Commit fa6661b7 authored by Thierry Reding's avatar Thierry Reding

drm/tegra: Optionally attach clients to the IOMMU

If a client is already attached to an IOMMU domain that is not the
shared domain, don't try to attach it again. This allows using the
IOMMU-backed DMA API.

Since the IOMMU-backed DMA API is now supported and there's no way
to detach from it on 64-bit ARM, don't bother to detach from it on
32-bit ARM either.
Signed-off-by: default avatarThierry Reding <treding@nvidia.com>
parent 2e8d8749
...@@ -20,10 +20,6 @@ ...@@ -20,10 +20,6 @@
#include <drm/drm_prime.h> #include <drm/drm_prime.h>
#include <drm/drm_vblank.h> #include <drm/drm_vblank.h>
#if IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU)
#include <asm/dma-iommu.h>
#endif
#include "drm.h" #include "drm.h"
#include "gem.h" #include "gem.h"
...@@ -908,30 +904,27 @@ int tegra_drm_unregister_client(struct tegra_drm *tegra, ...@@ -908,30 +904,27 @@ int tegra_drm_unregister_client(struct tegra_drm *tegra,
int host1x_client_iommu_attach(struct host1x_client *client) int host1x_client_iommu_attach(struct host1x_client *client)
{ {
struct iommu_domain *domain = iommu_get_domain_for_dev(client->dev);
struct drm_device *drm = dev_get_drvdata(client->parent); struct drm_device *drm = dev_get_drvdata(client->parent);
struct tegra_drm *tegra = drm->dev_private; struct tegra_drm *tegra = drm->dev_private;
struct iommu_group *group = NULL; struct iommu_group *group = NULL;
int err; int err;
if (tegra->domain) { /*
struct iommu_domain *domain; * If the host1x client is already attached to an IOMMU domain that is
* not the shared IOMMU domain, don't try to attach it to a different
* domain. This allows using the IOMMU-backed DMA API.
*/
if (domain && domain != tegra->domain)
return 0;
if (tegra->domain) {
group = iommu_group_get(client->dev); group = iommu_group_get(client->dev);
if (!group) { if (!group) {
dev_err(client->dev, "failed to get IOMMU group\n"); dev_err(client->dev, "failed to get IOMMU group\n");
return -ENODEV; return -ENODEV;
} }
#if IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU)
if (client->dev->archdata.mapping) {
struct dma_iommu_mapping *mapping =
to_dma_iommu_mapping(client->dev);
arm_iommu_detach_device(client->dev);
arm_iommu_release_mapping(mapping);
}
#endif
domain = iommu_get_domain_for_dev(client->dev);
if (domain != tegra->domain) { if (domain != tegra->domain) {
err = iommu_attach_group(tegra->domain, group); err = iommu_attach_group(tegra->domain, group);
if (err < 0) { if (err < 0) {
...@@ -939,6 +932,8 @@ int host1x_client_iommu_attach(struct host1x_client *client) ...@@ -939,6 +932,8 @@ int host1x_client_iommu_attach(struct host1x_client *client)
return err; return err;
} }
} }
tegra->use_explicit_iommu = true;
} }
client->group = group; client->group = group;
...@@ -963,6 +958,7 @@ void host1x_client_iommu_detach(struct host1x_client *client) ...@@ -963,6 +958,7 @@ void host1x_client_iommu_detach(struct host1x_client *client)
iommu_detach_group(tegra->domain, client->group); iommu_detach_group(tegra->domain, client->group);
iommu_group_put(client->group); iommu_group_put(client->group);
client->group = NULL;
} }
} }
...@@ -1046,6 +1042,7 @@ void tegra_drm_free(struct tegra_drm *tegra, size_t size, void *virt, ...@@ -1046,6 +1042,7 @@ void tegra_drm_free(struct tegra_drm *tegra, size_t size, void *virt,
static int host1x_drm_probe(struct host1x_device *dev) static int host1x_drm_probe(struct host1x_device *dev)
{ {
struct drm_driver *driver = &tegra_drm_driver; struct drm_driver *driver = &tegra_drm_driver;
struct iommu_domain *domain;
struct tegra_drm *tegra; struct tegra_drm *tegra;
struct drm_device *drm; struct drm_device *drm;
int err; int err;
...@@ -1060,7 +1057,36 @@ static int host1x_drm_probe(struct host1x_device *dev) ...@@ -1060,7 +1057,36 @@ static int host1x_drm_probe(struct host1x_device *dev)
goto put; goto put;
} }
if (iommu_present(&platform_bus_type)) { /*
* If the Tegra DRM clients are backed by an IOMMU, push buffers are
* likely to be allocated beyond the 32-bit boundary if sufficient
* system memory is available. This is problematic on earlier Tegra
* generations where host1x supports a maximum of 32 address bits in
* the GATHER opcode. In this case, unless host1x is behind an IOMMU
* as well it won't be able to process buffers allocated beyond the
* 32-bit boundary.
*
* The DMA API will use bounce buffers in this case, so that could
* perhaps still be made to work, even if less efficient, but there
* is another catch: in order to perform cache maintenance on pages
* allocated for discontiguous buffers we need to map and unmap the
* SG table representing these buffers. This is fine for something
* small like a push buffer, but it exhausts the bounce buffer pool
* (typically on the order of a few MiB) for framebuffers (many MiB
* for any modern resolution).
*
* Work around this by making sure that Tegra DRM clients only use
* an IOMMU if the parent host1x also uses an IOMMU.
*
* Note that there's still a small gap here that we don't cover: if
* the DMA API is backed by an IOMMU there's no way to control which
* device is attached to an IOMMU and which isn't, except via wiring
* up the device tree appropriately. This is considered an problem
* of integration, so care must be taken for the DT to be consistent.
*/
domain = iommu_get_domain_for_dev(drm->dev->parent);
if (domain && iommu_present(&platform_bus_type)) {
tegra->domain = iommu_domain_alloc(&platform_bus_type); tegra->domain = iommu_domain_alloc(&platform_bus_type);
if (!tegra->domain) { if (!tegra->domain) {
err = -ENOMEM; err = -ENOMEM;
...@@ -1104,7 +1130,7 @@ static int host1x_drm_probe(struct host1x_device *dev) ...@@ -1104,7 +1130,7 @@ static int host1x_drm_probe(struct host1x_device *dev)
if (err < 0) if (err < 0)
goto fbdev; goto fbdev;
if (tegra->domain) { if (tegra->use_explicit_iommu) {
u64 carveout_start, carveout_end, gem_start, gem_end; u64 carveout_start, carveout_end, gem_start, gem_end;
u64 dma_mask = dma_get_mask(&dev->dev); u64 dma_mask = dma_get_mask(&dev->dev);
dma_addr_t start, end; dma_addr_t start, end;
...@@ -1132,6 +1158,10 @@ static int host1x_drm_probe(struct host1x_device *dev) ...@@ -1132,6 +1158,10 @@ static int host1x_drm_probe(struct host1x_device *dev)
DRM_DEBUG_DRIVER(" GEM: %#llx-%#llx\n", gem_start, gem_end); DRM_DEBUG_DRIVER(" GEM: %#llx-%#llx\n", gem_start, gem_end);
DRM_DEBUG_DRIVER(" Carveout: %#llx-%#llx\n", carveout_start, DRM_DEBUG_DRIVER(" Carveout: %#llx-%#llx\n", carveout_start,
carveout_end); carveout_end);
} else if (tegra->domain) {
iommu_domain_free(tegra->domain);
tegra->domain = NULL;
iova_cache_put();
} }
if (tegra->hub) { if (tegra->hub) {
......
...@@ -36,6 +36,7 @@ struct tegra_drm { ...@@ -36,6 +36,7 @@ struct tegra_drm {
struct drm_device *drm; struct drm_device *drm;
struct iommu_domain *domain; struct iommu_domain *domain;
bool use_explicit_iommu;
struct mutex mm_lock; struct mutex mm_lock;
struct drm_mm mm; struct drm_mm mm;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment