[PATCH 4/4] io_uring: wire up min batch wake timeout

[Jens Axboe <axboe@kernel.dk>]

Expose min_wait_usec in io_uring_getevents_arg, replacing the pad member
that is currently in there. The value is in usecs, which is explained in
the name as well.

Note that if min_wait_usec and a normal timeout is used in conjunction,
the normal timeout is still relative to the base time. For example, if
min_wait_usec is set to 100 and the normal timeout is 1000, the max
total time waited is still 1000. This also means that if the normal
timeout is shorter than min_wait_usec, then only the min_wait_usec will
take effect.

See previous commit for an explanation of how this works.

IORING_FEAT_MIN_TIMEOUT is added as a feature flag for this, as
applications doing submit_and_wait_timeout() style operations will
generally not see the -EINVAL from the wait side as they return the
number of IOs submitted. Only if no IOs are submitted will the -EINVAL
bubble back up to the application.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 include/uapi/linux/io_uring.h |  3 ++-
 io_uring/io_uring.c           | 21 +++++++++++++--------
 2 files changed, 15 insertions(+), 9 deletions(-)

diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h
index 7bd10201a02b..dbefda14d087 100644
--- a/include/uapi/linux/io_uring.h
+++ b/include/uapi/linux/io_uring.h
@@ -522,6 +522,7 @@ struct io_uring_params {
 #define IORING_FEAT_CQE_SKIP		(1U << 11)
 #define IORING_FEAT_LINKED_FILE		(1U << 12)
 #define IORING_FEAT_REG_REG_RING	(1U << 13)
+#define IORING_FEAT_MIN_TIMEOUT		(1U << 14)
 
 /*
  * io_uring_register(2) opcodes and arguments
@@ -738,7 +739,7 @@ enum {
 struct io_uring_getevents_arg {
 	__u64	sigmask;
 	__u32	sigmask_sz;
-	__u32	pad;
+	__u32	min_wait_usec;
 	__u64	ts;
 };
 
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index e72261f280a7..8dd5eb647b43 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -2647,7 +2647,8 @@ static inline int io_cqring_wait_schedule(struct io_ring_ctx *ctx,
  */
 static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events,
 			  const sigset_t __user *sig, size_t sigsz,
-			  struct __kernel_timespec __user *uts)
+			  struct __kernel_timespec __user *uts,
+			  ktime_t min_time)
 {
 	struct io_wait_queue iowq;
 	struct io_rings *rings = ctx->rings;
@@ -2684,7 +2685,7 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events,
 	iowq.nr_timeouts = atomic_read(&ctx->cq_timeouts);
 	iowq.cq_min_tail = READ_ONCE(ctx->rings->cq.tail);
 	iowq.cq_tail = READ_ONCE(ctx->rings->cq.head) + min_events;
-	iowq.min_timeout = KTIME_MAX;
+	iowq.min_timeout = min_time;
 	iowq.timeout = KTIME_MAX;
 	start_time = ktime_get_ns();
 
@@ -3634,10 +3635,12 @@ static int io_validate_ext_arg(unsigned flags, const void __user *argp, size_t a
 
 static int io_get_ext_arg(unsigned flags, const void __user *argp, size_t *argsz,
 			  struct __kernel_timespec __user **ts,
-			  const sigset_t __user **sig)
+			  const sigset_t __user **sig, ktime_t *min_time)
 {
 	struct io_uring_getevents_arg arg;
 
+	*min_time = KTIME_MAX;
+
 	/*
 	 * If EXT_ARG isn't set, then we have no timespec and the argp pointer
 	 * is just a pointer to the sigset_t.
@@ -3656,8 +3659,8 @@ static int io_get_ext_arg(unsigned flags, const void __user *argp, size_t *argsz
 		return -EINVAL;
 	if (copy_from_user(&arg, argp, sizeof(arg)))
 		return -EFAULT;
-	if (arg.pad)
-		return -EINVAL;
+	if (arg.min_wait_usec)
+		*min_time = arg.min_wait_usec * NSEC_PER_USEC;
 	*sig = u64_to_user_ptr(arg.sigmask);
 	*argsz = arg.sigmask_sz;
 	*ts = u64_to_user_ptr(arg.ts);
@@ -3769,13 +3772,14 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit,
 		} else {
 			const sigset_t __user *sig;
 			struct __kernel_timespec __user *ts;
+			ktime_t min_time;
 
-			ret2 = io_get_ext_arg(flags, argp, &argsz, &ts, &sig);
+			ret2 = io_get_ext_arg(flags, argp, &argsz, &ts, &sig, &min_time);
 			if (likely(!ret2)) {
 				min_complete = min(min_complete,
 						   ctx->cq_entries);
 				ret2 = io_cqring_wait(ctx, min_complete, sig,
-						      argsz, ts);
+						      argsz, ts, min_time);
 			}
 		}
 
@@ -4058,7 +4062,8 @@ static __cold int io_uring_create(unsigned entries, struct io_uring_params *p,
 			IORING_FEAT_POLL_32BITS | IORING_FEAT_SQPOLL_NONFIXED |
 			IORING_FEAT_EXT_ARG | IORING_FEAT_NATIVE_WORKERS |
 			IORING_FEAT_RSRC_TAGS | IORING_FEAT_CQE_SKIP |
-			IORING_FEAT_LINKED_FILE | IORING_FEAT_REG_REG_RING;
+			IORING_FEAT_LINKED_FILE | IORING_FEAT_REG_REG_RING |
+			IORING_FEAT_MIN_TIMEOUT;
 
 	if (copy_to_user(params, p, sizeof(*p))) {
 		ret = -EFAULT;
-- 
2.43.0