Received: from malur.postgresql.org ([217.196.149.56]) by arkaria.postgresql.org with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1m6bXI-0004S0-GO for pgsql-performance@arkaria.postgresql.org; Thu, 22 Jul 2021 16:27:00 +0000 Received: from localhost ([127.0.0.1] helo=malur.postgresql.org) by malur.postgresql.org with esmtp (Exim 4.92) (envelope-from ) id 1m6bXH-0003xr-G2 for pgsql-performance@arkaria.postgresql.org; Thu, 22 Jul 2021 16:26:59 +0000 Received: from magus.postgresql.org ([2a02:c0:301:0:ffff::29]) by malur.postgresql.org with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1m6bXH-0003xe-7n for pgsql-performance@lists.postgresql.org; Thu, 22 Jul 2021 16:26:59 +0000 Received: from sss.pgh.pa.us ([66.207.139.130]) by magus.postgresql.org with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1m6bXE-0000zb-MN for pgsql-performance@postgresql.org; Thu, 22 Jul 2021 16:26:58 +0000 Received: from sss1.sss.pgh.pa.us (localhost [127.0.0.1]) by sss.pgh.pa.us (8.15.2/8.15.2) with ESMTP id 16MGQogo786911; Thu, 22 Jul 2021 12:26:50 -0400 From: Tom Lane To: David Rowley cc: "ldh@laurent-hasson.com" , Peter Geoghegan , Justin Pryzby , "pgsql-performance@postgresql.org" Subject: Re: Big performance slowdown from 11.2 to 13.3 In-reply-to: <786600.1626970898@sss.pgh.pa.us> References: <20210721191528.GC19620@telsasoft.com> <732758.1626911007@sss.pgh.pa.us> <784576.1626968702@sss.pgh.pa.us> <785218.1626969409@sss.pgh.pa.us> <786600.1626970898@sss.pgh.pa.us> Comments: In-reply-to Tom Lane message dated "Thu, 22 Jul 2021 12:21:38 -0400" MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-ID: <786909.1626971210.1@sss.pgh.pa.us> Date: Thu, 22 Jul 2021 12:26:50 -0400 Message-ID: <786910.1626971210@sss.pgh.pa.us> List-Id: List-Help: List-Subscribe: List-Post: List-Owner: List-Archive: Archived-At: Precedence: bulk I wrote: > I think the right fix here is to remove the cap, which will require > changing get_hash_mem to return double, and then maybe some cascading > changes --- I've not looked at its callers. Or, actually, returning size_t would likely make the most sense. We'd fold the 1024L multiplier in here too instead of doing that at the callers. regards, tom lane