Skip to content

Releases: intel/xpumanager

V1.2.8

22 Apr 13:05
@taotod taotod
V1.2.8_AMC_On_DNP
5f03bd2
Compare
Choose a tag to compare
V1.2.8

New features in this release

  • [SMI] Improve the performance of some SMI commands
  • [SMI] Support Debian 10.13 OS
  • [XPU Manager Windows CLI] Count the vGPU render engine utilization into vGPU utilization.
  • [XPU Manager/SMI] Support Max Series AMC firmware update on Intel D50DNP server (BMC firmware version should be v1.53 or newer)

Supported devices

  • Intel(R) Data Center GPU Flex Series
  • Intel(R) Data Center GPU Max Series

Supported OSes

  • XPU-SMI

    • CentOS 8/9 Stream (installer files: xpu-smi-*.el8/el9.x86_64.rpm)
    • CentOS 7.4/7.9 (installer files: xpu-smi-*.el7.x86_64.rpm)
    • Ubuntu 20.04.3/22.04 (installer files: xpu-smi_*.u22.04/22.04_amd64.rpm)
    • RHEL 8.5/8.6 (installer files: xpu-smi-*.el8.x86_64.rpm)
    • SLES 15 SP3/SP4 (installer files: xpu-smi-*.x86_64.rpm)
    • Debian 10.13 (installer files: xpu-smi_*+deb10u1_amd64.deb)

  • XPU Manager

    • Ubuntu 20.04.3/22.04 (installer files: xpumanager_*.u20.04/u22.04_amd64.deb)
    • RHEL 8.5/8.6 (installer files: xpumanager-*.el8.x86_64.rpm)
    • CentOS 8/9 Stream (installer files: xpumanager-*.el8/el9.x86_64.rpm)
    • CentOS 7.4/7.9 (installer files: xpumanager-*.el7.x86_64.rpm)
    • SLES 15 SP3/SP4 (installer files: xpumanager-*.x86_64.rpm)
    • Windows Server 2022/2019 (file: xpumcli.exe, limited supporting: GPU device info, GPU telemetries, GPU settings, GPU firmware update)

Notice

  • How to enable AMC firmware update on HPE DL380 Gen10 server
    • iLO5 firmware version should be v2.78 or newer
    • "Virtual NIC" should be enabled on iLO web console
    • "cdc_eem" should be built in the Linux kernel which can be checked with the command "lsmod | grep cdc_eem"
    • The virtual NIC (usb*) should be manually activated in Linux OS with the commands: "ifconfig usb* up" and "dhclient usb*"
  • How to enable AMC firmware update on Dell PowerEdge R750 server
    • iDRAC firmware version should be v6.10.80.00 or newer
    • Set "Pass-through Mode" of "OS to iDRAC Pass-through" to "USB NIC"
  • How to enable AMC firmware update on Intel D50DNP server
    • BMC firmware version should be v1.53 or newer
    • Enable "Host Interface" in "Configuration" -> "Security Settings" page and save.
    • Set the Host Interface IP address (may like eth2) and save. The IP address may like 192.168.10.1 and subnet mask may be 255.255.255.0.
    • Reboot OS to make Redfish host interface in OS. You may run the command "sudo dmidecode -t 42" to check.
Assets 2
Loading

V1.2.7

08 Apr 04:22
@taotod taotod
V1.2.7
492b124
Compare
Choose a tag to compare
V1.2.7

New features in this release

  • [XPU Manager/SMI/amcmcli] Improve the compatibility of Flex series GPU AMC firmware update on Inspur NF5280M6 server
  • [XPU Manager/SMI/amcmcli] Improve the performance of Flex series GPU AMC firmware update through IPMI protocol
  • [XPU Manager/SMI] Support Flex series GPU AMC firmware update on Dell R750 server
  • [XPU Manager/SMI] Add the BIOS setting checking for SRIOV configuration
  • [SMI] Support CentOS 7.4/7.9 OS

Supported devices

  • Intel(R) Data Center GPU Flex Series
  • Intel(R) Data Center GPU Max Series

Supported OSes

  • XPU-SMI

    • CentOS 8/9 Stream (installer files: xpu-smi-*.el8/el9.x86_64.rpm)
    • CentOS 7.4/7.9 (installer files: xpu-smi-*.el7.x86_64.rpm)
    • Ubuntu 20.04.3/22.04 (installer files: xpu-smi_*.u22.04/22.04_amd64.rpm)
    • RHEL 8.5/8.6 (installer files: xpu-smi-*.el8.x86_64.rpm)
    • SLES 15 SP3/SP4 (installer files: xpu-smi-*.x86_64.rpm)

  • XPU Manager

    • Ubuntu 20.04.3/22.04 (installer files: xpumanager_*.u20.04/u22.04_amd64.deb)
    • RHEL 8.5/8.6 (installer files: xpumanager-*.el8.x86_64.rpm)
    • CentOS 8/9 Stream (installer files: xpumanager-*.el8/el9.x86_64.rpm)
    • CentOS 7.4/7.9 (installer files: xpumanager-*.el7.x86_64.rpm)
    • SLES 15 SP3/SP4 (installer files: xpumanager-*.x86_64.rpm)
    • Windows Server 2022/2019 (file: xpumcli.exe, limited supporting: GPU device info, GPU telemetries, GPU settings, GPU firmware update)

Notice

  • How to enable AMC firmware update on HPE DL380 Gen10 server
    • iLO5 firmware version should be v2.78 or newer
    • "Virtual NIC" should be enabled on iLO web console
    • "cdc_eem" should be built in the Linux kernel which can be checked with the command "lsmod | grep cdc_eem"
    • The virtual NIC (usb*) should be manually activated in Linux OS with the commands: "ifconfig usb* up" and "dhclient usb*"
  • How to enable AMC firmware update on Dell PowerEdge R750 server
    • iDRAC firmware version should be v6.10.80.00 or newer
    • Set "Pass-through Mode" of "OS to iDRAC Pass-through" to "USB NIC"
Assets 2
Loading

V1.2.6

24 Mar 02:45
@taotod taotod
V1.2.6
36f1eb8
Compare
Choose a tag to compare
V1.2.6

New features in this release

  • [XPU Manager/SMI] Add compute/codec functional test into GPU diagnostics particular test suite
  • [XPU Manager/SMI] Improve the performance of GPU diagnostics pre-check
  • [XPU Manager/SMI] Add the GPU "PCI Device ID" and "PCI Vendor ID" into discovery command
  • [XPU Manager/SMI] Report the Flex series GPU memory bandwidth on the latest GPU driver
  • [XPU Manager/SMI] Show the ECC state for Max series GPU
  • [XPU Manager/SMI] Add CentOS 9 Stream support
  • [XPU Manager/SMI] Support Flex series GPU AMC firmware update on Supermicro SYS-420GP-TNR server (BMC: v09.93.09)
  • [Windows CLI tool] Support Flex series GPU AMC firmware update on Intel M50CYP server

Supported devices

  • Intel(R) Data Center GPU Flex Series
  • Intel(R) Data Center GPU Max Series

Supported OSes

  • XPU Manager

    • Ubuntu 20.04.3/22.04
    • RHEL 8.5/8.6
    • CentOS 8/9 Stream
    • CentOS 7.4/7.9
    • SLES 15 SP3/SP4
    • Windows Server 2022/2019 (limited supporting: GPU device info, GPU telemetries, GPU settings, GPU GFX firmware update)

  • XPU-SMI

    • CentOS 8/9 Stream
    • Ubuntu 20.04.3/22.04
    • RHEL 8.6
    • SLES 15 SP4

Notice

  • How to enable AMC firmware update on HPE DL380 Gen10 server
    • iLO5 firmware version should be v2.78 or newer
    • "Virtual NIC" should be enabled on iLO web console
    • "cdc_eem" should be built in the Linux kernel which can be checked with the command "lsmod | grep cdc_eem"
    • The virtual NIC (usb*) should be manually activated in Linux OS with the commands: "ifconfig usb* up" and "dhclient usb*"
Assets 2
Loading

V1.2.5

13 Mar 07:17
@taotod taotod
V1.2.5
abe6b3a
Compare
Choose a tag to compare
V1.2.5

New features in this release

  • [XPU Manager/SMI] Provide the device name and PCIe slot location in discovery dump command for the problematic GPUs
  • [XPU Manager/SMI] Make level-1 diagnostics work on PF when SRIOV is enabled.
  • [XPU Manager/SMI] Add the parameter in diagnostic precheck command to filter CPU status
  • [XPU Manager/SMI] Diagnostic single test can be multiply selected.
  • [XPU Manager/SMI] Show the SKU type in discovery command
  • [XPU Manager/SMI] Dump command can report multi-tile telemetries
  • [Windows CLI tool] Provide the vGPU engine utilization and memory usage in SRIOV configuration

Supported devices

  • Intel(R) Data Center GPU Flex Series
  • Intel(R) Data Center GPU Max Series

Supported OSes

  • XPU Manager

    • Ubuntu 20.04.3/22.04
    • RHEL 8.5/8.6
    • CentOS 8 Stream
    • CentOS 7.4/7.9
    • SLES 15 SP3/SP4
    • Windows Server 2022/2019 (limited supporting: GPU device info, GPU telemetries, GPU settings, GPU GFX firmware update)

  • XPU-SMI

    • CentOS 8 Stream
    • Ubuntu 20.04.3/22.04
    • RHEL 8.6
    • SLES 15 SP4

Notice

  • How to enable AMC firmware update on HPE DL380 Gen10 server
    • iLO5 firmware version should be v2.78 or newer
    • "Virtual NIC" should be enabled on iLO web console
    • "cdc_eem" should be built in the Linux kernel which can be checked with the command "lsmod | grep cdc_eem"
    • The virtual NIC (usb*) should be manually activated in Linux OS with the commands: "ifconfig usb* up" and "dhclient usb*"
Assets 2
Loading

V1.2.4

28 Feb 06:53
@taotod taotod
V1.2.4
a3d36a2
Compare
Choose a tag to compare
V1.2.4

New features in this release

  • [XPU Manager/SMI] Show the problematic GPU BDF/firmware versions and status in discovery --dump command when some GPUs don't work.
  • [XPU Manager/SMI] Change the package names to follow the common Linux package naming

Supported devices

  • Intel(R) Data Center GPU Flex Series
  • Intel(R) Data Center GPU Max Series

Supported OSes

  • XPU Manager

    • Ubuntu 20.04.3/22.04
    • RHEL 8.5/8.6
    • CentOS 8 Stream
    • CentOS 7.4/7.9
    • SLES 15 SP3/SP4
    • Windows Server 2022/2019 (limited supporting: GPU device info, GPU telemetries, GPU settings, GPU GFX firmware update)

  • XPU-SMI

    • CentOS 8 Stream
    • Ubuntu 20.04.3/22.04
    • RHEL 8.6
    • SLES 15 SP4

Notice

  • How to enable AMC firmware update on HPE DL380 Gen10 server
    • iLO5 firmware version should be v2.78 or newer
    • "Virtual NIC" should be enabled on iLO web console
    • "cdc_eem" should be built in the Linux kernel which can be checked with the command "lsmod | grep cdc_eem"
    • The virtual NIC (usb*) should be manually activated in Linux OS with the commands: "ifconfig usb* up" and "dhclient usb*"
Assets 2
Loading

V1.2.3

23 Feb 06:22
@taotod taotod
V1.2.3
0281ccc
Compare
Choose a tag to compare
V1.2.3

New features in this release

  • [XPU Manager/SMI] Add GPU basic functional test into diagnostics level 1 test
  • [XPU Manager/SMI] Provide the particular test in GPU diagnostics
  • [XPU Manager/SMI] Automatically set up Redfish host virtual NIC on HPE server
  • [XPU Manager/SMI] Refine the format of GPU diag precheck output
  • [XPU Manager/SMI] Make diag precheck feature independent of dmesg

Supported devices

  • Intel(R) Data Center GPU Flex Series
  • Intel(R) Data Center GPU Max Series

Supported OSes

  • XPU Manager

    • Ubuntu 20.04.3/22.04
    • RHEL 8.5/8.6
    • CentOS 8 Stream
    • CentOS 7.4/7.9
    • SLES 15 SP3/SP4
    • Windows Server 2022/2019 (limited supporting: GPU device info, GPU telemetries, GPU settings, GPU GFX firmware update)

  • XPU-SMI

    • CentOS 8 Stream
    • Ubuntu 20.04.3/22.04
    • RHEL 8.6
    • SLES 15 SP4

Notice

  • How to enable AMC firmware update on HPE DL380 Gen10 server
    • iLO5 firmware version should be v2.78 or newer
    • "Virtual NIC" should be enabled on iLO web console
    • "cdc_eem" should be built in the Linux kernel which can be checked with the command "lsmod | grep cdc_eem"
    • The virtual NIC (usb*) should be manually activated in Linux OS with the commands: "ifconfig usb* up" and "dhclient usb*"
Assets 2
Loading

V1.2.2

11 Feb 03:47
@taotod taotod
V1.2.2Hotfix
b649944
Compare
Choose a tag to compare
V1.2.2

New features in this release

  • [XPU Manager/SMI] Change the default installation folder to /usr/bin and /usr/lib
  • [XPU Manager/SMI] Fix the topology issue: tile 1 can't be shown for Max Series
  • [XPU Manager/SMI] Improve the i915 driver display
  • [XPU Manager/SMI] Discovery command only shows PF by default
  • [XPU Manager/SMI] Support updating AMC firmware update on HPE DL380 Gen 10 server (with iLO5 firmware v2.78)
  • [SMI] Add AMC firmware update on Intel M50CYP and Supermicro SYS-620C-TN12R servers.
  • [SMI] Improve the stats command performance for Max Series.

Supported devices

  • Intel(R) Data Center GPU Flex Series
  • Intel(R) Data Center GPU Max Series

Supported OSes

  • XPU Manager

    • Ubuntu 20.04.3/22.04
    • RHEL 8.5/8.6
    • CentOS 8 Stream
    • CentOS 7.4/7.9
    • SLES 15 SP3/SP4
    • Windows Server 2022/2019 (limited supporting: GPU device info, GPU telemetries, GPU settings, GPU GFX firmware update)

  • XPU-SMI

    • CentOS 8 Stream
    • Ubuntu 20.04.3/22.04
    • RHEL 8.6
    • SLES 15 SP4

Notice

  • How to enable AMC firmware update on HPE DL380 Gen10 server
    • iLO5 firmware version should be v2.78 or newer
    • "Virtual NIC" should be enabled on iLO web console
    • "cdc_eem" should be built in the Linux kernel which can be checked with the command "lsmod | grep cdc_eem"
    • The virtual NIC (usb*) should be manually activated in Linux OS with the commands: "ifconfig usb* up" and "dhclient usb*"
Assets 2
Loading

V1.2.1

13 Jan 05:45
@taotod taotod
V1.2.1Hotfix
5e670f4
Compare
Choose a tag to compare
V1.2.1

New features in this release

  • [XPU Manager/SMI] Report the per-process GPU memory usage.
  • [XPU Manager/SMI] Report the ATSM firmware status and add the new parameter recover the GPU firmware.
  • [XPU Manager/SMI] Use the locate time in CLI stats and dump command.
  • [XPU Manager] Report ATSM serial number on Coyote Pass server.
  • [XPU Manager] Dump the GPU telemetries to CSV file in XPU Manager Windows CLI tool
  • [XPU Manager/SMI] Fix the issue: GFX Data can't be updated in 1.2
  • [AMC Manager CLI] Provide the separate ATSM AMC FW Management CLI tool to remove GPU driver dependency.

Supported devices

  • Intel(R) Data Center GPU Flex Series
  • Intel(R) Data Center GPU Max Series

Supported OSes

  • XPU Manager
    -- Ubuntu 20.04.3/22.04
    -- RHEL 8.5/8.6
    -- CentOS 8 Stream
    -- CentOS 7.4/7.9
    -- SLES 15 SP3/SP4
    -- Windows Server 2022/2019 (limited supporting: GPU device info, GPU telemetries, GPU settings, GPU GFX firmware update)

  • XPU-SMI
    -- CentOS 8 Stream
    -- Ubuntu 20.04.3/22.04
    -- RHEL 8.6
    -- SLES 15 SP4

Assets 2
Loading

1.2.0 Sprint 17 Release

28 Dec 02:26
@taotod taotod
1.2_sprint17
04ebef9
Compare
Choose a tag to compare
1.2.0 Sprint 17 Release

New features in this release

  • [XPU Manager/SMI] Support the GPU memory error check in GPU diagnostics.
  • [XPU Manager/SMI] Collect GPU log info for the issue investigation.
  • [XPU Manager/SMI] Better model vGPU in SR-IOV mode.
  • [XPU Manager] Make AMC FW update through IPMI KCS work on non Intel servers.
  • [XPU Manager] Provide the Xe-Link topology in the Prometheus Exporter.

Supported devices

  • Intel(R) Data Center GPU Flex Series

Supported OSes

  • XPU Manager
    -- Ubuntu 20.04.3/22.04
    -- RHEL 8.5/8.6
    -- CentOS 8 Stream
    -- CentOS 7.4/7.9
    -- SLES 15 SP3/SP4
    -- Windows Server 2022/2019 (limited supporting: GPU device info, GPU telemetries, GPU settings, GPU GFX firmware update)

  • XPU-SMI
    -- CentOS 8 Stream
    -- Ubuntu 20.04.3/22.04
    -- RHEL 8.6
    -- SLES 15 SP4

Assets 2
Loading

1.2.0 Golden Release

25 Nov 11:48
@taotod taotod
v1.2.0_golden
9508677
Compare
Choose a tag to compare
1.2.0 Golden Release

New features in this release

  • [XPU Manager/XPU-SMI] Add the throttle reason into dump feature.
  • [XPU Manager/XPU-SMI] Show the GPU memory ECC state in discovery command
  • [XPU Manager/XPU-SMI] Support multiple GPU in dump stats library API
  • [XPU Manager/XPU-SMI] Report the socket ID for the GPU OAM configuration
  • [XPU Manager/XPU-SMI] Resolve the discovery JSON output compatibility issue.

Supported devices

  • Intel(R) Data Center GPU Flex Series

Supported OSes

  • XPU Manager
    -- Ubuntu 20.04.3/22.04
    -- RHEL 8.5/8.6
    -- CentOS 8 Stream
    -- CentOS 7.4/7.9
    -- SLES 15 SP3
    -- Windows Server 2022/2019 (limited supporting: GPU device info, GPU telemetries, GPU settings, GPU GFX firmware update)

  • XPU-SMI
    -- CentOS 8 Stream
    -- Ubuntu 20.04.3/22.04
    -- RHEL 8.6
    -- SLES 15 SP4

Notice

  • For Flex Series GPU GSC firmware update, please make sure to install iGSC v0.8.4 or newer. If you still use the old iGSC version (such as v0.7.1), please disable GPU power management by ""/sys/bus/platform/devices/i915.mei-gscfi.(xxxxx)/power/control" on all installed GPU devices before updating Flex Series GSC firmware
  • Since this 1.1.0 sprint version, XPU Manager Windows CLI tool supports Flex Series GPU memory ECC on/off and Flex Series GPU firmware update features. Please install igsc.dll v0.8.4 or newer versions in your Windows OS
Assets 13
Loading