802777 - [B2G] Lower image.mem.max_decoded_image_kb

Status: RESOLVED FIXED

(Firefox OS Graveyard :: General, defect)

Description

[Justin Lebar (not reading bugmail)]

image.mem.max_decoded_image_kb tracks the amount of decoded image data we "willingly" keep around.

Even if you set this value to 0, we often unwillingly keep around a lot more than this.  For example, all images on a page are always decoded (bug 689623).

But if we're using more than image.mem.max_decoded_image_kb memory for decoded image data and we are able to throw any of our decoded data, we'll do so.

Comment 1

[Justin Lebar (not reading bugmail)]

Attachment: Patch, v1

Comment 2

[Justin Lebar (not reading bugmail)]

I can't ask for approval-aurora on patches in the b2g component, so I guess I have to ask for blocking.

Comment 3

[Justin Lebar (not reading bugmail)]

Thanks, Joe.

https://hg.mozilla.org/integration/mozilla-inbound/rev/bd5e49f6d2fa

Comment 4

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/bd5e49f6d2fa

Comment 5

[Justin Lebar (not reading bugmail)]

https://hg.mozilla.org/releases/mozilla-aurora/rev/84573f69513c

Comment 6

[Timothy Nikkel (:tnikkel)]

(In reply to Justin Lebar [:jlebar] from comment #0)
> Even if you set this value to 0, we often unwillingly keep around a lot more
> than this.  For example, all images on a page are always decoded (bug
> 689623).

Doesn't pref("content.image.allow_locking", false); on b2g mean this isn't true for b2g?

Comment 7

[Justin Lebar (not reading bugmail)]

(In reply to Timothy Nikkel (:tn) from comment #6)
> (In reply to Justin Lebar [:jlebar] from comment #0)
> > Even if you set this value to 0, we often unwillingly keep around a lot more
> > than this.  For example, all images on a page are always decoded (bug
> > 689623).
> 
> Doesn't pref("content.image.allow_locking", false); on b2g mean this isn't
> true for b2g?

Correct.