git: 12857d8f4269 - main - vm_object: Augment an assertion in vm_object_split()

The branch main has been updated by markj:

URL: https://cgit.FreeBSD.org/src/commit/?id=12857d8f4269af7df85605b48ae5e7b2fd810fde

commit 12857d8f4269af7df85605b48ae5e7b2fd810fde
Author:     Mark Johnston <[email protected]>
AuthorDate: 2026-08-11 14:10:23 +0000
Commit:     Mark Johnston <[email protected]>
CommitDate: 2026-08-11 14:10:23 +0000

    vm_object: Augment an assertion in vm_object_split()
    
    In some private discussion it was pointed out that vm_object_split()'s
    pattern of dropping the source object lock looks dangerous in that the
    initial assumption that OBJ_ONEMAPPING is set may become false.  In
    practice I believe that the map lock holds this flag stable, but let's
    assert that.
    
    Reviewed by:    alc, kib
    MFC after:      1 week
    Differential Revision:  https://reviews.freebsd.org/D58766
---
 sys/vm/vm_object.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/sys/vm/vm_object.c b/sys/vm/vm_object.c
index aa2d7676e6a8..1e6e25d26a54 100644
--- a/sys/vm/vm_object.c
+++ b/sys/vm/vm_object.c
@@ -1515,7 +1515,8 @@ vm_object_split(vm_map_entry_t entry)
 
 	orig_object = entry->object.vm_object;
 	KASSERT((orig_object->flags & OBJ_ONEMAPPING) != 0,
-	    ("vm_object_split:  Splitting object with multiple mappings."));
+	    ("%s: splitting object %p with multiple mappings",
+	    __func__, orig_object));
 	if ((orig_object->flags & OBJ_ANON) == 0)
 		return;
 	if (orig_object->ref_count <= 1)
@@ -1574,6 +1575,8 @@ vm_object_split(vm_map_entry_t entry)
 	vm_object_set_flag(orig_object, OBJ_SPLIT);
 	vm_page_iter_limit_init(&pages, orig_object, offidxstart + size);
 retry:
+	KASSERT((orig_object->flags & OBJ_ONEMAPPING) != 0,
+	    ("%s: object %p lost ONEMAPPING", __func__, orig_object));
 	KASSERT(pctrie_iter_is_reset(&pages),
 	    ("%s: pctrie_iter not reset for retry", __func__));
 	for (m = vm_radix_iter_lookup_ge(&pages, offidxstart); m != NULL;