From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:41945) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Zw7M9-0000Ux-SG for qemu-devel@nongnu.org; Tue, 10 Nov 2015 06:45:02 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Zw7M5-0005S9-O4 for qemu-devel@nongnu.org; Tue, 10 Nov 2015 06:44:57 -0500 Received: from mx2.parallels.com ([199.115.105.18]:52256) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Zw7M5-0005RT-FB for qemu-devel@nongnu.org; Tue, 10 Nov 2015 06:44:53 -0500 References: <563E16AF.8070209@openvz.org> <20151110110025.GB19772@stefanha-x1.localdomain> From: "Denis V. Lunev" Message-ID: <5641D8A6.8030507@openvz.org> Date: Tue, 10 Nov 2015 14:44:38 +0300 MIME-Version: 1.0 In-Reply-To: <20151110110025.GB19772@stefanha-x1.localdomain> Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] assert during internal snapshot List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Stefan Hajnoczi , "Li, Liang Z" Cc: Amit Shah , Paolo Bonzini , QEMU , Juan Quintela On 11/10/2015 02:00 PM, Stefan Hajnoczi wrote: > On Mon, Nov 09, 2015 at 03:29:13AM +0000, Li, Liang Z wrote: >>> -----Original Message----- >>> From: Denis V. Lunev [mailto:den@openvz.org] >>> Sent: Saturday, November 07, 2015 11:20 PM >>> To: Li, Liang Z; Paolo Bonzini; Juan Quintela; Amit Shah >>> Cc: QEMU >>> Subject: assert during internal snapshot >>> >>> Hello, All! >>> >>> This commit >>> >>> commit 94f5a43704129ca4995aa3385303c5ae225bde42 >>> Author: Liang Li >>> Date: Mon Nov 2 15:37:00 2015 +0800 >>> >>> migration: defer migration_end & blk_mig_cleanup >>> >>> Because of the patch 3ea3b7fa9af067982f34b of kvm, which introduces a >>> lazy collapsing of small sptes into large sptes mechanism, now >>> migration_end() is a time consuming operation because it calls >>> memroy_global_dirty_log_stop(), which will trigger the dropping of small >>> sptes operation and takes about dozens of milliseconds, so call >>> migration_end() before all the vmsate data has already been transferred >>> to the destination will prolong VM downtime. This operation should be >>> deferred after all the data has been transferred to the destination. >>> >>> blk_mig_cleanup() can be deferred too. >>> >>> For a VM with 8G RAM, this patch can reduce the VM downtime about >>> 30 ms. >>> >>> Signed-off-by: Liang Li >>> Reviewed-by: Paolo Bonzini >>> Reviewed-by: Juan Quintela al3 >>> Reviewed-by: Amit Shah al3 >>> Signed-off-by: Juan Quintela al3 >>> >>> introduces the following regression >>> >>> (gdb) bt >>> #0 0x00007fd5d314a267 in __GI_raise (sig=sig@entry=6) >>> at ../sysdeps/unix/sysv/linux/raise.c:55 >>> #1 0x00007fd5d314beca in __GI_abort () at abort.c:89 >>> #2 0x00007fd5d314303d in __assert_fail_base ( >>> fmt=0x7fd5d32a5028 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", >>> assertion=assertion@entry=0x557288ed5b69 "i != mr->ioeventfd_nb", >>> file=file@entry=0x557288ed5a36 "/home/den/src/qemu/memory.c", >>> line=line@entry=1731, >>> function=function@entry=0x557288ed5fb0 <__PRETTY_FUNCTION__.32545> >>> "memory_region_del_eventfd") at assert.c:92 >>> #3 0x00007fd5d31430f2 in __GI___assert_fail ( >>> assertion=0x557288ed5b69 "i != mr->ioeventfd_nb", >>> file=0x557288ed5a36 "/home/den/src/qemu/memory.c", line=1731, >>> function=0x557288ed5fb0 <__PRETTY_FUNCTION__.32545> >>> "memory_region_del_eventfd") at assert.c:101 >>> #4 0x0000557288b108fa in memory_region_del_eventfd >>> (mr=0x55728ad83700, >>> addr=16, size=2, match_data=true, data=0, e=0x55728b21ff40) >>> at /home/den/src/qemu/memory.c:1731 >>> #5 0x0000557288d9fc18 in virtio_pci_set_host_notifier_internal ( >>> proxy=0x55728ad82e80, n=0, assign=false, set_handler=false) >>> at hw/virtio/virtio-pci.c:178 >>> #6 0x0000557288da19a9 in virtio_pci_set_host_notifier (d=0x55728ad82e80, >>> n=0, >>> assign=false) at hw/virtio/virtio-pci.c:984 >>> #7 0x0000557288b523df in virtio_scsi_dataplane_start (s=0x55728ad8afa0) >>> at /home/den/src/qemu/hw/scsi/virtio-scsi-dataplane.c:268 >>> #8 0x0000557288b50210 in virtio_scsi_handle_cmd (vdev=0x55728ad8afa0, >>> vq=0x55728b21ffc0) at /home/den/src/qemu/hw/scsi/virtio-scsi.c:574 >>> #9 0x0000557288b65cb7 in virtio_queue_notify_vq (vq=0x55728b21ffc0) >>> at /home/den/src/qemu/hw/virtio/virtio.c:966 >>> #10 0x0000557288b67bbf in virtio_queue_host_notifier_read >>> (n=0x55728b220010) >>> at /home/den/src/qemu/hw/virtio/virtio.c:1643 >>> #11 0x0000557288e12a2b in aio_dispatch (ctx=0x55728acaeab0) at >>> aio-posix.c:160 >>> #12 0x0000557288e03194 in aio_ctx_dispatch (source=0x55728acaeab0, >>> callback=0x0, user_data=0x0) at async.c:226 >>> #13 0x00007fd5d409fff7 in g_main_context_dispatch () >>> from /lib/x86_64-linux-gnu/libglib-2.0.so.0 >>> ---Type to continue, or q to quit--- >>> #14 0x0000557288e1110d in glib_pollfds_poll () at main-loop.c:211 >>> #15 0x0000557288e111e8 in os_host_main_loop_wait (timeout=0) at >>> main-loop.c:256 >>> #16 0x0000557288e11295 in main_loop_wait (nonblocking=0) at main- >>> loop.c:504 >>> #17 0x0000557288c1c31c in main_loop () at vl.c:1890 >>> #18 0x0000557288c23dec in main (argc=105, argv=0x7ffca9a6fa08, >>> envp=0x7ffca9a6fd58) at vl.c:4644 >>> (gdb) >>> >>> during 'virsh create-snapshot' operation over alive VM. >>> It happens 100% of time when VM is run using the following command line: >>> >>> 7498 ? tl 0:37 qemu-system-x86_64 -enable-kvm -name rhel7 >>> -S -machine pc-i440fx-2.2,accel=kvm,usb=off -cpu SandyBridge -m 1024 - >>> realtime mlock=off -smp 1,sockets=1,cores=1,threads=1 -object >>> iothread,id=iothread1 -uuid 456af4d3-5d67-41c6-a229-c55ded6098e9 >>> -no-user-config -nodefaults -chardev >>> socket,id=charmonitor,path=/var/lib/libvirt/qemu/rhel7.monitor,server,nowait >>> -mon chardev=charmonitor,id=monitor,mode=control -rtc >>> base=utc,driftfix=slew -global kvm-pit.lost_tick_policy=discard -no-hpet -no- >>> shutdown -global PIIX4_PM.disable_s3=1 -global PIIX4_PM.disable_s4=1 -boot >>> strict=on -device ich9-usb-ehci1,id=usb,bus=pci.0,addr=0x6.0x7 >>> -device >>> ich9-usb- >>> uhci1,masterbus=usb.0,firstport=0,bus=pci.0,multifunction=on,addr=0x6 >>> -device >>> ich9-usb-uhci2,masterbus=usb.0,firstport=2,bus=pci.0,addr=0x6.0x1 >>> -device >>> ich9-usb-uhci3,masterbus=usb.0,firstport=4,bus=pci.0,addr=0x6.0x2 >>> -device virtio-scsi-pci,id=scsi0,bus=pci.0,addr=0x5 -device >>> virtio-serial-pci,id=virtio-serial0,bus=pci.0,addr=0x7 -drive >>> file=/var/lib/libvirt/images/rhel7.qcow2,if=none,id=drive-scsi0-0-0- >>> 0,format=qcow2,cache=none,aio=native >>> -device >>> scsi-hd,bus=scsi0.0,channel=0,scsi-id=0,lun=0,drive=drive-scsi0-0-0-0,id=scsi0-0- >>> 0-0,bootindex=1 >>> -drive >>> file=/home/den/tmp/CentOS-7.0-1406-x86_64-DVD.iso,if=none,id=drive-ide0-0- >>> 0,readonly=on,format=raw >>> -device ide-cd,bus=ide.0,unit=0,drive=drive-ide0-0-0,id=ide0-0-0 -netdev >>> tap,fd=24,id=hostnet0,vhost=on,vhostfd=25 -device >>> virtio-net- >>> pci,netdev=hostnet0,id=net0,mac=52:54:00:97:2f:1d,bus=pci.0,addr=0x3 >>> -chardev >>> socket,id=charchannel0,path=/var/lib/libvirt/qemu/channel/target/rhel7.org.qe >>> mu.guest_agent.0,server,nowait >>> -device >>> virtserialport,bus=virtio- >>> serial0.0,nr=1,chardev=charchannel0,id=channel0,name=org.qemu.guest_agen >>> t.0 >>> -chardev spicevmc,id=charchannel1,name=vdagent -device >>> virtserialport,bus=virtio- >>> serial0.0,nr=2,chardev=charchannel1,id=channel1,name=com.redhat.spice.0 >>> -chardev pty,id=charconsole0 -device >>> virtconsole,chardev=charconsole0,id=console0 -device >>> usb-tablet,id=input0 -spice >>> port=5900,addr=127.0.0.1,disable-ticketing,seamless-migration=on -device >>> qxl- >>> vga,id=video0,ram_size=67108864,vram_size=67108864,vgamem_mb=16,bus= >>> pci.0,addr=0x2 >>> -device intel-hda,id=sound0,bus=pci.0,addr=0x4 -device >>> hda-duplex,id=sound0-codec0,bus=sound0.0,cad=0 -chardev >>> spicevmc,id=charredir0,name=usbredir -device >>> usb-redir,chardev=charredir0,id=redir0 -chardev >>> spicevmc,id=charredir1,name=usbredir -device >>> usb-redir,chardev=charredir1,id=redir1 -chardev >>> spicevmc,id=charredir2,name=usbredir -device >>> usb-redir,chardev=charredir2,id=redir2 -chardev >>> spicevmc,id=charredir3,name=usbredir -device >>> usb-redir,chardev=charredir3,id=redir3 -device >>> virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x8 -set >>> device.scsi0.iothread=iothread1 -msg timestamp=on >>> >>> Minimal dumb (!!!) fix is the following: >>> >>> hades ~/src/qemu $ cat 1.diff >>> diff --git a/migration/ram.c b/migration/ram.c index 25e9eeb..25ebf0d 100644 >>> --- a/migration/ram.c >>> +++ b/migration/ram.c >>> @@ -1100,7 +1100,8 @@ static void migration_bitmap_free(struct BitmapRcu >>> *bmap) >>> g_free(bmap); >>> } >>> >>> -static void migration_end(void) >>> +extern void migration_end(void); >>> +void migration_end(void) >>> { >>> /* caller have hold iothread lock or is in a bh, so there is >>> * no writing race against this migration_bitmap diff --git >>> a/migration/savevm.c b/migration/savevm.c index dbcc39a..01da865 100644 >>> --- a/migration/savevm.c >>> +++ b/migration/savevm.c >>> @@ -914,6 +914,7 @@ void qemu_savevm_state_cancel(void) >>> } >>> } >>> >>> +extern void migration_end(void); >>> static int qemu_savevm_state(QEMUFile *f, Error **errp) >>> { >>> int ret; >>> @@ -942,6 +943,8 @@ static int qemu_savevm_state(QEMUFile *f, Error >>> **errp) >>> qemu_savevm_state_complete(f); >>> ret = qemu_file_get_error(f); >>> } >>> + migration_end(); >>> + >>> if (ret != 0) { >>> qemu_savevm_state_cancel(); >>> error_setg_errno(errp, -ret, "Error while writing VM state"); hades >>> ~/src/qemu $ >>> >>> (patch attached to start a discussion). >>> >>> For me it looks like commit is wrong and should be reverted and reworked. >>> >>> Den >> Cc to : Stefan >> >> It seems the 'bdrv_drain_all()' in ' blk_mig_cleanup ()' should not be deferred. Stefan, is that right? > Denis, do you still experience this crash with your own savevm patch > series applied? Since that series does AioContext acquire/release where > missing in savevm, it might fix this bug. > > Stefan this patch [Qemu-devel] [PULL 57/57] migration: qemu_savevm_state_cleanup becomes mandatory operation should be applied before or on top of my patches. This problem is completely orthogonal and comes from the different source. Anyway, thank you for paying attention :) Den