git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: What's cooking in git.git (Jul 2016, #05; Wed, 13)

From
Johannes Schindelin <johannes.schindelin@gmx.de>
Date
Jul 14, 2016, 13:46 UTC
Message-ID
<alpine.DEB.2.20.1607141542250.6426@virtualbox>
In-Reply-To
<xmqqy4551nph.fsf@gitster.mtv.corp.google.com>
Hi Junio,
On Wed, 13 Jul 2016, Junio C Hamano wrote:
Show 32 quoted lines
> * dt/index-helper (2016-07-06) 21 commits
>  - index-helper: indexhelper.exitAfter config
>  - trace: measure where the time is spent in the index-heavy operations
>  - index-helper: optionally automatically run
>  - index-helper: autorun mode
>  - index-helper: don't run if already running
>  - index-helper: kill mode
>  - watchman: add a config option to enable the extension
>  - unpack-trees: preserve index extensions
>  - update-index: enable/disable watchman support
>  - index-helper: use watchman to avoid refreshing index with lstat()
>  - watchman: support watchman to reduce index refresh cost
>  - read-cache: add watchman 'WAMA' extension
>  - index-helper: log warnings
>  - index-helper: add --detach
>  - daemonize(): set a flag before exiting the main process
>  - index-helper: add --strict
>  - index-helper: new daemon for caching index and related stuff
>  - unix-socket.c: add stub implementation when unix sockets are not supported
>  - pkt-line: add gentle version of packet_write
>  - read-cache: allow to keep mmap'd memory after reading
>  - read-cache.c: fix constness of verify_hdr()
> 
>  A new "index-helper" daemon has been introduced to give newly
>  spawned Git process a quicker access to the data in the index, and
>  optionally interface with the watchman daemon to further reduce the
>  refresh cost.
> 
>  Will merge to 'next'.
> 
>  Is everybody happy with this version?
>  At v14.

I am trying to get back to working on the Windows support for this. I had something that compiled but did not quite manage to do the IPC on Monday evening, and have been scrambling with too much other stuff preventing me from getting back to that.

The reason I mention this is: I wanted to get it working first just to be able to determine whether there are possible improvements to the design that would make things much easier to port to non-Unix-sockets-capable systems.

Oh, and v14 has a bug that I reported already: http://article.gmane.org/gmane.comp.version-control.git/298949

Ciao, Dscho

Previous: Junio C HamanoNext: Junio C Hamano
Message 13 of 21 in “What's cooking in git.git (Jul 2016, #05; Wed, 13)”
  1. Junio C HamanoJul 13, 2016
  2. Stefan BellerJul 13, 2016
  3. Junio C HamanoJul 13, 2016
  4. Stefan BellerJul 13, 2016
  5. Junio C HamanoJul 13, 2016
  6. Junio C HamanoJul 13, 2016
  7. Jeff KingJul 13, 2016
  8. Duy NguyenJul 13, 2016
  9. Junio C HamanoJul 13, 2016
  10. Johannes SchindelinJul 14, 2016
  11. Johannes SchindelinJul 14, 2016
  12. Junio C HamanoJul 14, 2016
  13. Johannes SchindelinJul 14, 2016
  14. Junio C HamanoJul 14, 2016
  15. Lars SchneiderJul 16, 2016
  16. Eric WongJul 16, 2016
  17. Lars SchneiderJul 16, 2016
  18. list: avoid incompatibility with *BSD sys/queue.hEric Wong, Jul 17, 2016
  19. Eric SunshineJul 17, 2016
  20. Eric WongJul 17, 2016
  21. Lars SchneiderJul 18, 2016

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.