regression: Web content does not adapt to window size when layout.css.devPixelsPerPx changes
Categories
(Core :: Layout, defect, P3)
Tracking
()
People
(Reporter: jan, Assigned: hiro)
References
Details
(Keywords: correctness, nightly-community, regression)
Attachments
(2 files)
Win10, GTX 1060, 2560x1440
Screencast:
- Run
mozregression --launch 2019-08-10 --pref layout.css.devPixelsPerPx:'0.7' browser.aboutConfig.showWarning:false browser.startup.homepage:'https://bugzilla.mozilla.org/show_bug.cgi?id=wr-70|about:config'
- Switch to about:config tab and reset layout.css.devPixelsPerPx to default.
- Switch back to Bugzilla tab and see the bug. Maximizing the window or reloading the page does not help.
Linux is affected with Basic, OpenGL and WebRender as well. So it might be a Layout issue?
Debian Testing, KDE (compositor disabled), X11, Macbook Pro A1502
mozregression --good 2017-08-01 --bad 2018-01-15 --pref layout.css.devPixelsPerPx:1.5 browser.aboutConfig.showWarning:false general.warnOnAboutConfig:false browser.startup.homepage:'https://bugzilla.mozilla.org/show_bug.cgi?id=wr-70|about:config?filter=layout.css.devPixelsPerPx'
7:03.63 INFO: Last good revision: 40df5dd35fdb7ce3652fe4448ac8961c075c928e (2017-11-07)
7:03.63 INFO: First bad revision: c10cf7082a9b84839e7c3fdda80bd468e091739a (2017-11-08)
7:03.63 INFO: Pushlog:
https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=40df5dd35fdb7ce3652fe4448ac8961c075c928e&tochange=c10cf7082a9b84839e7c3fdda80bd468e091739a7:41.18 INFO: There are no build artifacts on inbound for these changesets (they are probably too old).
Disabling layout.css.servo.enabled and layout.display-list.retain for first bad build does not help.
Win10 has the same regression range.
Reporter | ||
Updated•6 years ago
|
Reporter | ||
Comment 1•6 years ago
|
||
The same problem might affect window content as a whole on startup.
Reporter | ||
Comment 2•6 years ago
|
||
Moving to CSS because it's a CSS pref.
Comment 3•6 years ago
|
||
The priority flag is not set for this bug.
:hiro, could you have a look please?
For more information, please visit auto_nag documentation.
Updated•6 years ago
|
Comment 4•6 years ago
|
||
Thanks for the report! Triaging as P3 for now, since we've apparently been shipping with this issue without noticing for ~2 years now.
(definitely would be nice to fix / get to the bottom of it, if someone gets a chance!)
Updated•6 years ago
|
Comment 5•6 years ago
|
||
(Also, this is more appropriately classified as Layout rather than CSS. Despite .css
in the pref-name, the actual implementation of this pixel scaling-factor is closer to layout code than it is to our css parser.)
Comment 6•6 years ago
|
||
Sean, we're making this as wontfix for 70 but affected for 71 in case you feel this should be included on your list.
Updated•6 years ago
|
Updated•6 years ago
|
Updated•6 years ago
|
Comment 7•5 years ago
|
||
I have this issue on Mac, as well.
Environment:
iMac with macOS Mojave
FF version: 75
STR:
- open a popup in a new window
- set the layout.css.devPixelsPerPx pref to 2
- check the popup window.
Expected:
Content covers all window.
Actual:
Content is displayed only on a section of the window.
Note: If I move the window or resize it, the content is displayed as expected.
Comment 8•5 years ago
|
||
Updated•5 years ago
|
Updated•3 years ago
|
Comment 9•3 years ago
|
||
The severity field for this bug is relatively low, S3. However, the bug has 5 See Also bugs.
:dholbert, could you consider increasing the bug severity?
For more information, please visit auto_nag documentation.
Assignee | ||
Comment 10•3 years ago
|
||
I did recently change relevant code in bug 1787079, is this still an issue? I haven't tried the Mac's popup case in comment 7, to me the STR in comment 0 doesn't reproduce.
Assignee | ||
Comment 11•3 years ago
|
||
Confirmed the popup window case works as expected.
Assignee | ||
Comment 12•3 years ago
|
||
I am pretty sure https://hg.mozilla.org/integration/autoland/rev/03f45da3ba4e changeset fixed the problem. I am unsure which changeset in the regression range in comment 0 regressed this. (I did skim the list twice, but couldn't find any suspicious change)
Updated•3 years ago
|
Description
•