linux-kernel - Re: [PATCH v5 6/6] vhost_scsi: Add check for inherit

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 for Android: free password hash cracker in your pocket

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

Message-ID: <cb7df64b-7a0a-47c6-9b62-76890edd32bf@oracle.com>
Date: Wed, 22 Jan 2025 14:18:57 -0600
From: Mike Christie <michael.christie@...cle.com>
To: Cindy Lu <lulu@...hat.com>, jasowang@...hat.com, mst@...hat.com,
        sgarzare@...hat.com, linux-kernel@...r.kernel.org,
        virtualization@...ts.linux-foundation.org, netdev@...r.kernel.org
Subject: Re: [PATCH v5 6/6] vhost_scsi: Add check for inherit_owner status

On 12/30/24 6:43 AM, Cindy Lu wrote:
> The vhost_scsi VHOST_NEW_WORKER requires the inherit_owner
> setting to be true. So we need to implement a check for this.
> 
> Signed-off-by: Cindy Lu <lulu@...hat.com>
> ---
>  drivers/vhost/scsi.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
> index 718fa4e0b31e..0d63b6b5c852 100644
> --- a/drivers/vhost/scsi.c
> +++ b/drivers/vhost/scsi.c
> @@ -2086,6 +2086,14 @@ vhost_scsi_ioctl(struct file *f,
>  			return -EFAULT;
>  		return vhost_scsi_set_features(vs, features);
>  	case VHOST_NEW_WORKER:
> +		/*
> +		 * vhost_tasks will account for worker threads under the parent's
> +		 * NPROC value but kthreads do not. To avoid userspace overflowing
> +		 * the system with worker threads inherit_owner must be true.
> +		 */
> +		if (!vs->dev.inherit_owner)> +			return -EFAULT;

Why did you keep this here? I had mentioned it belonged in the common code:

https://lore.kernel.org/virtualization/3864ae3b-d6cd-4227-b4bb-56e014c71667@oracle.com/T/#u

because the limitation applies to all drivers. I didn't see a reply about it
but saw you did fix up the comment. Not sure if I missed it or there was
a technical issue you hit.


> +		fallthrough;
>  	case VHOST_FREE_WORKER:
>  	case VHOST_ATTACH_VRING_WORKER:
>  	case VHOST_GET_VRING_WORKER: