From: Michael Haggerty <mhagger@alum.mit.edu>
To: Junio C Hamano <gitster@pobox.com>
Cc: "Nguyễn Thái Ngọc Duy" <pclouds@gmail.com>,
"Stefan Beller" <sbeller@google.com>, "Jeff King" <peff@peff.net>,
"Ævar Arnfjörð Bjarmason" <avarab@gmail.com>,
"David Turner" <novalis@novalis.org>,
git@vger.kernel.org, "Michael Haggerty" <mhagger@alum.mit.edu>
Subject: [PATCH 17/23] get_packed_ref_cache(): assume "packed-refs" won't change while locked
Date: Wed, 17 May 2017 14:05:40 +0200 [thread overview]
Message-ID: <eedc8f85c8237515bc66584649b1e3f3d547049c.1495014840.git.mhagger@alum.mit.edu> (raw)
In-Reply-To: <cover.1495014840.git.mhagger@alum.mit.edu>
If we've got the "packed-refs" file locked, then it can't change;
there's no need to keep calling `stat_validity_check()` on it.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
---
refs/files-backend.c | 16 +++++++++++-----
1 file changed, 11 insertions(+), 5 deletions(-)
diff --git a/refs/files-backend.c b/refs/files-backend.c
index c549a8d956..a910c44f38 100644
--- a/refs/files-backend.c
+++ b/refs/files-backend.c
@@ -354,13 +354,18 @@ static void files_ref_path(struct files_ref_store *refs,
/*
* Get the packed_ref_cache for the specified files_ref_store,
- * creating it if necessary.
+ * creating and populating it if it hasn't been read before or if the
+ * file has been changed (according to its `validity` field) since it
+ * was last read. On the other hand, if we hold the lock, then assume
+ * that the file hasn't been changed out from under us, so skip the
+ * extra `stat()` call in `stat_validity_check()`.
*/
static struct packed_ref_cache *get_packed_ref_cache(struct files_ref_store *refs)
{
const char *packed_refs_file = files_packed_refs_path(refs);
if (refs->packed &&
+ !is_lock_file_locked(&refs->packlock) &&
!stat_validity_check(&refs->packed->validity, packed_refs_file))
clear_packed_ref_cache(refs);
@@ -1300,10 +1305,11 @@ static int lock_packed_refs(struct files_ref_store *refs, int flags)
flags, timeout_value) < 0)
return -1;
/*
- * Get the current packed-refs while holding the lock. If the
- * packed-refs file has been modified since we last read it,
- * this will automatically invalidate the cache and re-read
- * the packed-refs file.
+ * Get the current packed-refs while holding the lock. It is
+ * important that we call `get_packed_ref_cache()` before
+ * setting `packed_ref_cache->lock`, because otherwise the
+ * former will see that the file is locked and assume that the
+ * cache can't be stale.
*/
packed_ref_cache = get_packed_ref_cache(refs);
/* Increment the reference count to prevent it from being freed: */
--
2.11.0
next prev parent reply other threads:[~2017-05-17 12:06 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-05-17 12:05 [PATCH 00/23] Prepare to separate out a packed_ref_store Michael Haggerty
2017-05-17 12:05 ` [PATCH 01/23] t3600: clean up permissions test properly Michael Haggerty
2017-05-17 12:42 ` Jeff King
2017-05-17 14:01 ` Michael Haggerty
2017-05-18 4:10 ` Junio C Hamano
2017-05-19 3:37 ` Michael Haggerty
2017-05-17 12:05 ` [PATCH 02/23] refs.h: clarify docstring for the ref_transaction_update()-related fns Michael Haggerty
2017-05-17 16:46 ` Stefan Beller
2017-05-18 4:13 ` Junio C Hamano
2017-05-17 12:05 ` [PATCH 03/23] ref_iterator_begin_fn(): fix docstring Michael Haggerty
2017-05-17 12:05 ` [PATCH 04/23] prefix_ref_iterator: don't trim too much Michael Haggerty
2017-05-17 12:55 ` Jeff King
2017-05-17 14:11 ` Michael Haggerty
2017-05-17 14:22 ` Jeff King
2017-05-18 4:19 ` Junio C Hamano
2017-05-18 4:50 ` Michael Haggerty
2017-05-17 12:05 ` [PATCH 05/23] refs_ref_iterator_begin(): don't check prefixes redundantly Michael Haggerty
2017-05-17 12:59 ` Jeff King
2017-05-17 14:21 ` Michael Haggerty
2017-05-17 12:05 ` [PATCH 06/23] refs: use `size_t` indexes when iterating over ref transaction updates Michael Haggerty
2017-05-17 16:59 ` Stefan Beller
2017-05-18 4:55 ` Michael Haggerty
2017-05-17 12:05 ` [PATCH 07/23] ref_store: take `logmsg` parameter when deleting references Michael Haggerty
2017-05-17 13:12 ` Jeff King
2017-05-17 15:01 ` Michael Haggerty
2017-05-17 15:03 ` Jeff King
2017-05-17 12:05 ` [PATCH 08/23] lockfile: add a new method, is_lock_file_locked() Michael Haggerty
2017-05-17 13:12 ` Jeff King
2017-05-17 12:05 ` [PATCH 09/23] files-backend: move `lock` member to `files_ref_store` Michael Haggerty
2017-05-17 13:15 ` Jeff King
2017-05-17 15:49 ` Michael Haggerty
2017-05-17 12:05 ` [PATCH 10/23] files_ref_store: put the packed files lock directly in this struct Michael Haggerty
2017-05-17 13:17 ` Jeff King
2017-05-17 15:05 ` Michael Haggerty
2017-05-17 17:18 ` Stefan Beller
2017-05-18 0:18 ` Brandon Williams
2017-05-19 4:00 ` Michael Haggerty
2017-05-18 0:17 ` Brandon Williams
2017-05-18 1:11 ` Jeff King
2017-05-18 15:42 ` Brandon Williams
2017-05-17 12:05 ` [PATCH 11/23] files_transaction_cleanup(): new helper function Michael Haggerty
2017-05-17 13:19 ` Jeff King
2017-05-19 4:49 ` Michael Haggerty
2017-05-17 17:26 ` Stefan Beller
2017-05-19 4:42 ` Michael Haggerty
2017-05-17 12:05 ` [PATCH 12/23] ref_transaction_commit(): break into multiple functions Michael Haggerty
2017-05-17 17:44 ` Stefan Beller
2017-05-19 7:58 ` Michael Haggerty
2017-05-17 12:05 ` [PATCH 13/23] ref_update_reject_duplicates(): expose function to whole refs module Michael Haggerty
2017-05-17 12:05 ` [PATCH 14/23] ref_update_reject_duplicates(): use `size_t` rather than `int` Michael Haggerty
2017-05-17 12:05 ` [PATCH 15/23] ref_update_reject_duplicates(): add a sanity check Michael Haggerty
2017-05-17 12:05 ` [PATCH 16/23] should_pack_ref(): new function, extracted from `files_pack_refs()` Michael Haggerty
2017-05-17 12:05 ` Michael Haggerty [this message]
2017-05-17 17:57 ` [PATCH 17/23] get_packed_ref_cache(): assume "packed-refs" won't change while locked Stefan Beller
2017-05-18 1:15 ` Jeff King
2017-05-18 16:58 ` Stefan Beller
2017-05-17 12:05 ` [PATCH 18/23] read_packed_refs(): do more of the work of reading packed refs Michael Haggerty
2017-05-17 12:05 ` [PATCH 19/23] read_packed_refs(): report unexpected fopen() failures Michael Haggerty
2017-05-17 13:28 ` Jeff King
2017-05-17 15:27 ` Michael Haggerty
2017-05-18 4:57 ` Junio C Hamano
2017-05-18 5:08 ` Jeff King
2017-05-17 12:05 ` [PATCH 20/23] refs_ref_iterator_begin(): handle `GIT_REF_PARANOIA` Michael Haggerty
2017-05-17 13:29 ` Jeff King
2017-05-17 12:05 ` [PATCH 21/23] create_ref_entry(): remove `check_name` option Michael Haggerty
2017-05-17 12:05 ` [PATCH 22/23] ref-filter: limit traversal to prefix Michael Haggerty
2017-05-17 13:38 ` Jeff King
2017-05-19 10:02 ` Michael Haggerty
2017-05-17 12:05 ` [PATCH 23/23] cache_ref_iterator_begin(): avoid priming unneeded directories Michael Haggerty
2017-05-17 13:42 ` [PATCH 00/23] Prepare to separate out a packed_ref_store Jeff King
2017-05-17 18:14 ` Stefan Beller
2017-05-18 17:14 ` Johannes Sixt
2017-05-18 17:22 ` Jeff King
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
List information: http://vger.kernel.org/majordomo-info.html
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=eedc8f85c8237515bc66584649b1e3f3d547049c.1495014840.git.mhagger@alum.mit.edu \
--to=mhagger@alum.mit.edu \
--cc=avarab@gmail.com \
--cc=git@vger.kernel.org \
--cc=gitster@pobox.com \
--cc=novalis@novalis.org \
--cc=pclouds@gmail.com \
--cc=peff@peff.net \
--cc=sbeller@google.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.
Code repositories for project(s) associated with this public inbox
https://80x24.org/mirrors/git.git
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 read-only IMAP folder(s) and NNTP newsgroup(s).