linux-kernel - Re: [PATCH v3 1/2] usbip: give back URBs for unsent unlink requests during cleanup

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]

Date:   Wed, 18 Aug 2021 12:36:11 -0600
From:   Shuah Khan <skhan@...uxfoundation.org>
To:     Greg KH <gregkh@...uxfoundation.org>
Cc:     Anirudh Rayabharam <mail@...rudhrb.com>,
        valentina.manea.m@...il.com, shuah@...nel.org,
        syzbot+74d6ef051d3d2eacf428@...kaller.appspotmail.com,
        linux-kernel-mentees@...ts.linuxfoundation.org,
        linux-usb@...r.kernel.org, linux-kernel@...r.kernel.org,
        Shuah Khan <skhan@...uxfoundation.org>
Subject: Re: [PATCH v3 1/2] usbip: give back URBs for unsent unlink requests
 during cleanup

On 8/17/21 11:39 PM, Greg KH wrote:
> On Tue, Aug 17, 2021 at 05:16:51PM -0600, Shuah Khan wrote:
>> On 8/13/21 12:25 PM, Anirudh Rayabharam wrote:
>>> In vhci_device_unlink_cleanup(), the URBs for unsent unlink requests are
>>> not given back. This sometimes causes usb_kill_urb to wait indefinitely
>>> for that urb to be given back. syzbot has reported a hung task issue [1]
>>> for this.
>>>
>>> To fix this, give back the urbs corresponding to unsent unlink requests
>>> (unlink_tx list) similar to how urbs corresponding to unanswered unlink
>>> requests (unlink_rx list) are given back.
>>>
>>> [1]: https://syzkaller.appspot.com/bug?id=08f12df95ae7da69814e64eb5515d5a85ed06b76
>>>
>>> Reported-by: syzbot+74d6ef051d3d2eacf428@...kaller.appspotmail.com
>>> Tested-by: syzbot+74d6ef051d3d2eacf428@...kaller.appspotmail.com
>>> Signed-off-by: Anirudh Rayabharam <mail@...rudhrb.com>
>>> ---
>>>    drivers/usb/usbip/vhci_hcd.c | 26 ++++++++++++++++++++++++++
>>>    1 file changed, 26 insertions(+)
>>>
>>> diff --git a/drivers/usb/usbip/vhci_hcd.c b/drivers/usb/usbip/vhci_hcd.c
>>> index 4ba6bcdaa8e9..6f3f374d4bbc 100644
>>> --- a/drivers/usb/usbip/vhci_hcd.c
>>> +++ b/drivers/usb/usbip/vhci_hcd.c
>>> @@ -957,8 +957,34 @@ static void vhci_device_unlink_cleanup(struct vhci_device *vdev)
>>>    	spin_lock(&vdev->priv_lock);
>>>    	list_for_each_entry_safe(unlink, tmp, &vdev->unlink_tx, list) {
>>> +		struct urb *urb;
>>> +
>>> +		/* give back URB of unsent unlink request */
>>>    		pr_info("unlink cleanup tx %lu\n", unlink->unlink_seqnum);
>>
>> I know this is an exiting one.
>> Let's make this pr_debug or remove it all together.
>>
>>> +
>>> +		urb = pickup_urb_and_free_priv(vdev, unlink->unlink_seqnum);
>>> +		if (!urb) {
>>> +			pr_info("the urb (seqnum %lu) was already given back\n",
>>> +				unlink->unlink_seqnum);
>>
>> Let's make this pr_debug or remove it all together.
> 
> As you have a struct device for all of these, please use dev_dbg() and
> friends, not pr_*(), for all of these.
> 

Yes. Makes perfect sense.

thanks,
-- Shuah