[PATCH v4 05/12] memory: Add NVIDIA Tegra memory controller support
From: acourbot@nvidia.com (Alexandre Courbot)
Date: 2014-10-30 13:35:07
Also in:
linux-iommu, linux-tegra
Possibly related (same subject, not in this thread)
- 2014-11-03 · [PATCH v4 05/12] memory: Add NVIDIA Tegra memory controller support · Alexandre Courbot <hidden>
- 2014-11-03 · [PATCH v4 05/12] memory: Add NVIDIA Tegra memory controller support · Thierry Reding <hidden>
- 2014-11-01 · [PATCH v4 05/12] memory: Add NVIDIA Tegra memory controller support · Alexandre Courbot <hidden>
- 2014-10-31 · [PATCH v4 05/12] memory: Add NVIDIA Tegra memory controller support · Thierry Reding <hidden>
- 2014-10-30 · [PATCH v4 05/12] memory: Add NVIDIA Tegra memory controller support · Thierry Reding <hidden>
On 10/30/2014 08:04 PM, Terje Bergstr?m wrote:
On 30.10.2014 12:22, Alexandre Courbot wrote:quoted
So should I understand that the GPU group is for addresses without bit 34 set (hence forcibly disabled) while GPUB is used when that bit is set? Or is it something else?That's exactly correct. And only GPUB can be programmed to be SMMU translated.
Great, thanks for confirming! Thierry, how do you want to address this? We could change the register for the GPU group, or (maybe preferable if we want to reflect the actual hardware state) add the GPUB group. I don't know if that would be easy though since we would have the problem of the gpusrd and gpuswr clients ownership (seems like they would belong to both groups?)