From: | Andres Freund <andres(at)anarazel(dot)de> |
---|---|
To: | pgsql-hackers(at)postgresql(dot)org, Noah Misch <noah(at)leadboat(dot)com> |
Subject: | static libpq (and other libraries) overwritten on aix |
Date: | 2022-08-17 19:01:54 |
Message-ID: | 20220817190154.nnu27rvcleu2764m@awork3.anarazel.de |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Hi,
I was hacking in making aix work with the meson patchset last night when I
noticed this delightful bit:
gmake -C src/interfaces/libpq
...
rm -f libpq.a
ar crs libpq.a fe-auth-scram.o fe-connect.o fe-exec.o fe-lobj.o fe-misc.o fe-print.o fe-protocol3.o fe-secure.o fe-trace.o legacy-pqsignal.o libpq-events.o pqexpbuffer.o fe-auth.o
touch libpq.a
( echo '#! libpq.so.5'; gawk '/^[^#]/ {printf "%s\n",$1}' /home/andres/src/postgres/build-ac/../src/interfaces/libpq/exports.txt ) >libpq.exp
gcc -maix64 -Wall -Wmissing-prototypes -Wpointer-arith -Wdeclaration-after-statement -Werror=vla -Wendif-labels -Wmissing-format-attribute -Wimplicit-fallthrough=3 -Wformat-security -fno-strict-aliasing -fwrapv -fexcess-precision=standard -Wno-format-truncation -O2 -pthread -D_REENTRANT -D_THREAD_SAFE -o libpq.so.5 libpq.a -Wl,-bE:libpq.exp -L../../../src/port -L../../../src/common -lpgcommon_shlib -lpgport_shlib -Wl,-bbigtoc -Wl,-blibpath:'/usr/local/pgsql/lib:/usr/lib:/lib' -Wl,-bnoentry -Wl,-H512 -Wl,-bM:SRE -lm
rm -f libpq.a
ar crs libpq.a libpq.so.5
we first create a static library libpq.a as normal, but then we overwrite it
with the special aix way of packing up shared libraries, by packing them up in
a static library. That part is correct, it's apparently the easiest way of
getting applications to link to shared libraries on AIX (I think the
-Wl,-bM:SRE is relevant for ensuring it'll be a dynamic link, rather than a
static one).
This likely has been going on for approximately forever.
Two questions:
1) Do we continue building static libraries for libpq etc?
2) Do we care about static libraries not suriving on AIX? There could also be
a race in the buildrules leading to sometimes static libs sometimes shared
libs winning, I think.
Greetings,
Andres Freund
From | Date | Subject | |
---|---|---|---|
Next Message | Robert Haas | 2022-08-17 19:21:55 | Re: hash_xlog_split_allocate_page: failed to acquire cleanup lock |
Previous Message | Andres Freund | 2022-08-17 18:45:34 | Re: hash_xlog_split_allocate_page: failed to acquire cleanup lock |