New issue
Advanced search Search tips
Note: Color blocks (like █ or █) mean that a user may not be available. Tooltip shows the reason.

Issue 807108 link

Starred by 1 user
Status: Assigned
Owner:
█ hjd@chromium.org
Last visit > 30 days ago
Cc:
erikc...@chromium.org
carlosk@chromium.org
primiano@chromium.org
█ picksi@chromium.org
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug



Sign in to add a comment

Private memory footprint should discount tracing overhead in chrome://tracing UI

Project Member Reported by ssid@chromium.org, Jan 30 2018 
Background context: go/memory-infra

For local debugging using memory-infra the value shown by "Private footprint" column includes overhead of memory infra.
We discount resident size of tracing from the Private dirty and resident size columns, and same should be done for private footprint column.

Comment 1 by ssid@chromium.org, Jan 30 2018 

I think this should not matter for the UMA calculations since the overhead of memory-infra is very small and we take only one global dump at a time.
In local debugging we could collect multiple memory dumps or we use heap profiling and the overhead accounts for significant value.

Comment 2 by erikc...@chromium.org, Jan 30 2018

Summary: Private memory footprint should discount tracing overhead in chrome://tracing UI (was: Private memory footprint should discount tracing overhead)

Comment 3 by primiano@chromium.org, Feb 1 2018

Owner: hjd@chromium.org
Status: Assigned (was: Untriaged)
Oh yes I forgot about this when we introduced PMF. We should definitely fix the discount. This is purely a change in TraceViewer, which will auto-propagate to both telemetry and UI.

assigning to hjd@, not urgent but at the same time this should be a (hopefully) minor change
►

Sign in to add a comment