The patch titled
Subject: mm, vmscan: simplify the logic deciding whether kswapd sleeps
has been added to the -mm tree. Its filename is
mm-vmscan-simplify-the-logic-deciding-whether-kswapd-sleeps.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/mm-vmscan-simplify-the-logic-deciding-whether-kswapd-sleeps.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/mm-vmscan-simplify-the-logic-deciding-whether-kswapd-sleeps.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>
Subject: mm, vmscan: simplify the logic deciding whether kswapd sleeps
kswapd goes through some complex steps trying to figure out if it should
stay awake based on the classzone_idx and the requested order. It is
unnecessarily complex and passes in an invalid classzone_idx to
balance_pgdat(). What matters most of all is whether a larger order has
been requsted and whether kswapd successfully reclaimed at the previous
order. This patch irons out the logic to check just that and the end
result is less headache inducing.
Link: http://lkml.kernel.org/r/1466518566-30034-9-git-send-email-mgorman@xxxxxxxxxxxxxxxxxxx
Signed-off-by: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>
Acked-by: Johannes Weiner <hannes@xxxxxxxxxxx>
Cc: Vlastimil Babka <vbabka@xxxxxxx>
Cc: Rik van Riel <riel@xxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/mmzone.h | 5 +
mm/memory_hotplug.c | 5 +
mm/page_alloc.c | 2
mm/vmscan.c | 104 +++++++++++++++++++++------------------
4 files changed, 63 insertions(+), 53 deletions(-)
diff -puN include/linux/mmzone.h~mm-vmscan-simplify-the-logic-deciding-whether-kswapd-sleeps include/linux/mmzone.h
--- a/include/linux/mmzone.h~mm-vmscan-simplify-the-logic-deciding-whether-kswapd-sleeps
+++ a/include/linux/mmzone.h
@@ -667,8 +667,9 @@ typedef struct pglist_data {
wait_queue_head_t pfmemalloc_wait;
struct task_struct *kswapd; /* Protected by
mem_hotplug_begin/end() */
- int kswapd_max_order;
- enum zone_type classzone_idx;
+ int kswapd_order;
+ enum zone_type kswapd_classzone_idx;
+
#ifdef CONFIG_COMPACTION
int kcompactd_max_order;
enum zone_type kcompactd_classzone_idx;
diff -puN mm/memory_hotplug.c~mm-vmscan-simplify-the-logic-deciding-whether-kswapd-sleeps mm/memory_hotplug.c
--- a/mm/memory_hotplug.c~mm-vmscan-simplify-the-logic-deciding-whether-kswapd-sleeps
+++ a/mm/memory_hotplug.c
@@ -1209,9 +1209,10 @@ static pg_data_t __ref *hotadd_new_pgdat
arch_refresh_nodedata(nid, pgdat);
} else {
- /* Reset the nr_zones and classzone_idx to 0 before reuse */
+ /* Reset the nr_zones, order and classzone_idx before reuse */
pgdat->nr_zones = 0;
- pgdat->classzone_idx = 0;
+ pgdat->kswapd_order = 0;
+ pgdat->kswapd_classzone_idx = 0;
}
/* we can use NODE_DATA(nid) from here */
diff -puN mm/page_alloc.c~mm-vmscan-simplify-the-logic-deciding-whether-kswapd-sleeps mm/page_alloc.c
--- a/mm/page_alloc.c~mm-vmscan-simplify-the-logic-deciding-whether-kswapd-sleeps
+++ a/mm/page_alloc.c
@@ -6046,7 +6046,7 @@ void __paginginit free_area_init_node(in
unsigned long end_pfn = 0;
/* pg_data_t should be reset to zero when it's allocated */
- WARN_ON(pgdat->nr_zones || pgdat->classzone_idx);
+ WARN_ON(pgdat->nr_zones || pgdat->kswapd_classzone_idx);
reset_deferred_meminit(pgdat);
pgdat->node_id = nid;
diff -puN mm/vmscan.c~mm-vmscan-simplify-the-logic-deciding-whether-kswapd-sleeps mm/vmscan.c
--- a/mm/vmscan.c~mm-vmscan-simplify-the-logic-deciding-whether-kswapd-sleeps
+++ a/mm/vmscan.c
@@ -2762,7 +2762,7 @@ static bool pfmemalloc_watermark_ok(pg_d
/* kswapd must be awake if processes are being throttled */
if (!wmark_ok && waitqueue_active(&pgdat->kswapd_wait)) {
- pgdat->classzone_idx = min(pgdat->classzone_idx,
+ pgdat->kswapd_classzone_idx = min(pgdat->kswapd_classzone_idx,
(enum zone_type)ZONE_NORMAL);
wake_up_interruptible(&pgdat->kswapd_wait);
}
@@ -3236,8 +3236,8 @@ out:
return sc.order;
}
-static void kswapd_try_to_sleep(pg_data_t *pgdat, int order,
- int classzone_idx, int balanced_classzone_idx)
+static void kswapd_try_to_sleep(pg_data_t *pgdat, int alloc_order, int reclaim_order,
+ int classzone_idx)
{
long remaining = 0;
DEFINE_WAIT(wait);
@@ -3247,9 +3247,19 @@ static void kswapd_try_to_sleep(pg_data_
prepare_to_wait(&pgdat->kswapd_wait, &wait, TASK_INTERRUPTIBLE);
+ /*
+ * If kswapd has not been woken recently, then kswapd goes fully
+ * to sleep. kcompactd may still need to wake if the original
+ * request was high-order.
+ */
+ if (classzone_idx == -1) {
+ wakeup_kcompactd(pgdat, alloc_order, classzone_idx);
+ classzone_idx = MAX_NR_ZONES - 1;
+ goto full_sleep;
+ }
+
/* Try to sleep for a short interval */
- if (prepare_kswapd_sleep(pgdat, order, remaining,
- balanced_classzone_idx)) {
+ if (prepare_kswapd_sleep(pgdat, reclaim_order, remaining, classzone_idx)) {
/*
* Compaction records what page blocks it recently failed to
* isolate pages from and skips them in the future scanning.
@@ -3262,19 +3272,19 @@ static void kswapd_try_to_sleep(pg_data_
* We have freed the memory, now we should compact it to make
* allocation of the requested order possible.
*/
- wakeup_kcompactd(pgdat, order, classzone_idx);
+ wakeup_kcompactd(pgdat, alloc_order, classzone_idx);
remaining = schedule_timeout(HZ/10);
finish_wait(&pgdat->kswapd_wait, &wait);
prepare_to_wait(&pgdat->kswapd_wait, &wait, TASK_INTERRUPTIBLE);
}
+full_sleep:
/*
* After a short sleep, check if it was a premature sleep. If not, then
* go fully to sleep until explicitly woken up.
*/
- if (prepare_kswapd_sleep(pgdat, order, remaining,
- balanced_classzone_idx)) {
+ if (prepare_kswapd_sleep(pgdat, reclaim_order, remaining, classzone_idx)) {
trace_mm_vmscan_kswapd_sleep(pgdat->node_id);
/*
@@ -3315,9 +3325,7 @@ static void kswapd_try_to_sleep(pg_data_
*/
static int kswapd(void *p)
{
- unsigned long order, new_order;
- int classzone_idx, new_classzone_idx;
- int balanced_classzone_idx;
+ unsigned int alloc_order, reclaim_order, classzone_idx;
pg_data_t *pgdat = (pg_data_t*)p;
struct task_struct *tsk = current;
@@ -3347,38 +3355,26 @@ static int kswapd(void *p)
tsk->flags |= PF_MEMALLOC | PF_SWAPWRITE | PF_KSWAPD;
set_freezable();
- order = new_order = 0;
- classzone_idx = new_classzone_idx = pgdat->nr_zones - 1;
- balanced_classzone_idx = classzone_idx;
+ pgdat->kswapd_order = alloc_order = reclaim_order = 0;
+ pgdat->kswapd_classzone_idx = classzone_idx = -1;
for ( ; ; ) {
bool ret;
- /*
- * While we were reclaiming, there might have been another
- * wakeup, so check the values.
- */
- new_order = pgdat->kswapd_max_order;
- new_classzone_idx = pgdat->classzone_idx;
- pgdat->kswapd_max_order = 0;
- pgdat->classzone_idx = pgdat->nr_zones - 1;
+kswapd_try_sleep:
+ kswapd_try_to_sleep(pgdat, alloc_order, reclaim_order,
+ classzone_idx);
- if (order < new_order || classzone_idx > new_classzone_idx) {
- /*
- * Don't sleep if someone wants a larger 'order'
- * allocation or has tigher zone constraints
- */
- order = new_order;
- classzone_idx = new_classzone_idx;
- } else {
- kswapd_try_to_sleep(pgdat, order, classzone_idx,
- balanced_classzone_idx);
- order = pgdat->kswapd_max_order;
- classzone_idx = pgdat->classzone_idx;
- new_order = order;
- new_classzone_idx = classzone_idx;
- pgdat->kswapd_max_order = 0;
- pgdat->classzone_idx = pgdat->nr_zones - 1;
- }
+ /*
+ * Read the new order and classzone_idx which may be -1 if
+ * kswapd_try_to_sleep() woke up after a short timeout instead
+ * of being woken by the page allocator.
+ */
+ alloc_order = reclaim_order = pgdat->kswapd_order;
+ classzone_idx = pgdat->kswapd_classzone_idx;
+ if (classzone_idx == -1)
+ classzone_idx = MAX_NR_ZONES - 1;
+ pgdat->kswapd_order = 0;
+ pgdat->kswapd_classzone_idx = -1;
ret = try_to_freeze();
if (kthread_should_stop())
@@ -3388,12 +3384,24 @@ static int kswapd(void *p)
* We can speed up thawing tasks if we don't call balance_pgdat
* after returning from the refrigerator
*/
- if (!ret) {
- trace_mm_vmscan_kswapd_wake(pgdat->node_id, order);
+ if (ret)
+ continue;
- /* return value ignored until next patch */
- balance_pgdat(pgdat, order, classzone_idx);
- }
+ /*
+ * Reclaim begins at the requested order but if a high-order
+ * reclaim fails then kswapd falls back to reclaiming for
+ * order-0. If that happens, kswapd will consider sleeping
+ * for the order it finished reclaiming at (reclaim_order)
+ * but kcompactd is woken to compact for the original
+ * request (alloc_order).
+ */
+ trace_mm_vmscan_kswapd_wake(pgdat->node_id, alloc_order);
+ reclaim_order = balance_pgdat(pgdat, alloc_order, classzone_idx);
+ if (reclaim_order < alloc_order)
+ goto kswapd_try_sleep;
+
+ alloc_order = reclaim_order = pgdat->kswapd_order;
+ classzone_idx = pgdat->kswapd_classzone_idx;
}
tsk->flags &= ~(PF_MEMALLOC | PF_SWAPWRITE | PF_KSWAPD);
@@ -3416,10 +3424,10 @@ void wakeup_kswapd(struct zone *zone, in
if (!cpuset_zone_allowed(zone, GFP_KERNEL | __GFP_HARDWALL))
return;
pgdat = zone->zone_pgdat;
- if (pgdat->kswapd_max_order < order) {
- pgdat->kswapd_max_order = order;
- pgdat->classzone_idx = min(pgdat->classzone_idx, classzone_idx);
- }
+ if (pgdat->kswapd_classzone_idx == -1)
+ pgdat->kswapd_classzone_idx = classzone_idx;
+ pgdat->kswapd_classzone_idx = max(pgdat->kswapd_classzone_idx, classzone_idx);
+ pgdat->kswapd_order = max(pgdat->kswapd_order, order);
if (!waitqueue_active(&pgdat->kswapd_wait))
return;
if (zone_balanced(zone, order, 0))
_
Patches currently in -mm which might be from mgorman@xxxxxxxxxxxxxxxxxxx are
mm-slaub-add-__gfp_atomic-to-the-gfp-reclaim-mask.patch
mm-vmstat-add-infrastructure-for-per-node-vmstats.patch
mm-vmscan-move-lru_lock-to-the-node.patch
mm-vmscan-move-lru-lists-to-node.patch
mm-vmscan-begin-reclaiming-pages-on-a-per-node-basis.patch
mm-vmscan-have-kswapd-only-scan-based-on-the-highest-requested-zone.patch
mm-vmscan-make-kswapd-reclaim-in-terms-of-nodes.patch
mm-vmscan-remove-balance-gap.patch
mm-vmscan-simplify-the-logic-deciding-whether-kswapd-sleeps.patch
mm-vmscan-by-default-have-direct-reclaim-only-shrink-once-per-node.patch
mm-vmscan-remove-duplicate-logic-clearing-node-congestion-and-dirty-state.patch
mm-vmscan-do-not-reclaim-from-kswapd-if-there-is-any-eligible-zone.patch
mm-vmscan-make-shrink_node-decisions-more-node-centric.patch
mm-memcg-move-memcg-limit-enforcement-from-zones-to-nodes.patch
mm-workingset-make-working-set-detection-node-aware.patch
mm-page_alloc-consider-dirtyable-memory-in-terms-of-nodes.patch
mm-move-page-mapped-accounting-to-the-node.patch
mm-rename-nr_anon_pages-to-nr_anon_mapped.patch
mm-move-most-file-based-accounting-to-the-node.patch
mm-move-vmscan-writes-and-file-write-accounting-to-the-node.patch
mm-vmscan-update-classzone_idx-if-buffer_heads_over_limit.patch
mm-vmscan-only-wakeup-kswapd-once-per-node-for-the-requested-classzone.patch
mm-convert-zone_reclaim-to-node_reclaim.patch
mm-vmscan-add-classzone-information-to-tracepoints.patch
mm-page_alloc-remove-fair-zone-allocation-policy.patch
mm-page_alloc-cache-the-last-node-whose-dirty-limit-is-reached.patch
mm-vmstat-replace-__count_zone_vm_events-with-a-zone-id-equivalent.patch
mm-vmstat-account-per-zone-stalls-and-pages-skipped-during-reclaim.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
