linux-kernel - Re: kvm: arm64: stable commit "Fix kernel BUG() due to bad backport of FPSIMD/SVE/SME fix" deadlocks host kernel

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]

Message-ID: <aOYkuatjNVyiQzU1@willie-the-truck>
Date: Wed, 8 Oct 2025 09:45:45 +0100
From: Will Deacon <will@...nel.org>
To: Sergey Senozhatsky <senozhatsky@...omium.org>
Cc: Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	Catalin Marinas <catalin.marinas@....com>,
	Mark Brown <broonie@...nel.org>,
	Mark Rutland <mark.rutland@....com>,
	Tomasz Figa <tfiga@...omium.org>,
	linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
	stable@...r.kernel.org
Subject: Re: kvm: arm64: stable commit "Fix kernel BUG() due to bad backport
 of FPSIMD/SVE/SME fix" deadlocks host kernel

On Wed, Oct 08, 2025 at 01:45:32PM +0900, Sergey Senozhatsky wrote:
> Commits 8f4dc4e54eed4 (6.1.y) and 23249dade24e6 (5.15.y) (maybe other
> stable kernels as well) deadlock the host kernel (presumably a
> recursive spinlock):
> 
>  queued_spin_lock_slowpath+0x274/0x358
>  raw_spin_rq_lock_nested+0x2c/0x48
>  _raw_spin_rq_lock_irqsave+0x30/0x4c
>  run_rebalance_domains+0x808/0x2e18
>  __do_softirq+0x104/0x550
>  irq_exit+0x88/0xe0
>  handle_domain_irq+0x7c/0xb0
>  gic_handle_irq+0x1cc/0x420
>  call_on_irq_stack+0x20/0x48
>  do_interrupt_handler+0x3c/0x50
>  el1_interrupt+0x30/0x58
>  el1h_64_irq_handler+0x18/0x24
>  el1h_64_irq+0x7c/0x80
>  kvm_arch_vcpu_ioctl_run+0x24c/0x49c
>  kvm_vcpu_ioctl+0xc4/0x614
> 
> We found out a similar report at [1], but it doesn't seem like a formal
> patch was ever posted.  Will, can you please send a formal patch so that
> stable kernels can run VMs again?

Yup, already queued up for stable:

https://lore.kernel.org/r/20251003183917.4209-1-will@kernel.org
https://lore.kernel.org/r/20251003184018.4264-1-will@kernel.org
https://lore.kernel.org/r/20251003184054.4286-1-will@kernel.org

Sorry for the breakage.

Will