Em seg., 3 de fev. de 2025 às 01:07, Michael Paquier <michael@paquier.xyz> escreveu:
On Fri, Jan 31, 2025 at 11:29:31AM +0300, Nazir Bilal Yavuz wrote: > On Wed, 29 Jan 2025 at 18:16, Bertrand Drouvot > <bertranddrouvot.pg@gmail.com> wrote: >> I think that's the main reason why ff99918c625 added this new GUC (looking at >> the commit message). I'd feel more comfortable if we keep it. > > As Michael suggested, I will run a couple of benchmarks to see the > actual effect of this change. Then let's see if this affects anything.
I've looked at bit at all that today, and something like the attached is what seems like the best streamlined version to me for the main feature. I am also planning to run some short benchmarks with track_io_timing=on on HEAD and with the patch, then see the difference, without any relationship to track_wal_io_timing.
The comment additions in pgstat_count_io_op_time() were worth a patch of their own. This part has been applied as b998fedab74c, after a few tweaks of my own.
Sorry, I couldn't find the email in this thread, linked to the commit:
