Commit 96e2db45 authored by Uladzislau Rezki (Sony)'s avatar Uladzislau Rezki (Sony) Committed by Linus Torvalds

mm/vmalloc: rework the drain logic

A current "lazy drain" model suffers from at least two issues.

First one is related to the unsorted list of vmap areas, thus in order to
identify the [min:max] range of areas to be drained, it requires a full
list scan.  What is a time consuming if the list is too long.

Second one and as a next step is about merging all fragments with a free
space.  What is also a time consuming because it has to iterate over
entire list which holds outstanding lazy areas.

See below the "preemptirqsoff" tracer that illustrates a high latency.  It
is ~24676us.  Our workloads like audio and video are effected by such long
latency:

<snip>
  tracer: preemptirqsoff

  preemptirqsoff latency trace v1.1.5 on 4.9.186-perf+
  --------------------------------------------------------------------
  latency: 24676 us, #4/4, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 P:8)
     -----------------
     | task: crtc_commit:112-261 (uid:0 nice:0 policy:1 rt_prio:16)
     -----------------
   => started at: __purge_vmap_area_lazy
   => ended at:   __purge_vmap_area_lazy

                   _------=> CPU#
                  / _-----=> irqs-off
                 | / _----=> need-resched
                 || / _---=> hardirq/softirq
                 ||| / _--=> preempt-depth
                 |||| /     delay
   cmd     pid   ||||| time  |   caller
      \   /      |||||  \    |   /
crtc_com-261     1...1    1us*: _raw_spin_lock <-__purge_vmap_area_lazy
[...]
crtc_com-261     1...1 24675us : _raw_spin_unlock <-__purge_vmap_area_lazy
crtc_com-261     1...1 24677us : trace_preempt_on <-__purge_vmap_area_lazy
crtc_com-261     1...1 24683us : <stack trace>
 => free_vmap_area_noflush
 => remove_vm_area
 => __vunmap
 => vfree
 => drm_property_free_blob
 => drm_mode_object_unreference
 => drm_property_unreference_blob
 => __drm_atomic_helper_crtc_destroy_state
 => sde_crtc_destroy_state
 => drm_atomic_state_default_clear
 => drm_atomic_state_clear
 => drm_atomic_state_free
 => complete_commit
 => _msm_drm_commit_work_cb
 => kthread_worker_fn
 => kthread
 => ret_from_fork
<snip>

To address those two issues we can redesign a purging of the outstanding
lazy areas.  Instead of queuing vmap areas to the list, we replace it by
the separate rb-tree.  In hat case an area is located in the tree/list in
ascending order.  It will give us below advantages:

a) Outstanding vmap areas are merged creating bigger coalesced blocks,
   thus it becomes less fragmented.

b) It is possible to calculate a flush range [min:max] without scanning
   all elements.  It is O(1) access time or complexity;

c) The final merge of areas with the rb-tree that represents a free
   space is faster because of (a).  As a result the lock contention is
   also reduced.

Link: https://lkml.kernel.org/r/20201116220033.1837-2-urezki@gmail.comSigned-off-by: default avatarUladzislau Rezki (Sony) <urezki@gmail.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Oleksiy Avramchenko <oleksiy.avramchenko@sonymobile.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: huang ying <huang.ying.caritas@gmail.com>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent 8945a723
...@@ -72,16 +72,14 @@ struct vmap_area { ...@@ -72,16 +72,14 @@ struct vmap_area {
struct list_head list; /* address sorted list */ struct list_head list; /* address sorted list */
/* /*
* The following three variables can be packed, because * The following two variables can be packed, because
* a vmap_area object is always one of the three states: * a vmap_area object can be either:
* 1) in "free" tree (root is vmap_area_root) * 1) in "free" tree (root is vmap_area_root)
* 2) in "busy" tree (root is free_vmap_area_root) * 2) or "busy" tree (root is free_vmap_area_root)
* 3) in purge list (head is vmap_purge_list)
*/ */
union { union {
unsigned long subtree_max_size; /* in "free" tree */ unsigned long subtree_max_size; /* in "free" tree */
struct vm_struct *vm; /* in "busy" tree */ struct vm_struct *vm; /* in "busy" tree */
struct llist_node purge_list; /* in purge list */
}; };
}; };
......
...@@ -413,10 +413,13 @@ static DEFINE_SPINLOCK(vmap_area_lock); ...@@ -413,10 +413,13 @@ static DEFINE_SPINLOCK(vmap_area_lock);
static DEFINE_SPINLOCK(free_vmap_area_lock); static DEFINE_SPINLOCK(free_vmap_area_lock);
/* Export for kexec only */ /* Export for kexec only */
LIST_HEAD(vmap_area_list); LIST_HEAD(vmap_area_list);
static LLIST_HEAD(vmap_purge_list);
static struct rb_root vmap_area_root = RB_ROOT; static struct rb_root vmap_area_root = RB_ROOT;
static bool vmap_initialized __read_mostly; static bool vmap_initialized __read_mostly;
static struct rb_root purge_vmap_area_root = RB_ROOT;
static LIST_HEAD(purge_vmap_area_list);
static DEFINE_SPINLOCK(purge_vmap_area_lock);
/* /*
* This kmem_cache is used for vmap_area objects. Instead of * This kmem_cache is used for vmap_area objects. Instead of
* allocating from slab we reuse an object from this cache to * allocating from slab we reuse an object from this cache to
...@@ -820,10 +823,17 @@ merge_or_add_vmap_area(struct vmap_area *va, ...@@ -820,10 +823,17 @@ merge_or_add_vmap_area(struct vmap_area *va,
if (!merged) if (!merged)
link_va(va, root, parent, link, head); link_va(va, root, parent, link, head);
/* return va;
* Last step is to check and update the tree. }
*/
augment_tree_propagate_from(va); static __always_inline struct vmap_area *
merge_or_add_vmap_area_augment(struct vmap_area *va,
struct rb_root *root, struct list_head *head)
{
va = merge_or_add_vmap_area(va, root, head);
if (va)
augment_tree_propagate_from(va);
return va; return va;
} }
...@@ -1138,7 +1148,7 @@ static void free_vmap_area(struct vmap_area *va) ...@@ -1138,7 +1148,7 @@ static void free_vmap_area(struct vmap_area *va)
* Insert/Merge it back to the free tree/list. * Insert/Merge it back to the free tree/list.
*/ */
spin_lock(&free_vmap_area_lock); spin_lock(&free_vmap_area_lock);
merge_or_add_vmap_area(va, &free_vmap_area_root, &free_vmap_area_list); merge_or_add_vmap_area_augment(va, &free_vmap_area_root, &free_vmap_area_list);
spin_unlock(&free_vmap_area_lock); spin_unlock(&free_vmap_area_lock);
} }
...@@ -1326,32 +1336,32 @@ void set_iounmap_nonlazy(void) ...@@ -1326,32 +1336,32 @@ void set_iounmap_nonlazy(void)
static bool __purge_vmap_area_lazy(unsigned long start, unsigned long end) static bool __purge_vmap_area_lazy(unsigned long start, unsigned long end)
{ {
unsigned long resched_threshold; unsigned long resched_threshold;
struct llist_node *valist; struct list_head local_pure_list;
struct vmap_area *va; struct vmap_area *va, *n_va;
struct vmap_area *n_va;
lockdep_assert_held(&vmap_purge_lock); lockdep_assert_held(&vmap_purge_lock);
valist = llist_del_all(&vmap_purge_list); spin_lock(&purge_vmap_area_lock);
if (unlikely(valist == NULL)) purge_vmap_area_root = RB_ROOT;
list_replace_init(&purge_vmap_area_list, &local_pure_list);
spin_unlock(&purge_vmap_area_lock);
if (unlikely(list_empty(&local_pure_list)))
return false; return false;
/* start = min(start,
* TODO: to calculate a flush range without looping. list_first_entry(&local_pure_list,
* The list can be up to lazy_max_pages() elements. struct vmap_area, list)->va_start);
*/
llist_for_each_entry(va, valist, purge_list) { end = max(end,
if (va->va_start < start) list_last_entry(&local_pure_list,
start = va->va_start; struct vmap_area, list)->va_end);
if (va->va_end > end)
end = va->va_end;
}
flush_tlb_kernel_range(start, end); flush_tlb_kernel_range(start, end);
resched_threshold = lazy_max_pages() << 1; resched_threshold = lazy_max_pages() << 1;
spin_lock(&free_vmap_area_lock); spin_lock(&free_vmap_area_lock);
llist_for_each_entry_safe(va, n_va, valist, purge_list) { list_for_each_entry_safe(va, n_va, &local_pure_list, list) {
unsigned long nr = (va->va_end - va->va_start) >> PAGE_SHIFT; unsigned long nr = (va->va_end - va->va_start) >> PAGE_SHIFT;
unsigned long orig_start = va->va_start; unsigned long orig_start = va->va_start;
unsigned long orig_end = va->va_end; unsigned long orig_end = va->va_end;
...@@ -1361,8 +1371,8 @@ static bool __purge_vmap_area_lazy(unsigned long start, unsigned long end) ...@@ -1361,8 +1371,8 @@ static bool __purge_vmap_area_lazy(unsigned long start, unsigned long end)
* detached and there is no need to "unlink" it from * detached and there is no need to "unlink" it from
* anything. * anything.
*/ */
va = merge_or_add_vmap_area(va, &free_vmap_area_root, va = merge_or_add_vmap_area_augment(va, &free_vmap_area_root,
&free_vmap_area_list); &free_vmap_area_list);
if (!va) if (!va)
continue; continue;
...@@ -1419,9 +1429,15 @@ static void free_vmap_area_noflush(struct vmap_area *va) ...@@ -1419,9 +1429,15 @@ static void free_vmap_area_noflush(struct vmap_area *va)
nr_lazy = atomic_long_add_return((va->va_end - va->va_start) >> nr_lazy = atomic_long_add_return((va->va_end - va->va_start) >>
PAGE_SHIFT, &vmap_lazy_nr); PAGE_SHIFT, &vmap_lazy_nr);
/* After this point, we may free va at any time */ /*
llist_add(&va->purge_list, &vmap_purge_list); * Merge or place it to the purge tree/list.
*/
spin_lock(&purge_vmap_area_lock);
merge_or_add_vmap_area(va,
&purge_vmap_area_root, &purge_vmap_area_list);
spin_unlock(&purge_vmap_area_lock);
/* After this point, we may free va at any time */
if (unlikely(nr_lazy > lazy_max_pages())) if (unlikely(nr_lazy > lazy_max_pages()))
try_purge_vmap_area_lazy(); try_purge_vmap_area_lazy();
} }
...@@ -3351,8 +3367,8 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets, ...@@ -3351,8 +3367,8 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
while (area--) { while (area--) {
orig_start = vas[area]->va_start; orig_start = vas[area]->va_start;
orig_end = vas[area]->va_end; orig_end = vas[area]->va_end;
va = merge_or_add_vmap_area(vas[area], &free_vmap_area_root, va = merge_or_add_vmap_area_augment(vas[area], &free_vmap_area_root,
&free_vmap_area_list); &free_vmap_area_list);
if (va) if (va)
kasan_release_vmalloc(orig_start, orig_end, kasan_release_vmalloc(orig_start, orig_end,
va->va_start, va->va_end); va->va_start, va->va_end);
...@@ -3401,8 +3417,8 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets, ...@@ -3401,8 +3417,8 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
for (area = 0; area < nr_vms; area++) { for (area = 0; area < nr_vms; area++) {
orig_start = vas[area]->va_start; orig_start = vas[area]->va_start;
orig_end = vas[area]->va_end; orig_end = vas[area]->va_end;
va = merge_or_add_vmap_area(vas[area], &free_vmap_area_root, va = merge_or_add_vmap_area_augment(vas[area], &free_vmap_area_root,
&free_vmap_area_list); &free_vmap_area_list);
if (va) if (va)
kasan_release_vmalloc(orig_start, orig_end, kasan_release_vmalloc(orig_start, orig_end,
va->va_start, va->va_end); va->va_start, va->va_end);
...@@ -3482,18 +3498,15 @@ static void show_numa_info(struct seq_file *m, struct vm_struct *v) ...@@ -3482,18 +3498,15 @@ static void show_numa_info(struct seq_file *m, struct vm_struct *v)
static void show_purge_info(struct seq_file *m) static void show_purge_info(struct seq_file *m)
{ {
struct llist_node *head;
struct vmap_area *va; struct vmap_area *va;
head = READ_ONCE(vmap_purge_list.first); spin_lock(&purge_vmap_area_lock);
if (head == NULL) list_for_each_entry(va, &purge_vmap_area_list, list) {
return;
llist_for_each_entry(va, head, purge_list) {
seq_printf(m, "0x%pK-0x%pK %7ld unpurged vm_area\n", seq_printf(m, "0x%pK-0x%pK %7ld unpurged vm_area\n",
(void *)va->va_start, (void *)va->va_end, (void *)va->va_start, (void *)va->va_end,
va->va_end - va->va_start); va->va_end - va->va_start);
} }
spin_unlock(&purge_vmap_area_lock);
} }
static int s_show(struct seq_file *m, void *p) static int s_show(struct seq_file *m, void *p)
...@@ -3551,10 +3564,7 @@ static int s_show(struct seq_file *m, void *p) ...@@ -3551,10 +3564,7 @@ static int s_show(struct seq_file *m, void *p)
seq_putc(m, '\n'); seq_putc(m, '\n');
/* /*
* As a final step, dump "unpurged" areas. Note, * As a final step, dump "unpurged" areas.
* that entire "/proc/vmallocinfo" output will not
* be address sorted, because the purge list is not
* sorted.
*/ */
if (list_is_last(&va->list, &vmap_area_list)) if (list_is_last(&va->list, &vmap_area_list))
show_purge_info(m); show_purge_info(m);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment