• Boot-time PCI issues with current-gen hardware


    Hi, I’ve just installed Antergos on a new laptop and I’m stuck with some errors during boot.

    It’s a Clevo P775, 8700K processor, NVidia 1070M, and the only special configuration I’ve done outside cnchi is to set up LVM on LUKs on 4-disk RAID10 using two Samsung 970 EVO drives and two Samsung 850 evo 2.5" drives. It’s set up for EFI boot and an encrypted /boot on the RAID10. I’m using the unlocked Prema BIOS but I haven’t messed with PCIE settings. CSM is off, and I’m in AHCI mode so linux can see the nvme drives.

    During boot I get two errors:

    Unknown controller

    kernel: sdhci-pci 0000:6f:00.0: SDHCI controller found [10ec:5250] (rev 1)
    kernel: sdhci-pci 0000:6f:00.0: enabling device (0000 -> 0002)
    kernel: mmc0: Unknown controller version (3). You may experience problems.
    kernel: mmc0: SDHCI controller on PCI [0000:6f:00.0] using ADMA
    

    And this PCIE bus error below that I can’t figure out. It references a root port, but I don’t know what’s on it.

    kernel: pcieport 0000:00:1c.0: AER: Corrected error received: 0000:00:1c.0
    kernel: pcieport 0000:00:1c.0: PCIe Bus Error: severity=Corrected, type=Physical Layer, (Receiver ID)
    kernel: pcieport 0000:00:1c.0:   device [8086:a290] error status/mask=00000001/00002000
    kernel: pcieport 0000:00:1c.0:    [ 0] Receiver Error         (First)
    kernel: pcieport 0000:04:00.0: Refused to change power state, currently in D3
    

    lspci -tv as pastebin link
    lspci -vv as pastebin link

    dmesg |grep failed:

    [    0.722675] pci 0000:04:02.0: BAR 15: failed to assign [mem size 0x00200000 64bit pref]
    [    0.722680] pci 0000:04:02.0: BAR 15: failed to assign [mem size 0x00200000 64bit pref]
    [   22.502050] thermal thermal_zone2: failed to read out thermal zone (-61)
    

    There are plenty of instructions online about how to mute the errors, but I would rather figure out what the issue is here.

  • Upon further inspection in bios, this port is set (by manufacturer) to topology x4 and has AER + PME SCI enabled. Most others are x1 and don’t have AER enabled. If it’s not NVME, SATA, or GPU, what sort of device could use 4 lanes?

    I don’t understand what PME SCI is, or how to determine when it should be enabled. My first instinct was to entirely disable power management on this link, but that had no effect. I’m not even really sure there’s any device there. TPM is disabled, so maybe that’s it, but TPM shouldn’t be a 4 lane hotplug device.

    Is there a general BIOS guide anywhere that says what these type of settings are meant for beyond “Enable/Disable <Name of thing we don’t explain what it is or does>”?

issues82 pci2 boot-time1 current-gen1 Posts 2Views 174
Log in to reply
Bloom Email Optin Plugin

Looks like your connection to Antergos Community Forum was lost, please wait while we try to reconnect.