- Virt - mailweb.openeuler.org

Re: Kunpeng920 virtualization performance
by Ying Fang 17 Sep '20

17 Sep '20

On 7/2/2020 3:18 PM, Haibin Zhang wrote: > Hi, Ying Fang. > Hi Haibin, Thanks for report this issue for us. We will verify and response to you later. > I use qemu create a VM (96 cores and 360g ram) on my kunpeng 920-4826 server, then do unixbench tests. > The performance of pipe-based context switching decreases severely, which shows below: > > Guest# ./context1 60 > COUNT|4779535|1|lps > COUNT|4779536|1|lps > > Host# ./context1 60 > COUNT|13990745|1|lps > COUNT|13990746|1|lps > Could you provide more details about your OS and virtualization software version information for us? > As you know, pipe-based context switching measures the number of times two processes can exchange an > increasing integer through a pipe in a duration (like 60 seconds). Times of guest is only 34% as that > of host. Attachment is the context1.c from unixbench. It looks simple, just spawn a child process with > which it carries on a bi-directional pipe conversation. Yes, we had knowledge about this context test now. Thanks. > > Guest# taskset -c 20 ./context1 60 > COUNT|18059443|1|lps > COUNT|18059443|1|lps > > Host# taskset -c 20 ./context1 60 > COUNT|17881974|1|lps > COUNT|17881975|1|lps > > Use taskset to set cpu affinity, which makes results better. > > Can you help me fix this issue? Maybe WFE/WFI traps lead the performance cost? Is there PLE(Pause Loop Exit) feature on Kunpeng? > We had some performance tuning techniques in the openEuler OS. And Xie XiangYou is the expert on this area, maybe he could help you with it. Moreover, you can drop us an issue on gitee via: https://gitee.com/src-openeuler/qemu >

2 20

vmtop文档推送openeuler
by chenjiajun 14 Sep '20

14 Sep '20

1 0

openeuler开源代码回合
by Peng Liang 09 Sep '20

09 Sep '20

Patch： block-qcow2-do-free-crypto_opts-in-qcow2_close.patch qemu-img-free-memory-before-re-assign.patch block-qcow2-threads-fix-qcow2_decompress.patch block-Avoid-memleak-on-qcow2-image-info-failure.patch block-bdrv_set_backing_bs-fix-use-after-free.patch hmp-vnc-Fix-info-vnc-list-leak.patch migration-colo-fix-use-after-free-of-local_err.patch migration-ram-fix-use-after-free-of-local_err.patch block-mirror-fix-use-after-free-of-local_err.patch block-fix-bdrv_root_attach_child-forget-to-unref-chi.patch virtio-serial-bus-Plug-memory-leak-on-realize-error-.patch virtio-blk-delete-vqs-on-the-error-path-in-realize.patch fix-vhost_user_blk_watch-crash.patch vhost-user-blk-delay-vhost_user_blk_disconnect.patch Changelog： - block/qcow2: do free crypto_opts in qcow2_close() - qemu-img: free memory before re-assign - block/qcow2-threads: fix qcow2_decompress - block: Avoid memleak on qcow2 image info failure - block: bdrv_set_backing_bs: fix use-after-free - hmp/vnc: Fix info vnc list leak - migration/colo: fix use after free of local_err - migration/ram: fix use after free of local_err - block/mirror: fix use after free of local_err - block: fix bdrv_root_attach_child forget to unref child_bs - virtio-serial-bus: Plug memory leak on realize() error paths - virtio-blk: delete vqs on the error path in realize() - fix vhost_user_blk_watch crash - vhost-user-blk: delay vhost_user_blk_disconnect

1 0

openeuler开源bugfix回合
by chenjiajun 09 Sep '20

09 Sep '20

1 0

CPU feature忽略evtstrm和cpuid
by Peng Liang 08 Sep '20

08 Sep '20

1 0

Virt sig会议议程纪要0904//RE: 【Meeting Notice】Virt sig meeting Time: 2020-09-04 14:30-15:30
by Chenzhendong (alex) 07 Sep '20

07 Sep '20

Virt sig会议议程纪要： 1、openEuler 20.09虚拟化特性代码已经合入，正在进行稳定性和可靠性等测试。 a) vcpu hotplug/mem hotplug/HPRE加速三个特性已经测试完成。 b) 虚拟机可信启动/安全启动/PMU NMI-Watchdog/双层调度等特性正在开展测试中; c) VMtop、ARM虚拟机支持超线程、ARM虚拟机cpu feature支持Custom三个特性尚未开展测试，正在分析及测试设计； 2、EulerRobot质量体系构建进展： a) EulerRobot已经开源，已有部分开发者参与，希望后续有更多开发者参与，共同完善和优化openEuler虚拟化质量体系。 b) crystal-ci当前存在物理机部署后卡死无法使用情况，crystal-ci团队解决中，等解决完后再进行对接。 c) openEuler下虚拟化qemu/libvirt等仓库，代码门禁已经OK，正在实现让qemu/libvirt仓库代码合入后自动化生成Patch同步到src-openEuler中便于版本rpm制作，后续代码提交将切换至openeuelr仓库。 3、openEuler 支持riscv虚拟化的进展如下，后续主要进行bugfix。 a) Qemu支持RISC-V架构的KVM； b) 支持RISC-V架构虚拟机热迁移； c) 支持vhost的ioeventfd方案； From: Chenzhendong (alex) via Virt [mailto:virt@openeuler.org] Sent: Friday, September 4, 2020 10:19 AM To: virt(a)openeuler.org; dev(a)openeuler.org Cc: Zhangxiaofeng (F) <victor.zhangxiaofeng(a)huawei.com> Subject: [Virt] 【Meeting Notice】Virt sig meeting Time: 2020-09-04 14:30-15:30 Virt sig会议议程： 1、openEuler 20.09 虚拟化特性进展 alexchen 2、EulerRobot质量体系构建进展 kuhnchen 3、openEuler RISC-V虚拟化支持进展 Jiangyifei ------- 原始邮件 -------- 发件人： Meeting Book <uMeeting(a)huawei.com<mailto:uMeeting@huawei.com>> 日期： 2020年9月4日周五 10:03 收件人： zdchenhn(a)163.com<mailto:zdchenhn@163.com> 主题：【Meeting Notice】Virt sig meeting Time: 2020-09-04 14:30-15:30 Topic Virt sig meeting Time 2020-09-04 14:30-15:30((UTC+08:00)Beijing) Join Conference Join (External) >><https://welink-meeting.zoom.us/j/258690350> Meeting ID 258 690 350 Convener 陈振东 Tips: Dial the access number to join conference<http://app.huawei.com/eshare/voiceMeetings> 注：仅支持部分国家和地区。( Only some countries and regions are supported.) 如果加入会议提示“会议已结束或会议号不存在”，请在官网下载最新版本入会。立即下载<https://zoom.us/download> If a message is displayed, indicating that the meeting has ended or the meeting ID does not exist, download the latest meeting client and try again.Download<https://zoom.us/download>

1 0

【Meeting Notice】Virt sig meeting Time: 2020-09-04 14:30-15:30
by Chenzhendong (alex) 04 Sep '20

04 Sep '20

Virt sig会议议程： 1、openEuler 20.09 虚拟化特性进展 alexchen 2、EulerRobot质量体系构建进展 kuhnchen 3、openEuler RISC-V虚拟化支持进展 Jiangyifei ------- 原始邮件 -------- 发件人： Meeting Book <uMeeting(a)huawei.com<mailto:uMeeting@huawei.com>> 日期： 2020年9月4日周五 10:03 收件人： zdchenhn(a)163.com<mailto:zdchenhn@163.com> 主题：【Meeting Notice】Virt sig meeting Time: 2020-09-04 14:30-15:30 Topic Virt sig meeting Time 2020-09-04 14:30-15:30((UTC+08:00)Beijing) Join Conference Join (External) >><https://welink-meeting.zoom.us/j/258690350> Meeting ID 258 690 350 Convener 陈振东 Tips: Dial the access number to join conference<http://app.huawei.com/eshare/voiceMeetings> 注：仅支持部分国家和地区。( Only some countries and regions are supported.) 如果加入会议提示“会议已结束或会议号不存在”，请在官网下载最新版本入会。立即下载<https://zoom.us/download> If a message is displayed, indicating that the meeting has ended or the meeting ID does not exist, download the latest meeting client and try again.Download<https://zoom.us/download>

1 0

Re: hugepages leads VM startup secondary cpu failure
by Ying Fang 04 Sep '20

04 Sep '20

On 9/3/2020 9:33 PM, Haibin Zhang wrote: > Hi, YingFang. > > I configure 4K PAGE_SIZE in order to use 1G hugepage in host. But VM cannot bring up secondary CPU > > Start VM using follow command: > qemu-system-aarch64 -name guest=12345,debug-threads=on \ > -machine virt,accel=kvm,usb=off,dump-guest-core=off,gic-version=3 \ > -cpu host \ > -m 122880 \ > -object iothread,id=iothread1 \ > -object memory-backend-file,id=ram-node0,prealloc=yes,mem-path=/dev/hugepages,share=yes,size=64424509440,host-nodes=0,policy=preferred \ > -numa node,nodeid=0,cpus=0-29,memdev=ram-node0 \ > -object memory-backend-file,id=ram-node1,prealloc=yes,mem-path=/dev/hugepages,share=yes,size=64424509440,host-nodes=1,policy=preferred \ > -numa node,nodeid=1,cpus=30-59,memdev=ram-node1 \ > -smp sockets=2,cores=30,threads=1 \ > -bios /usr/share/AAVMF/AAVMF_CODE.fd \ > -drive file=/data/haibin/centos8.qcow2,format=qcow2,if=none,id=drive-virtio-disk0,cache=none \ > -device virtio-blk-pci,scsi=off,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1 \ > -net none -serial telnet::9001,server,nowait -monitor stdio > Thanks for your reporting, I will CC Zeyu Jin to lookup into it. > VM kernel logs: > [ 0.010708] Mountpoint-cache hash table entries: 65536 (order: 3, 524288 bytes) > [ 0.012517] ASID allocator initialised with 32768 entries > [ 0.013105] rcu: Hierarchical SRCU implementation. > [ 0.013843] Platform MSI: ITS@0x8080000 domain created > [ 0.014375] PCI/MSI: ITS@0x8080000 domain created > [ 0.014850] Remapping and enabling EFI services. > [ 0.015946] smp: Bringing up secondary CPUs ... > [ 1.039238] CPU1: failed to come online > [ 1.039698] CPU1: failed in unknown state : 0x0 > [ 1.051285] Detected VIPT I-cache on CPU2 > [ 1.051303] GICv3: CPU2: found redistributor 2 region 0:0x00000000080e0000 > [ 1.051520] GICv3: CPU2: using allocated LPI pending table @0x00000003c06a0000 > [ 1.051545] CPU2: Booted secondary processor 0x0000000001 [0x481fd010] > [ 2.079879] CPU3: failed to come online > [ 2.082736] CPU3: failed in unknown state : 0x0 > [ 2.085620] Detected VIPT I-cache on CPU4 > [ 2.085638] GICv3: CPU4: found redistributor 4 region 0:0x0000000008120000 > [ 2.085785] GICv3: CPU4: using allocated LPI pending table @0x00000003c06c0000 > [ 2.085808] CPU4: Booted secondary processor 0x0000000002 [0x481fd010] > [ 3.120358] Detected VIPT I-cache on CPU5 > [ 3.120379] GICv3: CPU5: found redistributor 5 region 0:0x0000000008140000 > [ 3.120607] GICv3: CPU5: using allocated LPI pending table @0x00000003c06d0000 > [ 3.120619] CPU5: failed to come online > [ 3.120632] CPU5: Booted secondary processor 0x0000000003 [0x481fd010] > [ 3.121069] ------------[ cut here ]------------ > [ 3.126345] kernel BUG at kernel/irq_work.c:147! > [ 3.126797] Internal error: Oops - BUG: 0 [#1] SMP > [ 3.127268] Modules linked in: > [ 3.127568] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.18.0-193.14.2.el8_2.aarch64 #1 > [ 3.128345] pstate: a0000005 (NzCv daif -PAN -UAO) > [ 3.128818] pc : irq_work_run_list+0xa8/0xb8 > [ 3.129239] lr : irq_work_run+0x24/0x48 > [ 3.129615] sp : ffff000012aefc40 > [ 3.129940] x29: ffff000012aefc40 x28: 0000000000000000 > [ 3.130458] x27: 0000000000000000 x26: 0000000000000000 > [ 3.131040] x25: 00008003ac000000 x24: ffff8003bd202130 > [ 3.131592] x23: 0000000000000000 x22: 0000000000000000 > [ 3.132119] x21: 0000000000000000 x20: ffff00001128e4c0 > [ 3.132677] x19: ffff00001128a400 x18: 0000000000000010 > [ 3.133247] x17: 0000000000000000 x16: ffff80038237ba00 > [ 3.133816] x15: ffffffffffffffff x14: ffff0000115d3708 > [ 3.134391] x13: ffff000092aef9a7 x12: ffff000012aef9b0 > [ 3.134991] x11: ffff00001160f000 x10: ffff000012aef930 > [ 3.135567] x9 : 00000000ffffffd0 x8 : ffff0000105dab10 > [ 3.136128] x7 : 000000000000000d x6 : ffff000011fc696c > [ 3.136685] x5 : 0000000000000015 x4 : 0000000000000000 > [ 3.137258] x3 : ffff8003bcf6e4c0 x2 : 0000000000000000 > [ 3.137825] x1 : 0000000000000000 x0 : ffff8003bcf6a400 > [ 3.138405] Process swapper/0 (pid: 1, stack limit = 0x(____ptrval____)) > [ 3.139096] Call trace: > [ 3.139364] irq_work_run_list+0xa8/0xb8 > [ 3.139773] irq_work_run+0x24/0x48 > [ 3.140180] flush_smp_call_function_queue+0xa4/0x150 > [ 3.140714] smpcfd_dying_cpu+0x18/0x28 > [ 3.141134] cpuhp_invoke_callback+0xa8/0x618 > [ 3.141621] _cpu_up+0x1c0/0x1d0 > [ 3.142013] do_cpu_up+0x7c/0xb8 > [ 3.142377] cpu_up+0x24/0x30 > [ 3.142717] smp_init+0xac/0x120 > [ 3.143089] kernel_init_freeable+0x168/0x314 > [ 3.143559] kernel_init+0x18/0x10c > [ 3.143942] ret_from_fork+0x10/0x18 > [ 3.144336] Code: a8c37bfd d65f03c0 a90153f3 a9025bf5 (d4210000) > [ 3.145028] ---[ end trace 392329bbeab599c5 ]--- > [ 3.145548] Kernel panic - not syncing: Fatal exception > [ 3.146149] SMP: stopping secondary CPUs > [ 4.183106] SMP: failed to stop secondary CPUs 0,2,5 > [ 4.183677] ---[ end Kernel panic - not syncing: Fatal exception ]---. >

1 0

add document about custom CPU mode
by Peng Liang 29 Aug '20

29 Aug '20

1 0

[PATCH 0/6] Add RISC-V vm migration support
by Yifei Jiang 26 Aug '20

26 Aug '20

This series adds RISC-V vm migration support. Yifei Jiang (6): target/riscv: raise exception to HS mode at get_physical_address target/riscv: Add sifive_plic vmstate target/riscv: Support riscv cpu vmstate target/riscv: Add kvm_riscv_get/put_regs_timer target/riscv: Implement virtual time adjusting with vm state changing target/riscv: Support virtual time context synchronization hw/riscv/sifive_plic.c | 24 +++++++++- include/hw/riscv/sifive_plic.h | 1 + target/riscv/cpu.c | 31 +++++++++++- target/riscv/cpu.h | 7 +++ target/riscv/cpu_helper.c | 12 ++++- target/riscv/kvm.c | 86 ++++++++++++++++++++++++++++++++++ 6 files changed, 156 insertions(+), 5 deletions(-) -- 2.19.1

1 6

2024

2023

2022

2021

2020

Virt