Date: Tue, 08 Dec 2009 09:29:04 -0600 From: Robert Noland <rnoland@FreeBSD.org> To: Greg Rowe <greg@rowes.org> Cc: freebsd-x11@freebsd.org, John Baldwin <jhb@freebsd.org> Subject: Re: Radeon HD4850 DRI Issues Message-ID: <1260286145.2315.288.camel@balrog.2hip.net> In-Reply-To: <200912071550.56061.greg@rowes.org> References: <200912071550.56061.greg@rowes.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 2009-12-07 at 15:50 -0500, Greg Rowe wrote: > (II) RADEON(0): [pci] ring handle = 0xebe4f000 > (II) RADEON(0): [pci] Ring mapped at 0x28a50000 > (II) RADEON(0): [pci] Ring contents 0x55555555 > (II) RADEON(0): [pci] ring read ptr handle = 0xebf50000 > (II) RADEON(0): [pci] Ring read ptr mapped at 0x286ff000 > (II) RADEON(0): [pci] Ring read ptr contents 0x55555555 This is the root of the problem... And I'm still baffled at how this occurs... It almost has to be either a BIOS issue or somehow a problem with cache flushing. What confuses me even further is that many i386 installs work fine, so it seems to point to the IOMMU possibly. We explicitly zero the ring space during allocation. I'm cc'ing jhb@ in case he has any suggestions on what to look for... robert. -- Robert Noland <rnoland@FreeBSD.org> FreeBSD
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1260286145.2315.288.camel>