Was just reading this Google Doc and noticed the text/letter spacing is inconsistent in Firefox. No issue in Chrome. https://docs.google.com/document/d/1BfIqMcJvmMFdk8e_KKeOEofz16610Mql6gmfeFhrsZk/preview

Not sure if it's specific to Google Docs or if it's just that font

Reminds me of these old Chromium bugs https://bugs.chromium.org/p/chromium/issues/detail?id=707713

https://bugs.chromium.org/p/chromium/issues/detail?id=645055

The Bugbug bot thinks this bug should belong to the 'Core::Layout: Text and Fonts' component, and is moving the bug to that component. Please correct in case you think the bot is wrong.

Thanks for the report! I can't reproduce on Linux, for what it's worth. So, there may potentially be some platform-specific aspect here. What OS are you using?

Oops sorry about that. Windows 10 version 21H2

Thanks! I can reproduce on Windows 10 (in my case with these system display settings in case it matters: 1920x1080 resolution, 100% scaling)

This seems to be a regression, though it's a little while back:
Last good revision: 17e3c9ff2cad9711445a6c39df85925c8f347336 (2020-04-14)
First bad revision: afa247928245d02b1305a1850eb39f19b84a2f35 (2020-04-15)
Pushlog: https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=17e3c9ff2cad9711445a6c39df85925c8f347336&tochange=afa247928245d02b1305a1850eb39f19b84a2f35

In that range, bug 1629787 stands out as a potential culprit, since it's the only commit to mention fonts / font-related stuff that I'm seeing.

Set release status flags based on info from the regressing bug 1629787

:jfkthame, since you are the author of the regressor, bug 1629787, could you take a look?
For more information, please visit auto_nag documentation.

(Also: bug 1770132 is ~similar in symptoms, but it goes back much further; e.g. I can reproduce that one in Nightly 2015-01-01, way further back than the range in comment 4 here.)

(In reply to Daniel Holbert [:dholbert] from comment #4)

Thanks! I can reproduce on Windows 10 (in my case with these system display settings in case it matters: 1920x1080 resolution, 100% scaling)

This seems to be a regression, though it's a little while back:
Last good revision: 17e3c9ff2cad9711445a6c39df85925c8f347336 (2020-04-14)
First bad revision: afa247928245d02b1305a1850eb39f19b84a2f35 (2020-04-15)
Pushlog: https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=17e3c9ff2cad9711445a6c39df85925c8f347336&tochange=afa247928245d02b1305a1850eb39f19b84a2f35

In that range, bug 1629787 stands out as a potential culprit, since it's the only commit to mention fonts / font-related stuff that I'm seeing.

I'd be a bit surprised if bug 1629787 was the actual cause here. Looking at the pushlog again, I wonder if it might be related to bug 1547286. Does setting gfx.canvas.remote to false make any difference?

You're right, it was the gfx.canvas.remote pref-flip. Thought that pref doesn't help in current Nightly -- we had a later change that regressed this even with that pref set to false.

This command...

mozregression.exe --pref "gfx.canvas.remote:false" --good 2020-04-16 -a https://docs.google.com/document/d/1BfIqMcJvmMFdk8e_KKeOEofz16610Mql6gmfeFhrsZk/preview

yields this more-recent regression pushlog for "bad" results here:
https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=21f27a8573cb738731685ad57bcf1ed8be456658&tochange=3ba451ee28d11b0456c319676dafa846a91523c7
which is "Bug 1730772: Do not use GDI classic rendering when rendering Canvas. r=jfkthame,jrmuizel" and does seem to be mostly font-related changes.

So: in Windows Nightlies, this initially regressed via the pref-flip in bug 1547286, and (if you manually flip that pref to false) it regressed later via bug 1730772.

Set release status flags based on info from the regressing bug 1547286