git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: Duy Nguyen <pclouds@gmail.com>
To: Thomas Gummerer <t.gummerer@gmail.com>
Cc: Git Mailing List <git@vger.kernel.org>,
	Junio C Hamano <gitster@pobox.com>,
	Elijah Newren <newren@gmail.com>
Subject: Re: [PATCH 4/8] read-cache: add invalidate parameter to remove_marked_cache_entries
Date: Mon, 10 Dec 2018 17:08:56 +0100	[thread overview]
Message-ID: <CACsJy8AiQvu8W4=2HLKMdg+n2HiDrcLvKPRurKvziXaJdqefRg@mail.gmail.com> (raw)
In-Reply-To: <20181209200449.16342-5-t.gummerer@gmail.com>

On Sun, Dec 9, 2018 at 9:05 PM Thomas Gummerer <t.gummerer@gmail.com> wrote:
>
> When marking cache entries for removal, and later removing them all at
> once using 'remove_marked_cache_entries()', cache entries currently
> have to be invalidated manually in the cache tree and in the untracked
> cache.
>
> Add an invalidate flag to the function.  With the flag set, the
> function will take care of invalidating the path in the cache tree and
> in the untracked cache.
>
> This will be useful in a subsequent commit.
>
> Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
> ---
>
> For the two current callsites, unpack-trees seems to do this
> invalidation itself internally.

I'm still a bit scared of this invalidation business in unpack-trees.
The thing is, we handle two separate index_state there, src_index and
result and invalidation has to be done on the right one (because index
extensions are on src_index until the very end of unpack-trees;
invalidating on 'result' would be no-op and wrong).
remove_marked_cache_entries() seems to be called on 'result' while
invalidate_ce_path() is on src_index, hm....

> I don't quite understand why we don't
> need it in split-index mode though.  I assume it's because the cache
> tree in the main index would already have been invalidated?  I didn't
> have much time to dig, but couldn't produce any failures with it
> either, so I assume not invalidating paths is the right thing to do
> here.

Yeah I think it's because cache-tree and untracked cache are already
properly invalidated. This merge base thingy is done when we load the
index files up, not when we write them down. The "front" index may
record that a few paths in the base index are no longer valid and need
to be deleted. But untracked cache and cache-tree both should have
recorded that same info when these paths are marked for delete at
index write time.
-- 
Duy

  reply	other threads:[~2018-12-10 16:09 UTC|newest]

Thread overview: 115+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-12-09 20:04 [PATCH 0/8] introduce no-overlay and cached mode in git checkout Thomas Gummerer
2018-12-09 20:04 ` [PATCH 1/8] move worktree tests to t24* Thomas Gummerer
2018-12-10  3:48   ` Junio C Hamano
2018-12-10 15:32   ` Duy Nguyen
2018-12-11 21:50     ` Thomas Gummerer
2018-12-12 13:26       ` Eric Sunshine
2018-12-12 17:07         ` Duy Nguyen
2018-12-09 20:04 ` [PATCH 2/8] entry: factor out unlink_entry function Thomas Gummerer
2018-12-10 15:49   ` Duy Nguyen
2018-12-10 17:23     ` Elijah Newren
2018-12-10 17:27       ` Duy Nguyen
2018-12-11  2:23     ` Junio C Hamano
2018-12-20 13:36     ` Thomas Gummerer
2018-12-09 20:04 ` [PATCH 3/8] entry: support CE_WT_REMOVE flag in checkout_entry Thomas Gummerer
2018-12-10 15:58   ` Duy Nguyen
2018-12-11  2:28     ` Junio C Hamano
2018-12-12  6:16       ` Duy Nguyen
2018-12-12  7:36         ` Junio C Hamano
2018-12-10 17:49   ` Elijah Newren
2018-12-11 22:00     ` Thomas Gummerer
2018-12-09 20:04 ` [PATCH 4/8] read-cache: add invalidate parameter to remove_marked_cache_entries Thomas Gummerer
2018-12-10 16:08   ` Duy Nguyen [this message]
2018-12-10 18:09     ` Elijah Newren
2018-12-10 18:19       ` Duy Nguyen
2018-12-10 18:25         ` Elijah Newren
2018-12-10 18:33           ` Duy Nguyen
2018-12-10 18:47             ` Elijah Newren
2018-12-11 21:59       ` Thomas Gummerer
2018-12-11  2:42     ` Junio C Hamano
2018-12-09 20:04 ` [PATCH 5/8] checkout: introduce --{,no-}overlay option Thomas Gummerer
2018-12-10 16:42   ` Duy Nguyen
2018-12-11 22:42     ` Thomas Gummerer
2018-12-10 18:19   ` Elijah Newren
2018-12-11  3:07     ` Junio C Hamano
2018-12-11  6:04       ` Elijah Newren
2018-12-11 22:07       ` Thomas Gummerer
2018-12-09 20:04 ` [PATCH 6/8] checkout: add --cached option Thomas Gummerer
2018-12-10 16:49   ` Duy Nguyen
2018-12-11  3:13     ` Junio C Hamano
2018-12-11  6:12       ` Elijah Newren
2018-12-11 19:23         ` Duy Nguyen
2019-01-31  5:54           ` Duy Nguyen
2019-01-31 19:05             ` Junio C Hamano
2019-02-01  6:48               ` Duy Nguyen
2019-02-01 17:57                 ` Junio C Hamano
2019-02-02 10:57                   ` Duy Nguyen
2019-02-19  4:20                   ` Duy Nguyen
2019-02-19 14:42                     ` Elijah Newren
2019-02-19 14:57                       ` Duy Nguyen
2019-02-19 19:07                       ` Junio C Hamano
2019-02-19 22:24                         ` Elijah Newren
2019-02-19 22:36                           ` Junio C Hamano
2019-02-19 23:00                             ` Elijah Newren
2019-02-20  1:53                             ` Duy Nguyen
2019-02-19 19:02                     ` Junio C Hamano
2019-02-19 19:10                       ` Junio C Hamano
2019-02-19 22:04                         ` Elijah Newren
2019-02-19 22:29                           ` Junio C Hamano
2019-02-20  2:32                             ` Duy Nguyen
2019-02-20  3:52                           ` Duy Nguyen
2019-02-20  2:19                         ` Duy Nguyen
2019-02-19 22:13                       ` Elijah Newren
2019-02-19 22:33                         ` Junio C Hamano
2018-12-10 18:42   ` Elijah Newren
2018-12-11 22:18     ` Thomas Gummerer
2018-12-09 20:04 ` [PATCH 7/8] checkout: allow ignoring unmatched pathspec Thomas Gummerer
2018-12-10 16:51   ` Duy Nguyen
2018-12-11 22:23     ` Thomas Gummerer
2018-12-10 20:25   ` Elijah Newren
2018-12-11 22:36     ` Thomas Gummerer
2018-12-09 20:04 ` [PATCH 8/8] stash: use git checkout --no-overlay Thomas Gummerer
2018-12-10 20:26   ` Elijah Newren
2018-12-10  3:47 ` [PATCH 0/8] introduce no-overlay and cached mode in git checkout Junio C Hamano
2018-12-20  8:43   ` Thomas Gummerer
2018-12-10 17:06 ` Duy Nguyen
2018-12-10 17:18 ` Elijah Newren
2018-12-10 18:37   ` Duy Nguyen
2018-12-11 22:52   ` Thomas Gummerer
2018-12-12  7:28     ` Junio C Hamano
2018-12-20 13:48 ` [PATCH v2 0/8] introduce no-overlay " Thomas Gummerer
2018-12-20 13:48   ` [PATCH v2 1/8] move worktree tests to t24* Thomas Gummerer
2018-12-20 13:48   ` [PATCH v2 2/8] entry: factor out unlink_entry function Thomas Gummerer
2018-12-20 13:48   ` [PATCH v2 3/8] entry: support CE_WT_REMOVE flag in checkout_entry Thomas Gummerer
2018-12-20 13:48   ` [PATCH v2 4/8] read-cache: add invalidate parameter to remove_marked_cache_entries Thomas Gummerer
2018-12-20 13:48   ` [PATCH v2 5/8] checkout: clarify comment Thomas Gummerer
2018-12-20 13:48   ` [PATCH v2 6/8] checkout: factor out mark_cache_entry_for_checkout function Thomas Gummerer
2018-12-20 13:48   ` [PATCH v2 7/8] checkout: introduce --{,no-}overlay option Thomas Gummerer
2018-12-23  8:05     ` Duy Nguyen
2018-12-23  9:44       ` Eric Sunshine
2019-01-06 18:18         ` Thomas Gummerer
2018-12-20 13:48   ` [PATCH v2 8/8] checkout: introduce checkout.overlayMode config Thomas Gummerer
2019-01-02 23:39     ` Junio C Hamano
2019-01-06 18:32       ` Thomas Gummerer
2019-01-07 17:00         ` Junio C Hamano
2019-01-08 21:52   ` [PATCH v3 0/8] introduce no-overlay mode in git checkout Thomas Gummerer
2019-01-08 21:52     ` [PATCH v3 1/8] move worktree tests to t24* Thomas Gummerer
2019-01-08 21:52     ` [PATCH v3 2/8] entry: factor out unlink_entry function Thomas Gummerer
2019-01-08 21:52     ` [PATCH v3 3/8] entry: support CE_WT_REMOVE flag in checkout_entry Thomas Gummerer
2019-01-08 21:52     ` [PATCH v3 4/8] read-cache: add invalidate parameter to remove_marked_cache_entries Thomas Gummerer
2019-01-08 21:52     ` [PATCH v3 5/8] checkout: clarify comment Thomas Gummerer
2019-01-08 21:52     ` [PATCH v3 6/8] checkout: factor out mark_cache_entry_for_checkout function Thomas Gummerer
2019-01-08 21:52     ` [PATCH v3 7/8] checkout: introduce --{,no-}overlay option Thomas Gummerer
2019-01-22 23:53       ` Jonathan Nieder
2019-01-23 19:05         ` Junio C Hamano
2019-01-23 20:21         ` Thomas Gummerer
2019-01-23 20:47           ` Jonathan Nieder
2019-01-24 22:08             ` Thomas Gummerer
2019-01-23 21:08           ` Junio C Hamano
2019-01-24  1:12             ` Jonathan Nieder
2019-01-24 22:02               ` Thomas Gummerer
2019-01-24 23:02               ` Junio C Hamano
2019-01-25  2:26                 ` Jonathan Nieder
2019-01-25  9:24                   ` Duy Nguyen
2019-02-09 18:54               ` Philip Oakley
2019-01-08 21:52     ` [PATCH v3 8/8] checkout: introduce checkout.overlayMode config Thomas Gummerer

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='CACsJy8AiQvu8W4=2HLKMdg+n2HiDrcLvKPRurKvziXaJdqefRg@mail.gmail.com' \
    --to=pclouds@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=newren@gmail.com \
    --cc=t.gummerer@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.
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).