On Mon, Jul 10, 2017 at 10:37:46AM +0200, wm4 wrote: > On Sun, 9 Jul 2017 22:03:21 +0200 > Reimar Döffinger <reimar.doeffin...@gmx.de> wrote: > > > On 09.07.2017, at 16:08, "Ronald S. Bultje" <rsbul...@gmail.com> wrote: > > > Hi, > > > > > > On Sun, Jul 9, 2017 at 4:39 AM, Reimar Döffinger > > > <reimar.doeffin...@gmx.de> > > > wrote: > > > > > >> On 09.07.2017, at 02:52, "Ronald S. Bultje" <rsbul...@gmail.com> wrote: > > >>> On Sat, Jul 8, 2017 at 5:17 PM, Michael Niedermayer > > >> <mich...@niedermayer.cc> > > >>> wrote: > > >>> > > >>>> > > >>>> Does anyone object to this patch ? > > >>>> Or does anyone have a better idea on how to fix this ? > > >>>> if not id like to apply it > > >>> > > >>> > > >>> I think Rostislav's point is: why fix it, if it can only happen with > > >>> corrupt input? The before and after situation is identical: garbage in, > > >>> garbage out. If the compiler does funny things that makes the garbage > > >>> slightly differently bad, is that really so devilishly bad? It's still > > >>> garbage. Is anything improved by this? > > >> > > >> The way C works, you MUST assume any undefined behaviour can at any point > > >> [..] become exploitable.[..] If you don't like that, C is the wrong > > >> language to use. > > > > > > > > > I think I've read "the boy who cried wolf" a few too many times to my > > > kids, > > > but the form of this discussion is currently too polarizing/political for > > > my taste. > > > > That is my reading of the C standard, is that political or even just > > controversial? > > I mean of course you can ignore standards (see MPEG-4 ASP, and in some ways > > that was actually fairly reasonable thing to do at the time), and I don't > > fix every undefined behaviour case in my code when I can't think of any > > reasonable solution. > > So there is a cost-benefit discussion in principle. > > I believe the cost of not fixing undefined behaviour, just by virtue of > > going outside what the standard guarantees should be considered fairly high. > > That is an opinion, but is there any disagreement that undefined behaviour > > is at least an issue of some degree? > > If we can agree on that, then the question would only be how much > > effort/code ugliness is reasonable. > > There is also the point (which I hope I mentioned in the parts cut out) > > that just making sure that these cases are not already exploitable right > > now with the current compiler can in many cases be quite a pain (and does > > not have tool support), so I think fixing it would often be the > > lowest-effort method. > > The controversial thing is actually the SUINT nonsense. A type is > either signed or unsigned, but not both incompletely intransparent
> ways. Michael keeps adding them even though many are against it. It is extreemly rude from you to make this claim. When in fact i try my best to respect every maitainer and authors preferrance and never add it when people object. [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB The educated differ from the uneducated as much as the living from the dead. -- Aristotle
signature.asc
Description: Digital signature
_______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
