The GPU Acceleration Situation In Firefox 6

[Bluefang]

The Mozilla wiki has detailed info on the blacklists for each platform & Firefox version.

https://wiki.mozilla.org/Blocklisting/Blocked_Graphics_Drivers

[hal2k1]

The bug Firefox had with the Gallium drivers was within Firefox itself. They tracked down and fixed that bug, but didn't want to change the whitelist in FF6 without a full round of testing first.

FF7 will still only require Mesa 7.10.3, so the drivers have been good at least since that point. If I recall, the Gallium drivers were fine in 7.10.0 as well, it was some bug affecting both classic and Gallium that made them bump it to 7.10.3.

Edit: the actual bugfix may have been ported from FF7 into 6, so if you run FF6 and bypass the whitelist it may work correctly. But i'm not 100% sure whether that bug got fixed there or not.

This is a test from Mozilla of browser hardware acceleration:

http://demos.hacks.mozilla.org/openweb/HWACCEL/

On two of my systems this test result went from about 6 FPS with Firefox 4 and 5 to about 33 FPS in Firefox 6 beta.

My desktop system has an AMD/ATI R710 GPU, and I use the open source r600g Gallium3D drivers.

My netbook system (Acer Aspire One 522) has an AMD/ATI C50 APU, and I use the open source r600g Gallium3D drivers.

These are both very modest systems in terms of GPU performance. On neither system does WebGL work with Firefox 6. Only hardware acceleration, to some extent.

I expect this to improve again in Firefox 7, but it is still true to say that it has already improved considerably in Firefox 6.

[matthew11093]

Nice post, people should find this very helpful.

[smitty3268]

IonMonkey Overview

[b15hop]

This is a test from Mozilla of browser hardware acceleration:

http://demos.hacks.mozilla.org/openweb/HWACCEL/

On two of my systems this test result went from about 6 FPS with Firefox 4 and 5 to about 33 FPS in Firefox 6 beta.

My desktop system has an AMD/ATI R710 GPU, and I use the open source r600g Gallium3D drivers.

My netbook system (Acer Aspire One 522) has an AMD/ATI C50 APU, and I use the open source r600g Gallium3D drivers.

These are both very modest systems in terms of GPU performance. On neither system does WebGL work with Firefox 6. Only hardware acceleration, to some extent.

I expect this to improve again in Firefox 7, but it is still true to say that it has already improved considerably in Firefox 6.

On win7 HPE x64 with 8GB of ram, onboard HD 5730, i7 720QM, I get 60+ fps with FF5. Not yet tried FF6... Will get back to you on this one...

[Sidicas]

That's incorrect. In a few days after FF6 is released, the Type Inference work will land on mozilla-central, which brings some pretty massive js performance improvements on some tests. Then they are working on improving their garbage collector to be incremental and generational, which will make another big improvement. Then will come IonMonkey, which will allow further optimizations.

There's a lot of work going on to improve performance.

Ya.. I just did a comparison between Firefox 3.5 and Firefox 6...

v8 Benchmark version 6

Score
Firefox 3.5: 226
Firefox 6: 2254

I also don't have any GPU accelerations enabled in Firefox 6.. So definitely there's been significant improvements.. So yea, according to this benchmark Firefox 6 is 10x faster than Firefox 3.5 on the exact same hardware (Intel Pentium M 2.0 Ghz).

[hal2k1]

On win7 HPE x64 with 8GB of ram, onboard HD 5730, i7 720QM, I get 60+ fps with FF5. Not yet tried FF6... Will get back to you on this one...

How is your comment relevant in relation to the topic, "The GPU Acceleration Situation In Firefox 6"?

My point was that GPU acceleration does work to some extent in Firefox 6 for the open source Linux Gallium3D drivers for AMD/ATI r600/r700/Evergreen/N.Islands GPUs (r600g).

Your point?

[smitty3268]

How is your comment relevant in relation to the topic, "The GPU Acceleration Situation In Firefox 6"?

My point was that GPU acceleration does work to some extent in Firefox 6 for the open source Linux Gallium3D drivers for AMD/ATI r600/r700/Evergreen/N.Islands GPUs (r600g).

Your point?

Actually, that test doesn't have anything to do with the Mesa 3D drivers. (r600g)

It's testing Cairo and XRender performance. On Win7 it will test Direct2D instead.

I believe there was some bug in the way they were using xrender they fixed which brought the big speedup. (using the PAD extension?)

[hal2k1]

Actually, that test doesn't have anything to do with the Mesa 3D drivers. (r600g)

It's testing Cairo and XRender performance. On Win7 it will test Direct2D instead.

I believe there was some bug in the way they were using xrender they fixed which brought the big speedup. (using the PAD extension?)

Xrender is an interface between the program which wishes to draw graphics and the GPU 3D drivers.

http://en.wikipedia.org/wiki/XRender

"It is designed to target the 3D graphics capabilities of newer video cards."

Xrender performance has everything to do with the 3D drivers.

Cairo graphics has several back-ends:

http://en.wikipedia.org/wiki/Cairo_%28graphics%29

"Cairo is designed to use hardware acceleration when available."

"Cairo supports output to a number of different backends, known as "surfaces" in its code. Backend support includes output to the X Window System, Win32 GDI, Mac OS X Quartz, the BeOS API, OS/2, OpenGL contexts (directly and via glitz)"

Cairo graphics rendering performance can therefore also be greatly enhanced by the GPU drivers.

Regarding "On Win7 it will test Direct2D instead" - Phoronix is a Linux forum. Off topic.

[smitty3268]

Xrender is an interface between the program which wishes to draw graphics and the GPU 3D drivers.

http://en.wikipedia.org/wiki/XRender

"It is designed to target the 3D graphics capabilities of newer video cards."

Xrender performance has everything to do with the 3D drivers.

I know what XRender is. Support for it uses the 3D hardware (at least for non-Intel or post-SBA) but it is coded in the DDX drivers that come with X.Org. It has absolutely nothing to do with Mesa or r600g, other than the fact that people working on one may improve some common code in the kernel and end up helping both.

Cairo graphics has several back-ends:

http://en.wikipedia.org/wiki/Cairo_%28graphics%29

"Cairo is designed to use hardware acceleration when available."

"Cairo supports output to a number of different backends, known as "surfaces" in its code. Backend support includes output to the X Window System, Win32 GDI, Mac OS X Quartz, the BeOS API, OS/2, OpenGL contexts (directly and via glitz)"

Cairo graphics rendering performance can therefore also be greatly enhanced by the GPU drivers.

Cairo uses XRender by default in this case (the X Window System support you mention), which is done by the 2D DDX drivers. It has an experimental OpenGL backend but that is not used and not ready to be used.

Regarding "On Win7 it will test Direct2D instead" - Phoronix is a Linux forum. Off topic.

I was referring to one of the previous commenters who stated they got 60+ fps on Win7. That is because of the D2D support Firefox added, and not because of fast D3D support. Similarly, on Linux it is testing the DDX driver through XRender and not OpenGL in the gallium drivers.