linux: libzfs: simplify module-loaded check

The short-path is now one access() call,
we always modprobe zfs (ZFS_MODULE_LOADING which doesn't use the libzfs
boolean parsing is gone),
and we use a simple inotify IN_CREATE loop with a timerfd timeout
rather than 10ms kernel-style polling

There's one substantial difference: ZFS_MODULE_TIMEOUT=-1
now means "never give up", rather than "wait 10 minutes"

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz>
Closes #13330

man/man8/zfs.8

@@ -736,7 +736,7 @@ Do note that any changes done with the
 command will be undone if the share is ever unshared (like via a reboot).
 .
 .Sh ENVIRONMENT VARIABLES
-.Bl -tag -width "ZFS_MOUNT_HELPER"
+.Bl -tag -width "ZFS_MODULE_TIMEOUT"
 .It Sy ZFS_MOUNT_HELPER
 Cause
 .Nm zfs Cm mount
@@ -744,14 +744,28 @@ to use
 .Xr mount 8
 to mount ZFS datasets.
 This option is provided for backwards compatibility with older ZFS versions.
-.El
-.Bl -tag -width "ZFS_SET_PIPE_MAX"
+.
 .It Sy ZFS_SET_PIPE_MAX
 Tells
 .Nm zfs
 to set the maximum pipe size for sends/recieves.
 Disabled by default on Linux
 due to an unfixed deadlock in Linux's pipe size handling code.
+.
+.\" Shared with zpool.8
+.It Sy ZFS_MODULE_TIMEOUT
+Time, in seconds, to wait for
+.Pa /dev/zfs
+to appear.
+Defaults to
+.Sy 10 ,
+max
+.Sy 600 Pq 10 minutes .
+If
+.Pf < Sy 0 ,
+wait forever; if
+.Sy 0 ,
+don't wait.
 .El
 .
 .Sh INTERFACE STABILITY

man/man8/zpool.8

@@ -524,6 +524,20 @@ If
 .Sy ZPOOL_SCRIPTS_ENABLED
 is not set, it is assumed that the user is allowed to run
 .Nm zpool Cm status Ns / Ns Cm iostat Fl c .
+.\" Shared with zfs.8
+.It Sy ZFS_MODULE_TIMEOUT
+Time, in seconds, to wait for
+.Pa /dev/zfs
+to appear.
+Defaults to
+.Sy 10 ,
+max
+.Sy 600 Pq 10 minutes .
+If
+.Pf < Sy 0 ,
+wait forever; if
+.Sy 0 ,
+don't wait.
 .El
 .
 .Sh INTERFACE STABILITY