Files
Gregory Price (Meta) e55119ab61 mm/mmzone: read each zoneref slot once when filtering by nodemask
__next_zones_zonelist() tests a slot's zone for NULL, then reads the
slot again to look up the zone's node. Offlining a zone's last block
rebuilds the zonelists in place, so the second read can return the
terminator:

  CPU 0                             CPU 1
  __next_zones_zonelist(nodemask)
    zonelist_zone(slot 1) = NORMAL
                                    offline_pages()
                                      build_all_zonelists()
                                        [MOVABLE, NORMAL, NULL]
                                        -> [NORMAL, NULL]
    zref_in_nodemask(slot 1)
      zone_to_nid(NULL)

Current GCC and clang merge the two reads, but nothing requires it.
With mm/mmzone.o built with -fno-inline -fno-ipa-sra, holding a
nodemask allocation between the reads while the node's last movable
block goes offline oopses every time:

  BUG: kernel NULL pointer dereference, address: 0000000000000060
  RIP: 0010:zone_to_nid+0x0/0x10
  Call Trace:
   zref_in_nodemask+0x9/0x20
   __next_zones_zonelist+0x23/0x70
   prepare_alloc_pages.constprop.0+0x16c/0x190
   __alloc_frozen_pages_noprof+0xaf/0x330

Read the slot once and test that zone. A NULL zone still ends the walk.
To achieve that, rename zref_in_nodemask() to zone_in_nodemask() and
pass a zone instead of zoneref pointer.

Fixes: 11cd8638c3 ("mm, page_alloc: remove stop_machine from build_all_zonelists")
Cc: [email protected]
Signed-off-by: Gregory Price (Meta) <[email protected]>
Reviewed-by: Vlastimil Babka (SUSE) <[email protected]>
Link: https://patch.msgid.link/[email protected]
Reviewed-by: Zi Yan <[email protected]>
Signed-off-by: Vlastimil Babka (SUSE) <[email protected]>
2026-11-05 10:58:52 +01:00

117 lines
2.7 KiB
C

// SPDX-License-Identifier: GPL-2.0
/*
* linux/mm/mmzone.c
*
* management codes for pgdats, zones and page flags
*/
#include <linux/stddef.h>
#include <linux/mm.h>
#include <linux/mmzone.h>
struct pglist_data *first_online_pgdat(void)
{
return NODE_DATA(first_online_node);
}
struct pglist_data *next_online_pgdat(struct pglist_data *pgdat)
{
int nid = next_online_node(pgdat->node_id);
if (nid == MAX_NUMNODES)
return NULL;
return NODE_DATA(nid);
}
/*
* next_zone - helper magic for for_each_zone()
*/
struct zone *next_zone(struct zone *zone)
{
pg_data_t *pgdat = zone->zone_pgdat;
if (zone < pgdat->node_zones + MAX_NR_ZONES - 1)
zone++;
else {
pgdat = next_online_pgdat(pgdat);
if (pgdat)
zone = pgdat->node_zones;
else
zone = NULL;
}
return zone;
}
/* A NULL zone terminates the zonelist, so it ends the search. */
static inline bool zone_in_nodemask(const struct zone *zone,
const nodemask_t *nodes)
{
#ifdef CONFIG_NUMA
return !zone || node_isset(zone_to_nid(zone), *nodes);
#else
return true;
#endif /* CONFIG_NUMA */
}
/* Returns the next zone at or below highest_zoneidx in a zonelist */
struct zoneref *__next_zones_zonelist(struct zoneref *z,
enum zone_type highest_zoneidx,
const nodemask_t *nodes)
{
/*
* Find the next suitable zone to use for the allocation.
* Only filter based on nodemask if it's set
*/
if (unlikely(nodes == NULL))
while (zonelist_zone_idx(z) > highest_zoneidx)
z++;
else
while (zonelist_zone_idx(z) > highest_zoneidx ||
!zone_in_nodemask(zonelist_zone(z), nodes))
z++;
return z;
}
void lruvec_init(struct lruvec *lruvec)
{
enum lru_list lru;
memset(lruvec, 0, sizeof(struct lruvec));
spin_lock_init(&lruvec->lru_lock);
spin_lock_init(&lruvec->cost_lock);
zswap_lruvec_state_init(lruvec);
for_each_lru(lru)
INIT_LIST_HEAD(&lruvec->lists[lru]);
/*
* The "Unevictable LRU" is imaginary: though its size is maintained,
* it is never scanned, and unevictable pages are not threaded on it
* (so that their lru fields can be reused to hold mlock_count).
* Poison its list head, so that any operations on it would crash.
*/
list_del(&lruvec->lists[LRU_UNEVICTABLE]);
lru_gen_init_lruvec(lruvec);
}
#if defined(CONFIG_NUMA_BALANCING) && !defined(LAST_CPUPID_NOT_IN_PAGE_FLAGS)
int folio_xchg_last_cpupid(struct folio *folio, int cpupid)
{
unsigned long old_flags, flags;
int last_cpupid;
old_flags = READ_ONCE(folio->flags.f);
do {
flags = old_flags;
last_cpupid = (flags >> LAST_CPUPID_PGSHIFT) & LAST_CPUPID_MASK;
flags &= ~(LAST_CPUPID_MASK << LAST_CPUPID_PGSHIFT);
flags |= (cpupid & LAST_CPUPID_MASK) << LAST_CPUPID_PGSHIFT;
} while (unlikely(!try_cmpxchg(&folio->flags.f, &old_flags, flags)));
return last_cpupid;
}
#endif