Thread (9 messages) flat view 9 messages, 2 authors, 2016-07-28

Re: [PATCH] Input: xpad - power off wireless 360 controllers on suspend

From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Date: 2016-07-27 21:32:25

On Mon, Jul 25, 2016 at 10:35:08PM -0700, Cameron Gutman wrote:
quoted hunk ↗ jump to hunk
When the USB wireless adapter is suspended, the controllers
lose their connection. This causes them to start flashing
their LED rings and searching for the wireless adapter
again, wasting the controller's battery power.

Instead, we will tell the controllers to power down when
we suspend. This mirrors the behavior of the controllers
when connected to the console itself and how the official
Xbox One wireless adapter behaves on Windows.

Signed-off-by: Cameron Gutman <redacted>
---
This patch is independent of the other xpad patch [0] that I
submitted (and decided to wait on). It applies against
unmodified xpad.c in master.

[0] http://www.spinics.net/lists/linux-input/msg46062.html
---
 drivers/input/joystick/xpad.c | 43 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 43 insertions(+)
diff --git a/drivers/input/joystick/xpad.c b/drivers/input/joystick/xpad.c
index a529a45..3408019 100644
--- a/drivers/input/joystick/xpad.c
+++ b/drivers/input/joystick/xpad.c
@@ -115,6 +115,10 @@ static bool sticks_to_null;
 module_param(sticks_to_null, bool, S_IRUGO);
 MODULE_PARM_DESC(sticks_to_null, "Do not map sticks at all for unknown pads");
 
+static bool disable_auto_poweroff;
+module_param(disable_auto_poweroff, bool, S_IRUGO);
+MODULE_PARM_DESC(disable_auto_poweroff, "Do not power off wireless controllers on suspend");
Why negating? Why not do

static bool xpad_auto_poweroff = true;

?

(No need to resubmit if agree/disagree, I can fix up on my side).
quoted hunk ↗ jump to hunk
+
 static const struct xpad_device {
 	u16 idVendor;
 	u16 idProduct;
@@ -1248,6 +1252,36 @@ static void xpad_stop_input(struct usb_xpad *xpad)
 	usb_kill_urb(xpad->irq_in);
 }
 
+static void xpad360w_poweroff_controller(struct usb_xpad *xpad)
+{
+	unsigned long flags;
+	struct xpad_output_packet *packet =
+			&xpad->out_packets[XPAD_OUT_CMD_IDX];
+
+	spin_lock_irqsave(&xpad->odata_lock, flags);
+
+	packet->data[0] = 0x00;
+	packet->data[1] = 0x00;
+	packet->data[2] = 0x08;
+	packet->data[3] = 0xC0;
+	packet->data[4] = 0x00;
+	packet->data[5] = 0x00;
+	packet->data[6] = 0x00;
+	packet->data[7] = 0x00;
+	packet->data[8] = 0x00;
+	packet->data[9] = 0x00;
+	packet->data[10] = 0x00;
+	packet->data[11] = 0x00;
+	packet->len = 12;
+	packet->pending = true;
I wonder of we don't want to convert commands to something like that:

	static const u8 power_off_cmd[] = {
		0x00, 0x00, 0x08, 0xc0, 0x00, 0x00, 0x00, 0x00,
		0x00, 0x00, 0x00, 0x00,
	};

	...

	memcpy(packet->data, power_off_cmd, sizeof(power_off_cmd));
	// if we need to change something
	// packet->data[3] += command;
	// packet->data[6] = id;
	packet->len = sizeof(power_off_cmd);
	...

This should be a separate patch though.

Thanks.

-- 
Dmitry
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help