Unhandled IRQ at boot or when unloading mt7630e
The issue https://github.com/kuba-moo/mt7630e/issues/19 seems to affect this variant of the driver as well.
Sometimes the following error is reported to the system log at boot time or when I try to unload mt7630e:
irq 16: nobody cared (try booting with the "irqpoll" option)
CPU: 0 PID: 0 Comm: swapper/0 Tainted: G I 4.1.15-nrj-desktop-1rosa-x86_64 #1
Hardware name: System manufacturer System Product Name/Rampage III GENE, BIOS 0704 09/21/2010
ffff8801b5ecfa8c ffff8801b9203e28 ffffffff8175549d 0000000000009898
ffff8801b5ecfa00 ffff8801b9203e58 ffffffff810d0ca6 ffff8801b92108c0
ffff8801b5ecfa00 0000000000000000 0000000000000010 ffff8801b9203e98
Call Trace:
<IRQ> [<ffffffff8175549d>] dump_stack+0x4c/0x6e
[<ffffffff810d0ca6>] __report_bad_irq+0x36/0xd0
[<ffffffff810d1047>] note_interrupt+0x247/0x290
[<ffffffff810ce62e>] handle_irq_event_percpu+0xae/0x1f0
[<ffffffff810ce7b1>] handle_irq_event+0x41/0x70
[<ffffffff810d1536>] handle_fasteoi_irq+0x86/0x140
[<ffffffff81018782>] handle_irq+0x22/0x40
[<ffffffff81017fcf>] do_IRQ+0x4f/0xf0
[<ffffffff8175d2ee>] common_interrupt+0x6e/0x6e
<EOI> [<ffffffff8164db0e>] ? cpuidle_enter_state+0xae/0x240
[<ffffffff8164dae0>] ? cpuidle_enter_state+0x80/0x240
[<ffffffff8164dcd7>] cpuidle_enter+0x17/0x20
[<ffffffff810b812e>] cpu_startup_entry+0x3ae/0x430
[<ffffffff817481ea>] rest_init+0x8a/0x90
[<ffffffff81d4df78>] start_kernel+0x434/0x441
[<ffffffff81d4d120>] ? early_idt_handler_array+0x120/0x120
[<ffffffff81d4d4d7>] x86_64_start_reservations+0x2a/0x2c
[<ffffffff81d4d614>] x86_64_start_kernel+0x13b/0x14a
handlers:
[<ffffffffc01f5cc0>] irq_handler [firewire_ohci]
[<ffffffff815d48d0>] usb_hcd_irq
Disabling IRQ #16
When it happens at boot, the system hangs and becomes unusable until reboot (even no reaction to the magic sysrq).
If I blacklist the driver at boot (modprobe.blacklist=mt7630e), the problem have not shown up so far.
OS: ROSA Linux R6 x86_64 Kernel: 4.1.15 WiFi adapter: pci:14c3-7630-103c-197c (MT7630e 802.11bgn Wireless Network Adapter)
More details and the logs: http://hw.rosalinux.ru/index.php?probe=050ffb018c I had to reboot the system before collecting them, so the full system log of the previous problematic boot is http://hw.rosalinux.ru/index.php?probe=050ffb018c&log=journalctl.1
Any ideas?