Reviewers: jochen,
Message:
PTAL.
That first version wasn't meant for review, as both you and the compiler
noticed.. :-P
This triggers a minor bug in the webkit tests, so I'll have to wait for
that to
be fixed before submitting.
Description:
Add defensive assert for having a weak, empty presistent in a
PersistentValueMap.
(This shouldn't happen in the first place, but if it does then this
would explain the stack trace in the referenced bug. If so, zhe
RELEASE_ASSERT should help us pinpoint the culprit.)
[email protected]
BUG=368095
LOG=N
Please review this at https://codereview.chromium.org/265823006/
SVN Base: https://v8.googlecode.com/svn/branches/bleeding_edge
Affected files (+1, -0 lines):
M src/global-handles.cc
Index: src/global-handles.cc
diff --git a/src/global-handles.cc b/src/global-handles.cc
index
d6cd47918115799311e4409c9c4500083265aead..82b5f3d96b581674b29c8255b95f1cda1594b23a
100644
--- a/src/global-handles.cc
+++ b/src/global-handles.cc
@@ -207,6 +207,7 @@ class GlobalHandles::Node {
void MakeWeak(void* parameter, WeakCallback weak_callback) {
ASSERT(weak_callback != NULL);
ASSERT(state() != FREE);
+ CHECK(object_ != NULL);
set_state(WEAK);
set_parameter(parameter);
weak_callback_ = weak_callback;
--
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
---
You received this message because you are subscribed to the Google Groups "v8-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
