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

Re: [PATCH] enable core.fsyncObjectFiles by default

From
CMChris Mason <clm@fb.com>
Date
Jan 21, 2018, 21:32 UTC
Message-ID
<f797983e-1890-d84c-c3c4-87904a0a8135@fb.com>
In-Reply-To
<20180118162721.GA26078@lst.de>
On 01/18/2018 11:27 AM, Christoph Hellwig wrote:
Show 16 quoted lines
> [adding Chris to the Cc list - this is about the awful ext3 data=ordered
> behavior of syncing the whole file system data and metadata on each
> fsync]
> 
> On Wed, Jan 17, 2018 at 03:57:53PM -0800, Linus Torvalds wrote:
>> On Wed, Jan 17, 2018 at 3:52 PM, Theodore Ts'o <tytso@mit.edu> wrote:
>>>
>>> Well, let's be fair; this is something *ext3* got wrong, and it was
>>> the default file system back them.
>>
>> I'm pretty sure reiserfs and btrfs did too..
> 
> I'm pretty sure btrfs never did, and reiserfs at least looks like
> it currently doesn't but I'd have to dig into history to check if
> it ever did.
>

Christoph has this right, btrfs only fsyncs the one file that you've asked for, and unrelated data/metadata won't be included.

We've seen big fsync stalls on ext4 caused by data=ordered, so it's still possible to trigger on ext4, but much better than ext3.

I do share Ted's concern about the perf impact of the fsyncs on tons of loose files, but the defaults should be safety first.

-chris
Previous: Jeff KingNext: Ævar Arnfjörð Bjarmason
Message 24 of 52 in “enable core.fsyncObjectFiles by default”
  1. enable core.fsyncObjectFiles by defaultChristoph Hellwig, Jan 17, 2018
  2. Junio C HamanoJan 17, 2018
  3. Christoph HellwigJan 17, 2018
  4. Andreas SchwabJan 17, 2018
  5. Matthew WilcoxJan 17, 2018
  6. Christoph HellwigJan 17, 2018
  7. Ævar Arnfjörð BjarmasonJan 17, 2018
  8. Linus TorvaldsJan 17, 2018
  9. Linus TorvaldsJan 17, 2018
  10. Ævar Arnfjörð BjarmasonJan 17, 2018
  11. Linus TorvaldsJan 17, 2018
  12. Theodore Ts'oJan 17, 2018
  13. Linus TorvaldsJan 17, 2018
  14. Christoph HellwigJan 18, 2018
  15. Junio C HamanoJan 19, 2018
  16. Theodore Ts'oJan 20, 2018
  17. Junio C HamanoJan 20, 2018
  18. Ævar Arnfjörð BjarmasonJan 22, 2018
  19. Theodore Ts'oJan 22, 2018
  20. Jeff KingJan 23, 2018
  21. Theodore Ts'oJan 23, 2018
  22. Jeff KingJan 23, 2018
  23. Jeff KingJan 23, 2018
  24. Chris MasonJan 21, 2018
  25. Ævar Arnfjörð BjarmasonSep 17, 2020
  26. 2/2 core.fsyncObjectFiles: make the docs less flippantÆvar Arnfjörð Bjarmason, Sep 17, 2020
  27. Junio C HamanoSep 17, 2020
  28. Johannes SixtSep 17, 2020
  29. Johannes SchindelinOct 8, 2020
  30. Ævar Arnfjörð BjarmasonOct 8, 2020
  31. Junio C HamanoOct 8, 2020
  32. Johannes SchindelinOct 9, 2020
  33. Christoph HellwigSep 17, 2020
  34. Marc BranchaudSep 17, 2020
  35. 0/2 should core.fsyncObjectFiles fsync the dir entry + docsÆvar Arnfjörð Bjarmason, Sep 17, 2020
  36. 1/2 sha1-file: fsync() loose dir entry when core.fsyncObjectFilesÆvar Arnfjörð Bjarmason, Sep 17, 2020
  37. Jeff KingSep 17, 2020
  38. Christoph HellwigSep 17, 2020
  39. Christoph HellwigSep 17, 2020
  40. Jeff KingSep 17, 2020
  41. Christoph HellwigSep 17, 2020
  42. Junio C HamanoSep 17, 2020
  43. Jeff KingSep 17, 2020
  44. Taylor BlauSep 17, 2020
  45. Ævar Arnfjörð BjarmasonSep 22, 2020
  46. Johannes SixtSep 17, 2020
  47. Ævar Arnfjörð BjarmasonSep 22, 2020
  48. Johannes SchindelinNov 19, 2020
  49. Christoph HellwigSep 17, 2020
  50. Junio C HamanoSep 17, 2020
  51. Jeff KingJan 17, 2018
  52. Christoph HellwigJan 17, 2018

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.