PITA, yes, but not life-threateningly so. Consider the following: X = 0 IF X=1 THEN DC C'THIS IS A CONSTANT THAT IS NOW "REACHABLE" SO THE COMPILER WILL NOT "DELETE" IT IN ORDER TO "OPTIMIZE" THE LOAD MODULE." ENDIF or whatever the equivalent syntax is for FORTRAN or other language which cannot tolerate an unreachable comment. Bill Fairchild Franklin, TN ----- Original Message -----
From: "Paul Gilmartin" <paulgboul...@aim.com> To: IBM-MAIN@LISTSERV.UA.EDU Sent: Wednesday, October 30, 2013 3:39:25 PM Subject: Re: Interesting? How _compilers_ are compromising application security On Wed, 30 Oct 2013 15:48:58 -0400, Gord Tomlin wrote: >On 2013-10-30 12:18, John Gilmore wrote: >> What are deleted are blocks of 'unreachable' code. > >Some compilers issue a warning or error message if they encounter >unreachable statements. For example, see messages CCN3472, CCN3520, >CCN6274 and CCN7640 in "z/OS V1R13.0 XL C/C++ Messages". > In FORTRAN II it was simply fatal. I understand it was because the optimization pass couldn't deal with the situation. (Perhaps it divided by zero when it tried to weight the paths.) PITA when, for testing, one wanted to branch unconditionally around a statement. -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
