Re: Obscure crashes due to gcc 4.9 -O2 => -fisolate-erroneous-paths-dereference

[Andrew Pinski <pinskia@gmail.com>]

On Wed, Feb 18, 2015 at 11:21 AM, Jeff Prothero <jprother@altera.com> wrote:
>
> Starting with gcc 4.9, -O2 implicitly invokes
>
>     -fisolate-erroneous-paths-dereference:
>
> which
>
>     https://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html
>
> documents as
>
>     Detect paths that trigger erroneous or undefined behavior due to
>     dereferencing a null pointer. Isolate those paths from the main control
>     flow and turn the statement with erroneous or undefined behavior into a
>     trap. This flag is enabled by default at -O2 and higher.
>
> This results in a sizable number of previously working embedded programs mysteriously
> crashing when recompiled under gcc 4.9.  The problem is that embedded
> programs will often have ram starting at address zero (think hardware-defined
> interrupt vectors, say) which gets initialized by code which the
> -fisolate-erroneous-paths-deference logic can recognize as reading and/or
> writing address zero.

You should have used -fno-delete-null-pointer-checks which has been
doing this optimization for a long time now, just it got better with
-fisolate-erroneous-paths-dereference pass.

Thanks,
Andrew Pinski



>
> What happens then is that the previously running program compiles without
> any warnings, but then typically locks up mysteriously (often disabling the
> remote debug link) due to the trap not being gracefully handled by the
> embedded runtime.
>
> Granted, such code is out-of-spec wrt to C standards.
>
> None the less, the problem is quite painful to track down and
> unexpected.
>
> Is there any good reason the
>
>     -fisolate-erroneous-paths-dereference
>
> logic could not issue a compiletime warning or error, instead of just
> silently generating code virtually certain to crash at runtime?
>
> Such a warning/error would save a lot of engineers significant amounts
> of time, energy and frustration tracking down this problem.
>
> I would like to think that the spirit of gcc is about helping engineers
> efficiently correct nonstandard pain, rather than inflicting maximal
> pain upon engineers violating C standards.  :-)
>
> -Jeff
>
> BTW, I'd also be curious to know what is regarded as engineering best
> practice for writing a value to address zero when this is architecturally
> required by the hardware platform at hand.  Obviously one can do various
> things to obscure the process sufficiently that the current gcc implementation
> won't detect it and complain, but as gcc gets smarter about optimization
> those are at risk of failing in a future release.  It would be nice to have
> a guaranteed-to-work future-proof idiom for doing this. Do we have one, short
> of retreating to assembly code?