Are you using the open-source ATI driver or the proprietary fglrx driver? If you are using the open-source driver, I notice that you have an ASUS M2V-MX motherboard with a VIA chipset, so you may be encoubntering the problem described at http://choon.net/forum/read.php?21,106131,115940
This problem has been fixed upstream in kernel 3.0-rc3 and the fix has also been cc'd to the stable kernel tree. From the changelog:
commit 62fff811d73095bd95579d72f558f03c78f7914a
Author: Daniel Haid
Date: Wed Jun 8 20:04:45 2011 +1000
drm/radeon/kms: fix for radeon on systems >4GB without hardware iommu
On my x86_64 system with >4GB of ram and swiotlb instead of
a hardware iommu (because I have a VIA chipset), the call
to pci_set_dma_mask (see below) with 40bits returns an error.
But it seems that the radeon driver is designed to have
need_dma32 = true exactly if pci_set_dma_mask is called
with 32 bits and false if it is called with 40 bits.
I have read somewhere that the default are 32 bits. So if the
call fails I suppose that need_dma32 should be set to true.
Are you using the open-source ATI driver or the proprietary fglrx driver? If you are using the open-source driver, I notice that you have an ASUS M2V-MX motherboard with a VIA chipset, so you may be encoubntering the problem described at http:// choon.net/ forum/read. php?21, 106131, 115940
This problem has been fixed upstream in kernel 3.0-rc3 and the fix has also been cc'd to the stable kernel tree. From the changelog:
commit 62fff811d73095b d95579d72f558f0 3c78f7914a
Author: Daniel Haid
Date: Wed Jun 8 20:04:45 2011 +1000
drm/radeon/kms: fix for radeon on systems >4GB without hardware iommu
On my x86_64 system with >4GB of ram and swiotlb instead of
a hardware iommu (because I have a VIA chipset), the call
to pci_set_dma_mask (see below) with 40bits returns an error.
But it seems that the radeon driver is designed to have
need_dma32 = true exactly if pci_set_dma_mask is called
with 32 bits and false if it is called with 40 bits.
I have read somewhere that the default are 32 bits. So if the
call fails I suppose that need_dma32 should be set to true.
And indeed the patch fixes the problem I have had before choon.net/ forum/read. php?21, 106131, 115940
and which I had described here:
http://
Acked-by: Alex Deucher
cc: <email address hidden>
Signed-off-by: Dave Airlie
A PPA of the 3.0-rc3 kernel is available at http:// kernel. ubuntu. com/~kernel- ppa/mainline/