Skip to content

_io.FileIO uses a quadratic-time buffer growth algorithm #57368

Closed
Closed
_io.FileIO uses a quadratic-time buffer growth algorithm#57368
Labels
performancePerformance or resource usagetopic-IO

Description

@nadeemvawda
nadeemvawda
mannequin
opened on Oct 12, 2011
BPO 13159
Nosy @pitrou, @vstinner, @benjaminp
Files
  • buffer-growth.diff
    		•

    Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.

    Show more details

    GitHub fields:

    assignee = None
closed_at = <Date 2011-10-13.14:01:59.975>
created_at = <Date 2011-10-12.16:02:00.072>
labels = ['expert-IO', 'performance']
title = '_io.FileIO uses a quadratic-time buffer growth algorithm'
updated_at = <Date 2011-10-13.14:01:59.973>
user = 'https://bugs.python.org/nadeemvawda'

    bugs.python.org fields:

    activity = <Date 2011-10-13.14:01:59.973>
actor = 'nadeem.vawda'
assignee = 'nadeem.vawda'
closed = True
closed_date = <Date 2011-10-13.14:01:59.975>
closer = 'nadeem.vawda'
components = ['IO']
creation = <Date 2011-10-12.16:02:00.072>
creator = 'nadeem.vawda'
dependencies = []
files = ['23389']
hgrepos = []
issue_num = 13159
keywords = ['patch']
message_count = 7.0
messages = ['145403', '145418', '145443', '145448', '145449', '145450', '145453']
nosy_count = 6.0
nosy_names = ['pitrou', 'vstinner', 'nadeem.vawda', 'benjamin.peterson', 'stutzbach', 'python-dev']
pr_nums = []
priority = 'normal'
resolution = 'fixed'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'performance'
url = 'https://bugs.python.org/issue13159'
versions = ['Python 2.7', 'Python 3.2', 'Python 3.3']

    Linked PRs

    Activity

    1. nadeemvawda commented on Oct 12, 2011

      @nadeemvawda
      nadeemvawdamannequin
      on Oct 12, 2011
      MannequinAuthor

      As mentioned in bpo-6715, the buffer growth strategy used by _io.FileIO
      has quadratic running time if each reallocation is O(n). The code in
      question is new_buffersize() from Modules/_io/fileio.c:

          if (currentsize > SMALLCHUNK) {
        /* Keep doubling until we reach BIGCHUNK;
           then keep adding BIGCHUNK. */
        if (currentsize <= BIGCHUNK)
            return currentsize + currentsize;
        else
            return currentsize + BIGCHUNK;
    }
    return currentsize + SMALLCHUNK;

      Is there a reason for this? One possible improvement would be to instead
      use the same formula as list.resize() in Objects/listobject.c:

          new_allocated = (newsize >> 3) + (newsize < 9 ? 3 : 6);

      which has amortized O(n) running time behaviour.

      Your thoughts?

    2. nadeemvawda
      added
      topic-IO
      performancePerformance or resource usage
      on Oct 12, 2011

    3. nadeemvawda commented on Oct 12, 2011

      @nadeemvawda
      nadeemvawdamannequin
      on Oct 12, 2011
      MannequinAuthor

      I've attached a patch that makes the suggested change to FileIO, and also
      to _bz2.BZ2Compressor/Decompressor (which currently have the same issue).

    4. pitrou commented on Oct 12, 2011

      @pitrou