From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: rm@romanrm.net Received: from krantz.zx2c4.com (localhost [127.0.0.1]) by krantz.zx2c4.com (ZX2C4 Mail Server) with ESMTP id 4343ab08 for ; Tue, 10 Jul 2018 14:47:19 +0000 (UTC) Received: from rin.romanrm.net (rin.romanrm.net [IPv6:2001:41d0:1:8b3b::1]) by krantz.zx2c4.com (ZX2C4 Mail Server) with ESMTP id 762253db for ; Tue, 10 Jul 2018 14:47:16 +0000 (UTC) Date: Tue, 10 Jul 2018 19:54:36 +0500 From: Roman Mamedov To: "Jason A. Donenfeld" Subject: Re: [ANNOUNCE] WireGuard Snapshot `0.0.20180708` Available Message-ID: <20180710195436.357fb972@natsu> In-Reply-To: References: MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Cc: WireGuard mailing list List-Id: Development discussion of WireGuard List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Sun, 08 Jul 2018 18:52:32 +0200 "Jason A. Donenfeld" wrote: > * receive: use NAPI on the receive path > > This is a big change that should both improve preemption latency (by not > disabling it unconditionally) and vastly improve rx performance on most > systems by using NAPI. The main purpose of this snapshot is to test out this > technique. Just ran a few tests, it appears the performance is about 5-10% higher. Great work! Two VMs running on the same host (non-WG iperf3 is 14 Gbit/sec), typical results (upgrading receiver machine only): Single core: Before: 1.06 Gbit/sec After: 1.13 Gbit/sec Dual core: Before: 1.25 Gbit/sec After: 1.35 Gbit/sec Note that my "before" is a bit non-stock, but with a patch which removed two calls of "simd_relax" (as I wanted to keep max performance over the interactivity changes). "After" is the new snapshot without any patches. -- With respect, Roman