On Thu, Jan 27, 2022 at 11:37:29AM +0100, Mark Wielaard wrote:
Hi,
On Thu, Jan 27, 2022 at 10:41:36AM +0100, Roberto Ragusa wrote:
> On 1/22/22 10:05 PM, Mark Wielaard wrote:
>
> > So I would give valgrind a 6/6 (100%) score :)
>
> But if the compiler starts copying zeros on uninitialized memory,
> valgrind loses any ability to detect the defect in the code.
Yes. So that is the compromise. You'll always get initialized zeros
for local variables, so any usage is defined (though probably buggy).
But some of the tools, like valgrind memcheck, will be unable to
detect the buggy code.
If you believe the tools we have are pretty bad to begin with and/or
not actually used by people to find such bugs then this is a good
compromise. If you believe the tools are pretty good for detecting
these issues (and I believe they are, the example given was just
unfortunate because some of the issues weren't actually bad code and
some others were rightfully optimized out, so would never trigger),
then it is a bad compromise. But we definitely need to encourage
people to use the tools more.
It isn't about whether the tools are good or bad. We have many
great tools, valgrind included. It about the practicality of
using the tools to achieve the same end goal - no use of uninitalized
memory.
Valgrind is only win, if it is practical to actually exercise every
codepath in question to detect the latent problems.
Unit testing coverage for most projects is nowhere near good enough
to exercise all codepaths. Nor is any other functional/integration
testing that is done, if any.
Error code paths are a particular problem since they're rarely
exercised by normal users, and difficult to exercirse in tests
without clever fault injection techniques.
If you have unlimited developer resources, you might write tests
to exercise every relevant code path and put this all through
valgrind and become confident you don't have uninitialized variables.
Even then, valgrind isn't practical for some apps because of the
performance penalty it imposes.
So awesome as valgrind is, from a real world practicality POV,
it is a clear win to let the compiler default initialize every
variable to zero. It will eliminate a whole class of bugs that
will never be reliably identified by the developers, allowing
them to spend more time dealing with more important classes of
bugs or work on features or whatever else deserves their very
limited time.
Regards,
Daniel
--
|:
https://berrange.com -o-
https://www.flickr.com/photos/dberrange :|
|:
https://libvirt.org -o-
https://fstop138.berrange.com :|
|:
https://entangle-photo.org -o-
https://www.instagram.com/dberrange :|