diff options
author | miletus@chromium.org <miletus@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2014-01-22 09:30:00 +0000 |
---|---|---|
committer | miletus@chromium.org <miletus@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2014-01-22 09:30:00 +0000 |
commit | 49e3b2cfe044dd2bfd1dac66f9145beefb2db32e (patch) | |
tree | 8a80ed16cd9d6b77456f5a47ef65cca62adb795f /ash/shell.cc | |
parent | b772fcea82cc6572c0c36c3618e827e4a8b206bc (diff) | |
download | chromium_src-49e3b2cfe044dd2bfd1dac66f9145beefb2db32e.zip chromium_src-49e3b2cfe044dd2bfd1dac66f9145beefb2db32e.tar.gz chromium_src-49e3b2cfe044dd2bfd1dac66f9145beefb2db32e.tar.bz2 |
Make sure INPUT_EVENT_LATENCY_BEGIN_RWH_COMPONENT is created when event passes through RWHI
We want every input event to have a INPUT_EVENT_LATENCY_BEGIN_RWH_COMPONENT
in its LatencyInfo as the starting point for measuring input latency. In some
test cases the synthetic mouse/mouse wheel event can miss the component:
For example, for non-aura platform,
SyntheticGestureTargetBase::DispatchWebMouseWheelEventToPlatform calls
RWHI::ForwardWheelEventWithLatencyInfo(), while we only enforcing creating
the begin component in RWHI::ForwardWheelEvent().
The fix in this CL is to move CreateRWHLatencyInfoIfNotExist() call, which
enforces that the begin component exists/is created in the LatencyInfo, from
RWHI::ForwardWheelEvent into RWHI::ForwardWheelEventWithLatencyInfo.
The same is applied to FowardMouseEvent(WithLatencyInfo)().
BUG=246034
TEST=run telemtry smoothness test agasint non-aura linux build, and make sure
that INPUT_EVENT_LATENCY_BEGIN_RWH_COMPONENT exists for mouse wheel event's
LatencyInfo.
Review URL: https://codereview.chromium.org/136203003
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246257 0039d316-1c4b-4281-b951-d872f2087c98
Diffstat (limited to 'ash/shell.cc')
0 files changed, 0 insertions, 0 deletions