Closing the handle in one thread while another thread is locked causes
undefined behavior (as closing does not obtain a lock, and can cause
use-after-free or other nasty problems to the thread that does own the
lock). However, it is not sensible to try and obtain a lock in
nbd_close, as POSIX says that it is also undefined for any other
thread to wait on a mutex that has already been destroyed. Therefore,
we don't need to change our code, but merely remind users that
nbd_close is not safe until all other uses of the handle have ceased.
---
generator/generator | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/generator/generator b/generator/generator
index 896ad2a..bdd8fd7 100755
--- a/generator/generator
+++ b/generator/generator
@@ -3532,7 +3532,8 @@ for how to get further details of the error.
Closes the handle and frees any associated resources. The final
status of any command that has not been retired (whether by
C<nbd_aio_command_completed> or by a low-level completion callback
-returning C<1>) is lost.
+returning C<1>) is lost. This function is not safe to call while
+any other thread is still using any C<nbd_*> API on the same handle.
=head1 GETTING THE LATEST ERROR MESSAGE IN THE THREAD
--
2.20.1