More stacktraces from an other computer.

Deleted: bug1.png 54.6 KB

	bug1.png 54.6 KB View Download

Deleted: bug2.png 60.5 KB

	bug2.png 60.5 KB View Download

Deleted: bug3.png 63.1 KB

	bug3.png 63.1 KB View Download

This leak is still showing up in recent trace:

os-arch: "x86_64",
os-name: "Windows NT",
revision: "cdd15784955039742fe9a8235581922d41b82d78-refs/heads/master@{#492239}",
product-version: "Chrome/62.0.3178.0",

See attached file.

Deleted: bug1.png 53.9 KB

	bug1.png 53.9 KB View Download

There is about 68K objects leaked, and 10 Megabytes leaked.

reveman@ can you take a look or re-assign.

ping?

Discardable memory is by design not supposed to be deleted unless needed so I'm not sure this is a real issue. Ca you explain more how this extension works?

I believe the leak is not "discardable memory" but objects used to keep track of the discardable memory used.

To clarify, the remaining objects are the amount allocated objects with the specific stackframe that are allocated between two points (beginning of tracing / end of tracing) and still exist at the end.

The experiment ran for more than a week. Which means that over the week, there are about 68K objects, accounting for 2.5M that is 'potentially; leaked. It's a potential leak because it's possible that the 68K objects are alive and useful.

To validate that this is a leak, will should ensure that there is no longer shared-memory used, and that these objects still exists?

This regression has been open for half a year. It's not very actionable and the regression has been in all Chrome user's hands for months.