2022-09-14 07:52:05

by Johnothan King

[permalink] [raw]
Subject: [PATCH] HID: nintendo: check analog user calibration for plausibility

Arne Wendt writes:
Cheap clone controllers may (falsely) report as having a user
calibration for the analog sticks in place, but return
wrong/impossible values for the actual calibration data.
In the present case at mine, the controller reports having a
user calibration in place and successfully executes the read
commands. The reported user calibration however is
min = center = max = 0.

This pull request addresses problems of this kind by checking the
provided user calibration-data for plausibility (min < center < max)
and falling back to the default values if implausible.

I'll note that I was experiencing a crash because of this bug when using
the GuliKit KingKong 2 controller. The crash manifests as a divide by
zero error in the kernel logs:
kernel: divide error: 0000 [#1] PREEMPT SMP NOPTI

Link: https://github.com/nicman23/dkms-hid-nintendo/pull/25
Link: https://github.com/DanielOgorchock/linux/issues/36
Co-authored-by: Arne Wendt <[email protected]>
Signed-off-by: Johnothan King <[email protected]>
---
drivers/hid/hid-nintendo.c | 24 ++++++++++++++++++++++--
1 file changed, 22 insertions(+), 2 deletions(-)

diff --git a/drivers/hid/hid-nintendo.c b/drivers/hid/hid-nintendo.c
index 6028af3c3aae..7f287f6a75f5 100644
--- a/drivers/hid/hid-nintendo.c
+++ b/drivers/hid/hid-nintendo.c
@@ -793,7 +793,17 @@ static int joycon_request_calibration(struct joycon_ctlr *ctlr)
&ctlr->left_stick_cal_x,
&ctlr->left_stick_cal_y,
true);
- if (ret) {
+
+ /*
+ * Check whether read succeeded and perform plausibility check
+ * for retrieved values.
+ */
+ if (ret ||
+ ctlr->left_stick_cal_x.min >= ctlr->left_stick_cal_x.center ||
+ ctlr->left_stick_cal_x.center >= ctlr->left_stick_cal_x.max ||
+ ctlr->left_stick_cal_y.min >= ctlr->left_stick_cal_y.center ||
+ ctlr->left_stick_cal_y.center >= ctlr->left_stick_cal_y.max
+ ) {
hid_warn(ctlr->hdev,
"Failed to read left stick cal, using dflts; e=%d\n",
ret);
@@ -812,7 +822,17 @@ static int joycon_request_calibration(struct joycon_ctlr *ctlr)
&ctlr->right_stick_cal_x,
&ctlr->right_stick_cal_y,
false);
- if (ret) {
+
+ /*
+ * Check whether read succeeded and perform plausibility check
+ * for retrieved values.
+ */
+ if (ret ||
+ ctlr->right_stick_cal_x.min >= ctlr->right_stick_cal_x.center ||
+ ctlr->right_stick_cal_x.center >= ctlr->right_stick_cal_x.max ||
+ ctlr->right_stick_cal_y.min >= ctlr->right_stick_cal_y.center ||
+ ctlr->right_stick_cal_y.center >= ctlr->right_stick_cal_y.max
+ ) {
hid_warn(ctlr->hdev,
"Failed to read right stick cal, using dflts; e=%d\n",
ret);
--
2.37.3



2022-09-14 20:50:48

by Silvan Jegen

[permalink] [raw]
Subject: Re: [PATCH] HID: nintendo: check analog user calibration for plausibility

Hi

Johnothan King <[email protected]> wrote:
> Arne Wendt writes:
> Cheap clone controllers may (falsely) report as having a user
> calibration for the analog sticks in place, but return
> wrong/impossible values for the actual calibration data.
> In the present case at mine, the controller reports having a
> user calibration in place and successfully executes the read
> commands. The reported user calibration however is
> min = center = max = 0.
>
> This pull request addresses problems of this kind by checking the
> provided user calibration-data for plausibility (min < center < max)
> and falling back to the default values if implausible.
>
> I'll note that I was experiencing a crash because of this bug when using
> the GuliKit KingKong 2 controller. The crash manifests as a divide by
> zero error in the kernel logs:
> kernel: divide error: 0000 [#1] PREEMPT SMP NOPTI
>
> Link: https://github.com/nicman23/dkms-hid-nintendo/pull/25
> Link: https://github.com/DanielOgorchock/linux/issues/36
> Co-authored-by: Arne Wendt <[email protected]>
> Signed-off-by: Johnothan King <[email protected]>
> ---
> drivers/hid/hid-nintendo.c | 24 ++++++++++++++++++++++--
> 1 file changed, 22 insertions(+), 2 deletions(-)

Reviewed-by: Silvan Jegen <[email protected]>

I have actually hit this issue as well ...


Cheers,
Silvan

>
> diff --git a/drivers/hid/hid-nintendo.c b/drivers/hid/hid-nintendo.c
> index 6028af3c3aae..7f287f6a75f5 100644
> --- a/drivers/hid/hid-nintendo.c
> +++ b/drivers/hid/hid-nintendo.c
> @@ -793,7 +793,17 @@ static int joycon_request_calibration(struct joycon_ctlr *ctlr)
> &ctlr->left_stick_cal_x,
> &ctlr->left_stick_cal_y,
> true);
> - if (ret) {
> +
> + /*
> + * Check whether read succeeded and perform plausibility check
> + * for retrieved values.
> + */
> + if (ret ||
> + ctlr->left_stick_cal_x.min >= ctlr->left_stick_cal_x.center ||
> + ctlr->left_stick_cal_x.center >= ctlr->left_stick_cal_x.max ||
> + ctlr->left_stick_cal_y.min >= ctlr->left_stick_cal_y.center ||
> + ctlr->left_stick_cal_y.center >= ctlr->left_stick_cal_y.max
> + ) {
> hid_warn(ctlr->hdev,
> "Failed to read left stick cal, using dflts; e=%d\n",
> ret);
> @@ -812,7 +822,17 @@ static int joycon_request_calibration(struct joycon_ctlr *ctlr)
> &ctlr->right_stick_cal_x,
> &ctlr->right_stick_cal_y,
> false);
> - if (ret) {
> +
> + /*
> + * Check whether read succeeded and perform plausibility check
> + * for retrieved values.
> + */
> + if (ret ||
> + ctlr->right_stick_cal_x.min >= ctlr->right_stick_cal_x.center ||
> + ctlr->right_stick_cal_x.center >= ctlr->right_stick_cal_x.max ||
> + ctlr->right_stick_cal_y.min >= ctlr->right_stick_cal_y.center ||
> + ctlr->right_stick_cal_y.center >= ctlr->right_stick_cal_y.max
> + ) {
> hid_warn(ctlr->hdev,
> "Failed to read right stick cal, using dflts; e=%d\n",
> ret);


2022-09-20 10:24:26

by Benjamin Tissoires

[permalink] [raw]
Subject: Re: [PATCH] HID: nintendo: check analog user calibration for plausibility

On Wed, Sep 14, 2022 at 9:51 AM Johnothan King
<[email protected]> wrote:
>
> Arne Wendt writes:
> Cheap clone controllers may (falsely) report as having a user
> calibration for the analog sticks in place, but return
> wrong/impossible values for the actual calibration data.
> In the present case at mine, the controller reports having a
> user calibration in place and successfully executes the read
> commands. The reported user calibration however is
> min = center = max = 0.
>
> This pull request addresses problems of this kind by checking the
> provided user calibration-data for plausibility (min < center < max)
> and falling back to the default values if implausible.
>
> I'll note that I was experiencing a crash because of this bug when using
> the GuliKit KingKong 2 controller. The crash manifests as a divide by
> zero error in the kernel logs:
> kernel: divide error: 0000 [#1] PREEMPT SMP NOPTI
>
> Link: https://github.com/nicman23/dkms-hid-nintendo/pull/25
> Link: https://github.com/DanielOgorchock/linux/issues/36
> Co-authored-by: Arne Wendt <[email protected]>
> Signed-off-by: Johnothan King <[email protected]>
> ---
> drivers/hid/hid-nintendo.c | 24 ++++++++++++++++++++++--
> 1 file changed, 22 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/hid/hid-nintendo.c b/drivers/hid/hid-nintendo.c
> index 6028af3c3aae..7f287f6a75f5 100644
> --- a/drivers/hid/hid-nintendo.c
> +++ b/drivers/hid/hid-nintendo.c
> @@ -793,7 +793,17 @@ static int joycon_request_calibration(struct joycon_ctlr *ctlr)
> &ctlr->left_stick_cal_x,
> &ctlr->left_stick_cal_y,
> true);
> - if (ret) {
> +
> + /*
> + * Check whether read succeeded and perform plausibility check
> + * for retrieved values.
> + */
> + if (ret ||
> + ctlr->left_stick_cal_x.min >= ctlr->left_stick_cal_x.center ||
> + ctlr->left_stick_cal_x.center >= ctlr->left_stick_cal_x.max ||
> + ctlr->left_stick_cal_y.min >= ctlr->left_stick_cal_y.center ||
> + ctlr->left_stick_cal_y.center >= ctlr->left_stick_cal_y.max
> + ) {
> hid_warn(ctlr->hdev,
> "Failed to read left stick cal, using dflts; e=%d\n",
> ret);
> @@ -812,7 +822,17 @@ static int joycon_request_calibration(struct joycon_ctlr *ctlr)
> &ctlr->right_stick_cal_x,
> &ctlr->right_stick_cal_y,
> false);
> - if (ret) {
> +
> + /*
> + * Check whether read succeeded and perform plausibility check
> + * for retrieved values.
> + */
> + if (ret ||
> + ctlr->right_stick_cal_x.min >= ctlr->right_stick_cal_x.center ||
> + ctlr->right_stick_cal_x.center >= ctlr->right_stick_cal_x.max ||
> + ctlr->right_stick_cal_y.min >= ctlr->right_stick_cal_y.center ||
> + ctlr->right_stick_cal_y.center >= ctlr->right_stick_cal_y.max

Wouldn't it make sense to have a function that takes a single struct
joycon_stick_cal pointer and does the check?
This would make things more readable IMO.

Cheers,
Benjamin

> + ) {
> hid_warn(ctlr->hdev,
> "Failed to read right stick cal, using dflts; e=%d\n",
> ret);
> --
> 2.37.3
>
>

2022-09-20 18:09:24

by Johnothan King

[permalink] [raw]
Subject: Re: [PATCH] HID: nintendo: check analog user calibration for plausibility

> Wouldn't it make sense to have a function that takes a single struct
> joycon_stick_cal pointer and does the check?

I acknowledge the current code can be simplified further. I will submit a second version of this patch that moves the plausibility check to joycon_read_stick_calibration() and adds a joycon_use_default_calibration() function. This will avoid the code duplication present in the current patch.

In the v2 patch the check will still use both the cal_x and cal_y pointers. However, since the check will be reduced to just one if statement, that should be enough to simplify the code. The new function in the v2 patch (joycon_use_default_calibration()) was implemented so that I could simplify the existing default stick calibration code for reuse with both the left and right sticks. The new function in effect reduces four lines of code to one:
-ctlr->left_stick_cal_x.center = DFLT_STICK_CAL_CEN;
-ctlr->left_stick_cal_y.center = DFLT_STICK_CAL_CEN;
-ctlr->right_stick_cal_x.center = DFLT_STICK_CAL_CEN;
-ctlr->right_stick_cal_y.center = DFLT_STICK_CAL_CEN;
+cal_x->center = cal_y->center = DFLT_STICK_CAL_CEN;

- Johnothan King

------- Original Message -------
On Tuesday, September 20th, 2022 at 3:21 AM, Benjamin Tissoires <[email protected]> wrote:


> On Wed, Sep 14, 2022 at 9:51 AM Johnothan King
> [email protected] wrote:
>
> > Arne Wendt writes:
> > Cheap clone controllers may (falsely) report as having a user
> > calibration for the analog sticks in place, but return
> > wrong/impossible values for the actual calibration data.
> > In the present case at mine, the controller reports having a
> > user calibration in place and successfully executes the read
> > commands. The reported user calibration however is
> > min = center = max = 0.
> >
> > This pull request addresses problems of this kind by checking the
> > provided user calibration-data for plausibility (min < center < max)
> > and falling back to the default values if implausible.
> >
> > I'll note that I was experiencing a crash because of this bug when using
> > the GuliKit KingKong 2 controller. The crash manifests as a divide by
> > zero error in the kernel logs:
> > kernel: divide error: 0000 [#1] PREEMPT SMP NOPTI
> >
> > Link: https://github.com/nicman23/dkms-hid-nintendo/pull/25
> > Link: https://github.com/DanielOgorchock/linux/issues/36
> > Co-authored-by: Arne Wendt [email protected]
> > Signed-off-by: Johnothan King [email protected]
> > ---
> > drivers/hid/hid-nintendo.c | 24 ++++++++++++++++++++++--
> > 1 file changed, 22 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/hid/hid-nintendo.c b/drivers/hid/hid-nintendo.c
> > index 6028af3c3aae..7f287f6a75f5 100644
> > --- a/drivers/hid/hid-nintendo.c
> > +++ b/drivers/hid/hid-nintendo.c
> > @@ -793,7 +793,17 @@ static int joycon_request_calibration(struct joycon_ctlr ctlr)
> > &ctlr->left_stick_cal_x,
> > &ctlr->left_stick_cal_y,
> > true);
> > - if (ret) {
> > +
> > + /
> > + * Check whether read succeeded and perform plausibility check
> > + * for retrieved values.
> > + */
> > + if (ret ||
> > + ctlr->left_stick_cal_x.min >= ctlr->left_stick_cal_x.center ||
> > + ctlr->left_stick_cal_x.center >= ctlr->left_stick_cal_x.max ||
> > + ctlr->left_stick_cal_y.min >= ctlr->left_stick_cal_y.center ||
> > + ctlr->left_stick_cal_y.center >= ctlr->left_stick_cal_y.max
> > + ) {
> > hid_warn(ctlr->hdev,
> > "Failed to read left stick cal, using dflts; e=%d\n",
> > ret);
> > @@ -812,7 +822,17 @@ static int joycon_request_calibration(struct joycon_ctlr ctlr)
> > &ctlr->right_stick_cal_x,
> > &ctlr->right_stick_cal_y,
> > false);
> > - if (ret) {
> > +
> > + /
> > + * Check whether read succeeded and perform plausibility check
> > + * for retrieved values.
> > + */
> > + if (ret ||
> > + ctlr->right_stick_cal_x.min >= ctlr->right_stick_cal_x.center ||
> > + ctlr->right_stick_cal_x.center >= ctlr->right_stick_cal_x.max ||
> > + ctlr->right_stick_cal_y.min >= ctlr->right_stick_cal_y.center ||
> > + ctlr->right_stick_cal_y.center >= ctlr->right_stick_cal_y.max
>
>
> Wouldn't it make sense to have a function that takes a single struct
> joycon_stick_cal pointer and does the check?
> This would make things more readable IMO.
>
> Cheers,
> Benjamin
>
> > + ) {
> > hid_warn(ctlr->hdev,
> > "Failed to read right stick cal, using dflts; e=%d\n",
> > ret);
> > --
> > 2.37.3