When protection of wired read-only mapping is changed to read-write,

install new shadow object behind the map entry and copy the pages
from the underlying objects to it. This makes the mprotect(2) call to
actually perform the requested operation instead of silently do nothing
and return success, that causes SIGSEGV on later write access to the
mapping.

Reuse vm_fault_copy_entry() to do the copying, modifying it to behave
correctly when src_entry == dst_entry.

Reviewed by:	alc
MFC after:	3 weeks

2 files changed, 56 insertions, 20 deletions

diff --git a/sys/vm/vm_fault.c b/sys/vm/vm_fault.c
index 508d617..68057fb 100644
--- a/sys/vm/vm_fault.c
+++ b/sys/vm/vm_fault.c
@@ -1119,7 +1119,10 @@ vm_fault_unwire(vm_map_t map, vm_offset_t start, vm_offset_t end,
  *	Routine:
  *		vm_fault_copy_entry
  *	Function:
- *		Copy all of the pages from a wired-down map entry to another.
+ *		Create new shadow object backing dst_entry with private copy of
+ *		all underlying pages. When src_entry is equal to dst_entry,
+ *		function implements COW for wired-down map entry. Otherwise,
+ *		it forks wired entry into dst_map.
  *
  *	In/out conditions:
  *		The source and destination maps must be locked for write.
@@ -1131,19 +1134,20 @@ vm_fault_copy_entry(vm_map_t dst_map, vm_map_t src_map,
     vm_map_entry_t dst_entry, vm_map_entry_t src_entry,
     vm_ooffset_t *fork_charge)
 {
-	vm_object_t backing_object, dst_object, object;
-	vm_object_t src_object;
+	vm_object_t backing_object, dst_object, object, src_object;
 	vm_pindex_t dst_pindex, pindex, src_pindex;
-	vm_prot_t prot;
+	vm_prot_t access, prot;
 	vm_offset_t vaddr;
 	vm_page_t dst_m;
 	vm_page_t src_m;
-	boolean_t src_readonly;
+	boolean_t src_readonly, upgrade;
 
 #ifdef	lint
 	src_map++;
 #endif	/* lint */
 
+	upgrade = src_entry == dst_entry;
+
 	src_object = src_entry->object.vm_object;
 	src_pindex = OFF_TO_IDX(src_entry->offset);
 	src_readonly = (src_entry->protection & VM_PROT_WRITE) == 0;
@@ -1160,17 +1164,34 @@ vm_fault_copy_entry(vm_map_t dst_map, vm_map_t src_map,
 #endif
 
 	VM_OBJECT_LOCK(dst_object);
-	KASSERT(dst_entry->object.vm_object == NULL,
+	KASSERT(upgrade || dst_entry->object.vm_object == NULL,
 	    ("vm_fault_copy_entry: vm_object not NULL"));
 	dst_entry->object.vm_object = dst_object;
 	dst_entry->offset = 0;
-	dst_object->uip = curthread->td_ucred->cr_ruidinfo;
-	uihold(dst_object->uip);
 	dst_object->charge = dst_entry->end - dst_entry->start;
-	KASSERT(dst_entry->uip == NULL,
-	    ("vm_fault_copy_entry: leaked swp charge"));
-	*fork_charge += dst_object->charge;
-	prot = dst_entry->max_protection;
+	if (fork_charge != NULL) {
+		KASSERT(dst_entry->uip == NULL,
+		    ("vm_fault_copy_entry: leaked swp charge"));
+		dst_object->uip = curthread->td_ucred->cr_ruidinfo;
+		uihold(dst_object->uip);
+		*fork_charge += dst_object->charge;
+	} else {
+		dst_object->uip = dst_entry->uip;
+		dst_entry->uip = NULL;
+	}
+	access = prot = dst_entry->max_protection;
+	/*
+	 * If not an upgrade, then enter the mappings in the pmap as
+	 * read and/or execute accesses.  Otherwise, enter them as
+	 * write accesses.
+	 *
+	 * A writeable large page mapping is only created if all of
+	 * the constituent small page mappings are modified. Marking
+	 * PTEs as modified on inception allows promotion to happen
+	 * without taking potentially large number of soft faults.
+	 */
+	if (!upgrade)
+		access &= ~VM_PROT_WRITE;
 
 	/*
 	 * Loop through all of the pages in the entry's range, copying each
@@ -1221,21 +1242,30 @@ vm_fault_copy_entry(vm_map_t dst_map, vm_map_t src_map,
 		VM_OBJECT_UNLOCK(dst_object);
 
 		/*
-		 * Enter it in the pmap as a read and/or execute access.
+		 * Enter it in the pmap. If a wired, copy-on-write
+		 * mapping is being replaced by a write-enabled
+		 * mapping, then wire that new mapping.
 		 */
-		pmap_enter(dst_map->pmap, vaddr, prot & ~VM_PROT_WRITE, dst_m,
-		    prot, FALSE);
+		pmap_enter(dst_map->pmap, vaddr, access, dst_m, prot, upgrade);
 
 		/*
 		 * Mark it no longer busy, and put it on the active list.
 		 */
 		VM_OBJECT_LOCK(dst_object);
 		vm_page_lock_queues();
-		vm_page_activate(dst_m);
+		if (upgrade) {
+			vm_page_unwire(src_m, 0);
+			vm_page_wire(dst_m);
+		} else
+			vm_page_activate(dst_m);
 		vm_page_unlock_queues();
 		vm_page_wakeup(dst_m);
 	}
 	VM_OBJECT_UNLOCK(dst_object);
+	if (upgrade) {
+		dst_entry->eflags &= ~(MAP_ENTRY_COW | MAP_ENTRY_NEEDS_COPY);
+		vm_object_deallocate(src_object);
+	}
 }
 
 
diff --git a/sys/vm/vm_map.c b/sys/vm/vm_map.c
index 1166712..06ae63e 100644
--- a/sys/vm/vm_map.c
+++ b/sys/vm/vm_map.c
@@ -1805,10 +1805,10 @@ int
 vm_map_protect(vm_map_t map, vm_offset_t start, vm_offset_t end,
 	       vm_prot_t new_prot, boolean_t set_max)
 {
-	vm_map_entry_t current;
-	vm_map_entry_t entry;
+	vm_map_entry_t current, entry;
 	vm_object_t obj;
 	struct uidinfo *uip;
+	vm_prot_t old_prot;
 
 	vm_map_lock(map);
 
@@ -1897,9 +1897,8 @@ vm_map_protect(vm_map_t map, vm_offset_t start, vm_offset_t end,
 	 */
 	current = entry;
 	while ((current != &map->header) && (current->start < end)) {
-		vm_prot_t old_prot;
-
 		old_prot = current->protection;
+
 		if (set_max)
 			current->protection =
 			    (current->max_protection = new_prot) &
@@ -1907,6 +1906,13 @@ vm_map_protect(vm_map_t map, vm_offset_t start, vm_offset_t end,
 		else
 			current->protection = new_prot;
 
+		if ((current->eflags & (MAP_ENTRY_COW | MAP_ENTRY_USER_WIRED))
+		     == (MAP_ENTRY_COW | MAP_ENTRY_USER_WIRED) &&
+		    (current->protection & VM_PROT_WRITE) != 0 &&
+		    (old_prot & VM_PROT_WRITE) == 0) {
+			vm_fault_copy_entry(map, map, current, current, NULL);
+		}
+
 		/*
 		 * Update physical map if necessary. Worry about copy-on-write
 		 * here.