evdev: Fix crashing with touch devices

evdev: Fix crashing with touch devices

The touchscreen crashes because of BTN_TOUCH events that hit NOTREACHED.

The touchpad crashes for similar reasons. We don't have touchpad support
here at all, so ignore these devices.

NOTRY=true

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=233431

[rjkroege, 2013-11-04 23:39:46]

https://codereview.chromium.org/58473003/diff/1/ui/events/ozone/evdev/event_d...
File ui/events/ozone/evdev/event_device_info.cc (right):

https://codereview.chromium.org/58473003/diff/1/ui/events/ozone/evdev/event_d...
ui/events/ozone/evdev/event_device_info.cc:62: if (ioctl(fd,
EVIOCGPROP(sizeof(prop_bits_)), prop_bits_))
you are not using the same ioctl as I was? 

And if it fails, some kind of diag would be desirable?

https://codereview.chromium.org/58473003/diff/1/ui/events/ozone/evdev/event_d...
File ui/events/ozone/evdev/event_device_info.h (right):

https://codereview.chromium.org/58473003/diff/1/ui/events/ozone/evdev/event_d...
ui/events/ozone/evdev/event_device_info.h:18: class EventDeviceInfo {
a class doc?

https://codereview.chromium.org/58473003/diff/1/ui/events/ozone/evdev/event_d...
ui/events/ozone/evdev/event_device_info.h:39: unsigned long
ev_bits_[EVDEV_BITS_TO_LONGS(EV_CNT)];
how big are these?

https://codereview.chromium.org/58473003/diff/1/ui/events/ozone/evdev/touch_e...
File ui/events/ozone/evdev/touch_event_converter.cc (right):

https://codereview.chromium.org/58473003/diff/1/ui/events/ozone/evdev/touch_e...
ui/events/ozone/evdev/touch_event_converter.cc:184: } else if (input.type ==
EV_KEY) {
it is possible to get key events on the touchscreen fd? does this actually ever
happen?

[spang, 2013-11-05 00:10:02]

https://codereview.chromium.org/58473003/diff/1/ui/events/ozone/evdev/event_d...
File ui/events/ozone/evdev/event_device_info.cc (right):

https://codereview.chromium.org/58473003/diff/1/ui/events/ozone/evdev/event_d...
ui/events/ozone/evdev/event_device_info.cc:62: if (ioctl(fd,
EVIOCGPROP(sizeof(prop_bits_)), prop_bits_))
On 2013/11/04 23:39:46, rjkroege wrote:
> you are not using the same ioctl as I was? 
> 
> And if it fails, some kind of diag would be desirable?

I am above in GetEventbits; this one is different.

Added DLOG for failures.

https://codereview.chromium.org/58473003/diff/1/ui/events/ozone/evdev/event_d...
File ui/events/ozone/evdev/event_device_info.h (right):

https://codereview.chromium.org/58473003/diff/1/ui/events/ozone/evdev/event_d...
ui/events/ozone/evdev/event_device_info.h:18: class EventDeviceInfo {
On 2013/11/04 23:39:46, rjkroege wrote:
> a class doc?

Done.

https://codereview.chromium.org/58473003/diff/1/ui/events/ozone/evdev/event_d...
ui/events/ozone/evdev/event_device_info.h:39: unsigned long
ev_bits_[EVDEV_BITS_TO_LONGS(EV_CNT)];
On 2013/11/04 23:39:46, rjkroege wrote:
> how big are these?

Whole thing is currently 152 bytes.

This is less than X does though (it has the same and more). It also keeps it
around forever, which we are not (yet).

https://codereview.chromium.org/58473003/diff/1/ui/events/ozone/evdev/touch_e...
File ui/events/ozone/evdev/touch_event_converter.cc (right):

https://codereview.chromium.org/58473003/diff/1/ui/events/ozone/evdev/touch_e...
ui/events/ozone/evdev/touch_event_converter.cc:184: } else if (input.type ==
EV_KEY) {
On 2013/11/04 23:39:46, rjkroege wrote:
> it is possible to get key events on the touchscreen fd? does this actually
ever
> happen?
> 

Yes it happens. Only the BTN_TOUCH "key".

[rjkroege, 2013-11-05 17:56:51]

lgtm

https://codereview.chromium.org/58473003/diff/70001/ui/events/ozone/evdev/eve...
File ui/events/ozone/evdev/event_device_info.cc (right):

https://codereview.chromium.org/58473003/diff/70001/ui/events/ozone/evdev/eve...
ui/events/ozone/evdev/event_device_info.cc:55:
base::ThreadRestrictions::AssertIOAllowed();
nit: move the assert to GetPropBits and GetEventBits

https://codereview.chromium.org/58473003/diff/70001/ui/events/ozone/evdev/tou...
File ui/events/ozone/evdev/touch_event_converter.cc (right):

https://codereview.chromium.org/58473003/diff/70001/ui/events/ozone/evdev/tou...
ui/events/ozone/evdev/touch_event_converter.cc:184: } else if (input.type ==
EV_KEY) {
I still think it's weird. :-)

[Yufeng Shen (Slow to review), 2013-11-05 18:28:02]

https://codereview.chromium.org/58473003/diff/70001/ui/events/ozone/evdev/tou...
File ui/events/ozone/evdev/touch_event_converter.cc (right):

https://codereview.chromium.org/58473003/diff/70001/ui/events/ozone/evdev/tou...
ui/events/ozone/evdev/touch_event_converter.cc:184: } else if (input.type ==
EV_KEY) {
On 2013/11/05 17:56:52, rjkroege wrote:
> I still think it's weird. :-)

Check the "EV_KEY" section in 
https://www.kernel.org/doc/Documentation/input/event-codes.txt

[spang, 2013-11-05 19:18:34]

https://codereview.chromium.org/58473003/diff/70001/ui/events/ozone/evdev/eve...
File ui/events/ozone/evdev/event_device_info.cc (right):

https://codereview.chromium.org/58473003/diff/70001/ui/events/ozone/evdev/eve...
ui/events/ozone/evdev/event_device_info.cc:55:
base::ThreadRestrictions::AssertIOAllowed();
On 2013/11/05 17:56:52, rjkroege wrote:
> nit: move the assert to GetPropBits and GetEventBits

Done.

[commit-bot: I haz the power, 2013-11-05 19:24:51]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/spang@chromium.org/58473003/140001

[commit-bot: I haz the power, 2013-11-06 00:09:03]

Retried try job too often on linux_aura for step(s) browser_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_aura...

[commit-bot: I haz the power, 2013-11-06 16:01:32]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/spang@chromium.org/58473003/140001

[commit-bot: I haz the power, 2013-11-06 19:45:03]

Retried try job too often on linux_aura for step(s) browser_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_aura...

[commit-bot: I haz the power, 2013-11-06 20:08:12]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/spang@chromium.org/58473003/140001

[commit-bot: I haz the power, 2013-11-06 22:00:13]

Failed to apply patch for ui/events/events.gyp:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  patching file ui/events/events.gyp
  Hunk #1 FAILED at 65.
  1 out of 1 hunk FAILED -- saving rejects to file ui/events/events.gyp.rej

Patch:       ui/events/events.gyp
Index: ui/events/events.gyp
diff --git a/ui/events/events.gyp b/ui/events/events.gyp
index
eb11794f97821e56e2a802eed36fe5dd5c86d055..673b06e5fb9d80a326612394a3cd26c2c19f404e
100644
--- a/ui/events/events.gyp
+++ b/ui/events/events.gyp
@@ -65,6 +65,8 @@
         'keycodes/keyboard_codes.h',
         'latency_info.cc',
         'latency_info.h',
+        'ozone/evdev/event_device_info.cc',
+        'ozone/evdev/event_device_info.h',
         'ozone/evdev/event_factory.cc',
         'ozone/evdev/event_factory.h',
         'ozone/evdev/event_modifiers.cc',

[commit-bot: I haz the power, 2013-11-06 23:12:41]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/spang@chromium.org/58473003/540001

[commit-bot: I haz the power, 2013-11-07 00:20:00]

Change committed as 233431