Re: Preventing abort() and exit() calls in libpq - Mailing list pgsql-hackers

From Jacob Champion
Subject Re: Preventing abort() and exit() calls in libpq
Date
Msg-id 647ee70e51621c8f35c8c06d20faf8c96345e6d0.camel@vmware.com
Whole thread Raw
In response to Re: Preventing abort() and exit() calls in libpq  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: Preventing abort() and exit() calls in libpq  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
On Wed, 2021-06-30 at 10:42 -0400, Tom Lane wrote:
> Alvaro Herrera <alvherre@alvh.no-ip.org> writes:
> > On 2021-Jun-30, Tom Lane wrote:
> > > You mentioned __gcov_exit, but I'm not sure if we need an
> > > exception for that.  I see it referenced by the individual .o
> > > files, but the completed .so has no such reference, so at least
> > > on RHEL8 it's apparently satisfied during .so linkage.  Do you
> > > see something different?
> > Well, not really.  I saw it but only after I removed -fprofile-arcs from
> > Makefile.shlib's link line; but per my other email, that doesn't really
> > work.
> > Everything seems to work well for me after removing abort from that grep.
> 
> OK, thanks, will push a fix momentarily.

With latest HEAD, building with --enable-coverage still fails on my
Ubuntu 20.04:

    ! nm -A -u libpq.so.5.15 2>/dev/null | grep -v __cxa_atexit | grep exit
    libpq.so.5.15:                 U exit@@GLIBC_2.2.5

I don't see any exit references in the libpq objects or in
libpgport_shlib, so it seems like libpgcommon_shlib is the culprit... I
assume turning off optimizations leads to less dead code elimination?

--Jacob

pgsql-hackers by date:

Previous
From: Fabien COELHO
Date:
Subject: Re: rand48 replacement
Next
From: Tom Lane
Date:
Subject: Re: Preventing abort() and exit() calls in libpq