Call For Vote: Project Tsan

[Jean Christophe Beyler]

Hi Florian,

I am not disagreeing with this conversation but could we perhaps wait until
the project opens and we have a tsan mailing list where we could debate and
talk about it there?

Then we could see what kind of solution could be done and have it recorded
there? When the mailing thread opens, I'll send an email with the top
points and we can go from there.

Any objections?

Thanks,
Jc

On Wed, Mar 13, 2019 at 8:53 AM Florian Weimer <fweimer at redhat.com> wrote:

> * Martin Buchholz:
>
> > There may be a cultural expectations gap.
> >
> > Companies (like Google) might make the toolchain a part of their basic
> > computing infrastructure; build everything from source, and use a tsan
> > that comes with the one blessed toolchain used to build everything
> > else.
> >
> > Linux distros (like Red Hat) probably also try to build most binaries
> > with a preferred toolchain, but there is a much wider promise of ABI
> > compatibility.  In particular, part of the value proposition of having
> > a Linux distro is users not having to rebuild from source, and being
> > able to combine binaries built by others, using different toolchains.
>
> If OpenJDK starts to poke at libc internals, like the sanitizers do
> today, OpenJDK will potentially need porting to each new libc version.
>
> You will not be able to run older OpenJDK versions on newer libcs, as it
> is possible today.
>
> It will likely make it more complicated to build binaries which are
> compatible with a wide range of systems.
>
> This is what we see with the sanitizers in LLVM and GCC today.  If you
> copy a variant of this code into OpenJDK, you will have the same
> problem.
>
> Thanks,
> Florian
>


-- 

Thanks,
Jc