Horton

Re: Atomic operations within spinlocks

Andres Freund <andres@anarazel.de>

View thread View on postgr.es
From: Andres Freund <andres@anarazel.de>
To: Robert Haas <robertmhaas@gmail.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2020-06-11T17:26:53Z
Lists: pgsql-hackers 
Hi,

On 2020-06-10 07:26:32 -0400, Robert Haas wrote:
> On Fri, Jun 5, 2020 at 8:19 PM Andres Freund <andres@anarazel.de> wrote:
> > Randomly noticed while looking at the code:
> >         uint64          flagbit = UINT64CONST(1) << (uint64) type;
> >
> > that shouldn't be 64bit, right?
> 
> I'm going to admit ignorance here. What's the proper coding rule?

Well, pss_barrierCheckMask member is just 32bit, so it seems odd to
declare the local variable 64bit?

	uint64		flagbit = UINT64CONST(1) << (uint64) type;
...
		pg_atomic_fetch_or_u32(&slot->pss_barrierCheckMask, flagbit);


Greetings,

Andres Freund

Commits

  1. Convert SpinLock* macros to static inline functions.

  2. Clean up includes of s_lock.h.

  3. Fix deadlock danger when atomic ops are done under spinlock.

  4. Add basic spinlock tests to regression tests.

  5. spinlock emulation: Fix bug when more than INT_MAX spinlocks are initialized.

  6. Avoid potential spinlock in a signal handler as part of global barriers.

  7. Make pg_stat_wal_receiver consistent with the WAL receiver's shmem info

  8. Improve 64bit atomics support.