tjh.dev / kernel
Linux kernel mirror (for testing) git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel os linux
1
fork

Configure Feed

Select the types of activity you want to include in your feed.

gpu: host1x: Avoid trying to use GART on Tegra20

Since commit c7e3ca515e78 ("iommu/tegra: gart: Do not register with
bus") quite some time ago, the GART driver has effectively disabled
itself to avoid issues with the GPU driver expecting it to work in ways
that it doesn't. As of commit 57365a04c921 ("iommu: Move bus setup to
IOMMU device registration") that bodge no longer works, but really the
GPU driver should be responsible for its own behaviour anyway. Make the
workaround explicit.

Reported-by: Jon Hunter <jonathanh@nvidia.com>
Suggested-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Tested-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>

authored by

Robin Murphy and committed by
Thierry Reding c2418f91 9abf2313

+8
+4
drivers/gpu/drm/tegra/drm.c
··· 1093 1093 struct host1x *host1x = dev_get_drvdata(dev->dev.parent); 1094 1094 struct iommu_domain *domain; 1095 1095 1096 + /* Our IOMMU usage policy doesn't currently play well with GART */ 1097 + if (of_machine_is_compatible("nvidia,tegra20")) 1098 + return false; 1099 + 1096 1100 /* 1097 1101 * If the Tegra DRM clients are backed by an IOMMU, push buffers are 1098 1102 * likely to be allocated beyond the 32-bit boundary if sufficient
+4
drivers/gpu/host1x/dev.c
··· 292 292 293 293 static bool host1x_wants_iommu(struct host1x *host1x) 294 294 { 295 + /* Our IOMMU usage policy doesn't currently play well with GART */ 296 + if (of_machine_is_compatible("nvidia,tegra20")) 297 + return false; 298 + 295 299 /* 296 300 * If we support addressing a maximum of 32 bits of physical memory 297 301 * and if the host1x firewall is enabled, there's no need to enable