On 01/03/2011 10:56 PM, Steven wrote:
Hi,
> Gerd,
>
> Have the recent changes fixed the PGN instability issue ?
Yes. They were caused by a (debug-only) buffer overflow in the
engine.cpp code.
> Joost wrote:
>> Yet what surprised me is that flag values> 3 really
>> could be
>> meaningless. Note that they are not taken into account in
>> the switch
>> statement to follow. Hence I wonder what is going on in the
>> hash
>> machinery itself. Including why the closing tte_ScoreFlag()
>> operation in
>> ProbeHash() masks the hash flag with 0x07.
>> I mean, it might as well be that this mask value should
>> actually be
>> equal to 0x03.... This would render a meaningful value and
>> explain the
>> buffer size of 4 uints at the same time!
> I had the same first impression , but Gerd hasn't changed this
> line : engine.cpp:
>
> inline scoreFlagT tte_ScoreFlag (transTableEntryT * tte)
> { return (tte->flags& 7); }
>
> Isn't it meant to be
>
> { return (tte->flags& 3); }
My suggestion was probably wrong. My current feeling is that values > 3
simply denote cache misses - this I concluded from somewhat more recent
crafty code, which includes some helpful comments.
> Alex, i think your recent cvs commit *comments* are erroneous.
> The engine.cpp changes are not fixes to scidlet,
> but Scid's built-in engine, which is different.
The comments are OK. The engine.cpp code is used for both the internal
engine and scidlet. Gerd's original fix to "the pgn stability issue"
broke scidlet's build.
> Gerd seems to be responsible
Is this what you feel, Gerd? :-)
> for some quite detailed
> technical programming. Perhaps he should have CVS write
> permissions ?? That way he can fix all his bugs
lol.
> easily enough ;>
Do agree, nevertheless.
Cheers,
Joost.
> Steve
------------------------------------------------------------------------------
Learn how Oracle Real Application Clusters (RAC) One Node allows customers
to consolidate database storage, standardize their database environment, and,
should the need arise, upgrade to a full multi-node Oracle RAC database
without downtime or disruption
http://p.sf.net/sfu/oracle-sfdevnl
_______________________________________________
Scid-users mailing list
Scid-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scid-users
