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

Re: Mozilla .git tree

From
Jon Smirl <jonsmirl@gmail.com>
Date
Sep 2, 2006, 14:20 UTC
Message-ID
<9e4733910609020720w3633aa0cq5016fb1e223fc4cb@mail.gmail.com>
In-Reply-To
<7vveo6zfx0.fsf@assigned-by-dhcp.cox.net>
On 9/2/06, Junio C Hamano <junkio@cox.net> wrote:
Show 9 quoted lines
> Shawn Pearce <spearce@spearce.org> writes:
>
> > Using a 28 byte index entry instead of a 32 byte index entry means
> > the Mozilla historical pack index will only be 52.4 MiB rather than
> > the slightly larger 59.9 MiB.
>
> Yup, that was one consideration.  One small worry is we will be
> placing u64 at 4-byte alignment by using 28-byte entries but I
> think that is Ok.

If you're going to redo the pack formats another big win for the Mozilla pack is to convert pack internal sha1 references into file offsets.within the pack. Doing that will take around 30MB off from the Mozilla pack size. sha1's are not compressible so this is a direct savings.

This might reduce memory usage too. The index is only needed to get the initial object from the pack. Since index use is lighter it could just be open/closed when needed.

You could also introduce a zlib dictionary object into the format and just leave it empty for now.

-- 
Jon Smirl
jonsmirl@gmail.com

-- 
VGER BF report: U 0.5
Previous: Junio C HamanoNext: Shawn Pearce
Message 13 of 52 in “Re: Mozilla .git tree”
  1. Shawn PearceAug 30, 2006
  2. Shawn PearceAug 30, 2006
  3. Jon SmirlAug 30, 2006
  4. Shawn PearceAug 30, 2006
  5. Jon SmirlAug 30, 2006
  6. Nicolas PitreAug 30, 2006
  7. Junio C HamanoAug 30, 2006
  8. Junio C HamanoSep 1, 2006
  9. Shawn PearceSep 2, 2006
  10. Junio C HamanoSep 2, 2006
  11. Shawn PearceSep 2, 2006
  12. Junio C HamanoSep 2, 2006
  13. Jon SmirlSep 2, 2006
  14. Shawn PearceSep 2, 2006
  15. Linus TorvaldsSep 2, 2006
  16. Junio C HamanoSep 2, 2006
  17. Shawn PearceSep 2, 2006
  18. Shawn PearceSep 2, 2006
  19. Junio C HamanoSep 2, 2006
  20. Shawn PearceSep 2, 2006
  21. Junio C HamanoSep 2, 2006
  22. Shawn PearceSep 3, 2006
  23. A Large Angry SCMSep 1, 2006
  24. Linus TorvaldsSep 1, 2006
  25. Junio C HamanoSep 1, 2006
  26. pack-objects: re-validate data we copy from elsewhere.Junio C Hamano, Sep 1, 2006
  27. Linus TorvaldsSep 2, 2006
  28. VGER BF report?Johannes Schindelin, Sep 2, 2006
  29. Sam RavnborgSep 2, 2006
  30. Junio C HamanoSep 2, 2006
  31. Junio C HamanoSep 2, 2006
  32. Shawn PearceSep 2, 2006
  33. Junio C HamanoSep 2, 2006
  34. Linus TorvaldsSep 2, 2006
  35. Junio C HamanoSep 2, 2006
  36. Shawn PearceSep 2, 2006
  37. Junio C HamanoSep 3, 2006
  38. Shawn PearceSep 4, 2006
  39. Junio C HamanoSep 4, 2006
  40. Shawn PearceSep 4, 2006
  41. Junio C HamanoSep 4, 2006
  42. Linus TorvaldsSep 3, 2006
  43. Junio C HamanoSep 3, 2006
  44. Junio C HamanoSep 5, 2006
  45. Linus TorvaldsSep 2, 2006
  46. Junio C HamanoSep 2, 2006
  47. Junio C HamanoSep 3, 2006
  48. Linus TorvaldsSep 3, 2006
  49. Linus TorvaldsSep 3, 2006
  50. Junio C HamanoSep 3, 2006
  51. Junio C HamanoSep 4, 2006
  52. Linus TorvaldsSep 4, 2006

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.