Bugs item #1894726, was opened at 2008-02-16 00:44
Message generated for change (Comment added) made by lebert
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=111005&aid=1894726&group_id=11005

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Bert Wesarg (lebert)
Assigned to: Nobody/Anonymous (nobody)
Summary: fix memoy leak in PromoteToGlobal()

Initial Comment:
InstallSymbol() allocates a new Symbol and put this to the GlobalSymList the 
Symbol passed to PromoteToGlobal() is lost.

This patch adds the Symbol directly to the GlobalSymList, without calling 
InstallSymbol().

The case in which the Symbol is already in the LocalSymbolList is not changed, 
because I don't know if this can a cause SIGSEGV but it should also be 
corrected (ie freeing sym).

----------------------------------------------------------------------

>Comment By: Bert Wesarg (lebert)
Date: 2008-03-01 02:07

Message:
Logged In: YES 
user_id=122956
Originator: YES

I will commit this fix by the next week. There is only the problem with
the extra LookupSymbol() call. I suspect that this would be a greater
error, if this call ever succeeded, because that would mean that a
LOCAL_SYM was not in the LocalSymList, or there was previously a non
LOCAL_SYM in the GlobalSymList. I would like to add a fprintf instead of
the return, and see if this case is ever triggered (see the new patch).


File Added: fix-mem-leak-in-PromoteToGlobal.patch

----------------------------------------------------------------------

Comment By: Nobody/Anonymous (nobody)
Date: 2008-02-16 23:10

Message:
Logged In: NO 

should be updated to CVS immediately

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=111005&aid=1894726&group_id=11005
-- 
NEdit Develop mailing list - [email protected]
http://www.nedit.org/mailman/listinfo/develop

Reply via email to