
if (get_timespec64(&tr->ts, u64_to_user_ptr(sqe->addr2)))
return -EFAULT;
struct io_ring_ctx *ctx = req->ctx;
int ret;
- spin_lock_irq(&ctx->timeout_lock);
- if (!(req->timeout_rem.flags & IORING_TIMEOUT_UPDATE))
+ if (!(req->timeout_rem.flags & IORING_TIMEOUT_UPDATE)) {
+ spin_lock(&ctx->completion_lock);
+ spin_lock_irq(&ctx->timeout_lock);
ret = io_timeout_cancel(ctx, tr->addr);
- else
- ret = io_timeout_update(ctx, tr->addr, &tr->ts,
- io_translate_timeout_mode(tr->flags));
- spin_unlock_irq(&ctx->timeout_lock);
+ spin_unlock_irq(&ctx->timeout_lock);
+ spin_unlock(&ctx->completion_lock);
+ } else {
+ enum hrtimer_mode mode = io_translate_timeout_mode(tr->flags);
+
+ spin_lock_irq(&ctx->timeout_lock);
+ if (tr->ltimeout)
+ ret = io_linked_timeout_update(ctx, tr->addr, &tr->ts, mode);
+ else
+ ret = io_timeout_update(ctx, tr->addr, &tr->ts, mode);
+ spin_unlock_irq(&ctx->timeout_lock);
+ }
if (ret < 0)
req_set_fail(req);
if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
return -EINVAL;
- if (sqe->ioprio || sqe->buf_index || sqe->len != 1)
+ if (sqe->ioprio || sqe->buf_index || sqe->len != 1 ||
+ sqe->splice_fd_in)
return -EINVAL;
if (off && is_timeout_link)
return -EINVAL;
flags = READ_ONCE(sqe->timeout_flags);
- if (flags & ~IORING_TIMEOUT_ABS)
+ if (flags & ~(IORING_TIMEOUT_ABS | IORING_TIMEOUT_CLOCK_MASK))
+ return -EINVAL;
+ /* more than one clock specified is invalid, obviously */
+ if (hweight32(flags & IORING_TIMEOUT_CLOCK_MASK) > 1)
return -EINVAL;
+ INIT_LIST_HEAD(&req->timeout.list);
req->timeout.off = off;
if (unlikely(off && !req->ctx->off_timeout_used))
req->ctx->off_timeout_used = true;
data = req->async_data;
data->req = req;
+ data->flags = flags;
if (get_timespec64(&data->ts, u64_to_user_ptr(sqe->addr)))
return -EFAULT;
data->mode = io_translate_timeout_mode(flags);
- hrtimer_init(&data->timer, CLOCK_MONOTONIC, data->mode);
+ hrtimer_init(&data->timer, io_timeout_get_clock(data), data->mode);
if (is_timeout_link) {
struct io_submit_link *link = &req->ctx->submit_state.link;
struct io_ring_ctx *ctx = req->ctx;
int ret;
- WARN_ON_ONCE(req->task != current);
+ WARN_ON_ONCE(!io_wq_current_is_worker() && req->task != current);
ret = io_async_cancel_one(req->task->io_uring, sqe_addr, ctx);
if (ret != -ENOENT)
return -EINVAL;
if (unlikely(req->flags & (REQ_F_FIXED_FILE | REQ_F_BUFFER_SELECT)))
return -EINVAL;
- if (sqe->ioprio || sqe->off || sqe->len || sqe->cancel_flags)
+ if (sqe->ioprio || sqe->off || sqe->len || sqe->cancel_flags ||
+ sqe->splice_fd_in)
return -EINVAL;
req->cancel.addr = READ_ONCE(sqe->addr);
{
if (unlikely(req->flags & (REQ_F_FIXED_FILE | REQ_F_BUFFER_SELECT)))
return -EINVAL;
- if (sqe->ioprio || sqe->rw_flags)
+ if (sqe->ioprio || sqe->rw_flags || sqe->splice_fd_in)
return -EINVAL;
req->rsrc_update.offset = READ_ONCE(sqe->off);
int ret;
u32 seq;
+ if (req->flags & REQ_F_FAIL) {
+ io_req_complete_fail_submit(req);
+ return true;
+ }
+
/*
* If we need to drain a request in the middle of a link, drain the
* head request and the next request/link after the current link.
if (!req_need_defer(req, seq) && list_empty(&ctx->defer_list)) {
spin_unlock(&ctx->completion_lock);
kfree(de);
- io_queue_async_work(req);
+ io_queue_async_work(req, NULL);
return true;
}
if (timeout)
io_queue_linked_timeout(timeout);
+ /* either cancelled or io-wq is dying, so don't touch tctx->iowq */
if (work->flags & IO_WQ_WORK_CANCEL)
ret = -ECANCELED;
return io_file_get_normal(ctx, req, fd);
}
-static void io_req_task_link_timeout(struct io_kiocb *req)
+static void io_req_task_link_timeout(struct io_kiocb *req, bool *locked)
{
struct io_kiocb *prev = req->timeout.prev;
int ret;
if (!req_ref_inc_not_zero(prev))
prev = NULL;
}
+ list_del(&req->timeout.list);
req->timeout.prev = prev;
spin_unlock_irqrestore(&ctx->timeout_lock, flags);
data->timer.function = io_link_timeout_fn;
hrtimer_start(&data->timer, timespec64_to_ktime(data->ts),
data->mode);
+ list_add_tail(&req->timeout.list, &ctx->ltimeout_list);
}
spin_unlock_irq(&ctx->timeout_lock);
/* drop submission reference */
* Queued up for async execution, worker will release
* submit reference when the iocb is actually submitted.
*/
- io_queue_async_work(req);
+ io_queue_async_work(req, NULL);
break;
}
if (unlikely(req->ctx->drain_active) && io_drain_req(req))
return;
- if (likely(!(req->flags & REQ_F_FORCE_ASYNC))) {
+ if (likely(!(req->flags & (REQ_F_FORCE_ASYNC | REQ_F_FAIL)))) {
__io_queue_sqe(req);
+ } else if (req->flags & REQ_F_FAIL) {
+ io_req_complete_fail_submit(req);
} else {
int ret = io_req_prep_async(req);
if (unlikely(ret))
io_req_complete_failed(req, ret);
else
- io_queue_async_work(req);
+ io_queue_async_work(req, NULL);
}
}
ret = io_init_req(ctx, req, sqe);
if (unlikely(ret)) {
fail_req:
+ /* fail even hard links since we don't submit */
if (link->head) {
- /* fail even hard links since we don't submit */
- req_set_fail(link->head);
- io_req_complete_failed(link->head, -ECANCELED);
- link->head = NULL;
+ /*
+ * we can judge a link req is failed or cancelled by if
+ * REQ_F_FAIL is set, but the head is an exception since
+ * it may be set REQ_F_FAIL because of other req's failure
+ * so let's leverage req->result to distinguish if a head
+ * is set REQ_F_FAIL because of its failure or other req's
+ * failure so that we can set the correct ret code for it.
+ * init result here to avoid affecting the normal path.
+ */
+ if (!(link->head->flags & REQ_F_FAIL))
+ req_fail_link_node(link->head, -ECANCELED);
+ } else if (!(req->flags & (REQ_F_LINK | REQ_F_HARDLINK))) {
+ /*
+ * the current req is a normal req, we should return
+ * error and thus break the submittion loop.
+ */
+ io_req_complete_failed(req, ret);
+ return ret;
}
- io_req_complete_failed(req, ret);
- return ret;
+ req_fail_link_node(req, ret);
+ } else {
+ ret = io_req_prep(req, sqe);
+ if (unlikely(ret))
+ goto fail_req;
}
- ret = io_req_prep(req, sqe);
- if (unlikely(ret))
- goto fail_req;
-
/* don't need @sqe from now on */
trace_io_uring_submit_sqe(ctx, req, req->opcode, req->user_data,
req->flags, true,
if (link->head) {
struct io_kiocb *head = link->head;
- ret = io_req_prep_async(req);
- if (unlikely(ret))
- goto fail_req;
+ if (!(req->flags & REQ_F_FAIL)) {
+ ret = io_req_prep_async(req);
+ if (unlikely(ret)) {
+ req_fail_link_node(req, ret);
+ if (!(head->flags & REQ_F_FAIL))
+ req_fail_link_node(head, -ECANCELED);
+ }
+ }
trace_io_uring_link(ctx, req, head);
link->last->link = req;
link->last = req;
static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr)
__must_hold(&ctx->uring_lock)
{
- struct io_uring_task *tctx;
int submitted = 0;
/* make sure SQ entry isn't read before tail */
nr = min3(nr, ctx->sq_entries, io_sqring_entries(ctx));
if (!percpu_ref_tryget_many(&ctx->refs, nr))
return -EAGAIN;
+ io_get_task_refs(nr);
- tctx = current->io_uring;
- tctx->cached_refs -= nr;
- if (unlikely(tctx->cached_refs < 0)) {
- unsigned int refill = -tctx->cached_refs + IO_TCTX_REFS_CACHE_NR;
-
- percpu_counter_add(&tctx->inflight, refill);
- refcount_add(refill, ¤t->usage);
- tctx->cached_refs += refill;
- }
io_submit_state_start(&ctx->submit_state, nr);
-
while (submitted < nr) {
const struct io_uring_sqe *sqe;
struct io_kiocb *req;
}
sqe = io_get_sqe(ctx);
if (unlikely(!sqe)) {
- kmem_cache_free(req_cachep, req);
+ list_add(&req->inflight_entry, &ctx->submit_state.free_list);
break;
}
/* will complete beyond this point, count as submitted */
size_t init_size = size;
void **table;
- table = kcalloc(nr_tables, sizeof(*table), GFP_KERNEL);
+ table = kcalloc(nr_tables, sizeof(*table), GFP_KERNEL_ACCOUNT);
if (!table)
return NULL;
for (i = 0; i < nr_tables; i++) {
unsigned int this_size = min_t(size_t, size, PAGE_SIZE);
- table[i] = kzalloc(this_size, GFP_KERNEL);
+ table[i] = kzalloc(this_size, GFP_KERNEL_ACCOUNT);
if (!table[i]) {
io_free_page_table(table, init_size);
return NULL;
static bool io_alloc_file_tables(struct io_file_table *table, unsigned nr_files)
{
- table->files = kvcalloc(nr_files, sizeof(table->files[0]), GFP_KERNEL);
+ table->files = kvcalloc(nr_files, sizeof(table->files[0]),
+ GFP_KERNEL_ACCOUNT);
return !!table->files;
}
return -EINVAL;
if (nr_args > IORING_MAX_FIXED_FILES)
return -EMFILE;
+ if (nr_args > rlimit(RLIMIT_NOFILE))
+ return -EMFILE;
ret = io_rsrc_node_switch_start(ctx);
if (ret)
return ret;
#endif
}
+static int io_install_fixed_file(struct io_kiocb *req, struct file *file,
+ unsigned int issue_flags, u32 slot_index)
+{
+ struct io_ring_ctx *ctx = req->ctx;
+ bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK;
+ struct io_fixed_file *file_slot;
+ int ret = -EBADF;
+
+ io_ring_submit_lock(ctx, !force_nonblock);
+ if (file->f_op == &io_uring_fops)
+ goto err;
+ ret = -ENXIO;
+ if (!ctx->file_data)
+ goto err;
+ ret = -EINVAL;
+ if (slot_index >= ctx->nr_user_files)
+ goto err;
+
+ slot_index = array_index_nospec(slot_index, ctx->nr_user_files);
+ file_slot = io_fixed_file_slot(&ctx->file_table, slot_index);
+ ret = -EBADF;
+ if (file_slot->file_ptr)
+ goto err;
+
+ *io_get_tag_slot(ctx->file_data, slot_index) = 0;
+ io_fixed_file_set(file_slot, file);
+ ret = io_sqe_file_register(ctx, file, slot_index);
+ if (ret) {
+ file_slot->file_ptr = 0;
+ goto err;
+ }
+
+ ret = 0;
+err:
+ io_ring_submit_unlock(ctx, !force_nonblock);
+ if (ret)
+ fput(file);
+ return ret;
+}
+
static int io_queue_rsrc_removal(struct io_rsrc_data *data, unsigned idx,
struct io_rsrc_node *node, void *rsrc)
{
sock_release(ctx->ring_sock);
}
#endif
+ WARN_ON_ONCE(!list_empty(&ctx->ltimeout_list));
io_mem_free(ctx->rings);
io_mem_free(ctx->sq_sqes);
* Must be after io_uring_del_task_file() (removes nodes under
* uring_lock) to avoid race with io_uring_try_cancel_iowq().
*/
- tctx->io_wq = NULL;
io_wq_put_and_exit(wq);
+ tctx->io_wq = NULL;
}
}
return io_wq_cpu_affinity(tctx->io_wq, NULL);
}
+static int io_register_iowq_max_workers(struct io_ring_ctx *ctx,
+ void __user *arg)
+{
+ struct io_uring_task *tctx = NULL;
+ struct io_sq_data *sqd = NULL;
+ __u32 new_count[2];
+ int i, ret;
+
+ if (copy_from_user(new_count, arg, sizeof(new_count)))
+ return -EFAULT;
+ for (i = 0; i < ARRAY_SIZE(new_count); i++)
+ if (new_count[i] > INT_MAX)
+ return -EINVAL;
+
+ if (ctx->flags & IORING_SETUP_SQPOLL) {
+ sqd = ctx->sq_data;
+ if (sqd) {
+ mutex_lock(&sqd->lock);
+ tctx = sqd->thread->io_uring;
+ }
+ } else {
+ tctx = current->io_uring;
+ }
+
+ ret = -EINVAL;
+ if (!tctx || !tctx->io_wq)
+ goto err;
+
+ ret = io_wq_max_workers(tctx->io_wq, new_count);
+ if (ret)
+ goto err;
+
+ if (sqd)
+ mutex_unlock(&sqd->lock);
+
+ if (copy_to_user(arg, new_count, sizeof(new_count)))
+ return -EFAULT;
+
+ return 0;
+err:
+ if (sqd)
+ mutex_unlock(&sqd->lock);
+ return ret;
+}
+
static bool io_register_op_must_quiesce(int op)
{
switch (op) {
case IORING_REGISTER_BUFFERS_UPDATE:
case IORING_REGISTER_IOWQ_AFF:
case IORING_UNREGISTER_IOWQ_AFF:
+ case IORING_REGISTER_IOWQ_MAX_WORKERS:
return false;
default:
return true;
break;
ret = io_unregister_iowq_aff(ctx);
break;
+ case IORING_REGISTER_IOWQ_MAX_WORKERS:
+ ret = -EINVAL;
+ if (!arg || nr_args != 2)
+ break;
+ ret = io_register_iowq_max_workers(ctx, arg);
+ break;
default:
ret = -EINVAL;
break;
BUILD_BUG_SQE_ELEM(40, __u16, buf_group);
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_ON(sizeof(struct io_uring_files_update) !=
sizeof(struct io_uring_rsrc_update));
BUILD_BUG_ON(sizeof(struct io_uring_rsrc_update) >
sizeof(struct io_uring_rsrc_update2));
+
+ /* ->buf_index is u16 */
+ BUILD_BUG_ON(IORING_MAX_REG_BUFFERS >= (1u << 16));
+
/* should fit into one byte */
BUILD_BUG_ON(SQE_VALID_FLAGS >= (1 << 8));
projects / linux-2.6-microblaze.git / blobdiff