zed: implement close_from() in terms of /proc/self/fd, if available

/dev/fd on Darwin

Consider the following strace output:
  prlimit64(0, RLIMIT_NOFILE, NULL, {rlim_cur=1024, rlim_max=1024*1024}) = 0

Yes, that is well over a million file descriptors!

This reduces the ZED start-up time from "at least a second" to
"instantaneous", and, under strace, from "don't even try" to "usable"
by simple virtue of doing five syscalls instead of over a million;
in most cases the main loop does nothing

Recent Linuxes (5.8+) have close_range(2) for this, but that's an
overoptimisation (and libcs don't have wrappers for it yet)

This is also run by the ZEDLET pre-exec. Compare:
  Finished "all-syslog.sh" eid=13 pid=6717 time=1.027100s exit=0
  Finished "history_event-zfs-list-cacher.sh" eid=13 pid=6718 time=1.046923s exit=0
to
  Finished "all-syslog.sh" eid=12 pid=4834 time=0.001836s exit=0
  Finished "history_event-zfs-list-cacher.sh" eid=12 pid=4835 time=0.001346s exit=0
lol

Reviewed-by: Brian Behlendorf <[email protected]>
Signed-off-by: Ahelenia Ziemiańska <[email protected]>
Closes #11834

1 files changed, 23 insertions, 9 deletions

diff --git a/cmd/zed/zed_file.c b/cmd/zed/zed_file.c
index 4437892e9..c7b0d00b6 100644
--- a/cmd/zed/zed_file.c
+++ b/cmd/zed/zed_file.c
@@ -12,6 +12,7 @@
  * You may not use this file except in compliance with the license.
  */
 
+#include <dirent.h>
 #include <errno.h>
 #include <fcntl.h>
 #include <limits.h>
@@ -160,6 +161,13 @@ zed_file_is_locked(int fd)
 	return (lock.l_pid);
 }
 
+
+#if __APPLE__
+#define	PROC_SELF_FD "/dev/fd"
+#else /* Linux-compatible layout */
+#define	PROC_SELF_FD "/proc/self/fd"
+#endif
+
 /*
  * Close all open file descriptors greater than or equal to [lowfd].
  * Any errors encountered while closing file descriptors are ignored.
@@ -167,17 +175,23 @@ zed_file_is_locked(int fd)
 void
 zed_file_close_from(int lowfd)
 {
-	const int maxfd_def = 256;
-	int errno_bak;
+	static const int maxfd_def = 256;
+	int errno_bak = errno;
 	struct rlimit rl;
-	int maxfd;
+	int maxfd = 0;
 	int fd;
-
-	errno_bak = errno;
-
-	if (getrlimit(RLIMIT_NOFILE, &rl) < 0) {
-		maxfd = maxfd_def;
-	} else if (rl.rlim_max == RLIM_INFINITY) {
+	DIR *fddir;
+	struct dirent *fdent;
+
+	if ((fddir = opendir(PROC_SELF_FD)) != NULL) {
+		while ((fdent = readdir(fddir)) != NULL) {
+			fd = atoi(fdent->d_name);
+			if (fd > maxfd && fd != dirfd(fddir))
+				maxfd = fd;
+		}
+		(void) closedir(fddir);
+	} else if (getrlimit(RLIMIT_NOFILE, &rl) < 0 ||
+	    rl.rlim_max == RLIM_INFINITY) {
 		maxfd = maxfd_def;
 	} else {
 		maxfd = rl.rlim_max;