Re: [PATCH v9 07/14] swiotlb: Bounce data from/to restricted DMA pool if available
From: Christoph Hellwig <hch@lst.de>
Date: 2021-06-14 06:25:40
Also in:
intel-gfx, linux-devicetree, linux-iommu, linux-pci, lkml, xen-devel
On Fri, Jun 11, 2021 at 11:26:52PM +0800, Claire Chang wrote:
quoted hunk ↗ jump to hunk
Regardless of swiotlb setting, the restricted DMA pool is preferred if available. The restricted DMA pools provide a basic level of protection against the DMA overwriting buffer contents at unexpected times. However, to protect against general data leakage and system memory corruption, the system needs to provide a way to lock down the memory access, e.g., MPU. Note that is_dev_swiotlb_force doesn't check if swiotlb_force == SWIOTLB_FORCE. Otherwise the memory allocation behavior with default swiotlb will be changed by the following patche ("dma-direct: Allocate memory from restricted DMA pool if available"). Signed-off-by: Claire Chang <redacted> --- include/linux/swiotlb.h | 10 +++++++++- kernel/dma/direct.c | 3 ++- kernel/dma/direct.h | 3 ++- kernel/dma/swiotlb.c | 1 + 4 files changed, 14 insertions(+), 3 deletions(-)diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h index 06cf17a80f5c..8200c100fe10 100644 --- a/include/linux/swiotlb.h +++ b/include/linux/swiotlb.h@@ -85,6 +85,7 @@ extern enum swiotlb_force swiotlb_force; * unmap calls. * @debugfs: The dentry to debugfs. * @late_alloc: %true if allocated using the page allocator + * @force_swiotlb: %true if swiotlb is forced */ struct io_tlb_mem { phys_addr_t start;@@ -95,6 +96,7 @@ struct io_tlb_mem { spinlock_t lock; struct dentry *debugfs; bool late_alloc; + bool force_swiotlb; struct io_tlb_slot { phys_addr_t orig_addr; size_t alloc_size;@@ -115,6 +117,11 @@ static inline void swiotlb_set_io_tlb_default_mem(struct device *dev) dev->dma_io_tlb_mem = io_tlb_default_mem; } +static inline bool is_dev_swiotlb_force(struct device *dev) +{ + return dev->dma_io_tlb_mem->force_swiotlb; +} + void __init swiotlb_exit(void); unsigned int swiotlb_max_segment(void); size_t swiotlb_max_mapping_size(struct device *dev);@@ -126,8 +133,9 @@ static inline bool is_swiotlb_buffer(struct device *dev, phys_addr_t paddr) { return false; } -static inline void swiotlb_set_io_tlb_default_mem(struct device *dev) +static inline bool is_dev_swiotlb_force(struct device *dev) { + return false; } static inline void swiotlb_exit(void) {diff --git a/kernel/dma/direct.c b/kernel/dma/direct.c index 7a88c34d0867..078f7087e466 100644 --- a/kernel/dma/direct.c +++ b/kernel/dma/direct.c@@ -496,7 +496,8 @@ size_t dma_direct_max_mapping_size(struct device *dev) { /* If SWIOTLB is active, use its maximum mapping size */ if (is_swiotlb_active(dev) && - (dma_addressing_limited(dev) || swiotlb_force == SWIOTLB_FORCE)) + (dma_addressing_limited(dev) || swiotlb_force == SWIOTLB_FORCE || + is_dev_swiotlb_force(dev)))
I think we can remove the extra swiotlb_force check here if the swiotlb_force setting is propagated into io_tlb_default_mem->force when that is initialized. This avoids an extra check in the fast path.
- if (unlikely(swiotlb_force == SWIOTLB_FORCE)) + if (unlikely(swiotlb_force == SWIOTLB_FORCE) || + is_dev_swiotlb_force(dev))
Same here.