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

Re: [PATCH 1/4] run-command: add pipe_output to run_processes_parallel

From
Junio C Hamano <gitster@pobox.com>
Date
Sep 27, 2022, 04:45 UTC
Message-ID
<xmqqpmfh8lxw.fsf@gitster.g>
In-Reply-To
<CAFySSZA=tThoHdTY7+bMStvC=xeeyMiv4aVDYt-eNW2mQE10qg@mail.gmail.com>
Calvin Wan <calvinwan@google.com> writes:
Show 8 quoted lines
> You are correct that storing unbounded output doesn't seem like a good
> idea. One idea I have is to parse output during the periodic collection rather
> than waiting till the end. The other idea I had was to add another
> "git status --porcelain" option that would only output the necessary
> pieces of information so we wouldn't have to bother with worrying about
> unbounded output.
>
> Any other thoughts as to how I can workaround this?
I wonder if you can arrange not to let them make unbounded progress?

In order to run diff-files with path A B C D E ... where B and D are submodules and others are not submodules, you do not have to run and finish comparison for B and D before you can do the comparison for other paths, in order to preserve the proper output order. You can start child task for B and D and arrange so that they will run for any other submodules, and then you

 - run comparison for A.  The child task for B and D may be running
   and starting to talk back to you, in which case their write may
   get stuck waiting for you to read from them, but that is OK, as
   you will read from them shortly.
 - wait for the child task for B.  This is done by reading from the
   pipe connected to it and waiting for its death synchronously.
   The child task for D is still running and may be making progress,
   but you are not obligated to read its output to the end.  You can
   postpone reading to conserve memory and it will fill the pipe and
   stall automatically.  Then accept the result for B.
 - run comparison for C.
 - wait for the child task for D.  You may have some data read
   already while dealing with B, but you may still have some reading
   to do at this point.  Let it finish synchronously.
 - run comparison for E.
etc., perhaps?
Previous: Calvin WanNext: Calvin Wan
Message 8 of 32 in “submodule: parallelize status”
  1. 0/4 submodule: parallelize statusCalvin Wan, Sep 22, 2022
  2. 1/4 run-command: add pipe_output to run_processes_parallelCalvin Wan, Sep 22, 2022
  3. Ævar Arnfjörð BjarmasonSep 23, 2022
  4. Calvin WanSep 26, 2022
  5. Ævar Arnfjörð BjarmasonSep 27, 2022
  6. Junio C HamanoSep 23, 2022
  7. Calvin WanSep 26, 2022
  8. Junio C HamanoSep 27, 2022
  9. Calvin WanSep 27, 2022
  10. Junio C HamanoSep 27, 2022
  11. Ævar Arnfjörð BjarmasonSep 27, 2022
  12. Calvin WanSep 27, 2022
  13. Ævar Arnfjörð BjarmasonSep 27, 2022
  14. Calvin WanSep 27, 2022
  15. Ævar Arnfjörð BjarmasonSep 28, 2022
  16. Calvin WanSep 29, 2022
  17. 2/4 submodule: move status parsing into functionCalvin Wan, Sep 22, 2022
  18. 3/4 diff-lib: refactor functionsCalvin Wan, Sep 22, 2022
  19. Junio C HamanoSep 23, 2022
  20. Calvin WanSep 26, 2022
  21. 4/4 diff-lib: parallelize run_diff_files for submodulesCalvin Wan, Sep 22, 2022
  22. Ævar Arnfjörð BjarmasonSep 23, 2022
  23. Junio C HamanoSep 24, 2022
  24. Calvin WanSep 26, 2022
  25. Junio C HamanoSep 23, 2022
  26. Calvin WanSep 26, 2022
  27. Phillip WoodSep 25, 2022
  28. Junio C HamanoSep 26, 2022
  29. Calvin WanSep 26, 2022
  30. Emily ShafferSep 27, 2022
  31. Junio C HamanoSep 23, 2022
  32. Calvin WanSep 26, 2022

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.