Re: [bug?] [powerpc] hid_logitech_dj: Wireless mouse does not work in xorg or gpm

[Jonathan Nieder <jrnieder@gmail.com>]

Hugo Osvaldo Barrera wrote:

> Here's the dmesg I got:

Thanks.

[...]
> [69358.971692] usbhid: USB HID core driver
> <Mouse receiver was plugged in at this point.  Mouse was off.>
> [69358.973309] [...]/drivers/hid/hid-logitech-dj.c: Logitech-DJ:logi_dj_init
[...]
> [69381.911438] logitech-djreceiver 0003:046D:C52B.0003: hiddev0,hidraw0: USB HID v1.11 Device [Logitech USB Receiver] on usb-0001:10:1b.1-1/input2
> [69381.918070] [...]/drivers/hid/hid-logitech-dj.c: logi_dj_raw_event, size:15
> [69381.918124] [...]/drivers/hid/hid-logitech-dj.c: delayedwork_callback
> [69381.918135] [...]/drivers/hid/hid-logitech-dj.c: logi_dj_recv_add_djhid_device: device list is empty
> [69381.920069] [...]/drivers/hid/hid-logitech-dj.c: logi_dj_raw_event, size:15
> [69381.920119] [...]/drivers/hid/hid-logitech-dj.c: delayedwork_callback
> [69381.920297] [...]/drivers/hid/hid-logitech-dj.c: logi_dj_ll_parse
> [69381.920308] [...]/drivers/hid/hid-logitech-dj.c: logi_dj_ll_start
> <Mouse was turned on at this point.>
> [69381.920430] logitech-djdevice 0003:046D:C52B.0004: hidraw1: USB HID v1.11 Device [Logitech Unifying Device. Wireless PID:101a] on usb-0001:10:1b.1-1:1

Still no "logi_dj_ll_parse: sending a mouse descriptor, reports_supported: 4"
so it looks like the patch didn't have the intended effect.  Alas.

The pairing (...add_djhid_device) happens during initialization,
before the mouse is turned on...

> Here's the capture of the pairing event.
[...]
> Reading from 0003:046D:C52B.0003 resulted in the following
>
> <Mouse turned on at this point>
> report (size 15) (numbered) =  20 01 42 00 00 00 00 00 00 00 00 00 00 00 00
> <Mouse turned off at this point>
> report (size 15) (numbered) =  20 01 42 01 00 00 00 00 00 00 00 00 00 00 00
> <EOF>

... so the events captured do not include pairing events (0x41).  What
we see here are connection status events (0x42).  Sorry about that.

Here's a new patch that prints more information to dmesg.  After
applying the patch, you can use the patched driver by running

	make drivers/hid/hid-logitech-dj.ko
	modprobe -r hid-logitech-dj
	insmod drivers/hid/hid-logitech-dj.ko
---
Thanks again,
Jonathan

 drivers/hid/hid-logitech-dj.c |   14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/hid/hid-logitech-dj.c b/drivers/hid/hid-logitech-dj.c
index e1c38bba4375..33a70cfbec77 100644
--- a/drivers/hid/hid-logitech-dj.c
+++ b/drivers/hid/hid-logitech-dj.c
@@ -23,6 +23,7 @@
 
 
 #include <linux/device.h>
+#include <asm/unaligned.h>
 #include <linux/hid.h>
 #include <linux/module.h>
 #include <linux/usb.h>
@@ -273,8 +274,8 @@ static void logi_dj_recv_add_djhid_device(struct dj_receiver_dev *djrcv_dev,
 		goto dj_device_allocate_fail;
 	}
 
-	dj_dev->reports_supported = le32_to_cpu(
-		dj_report->report_params[DEVICE_PAIRED_RF_REPORT_TYPE]);
+	dj_dev->reports_supported = get_unaligned_le32(
+		dj_report->report_params + DEVICE_PAIRED_RF_REPORT_TYPE);
 	dj_dev->hdev = dj_hiddev;
 	dj_dev->dj_receiver_dev = djrcv_dev;
 	dj_dev->device_index = dj_report->device_index;
@@ -618,9 +619,18 @@ static int logi_dj_raw_event(struct hid_device *hdev,
 	struct dj_report *dj_report = (struct dj_report *) data;
 	unsigned long flags;
 	bool report_processed = false;
+	char buf[32 * 3 + 1];
+	int i;
 
 	dbg_hid("%s, size:%d\n", __func__, size);
 
+	if (0 < size && size <= 32) {
+		for (i = 0; i < size; i++)
+			snprintf(buf + 3*i, sizeof(buf) - 3*i,
+					" %02x", data[i]);
+		dbg_hid("%s: report = %s\n", __func__, buf);
+	}
+
 	/* Here we receive all data coming from iface 2, there are 4 cases:
 	 *
 	 * 1) Data should continue its normal processing i.e. data does not
-- 
1.7.10.1