Horton

Re: EINTR in ftruncate()

Thomas Munro <thomas.munro@gmail.com>

View thread View on postgr.es
From: Thomas Munro <thomas.munro@gmail.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Alvaro Herrera <alvherre@alvh.no-ip.org>, Andres Freund <andres@anarazel.de>, Pg Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-11-09T00:11:01Z
Lists: pgsql-hackers 
On Wed, Aug 17, 2022 at 7:51 AM Thomas Munro <thomas.munro@gmail.com> wrote:
> On Sat, Jul 16, 2022 at 5:18 PM Thomas Munro <thomas.munro@gmail.com> wrote:
> > On Sat, Jul 16, 2022 at 1:28 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> > > Thomas Munro <thomas.munro@gmail.com> writes:
> > > > On Fri, Jul 15, 2022 at 9:34 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> > > >> (Someday we oughta go ahead and make our Windows signal API look more
> > > >> like POSIX, as I suggested back in 2015.  I'm still not taking
> > > >> point on that, though.)
> > >
> > > > For the sigprocmask() part, here's a patch that passes CI.  Only the
> > > > SIG_SETMASK case is actually exercised by our current code, though.
> > >
> > > Passes an eyeball check, but I can't actually test it.
> >
> > Thanks.  Pushed.
> >
> > I'm not brave enough to try to write a replacement sigaction() yet,
> > but it does appear that we could rip more ugliness and inconsistencies
> > that way, eg sa_mask.
>
> Here's a draft patch that adds a minimal sigaction() implementation
> for Windows.  It doesn't implement stuff we're not using, for sample
> sa_sigaction functions, but it has the sa_mask feature so we can
> harmonize the stuff that I believe you were talking about.

Pushed.

As discussed before, a much better idea would probably be to use
latch-based wakeup instead of putting postmaster logic in signal
handlers, but in the meantime this gets rid of the extra
Windows-specific noise.

Commits

  1. Provide sigaction() for Windows.

  2. Emulate sigprocmask(), not sigsetmask(), on Windows.

  3. Make dsm_impl_posix_resize more future-proof.

  4. Don't clobber postmaster sigmask in dsm_impl_resize.

  5. Create a distinct wait event for POSIX DSM allocation.

  6. Remove redundant ftruncate() for POSIX DSM memory.

  7. Block signals while allocating DSM memory.

  8. Remove dsm_resize() and dsm_remap().

  9. XLOG (also known as WAL -:)) Bootstrap/Startup/Shutdown.