linux-stable/include/linux/iova.h

111 lines
3.3 KiB
C
Raw Normal View History

/*
* Copyright (c) 2006, Intel Corporation.
*
* This file is released under the GPLv2.
*
* Copyright (C) 2006-2008 Intel Corporation
* Author: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
*
*/
#ifndef _IOVA_H_
#define _IOVA_H_
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/rbtree.h>
#include <linux/dma-mapping.h>
/* iova structure */
struct iova {
struct rb_node node;
iommu/iova: introduce per-cpu caching to iova allocation IOVA allocation has two problems that impede high-throughput I/O. First, it can do a linear search over the allocated IOVA ranges. Second, the rbtree spinlock that serializes IOVA allocations becomes contended. Address these problems by creating an API for caching allocated IOVA ranges, so that the IOVA allocator isn't accessed frequently. This patch adds a per-CPU cache, from which CPUs can alloc/free IOVAs without taking the rbtree spinlock. The per-CPU caches are backed by a global cache, to avoid invoking the (linear-time) IOVA allocator without needing to make the per-CPU cache size excessive. This design is based on magazines, as described in "Magazines and Vmem: Extending the Slab Allocator to Many CPUs and Arbitrary Resources" (currently available at https://www.usenix.org/legacy/event/usenix01/bonwick.html) Adding caching on top of the existing rbtree allocator maintains the property that IOVAs are densely packed in the IO virtual address space, which is important for keeping IOMMU page table usage low. To keep the cache size reasonable, we bound the IOVA space a CPU can cache by 32 MiB (we cache a bounded number of IOVA ranges, and only ranges of size <= 128 KiB). The shared global cache is bounded at 4 MiB of IOVA space. Signed-off-by: Omer Peleg <omer@cs.technion.ac.il> [mad@cs.technion.ac.il: rebased, cleaned up and reworded the commit message] Signed-off-by: Adam Morrison <mad@cs.technion.ac.il> Reviewed-by: Shaohua Li <shli@fb.com> Reviewed-by: Ben Serebrin <serebrin@google.com> [dwmw2: split out VT-d part into a separate patch] Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2016-04-20 08:34:11 +00:00
unsigned long pfn_hi; /* Highest allocated pfn */
unsigned long pfn_lo; /* Lowest allocated pfn */
};
struct iova_magazine;
struct iova_cpu_rcache;
#define IOVA_RANGE_CACHE_MAX_SIZE 6 /* log of max cached IOVA range size (in pages) */
#define MAX_GLOBAL_MAGS 32 /* magazines per bin */
struct iova_rcache {
spinlock_t lock;
unsigned long depot_size;
struct iova_magazine *depot[MAX_GLOBAL_MAGS];
struct iova_cpu_rcache __percpu *cpu_rcaches;
};
/* holds all the iova translations for a domain */
struct iova_domain {
spinlock_t iova_rbtree_lock; /* Lock to protect update of rbtree */
struct rb_root rbroot; /* iova domain rbtree root */
struct rb_node *cached32_node; /* Save last alloced node */
unsigned long granule; /* pfn granularity for this domain */
unsigned long start_pfn; /* Lower limit for this domain */
unsigned long dma_32bit_pfn;
iommu/iova: introduce per-cpu caching to iova allocation IOVA allocation has two problems that impede high-throughput I/O. First, it can do a linear search over the allocated IOVA ranges. Second, the rbtree spinlock that serializes IOVA allocations becomes contended. Address these problems by creating an API for caching allocated IOVA ranges, so that the IOVA allocator isn't accessed frequently. This patch adds a per-CPU cache, from which CPUs can alloc/free IOVAs without taking the rbtree spinlock. The per-CPU caches are backed by a global cache, to avoid invoking the (linear-time) IOVA allocator without needing to make the per-CPU cache size excessive. This design is based on magazines, as described in "Magazines and Vmem: Extending the Slab Allocator to Many CPUs and Arbitrary Resources" (currently available at https://www.usenix.org/legacy/event/usenix01/bonwick.html) Adding caching on top of the existing rbtree allocator maintains the property that IOVAs are densely packed in the IO virtual address space, which is important for keeping IOMMU page table usage low. To keep the cache size reasonable, we bound the IOVA space a CPU can cache by 32 MiB (we cache a bounded number of IOVA ranges, and only ranges of size <= 128 KiB). The shared global cache is bounded at 4 MiB of IOVA space. Signed-off-by: Omer Peleg <omer@cs.technion.ac.il> [mad@cs.technion.ac.il: rebased, cleaned up and reworded the commit message] Signed-off-by: Adam Morrison <mad@cs.technion.ac.il> Reviewed-by: Shaohua Li <shli@fb.com> Reviewed-by: Ben Serebrin <serebrin@google.com> [dwmw2: split out VT-d part into a separate patch] Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2016-04-20 08:34:11 +00:00
struct iova_rcache rcaches[IOVA_RANGE_CACHE_MAX_SIZE]; /* IOVA range caches */
};
static inline unsigned long iova_size(struct iova *iova)
{
return iova->pfn_hi - iova->pfn_lo + 1;
}
static inline unsigned long iova_shift(struct iova_domain *iovad)
{
return __ffs(iovad->granule);
}
static inline unsigned long iova_mask(struct iova_domain *iovad)
{
return iovad->granule - 1;
}
static inline size_t iova_offset(struct iova_domain *iovad, dma_addr_t iova)
{
return iova & iova_mask(iovad);
}
static inline size_t iova_align(struct iova_domain *iovad, size_t size)
{
return ALIGN(size, iovad->granule);
}
static inline dma_addr_t iova_dma_addr(struct iova_domain *iovad, struct iova *iova)
{
return (dma_addr_t)iova->pfn_lo << iova_shift(iovad);
}
static inline unsigned long iova_pfn(struct iova_domain *iovad, dma_addr_t iova)
{
return iova >> iova_shift(iovad);
}
int iova_cache_get(void);
void iova_cache_put(void);
struct iova *alloc_iova_mem(void);
void free_iova_mem(struct iova *iova);
void free_iova(struct iova_domain *iovad, unsigned long pfn);
void __free_iova(struct iova_domain *iovad, struct iova *iova);
struct iova *alloc_iova(struct iova_domain *iovad, unsigned long size,
intel-iommu: optimize sg map/unmap calls This patch adds PageSelectiveInvalidation support replacing existing DomainSelectiveInvalidation for intel_{map/unmap}_sg() calls and also enables to mapping one big contiguous DMA virtual address which is mapped to discontiguous physical address for SG map/unmap calls. "Doamin selective invalidations" wipes out the IOMMU address translation cache based on domain ID where as "Page selective invalidations" wipes out the IOMMU address translation cache for that address mask range which is more cache friendly when compared to Domain selective invalidations. Here is how it is done. 1) changes to iova.c alloc_iova() now takes a bool size_aligned argument, which when when set, returns the io virtual address that is naturally aligned to 2 ^ x, where x is the order of the size requested. Returning this io vitual address which is naturally aligned helps iommu to do the "page selective invalidations" which is IOMMU cache friendly over "domain selective invalidations". 2) Changes to driver/pci/intel-iommu.c Clean up intel_{map/unmap}_{single/sg} () calls so that s/g map/unamp calls is no more dependent on intel_{map/unmap}_single() intel_map_sg() now computes the total DMA virtual address required and allocates the size aligned total DMA virtual address and maps the discontiguous physical address to the allocated contiguous DMA virtual address. In the intel_unmap_sg() case since the DMA virtual address is contiguous and size_aligned, PageSelectiveInvalidation is used replacing earlier DomainSelectiveInvalidations. Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com> Cc: Greg KH <greg@kroah.com> Cc: Ashok Raj <ashok.raj@intel.com> Cc: Suresh B <suresh.b.siddha@intel.com> Cc: Andi Kleen <ak@suse.de> Cc: Arjan van de Ven <arjan@infradead.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-21 23:41:58 +00:00
unsigned long limit_pfn,
bool size_aligned);
iommu/iova: introduce per-cpu caching to iova allocation IOVA allocation has two problems that impede high-throughput I/O. First, it can do a linear search over the allocated IOVA ranges. Second, the rbtree spinlock that serializes IOVA allocations becomes contended. Address these problems by creating an API for caching allocated IOVA ranges, so that the IOVA allocator isn't accessed frequently. This patch adds a per-CPU cache, from which CPUs can alloc/free IOVAs without taking the rbtree spinlock. The per-CPU caches are backed by a global cache, to avoid invoking the (linear-time) IOVA allocator without needing to make the per-CPU cache size excessive. This design is based on magazines, as described in "Magazines and Vmem: Extending the Slab Allocator to Many CPUs and Arbitrary Resources" (currently available at https://www.usenix.org/legacy/event/usenix01/bonwick.html) Adding caching on top of the existing rbtree allocator maintains the property that IOVAs are densely packed in the IO virtual address space, which is important for keeping IOMMU page table usage low. To keep the cache size reasonable, we bound the IOVA space a CPU can cache by 32 MiB (we cache a bounded number of IOVA ranges, and only ranges of size <= 128 KiB). The shared global cache is bounded at 4 MiB of IOVA space. Signed-off-by: Omer Peleg <omer@cs.technion.ac.il> [mad@cs.technion.ac.il: rebased, cleaned up and reworded the commit message] Signed-off-by: Adam Morrison <mad@cs.technion.ac.il> Reviewed-by: Shaohua Li <shli@fb.com> Reviewed-by: Ben Serebrin <serebrin@google.com> [dwmw2: split out VT-d part into a separate patch] Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2016-04-20 08:34:11 +00:00
void free_iova_fast(struct iova_domain *iovad, unsigned long pfn,
unsigned long size);
unsigned long alloc_iova_fast(struct iova_domain *iovad, unsigned long size,
unsigned long limit_pfn);
struct iova *reserve_iova(struct iova_domain *iovad, unsigned long pfn_lo,
unsigned long pfn_hi);
void copy_reserved_iova(struct iova_domain *from, struct iova_domain *to);
void init_iova_domain(struct iova_domain *iovad, unsigned long granule,
unsigned long start_pfn, unsigned long pfn_32bit);
struct iova *find_iova(struct iova_domain *iovad, unsigned long pfn);
void put_iova_domain(struct iova_domain *iovad);
struct iova *split_and_remove_iova(struct iova_domain *iovad,
struct iova *iova, unsigned long pfn_lo, unsigned long pfn_hi);
iommu/iova: introduce per-cpu caching to iova allocation IOVA allocation has two problems that impede high-throughput I/O. First, it can do a linear search over the allocated IOVA ranges. Second, the rbtree spinlock that serializes IOVA allocations becomes contended. Address these problems by creating an API for caching allocated IOVA ranges, so that the IOVA allocator isn't accessed frequently. This patch adds a per-CPU cache, from which CPUs can alloc/free IOVAs without taking the rbtree spinlock. The per-CPU caches are backed by a global cache, to avoid invoking the (linear-time) IOVA allocator without needing to make the per-CPU cache size excessive. This design is based on magazines, as described in "Magazines and Vmem: Extending the Slab Allocator to Many CPUs and Arbitrary Resources" (currently available at https://www.usenix.org/legacy/event/usenix01/bonwick.html) Adding caching on top of the existing rbtree allocator maintains the property that IOVAs are densely packed in the IO virtual address space, which is important for keeping IOMMU page table usage low. To keep the cache size reasonable, we bound the IOVA space a CPU can cache by 32 MiB (we cache a bounded number of IOVA ranges, and only ranges of size <= 128 KiB). The shared global cache is bounded at 4 MiB of IOVA space. Signed-off-by: Omer Peleg <omer@cs.technion.ac.il> [mad@cs.technion.ac.il: rebased, cleaned up and reworded the commit message] Signed-off-by: Adam Morrison <mad@cs.technion.ac.il> Reviewed-by: Shaohua Li <shli@fb.com> Reviewed-by: Ben Serebrin <serebrin@google.com> [dwmw2: split out VT-d part into a separate patch] Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2016-04-20 08:34:11 +00:00
void free_cpu_cached_iovas(unsigned int cpu, struct iova_domain *iovad);
#endif