Skip to contentSkip to conversation

Fix hiddev's use of usb_find_interface

PATCH In review Guillaume Chazarain 3 participants 6 messages Sep 12, 2010 – Sep 14, 2010

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 Fix hiddev's use of usb_find_interface

My macbook infrared remote control was broken by commit bd25f4dd6972755579d0ea50d1a5ace2e9b00d1a:
HID: hiddev: use usb_find_interface, get rid of BKL

This device appears in dmesg as:
apple 0003:05AC:8242.0001: hiddev0,hidraw0: USB HID v1.11 Device [Apple Computer, Inc. IR Receiver] on usb-0000:00:1d.2-1/input0

It stopped working as lircd was getting ENODEV when opening /dev/usb/hiddev0.

AFAICS hiddev_driver is a dummy driver so usb_find_interface(&hiddev_driver) does not find anything.

The device is associated with the usbhid driver, so let's do usb_find_interface(&hid_driver) instead.

$ ls -l /sys/devices/pci0000:00/0000:00:1d.2/usb7/7-1/7-1:1.0/usb/hiddev0/device/driver
lrwxrwxrwx 1 root root 0 2010-09-12 16:28 /sys/devices/pci0000:00/0000:00:1d.2/usb7/7-1/7-1:1.0/usb/hiddev0/device/driver -> ../../../../../../bus/usb/drivers/usbhid
---
 drivers/hid/usbhid/hid-core.c |    5 +++++
 drivers/hid/usbhid/hiddev.c   |    2 +-
 drivers/hid/usbhid/usbhid.h   |    1 +
 3 files changed, 7 insertions(+), 1 deletions(-)
drivers/hid/usbhid/hid-core.c +5 -0
index b729c02..7b2f7f6 100644
@@ -1445,6 +1445,11 @@ static const struct hid_device_id hid_usb_table[] = {
 	{ }
 };
 
+struct usb_interface *usbhid_find_interface(int minor)
+{
+	return usb_find_interface(&hid_driver, minor);
+}
+
 static struct hid_driver hid_usb_driver = {
 	.name = "generic-usb",
 	.id_table = hid_usb_table,
drivers/hid/usbhid/hiddev.c +1 -1
index 0a29c51..681e620 100644
@@ -270,7 +270,7 @@ static int hiddev_open(struct inode *inode, struct file *file)
 	struct hiddev *hiddev;
 	int res;
 
-	intf = usb_find_interface(&hiddev_driver, iminor(inode));
+	intf = usbhid_find_interface(iminor(inode));
 	if (!intf)
 		return -ENODEV;
 	hid = usb_get_intfdata(intf);
drivers/hid/usbhid/usbhid.h +1 -0
index 693fd3e..89d2e84 100644
@@ -42,6 +42,7 @@ void usbhid_submit_report
 (struct hid_device *hid, struct hid_report *report, unsigned char dir);
 int usbhid_get_power(struct hid_device *hid);
 void usbhid_put_power(struct hid_device *hid);
+struct usb_interface *usbhid_find_interface(int minor);
 
 /* iofl flags */
 #define HID_CTRL_RUNNING	1
Signature · 1 line
1.7.0.4
#1 lore raw
Quoted 3 lines · Guillaume Chazarain

My macbook infrared remote control was broken by commit bd25f4dd6972755579d0ea50d1a5ace2e9b00d1a:
HID: hiddev: use usb_find_interface, get rid of BKL

Right, my bug. Sorry for the trouble you went through.

A few other people have reported the same and Jiri has done a fix that was merged in 2.6.36-rc2.

Arnd

#2 lore raw
Quoted 8 lines · Arnd Bergmann
Quoted 3 lines

My macbook infrared remote control was broken by commit bd25f4dd6972755579d0ea50d1a5ace2e9b00d1a:
HID: hiddev: use usb_find_interface, get rid of BKL

Right, my bug. Sorry for the trouble you went through.

A few other people have reported the same and Jiri has done a fix that was merged in 2.6.36-rc2.

Yeah -- Guillaume, I think that you are actually seeing symptoms of intfdata being bogus. Could you please try with any kernel with commit 9c9e54a8df0be48 ("HID: hiddev: fix memory corruption due to invalid intfdata") applied (which is the one Arnd is talking about) and report back whether the problem is fixed by it?

Thanks,

Signature · 2 lines
Jiri Kosina
SUSE Labs, Novell Inc.
#3 lore raw
Quoted 5 lines · Jiri Kosina

Yeah -- Guillaume, I think that you are actually seeing symptoms of intfdata being bogus. Could you please try with any kernel with commit 9c9e54a8df0be48 ("HID: hiddev: fix memory corruption due to invalid intfdata") applied (which is the one Arnd is talking about) and report back whether the problem is fixed by it?

I also saw the problem with 2.6.36-rc3. My problem is with usb_find_interface(&hiddev_driver) not finding anything and 9c9e54a8df0be48aa359744f412377cc55c3b7d2 did not change this line.

In its current state I don't understand how hiddev_open() can possibly work. I see people reporting suspend issues and stack traces in dmesg but nobody reporting the driver being non-functioning, which is what I'm doing here.

Thanks.

Signature · 1 line
Guillaume
#4 lore raw
Quoted 1 line · Guillaume Chazarain

I also saw the problem with 2.6.36-rc3.

For completeness, I also reproduced the problem with 2.6.36-rc4. Not being familiar with the driver model, I'm not attached to the approach I took in the patch, but it solves a real regression: hiddev not working.

Signature · 1 line
Guillaume
#5 lore raw
In reply to Guillaume Chazarain
Quoted 6 lines · Guillaume Chazarain
Quoted 5 lines

Yeah -- Guillaume, I think that you are actually seeing symptoms of intfdata being bogus. Could you please try with any kernel with commit 9c9e54a8df0be48 ("HID: hiddev: fix memory corruption due to invalid intfdata") applied (which is the one Arnd is talking about) and report back whether the problem is fixed by it?

Quoted 8 lines

I also saw the problem with 2.6.36-rc3. My problem is with usb_find_interface(&hiddev_driver) not finding anything and 9c9e54a8df0be48aa359744f412377cc55c3b7d2 did not change this line.

In its current state I don't understand how hiddev_open() can possibly work. I see people reporting suspend issues and stack traces in dmesg but nobody reporting the driver being non-functioning, which is what I'm doing here.

You are right. I have now applied your patch, thanks a lot for tracking it down.

Signature · 2 lines
Jiri Kosina
SUSE Labs, Novell Inc.