[ragel-users] wide finder in ragel

Adrian Thurston thurs... at cs.queensu.ca
Fri Nov 9 20:01:16 UTC 2007


There is an error in this which affects correctness seriously and
performance slightly. The :> operator should really be :>>. Meaning that
we stop the [^\n]* machine when the search pattern finishes, not when it
starts. In the previous version any "G" ahead of the "GET ..." will fail
the pattern.

Adrian

Adrian Thurston wrote:
> Hi, here is a completed version of the wide finder benchmark.
> 
> Summary: ragel + mmap + pthreads + std::map.
> 
> I don't have a multi-core machine to test on, but on a single core it
> goes about 10x faster than the ruby version here:
> 
> http://www.tbray.org/ongoing/When/200x/2007/09/20/Wide-Finder
> 
> Adrian
> 
> 
> > 
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: Makefile
URL: <http://www.colm.net/pipermail/ragel-users/attachments/20071109/cc92f001/attachment-0002.ksh>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rwf.cpp
Type: text/x-c++src
Size: 10522 bytes
Desc: not available
URL: <http://www.colm.net/pipermail/ragel-users/attachments/20071109/cc92f001/attachment-0001.cpp>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: rwf.rl
URL: <http://www.colm.net/pipermail/ragel-users/attachments/20071109/cc92f001/attachment-0003.ksh>


More information about the ragel-users mailing list