Dave Mitchell wrote:

> I also 'fixed' what I thought might be a leak in the $^V getting code,
> although it didn't make any obvious difference.
> 
I looked at that myself but, as you say, it didn't seem to fix it.

The patch below makes the same change to the $] getting code.  Should I 
commit this too, or did you deliberately not do that case for some 
reason that I've missed?

- Steve

==== //depot/perl/gv.c#205 - C:\p5p\bleadperl\gv.c ====
1066,1068c1066,1068
<           sv = vnumify(PL_patchlevel);
<           SvREADONLY_on(sv);
<           GvSV(gv) = sv;
---
 >            GvSV(gv) = vnumify(PL_patchlevel);
 >            SvREADONLY_on(GvSV(gv));
 >            SvREFCNT_dec(sv);




------------------------------------------------
Radan Computational Ltd.

The information contained in this message and any files transmitted with it are confidential
 and intended for the addressee(s) only.  If you have received this message in error or ther
e are any problems, please notify the sender immediately.  The unauthorized use, disclosure,
 copying or alteration of this message is strictly forbidden.  Note that any views or opinio
ns presented in this email are solely those of the author and do not necessarily represent t
hose of Radan Computational Ltd.  The recipient(s) of this message should check it and any a
ttached files for viruses: Radan Computational will accept no liability for any damage cause
d by any virus transmitted by this email.