From owner-freebsd-virtualization@freebsd.org Sun Apr 11 23:54:55 2021 Return-Path: Delivered-To: freebsd-virtualization@mailman.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.nyi.freebsd.org (Postfix) with ESMTP id 6A6D65E2CFE for ; Sun, 11 Apr 2021 23:54:55 +0000 (UTC) (envelope-from jtubnor@gmail.com) Received: from mail-wm1-f50.google.com (mail-wm1-f50.google.com [209.85.128.50]) (using TLSv1.3 with cipher TLS_AES_128_GCM_SHA256 (128/128 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256 client-signature RSA-PSS (2048 bits) client-digest SHA256) (Client CN "smtp.gmail.com", Issuer "GTS CA 1O1" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 4FJTHy5HKNz4mJH; Sun, 11 Apr 2021 23:54:54 +0000 (UTC) (envelope-from jtubnor@gmail.com) Received: by mail-wm1-f50.google.com with SMTP id 12so5770968wmf.5; Sun, 11 Apr 2021 16:54:54 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=eKTHHNNd2NnKtcWpd4wEu4WO6TdIRY8IYuBd86GA2eY=; b=PSiPZ+XDrUuuxa04FssKlRtIhgQohbW1Oa5RxZDB8NHdCPploGTcvDZFIAIkmjZjPk XaVKc9bZDiFKIyufopgZ7xD4lWJnYk7w6iZUw5YCi/T01VLT5WR6fdzMrAQdAhg8YaPg Ldl1X9r+1ERpNASV9o8oOI/riMzZddQXsXsUbf+vOOVtprFHrGKKgPzNBDsxm73ddAKe XfSHYrBlSehpqPoWPZNio0/ZdjB7pruTXlJNKfA3wa5CIWzXtp6lsWdsxvQg9H6YJoln uhpgiqKuBJLtK3K/PzyFHiItJHdbOgREm6FiT6w5qoPUpFNIas2oC2KGuf7tDHFT85Dk xrHw== X-Gm-Message-State: AOAM533Kz5ZCrm3dpOV+ld9l7LB6uluwSrr1IlL6bW8UV+scph0gNUmf MDnOlCY2IVbdh1ohsjQ0w/5V9pFl/R8NKjF26FTU2ypq X-Google-Smtp-Source: ABdhPJxzoS0kfb+m4hwmoPfzVvzvc4smnfAN3hft7G/D0R90+C2gXz5HFfY+xf7h4F39sVmwAkT2o2N2UvyEMyyuHXc= X-Received: by 2002:a1c:7fcd:: with SMTP id a196mr24149461wmd.180.1618185293397; Sun, 11 Apr 2021 16:54:53 -0700 (PDT) MIME-Version: 1.0 References: <7850c18aba62e6150f227f3c1168974c@userve.net> <4d863f34-6df0-0b0a-f487-e492324e8752@freebsd.org> In-Reply-To: From: Jason Tubnor Date: Mon, 12 Apr 2021 09:54:41 +1000 Message-ID: Subject: Re: bhyve current windows status To: Robert Crowston Cc: Peter Grehan , Matt Churchyard , FreeBSD virtualization X-Rspamd-Queue-Id: 4FJTHy5HKNz4mJH X-Spamd-Bar: + Authentication-Results: mx1.freebsd.org; dkim=none; dmarc=none; spf=pass (mx1.freebsd.org: domain of jtubnor@gmail.com designates 209.85.128.50 as permitted sender) smtp.mailfrom=jtubnor@gmail.com X-Spamd-Result: default: False [1.69 / 15.00]; RCVD_TLS_ALL(0.00)[]; FREEMAIL_ENVFROM(0.00)[gmail.com]; ARC_NA(0.00)[]; FROM_HAS_DN(0.00)[]; RCPT_COUNT_THREE(0.00)[4]; R_SPF_ALLOW(-0.20)[+ip4:209.85.128.0/17]; NEURAL_SPAM_SHORT(1.00)[1.000]; MIME_GOOD(-0.10)[multipart/alternative,text/plain]; DMARC_NA(0.00)[tubnor.net]; RBL_DBL_DONT_QUERY_IPS(0.00)[209.85.128.50:from]; SPAMHAUS_ZRD(0.00)[209.85.128.50:from:127.0.2.255]; TO_MATCH_ENVRCPT_SOME(0.00)[]; TO_DN_ALL(0.00)[]; NEURAL_SPAM_LONG(1.00)[1.000]; RCVD_IN_DNSWL_NONE(0.00)[209.85.128.50:from]; NEURAL_HAM_MEDIUM(-0.31)[-0.306]; FREEMAIL_TO(0.00)[protonmail.com]; FORGED_SENDER(0.30)[jason@tubnor.net,jtubnor@gmail.com]; RWL_MAILSPIKE_POSSIBLE(0.00)[209.85.128.50:from]; R_DKIM_NA(0.00)[]; MIME_TRACE(0.00)[0:+,1:+,2:~]; ASN(0.00)[asn:15169, ipnet:209.85.128.0/17, country:US]; FROM_NEQ_ENVFROM(0.00)[jason@tubnor.net,jtubnor@gmail.com]; MAILMAN_DEST(0.00)[freebsd-virtualization]; RCVD_COUNT_TWO(0.00)[2] Content-Type: text/plain; charset="UTF-8" X-Content-Filtered-By: Mailman/MimeDel 2.1.34 X-BeenThere: freebsd-virtualization@freebsd.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: "Discussion of various virtualization techniques FreeBSD supports." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 11 Apr 2021 23:54:55 -0000 On Sun, 11 Apr 2021 at 20:07, Robert Crowston wrote: > nvme is faster than virtio-blk? It seems strange that a paravirtualized > driver would be slower. Is that because of the regression you mention? > > > Not saying it is slower, just not any faster than the NVMe presentation that we have discovered in production. NVMe is the general direction storage is taking, doesn't require the drivers to be injected at installation time and lower usability friction. The regression was due to a bug in the virtio-stor code that would nuke your storage as soon as the driver loaded after upgrading it the following boot. That was discussed on a bhyve call last year. It is no longer an issue with the latest driver if you wish to continue to use virtio-stor. Cheers, Jason.