[Bug 287342] kern: Intel ArrowLake / AlderLake / *Lake stability and efficiency issues.

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=287342

            Bug ID: 287342
           Summary: kern: Intel ArrowLake / AlderLake / *Lake stability
                    and efficiency issues.
           Product: Base System
           Version: 14.2-RELEASE
          Hardware: amd64
                OS: Any
            Status: New
          Severity: Affects Some People
          Priority: ---
         Component: kern
          Assignee: bugs@FreeBSD.org
          Reporter: tomek@cedro.info

Hello world :-)

I just switched my 14-2 RELEASE AMD64 to Intel Ultra9 285K CPU (Arrow Lake)
with ASUS TUF Z890 PRO WIFI motherboard, I am using NVME Samsung 9100PRO disk
at PCI-E Gen5 nvme lane/slot (there is room for 3 other disks at Gen4 slots),
not using onboard GPU but NVIDIA GTX1060 on main PCI-Ex16.Gen5, I have one
additional PCI-Ex1 USB 3.0 controller (uPD720200) and additional 6 port SATA
controller for old HDD/BD drives at PCI-Ex4.

https://bsd-hardware.info/?probe=d20947a825

I have noticed some performance issues (audio and video glitches / hangs /
clicks) and stability issues (i.e. all onboard USB controllers hangs randomly).
Installed windoze for comparison to exclude hardware problems and the issues
are not there.

The partial quick-fix for this is to set PCI-Ex16 lane to 8x/8x mode and
disable CPU C-States (at least below C2) in BIOS then machine is kinda usable.

This is new hardware, but I realized other people have similar problems with
Intel ArrowLake and a bit older AlderLake based devices, so I am opening a
dedicated bug report for this issue.

After some talk on mailing lists (i.e. multimedia) it turns out the audio and
video glitches cause may be somewhere below and only gets hit by the problem,
it is highly correlated with computer IO activity, tuning all sorts of sysctls
for kern ipc vfs snd does not help.

https://lists.freebsd.org/archives/freebsd-multimedia/2025-June/003107.html

Some related bug reports:
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=263385
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=237666
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=285156
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=277534
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=274814
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=277437

dmesg shows some problems but not sure if this is related:

smbus0: <System Management Bus> on ichsmb0
acpi_wmi0: <ACPI-WMI mapping> on acpi0
acpi_wmi0: Embedded MOF found
ACPI: \134AWW0.WQMO: 1 arguments were passed to a non-method ACPI object
(Buffer) (20221020/nsarguments-361)
acpi_wmi1: <ACPI-WMI mapping> on acpi0
acpi_wmi1: Embedded MOF found
ACPI: \134ALCW.WQMO: 1 arguments were passed to a non-method ACPI object
(Buffer) (20221020/nsarguments-361)
acpi_wmi2: <ACPI-WMI mapping> on acpi0
acpi_wmi2: Embedded MOF found
ACPI: \134RMTW.WQMO: 1 arguments were passed to a non-method ACPI object
(Buffer) (20221020/nsarguments-361)
acpi_wmi3: <ACPI-WMI mapping> on acpi0
acpi_wmi3: Embedded MOF found
ACPI: \134ABAW.WQMO: 1 arguments were passed to a non-method ACPI object
(Buffer) (20221020/nsarguments-361)
acpi_wmi4: <ACPI-WMI mapping> on acpi0
acpi_wmi4: Embedded MOF found
ACPI: \134AMZW.WQMO: 1 arguments were passed to a non-method ACPI object
(Buffer) (20221020/nsarguments-361)
acpi_wmi5: <ACPI-WMI mapping> on acpi0
acpi_wmi5: Embedded MOF found
ACPI: \134_SB.WFDE.WQCC: 1 arguments were passed to a non-method ACPI object
(Buffer) (20221020/nsarguments-361)
acpi_wmi6: <ACPI-WMI mapping> on acpi0
acpi_wmi6: Embedded MOF found
ACPI: \134_SB.WFTE.WQCC: 1 arguments were passed to a non-method ACPI object
(Buffer) (20221020/nsarguments-361)

ACPI Warning: \134_SB.PC00.RP01.PXSX._DSM: Argument #4 type mismatch - Found
[Buffer], ACPI requires [Package] (20221020/nsarguments-212)
Firmware Error (ACPI): Failure creating named object
[\134_SB.PC00.RP01.PXSX._DSM.USRG], AE_ALREADY_EXISTS (20221020/dsfield-352)
ACPI Error: AE_ALREADY_EXISTS, CreateBufferField failure
(20221020/dswload2-639)
ACPI Error: Aborting method \134_SB.PC00.RP01.PXSX._DSM due to previous error
(AE_ALREADY_EXISTS) (20221020/psparse-689)
Firmware Error (ACPI): Failure creating named object
[\134_SB.PC00.RP01.PXSX._DSM.USRG], AE_ALREADY_EXISTS (20221020/dsfield-352)
ACPI Error: AE_ALREADY_EXISTS, CreateBufferField failure
(20221020/dswload2-639)
ACPI Error: Aborting method \134_SB.PC00.RP01.PXSX._DSM due to previous error
(AE_ALREADY_EXISTS) (20221020/psparse-689)

-- 
You are receiving this mail because:
You are the assignee for the bug.