Using the per-revision bisect providing the bisect results,
Good build: 56.0.2900.0 (Revision: 427219).
Bad build: 56.0.2902.0 (Revision: 427892).

You are probably looking for a change made after 427365 (known good), but no later than 427366 (first known bad).
CHANGELOG URL:
The script might not always return single CL as suspectas some perf builds might get missing due to failure.
  https://chromium.googlesource.com/chromium/src/+log/ba26074e724e522df676e0d3133bbe5f3f8d28b4..61fa0fb418a3f77665e0bc1be88ec5a2984c6dc6

@jaydasika -- Could you please look into the issue, pardon me if it has nothing to do with your changes and if possible please assign it to concern owner.
Thank You.

I have verified that https://chromium.googlesource.com/chromium/src/+log/ba26074e724e522df676e0d3133bbe5f3f8d28b4..61fa0fb418a3f77665e0bc1be88ec5a2984c6dc6 is not the cause for this bug

Uh, don't leave a blank status.

jaydasika: Were you able to reproduce the bug? If yes, can you help do a quick bisect to figure out who the possible suspects are? I don't have a Retina Mac here. Otherwise we can wait for QA team to do another round of bisecting.

Did a bisect again. The bisect script is pointing to my CL.
But, I have locally reverted it and the bug is reproducible even on top of the revert. I am not sure what's going on.

Can you try syncing your local build back to r427366, and then try r427365 for comparison? If you are way past r427366, there could be that some other CL that also affects the rendering and thus negating your CL. You'll have to try some local builds with your CL rolled back to see what's interacting with what.

With response to comment #2:
Re-bisected and got the following narrow bisect:
https://chromium.googlesource.com/chromium/src/+log/3199bb68eb9c51ba1f67e75c49050da47a52e51d..579a409d89d9da9a10a924e3d4704fc1e1144d77?pretty=fuller&n=100

Suspecting: r427366 ?

Verified that the bug got introduced first when crrev.com/2408243002 landed. But, the bug persists after that patch got reverted in crrev.com/2452303003/. So, some CL between those two builds also causes the bug. I am trying to bisect that now. (I am having issues with goma on my Mac and so the bisecting process has been painfully slow)
If you know any potential culprits between those two build, let me know.

Could this be related to  issue 663499 ?

https://chromium.googlesource.com/chromium/src.git/+/b433552b9bfc3ade27dfe7a4b95496f609ba4c35 is the culprit.
Re #9 : Could be related.

This bug got accidentally fixed in https://codereview.chromium.org/2163823003 (CL1). That CL has a bug which inflates the clip rects. Now, the inflation bug is fixed by https://codereview.chromium.org/2408243002 (CL2) and that re-introduces this bug.

I can see this bug in M53. CL1 is a part of M54, so I don't see it in M54 and I see it again in M55 as CL2 is in it.

dsinclair : Do you think this is a PDFium issue ? If yes, please assign it to the right owner.
 

Still able to reproduce the issue on mac 10.11.6 using chrome version 57.0.2926.0

dsinclair@, Could you please respond as per comment #11

I'm confused, if the two CLs are in CC that fix/break the code, isn't that where the problem is?

Just to update:

Still able to reproduce the issue on mac 10.11.6 using chrome version 57.0.2935.0

Re #13: The CL that fixed it was calculating inflated clip rects in CC and that fixed it. I think the actual bug is in blink/PDFium in setting the layer's bounds (which CC uses to compute clip rects)

jaydasika@, could you work w/ dsinclair@ to see what the correct bounds blink/PDFium should have set?

Just to update:

Still able to reproduce the issue on mac retina 10.12.1 using chrome latest version 57.0.2953.0

@jaydasika : Could you please look in to it.

Thank you!

Just to update:

Still able to reproduce the issue on mac retina 10.12.1 using chrome latest version 57.0.2969.0

@jaydasika : Could you please look in to it.

Thank you!

As I said in comment 15, I think this bug is in blink/Pdfium, I will try to figure out where exactly it is and pass it own to the right owner.

jaydasika@ - Could you please provide any update on this issue.

Thanks...!!

Still able to reproduce this issue on Mac 10.12.2 using latest canary #57.0.2986.0.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/44b074ebba30c2f7c20c70f6a3a6a95ebc341037

commit 44b074ebba30c2f7c20c70f6a3a6a95ebc341037
Author: jaydasika <jaydasika@chromium.org>
Date: Tue Mar 07 21:11:38 2017

cc: Add a helper function for transforming rects between surface spaces

And use it everywhere we convert rects in draw_property_utils.cc between
surface spaces. Using this helper function also fixes a bug in
caclulating render surface's clip rect.

BUG= 663309 
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2734983004
Cr-Commit-Position: refs/heads/master@{#455222}

[modify] https://crrev.com/44b074ebba30c2f7c20c70f6a3a6a95ebc341037/cc/trees/draw_property_utils.cc
[modify] https://crrev.com/44b074ebba30c2f7c20c70f6a3a6a95ebc341037/cc/trees/layer_tree_host_common_unittest.cc