linux-kernel - Re: [PATCH] usb: xhci: fix incorrect memset()

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] [day] [month] [year] [list]

Message-ID: <f984c9e0-e0b6-b6b9-0491-92c64363ad47@linux.intel.com>
Date:   Tue, 12 Dec 2017 12:47:34 +0200
From:   Mathias Nyman <mathias.nyman@...ux.intel.com>
To:     Arnd Bergmann <arnd@...db.de>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc:     Mathias Nyman <mathias.nyman@...el.com>,
        Lu Baolu <baolu.lu@...ux.intel.com>, linux-usb@...r.kernel.org,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] usb: xhci: fix incorrect memset()

On 11.12.2017 19:05, Arnd Bergmann wrote:
> On Mon, Dec 11, 2017 at 6:01 PM, Greg Kroah-Hartman
> <gregkh@...uxfoundation.org> wrote:
>> On Mon, Dec 11, 2017 at 02:59:13PM +0200, Mathias Nyman wrote:
>>> On 11.12.2017 13:27, Arnd Bergmann wrote:
>>>> gcc-8 warnings about the new driver using a memset with a bogus length:
>>>>
>>>> drivers/usb/host/xhci-dbgcap.c: In function 'xhci_dbc_eps_exit':
>>>> drivers/usb/host/xhci-dbgcap.c:369:2: error: 'memset' used with length equal to number of elements without multiplication by element size [-Werror=memset-elt-size]
>>>>
>>>> It looks like the author meant to use sizeof() rather than ARRAY_SIZE()
>>>> here, so use that.
>>>>
>>>> Fixes: dfba2174dc42 ("usb: xhci: Add DbC support in xHCI driver")
>>>> Signed-off-by: Arnd Bergmann <arnd@...db.de>
>>>> ---
>>>
>>> Another patch to fix the same thing was sent earlier as a follow up to the original series.
>>> https://marc.info/?l=linux-usb&m=151298133524873&w=2
>>>
>>> But your patch includes the Fixes line with the commit id, which is nice.
>>>
>>> Both fix the problem, It doesn't matter for me which one gets applied
>>
>> They fix it in different ways, which is correct?
> 
> Both are correct, 'sizeof(dbc->eps)' is the same as 'sizeof(struct
> dbc_ep) * ARRAY_SIZE(dbc->eps)'.
> 
>> And Arnd probably doesn't have the hardware to do so?
> 
> Right.
> 

Just to test and confirm boyh work I added a couple printk messages after memset()
showing bytes zeroed and method used:

# echo enabled > dbc
# cat dbc
enabled
# [  219.050277] xhci_hcd 0000:00:14.0: DbC connected
[  219.307250] xhci_hcd 0000:00:14.0: DbC configured
[  219.313601] xhci_hcd 0000:00:14.0: DbC now attached to /dev/ttyDBC0

# echo disabled > dbc
[  255.745590] xhci_hcd 0000:00:14.0: set 80 bytes to zero sizeof(struct dbc_ep) * ARRAY_SIZE(dbc->eps)
[  255.755970] xhci_hcd 0000:00:14.0: set 80 bytes to zero sizeof(dbc->eps)

-Mathias