All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=805585

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=5a680b863c0f9b30f9dea6cb92b5d2eda5faeea23a8621211cb014cd34ca6bc9


Bot(s) for this bug's original alert(s):

chromium-rel-mac-retina
chromium-rel-mac11
chromium-rel-mac11-air
chromium-rel-mac11-pro
chromium-rel-mac12
chromium-rel-win10
chromium-rel-win7-dual
chromium-rel-win7-gpu-ati
chromium-rel-win7-x64-dual
chromium-rel-win8-dual

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14c01cac840000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/14c01cac840000

cc/ipc: Don't reject CFs in case of a filter serialization error.
By khushalsagar@chromium.org · Tue Jan 23 02:17:46 2018
chromium @ fb25d8a626538f7dd54f3efb381db9fe20927650

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

Huh. The frame is taking ~10ms to serialize now, as compared to 1ms earlier. This change didn't affect anything in the filter's serialization itself, just how the data is represented in mojom. I'll try to do more tracing to narrow this down.

It's dubious that such a small change would have a 10x impact on
serialization cost.

Fix in review: https://chromium-review.googlesource.com/c/chromium/src/+/885511. This will also need a merge to 65.

Pls apply appropriate OSs label. Also apply M-65 label if it applicable to M65.
Please request a merge to M65 once cl lands and it looks good in canary and wait for approval. Thank you.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/793792516e90a2f694f874949525be0b1f079da4

commit 793792516e90a2f694f874949525be0b1f079da4
Author: Khushal <khushalsagar@chromium.org>
Date: Sat Jan 27 16:46:17 2018

cc/ipc: Fix PaintFilter serialization.

Ensure that the memory returned when serializing a filter is resized to
the number of bytes written. Otherwise the message serializes the
complete buffer allocated initially.

R=rockot@chromium.org, tsepez@chromium.org

Bug:  805585 
Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel
Change-Id: Ia85688cbd77529c2f6e9c044de9a83a254787442
Reviewed-on: https://chromium-review.googlesource.com/885511
Reviewed-by: enne <enne@chromium.org>
Reviewed-by: Tom Sepez <tsepez@chromium.org>
Reviewed-by: Ken Rockot <rockot@chromium.org>
Commit-Queue: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532228}
[modify] https://crrev.com/793792516e90a2f694f874949525be0b1f079da4/cc/paint/paint_op_reader.h
[modify] https://crrev.com/793792516e90a2f694f874949525be0b1f079da4/services/viz/public/cpp/compositing/paint_filter_struct_traits.h

khushalsagar@, pls apply appropriate OSs label and request a merge to M65 once change listed at #8 is well baked/verified in canary.

Does this also need merge to M64?

No, it should only be M65. The change that resulted in this regression was merged to 65 in issue 803777. I'll wait for feedback from canary and ensure the bots have recovered before requesting a merge.

And looks like all the bots have recovered.

Nothing bad from canary so far. 

Adding OS labels. The code path affected by this change affects all platforms.

friendly ping for merge approval.

Your change meets the bar and is auto-approved for M65. Please go ahead and merge the CL to branch 3325 manually. Please contact milestone owner if you have questions.
Owners: cmasso@(Android), cmasso@(iOS), bhthompson@(ChromeOS), govind@(Desktop)

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

Pls merge your change to M65 branch 3325 ASAP so we can pick it up for next M65 dev release. Thank you.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/4da89adcd6990d1b43f9e773f0cfaa06efa0dd14

commit 4da89adcd6990d1b43f9e773f0cfaa06efa0dd14
Author: Khushal <khushalsagar@chromium.org>
Date: Tue Jan 30 22:01:10 2018

cc/ipc: Fix PaintFilter serialization.

Ensure that the memory returned when serializing a filter is resized to
the number of bytes written. Otherwise the message serializes the
complete buffer allocated initially.

R=​rockot@chromium.org, tsepez@chromium.org

Bug:  805585 
Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel
Change-Id: Ia85688cbd77529c2f6e9c044de9a83a254787442
Reviewed-on: https://chromium-review.googlesource.com/885511
Reviewed-by: enne <enne@chromium.org>
Reviewed-by: Tom Sepez <tsepez@chromium.org>
Reviewed-by: Ken Rockot <rockot@chromium.org>
Commit-Queue: Khushal <khushalsagar@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#532228}(cherry picked from commit 793792516e90a2f694f874949525be0b1f079da4)
Reviewed-on: https://chromium-review.googlesource.com/894270
Reviewed-by: Khushal <khushalsagar@chromium.org>
Cr-Commit-Position: refs/branch-heads/3325@{#185}
Cr-Branched-From: bc084a8b5afa3744a74927344e304c02ae54189f-refs/heads/master@{#530369}
[modify] https://crrev.com/4da89adcd6990d1b43f9e773f0cfaa06efa0dd14/cc/paint/paint_op_reader.h
[modify] https://crrev.com/4da89adcd6990d1b43f9e773f0cfaa06efa0dd14/services/viz/public/cpp/compositing/paint_filter_struct_traits.h