Intel wifi 7265 only appears after sleeping and waking laptop

Baeldaeg · 2026-04-03 08:56:22

As far as I can see there's no sign of the wifi device until the laptop is put to sleep and then awakened.

lspci -k before sleeping:
https://pastebin.com/xhqtTrsP

lspci -k after sleeping:
https://pastebin.com/N1dw0Svx

journalctl before sleeping:
https://pastebin.com/FSEJFK2z

journalctl after sleeping:
https://pastebin.com/9vpRWava

seth · 2026-04-03 13:06:41

Before:

Apr 03 08:43:11 archlinux kernel: Linux version 6.18.20-1-lts (linux-lts@archlinux) (gcc (GCC) 15.2.1 20260209, GNU ld (GNU Binutils) 2.46) #1 SMP PREEMPT_DYNAMIC Wed, 25 Mar 2026 12:17:34 +0000
Apr 03 08:43:11 archlinux kernel: pci 0000:01:00.0: [8086:095a] type 00 class 0x028000 PCIe Endpoint
Apr 03 08:43:11 archlinux kernel: pci 0000:01:00.0: BAR 0 [mem 0xcef00000-0xcef01fff 64bit]
Apr 03 08:43:11 archlinux kernel: pci 0000:01:00.0: PME# supported from D0 D3hot D3cold

After:

Apr 03 09:01:11 Sarah-laptop kernel: pci 0000:01:00.0: [8086:095a] type 00 class 0x028000 PCIe Endpoint
Apr 03 09:01:11 Sarah-laptop kernel: pci 0000:01:00.0: BAR 0 [mem 0xcef00000-0xcef01fff 64bit]
Apr 03 09:01:11 Sarah-laptop kernel: pci 0000:01:00.0: PME# supported from D0 D3hot D3cold
Apr 03 09:01:11 Sarah-laptop kernel: pci 0000:01:00.0: BAR 0 [mem 0xcef00000-0xcef01fff 64bit]: assigned
Apr 03 09:01:12 Sarah-laptop kernel: iwlwifi 0000:01:00.0: Detected crf-id 0x0, cnv-id 0x0 wfpm id 0x0
Apr 03 09:01:12 Sarah-laptop kernel: iwlwifi 0000:01:00.0: PCI dev 095a/9e10, rev=0x210, rfid=0xd55555d5
Apr 03 09:01:12 Sarah-laptop kernel: iwlwifi 0000:01:00.0: Detected Intel(R) Dual Band Wireless-AC 7265
Apr 03 09:01:12 Sarah-laptop kernel: iwlwifi 0000:01:00.0: Found debug destination: EXTERNAL_DRAM
Apr 03 09:01:12 Sarah-laptop kernel: iwlwifi 0000:01:00.0: Found debug configuration: 0
Apr 03 09:01:12 Sarah-laptop kernel: iwlwifi 0000:01:00.0: loaded firmware version 29.9ef079ed.0 7265D-29.ucode op_mode iwlmvm
Apr 03 09:01:13 Sarah-laptop kernel: iwlwifi 0000:01:00.0: Applying debug destination EXTERNAL_DRAM
Apr 03 09:01:13 Sarah-laptop kernel: iwlwifi 0000:01:00.0: Allocated 0x00400000 bytes for firmware monitor.
Apr 03 09:01:13 Sarah-laptop kernel: iwlwifi 0000:01:00.0: base HW address: b4:d5:bd:c4:f9:f0, OTP minor version: 0x0
Apr 03 09:01:13 Sarah-laptop NetworkManager[565]: <info>  [1775203273.3051] rfkill2: found Wi-Fi radio killswitch (at /sys/devices/pci0000:00/0000:00:1c.0/0000:01:00.0/ieee80211/phy0/rfkill2) (driver iwlwifi)
Apr 03 09:01:13 Sarah-laptop kernel: iwlwifi 0000:01:00.0 wlp1s0: renamed from wlan0
Apr 03 09:01:13 Sarah-laptop kernel: iwlwifi 0000:01:00.0: Applying debug destination EXTERNAL_DRAM
Apr 03 09:01:13 Sarah-laptop kernel: iwlwifi 0000:01:00.0: Applying debug destination EXTERNAL_DRAM
Apr 03 09:01:13 Sarah-laptop kernel: iwlwifi 0000:01:00.0: FW already configured (0) - re-configuring
Apr 03 09:01:13 Sarah-laptop kernel: iwlwifi 0000:01:00.0: Applying debug destination EXTERNAL_DRAM
Apr 03 09:01:13 Sarah-laptop kernel: iwlwifi 0000:01:00.0: Applying debug destination EXTERNAL_DRAM
Apr 03 09:01:13 Sarah-laptop kernel: iwlwifi 0000:01:00.0: FW already configured (0) - re-configuring

Apr 03 08:43:11 archlinux kernel: DMI: Google Eve/Eve, BIOS MrChromebox-2512.2 01/25/2026

See eg. https://github.com/MrChromebox/firmware/issues/795

1. is this a regression (hinging on the mrchromebox firmware version)?
2. Is the 6.19 kernel affected?
3. is the install iso (or https://grml.org/ )
4. Do you use a bootmanager (liek grub)? How much time passes from powering the device til the kernel loading?

Baeldaeg · 2026-04-03 18:05:18

1. I don't think so as I updated the firmware after the wifi started misbehaving.
2. Yes.
3. GRML shows the same output to lspci
4. I use GRUB. Is there a program to measure the time? Or do I just use a stopwatch?

If I boot off a Windows PE memory stick the wifi works normally on that.

seth · 2026-04-03 19:28:07

The idea would be to just stall the boot a bit in order the HW needs a moment to power up.

You can also try to lie to the firmware, eg

acpi_osi=! acpi_osi="Windows 2015"

https://learn.microsoft.com/en-us/windo … inacpi-osi

If I boot off a Windows PE memory stick the wifi works normally on that.

Did you try this only after the fact or have you been running windows on this device ("sorta dual boot")?

See http://wiki.archlinux.org/title/Dual_bo … ibernation - it would be a rather typical symptom but since there were no further partitions I disregarded that scenario.
In doubt try to properly shut down windows from the PE boot once.

Baeldaeg · 2026-04-05 09:53:30

Neither of those had any effect. I also tried

acpi_rev_override

to no effect.

However installing linux-lts-6.12.75-1 from the archive gets it working properly.
Going up to linux-lts-6.18.16-1 breaks it again.

I might try the non-lts kernels in-between those and see exactly which version it stops working on.

Baeldaeg · 2026-04-05 10:40:52

linux-6.16.9.arch1-1 the wifi works as it should.

linux-6.17.1.arch1-1 and linux-6.17.2.arch1-1 boot to a black screen.

linux-6.17.3.arch1-1 the wifi misbehaves as above.

seth · 2026-04-05 17:42:27

Can you test the 6.17 rc bisection kernels from https://pkgbuild.com//~gromit/linux-bisection-kernels/ ?

About the black screen, try adding "nomodeset" (but this will only work w/ the multi-user.target, see the 2nd link below)

Baeldaeg · 2026-04-05 18:38:00

Same misbehaviour with 6.17.1 rc.
Should I try 6.17.2 etc too?

seth · 2026-04-05 18:58:11

At this point it's not very likely that kernels between 6.17rc1 and 6.17.3 do not exhibit the behavior.

Instead of a S3 cycle, does it suffice to rescan the bus?

echo 1 |sudo tee  /sys/bus/pci/rescan

https://stackoverflow.com/questions/323 … f-pcie-bus

To isolate the offending commit you'll have to https://wiki.archlinux.org/title/Bisect … s_with_Git https://github.com/archlinux/linux
(I'm afraid the issue is too odd and unspecific to make an informed guess as to why the initial probe might fail/happen)

Ftr, did you test 6.16.10.arch1-1 (though I assume it'll also work)

Baeldaeg · 2026-04-06 10:29:17

The bus rescan didn't seem to do anything.

I hadn't realised there was a 6.16.10. I've now tried that and the wifi misbehaves. I then went back to 6.16.9 and it misbehaves on there too now. I downgraded a couple more times and found 6.15.9-arch1-1 is the highest version it works normally on.

linux-lts-6.12.75-1 is still working fine.

Last edited by Baeldaeg (2026-04-06 10:40:57)

seth · 2026-04-06 19:47:47

I downgraded a couple more times and found 6.15.9-arch1-1 is the highest version it works normally on.

Sure this isn't a race condition and works by fluke?
gromit also has bisection kernels for the 6.16 rc's

The bus rescan didn't seem to do anything.

You'd expect the device to show up in lspci, also the rescan should trigger some output in

dmesg -W

The wifi would not magically work and pot. even be rfkill'd

Arch Linux

#1 2026-04-03 08:56:22

Intel wifi 7265 only appears after sleeping and waking laptop

#2 2026-04-03 13:06:41

Re: Intel wifi 7265 only appears after sleeping and waking laptop

#3 2026-04-03 18:05:18

Re: Intel wifi 7265 only appears after sleeping and waking laptop

#4 2026-04-03 19:28:07

Re: Intel wifi 7265 only appears after sleeping and waking laptop

#5 2026-04-05 09:53:30

Re: Intel wifi 7265 only appears after sleeping and waking laptop

#6 2026-04-05 10:40:52

Re: Intel wifi 7265 only appears after sleeping and waking laptop

#7 2026-04-05 17:42:27

Re: Intel wifi 7265 only appears after sleeping and waking laptop

#8 2026-04-05 18:38:00

Re: Intel wifi 7265 only appears after sleeping and waking laptop

#9 2026-04-05 18:58:11

Re: Intel wifi 7265 only appears after sleeping and waking laptop

#10 2026-04-06 10:29:17

Re: Intel wifi 7265 only appears after sleeping and waking laptop

#11 2026-04-06 19:47:47

Re: Intel wifi 7265 only appears after sleeping and waking laptop

Board footer