oss-sec mailing list archives
Re: How serious is undefined behavior?
From: Solar Designer <solar () openwall com>
Date: Mon, 13 Jul 2015 03:19:14 +0300
On Mon, Jul 06, 2015 at 06:17:34PM +0200, Hanno B??ck wrote:
However I wonder how practically relevant these issues are and also how much focus should be given to them.
Related: "What is C in practice? (Cerberus survey): Analysis of Responses" http://www.cl.cam.ac.uk/~pes20/cerberus/notes50-2015-05-24-survey-discussion.html A productive direction may be for the free software community (or an even wider community, if possible) to agree on de facto mainstream C standard, where certain kinds of UB and such would in fact be defined in specific ways. There would still remain many kinds of UB and such, but fewer of them and the easier avoidable ones. As things currently are, non-trivial programs sort of have to make certain assumptions beyond what's guaranteed by C standards anyway. Alexander
Current thread:
- How serious is undefined behavior? Hanno Böck (Jul 06)
- Re: How serious is undefined behavior? John Haxby (Jul 06)
- Re: How serious is undefined behavior? Daniel Micay (Jul 06)
- Re: How serious is undefined behavior? Solar Designer (Jul 06)
- Re: How serious is undefined behavior? Alexander Cherepanov (Jul 09)
- Re: How serious is undefined behavior? Solar Designer (Jul 12)
- Re: How serious is undefined behavior? Alexander Cherepanov (Jul 13)
- Re: How serious is undefined behavior? Xi Wang (Jul 13)