tail: avoid a race where we could miss new data with --pid

* src/tail.c (tail_forever, tail_forever_inotify): Close a race in
tail_forever_inotify where new data written after the file check by
a now dead process, but before the pid check, is not output.  We use
the POSIX guarantee that read() and write() are serialized wrt each
other even in separate processes, to assume full file consistency
after exit() and so poll for new data _after_ the writer has exited.
This also allows us to not redundantly _wait_ for new data if the
process is dead.
* tests/tail-2/pid: Remove the now partially invalid sub second sleep
check as we now don't unconditionally wait, and replace it with a check
for the redundant sleep.  Also clarify some of the existing comments.
* NEWS: Mention the fix.

1 files changed, 8 insertions, 7 deletions

diff --git a/tests/tail-2/pid b/tests/tail-2/pid
index 90f168430..760e289bc 100755
--- a/tests/tail-2/pid
+++ b/tests/tail-2/pid
@@ -29,26 +29,27 @@ touch here || framework_failure
 fail=0
 
 for inotify in ---disable-inotify ''; do
-  # Use tail itself to create a background process to monitor.
+  # Use tail itself to create a background process to monitor,
+  # which will auto exit when "here" is removed.
   tail -f $inotify here &
   bg_pid=$!
 
   # Ensure that tail --pid=PID does not exit when PID is alive.
-  timeout 1 tail -s.1 -f $inotify here --pid=$bg_pid
+  timeout 1 tail -f -s.1 --pid=$bg_pid $inotify here
   test $? = 124 || fail=1
 
   # Cleanup background process
   kill $bg_pid
 
-  # Ensure that tail --pid=PID exits successfully when PID is dead.
+  # Ensure that tail --pid=PID exits with success status when PID is dead.
   # Use an unlikely-to-be-live PID
-  timeout 3 tail -s.1 --pid=$PID_T_MAX -f $inotify /dev/null
+  timeout 3 tail -f -s.1 --pid=$PID_T_MAX $inotify /dev/null
   ret=$?
-  test $ret = 124 && skip_test_ "pid $PID_T_MAX present"
+  test $ret = 124 && skip_test_ "pid $PID_T_MAX present or tail too slow"
   test $ret = 0 || fail=1
 
-  # Ensure fractional sleep parameter is honored with --pid
-  timeout 3 tail -s.1 -f $inotify /dev/null --pid=$PID_T_MAX
+  # Ensure tail doesn't wait for data when PID is dead
+  timeout 3 tail -f -s10 --pid=$PID_T_MAX $inotify /dev/null
   test $? = 124 && fail=1
 done