diff options
author | David Mitchell <davem@iabyn.com> | 2010-07-18 15:07:08 +0100 |
---|---|---|
committer | David Mitchell <davem@iabyn.com> | 2010-07-18 18:39:18 +0100 |
commit | cfc1e951d98ba2b9a0e066aba9aadba4cd919eec (patch) | |
tree | 035e1c687d0f681990220c6e3f4db9317252fd3e /sv.c | |
parent | 00c0cb6d254eaba165c8445a6e68686b8285b5a3 (diff) | |
download | perl-cfc1e951d98ba2b9a0e066aba9aadba4cd919eec.tar.gz |
add CVf_CVGV_RC flag
after the recent commit 803f274831f937654d48f8cf0468521cbf8f5dff,
the CvGV field is sometimes reference counted. Since it was intended that
the reference counting would happen only for anonymous CVs, the CVf_ANON
flag was co-opted to indicate whether RC was being used. This is not
entirely robust; for example, sub __ANON__ {} is a non-anon sub which
points to the same GV used by anon subs, which while itself doesn't
directly break things, shows that the potential for breakage is there.
So add a separate flag just to indicate the reference count status of the
CvGV field.
Diffstat (limited to 'sv.c')
-rw-r--r-- | sv.c | 3 |
1 files changed, 2 insertions, 1 deletions
@@ -5687,6 +5687,7 @@ S_anonymise_cv_maybe(pTHX_ GV *gv, CV* cv) SvREFCNT_dec(gvname); CvANON_on(cv); + CvCVGV_RC_on(cv); CvGV(cv) = MUTABLE_GV(SvREFCNT_inc(anongv)); } @@ -11438,7 +11439,7 @@ S_sv_dup_common(pTHX_ const SV *const sstr, CLONE_PARAMS *const param) /* don't dup if copying back - CvGV isn't refcounted, so the * duped GV may never be freed. A bit of a hack! DAPM */ CvGV(dstr) = - CvANON(dstr) + CvCVGV_RC(dstr) ? gv_dup_inc(CvGV(sstr), param) : (param->flags & CLONEf_JOIN_IN) ? NULL |