Re: [PATCH] vt: Fix a missing-check bug in drivers/tty/vt/vt.c file of Linux 5.0.14

[Greg KH <gregkh@linuxfoundation.org>]

On Fri, May 10, 2019 at 10:24:50PM +0800, Gen Zhang wrote:
> On Fri, May 10, 2019 at 13:14:02PM +0800, Greg KH <
> gregkh@linuxfoundation.org> wrote:
> >Note, your email client ate all of the tabs and made the patch
> >impossible to apply, so please fix this up before you resend it.
> >
> >thanks,
> >
> >greg k-h
> From: Gen Zhang <blackgod016574@gmail.com>
> Date: Fri, 10 May 2019 09:31:30 +0000
> Subject: [PATCH] vt: Fix a missing-check bug in drivers/tty/vt/vt.c file of
> Linux 5.0.14
> 
> Hi,
> I found this missing-check bug in Linux-5.0.14/drivers/tty/vt/vt.c when I
> was examining the source code.
> 
> In function con_init(), the pointer variable vc_cons[currcons].d, vc and
> vc->vc_screenbuf is allocated a memory space via kzalloc().
> And they are used in the following codes.
> 
> However, when there is a memory allocation error, kzalloc can  be failed.
> Thus null pointer (vc_cons[currcons].d, vc and vc->vc_screenbuf)
> dereference may happen.
> And it will cause the kernel to crash. Therefore, we should check return
> value and handle an error.
> 
> Below is the patch file, and I am ready to sumbit it to the kernel tree.
> I am looking forward to a reply on this, thank you!
> 
> Kind regards
> Gen
> 
> Signed-off-by: Gen Zhang <blackgod016574@gmail.com>
> ---
> 
> --- a/drivers/tty/vt/vt.c
> +++ b/drivers/tty/vt/vt.c
> @@ -3322,10 +3322,14 @@ static int __init con_init(void)
> 
>   for (currcons = 0; currcons < MIN_NR_CONSOLES; currcons++) {
>   vc_cons[currcons].d = vc = kzalloc(sizeof(struct vc_data), GFP_NOWAIT);
> + if (!vc_cons[currcons].d || !vc)
> + goto err_vc;
>   INIT_WORK(&vc_cons[currcons].SAK_work, vc_SAK);
>   tty_port_init(&vc->port);
>   visual_init(vc, currcons, 1);
>   vc->vc_screenbuf = kzalloc(vc->vc_screenbuf_size, GFP_NOWAIT);
> + if (!vc->vc_screenbuf)
> + goto err_vc_screenbuf;
>   vc_init(vc, vc->vc_rows, vc->vc_cols,
>   currcons || !vc->vc_sw->con_save_screen);
>   }
> @@ -3347,6 +3351,14 @@ static int __init con_init(void)
>   register_console(&vt_console_driver);
>  #endif
>   return 0;
> +err_vc:
> + console_unlock();
> + return -ENOMEM;
> +err_vc_screenbuf:
> + console_unlock();
> + kfree(vc);
> + vc_cons[currcons].d = NULL;
> + return -ENOMEM;
>  }
>  console_initcall(con_init);

Still impossible to apply :(

Also, what about Dave's response to you?  This really can never be hit,
like other early-init tty allocations that we do not check because of
this issue, correct?

thanks,

greg k-h