From 54624acf8843375a6de3717ac18df3b5104c39c5 Mon Sep 17 00:00:00 2001 From: Yicong Yang Date: Thu, 20 Jun 2024 17:28:55 +0800 Subject: dma-mapping: benchmark: Don't starve others when doing the test The test thread will start N benchmark kthreads and then schedule out until the test time finished and notify the benchmark kthreads to stop. The benchmark kthreads will keep running until notified to stop. There's a problem with current implementation when the benchmark kthreads number is equal to the CPUs on a non-preemptible kernel: since the scheduler will balance the kthreads across the CPUs and when the test time's out the test thread won't get a chance to be scheduled on any CPU then cannot notify the benchmark kthreads to stop. This can be easily reproduced on a VM (simulated with 16 CPUs) with PREEMPT_VOLUNTARY: estuary:/mnt$ ./dma_map_benchmark -t 16 -s 1 rcu: INFO: rcu_sched self-detected stall on CPU rcu: 10-...!: (5221 ticks this GP) idle=ed24/1/0x4000000000000000 softirq=142/142 fqs=0 rcu: (t=5254 jiffies g=-559 q=45 ncpus=16) rcu: rcu_sched kthread starved for 5255 jiffies! g-559 f0x0 RCU_GP_WAIT_FQS(5) ->state=0x0 ->cpu=12 rcu: Unless rcu_sched kthread gets sufficient CPU time, OOM is now expected behavior. rcu: RCU grace-period kthread stack dump: task:rcu_sched state:R running task stack:0 pid:16 tgid:16 ppid:2 flags:0x00000008 Call trace __switch_to+0xec/0x138 __schedule+0x2f8/0x1080 schedule+0x30/0x130 schedule_timeout+0xa0/0x188 rcu_gp_fqs_loop+0x128/0x528 rcu_gp_kthread+0x1c8/0x208 kthread+0xec/0xf8 ret_from_fork+0x10/0x20 Sending NMI from CPU 10 to CPUs 0: NMI backtrace for cpu 0 CPU: 0 PID: 332 Comm: dma-map-benchma Not tainted 6.10.0-rc1-vanilla-LSE #8 Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015 pstate: 20400005 (nzCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : arm_smmu_cmdq_issue_cmdlist+0x218/0x730 lr : arm_smmu_cmdq_issue_cmdlist+0x488/0x730 sp : ffff80008748b630 x29: ffff80008748b630 x28: 0000000000000000 x27: ffff80008748b780 x26: 0000000000000000 x25: 000000000000bc70 x24: 000000000001bc70 x23: ffff0000c12af080 x22: 0000000000010000 x21: 000000000000ffff x20: ffff80008748b700 x19: ffff0000c12af0c0 x18: 0000000000010000 x17: 0000000000000001 x16: 0000000000000040 x15: ffffffffffffffff x14: 0001ffffffffffff x13: 000000000000ffff x12: 00000000000002f1 x11: 000000000001ffff x10: 0000000000000031 x9 : ffff800080b6b0b8 x8 : ffff0000c2a48000 x7 : 000000000001bc71 x6 : 0001800000000000 x5 : 00000000000002f1 x4 : 01ffffffffffffff x3 : 000000000009aaf1 x2 : 0000000000000018 x1 : 000000000000000f x0 : ffff0000c12af18c Call trace: arm_smmu_cmdq_issue_cmdlist+0x218/0x730 __arm_smmu_tlb_inv_range+0xe0/0x1a8 arm_smmu_iotlb_sync+0xc0/0x128 __iommu_dma_unmap+0x248/0x320 iommu_dma_unmap_page+0x5c/0xe8 dma_unmap_page_attrs+0x38/0x1d0 map_benchmark_thread+0x118/0x2c0 kthread+0xec/0xf8 ret_from_fork+0x10/0x20 Solve this by adding scheduling point in the kthread loop, so if there're other threads in the system they may have a chance to run, especially the thread to notify the test end. However this may degrade the test concurrency so it's recommended to run this on an idle system. Signed-off-by: Yicong Yang Acked-by: Barry Song Signed-off-by: Christoph Hellwig --- kernel/dma/map_benchmark.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/kernel/dma/map_benchmark.c b/kernel/dma/map_benchmark.c index 4950e0b622b1..cc19a3efea89 100644 --- a/kernel/dma/map_benchmark.c +++ b/kernel/dma/map_benchmark.c @@ -89,6 +89,22 @@ static int map_benchmark_thread(void *data) atomic64_add(map_sq, &map->sum_sq_map); atomic64_add(unmap_sq, &map->sum_sq_unmap); atomic64_inc(&map->loops); + + /* + * We may test for a long time so periodically check whether + * we need to schedule to avoid starving the others. Otherwise + * we may hangup the kernel in a non-preemptible kernel when + * the test kthreads number >= CPU number, the test kthreads + * will run endless on every CPU since the thread resposible + * for notifying the kthread stop (in do_map_benchmark()) + * could not be scheduled. + * + * Note this may degrade the test concurrency since the test + * threads may need to share the CPU time with other load + * in the system. So it's recommended to run this benchmark + * on an idle system. + */ + cond_resched(); } out: -- cgit v1.3.1 From 7296f2301a057493e97b07739213c6e864f76891 Mon Sep 17 00:00:00 2001 From: Michael Kelley Date: Mon, 8 Jul 2024 12:41:00 -0700 Subject: swiotlb: reduce swiotlb pool lookups With CONFIG_SWIOTLB_DYNAMIC enabled, each round-trip map/unmap pair in the swiotlb results in 6 calls to swiotlb_find_pool(). In multiple places, the pool is found and used in one function, and then must be found again in the next function that is called because only the tlb_addr is passed as an argument. These are the six call sites: dma_direct_map_page: 1. swiotlb_map -> swiotlb_tbl_map_single -> swiotlb_bounce dma_direct_unmap_page: 2. dma_direct_sync_single_for_cpu -> is_swiotlb_buffer 3. dma_direct_sync_single_for_cpu -> swiotlb_sync_single_for_cpu -> swiotlb_bounce 4. is_swiotlb_buffer 5. swiotlb_tbl_unmap_single -> swiotlb_del_transient 6. swiotlb_tbl_unmap_single -> swiotlb_release_slots Reduce the number of calls by finding the pool at a higher level, and passing it as an argument instead of searching again. A key change is for is_swiotlb_buffer() to return a pool pointer instead of a boolean, and then pass this pool pointer to subsequent swiotlb functions. There are 9 occurrences of is_swiotlb_buffer() used to test if a buffer is a swiotlb buffer before calling a swiotlb function. To reduce code duplication in getting the pool pointer and passing it as an argument, introduce inline wrappers for this pattern. The generated code is essentially unchanged. Since is_swiotlb_buffer() no longer returns a boolean, rename some functions to reflect the change: * swiotlb_find_pool() becomes __swiotlb_find_pool() * is_swiotlb_buffer() becomes swiotlb_find_pool() * is_xen_swiotlb_buffer() becomes xen_swiotlb_find_pool() With these changes, a round-trip map/unmap pair requires only 2 pool lookups (listed using the new names and wrappers): dma_direct_unmap_page: 1. dma_direct_sync_single_for_cpu -> swiotlb_find_pool 2. swiotlb_tbl_unmap_single -> swiotlb_find_pool These changes come from noticing the inefficiencies in a code review, not from performance measurements. With CONFIG_SWIOTLB_DYNAMIC, __swiotlb_find_pool() is not trivial, and it uses an RCU read lock, so avoiding the redundant calls helps performance in a hot path. When CONFIG_SWIOTLB_DYNAMIC is *not* set, the code size reduction is minimal and the perf benefits are likely negligible, but no harm is done. No functional change is intended. Signed-off-by: Michael Kelley Reviewed-by: Petr Tesarik Signed-off-by: Christoph Hellwig --- drivers/iommu/dma-iommu.c | 11 ++--- drivers/xen/swiotlb-xen.c | 31 ++++++++----- include/linux/scatterlist.h | 2 +- include/linux/swiotlb.h | 105 ++++++++++++++++++++++++++------------------ kernel/dma/direct.c | 10 ++--- kernel/dma/direct.h | 9 ++-- kernel/dma/swiotlb.c | 67 +++++++++++++++------------- 7 files changed, 129 insertions(+), 106 deletions(-) diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c index 43520e7275cc..7b4486238427 100644 --- a/drivers/iommu/dma-iommu.c +++ b/drivers/iommu/dma-iommu.c @@ -1081,8 +1081,7 @@ static void iommu_dma_sync_single_for_cpu(struct device *dev, if (!dev_is_dma_coherent(dev)) arch_sync_dma_for_cpu(phys, size, dir); - if (is_swiotlb_buffer(dev, phys)) - swiotlb_sync_single_for_cpu(dev, phys, size, dir); + swiotlb_sync_single_for_cpu(dev, phys, size, dir); } static void iommu_dma_sync_single_for_device(struct device *dev, @@ -1094,8 +1093,7 @@ static void iommu_dma_sync_single_for_device(struct device *dev, return; phys = iommu_iova_to_phys(iommu_get_dma_domain(dev), dma_handle); - if (is_swiotlb_buffer(dev, phys)) - swiotlb_sync_single_for_device(dev, phys, size, dir); + swiotlb_sync_single_for_device(dev, phys, size, dir); if (!dev_is_dma_coherent(dev)) arch_sync_dma_for_device(phys, size, dir); @@ -1189,7 +1187,7 @@ static dma_addr_t iommu_dma_map_page(struct device *dev, struct page *page, arch_sync_dma_for_device(phys, size, dir); iova = __iommu_dma_map(dev, phys, size, prot, dma_mask); - if (iova == DMA_MAPPING_ERROR && is_swiotlb_buffer(dev, phys)) + if (iova == DMA_MAPPING_ERROR) swiotlb_tbl_unmap_single(dev, phys, size, dir, attrs); return iova; } @@ -1209,8 +1207,7 @@ static void iommu_dma_unmap_page(struct device *dev, dma_addr_t dma_handle, __iommu_dma_unmap(dev, dma_handle, size); - if (unlikely(is_swiotlb_buffer(dev, phys))) - swiotlb_tbl_unmap_single(dev, phys, size, dir, attrs); + swiotlb_tbl_unmap_single(dev, phys, size, dir, attrs); } /* diff --git a/drivers/xen/swiotlb-xen.c b/drivers/xen/swiotlb-xen.c index 6579ae3f6dac..35155258a7e2 100644 --- a/drivers/xen/swiotlb-xen.c +++ b/drivers/xen/swiotlb-xen.c @@ -88,7 +88,8 @@ static inline int range_straddles_page_boundary(phys_addr_t p, size_t size) return 0; } -static int is_xen_swiotlb_buffer(struct device *dev, dma_addr_t dma_addr) +static struct io_tlb_pool *xen_swiotlb_find_pool(struct device *dev, + dma_addr_t dma_addr) { unsigned long bfn = XEN_PFN_DOWN(dma_to_phys(dev, dma_addr)); unsigned long xen_pfn = bfn_to_local_pfn(bfn); @@ -99,8 +100,8 @@ static int is_xen_swiotlb_buffer(struct device *dev, dma_addr_t dma_addr) * in our domain. Therefore _only_ check address within our domain. */ if (pfn_valid(PFN_DOWN(paddr))) - return is_swiotlb_buffer(dev, paddr); - return 0; + return swiotlb_find_pool(dev, paddr); + return NULL; } #ifdef CONFIG_X86 @@ -227,8 +228,9 @@ static dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page, * Ensure that the address returned is DMA'ble */ if (unlikely(!dma_capable(dev, dev_addr, size, true))) { - swiotlb_tbl_unmap_single(dev, map, size, dir, - attrs | DMA_ATTR_SKIP_CPU_SYNC); + __swiotlb_tbl_unmap_single(dev, map, size, dir, + attrs | DMA_ATTR_SKIP_CPU_SYNC, + swiotlb_find_pool(dev, map)); return DMA_MAPPING_ERROR; } @@ -254,6 +256,7 @@ static void xen_swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr, size_t size, enum dma_data_direction dir, unsigned long attrs) { phys_addr_t paddr = xen_dma_to_phys(hwdev, dev_addr); + struct io_tlb_pool *pool; BUG_ON(dir == DMA_NONE); @@ -265,8 +268,10 @@ static void xen_swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr, } /* NOTE: We use dev_addr here, not paddr! */ - if (is_xen_swiotlb_buffer(hwdev, dev_addr)) - swiotlb_tbl_unmap_single(hwdev, paddr, size, dir, attrs); + pool = xen_swiotlb_find_pool(hwdev, dev_addr); + if (pool) + __swiotlb_tbl_unmap_single(hwdev, paddr, size, dir, + attrs, pool); } static void @@ -274,6 +279,7 @@ xen_swiotlb_sync_single_for_cpu(struct device *dev, dma_addr_t dma_addr, size_t size, enum dma_data_direction dir) { phys_addr_t paddr = xen_dma_to_phys(dev, dma_addr); + struct io_tlb_pool *pool; if (!dev_is_dma_coherent(dev)) { if (pfn_valid(PFN_DOWN(dma_to_phys(dev, dma_addr)))) @@ -282,8 +288,9 @@ xen_swiotlb_sync_single_for_cpu(struct device *dev, dma_addr_t dma_addr, xen_dma_sync_for_cpu(dev, dma_addr, size, dir); } - if (is_xen_swiotlb_buffer(dev, dma_addr)) - swiotlb_sync_single_for_cpu(dev, paddr, size, dir); + pool = xen_swiotlb_find_pool(dev, dma_addr); + if (pool) + __swiotlb_sync_single_for_cpu(dev, paddr, size, dir, pool); } static void @@ -291,9 +298,11 @@ xen_swiotlb_sync_single_for_device(struct device *dev, dma_addr_t dma_addr, size_t size, enum dma_data_direction dir) { phys_addr_t paddr = xen_dma_to_phys(dev, dma_addr); + struct io_tlb_pool *pool; - if (is_xen_swiotlb_buffer(dev, dma_addr)) - swiotlb_sync_single_for_device(dev, paddr, size, dir); + pool = xen_swiotlb_find_pool(dev, dma_addr); + if (pool) + __swiotlb_sync_single_for_device(dev, paddr, size, dir, pool); if (!dev_is_dma_coherent(dev)) { if (pfn_valid(PFN_DOWN(dma_to_phys(dev, dma_addr)))) diff --git a/include/linux/scatterlist.h b/include/linux/scatterlist.h index 77df3d7b18a6..e61d164622db 100644 --- a/include/linux/scatterlist.h +++ b/include/linux/scatterlist.h @@ -332,7 +332,7 @@ static inline void sg_dma_unmark_bus_address(struct scatterlist *sg) * Description: * Returns true if the scatterlist was marked for SWIOTLB bouncing. Not all * elements may have been bounced, so the caller would have to check - * individual SG entries with is_swiotlb_buffer(). + * individual SG entries with swiotlb_find_pool(). */ static inline bool sg_dma_is_swiotlb(struct scatterlist *sg) { diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h index 14bc10c1bb23..3dae0f592063 100644 --- a/include/linux/swiotlb.h +++ b/include/linux/swiotlb.h @@ -42,24 +42,6 @@ int swiotlb_init_late(size_t size, gfp_t gfp_mask, int (*remap)(void *tlb, unsigned long nslabs)); extern void __init swiotlb_update_mem_attributes(void); -phys_addr_t swiotlb_tbl_map_single(struct device *hwdev, phys_addr_t phys, - size_t mapping_size, - unsigned int alloc_aligned_mask, enum dma_data_direction dir, - unsigned long attrs); - -extern void swiotlb_tbl_unmap_single(struct device *hwdev, - phys_addr_t tlb_addr, - size_t mapping_size, - enum dma_data_direction dir, - unsigned long attrs); - -void swiotlb_sync_single_for_device(struct device *dev, phys_addr_t tlb_addr, - size_t size, enum dma_data_direction dir); -void swiotlb_sync_single_for_cpu(struct device *dev, phys_addr_t tlb_addr, - size_t size, enum dma_data_direction dir); -dma_addr_t swiotlb_map(struct device *dev, phys_addr_t phys, - size_t size, enum dma_data_direction dir, unsigned long attrs); - #ifdef CONFIG_SWIOTLB /** @@ -143,37 +125,27 @@ struct io_tlb_mem { #endif }; -#ifdef CONFIG_SWIOTLB_DYNAMIC - -struct io_tlb_pool *swiotlb_find_pool(struct device *dev, phys_addr_t paddr); - -#else - -static inline struct io_tlb_pool *swiotlb_find_pool(struct device *dev, - phys_addr_t paddr) -{ - return &dev->dma_io_tlb_mem->defpool; -} - -#endif +struct io_tlb_pool *__swiotlb_find_pool(struct device *dev, phys_addr_t paddr); /** - * is_swiotlb_buffer() - check if a physical address belongs to a swiotlb + * swiotlb_find_pool() - find swiotlb pool to which a physical address belongs * @dev: Device which has mapped the buffer. * @paddr: Physical address within the DMA buffer. * - * Check if @paddr points into a bounce buffer. + * Find the swiotlb pool that @paddr points into. * * Return: - * * %true if @paddr points into a bounce buffer - * * %false otherwise + * * pool address if @paddr points into a bounce buffer + * * NULL if @paddr does not point into a bounce buffer. As such, this function + * can be used to determine if @paddr denotes a swiotlb bounce buffer. */ -static inline bool is_swiotlb_buffer(struct device *dev, phys_addr_t paddr) +static inline struct io_tlb_pool *swiotlb_find_pool(struct device *dev, + phys_addr_t paddr) { struct io_tlb_mem *mem = dev->dma_io_tlb_mem; if (!mem) - return false; + return NULL; #ifdef CONFIG_SWIOTLB_DYNAMIC /* @@ -182,16 +154,19 @@ static inline bool is_swiotlb_buffer(struct device *dev, phys_addr_t paddr) * If a SWIOTLB address is checked on another CPU, then it was * presumably loaded by the device driver from an unspecified private * data structure. Make sure that this load is ordered before reading - * dev->dma_uses_io_tlb here and mem->pools in swiotlb_find_pool(). + * dev->dma_uses_io_tlb here and mem->pools in __swiotlb_find_pool(). * * This barrier pairs with smp_mb() in swiotlb_find_slots(). */ smp_rmb(); - return READ_ONCE(dev->dma_uses_io_tlb) && - swiotlb_find_pool(dev, paddr); + if (READ_ONCE(dev->dma_uses_io_tlb)) + return __swiotlb_find_pool(dev, paddr); #else - return paddr >= mem->defpool.start && paddr < mem->defpool.end; + if (paddr >= mem->defpool.start && paddr < mem->defpool.end) + return &mem->defpool; #endif + + return NULL; } static inline bool is_swiotlb_force_bounce(struct device *dev) @@ -219,9 +194,10 @@ static inline void swiotlb_dev_init(struct device *dev) { } -static inline bool is_swiotlb_buffer(struct device *dev, phys_addr_t paddr) +static inline struct io_tlb_pool *swiotlb_find_pool(struct device *dev, + phys_addr_t paddr) { - return false; + return NULL; } static inline bool is_swiotlb_force_bounce(struct device *dev) { @@ -260,6 +236,49 @@ static inline phys_addr_t default_swiotlb_limit(void) } #endif /* CONFIG_SWIOTLB */ +phys_addr_t swiotlb_tbl_map_single(struct device *hwdev, phys_addr_t phys, + size_t mapping_size, unsigned int alloc_aligned_mask, + enum dma_data_direction dir, unsigned long attrs); +dma_addr_t swiotlb_map(struct device *dev, phys_addr_t phys, + size_t size, enum dma_data_direction dir, unsigned long attrs); + +void __swiotlb_tbl_unmap_single(struct device *hwdev, phys_addr_t tlb_addr, + size_t mapping_size, enum dma_data_direction dir, + unsigned long attrs, struct io_tlb_pool *pool); +static inline void swiotlb_tbl_unmap_single(struct device *dev, + phys_addr_t addr, size_t size, enum dma_data_direction dir, + unsigned long attrs) +{ + struct io_tlb_pool *pool = swiotlb_find_pool(dev, addr); + + if (unlikely(pool)) + __swiotlb_tbl_unmap_single(dev, addr, size, dir, attrs, pool); +} + +void __swiotlb_sync_single_for_device(struct device *dev, phys_addr_t tlb_addr, + size_t size, enum dma_data_direction dir, + struct io_tlb_pool *pool); +static inline void swiotlb_sync_single_for_device(struct device *dev, + phys_addr_t addr, size_t size, enum dma_data_direction dir) +{ + struct io_tlb_pool *pool = swiotlb_find_pool(dev, addr); + + if (unlikely(pool)) + __swiotlb_sync_single_for_device(dev, addr, size, dir, pool); +} + +void __swiotlb_sync_single_for_cpu(struct device *dev, phys_addr_t tlb_addr, + size_t size, enum dma_data_direction dir, + struct io_tlb_pool *pool); +static inline void swiotlb_sync_single_for_cpu(struct device *dev, + phys_addr_t addr, size_t size, enum dma_data_direction dir) +{ + struct io_tlb_pool *pool = swiotlb_find_pool(dev, addr); + + if (unlikely(pool)) + __swiotlb_sync_single_for_cpu(dev, addr, size, dir, pool); +} + extern void swiotlb_print_info(void); #ifdef CONFIG_DMA_RESTRICTED_POOL diff --git a/kernel/dma/direct.c b/kernel/dma/direct.c index 4d543b1e9d57..4480a3cd92e0 100644 --- a/kernel/dma/direct.c +++ b/kernel/dma/direct.c @@ -404,9 +404,7 @@ void dma_direct_sync_sg_for_device(struct device *dev, for_each_sg(sgl, sg, nents, i) { phys_addr_t paddr = dma_to_phys(dev, sg_dma_address(sg)); - if (unlikely(is_swiotlb_buffer(dev, paddr))) - swiotlb_sync_single_for_device(dev, paddr, sg->length, - dir); + swiotlb_sync_single_for_device(dev, paddr, sg->length, dir); if (!dev_is_dma_coherent(dev)) arch_sync_dma_for_device(paddr, sg->length, @@ -430,9 +428,7 @@ void dma_direct_sync_sg_for_cpu(struct device *dev, if (!dev_is_dma_coherent(dev)) arch_sync_dma_for_cpu(paddr, sg->length, dir); - if (unlikely(is_swiotlb_buffer(dev, paddr))) - swiotlb_sync_single_for_cpu(dev, paddr, sg->length, - dir); + swiotlb_sync_single_for_cpu(dev, paddr, sg->length, dir); if (dir == DMA_FROM_DEVICE) arch_dma_mark_clean(paddr, sg->length); @@ -640,7 +636,7 @@ size_t dma_direct_max_mapping_size(struct device *dev) bool dma_direct_need_sync(struct device *dev, dma_addr_t dma_addr) { return !dev_is_dma_coherent(dev) || - is_swiotlb_buffer(dev, dma_to_phys(dev, dma_addr)); + swiotlb_find_pool(dev, dma_to_phys(dev, dma_addr)); } /** diff --git a/kernel/dma/direct.h b/kernel/dma/direct.h index 18d346118fe8..d2c0b7e632fc 100644 --- a/kernel/dma/direct.h +++ b/kernel/dma/direct.h @@ -58,8 +58,7 @@ static inline void dma_direct_sync_single_for_device(struct device *dev, { phys_addr_t paddr = dma_to_phys(dev, addr); - if (unlikely(is_swiotlb_buffer(dev, paddr))) - swiotlb_sync_single_for_device(dev, paddr, size, dir); + swiotlb_sync_single_for_device(dev, paddr, size, dir); if (!dev_is_dma_coherent(dev)) arch_sync_dma_for_device(paddr, size, dir); @@ -75,8 +74,7 @@ static inline void dma_direct_sync_single_for_cpu(struct device *dev, arch_sync_dma_for_cpu_all(); } - if (unlikely(is_swiotlb_buffer(dev, paddr))) - swiotlb_sync_single_for_cpu(dev, paddr, size, dir); + swiotlb_sync_single_for_cpu(dev, paddr, size, dir); if (dir == DMA_FROM_DEVICE) arch_dma_mark_clean(paddr, size); @@ -121,8 +119,7 @@ static inline void dma_direct_unmap_page(struct device *dev, dma_addr_t addr, if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC)) dma_direct_sync_single_for_cpu(dev, addr, size, dir); - if (unlikely(is_swiotlb_buffer(dev, phys))) - swiotlb_tbl_unmap_single(dev, phys, size, dir, + swiotlb_tbl_unmap_single(dev, phys, size, dir, attrs | DMA_ATTR_SKIP_CPU_SYNC); } #endif /* _KERNEL_DMA_DIRECT_H */ diff --git a/kernel/dma/swiotlb.c b/kernel/dma/swiotlb.c index fe1ccb53596f..b4d8167b2f8d 100644 --- a/kernel/dma/swiotlb.c +++ b/kernel/dma/swiotlb.c @@ -763,16 +763,18 @@ static void swiotlb_dyn_free(struct rcu_head *rcu) } /** - * swiotlb_find_pool() - find the IO TLB pool for a physical address + * __swiotlb_find_pool() - find the IO TLB pool for a physical address * @dev: Device which has mapped the DMA buffer. * @paddr: Physical address within the DMA buffer. * * Find the IO TLB memory pool descriptor which contains the given physical - * address, if any. + * address, if any. This function is for use only when the dev is known to + * be using swiotlb. Use swiotlb_find_pool() for the more general case + * when this condition is not met. * * Return: Memory pool which contains @paddr, or %NULL if none. */ -struct io_tlb_pool *swiotlb_find_pool(struct device *dev, phys_addr_t paddr) +struct io_tlb_pool *__swiotlb_find_pool(struct device *dev, phys_addr_t paddr) { struct io_tlb_mem *mem = dev->dma_io_tlb_mem; struct io_tlb_pool *pool; @@ -855,9 +857,8 @@ static unsigned int swiotlb_align_offset(struct device *dev, * Bounce: copy the swiotlb buffer from or back to the original dma location */ static void swiotlb_bounce(struct device *dev, phys_addr_t tlb_addr, size_t size, - enum dma_data_direction dir) + enum dma_data_direction dir, struct io_tlb_pool *mem) { - struct io_tlb_pool *mem = swiotlb_find_pool(dev, tlb_addr); int index = (tlb_addr - mem->start) >> IO_TLB_SHIFT; phys_addr_t orig_addr = mem->slots[index].orig_addr; size_t alloc_size = mem->slots[index].alloc_size; @@ -1243,7 +1244,7 @@ found: * that was made by swiotlb_dyn_alloc() on a third CPU (cf. multicopy * atomicity). * - * See also the comment in is_swiotlb_buffer(). + * See also the comment in swiotlb_find_pool(). */ smp_mb(); @@ -1435,13 +1436,13 @@ phys_addr_t swiotlb_tbl_map_single(struct device *dev, phys_addr_t orig_addr, * hardware behavior. Use of swiotlb is supposed to be transparent, * i.e. swiotlb must not corrupt memory by clobbering unwritten bytes. */ - swiotlb_bounce(dev, tlb_addr, mapping_size, DMA_TO_DEVICE); + swiotlb_bounce(dev, tlb_addr, mapping_size, DMA_TO_DEVICE, pool); return tlb_addr; } -static void swiotlb_release_slots(struct device *dev, phys_addr_t tlb_addr) +static void swiotlb_release_slots(struct device *dev, phys_addr_t tlb_addr, + struct io_tlb_pool *mem) { - struct io_tlb_pool *mem = swiotlb_find_pool(dev, tlb_addr); unsigned long flags; unsigned int offset = swiotlb_align_offset(dev, 0, tlb_addr); int index, nslots, aindex; @@ -1505,11 +1506,9 @@ static void swiotlb_release_slots(struct device *dev, phys_addr_t tlb_addr) * * Return: %true if @tlb_addr belonged to a transient pool that was released. */ -static bool swiotlb_del_transient(struct device *dev, phys_addr_t tlb_addr) +static bool swiotlb_del_transient(struct device *dev, phys_addr_t tlb_addr, + struct io_tlb_pool *pool) { - struct io_tlb_pool *pool; - - pool = swiotlb_find_pool(dev, tlb_addr); if (!pool->transient) return false; @@ -1522,7 +1521,7 @@ static bool swiotlb_del_transient(struct device *dev, phys_addr_t tlb_addr) #else /* !CONFIG_SWIOTLB_DYNAMIC */ static inline bool swiotlb_del_transient(struct device *dev, - phys_addr_t tlb_addr) + phys_addr_t tlb_addr, struct io_tlb_pool *pool) { return false; } @@ -1532,36 +1531,39 @@ static inline bool swiotlb_del_transient(struct device *dev, /* * tlb_addr is the physical address of the bounce buffer to unmap. */ -void swiotlb_tbl_unmap_single(struct device *dev, phys_addr_t tlb_addr, - size_t mapping_size, enum dma_data_direction dir, - unsigned long attrs) +void __swiotlb_tbl_unmap_single(struct device *dev, phys_addr_t tlb_addr, + size_t mapping_size, enum dma_data_direction dir, + unsigned long attrs, struct io_tlb_pool *pool) { /* * First, sync the memory before unmapping the entry */ if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC) && (dir == DMA_FROM_DEVICE || dir == DMA_BIDIRECTIONAL)) - swiotlb_bounce(dev, tlb_addr, mapping_size, DMA_FROM_DEVICE); + swiotlb_bounce(dev, tlb_addr, mapping_size, + DMA_FROM_DEVICE, pool); - if (swiotlb_del_transient(dev, tlb_addr)) + if (swiotlb_del_transient(dev, tlb_addr, pool)) return; - swiotlb_release_slots(dev, tlb_addr); + swiotlb_release_slots(dev, tlb_addr, pool); } -void swiotlb_sync_single_for_device(struct device *dev, phys_addr_t tlb_addr, - size_t size, enum dma_data_direction dir) +void __swiotlb_sync_single_for_device(struct device *dev, phys_addr_t tlb_addr, + size_t size, enum dma_data_direction dir, + struct io_tlb_pool *pool) { if (dir == DMA_TO_DEVICE || dir == DMA_BIDIRECTIONAL) - swiotlb_bounce(dev, tlb_addr, size, DMA_TO_DEVICE); + swiotlb_bounce(dev, tlb_addr, size, DMA_TO_DEVICE, pool); else BUG_ON(dir != DMA_FROM_DEVICE); } -void swiotlb_sync_single_for_cpu(struct device *dev, phys_addr_t tlb_addr, - size_t size, enum dma_data_direction dir) +void __swiotlb_sync_single_for_cpu(struct device *dev, phys_addr_t tlb_addr, + size_t size, enum dma_data_direction dir, + struct io_tlb_pool *pool) { if (dir == DMA_FROM_DEVICE || dir == DMA_BIDIRECTIONAL) - swiotlb_bounce(dev, tlb_addr, size, DMA_FROM_DEVICE); + swiotlb_bounce(dev, tlb_addr, size, DMA_FROM_DEVICE, pool); else BUG_ON(dir != DMA_TO_DEVICE); } @@ -1585,8 +1587,9 @@ dma_addr_t swiotlb_map(struct device *dev, phys_addr_t paddr, size_t size, /* Ensure that the address returned is DMA'ble */ dma_addr = phys_to_dma_unencrypted(dev, swiotlb_addr); if (unlikely(!dma_capable(dev, dma_addr, size, true))) { - swiotlb_tbl_unmap_single(dev, swiotlb_addr, size, dir, - attrs | DMA_ATTR_SKIP_CPU_SYNC); + __swiotlb_tbl_unmap_single(dev, swiotlb_addr, size, dir, + attrs | DMA_ATTR_SKIP_CPU_SYNC, + swiotlb_find_pool(dev, swiotlb_addr)); dev_WARN_ONCE(dev, 1, "swiotlb addr %pad+%zu overflow (mask %llx, bus limit %llx).\n", &dma_addr, size, *dev->dma_mask, dev->bus_dma_limit); @@ -1764,7 +1767,7 @@ struct page *swiotlb_alloc(struct device *dev, size_t size) if (unlikely(!PAGE_ALIGNED(tlb_addr))) { dev_WARN_ONCE(dev, 1, "Cannot allocate pages from non page-aligned swiotlb addr 0x%pa.\n", &tlb_addr); - swiotlb_release_slots(dev, tlb_addr); + swiotlb_release_slots(dev, tlb_addr, pool); return NULL; } @@ -1774,11 +1777,13 @@ struct page *swiotlb_alloc(struct device *dev, size_t size) bool swiotlb_free(struct device *dev, struct page *page, size_t size) { phys_addr_t tlb_addr = page_to_phys(page); + struct io_tlb_pool *pool; - if (!is_swiotlb_buffer(dev, tlb_addr)) + pool = swiotlb_find_pool(dev, tlb_addr); + if (!pool) return false; - swiotlb_release_slots(dev, tlb_addr); + swiotlb_release_slots(dev, tlb_addr, pool); return true; } -- cgit v1.3.1 From b69bdba5a37eb6224039e9572e0e98fc3a931fee Mon Sep 17 00:00:00 2001 From: Yang Li Date: Fri, 12 Jul 2024 15:38:16 +0800 Subject: swiotlb: fix kernel-doc description for swiotlb_del_transient Describe the pool argument in the kernel-doc comment for swiotlb_del_transient. Reported-by: Abaci Robot Signed-off-by: Yang Li Signed-off-by: Christoph Hellwig --- kernel/dma/swiotlb.c | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/dma/swiotlb.c b/kernel/dma/swiotlb.c index b4d8167b2f8d..df68d29740a0 100644 --- a/kernel/dma/swiotlb.c +++ b/kernel/dma/swiotlb.c @@ -1500,6 +1500,7 @@ static void swiotlb_release_slots(struct device *dev, phys_addr_t tlb_addr, * swiotlb_del_transient() - delete a transient memory pool * @dev: Device which mapped the buffer. * @tlb_addr: Physical address within a bounce buffer. + * @pool: Pointer to the transient memory pool to be checked and deleted. * * Check whether the address belongs to a transient SWIOTLB memory pool. * If yes, then delete the pool. -- cgit v1.3.1