linux-kernel - Re: [PATCH] usb: xhci: Check kcalloc_node() when allocating interrupter array in xhci_mem

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: <20251103094036.2d1593bc.michal.pecio@gmail.com>
Date: Mon, 3 Nov 2025 09:40:36 +0100
From: Michal Pecio <michal.pecio@...il.com>
To: Mathias Nyman <mathias.nyman@...el.com>, Greg Kroah-Hartman
 <gregkh@...uxfoundation.org>
Cc: Guangshuo Li <lgs201920130244@...il.com>, Wesley Cheng
 <quic_wcheng@...cinc.com>, linux-usb@...r.kernel.org,
 linux-kernel@...r.kernel.org, stable@...r.kernel.org
Subject: Re: [PATCH] usb: xhci: Check kcalloc_node() when allocating
 interrupter array in xhci_mem_init()

On Thu, 18 Sep 2025 21:08:38 +0800, Guangshuo Li wrote:
> kcalloc_node() may fail. When the interrupter array allocation returns
> NULL, subsequent code uses xhci->interrupters (e.g. in xhci_add_interrupter()
> and in cleanup paths), leading to a potential NULL pointer dereference.
> 
> Check the allocation and bail out to the existing fail path to avoid
> the NULL dereference.
> 
> Fixes: c99b38c412343 ("xhci: add support to allocate several interrupters")
> Cc: stable@...r.kernel.org
> Signed-off-by: Guangshuo Li <lgs201920130244@...il.com>
> ---
>  drivers/usb/host/xhci-mem.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c
> index d698095fc88d..da257856e864 100644
> --- a/drivers/usb/host/xhci-mem.c
> +++ b/drivers/usb/host/xhci-mem.c
> @@ -2505,7 +2505,8 @@ int xhci_mem_init(struct xhci_hcd *xhci, gfp_t flags)
>  		       "Allocating primary event ring");
>  	xhci->interrupters = kcalloc_node(xhci->max_interrupters, sizeof(*xhci->interrupters),
>  					  flags, dev_to_node(dev));
> -
> +	if (!xhci->interrupters)
> +		goto fail;
>  	ir = xhci_alloc_interrupter(xhci, 0, flags);
>  	if (!ir)
>  		goto fail;
> -- 
> 2.43.0

Hi Greg and Mathias,

I noticed that this bug still exists in current 6.6 and 6.12 releases,
what would be the sensible course of action to fix it?

The patch from Guangshuo Li is a specific fix and it applies cleanly on
those branches. By simulating allocation failure, I confirmed the bug
and the fix on 6.6.113, which is identical to the current 6.6.116.

Mainline added an identical check in 83d98dea48eb ("usb: xhci: add
individual allocation checks in xhci_mem_init()") which is a cleanup
that has a merge conflict at least with 6.6.

The conflict seems superficial and probably the cleanup would have no
side effects on 6.6/6.12, but I haven't really reviewed this and maybe
it would be simpler to just take the targeted fix?

Regards,
Michal