Yes, sorry, but this doesn't change the matter.
The result is that get_linear_isa runs 10x faster on 5.10.0 while C::A::G
runs up to 25% slower on 5.10.0 and therefore the root of the problem is
still somewhere in C::A::G.
2007/12/21, Aaron Crane <[EMAIL PROTECTED]>:
>
> Oleg Pronin writes:
> > Benchmark: running aa, bb, cc, dd, ee for at least 1 CPU seconds...
> > aa: 1 wallclock secs ( 1.10 usr + 0.00 sys = 1.10 CPU) @
> 235174.13/s (n=259059)
> > bb: 1 wallclock secs ( 1.09 usr + 0.00 sys = 1.09 CPU) @
> 100662.86/s (n=110100)
> > cc: 2 wallclock secs ( 1.07 usr + 0.00 sys = 1.07 CPU) @
> 102867.15/s (n=110100)
> > dd: 1 wallclock secs ( 1.08 usr + 0.00 sys = 1.08 CPU) @
> 102121.74/s (n=110100)
> > ee: 1 wallclock secs ( 1.06 usr + 0.00 sys = 1.06 CPU) @
> 103623.53/s (n=110100)
>
> I'm afraid I haven't been paying enough attention to this to know if
> I'm missing something, but this code in your benchmark looks very odd:
>
> timethese(-1, {
> aa => sub {mro::get_linear_isa('AA', 'c3')},
> bb => sub {mro::get_linear_isa('BB', 'c3')},
> cc => sub {mro::get_linear_isa('BB', 'c3')},
> dd => sub {mro::get_linear_isa('BB', 'c3')},
> ee => sub {mro::get_linear_isa('BB', 'c3')},
> });
>
> Presumably the last three occurrences of 'BB' should be 'CC', 'DD',
> and 'EE', respectively, no?
>
> The two benchmark scripts behave the same in that respect.
>
> --
> Aaron Crane
>
> _______________________________________________
> List: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/dbix-class
> IRC: irc.perl.org#dbix-class
> SVN: http://dev.catalyst.perl.org/repos/bast/DBIx-Class/
> Searchable Archive:
> http://www.grokbase.com/group/[EMAIL PROTECTED]
>
_______________________________________________
List: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/dbix-class
IRC: irc.perl.org#dbix-class
SVN: http://dev.catalyst.perl.org/repos/bast/DBIx-Class/
Searchable Archive: http://www.grokbase.com/group/[EMAIL PROTECTED]
