* [lttng-dev] [PATCH lttng-tools] Fix: hash table growth (for small tables) should be limited
@ 2013-08-27 22:01 Mathieu Desnoyers
2013-08-27 22:11 ` [lttng-dev] [PATCH lttng-tools] Fix: hash table growth (for small tables) should be limited (v2) Mathieu Desnoyers
0 siblings, 1 reply; 3+ messages in thread
From: Mathieu Desnoyers @ 2013-08-27 22:01 UTC (permalink / raw)
Buckets with many entries encountered in a hash table could cause it to
grow to a large size, beyond the scope for which this mechanism is
expected to play a role when node accounting is available. Indeed, when
the hash table grows to larger size, split-counter node accounting is
expected to deal with resize/shrink rather than relying on an heuristic
based on the largest bucket size.
This is fixing an issue where we see hash tables sometimes reaching 65k
entries index (65536*8 = 524288 bytes) for a workload limited to adding
1000 entries and then removing all of them, done in a loop (random
keys).
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers at efficios.com>
---
diff --git a/src/common/hashtable/rculfhash.c b/src/common/hashtable/rculfhash.c
index b430a3d..eeb5c43 100644
--- a/src/common/hashtable/rculfhash.c
+++ b/src/common/hashtable/rculfhash.c
@@ -726,9 +726,32 @@ void check_resize(struct cds_lfht *ht, unsigned long size, uint32_t chain_len)
if (chain_len > 100)
dbg_printf("WARNING: large chain length: %u.\n",
chain_len);
- if (chain_len >= CHAIN_LEN_RESIZE_THRESHOLD)
- cds_lfht_resize_lazy_grow(ht, size,
- cds_lfht_get_count_order_u32(chain_len - (CHAIN_LEN_TARGET - 1)));
+ if (chain_len >= CHAIN_LEN_RESIZE_THRESHOLD) {
+ int growth;
+
+ /*
+ * Ideal growth calculated based on chain length.
+ */
+ growth = cds_lfht_get_count_order_u32(chain_len
+ - (CHAIN_LEN_TARGET - 1));
+ if (!(ht->flags & CDS_LFHT_ACCOUNTING)
+ && (size << growth) >= (1UL << COUNT_COMMIT_ORDER)) {
+ /*
+ * If ideal growth expands the hash table size
+ * beyond the "small hash table" sizes, use the
+ * maximum small hash table size to attempt
+ * expanding the hash table. This only applies
+ * when node accounting is available, otherwise
+ * the chain length is used to expand the hash
+ * table in every case.
+ */
+ growth = COUNT_COMMIT_ORDER -
+ cds_lfht_get_count_order_u32(size);
+ if (growth <= 0)
+ return;
+ }
+ cds_lfht_resize_lazy_grow(ht, size, growth);
+ }
}
static
--
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com
^ permalink raw reply [flat|nested] 3+ messages in thread
* [lttng-dev] [PATCH lttng-tools] Fix: hash table growth (for small tables) should be limited (v2)
2013-08-27 22:01 [lttng-dev] [PATCH lttng-tools] Fix: hash table growth (for small tables) should be limited Mathieu Desnoyers
@ 2013-08-27 22:11 ` Mathieu Desnoyers
2013-08-29 16:54 ` David Goulet
0 siblings, 1 reply; 3+ messages in thread
From: Mathieu Desnoyers @ 2013-08-27 22:11 UTC (permalink / raw)
Buckets with many entries encountered in a hash table could cause it to
grow to a large size, beyond the scope for which this mechanism is
expected to play a role when node accounting is available. Indeed, when
the hash table grows to larger size, split-counter node accounting is
expected to deal with resize/shrink rather than relying on an heuristic
based on the largest bucket size.
This is fixing an issue where we see hash tables sometimes reaching 65k
entries index (65536*8 = 524288 bytes) for a workload limited to adding
1000 entries and then removing all of them, done in a loop (random
keys).
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers at efficios.com>
---
diff --git a/src/common/hashtable/rculfhash.c b/src/common/hashtable/rculfhash.c
index b430a3d..ebdc4ff 100644
--- a/src/common/hashtable/rculfhash.c
+++ b/src/common/hashtable/rculfhash.c
@@ -726,9 +726,32 @@ void check_resize(struct cds_lfht *ht, unsigned long size, uint32_t chain_len)
if (chain_len > 100)
dbg_printf("WARNING: large chain length: %u.\n",
chain_len);
- if (chain_len >= CHAIN_LEN_RESIZE_THRESHOLD)
- cds_lfht_resize_lazy_grow(ht, size,
- cds_lfht_get_count_order_u32(chain_len - (CHAIN_LEN_TARGET - 1)));
+ if (chain_len >= CHAIN_LEN_RESIZE_THRESHOLD) {
+ int growth;
+
+ /*
+ * Ideal growth calculated based on chain length.
+ */
+ growth = cds_lfht_get_count_order_u32(chain_len
+ - (CHAIN_LEN_TARGET - 1));
+ if ((ht->flags & CDS_LFHT_ACCOUNTING)
+ && (size << growth) >= (1UL << COUNT_COMMIT_ORDER)) {
+ /*
+ * If ideal growth expands the hash table size
+ * beyond the "small hash table" sizes, use the
+ * maximum small hash table size to attempt
+ * expanding the hash table. This only applies
+ * when node accounting is available, otherwise
+ * the chain length is used to expand the hash
+ * table in every case.
+ */
+ growth = COUNT_COMMIT_ORDER -
+ cds_lfht_get_count_order_u32(size);
+ if (growth <= 0)
+ return;
+ }
+ cds_lfht_resize_lazy_grow(ht, size, growth);
+ }
}
static
--
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com
^ permalink raw reply [flat|nested] 3+ messages in thread
* [lttng-dev] [PATCH lttng-tools] Fix: hash table growth (for small tables) should be limited (v2)
2013-08-27 22:11 ` [lttng-dev] [PATCH lttng-tools] Fix: hash table growth (for small tables) should be limited (v2) Mathieu Desnoyers
@ 2013-08-29 16:54 ` David Goulet
0 siblings, 0 replies; 3+ messages in thread
From: David Goulet @ 2013-08-29 16:54 UTC (permalink / raw)
Merged!
Mathieu Desnoyers:
> Buckets with many entries encountered in a hash table could cause it to
> grow to a large size, beyond the scope for which this mechanism is
> expected to play a role when node accounting is available. Indeed, when
> the hash table grows to larger size, split-counter node accounting is
> expected to deal with resize/shrink rather than relying on an heuristic
> based on the largest bucket size.
>
> This is fixing an issue where we see hash tables sometimes reaching 65k
> entries index (65536*8 = 524288 bytes) for a workload limited to adding
> 1000 entries and then removing all of them, done in a loop (random
> keys).
>
> Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers at efficios.com>
> ---
> diff --git a/src/common/hashtable/rculfhash.c b/src/common/hashtable/rculfhash.c
> index b430a3d..ebdc4ff 100644
> --- a/src/common/hashtable/rculfhash.c
> +++ b/src/common/hashtable/rculfhash.c
> @@ -726,9 +726,32 @@ void check_resize(struct cds_lfht *ht, unsigned long size, uint32_t chain_len)
> if (chain_len > 100)
> dbg_printf("WARNING: large chain length: %u.\n",
> chain_len);
> - if (chain_len >= CHAIN_LEN_RESIZE_THRESHOLD)
> - cds_lfht_resize_lazy_grow(ht, size,
> - cds_lfht_get_count_order_u32(chain_len - (CHAIN_LEN_TARGET - 1)));
> + if (chain_len >= CHAIN_LEN_RESIZE_THRESHOLD) {
> + int growth;
> +
> + /*
> + * Ideal growth calculated based on chain length.
> + */
> + growth = cds_lfht_get_count_order_u32(chain_len
> + - (CHAIN_LEN_TARGET - 1));
> + if ((ht->flags & CDS_LFHT_ACCOUNTING)
> + && (size << growth) >= (1UL << COUNT_COMMIT_ORDER)) {
> + /*
> + * If ideal growth expands the hash table size
> + * beyond the "small hash table" sizes, use the
> + * maximum small hash table size to attempt
> + * expanding the hash table. This only applies
> + * when node accounting is available, otherwise
> + * the chain length is used to expand the hash
> + * table in every case.
> + */
> + growth = COUNT_COMMIT_ORDER -
> + cds_lfht_get_count_order_u32(size);
> + if (growth <= 0)
> + return;
> + }
> + cds_lfht_resize_lazy_grow(ht, size, growth);
> + }
> }
>
> static
>
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2013-08-29 16:54 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-08-27 22:01 [lttng-dev] [PATCH lttng-tools] Fix: hash table growth (for small tables) should be limited Mathieu Desnoyers
2013-08-27 22:11 ` [lttng-dev] [PATCH lttng-tools] Fix: hash table growth (for small tables) should be limited (v2) Mathieu Desnoyers
2013-08-29 16:54 ` David Goulet
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox