1359263 - 2.17ms uninterruptible reflow at receiveMessage@resource://gre/modules/SelectParentHelper.jsm:214:7

Status: NEW

(Core :: Layout: Form Controls, defect, P4)

Description

[Trevor Rowbotham [:rowbot]]

Here's the stack:

receiveMessage@resource://gre/modules/SelectParentHelper.jsm:214:7

Comment 1

[Florian Quèze [:florian]]

http://searchfox.org/mozilla-central/rev/baf47b352e873d4516d7656186d6d7c7447d3873/toolkit/modules/SelectParentHelper.jsm#213
      // Restore scroll position to what it was prior to the update.
      scrollBox.scrollTop = scrollTop;

Comment 3

[Daniel Holbert [:dholbert]]

Do we have any steps to reproduce here?  i.e. how do you trigger the reported error, and where do you see it? (Perhaps the browser console, I'm guessing? I'm not familiar with this "X ms uninterruptible reflow" message, offhand.)

I'm guessing the STR involve triggering a select dropdown -- but does this happen on a particular page, or any page with a select dropdown, or what?  And does it happpen every time, or did you just see this once?

(Depending on how complicated the page in question is & how much has changed since the last layout, 2ms may or may not be an unreasonable amount of time for a forced layout flush -- I'm not sure offhand.

Comment 4

[Florian Quèze [:florian]]

(In reply to Daniel Holbert [:dholbert] (AFK May 3-7, 13-21) from comment #3)

> (Depending on how complicated the page in question is & how much has changed
> since the last layout, 2ms may or may not be an unreasonable amount of time
> for a forced layout flush -- I'm not sure offhand.

The "2ms" value is also not very significant without knowing if it's high end of low end hardware.

Usually in these bugs the question isn't so much "is the amount of time during the uninterruptible reflow reasonable?", but "can we avoid the uninterrupted reflow in this case?".

Comment 5

[Daniel Holbert [:dholbert]]

Thanks for the "see-also" linkage - I'll bet this is indeed largely a duplicate of bug 1118086.

Comment 6

[Trevor Rowbotham [:rowbot]]

Sorry, I don't have STR. I filed this at the end of a browser session and don't know when it occurred since I had the sound disabled. If I find STR, I'll be sure to update this bug.

Comment 7

[Trevor Rowbotham [:rowbot]]

Sorry Florian, I must have missed your question about my hardware. This was recorded on a machine with an Intel Core i7 930 CPU, 12 GB RAM, nvidia GTX 660 GPU, and SSD. The CPU itself is 7 years old, but likely still falls into the high end I would think.

Daniel, the "X ms uninterruptible reflow" message comes from the Oh no! Reflow! addon, which you can find here: https://mikeconley.github.io/ohnoreflow/

Comment 8

[Daniel Holbert [:dholbert]]

Trying to make our layout [qf:p1] bug list understandable/actionable/actually-high-priority.

For this particular bug, I don't think it makes sense to call this [qf:p1], given that:
 (a) The reported ~2ms reflow isn't necessarily an unreasonable duration for a reflow, depending on hardware & content (per comment 4)
 (b) We don't know anything about what page-content triggered this, per comment 6. So we can't really investigate very easily.
 (c) Based on the JS involved (SelectParentHelper.jsm), it seems likely that whatever was going on here would be addressed by the much-more-painful bug 1118086, which has clear STR and which itself is flagged as [qf:p1]

So at this point, I think the right status for this bug is "depends-on/possible-dupe-of bug 1118086" & we should retriage after that bug is fixed.

Comment 9

[Daniel Holbert [:dholbert]]

(Sorry, to avoid getting this back into triage, I should've bumped this to [qf:p3]. Fixing that now.)