Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.python > #34302

Re: why does dead code costs time?

Show all headers | View raw

On Wed, Dec 5, 2012 at 10:34 AM, Steven D'Aprano
<steve+comp.lang.python@pearwood.info> wrote:
> The difference is almost certain between the LOAD_CONST and the
> LOAD_GLOBAL.
>
> As to *why* there is such a difference, I believe that's a leftover from
> early Python days when None was not a keyword and could be reassigned.

I think this should even be considered a bug, not just a missing
optimization.  Consider:

>>> globals()['None'] = 42
>>> def f(x):
...     return None
...     print(x)
...
>>> f('test')
42

The use of the LOAD_GLOBAL allows None to effectively be reassigned.

It's also worth noting that:

>>> def f(x):
...     return
...     print(x)
...
>>> dis.dis(f)
  2           0 LOAD_CONST               0 (None)
              3 RETURN_VALUE

  3           4 LOAD_GLOBAL              0 (print)
              7 LOAD_FAST                0 (x)
             10 CALL_FUNCTION            1 (1 positional, 0 keyword pair)
             13 POP_TOP

So if you just write 'return' rather than 'return None', you get the
correct bytecode.  Additionally, the use of LOAD_GLOBAL instead of
LOAD_CONST seems to be linked to having unreachable code at the end of
the function.  This is fine:

>>> def f(x):
...     if x:
...         return None
...     print(x)
...
>>> dis.dis(f)
  2           0 LOAD_FAST                0 (x)
              3 POP_JUMP_IF_FALSE       10

  3           6 LOAD_CONST               0 (None)
              9 RETURN_VALUE

  4     >>   10 LOAD_GLOBAL              1 (print)
             13 LOAD_FAST                0 (x)
             16 CALL_FUNCTION            1 (1 positional, 0 keyword pair)
             19 POP_TOP
             20 LOAD_CONST               0 (None)
             23 RETURN_VALUE

But this is not.  Note here that *both* loads of None become
LOAD_GLOBAL in this case:

>>> def f(x):
...     if x:
...         return None
...     return None
...     print(x)
...
>>> dis.dis(f)
  2           0 LOAD_FAST                0 (x)
              3 POP_JUMP_IF_FALSE       13

  3           6 LOAD_GLOBAL              0 (None)
              9 RETURN_VALUE
             10 JUMP_FORWARD             0 (to 13)

  4     >>   13 LOAD_GLOBAL              0 (None)
             16 RETURN_VALUE

  5          17 LOAD_GLOBAL              1 (print)
             20 LOAD_FAST                0 (x)
             23 CALL_FUNCTION            1 (1 positional, 0 keyword pair)
             26 POP_TOP

Back to comp.lang.python | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

why does dead code costs time? Bruno Dupuis <python.ml.bruno.dupuis@lisael.org> - 2012-12-05 16:46 +0100
  Re: why does dead code costs time? Neil Cerutti <neilc@norwich.edu> - 2012-12-05 16:15 +0000
    Re: why does dead code costs time? Bruno Dupuis <python.ml.bruno.dupuis@lisael.org> - 2012-12-05 17:40 +0100
      Re: why does dead code costs time? Ramchandra Apte <maniandram01@gmail.com> - 2012-12-09 06:11 -0800
        Re: why does dead code costs time? Mark Lawrence <breamoreboy@yahoo.co.uk> - 2012-12-09 16:47 +0000
          Re: why does dead code costs time? Ramchandra Apte <maniandram01@gmail.com> - 2012-12-12 21:23 -0800
            Re: why does dead code costs time? Steven D'Aprano <steve+comp.lang.python@pearwood.info> - 2012-12-13 06:01 +0000
              Re: why does dead code costs time? rusi <rustompmody@gmail.com> - 2012-12-12 22:27 -0800
                Re: why does dead code costs time? Cameron Simpson <cs@zip.com.au> - 2012-12-13 17:51 +1100
                Re: why does dead code costs time? rusi <rustompmody@gmail.com> - 2012-12-12 22:59 -0800
                Re: why does dead code costs time? Chris Angelico <rosuav@gmail.com> - 2012-12-13 18:03 +1100
          Re: why does dead code costs time? Ramchandra Apte <maniandram01@gmail.com> - 2012-12-12 21:23 -0800
    Re: why does dead code costs time? Bruno Dupuis <python.ml.bruno.dupuis@lisael.org> - 2012-12-05 18:19 +0100
      Re: why does dead code costs time? Tim Roberts <timr@probo.com> - 2012-12-05 21:20 -0800
  Re: why does dead code costs time? Steven D'Aprano <steve+comp.lang.python@pearwood.info> - 2012-12-05 17:34 +0000
    Re: why does dead code costs time? Steven D'Aprano <steve+comp.lang.python@pearwood.info> - 2012-12-05 17:39 +0000
    Re: why does dead code costs time? Ian Kelly <ian.g.kelly@gmail.com> - 2012-12-05 10:59 -0700
    Re: why does dead code costs time? Bruno Dupuis <python.ml.bruno.dupuis@lisael.org> - 2012-12-05 19:24 +0100
    Re: why does dead code costs time? Terry Reedy <tjreedy@udel.edu> - 2012-12-05 15:41 -0500
    Re: why does dead code costs time? Chris Kaynor <ckaynor@zindagigames.com> - 2012-12-05 12:49 -0800
    Re: why does dead code costs time? Bruno Dupuis <python.ml.bruno.dupuis@lisael.org> - 2012-12-05 21:50 +0100
    Re: why does dead code costs time? Bruno Dupuis <python.ml.bruno.dupuis@lisael.org> - 2012-12-05 22:58 +0100

csiph-web