[liburing,4/4] man: document clock id and IORING_ENTER_ABS_TIMER

Commit Message

Pavel Begunkov Aug. 18, 2024, 6:55 p.m. UTC

Signed-off-by: Pavel Begunkov <[e-mail]>
---
 man/io_uring_enter.2    | 12 ++++++++++++
 man/io_uring_register.2 | 20 ++++++++++++++++++++
 2 files changed, 32 insertions(+)

Patch

diff --git a/man/io_uring_enter.2 b/man/io_uring_enter.2
index 5e4121b..8c79771 100644
--- a/man/io_uring_enter.2
+++ b/man/io_uring_enter.2
@@ -105,6 +105,18 @@  then setting this flag will tell the kernel that the
 .I ring_fd
 passed in is the registered ring offset rather than a normal file descriptor.
 
+.TP
+.B IORING_ENTER_ABS_TIMER
+
+When this flag is set, the timeout argument passed in
+.I struct io_uring_getevents_arg
+will be interpreted as an absolute
+time of the registered clock (see
+.B IORING_REGISTER_CLOCK
+) until which the waiting should end.
+
+Available since 6.12
+
 .PP
 .PP
 If the io_uring instance was configured for polling, by specifying
diff --git a/man/io_uring_register.2 b/man/io_uring_register.2
index 4590588..c8521b7 100644
--- a/man/io_uring_register.2
+++ b/man/io_uring_register.2
@@ -720,6 +720,26 @@  The application must have registered a file table first.
 
 Available since 6.0.
 
+.TP
+.B IORING_REGISTER_CLOCK
+Specifies which clock id io_uring will use for timers while waiting for
+completion events with
+.B IORING_ENTER_GETEVENTS.
+It's only effective if the timeout argument in
+.I struct io_uring_getevents_arg
+is passed, ignored otherwise.
+When used in conjunction with
+.B IORING_ENTER_ABS_TIMER,
+interprets the timeout argument as absolute time of the specified clock.
+
+The default clock is
+.B CLOCK_MONOTONIC.
+
+Available since 6.12 and supports
+.B CLOCK_MONOTONIC
+and
+.B CLOCK_BOOTTIME.
+
 .SH RETURN VALUE
 
 On success,