I'm curious if small >4K textures are getting passed back as these mysterious 64K pages. What does that look like on your ATI machine?
--Jet ----- Original Message ----- From: "Jeff Muizelaar" <jmuizel...@mozilla.com> To: "Jet Villegas" <j...@mozilla.com> Cc: "Justin Lebar" <jle...@mozilla.com>, "Aaron Klotz" <akl...@mozilla.com>, dev-platform@lists.mozilla.org, "Nicholas Nethercote" <nnetherc...@mozilla.com>, "Brian R. Bondy" <bbo...@mozilla.com>, "Benjamin Smedberg" <benja...@smedbergs.us>, "Bas Schouten" <bschou...@mozilla.com> Sent: Tuesday, April 9, 2013 12:45:51 PM Subject: Re: Virtual Memory fragmentation issues Just allocating a bunch of Texture memory and looking at the process in VMMap should show the problem. Process Explorer also lets you view per process GPU memory consumption which might be interesting to look at. -Jeff On 2013-04-09, at 3:29 PM, Jet Villegas wrote: > Jeff: how mysterious is this phenomena? Can we write a test that forces it to > happen? I'm curious how it behaves on ATI vs. other drivers. > > --Jet > > > ----- Original Message ----- > From: "Jeff Muizelaar" <jmuizel...@mozilla.com> > To: "Benjamin Smedberg" <benja...@smedbergs.us> > Cc: "Justin Lebar" <jle...@mozilla.com>, "Aaron Klotz" <akl...@mozilla.com>, > dev-platform@lists.mozilla.org, "Nicholas Nethercote" > <nnetherc...@mozilla.com>, "Brian R. Bondy" <bbo...@mozilla.com> > Sent: Tuesday, April 9, 2013 12:24:32 PM > Subject: Re: Virtual Memory fragmentation issues > > > On 2013-04-08, at 7:46 PM, Benjamin Smedberg wrote: > >> In stability-land we're starting to see some "interesting" problems relating >> to virtual memory usage in Firefox on Windows. > > >> Either our code or the ATI driver is leaking mapped memory references in a >> way that chews up VM space without actually using memory. We need to figure >> out why and fix our work around the problem. I'm hoping that one of you on >> the "To" of this email may be able to help figure out a way to either patch >> Firefox or run an external debugging tool which can help record information >> about the stacks when these mappings are typically created and destroyed, >> and help create a detailed understanding of the problem so that we can >> either fix our code or hand the problem to the nvidia developers with the >> most reproducible/detailed report possible > > One interesting phenomenon that I have seen is pages being mysteriously added > to firefox's address space. I believe this can happen when graphics memory is > paged into system memory and is added to our address space. I'm not sure what > OS accounting all knows about this and what doesn't. > > -Jeff > > _______________________________________________ > dev-platform mailing list > dev-platform@lists.mozilla.org > https://lists.mozilla.org/listinfo/dev-platform _______________________________________________ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
