From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 40753C7EE22 for ; Mon, 15 May 2023 06:13:52 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235535AbjEOGNu (ORCPT ); Mon, 15 May 2023 02:13:50 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:48138 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231493AbjEOGNu (ORCPT ); Mon, 15 May 2023 02:13:50 -0400 Received: from dfw.source.kernel.org (dfw.source.kernel.org [IPv6:2604:1380:4641:c500::1]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 22B8CE1; Sun, 14 May 2023 23:13:49 -0700 (PDT) Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by dfw.source.kernel.org (Postfix) with ESMTPS id AA5C361F38; Mon, 15 May 2023 06:13:48 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id B5F42C433EF; Mon, 15 May 2023 06:13:47 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1684131228; bh=Y45E+N01pMi2CMlWB0B+OganKrfzdE6yeEf0dWAO4yU=; h=Date:From:To:Cc:Subject:References:In-Reply-To:From; b=R3Y89wxdF1A1pzNHqxrT7L4mihy9RdQllcqFtvj2Ja+ux0lx5o9lynlUAuoHXNvA3 wx9hwEOsJFlXM3ec3IRUpAat/xUwFv8w7TJPO2FPZI0xYfXnIwG2mjgpArZIx+Cf4u sZ3ZuILtK0r7DXqsvUsNH2lSLDipgk5anqaV/yljSxXaVSySMpLeL+BQC44hBnoqmD pcH/qj625BS6iurlQoDR/YMyW5D+3qi9RpSjSBs3R/Xw6jh7zN3PDw5h6zzwhz8b4p VEQ1t7MDdFQT8RxLIOHY2kRuoOX9DfhQjxUnT9yxY74Ovoc9/tWJC0BCHf1vKDW3Ts 2/xioqiq8OsAg== Date: Sun, 14 May 2023 23:13:46 -0700 From: Eric Biggers To: Kent Overstreet Cc: Lorenzo Stoakes , Christoph Hellwig , linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, linux-bcachefs@vger.kernel.org, Kent Overstreet , Andrew Morton , Uladzislau Rezki , linux-mm@kvack.org Subject: Re: [PATCH 07/32] mm: Bring back vmalloc_exec Message-ID: <20230515061346.GB15871@sol.localdomain> References: <20230509165657.1735798-8-kent.overstreet@linux.dev> <20230510064849.GC1851@quark.localdomain> <20230513015752.GC3033@quark.localdomain> <20230514184325.GB9528@sol.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Precedence: bulk List-ID: X-Mailing-List: linux-bcachefs@vger.kernel.org On Mon, May 15, 2023 at 01:38:51AM -0400, Kent Overstreet wrote: > On Sun, May 14, 2023 at 11:43:25AM -0700, Eric Biggers wrote: > > I think it would also help if the generated assembly had the handling of the > > fields interleaved. To achieve that, it might be necessary to interleave the C > > code. > > No, that has negligable effect on performance - as expected, for an out > of order processor. < 1% improvement. > > It doesn't look like this approach is going to work here. Sadly. I'd be glad to take a look at the code you actually tried. It would be helpful if you actually provided it, instead of just this "I tried it, I'm giving up now" sort of thing. I was also hoping you'd take the time to split this out into a userspace micro-benchmark program that we could quickly try different approaches on. BTW, even if people are okay with dynamic code generation (which seems unlikely?), you'll still need a C version for architectures that you haven't implemented the dynamic code generation for. - Eric