Re: Refactoring the checkpointer's fsync request queue

Kevin Grittner <kgrittn@gmail.com>

From: Kevin Grittner <kgrittn@gmail.com>
To: Thomas Munro <thomas.munro@enterprisedb.com>
Cc: Dmitry Dolgov <9erthalion6@gmail.com>, Robert Haas <robertmhaas@gmail.com>, Andres Freund <andres@anarazel.de>, PostgreSQL-development <pgsql-hackers@postgresql.org>, Shawn Debnath <sdn@amazon.com>
Date: 2019-01-22T14:33:48Z
Lists: pgsql-hackers
On Tue, Jan 22, 2019 at 8:27 AM Kevin Grittner <kgrittn@gmail.com> wrote:

> With the help of VMware's Dirk Hohndel (VMware's Chief Open Source
> Officer, a VP position near the top of the organization, and a
> personal friend of Linus), I have been fortunate enough to make
> contact directly with Linus Torvalds to discuss this issue.  In emails
> to me he has told me that this patch is no longer provisional:
>
>   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=fff75eb2a08c2ac96404a2d79685668f3cf5a7a3
>
> Linus has given me permission to quote him, so here is a quote from an
> email he sent 2019-01-17:

> > That commit (b4678df184b3: "errseq: Always report a writeback error
> > once") was already backported to the stable trees (4.14 and 4.16), so
> > yes, everything should be fine. We did indeed miss old errors for a
> > while.

Sorry, but somehow I got the parent link rather that the intended
commit.  Linus got it right, of course.

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=b4678df184b314a2bd47d2329feca2c2534aa12b
> --
> Kevin Grittner
> VMware vCenter Server
> https://www.vmware.com/



--
Kevin Grittner
VMware vCenter Server
https://www.vmware.com/


Commits

  1. Fix bugs in mdsyncfiletag().

  2. Refactor the fsync queue for wider use.

  3. Don't forget about failed fsync() requests.

  4. PANIC on fsync() failure.

  5. Move LockClauseStrength, LockWaitPolicy into new file nodes/lockoptions.h.

  6. Add new file for checkpointer.c

  7. Split work of bgwriter between 2 processes: bgwriter and checkpointer.

  8. Install infrastructure for shared-memory free space map. Doesn't actually