981257 - We're running out of node bits

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, defect)

Description

[Boris Zbarsky [:bzbarsky]]

We have two bool bits left, and only one non-bool bit.  That's not counting the event state, which we have 18 free bits in, if we want to try using them somehow, of course.

Looking at our existing bits, we can nix ParentIsContent and replace its one use with:

  nsIContent* GetParent() const {
    return mParent && mParent->IsContent() ? mParent : nullptr;
  }

That does mean reading stuff from the parent, which may mean more cache misse, but the caller was probably going to anyway.  Worth doing that?

Any other bits we can obviously kill off?

Or should I just not be worrying about this until the next poor sod tries to add flags and gets compile errors?

Comment 1

[Kyle Huey (Exited; not receiving bugmail, old account, do not use)]

Killing XBL would free up several ...

Is directionality important enough to chew up two of them?  

NODE_ATTACH_BINDING_ON_POSTCREATE appears to be unused.  That's almost too good to be true.

Comment 2

[Boris Zbarsky [:bzbarsky]]

> Killing XBL would free up several ...

Yes, but that's not a short-term project.

> Is directionality important enough to chew up two of them?  

Directionality uses two non-boolean flags and 6 boolean flags.  I couldn't find obvious ways to nix some of them.  :(

> NODE_ATTACH_BINDING_ON_POSTCREATE appears to be unused.

Oho!  It is in fact unused.  It went away in bug 883892.  We used to set the flag in PreCreate (because we had to check for XBL there to decide whether we can do slimwrappers) and check it in PostCreate (to avoid redoing all that checking work), but now Element::WrapObject just does all the work involved, so there's no need to smuggle state along.

Comment 3

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Attachment: Patch v1

Comment 4

[Kyle Huey (Exited; not receiving bugmail, old account, do not use)]

Comment on attachment 8388060 [details] [diff] [review]
Patch v1

I would prefer that bz reviewed this, mostly because I'm not sure if there's anywhere else that depends on the numbers involved here.

Comment 5

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 8388060 [details] [diff] [review]
Patch v1

This should be good.  r=me

Comment 6

[Boris Zbarsky [:bzbarsky]]

ms2ger, do you plan to land that?

Comment 7

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Can we wait to land it until I'm ready to land bug 996796?

Comment 8

[Boris Zbarsky [:bzbarsky]]

Sure.  You're the main consumer for this patch right now!

Comment 9

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

I forgot about this... Let me know when/if I should land it.

Comment 10

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/267e2477f763

Comment 11

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

https://hg.mozilla.org/mozilla-central/rev/267e2477f763