make dma_addressing_limited work for memory encryption setups v2
Subsystems (2)
Hi all,
this little series fixes dma_addressing_limited to return true for systems that use bounce buffers due to memory encryption.
Changes since v1:
- take SWIOTLB_FORCE into account
PATCH 1/2 dma-mapping: move dma_addressing_limited out of line
This function isn't used in the fast path, and moving it out of line will reduce include clutter with the next change.
--- include/linux/dma-mapping.h | 14 +------------- kernel/dma/mapping.c | 15 +++++++++++++++ 2 files changed, 16 insertions(+), 13 deletions(-)
include/linux/dma-mapping.h +1 -13
@@ -684,19 +684,7 @@ static inline int dma_coerce_mask_and_coherent(struct device *dev, u64 mask)return dma_set_mask_and_coherent (dev, mask); } -/** -* dma_addressing_limited - return if the device is addressing limited -* @dev: device to check -* -* Return %true if the devices DMA mask is too small to address all memory in -* the system, else %false. Lack of addressing bits is the prime reason for -* bounce buffering, but might not be the only one. -*/ -static inline bool dma_addressing_limited (struct device *dev) -{ -return min_not_zero (dma_get_mask (dev), dev->bus_dma_limit) < -dma_get_required_mask (dev); -} +bool dma_addressing_limited (struct device *dev);#ifdef CONFIG_ARCH_HAS_SETUP_DMA_OPS void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
kernel/dma/mapping.c +15 -0
@@ -405,3 +405,18 @@ unsigned long dma_get_merge_boundary(struct device *dev)return ops->get_merge_boundary (dev); }EXPORT_SYMBOL_GPL (dma_get_merge_boundary); + +/** +* dma_addressing_limited - return if the device is addressing limited +* @dev: device to check +* +* Return %true if the devices DMA mask is too small to address all memory in +* the system, else %false. Lack of addressing bits is the prime reason for +* bounce buffering, but might not be the only one. +*/ +bool dma_addressing_limited (struct device *dev) +{ +return min_not_zero (dma_get_mask (dev), dev->bus_dma_limit) < +dma_get_required_mask (dev); +} +EXPORT_SYMBOL_GPL (dma_addressing_limited);
Signature · 1 line
2.20.1
PATCH 2/2 dma-mapping: force unencryped devices are always addressing limited
In reply to Christoph HellwigDevices that are forced to DMA through swiotlb need to be treated as if they are addressing limited.
--- include/linux/dma-direct.h | 1 + kernel/dma/direct.c | 8 ++++++-- kernel/dma/mapping.c | 3 +++ 3 files changed, 10 insertions(+), 2 deletions(-)
include/linux/dma-direct.h +1 -0
@@ -85,4 +85,5 @@ int dma_direct_mmap(struct device *dev, struct vm_area_struct *vma, void *cpu_addr, dma_addr_t dma_addr, size_t size, unsigned long attrs);int dma_direct_supported (struct device *dev,u64 mask); +bool dma_direct_addressing_limited (struct device *dev);#endif /* _LINUX_DMA_DIRECT_H */
kernel/dma/direct.c +6 -2
@@ -497,11 +497,15 @@ int dma_direct_supported(struct device *dev, u64 mask)return mask >=__phys_to_dma (dev, min_mask); } +bool dma_direct_addressing_limited (struct device *dev) +{ +return force_dma_unencrypted (dev) || swiotlb_force ==SWIOTLB_FORCE ; +} +size_t dma_direct_max_mapping_size (struct device *dev) {/* If SWIOTLB is active, use its maximum mapping size */ -if (is_swiotlb_active () && - (dma_addressing_limited (dev) || swiotlb_force ==SWIOTLB_FORCE )) +if (is_swiotlb_active () &&dma_addressing_limited (dev))return swiotlb_max_mapping_size (dev);return SIZE_MAX ; }
kernel/dma/mapping.c +3 -0
@@ -416,6 +416,9 @@ EXPORT_SYMBOL_GPL(dma_get_merge_boundary);*/ bool dma_addressing_limited (struct device *dev) { +if (dma_is_direct (get_dma_ops (dev)) && +dma_direct_addressing_limited (dev)) +return true;return min_not_zero (dma_get_mask (dev), dev->bus_dma_limit) <dma_get_required_mask (dev); }
Signature · 6 lines
2.20.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Re: [PATCH 2/2] dma-mapping: force unencryped devices are always addressing limited
Hi, Christoph.
Quoted 52 lines · Christoph Hellwig
Devices that are forced to DMA through swiotlb need to be treated as if
they are addressing limited.Signed-off-by Christoph Hellwig <[email protected]>--- include/linux/dma-direct.h | 1 + kernel/dma/direct.c | 8 ++++++-- kernel/dma/mapping.c | 3 +++ 3 files changed, 10 insertions(+), 2 deletions(-)include/linux/dma-direct.h +1
-0@@ -85,4 +85,5 @@ int dma_direct_mmap(struct device *dev, struct vm_area_struct *vma, void *cpu_addr, dma_addr_t dma_addr, size_t size, unsigned long attrs);int dma_direct_supported (struct device *dev,u64 mask); +bool dma_direct_addressing_limited (struct device *dev);#endif /*
_LINUX_DMA_DIRECT_H*/kernel/dma/direct.c +6
-2@@ -497,11 +497,15 @@ int dma_direct_supported(struct device *dev, u64 mask)return mask >=__phys_to_dma (dev, min_mask); } +bool dma_direct_addressing_limited (struct device *dev) +{ +return force_dma_unencrypted (dev) || swiotlb_force ==SWIOTLB_FORCE ; +} +size_t dma_direct_max_mapping_size (struct device *dev) {/* If SWIOTLB is active, use its maximum mapping size */ -if (is_swiotlb_active () && - (dma_addressing_limited (dev) || swiotlb_force == SWIOTLB_FORCE)) +if (is_swiotlb_active () &&dma_addressing_limited (dev))kernel/dma/direct.c +0
-0return swiotlb_max_mapping_size (dev);return SIZE_MAX ; }kernel/dma/mapping.c +0
-0
kernel/dma/mapping.c EXPORT_SYMBOL_GPL(dma_get_merge_boundary); · line 421
Quoted 6 lines
kernel/dma/mapping.c +3
-0@@ -416,6 +416,9 @@ EXPORT_SYMBOL_GPL(dma_get_merge_boundary);*/ bool dma_addressing_limited (struct device *dev) { +if (dma_is_direct (get_dma_ops (dev)) && +dma_direct_addressing_limited (dev)) +return true;
This works fine for vmwgfx, for which the below expression is always 0. But it looks like the only current user of dma_addressing_limited outside of the dma code, radeon, actually wants only the below expression to force GFP_DMA32 page allocations when the devices have limited dma address space. Perhaps Christian can elaborate on that.
So in the end it looks like we have two different use cases. One to force coherent memory (vmwgfx, possibly other grahpics drivers) or reduced queue depth (vmw_pvscsi) when we have bounce-buffering.
The other one is to force GFP_DMA32 page allocation when the device dma-addressing is limited. Perhaps this mode can be replaced by using dma_coherent memory and stripped that functionality from TTM?
kernel/dma/mapping.c EXPORT_SYMBOL_GPL(dma_get_merge_boundary); · line 424
Quoted 3 lines
kernel/dma/mapping.c +0
-0return min_not_zero (dma_get_mask (dev), dev->bus_dma_limit) <dma_get_required_mask (dev); }
Thanks,
Thomas