It’s an ar750. I was watching YouTube and noticed something so i checked the log. I really can’t remember what made me check. Is there a way to export the log to a file?
Edit: i think i remember wlan-sta being deauthenticated before this. I also found this same problem here:
Driver problem maybe?
Edit 2:
I also notice in the kernel a lot of errors like this:
35.382241] firmware ath10k!pre-cal-pci-0000:00:00.0.bin: firmware_loading_store: map pages failed
[ 35.411410] ath10k_pci 0000:00:00.0: qca9887 hw1.0 target 0x4100016d chip_id 0x004000ff sub 0000:0000
[ 35.420986] ath10k_pci 0000:00:00.0: kconfig debug 0 debugfs 1 tracing 0 dfs 1 testmode 1
[ 35.434886] ath10k_pci 0000:00:00.0: firmware ver 10.2.4-1.0-00013 api 5 features no-p2p,ignore-otp,skip-clock-init,mfp crc32 14a57ac6
[ 35.447514] ath10k_pci 0000:00:00.0: Direct firmware load for ath10k/QCA9887/hw1.0/board-2.bin failed with error -2
[ 35.458322] ath10k_pci 0000:00:00.0: Falling back to user helper
[ 35.632503] firmware ath10k!QCA9887!hw1.0!board-2.bin: firmware_loading_store: map pages failed
[ 35.653574] ath10k_pci 0000:00:00.0: board_file api 1 bmi_id N/A crc32 546cca0d
[ 36.729878] ath10k_pci 0000:00:00.0: htt-ver 2.1 wmi-op 5 htt-op 2 cal file max-sta 128 raw 0 hwcrypto 1
[ 36.842739] ath: EEPROM regdomain: 0x0
[ 36.842757] ath: EEPROM indicates default country code should be used
[ 36.842766] ath: doing EEPROM country->regdmn map search
[ 36.842785] ath: country maps to regdmn code: 0x3a
[ 36.842795] ath: Country alpha2 being used: US
[ 36.842803] ath: Regpair used: 0x3a
This happens several times during boot, but eventually stops. I’m curious about something tho. Why use the ath10k driver? Isn’t it an ath9k board?
Sounds like you mau have run out of (contiguous?) kernel memory when trying to reload the firmware into the Atheros WiFI chip (presumably as part of recovery from a firmware error or device crash); how long had the GL750 been running, and about how much traffic do you think had you passed at that point?
This is actually during the boot of the ar750, from the time stamp I’d say 35 seconds in. I’m attaching a complete kernel log from last night to look at, it’s from a fresh boot, you’ll see what’s confusing i think.kernel.log.zip (10.8 KB)
OK, I see what’s going on here. These two errors are benign:
ath10k_pci 0000:00:00.0: Direct firmware load for ath10k/pre-cal-pci-0000:00:00.0.bin failed with error -2
ath10k_pci 0000:00:00.0: Direct firmware load for ath10k/QCA9887/hw1.0/board-2.bin failed with error -2
The issue seems to be here:
[ 7306.844396] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 83 (0x53)
[ 7307.210539] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 115 (0x73)
[ 7307.577946] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 88 (0x58)
[ 7642.796977] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 109 (0x6d)
[ 7643.753010] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 90 (0x5a)
[ 7644.450611] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 90 (0x5a)
[ 8246.046588] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 103 (0x67)
[ 9454.161365] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 114 (0x72)
[ 9455.222623] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 99 (0x63)
[ 9455.589936] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 116 (0x74)
[ 9755.030726] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 123 (0x7b)
[10356.148302] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 80 (0x50)
[10657.940411] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 81 (0x51)
[10658.308291] Rate marked as an HT rate but passed status->rate_idx is not an MCS index [0-76]: 94 (0x5e)
It’s been a few years since I’ve played with WiFi drivers, but it appears we’re using the “wrong” rate for that radio when scanning for SSIDs? @alzhao will need to chime in here, but perhaps this webpage will help: http://mcsindex.com/
I’m not sure if it’s a problem either. Over the last week I’ve had a few instances of going to bed watching YouTube or Netflix, and waking up in the morning with no connection and unable to connect without a reboot (unable to check logs) I’m trying to track down the issue. I was still on this morning, so checked the logs and this is what i saw. Don’t know if it’s related or not to my actual problem tho.
Yeah, I don’t think these kernel WARN_ON()s (which are up to the discretion of the driver writer, and sometimes are “noisier” than the issue they’re reporting would seem to justify) are reflective of a symptom of that issue, either.