Atul et al,

> Judy worked perfectly fine for me with 32 bit compilation.
>
> This is the first time i am compiling it for 64 bit on solaris.  So
> wanted to check if somebody has already compiled and using it on 64
> bit platform.

Hmm...  I don't recall trying it on Solaris at all back when we
developed it...  Hopefully someone else on this list will have more to
say.

> I have also checked with UMEM and i found no leaks with 32 bit
> libraries, but with 64 bit it shows heap corruption.  Also tried
> changing offset variable (which is declared as int) in Judy1Set.c to
> Word_t.  It fixed the issue but the final results (bit counts) are not
> OK.

Weird.  Now, libJudy does some odd things with bits and bytes, but we
tried to ensure it was generic and safe as possible, following the C
language rules, although as you can see in some of the comments it says,
"modern compilers do this or do that" with an expectation about
performance of the assembly code.

Of course int is signed and Word_t is unsigned, but offhand I can only
guess why that would make a difference.  I haven't looked into the
source code in years.  I hope Doug is listening.

Alan

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Judy-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/judy-devel

Reply via email to