This is an stock version of AdBlock Plus 3.5.2 installed on the 2018 reference. Here's a profile:

https://perfht.ml/2YaBdeL

We should see if there's anything that can be done to bring this down a bit - especially considering how many AdBlock Plus users there are.

Hey Caitlin, do we have a contact at ABP we can flag to take a look perhaps?

Unfortunately I don't have a direct contact name, but I just emailed their support channel and linked to this bug. Hopefully they will look into it.

Maybe we can get a contact from AMO?

Or https://eyeo.com/contact/.

We are going to investigate this. I am trying to figure out who would be the right Eyeo contact person in the future.

https://gitlab.com/eyeo/adblockplus/adblockpluschrome/issues/46

Thanks for letting us know. Manish is working on this, please see https://gitlab.com/eyeo/adblockplus/adblockpluscore/issues/53

Note that this doesn't seem to involve indexeddb beyond the fact that that's where the data came from. That is, the data is read, and after that happens, we spend an additional ~4s processing it. It looks like ABP folks are picking this up, and at least at the moment, it doesn't look like there's much for Firefox to do so I'm going to take this out of the fxperf list - though we will obviously reconsider fxperf priority if there turns out to be a significant issue with how the platform supports what's happening here.

Aside: seems BMO doesn't support adding gitlab issues as "see also" references...

(In reply to :Gijs (he/him) from comment #7)

Aside: seems BMO doesn't support adding gitlab issues as "see also" references...

I filed bug 1575003 for this.

outreach complete.