You are not logged in.
I just got a new laptop with intel i5 13500H cpu for arch installation. However during the procedure I find the wireless module cannot be recognized by current linux kernel. The information is post below
# ip link
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
# lspci -k (of network controller info)
0000:00:14.3 Network controller: Intel Corporation Device 51f1 (rev 01)
Subsystem: Rivet Networks Device 1672
Kernel modules: iwlwifi, wl
# lsmod | grep iwlwifi
iwlwifi 548864 0
cfg80211 1286144 2 wl,iwlwifi
# rfkill
ID TYPE DEVICE SOFT HARD
0 wlan acer-wireless unblocked unblocked
1 bluetooth acer-bluetooth unblocked unblocked
2 bluetooth hci0 unblocked unblocked
# dmesg (of iwlwifi info)
[ 29.343682] iwlwifi 0000:00:14.3: enabling device (0000 -> 0002)
[ 29.348772] iwlwifi: No config found for PCI dev 51f1/1672, rev=0x370, rfid=0x2010d000
[ 29.350404] iwlwifi: probe of 0000:00:14.3 failed with error -22
I guess the wireless module is wifi 6E series and driven by the same driver as AX1675i/s (AX211)s. The bug might be led by a upstream bug of kernel, I find a patcn here. However I have no idea on how to use this to modify the installation iso.
Currently, I use the windows installation and post the problem, while I can boot with archiso livecd to gather more information if anyone needs. I want to know how to submit this bug to upstream and when it will get fixed since the patch is published in April.
Thanks guys. If anything I described not clear or wrong or this should not be post here please let me know.
Last edited by wzdlc1996 (2023-06-08 06:29:02)
Offline
See Kernel#Compilation to build a kernel with the patch applied then Archiso#Custom_local_repository.
The ISO I generated with the patched kernel is linked below
https://drive.google.com/file/d/1Psqzb- … share_link archlinux-2023.06.06-x86_64.iso
The patch comment [1] asked for it to be resubmitted with an adjusted title which it does not appear to have been so no further action was taken.
Offline
Thank you for the reply!
I have tried the iso you generated. During livecd the system successfully recognizes wlan0. The partial output of dmesg is listed
# dmesg | grep iwlwifi
[ 17.789017] iwlwifi 0000:00:14.3: enabling device (0000 -> 0002)
[ 18.006244] iwlwifi 0000:00:14.3: api flags index 2 larger than supported by driver
[ 18.006264] iwlwifi 0000:00:14.3: TLV_FW_FSEQ_VERSION: FSEQ Version: 0.0.2.36
[ 18.033642] iwlwifi 0000:00:14.3: loaded firmware version 74.fe17486e.0 so-a0-gf-a0-74.ucode op_mode iwlmvm
[ 18.245383] iwlwifi 0000:00:14.3: Detected Killer(R) Wi-Fi 6E AX1675i 160MHz Wireless Network Adapter (211NGW), REV=0x370
[ 18.254148] iwlwifi 0000:00:14.3: WRT: Invalid buffer destination
[ 18.419921] iwlwifi 0000:00:14.3: WFPM_UMAC_PD_NOTIFICATION: 0x1f
[ 18.419983] iwlwifi 0000:00:14.3: WFPM_LMAC2_PD_NOTIFICATION: 0x1f
[ 18.419992] iwlwifi 0000:00:14.3: WFPM_AUTH_KEY_0: 0x80
[ 18.420001] iwlwifi 0000:00:14.3: CNVI_SCU_SEQ_DATA_DW9: 0x0
[ 18.432015] iwlwifi 0000:00:14.3: loaded PNVM version e4a49534
[ 18.447799] iwlwifi 0000:00:14.3: Detected RF GF, rfid=0x2010d000
[ 18.517368] iwlwifi 0000:00:14.3: base HW address: 8c:f8:c5:03:87:2f
[ 18.610447] iwlwifi 0000:00:14.3: WRT: Invalid buffer destination
[ 18.771207] iwlwifi 0000:00:14.3: WFPM_UMAC_PD_NOTIFICATION: 0x1f
[ 18.771264] iwlwifi 0000:00:14.3: WFPM_LMAC2_PD_NOTIFICATION: 0x1f
[ 18.771273] iwlwifi 0000:00:14.3: WFPM_AUTH_KEY_0: 0x80
[ 18.771282] iwlwifi 0000:00:14.3: CNVI_SCU_SEQ_DATA_DW9: 0x0
[ 21.934067] iwlwifi 0000:00:14.3: WRT: Invalid buffer destination
[ 22.095767] iwlwifi 0000:00:14.3: WFPM_UMAC_PD_NOTIFICATION: 0x1f
[ 22.095797] iwlwifi 0000:00:14.3: WFPM_LMAC2_PD_NOTIFICATION: 0x1f
[ 22.095902] iwlwifi 0000:00:14.3: WFPM_AUTH_KEY_0: 0x80
[ 22.095915] iwlwifi 0000:00:14.3: CNVI_SCU_SEQ_DATA_DW9: 0x0
The ip link also shows wlan0 with state DOWN. The output message shows "WRT: Invalid buffer destination" seems some issues tho. However I think this would be easy to resolved. The patch seems necessary and indeed a solution to this issue. Can we do something help to make it work on current official iso and kernel?
See Kernel#Compilation to build a kernel with the patch applied then Archiso#Custom_local_repository.
The ISO I generated with the patched kernel is linked below
https://drive.google.com/file/d/1Psqzb- … share_link archlinux-2023.06.06-x86_64.isoThe patch comment [1] asked for it to be resubmitted with an adjusted title which it does not appear to have been so no further action was taken.
Offline
I would start by seeing if you can get the interface to connect without further issues.
I do not know what the exact process is for getting a patch into the kernel that the original author has dropped. I would start by trying to contact the original patches author and see if they have any plans to resend as that would be easiest.
If not someone else would need to send it. Arch will not take the patch before at has been accepted into the kernel.
Offline
Thanks!
I complete the installation based on your ISO. As the officially distributed kernel is still original 6.3.6, I cannot use wifi after the installation.
So I recompiled the kernel with that patch and switched to the fixed kernel. Then the wifi works. Till now, I have not seen any issues on it, everything works fine.
I make some search on the internet and find recently this issue also bothers other users from different Linux distros like [1][2][3] . Maybe this patch is necessary as this hardware might get more popular?
Thanks for the help! I learnt a lot about kernel.
ps: If Arch does not apply this patch in the official package, does it mean I should recompile kernel by myself every time it updates? I find it really time consuming : (
[1]: https://community.intel.com/t5/Wireless … -p/1379155
[2]: https://askubuntu.com/questions/1461041 … predator-p
[3]: https://community.intel.com/t5/Wireless … -p/1297466
I would start by seeing if you can get the interface to connect without further issues.
I do not know what the exact process is for getting a patch into the kernel that the original author has dropped. I would start by trying to contact the original patches author and see if they have any plans to resend as that would be easiest.
If not someone else would need to send it. Arch will not take the patch before at has been accepted into the kernel.
Last edited by wzdlc1996 (2023-06-07 06:46:13)
Offline
ps: If Arch does not apply this patch in the official package, does it mean I should recompile kernel by myself every time it updates? I find it really time consuming : (
[1]: https://community.intel.com/t5/Wireless … -p/1379155
[2]: https://askubuntu.com/questions/1461041 … predator-p
[3]: https://community.intel.com/t5/Wireless … -p/1297466
Yes you will have to recompile the kernel on updates. If the patch gets more popular perhaps that will increase the chances of someone submitting it upstream?
loqs wrote:I would start by seeing if you can get the interface to connect without further issues.
I do not know what the exact process is for getting a patch into the kernel that the original author has dropped. I would start by trying to contact the original patches author and see if they have any plans to resend as that would be easiest.
If not someone else would need to send it. Arch will not take the patch before at has been accepted into the kernel.
See https://github.com/archlinux/linux/commits/v6.3.6-arch1 where Arch is applying two commits that are not upstreamed, one allows preventing unprivileged users to create namespaces and the second marks the release as -arch1.
$ perl scripts/get_maintainer.pl drivers/net/wireless/intel/iwlwifi/pcie/drv.c
Gregory Greenman <gregory.greenman@intel.com> (supporter:INTEL WIRELESS WIFI LINK (iwlwifi),commit_signer:13/12=100%,authored:1/12=8%)
Kalle Valo <kvalo@kernel.org> (maintainer:NETWORKING DRIVERS (WIRELESS))
Johannes Berg <johannes.berg@intel.com> (commit_signer:10/12=83%)
Mukesh Sisodiya <mukesh.sisodiya@intel.com> (commit_signer:8/12=67%,authored:8/12=67%,added_lines:287/312=92%,removed_lines:197/200=98%)
Daniel Gabay <daniel.gabay@intel.com> (commit_signer:1/12=8%,authored:1/12=8%)
Golan Ben Ami <golan.ben.ami@intel.com> (commit_signer:1/12=8%,authored:1/12=8%)
Yaara Baruch <yaara.baruch@intel.com> (authored:1/12=8%,added_lines:17/312=5%)
linux-wireless@vger.kernel.org (open list:INTEL WIRELESS WIFI LINK (iwlwifi))
linux-kernel@vger.kernel.org (open list)
You could contact Gregory Greenman if you received no response from Yi Kuo although do note he was the recipient of the original patch addressing this issue sent by Yi Kuo.
Upstream bug report https://bugzilla.kernel.org/show_bug.cgi?id=215876
Offline
Thanks!
I will contact to Yi Kuo via email with the link to this post. If he/she do not reply, I will try to contact to Gregory Greeman about this issue!
I will also mark this post [SOLVED]. I am appreciate for your help!
You could contact Gregory Greenman if you received no response from Yi Kuo although do note he was the recipient of the original patch addressing this issue sent by Yi Kuo.
Upstream bug report https://bugzilla.kernel.org/show_bug.cgi?id=215876
Offline
Do you guys still have a copy of the patched kernel? I am having the same issue and I've never patched or compile a kernel before.
Offline
The following step by step is based on Kernel/Arch_build_system it requires devtools in addition to base-devel and applies https://lore.kernel.org/linux-wireless/ … yikuo.dev/ to the current linux package, I put the required changes in a diff and upload it to a pastebin, which the step by step downloads and uses.
I recommend you enable Makepkg#Parallel_compilation before building the kernel.
$ pkgctl repo clone --protocol=https linux #obtain the PKGBUILD for linux
$ cd linux/
$ curl -o PKGBUILD.diff http://ix.io/4yuG # download diff of changes
$ git apply -v PKGBUILD.diff
$ makepkg -rsi # build package
Built package
https://drive.google.com/file/d/1Xtb-nF … sp=sharing linux-6.3.8.arch1-1.1-x86_64.pkg.tar.zst
https://drive.google.com/file/d/1C2xOM8 … sp=sharing linux-headers-6.3.8.arch1-1.1-x86_64.pkg.tar.zst
Last edited by loqs (2023-06-17 16:41:22)
Offline
Thanks loqs. I'll work on this.
Offline
Thanks loqs. The patched kernel works.
Offline
The commit should land in 6.5 https://git.kernel.org/pub/scm/linux/ke … d520d009d2
Offline
With kernel 6.6.5 I'm still seeing the issue:
❯ uname -r
6.6.5-arch1-1
❯ journalctl -b -p err
Dec 11 15:20:40 robin kernel:
Dec 11 15:20:40 robin kernel: NVRM: loading NVIDIA UNIX x86_64 Kernel Module 545.29.0
6 Thu Nov 16 01:59:08 UTC 2023
Dec 11 15:20:41 robin kernel: spi-nor spi0.0: unrecognized JEDEC id bytes: f7 f0 30 09
03 00
Dec 11 15:20:41 robin kernel: iwlwifi 0000:00:14.3: WRT: Invalid buffer destination
Dec 11 15:20:43 robin kernel: Bluetooth: hci0: Malformed MSFT vendor event: 0x02
Dec 11 15:20:43 robin kernel: iwlwifi 0000:00:14.3: WRT: Invalid buffer destination
Dec 11 15:20:44 robin kernel: iwlwifi 0000:00:14.3: WRT: Invalid buffer destination
Dec 11 15:20:56 robin pulseaudio[2663]: GetManagedObjects() failed: org.freedesktop.sy
stemd1.NoSuchUnit: Unit dbus-org.bluez.service not found.
Dec 11 15:20:57 robin systemd[2153]: Failed to start Flameshot - Screenshot tool.
Dec 11 15:31:01 robin bluetoothd[8886]: src/adapter.c:reset_adv_monitors_complete() Fa
iled to reset Adv Monitors: Failed (0x03)
Dec 11 15:31:01 robin bluetoothd[8886]: Failed to clear UUIDs: Failed (0x03)
Dec 11 15:31:01 robin bluetoothd[8886]: Failed to add UUID: Failed (0x03)
Dec 11 15:31:01 robin bluetoothd[8886]: Failed to add UUID: Failed (0x03)
Dec 11 15:31:01 robin bluetoothd[8886]: Failed to add UUID: Failed (0x03)
Dec 11 15:43:15 robin sudo[9981]: aemonge : a password is required ; PWD=/home/aemong
e ; USER=root ; COMMAND=/usr/bin/dd if=/tmp/nvim.aemonge/YZRpM9/1 of=/etc/modprobe.d/iwlwifi.conf bs=1048576
Any idea why? or how to fix it?
Last edited by aemonge (2023-12-11 14:47:39)
Offline
@aemonge why do you believe it is the same issue; when the output you posted does not contain:
[ 29.348772] iwlwifi: No config found for PCI dev 51f1/1672, rev=0x370, rfid=0x2010d000
Offline
Hey @loqs, you're completely right. This issue is not the same
I've ended up in this forum, looking for answers to my issue, and didn't read comprehensively. Sorry about that.
Offline