From: David Turner <dturner@twopensource.com>
To: Johannes Schindelin <Johannes.Schindelin@gmx.de>,
Duy Nguyen <pclouds@gmail.com>
Cc: Git Mailing List <git@vger.kernel.org>
Subject: Re: [PATCH 18/19] index-helper: autorun
Date: Thu, 17 Mar 2016 14:31:07 -0400 [thread overview]
Message-ID: <1458239467.9385.21.camel@twopensource.com> (raw)
In-Reply-To: <alpine.DEB.2.20.1603171536420.4690@virtualbox>
On Thu, 2016-03-17 at 15:43 +0100, Johannes Schindelin wrote:
> Hi Duy,
>
> On Thu, 17 Mar 2016, Duy Nguyen wrote:
>
> > On Thu, Mar 17, 2016 at 1:27 AM, Johannes Schindelin
> > <Johannes.Schindelin@gmx.de> wrote:
> > > I am much more concerned about concurrent accesses and the
> > > communication
> > > between the Git processes and the index-helper. Writing to the
> > > .pid file
> > > sounds very fragile to me, in particular when multiple processes
> > > can poke
> > > the index-helper in succession and some readers are unaware that
> > > the index
> > > is being refreshed.
> >
> > It's not that bad.
>
> Well, the way I read the code it is possible that:
>
> 1. Git process 1 starts, reading the index
> 2. Git process 2 starts, poking the index-helper
> 3. The index-helper updates the .pid file (why not set a bit in the
> shared
> memory?) with a prefix "W"
> 4. Git process 2 reads the .pid file and waits for the "W" to go away
> (what if index-helper is not fast enough to write the "W"?)
> 5. Git process 1 access the index, happily oblivious that it is being
> updated and the data is in an inconsistent state
That's not quite how I understand it. It's more like MVCC. Writes to
the index go to a new index file. Index files are identified by their
SHA. Reads from the index go into a new shm, identified by SHA.
The "W" is set only once -- it just means "this index helper knows how
to talk to watchman". It's a compile-time option.
(I'm going to change this anyway when I switch to named pipes).
The watchman data is shared independently; if it's not ready in time
(whatever that means -- it's 1s in the current code), then read-cache
should fall back to brute-force checking every file.
> > We should have protection in place to deal with this and fall back
> > to
> > reading directly from file when things get suspicious.
>
> I really want to prevent that. I know of use cases where the index
> weighs
> 300MB, and falling back to reading it directly *really* hurts.
> > But I agree that sending UNIX signals (or PostMessage) is not
> > really
> > good communication.
>
> Yeah, I really would like two-way communication instead. Named pipes?
> They'd have the advantage that you could use the full path to the
> index as
> identifier.
>
> The way I read the current code, we would actually create a different
> shared memory every time the index changes because its checksum is
> part of
> the shared memory's "path"...
>
> Ciao,
> Dscho
next prev parent reply other threads:[~2016-03-17 18:32 UTC|newest]
Thread overview: 60+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-03-09 18:36 [PATCH 00/19] index-helper, watchman David Turner
2016-03-09 18:36 ` [PATCH 01/19] trace.c: add GIT_TRACE_PACK_STATS for pack usage statistics David Turner
2016-03-09 22:58 ` Junio C Hamano
2016-03-10 0:05 ` David Turner
2016-03-10 10:59 ` Duy Nguyen
2016-03-09 18:36 ` [PATCH 02/19] read-cache.c: fix constness of verify_hdr() David Turner
2016-03-09 18:36 ` [PATCH 03/19] read-cache: allow to keep mmap'd memory after reading David Turner
2016-03-09 23:02 ` Junio C Hamano
2016-03-10 0:09 ` David Turner
2016-03-09 18:36 ` [PATCH 04/19] index-helper: new daemon for caching index and related stuff David Turner
2016-03-09 23:09 ` Junio C Hamano
2016-03-09 23:21 ` Junio C Hamano
2016-03-10 0:01 ` David Turner
2016-03-10 11:17 ` Duy Nguyen
2016-03-10 20:22 ` David Turner
2016-03-11 1:11 ` Duy Nguyen
2016-03-10 0:18 ` David Turner
2016-03-15 11:56 ` Duy Nguyen
2016-03-15 15:56 ` Junio C Hamano
2016-03-15 11:52 ` Duy Nguyen
2016-03-09 18:36 ` [PATCH 05/19] trace.c: add GIT_TRACE_INDEX_STATS for index statistics David Turner
2016-03-09 18:36 ` [PATCH 06/19] index-helper: add --strict David Turner
2016-03-09 18:36 ` [PATCH 07/19] daemonize(): set a flag before exiting the main process David Turner
2016-03-09 18:36 ` [PATCH 08/19] index-helper: add --detach David Turner
2016-03-09 18:36 ` [PATCH 09/19] index-helper: add Windows support David Turner
2016-03-16 11:42 ` Duy Nguyen
2016-03-17 12:18 ` Johannes Schindelin
2016-03-17 12:59 ` Duy Nguyen
2016-03-09 18:36 ` [PATCH 10/19] read-cache: add watchman 'WAMA' extension David Turner
2016-03-09 18:36 ` [PATCH 11/19] Add watchman support to reduce index refresh cost David Turner
2016-03-09 18:36 ` [PATCH 12/19] read-cache: allow index-helper to prepare shm before git reads it David Turner
2016-03-09 18:36 ` [PATCH 13/19] index-helper: use watchman to avoid refreshing index with lstat() David Turner
2016-03-09 18:36 ` [PATCH 14/19] update-index: enable/disable watchman support David Turner
2016-03-09 18:36 ` [PATCH 15/19] unpack-trees: preserve index extensions David Turner
2016-03-09 18:36 ` [PATCH 16/19] index-helper: rewrite pidfile after daemonizing David Turner
2016-03-09 18:36 ` [PATCH 17/19] index-helper: process management David Turner
2016-03-09 18:36 ` [PATCH 18/19] index-helper: autorun David Turner
2016-03-15 12:12 ` Duy Nguyen
2016-03-15 14:26 ` Johannes Schindelin
2016-03-16 11:37 ` Duy Nguyen
2016-03-16 18:11 ` David Turner
2016-03-16 18:27 ` Johannes Schindelin
2016-03-17 13:02 ` Duy Nguyen
2016-03-17 14:43 ` Johannes Schindelin
2016-03-17 18:31 ` David Turner [this message]
2016-03-18 0:50 ` Duy Nguyen
2016-03-18 7:14 ` Johannes Schindelin
2016-03-18 7:44 ` Duy Nguyen
2016-03-18 17:22 ` David Turner
2016-03-18 23:09 ` Duy Nguyen
2016-03-18 7:17 ` Johannes Schindelin
2016-03-18 7:34 ` Duy Nguyen
2016-03-18 15:57 ` Johannes Schindelin
2016-03-09 18:36 ` [PATCH 19/19] hack: watchman/untracked cache mashup David Turner
2016-03-15 12:31 ` Duy Nguyen
2016-03-17 0:56 ` David Turner
2016-03-17 13:06 ` Duy Nguyen
2016-03-17 18:08 ` David Turner
2016-03-29 17:09 ` [PATCH 00/19] index-helper, watchman Torsten Bögershausen
2016-03-29 21:51 ` David Turner
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=1458239467.9385.21.camel@twopensource.com \
--to=dturner@twopensource.com \
--cc=Johannes.Schindelin@gmx.de \
--cc=git@vger.kernel.org \
--cc=pclouds@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 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).