pot8oe
(Thomas Kenny Jr)
April 16, 2023, 1:59am
43
I noticed if I remove and add the iwlwifi driver a couple times it will start working. I wrote a script that runs until the driver loads successfully. Hack but usually only takes a couple quick iterations for success. Not at my workstation now but can share the simple script tomorrow if you want
-Tom
gnwiii
(George N. White III)
April 16, 2023, 12:59pm
44
This sounds like a âraceâ condition. The iwlwifi driver supports many models so may spend more time probing the card than typical drivers, and
systemd is loading stuff in parallel so it may get less time.
Some use cases need network enabled earlier than others. You could try turning off wifi so it only loads manually (e.g., using the NM gui or command-line). A bug report is in order.
1 Like
mutenroid
(Alberto Moro)
April 16, 2023, 8:55pm
45
`⯠rfkill
ID TYPE DEVICE SOFT HARD
0 bluetooth tpacpi_bluetooth_sw unblocked unblocked
1 bluetooth hci0 unblocked unblocked
2 wlan phy0 unblocked unblocked
`
mutenroid
(Alberto Moro)
April 16, 2023, 9:05pm
46
Interesting appointment about this.
I mentioned in the past in this same thread, that with and old Manjaro version works perfectly.
In particular with the version 21.3.7 (https://download.manjaro.org/gnome/21.3.7/manjaro-gnome-21.3.7-220816-linux515.iso ). These are dmesg, journal log for this version from live usb:
[manjaro-gnome ~]# dmesg | grep iwlwifi
[ 11.196724] iwlwifi 0000:00:14.3: enabling device (0000 -> 0002)
[ 11.198921] iwlwifi 0000:00:14.3: Direct firmware load for iwlwifi-so-a0-gf-a0-66.ucode failed with error -2
[ 11.198960] iwlwifi 0000:00:14.3: Direct firmware load for iwlwifi-so-a0-gf-a0-65.ucode failed with error -2
[ 11.248512] iwlwifi 0000:00:14.3: api flags index 2 larger than supported by driver
[ 11.248524] iwlwifi 0000:00:14.3: TLV_FW_FSEQ_VERSION: FSEQ Version: 0.0.2.25
[ 11.248787] iwlwifi 0000:00:14.3: loaded firmware version 64.97bbee0a.0 so-a0-gf-a0-64.ucode op_mode iwlmvm
[ 11.616374] iwlwifi 0000:00:14.3: Detected Intel(R) Wi-Fi 6E AX211 160MHz, REV=0x370
[ 11.824564] iwlwifi 0000:00:14.3: retry init count 0
[ 11.833882] iwlwifi 0000:00:14.3: Detected Intel(R) Wi-Fi 6E AX211 160MHz, REV=0x370
[ 12.034157] iwlwifi 0000:00:14.3: loaded PNVM version 0x881c99e1
[ 12.049256] iwlwifi 0000:00:14.3: Detected RF GF, rfid=0x2010d000
[ 12.119618] iwlwifi 0000:00:14.3: base HW address: 60:dd:8e:a5:ed:7b
[ 16.687137] iwlwifi 0000:00:14.3 wlp0s20f3: renamed from wlan0
[manjaro-gnome ~]# uname -a
Linux manjaro-gnome 5.15.60-1-MANJARO #1 SMP PREEMPT Thu Aug 11 13:14:05 UTC 2022 x86_64 GNU/Linux
[manjaro-gnome ~]# cat /etc/os-release
NAME="Manjaro Linux"
ID=manjaro
ID_LIKE=arch
BUILD_ID=rolling
PRETTY_NAME="Manjaro Linux"
ANSI_COLOR="32;1;24;144;200"
HOME_URL="https://manjaro.org/"
DOCUMENTATION_URL="https://wiki.manjaro.org/"
SUPPORT_URL="https://manjaro.org/"
BUG_REPORT_URL="https://bugs.manjaro.org/"
LOGO=manjarolinux
[manjaro-gnome ~]# journalctl | head
Apr 16 20:38:25 manjaro-gnome kernel: Linux version 5.15.60-1-MANJARO (builduser@fv-az462-678) (gcc (GCC) 12.1.1 20220730, GNU ld (GNU Binutils) 2.39) #1 SMP PREEMPT Thu Aug 11 13:14:05 UTC 2022
Apr 16 20:38:25 manjaro-gnome kernel: Command line: BOOT_IMAGE=/boot/vmlinuz-x86_64 lang=en_US keytable=us tz=UTC misobasedir=manjaro misolabel=MANJARO_GNOME_2137 quiet systemd.show_status=1 driver=free nouveau.modeset=1 i915.modeset=1 radeon.modeset=1
Apr 16 20:38:25 manjaro-gnome kernel: x86/split lock detection: #AC: crashing the kernel on kernel split_locks and warning on user-space split_locks
Apr 16 20:38:25 manjaro-gnome kernel: x86/fpu: Supporting XSAVE feature 0x001: 'x87 floating point registers'
Apr 16 20:38:25 manjaro-gnome kernel: x86/fpu: Supporting XSAVE feature 0x002: 'SSE registers'
Apr 16 20:38:25 manjaro-gnome kernel: x86/fpu: Supporting XSAVE feature 0x004: 'AVX registers'
Apr 16 20:38:25 manjaro-gnome kernel: x86/fpu: Supporting XSAVE feature 0x200: 'Protection Keys User registers'
Apr 16 20:38:25 manjaro-gnome kernel: x86/fpu: xstate_offset[2]: 576, xstate_sizes[2]: 256
Apr 16 20:38:25 manjaro-gnome kernel: x86/fpu: xstate_offset[9]: 832, xstate_sizes[9]: 8
Apr 16 20:38:25 manjaro-gnome kernel: x86/fpu: Enabled xstate features 0x207, context size is 840 bytes, using 'compacted' format.
If i try latest available Manjaro iso version (22.0.5) doesnât work, and shows the same trace error like Fedora.
IMHO, there must be some problem with the new kernels starting with the version > 5.15
Regards
gnwiii
(George N. White III)
April 16, 2023, 9:24pm
47
Manjaro loaded so-a0-gf-a0-**64**.ucode
, but that isnât available for more recent kernels. Intel is responsible for the drivers and microcode. I imagine the business case for porting an old driver to recent kernels is not great (as failing to update the driver means users buy fewer new cards), and there can be issues (kept private) related to security that can only be mitigated using newer hardware.
That is unlikely. Although the perception of âworkingâ does depend on the use case there may be issues that arenât apparent until bad things happen. New wifi cards are not expensive, and all but the thinnest and lightest systems support swapping wifi cards.
mutenroid
(Alberto Moro)
April 16, 2023, 10:03pm
48
But this error, seems to be a problem with NIC initialization code. Firmware its not involved along boot, because hw initialization failed at all.
Makeing search about this string, we can see the loop and timeout configuration for harware ready consideration:
/* Note: returns standard 0/-ERROR code */
int iwl_pcie_prepare_card_hw(struct iwl_trans *trans)
{
int ret;
int iter;
IWL_DEBUG_INFO(trans, "iwl_trans_prepare_card_hw enter\n");
ret = iwl_pcie_set_hw_ready(trans);
/* If the card is ready, exit 0 */
if (ret >= 0) {
trans->csme_own = false;
return 0;
}
iwl_set_bit(trans, CSR_DBG_LINK_PWR_MGMT_REG,
CSR_RESET_LINK_PWR_MGMT_DISABLED);
usleep_range(1000, 2000);
This file has been truncated. show original
mutenroid
(Alberto Moro)
April 16, 2023, 10:08pm
49
Sorry, but I canât believe that such modern equipment (Thinkpad T16 reach production 1 year ago) with such a modern hw (full Intel 12th Gen chipset & Intel(R) Wi-Fi 6E AX211) is not well supported.
Itâs very dissapointment
Regards
mutenroid
(Alberto Moro)
May 10, 2023, 10:40am
50
i have found a solution to the problem.
The tip is as follows:
Enter BIOS and go to Config section
Disable Intelâ(R) AMT (enabled by default) and select disabled (not permanent disabled)
Save and restart
After a loooooong postâŚVoila!..wifi it works like a charm.
⯠dmesg | grep iwlwifi
[ 11.100435] iwlwifi 0000:00:14.3: enabling device (0000 â 0002)
[ 11.184234] iwlwifi 0000:00:14.3: api flags index 2 larger than supported by driver
[ 11.184253] iwlwifi 0000:00:14.3: TLV_FW_FSEQ_VERSION: FSEQ Version: 0.0.2.36
[ 11.184606] iwlwifi 0000:00:14.3: loaded firmware version 72.daa05125.0 so-a0-gf-a0-72.ucode op_mode iwlmvm
[ 11.360757] iwlwifi 0000:00:14.3: Detected Intel(R) Wi-Fi 6E AX211 160MHz, REV=0x370
[ 11.530476] iwlwifi 0000:00:14.3: WFPM_UMAC_PD_NOTIFICATION: 0x1f
[ 11.530485] iwlwifi 0000:00:14.3: WFPM_LMAC2_PD_NOTIFICATION: 0x1f
[ 11.530493] iwlwifi 0000:00:14.3: WFPM_AUTH_KEY_0: 0x80
[ 11.530502] iwlwifi 0000:00:14.3: CNVI_SCU_SEQ_DATA_DW9: 0x0
[ 11.532034] iwlwifi 0000:00:14.3: loaded PNVM version 64acdc51
[ 11.546821] iwlwifi 0000:00:14.3: Detected RF GF, rfid=0x2010d000
[ 11.617017] iwlwifi 0000:00:14.3: base HW address: 60:dd:8e:a5:ed:7b
[ 11.634140] iwlwifi 0000:00:14.3 wlp0s20f3: renamed from wlan0
[ 13.326578] iwlwifi 0000:00:14.3: WFPM_UMAC_PD_NOTIFICATION: 0x1f
[ 13.326588] iwlwifi 0000:00:14.3: WFPM_LMAC2_PD_NOTIFICATION: 0x1f
[ 13.326596] iwlwifi 0000:00:14.3: WFPM_AUTH_KEY_0: 0x80
[ 13.326604] iwlwifi 0000:00:14.3: CNVI_SCU_SEQ_DATA_DW9: 0x0
[ 13.607888] iwlwifi 0000:00:14.3: WFPM_UMAC_PD_NOTIFICATION: 0x1f
[ 13.607898] iwlwifi 0000:00:14.3: WFPM_LMAC2_PD_NOTIFICATION: 0x1f
[ 13.607906] iwlwifi 0000:00:14.3: WFPM_AUTH_KEY_0: 0x80
[ 13.607914] iwlwifi 0000:00:14.3: CNVI_SCU_SEQ_DATA_DW9: 0x0
Switch to enabled again Intel AMT option, and wifi it still works flawlessly.
gnwiii
(George N. White III)
May 10, 2023, 11:42am
51
Most systems are never completely off, and many systems aimed at large enterprises, student labs, call centers, etc. support network booting over wifi, so it is conceivable that old firmware hangs around thru linux upgrades. BIOS control for such things is hit-or-miss. I wonder how many people swapped laptop network cards when they could have just disconnected the battery to force a wifi reset.