From: Derrick Stolee <dstolee@microsoft.com>
To: git@vger.kernel.org
Cc: jonathantanmy@google.com, stolee@gmail.com,
Derrick Stolee <dstolee@microsoft.com>
Subject: [PATCH] sha1_name: use bsearch_hash() for abbreviations
Date: Tue, 20 Mar 2018 16:03:25 -0400 [thread overview]
Message-ID: <20180320200325.168147-1-dstolee@microsoft.com> (raw)
This patch updates the abbreviation code to use bsearch_hash() as defined
in [1]. It gets a nice speedup since the old implementation did not use
the fanout table at all.
One caveat about the patch: there is a place where I cast a sha1 hash
into a struct object_id pointer. This is because the abbreviation code
still uses 'const unsigned char *' instead of structs. I wanted to avoid
a hashcpy() in these calls, but perhaps that is not too heavy a cost.
I look forward to feedback on this.
Thanks,
-Stolee
[1] https://public-inbox.org/git/007f3a4c84cb1c07255404ad1ea9f797129c5cf0.1517609773.git.jonathantanmy@google.com/
[PATCH 2/2] packfile: refactor hash search with fanout table
-- >8 --
When computing abbreviation lengths for an object ID against a single
packfile, the method find_abbrev_len_for_pack() currently implements
binary search. This is one of several implementations. One issue with
this implementation is that it ignores the fanout table in the pack-
index.
Translate this binary search to use the existing bsearch_hash() method
that correctly uses a fanout table. To keep the details about pack-
index version 1 or 2 out of sha1_name.c, create a bsearch_pack() method
in packfile.c.
Due to the use of the fanout table, the abbreviation computation is
slightly faster than before. For a fully-repacked copy of the Linux
repo, the following 'git log' commands improved:
* git log --oneline --parents --raw
Before: 66.83s
After: 64.85s
Rel %: -2.9%
* git log --oneline --parents
Before: 7.85s
After: 7.29s
Rel %: -7.1%
Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
---
packfile.c | 23 +++++++++++++++++++++++
packfile.h | 8 ++++++++
sha1_name.c | 24 ++++--------------------
3 files changed, 35 insertions(+), 20 deletions(-)
diff --git a/packfile.c b/packfile.c
index 7c1a2519fc..ea0388f6dd 100644
--- a/packfile.c
+++ b/packfile.c
@@ -1654,6 +1654,29 @@ void *unpack_entry(struct packed_git *p, off_t obj_offset,
return data;
}
+int bsearch_pack(const struct object_id *oid, const struct packed_git *p, uint32_t *result)
+{
+ const unsigned char *index_fanout = p->index_data;
+ const unsigned char *index_lookup;
+ int index_lookup_width;
+
+ if (!index_fanout)
+ BUG("bsearch_pack called without a valid pack-index");
+
+ index_lookup = index_fanout + 4 * 256;
+ if (p->index_version == 1) {
+ index_lookup_width = 24;
+ index_lookup += 4;
+ } else {
+ index_lookup_width = 20;
+ index_fanout += 8;
+ index_lookup += 8;
+ }
+
+ return bsearch_hash(oid->hash, (const uint32_t*)index_fanout,
+ index_lookup, index_lookup_width, result);
+}
+
const unsigned char *nth_packed_object_sha1(struct packed_git *p,
uint32_t n)
{
diff --git a/packfile.h b/packfile.h
index a7fca598d6..ec08cb2bb0 100644
--- a/packfile.h
+++ b/packfile.h
@@ -78,6 +78,14 @@ extern struct packed_git *add_packed_git(const char *path, size_t path_len, int
*/
extern void check_pack_index_ptr(const struct packed_git *p, const void *ptr);
+/*
+ * Perform binary search on a pack-index for a given oid. Packfile is expected to
+ * have a valid pack-index.
+ *
+ * See 'bsearch_hash' for more information.
+ */
+int bsearch_pack(const struct object_id *oid, const struct packed_git *p, uint32_t *result);
+
/*
* Return the SHA-1 of the nth object within the specified packfile.
* Open the index if it is not already open. The return value points
diff --git a/sha1_name.c b/sha1_name.c
index 735c1c0b8e..be3627b915 100644
--- a/sha1_name.c
+++ b/sha1_name.c
@@ -512,32 +512,16 @@ static void find_abbrev_len_for_pack(struct packed_git *p,
struct min_abbrev_data *mad)
{
int match = 0;
- uint32_t num, last, first = 0;
+ uint32_t num, first = 0;
struct object_id oid;
+ const struct object_id *mad_oid;
if (open_pack_index(p) || !p->num_objects)
return;
num = p->num_objects;
- last = num;
- while (first < last) {
- uint32_t mid = first + (last - first) / 2;
- const unsigned char *current;
- int cmp;
-
- current = nth_packed_object_sha1(p, mid);
- cmp = hashcmp(mad->hash, current);
- if (!cmp) {
- match = 1;
- first = mid;
- break;
- }
- if (cmp > 0) {
- first = mid + 1;
- continue;
- }
- last = mid;
- }
+ mad_oid = (const struct object_id *)mad->hash;
+ match = bsearch_pack(mad_oid, p, &first);
/*
* first is now the position in the packfile where we would insert
--
2.17.0.rc0
next reply other threads:[~2018-03-20 20:03 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-20 20:03 Derrick Stolee [this message]
2018-03-20 22:25 ` [PATCH] sha1_name: use bsearch_hash() for abbreviations Jonathan Tan
2018-03-21 13:24 ` Derrick Stolee
2018-03-21 22:42 ` brian m. carlson
2018-03-22 17:40 ` [PATCH v2 0/3] Use " Derrick Stolee
2018-03-22 17:40 ` [PATCH v2 1/3] sha1_name: convert struct min_abbrev_data to object_id Derrick Stolee
2018-03-22 17:40 ` [PATCH v2 2/3] packfile: define and use bsearch_pack() Derrick Stolee
2018-03-22 17:40 ` [PATCH v2 3/3] sha1_name: use bsearch_pack() for abbreviations Derrick Stolee
2018-03-24 16:41 ` [PATCH 4/3] sha1_name: use bsearch_pack() in unique_in_pack() René Scharfe
2018-03-25 16:19 ` Junio C Hamano
2018-03-25 16:32 ` René Scharfe
2018-03-25 18:21 ` Derrick Stolee
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=20180320200325.168147-1-dstolee@microsoft.com \
--to=dstolee@microsoft.com \
--cc=git@vger.kernel.org \
--cc=jonathantanmy@google.com \
--cc=stolee@gmail.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.