1 files changed, 108 insertions, 30 deletions

diff --git a/io_uring/sqpoll.c b/io_uring/sqpoll.c
index 65b5dbe3c850..03c699493b5a 100644
--- a/io_uring/sqpoll.c
+++ b/io_uring/sqpoll.c
@@ -10,14 +10,17 @@
 #include <linux/slab.h>
 #include <linux/audit.h>
 #include <linux/security.h>
+#include <linux/cpuset.h>
 #include <linux/io_uring.h>
 
 #include <uapi/linux/io_uring.h>
 
 #include "io_uring.h"
+#include "napi.h"
 #include "sqpoll.h"
 
 #define IORING_SQPOLL_CAP_ENTRIES_VALUE 8
+#define IORING_TW_CAP_ENTRIES_VALUE	32
 
 enum {
 	IO_SQ_THREAD_SHOULD_STOP = 0,
@@ -37,12 +40,13 @@ void io_sq_thread_unpark(struct io_sq_data *sqd)
 	if (atomic_dec_return(&sqd->park_pending))
 		set_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state);
 	mutex_unlock(&sqd->lock);
+	wake_up(&sqd->wait);
 }
 
 void io_sq_thread_park(struct io_sq_data *sqd)
 	__acquires(&sqd->lock)
 {
-	WARN_ON_ONCE(sqd->thread == current);
+	WARN_ON_ONCE(data_race(sqd->thread) == current);
 
 	atomic_inc(&sqd->park_pending);
 	set_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state);
@@ -103,29 +107,21 @@ static struct io_sq_data *io_attach_sq_data(struct io_uring_params *p)
 {
 	struct io_ring_ctx *ctx_attach;
 	struct io_sq_data *sqd;
-	struct fd f;
+	CLASS(fd, f)(p->wq_fd);
 
-	f = fdget(p->wq_fd);
-	if (!f.file)
+	if (fd_empty(f))
 		return ERR_PTR(-ENXIO);
-	if (!io_is_uring_fops(f.file)) {
-		fdput(f);
+	if (!io_is_uring_fops(fd_file(f)))
 		return ERR_PTR(-EINVAL);
-	}
 
-	ctx_attach = f.file->private_data;
+	ctx_attach = fd_file(f)->private_data;
 	sqd = ctx_attach->sq_data;
-	if (!sqd) {
-		fdput(f);
+	if (!sqd)
 		return ERR_PTR(-EINVAL);
-	}
-	if (sqd->task_tgid != current->tgid) {
-		fdput(f);
+	if (sqd->task_tgid != current->tgid)
 		return ERR_PTR(-EPERM);
-	}
 
 	refcount_inc(&sqd->refs);
-	fdput(f);
 	return sqd;
 }
 
@@ -174,7 +170,7 @@ static int __io_sq_thread(struct io_ring_ctx *ctx, bool cap_entries)
 	if (cap_entries && to_submit > IORING_SQPOLL_CAP_ENTRIES_VALUE)
 		to_submit = IORING_SQPOLL_CAP_ENTRIES_VALUE;
 
-	if (!wq_list_empty(&ctx->iopoll_list) || to_submit) {
+	if (to_submit || !wq_list_empty(&ctx->iopoll_list)) {
 		const struct cred *creds = NULL;
 
 		if (ctx->sq_creds != current_cred())
@@ -212,21 +208,73 @@ static bool io_sqd_handle_event(struct io_sq_data *sqd)
 		mutex_unlock(&sqd->lock);
 		if (signal_pending(current))
 			did_sig = get_signal(&ksig);
-		cond_resched();
+		wait_event(sqd->wait, !atomic_read(&sqd->park_pending));
 		mutex_lock(&sqd->lock);
 		sqd->sq_cpu = raw_smp_processor_id();
 	}
 	return did_sig || test_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state);
 }
 
+/*
+ * Run task_work, processing the retry_list first. The retry_list holds
+ * entries that we passed on in the previous run, if we had more task_work
+ * than we were asked to process. Newly queued task_work isn't run until the
+ * retry list has been fully processed.
+ */
+static unsigned int io_sq_tw(struct llist_node **retry_list, int max_entries)
+{
+	struct io_uring_task *tctx = current->io_uring;
+	unsigned int count = 0;
+
+	if (*retry_list) {
+		*retry_list = io_handle_tw_list(*retry_list, &count, max_entries);
+		if (count >= max_entries)
+			goto out;
+		max_entries -= count;
+	}
+	*retry_list = tctx_task_work_run(tctx, max_entries, &count);
+out:
+	if (task_work_pending(current))
+		task_work_run();
+	return count;
+}
+
+static bool io_sq_tw_pending(struct llist_node *retry_list)
+{
+	struct io_uring_task *tctx = current->io_uring;
+
+	return retry_list || !llist_empty(&tctx->task_list);
+}
+
+static void io_sq_update_worktime(struct io_sq_data *sqd, struct rusage *start)
+{
+	struct rusage end;
+
+	getrusage(current, RUSAGE_SELF, &end);
+	end.ru_stime.tv_sec -= start->ru_stime.tv_sec;
+	end.ru_stime.tv_usec -= start->ru_stime.tv_usec;
+
+	sqd->work_time += end.ru_stime.tv_usec + end.ru_stime.tv_sec * 1000000;
+}
+
 static int io_sq_thread(void *data)
 {
+	struct llist_node *retry_list = NULL;
 	struct io_sq_data *sqd = data;
 	struct io_ring_ctx *ctx;
+	struct rusage start;
 	unsigned long timeout = 0;
-	char buf[TASK_COMM_LEN];
+	char buf[TASK_COMM_LEN] = {};
 	DEFINE_WAIT(wait);
 
+	/* offload context creation failed, just exit */
+	if (!current->io_uring) {
+		mutex_lock(&sqd->lock);
+		sqd->thread = NULL;
+		mutex_unlock(&sqd->lock);
+		goto err_out;
+	}
+
 	snprintf(buf, sizeof(buf), "iou-sqp-%d", sqd->task_pid);
 	set_task_comm(current, buf);
 
@@ -240,6 +288,14 @@ static int io_sq_thread(void *data)
 		sqd->sq_cpu = raw_smp_processor_id();
 	}
 
+	/*
+	 * Force audit context to get setup, in case we do prep side async
+	 * operations that would trigger an audit call before any issue side
+	 * audit has been done.
+	 */
+	audit_uring_entry(IORING_OP_NOP);
+	audit_uring_exit(true, 0);
+
 	mutex_lock(&sqd->lock);
 	while (1) {
 		bool cap_entries, sqt_spin = false;
@@ -251,18 +307,25 @@ static int io_sq_thread(void *data)
 		}
 
 		cap_entries = !list_is_singular(&sqd->ctx_list);
+		getrusage(current, RUSAGE_SELF, &start);
 		list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) {
 			int ret = __io_sq_thread(ctx, cap_entries);
 
 			if (!sqt_spin && (ret > 0 || !wq_list_empty(&ctx->iopoll_list)))
 				sqt_spin = true;
 		}
-		if (io_run_task_work())
+		if (io_sq_tw(&retry_list, IORING_TW_CAP_ENTRIES_VALUE))
 			sqt_spin = true;
 
+		list_for_each_entry(ctx, &sqd->ctx_list, sqd_list)
+			if (io_napi(ctx))
+				io_napi_sqpoll_busy_poll(ctx);
+
 		if (sqt_spin || !time_after(jiffies, timeout)) {
-			if (sqt_spin)
+			if (sqt_spin) {
+				io_sq_update_worktime(sqd, &start);
 				timeout = jiffies + sqd->sq_thread_idle;
+			}
 			if (unlikely(need_resched())) {
 				mutex_unlock(&sqd->lock);
 				cond_resched();
@@ -273,7 +336,7 @@ static int io_sq_thread(void *data)
 		}
 
 		prepare_to_wait(&sqd->wait, &wait, TASK_INTERRUPTIBLE);
-		if (!io_sqd_events_pending(sqd) && !task_work_pending(current)) {
+		if (!io_sqd_events_pending(sqd) && !io_sq_tw_pending(retry_list)) {
 			bool needs_sched = true;
 
 			list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) {
@@ -312,13 +375,16 @@ static int io_sq_thread(void *data)
 		timeout = jiffies + sqd->sq_thread_idle;
 	}
 
+	if (retry_list)
+		io_sq_tw(&retry_list, UINT_MAX);
+
 	io_uring_cancel_generic(true, sqd);
 	sqd->thread = NULL;
 	list_for_each_entry(ctx, &sqd->ctx_list, sqd_list)
 		atomic_or(IORING_SQ_NEED_WAKEUP, &ctx->rings->sq_flags);
 	io_run_task_work();
 	mutex_unlock(&sqd->lock);
-
+err_out:
 	complete(&sqd->exited);
 	do_exit(0);
 }
@@ -343,21 +409,17 @@ void io_sqpoll_wait_sq(struct io_ring_ctx *ctx)
 __cold int io_sq_offload_create(struct io_ring_ctx *ctx,
 				struct io_uring_params *p)
 {
+	struct task_struct *task_to_put = NULL;
 	int ret;
 
 	/* Retain compatibility with failing for an invalid attach attempt */
 	if ((ctx->flags & (IORING_SETUP_ATTACH_WQ | IORING_SETUP_SQPOLL)) ==
 				IORING_SETUP_ATTACH_WQ) {
-		struct fd f;
-
-		f = fdget(p->wq_fd);
-		if (!f.file)
+		CLASS(fd, f)(p->wq_fd);
+		if (fd_empty(f))
 			return -ENXIO;
-		if (!io_is_uring_fops(f.file)) {
-			fdput(f);
+		if (!io_is_uring_fops(fd_file(f)))
 			return -EINVAL;
-		}
-		fdput(f);
 	}
 	if (ctx->flags & IORING_SETUP_SQPOLL) {
 		struct task_struct *tsk;
@@ -393,11 +455,22 @@ __cold int io_sq_offload_create(struct io_ring_ctx *ctx,
 			return 0;
 
 		if (p->flags & IORING_SETUP_SQ_AFF) {
+			cpumask_var_t allowed_mask;
 			int cpu = p->sq_thread_cpu;
 
 			ret = -EINVAL;
 			if (cpu >= nr_cpu_ids || !cpu_online(cpu))
 				goto err_sqpoll;
+			ret = -ENOMEM;
+			if (!alloc_cpumask_var(&allowed_mask, GFP_KERNEL))
+				goto err_sqpoll;
+			ret = -EINVAL;
+			cpuset_cpus_allowed(current, allowed_mask);
+			if (!cpumask_test_cpu(cpu, allowed_mask)) {
+				free_cpumask_var(allowed_mask);
+				goto err_sqpoll;
+			}
+			free_cpumask_var(allowed_mask);
 			sqd->sq_cpu = cpu;
 		} else {
 			sqd->sq_cpu = -1;
@@ -412,6 +485,7 @@ __cold int io_sq_offload_create(struct io_ring_ctx *ctx,
 		}
 
 		sqd->thread = tsk;
+		task_to_put = get_task_struct(tsk);
 		ret = io_uring_alloc_task_context(tsk, ctx);
 		wake_up_new_task(tsk);
 		if (ret)
@@ -422,11 +496,15 @@ __cold int io_sq_offload_create(struct io_ring_ctx *ctx,
 		goto err;
 	}
 
+	if (task_to_put)
+		put_task_struct(task_to_put);
 	return 0;
 err_sqpoll:
 	complete(&ctx->sq_data->exited);
 err:
 	io_sq_thread_finish(ctx);
+	if (task_to_put)
+		put_task_struct(task_to_put);
 	return ret;
 }