git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Nguyễn Thái Ngọc Duy" <pclouds@gmail.com>
To: git@vger.kernel.org
Cc: "Nguyễn Thái Ngọc Duy" <pclouds@gmail.com>
Subject: [PATCH v3 11/13] Preallocate hash tables when the number of inserts are known in advance
Date: Tue, 12 Mar 2013 20:04:58 +0700	[thread overview]
Message-ID: <1363093500-16796-12-git-send-email-pclouds@gmail.com> (raw)
In-Reply-To: <1363093500-16796-1-git-send-email-pclouds@gmail.com>

This avoids unnecessary allocations and reinsertions. On webkit.git
(i.e. about 100k inserts to the name hash table), this reduces about
100ms out of 3s user time.

treat_leading_path:   0.000  0.000
read_directory:       1.299  1.305
+treat_one_path:      0.599  0.599
++is_excluded:        0.103  0.103
+++prep_exclude:      0.040  0.040
+++matching:          0.036  0.035
++dir_exist:          0.036  0.035
++index_name_exists:  0.291  0.292
lazy_init_name_hash:  0.257  0.155
+simplify_away:       0.082  0.083
+dir_add_name:        0.000  0.000


Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
---
 diffcore-rename.c | 1 +
 hash.h            | 7 +++++++
 name-hash.c       | 1 +
 3 files changed, 9 insertions(+)

diff --git a/diffcore-rename.c b/diffcore-rename.c
index 512d0ac..8d3d9bb 100644
--- a/diffcore-rename.c
+++ b/diffcore-rename.c
@@ -389,6 +389,7 @@ static int find_exact_renames(struct diff_options *options)
 	struct hash_table file_table;
 
 	init_hash(&file_table);
+	preallocate_hash(&file_table, (rename_src_nr + rename_dst_nr) * 2);
 	for (i = 0; i < rename_src_nr; i++)
 		insert_file_table(&file_table, -1, i, rename_src[i].p->one);
 
diff --git a/hash.h b/hash.h
index b875ce6..244d1fe 100644
--- a/hash.h
+++ b/hash.h
@@ -40,4 +40,11 @@ static inline void init_hash(struct hash_table *table)
 	table->array = NULL;
 }
 
+static inline void preallocate_hash(struct hash_table *table, unsigned int size)
+{
+	assert(table->size == 0 && table->nr == 0 && table->array == NULL);
+	table->size = size;
+	table->array = xcalloc(sizeof(struct hash_table_entry), size);
+}
+
 #endif
diff --git a/name-hash.c b/name-hash.c
index 942c459..1287a19 100644
--- a/name-hash.c
+++ b/name-hash.c
@@ -92,6 +92,7 @@ static void lazy_init_name_hash(struct index_state *istate)
 
 	if (istate->name_hash_initialized)
 		return;
+	preallocate_hash(&istate->name_hash, istate->cache_nr * 2);
 	for (nr = 0; nr < istate->cache_nr; nr++)
 		hash_index_entry(istate, istate->cache[nr]);
 	istate->name_hash_initialized = 1;
-- 
1.8.1.2.536.gf441e6d

  parent reply	other threads:[~2013-03-12 13:07 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-03-09  4:09 [PATCH 0/3] Trivial (and small) exclude optimizations Nguyễn Thái Ngọc Duy
2013-03-09  4:09 ` [PATCH 1/3] match_pathname: avoid calling strncmp if baselen is 0 Nguyễn Thái Ngọc Duy
2013-03-09  9:06   ` Antoine Pelisse
2013-03-09  4:09 ` [PATCH 2/3] dir.c: inline convenient *_icase helpers Nguyễn Thái Ngọc Duy
2013-03-09  4:09 ` [PATCH 3/3] match_basename: use strncmp instead of strcmp Nguyễn Thái Ngọc Duy
2013-03-09  7:50   ` Junio C Hamano
2013-03-09  8:47     ` Fredrik Gustafsson
2013-03-09  9:58     ` Duy Nguyen
2013-03-10  6:14 ` [PATCH v2 0/6] Exclude optimizations Nguyễn Thái Ngọc Duy
2013-03-10  6:14   ` [PATCH v2 1/6] match_pathname: avoid calling strncmp if baselen is 0 Nguyễn Thái Ngọc Duy
2013-03-10  6:14   ` [PATCH v2 2/6] dir.c: inline convenient *_icase helpers Nguyễn Thái Ngọc Duy
2013-03-10  6:14   ` [PATCH v2 3/6] match_basename: use strncmp instead of strcmp Nguyễn Thái Ngọc Duy
2013-03-10  7:34     ` Junio C Hamano
2013-03-10 10:38       ` Duy Nguyen
2013-03-10 11:43         ` Antoine Pelisse
2013-03-10 11:54           ` Antoine Pelisse
2013-03-10 12:06             ` Duy Nguyen
2013-03-10 12:11               ` Antoine Pelisse
2013-03-10 12:14                 ` Duy Nguyen
2013-03-12 20:59         ` Junio C Hamano
2013-03-13  1:11           ` Duy Nguyen
2013-03-10  6:14   ` [PATCH v2 4/6] match_{base,path}name: replace strncmp_icase with strnequal_icase Nguyễn Thái Ngọc Duy
2013-03-10  6:14   ` [PATCH v2 5/6] dir.c: pass pathname length to last_exclude_matching Nguyễn Thái Ngọc Duy
2013-03-10  6:14   ` [PATCH v2 6/6] exclude: filter patterns by directory level Nguyễn Thái Ngọc Duy
2013-03-10  8:20     ` Junio C Hamano
2013-03-10 10:18       ` Duy Nguyen
2013-03-10 10:58       ` Junio C Hamano
2013-03-10 11:14         ` Duy Nguyen
2013-03-11 15:11   ` [PATCH v2 0/6] Exclude optimizations Duy Nguyen
2013-03-12 13:04   ` [PATCH v3 00/13] " Nguyễn Thái Ngọc Duy
2013-03-12 13:04     ` [PATCH v3 01/13] dir.c: add MEASURE_EXCLUDE code for tracking exclude performance Nguyễn Thái Ngọc Duy
2013-03-12 13:04     ` [PATCH v3 02/13] match_pathname: avoid calling strncmp if baselen is 0 Nguyễn Thái Ngọc Duy
2013-03-12 13:04     ` [PATCH v3 03/13] dir.c: inline convenient *_icase helpers Nguyễn Thái Ngọc Duy
2013-03-12 13:04     ` [PATCH v3 04/13] match_basename: use strncmp instead of strcmp Nguyễn Thái Ngọc Duy
2013-03-12 17:40       ` Antoine Pelisse
2013-03-13  1:05         ` Duy Nguyen
2013-03-12 13:04     ` [PATCH v3 05/13] match_{base,path}name: replace strncmp_icase with memequal_icase Nguyễn Thái Ngọc Duy
2013-03-13  1:14       ` Duy Nguyen
2013-03-12 13:04     ` [PATCH v3 06/13] dir: pass pathname length to last_exclude_matching Nguyễn Thái Ngọc Duy
2013-03-12 13:04     ` [PATCH v3 07/13] exclude: avoid calling prep_exclude on entries of the same directory Nguyễn Thái Ngọc Duy
2013-03-12 13:04     ` [PATCH v3 08/13] exclude: record baselen in the pattern Nguyễn Thái Ngọc Duy
2013-03-12 13:04     ` [PATCH v3 09/13] exclude: filter out patterns not applicable to the current directory Nguyễn Thái Ngọc Duy
2013-03-12 23:13       ` Eric Sunshine
2013-03-12 13:04     ` [PATCH v3 10/13] read_directory: avoid invoking exclude machinery on tracked files Nguyễn Thái Ngọc Duy
2013-03-12 13:04     ` Nguyễn Thái Ngọc Duy [this message]
2013-03-12 13:04     ` [PATCH v3 12/13] name-hash: allow to lookup a name with precalculated base hash Nguyễn Thái Ngọc Duy
2013-03-12 13:05     ` [PATCH v3 13/13] read_directory: calculate name hashes incrementally Nguyễn Thái Ngọc Duy
2013-03-14 13:05     ` [PATCH v3 00/13] Exclude optimizations Duy Nguyen

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=1363093500-16796-12-git-send-email-pclouds@gmail.com \
    --to=pclouds@gmail.com \
    --cc=git@vger.kernel.org \
    /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;
as well as URLs for NNTP newsgroup(s).