Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 30 Mar 2011 13:19:58 -0400 (EDT)
From: "Steven M. Christey" <>
To: oss-security <>
cc: "Steven M. Christey" <>,
        Rickard Green <>,
        Bjorn-Egil Dahlberg <>,
        Sverker Eriksson <>, Patrik Nyblom <>,
        Raimo Niskanen <>, Bjorn Gustavsson <>,
        Niclas Axelsson <>, Hans Bolinder <>
Subject: Re: CVE Request -- Erlang/OTP R14, Erlang/OTP R14B01, Erlang/OTP
 R14B02 -- multiple security fixes

Some informal guidance on vulnerabilities in language 
interpreters/compilers: if there's a reasonable chance that an API 
function's correctness is affected, and that API function could be used by 
an application to process untrusted data (and/or affect the application's 
control flow), then it is generally treated as a security concern.  When 
API correctness is *not* affected - but applications could just use it in 
an insecure way - then the applications are "blamed" for the issue (the 
classic example is C's strcpy() function, which has a significant design 
limitation that many application programmers don't take into account, 
leading to buffer overflows.)

So for issues like "inexact comparisons" (whatever those are ;-) there is 
the consideration of whether such functionality is likely to be used when 
implementing security-related functionality.  For issues like incorrectly 
reporting error status from an API function, that may be a candidate for a 
CVE if the incorrect status report could have downstream effects on an 
application's correctness.

- Steve

Powered by blists - more mailing lists

Please check out the Open Source Software Security Wiki, which is counterpart to this mailing list.

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.