This commit fixes a use-after-free that causes the application to crash
on shutdown (detected by ASAN).
The vhost library uses a background event dispatch thread that monitors
fds with epoll. It runs in an infinite loop, waiting for I/O events
and calling callbacks when they occur.
During cleanup, a race condition existed:
Main Thread: Event Dispatch Thread:
1. Remove fds from fdset while (1) {
2. Close file descriptors epoll_wait() [gets interrupted]
3. rte_eal_cleanup() [continues loop]
4. Unmap hugepages Accesses fdset... CRASH
}
There was no explicit cleanup of the fdset structure.
The fdset structure is allocated with rte_zmalloc() and the memory would
only be reclaimed at application shutdown when rte_eal_cleanup() is called,
which invokes rte_eal_memory_detach() to unmap all the hugepage memory.
Meanwhile, the event dispatch thread could still be running and accessing
the fdset.
The code had a `destroy` flag that the event dispatch thread checked,
but it was never set during cleanup, and the code never waited for
the thread to actually exit before freeing memory.
To fix this, the commit implements fdset_destroy() that sets the destroy
flag with mutex protection, waits for thread termination, and cleans up
all resources including the fdset memory.
Update socket.c to call fdset_destroy() when the last vhost-user socket
is unregistered.
Fixes: 0e38b42bf61c ("vhost: manage FD with epoll")
Cc: [email protected]
Signed-off-by: Shani Peretz <[email protected]>
-----------------
v3:
removed vduse implementation from this fix
---
lib/vhost/fd_man.c | 45 ++++++++++++++++++++++++++++++++++++++++++++-
lib/vhost/fd_man.h | 1 +
lib/vhost/socket.c | 7 +++++++
3 files changed, 52 insertions(+), 1 deletion(-)
diff --git a/lib/vhost/fd_man.c b/lib/vhost/fd_man.c
index f9147edee7..b4597dec75 100644
--- a/lib/vhost/fd_man.c
+++ b/lib/vhost/fd_man.c
@@ -387,9 +387,52 @@ fdset_event_dispatch(void *arg)
}
}
- if (pfdset->destroy)
+ pthread_mutex_lock(&pfdset->fd_mutex);
+ bool should_destroy = pfdset->destroy;
+ pthread_mutex_unlock(&pfdset->fd_mutex);
+ if (should_destroy)
break;
}
return 0;
}
+
+/**
+ * Destroy the fdset and stop its event dispatch thread.
+ */
+void
+fdset_destroy(struct fdset *pfdset)
+{
+ uint32_t val;
+ int i;
+
+ if (pfdset == NULL)
+ return;
+
+ /* Signal the event dispatch thread to stop */
+ pthread_mutex_lock(&pfdset->fd_mutex);
+ pfdset->destroy = true;
+ pthread_mutex_unlock(&pfdset->fd_mutex);
+
+ /* Wait for the event dispatch thread to finish */
+ rte_thread_join(pfdset->tid, &val);
+
+ /* Close the epoll file descriptor */
+ close(pfdset->epfd);
+
+ /* Destroy the mutex */
+ pthread_mutex_destroy(&pfdset->fd_mutex);
+
+ /* Remove from global registry */
+ pthread_mutex_lock(&fdsets_mutex);
+ for (i = 0; i < MAX_FDSETS; i++) {
+ if (fdsets[i] == pfdset) {
+ fdsets[i] = NULL;
+ break;
+ }
+ }
+ pthread_mutex_unlock(&fdsets_mutex);
+
+ /* Free the fdset structure */
+ rte_free(pfdset);
+}
diff --git a/lib/vhost/fd_man.h b/lib/vhost/fd_man.h
index eadcc6fb42..ed2109f3c8 100644
--- a/lib/vhost/fd_man.h
+++ b/lib/vhost/fd_man.h
@@ -21,5 +21,6 @@ int fdset_add(struct fdset *pfdset, int fd,
void fdset_del(struct fdset *pfdset, int fd);
int fdset_try_del(struct fdset *pfdset, int fd);
+void fdset_destroy(struct fdset *pfdset);
#endif
diff --git a/lib/vhost/socket.c b/lib/vhost/socket.c
index ae95e7e6b0..eb01231478 100644
--- a/lib/vhost/socket.c
+++ b/lib/vhost/socket.c
@@ -1141,6 +1141,13 @@ rte_vhost_driver_unregister(const char *path)
count = --vhost_user.vsocket_cnt;
vhost_user.vsockets[i] = vhost_user.vsockets[count];
vhost_user.vsockets[count] = NULL;
+
+ /* Check if we need to destroy the vhost fdset */
+ if (vhost_user.vsocket_cnt == 0 && vhost_user.fdset != NULL) {
+ fdset_destroy(vhost_user.fdset);
+ vhost_user.fdset = NULL;
+ }
+
pthread_mutex_unlock(&vhost_user.mutex);
return 0;
}
--
2.43.0
