[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
## Re: speed of octave symbol table code

**From**: |
John W. Eaton |

**Subject**: |
Re: speed of octave symbol table code |

**Date**: |
Tue, 23 Oct 2007 10:25:55 -0400 |

On 23-Oct-2007, David Bateman wrote:
| It really depends on the real life code and the number of symbol table
| lookups per function, and as you say how many symbols per function. What
| is the average number of symbols per function with your code for "make
| check".
The numbers are not necessarily unique for each function because
different paths through the code may result in different numbers of
symbols being stored in the table and distribution of chains. In any
case, the average for unique calls is around 38 symbols, but if we
weight the average by the number of calls made for a given number of
symbols/distribution of chains, the average drops to around 17. So
there are many calls to functions with just a few symbols. So maybe
we could at least cut the table size down to 64.
jwe

**speed of octave symbol table code**, *David Bateman*, `2007/10/22`
**speed of octave symbol table code**, *John W. Eaton*, `2007/10/22`
**Re: speed of octave symbol table code**, *David Bateman*, `2007/10/22`
**Re: speed of octave symbol table code**, *John W. Eaton*, `2007/10/22`
**Re: speed of octave symbol table code**, *John W. Eaton*, `2007/10/23`
**Re: speed of octave symbol table code**, *David Bateman*, `2007/10/23`
**Re: speed of octave symbol table code**,
*John W. Eaton* **<=**
**Re: speed of octave symbol table code**, *David Bateman*, `2007/10/23`
**Re: speed of octave symbol table code**, *Tom Holroyd*, `2007/10/23`
**Re: speed of octave symbol table code**, *John W. Eaton*, `2007/10/23`