Almost 2 month later and I've succeeded in solving that issue.
Created file /etc/modprobe.d/iwl.conf:
options iwlwifi 11n_disable=1 swcrypto=0 bt_coex_active=0 power_save=0 options iwlmvm power_scheme=1 options iwlwifi d0i3_disable=1 options iwlwifi uapsd_disable=1 options iwlwifi lar_disable=1
And reboot. No microcode errors found for two days. Maybe it's not necessary to disable all of those features.
A credit goes to linux mint & manjaro forums:
https://forums.linuxmint.com/viewtopic.php?t=295219
https://forum.manjaro.org/t/sovled-iwlw … blem/82689
This worked like a charm for me.
My issue was every time I connected my laptop to an external monitor my internet connection was down.
Sorry for the abruptness.
In /etc/modprobe.d/specify_name.conf
'options iwlmvm power_scheme=1'
and reboot.My computer is equipped with Intel AX200.
Until recently I was having the same problem.I hope that helps to solve the problem.
Looks like it solved the issue! Thanks for sharing this workaround.
]]>@04tm34l: I will test it, thanks for the suggestion.
]]>In /etc/modprobe.d/specify_name.conf
'options iwlmvm power_scheme=1'
and reboot.
My computer is equipped with Intel AX200.
Until recently I was having the same problem.
I hope that helps to solve the problem.
]]>Thanks for the suggestion, I'll try that and let you know if it helps (my connection is slower than 150 Mbps anyway).
]]>I went into router + extender and set the channel width to 20Mhz instead of auto (20/40MHZ) and it doesn't happen now. The disk still errors (but works), but iwlwifi doesnt crash.
edit: Am using 11n_disable=8 as it gives me higher speeds.
I have Intel 7260 card. This problem seems to happen (at least for me) only on 802.11n 40MHz channel. I mean only 40MHz, as I do not see the problem on slower 20MHz, 802.11g, or even 802.11ac. My wifi works seamlessly on 5GHz 802.11ac. For me, I do not always see the microcode sw error, but what I always see is that 40MHz connection is so bad, that is is not even usable. The ping goes up to 4000 and timeout. A temporary workaround is to disable 40MHz yes, but this means using slower connection (less than 150 Mbps). Btw you don't need to disable 40MHz from the router, you can do it by adding `cfg80211_disable_40mhz_24ghz=Y` options, which will, well, disable 40MHz for all networks. This option is available for the `cfg80211` module. Add this in /etc/modprobe.d/cfg80211.conf:
options cfg80211 cfg80211_disable_40mhz_24ghz=Y
I don't know if others have tried it, but I do not need to add any module option to iwlwifi for this to work, not even disabling 802.11n, which is really unacceptable as it disables 802.11ac too and that means only using slower 802.11g. Turning on software encryption, disabling bluetooth coexistence, disabling power save (which by default has been disabled) do not really do it. Using
11n_disable=8
does not really make a difference, still 40MHz channel width is unusable (maybe latency is better but maybe I don't feel it).
Also for you who may have the same case for me, my laptop originally comes with 2.4GHz only wifi adapter. I then replaced it with Intel 7260, with the same antenna, and 5GHz works flawlessly. I don't know if antenna seems to be the problem for 40MHz channel, need others to confirm.
]]>Kewl wrote:Thanks seth for your help.
I am now testing the 9260NGW with LTS 5.4.87, iwlwifi has not crashed for the past 24 hours (which does not mean it will not), I am testing in ac mode without any parameter.Otherwise I will switch to a Qualcomm card, I have a QCA9377 running in ac at 433Mbps which has never caused any issue and they have the QCA6174A with 2 streams (867Mbps) that would be worth testing, both cards have BT 5.0 and are supported by the ath10k driver. This is out of scope of this thread but to give some alternatives for the ones who would be really stuck with this crash.
I have a I'd like to jump ship to a Qualcomm card, but I can't find one that fits my needs. I need it to be PCIE, and really fast. It seems like Intel kind of dominates in that category, but the cards are unreliable on Linux.
If I search for QCA6174A on amazon, or anywhere, I either get m.2 type cards, or more Intel cards. Maybe Qualcomm wifi chips don't often make their way into Desktop parts?
PS: I currently have a Wi-Fi 6 AX200, which is plenty fast, but it keeps crashing my computer. This is my second Intel device, I replaced the old one thinking that it was a hardware problem, not a driver problem. So I'd recommend just staying away from Intel chips if you need performance at 5ghz (if an appropriate non Intel part can be found)
You could get a M.2 to PCIe adapter for about $15 (adapter and antennas) it could give you more choice of cards.
I gave up with the 9260NGW which started to crash with Linux LTS as well. I am testing 8260NGW which seems to be a bit more stable so far in 802.11ac (in 2 * 80MHz mode = 866mbps). The main difference is that it does not support 160MHz and has Bluetooth 4.2 vs 5.1.
Thanks seth for your help.
I am now testing the 9260NGW with LTS 5.4.87, iwlwifi has not crashed for the past 24 hours (which does not mean it will not), I am testing in ac mode without any parameter.Otherwise I will switch to a Qualcomm card, I have a QCA9377 running in ac at 433Mbps which has never caused any issue and they have the QCA6174A with 2 streams (867Mbps) that would be worth testing, both cards have BT 5.0 and are supported by the ath10k driver. This is out of scope of this thread but to give some alternatives for the ones who would be really stuck with this crash.
I have a I'd like to jump ship to a Qualcomm card, but I can't find one that fits my needs. I need it to be PCIE, and really fast. It seems like Intel kind of dominates in that category, but the cards are unreliable on Linux.
If I search for QCA6174A on amazon, or anywhere, I either get m.2 type cards, or more Intel cards. Maybe Qualcomm wifi chips don't often make their way into Desktop parts?
PS: I currently have a Wi-Fi 6 AX200, which is plenty fast, but it keeps crashing my computer. This is my second Intel device, I replaced the old one thinking that it was a hardware problem, not a driver problem. So I'd recommend just staying away from Intel chips if you need performance at 5ghz (if an appropriate non Intel part can be found)
]]>jan 17 10:35:23 arch kernel: usb-storage 2-2:1.0: USB Mass Storage device detected
jan 17 10:35:23 arch kernel: scsi host2: usb-storage 2-2:1.0
jan 17 10:35:24 arch kernel: scsi 2:0:0:0: Direct-Access WD My Passport 259F 1004 PQ: 0 ANSI: 6
jan 17 10:35:24 arch kernel: scsi 2:0:0:1: Enclosure WD SES Device 1004 PQ: 0 ANSI: 6
jan 17 10:35:24 arch kernel: ses 2:0:0:1: Attached Enclosure device
jan 17 10:35:24 arch kernel: sd 2:0:0:0: [sdb] Spinning up disk...
jan 17 10:35:28 arch kernel: ses 2:0:0:1: Wrong diagnostic page; asked for 1 got 8
jan 17 10:35:28 arch kernel: ses 2:0:0:1: Failed to get diagnostic page 0x1
jan 17 10:35:28 arch kernel: ses 2:0:0:1: Failed to bind enclosure -19
jan 17 10:35:29 arch kernel: .ready
jan 17 10:35:29 arch kernel: sd 2:0:0:0: [sdb] 1953458176 512-byte logical blocks: (1.00 TB/931 GiB)
jan 17 10:35:29 arch kernel: sd 2:0:0:0: [sdb] Write Protect is off
jan 17 10:35:29 arch kernel: sd 2:0:0:0: [sdb] Mode Sense: 47 00 10 08
jan 17 10:35:29 arch kernel: sd 2:0:0:0: [sdb] No Caching mode page found
jan 17 10:35:29 arch kernel: sd 2:0:0:0: [sdb] Assuming drive cache: write through
jan 17 10:35:30 arch kernel: sdb: sdb1
jan 17 10:35:30 arch kernel: sd 2:0:0:0: [sdb] Attached SCSI disk
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: Queue 11 is active on fifo 1 and stuck for 10000 ms. SW [171, 181] HW [171, 181] FH TRB=0x0c010b0b4
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: Microcode SW error detected. Restarting 0x2000000.
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: Start IWL Error Log Dump:
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: Status: 0x00000040, count: 6
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: Loaded firmware version: 29.198743027.0 7265D-29.ucode
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000084 | NMI_INTERRUPT_UNKNOWN
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00800634 | trm_hw_status0
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | trm_hw_status1
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00043D6C | branchlink2
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x0004AFD6 | interruptlink1
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x0000AD8C | interruptlink2
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | data1
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000080 | data2
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x07030000 | data3
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x28002E5B | beacon time
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x426471A4 | tsf low
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000004 | tsf hi
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | time gp1
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x141A0876 | time gp2
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000001 | uCode revision type
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x0000001D | uCode version major
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x0BD893F3 | uCode version minor
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000210 | hw version
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00489200 | board version
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x0BAB001C | hcmd
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0xA4022002 | isr0
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x01000000 | isr1
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x0000000A | isr2
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x004138C5 | isr3
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | isr4
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x008D019C | last cmd Id
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | wait_event
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000080 | l2p_control
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00012030 | l2p_duration
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x0000003F | l2p_mhvalid
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x000000CE | l2p_addr_match
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000007 | lmpm_pmg_sel
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x14101241 | timestamp
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00348898 | flow_handler
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: Fseq Registers:
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | FSEQ_ERROR_CODE
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | FSEQ_TOP_INIT_VERSION
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | FSEQ_CNVIO_INIT_VERSION
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | FSEQ_OTP_VERSION
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | FSEQ_TOP_CONTENT_VERSION
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | FSEQ_ALIVE_TOKEN
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | FSEQ_CNVI_ID
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | FSEQ_CNVR_ID
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | CNVI_AUX_MISC_CHIP
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | CNVR_AUX_MISC_CHIP
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | CNVR_SCU_SD_REGS_SD_REG_DIG_DCDC_VTRIM
jan 17 10:35:33 arch kernel: iwlwifi 0000:02:00.0: 0x00000000 | CNVR_SCU_SD_REGS_SD_REG_ACTIVE_VDIG_MIRROR
I went into router + extender and set the channel width to 20Mhz instead of auto (20/40MHZ) and it doesn't happen now. The disk still errors (but works), but iwlwifi doesnt crash.
edit: Am using 11n_disable=8 as it gives me higher speeds.
]]>seth wrote:Notice that a bunch of module options are set by the people claiming them to be an effective mitigation and a different option or a combination could be the relevant one.
Just tested with all the parameters proposed by sinatosk, and the issue still occurs.
I suggest you try with Linux LTS, it has not crashed for a few days with LTS on my side then I am facing some new connection issues after suspend I did not have with the most recent kernel I have not resolved yet.
]]>Notice that a bunch of module options are set by the people claiming them to be an effective mitigation and a different option or a combination could be the relevant one.
Just tested with all the parameters proposed by sinatosk, and the issue still occurs.
]]>Otherwise I will switch to a Qualcomm card, I have a QCA9377 running in ac at 433Mbps which has never caused any issue and they have the QCA6174A with 2 streams (867Mbps) that would be worth testing, both cards have BT 5.0 and are supported by the ath10k driver. This is out of scope of this thread but to give some alternatives for the ones who would be really stuck with this crash.
]]>