Opened 10 years ago

Closed 10 years ago

#2194 closed defect (fixed)

Performance problem: new channel replacement code seems to chew CPU

Reported by: pde Owned by: pde
Priority: Medium Milestone:
Component: HTTPS Everywhere/EFF-HTTPS Everywhere Version:
Severity: Keywords:
Cc: noncombatant, mikeperry Actual Points:
Parent ID: Points:
Reviewer: Sponsor:

Description

Both Chris Palmer and I have the subjective fear that firefox is more prone to runaway CPU consumption since we merged mikeperry's branch incorporating Giorgio's channel replacement fixes from here:

https://bugzilla.mozilla.org/show_bug.cgi?id=548102

Child Tickets

Change History (2)

comment:1 Changed 10 years ago by pde

This is a really hard bug to confirm. Right now, my browsers seem to use 8-15% cpu. High, but not insane. I think I saw closer to 100% when I first merged those fixes from mikeperry/giorgio, but my memory is untrustworthy. Chris, what do you see with 0.3.0.development.1?

comment:2 Changed 10 years ago by pde

Resolution: fixed
Status: newclosed

This seems to have been a simple performance bottleneck, possibly triggered by expanding rule scope. It seems to have been fixed by the efficient rulesets lookup table.

Note: See TracTickets for help on using tickets.