Issue with some calls to GetMultiXactIdMembers() - Mailing list pgsql-hackers

From Greg Nancarrow
Subject Issue with some calls to GetMultiXactIdMembers()
Date
Msg-id CAJcOf-dO5kVosn+bS4EnYKUnjSZOa7f6duE2TRraP92qvk1piw@mail.gmail.com
Whole thread Raw
Responses Re: Issue with some calls to GetMultiXactIdMembers()  (Heikki Linnakangas <hlinnaka@iki.fi>)
List pgsql-hackers
Hi,

There's a couple of calls to GetMultiXactIdMembers() in heapam.c which
subsequently pfree() the returned "members" pointer (pass-by-reference
parameter) if it's non-NULL.
However, there's an error return within GetMultiXactIdMembers() that
returns -1 without NULLing out "members", and the callers have simply
allocated that pointer on the stack without initializing it to NULL.
If that error condition were to ever happen, pfree() would likely be
called with a junk value.
Also note that there's another error return (about 15 lines further
down) in GetMultiXactIdMembers() that returns -1 and does NULL out
"members", so the handling is inconsistent.
The attached patch adds the NULLing out of the "members" pointer in
the first error case, to fix that and guard against possible pfree()
on error by such callers.

I also note that there are other callers which pfree() "members" based
on the returned "nmembers" value, and this is also inconsistent.
Some pfree() "members" if nmembers>= 0, while others pfree() it if nmembers>0.
After looking at the code for a while, it looks like the "nmembers ==
0" case can't actually happen (right?). I decided not to mess with any
of the calling code.

Regards,
Greg Nancarrow
Fujitsu Australia

Attachment

pgsql-hackers by date:

Previous
From: Amit Kapila
Date:
Subject: Re: locking [user] catalog tables vs 2pc vs logical rep
Next
From: Yagiz Nizipli
Date:
Subject: Re: [PATCH] rename column if exists