From owner-freebsd-xen@freebsd.org Mon Feb 13 07:58:24 2017 Return-Path: Delivered-To: freebsd-xen@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id E3B92CDDD6A for ; Mon, 13 Feb 2017 07:58:24 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from kenobi.freebsd.org (kenobi.freebsd.org [IPv6:2001:1900:2254:206a::16:76]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id C88A4D1C for ; Mon, 13 Feb 2017 07:58:24 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from bugs.freebsd.org ([127.0.1.118]) by kenobi.freebsd.org (8.15.2/8.15.2) with ESMTP id v1D7wNnO023771 for ; Mon, 13 Feb 2017 07:58:24 GMT (envelope-from bugzilla-noreply@freebsd.org) From: bugzilla-noreply@freebsd.org To: freebsd-xen@FreeBSD.org Subject: [Bug 212681] I/O is slow for FreeBSD DOMu on XenServer Date: Mon, 13 Feb 2017 07:58:24 +0000 X-Bugzilla-Reason: AssignedTo X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: Base System X-Bugzilla-Component: kern X-Bugzilla-Version: 10.3-RELEASE X-Bugzilla-Keywords: performance X-Bugzilla-Severity: Affects Only Me X-Bugzilla-Who: rainer@ultra-secure.de X-Bugzilla-Status: New X-Bugzilla-Resolution: X-Bugzilla-Priority: --- X-Bugzilla-Assigned-To: freebsd-xen@FreeBSD.org X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: https://bugs.freebsd.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Feb 2017 07:58:25 -0000 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D212681 --- Comment #50 from rainer@ultra-secure.de --- Output from the xen-server: xe vm-list name-label=3Di-129-1591-VM params=3Dall uuid ( RO) : 30f354c0-6a14-0dea-2be2-070a38ca2fc0 name-label ( RW): i-129-1591-VM name-description ( RW): Template which allows VM installation from install media user-version ( RW): 1 is-a-template ( RW): false is-a-snapshot ( RO): false snapshot-of ( RO): snapshots ( RO):=20 snapshot-time ( RO): 19700101T00:00:00Z snapshot-info ( RO):=20 parent ( RO): children ( RO):=20 is-control-domain ( RO): false power-state ( RO): running memory-actual ( RO): 8589938688 memory-target ( RO): memory-overhead ( RO): 71303168 memory-static-max ( RW): 8589934592 memory-dynamic-max ( RW): 8589934592 memory-dynamic-min ( RW): 8589934592 memory-static-min ( RW): 8589934592 suspend-VDI-uuid ( RW): suspend-SR-uuid ( RW): VCPUs-params (MRW): weight: 51; cap: 160 VCPUs-max ( RW): 2 VCPUs-at-startup ( RW): 2 actions-after-shutdown ( RW): Destroy actions-after-reboot ( RW): Restart actions-after-crash ( RW): Destroy console-uuids (SRO): 502503d4-ace2-f4c5-d6f7-5b7750ab5b18 platform (MRW): timeoffset: -2; viridian: false; acpi= : 1; apic: true; pae: true; nx: true allowed-operations (SRO): changing_dynamic_range; hard_reboot; hard_shutdown; pause; snapshot current-operations (SRO):=20 blocked-operations (MRW):=20 allowed-VBD-devices (SRO): allowed-VIF-devices (SRO): possible-hosts ( RO): HVM-boot-policy ( RW): BIOS order HVM-boot-params (MRW): order: dc HVM-shadow-multiplier ( RW): 1.000 PV-kernel ( RW):=20 PV-ramdisk ( RW):=20 PV-args ( RW):=20 PV-legacy-args ( RW):=20 PV-bootloader ( RW):=20 PV-bootloader-args ( RW):=20 last-boot-CPU-flags ( RO): vendor: GenuineIntel; features: 77bee3ff-bfebfbff-00000001-2c100800 last-boot-record ( RO): resident-on ( RO): b2615dd7-9308-4e9c-912b-1f302920ea63 affinity ( RW): b2615dd7-9308-4e9c-912b-1f302920ea63 other-config (MRW): vgpu_pci: ; mac_seed: ed5148a2-d9cd-b180-697d-e9998e496ee1; install-methods: cdrom; vm_uuid: d61de64b-8c67-4c9c-82ab-71a197ff1530 dom-id ( RO): 52 recommendations ( RO): xenstore-data (MRW): vm-data:=20 ha-always-run ( RW) [DEPRECATED]: false ha-restart-priority ( RW):=20 blobs ( RO):=20 start-time ( RO): 20170211T14:23:10Z install-time ( RO): 19700101T00:00:00Z VCPUs-number ( RO): 2 VCPUs-utilisation (MRO): os-version (MRO): PV-drivers-version (MRO): PV-drivers-up-to-date ( RO): memory (MRO): disks (MRO): networks (MRO): other (MRO): live ( RO): guest-metrics-last-updated ( RO): cooperative ( RO) [DEPRECATED]: tags (SRW):=20 appliance ( RW): start-delay ( RW): 0 shutdown-delay ( RW): 0 order ( RW): 0 version ( RO): 0 generation-id ( RO):=20 hardware-platform-version ( RO): 0 --=20 You are receiving this mail because: You are the assignee for the bug.= From owner-freebsd-xen@freebsd.org Mon Feb 13 09:49:44 2017 Return-Path: Delivered-To: freebsd-xen@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 98AEECDDC41 for ; Mon, 13 Feb 2017 09:49:44 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from kenobi.freebsd.org (kenobi.freebsd.org [IPv6:2001:1900:2254:206a::16:76]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 7B2BB1FA0 for ; Mon, 13 Feb 2017 09:49:44 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from bugs.freebsd.org ([127.0.1.118]) by kenobi.freebsd.org (8.15.2/8.15.2) with ESMTP id v1D9ng9c025212 for ; Mon, 13 Feb 2017 09:49:44 GMT (envelope-from bugzilla-noreply@freebsd.org) From: bugzilla-noreply@freebsd.org To: freebsd-xen@FreeBSD.org Subject: [Bug 212681] I/O is slow for FreeBSD DOMu on XenServer Date: Mon, 13 Feb 2017 09:49:43 +0000 X-Bugzilla-Reason: AssignedTo X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: Base System X-Bugzilla-Component: kern X-Bugzilla-Version: 10.3-RELEASE X-Bugzilla-Keywords: performance X-Bugzilla-Severity: Affects Only Me X-Bugzilla-Who: royger@freebsd.org X-Bugzilla-Status: New X-Bugzilla-Resolution: X-Bugzilla-Priority: --- X-Bugzilla-Assigned-To: freebsd-xen@FreeBSD.org X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: https://bugs.freebsd.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Feb 2017 09:49:44 -0000 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D212681 --- Comment #51 from Roger Pau Monn=C3=83=C2=A9 --- (In reply to rainer from comment #49) So performance is slightly better with this patch? (IIRC you where getting 17M/s and with the patch you get 26M/s) --=20 You are receiving this mail because: You are the assignee for the bug.= From owner-freebsd-xen@freebsd.org Mon Feb 13 09:59:24 2017 Return-Path: Delivered-To: freebsd-xen@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id A31ACCDDE4C for ; Mon, 13 Feb 2017 09:59:24 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from kenobi.freebsd.org (kenobi.freebsd.org [IPv6:2001:1900:2254:206a::16:76]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 88339400 for ; Mon, 13 Feb 2017 09:59:24 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from bugs.freebsd.org ([127.0.1.118]) by kenobi.freebsd.org (8.15.2/8.15.2) with ESMTP id v1D9xObe050325 for ; Mon, 13 Feb 2017 09:59:24 GMT (envelope-from bugzilla-noreply@freebsd.org) From: bugzilla-noreply@freebsd.org To: freebsd-xen@FreeBSD.org Subject: [Bug 212681] I/O is slow for FreeBSD DOMu on XenServer Date: Mon, 13 Feb 2017 09:59:24 +0000 X-Bugzilla-Reason: AssignedTo X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: Base System X-Bugzilla-Component: kern X-Bugzilla-Version: 10.3-RELEASE X-Bugzilla-Keywords: performance X-Bugzilla-Severity: Affects Only Me X-Bugzilla-Who: rainer@ultra-secure.de X-Bugzilla-Status: New X-Bugzilla-Resolution: X-Bugzilla-Priority: --- X-Bugzilla-Assigned-To: freebsd-xen@FreeBSD.org X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: https://bugs.freebsd.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Feb 2017 09:59:24 -0000 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D212681 --- Comment #52 from rainer@ultra-secure.de --- Yes. So, 60%-70% increase. --=20 You are receiving this mail because: You are the assignee for the bug.= From owner-freebsd-xen@freebsd.org Mon Feb 13 10:14:50 2017 Return-Path: Delivered-To: freebsd-xen@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 60BD2CDC410 for ; Mon, 13 Feb 2017 10:14:50 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from kenobi.freebsd.org (kenobi.freebsd.org [IPv6:2001:1900:2254:206a::16:76]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 5051011E3 for ; Mon, 13 Feb 2017 10:14:50 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from bugs.freebsd.org ([127.0.1.118]) by kenobi.freebsd.org (8.15.2/8.15.2) with ESMTP id v1DAEnjF007574 for ; Mon, 13 Feb 2017 10:14:50 GMT (envelope-from bugzilla-noreply@freebsd.org) From: bugzilla-noreply@freebsd.org To: freebsd-xen@FreeBSD.org Subject: [Bug 212681] I/O is slow for FreeBSD DOMu on XenServer Date: Mon, 13 Feb 2017 10:14:50 +0000 X-Bugzilla-Reason: AssignedTo X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: Base System X-Bugzilla-Component: kern X-Bugzilla-Version: 10.3-RELEASE X-Bugzilla-Keywords: performance X-Bugzilla-Severity: Affects Only Me X-Bugzilla-Who: royger@freebsd.org X-Bugzilla-Status: New X-Bugzilla-Resolution: X-Bugzilla-Priority: --- X-Bugzilla-Assigned-To: freebsd-xen@FreeBSD.org X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: https://bugs.freebsd.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Feb 2017 10:14:50 -0000 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D212681 --- Comment #53 from Roger Pau Monn=C3=83=C2=A9 --- I'm attaching another patch that will allow to selectively disable some PV optimizations, you will have to play with the following tunables, and see if you can find which one(s) causes the VM to go faster: hw.xen.disable_pv_ipi hw.xen.disable_pv_et hw.xen.disable_pv_tc hw.xen.disable_pv_clk hw.xen.disable_pv_disks hw.xen.disable_pv_nics Those tunables should be set in /boot/loader.conf, like: hw.xen.disable_pv_ipi=3D1 Thanks, Roger. --=20 You are receiving this mail because: You are the assignee for the bug.= From owner-freebsd-xen@freebsd.org Mon Feb 13 10:15:24 2017 Return-Path: Delivered-To: freebsd-xen@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 1D952CDC43D for ; Mon, 13 Feb 2017 10:15:24 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from kenobi.freebsd.org (kenobi.freebsd.org [IPv6:2001:1900:2254:206a::16:76]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 0D4ED1234 for ; Mon, 13 Feb 2017 10:15:24 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from bugs.freebsd.org ([127.0.1.118]) by kenobi.freebsd.org (8.15.2/8.15.2) with ESMTP id v1DAFNQx008895 for ; Mon, 13 Feb 2017 10:15:23 GMT (envelope-from bugzilla-noreply@freebsd.org) From: bugzilla-noreply@freebsd.org To: freebsd-xen@FreeBSD.org Subject: [Bug 212681] I/O is slow for FreeBSD DOMu on XenServer Date: Mon, 13 Feb 2017 10:15:23 +0000 X-Bugzilla-Reason: AssignedTo X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: Base System X-Bugzilla-Component: kern X-Bugzilla-Version: 10.3-RELEASE X-Bugzilla-Keywords: performance X-Bugzilla-Severity: Affects Only Me X-Bugzilla-Who: royger@freebsd.org X-Bugzilla-Status: New X-Bugzilla-Resolution: X-Bugzilla-Priority: --- X-Bugzilla-Assigned-To: freebsd-xen@FreeBSD.org X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: attachments.created Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: https://bugs.freebsd.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Feb 2017 10:15:24 -0000 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D212681 --- Comment #54 from Roger Pau Monn=C3=83=C2=A9 --- Created attachment 179940 --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=3D179940&action= =3Dedit Selectively disable PV optimizations --=20 You are receiving this mail because: You are the assignee for the bug.= From owner-freebsd-xen@freebsd.org Mon Feb 13 13:19:51 2017 Return-Path: Delivered-To: freebsd-xen@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 48B9FCDBA4F for ; Mon, 13 Feb 2017 13:19:51 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from kenobi.freebsd.org (kenobi.freebsd.org [IPv6:2001:1900:2254:206a::16:76]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 3885C1A1B for ; Mon, 13 Feb 2017 13:19:51 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from bugs.freebsd.org ([127.0.1.118]) by kenobi.freebsd.org (8.15.2/8.15.2) with ESMTP id v1DDJpla020454 for ; Mon, 13 Feb 2017 13:19:51 GMT (envelope-from bugzilla-noreply@freebsd.org) From: bugzilla-noreply@freebsd.org To: freebsd-xen@FreeBSD.org Subject: [Bug 212681] I/O is slow for FreeBSD DOMu on XenServer Date: Mon, 13 Feb 2017 13:19:51 +0000 X-Bugzilla-Reason: AssignedTo X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: Base System X-Bugzilla-Component: kern X-Bugzilla-Version: 10.3-RELEASE X-Bugzilla-Keywords: performance X-Bugzilla-Severity: Affects Only Me X-Bugzilla-Who: rainer@ultra-secure.de X-Bugzilla-Status: New X-Bugzilla-Resolution: X-Bugzilla-Priority: --- X-Bugzilla-Assigned-To: freebsd-xen@FreeBSD.org X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: https://bugs.freebsd.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Feb 2017 13:19:51 -0000 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D212681 --- Comment #55 from rainer@ultra-secure.de --- I can't see these tunables in sysctl. But: hw.xen.disable_pv_disks=3D1 is responsible for the slight increases in disk-performance. --=20 You are receiving this mail because: You are the assignee for the bug.= From owner-freebsd-xen@freebsd.org Mon Feb 13 13:54:57 2017 Return-Path: Delivered-To: freebsd-xen@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id ADE41CDD778 for ; Mon, 13 Feb 2017 13:54:57 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from kenobi.freebsd.org (kenobi.freebsd.org [IPv6:2001:1900:2254:206a::16:76]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 9D8B61223 for ; Mon, 13 Feb 2017 13:54:57 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from bugs.freebsd.org ([127.0.1.118]) by kenobi.freebsd.org (8.15.2/8.15.2) with ESMTP id v1DDsu5M012400 for ; Mon, 13 Feb 2017 13:54:57 GMT (envelope-from bugzilla-noreply@freebsd.org) From: bugzilla-noreply@freebsd.org To: freebsd-xen@FreeBSD.org Subject: [Bug 212681] I/O is slow for FreeBSD DOMu on XenServer Date: Mon, 13 Feb 2017 13:54:57 +0000 X-Bugzilla-Reason: AssignedTo X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: Base System X-Bugzilla-Component: kern X-Bugzilla-Version: 10.3-RELEASE X-Bugzilla-Keywords: performance X-Bugzilla-Severity: Affects Only Me X-Bugzilla-Who: royger@freebsd.org X-Bugzilla-Status: New X-Bugzilla-Resolution: X-Bugzilla-Priority: --- X-Bugzilla-Assigned-To: freebsd-xen@FreeBSD.org X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: https://bugs.freebsd.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Feb 2017 13:54:57 -0000 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D212681 --- Comment #56 from Roger Pau Monn=C3=83=C2=A9 --- (In reply to rainer from comment #55) Yes, you won't see those tunables in sysctl. Then again I'm quite lost, because you did test a plain dd, and that was actually working fine (and yielding results in line with Linux). Can you tr= y a more complete benchmark, like unixbench (available in ports) without any tunables and report it's results? --=20 You are receiving this mail because: You are the assignee for the bug.= From owner-freebsd-xen@freebsd.org Mon Feb 13 14:39:30 2017 Return-Path: Delivered-To: freebsd-xen@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 8743DCDD08A for ; Mon, 13 Feb 2017 14:39:30 +0000 (UTC) (envelope-from alexander.nusov@nfvexpress.com) Received: from sender-of-o52.zoho.com (sender-of-o52.zoho.com [135.84.80.217]) (using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 40931B9A; Mon, 13 Feb 2017 14:39:29 +0000 (UTC) (envelope-from alexander.nusov@nfvexpress.com) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; t=1486996756; s=zoho; d=nfvexpress.com; i=alexander.nusov@nfvexpress.com; h=Date:From:To:Cc:Message-Id:In-Reply-To:References:Subject:MIME-Version:Content-Type; l=2904; bh=VPcyrZF408L+z1TDZ1eau+Dde4cwSwn/n2+z6PJy9pk=; b=cL7SCZNWc2i3bS/IwdVCUxAfPw8DJwkDJ/IlxKt8WYCahiNjqi3DrVpj+dwMdIgC /nkW69TXhGNZ28FO4YUSnHEs11J5msv+WnRQRkeZ7TWo2CmfCQ7ij2jRS8UBrCjv0BN RIsXcfdjaQiwdqhykXAxL8u8hZJ5WHVnRV2XIUEc= Received: from mail.zoho.com by mx.zohomail.com with SMTP id 1486996756207973.8292327939075; Mon, 13 Feb 2017 06:39:16 -0800 (PST) Date: Mon, 13 Feb 2017 06:39:16 -0800 From: Alexander Nusov To: Cc: , Message-Id: <15a37e9d6eb.11239b0492614.5094818896052410185@nfvexpress.com> In-Reply-To: <20170203101402.dto7lzj2sldvuxl5@MacBook-Pro-de-Roger.local> References: <20170124114444.xdl3qj35lwebkso7@dhcp-3-221.uk.xensource.com> <159d0f04b55.10bbf935114648.7927688075504705395@nfvexpress.com> <20170124165621.iidjypfoyp4ccysi@dhcp-3-221.uk.xensource.com> <159d55b6b65.dc9b3b1720453.5847395326743424395@nfvexpress.com> <20170125115051.sasgqxwgt47p7pwi@dhcp-3-221.uk.xensource.com> <159d57da8bb.e9fea5dc25161.6113276104371162720@nfvexpress.com> <20170203101402.dto7lzj2sldvuxl5@MacBook-Pro-de-Roger.local> Subject: Re: Xen on FreeBSD 11: Cannot boot from QCOW2 properly (waiting for XENBUS, xenbusb_nop_confighook_cb) MIME-Version: 1.0 X-Priority: Normal User-Agent: Zoho Mail X-Mailer: Zoho Mail Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit X-Content-Filtered-By: Mailman/MimeDel 2.1.23 X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Feb 2017 14:39:30 -0000 Hi! Please notify me when the issue will be fixed so I could start backporting patches to 11.0 to test qcow backend. ---- On Fri, 03 Feb 2017 13:14:02 +0300 roger.pau@citrix.com wrote ---- On Thu, Feb 02, 2017 at 11:37:02PM +0000, Akshay Jaggi wrote: > [-Alex] > > Hi Roger, > > Did this get solved with the change you submitted to Xen? This issue is cased by a bug in the VM subsystem, kib is currently looking into it, and patches will be committed soon hopefully. The Xen patch I've sent is the FreeBSD handlers for the gnttab lib, with that everything is already upstream :). I will backport the Xen patch to your Xen package once the VM issue is solved. Thanks, Roger. _______________________________________________ freebsd-xen@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-xen To unsubscribe, send any mail to "freebsd-xen-unsubscribe@freebsd.org" From owner-freebsd-xen@freebsd.org Mon Feb 13 15:58:56 2017 Return-Path: Delivered-To: freebsd-xen@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 28A34CDDC52 for ; Mon, 13 Feb 2017 15:58:56 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from kenobi.freebsd.org (kenobi.freebsd.org [IPv6:2001:1900:2254:206a::16:76]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 186AAF1 for ; Mon, 13 Feb 2017 15:58:56 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from bugs.freebsd.org ([127.0.1.118]) by kenobi.freebsd.org (8.15.2/8.15.2) with ESMTP id v1DFwtaK051664 for ; Mon, 13 Feb 2017 15:58:55 GMT (envelope-from bugzilla-noreply@freebsd.org) From: bugzilla-noreply@freebsd.org To: freebsd-xen@FreeBSD.org Subject: [Bug 212681] I/O is slow for FreeBSD DOMu on XenServer Date: Mon, 13 Feb 2017 15:58:55 +0000 X-Bugzilla-Reason: AssignedTo X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: Base System X-Bugzilla-Component: kern X-Bugzilla-Version: 10.3-RELEASE X-Bugzilla-Keywords: performance X-Bugzilla-Severity: Affects Only Me X-Bugzilla-Who: rainer@ultra-secure.de X-Bugzilla-Status: New X-Bugzilla-Resolution: X-Bugzilla-Priority: --- X-Bugzilla-Assigned-To: freebsd-xen@FreeBSD.org X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: https://bugs.freebsd.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Feb 2017 15:58:56 -0000 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D212681 --- Comment #57 from rainer@ultra-secure.de --- Well, I did do dd test, but they only write on a filesystem. It was (back then) most likely on ZFS, with compression etc. that changed t= he results. Esp. if I just write zeros from /dev/null. That's why I switched to dc3dd because it does the same thing on Linux and = on FreeBSD and completely eliminates the filesystem (and caching) layer as wel= l as any other kinds of write-optimization. I've now created an empty UFS filesystem on my 50G volume and run unixbench= on it. --=20 You are receiving this mail because: You are the assignee for the bug.= From owner-freebsd-xen@freebsd.org Mon Feb 13 16:15:46 2017 Return-Path: Delivered-To: freebsd-xen@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 740F8CDD23C for ; Mon, 13 Feb 2017 16:15:46 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from kenobi.freebsd.org (kenobi.freebsd.org [IPv6:2001:1900:2254:206a::16:76]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 63EAC10E9 for ; Mon, 13 Feb 2017 16:15:46 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from bugs.freebsd.org ([127.0.1.118]) by kenobi.freebsd.org (8.15.2/8.15.2) with ESMTP id v1DGFj0h010211 for ; Mon, 13 Feb 2017 16:15:46 GMT (envelope-from bugzilla-noreply@freebsd.org) From: bugzilla-noreply@freebsd.org To: freebsd-xen@FreeBSD.org Subject: [Bug 212681] I/O is slow for FreeBSD DOMu on XenServer Date: Mon, 13 Feb 2017 16:15:46 +0000 X-Bugzilla-Reason: AssignedTo X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: Base System X-Bugzilla-Component: kern X-Bugzilla-Version: 10.3-RELEASE X-Bugzilla-Keywords: performance X-Bugzilla-Severity: Affects Only Me X-Bugzilla-Who: royger@freebsd.org X-Bugzilla-Status: New X-Bugzilla-Resolution: X-Bugzilla-Priority: --- X-Bugzilla-Assigned-To: freebsd-xen@FreeBSD.org X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: https://bugs.freebsd.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Feb 2017 16:15:46 -0000 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D212681 --- Comment #58 from Roger Pau Monn=C3=83=C2=A9 --- (In reply to rainer from comment #57) You can also use plain dd to write to a block device, just like you do with dc3dd. Can you actually also try if plain dd shows the same slowness with writing to a block device directly? --=20 You are receiving this mail because: You are the assignee for the bug.= From owner-freebsd-xen@freebsd.org Mon Feb 13 17:46:35 2017 Return-Path: Delivered-To: freebsd-xen@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id C6FB6CDDCB5 for ; Mon, 13 Feb 2017 17:46:35 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from kenobi.freebsd.org (kenobi.freebsd.org [IPv6:2001:1900:2254:206a::16:76]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id ADBBE1546 for ; Mon, 13 Feb 2017 17:46:35 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from bugs.freebsd.org ([127.0.1.118]) by kenobi.freebsd.org (8.15.2/8.15.2) with ESMTP id v1DHkZae050825 for ; Mon, 13 Feb 2017 17:46:35 GMT (envelope-from bugzilla-noreply@freebsd.org) From: bugzilla-noreply@freebsd.org To: freebsd-xen@FreeBSD.org Subject: [Bug 212681] I/O is slow for FreeBSD DOMu on XenServer Date: Mon, 13 Feb 2017 17:46:35 +0000 X-Bugzilla-Reason: AssignedTo X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: Base System X-Bugzilla-Component: kern X-Bugzilla-Version: 10.3-RELEASE X-Bugzilla-Keywords: performance X-Bugzilla-Severity: Affects Only Me X-Bugzilla-Who: rainer@ultra-secure.de X-Bugzilla-Status: New X-Bugzilla-Resolution: X-Bugzilla-Priority: --- X-Bugzilla-Assigned-To: freebsd-xen@FreeBSD.org X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: https://bugs.freebsd.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Feb 2017 17:46:35 -0000 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D212681 --- Comment #59 from rainer@ultra-secure.de --- BYTE UNIX Benchmarks (Version 4.1.0) System -- freebsd11 Start Benchmark Run: Mon Feb 13 17:28:56 CET 2017 3 interactive users. 5:28PM up 3:06, 3 users, load averages: 0.62, 0.69, 0.71 -r-xr-xr-x 1 root wheel 153744 Sep 27 18:03 /bin/sh /bin/sh: ELF 64-bit LSB executable, x86-64, version 1 (FreeBSD), dynamica= lly linked, interpreter /libexec/ld-elf.so.1, for FreeBSD 11.0 (1100122), FreeBSD-style, stripped zroot/ROOT/default 11262956 6353528 4909428 56% / File Read 4096 bufsize 8000 maxblocks 2313874.0 KBps (30.0 secs, 3 samp= les) File Write 4096 bufsize 8000 maxblocks 120869.0 KBps (30.0 secs, 3 sampl= es) File Copy 4096 bufsize 8000 maxblocks 126921.0 KBps (30.0 secs, 3 sampl= es) Shell Scripts (1 concurrent) 2137.7 lpm (60.0 secs, 3 sampl= es) Shell Scripts (8 concurrent) 314.1 lpm (60.0 secs, 3 sampl= es) Shell Scripts (16 concurrent) 154.5 lpm (60.0 secs, 3 sampl= es) Arithmetic Test (type =3D short) 1.0 lps (0.0 secs, 3 samp= les) Arithmetic Test (type =3D int) 1.0 lps (0.0 secs, 3 samp= les) Arithmetic Test (type =3D long) 1.0 lps (0.0 secs, 3 samp= les) Arithmetic Test (type =3D float) 1.0 lps (0.0 secs, 3 samp= les) Arithmetic Test (type =3D double) 1.0 lps (0.0 secs, 3 samp= les) Arithoh 1.0 lps (0.0 secs, 3 sample= s) C Compiler Throughput 499.1 lpm (60.0 secs, 3 sampl= es) Dc: sqrt(2) to 99 decimal places 78631.1 lpm (30.0 secs, 3 sampl= es) Recursion Test--Tower of Hanoi 192648.9 lps (20.0 secs, 3 sampl= es) INDEX VALUES=20=20=20=20=20=20=20=20=20=20=20=20 TEST BASELINE RESULT INDEX File Copy 4096 bufsize 8000 maxblocks 5800.0 126921.0 218.8 Shell Scripts (8 concurrent) 6.0 314.1 523.5 =3D=3D=3D= =3D=3D=3D=3D=3D=3D FINAL SCORE 338.5 on a 50G UFS partition. --=20 You are receiving this mail because: You are the assignee for the bug.= From owner-freebsd-xen@freebsd.org Wed Feb 15 12:08:17 2017 Return-Path: Delivered-To: freebsd-xen@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id C3F5DCE0B30 for ; Wed, 15 Feb 2017 12:08:17 +0000 (UTC) (envelope-from liuyingdong@huawei.com) Received: from szxga01-in.huawei.com (szxga01-in.huawei.com [58.251.152.64]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "myname.my.domain", Issuer "www.mirapoint.com" (not verified)) by mx1.freebsd.org (Postfix) with ESMTPS id 81B548D5 for ; Wed, 15 Feb 2017 12:08:13 +0000 (UTC) (envelope-from liuyingdong@huawei.com) Received: from 172.24.1.47 (EHLO SZXEML423-HUB.china.huawei.com) ([172.24.1.47]) by szxrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id DZA28921; Wed, 15 Feb 2017 20:06:46 +0800 (CST) Received: from [127.0.0.1] (10.52.5.141) by SZXEML423-HUB.china.huawei.com (10.82.67.154) with Microsoft SMTP Server id 14.3.235.1; Wed, 15 Feb 2017 20:06:42 +0800 Subject: =?UTF-8?Q?Re:_Multiple_patch_review_=28was:_Re:_=e7=ad=94=e5=a4=8d:?= =?UTF-8?Q?_=e7=ad=94=e5=a4=8d:_[PATCH]netfront:_need_release_all_resources?= =?UTF-8?Q?=29_after_adding_and_removing_NICs_time_and_again?= To: , , <"suoben@huawei.com; zhao.zhaojun@huawei.com; wanglinkai@huawei.com; leo.gaoxiaodong@huawei.com; chuzhaosong@huawei.com; chunfeng.wang"@huawei.com> References: <1676c942-84ec-e573-42e9-8debd47aeb1b@huawei.com> From: liuyingdong Message-ID: Date: Wed, 15 Feb 2017 14:06:36 +0200 User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:45.0) Gecko/20100101 Thunderbird/45.7.1 MIME-Version: 1.0 In-Reply-To: <1676c942-84ec-e573-42e9-8debd47aeb1b@huawei.com> Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8bit X-Originating-IP: [10.52.5.141] X-CFilter-Loop: Reflected X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A020203.58A44457.0257, ss=1, re=0.000, recu=0.000, reip=0.000, cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32 X-Mirapoint-Loop-Id: 2e888baaf47801025f90c7958475e701 X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 15 Feb 2017 12:08:17 -0000 On 2017/2/13 12:05, liuyingdong wrote: > 发件人: roger.pau@citrix.com [mailto:roger.pau@citrix.com] > 发送时间: 2017年2月10日 17:12 > 收件人: Liuyingdong > 抄送: freebsd-xen@freebsd.org; Suoben ; Zhaojun (Euler) ; Wanglinkai ; chuzhaosong ; Wangchunfeng (Ivan) ; Gaoxiaodong (Leo) > 主题: Multiple patch review (was: Re: 答复: 答复: [PATCH]netfront: need release all resources) after adding and removing NICs time and again > > On Tue, Feb 07, 2017 at 04:55:34PM +0000, Liuyingdong wrote: >> Hi Roger, >> I am so sorry and please review tne below URL: >> https://lists.freebsd.org/pipermail/freebsd-xen/2017-February/002957.h >> tml > > Hello, > > Thanks for the patches, and sorry for the delay. It seems like you have not applied some of my comments, so I will have to re-post them here. If some of the comments don't apply for whatever reason, please reply back and explain why, or else this is not going to progress in an useful way for any of us. > > I would also request you to look into using `git send-email`, reviewing your patches as attachments is not very comfortable. Or else, you could create an account to https://reviews.freebsd.org/ and upload the patches there assigning me as a reviewer. > I have created an account to https://reviews.freebsd.org/ and uploaded the patches but I cann't assign you as a reviewer so I assign visible to All Users. The three modified patches are as follows: 1.https://reviews.freebsd.org/differential/diff/25207/ 2.https://reviews.freebsd.org/differential/diff/25208/ 3.https://reviews.freebsd.org/differential/diff/25209/ > When replying, can you also make sure that you reply in-line to my comments, or else it's very hard to follow the conversation. > ok > --- >> From 0af05ac01be853340b3b53862de9853d8d44c0c6 Mon Sep 17 00:00:00 2001 >> From: liuyingdong >> Date: Thu, 2 Feb 2017 19:40:48 +0200 >> Subject: introduce suspend_cancel mechanism for frontend devices. >> 1.1 On a cancelled suspend, xen frontend devices need know their state is invariant. >> 1.2 On a cancelled suspend the vcpu_info location does not change >> (it's still in the per-cpu area registered by xen_hvm_cpu_init()). >> So do not call xen_hvm_init_shared_info_page() which would make the kernel think its back in the shared info. >> With the wrong vcpu_info, events cannot be received and the domain will hang after a cancelled suspend. >> >> --- >> dev/xen/blkfront/blkfront.c | 5 +++++ >> dev/xen/control/control.c | 11 +++++++++-- >> dev/xen/netfront/netfront.c | 21 +++++++++++++++++++++ >> x86/xen/hvm.c | 16 ++++++++++------ >> xen/xenbus/xenbusb.c | 35 +++++++++++++++++++---------------- >> xen/xenbus/xenbusvar.h | 2 ++ >> 6 files changed, 66 insertions(+), 24 deletions(-) >> >> diff --git a/dev/xen/blkfront/blkfront.c b/dev/xen/blkfront/blkfront.c >> index 9eca220..47cd83f 100644 >> --- a/dev/xen/blkfront/blkfront.c >> +++ b/dev/xen/blkfront/blkfront.c >> @@ -1537,6 +1537,11 @@ xbd_resume(device_t dev) { >> struct xbd_softc *sc = device_get_softc(dev); >> >> + if(g_suspend_cancelled) { >> + sc->xbd_state = XBD_STATE_CONNECTED; >> + return (0); >> + } >> + >> DPRINTK("xbd_resume: %s\n", xenbus_get_node(dev)); >> >> xbd_free(sc); >> diff --git a/dev/xen/control/control.c b/dev/xen/control/control.c >> index 58fefcc..4f8471f 100644 >> --- a/dev/xen/control/control.c >> +++ b/dev/xen/control/control.c >> @@ -190,6 +190,11 @@ xctrl_reboot() >> shutdown_nice(0); >> } >> >> +static void _set_suspend_cancelled(bool suspend_cancelled) { >> + g_suspend_cancelled = suspend_cancelled; } >> + >> static void >> xctrl_suspend() >> { >> @@ -278,7 +283,9 @@ xctrl_suspend() >> /* >> * Reset grant table info. >> */ >> - gnttab_resume(NULL); >> + if(suspend_cancelled == 0) { >> + gnttab_resume(NULL); >> + } >> >> #ifdef SMP >> if (!CPU_EMPTY(&cpu_suspend_map)) { >> @@ -296,6 +303,7 @@ xctrl_suspend() >> * FreeBSD really needs to add DEVICE_SUSPEND_CANCEL or >> * similar. >> */ >> + _set_suspend_cancelled(suspend_cancelled != 0); > > No need for this helper, you can just s/suspend_cancelled/xen_suspend_called/ > and make it global. > ok, I agree with you. >> DEVICE_RESUME(root_bus); >> mtx_unlock(&Giant); >> >> @@ -319,7 +327,6 @@ xctrl_suspend() >> #endif >> >> resume_all_proc(); >> - > > Stray change. > ok, I agree with you. >> EVENTHANDLER_INVOKE(power_resume); >> >> if (bootverbose) >> diff --git a/dev/xen/netfront/netfront.c b/dev/xen/netfront/netfront.c >> index 459712a..537018a 100644 >> --- a/dev/xen/netfront/netfront.c >> +++ b/dev/xen/netfront/netfront.c >> @@ -153,6 +153,10 @@ static int xn_get_responses(struct netfront_rxq *, >> struct netfront_rx_info *, RING_IDX, RING_IDX *, >> struct mbuf **); >> >> +#ifdef INET >> +static void netfront_send_fake_arp(device_t dev, struct netfront_info >> +*info); #endif >> + >> #define virt_to_mfn(x) (vtophys(x) >> PAGE_SHIFT) >> >> #define INVALID_P2M_ENTRY (~0UL) >> @@ -440,6 +444,23 @@ netfront_resume(device_t dev) { >> struct netfront_info *info = device_get_softc(dev); >> >> + if(g_suspend_cancelled) { >> + u_int i; >> + for (i = 0; i < info->num_queues; i++) { >> + XN_RX_LOCK(&info->rxq[i]); >> + XN_TX_LOCK(&info->txq[i]); > > Bad indentation, you should use tabs (not spaces). > ok, I agree with you. >> + } >> + netfront_carrier_on(info); >> + for (i = 0; i < info->num_queues; i++) { >> + XN_RX_UNLOCK(&info->rxq[i]); >> + XN_TX_UNLOCK(&info->txq[i]); > > Bad indentation, you should use tabs (not spaces). > ok, I agree with you. >> + } >> +#ifdef INET >> + netfront_send_fake_arp(dev, info); >> +#endif > > I don't think you need to send an ARP on a cancelled suspend, the bridge should already have the mac address of the virtual interface in it's cache right? > For FreeBSD 10.X and FreeBSD 11.0, I have tested and found ping was not available if I don't send an ARP on a cancelled suspend. For 12.0-CURRENT,I don't think I need to send an ARP on a cancelled suspend. But I don't know why. >> + return (0); >> + } >> + >> netif_disconnect_backend(info); >> return (0); >> } >> diff --git a/x86/xen/hvm.c b/x86/xen/hvm.c index e10659e..c96b7be >> 100644 >> --- a/x86/xen/hvm.c >> +++ b/x86/xen/hvm.c >> @@ -297,10 +297,9 @@ xen_hvm_init(enum xen_hvm_init_type init_type) >> int error; >> int i; >> >> - if (init_type == XEN_HVM_INIT_CANCELLED_SUSPEND) >> - return; >> - >> - error = xen_hvm_init_hypercall_stubs(init_type); >> + if (init_type != XEN_HVM_INIT_CANCELLED_SUSPEND) { >> + error = xen_hvm_init_hypercall_stubs(init_type); >> + } >> >> switch (init_type) { >> case XEN_HVM_INIT_COLD: >> @@ -331,6 +330,8 @@ xen_hvm_init(enum xen_hvm_init_type init_type) >> CPU_FOREACH(i) >> DPCPU_ID_SET(i, vcpu_info, NULL); >> break; >> + case XEN_HVM_INIT_CANCELLED_SUSPEND: >> + break; >> default: >> panic("Unsupported HVM initialization type"); >> } >> @@ -344,7 +345,9 @@ xen_hvm_init(enum xen_hvm_init_type init_type) >> * is passed inside the start_info struct and is already set, so this >> * functions are no-ops. >> */ >> - xen_hvm_init_shared_info_page(); >> + if (init_type != XEN_HVM_INIT_CANCELLED_SUSPEND) { >> + xen_hvm_init_shared_info_page(); >> + } >> xen_hvm_disable_emulated_devices(); > > I'm not sure I follow why this change to xen_hvm_init is needed, on a cancelled suspend you shouldn't need to re-set the callback vector or unplug any emulated devices. > you are right,I need not reset the callback vector or unplug any emulated devices on a cancelled suspend. >> >> @@ -361,7 +364,8 @@ xen_hvm_resume(bool suspend_cancelled) >> XEN_HVM_INIT_CANCELLED_SUSPEND : XEN_HVM_INIT_RESUME); >> >> /* Register vcpu_info area for CPU#0. */ >> - xen_hvm_cpu_init(); >> + if(!suspend_cancelled) >> + xen_hvm_cpu_init(); > > Seeing this here, why don't we just avoid calling xen_hvm_resume from xctrl_suspend on a cancelled suspend? > I have avoided calling xen_hvm_resume on a cancelled suspend. >> >> static void >> diff --git a/xen/xenbus/xenbusb.c b/xen/xenbus/xenbusb.c index >> 3a0e543..fa9ba61 100644 >> --- a/xen/xenbus/xenbusb.c >> +++ b/xen/xenbus/xenbusb.c >> @@ -791,29 +791,32 @@ xenbusb_resume(device_t dev) >> if (device_get_state(kids[i]) == DS_NOTPRESENT) >> continue; >> >> - ivars = device_get_ivars(kids[i]); >> + if(!g_suspend_cancelled) { >> + ivars = device_get_ivars(kids[i]); >> >> - xs_unregister_watch(&ivars->xd_otherend_watch); >> - xenbus_set_state(kids[i], XenbusStateInitialising); >> + xs_unregister_watch(&ivars->xd_otherend_watch); >> + xenbus_set_state(kids[i], XenbusStateInitialising); >> >> - /* >> - * Find the new backend details and >> - * re-register our watch. >> - */ >> - error = XENBUSB_GET_OTHEREND_NODE(dev, ivars); >> - if (error) >> - return (error); >> + /* >> + * Find the new backend details and >> + * re-register our watch. >> + */ >> + error = XENBUSB_GET_OTHEREND_NODE(dev, ivars); >> + if (error) >> + return (error); >> >> - statepath = malloc(ivars->xd_otherend_path_len >> - + strlen("/state") + 1, M_XENBUS, M_WAITOK); >> - sprintf(statepath, "%s/state", ivars->xd_otherend_path); >> + statepath = malloc(ivars->xd_otherend_path_len >> + + strlen("/state") + 1, M_XENBUS, M_WAITOK); >> + sprintf(statepath, "%s/state", ivars->xd_otherend_path); >> >> - free(ivars->xd_otherend_watch.node, M_XENBUS); >> - ivars->xd_otherend_watch.node = statepath; >> + free(ivars->xd_otherend_watch.node, M_XENBUS); >> + ivars->xd_otherend_watch.node = statepath; >> + } >> >> DEVICE_RESUME(kids[i]); >> >> - xs_register_watch(&ivars->xd_otherend_watch); >> + if(!g_suspend_cancelled) >> + xs_register_watch(&ivars->xd_otherend_watch); >> #if 0 > > Why don't you just add: > > if (xenbusb_suspend_cancelled == 1) { > DEVICE_RESUME(kids[i]); > continue; > } > > To the top of the loop? This would prevent adding a bunch of nested conditionals. > I don't think so. DEVICE_RESUME(kids[i]) need be called whether the xenbusb_suspend_cancelled is 0 or 1. >> /* >> * Can't do this yet since we are running in diff --git >> a/xen/xenbus/xenbusvar.h b/xen/xenbus/xenbusvar.h index >> 377d60c..02a5bfa 100644 >> --- a/xen/xenbus/xenbusvar.h >> +++ b/xen/xenbus/xenbusvar.h >> @@ -99,6 +99,8 @@ XENBUS_ACCESSOR(state, STATE, enum xenbus_state) >> XENBUS_ACCESSOR(otherend_id, OTHEREND_ID, int) >> XENBUS_ACCESSOR(otherend_path, OTHEREND_PATH, const char *) >> >> +bool g_suspend_cancelled; > > This should be called xen_suspend_cancelled, and it would be better to place it in xen/xen-os.h instead. Also I'm not sure how that even compiles, isn't this missing an "extern" keyword in front? > defining a bool xen_suspend_cancelled need add an "extern" keyword in the front of xen/xen-os.h. >> + >> /** >> * Return the state of a XenBus device. >> * >> -- >> 2.11.0.windows.3 > --- >> From f4a0660ce77f72e0d0c08f3316ed4d8f78f1b8ab Mon Sep 17 00:00:00 2001 >> From: liuyingdong >> Date: Thu, 2 Feb 2017 18:51:08 +0200 >> Subject: If there is a user process which maybe often reads and writes >> xenstore, the application has been suspended after stop_all_proc is >> called. It held xs.request_mutex lock but in the following functions xs_write and xs_suspend will not get the lock. So the VM will hang. >> In order to prevent this from happening, this process need wait until >> stop_all_proc has finished during live migration. >> >> --- >> dev/xen/control/control.c | 2 ++ >> dev/xen/xenstore/xenstore.c | 31 +++++++++++++++++++++++++++++++ >> dev/xen/xenstore/xenstore_dev.c | 7 +++++++ >> xen/xenstore/xenstore_internal.h | 4 ++++ >> 4 files changed, 44 insertions(+) >> >> diff --git a/dev/xen/control/control.c b/dev/xen/control/control.c >> index ae13c6c..58fefcc 100644 >> --- a/dev/xen/control/control.c >> +++ b/dev/xen/control/control.c >> @@ -147,6 +147,7 @@ __FBSDID("$FreeBSD$"); #include >> >> >> #include >> +#include >> >> /*--------------------------- Forward Declarations >> --------------------------*/ >> /** Function signature for shutdown event handlers. */ @@ -199,7 >> +199,9 @@ xctrl_suspend() >> int suspend_cancelled; >> >> EVENTHANDLER_INVOKE(power_suspend_early); >> + xs_lock(); >> stop_all_proc(); >> + xs_unlock(); > > This seems fine, although I think that a better solution would be to simply don't use such kind of locks, > and allow concurrent access to the ring by properly using the xenstore ids, but that involves a non-trivial amount of code. > I think so but this solution is more simple. >> EVENTHANDLER_INVOKE(power_suspend); >> >> #ifdef EARLY_AP_STARTUP >> diff --git a/dev/xen/xenstore/xenstore.c b/dev/xen/xenstore/xenstore.c >> index 4f89b74..d44f064 100644 >> --- a/dev/xen/xenstore/xenstore.c >> +++ b/dev/xen/xenstore/xenstore.c >> @@ -1255,6 +1255,37 @@ xs_suspend(device_t dev) >> return (0); >> } >> >> +int >> +xs_try_lock(void) >> +{ >> + /* >> + sx_try_slock() and sx_try_xlock() will return 0 if the shared/exclusive >> + lock cannot be acquired immediately; otherwise the shared/exclusive lock >> + will be acquired and a non-zero value will be returned. >> + */ >> + return sx_try_xlock(&xs.request_mutex); } > > I don't think you need this at all, see below... > > [...] >> diff --git a/dev/xen/xenstore/xenstore_dev.c >> b/dev/xen/xenstore/xenstore_dev.c index ce62140..d6b97c0 100644 >> --- a/dev/xen/xenstore/xenstore_dev.c >> +++ b/dev/xen/xenstore/xenstore_dev.c >> @@ -128,6 +128,13 @@ xs_dev_write(struct cdev *dev, struct uio *uio, int ioflag) >> if (error != 0) >> return (error); >> >> + while(!xs_try_lock()) { >> + error = tsleep(u, PCATCH, "xsdwrite", hz/10); >> + if (error && error != EWOULDBLOCK) >> + return (error); >> + } >> + xs_unlock(); > > Why do you need this try_lock/unlock construction here? It is not protecting you against anything, the more that you don't perform any actual work with the lock held. > yes,I think so. I have replaced xs_try_lock with xs_is_lock: int xs_is_lock(void) { return sx_xlocked(&xs.request_mutex); } >> if ((len + u->len) > sizeof(u->u.buffer)) >> return (EINVAL); >> >> diff --git a/xen/xenstore/xenstore_internal.h >> b/xen/xenstore/xenstore_internal.h >> index 3355c27..31db935 100644 >> --- a/xen/xenstore/xenstore_internal.h >> +++ b/xen/xenstore/xenstore_internal.h >> @@ -34,3 +34,7 @@ >> >> /* Used by the XenStore character device to borrow kernel's store >> connection. */ int xs_dev_request_and_reply(struct xsd_sockmsg *msg, >> void **result); >> + >> +int xs_try_lock(void); >> +void xs_lock(void); >> +void xs_unlock(void); >> -- >> 2.11.0.windows.3 > --- >> From 78215c0b117fb7b651cf7e57a6a323407413ddde Mon Sep 17 00:00:00 2001 >> From: liuyingdong >> Date: Thu, 2 Feb 2017 19:52:33 +0200 >> Subject: Because wrong order of device resume, VM will hang after live migration. >> attach the Xen PV timer to the nexus directly as it was done before. >> >> --- >> dev/xen/timer/timer.c | 4 ++-- >> 1 file changed, 2 insertions(+), 2 deletions(-) >> >> diff --git a/dev/xen/timer/timer.c b/dev/xen/timer/timer.c index >> 0b26847..ae8f603 100644 >> --- a/dev/xen/timer/timer.c >> +++ b/dev/xen/timer/timer.c >> @@ -546,5 +546,5 @@ static driver_t xentimer_driver = { >> sizeof(struct xentimer_softc), >> }; >> >> -DRIVER_MODULE(xentimer, xenpv, xentimer_driver, xentimer_devclass, 0, >> 0); -MODULE_DEPEND(xentimer, xenpv, 1, 1, 1); >> +DRIVER_MODULE(xentimer, nexus, xentimer_driver, xentimer_devclass, 0, >> +0); MODULE_DEPEND(xentimer, nexus, 1, 1, 1); > > No, this is not how this should be solved. I understand that you might not want to implement the full solution proposed in freebsd-arch [0], but going back to Xen devices attaching directly to the nexus is also not an option IMHO. > > What I would accept is marking the Xen time-counter as not safe for suspension, so that FreeBSD itself will switch to a different timer when suspending. This simply means removing the TC_FLAGS_SUSPEND_SAFE from the timecounter flags. > Please add a "FIXME" comment, explaining why this is disabled, and when it could be enabled again. Could you please test that, and see if it solves your issues? > > Thanks, Roger > > [0] https://lists.freebsd.org/pipermail/freebsd-arch/2016-December/018041.html > I have tested that and find it may solve my issue. Thanks, Terry. From owner-freebsd-xen@freebsd.org Wed Feb 15 12:31:53 2017 Return-Path: Delivered-To: freebsd-xen@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 38468CDF235 for ; Wed, 15 Feb 2017 12:31:53 +0000 (UTC) (envelope-from prvs=21221c659=roger.pau@citrix.com) Received: from SMTP.EU.CITRIX.COM (smtp.eu.citrix.com [185.25.65.24]) (using TLSv1.2 with cipher RC4-SHA (128/128 bits)) (Client CN "mail.citrix.com", Issuer "DigiCert SHA2 Secure Server CA" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 89F071594 for ; Wed, 15 Feb 2017 12:31:51 +0000 (UTC) (envelope-from prvs=21221c659=roger.pau@citrix.com) X-IronPort-AV: E=Sophos;i="5.35,165,1484006400"; d="scan'208";a="40811483" Date: Wed, 15 Feb 2017 12:31:45 +0000 From: Roger Pau =?iso-8859-1?Q?Monn=E9?= To: liuyingdong CC: , , , , , , Subject: Re: Multiple patch review =?utf-8?Q?=28was?= =?utf-8?B?OiBSZTog562U5aSNOiDnrZTlpI06IFtQQVRDSF1uZXRmcm9udDogbmVlZCBy?= =?utf-8?Q?eleas?= =?utf-8?Q?e?= all resources) after adding and removing NICs time and again Message-ID: <20170215123145.ucobkwzft6jrgcsu@dhcp-3-221.uk.xensource.com> References: <1676c942-84ec-e573-42e9-8debd47aeb1b@huawei.com> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: User-Agent: NeoMutt/20170206 (1.7.2) X-ClientProxiedBy: AMSPEX02CAS02.citrite.net (10.69.22.113) To AMSPEX02CL02.citrite.net (10.69.22.126) X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 15 Feb 2017 12:31:53 -0000 On Wed, Feb 15, 2017 at 02:06:36PM +0200, liuyingdong wrote: > On 2017/2/13 12:05, liuyingdong wrote: > > 发件人: roger.pau@citrix.com [mailto:roger.pau@citrix.com] > > 发送时间: 2017年2月10日 17:12 > > 收件人: Liuyingdong > > 抄送: freebsd-xen@freebsd.org; Suoben ; Zhaojun (Euler) ; Wanglinkai ; chuzhaosong ; Wangchunfeng (Ivan) ; Gaoxiaodong (Leo) > > 主题: Multiple patch review (was: Re: 答复: 答复: [PATCH]netfront: need release all resources) after adding and removing NICs time and again > > > > On Tue, Feb 07, 2017 at 04:55:34PM +0000, Liuyingdong wrote: > >> Hi Roger, > >> I am so sorry and please review tne below URL: > >> https://lists.freebsd.org/pipermail/freebsd-xen/2017-February/002957.h > >> tml > > > > Hello, > > > > Thanks for the patches, and sorry for the delay. It seems like you have not applied some of my comments, so I will have to re-post them here. If some of the comments don't apply for whatever reason, please reply back and explain why, or else this is not going to progress in an useful way for any of us. > > > > I would also request you to look into using `git send-email`, reviewing your patches as attachments is not very comfortable. Or else, you could create an account to https://reviews.freebsd.org/ and upload the patches there assigning me as a reviewer. > > > I have created an account to https://reviews.freebsd.org/ and uploaded the patches but I cann't assign you as a reviewer so I assign visible to All Users. > The three modified patches are as follows: > 1.https://reviews.freebsd.org/differential/diff/25207/ > 2.https://reviews.freebsd.org/differential/diff/25208/ > 3.https://reviews.freebsd.org/differential/diff/25209/ You need to finish creating the revision, click on "Continue", and add a title/description and reviewers, see https://wiki.freebsd.org/Phabricator#Create_a_Revision_via_Web_Interface for more information. Roger. From owner-freebsd-xen@freebsd.org Thu Feb 16 18:54:00 2017 Return-Path: Delivered-To: freebsd-xen@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id D2DD6CE02D0 for ; Thu, 16 Feb 2017 18:54:00 +0000 (UTC) (envelope-from liuyingdong@huawei.com) Received: from szxga03-in.huawei.com (szxga03-in.huawei.com [119.145.14.66]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "myname.my.domain", Issuer "www.mirapoint.com" (not verified)) by mx1.freebsd.org (Postfix) with ESMTPS id BAF501B49 for ; Thu, 16 Feb 2017 18:53:59 +0000 (UTC) (envelope-from liuyingdong@huawei.com) Received: from 172.24.1.136 (EHLO szxeml426-hub.china.huawei.com) ([172.24.1.136]) by szxrg03-dlp.huawei.com (MOS 4.4.3-GA FastPath queued) with ESMTP id CPH00799; Fri, 17 Feb 2017 02:47:17 +0800 (CST) Received: from [127.0.0.1] (10.52.13.201) by szxeml426-hub.china.huawei.com (10.82.67.181) with Microsoft SMTP Server id 14.3.235.1; Fri, 17 Feb 2017 02:47:11 +0800 Subject: Re: Multiple patch review References: To: roger freebsd xen maintainer , From: liuyingdong Message-ID: Date: Thu, 16 Feb 2017 20:47:05 +0200 User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:45.0) Gecko/20100101 Thunderbird/45.7.1 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8bit X-Originating-IP: [10.52.13.201] X-CFilter-Loop: Reflected X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 16 Feb 2017 18:54:00 -0000 > 发件人: Roger Pau Monné [mailto:roger.pau@citrix.com] > 发送时间: 2017年2月15日 14:32 > 收件人: Liuyingdong > 抄送: freebsd-xen@freebsd.org; > 主题: Re: Multiple patch review (was: Re:[PATCH]netfront: need release all resources) after adding and removing NICs time and again > > On Wed, Feb 15, 2017 at 02:06:36PM +0200, liuyingdong wrote: >> On 2017/2/13 12:05, liuyingdong wrote: >>> 发件人: roger.pau@citrix.com [mailto:roger.pau@citrix.com] >>> 发送时间: 2017年2月10日 17:12 >>> 收件人: Liuyingdong >>> 抄送: freebsd-xen@freebsd.org; Suoben ; Zhaojun >>> (Euler) ; Wanglinkai >>> ; chuzhaosong ; >>> Wangchunfeng (Ivan) ; Gaoxiaodong (Leo) >>> >>> 主题: Multiple patch review (was: Re:[PATCH]netfront: need >>> release all resources) after adding and removing NICs time and again >>> >>> On Tue, Feb 07, 2017 at 04:55:34PM +0000, Liuyingdong wrote: >>>> Hi Roger, >>>> I am so sorry and please review tne below URL: >>>> https://lists.freebsd.org/pipermail/freebsd-xen/2017-February/002957.html >>> >>> Hello, >>> >>> Thanks for the patches, and sorry for the delay. It seems like you have not applied some of my comments, so I will have to re-post them here. If some of the comments don't apply for whatever reason, please reply back and explain why, or else this is not going to progress in an useful way for any of us. >>> >>> I would also request you to look into using `git send-email`, reviewing your patches as attachments is not very comfortable. Or else, you could create an account to https://reviews.freebsd.org/ and upload the patches there assigning me as a reviewer. >>> >> I have created an account to https://reviews.freebsd.org/ and uploaded the patches but I cann't assign you as a reviewer so I assign visible to All Users. >> The three modified patches are as follows: >> 1.https://reviews.freebsd.org/differential/diff/25207/ >> 2.https://reviews.freebsd.org/differential/diff/25208/ >> 3.https://reviews.freebsd.org/differential/diff/25209/ > > You need to finish creating the revision, click on "Continue", and add a title/description and reviewers, see https://wiki.freebsd.org/Phabricator#Create_a_Revision_via_Web_Interface for more information. > Thank you for the time and patience you devoted to guide me how to create a revision. I have created three revisions and please review again. https://reviews.freebsd.org/D9635 https://reviews.freebsd.org/D9638 https://reviews.freebsd.org/D9639 > Roger. > Thanks, Terry.