On Fri, Jun 7, 2024 at 8:05 PM Alvaro Herrera <alvhe...@alvh.no-ip.org> wrote: > In passing, I noticed that WaitReadBuffers has zero comments, which > seems an insufficient number of them.
Ack. Here is a patch for that. I guess I hadn't put a comment there because it's hard to write anything without sort of duplicating what is already said by the StartReadBuffers() comment and doubling down on descriptions of future plans... but, in for a penny, in for a pound as they say.
From db5e56825e4b4c595885373c413011c50fedd3e8 Mon Sep 17 00:00:00 2001 From: Thomas Munro <thomas.munro@gmail.com> Date: Sun, 9 Jun 2024 09:51:02 +1200 Subject: [PATCH] Add comment for WaitReadBuffers(). Per complaint from Alvaro while reviewing something else. Reported-by: Alvaro Herrera <alvherre@alvh.no-ip.org> Discussion: https://postgr.es/m/202406070805.icofqromovvn%40alvherre.pgsql --- src/backend/storage/buffer/bufmgr.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c index a8e3377263f..409210a6ed2 100644 --- a/src/backend/storage/buffer/bufmgr.c +++ b/src/backend/storage/buffer/bufmgr.c @@ -1371,6 +1371,17 @@ WaitReadBuffersCanStartIO(Buffer buffer, bool nowait) return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait); } +/* + * The second step of a StartReadBuffers(), WaitReadBuffers() sequence. It is + * only necessary to call this if StartReadBuffers() returned true, indicating + * that I/O was necessary. + * + * Currently, this function performs synchronous I/O system calls. The earlier + * StartReadBuffers() call might have issued advice to give the OS a head + * start so that it completes quickly. In future work, a true asynchronous I/O + * could be started by StartReadBuffers(), and this function would wait for it + * to complete. + */ void WaitReadBuffers(ReadBuffersOperation *operation) { -- 2.45.1
