Re: [PATCH] vsock: reject negative tv_usec in SO_VM_SOCKETS_CONNECT_TIMEOUT

[Bobby Eshleman <bobbyeshleman@meta.com>]

On Fri, Oct 30, 2026 at 02:43:50PM +0800, phind.uet@gmail.com wrote:
> From: Nguyen Dinh Phi <phind.uet@gmail.com>
> 
> The timeval validation checks tv_usec < USEC_PER_SEC but not
> tv_usec >= 0, so a negative tv_usec makes the timeout computation
> wrap into a bogus value, later tripping "schedule_timeout: wrong
> timeout value" instead of being rejected.
> 
> Fixes: d021c344051a ("VSOCK: Introduce VM Sockets")
> 
> Signed-off-by: Nguyen Dinh Phi <phind.uet@gmail.com>
> ---
>  net/vmw_vsock/af_vsock.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
> index d0508887a2fb..1d7d53542724 100644
> --- a/net/vmw_vsock/af_vsock.c
> +++ b/net/vmw_vsock/af_vsock.c
> @@ -2156,7 +2156,8 @@ static int vsock_connectible_setsockopt(struct socket *sock,
>  					     optname == SO_VM_SOCKETS_CONNECT_TIMEOUT_OLD);
>  		if (err)
>  			break;
> -		if (tv.tv_sec >= 0 && tv.tv_usec < USEC_PER_SEC &&
> +		if (tv.tv_sec >= 0 && tv.tv_usec >= 0 &&
> +		    tv.tv_usec < USEC_PER_SEC &&
>  		    tv.tv_sec < (MAX_SCHEDULE_TIMEOUT / HZ - 1)) {
>  			vsk->connect_timeout = tv.tv_sec * HZ +
>  				DIV_ROUND_UP((unsigned long)tv.tv_usec, (USEC_PER_SEC / HZ));
> -- 
> 2.43.0
> 

LGTM. Maybe we could think about adding some easy opt range tests into
vsock_test in the future.

Reviewed-by: Bobby Eshleman <bobbyeshleman@meta.com>