<div dir="ltr">Actually there is still some issue around this. When trying to start another VM, so using even more pmem, there is a different issue now. Error copied below.[]# virsh start manual_clone error: Failed to start domain manual_clone error: internal error: qemu unexpectedly closed the monitor: ftruncate: Invalid argument 2019-09-03T21:28:41.031924Z qemu-system-x86_64: -object memory-backend-file,id=memnvdimm0,prealloc=yes,mem-path=/dev/dax1.1,share=yes,size=73014444032: os_mem_prealloc: Insufficient free host memory pages available to allocate guest RAM Old error was Error: # virsh start test error: Failed to start domain test error: internal error: qemu unexpectedly closed the monitor: ftruncate: Invalid argument 2019-08-22T04:16:08.744402Z qemu-system-x86_64: -object memory-backend-file,id=memnvdimm0,prealloc=yes,mem-path=/dev/dax1.0,size=403726925824: unable to map backing store for guest RAM: Cannot allocate memory</div> <div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, Sep 3, 2019 at 9:53 AM Dan Williams <<a href="mailto:dan.j.williams@intel.com">dan.j.williams@intel.com</a>> wrote: </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">On Mon, Sep 2, 2019 at 10:10 AM Seema Pandit <<a href="mailto:pan.blr.17@gmail.com" target="_blank">pan.blr.17@gmail.com</a>> wrote: > > After adding the memoryBacking tag in xml as below (in addition, to other xml changes to add nvdimm), virsh could allocate AD memory larger than the system RAM and VMs could start successfully. > > <memoryBacking> > > <access mode='shared'/> > > <discard/> > > </memoryBacking> > > > > This adds share=yes in command line. > > > > -object memory-backend-file,id=memnvdimm0,prealloc=yes,mem-path=/mnt/pmem0/file1,share=yes,size=414464344064 -device nvdimm,node=0,label-size=131072,memdev=memnvdimm0,id=nvdimm0,slot=0 > > > > > > For reference qemu command line where VM starts quickly: > > qemu-system-x86_64 \ > > -name qemu-gold29 \ > > -drive file=/var/lib/libvirt/images/gold29-ad.qcow2,format=qcow2,index=0,media=disk \ -m 2G,slots=4,maxmem=428G \ -smp 2 \ -machine pc,accel=kvm,nvdimm=on \ -enable-kvm \ -object memory-backend-file,id=pmem1,share=on,mem-path=/mnt/pmem0/file1,size=386G,align=4K \ -device nvdimm,memdev=pmem1,id=nv1 \ -daemonize > > > > > > Qemu command line generated from virsh: (please note VM now starts with this command line, shared=yes.) > > > > /usr/bin/qemu-system-x86_64 -machine accel=kvm -name guest=mix-test,debug-threads=on -S -object secret,id=masterKey0,format=raw,file=/var/lib/libvirt/qemu/domain-18-mix-test/master-key.aes -machine pc-i440fx-3.0,accel=kvm,usb=off,vmport=off,dump-guest-core=off,nvdimm=on -cpu Skylake-Server-IBRS,hypervisor=on -m size=2097152k,slots=16,maxmem=419430400k -realtime mlock=off -smp 2,sockets=2,cores=1,threads=1 -object memory-backend-file,id=ram-node0,mem-path=/var/lib/libvirt/qem /ram/libvirt/qemu/18-mix-test/ram-node0,discard-data=yes,share=yes,size=2147483648 -numa node,nodeid=0,cpus=0-1,memdev=ram-node0 -object memory-backend-file,id=memnvdimm0,prealloc=yes,mem-path=/mnt/pmem0/file1,share=yes,size=414464344064 -device nvdimm,node=0,label-size=131072,memdev=memnvdimm0,id=nvdimm0,slot=0 -uuid 318c0529-0330-460b-8d0a-3b253e9decdd -no-user-config -nodefaults -chardev socket,id=charmonitor,fd=32,server,nowait -mon chardev=charmonitor,id=monitor,mode=control -rtc base=utc,driftfix=slew -global kvm-pit.lost_tick_policy=delay -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=0x5.0x7 -device ich9-usb-uhci1,masterbus=usb.0,firstport=0,bus=pci.0,multifunction=on,addr=0x5 -device ich9-usb-uhci2,masterbus=usb.0,firstport=2,bus=pci.0,addr=0x5.0x1 -device ich9-usb-uhci3,masterbus=usb.0,firstport=4,bus=pci.0,addr=0x5.0x2 -device virtio-serial-pci,id=virtio-serial0,bus=pci.0,addr=0x6 -drive file=/var/lib/libvirt/images/mix-test.qcow2,format=qcow2,if=none,id=drive-virtio-disk0 -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x7,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1 -drive if=none,id=drive-ide0-0-0,readonly=on -device ide-cd,bus=ide.0,unit=0,drive=drive-ide0-0-0,id=ide0-0-0 -netdev tap,fd=34,id=hostnet0 -device e1000,netdev=hostnet0,id=net0,mac=52:54:00:06:db:55,bus=pci.0,addr=0xa -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -chardev socket,id=charchannel0,fd=35,server,nowait -device virtserialport,bus=virtio-serial0.0,nr=1,chardev=charchannel0,id=channel0,name=org.qemu.guest_agent.0 -chardev spicevmc,id=charchannel1,name=vdagent -device virtserialport,bus=virtio-serial0.0,nr=2,chardev=charchannel1,id=channel1,name=com.redhat.spice.0 -device usb-tablet,id=input0,bus=usb.0,port=1 -spice port=5901,addr=127.0.0.1,disable-ticketing,image-compression=off,seamless-migration=on -device qxl-vga,id=video0,ram_size=67108864,vram_size=67108864,vram64_size_mb=0,vgamem_mb=16,max_outputs=1,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,bus=usb.0,port=2 -chardev spicevmc,id=charredir1,name=usbredir -device usb-redir,chardev=charredir1,id=redir1,bus=usb.0,port=3 -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x8 -object rng-random,id=objrng0,filename=/dev/urandom -device virtio-rng-pci,rng=objrng0,id=rng0,bus=pci.0,addr=0x9 -sandbox on,obsolete=deny,elevateprivileges=deny,spawn=deny,resourcecontrol=deny -msg timestamp=on > > > VM start takes longer that qemu. Any thoughts?? Why is prealloc=yes default on for nvdimm? Any other important deltas? > The "prealloc" option is a "pay me now" vs "pay me later" type of decision. If the guest workload is ok to absorb fault latency at run time then disable prealloc, if it would prefer more predictable latency and pay all the fault penalty up front then specify prealloc. The "shared" parameter, if it means MAP_SHARED vs MAP_PRIVATE, must be set to "yes" if the guest expects persistence. MAP_PRIVATE is otherwise unacceptable for emulating persistent memory because writes to private memory are discarded, and as you have seen require volatile DRAM backing even in the DAX case. </blockquote></div>