summaryrefslogtreecommitdiff
path: root/builtin/grep.c
diff options
context:
space:
mode:
authorJeff King <peff@peff.net>2012-02-02 03:24:28 -0500
committerJunio C Hamano <gitster@pobox.com>2012-02-02 10:36:08 -0800
commit9dd5245c1043dd18fd7b3f44b9e51eef7e4b58d8 (patch)
tree69cb585a0fb10df11ca839fdaca7795193fbe12c /builtin/grep.c
parent08265798e1ff6abc1b0aaff31c1471f83bd51425 (diff)
downloadgit-9dd5245c1043dd18fd7b3f44b9e51eef7e4b58d8.tar.gz
grep: pre-load userdiff drivers when threadedjk/grep-binary-attribute
The low-level grep_source code will automatically load the userdiff driver to see whether a file is binary. However, when we are threaded, it will load the drivers in a non-deterministic order, handling each one as its assigned thread happens to be scheduled. Meanwhile, the attribute lookup code (which underlies the userdiff driver lookup) is optimized to handle paths in sequential order (because they tend to share the same gitattributes files). Multi-threading the lookups destroys the locality and makes this optimization less effective. We can fix this by pre-loading the userdiff driver in the main thread, before we hand off the file to a worker thread. My best-of-five for "git grep foo" on the linux-2.6 repository went from: real 0m0.391s user 0m1.708s sys 0m0.584s to: real 0m0.360s user 0m1.576s sys 0m0.572s Not a huge speedup, but it's quite easy to do. The only trick is that we shouldn't perform this optimization if "-a" was used, in which case we won't bother checking whether the files are binary at all. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Diffstat (limited to 'builtin/grep.c')
-rw-r--r--builtin/grep.c10
1 files changed, 6 insertions, 4 deletions
diff --git a/builtin/grep.c b/builtin/grep.c
index c9b6a138fe..e741aca18c 100644
--- a/builtin/grep.c
+++ b/builtin/grep.c
@@ -85,8 +85,8 @@ static pthread_cond_t cond_result;
static int skip_first_line;
-static void add_work(enum grep_source_type type, const char *name,
- const void *id)
+static void add_work(struct grep_opt *opt, enum grep_source_type type,
+ const char *name, const void *id)
{
grep_lock();
@@ -95,6 +95,8 @@ static void add_work(enum grep_source_type type, const char *name,
}
grep_source_init(&todo[todo_end].source, type, name, id);
+ if (opt->binary != GREP_BINARY_TEXT)
+ grep_source_load_driver(&todo[todo_end].source);
todo[todo_end].done = 0;
strbuf_reset(&todo[todo_end].out);
todo_end = (todo_end + 1) % ARRAY_SIZE(todo);
@@ -333,7 +335,7 @@ static int grep_sha1(struct grep_opt *opt, const unsigned char *sha1,
#ifndef NO_PTHREADS
if (use_threads) {
- add_work(GREP_SOURCE_SHA1, pathbuf.buf, sha1);
+ add_work(opt, GREP_SOURCE_SHA1, pathbuf.buf, sha1);
strbuf_release(&pathbuf);
return 0;
} else
@@ -362,7 +364,7 @@ static int grep_file(struct grep_opt *opt, const char *filename)
#ifndef NO_PTHREADS
if (use_threads) {
- add_work(GREP_SOURCE_FILE, buf.buf, filename);
+ add_work(opt, GREP_SOURCE_FILE, buf.buf, filename);
strbuf_release(&buf);
return 0;
} else