3.7-rc3: mouses stopped working

Commit Message

Dmitry Torokhov Nov. 18, 2012, 8:56 a.m. UTC

On Fri, Nov 16, 2012 at 01:37:31PM +0100, Jiri Kosina wrote:
> On Fri, 16 Nov 2012, Pavel Machek wrote:
> 
> > > > root@amd:/tmp# uname -a
> > > > Linux amd 3.7.0-rc5+ #236 SMP Thu Nov 15 20:07:26 CET 2012 i686 GNU/Linux
> > > > root@amd:/tmp# cat /dev/input/mice
> > > > cat: /dev/input/mice: No such device or address
> > > > root@amd:/tmp# ls -al /dev/input/mice
> > > > crw-rw---- 1 root root 13, 63 Apr 25  2010 /dev/input/mice
> > > > root@amd:/tmp# 
> > > 
> > > Let's see whether this is /dev/input/mice issue or something else first.
> > > 
> > > Is anything coming out of corresponding /dev/input/eventX node?
> > 
> > /dev/input/mouse0 and /dev/input/mouse1 generate events when I move
> > internal/USB mouse.
> > 
> > ls -al /dev/input/mouse0 
> > crw-rw---- 1 root root 13, 32 Apr 25  2010 /dev/input/mouse0
> > pavel@amd:/tmp$ ls -al /dev/input/mouse1
> > crw------T 1 root root 13, 33 Aug 30 16:41 /dev/input/mouse1
> > pavel@amd:/tmp$ 
> > 
> > > And /dev/hidrawX?
> > 
> > /dev/hidraw0 seems to correspond to USB mouse and generates
> > events. Could not find hidraw corresponding to PS/2 mouse.
> 
> That is correct and expected behavior.
> 
> > So it seems to be "/dev/input/mice" issue. Does it work for you?
> 
> Yes, it works for me properly (testing with rc4+, 1b42fc4 more precisely). 
> Even after disconnect/reconnect cycle of USB mouse, I am getting events 
> properly from /dev/input/mice (frankly I am surprised that you noticed ... 
> Xorg has been using evdev driver for ages already by default, right?)
> 
> What was the last kernel version that worked for you in this respect?
> 
> Adding Dmitry and linux-input.

Does the patch below help?

Thanks.

Comments

Krzysztof Mazur Nov. 18, 2012, 1:34 p.m. UTC | #1

On Sun, Nov 18, 2012 at 12:56:38AM -0800, Dmitry Torokhov wrote:
> 
> Input: mousedev - move /deb/input/mice to teh correct minor

s/teh/the/ ??

> 
> From: Dmitry Torokhov <[e-mail]>
> 
> When doing conversion to dynamic input numbers I inadvertently moved
> /dev/input/mice form c,13,63 to c,13,31. We need to fix this so that
> setups with statically populated /dev continue working.
> 
> Signed-off-by: Dmitry Torokhov <[e-mail]>

I had the same problem and this patch fixes it (I even concurrently fixed
that in similar way).

If you need that:
Tested-by: Krzysztof Mazur <[e-mail]>

Thanks,

Krzysiek
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to [e-mail]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Pavel Machek Nov. 18, 2012, 6:53 p.m. UTC | #2

Hi!


> > That is correct and expected behavior.
> > 
> > > So it seems to be "/dev/input/mice" issue. Does it work for you?
> > 
> > Yes, it works for me properly (testing with rc4+, 1b42fc4 more precisely). 
> > Even after disconnect/reconnect cycle of USB mouse, I am getting events 
> > properly from /dev/input/mice (frankly I am surprised that you noticed ... 
> > Xorg has been using evdev driver for ages already by default, right?)
> > 
> > What was the last kernel version that worked for you in this respect?
> > 
> > Adding Dmitry and linux-input.
> 
> Does the patch below help?

Yes. With this patch, my mice are back.

Tested-by: Pavel Machek <[e-mail]>

								Pavel

Patch

diff --git a/drivers/input/mousedev.c b/drivers/input/mousedev.c
index 8f02e3d..4c842c3 100644
--- a/drivers/input/mousedev.c
+++ b/drivers/input/mousedev.c
@@ -12,8 +12,8 @@ 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #define MOUSEDEV_MINOR_BASE	32
-#define MOUSEDEV_MINORS		32
-#define MOUSEDEV_MIX		31
+#define MOUSEDEV_MINORS		31
+#define MOUSEDEV_MIX		63
 
 #include <linux/sched.h>
 #include <linux/slab.h>