MIME-Version: 1.0
References: <19095-6d8256d0c37d4be2@postgresql.org> <CAAh00ETwx8_AEM0wgoi-v875uC-FLuGGOMTaP9fCNdVw4Zq=Vg@mail.gmail.com>
 <2305435.1762929527@sss.pgh.pa.us> <7C7B32ED-5742-49D5-9380-5E02AA5B03C4@yesql.se>
 <aRRCE_2ydBfCnO7Q@paquier.xyz> <057A2AED-0116-446B-8986-57FDA70CA5A3@yesql.se>
 <2wdijz5mucqx33vqzyep44uyucauw33egvwa7cbociesqlks5x@yj63saikpzwu> <CAAh00ETETkxbHmFff+2j4Lj6bbr5fWzMy=rhTg-yg4CzD=HiNA@mail.gmail.com>
In-Reply-To: <CAAh00ETETkxbHmFff+2j4Lj6bbr5fWzMy=rhTg-yg4CzD=HiNA@mail.gmail.com>
From: BharatDB <bharatdbpg@gmail.com>
Date: Wed, 19 Nov 2025 17:59:09 +0530
Message-ID: <CAAh00ERij397C5BXe=R6T38Ki7_G=0svdcFuTtz7eG-qs7vtuw@mail.gmail.com>
Subject: Re: BUG #19095: Test if function exit() is used fail when linked static
To: Andres Freund <andres@anarazel.de>, Tom Lane <tgl@sss.pgh.pa.us>
Cc: torsten.rupp@gmx.net, pgsql-bugs@lists.postgresql.org, 
	VASUKI M <vasukim1992002@gmail.com>, daniel@yesql.se, michael@paquier.xyz
Content-Type: multipart/mixed; boundary="000000000000147bf40643f1b952"
Archived-At: <https://www.postgresql.org/message-id/CAAh00ERij397C5BXe%3DR6T38Ki7_G%3D0svdcFuTtz7eG-qs7vtuw%40mail.gmail.com>
Precedence: bulk

--000000000000147bf40643f1b952
Content-Type: multipart/alternative; boundary="000000000000147bf10643f1b950"

--000000000000147bf10643f1b950
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Sorry,for the wrong patch this is the correct one.kindly find the attached
patch for testing
I would love to hear the feedback from you committers.

regards,
Vasuki M
BharatDB,C-DAC Chennai.

On Wed, Nov 19, 2025 at 2:38=E2=80=AFPM BharatDB <bharatdbpg@gmail.com> wro=
te:

> Hi Andres and hackers,
> I came up with the solution
> =E2=80=94 short follow-up with what I changed and how the Meson check act=
ually
> runs.
>
> Summary of what I did
>   - Added a Meson custom_target in src/interfaces/libpq/meson.build that
>     scans libpq's object files for direct exit() references.
>   - Added pthread_exit to the Makefile whitelist (the Makefile test
>     already existed; this just updates the whitelist).
>   -Added the custom target in the top level meson.build
>
> Where the Meson check lives and when it runs
>   - I put the custom_target in src/interfaces/libpq/meson.build,
>     immediately after the libpq shared/static library targets are
>     defined and after the declare_dependency for libpq.  That is the
>     correct location in the libpq build file.
>   - The custom_target declares `depends: [libpq_st, libpq_so]`.  This
>     is important: it tells Meson to build those targets first so the
>     .o files are present in the build directory before the check runs.
>     The check itself does not scan libpq_st or libpq_so; it scans
>     files in build/src/interfaces/libpq/*.o.
>   - The command uses `find <builddir>/src/interfaces/libpq -name '*.o'`
>     and runs `nm -u` on each .o, piping through `grep -v` for the
>     whitelisted names and finally checking for exit.
>   - The check is skipped for:
>       * cross-builds
>       * when coverage (b_coverage) is enabled
>       * on Windows (no nm/grep in the same form there) as Andrew question=
ed
>
> Why pthread_exit is whitelisted
>   - pthread_exit can legitimately appear in a few build/runtime
>     configurations (thread runtimes or link-time glue), and the
>     Makefile test was updated to whitelist it.  The Meson check has the
>     same whitelist so both build systems behave the same.
>   - Whitelisting pthread_exit doesn't remove the value of the test:
>     it only avoids false positives for legitimate thread shutdown code.
>     We still catch direct exit()/ _exit()/abort()-style calls as tom said=
.
>
> How to reproduce locally
>   - From repo root:
>       rm -rf build
>       meson setup build
>       cd build
>       ninja
>     The custom_target runs as part of the normal build and will fail
>     the build if any .o contains an un-whitelisted exit() reference.
>
> On Fri, Nov 14, 2025 at 7:31=E2=80=AFPM Andres Freund <andres@anarazel.de=
> wrote:
>
> But more generally: If we allow pthread_exit(), what's the point of this
>> test?
>> That's one of the functions we better avoid calling, no?
>>
>  I agree it's worth questioning which functions we allow.  The
>  current choice (whitelist pthread_exit) mirrors the Makefile
>  behavior, avoids false positives, and keeps the test focused on
>  direct process-termination calls authored in our code.  If the
>  community prefers a stricter policy so we can adjust the whitelist.
>
>
>> ISTM that if we do want to continue having this test, the issue is that
>> we're
>> testing the shared library - which will have already linked against stat=
ic
>> libraries like the sanitizer ones or in this case libcrypto. What we
>> ought to
>> do is to test the .o files constituting libpq.so, rather than the alread=
y
>> linked .so. That way we will find our own calls to exit etc, but not one=
s
>> in
>> static libraries.
>>
>
> TBH After so many test runs I have finalized
> Why scan .o files (not the final .so)?
>   - A shared library is usually linked with other static libraries
>     (libcrypto, sanitizer runtimes).  If we scan the final .so we'll
>     see references that originate in those static libraries and produce
>     false positives.
>   - If we scan the .o files that make up libpq, we only inspect our
>     own compilation units and will catch only exit() calls introduced
>     by our code.
>
> HTH! I attached the patch and also added the meson test output before
> /after of any file containing an 'exit' explicitly; it fails the
> build[ninja].
>
> Regards,
> Vasuki M
> BharatDB[CDAC chennai]
>

--000000000000147bf10643f1b950
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Sorry,for the wrong patch this is the correct=C2=A0one.kin=
dly find the attached patch for testing=C2=A0<br>I would love to hear the f=
eedback from you committers.<div><br></div><div>regards,</div><div>Vasuki M=
</div><div>BharatDB,C-DAC Chennai.=C2=A0</div></div><br><div class=3D"gmail=
_quote gmail_quote_container"><div dir=3D"ltr" class=3D"gmail_attr">On Wed,=
 Nov 19, 2025 at 2:38=E2=80=AFPM BharatDB &lt;<a href=3D"mailto:bharatdbpg@=
gmail.com">bharatdbpg@gmail.com</a>&gt; wrote:<br></div><blockquote class=
=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rg=
b(204,204,204);padding-left:1ex"><div dir=3D"ltr"><div dir=3D"ltr">Hi Andre=
s and hackers,<br>I came up with the solution<br>=E2=80=94 short follow-up =
with what I changed and how the Meson check actually runs.<br><br>Summary o=
f what I did<br>=C2=A0 - Added a Meson custom_target in src/interfaces/libp=
q/meson.build that<br>=C2=A0 =C2=A0 scans libpq&#39;s object files for dire=
ct exit() references.<br>=C2=A0 - Added pthread_exit to the Makefile whitel=
ist (the Makefile test<br>=C2=A0 =C2=A0 already existed; this just updates =
the whitelist).</div><div>=C2=A0 -Added the custom target in the top level =
meson.build</div><div><br></div><div dir=3D"ltr">Where the Meson check live=
s and when it runs<br>=C2=A0 - I put the custom_target in src/interfaces/li=
bpq/meson.build,<br>=C2=A0 =C2=A0 immediately after the libpq shared/static=
 library targets are<br>=C2=A0 =C2=A0 defined and after the declare_depende=
ncy for libpq.=C2=A0 That is the<br>=C2=A0 =C2=A0 correct location in the l=
ibpq build file.<br>=C2=A0 - The custom_target declares `depends: [libpq_st=
, libpq_so]`.=C2=A0 This<br>=C2=A0 =C2=A0 is important: it tells Meson to b=
uild those targets first so the<br>=C2=A0 =C2=A0 .o files are present in th=
e build directory before the check runs.<br>=C2=A0 =C2=A0 The check itself =
does not scan libpq_st or libpq_so; it scans<br>=C2=A0 =C2=A0 files in buil=
d/src/interfaces/libpq/*.o.<br>=C2=A0 - The command uses `find &lt;builddir=
&gt;/src/interfaces/libpq -name &#39;*.o&#39;`<br>=C2=A0 =C2=A0 and runs `n=
m -u` on each .o, piping through `grep -v` for the<br>=C2=A0 =C2=A0 whiteli=
sted names and finally checking for exit.<br>=C2=A0 - The check is skipped =
for:<br>=C2=A0 =C2=A0 =C2=A0 * cross-builds<br>=C2=A0 =C2=A0 =C2=A0 * when =
coverage (b_coverage) is enabled<br>=C2=A0 =C2=A0 =C2=A0 * on Windows (no n=
m/grep in the same form there) as Andrew questioned<br><br>Why pthread_exit=
 is whitelisted<br>=C2=A0 - pthread_exit can legitimately appear in a few b=
uild/runtime<br>=C2=A0 =C2=A0 configurations (thread runtimes or link-time =
glue), and the<br>=C2=A0 =C2=A0 Makefile test was updated to whitelist it.=
=C2=A0 The Meson check has the<br>=C2=A0 =C2=A0 same whitelist so both buil=
d systems behave the same.<br>=C2=A0 - Whitelisting pthread_exit doesn&#39;=
t remove the value of the test:<br>=C2=A0 =C2=A0 it only avoids false posit=
ives for legitimate thread shutdown code.<br>=C2=A0 =C2=A0 We still catch d=
irect exit()/ _exit()/abort()-style calls as tom said.<br><br>How to reprod=
uce locally<br>=C2=A0 - From repo root:<br>=C2=A0 =C2=A0 =C2=A0 rm -rf buil=
d<br>=C2=A0 =C2=A0 =C2=A0 meson setup build<br>=C2=A0 =C2=A0 =C2=A0 cd buil=
d<br>=C2=A0 =C2=A0 =C2=A0 ninja<br>=C2=A0 =C2=A0 The custom_target runs as =
part of the normal build and will fail<br>=C2=A0 =C2=A0 the build if any .o=
 contains an un-whitelisted exit() reference.<br><br></div><div class=3D"gm=
ail_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Fri, Nov 14, 2025 at 7:=
31=E2=80=AFPM Andres Freund &lt;<a href=3D"mailto:andres@anarazel.de" targe=
t=3D"_blank">andres@anarazel.de</a>&gt; wrote:</div><div dir=3D"ltr" class=
=3D"gmail_attr"><br></div><blockquote class=3D"gmail_quote" style=3D"margin=
:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"=
>
But more generally: If we allow pthread_exit(), what&#39;s the point of thi=
s test?<br>
That&#39;s one of the functions we better avoid calling, no?<br></blockquot=
e>=C2=A0I agree it&#39;s worth questioning which functions we allow.=C2=A0 =
The<br>=C2=A0current choice (whitelist pthread_exit) mirrors the Makefile<b=
r>=C2=A0behavior, avoids false positives, and keeps the test focused on<br>=
=C2=A0direct process-termination calls authored in our code.=C2=A0 If the<b=
r>=C2=A0community prefers a stricter policy so we can adjust the whitelist.=
<br><div>=C2=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0px =
0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
ISTM that if we do want to continue having this test, the issue is that we&=
#39;re<br>
testing the shared library - which will have already linked against static<=
br>
libraries like the sanitizer ones or in this case libcrypto. What we ought =
to<br>
do is to test the .o files constituting libpq.so, rather than the already<b=
r>
linked .so. That way we will find our own calls to exit etc, but not ones i=
n<br>
static libraries.<br></blockquote><div>=C2=A0</div><div>TBH After so many t=
est runs I have finalized</div><div>Why scan .o files (not the final .so)?<=
br>=C2=A0 - A shared library is usually linked with other static libraries<=
br>=C2=A0 =C2=A0 (libcrypto, sanitizer runtimes).=C2=A0 If we scan the fina=
l .so we&#39;ll<br>=C2=A0 =C2=A0 see references that originate in those sta=
tic libraries and produce<br>=C2=A0 =C2=A0 false positives.<br>=C2=A0 - If =
we scan the .o files that make up libpq, we only inspect our<br>=C2=A0 =C2=
=A0 own compilation units and will catch only exit() calls introduced<br>=
=C2=A0 =C2=A0 by our code.</div><div><br>HTH! I attached the patch and also=
 added the meson test output before /after of any file containing an &#39;e=
xit&#39; explicitly; it fails the build[ninja].<br><br>Regards,<br>Vasuki M=
<br>BharatDB[CDAC chennai]</div></div></div>
</blockquote></div>

--000000000000147bf10643f1b950--
--000000000000147bf40643f1b952
Content-Type: text/x-patch; charset="UTF-8"; 
	name="0001-libpq-Add-exit-function-check-for-Meson-build-and-wh.patch"
Content-Disposition: attachment; 
	filename="0001-libpq-Add-exit-function-check-for-Meson-build-and-wh.patch"
Content-Transfer-Encoding: base64
Content-ID: <f_mi5z31l60>
X-Attachment-Id: f_mi5z31l60
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--000000000000147bf40643f1b952--