Andi Kleen <ak@linux.intel.com> <ak@suse.de>
Andi Shyti <andi@etezian.org> <andi.shyti@samsung.com>
Andreas Herrmann <aherrman@de.ibm.com>
+Andrej Shadura <andrew.shadura@collabora.co.uk>
+Andrej Shadura <andrew@shadura.me> <andrew@beldisplaytech.com>
Andrew Morton <akpm@linux-foundation.org>
Andrew Murray <amurray@thegoodpenguin.co.uk> <amurray@embedded-bits.co.uk>
Andrew Murray <amurray@thegoodpenguin.co.uk> <andrew.murray@arm.com>
struct fd f = fdget(fd);
int ret = -EBADF;
- if (!f.file)
+ if (!f.file || !(f.file->f_mode & FMODE_READ))
goto out;
ret = kernel_read_file(f.file, offset, buf, buf_size, file_size, id);
int ocfs2_convert_inline_data_to_extents(struct inode *inode,
struct buffer_head *di_bh)
{
- int ret, i, has_data, num_pages = 0;
+ int ret, has_data, num_pages = 0;
int need_free = 0;
u32 bit_off, num;
handle_t *handle;
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
struct ocfs2_dinode *di = (struct ocfs2_dinode *)di_bh->b_data;
struct ocfs2_alloc_context *data_ac = NULL;
- struct page **pages = NULL;
- loff_t end = osb->s_clustersize;
+ struct page *page = NULL;
struct ocfs2_extent_tree et;
int did_quota = 0;
has_data = i_size_read(inode) ? 1 : 0;
if (has_data) {
- pages = kcalloc(ocfs2_pages_per_cluster(osb->sb),
- sizeof(struct page *), GFP_NOFS);
- if (pages == NULL) {
- ret = -ENOMEM;
- mlog_errno(ret);
- return ret;
- }
-
ret = ocfs2_reserve_clusters(osb, 1, &data_ac);
if (ret) {
mlog_errno(ret);
- goto free_pages;
+ goto out;
}
}
}
if (has_data) {
- unsigned int page_end;
+ unsigned int page_end = min_t(unsigned, PAGE_SIZE,
+ osb->s_clustersize);
u64 phys;
ret = dquot_alloc_space_nodirty(inode,
*/
block = phys = ocfs2_clusters_to_blocks(inode->i_sb, bit_off);
- /*
- * Non sparse file systems zero on extend, so no need
- * to do that now.
- */
- if (!ocfs2_sparse_alloc(osb) &&
- PAGE_SIZE < osb->s_clustersize)
- end = PAGE_SIZE;
-
- ret = ocfs2_grab_eof_pages(inode, 0, end, pages, &num_pages);
+ ret = ocfs2_grab_eof_pages(inode, 0, page_end, &page,
+ &num_pages);
if (ret) {
mlog_errno(ret);
need_free = 1;
* This should populate the 1st page for us and mark
* it up to date.
*/
- ret = ocfs2_read_inline_data(inode, pages[0], di_bh);
+ ret = ocfs2_read_inline_data(inode, page, di_bh);
if (ret) {
mlog_errno(ret);
need_free = 1;
goto out_unlock;
}
- page_end = PAGE_SIZE;
- if (PAGE_SIZE > osb->s_clustersize)
- page_end = osb->s_clustersize;
-
- for (i = 0; i < num_pages; i++)
- ocfs2_map_and_dirty_page(inode, handle, 0, page_end,
- pages[i], i > 0, &phys);
+ ocfs2_map_and_dirty_page(inode, handle, 0, page_end, page, 0,
+ &phys);
}
spin_lock(&oi->ip_lock);
}
out_unlock:
- if (pages)
- ocfs2_unlock_and_free_pages(pages, num_pages);
+ if (page)
+ ocfs2_unlock_and_free_pages(&page, num_pages);
out_commit:
if (ret < 0 && did_quota)
out:
if (data_ac)
ocfs2_free_alloc_context(data_ac);
-free_pages:
- kfree(pages);
return ret;
}
}
if (ocfs2_clusterinfo_valid(osb)) {
+ /*
+ * ci_stack and ci_cluster in ocfs2_cluster_info may not be null
+ * terminated, so make sure no overflow happens here by using
+ * memcpy. Destination strings will always be null terminated
+ * because osb is allocated using kzalloc.
+ */
osb->osb_stackflags =
OCFS2_RAW_SB(di)->s_cluster_info.ci_stackflags;
- strlcpy(osb->osb_cluster_stack,
+ memcpy(osb->osb_cluster_stack,
OCFS2_RAW_SB(di)->s_cluster_info.ci_stack,
- OCFS2_STACK_LABEL_LEN + 1);
+ OCFS2_STACK_LABEL_LEN);
if (strlen(osb->osb_cluster_stack) != OCFS2_STACK_LABEL_LEN) {
mlog(ML_ERROR,
"couldn't mount because of an invalid "
status = -EINVAL;
goto bail;
}
- strlcpy(osb->osb_cluster_name,
+ memcpy(osb->osb_cluster_name,
OCFS2_RAW_SB(di)->s_cluster_info.ci_cluster,
- OCFS2_CLUSTER_NAME_LEN + 1);
+ OCFS2_CLUSTER_NAME_LEN);
} else {
/* The empty string is identical with classic tools that
* don't know about s_cluster_info. */
if (mode_wp && mode_dontwake)
return -EINVAL;
- ret = mwriteprotect_range(ctx->mm, uffdio_wp.range.start,
- uffdio_wp.range.len, mode_wp,
- &ctx->mmap_changing);
+ if (mmget_not_zero(ctx->mm)) {
+ ret = mwriteprotect_range(ctx->mm, uffdio_wp.range.start,
+ uffdio_wp.range.len, mode_wp,
+ &ctx->mmap_changing);
+ mmput(ctx->mm);
+ } else {
+ return -ESRCH;
+ }
+
if (ret)
return ret;
CPUHP_SLUB_DEAD,
CPUHP_DEBUG_OBJ_DEAD,
CPUHP_MM_WRITEBACK_DEAD,
+ /* Must be after CPUHP_MM_VMSTAT_DEAD */
+ CPUHP_MM_DEMOTION_DEAD,
CPUHP_MM_VMSTAT_DEAD,
CPUHP_SOFTIRQ_DEAD,
CPUHP_NET_MVNETA_DEAD,
CPUHP_AP_BASE_CACHEINFO_ONLINE,
CPUHP_AP_ONLINE_DYN,
CPUHP_AP_ONLINE_DYN_END = CPUHP_AP_ONLINE_DYN + 30,
+ /* Must be after CPUHP_AP_ONLINE_DYN for node_states[N_CPU] update */
+ CPUHP_AP_MM_DEMOTION_ONLINE,
CPUHP_AP_X86_HPET_ONLINE,
CPUHP_AP_X86_KVM_CLK_ONLINE,
CPUHP_AP_DTPM_CPU_ONLINE,
#endif
}
-#if defined(CONFIG_UM) || defined(CONFIG_IA64)
+#if (defined(CONFIG_UML) && defined(CONFIG_X86_32)) || defined(CONFIG_IA64)
/*
* These functions parameterize elf_core_dump in fs/binfmt_elf.c to write out
* extra segments containing the gate DSO contents. Dumping its
#define register_hotmemory_notifier(nb) register_memory_notifier(nb)
#define unregister_hotmemory_notifier(nb) unregister_memory_notifier(nb)
#else
-#define hotplug_memory_notifier(fn, pri) ({ 0; })
+static inline int hotplug_memory_notifier(notifier_fn_t fn, int pri)
+{
+ return 0;
+}
/* These aren't inline functions due to a GCC bug. */
#define register_hotmemory_notifier(nb) ({ (void)(nb); 0; })
#define unregister_hotmemory_notifier(nb) ({ (void)(nb); })
mapping = (struct address_space *)
((unsigned long)page->mapping & ~PAGE_MAPPING_FLAGS);
- if (mapping != page->mapping)
+ if (!mapping || mapping != page->mapping)
return false;
return mapping->a_ops == &secretmem_aops;
if (mapping) {
int nr = thp_nr_pages(head);
- if (PageSwapBacked(head))
+ if (PageSwapBacked(head)) {
__mod_lruvec_page_state(head, NR_SHMEM_THPS,
-nr);
- else
+ } else {
__mod_lruvec_page_state(head, NR_FILE_THPS,
-nr);
+ filemap_nr_thps_dec(mapping);
+ }
}
__split_huge_page(page, list, end);
if (!size)
return;
- if (memblock.memory.cnt <= 1) {
+ if (!memblock_memory->total_size) {
pr_warn("%s: No memory registered yet\n", __func__);
return;
}
goto out;
}
- if (flags & MPOL_F_NUMA_BALANCING) {
- if (new && new->mode == MPOL_BIND) {
- new->flags |= (MPOL_F_MOF | MPOL_F_MORON);
- } else {
- ret = -EINVAL;
- mpol_put(new);
- goto out;
- }
- }
-
ret = mpol_set_nodemask(new, nodes, scratch);
if (ret) {
mpol_put(new);
return -EINVAL;
if ((*flags & MPOL_F_STATIC_NODES) && (*flags & MPOL_F_RELATIVE_NODES))
return -EINVAL;
-
+ if (*flags & MPOL_F_NUMA_BALANCING) {
+ if (*mode != MPOL_BIND)
+ return -EINVAL;
+ *flags |= (MPOL_F_MOF | MPOL_F_MORON);
+ }
return 0;
}
EXPORT_SYMBOL(migrate_vma_finalize);
#endif /* CONFIG_DEVICE_PRIVATE */
-#if defined(CONFIG_MEMORY_HOTPLUG)
+#if defined(CONFIG_HOTPLUG_CPU)
/* Disable reclaim-based migration. */
static void __disable_all_migrate_targets(void)
{
put_online_mems();
}
-/*
- * React to hotplug events that might affect the migration targets
- * like events that online or offline NUMA nodes.
- *
- * The ordering is also currently dependent on which nodes have
- * CPUs. That means we need CPU on/offline notification too.
- */
-static int migration_online_cpu(unsigned int cpu)
-{
- set_migration_target_nodes();
- return 0;
-}
-
-static int migration_offline_cpu(unsigned int cpu)
-{
- set_migration_target_nodes();
- return 0;
-}
-
/*
* This leaves migrate-on-reclaim transiently disabled between
* the MEM_GOING_OFFLINE and MEM_OFFLINE events. This runs
* set_migration_target_nodes().
*/
static int __meminit migrate_on_reclaim_callback(struct notifier_block *self,
- unsigned long action, void *arg)
+ unsigned long action, void *_arg)
{
+ struct memory_notify *arg = _arg;
+
+ /*
+ * Only update the node migration order when a node is
+ * changing status, like online->offline. This avoids
+ * the overhead of synchronize_rcu() in most cases.
+ */
+ if (arg->status_change_nid < 0)
+ return notifier_from_errno(0);
+
switch (action) {
case MEM_GOING_OFFLINE:
/*
return notifier_from_errno(0);
}
+/*
+ * React to hotplug events that might affect the migration targets
+ * like events that online or offline NUMA nodes.
+ *
+ * The ordering is also currently dependent on which nodes have
+ * CPUs. That means we need CPU on/offline notification too.
+ */
+static int migration_online_cpu(unsigned int cpu)
+{
+ set_migration_target_nodes();
+ return 0;
+}
+
+static int migration_offline_cpu(unsigned int cpu)
+{
+ set_migration_target_nodes();
+ return 0;
+}
+
static int __init migrate_on_reclaim_init(void)
{
int ret;
- ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "migrate on reclaim",
- migration_online_cpu,
- migration_offline_cpu);
+ ret = cpuhp_setup_state_nocalls(CPUHP_MM_DEMOTION_DEAD, "mm/demotion:offline",
+ NULL, migration_offline_cpu);
/*
* In the unlikely case that this fails, the automatic
* migration targets may become suboptimal for nodes
* rare case, do not bother trying to do anything special.
*/
WARN_ON(ret < 0);
+ ret = cpuhp_setup_state(CPUHP_AP_MM_DEMOTION_ONLINE, "mm/demotion:online",
+ migration_online_cpu, NULL);
+ WARN_ON(ret < 0);
hotplug_memory_notifier(migrate_on_reclaim_callback, 100);
return 0;
}
late_initcall(migrate_on_reclaim_init);
-#endif /* CONFIG_MEMORY_HOTPLUG */
+#endif /* CONFIG_HOTPLUG_CPU */
total_usage += table_size;
return 0;
}
-#ifdef CONFIG_MEMORY_HOTPLUG
+
static void free_page_ext(void *addr)
{
if (is_vmalloc_addr(addr)) {
return notifier_from_errno(ret);
}
-#endif
-
void __init page_ext_init(void)
{
unsigned long pfn;
return 0;
}
-#if defined(CONFIG_NUMA) && defined(CONFIG_MEMORY_HOTPLUG)
+#if defined(CONFIG_NUMA)
/*
* Drains freelist for a node on each slab cache, used for memory hot-remove.
* Returns -EBUSY if all objects cannot be drained so that the node is not
out:
return notifier_from_errno(ret);
}
-#endif /* CONFIG_NUMA && CONFIG_MEMORY_HOTPLUG */
+#endif /* CONFIG_NUMA */
/*
* swap the static kmem_cache_node with kmalloced memory
}
static inline bool slab_free_freelist_hook(struct kmem_cache *s,
- void **head, void **tail)
+ void **head, void **tail,
+ int *cnt)
{
void *object;
*head = object;
if (!*tail)
*tail = object;
+ } else {
+ /*
+ * Adjust the reconstructed freelist depth
+ * accordingly if object's reuse is delayed.
+ */
+ --(*cnt);
}
} while (object != old_tail);
struct kmem_cache_cpu *c;
unsigned long tid;
- memcg_slab_free_hook(s, &head, 1);
+ /* memcg_slab_free_hook() is already called for bulk free. */
+ if (!tail)
+ memcg_slab_free_hook(s, &head, 1);
redo:
/*
* Determine the currently cpus per cpu slab.
* With KASAN enabled slab_free_freelist_hook modifies the freelist
* to remove objects, whose reuse must be delayed.
*/
- if (slab_free_freelist_hook(s, &head, &tail))
+ if (slab_free_freelist_hook(s, &head, &tail, &cnt))
do_slab_free(s, page, head, tail, cnt, addr);
}
if (alloc_kmem_cache_cpus(s))
return 0;
- free_kmem_cache_nodes(s);
error:
+ __kmem_cache_release(s);
return -EINVAL;
}
return 0;
err = sysfs_slab_add(s);
- if (err)
+ if (err) {
__kmem_cache_release(s);
+ return err;
+ }
if (s->flags & SLAB_STORE_USER)
debugfs_slab_add(s);
- return err;
+ return 0;
}
void *__kmalloc_track_caller(size_t size, gfp_t gfpflags, unsigned long caller)
struct kmem_cache *s = file_inode(filep)->i_private;
unsigned long *obj_map;
+ if (!t)
+ return -ENOMEM;
+
obj_map = bitmap_alloc(oo_objects(s->oo), GFP_KERNEL);
- if (!obj_map)
+ if (!obj_map) {
+ seq_release_private(inode, filep);
return -ENOMEM;
+ }
if (strcmp(filep->f_path.dentry->d_name.name, "alloc_traces") == 0)
alloc = TRACK_ALLOC;
if (!alloc_loc_track(t, PAGE_SIZE / sizeof(struct location), GFP_KERNEL)) {
bitmap_free(obj_map);
+ seq_release_private(inode, filep);
return -ENOMEM;
}
uffd_test_ops->allocate_area((void **)&area_src);
uffd_test_ops->allocate_area((void **)&area_dst);
- uffd_test_ops->release_pages(area_src);
- uffd_test_ops->release_pages(area_dst);
-
userfaultfd_open(features);
count_verify = malloc(nr_pages * sizeof(unsigned long long));
*(area_count(area_src, nr) + 1) = 1;
}
+ /*
+ * After initialization of area_src, we must explicitly release pages
+ * for area_dst to make sure it's fully empty. Otherwise we could have
+ * some area_dst pages be errornously initialized with zero pages,
+ * hence we could hit memory corruption later in the test.
+ *
+ * One example is when THP is globally enabled, above allocate_area()
+ * calls could have the two areas merged into a single VMA (as they
+ * will have the same VMA flags so they're mergeable). When we
+ * initialize the area_src above, it's possible that some part of
+ * area_dst could have been faulted in via one huge THP that will be
+ * shared between area_src and area_dst. It could cause some of the
+ * area_dst won't be trapped by missing userfaults.
+ *
+ * This release_pages() will guarantee even if that happened, we'll
+ * proactively split the thp and drop any accidentally initialized
+ * pages within area_dst.
+ */
+ uffd_test_ops->release_pages(area_dst);
+
pipefd = malloc(sizeof(int) * nr_cpus * 2);
if (!pipefd)
err("pipefd");