Release Notes

Tracking number 141837
Servers with the AMD Rome processor are unable to
read the boost states using cpupower
Description Servers with the AMD Rome processor are unable to read the boost states using the cpupower tool. The
cpupower tool is unable to read the mperf/aperf registers in the AMD Rome processor, and thus unable
to collect the information about boost states. It reports the number of boost states as zero.
Applies to SUSE Linux Enterprise Server 15 SP1
Solution: The issue is resolved in cpupower-4.19-6.5.1.x86_64.rpm version. For more information, see SUSE Linux
Enterprise Server Knowledge Base article 000019637.
Systems affected All Dell EMC PowerEdge servers
Tracking number 149656
PowerEdge servers with the AMD Rome processor fail
to detect an NVMe drive after multiple hot plugs
Description:
The PowerEdge servers with the AMD Rome processor fail to detect an NVMe drive after multiple hot
plugs.
Applies to:
SUSE Linux Enterprise Server 15 SP1
Cause: Issue in pciehp driver.
Workaround:
NOTE: The output of each command varies based on your system configuration.
Follow the steps:
1. Identify the parent device to which the NVMe device is connected using the following command:
lspci -t
-+-[0000:e0]-+-00.0
| +-00.2
| +-01.0
| +-02.0
| +-03.0
| +-03.1-[e2]----00.0
From the above snippet, the device e2:00.0 is the NVMe device and the device e0:03.1 is the
parent device. The parent device is used for the steps that are described subsequently.
2. Read the Slot Status Register in the PCI Express Capability structure by running the following
command:
setpci -s e0:03.1 CAP_EXP+0x1a.w
Output: 0148 (value returned)
3. Clear the event bits that are impacted by running the following command:
setpci -s e0:03.1 CAP_EXP+0x1a.w=0x0108
4. Re-read the Slot Status Register to confirm that event bits are cleared by running the following
command:
setpci -s e0:03.1 CAP_EXP+0x1a.w
Fixes
13