# Re: \[BUG v3.6.2, 3.7-rc1\] Unable to rmmod any hid special driver

- From: Jiri Kosina \<hidden\>
- Date: 2012-11-01 10:14:55 
- List: linux-input
- Page: https://ratatoskr.run/linux-input/2012/11/8920512
- Thread: https://ratatoskr.run/linux-input/2012/10/8920422/t.md

```text
On Thu, 18 Oct 2012, Benjamin Tissoires wrote:

> Hi Guys,
> 
> Commit 4fe9f8e203fdad1524c04beb390f3c6099781ed9 upstream
> Commit 57a38d925e665fcd729cd9ad4a75c84e11bd95c9 linux-stable
> 
> This commit "HID: hidraw: don't deallocate memory when it is in use"
> prevents users to rmmod then insmod  any hid special driver without
> either physically unplugging the device or without rmmod-ing the hid
> bus itself.
> 
> When the user does a rmmod, hidinput_disconnect is called before
> hidraw_disconnect, thus, the hidraw device is still there, but the
> input one not. It leads to the fact that the device is not working
> anymore even if you call modprobe.
> 
> Any idea to solve that?

Guys, how about this?


---
 drivers/hid/hidraw.c |    5 +++--
 1 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/hid/hidraw.c b/drivers/hid/hidraw.c
index 17d15bb..fce4256 100644
--- a/drivers/hid/hidraw.c
+++ b/drivers/hid/hidraw.c
@@ -560,10 +560,11 @@ static void drop_ref(struct hidraw *hidraw, int exists_bit)
 {
 	mutex_lock(&minors_lock);
 	if (exists_bit) {
-		hid_hw_close(hidraw->hid);
 		hidraw->exist = 0;
-		if (hidraw->open)
+		if (hidraw->open) {
+			hid_hw_close(hidraw->hid);
 			wake_up_interruptible(&hidraw->wait);
+		}
 	} else {
 		--hidraw->open;
 	}

-- 
Jiri Kosina
SUSE Labs
```