[Bug 293382] Dead lock and kernel crash around closefp_impl

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=293382

--- Comment #52 from Paul <devgs@ukr.net> ---
(In reply to Konstantin Belousov from comment #50)

Hi,

Thanks for the patch. Though, could't help but also include the assert implied
by Kyle. It should't hurt to have an extra check, I hope. This is the final
patch that we've applied to the current HEAD:


diff --git a/sys/kern/kern_event.c b/sys/kern/kern_event.c
index 01731ca46b6b..38b67d81dff2 100644
--- a/sys/kern/kern_event.c
+++ b/sys/kern/kern_event.c
@@ -2037,8 +2037,13 @@ kqueue_expand(struct kqueue *kq, const struct filterops
*fops, uintptr_t ident,
                                to_free = list;
                        } else {
                                if (kq->kq_knlist != NULL) {
-                                       bcopy(kq->kq_knlist, list,
-                                           kq->kq_knlistsize * sizeof(*list));
+                                       volatile char *src, *dst;
+                                       src = (void *)kq->kq_knlist;
+                                       dst = (void *)list;
+                                       for (size_t i = 0;
+                                           i < kq->kq_knlistsize *
sizeof(*list);
+                                           i++)
+                                               dst[i] = src[i];
                                        to_free = kq->kq_knlist;
                                        kq->kq_knlist = NULL;
                                }
@@ -2953,8 +2958,9 @@ knote_drop_detached(struct knote *kn, struct thread *td)
        else
                list = &kq->kq_knhash[KN_HASH(kn->kn_id, kq->kq_knhashmask)];

-       if (!SLIST_EMPTY(list))
-               SLIST_REMOVE(list, kn, knote, kn_link);
+       KASSERT(!SLIST_EMPTY(list),
+               ("knote %p is not on the list of %lu", kn, kn->kn_id));
+       SLIST_REMOVE(list, kn, knote, kn_link);
        if (kn->kn_status & KN_QUEUED)
                knote_dequeue(kn);
        KQ_UNLOCK_FLUX(kq);

-- 
You are receiving this mail because:
You are the assignee for the bug.