From: Duy Nguyen <email@example.com> To: Thomas Gummerer <firstname.lastname@example.org> Cc: Git Mailing List <email@example.com>, Thomas Rast <firstname.lastname@example.org>, Michael Haggerty <email@example.com>, Junio C Hamano <firstname.lastname@example.org>, Robin Rosenberg <email@example.com> Subject: Re: [PATCH 13/22] documentation: add documentation of the index-v5 file format Date: Thu, 11 Jul 2013 19:50:38 +0700 [thread overview] Message-ID: <CACsJy8B+y1pcnfNy_kuMXBZ5J72SR71TxE0RtV3a41HgnuDGaA@mail.gmail.com> (raw) In-Reply-To: <firstname.lastname@example.org> On Thu, Jul 11, 2013 at 7:26 PM, Thomas Gummerer <email@example.com> wrote: > Duy Nguyen <firstname.lastname@example.org> writes: > >> On Thu, Jul 11, 2013 at 6:39 PM, Thomas Gummerer <email@example.com> wrote: >>>> Question about the possibility of updating index file directly. If git >>>> updates a few fields of an entry (but not entrycrc yet) and crashes, >>>> the entry would become corrupt because its entrycrc does not match the >>>> content. What do we do? Do we need to save a copy of the entry >>>> somewhere in the index file (maybe in the conflict data section), so >>>> that the reader can recover the index? Losing the index because of >>>> bugs is big deal in my opinion. pre-v5 never faces this because we >>>> keep the original copy til the end. >>>> >>>> Maybe entrycrc should not cover stat fields and statcrc. It would make >>>> refreshing safer. If the above happens during refresh, only statcrc is >>>> corrupt and we can just refresh the entry. entrycrc still says the >>>> other fields are good (and they are). >>> >>> The original idea was to change the lock-file for partial writing to >>> make it work for this case. The exact structure of the file still has >>> to be defined, but generally it would be done in the following steps: >>> >>> 1. Write the changed entry to the lock-file >>> 2. Change the entry in the index >>> 3. If we succeed delete the lock-file (commit the transaction) >>> >>> If git crashes, and leaves the index corrupted, we can recover the >>> information from the lock-file and write the new information to the >>> index file and then delete the lock-file. >> >> Ah makes sense. Still concerned about refreshing though. Updated files >> are usually few while refreshed files could be a lot more, increasing >> the cost at #1. > > Any idea how common refreshing a big part of the cache is? No, probably not common. Anyone who does "find|xargs touch" deserves to be punished. Files can be edited, then reverted by an editor, but there should not be many of those. The only sensible case is "git checkout <path>" with lots of modified files. But that can't happen often. > If it's not to common, I'd prefer to leave the stat data and stat crc in the > entrycrc, as we can inform the user if something is wrong with the > index, be it from git failing, or from disk corruption. > > On the other hand if refresh_cache is relatively common and usually > changes a big part of the index we should leave them out, as git can > still run correctly with incorrect stat data, but takes a little longer, > because it may have to check the file contents. That will be trade-off > to make here. -- Duy
next prev parent reply other threads:[~2013-07-11 12:51 UTC|newest] Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top 2013-07-07 8:11 [PATCH 00/22] Index v5 Thomas Gummerer 2013-07-07 8:11 ` [PATCH 01/22] t2104: Don't fail for index versions other than  Thomas Gummerer 2013-07-07 8:11 ` [PATCH 02/22] read-cache: split index file version specific functionality Thomas Gummerer 2013-07-07 8:11 ` [PATCH 03/22] read-cache: move index v2 specific functions to their own file Thomas Gummerer 2013-07-07 8:11 ` [PATCH 04/22] read-cache: Re-read index if index file changed Thomas Gummerer 2013-07-07 8:11 ` [PATCH 05/22] read-cache: add index reading api Thomas Gummerer 2013-07-08 2:01 ` Duy Nguyen 2013-07-08 11:40 ` Thomas Gummerer 2013-07-08 2:19 ` Duy Nguyen 2013-07-08 11:20 ` Thomas Gummerer 2013-07-08 12:45 ` Duy Nguyen 2013-07-08 13:37 ` Thomas Gummerer 2013-07-08 20:54 ` [PATCH 5.5/22] Add documentation for the index api Thomas Gummerer 2013-07-09 15:42 ` Duy Nguyen 2013-07-09 20:10 ` Thomas Gummerer 2013-07-10 5:28 ` Duy Nguyen 2013-07-11 11:30 ` Thomas Gummerer 2013-07-11 11:42 ` Duy Nguyen 2013-07-11 12:27 ` Duy Nguyen 2013-07-08 16:36 ` [PATCH 05/22] read-cache: add index reading api Junio C Hamano 2013-07-08 20:10 ` Thomas Gummerer 2013-07-08 23:09 ` Junio C Hamano 2013-07-09 20:13 ` Thomas Gummerer 2013-07-07 8:11 ` [PATCH 06/22] make sure partially read index is not changed Thomas Gummerer 2013-07-08 16:31 ` Junio C Hamano 2013-07-08 18:33 ` Thomas Gummerer 2013-07-07 8:11 ` [PATCH 07/22] dir.c: use index api Thomas Gummerer 2013-07-07 8:11 ` [PATCH 08/22] tree.c: " Thomas Gummerer 2013-07-07 8:11 ` [PATCH 09/22] name-hash.c: " Thomas Gummerer 2013-07-07 8:11 ` [PATCH 10/22] grep.c: Use " Thomas Gummerer 2013-07-07 8:11 ` [PATCH 11/22] ls-files.c: use the " Thomas Gummerer 2013-07-07 8:11 ` [PATCH 12/22] read-cache: make read_blob_data_from_index use " Thomas Gummerer 2013-07-07 8:11 ` [PATCH 13/22] documentation: add documentation of the index-v5 file format Thomas Gummerer 2013-07-11 10:39 ` Duy Nguyen 2013-07-11 11:39 ` Thomas Gummerer 2013-07-11 11:47 ` Duy Nguyen 2013-07-11 12:26 ` Thomas Gummerer 2013-07-11 12:50 ` Duy Nguyen [this message] 2013-07-07 8:11 ` [PATCH 14/22] read-cache: make in-memory format aware of stat_crc Thomas Gummerer 2013-07-07 8:11 ` [PATCH 15/22] read-cache: read index-v5 Thomas Gummerer 2013-07-07 20:18 ` Eric Sunshine 2013-07-08 11:40 ` Thomas Gummerer 2013-07-07 8:11 ` [PATCH 16/22] read-cache: read resolve-undo data Thomas Gummerer 2013-07-07 8:11 ` [PATCH 17/22] read-cache: read cache-tree in index-v5 Thomas Gummerer 2013-07-07 20:41 ` Eric Sunshine 2013-07-07 8:11 ` [PATCH 18/22] read-cache: write index-v5 Thomas Gummerer 2013-07-07 20:43 ` Eric Sunshine 2013-07-07 8:11 ` [PATCH 19/22] read-cache: write index-v5 cache-tree data Thomas Gummerer 2013-07-07 8:11 ` [PATCH 20/22] read-cache: write resolve-undo data for index-v5 Thomas Gummerer 2013-07-07 8:11 ` [PATCH 21/22] update-index.c: rewrite index when index-version is given Thomas Gummerer 2013-07-07 8:12 ` [PATCH 22/22] p0003-index.sh: add perf test for the index formats 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=CACsJy8B+y1pcnfNy_kuMXBZ5J72SR71TxE0RtV3a41HgnuDGaA@mail.gmail.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --subject='Re: [PATCH 13/22] documentation: add documentation of the index-v5 file format' \ /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
Code repositories for project(s) associated with this 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).