1056116 - No HWA-supported font-rendering on IntelHD since 2014-08-14 (Optimus)

Status: RESOLVED INCOMPLETE

(Core :: Graphics, defect)

Description

[Elbart]

Attachment: intelhd.png

Open URL and notice different font-rendering in Nightlies of 08-14 and newer compared to 08-13 and older.

Regression-range:

Last good revision: 7fc96293ada8 (2014-08-13)
First bad revision: 5299864050ee (2014-08-14)
Pushlog:
https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=7fc96293ada8&tochange=5299864050ee

Last good revision: 81c93f4a6212
First bad revision: 043edc92814e
Pushlog:
https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=81c93f4a6212&tochange=043edc92814e

9c339e14c720	David Parks — Bug 1008759 - Blocklist D2D on NVidia 310M, even when it is believed to be the secondary GPU. r=bjacob

Note: I don't understand the reasoning behind blocking the 310m by Firefox in the first place. The nVidia-driver's default setting for Firefox is to use the integrated GPU, meaning IntelHD. The user has to manually override this restriction, which obviously isn't recommended, otherwise blocking the GPU wouldn't be the default setting in the first place. And, imho, Firefox was running even slower on the nVidia-GPU than on the IntelHD.

Why caring about such a fringe case?

Comment 1

[Elbart]

Attachment: about-support 0813.txt

Comment 2

[Elbart]

Attachment: about-support 0814.txt

Comment 3

[Bas Schouten (:bas.schouten)]

(In reply to Elbart from comment #0)
> Created attachment 8475912 [details]
> intelhd.png
> 
> Open URL and notice different font-rendering in Nightlies of 08-14 and newer
> compared to 08-13 and older.
> 
> Regression-range:
> 
> Last good revision: 7fc96293ada8 (2014-08-13)
> First bad revision: 5299864050ee (2014-08-14)
> Pushlog:
> https://hg.mozilla.org/mozilla-central/
> pushloghtml?fromchange=7fc96293ada8&tochange=5299864050ee
> 
> Last good revision: 81c93f4a6212
> First bad revision: 043edc92814e
> Pushlog:
> https://hg.mozilla.org/integration/mozilla-inbound/
> pushloghtml?fromchange=81c93f4a6212&tochange=043edc92814e
> 
> 9c339e14c720	David Parks — Bug 1008759 - Blocklist D2D on NVidia 310M, even
> when it is believed to be the secondary GPU. r=bjacob
> 
> Note: I don't understand the reasoning behind blocking the 310m by Firefox
> in the first place. The nVidia-driver's default setting for Firefox is to
> use the integrated GPU, meaning IntelHD. The user has to manually override
> this restriction, which obviously isn't recommended, otherwise blocking the
> GPU wouldn't be the default setting in the first place. And, imho, Firefox
> was running even slower on the nVidia-GPU than on the IntelHD.
> 
> Why caring about such a fringe case?

The problem is the exact problematic scenario could not be detected in this case. And there was crashes coming in with this usage pattern. In that situation our hands are sort of tied, I'll leave it to bjacob to decide if we're hurting enough users to consider reverting this blacklist entry.

Comment 4

[Benoit Jacob [:bjacob] (mostly away)]

This was crashing e10s, which is why David cared. I'm not the one to decide how to weigh e10s crashiness against font quality on Windows. It's a conversation for David and Bas to have, I suppose. Maybe I should have forwarded the review request on bug 1008759 to Bas.

Comment 5

[BugBot [:suhaib / :marco/ :calixte]]

In the process of migrating remaining bugs to the new severity system, the severity for this bug cannot be automatically determined. Please retriage this bug using the new severity system.

Comment 6

[BugBot [:suhaib / :marco/ :calixte]]

The severity field is not set for this bug.
:bhood, could you have a look please?

For more information, please visit BugBot documentation.

Comment 7

[Bob Hood [:bhood]]

Elbart, is this still an issue with the current release (116 as of this writing) of Firefox?

Comment 8

[BugBot [:suhaib / :marco/ :calixte]]

Redirect a needinfo that is pending on an inactive user to the triage owner.
:bhood, since the bug has recent activity, could you please find another way to get the information or close the bug as INCOMPLETE if it is not actionable?

For more information, please visit BugBot documentation.