Skip to contentSkip to conversation

New: Unable to power off iMac16,2]

[email protected]: Bjorn Helgaas 1 participant 1 message Aug 2, 2021 PCI

lore
Share & tools
lore
#0 lore raw

Thanks for the report, Kevin.

Could you also attach the dmesg log from booting with "pci=earlydump"? The kernel really should log the BARs and windows we get from firmware before we change anything, but we currently do not, so "pci=earlydump" is the only way to get that.

----- Forwarded message from [email protected] -----

Date: Mon, 02 Aug 2021 16:02:47 +0000
From: [email protected]
To: [email protected]
Subject: [Bug 213945] New: Unable to power off iMac16,2
Message-ID: <[email protected]/>

https://bugzilla.kernel.org/show_bug.cgi?id=213945

            Bug ID: 213945
           Summary: Unable to power off iMac16,2
           Product: Drivers
           Version: 2.5
    Kernel Version: 5.4.0
          Hardware: Intel
                OS: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: PCI
          Assignee: [email protected]
          Reporter: [email protected]
        Regression: No

Created attachment 298155
--> https://bugzilla.kernel.org/attachment.cgi?id=298155&action=edit
dmesg

I believe this bug is very similar if not the same to
https://bugzilla.kernel.org/show_bug.cgi?id=103211.

I have managed to work around it and have the computer power off by using the two setpci commands from comment 298 in bug report 103211 in a systemd shutdown script placed in /lib/systemd/system-shutdown.

#!/bin/sh

case $1 in

        poweroff)
        setpci -s1c.0 0x20.l=0xf011f001
        setpci -s1c.0 0x24.l=0xf031f021
        ;;

esac

Without the script the computer will just hang at shutdown. It also will not suspend properly. It looks like it does, but then it will not wake up and I have to hard power it down by holding the power button for 10 seconds before I can turn it back on again.

I am attaching the output of dmesg, lspci, dmidecode and /proc/iomem and /proc/ioports to aid in possibly extending the quirk fix from the previous bug report. Hopefully the issue is similar enough that it will be a quick fix.

I do believe I am running the latest BIOS (427.140.8.0.0 06/13/2021) which I believe was installed when I installed Big Sur.

Signature · 6 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.

----- End forwarded message -----