I wrote:
> OK, I'll make it so. Thanks for looking at it!
Or not. My idea worked okay in v17, but not in older branches.
Pre-v17, libpq itself can call pqsignal (though only in non-
thread-safe builds). With this patch, that would have resulted
in pulling src/port/pqsignal.o into libpq. libpq
Nathan Bossart writes:
> On Tue, Jan 14, 2025 at 02:52:29PM -0500, Tom Lane wrote:
>> So now I'm inclined to include the ABI-compatible wrapper, which
>> will ensure that extensions continue to link to libpgport's pqsignal.
> Fine by me.
OK, I'll make it so. Thanks for looking at it!
On Tue, Jan 14, 2025 at 02:52:29PM -0500, Tom Lane wrote:
> After more thought I've realized that the asymmetrical detection
> here isn't all that bad, because the outcomes are different.
> If we fail to catch old-headers-and-new-library, the result will
> either be a link failure or (if the extens
Nathan Bossart writes:
> On Mon, Jan 13, 2025 at 05:51:54PM -0500, Tom Lane wrote:
>> (plus or minus an extern or so, but you get the idea). The point of
>> this is that compiling against old headers and then linking against
>> newer libpgport.a would still work. It does nothing however for the
On Mon, Jan 13, 2025 at 05:51:54PM -0500, Tom Lane wrote:
> It's fair to worry about this, but I don't think my testing that would
> prove a lot. AFAICS, whether somebody runs into trouble would depend
> on many factors like their specific build process and what versions of
> which packages they h
Nathan Bossart writes:
> On Sat, Jan 11, 2025 at 02:04:13PM -0500, Tom Lane wrote:
>> What I propose doing in the released branches is what's shown in
>> the attached patch for v17: rename port/pqsignal.c's function to
>> pqsignal_fe in frontend, but leave it as pqsignal in the backend.
>> Leaving
On Sat, Jan 11, 2025 at 02:04:13PM -0500, Tom Lane wrote:
> After studying this more, I think what we should do in HEAD is
> even more aggressive: let's make the real name of port/pqsignal.c's
> function be either pqsignal_fe in frontend, or pqsignal_be in backend.
> This positively ensures that th
I wrote:
> Actually, after re-reading the thread that led to 06843df4a [1],
> I think a better idea is to introduce some macro magic to force
> frontend clients to use libpgport's version of pqsignal() instead
> of the one from libpq. We mustn't change the real name of libpq's
> version, but I thi
I wrote:
> Since legacy-pqsignal is really not supposed to be used by clients
> anymore, maybe we could just adjust it to set SA_RESTART for SIGALRM?
> The other alternatives I can think of amount to re-introducing
> link order dependencies, which would be horrid.
Actually, after re-reading the th
Hi,
On Fri, 10 Jan 2025 at 17:58, Tom Lane wrote:
>
> Nazir Bilal Yavuz writes:
> > I ran it ~500 times on HEAD but the issue did not occur on my machine.
>
> What platform are you testing on?
My local machine is: Linux 6.12.8-amd64 #1 SMP PREEMPT_DYNAMIC Debian
6.12.8-1 (2025-01-02) x86_64 GNU
On 2025/01/10 21:41, Fujii Masao wrote:
On 2025/01/10 16:09, Andy Fan wrote:
Andy Fan writes:
Hi:
I run into the {subject} issue with the below setup.
cat foo.sql
\setshell txn_mode echo ${TXN_MODE}
\setshell speed echo ${SPEED}
\setshell sleep_ms echo ${SLEEP_MS}
\setshell subtxn_mod
Nazir Bilal Yavuz writes:
> I ran it ~500 times on HEAD but the issue did not occur on my machine.
What platform are you testing on?
regards, tom lane
Fujii Masao writes:
> Before this commit, pgbench used pqsignal() from port/pqsignal.c
> to set the signal handler for SIGALRM. This version of pqsignal()
> sets SA_RESTART for frontend code, so fgets() in runShellCommand()
> wouldn't return NULL even if SIGALRM arrived during fgets(),
> preventin
Hi,
On Fri, 10 Jan 2025 at 10:10, Andy Fan wrote:
>
> Andy Fan writes:
>
> > Hi:
> >
> > I run into the {subject} issue with the below setup.
> >
> > cat foo.sql
> >
> > \setshell txn_mode echo ${TXN_MODE}
> > \setshell speed echo ${SPEED}
> > \setshell sleep_ms echo ${SLEEP_MS}
> > \setshell su
On 2025/01/10 16:09, Andy Fan wrote:
Andy Fan writes:
Hi:
I run into the {subject} issue with the below setup.
cat foo.sql
\setshell txn_mode echo ${TXN_MODE}
\setshell speed echo ${SPEED}
\setshell sleep_ms echo ${SLEEP_MS}
\setshell subtxn_mode echo ${SUBTXN_MODE}
select 1;
$ TXN_MOD
Andy Fan writes:
> Hi:
>
> I run into the {subject} issue with the below setup.
>
> cat foo.sql
>
> \setshell txn_mode echo ${TXN_MODE}
> \setshell speed echo ${SPEED}
> \setshell sleep_ms echo ${SLEEP_MS}
> \setshell subtxn_mode echo ${SUBTXN_MODE}
>
> select 1;
>
> $ TXN_MODE=-1 SPEED=1 SLEEP_M
Hi:
I run into the {subject} issue with the below setup.
cat foo.sql
\setshell txn_mode echo ${TXN_MODE}
\setshell speed echo ${SPEED}
\setshell sleep_ms echo ${SLEEP_MS}
\setshell subtxn_mode echo ${SUBTXN_MODE}
select 1;
$ TXN_MODE=-1 SPEED=1 SLEEP_MS=0 SUBTXN_MODE=-1 pgbench -n -ffoo.sql
17 matches
Mail list logo
