由于 DMA 写入导致的启动缓慢/由于未设置 PTE 写入访问权限导致的启动缓慢

由于 DMA 写入导致的启动缓慢/由于未设置 PTE 写入访问权限导致的启动缓慢

我遇到了 Ubuntu 20.04 启动缓慢的问题,大约需要 2 分钟。我已经确定了问题所在。但是,我不明白错误是什么。有几个问题。但是,在互联网上进行一些搜索并没有给出解决方案或更好地理解这个问题。

系统(笔记本):

  • BTO 的 TOXIC AURIX 17" MK-III
  • 英特尔® 酷睿 i9-10900K 处理器
  • 操作系统 Ubuntu 20.04.3 LTS
  • 底板信息:笔记本产品名称 X170KM-G
  • 建造年份:2021 年 8 月
  • Bios 供应商 os INSYDE Corp. 版本 1.07.04TNB(发布日期 2020 年 1 月 11 日)
  • NVIDIA®GeForce®RTX 3080
  • NVIDIA-SMI 470.63.01 / CUDA 版本 11.4

在启动过程中我获得以下信息:

DMAR: DRHD: handling fault status reg 2

DMAR: [DMA Write] Request device [05:00.0] PASID ffffffff fault addr 6d95f000 [fault reason 05] PTE Write access is not set

DMAR: [DMA Write] Request device [05:00.0] PASID ffffffff fault addr 6d95f000 [fault reason 05] PTE Write access is not set

DMAR: DRHD: handling fault status reg 2

DMAR: [DMA Write] Request device [05:00.0] PASID ffffffff fault addr 6d95f000 [fault reason 05] PTE Write access is not set

DMAR: DRHD: handling fault status reg 2

DMAR: [DMA Write] Request device [05:00.0] PASID ffffffff fault addr 6d95f000 [fault reason 05] PTE Write access is not set

输出结果为journalctl -p 3 -xb

x86/cpu: SGX disabled by BIOS

mce: [Hardware Error]: CPU 0: Machine Check: 0 Bank 6: ee0000000040110a

mce: [Hardware Error]: TSC 0 ADDR fef20300 MISC 3880000086 

mce: [Hardware Error]: PROCESSOR 0:a0655 TIME 1630579052 SOCKET 0 APIC 0 microcode ec

tpm tpm0: [Firmware Bug]: TPM interrupt not working, polling instead

DMAR: DRHD: handling fault status reg 2

DMAR: [DMA Write] Request device [05:00.0] PASID ffffffff fault addr 6d95f000 [fault reason 05] PTE Write access is not set

DMAR: DRHD: handling fault status reg 2

DMAR: [DMA Write] Request device [05:00.0] PASID ffffffff fault addr 6d95f000 [fault reason 05] PTE Write access is not set

DMAR: DRHD: handling fault status reg 2

DMAR: [DMA Write] Request device [05:00.0] PASID ffffffff fault addr 6d95f000 [fault reason 05] PTE Write access is not set

DMAR: DRHD: handling fault status reg 2

DMAR: [DMA Write] Request device [05:00.0] PASID ffffffff fault addr 6d95f000 [fault reason 05] PTE Write access is not set

thunderbolt 0000:05:00.0: failed to send driver ready to ICM

额外的步骤,但没有解决方案:

在 /etc/default/grub 中我添加

GRUB_CMDLINE_LINUX_DEFAULT="quiet splash intel_iommu=igfx_off" 

并执行sudo update-grub但它并不能解决问题。

更新到 Ubuntu 21.04 并不能解决问题。我重新安装了 Ubuntu 20.04。我正在运行 Lambda Stack,目前支持 Ubuntu 20.04 LTS、18.04 LTS 和 16.04 LTS

输出lspci -tv

-[0000:00]-+-00.0  Intel Corporation Device 9b33
           +-01.0-[01]--+-00.0  NVIDIA Corporation Device 24dc
           |            \-00.1  NVIDIA Corporation Device 228b
           +-08.0  Intel Corporation Xeon E3-1200 v5/v6 / E3-1500 v5 / 6th/7th/8th Gen Core Processor Gaussian Mixture Model
           +-14.0  Intel Corporation Device 43ed
           +-14.2  Intel Corporation Device 43ef
           +-14.3  Intel Corporation Device 43f0
           +-15.0  Intel Corporation Device 43e8
           +-15.1  Intel Corporation Device 43e9
           +-15.2  Intel Corporation Device 43ea
           +-16.0  Intel Corporation Device 43e0
           +-17.0  Intel Corporation Device 43d2
           +-19.0  Intel Corporation Device 43ad
           +-19.2  Intel Corporation Device 43a7
           +-1b.0-[02]----00.0  Samsung Electronics Co Ltd Device a809
           +-1c.0-[03-6b]----00.0-[04-6a]--+-00.0-[05]----00.0  Intel Corporation Device 1137
           |                               +-01.0-[06-37]--
           |                               +-02.0-[38]----00.0  Intel Corporation Device 1138
           |                               \-03.0-[39-6a]--
           +-1d.0-[6c]----00.0  Samsung Electronics Co Ltd Device a809
           +-1d.5-[6d]----00.0  Realtek Semiconductor Co., Ltd. Device 3000
           +-1d.6-[6e]----00.0  Realtek Semiconductor Co., Ltd. RTS5260 PCI Express Card Reader
           +-1f.0  Intel Corporation Device 4385
           +-1f.3  Intel Corporation Device f0c8
           +-1f.4  Intel Corporation Device 43a3
           \-1f.5  Intel Corporation Device 43a4
  • 内存测试:通过

  • “quiet splash” 并将其更改为“quiet splash nomodeset” -> 没有成功,重启时间仍然很长。 的输出journalctl -p 3 -xb仍然给出上述错误。 我还修改了 /etc/default/grub 并执行sudo update-grub。 没有成功。

  • sudo lshw -c storage -c disk提供以下输出:

*-sata                    
       description: SATA controller
       product: Intel Corporation
       vendor: Intel Corporation
       physical id: 17
       bus info: pci@0000:00:17.0
       version: 11
       width: 32 bits
       clock: 66MHz
       capabilities: sata msi pm ahci_1.0 bus_master cap_list
       configuration: driver=ahci latency=0
       resources: irq:132 memory:a2590000-a2591fff memory:a2594000-a25940ff ioport:5040(size=8) ioport:5048(size=4) ioport:5020(size=32) memory:a2593000-a25937ff
  *-storage
       description: Non-Volatile memory controller
       product: Samsung Electronics Co Ltd
       vendor: Samsung Electronics Co Ltd
       physical id: 0
       bus info: pci@0000:02:00.0
       version: 00
       width: 64 bits
       clock: 33MHz
       capabilities: storage pm msi pciexpress msix nvm_express bus_master cap_list
       configuration: driver=nvme latency=0
       resources: irq:16 memory:a2400000-a2403fff
     *-nvme0
          description: NVMe device
          product: Samsung SSD 980 1TB
          physical id: 0
          logical name: /dev/nvme0
          version: 1B4QFXO7
          serial: S649NF0R594587R
          configuration: nqn=nqn.1994-11.com.samsung:nvme:980M.2:S649NF0R594587R state=live
        *-namespace
             description: NVMe namespace
             physical id: 1
             logical name: /dev/nvme0n1
             size: 931GiB (1TB)
             capabilities: gpt-1.00 partitioned partitioned:gpt
             configuration: guid=c1d7d756-6025-4936-ae78-8096dd1aa5b1 logicalsectorsize=512 sectorsize=512
*-storage
       description: Non-Volatile memory controller
       product: Samsung Electronics Co Ltd
       vendor: Samsung Electronics Co Ltd
       physical id: 0
       bus info: pci@0000:6c:00.0
       version: 00
       width: 64 bits
       clock: 33MHz
       capabilities: storage pm msi pciexpress msix nvm_express bus_master cap_list
       configuration: driver=nvme latency=0
       resources: irq:16 memory:a2300000-a2303fff
     *-nvme1
          description: NVMe device
          product: Samsung SSD 980 1TB
          physical id: 0
          logical name: /dev/nvme1
          version: 1B4QFXO7
          serial: S649NF1R501393P
          configuration: nqn=nqn.1994-11.com.samsung:nvme:980M.2:S649NF1R501393P state=live
        *-namespace
             description: NVMe namespace
             physical id: 1
             logical name: /dev/nvme1n1
             size: 931GiB (1TB)
             capabilities: partitioned partitioned:dos
             configuration: logicalsectorsize=512 sectorsize=512 signature=b7d3e066
  • sudo dmidecode -t 8给出:
# dmidecode 3.2
Getting SMBIOS data from sysfs.
SMBIOS 3.3 present.
# SMBIOS implementations newer than version 3.2.0 are not
# fully supported by this version of dmidecode.

Handle 0x0008, DMI type 8, 9 bytes
Port Connector Information
    Internal Reference Designator: J_HDMI1
    Internal Connector Type: None
    External Reference Designator: <BAD INDEX>
    External Connector Type: Other
    Port Type: Video Port

Handle 0x0009, DMI type 8, 9 bytes
Port Connector Information
    Internal Reference Designator: J_CRT1
    Internal Connector Type: None
    External Reference Designator: CRT
    External Connector Type: Other
    Port Type: Video Port

Handle 0x000A, DMI type 8, 9 bytes
Port Connector Information
    Internal Reference Designator: J_DP2
    Internal Connector Type: None
    External Reference Designator: DP2
    External Connector Type: Other
    Port Type: Video Port

Handle 0x000B, DMI type 8, 9 bytes
Port Connector Information
    Internal Reference Designator: J_AUDIO2
    Internal Connector Type: None
    External Reference Designator: Headphone
    External Connector Type: Mini Jack (headphones)
    Port Type: Audio Port

Handle 0x000C, DMI type 8, 9 bytes
Port Connector Information
    Internal Reference Designator: J_INTMIC1
    Internal Connector Type: None
    External Reference Designator: MIC In
    External Connector Type: Other
    Port Type: Audio Port

Handle 0x000D, DMI type 8, 9 bytes
Port Connector Information
    Internal Reference Designator: J_USB3_1
    Internal Connector Type: None
    External Reference Designator: USB Port1
    External Connector Type: Access Bus (USB)
    Port Type: USB

Handle 0x000E, DMI type 8, 9 bytes
Port Connector Information
    Internal Reference Designator: J_USB3_2
    Internal Connector Type: None
    External Reference Designator: USB Port2
    External Connector Type: Access Bus (USB)
    Port Type: USB

Handle 0x000F, DMI type 8, 9 bytes
Port Connector Information
    Internal Reference Designator: J_USB3_3
    Internal Connector Type: None
    External Reference Designator: USB Port3
    External Connector Type: Access Bus (USB)
    Port Type: USB

Handle 0x0010, DMI type 8, 9 bytes
Port Connector Information
    Internal Reference Designator: J_TYPEC1
    Internal Connector Type: None
    External Reference Designator: USB Port1
    External Connector Type: Access Bus (USB)
    Port Type: USB

Handle 0x0011, DMI type 8, 9 bytes
Port Connector Information
    Internal Reference Designator: J_RJ_1
    Internal Connector Type: None
    External Reference Designator: Giga Lan
    External Connector Type: RJ-45
    Port Type: Network Port

Handle 0x0012, DMI type 8, 9 bytes
Port Connector Information
    Internal Reference Designator: J_CARD-REV1
    Internal Connector Type: None
    External Reference Designator: Card Reader
    External Connector Type: Other
    Port Type: Other

似乎没有 Thunderbolt 支持,即无法确定安全级别。

在 BIOS 中,我禁用了独立 Thunderbolt 支持。我仍然没有 Thunderbolt 支持(无法确定安全级别)。但是,系统在不到 30 秒的时间内启动(过去需要 120 秒),我没有收到 DMAR 错误,也没有 thunderbolt 0000:05:00.0:无法将驱动程序准备好发送到 ICM 消息。

据称 Thunderbolt 是问题的根源。


答案1

我无法识别“设备 [05:00.0]”,但仔细阅读后,我想我找到了答案。我看到了“雷电 0000:05:00.0”……您是否连接了任何雷电设备?

DMAR: DRHD: handling fault status reg 2

DMAR: [DMA Write] Request device [05:00.0] PASID ffffffff fault addr 6d95f000 [fault reason 05] PTE Write access is not set

thunderbolt 0000:05:00.0: failed to send driver ready to ICM

Ubuntu 20.04。在 BIOS 中关闭 Thunderbolt 支持,现在笔记本电脑可以在不到 30 秒的时间内启动,而不是花 2 分钟并出现大量错误消息。

怀疑有了这样的新硬件,您需要更新到 Ubuntu 21.04 才能获得全面支持。

相关内容