Re: Regression tests fail with musl libc because libpq.so can't be loaded

On Tue, Mar 19, 2024 at 11:48:50AM +1300, Thomas Munro wrote:
> On Tue, Mar 19, 2024 at 10:17 AM Thomas Munro <thomas(dot)munro(at)gmail(dot)com> wrote:
> > ... (though my patch could be a little sneakier and steal
> > all the bytes right up to the = sign to get more space for our
> > message!).
>
> Here's one like that. No musl here -- does this work Wolfgang? Do we
> think it's generous enough with space in practice that we could just
> always do this for __linux__ systems without anyone noticing (ie
> including glibc users)? Should we be more specific about which LD_*
> variables? Do people not doing hacking/testing ever really set those,
> eg on production servers? This code path was once used by up to a
> dozen or so OSes but they're all dead, only Linux, Solaris and macOS
> left, and I don't have any reason to think they suffer from this
> problem and Macs don't even follow the SysV LD_ naming convention,
> hence gating on Linux.

So this would truncate the process title on all Linux that have an LD_
environment entry, even those without musl?

--
Bruce Momjian <bruce(at)momjian(dot)us> https://momjian.us
EDB https://enterprisedb.com

Only you can decide what is important to you.