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

Re: [RFC] Convert builin-mailinfo.c to use The Better String Library.

From
WBWalter Bright <boost@digitalmars.com>
Date
Sep 7, 2007, 08:36 UTC
Message-ID
<fbr2iv$ugg$1@sea.gmane.org>
In-Reply-To
<D7BEA87D-1DCF-4A48-AD5B-0A3FDC973C8A@wincent.com>
Wincent Colaiuta wrote:
> Git is all about speed, and C is the best 
> choice for speed, especially in context of Git's workload.

I can appreciate that. I originally got into writing compilers because my game (Empire) ran too slowly and I thought the existing compilers could be dramatically improved.

And technically, yes, you can write code in C that is >= the speed of any other language (other than asm). But practically, this isn't necessarily so, for the following reasons:

1) You wind up having to implement the complex, dirty details of things 
yourself. The consequences of this are:
    a) you pick a simpler algorithm (which is likely less efficient - I 
run across bubble sorts all the time in code)
    b) once you implement, tune, and squeeze all the bugs out of those 
complex, dirty details, you're reluctant to change it. You're reluctant 
to try a different algorithm to see if it's faster. I've seen this 
effect a lot in my own code. (I translated a large body of my own C++ 
code that I'd spent months tuning to D, and quickly managed to get 
significantly more speed out of it, because it was much simpler to try 
out different algorithms/data structures.)
2) Garbage collection has an interesting and counterintuitive 
consequence. If you compare n malloc/free's with n gcnew/collections, 
the malloc/free will come out faster, and you conclude that gc is slow. 
But that misses one huge speed advantage of gc - you can do FAR fewer 
allocations! For example, I've done a lot of string manipulating 
programs in C. The basic problem is keeping track of who owns each 
string. This is done by, when in doubt, make a copy of the string.

But if you have gc, you don't worry about who owns the string. You just make another pointer to it. D takes this a step further with the concept of array slicing, where one creates windows on existing arrays, or windows on windows on windows, and no allocations are ever done. It's just pointer fiddling.

------ Walter Bright http://www.digitalmars.com C, C++, D programming language compilers http://www.astoriaseminar.com Extraordinary C++

Previous: Wincent ColaiutaNext: Andreas Ericsson
Message 77 of 102 in “[RFC] Convert builin-mailinfo.c to use The Better String Library.”
  1. Lukas SandströmSep 4, 2007
  2. Alex RiesenSep 4, 2007
  3. Pierre HabouzitSep 4, 2007
  4. Kristian HøgsbergSep 5, 2007
  5. Matthieu MoySep 5, 2007
  6. Miles BaderSep 6, 2007
  7. Dmitry KakurinSep 6, 2007
  8. Shawn O. PearceSep 6, 2007
  9. Andreas EricssonSep 6, 2007
  10. Junio C HamanoSep 6, 2007
  11. Andreas EricssonSep 6, 2007
  12. David KastrupSep 6, 2007
  13. Miles BaderSep 6, 2007
  14. Johannes SchindelinSep 6, 2007
  15. Linus TorvaldsSep 6, 2007
  16. Dmitry KakurinSep 7, 2007
  17. Linus TorvaldsSep 7, 2007
  18. Dmitry KakurinSep 7, 2007
  19. Linus TorvaldsSep 7, 2007
  20. Dmitry KakurinSep 7, 2007
  21. David SymondsSep 7, 2007
  22. Theodore TsoSep 7, 2007
  23. Steven BurnsSep 20, 2007
  24. Andreas EricssonSep 20, 2007
  25. Andreas EricssonSep 7, 2007
  26. Dmitry KakurinSep 7, 2007
  27. David KastrupSep 7, 2007
  28. Dmitry KakurinSep 8, 2007
  29. David KastrupSep 8, 2007
  30. Andreas EricssonSep 9, 2007
  31. David KastrupSep 7, 2007
  32. Johannes SchindelinSep 7, 2007
  33. Johannes SchindelinSep 7, 2007
  34. David KastrupSep 7, 2007
  35. Linus TorvaldsSep 7, 2007
  36. alanSep 7, 2007
  37. Walter BrightSep 7, 2007
  38. David KastrupSep 7, 2007
  39. Walter BrightSep 7, 2007
  40. David KastrupSep 7, 2007
  41. Walter BrightSep 7, 2007
  42. David KastrupSep 7, 2007
  43. Walter BrightSep 7, 2007
  44. David KastrupSep 7, 2007
  45. Walter BrightSep 7, 2007
  46. Andreas EricssonSep 8, 2007
  47. Pierre HabouzitSep 9, 2007
  48. Andreas EricssonSep 9, 2007
  49. Wincent ColaiutaSep 7, 2007
  50. Pierre HabouzitSep 7, 2007
  51. Walter BrightSep 7, 2007
  52. David KastrupSep 7, 2007
  53. Walter BrightSep 7, 2007
  54. Pierre HabouzitSep 7, 2007
  55. David KastrupSep 7, 2007
  56. Pierre HabouzitSep 7, 2007
  57. Walter BrightSep 7, 2007
  58. Pierre HabouzitSep 7, 2007
  59. Walter BrightSep 7, 2007
  60. John 'Z-Bo' ZabroskiSep 8, 2007
  61. David KastrupSep 8, 2007
  62. Steven BurnsSep 19, 2007
  63. Wincent ColaiutaSep 7, 2007
  64. Paul WankadiaSep 7, 2007
  65. Nicolas PitreSep 7, 2007
  66. Wincent ColaiutaSep 7, 2007
  67. Andreas EricssonSep 7, 2007
  68. Johannes SchindelinSep 7, 2007
  69. Andreas EricssonSep 7, 2007
  70. Wincent ColaiutaSep 7, 2007
  71. Karl HasselströmSep 7, 2007
  72. Andreas EricssonSep 7, 2007
  73. Wincent ColaiutaSep 7, 2007
  74. Andreas EricssonSep 9, 2007
  75. David KastrupSep 7, 2007
  76. Wincent ColaiutaSep 7, 2007
  77. Walter BrightSep 7, 2007
  78. Andreas EricssonSep 7, 2007
  79. Walter BrightSep 7, 2007
  80. David KastrupSep 7, 2007
  81. Andreas EricssonSep 9, 2007
  82. Bernd JendrissekSep 17, 2009
  83. Wincent ColaiutaSep 7, 2007
  84. Walter BrightSep 7, 2007
  85. Steven BurnsSep 22, 2007
  86. David KastrupSep 7, 2007
  87. Andy ParkinsSep 7, 2007
  88. David KastrupSep 7, 2007
  89. Johannes SchindelinSep 7, 2007
  90. Dmitry KakurinSep 8, 2007
  91. David KastrupSep 8, 2007
  92. Alex RiesenSep 8, 2007
  93. figoSep 24, 2007
  94. David KastrupSep 24, 2007
  95. Steven BurnsSep 25, 2007
  96. David KastrupSep 25, 2007
  97. Syed M RaihanMay 22, 2012
  98. Ian MoltonJun 10, 2010
  99. Jakub NarebskiJun 11, 2010
  100. Dario RodriguezJun 11, 2010
  101. Kristian HøgsbergSep 5, 2007
  102. Lukas SandströmSep 7, 2007

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.