960899 - add debug function and lldb command to print frame state symbolically

Status: RESOLVED FIXED

Description

[Cameron McCormack (:heycam)]

Created attachment 8361528 [details] [diff] [review]
patch

Comment 1

[Cameron McCormack (:heycam)]

Created attachment 8361531 [details] [diff] [review]
patch v1.1

Add some docs to python/lldbutils/README.txt too.

Comment 2

[David Baron :dbaron: 🏴󠁵󠁳󠁣󠁡󠁿 ⌚UTC-8 (if account gets disabled due to email bounces, ask a bugzilla admin to reenable it)]

Comment on attachment 8361531 [details] [diff] [review]
patch v1.1

Please actually test that:
  T* t = do_QueryFrame(f); 
fails if T isn't *itself* a queryframe target?  (It looks like we
have the necessary static_asserts in nsQueryFrame.h)

r=dbaron

Comment 3

[Cameron McCormack (:heycam)]

Yes the static_assert fires, which is how I found I needed to add the couple of queryframe targets in the patch.

Comment 4

[Cameron McCormack (:heycam)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/1f72556326be

Comment 5

[Cameron McCormack (:heycam)]

Backed out in https://hg.mozilla.org/integration/mozilla-inbound/rev/334a63e35b74 since bug 960848 got backed out.

Comment 6

[Timothy Nikkel (:tnikkel)]

This is great! It'd be even greater to use this in frame tree dumps (nsIFrame::List)!

Comment 7

[Cameron McCormack (:heycam)]

We could do that although it will make the line lengths even longer.  At the moment, with a modest frame depth, I can maximise my terminal and not get (many) line wraps when I dump a frame tree.

Comment 8

[Timothy Nikkel (:tnikkel)]

I had given up on making frame tree dumps not wrap, so more information was only a plus for me.

Comment 9

[Cameron McCormack (:heycam)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/191a71ff02bf

tn, can you file a separate bug for sticking this in nsIFrame::List?

Comment 10

[Phil Ringnalda (:philor)]

Backed out in https://hg.mozilla.org/integration/mozilla-inbound/rev/ec60c69dafc2 for https://tbpl.mozilla.org/php/getParsedLog.php?id=34109972&tree=Mozilla-Inbound

Comment 11

[Cameron McCormack (:heycam)]

https://tbpl.mozilla.org/?tree=Try&rev=34b613903d9a

https://hg.mozilla.org/integration/mozilla-inbound/rev/8eca8d49dcb1

Comment 12

[Timothy Nikkel (:tnikkel)]

(In reply to Cameron McCormack (:heycam) from comment #9)
> https://hg.mozilla.org/integration/mozilla-inbound/rev/191a71ff02bf
> 
> tn, can you file a separate bug for sticking this in nsIFrame::List?

After I mentioned that I realized that it's not very often I need to look at frame state lately, so if it interferes with others' workflows then I don't feel strongly about it.

Comment 13

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/8eca8d49dcb1