2019-12-10 11:40:36

by Johan Hovold

[permalink] [raw]
Subject: [PATCH 0/7] Input: fix USB alsetting bugs

We had quite a few driver using the first alternate setting instead of
the current one when doing descriptor sanity checks. This is mostly an
issue on kernels with panic_on_warn set due to a WARN() in
usb_submit_urn(). Since we've started backporting such fixes (e.g. as
reported by syzbot), I've marked these for stable as well.

Included are also a couple of related clean ups to prevent future
issues.

Johan


Johan Hovold (7):
Input: pegasus_notetaker: fix endpoint sanity check
Input: aiptek: fix endpoint sanity check
Input: aiptek: use descriptors of current altsetting
Input: gtco: fix endpoint sanity check
Input: gtco: fix extra-descriptor debug message
Input: gtco: drop redundant variable reinit
Input: sur40: fix interface sanity checks

drivers/input/tablet/aiptek.c | 8 ++++----
drivers/input/tablet/gtco.c | 13 ++++---------
drivers/input/tablet/pegasus_notetaker.c | 2 +-
drivers/input/touchscreen/sur40.c | 2 +-
4 files changed, 10 insertions(+), 15 deletions(-)

--
2.24.0


2019-12-12 15:27:26

by Vladis Dronov

[permalink] [raw]
Subject: Re: [PATCH 0/7] Input: fix USB alsetting bugs

Hello,

On 10.12.19 12:37, Johan Hovold wrote:

> We had quite a few driver using the first alternate setting instead of
> the current one when doing descriptor sanity checks. This is mostly an
> issue on kernels with panic_on_warn set due to a WARN() in
> usb_submit_urn(). Since we've started backporting such fixes (e.g. as
> reported by syzbot), I've marked these for stable as well.
>
> Included are also a couple of related clean ups to prevent future
> issues.

For the series:

Acked-by: Vladis Dronov <[email protected]>

Best regards,
Vladis Dronov | Red Hat, Inc. | The Core Kernel | Senior Software Engineer

2020-01-10 22:44:44

by Dmitry Torokhov

[permalink] [raw]
Subject: Re: [PATCH 0/7] Input: fix USB alsetting bugs

On Thu, Dec 12, 2019 at 04:25:18PM +0100, Vladis Dronov wrote:
> Hello,
>
> On 10.12.19 12:37, Johan Hovold wrote:
>
> > We had quite a few driver using the first alternate setting instead of
> > the current one when doing descriptor sanity checks. This is mostly an
> > issue on kernels with panic_on_warn set due to a WARN() in
> > usb_submit_urn(). Since we've started backporting such fixes (e.g. as
> > reported by syzbot), I've marked these for stable as well.
> >
> > Included are also a couple of related clean ups to prevent future
> > issues.
>
> For the series:
>
> Acked-by: Vladis Dronov <[email protected]>

Applied the lot, thank you.

--
Dmitry