Dell Inspiron 3567 - WiFi disconnect + Power Cable error

gnome
#1

Hi everyone, it’s from a while that the wifi disconnect randomly: this is the output of dmesg after the wifi reconnected:

> [  160.008698] Failed to release pages: bind_count=1, pages_pin_count=1, pin_display=0
> [  160.008735] ------------[ cut here ]------------
> [  160.008833] WARNING: CPU: 3 PID: 145 at drivers/gpu/drm/i915/i915_gem_userptr.c:89 cancel_userptr.cold.17+0x25/0x2c [i915]
> [  160.008841] Modules linked in: fuse cmac rfcomm ccm bnep nls_iso8859_1 nls_cp437 vfat fat joydev hid_rmi rmi_core videobuf2_vmalloc videobuf2_memops videobuf2_v4l2 videobuf2_core snd_soc_skl videodev snd_soc_skl_ipc media intel_rapl snd_soc_sst_ipc snd_soc_sst_dsp x86_pkg_temp_thermal intel_powerclamp snd_hda_codec_hdmi coretemp snd_hda_codec_realtek snd_hda_codec_generic kvm_intel snd_hda_ext_core kvm snd_soc_sst_match ath3k snd_soc_core irqbypass snd_compress snd_pcm_dmaengine crct10dif_pclmul btusb btrtl btbcm btintel crc32_pclmul ghash_clmulni_intel pcbc wmi_bmof aesni_intel dell_wmi aes_x86_64 ac97_bus crypto_simd glue_helper cryptd i915 arc4 rtsx_usb_ms intel_cstate ath9k memstick intel_rapl_perf ath9k_common snd_hda_intel ath9k_hw bluetooth ecdh_generic snd_hda_codec dell_laptop dell_smbios
> [  160.008902]  ath dcdbas i2c_algo_bit mousedev iTCO_wdt mac80211 drm_kms_helper snd_hda_core iTCO_vendor_support dell_smm_hwmon r8169 mii snd_hwdep snd_pcm cfg80211 snd_timer psmouse evdev input_leds mac_hid i2c_hid rfkill snd battery pcspkr soundcore wmi thermal intel_hid pinctrl_sunrisepoint pinctrl_intel video drm sparse_keymap int3400_thermal acpi_thermal_rel intel_gtt agpgart pcc_cpufreq acpi_pad button ac syscopyarea processor_thermal_device sysfillrect int340x_thermal_zone shpchp sysimgblt fb_sys_fops intel_soc_dts_iosf idma64 intel_pch_thermal i2c_i801 mei_me mei intel_lpss_pci intel_lpss sch_fq_codel uinput sg crypto_user ip_tables x_tables ext4 crc32c_generic crc16 mbcache jbd2 fscrypto sr_mod cdrom sd_mod rtsx_usb_sdmmc led_class mmc_core rtsx_usb hid_generic usbhid hid serio_raw atkbd
> [  160.008968]  libps2 ahci xhci_pci libahci xhci_hcd crc32c_intel libata scsi_mod usbcore usb_common i8042 serio
> [  160.008983] CPU: 3 PID: 145 Comm: kworker/u8:2 Not tainted 4.14.67-1-MANJARO #1
> [  160.008984] Hardware name: Dell Inc. Inspiron 15-3567/0FGN4M, BIOS 2.6.0 07/17/2018
> [  160.009030] Workqueue: i915-userptr-release cancel_userptr [i915]
> [  160.009033] task: ffff91bb836e0f00 task.stack: ffffb4e58107c000
> [  160.009081] RIP: 0010:cancel_userptr.cold.17+0x25/0x2c [i915]
> [  160.009083] RSP: 0018:ffffb4e58107fe80 EFLAGS: 00010246
> [  160.009086] RAX: 0000000000000047 RBX: ffff91bb394be3c0 RCX: 0000000000000000
> [  160.009088] RDX: 0000000000000000 RSI: ffff91bb8f5965d8 RDI: ffff91bb8f5965d8
> [  160.009089] RBP: ffff91bb394be570 R08: 0000000000000001 R09: 000000000000032d
> [  160.009091] R10: 0000000000000004 R11: 0000000000000000 R12: 0000000000000000
> [  160.009092] R13: 0000000000000000 R14: 0ffff91bb6c9f120 R15: ffff91bb8337e780
> [  160.009095] FS:  0000000000000000(0000) GS:ffff91bb8f580000(0000) knlGS:0000000000000000
> [  160.009097] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [  160.009099] CR2: 00007f50b78206e0 CR3: 000000002d00a003 CR4: 00000000003606e0
> [  160.009100] Call Trace:
> [  160.009110]  process_one_work+0x1e0/0x3e0
> [  160.009114]  worker_thread+0x2b/0x3d0
> [  160.009117]  ? rescuer_thread+0x340/0x340
> [  160.009121]  kthread+0x119/0x130
> [  160.009125]  ? kthread_create_on_node+0x60/0x60
> [  160.009130]  ret_from_fork+0x35/0x40
> [  160.009133] Code: ff e9 8a b0 f5 ff 8b 93 d0 01 00 00 8b 8b ac 01 00 00 48 c7 c7 88 0f df c0 c6 05 95 da 0a 00 01 8b b3 a4 01 00 00 e8 9b 41 31 db <0f> 0b e9 6c b5 f5 ff 48 c7 c6 08 15 df c0 48 c7 c7 09 a7 e0 c0 
> [  160.009187] ---[ end trace 1af3b9d346b63d51 ]---
> [  483.033786] wlp1s0: authenticate with 50:a7:2b:21:c3:34
> [  483.047934] wlp1s0: send auth to 50:a7:2b:21:c3:34 (try 1/3)
> [  483.056011] wlp1s0: authenticated
> [  483.058638] wlp1s0: associate with 50:a7:2b:21:c3:34 (try 1/3)
> [  483.062983] wlp1s0: RX AssocResp from 50:a7:2b:21:c3:34 (capab=0x431 status=0 aid=2)
> [  483.063087] wlp1s0: associated
> [  483.068130] ath: EEPROM regdomain: 0x817c
> [  483.068131] ath: EEPROM indicates we should expect a country code
> [  483.068132] ath: doing EEPROM country->regdmn map search
> [  483.068133] ath: country maps to regdmn code: 0x37
> [  483.068134] ath: Country alpha2 being used: IT
> [  483.068134] ath: Regpair used: 0x37
> [  483.068135] ath: regdomain 0x817c dynamically updated by country IE
> [ 1366.705340] acpi INT3400:00: Unsupported event [0x86]
> [ 1367.130130] ACPI Error: Cannot release Mutex [PATM], not acquired (20170728/exmutex-393)
> [ 1367.130149] ACPI Error: Method parse/execution failed \_SB.PCI0.LPCB.ECDV._Q66, AE_AML_MUTEX_NOT_ACQUIRED (20170728/psparse-550)
> [ 1374.984966] acpi INT3400:00: Unsupported event [0x86]
> [ 1375.406113] ACPI Error: Cannot release Mutex [PATM], not acquired (20170728/exmutex-393)
> [ 1375.406141] ACPI Error: Method parse/execution failed \_SB.PCI0.LPCB.ECDV._Q66, AE_AML_MUTEX_NOT_ACQUIRED (20170728/psparse-550)
> [ 1375.411159] ACPI Error: Cannot release Mutex [PATM], not acquired (20170728/exmutex-393)
> [ 1375.411180] ACPI Error: Method parse/execution failed \_SB.PCI0.LPCB.ECDV._Q66, AE_AML_MUTEX_NOT_ACQUIRED (20170728/psparse-550)

To make matters worse, the last 6 lines of output appears when i unplug (the first 3 lines) and plug (the last 4 lines).
How can i resolve? Expecially the wifi issue. Thanks in advance.

0 Likes

#2

So ok, i fixed the wifi issue.
The Inspiron 3567, has a Qualcomm Atheros QCA9565 / AR9565 wifi card, the problem is with the arh9k driver, and it can easily fixed with:

sudo gedit /etc/modprobe.d/ath9k.conf
Adding the line: options ath9k nohwcrypt=1
Save + Reboot.

But remain the issue with the power cable, and the error that it can be seen in dmseg posted (where there is “cut here”). Thanks in advance.

UPDATE: The wifi issue, still appear… Anyone has suggestion?
UPDATE 2: In practice, the wifi oscillates between a good signal with a stable connection, and moments when it decreases and comes off restarting. I’m no more than 10 meters from the modem, and there’s only one wall to separate. The problem is most probably from the SO.

0 Likes

#3

No one can help me? This wifi problem is gonna make me cry…

0 Likes

#4

From your errors I would suggest making sure your Regulatory Domain is set with your correct country code.

You can search the Manjaro forum or the Arch Wiki for information on setting the correct country code.

I’m not sure that will fix your issue, but it is a good place to start.

You can also try disabling Mac address randomization:

sudo nano /etc/NetworkManager/NetworkManager.conf

Add these lines to the conf file:

[device]
wifi.scan-rand-mac-address=no

Your adapter is a combo card with integrated bluetooth. Bluetooth can often interfere with wifi when they are in close proximity. If Bluetooth is not essential, you can try disabling it to see if your wifi improves. The best way to disable bluetooth is to create or edit:

/etc/modprobe.d/blacklist.conf

Add the following entry:

install bluetooth /bin/false 

Then disable the Bluetooth service:

sudo systemctl disable bluetooth

Then, restart.

That will disable bluetooth by preventing its dependencies from starting any bluetooth module. You can easily enable bluetooth again by commenting out (or delete) the entry in /etc/modprobe.d/blacklist.conf and then re-enable the Bluetooth service.

Testing out kernels 4.18 & 4.19 may also be a good idea.

Does your wifi work better or worse with your power disconnected? This could perhaps indicate a powersaving issue. Try starting your laptop with the power unplugged, then plug the power in after your computer has fully started. See if your wifi behaves differently when plugging/unplugging as I suggested.

1 Like

#5

Hi and thanks for your reply!
I investigated to set a properly Regulatory Domain, thath i think is the real problem. Previously at this post, i changed it by “sudo iw reg set IT” but i read online that it is hardcoded in EEPROM and it doesn’t change by userland.
So i searched online, but it seems to be a kernel problem, so i update to 4.18 how you suggested, but the problem is aggravated.
Now, I think to return at 4.14 and try with the other solutions you have posted, one by one.
P.s. the stability of the net it seems not to depend by the power cable.

1 Like

closed #6

This topic was automatically closed after 180 days. New replies are no longer allowed.

0 Likes