Bug 2401216 - Regression in mediatek mt7922 wifi
Summary: Regression in mediatek mt7922 wifi
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Fedora
Classification: Fedora
Component: linux-firmware
Version: 42
Hardware: x86_64
OS: Linux
unspecified
high
Target Milestone: ---
Assignee: David Woodhouse
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard: AcceptedBlocker
: 2400176 2402432 (view as bug list)
Depends On:
Blocks: F43FinalBlocker
TreeView+ depends on / blocked
 
Reported: 2025-10-03 11:36 UTC by Yegor
Modified: 2025-11-05 08:29 UTC (History)
40 users (show)

Fixed In Version: linux-firmware-20251021-1.fc43 linux-firmware-20251021-1.fc42 linux-firmware-20251021-1.fc41
Clone Of:
Environment:
Last Closed: 2025-10-23 23:54:17 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)

Description Yegor 2025-10-03 11:36:35 UTC
After update to  20250917-2.fc42. Wifi page in settings is gone. Wifi tile in a quick settings is gone too. My device is ASUS Zenbook UM3406KA. My wifi adapter is mediatek mt7922. Here is a dmesg:


[    6.203361] mt7921e 0000:62:00.0: ASIC revision: 79220010
[    6.279975] mt7921e 0000:62:00.0: HW/SW Version: 0x8a108a10, Build Time: 20250903123148a
[    6.300996] mt7921e 0000:62:00.0: WM Firmware Version: ____000000, Build Time: 20250903123243
[    6.375450] Modules linked in: snd_acp70 snd_acp_i2s snd_acp_pdm snd_acp_pcm snd_sof_amd_acp70 snd_sof_amd_acp63 snd_sof_amd_vangogh snd_sof_amd_rembrandt snd_sof_amd_renoir snd_sof_amd_acp snd_sof_pci snd_sof_xtensa_dsp amd_atl intel_rapl_msr snd_sof intel_rapl_common snd_hda_codec_realtek(+) mt7921e snd_sof_utils mt7921_common snd_hda_codec_generic snd_pci_ps snd_hda_scodec_component snd_soc_acpi_amd_match mt792x_lib snd_hda_codec_hdmi snd_amd_sdw_acpi soundwire_amd mt76_connac_lib uvcvideo edac_mce_amd soundwire_generic_allocation btusb mt76 soundwire_bus uvc btrtl snd_hda_intel videobuf2_vmalloc btintel kvm_amd videobuf2_memops btbcm snd_intel_dspcfg videobuf2_v4l2 snd_soc_sdca btmtk snd_intel_sdw_acpi snd_soc_core snd_hda_scodec_cs35l41_spi snd_ctl_led videobuf2_common snd_compress snd_hda_codec kvm mac80211 ac97_bus regmap_spi bluetooth snd_pcm_dmaengine videodev snd_hda_core snd_rpl_pci_acp6x snd_acp_pci irqbypass snd_hwdep mc asus_nb_wmi snd_amd_acpi_mach snd_seq rapl asus_wmi libarc4 snd_acp_legacy_common
[    6.375700]  mt76_dma_tx_cleanup+0x15a/0x330 [mt76]
[    6.375713]  mt792x_poll_tx+0x58/0x150 [mt792x_lib]
[   21.671513] mt7921e 0000:62:00.0: Message 0004005c (seq 12) timeout
[   24.808563] mt7921e 0000:62:00.0: Message 00000010 (seq 13) timeout
[   24.808569] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   27.944334] mt7921e 0000:62:00.0: Message 00000010 (seq 14) timeout
[   27.944351] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   31.015698] mt7921e 0000:62:00.0: Message 00000010 (seq 15) timeout
[   31.015716] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   34.087684] mt7921e 0000:62:00.0: Message 00000010 (seq 1) timeout
[   34.087694] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   37.095901] mt7921e 0000:62:00.0: Message 00000010 (seq 2) timeout
[   37.095908] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   39.932799] mt7921e 0000:62:00.0: Message 00000010 (seq 3) timeout
[   39.932815] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   42.767365] mt7921e 0000:62:00.0: Message 00000010 (seq 4) timeout
[   42.767380] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   45.631246] mt7921e 0000:62:00.0: Message 00000010 (seq 5) timeout
[   45.631257] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   45.631312] mt7921e 0000:62:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0010 address=0xffe57940 flags=0x0050]
[   45.631326] mt7921e 0000:62:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0010 address=0xffbc6940 flags=0x0050]
[   45.631334] mt7921e 0000:62:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0010 address=0xffbde940 flags=0x0000]
[   45.631341] mt7921e 0000:62:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0010 address=0xffbde980 flags=0x0000]
[   48.770168] mt7921e 0000:62:00.0: Message 00000010 (seq 6) timeout
[   48.770183] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   48.770280] mt7921e 0000:62:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0010 address=0xffdce940 flags=0x0000]
[   48.834849] mt7921e 0000:62:00.0: hardware init failed


Reproducible: Always

Comment 1 Yegor 2025-10-03 11:44:26 UTC
@dracut-maint-list

Comment 2 moosager90 2025-10-03 14:24:51 UTC
I can confirm this happened to me as well on that same laptop (ASUS Zenbook UM3406KA). I will post logs when I can get them from my offline laptop.

Comment 3 Peter Robinson 2025-10-03 14:40:34 UTC
Can you 'rpm -qa \*firmware\*', I'm guessing you have mt7xxx-firmware but not mediatek-firmware. I split out the actual mediatek SoC firmware from linux-firmware/mt7xxx, but may have got the split wrong or misidentified a wifi FW as a SoC FW.

Comment 4 Peter Robinson 2025-10-03 14:41:36 UTC
Sadly the dmesg doesn't seem to identify the file that it's failing to load

Comment 5 moosager90 2025-10-03 14:58:29 UTC
I did not have mediatek-firmware installed before, but the issue persists even after installing it.

rpm -qa \*firmware\* before installing mediatek-firmware:

alsa-sof-firmware-2025.05.1-1.fc42.noarch
linux-firmware-whence-20250917-2.fc42.noarch
iwlwifi-mld-firmware-20250917-2.fc42.noarch
linux-firmware-20250917-2.fc42.noarch
iwlwifi-mvm-firmware-20250917-2.fc42.noarch
amd-gpu-firmware-20250917-2.fc42.noarch
amd-ucode-firmware-20250917-2.fc42.noarch
atheros-firmware-20250917-2.fc42.noarch
brcmfmac-firmware-20250917-2.fc42.noarch
cirrus-audio-firmware-20250917-2.fc42.noarch
intel-audio-firmware-20250917-2.fc42.noarch
intel-gpu-firmware-20250917-2.fc42.noarch
intel-vsc-firmware-20250917-2.fc42.noarch
iwlegacy-firmware-20250917-2.fc42.noarch
iwlwifi-dvm-firmware-20250917-2.fc42.noarch
libertas-firmware-20250917-2.fc42.noarch
mt7xxx-firmware-20250917-2.fc42.noarch
nvidia-gpu-firmware-20250917-2.fc42.noarch
nxpwireless-firmware-20250917-2.fc42.noarch
qcom-wwan-firmware-20250917-2.fc42.noarch
realtek-firmware-20250917-2.fc42.noarch
tiwilink-firmware-20250917-2.fc42.noarch

After installing mediatek-firmware, a new line was added to that as expected:
mediatek-firmware-20250917-2.fc42.noarch

After installing mediatek-firmware, my dmesg matches that of the original poster:

[   14.064786] mt7921e 0000:62:00.0: enabling device (0000 -> 0002)
[   14.076365] mt7921e 0000:62:00.0: ASIC revision: 79220010
[   14.152423] mt7921e 0000:62:00.0: HW/SW Version: 0x8a108a10, Build Time: 20250903123148a
[   14.534356] mt7921e 0000:62:00.0: WM Firmware Version: ____000000, Build Time: 20250903123243
[   16.211375] mt7921e 0000:62:00.0: driver own failed
[   17.342504] mt7921e 0000:62:00.0: driver own failed
[   17.959393] mt7921e 0000:62:00.0: Message 0004005c (seq 14) timeout
[   21.096511] mt7921e 0000:62:00.0: Message 00000010 (seq 15) timeout
[   21.096524] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   24.231462] mt7921e 0000:62:00.0: Message 00000010 (seq 1) timeout
[   24.231476] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   27.305878] mt7921e 0000:62:00.0: Message 00000010 (seq 2) timeout
[   27.305893] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   30.439535] mt7921e 0000:62:00.0: Message 00000010 (seq 3) timeout
[   30.439551] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   33.511582] mt7921e 0000:62:00.0: Message 00000010 (seq 4) timeout
[   33.511604] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   36.649758] mt7921e 0000:62:00.0: Message 00000010 (seq 5) timeout
[   36.649775] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   39.783632] mt7921e 0000:62:00.0: Message 00000010 (seq 6) timeout
[   39.783651] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   42.855669] mt7921e 0000:62:00.0: Message 00000010 (seq 7) timeout
[   42.855684] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   45.927701] mt7921e 0000:62:00.0: Message 00000010 (seq 8) timeout
[   45.927717] mt7921e 0000:62:00.0: Failed to get patch semaphore
[   45.999577] mt7921e 0000:62:00.0: hardware init failed


My dmesg before installing mediatek-firmware only had lines like these (note: I didn't save them before installing mediatek-firmware, so I extracted these from my current dmesg log just for reference):

[   14.064786] mt7921e 0000:62:00.0: enabling device (0000 -> 0002)
[   14.076365] mt7921e 0000:62:00.0: ASIC revision: 79220010
[   14.152423] mt7921e 0000:62:00.0: HW/SW Version: 0x8a108a10, Build Time: 20250903123148a
[   14.534356] mt7921e 0000:62:00.0: WM Firmware Version: ____000000, Build Time: 20250903123243

For reference: I had the LVFS repository disabled before, and I only enabled it and ran the associated updates today in an attempt to fix this kernel panic: https://bugzilla.redhat.com/show_bug.cgi?id=2360340

Comment 6 Peter Robinson 2025-10-03 15:27:27 UTC
> For reference: I had the LVFS repository disabled before, and I only enabled
> it and ran the associated updates today in an attempt to fix this kernel
> panic: https://bugzilla.redhat.com/show_bug.cgi?id=2360340

Is that meant to be related? If not please don't provide confusion for this report.

Comment 7 Peter Robinson 2025-10-03 15:28:30 UTC
(In reply to Yegor from comment #0)
> After update to  20250917-2.fc42. Wifi page in settings is gone. Wifi tile
> in a quick settings is gone too. My device is ASUS Zenbook UM3406KA. My wifi
> adapter is mediatek mt7922. Here is a dmesg:

What was the prior revision?

Comment 8 Yegor 2025-10-03 16:02:43 UTC
(In reply to Peter Robinson from comment #7)
> (In reply to Yegor from comment #0)
> > After update to  20250917-2.fc42. Wifi page in settings is gone. Wifi tile
> > in a quick settings is gone too. My device is ASUS Zenbook UM3406KA. My wifi
> > adapter is mediatek mt7922. Here is a dmesg:
> 
> What was the prior revision?

Sadly I don't remember, but it was last version before 20250917-2.fc42 available in fedora repos (I checked fore updates yesterday and there was nothing to update)

Comment 9 egponent 2025-10-03 20:25:10 UTC
I also experienced this bug. After BTRFS rollback last working version is 20250808-1.fc42. Non-working version for me is 20250917-2.fc42. Reply if additional info is desired.

Comment 10 D. Vale 2025-10-04 22:50:14 UTC
Confirmed regression on ASUS Vivobook S14 (Ryzen AI HX 370 / MT7922): Wi-Fi breaks with linux-firmware 20250917-2.fc42, fixed by rollback to 20250708

System:
ASUS Vivobook S14 (Ryzen AI 9 HX 370, XDNA NPU)
Fedora 42 x86_64
Linux 6.16.9-200.fc42.x86_64

Adapter details (from lspci -nn):
62:00.0 Network controller [0280]: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter [14c3:0616]
Subsystem: Foxconn International, Inc. Device [105b:e0de]

Firmware packages currently installed (after rollback):
linux-firmware-20250708-1.fc42.noarch
mt7xxx-firmware-20250708-1.fc42.noarch
linux-firmware-whence-20250917-2.fc42.noarch

dmesg (working state):
mt7921e 0000:62:00.0: ASIC revision: 79220010
mt7921e 0000:62:00.0: HW/SW Version: 0x8a108a10, Build Time: 20250523103150a
mt7921e 0000:62:00.0: WM Firmware Version: ____000000, Build Time: 20250523103234

Details:
After updating to 20250917-2.fc42 (both linux-firmware and mt7xxx-firmware), Wi-Fi disappeared completely.
No networks were shown and the Wi-Fi page was missing from Settings.
dmesg displayed "WM Firmware Version: ____000000" and repeated "Failed to get patch semaphore" messages.
The kernel version remained the same (6.16.9-200.fc42) before and after rollback, confirming the regression is firmware-related.

Rolling back both packages to 20250708-1.fc42 allowed the driver to load successfully again after reboot.
Wi-Fi remained soft-blocked in rfkill until manually unblocked, after which scan results and connectivity returned to normal.

This confirms the regression affects MediaTek MT7922 adapters on newer AMD Ryzen AI systems (not just Zenbook UM3406KA).

Observation:
Behavior matches the regression Peter Robinson described earlier regarding the firmware split.
Rolling back to the July (20250708) build restores normal firmware loading on kernel 6.16.9.

Comment 11 hackintoshfive 2025-10-05 15:24:55 UTC
Affects Lenovo ThinkPad L13 Yoga Gen 4, which uses MT7922 too.

I tried installing mediatek-firmware as suggested and this changed the dmesg for me, removing all the mt7921e error messages. However it still doesn't work, with no sign of the adaptor in rfkill or nmcli.

After upgrade, before installing mediatek-firmware:
~$ cat log\ messages20251004 | grep mt
22:58:49 kernel: mt7921e 0000:06:00.0: hardware init failed
22:58:46 kernel: mt7921e 0000:06:00.0: Failed to get patch semaphore
22:58:43 kernel: mt7921e 0000:06:00.0: Failed to get patch semaphore
22:58:40 kernel: mt7921e 0000:06:00.0: Failed to get patch semaphore
22:58:36 kernel: mt7921e 0000:06:00.0: Failed to get patch semaphore
22:58:33 kernel: mt7921e 0000:06:00.0: Failed to get patch semaphore
22:58:30 kernel: mt7921e 0000:06:00.0: Failed to get patch semaphore
22:58:27 kernel: mt7921e 0000:06:00.0: Failed to get patch semaphore
22:58:24 kernel: mt7921e 0000:06:00.0: Failed to get patch semaphore
22:58:21 kernel: mt7921e 0000:06:00.0: Message 0004005c (seq 12) timeout
22:58:20 kernel: mt7921e 0000:06:00.0: driver own failed
22:58:19 kernel: mt7921e 0000:06:00.0: driver own failed
22:58:18 kernel: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=systemd-binfmt comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
22:58:18 systemd: Finished systemd-binfmt.service - Set Up Additional Binary Formats.
22:58:18 kernel: mt7921e 0000:06:00.0: WM Firmware Version: ____000000, Build Time: 20250903123243
22:58:18 kernel: mt7921e 0000:06:00.0: HW/SW Version: 0x8a108a10, Build Time: 20250903123148a
22:58:18 kernel: mt7921e 0000:06:00.0: ASIC revision: 79220010
22:58:17 kernel: mt7921e 0000:06:00.0: enabling device (0000 -> 0002)
22:58:17 mtp-probe: bus: 1, device: 2 was not an MTP device
22:58:17 mtp-probe: checking bus 1, device 2: "/sys/devices/pci0000:00/0000:00:08.1/0000:07:00.3/usb1/1-3"
22:58:16 kernel: Set up automount proc-sys-fs-binfmt_misc.automount - Arbitrary Executable File Formats File System Automount Point.

After upgrade and installing mediatek-firmware:
~$ cat log\ messages20251004\ mediatek-firmware | grep mt
15:08:15 mtp-probe: bus: 3, device: 4 was not an MTP device
15:08:15 mtp-probe: bus: 3, device: 4 was not an MTP device
15:07:35 kernel: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=systemd-binfmt comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
15:07:35 systemd: Finished systemd-binfmt.service - Set Up Additional Binary Formats.
15:07:35 kernel: mt7921e 0000:06:00.0: WM Firmware Version: ____000000, Build Time: 20250903123243
15:07:35 kernel: mt7921e 0000:06:00.0: HW/SW Version: 0x8a108a10, Build Time: 20250903123148a
15:07:35 kernel: mt7921e 0000:06:00.0: ASIC revision: 79220010
15:07:34 mtp-probe: bus: 1, device: 3 was not an MTP device
15:07:34 mtp-probe: bus: 3, device: 2 was not an MTP device
15:07:34 mtp-probe: checking bus 3, device 2: "/sys/devices/pci0000:00/0000:00:08.1/0000:07:00.4/usb3/3-3"
15:07:34 mtp-probe: checking bus 1, device 3: "/sys/devices/pci0000:00/0000:00:08.1/0000:07:00.3/usb1/1-3"
15:07:34 mtp-probe: bus: 1, device: 2 was not an MTP device
15:07:33 kernel: Set up automount proc-sys-fs-binfmt_misc.automount - Arbitrary Executable File Formats File System Automount Point.

The wifi does not work in either of the two logs above.


Before the upgrades, wifi worked, as in the following log:

~$ cat log\ messages20251002 | grep mt
22:57:39 systemd: proc-sys-fs-binfmt_misc.mount: Deactivated successfully.
22:57:39 kernel: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=systemd-binfmt comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
22:57:39 systemd: Stopped systemd-binfmt.service - Set Up Additional Binary Formats.
22:57:24 kernel: mt7921e 0000:06:00.0 wlp6s0: renamed from wlan0
22:57:23 kernel: mt7921e 0000:06:00.0: WM Firmware Version: ____000000, Build Time: 20250523103234
22:57:23 kernel: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=systemd-binfmt comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
22:57:23 systemd: Finished systemd-binfmt.service - Set Up Additional Binary Formats.
22:57:23 kernel: mt7921e 0000:06:00.0: HW/SW Version: 0x8a108a10, Build Time: 20250523103150a
22:57:23 kernel: mt7921e 0000:06:00.0: ASIC revision: 79220010
22:57:23 kernel: mt7921e 0000:06:00.0: enabling device (0000 -> 0002)
22:57:22 mtp-probe: bus: 1, device: 2 was not an MTP device
22:57:22 mtp-probe: bus: 3, device: 2 was not an MTP device
22:57:21 kernel: Set up automount proc-sys-fs-binfmt_misc.automount - Arbitrary Executable File Formats File System Automount Point.

I am using Silverblue, so can provide an exact diff. The issue begins after the following diff:

Upgraded:
  amd-gpu-firmware 20250808-1.fc42 -> 20250917-2.fc42
  amd-ucode-firmware 20250808-1.fc42 -> 20250917-2.fc42
  apr-util 1.6.3-22.fc42 -> 1.6.3-25.fc42
  apr-util-lmdb 1.6.3-22.fc42 -> 1.6.3-25.fc42
  apr-util-openssl 1.6.3-22.fc42 -> 1.6.3-25.fc42
  atheros-firmware 20250808-1.fc42 -> 20250917-2.fc42
  brcmfmac-firmware 20250808-1.fc42 -> 20250917-2.fc42
  buildah 2:1.41.4-1.fc42 -> 2:1.41.5-1.fc42
  cirrus-audio-firmware 20250808-1.fc42 -> 20250917-2.fc42
  gvfs 1.57.2-1.fc42 -> 1.58.0-1.fc42
  gvfs-afc 1.57.2-1.fc42 -> 1.58.0-1.fc42
  gvfs-afp 1.57.2-1.fc42 -> 1.58.0-1.fc42
  gvfs-archive 1.57.2-1.fc42 -> 1.58.0-1.fc42
  gvfs-client 1.57.2-1.fc42 -> 1.58.0-1.fc42
  gvfs-fuse 1.57.2-1.fc42 -> 1.58.0-1.fc42
  gvfs-goa 1.57.2-1.fc42 -> 1.58.0-1.fc42
  gvfs-gphoto2 1.57.2-1.fc42 -> 1.58.0-1.fc42
  gvfs-mtp 1.57.2-1.fc42 -> 1.58.0-1.fc42
  gvfs-smb 1.57.2-1.fc42 -> 1.58.0-1.fc42
  ibus-typing-booster 2.27.75-2.fc42 -> 2.28.0-1.fc42
  intel-audio-firmware 20250808-1.fc42 -> 20250917-2.fc42
  intel-gpu-firmware 20250808-1.fc42 -> 20250917-2.fc42
  intel-vpl-gpu-rt 25.3.2-1.fc42 -> 25.3.4-1.fc42
  intel-vsc-firmware 20250808-1.fc42 -> 20250917-2.fc42
  iwlegacy-firmware 20250808-1.fc42 -> 20250917-2.fc42
  iwlwifi-dvm-firmware 20250808-1.fc42 -> 20250917-2.fc42
  iwlwifi-mvm-firmware 20250808-1.fc42 -> 20250917-2.fc42
  kernel 6.16.8-200.fc42 -> 6.16.9-200.fc42
  kernel-core 6.16.8-200.fc42 -> 6.16.9-200.fc42
  kernel-modules 6.16.8-200.fc42 -> 6.16.9-200.fc42
  kernel-modules-core 6.16.8-200.fc42 -> 6.16.9-200.fc42
  kernel-modules-extra 6.16.8-200.fc42 -> 6.16.9-200.fc42
  kernel-tools 6.16.8-200.fc42 -> 6.16.9-200.fc42
  kernel-tools-libs 6.16.8-200.fc42 -> 6.16.9-200.fc42
  libertas-firmware 20250808-1.fc42 -> 20250917-2.fc42
  libjcat 0.2.3-1.fc42 -> 0.2.4-1.fc42
  linux-firmware 20250808-1.fc42 -> 20250917-2.fc42
  linux-firmware-whence 20250808-1.fc42 -> 20250917-2.fc42
  mt7xxx-firmware 20250808-1.fc42 -> 20250917-2.fc42
  nvidia-gpu-firmware 20250808-1.fc42 -> 20250917-2.fc42
  nxpwireless-firmware 20250808-1.fc42 -> 20250917-2.fc42
  podman 5:5.6.1-1.fc42 -> 5:5.6.2-1.fc42
  python3-boto3 1.40.39-1.fc42 -> 1.40.43-1.fc42
  python3-botocore 1.40.39-1.fc42 -> 1.40.43-1.fc42
  qcom-wwan-firmware 20250808-1.fc42 -> 20250917-2.fc42
  realtek-firmware 20250808-1.fc42 -> 20250917-2.fc42
  tiwilink-firmware 20250808-1.fc42 -> 20250917-2.fc42
Added:
  iwlwifi-mld-firmware-20250917-2.fc42.noarch
  mediatek-firmware-20250917-2.fc42.noarch

Running  sudo rpm-ostree override replace https://koji.fedoraproject.org/koji/buildinfo?buildID=2788336 --remove iwlwifi-mld-firmware  resolved my issue by downgrading linux-firmware to 20250808.

Working setup:

● fedora:fedora/42/x86_64/silverblue
                  Version: 42.20251005.0 (2025-10-05T00:26:32Z)
               BaseCommit: 6dd56f0b787ec5c3527a6f2a9068bbc210877da5641f91b11df15d50b1b78b53
             GPGSignature: Valid signature by B0F4950458F69E1150C6C5EDC8AC4916105EF944
      RemovedBasePackages: iwlwifi-mld-firmware 20250917-2.fc42 noopenh264 2.5.0-2.fc42
           LocalOverrides: amd-gpu-firmware amd-ucode-firmware linux-firmware qcom-wwan-firmware intel-audio-firmware nvidia-gpu-firmware libertas-firmware atheros-firmware iwlegacy-firmware tiwilink-firmware brcmfmac-firmware nxpwireless-firmware mt7xxx-firmware iwlwifi-mvm-firmware intel-vsc-firmware cirrus-audio-firmware iwlwifi-dvm-firmware linux-firmware-whence intel-gpu-firmware realtek-firmware 20250917-2.fc42 -> 20250808-1.fc42
          LayeredPackages: adb gstreamer1-plugin-openh264 mozilla-openh264 openh264 openssl pam-u2f pamu2fcfg tailscale xsensors
             InitramfsEtc: /etc/vconsole.conf


As observed by others, rolling back linux-firmware sorts this out. However it seems that the presence of mediatek-firmware (or some unrelated condition that corresponds by chance) affects the exact error that is shown. A similar difference was also reported in comment #5 but in the reverse direction. Either one of us has made a mistake, or something weird is going on here. I also did not observe the  enabling device (0000 -> 0002)  message in my logs after installing mediatek-firmware, whereas #5 did.

Comment 12 Peter Robinson 2025-10-05 17:11:18 UTC
> I tried installing mediatek-firmware as suggested and this changed the dmesg
> for me, removing all the mt7921e error messages. However it still doesn't
> work, with no sign of the adaptor in rfkill or nmcli.

Does it report which files are actually loaded, or complain a particular file name is missing?

Comment 13 hackintoshfive 2025-10-05 17:23:11 UTC
(In reply to Peter Robinson from comment #12)

> Does it report which files are actually loaded, or complain a particular
> file name is missing?

Not in a way that I can see in the log with grep -C 10 mt|firmware. Do you know what the message would look like, or whether I need to increase the loglevel via the cmdline?

Comment 14 Peter Robinson 2025-10-05 18:42:42 UTC
> Not in a way that I can see in the log with grep -C 10 mt|firmware. Do you
> know what the message would look like, or whether I need to increase the
> loglevel via the cmdline?

No idea, I've never used the mediatek wifi modules.

Comment 15 William Oprandi 2025-10-07 10:47:27 UTC
Same here, tried to install mediatek-firmware but no changes.

# lspci | grep Network
02:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter

# dmesg

[   16.013453] mt7921e 0000:02:00.0: driver own failed
[   16.013528] mt7921e 0000:02:00.0: probe with driver mt7921e failed with error -5

I'm unable to rollback with dnf

Comment 16 Peter Robinson 2025-10-07 10:47:53 UTC
*** Bug 2400176 has been marked as a duplicate of this bug. ***

Comment 17 Peter Robinson 2025-10-07 10:52:07 UTC
Can people please add the revision(s) of the Mediatek hardware they have, I am trying to cross verify which HW is actually affected. The single line output from lspci/lsusb is likely enough.

Please be concise, I really don't care about all the pci/usb devices on the platform, won't dig through pages of attachments.

Comment 18 Chris Irwin 2025-10-07 14:04:59 UTC
$ lspci | grep Net
01:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter

Comment 19 hackintoshfive 2025-10-07 14:32:27 UTC
$ lspci | grep Net
06:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter

Comment 20 tavisirbu2310 2025-10-07 19:15:25 UTC
lspci | grep Net
01:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter

Comment 21 moosager90 2025-10-07 22:22:58 UTC
$ lspci | grep Net
62:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter

Comment 22 Peter Robinson 2025-10-08 08:35:41 UTC
*** Bug 2402432 has been marked as a duplicate of this bug. ***

Comment 23 Yegor 2025-10-08 12:43:34 UTC
(In reply to Peter Robinson from comment #17)
> Can people please add the revision(s) of the Mediatek hardware they have

Same here.
$ lspci | grep Net
62:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter

Comment 24 Fedora Blocker Bugs Application 2025-10-08 13:27:04 UTC
Proposed as a Blocker for 43-final by Fedora user pbrobinson using the blocker tracking app because:

 This is a common WiFi module on a number of platforms.

I've reached out to the upstream maintainer for engagement to see whether a fix is coming or whether I need to investigate a revert is the best route.

Comment 25 Chris Shiflet 2025-10-08 15:37:36 UTC
I experienced the same issue and was able to restore functionality of my Mediatek MT7922 by downgrading the `mt7xxx-firmware` package from `20250917-2.fc42` to `20250311-1.fc42` that appears to be generated from the `linux-firmware` source RPM.
```
lspci | grep Net
62:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter
```

```
sudo dnf downgrade mt7xxx-firmware
```

```
dnf list --showduplicates mt7xxx-firmware
Updating and loading repositories:
Repositories loaded.
Installed packages
mt7xxx-firmware.noarch 20250311-1.fc42 fedora

Available packages
mt7xxx-firmware.noarch 20250311-1.fc42 fedora
mt7xxx-firmware.noarch 20250917-2.fc42 updates
```

```
dnf info mt7xxx-firmware
Updating and loading repositories:
Repositories loaded.
Installed packages
Name            : mt7xxx-firmware
Epoch           : 0
Version         : 20250311
Release         : 1.fc42
Architecture    : noarch
Installed size  : 20.2 MiB
Source          : linux-firmware-20250311-1.fc42.src.rpm
From repository : fedora
Summary         : Firmware for Mediatek 7600/7900 series WiFi/Bluetooth adapters
URL             : http://www.kernel.org/
License         : LicenseRef-Callaway-Redistributable-no-modification-permitted
Description     : Firmware for Mediatek 7600/7900 series WiFi/Bluetooth adapters
Vendor          : Fedora Project

Available packages
Name           : mt7xxx-firmware
Epoch          : 0
Version        : 20250917
Release        : 2.fc42
Architecture   : noarch
Download size  : 17.7 MiB
Installed size : 18.4 MiB
Source         : linux-firmware-20250917-2.fc42.src.rpm
Repository     : updates
Summary        : Firmware for Mediatek 7600/7900 series WiFi/Bluetooth adapters
URL            : http://www.kernel.org/
License        : LicenseRef-Callaway-Redistributable-no-modification-permitted
Description    : Firmware for Mediatek 7600/7900 series WiFi/Bluetooth adapters
Vendor         : Fedora Project
```

Comment 26 Matthias Saou 2025-10-09 09:20:16 UTC
On a Lenovo ThinkPad Z13 Gen 2 (21JVCTO1WW), I had the exact same issue, but downgrading the mt7xxx-firmware package wasn't enough, as I was still getting a failure:

[   69.019040] mt7921e 0000:01:00.0: enabling device (0000 -> 0002)
[   70.091451] mt7921e 0000:01:00.0: driver own failed
[   70.091566] mt7921e 0000:01:00.0: probe with driver mt7921e failed with error -5

But I was also seeing some other kernel issues related to usb 1-3 ... not sure what those were, but what worked for me was also downgrading the kernel:

kernel-6.16.10-200.fc42.x86_64
kernel-6.16.9-200.fc42.x86_64

Both the above (latest and previous) were exhibiting those usb 1-3 errors and WiFi still not working. The following kernel made everything work again, all usb issues gone and WiFi working!

kernel-6.16.8-200.fc42.x86_64

Comment 27 stexads 2025-10-09 09:38:13 UTC
> Please be concise, I really don't care about all the pci/usb devices on the
> platform, won't dig through pages of attachments.


01:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter

Comment 28 Peter Robinson 2025-10-09 10:37:45 UTC
Right so it's the "Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter" in all cases as far as I can see.

Comment 29 Silvio Bierman 2025-10-09 14:30:47 UTC
Yes, that is the one for me as well (on an ASUS Zenbook 14 UM3406KA_UM3406KA).

Comment 30 Silvio Bierman 2025-10-09 14:34:57 UTC
BTW: I am currently running Fedora 43 beta. There I could not simply downgrade but had to download an older mt7xxx-firmware package from V42 and downgrade to that one. Simply excluded the 0917 version of the package and staying on 0311 (from f42) works like a charm.

Comment 31 Peter Robinson 2025-10-09 15:12:28 UTC
> Simply excluded the 0917 version of the package and staying on 0311 (from f42) works like a charm.

Which would be fine if there was only this device in the package, linux-firmware is a collection of 1000s of firmwares.

Comment 32 Silvio Bierman 2025-10-09 16:09:14 UTC
I do understand and appreciate that. But have no Wifi is quite a turn-off from a brand new laptop and I am happy to get it working again. I do not mean that other should do the same as I did unless they are in exactly the same spot and want a quick fix (hack).

Comment 33 Nikolai Vincent Vaags 2025-10-10 05:37:28 UTC
Experienced this same issue on my desktop with a
11:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter

spending a long time booting/shutting down with errors like
kernel: usb 3-7: device descriptor read/64, error -110
kernel: usb 3-7: device not accepting address 5, error -71

kernel: mt7921e 0000:11:00.0: Message 00000010 (seq 1) timeout
kernel: mt7921e 0000:11:00.0: Failed to start patch


It also led my machine to hard-lock when attempting to sleep.

However, reading the following thread, people mentioned having to unplug power (after downgrading) in order to resolve the issue:
https://bbs.archlinux.org/viewtopic.php?id=308360&p=2


..so I decided to test completely unplugging my desktop for a few minutes, _without_ downgrading. And it appears to have resolved the issue -- bluetooth and wifi now show up, machine boots and shuts down normally, and sleep works again.

Comment 34 Alex Vojtko-Proctor 2025-10-10 13:40:00 UTC
I experienced the same thing with the MT7922, including the freeze on shutdown.

 - I wasn't able to reproduce Nikolai's report of fixing this by completely unplugging my battery without a downgrade.
 - Rolling back linux-firmware (and removing iwlwifi-mld-firmware) worked for me, as it did for hackintoshfive.
 - I didn't get the "Failed to get patch semaphore" kernel message.

    # lspci
    01:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter

    # dmesg | grep mt7921e
    mt7921e 0000:01:00.0: enabling device (0000 -> 0002)
    mt7921e 0000:01:00.0: ASIC revision: 79220010
    mt7921e 0000:01:00.0: HW/SW Version: 0x8a108a10, Build Time: 20250903123148a
    mt7921e 0000:01:00.0: WM Firmware Version: ____000000, Build Time: 20250903123243

    $ cat dmesg_working_short.txt | grep mt7921e
    mt7921e 0000:01:00.0: HW/SW Version: 0x8a108a10, Build Time: 20250523103150a
    mt7921e 0000:01:00.0: WM Firmware Version: ____000000, Build Time: 20250523103234
    mt7921e 0000:01:00.0 wlp1s0: renamed from wlan0

    # rpm-ostree db diff 559a02c d5b0489
    Upgraded:
      amd-gpu-firmware 20250808-1.fc42 -> 20250917-2.fc42
      amd-ucode-firmware 20250808-1.fc42 -> 20250917-2.fc42
      atheros-firmware 20250808-1.fc42 -> 20250917-2.fc42
      brcmfmac-firmware 20250808-1.fc42 -> 20250917-2.fc42
      cirrus-audio-firmware 20250808-1.fc42 -> 20250917-2.fc42
      intel-audio-firmware 20250808-1.fc42 -> 20250917-2.fc42
      intel-gpu-firmware 20250808-1.fc42 -> 20250917-2.fc42
      intel-vsc-firmware 20250808-1.fc42 -> 20250917-2.fc42
      iwlegacy-firmware 20250808-1.fc42 -> 20250917-2.fc42
      iwlwifi-dvm-firmware 20250808-1.fc42 -> 20250917-2.fc42
      iwlwifi-mvm-firmware 20250808-1.fc42 -> 20250917-2.fc42
      kernel 6.16.8-200.fc42 -> 6.16.9-200.fc42
      kernel-core 6.16.8-200.fc42 -> 6.16.9-200.fc42
      kernel-modules 6.16.8-200.fc42 -> 6.16.9-200.fc42
      kernel-modules-core 6.16.8-200.fc42 -> 6.16.9-200.fc42
      kernel-modules-extra 6.16.8-200.fc42 -> 6.16.9-200.fc42
      libertas-firmware 20250808-1.fc42 -> 20250917-2.fc42
      linux-firmware 20250808-1.fc42 -> 20250917-2.fc42
      linux-firmware-whence 20250808-1.fc42 -> 20250917-2.fc42
      mt7xxx-firmware 20250808-1.fc42 -> 20250917-2.fc42
      nvidia-gpu-firmware 20250808-1.fc42 -> 20250917-2.fc42
      nxpwireless-firmware 20250808-1.fc42 -> 20250917-2.fc42
      qcom-wwan-firmware 20250808-1.fc42 -> 20250917-2.fc42
      realtek-firmware 20250808-1.fc42 -> 20250917-2.fc42
      tiwilink-firmware 20250808-1.fc42 -> 20250917-2.fc42
    Added:
      iwlwifi-mld-firmware-20250917-2.fc42.noarch

For the benefit of other Fedora Atomic users experiencing the same thing:

    # rpm-ostree override replace https://koji.fedoraproject.org/koji/buildinfo?buildID=2788336 --remove iwlwifi-mld-firmware

Thanks for your help.

Comment 35 mcternal83747 2025-10-11 09:56:16 UTC
Framework 13 AMD 7840U tested and working fine from latest live ISO? MT7922

Comment 36 Sayan 2025-10-11 14:21:44 UTC
I can confirm that I'm having the same issue. Downgrading mt7xxx-firmware from 20250917-2.fc42 to 20250311-1.fc42 seems to be a temporary workaround. Also a restart doesn't immediately fix the issue. Instead, a complete shutdown followed by a fresh boot, following the downgrade fixes it.

Comment 37 Chris Irwin 2025-10-12 05:49:35 UTC
Note that linux-firmware-20250808-1 also works fine.

Comment 38 Kashyap Chamarthy 2025-10-13 09:34:06 UTC
(In reply to Chris Irwin from comment #37)
> Note that linux-firmware-20250808-1 also works fine.

Yeah, that's much better; I wonder if linux-firmware-20250917-1 from here also works.

Meanwhile, I just saw that Peter reverted[2] the patch he added in  20250917-2.  Maybe that'll do the trick.

He's got a new build here[2], if those affected can test it.


[1] https://src.fedoraproject.org/rpms/linux-firmware/c/25ca2e4

[2] https://koji.fedoraproject.org/koji/buildinfo?buildID=2842187

Comment 39 Peter Robinson 2025-10-13 09:52:50 UTC
> Meanwhile, I just saw that Peter reverted[2] the patch he added in 
> 20250917-2.  Maybe that'll do the trick.

I didn't revert any patch, it was dropped because it landed upstream in the new release. Also it's completely unrelated hardware/driver. So no, it won't do the trick, also that's why this bug was not added to the update, because that update doesn't do anything that I am aware of to fix the bug. I am still awaiting a response from the vendor.

Comment 41 Kashyap Chamarthy 2025-10-13 12:21:11 UTC
(In reply to Peter Robinson from comment #39)
> > Meanwhile, I just saw that Peter reverted[2] the patch he added in 
> > 20250917-2.  Maybe that'll do the trick.
> 
> I didn't revert any patch, it was dropped because it landed upstream in the
> new release. Also it's completely unrelated hardware/driver. 

Oops, perils of skimming that patch in a hurry.  I saw "mediatek" in the
'dist-git' commit and guessed the driver was related.

> So no, it won't
> do the trick, also that's why this bug was not added to the update, because
> that update doesn't do anything that I am aware of to fix the bug. I am
> still awaiting a response from the vendor.

Noted; thanks for the correction.

You might know all of this:

FWIW, I did some more digging.  Based on the upstream kernel merge 
commit from comment#40 above, I looked at the MR[1] for MT7922.  There's 
a breakage report[2] on Arch Linux (and Alpine too): "causes network to 
fail startup and dramatically slows down basically every operation in
the kernel".  It's also being discussed on the 'linux-firmware' upstream
list[3].  Arch did revert[4] the 'mediatek' update for MT7922.

Besides switching to a working firmware, there seems to be a 
workaround by disabling a feature called "CLC"; someone confirmed[5] on
the upstream list that this also works:

    mt7921_common disable_clc=1

[1] https://gitlab.com/kernel-firmware/linux-firmware/-/merge_requests/684
[2] https://gitlab.archlinux.org/archlinux/packaging/packages/linux-firmware/-/issues/29
[3] https://lore.kernel.org/linux-firmware/87bjn6fgr6.fsf@gentoo.org/
[4] https://gitlab.archlinux.org/archlinux/packaging/packages/linux-firmware/-/commit/ce6367176
[5] https://lore.kernel.org/linux-firmware/d0a06fe1-abd0-49b2-8867-6c01d6551e54@gmail.com/

Comment 42 Lukas Ruzicka 2025-10-13 18:38:40 UTC
AGREED AcceptedFinalBlocker

Discussed at the 2025-10-13 (blocker / freeze exception) review meeting:

This seems a wifi chip that is widespread enough to justify a blocker. All networking criteria are violated for affected hardware: https://fedoraproject.org/wiki/Basic_Release_Criteria#Network_requirements

https://meetbot-raw.fedoraproject.org//blocker-review_matrix_fedoraproject-org/2025-10-13/f43-blocker-review.2025-10-13-16.00.txt

Comment 43 Kamil Páral 2025-10-15 08:42:25 UTC
I tested this with Mediatek 7925 and that doesn't seem to be affected.

c2:00.0 Network controller [0280]: MEDIATEK Corp. Device [14c3:7925]

Comment 44 Lukas Ruzicka 2025-10-15 09:49:39 UTC
I tested with two Mediateks, the 7925 and 7922 and none of them is affected.

c0:00.0 Network controller: MEDIATEK Corp. MT7925 (RZ717) Wi-Fi 7 160Mhz
01:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter

Comment 45 Lukas Ruzicka 2025-10-15 13:23:15 UTC
(In reply to Lukas Ruzicka from comment #44)
> I tested with two Mediateks, the 7925 and 7922 and none of them is affected.
> 
c0:00.0 Network controller: MEDIATEK Corp. MT7925 (RZ717) Wi-Fi 7 160Mhz [14c3:0717]
01:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter [14c3:0616]

Comment 46 tavisirbu2310 2025-10-16 06:34:55 UTC
(In reply to Lukas Ruzicka from comment #45)
> (In reply to Lukas Ruzicka from comment #44)
> > I tested with two Mediateks, the 7925 and 7922 and none of them is affected.
> > 
> c0:00.0 Network controller: MEDIATEK Corp. MT7925 (RZ717) Wi-Fi 7 160Mhz
> [14c3:0717]
> 01:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express
> Wireless Network Adapter [14c3:0616]


how do you not have the issue on MT7922? Everyone's been having issues on that model. Did you do anything different? What version of linux-firmware do you have installed?

Comment 47 tavisirbu2310 2025-10-16 06:35:07 UTC
(In reply to Lukas Ruzicka from comment #45)
> (In reply to Lukas Ruzicka from comment #44)
> > I tested with two Mediateks, the 7925 and 7922 and none of them is affected.
> > 
> c0:00.0 Network controller: MEDIATEK Corp. MT7925 (RZ717) Wi-Fi 7 160Mhz
> [14c3:0717]
> 01:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express
> Wireless Network Adapter [14c3:0616]


how do you not have the issue on MT7922? Everyone's been having issues on that model. Did you do anything different? What version of linux-firmware do you have installed?

Comment 48 Lukas Ruzicka 2025-10-16 10:57:45 UTC
I do not have problems. The MT7922 module lives in my Framework 13 (AMD based) laptop and now I am connected through wi-fi, the connection is stable and reasonably fast. I am not doing anything different, I just switched to Fedora 43 when it got branched and I keep it updated all the time. Nothing else, really.

My firmware version is: linux-firmware-20250917-2

However I have found a minor discrepancy in naming the device:

sudo dmesg | grep mt79

[   85.705864] mt7921e 0000:01:00.0: enabling device (0000 -> 0002)
[   85.718452] mt7921e 0000:01:00.0: ASIC revision: 79220010
[   85.795467] mt7921e 0000:01:00.0: HW/SW Version: 0x8a108a10, Build Time: 20250903123148a
[   86.186462] mt7921e 0000:01:00.0: WM Firmware Version: ____000000, Build Time: 20250903123243
[   87.362202] mt7921e 0000:01:00.0 wlp1s0: renamed from wlan0



lspci -nn | grep MEDIATEK

01:00.0 Network controller [0280]: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter [14c3:0616]

Comment 49 zalnars 2025-10-17 03:59:25 UTC
I have a ideapad 14akp10 which has a Ryzen AI 7 350 with MT7925(e) network controller. WiFi works fine, but Bluetooth doesn't work and I need to run this command, after every wake from sleep:

```
sudo modprobe -r btusb && sleep 1 && sudo modprobe btusb
```


I also have other off-topic issues, particularly with it's GPU/APU integrated audio controller, some kind of senary/conexant/something Intel HD audio-like, which has:

- digital microphone input (maxes out the vu meter, but no sound)
- stereo microphone input (works, but really really quiet)

Comment 50 zalnars 2025-10-17 04:02:49 UTC
I can only select digital in most apps (including gnome), so to use the other one I need to patch it manually with helvum.

Comment 51 Peter Robinson 2025-10-17 09:16:58 UTC
This is not a bug about bluetooth, please don't hijack the bug for completely unrelated issues.

Comment 52 Hailazy 2025-10-17 14:52:55 UTC
I can confirm I am experiencing the same issue on my system.

**My System:**
- Motherboard: ASUS TUF GAMING B650M-E WIFI
- Wireless Chip: MediaTek MT7922 (RZ616)
- Kernel: 6.16.12-200.fc42.x86_64
- BIOS: 3278

**Logs:**
# dmesg | grep -i bluetooth
[    7.912337] Bluetooth: hci0: Opcode 0x0c03 failed: -110

# lsusb
Bus 001 Device 002: ID 13d3:3596 IMC Networks Wireless_Device

---

**Attempted Solutions (without success):**
- Performed multiple cold boots (full shutdown, unplug power).
- The system is fully updated via `sudo dnf upgrade`.
- Attempted to downgrade the `mt7xxx-firmware` package to a known-good version as suggested in this bug report. However, all older packages seem to have been removed from Fedora's servers, leading to "Not Found" errors when trying to download them. This makes the downgrade workaround impossible to apply.

Comment 53 mbeacher 2025-10-17 20:24:40 UTC
I am seeing this issue as well

Upgraded to mt7xxx-firmware-20251011-2.fc42, Wifi showed the same behavior as mt7xxx-firmware-20250917-2.fc42

Downgrading to mt7xxx-firmware-20250311-1.fc42 resolved the issue.

lspci | grep Net
62:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter

Operating System: Fedora Linux 42
KDE Plasma Version: 6.4.5
KDE Frameworks Version: 6.19.0
Qt Version: 6.9.2
Kernel Version: 6.16.12-200.fc42.x86_64 (64-bit)
Graphics Platform: Wayland
Processors: 20 × AMD Ryzen AI 9 365 w/ Radeon 880M
Memory: 24 GiB of RAM (22.6 GiB usable)
Graphics Processor: AMD Radeon 890M Graphics
Manufacturer: ASUSTeK COMPUTER INC.
Product Name: ASUS Vivobook S 14 M5406WA_M5406WA
System Version: 1.0

Comment 54 hayo 2025-10-20 09:14:27 UTC
I have the same issue as above, even with the latest update 20251011 firmware:

Only twice Downgrading to mt7xxx-firmware- to 20250917-2 worked for me to use WIFI again. 

My spec:

01:00.0 Network controller: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter

Operating System: Fedora Linux 42
Qt Version: 6.9.2
Kernel Version: 6.16.12-200.fc42.x86_64 (x86_64)
Graphics Platform: Waylandwayland
Processors: 16 ×  AMD Ryzen 7 PRO 7840U w/ Radeon 780M Graphics
Memory: 61Gi of RAM (55Gi usable)
Graphics Processor: Advanced Micro Devices, Inc. [AMD/ATI] Phoenix1 (rev dd)
Manufacturer: LENOVO
System Version: ThinkPad Z13 Gen 2

Comment 55 Silvio Bierman 2025-10-20 12:40:36 UTC
Are you sure you are running this version (the September 17th one)? For most of us (including me) this version was the first one that caused issues.

Comment 56 hayo 2025-10-20 12:46:21 UTC
(In reply to Silvio Bierman from comment #55)
> Are you sure you are running this version (the September 17th one)? For most
> of us (including me) this version was the first one that caused issues.

Yes, this is the output of sudo dnf update iwlwifi-mvm-firmware

Upgrading:
 amd-gpu-firmware                      noarch    20251011-1.fc42                        updates                 25.7 MiB
   replacing amd-gpu-firmware          noarch    20250917-2.fc42                        updates-archive         25.7 MiB
 amd-ucode-firmware                    noarch    20251011-1.fc42                        updates                419.9 KiB
   replacing amd-ucode-firmware        noarch    20250917-2.fc42                        updates-archive        419.9 KiB
 atheros-firmware                      noarch    20251011-1.fc42                        updates                 40.9 MiB
   replacing atheros-firmware          noarch    20250917-2.fc42                        updates-archive         40.9 MiB
 brcmfmac-firmware                     noarch    20251011-1.fc42                        updates                  9.5 MiB
   replacing brcmfmac-firmware         noarch    20250917-2.fc42                        updates-archive          9.5 MiB
 cirrus-audio-firmware                 noarch    20251011-1.fc42                        updates                  1.9 MiB
   replacing cirrus-audio-firmware     noarch    20250917-2.fc42                        updates-archive          1.9 MiB
 intel-audio-firmware                  noarch    20251011-1.fc42                        updates                  3.3 MiB
   replacing intel-audio-firmware      noarch    20250917-2.fc42                        updates-archive          3.3 MiB
 intel-gpu-firmware                    noarch    20251011-1.fc42                        updates                  8.8 MiB
   replacing intel-gpu-firmware        noarch    20250917-2.fc42                        updates-archive          8.8 MiB
 iwlegacy-firmware                     noarch    20251011-1.fc42                        updates                123.2 KiB
   replacing iwlegacy-firmware         noarch    20250917-2.fc42                        updates-archive        123.2 KiB
 iwlwifi-dvm-firmware                  noarch    20251011-1.fc42                        updates                  1.8 MiB
   replacing iwlwifi-dvm-firmware      noarch    20250917-2.fc42                        updates-archive          1.8 MiB
 iwlwifi-mld-firmware                  noarch    20251011-1.fc42                        updates                  5.9 MiB
   replacing iwlwifi-mld-firmware      noarch    20250917-2.fc42                        updates-archive          5.9 MiB
 iwlwifi-mvm-firmware                  noarch    20251011-1.fc42                        updates                 62.7 MiB
   replacing iwlwifi-mvm-firmware      noarch    20250917-2.fc42                        updates-archive         62.7 MiB
 libertas-firmware                     noarch    20251011-1.fc42                        updates                  1.3 MiB
   replacing libertas-firmware         noarch    20250917-2.fc42                        updates-archive          1.3 MiB
 linux-firmware                        noarch    20251011-1.fc42                        updates                 41.4 MiB
   replacing linux-firmware            noarch    20250917-2.fc42                        updates-archive         40.8 MiB
 linux-firmware-whence                 noarch    20251011-1.fc42                        updates                346.5 KiB
   replacing linux-firmware-whence     noarch    20250917-2.fc42                        updates-archive        345.8 KiB
 nvidia-gpu-firmware                   noarch    20251011-1.fc42                        updates                101.0 MiB
   replacing nvidia-gpu-firmware       noarch    20250917-2.fc42                        updates-archive        101.0 MiB
 nxpwireless-firmware                  noarch    20251011-1.fc42                        updates                905.2 KiB
   replacing nxpwireless-firmware      noarch    20250917-2.fc42                        updates-archive        905.2 KiB
 qcom-wwan-firmware                    noarch    20251011-1.fc42                        updates                300.4 KiB
   replacing qcom-wwan-firmware        noarch    20250917-2.fc42                        updates-archive        300.4 KiB
 realtek-firmware                      noarch    20251011-1.fc42                        updates                  5.2 MiB
   replacing realtek-firmware          noarch    20250917-2.fc42                        updates-archive          5.2 MiB
 tiwilink-firmware                     noarch    20251011-1.fc42                        updates                  4.6 MiB
   replacing tiwilink-firmware         noarch    20250917-2.fc42                        updates-archive          4.6 MiB

Comment 57 nihui 2025-10-20 13:44:24 UTC
I confirm that the WiFi and Bluetooth can function properly after downgrading the mt7xxx-firmware to version 20250311 and performing a cold boot by unplugging the power.

```
09:00.0 Network controller: MEDIATEK Corp. Device 7925
        Subsystem: Foxconn International, Inc. Device e112
        Flags: bus master, fast devsel, latency 0, IRQ 145, IOMMU group 21
        Memory at dca00000 (64-bit, non-prefetchable) [size=2M]
        Memory at dcc00000 (64-bit, non-prefetchable) [size=32K]
        Capabilities: [80] Express Endpoint, IntMsgNum 0
        Capabilities: [e0] MSI: Enable+ Count=1/32 Maskable+ 64bit+
        Capabilities: [f8] Power Management version 3
        Capabilities: [100] Vendor Specific Information: ID=1556 Rev=1 Len=008 <?>
        Capabilities: [108] Latency Tolerance Reporting
        Capabilities: [110] L1 PM Substates
        Capabilities: [200] Advanced Error Reporting
        Kernel driver in use: mt7925e
        Kernel modules: mt7925e
```

System:
  Host: nihui-pc Kernel: 6.16.12-200.fc42.x86_64 arch: x86_64 bits: 64
  Console: pty pts/1 Distro: Fedora Linux 42 (KDE Plasma Desktop Edition)
Machine:
  Type: Desktop Mobo: Gigabyte model: B650E AORUS PRO X USB4 serial: N/A
    UEFI: American Megatrends LLC. v: F9d date: 09/18/2025
CPU:
  Info: 16-core model: AMD Ryzen 9 9950X bits: 64 type: MT MCP

Comment 58 Adam Williamson 2025-10-20 18:31:28 UTC
Per Peter, "I'm awaiting a new release to be tagged upstream, I had it reverted" - so marking as POST. We can't backport as it would require a binary patch which is a giant pain to deal with in dist-git, so we need the new release.

Comment 60 Fedora Update System 2025-10-21 23:22:43 UTC
FEDORA-2025-1ca4c970c1 (linux-firmware-20251021-1.fc41) has been submitted as an update to Fedora 41.
https://bodhi.fedoraproject.org/updates/FEDORA-2025-1ca4c970c1

Comment 61 Fedora Update System 2025-10-21 23:22:45 UTC
FEDORA-2025-8fdb6f9fe8 (linux-firmware-20251021-1.fc42) has been submitted as an update to Fedora 42.
https://bodhi.fedoraproject.org/updates/FEDORA-2025-8fdb6f9fe8

Comment 62 Fedora Update System 2025-10-21 23:22:49 UTC
FEDORA-2025-0b6e376195 (linux-firmware-20251021-1.fc43) has been submitted as an update to Fedora 43.
https://bodhi.fedoraproject.org/updates/FEDORA-2025-0b6e376195

Comment 63 Fedora Update System 2025-10-22 01:08:35 UTC
FEDORA-2025-8fdb6f9fe8 has been pushed to the Fedora 42 testing repository.
Soon you'll be able to install the update with the following command:
`sudo dnf upgrade --enablerepo=updates-testing --refresh --advisory=FEDORA-2025-8fdb6f9fe8`
You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2025-8fdb6f9fe8

See also https://fedoraproject.org/wiki/QA:Updates_Testing for more information on how to test updates.

Comment 64 Fedora Update System 2025-10-22 02:27:58 UTC
FEDORA-2025-0b6e376195 has been pushed to the Fedora 43 testing repository.
Soon you'll be able to install the update with the following command:
`sudo dnf upgrade --enablerepo=updates-testing --refresh --advisory=FEDORA-2025-0b6e376195`
You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2025-0b6e376195

See also https://fedoraproject.org/wiki/QA:Updates_Testing for more information on how to test updates.

Comment 65 Fedora Update System 2025-10-22 02:52:45 UTC
FEDORA-2025-1ca4c970c1 has been pushed to the Fedora 41 testing repository.
Soon you'll be able to install the update with the following command:
`sudo dnf upgrade --enablerepo=updates-testing --refresh --advisory=FEDORA-2025-1ca4c970c1`
You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2025-1ca4c970c1

See also https://fedoraproject.org/wiki/QA:Updates_Testing for more information on how to test updates.

Comment 66 Chris Irwin 2025-10-22 05:51:37 UTC
linux-firmware-20251021-1.fc42 resolves the issue for me. It looks like this is the upstream revert of the mt7922 firmware.

Comment 67 Peter Robinson 2025-10-22 09:26:03 UTC
(In reply to Chris Irwin from comment #66)
> linux-firmware-20251021-1.fc42 resolves the issue for me. It looks like this
> is the upstream revert of the mt7922 firmware.

That is correct. With radio silence from Mediatek it was the most straight forward solution for the wider ecosystem. I will keep an eye out upstream (I do anyway) for any PR for this device in the short term to ensure they don't just re-push the broken firmware.

Comment 68 Adam Williamson 2025-10-22 15:54:37 UTC
Marking VERIFIED with Chris' feedback.

Comment 69 Dimitris 2025-10-22 18:39:38 UTC
Just a quick (late) note:  Also on a Framework 13 AMD 7840U with

01:00.0 Network controller [0280]: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter [14c3:0616]

and like Lukas I didn't ever see any issues with WiFi on either 20250917 or 20251011 (F42).  Access point is a single SSID (same password, WPA3 SAE) on both 2.4GHz (WiFi4) and 5GHz (WiFi6), and the client behaves as expected (sometimes starts on the 2.4GHz band but eventually will move to the 5GHz which has more bandwidth).

Now on 20251021, no regressions on that either.

Comment 70 Peter Robinson 2025-10-23 13:17:12 UTC
Note there is now new 20251020143135a/20251020143009 version upstream, if someone with one of these devices would be happy to assist me in testing of this FW out of band of this bug to verify we don't have another regression upstream I would appreciate that. Please reach out out of band (email/irc/matrix).

Comment 71 Fedora Update System 2025-10-23 23:54:17 UTC
FEDORA-2025-0b6e376195 (linux-firmware-20251021-1.fc43) has been pushed to the Fedora 43 stable repository.
If problem still persists, please make note of it in this bug report.

Comment 72 Richard Kimberly Heck 2025-10-24 01:54:37 UTC
So, just for clarity: The update to linux-firmware solves the problem for people, even with the mediatek-firmware-0251011? I ask because it is a real pain for me to downgrade from this laptop if the wireless isn't working, since it has no ethernet port.

Comment 73 tavisirbu2310 2025-10-24 06:33:58 UTC
(In reply to Richard Kimberly Heck from comment #72)
> So, just for clarity: The update to linux-firmware solves the problem for
> people, even with the mediatek-firmware-0251011? I ask because it is a real
> pain for me to downgrade from this laptop if the wireless isn't working,
> since it has no ethernet port.

You can easily connect your phone via USB-C to your laptop and have internet via your phone.

Comment 74 moosager90 2025-10-24 08:07:30 UTC
I can confirm linux-firmware-20251021-1.fc42 fixes the issue for me.

Comment 75 Peter Robinson 2025-10-24 10:03:39 UTC
(In reply to Richard Kimberly Heck from comment #72)
> So, just for clarity: The update to linux-firmware solves the problem for
> people, even with the mediatek-firmware-0251011? I ask because it is a real
> pain for me to downgrade from this laptop if the wireless isn't working,
> since it has no ethernet port.

It was reverted to the working one, so yes it should be fine. The easiest way to make it straight forward is to download the rollback ones first before you upgrade to have them on the local filesystem so you can do 'sudo dnf upgrade ~/backup-fw/*firmware*rpm'

Comment 76 Silvio Bierman 2025-10-24 10:54:10 UTC
(In reply to Fedora Update System from comment #71)
> FEDORA-2025-0b6e376195 (linux-firmware-20251021-1.fc43) has been pushed to
> the Fedora 43 stable repository.
> If problem still persists, please make note of it in this bug report.

I still see the 20250917 version in stable on: https://packages.fedoraproject.org/pkgs/linux-firmware/mt7xxx-firmware/

Comment 77 Richard Kimberly Heck 2025-10-26 23:13:41 UTC
Confirming that it works for me, too. Someone should push this to the F42 updates repo?

Comment 78 Fedora Update System 2025-10-27 00:59:00 UTC
FEDORA-2025-8fdb6f9fe8 (linux-firmware-20251021-1.fc42) has been pushed to the Fedora 42 stable repository.
If problem still persists, please make note of it in this bug report.

Comment 79 Fedora Update System 2025-10-27 01:18:57 UTC
FEDORA-2025-1ca4c970c1 (linux-firmware-20251021-1.fc41) has been pushed to the Fedora 41 stable repository.
If problem still persists, please make note of it in this bug report.

Comment 80 nihui 2025-10-27 13:11:08 UTC
new firmware works for me

root@nihui-pc:~# rpm -qa | grep firmware | sort
alsa-sof-firmware-2025.05.1-1.fc42.noarch
amd-gpu-firmware-20251021-1.fc42.noarch
amd-ucode-firmware-20251021-1.fc42.noarch
atheros-firmware-20251021-1.fc42.noarch
atmel-firmware-1.3-34.fc42.noarch
brcmfmac-firmware-20251021-1.fc42.noarch
cirrus-audio-firmware-20251021-1.fc42.noarch
intel-audio-firmware-20251021-1.fc42.noarch
intel-gpu-firmware-20251021-1.fc42.noarch
ipw2100-firmware-1.3-29.fc35.noarch
ipw2200-firmware-3.1-22.fc35.noarch
iwlegacy-firmware-20251021-1.fc42.noarch
iwlwifi-dvm-firmware-20251021-1.fc42.noarch
iwlwifi-mld-firmware-20251021-1.fc42.noarch
iwlwifi-mvm-firmware-20251021-1.fc42.noarch
libertas-firmware-20251021-1.fc42.noarch
linux-firmware-20251021-1.fc42.noarch
linux-firmware-whence-20251021-1.fc42.noarch
mt7xxx-firmware-20251021-1.fc42.noarch
nxpwireless-firmware-20251021-1.fc42.noarch
qcom-wwan-firmware-20251021-1.fc42.noarch
realtek-firmware-20251021-1.fc42.noarch
tiwilink-firmware-20251021-1.fc42.noarch
zd1211-firmware-1.5-18.fc42.noarch
root@nihui-pc:~# uname -a
Linux nihui-pc 6.17.4-200.fc42.x86_64 #1 SMP PREEMPT_DYNAMIC Sun Oct 19 18:47:49 UTC 2025 x86_64 GNU/Linux

Comment 81 sparrow 2025-10-28 17:59:39 UTC
This firmware (mt7xxx-firmware-20251021-1.fc42.noarch) works for me as well!
Previously I had to revert to the 20250311 version.

Host: ASUS Vivobook S 14 M5406WA_M5406WA (1.0)
Kernel: Linux 6.17.4-200.fc42.x86_64
Network: MEDIATEK Corp. MT7922 802.11ax PCI Express Wireless Network Adapter

Comment 82 Peter Robinson 2025-10-28 18:03:56 UTC
(In reply to Richard Kimberly Heck from comment #77)
> Confirming that it works for me, too. Someone should push this to the F42
> updates repo?

It was already pushed, if you read the comments above you would have seen the links from the update platform.

Comment 83 Hugo Posnic 2025-11-05 08:29:25 UTC
Still errors on latest firmware:
https://bugzilla.redhat.com/show_bug.cgi?id=2411854


Note You need to log in before you can comment on or make changes to this bug.