Horton

Re: Atomics for heap_parallelscan_nextpage()

Heikki Linnakangas <hlinnaka@iki.fi>

View thread View on postgr.es
From: Heikki Linnakangas <hlinnaka@iki.fi>
To: Andres Freund <andres@anarazel.de>, Tom Lane <tgl@sss.pgh.pa.us>
Cc: David Rowley <david.rowley@2ndquadrant.com>, PostgreSQL-development <pgsql-hackers@postgresql.org>, Amit Khandekar <amitdkhan.pg@gmail.com>, Ashutosh Bapat <ashutosh.bapat@enterprisedb.com>, Robert Haas <robertmhaas@gmail.com>
Date: 2017-08-16T17:19:29Z
Lists: pgsql-hackers

Attachments

On 08/16/2017 08:10 PM, Andres Freund wrote:
>> Afaict shm_create/shm_toc_allocate don't actually guarantee that the end
>> of the toc's memory is suitably aligned.  But I didn't yet have any
>> coffee, so ...
> 
> Robert, I'm not quite sure what the intended behaviour of shm_toc is wrt
> alignment. I see that individual chunks are BUFFERALIGNed (both during
> estimation, and allocation). But I don't see how the size of the entire
> toc is aligned, which seems a requirement, given we allocate from the
> end.
> Seems like we'd just have to add alignment of the total size to
> shm_toc_estimate()?

Yeah, that's the gist of it.

The attached patch seems to fix this. I'm not too familiar with this DSM 
stuff, but this seems right to me. Unless someone has a better idea 
soon, I'll commit this to make the buildfarm happy.

- Heikki

Commits

  1. Fix pg_atomic_u64 initialization.

  2. Fix shm_toc.c to always return buffer-aligned memory.

  3. Use atomic ops to hand out pages to scan in parallel scan.

  4. Improve 64bit atomics support.