From: mathieu.desnoyers@efficios.com (Mathieu Desnoyers)
Subject: [lttng-dev] [PATCH lttng-tools] Fix: hashtable: take split_count_order into account
Date: Fri, 30 Aug 2013 14:31:48 -0400 [thread overview]
Message-ID: <20130830183148.GA12014@Krystal> (raw)
This makes check_resize() handle split-counters more precisely.
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 ebdc4ff..0ddd21a 100644
--- a/src/common/hashtable/rculfhash.c
+++ b/src/common/hashtable/rculfhash.c
@@ -570,6 +570,7 @@ void cds_lfht_resize_lazy_count(struct cds_lfht *ht, unsigned long size,
static long nr_cpus_mask = -1;
static long split_count_mask = -1;
+static int split_count_order = -1;
#if defined(HAVE_SYSCONF)
static void ht_init_nr_cpus_mask(void)
@@ -606,6 +607,8 @@ void alloc_split_items_count(struct cds_lfht *ht)
split_count_mask = DEFAULT_SPLIT_COUNT_MASK;
else
split_count_mask = nr_cpus_mask;
+ split_count_order =
+ cds_lfht_get_count_order_ulong(split_count_mask + 1);
}
assert(split_count_mask >= 0);
@@ -721,7 +724,7 @@ void check_resize(struct cds_lfht *ht, unsigned long size, uint32_t chain_len)
* Use bucket-local length for small table expand and for
* environments lacking per-cpu data support.
*/
- if (count >= (1UL << COUNT_COMMIT_ORDER))
+ if (count >= (1UL << (COUNT_COMMIT_ORDER + split_count_order)))
return;
if (chain_len > 100)
dbg_printf("WARNING: large chain length: %u.\n",
@@ -735,7 +738,9 @@ void check_resize(struct cds_lfht *ht, unsigned long size, uint32_t chain_len)
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)) {
+ && (size << growth)
+ >= (1UL << (COUNT_COMMIT_ORDER
+ + split_count_order))) {
/*
* If ideal growth expands the hash table size
* beyond the "small hash table" sizes, use the
@@ -745,8 +750,8 @@ void check_resize(struct cds_lfht *ht, unsigned long size, uint32_t chain_len)
* the chain length is used to expand the hash
* table in every case.
*/
- growth = COUNT_COMMIT_ORDER -
- cds_lfht_get_count_order_u32(size);
+ growth = COUNT_COMMIT_ORDER + split_count_order
+ - cds_lfht_get_count_order_ulong(size);
if (growth <= 0)
return;
}
--
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com
reply other threads:[~2013-08-30 18:31 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20130830183148.GA12014@Krystal \
--to=mathieu.desnoyers@efficios.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox