linux-kernel - Re: [PATCH] uvc-gadget: Fix Set Interface (alternate setting) response behaviour

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:   Tue, 07 Mar 2017 14:29:16 +0200
From:   Laurent Pinchart <laurent.pinchart@...asonboard.com>
To:     Felipe Balbi <balbi@...nel.org>
Cc:     Roger Quadros <rogerq@...com>, b-liu@...com, nsekhar@...com,
        linux-usb@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] uvc-gadget: Fix Set Interface (alternate setting) response behaviour

Hi Felipe,

On Tuesday 07 Mar 2017 12:57:40 Felipe Balbi wrote:
> Laurent Pinchart writes:
> > On Friday 03 Mar 2017 13:17:15 Roger Quadros wrote:
> >> On alternate setting change, webcam gadget sends us a UVC_EVENT_STREAMON
> >> or UVC_EVENT_STREAMOFF event. It expects delayed status response on
> >> STREAMON event only but doesn't expect us to send that response over USB.
> >> It sends the delayed response when we issue the VIDIOC_STREAMON ioctl.
> >> 
> >> So we must not send UVCIOC_SEND_RESPONSE ioctl in these cases that too
> >> with invalid response length.
> > 
> > The commit message only explains why we should not call
> > UVCIOC_SEND_RESPONSE in response to a STREAMON event, but not why we
> > shouldn't either in response to a STREAMOFF event. The patch is correct
> > changing both, but I propose wording the above two paragraphs as follows.
> > 
> > "uvc-gadget: Do not send Set Interface (alternate setting) response twice
> > 
> > On alternate setting change, the webcam gadget sends us a
> > UVC_EVENT_STREAMON or UVC_EVENT_STREAMOFF event. In the first case, the
> > driver will issue a delayed status response automatically when we call
> > the VIDIOC_STREAMON ioctl. In the second case, the driver sends the
> > status response immediately. We must thus not send the status response
> > manually with UVCIOC_SEND_RESPONSE in any of those cases."
> > 
> > If you're fine with that I'll change the message when applying, there's no
> > need to resend the patch.
> 
> I have this in my testing/fixes and was planning to send it to Greg this
> week. I can drop it from my queue, no problem, but then let me know as
> you would need my acked-by.

This is a userspace application patch. Feel free to send it to Greg, but I 
don't think he will know what to do with it :-) Were you maybe confusing this 
patch with the kernel fix that Roger sent a few days ago ? That one should be 
queued, please keep it in your tree.

-- 
Regards,

Laurent Pinchart