From mboxrd@z Thu Jan 1 00:00:00 1970 From: Robert Kaiser Subject: Re: Status of WebGL with open-source intel drivers Date: Sun, 15 May 2011 23:50:27 +0200 Message-ID: References: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: Received: from lo.gmane.org (lo.gmane.org [80.91.229.12]) by gabe.freedesktop.org (Postfix) with ESMTP id 15C859E7BA for ; Sun, 15 May 2011 14:50:45 -0700 (PDT) Received: from list by lo.gmane.org with local (Exim 4.69) (envelope-from ) id 1QLjD1-0004oU-GI for intel-gfx@lists.freedesktop.org; Sun, 15 May 2011 23:50:43 +0200 Received: from 91-113-91-9.adsl.highway.telekom.at ([91.113.91.9]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sun, 15 May 2011 23:50:43 +0200 Received: from kairo by 91-113-91-9.adsl.highway.telekom.at with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sun, 15 May 2011 23:50:43 +0200 In-Reply-To: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org Errors-To: intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org To: intel-gfx@lists.freedesktop.org List-Id: intel-gfx@lists.freedesktop.org Clemens Eisserer schrieb: > Whats the current state of WebGL on various browsers with the > open-source intel driver? > I have a i945GM based machine running Fedora-14, and no matter which > browser I try (FireFox4, Chrome, Opera) - all tell me WebGL has been > disabled. > Even disabling sanity checks in FireFox that prevent it from enabling > WebGL whith intel drivers didn't help. > > Are there any plans to support the missing OpenGL features that > prevent browsers from enabling webgl on my system? I can't speak for the others, but as I'm working for a "browser manufacturer" and I have followed WebGL somewhat around here, what I can say is that Mozilla has only enabled WebGL on Linux for relatively current versions of the binary NVidia driver in Firefox 4. This is because there were some instabilities in Mesa and other drivers with our WebGL code and we wanted to get something shipped for once. There is some environment variable that can be set to enable it despite, but there's a better way if you want to test in a way that hopefully can help both the Intel driver team and our team: try "Nightly" builds (currently the code on the way to Firefox 6, which is to be released in August). There we enable WebGL for all drivers that are new enough and not explicitely blacklisted (and do not crash when instantiating a GLX context), which makes the Intel drivers be enabled as well. I have a 945GM-based laptop where those builds work to enable WebGL, but something either in how our code works or how the driver works does make that very inefficient, so that I only get one frame rendered every roughly 2 seconds. Would be interesting to have experts on both sides work together to find out how to improve on that. Unfortunately, I can't test on a newer Intel chipset as my SandBridge system only works with vesafb and nobody seems to be willing to even look into https://bugs.freedesktop.org/show_bug.cgi?id=37010 so I'm starting to think if I should kick out Intel from my board and graphics setup again. Sorry for the somewhat offtopic rant, but I'm getting somewhat frustrated - esp. as I'd expect that WebGL could work decently there if I could just use a proper driver. Robert Kaiser