1. 09 Jan, 2019 12 commits
  2. 12 Dec, 2018 1 commit
  3. 11 Dec, 2018 1 commit
  4. 02 Dec, 2018 1 commit
  5. 16 Oct, 2018 1 commit
  6. 10 Oct, 2018 1 commit
  7. 09 Oct, 2018 2 commits
  8. 19 Sep, 2018 1 commit
  9. 12 Sep, 2018 1 commit
  10. 11 Sep, 2018 1 commit
  11. 03 Sep, 2018 2 commits
  12. 27 Aug, 2018 1 commit
  13. 14 Aug, 2018 1 commit
  14. 13 Aug, 2018 1 commit
    • zhaozhao.zz's avatar
      pipeline: do not sdsrange querybuf unless all commands processed · 14c4ddb5
      zhaozhao.zz authored
      This is an optimization for processing pipeline, we discussed a
      problem in issue #5229: clients may be paused if we apply `CLIENT
      PAUSE` command, and then querybuf may grow too large, the cost of
      memmove in sdsrange after parsing a completed command will be
      horrible. The optimization is that parsing all commands in queyrbuf
      , after that we can just call sdsrange only once.
      14c4ddb5
  15. 31 Jul, 2018 2 commits
  16. 30 Jul, 2018 1 commit
  17. 23 Jul, 2018 2 commits
  18. 22 Jul, 2018 1 commit
  19. 16 Jul, 2018 2 commits
    • antirez's avatar
      Hopefully improve commenting of #5126. · f9c84d6d
      antirez authored
      Reading the PR gave me the opportunity to better specify what the code
      was doing in places where I was not immediately sure about what was
      going on. Moreover I documented the structure in server.h so that people
      reading the header file will immediately understand what the structure
      is useful for.
      f9c84d6d
    • Oran Agra's avatar
      slave buffers were wasteful and incorrectly counted causing eviction · bf680b6f
      Oran Agra authored
      A) slave buffers didn't count internal fragmentation and sds unused space,
         this caused them to induce eviction although we didn't mean for it.
      
      B) slave buffers were consuming about twice the memory of what they actually needed.
      - this was mainly due to sdsMakeRoomFor growing to twice as much as needed each time
        but networking.c not storing more than 16k (partially fixed recently in 237a38737).
      - besides it wasn't able to store half of the new string into one buffer and the
        other half into the next (so the above mentioned fix helped mainly for small items).
      - lastly, the sds buffers had up to 30% internal fragmentation that was wasted,
        consumed but not used.
      
      C) inefficient performance due to starting from a small string and reallocing many times.
      
      what i changed:
      - creating dedicated buffers for reply list, counting their size with zmalloc_size
      - when creating a new reply node from, preallocate it to at least 16k.
      - when appending a new reply to the buffer, first fill all the unused space of the
        previous node before starting a new one.
      
      other changes:
      - expose mem_not_counted_for_evict info field for the benefit of the test suite
      - add a test to make sure slave buffers are counted correctly and that they don't cause eviction
      bf680b6f
  20. 09 Jul, 2018 1 commit
  21. 04 Jul, 2018 1 commit
  22. 03 Jul, 2018 2 commits
  23. 02 Jul, 2018 1 commit