Run bisect and find first known bad CL: 
https://chromium.googlesource.com/chromium/src/+log/5562d23c289646cfd767d5c2597ad87fc34866e1..de7e0c3e9bf26b7c3bad40c6aa8883eb47159989

 Issue 625029  has been merged into this issue.

Moving this nonessential bug to the next milestone.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Cannot repro on linux-cros build, but can repro on cros machine. Taking a closer look now.

Also my files window still visible in screenshot, the user tray is however blank.

Only repro on high-dpi, doing that with linux-cros build repros.

So with device scale factor, while calculation clip and transforming the rect, scale got applied twice, and the offset that should be cancelled out transform back and forth didn't. The rect with offset intersecting with layer bounds returns empty rect. I have a CL for this. 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/296dd9f02b09445ad5976396a86c00ff3181c253

commit 296dd9f02b09445ad5976396a86c00ff3181c253
Author: weiliangc <weiliangc@chromium.org>
Date: Tue Jul 05 14:59:51 2016

cc: Fix non root copy request visible rect in HighDPI

Before while transform clip into layer space, incorrectly used wrong
helper function and applied device scale twice. Create correct helper
function and add unittest.

BUG= 622083 
R=ajuma
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review-Url: https://codereview.chromium.org/2117183002
Cr-Commit-Position: refs/heads/master@{#403782}

[modify] https://crrev.com/296dd9f02b09445ad5976396a86c00ff3181c253/cc/test/layer_tree_host_common_test.cc
[modify] https://crrev.com/296dd9f02b09445ad5976396a86c00ff3181c253/cc/trees/draw_property_utils.cc
[modify] https://crrev.com/296dd9f02b09445ad5976396a86c00ff3181c253/cc/trees/layer_tree_host_common_unittest.cc

 Issue 626841  has been merged into this issue.

#11, is there more work required for this? Also should the fix be merged in M-53? 

The fix is done and is in Canary. No more work is needed. Adding merge request to M53.

Your change meets the bar and is auto-approved for M53 (branch: 2785)

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/6196d6ec443cd6290a1eff911532d62b52f56d38

commit 6196d6ec443cd6290a1eff911532d62b52f56d38
Author: Weiliang Chen <weiliangc@chromium.org>
Date: Wed Jul 13 20:00:53 2016

cc: Fix non root copy request visible rect in HighDPI

Before while transform clip into layer space, incorrectly used wrong
helper function and applied device scale twice. Create correct helper
function and add unittest.

BUG= 622083 
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review-Url: https://codereview.chromium.org/2117183002
Cr-Commit-Position: refs/heads/master@{#403782}
(cherry picked from commit 296dd9f02b09445ad5976396a86c00ff3181c253)

Review URL: https://codereview.chromium.org/2140313004 .

Cr-Commit-Position: refs/branch-heads/2785@{#106}
Cr-Branched-From: 68623971be0cfc492a2cb0427d7f478e7b214c24-refs/heads/master@{#403382}

[modify] https://crrev.com/6196d6ec443cd6290a1eff911532d62b52f56d38/cc/test/layer_tree_host_common_test.cc
[modify] https://crrev.com/6196d6ec443cd6290a1eff911532d62b52f56d38/cc/trees/draw_property_utils.cc
[modify] https://crrev.com/6196d6ec443cd6290a1eff911532d62b52f56d38/cc/trees/layer_tree_host_common_unittest.cc

Chrome OS  54.0.2824.0/8696.0.0 samus