975434 - Remove the ability to cancel memory minimizations

Status: RESOLVED FIXED

(Firefox OS Graveyard :: General, defect)

Description

[Gabriele Svelto [:gsvelto]]

+++ This bug was initially created as a clone of Bug #971728 +++

In bug 814771 we had introduced the ability to cancel a memory minimization and used it to interrupt the process when the homescreen was quickly sent to the background and then to the foreground again. Now that bug 971728 has landed we don't use this functionality anymore so let's remove it as it's just become unnecessary complexity in the relevant code.

Comment 1

[Gabriele Svelto [:gsvelto]]

Attachment: [PATCH] Remove the ability to cancel memory minimizations

Comment 2

[Gabriele Svelto [:gsvelto]]

Comment on attachment 8381435 [details] [diff] [review]
[PATCH] Remove the ability to cancel memory minimizations

This effectively removes the changes introduced in bug 814771. The ability to cancel memory minimizations is not needed anymore now that we're using regular memory-pressure events for this task, removing it simplifies the code which IMHO is a good thing.

The try run is here:

https://tbpl.mozilla.org/?tree=Try&rev=1525dfd4e28f

Comment 3

[Nicholas Nethercote [inactive]]

Comment on attachment 8381435 [details] [diff] [review]
[PATCH] Remove the ability to cancel memory minimizations

Review of attachment 8381435 [details] [diff] [review]:
-----------------------------------------------------------------

::: dom/ipc/ContentChild.cpp
@@ +1511,5 @@
>      }
>  #endif
>      nsCOMPtr<nsIMemoryReporterManager> mgr =
>          do_GetService("@mozilla.org/memory-reporter-manager;1");
>      NS_ENSURE_TRUE(mgr, true);

This function always returns true, so if you want to change it to return void that'd be fine by me.

Comment 4

[Gabriele Svelto [:gsvelto]]

(In reply to Nicholas Nethercote [:njn] from comment #3)
> This function always returns true, so if you want to change it to return
> void that'd be fine by me.

The prototype for that method is generated from IPDL and IIRC it's not possible to force it to return void. It's a bit silly but we'll have to live with it. Thanks for the review!

Comment 5

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/a21fc292b2ea

Comment 6

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/mozilla-central/rev/a21fc292b2ea