[BUG] mt7925e: MSI Vector A16 HX A8WHG-004US reports permanent hardware rfkill block under Linux; ACPI GPPA.WLAN AE_ALREADY_EXISTS

[Sean Wang]

Hi,

On Mon, Mar 9, 2026 at 2:20 PM Ross <rosspayant at gmail.com> wrote:
>
> Hello, I want to report a wifi bug. This bug makes the wifi hard lock
> on my computer under Arch Linux as well as Fedora. Wifi does work on
> Linux LTE and Windows 11. The computer is a Vector A16 HX A8WHG-004US
> and this it's actually the second I bought. I returned the first one
> because of the exact same bug. Since I got the second one I've stuck
> it through hoping there would be a fix, it's been three months now.
> I've tried all the possible fixes (on both machines), nothing. "sudo
> rfkill unblock all" and "sudo rfkill unblock wifi" do not work. There
> is no switch in the bios or "airplane mode" hotkey to toggle either.
> The wifi just does not work in Linux (apart from LTE). I've updated
> the bios to the new version every time a new one drops (including the
> most recent one from a couple days ago) and nothing has changed
> anything. I've updated everything else frequently. No dice.
>
> Anyway, I suppose this is where I go to file report bugs like this?
> I've never issued a bug report before so I'm not sure. Thank you so
> much for giving this the time of day, as well as all the other work
> you do on the LInux project. If you need further information I'll give
> you whatever you need.
>
> I had the following written for me by AI but I can confirm it's
> accurate to the problem:
>
>
> I am reporting what appears to be an MSI-platform-specific MT7925
> hard-rfkill issue on Linux.
>
> Hardware
> - Laptop: MSI Vector A16 HX A8WHG-004US
> - WLAN: MEDIATEK MT7925 802.11be 160MHz 2x2 PCIe Wireless Network
> Adapter [14c3:7925]
> - Subsystem: Foxconn International, Inc. Device [105b:e138]
>
> Software
> - BIOS: E15MMAMS.107
> - Kernel: 6.19.6-arch1-1
> - linux-firmware: 20260221-1
> - Distro: Arch Linux
>
> Problem
> - The Wi-Fi device is detected correctly.
> - The mt7925e driver binds successfully.
> - Firmware appears to load.
> - The interface is created/renamed to wlp5s0.
> - However, rfkill always reports the WLAN as hard blocked.
>
> Current rfkill state
> - Bluetooth: Soft blocked: no / Hard blocked: no
> - Wireless LAN: Soft blocked: no / Hard blocked: yes
>
> Attempted commands
> - sudo rfkill unblock wifi
> - sudo rfkill unblock all
>
> Neither command changes the WLAN hard block.
>
> Relevant observations
> - lspci shows:
> MEDIATEK Corp. MT7925 802.11be 160MHz 2x2 PCIe Wireless Network
> Adapter [Filogic 360] [14c3:7925]
> Kernel driver in use: mt7925e
> - The boot log shows the device being enabled, ASIC revision detected,
> firmware information printed, and the interface renamed from wlan0 to
> wlp5s0.
> - The same boot log also shows repeated ACPI errors involving WLAN
> objects under \_SB.PCI0.GPPA.WLAN..., including repeated
> AE_ALREADY_EXISTS failures.
>
> This makes it look like:
> - the device is present,
> - the mt7925e driver is loading,
> - but platform/ACPI state may be forcing or misreporting hardware rfkill.
>
> Relevant kernel log excerpt
>
> mt7925e 0000:05:00.0: enabling device (0000 -> 0002)
> mt7925e 0000:05:00.0: ASIC revision: 79250000
> mt7925e 0000:05:00.0: HW/SW Version: 0x8a108a10, Build Time: 20260106153007a
> mt7925e 0000:05:00.0: WM Firmware Version: ____000000, Build Time:
> 20260106153120
> mt7925e 0000:05:00.0 wlp5s0: renamed from wlan0
>

>From the logs so far, this does not look like a basic mt7925e
initialization failure. The device probes normally, firmware loads,
and the interface is created, but the WLAN still ends up reported as
hard blocked, so Wi-Fi never becomes usable. Typically, this kind of
blocked state comes from a platform-level source, such as a hardware
radio switch or platform-provided radio state, rather than from the
mt7925e path itself.

> ACPI BIOS Error (bug): Failure creating named object
> [\_SB.PCI0.GPPA.WLAN._S0W], AE_ALREADY_EXISTS
> ACPI BIOS Error (bug): Failure creating named object
> [\_SB.PCI0.GPPA.WLAN._S4W], AE_ALREADY_EXISTS
> ACPI BIOS Error (bug): Failure creating named object
> [\_SB.PCI0.GPPA.WLAN._DSM], AE_ALREADY_EXISTS
> ACPI BIOS Error (bug): Failure creating named object
> [\_SB.PCI0.GPPA.WLAN.PCIF], AE_ALREADY_EXISTS
> ACPI BIOS Error (bug): Failure creating named object
> [\_SB.PCI0.GPPA.WLAN.NVID], AE_ALREADY_EXISTS
> ACPI BIOS Error (bug): Failure creating named object
> [\_SB.PCI0.GPPA.WLAN.NDID], AE_ALREADY_EXISTS
> ACPI BIOS Error (bug): Failure creating named object
> [\_SB.PCI0.GPPA.WLAN.PWR1], AE_ALREADY_EXISTS
> ACPI BIOS Error (bug): Failure creating named object
> [\_SB.PCI0.GPPA.WLAN._PR0], AE_ALREADY_EXISTS
> ACPI BIOS Error (bug): Failure creating named object
> [\_SB.PCI0.GPPA.WLAN._PR2], AE_ALREADY_EXISTS
> ACPI BIOS Error (bug): Failure creating named object
> [\_SB.PCI0.GPPA.WLAN._PR3], AE_ALREADY_EXISTS
> ACPI BIOS Error (bug): Failure creating named object
> [\_SB.PCI0.GPPA.WLAN.PWFR], AE_ALREADY_EXISTS
> ACPI BIOS Error (bug): Failure creating named object
> [\_SB.PCI0.GPPA.WLAN._PRR], AE_ALREADY_EXISTS
>
> Reproduction
> 1. Boot Linux normally.
> 2. Run:
> rfkill list all
> 3. Wireless LAN is always reported as:
> Soft blocked: no
> Hard blocked: yes
>
> Additional notes
> - Windows 11 on the same laptop can use Wi-Fi normally.
> - Updating BIOS from E15MMAMS.106 to E15MMAMS.107 did not resolve the issue.
> - This may be similar in shape to other MT7925 hard-rfkill issues that
> required a machine-specific quirk, but on this MSI model the prominent
> clue is the repeated ACPI GPPA.WLAN object collision.
>
> Please let me know if you want:
> - full lspci -nnk output
> - full dmesg/journalctl -b -k
> - acpidump / decoded DSDT
> - additional testing with a proposed patch or boot parameter
>
> --
> Ross Payant
>