linux-kernel - Re: [BUG 4.15-rc7] IRQ matrix management errors

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 16 Jan 2018 11:33:10 +0100 (CET)
From:   Thomas Gleixner <tglx@...utronix.de>
To:     Keith Busch <keith.busch@...el.com>
cc:     LKML <linux-kernel@...r.kernel.org>
Subject: Re: [BUG 4.15-rc7] IRQ matrix management errors

On Tue, 16 Jan 2018, Keith Busch wrote:

> This is all way over my head, but the part that obviously shows
> something's gone wrong:
> 
>   kworker/u674:3-1421  [028] d...   335.307051: irq_matrix_reserve_managed: bit=56 cpu=0 online=1 avl=86 alloc=116 managed=3 online_maps=112 global_avl=22084, global_rsvd=157, total_alloc=570
>   kworker/u674:3-1421  [028] d...   335.307053: irq_matrix_remove_managed: bit=56 cpu=0 online=1 avl=87 alloc=116 managed=2 online_maps=112 global_avl=22085, global_rsvd=157, total_alloc=570
>   kworker/u674:3-1421  [028] ....   335.307054: vector_reserve_managed: irq=45 ret=-28
>   kworker/u674:3-1421  [028] ....   335.307054: vector_setup: irq=45 is_legacy=0 ret=-28
>   kworker/u674:3-1421  [028] d...   335.307055: vector_teardown: irq=45 is_managed=1 has_reserved=0
> 
> Which leads me to x86_vector_alloc_irqs goto error:
> 
> error:
> 	x86_vector_free_irqs(domain, virq, i + 1);
> 
> The last parameter looks weird. It's the nr_irqs, and since we failed and
> bailed, I would think we'd need to subtract 1 rather than add 1. Adding
> 1 would doublely remove the failed one, and remove the next one that
> was never setup, right?

Right. That's fishy. Let me stare at it.

> Or maybe irq_matrix_reserve_managed wasn't expected to fail in the
> first place?

Well, it can faul. I don't know why it fails in that case, but let me look
a bit more.

Thanks,

	tglx