From 5827a03545663f6d6b491a35edb313900608568b Mon Sep 17 00:00:00 2001 From: Jeff King Date: Thu, 13 Oct 2016 12:53:44 -0400 Subject: [PATCH] fetch: use "quick" has_sha1_file for tag following When we auto-follow tags in a fetch, we look at all of the tags advertised by the remote and fetch ones where we don't already have the tag, but we do have the object it peels to. This involves a lot of calls to has_sha1_file(), some of which we can reasonably expect to fail. Since 45e8a74 (has_sha1_file: re-check pack directory before giving up, 2013-08-30), this may cause many calls to reprepare_packed_git(), which is potentially expensive. This has gone unnoticed for several years because it requires a fairly unique setup to matter: 1. You need to have a lot of packs on the client side to make reprepare_packed_git() expensive (the most expensive part is finding duplicates in an unsorted list, which is currently quadratic). 2. You need a large number of tag refs on the server side that are candidates for auto-following (i.e., that the client doesn't have). Each one triggers a re-read of the pack directory. 3. Under normal circumstances, the client would auto-follow those tags and after one large fetch, (2) would no longer be true. But if those tags point to history which is disconnected from what the client otherwise fetches, then it will never auto-follow, and those candidates will impact it on every fetch. So when all three are true, each fetch pays an extra O(nr_tags * nr_packs^2) cost, mostly in string comparisons on the pack names. This was exacerbated by 47bf4b0 (prepare_packed_git_one: refactor duplicate-pack check, 2014-06-30) which uses a slightly more expensive string check, under the assumption that the duplicate check doesn't happen very often (and it shouldn't; the real problem here is how often we are calling reprepare_packed_git()). This patch teaches fetch to use HAS_SHA1_QUICK to sacrifice accuracy for speed, in cases where we might be racy with a simultaneous repack. This is similar to the fix in 0eeb077 (index-pack: avoid excessive re-reading of pack directory, 2015-06-09). As with that case, it's OK for has_sha1_file() occasionally say "no I don't have it" when we do, because the worst case is not a corruption, but simply that we may fail to auto-follow a tag that points to it. Here are results from the included perf script, which sets up a situation similar to the one described above: Test HEAD^ HEAD ---------------------------------------------------------- 5550.4: fetch 11.21(10.42+0.78) 0.08(0.04+0.02) -99.3% Reported-by: Vegard Nossum Signed-off-by: Jeff King Signed-off-by: Junio C Hamano --- builtin/fetch.c | 11 +++-- cache.h | 1 + sha1_file.c | 5 ++ t/perf/p5550-fetch-tags.sh | 99 ++++++++++++++++++++++++++++++++++++++ 4 files changed, 112 insertions(+), 4 deletions(-) create mode 100755 t/perf/p5550-fetch-tags.sh diff --git a/builtin/fetch.c b/builtin/fetch.c index f896aa1f88..3e1a2668c6 100644 --- a/builtin/fetch.c +++ b/builtin/fetch.c @@ -232,9 +232,10 @@ static void find_non_local_tags(struct transport *transport, * as one to ignore by setting util to NULL. */ if (ends_with(ref->name, "^{}")) { - if (item && !has_object_file(&ref->old_oid) && + if (item && + !has_object_file_with_flags(&ref->old_oid, HAS_SHA1_QUICK) && !will_fetch(head, ref->old_oid.hash) && - !has_sha1_file(item->util) && + !has_sha1_file_with_flags(item->util, HAS_SHA1_QUICK) && !will_fetch(head, item->util)) item->util = NULL; item = NULL; @@ -247,7 +248,8 @@ static void find_non_local_tags(struct transport *transport, * to check if it is a lightweight tag that we want to * fetch. */ - if (item && !has_sha1_file(item->util) && + if (item && + !has_sha1_file_with_flags(item->util, HAS_SHA1_QUICK) && !will_fetch(head, item->util)) item->util = NULL; @@ -267,7 +269,8 @@ static void find_non_local_tags(struct transport *transport, * We may have a final lightweight tag that needs to be * checked to see if it needs fetching. */ - if (item && !has_sha1_file(item->util) && + if (item && + !has_sha1_file_with_flags(item->util, HAS_SHA1_QUICK) && !will_fetch(head, item->util)) item->util = NULL; diff --git a/cache.h b/cache.h index 1fd2d5dc3e..68eb8a6415 100644 --- a/cache.h +++ b/cache.h @@ -1116,6 +1116,7 @@ static inline int has_sha1_file(const unsigned char *sha1) /* Same as the above, except for struct object_id. */ extern int has_object_file(const struct object_id *oid); +extern int has_object_file_with_flags(const struct object_id *oid, int flags); /* * Return true iff an alternate object database has a loose object diff --git a/sha1_file.c b/sha1_file.c index cb571ac6e8..2fdafa6a99 100644 --- a/sha1_file.c +++ b/sha1_file.c @@ -3239,6 +3239,11 @@ int has_object_file(const struct object_id *oid) return has_sha1_file(oid->hash); } +int has_object_file_with_flags(const struct object_id *oid, int flags) +{ + return has_sha1_file_with_flags(oid->hash, flags); +} + static void check_tree(const void *buf, size_t size) { struct tree_desc desc; diff --git a/t/perf/p5550-fetch-tags.sh b/t/perf/p5550-fetch-tags.sh new file mode 100755 index 0000000000..a5dc39f86a --- /dev/null +++ b/t/perf/p5550-fetch-tags.sh @@ -0,0 +1,99 @@ +#!/bin/sh + +test_description='performance of tag-following with many tags + +This tests a fairly pathological case, so rather than rely on a real-world +case, we will construct our own repository. The situation is roughly as +follows. + +The parent repository has a large number of tags which are disconnected from +the rest of history. That makes them candidates for tag-following, but we never +actually grab them (and thus they will impact each subsequent fetch). + +The child repository is a clone of parent, without the tags, and is at least +one commit behind the parent (meaning that we will fetch one object and then +examine the tags to see if they need followed). Furthermore, it has a large +number of packs. + +The exact values of "large" here are somewhat arbitrary; I picked values that +start to show a noticeable performance problem on my machine, but without +taking too long to set up and run the tests. +' +. ./perf-lib.sh + +# make a long nonsense history on branch $1, consisting of $2 commits, each +# with a unique file pointing to the blob at $2. +create_history () { + perl -le ' + my ($branch, $n, $blob) = @ARGV; + for (1..$n) { + print "commit refs/heads/$branch"; + print "committer nobody now"; + print "data 4"; + print "foo"; + print "M 100644 $blob $_"; + } + ' "$@" | + git fast-import --date-format=now +} + +# make a series of tags, one per commit in the revision range given by $@ +create_tags () { + git rev-list "$@" | + perl -lne 'print "create refs/tags/$. $_"' | + git update-ref --stdin +} + +# create $1 nonsense packs, each with a single blob +create_packs () { + perl -le ' + my ($n) = @ARGV; + for (1..$n) { + print "blob"; + print "data <