In sleep(3), if seconds is zero we don't call nanosleep(2).
I don't like this. If sleep(3) really is a simplified interface to
nanosleep(2) (as we claim in the manpage) we should let nanosleep(2)
handle the input and make decisions.
Other benefits:
- sleep(3) now *always* shows up in ktrace.
- sleep(3) with a zero input now blocks for up to 1 tick, just like
nanosleep(2) does with a zero input (more intuitive behavior).
- Neither NetBSD nor FreeBSD bypass nanosleep(2) like this, so now our
sleep(3) is more like theirs.
ok?
Index: sleep.c
===================================================================
RCS file: /cvs/src/lib/libc/gen/sleep.c,v
retrieving revision 1.12
diff -u -p -r1.12 sleep.c
--- sleep.c 14 Dec 2009 05:10:13 -0000 1.12
+++ sleep.c 6 Jan 2021 16:21:11 -0000
@@ -36,9 +36,6 @@ sleep(unsigned int seconds)
{
struct timespec rqt, rmt;
- if (seconds == 0)
- return(0);
-
rqt.tv_sec = seconds;
rqt.tv_nsec = 0;
