All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=896277

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=e2332907301cfdd59801c3d4cb16a9fd0d8eb9d7584b1d7051e0b59a598cd7a4


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

linux-perf

system_health.memory_desktop - Benchmark documentation link:
  https://bit.ly/system-health-benchmarks

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/16a637bce40000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/16a637bce40000

Disallow libjpeg_turbo downscaling of images with non-whole MCUs. by andrescj@chromium.org
https://chromium.googlesource.com/chromium/src/+/8b9172a61f4e507b2eef30b2af438d2c19c0992d
memory:chrome:all_processes:reported_by_chrome:skia:effective_size: 3.631e+05 → 8.241e+05 (+4.61e+05)

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

Benchmark documentation link:
  https://bit.ly/system-health-benchmarks

Same justification as  https://crbug.com/897115 : the memory regression is expected because my CL effectively (and deliberately) negates some of the benefits of https://chromium-review.googlesource.com/c/chromium/src/+/1102025: some JPEG images need to be decoded to full size instead of a smaller size in order to avoid artifacts as described in  https://crbug.com/890745 .

To investigate this, I effectively reverted both:

- https://chromium-review.googlesource.com/c/chromium/src/+/1274845 (my CL)
- https://chromium-review.googlesource.com/c/chromium/src/+/1102025 (the earlier CL negated by mine)

Then, I ran a system_health.memory_desktop Pinpoint job using this revert for the bots and stories listed in https://chromeperf.appspot.com/group_report?bug_id=896277. The results for memory:chrome:all_processes:reported_by_chrome:skia:effective_size are:

Bot          Story                                        Diff.
===============================================================
linux-perf   load:news:qq:2018                            +0.9%
linux-perf   browse:news:nytimes:2018                     +0.3%
linux-perf   browse:social:twitter:infinite:scroll:2018   -1.5%
linux-perf   load:tools:weather                           +0.1%

A positive % indicates a regression with respect to the revert, but none of the regressions were marked as significant.

Furthermore, with respect to the revert, for linux-perf, the difference for memory:chrome:all_processes:reported_by_chrome:skia:effective_size is -2.3%.

So, this suggests that at worst, my new CL simply takes us to the state before the CL that triggered the artifacts in  https://crbug.com/890745 .

There were other regressions marked as significant with respect to the revert (although significance tends to change between Results2 refreshes). The only ones that caught my attention due to a large percentage were:

memory:chrome:unknown_processes:reported_by_os:system_memory:stack:private_dirty_size (+72.1%)
memory:chrome:unknown_processes:reported_by_os:system_memory:stack:proportional_resident_size (+70.3%)

I re-ran the linux-perf job which did not reproduce these regressions. Instead I got different outstanding ones (which were not reproduced in the first run):

memory:chrome:browser_process:reported_by_os:system_memory:stack:private_dirty_size (+64.2%)
memory:chrome:browser_process:reported_by_os:system_memory:stack:proportional_resident_size (+64.2%)

So, I'll assume this reported_by_os:system_memory:stack:{private_dirty_size, proportional_resident_size} family is noisy.

For reference:

CL with the reverts: https://chromium-review.googlesource.com/c/chromium/src/+/1305495
linux-perf all results: https://pinpoint-dot-chromeperf.appspot.com/job/16d9f3bde40000
linux-perf load:news:qq:2018 results: https://pinpoint-dot-chromeperf.appspot.com/job/139a6e4de40000
linux-perf browse:news:nytimes:2018 results: https://pinpoint-dot-chromeperf.appspot.com/job/14eb6499e40000
linux-perf browse:social:twitter:infinite:scroll:2018 results: https://pinpoint-dot-chromeperf.appspot.com/job/10d35d11e40000
linux-perf load:tools:weather results: https://pinpoint-dot-chromeperf.appspot.com/job/12259086e40000
linux-perf all results (second run): https://pinpoint-dot-chromeperf.appspot.com/job/10967fd9e40000