Kernel KVM virtualization development
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Steffen Eiden" <seiden@linux.ibm.com>
Cc: kvmarm@lists.linux.dev, linux-s390@vger.kernel.org,
	"Vasily Gorbik" <gor@linux.ibm.com>,
	"Marc Zyngier" <maz@kernel.org>,
	kvm@vger.kernel.org, "Alexander Gordeev" <agordeev@linux.ibm.com>,
	"Heiko Carstens" <hca@linux.ibm.com>,
	"Oliver Upton" <oupton@kernel.org>
Subject: Re: [PATCH v4 01/27] VFIO: take reference to the KVM module
Date: Mon, 06 Jul 2026 09:12:23 +0000	[thread overview]
Message-ID: <20260706091223.E32F31F000E9@smtp.kernel.org> (raw)
In-Reply-To: <20260706085229.979525-2-seiden@linux.ibm.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] In the VFIO cdev binding path, concurrent or repeated calls to `VFIO_DEVICE_BIND_IOMMUFD` on multiple file descriptors for the same device cause KVM reference leaks and state corruption.
--

commit 298282ceb4d6390cae8f782bcdd7f84408cf4d30
Author: Paolo Bonzini <pbonzini@redhat.com>

VFIO: take reference to the KVM module

This patch modifies VFIO to store a pointer to the KVM module alongside the
KVM pointer. It updates the get/set paths to take and release a module
reference to prevent the KVM module from unloading while in use, in
preparation for removing symbol_get and symbol_put.

> diff --git a/drivers/vfio/vfio_main.c b/drivers/vfio/vfio_main.c
> index ed538aebb0b8e..8289aca97b6be 100644
> --- a/drivers/vfio/vfio_main.c
> +++ b/drivers/vfio/vfio_main.c
[ ... ]
> @@ -448,7 +448,8 @@ void vfio_unregister_group_dev(struct vfio_device *device)
>  EXPORT_SYMBOL_GPL(vfio_unregister_group_dev);
>  
>  #if IS_ENABLED(CONFIG_KVM)
> -void vfio_device_get_kvm_safe(struct vfio_device *device, struct kvm *kvm)
> +void vfio_device_get_kvm_safe(struct vfio_device *device, struct kvm *kvm,
> +			      struct module *kvm_module)
>  {
>  	void (*pfn)(struct kvm *kvm);
>  	bool (*fn)(struct kvm *kvm);
> @@ -459,25 +460,31 @@ void vfio_device_get_kvm_safe(struct vfio_device *device, struct kvm *kvm)
>  	if (!kvm)
>  		return;
>  
> +	if (!try_module_get(kvm_module))
> +		return;
> +
>  	pfn = symbol_get(kvm_put_kvm);
>  	if (WARN_ON(!pfn))
> -		return;
> +		goto out_put_mod;
>  
>  	fn = symbol_get(kvm_get_kvm_safe);
> -	if (WARN_ON(!fn)) {
> -		symbol_put(kvm_put_kvm);
> -		return;
> -	}
> +	if (WARN_ON(!fn))
> +		goto out_put_sym;
>  
>  	ret = fn(kvm);
>  	symbol_put(kvm_get_kvm_safe);
> -	if (!ret) {
> -		symbol_put(kvm_put_kvm);
> -		return;
> -	}
> +	if (!ret)
> +		goto out_put_sym;
>  
>  	device->put_kvm = pfn;
>  	device->kvm = kvm;
> +	device->kvm_module = kvm_module;
> +	return;

[Severity: High]
This is a pre-existing issue, but can concurrent or repeated calls to
VFIO_DEVICE_BIND_IOMMUFD on multiple file descriptors for the same device
cause KVM reference leaks and state corruption here?

In vfio_device_get_kvm_safe() in drivers/vfio/vfio_main.c, the assignments
above blindly overwrite device->kvm (and now device->kvm_module) without
dropping any existing references.

If a user opens the VFIO device node twice (fd1 and fd2) and binds fd1,
device->kvm is set and the reference is taken. If they then call
VFIO_DEVICE_BIND_IOMMUFD on fd2, this function overwrites the pointers,
leaking the initial KVM and module references.

Furthermore, when vfio_df_open() subsequently fails in the fd2 bind path
because device->open_count > 0, the error path in vfio_df_ioctl_bind_iommufd()
unconditionally calls vfio_device_put_kvm(device). This drops the newly
overwritten reference and clears device->kvm to NULL, entirely breaking the
KVM integration for the still-active fd1.

> +
> +out_put_sym:
> +	symbol_put(kvm_put_kvm);
> +out_put_mod:
> +	module_put(kvm_module);
>  }
>  
>  void vfio_device_put_kvm(struct vfio_device *device)

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260706085229.979525-1-seiden@linux.ibm.com?part=1

  reply	other threads:[~2026-07-06  9:12 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-07-06  8:52 [PATCH v4 00/27] KVM: s390: Introduce arm64 KVM Steffen Eiden
2026-07-06  8:52 ` [PATCH v4 01/27] VFIO: take reference to the KVM module Steffen Eiden
2026-07-06  9:12   ` sashiko-bot [this message]
2026-07-06 19:53   ` Jason Gunthorpe
2026-07-06 21:12     ` Sean Christopherson
2026-07-06  8:52 ` [PATCH v4 02/27] KVM, vfio: remove symbol_get(kvm_get_kvm_safe) from vfio Steffen Eiden
2026-07-06  8:52 ` [PATCH v4 03/27] KVM, vfio: remove symbol_get(kvm_put_kvm) " Steffen Eiden
2026-07-06  9:25   ` sashiko-bot
2026-07-06  8:52 ` [PATCH v4 04/27] KVM: Make device name configurable Steffen Eiden
2026-07-06  9:14   ` sashiko-bot
2026-07-06  8:52 ` [PATCH v4 05/27] KVM: Remove KVM_MMIO as config option Steffen Eiden
2026-07-06 20:53   ` Marc Zyngier
2026-07-06  8:52 ` [PATCH v4 06/27] arm64: Use proper include variant Steffen Eiden
2026-07-06  8:52 ` [PATCH v4 07/27] arm64: Prepare sharing arm64 headers with s390 Steffen Eiden
2026-07-06 21:22   ` Marc Zyngier
2026-07-06  8:52 ` [PATCH v4 08/27] KVM: arm64: Prepare sharing arm64 code " Steffen Eiden
2026-07-06  9:21   ` sashiko-bot
2026-07-06 21:28   ` Marc Zyngier
2026-07-06  8:52 ` [PATCH v4 09/27] KVM: arm64: Access elements of vcpu_gp_regs individually Steffen Eiden
2026-07-06 17:44   ` Marc Zyngier
2026-07-06  8:52 ` [PATCH v4 10/27] KVM: arm64: Refactor core-reset into a separate function Steffen Eiden
2026-07-06  8:52 ` [PATCH v4 11/27] s390: Use arm64 headers Steffen Eiden
2026-07-06  9:31   ` sashiko-bot
2026-07-06  8:52 ` [PATCH v4 12/27] KVM: s390: Use arm64 code Steffen Eiden
2026-07-06  9:26   ` sashiko-bot
2026-07-06  8:52 ` [PATCH v4 13/27] KVM: s390: Prepare KVM/s390 for a second KVM module Steffen Eiden
2026-07-06  8:52 ` [PATCH v4 14/27] KVM: s390: Move s390 kvm code into a subdirectory Steffen Eiden
2026-07-06  9:35   ` sashiko-bot
2026-07-06  8:52 ` [PATCH v4 15/27] KVM: S390: Prepare gmap for a second KVM implementation Steffen Eiden
2026-07-06  9:36   ` sashiko-bot
2026-07-06  8:52 ` [PATCH v4 16/27] KVM: s390: gmap: Refactor storage key and CMMA code into separate files Steffen Eiden
2026-07-06  8:52 ` [PATCH v4 17/27] KVM: s390: Refactor prefix handling into a separate file Steffen Eiden
2026-07-06  8:52 ` [PATCH v4 18/27] KVM: s390: Prepare kvm-s390 for a second kvm module Steffen Eiden
2026-07-06  8:52 ` [PATCH v4 19/27] s390: Introduce Start Arm Execution instruction Steffen Eiden
2026-07-06 11:17   ` Janosch Frank
2026-07-06 11:30     ` Christian Borntraeger
2026-07-06 11:58       ` Steffen Eiden
2026-07-06 11:53     ` Steffen Eiden
2026-07-06  8:52 ` [PATCH v4 20/27] KVM: s390: arm64: Introduce host definitions Steffen Eiden
2026-07-06  8:52 ` [PATCH v4 21/27] s390/hwcaps: Report SAE support as hwcap Steffen Eiden
2026-07-06 11:50   ` Janosch Frank
2026-07-06  8:52 ` [PATCH v4 22/27] KVM: s390: Add basic arm64 kvm module Steffen Eiden
2026-07-06  9:51   ` sashiko-bot
2026-07-06  8:52 ` [PATCH v4 23/27] KVM: s390: arm64: Implement required functions Steffen Eiden
2026-07-06  9:55   ` sashiko-bot
2026-07-06 12:15   ` Janosch Frank
2026-07-06 20:35   ` Marc Zyngier
2026-07-06  8:52 ` [PATCH v4 24/27] KVM: s390: arm64: Implement vm/vcpu create destroy Steffen Eiden
2026-07-06 10:04   ` sashiko-bot
2026-07-06  8:52 ` [PATCH v4 25/27] KVM: s390: arm64: Implement vCPU IOCTLs Steffen Eiden
2026-07-06 10:16   ` sashiko-bot
2026-07-06  8:52 ` [PATCH v4 26/27] KVM: s390: arm64: Implement basic page fault handler Steffen Eiden
2026-07-06 10:03   ` sashiko-bot
2026-07-06  8:52 ` [PATCH v4 27/27] KVM: s390: arm64: Enable KVM_ARM64 config and Kbuild Steffen Eiden
2026-07-06 10:07   ` sashiko-bot

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260706091223.E32F31F000E9@smtp.kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=agordeev@linux.ibm.com \
    --cc=gor@linux.ibm.com \
    --cc=hca@linux.ibm.com \
    --cc=kvm@vger.kernel.org \
    --cc=kvmarm@lists.linux.dev \
    --cc=linux-s390@vger.kernel.org \
    --cc=maz@kernel.org \
    --cc=oupton@kernel.org \
    --cc=sashiko-reviews@lists.linux.dev \
    --cc=seiden@linux.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox