diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index 69ff0b64ba591d28b79da3a8b66c446b3dafecac..fe3b216e39d54b578bf9fa8d99984cac41e998c9 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -5663,6 +5663,7 @@ CONFIG_INFINIBAND_ON_DEMAND_PAGING=y CONFIG_INFINIBAND_ADDR_TRANS=y CONFIG_INFINIBAND_ADDR_TRANS_CONFIGFS=y CONFIG_INFINIBAND_VIRT_DMA=y +CONFIG_INFINIBAND_PEER_MEMORY=y # CONFIG_INFINIBAND_MTHCA is not set CONFIG_INFINIBAND_CXGB4=m # CONFIG_INFINIBAND_EFA is not set diff --git a/arch/x86/configs/openeuler_defconfig b/arch/x86/configs/openeuler_defconfig index f3b810d0cf47c277830cac0708cb5e4e7df5350b..d3de89970c1f96b5f0b0b0507ff9350b92dc6b7b 100644 --- a/arch/x86/configs/openeuler_defconfig +++ b/arch/x86/configs/openeuler_defconfig @@ -6286,6 +6286,7 @@ CONFIG_INFINIBAND_ON_DEMAND_PAGING=y CONFIG_INFINIBAND_ADDR_TRANS=y CONFIG_INFINIBAND_ADDR_TRANS_CONFIGFS=y CONFIG_INFINIBAND_VIRT_DMA=y +CONFIG_INFINIBAND_PEER_MEMORY=y # CONFIG_INFINIBAND_MTHCA is not set # CONFIG_INFINIBAND_QIB is not set CONFIG_INFINIBAND_CXGB4=m diff --git a/drivers/infiniband/Kconfig b/drivers/infiniband/Kconfig index 194e824832d6bc848187f768e9ba7f673c0cd0fc..1213a8c334005bb44cf1fe7ad720690b74b9f062 100644 --- a/drivers/infiniband/Kconfig +++ b/drivers/infiniband/Kconfig @@ -73,6 +73,16 @@ config INFINIBAND_ADDR_TRANS_CONFIGFS This allows the user to config the default GID type that the CM uses for each device, when initiaing new connections. +config INFINIBAND_PEER_MEMORY + bool "InfiniBand peer memory support" + depends on INFINIBAND_USER_MEM + default n + help + Peer memory support for the InfinBand subsystem. This + enables GPU drivers to provide peer memory operations + and allows InfiniBand hardware drivers to utilize GPU + peer memory. + config INFINIBAND_VIRT_DMA def_bool !HIGHMEM diff --git a/drivers/infiniband/core/Makefile b/drivers/infiniband/core/Makefile index ccf2670ef45efe83c4b1d88806ff23e4de2add63..3cc7c1837170a1cc68d0082018043646984cc76e 100644 --- a/drivers/infiniband/core/Makefile +++ b/drivers/infiniband/core/Makefile @@ -41,4 +41,5 @@ ib_uverbs-y := uverbs_main.o uverbs_cmd.o uverbs_marshall.o \ uverbs_std_types_wq.o \ uverbs_std_types_qp.o ib_uverbs-$(CONFIG_INFINIBAND_USER_MEM) += umem.o +ib_uverbs-$(CONFIG_INFINIBAND_PEER_MEMORY) += peer_mem.o ib_uverbs-$(CONFIG_INFINIBAND_ON_DEMAND_PAGING) += umem_odp.o diff --git a/drivers/infiniband/core/ib_peer_mem.h b/drivers/infiniband/core/ib_peer_mem.h new file mode 100644 index 0000000000000000000000000000000000000000..248530a7f931a561727fbfb5f36a045ac07915f6 --- /dev/null +++ b/drivers/infiniband/core/ib_peer_mem.h @@ -0,0 +1,65 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ +/* + * Copyright (c) 2014-2020, Mellanox Technologies. All rights reserved. + */ +#ifndef RDMA_IB_PEER_MEM_H +#define RDMA_IB_PEER_MEM_H + +#include +#include +#include +#include + +struct ib_peer_memory_statistics { + atomic64_t num_alloc_mrs; + atomic64_t num_dealloc_mrs; + atomic64_t num_reg_pages; + atomic64_t num_dereg_pages; + atomic64_t num_reg_bytes; + atomic64_t num_dereg_bytes; + unsigned long num_free_callbacks; +}; + +struct ib_peer_memory_client { + struct kobject kobj; + refcount_t usecnt; + struct completion usecnt_zero; + const struct peer_memory_client *peer_mem; + struct list_head core_peer_list; + struct ib_peer_memory_statistics stats; + struct xarray umem_xa; + u32 xa_cyclic_next; + bool invalidation_required; +}; + +enum ib_umem_mapped_state { + UMEM_PEER_UNMAPPED, + UMEM_PEER_MAPPED, + UMEM_PEER_INVALIDATED, +}; + +struct ib_umem_peer { + struct ib_umem umem; + struct kref kref; + /* peer memory that manages this umem */ + struct ib_peer_memory_client *ib_peer_client; + void *peer_client_context; + umem_invalidate_func_t invalidation_func; + void *invalidation_private; + struct mutex mapping_lock; + enum ib_umem_mapped_state mapped_state; + u32 xa_id; + struct scatterlist *first_sg; + dma_addr_t first_dma_address; + unsigned int first_dma_length; + unsigned int first_length; + struct scatterlist *last_sg; + unsigned int last_dma_length; + unsigned int last_length; +}; + +struct ib_umem *ib_peer_umem_get(struct ib_umem *old_umem, int old_ret, + unsigned long peer_mem_flags); +void ib_peer_umem_release(struct ib_umem *umem); + +#endif diff --git a/drivers/infiniband/core/peer_mem.c b/drivers/infiniband/core/peer_mem.c new file mode 100644 index 0000000000000000000000000000000000000000..0108e6134979d028f7cb02b9f567cdf12009100d --- /dev/null +++ b/drivers/infiniband/core/peer_mem.c @@ -0,0 +1,682 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* + * Copyright (c) 2014-2020, Mellanox Technologies. All rights reserved. + */ + +#include +#include +#include +#include "ib_peer_mem.h" + +static DEFINE_MUTEX(peer_memory_mutex); +static LIST_HEAD(peer_memory_list); +static struct kobject *peers_kobj; +#define PEER_NO_INVALIDATION_ID U32_MAX + +static int ib_invalidate_peer_memory(void *reg_handle, u64 core_context); + +struct peer_mem_attribute { + struct attribute attr; + ssize_t (*show)(struct ib_peer_memory_client *ib_peer_client, + struct peer_mem_attribute *attr, char *buf); + ssize_t (*store)(struct ib_peer_memory_client *ib_peer_client, + struct peer_mem_attribute *attr, const char *buf, + size_t count); +}; + +#define PEER_ATTR_RO(_name) \ + struct peer_mem_attribute peer_attr_ ## _name = __ATTR_RO(_name) + +static ssize_t version_show(struct ib_peer_memory_client *ib_peer_client, + struct peer_mem_attribute *attr, char *buf) +{ + return scnprintf(buf, PAGE_SIZE, "%s\n", + ib_peer_client->peer_mem->version); +} +static PEER_ATTR_RO(version); + +static ssize_t num_alloc_mrs_show(struct ib_peer_memory_client *ib_peer_client, + struct peer_mem_attribute *attr, char *buf) +{ + return scnprintf(buf, PAGE_SIZE, "%llu\n", + (u64)atomic64_read(&ib_peer_client->stats.num_alloc_mrs)); +} +static PEER_ATTR_RO(num_alloc_mrs); + +static ssize_t +num_dealloc_mrs_show(struct ib_peer_memory_client *ib_peer_client, + struct peer_mem_attribute *attr, char *buf) + +{ + return scnprintf(buf, PAGE_SIZE, "%llu\n", + (u64)atomic64_read(&ib_peer_client->stats.num_dealloc_mrs)); +} +static PEER_ATTR_RO(num_dealloc_mrs); + +static ssize_t num_reg_pages_show(struct ib_peer_memory_client *ib_peer_client, + struct peer_mem_attribute *attr, char *buf) +{ + return scnprintf(buf, PAGE_SIZE, "%llu\n", + (u64)atomic64_read(&ib_peer_client->stats.num_reg_pages)); +} +static PEER_ATTR_RO(num_reg_pages); + +static ssize_t +num_dereg_pages_show(struct ib_peer_memory_client *ib_peer_client, + struct peer_mem_attribute *attr, char *buf) +{ + return scnprintf(buf, PAGE_SIZE, "%llu\n", + (u64)atomic64_read(&ib_peer_client->stats.num_dereg_pages)); +} +static PEER_ATTR_RO(num_dereg_pages); + +static ssize_t num_reg_bytes_show(struct ib_peer_memory_client *ib_peer_client, + struct peer_mem_attribute *attr, char *buf) +{ + return scnprintf(buf, PAGE_SIZE, "%llu\n", + (u64)atomic64_read(&ib_peer_client->stats.num_reg_bytes)); +} +static PEER_ATTR_RO(num_reg_bytes); + +static ssize_t +num_dereg_bytes_show(struct ib_peer_memory_client *ib_peer_client, + struct peer_mem_attribute *attr, char *buf) +{ + return scnprintf(buf, PAGE_SIZE, "%llu\n", + (u64)atomic64_read(&ib_peer_client->stats.num_dereg_bytes)); +} +static PEER_ATTR_RO(num_dereg_bytes); + +static ssize_t +num_free_callbacks_show(struct ib_peer_memory_client *ib_peer_client, + struct peer_mem_attribute *attr, char *buf) +{ + return scnprintf(buf, PAGE_SIZE, "%lu\n", + ib_peer_client->stats.num_free_callbacks); +} +static PEER_ATTR_RO(num_free_callbacks); + +static struct attribute *peer_mem_attrs[] = { + &peer_attr_version.attr, + &peer_attr_num_alloc_mrs.attr, + &peer_attr_num_dealloc_mrs.attr, + &peer_attr_num_reg_pages.attr, + &peer_attr_num_dereg_pages.attr, + &peer_attr_num_reg_bytes.attr, + &peer_attr_num_dereg_bytes.attr, + &peer_attr_num_free_callbacks.attr, + NULL, +}; + +static const struct attribute_group peer_mem_attr_group = { + .attrs = peer_mem_attrs, +}; + +static ssize_t peer_attr_show(struct kobject *kobj, struct attribute *attr, + char *buf) +{ + struct peer_mem_attribute *peer_attr = + container_of(attr, struct peer_mem_attribute, attr); + + if (!peer_attr->show) + return -EIO; + return peer_attr->show(container_of(kobj, struct ib_peer_memory_client, + kobj), + peer_attr, buf); +} + +static const struct sysfs_ops peer_mem_sysfs_ops = { + .show = peer_attr_show, +}; + +static void ib_peer_memory_client_release(struct kobject *kobj) +{ + struct ib_peer_memory_client *ib_peer_client = + container_of(kobj, struct ib_peer_memory_client, kobj); + + kfree(ib_peer_client); +} + +static struct kobj_type peer_mem_type = { + .sysfs_ops = &peer_mem_sysfs_ops, + .release = ib_peer_memory_client_release, +}; + +static int ib_memory_peer_check_mandatory(const struct peer_memory_client + *peer_client) +{ +#define PEER_MEM_MANDATORY_FUNC(x) {offsetof(struct peer_memory_client, x), #x} + int i; + static const struct { + size_t offset; + char *name; + } mandatory_table[] = { + PEER_MEM_MANDATORY_FUNC(acquire), + PEER_MEM_MANDATORY_FUNC(get_pages), + PEER_MEM_MANDATORY_FUNC(put_pages), + PEER_MEM_MANDATORY_FUNC(dma_map), + PEER_MEM_MANDATORY_FUNC(dma_unmap), + }; + + for (i = 0; i < ARRAY_SIZE(mandatory_table); ++i) { + if (!*(void **)((void *)peer_client + + mandatory_table[i].offset)) { + pr_err("Peer memory %s is missing mandatory function %s\n", + peer_client->name, mandatory_table[i].name); + return -EINVAL; + } + } + + return 0; +} + +void * +ib_register_peer_memory_client(const struct peer_memory_client *peer_client, + invalidate_peer_memory *invalidate_callback) +{ + struct ib_peer_memory_client *ib_peer_client; + int ret; + + if (ib_memory_peer_check_mandatory(peer_client)) + return NULL; + + ib_peer_client = kzalloc(sizeof(*ib_peer_client), GFP_KERNEL); + if (!ib_peer_client) + return NULL; + kobject_init(&ib_peer_client->kobj, &peer_mem_type); + refcount_set(&ib_peer_client->usecnt, 1); + init_completion(&ib_peer_client->usecnt_zero); + ib_peer_client->peer_mem = peer_client; + xa_init_flags(&ib_peer_client->umem_xa, XA_FLAGS_ALLOC); + + /* + * If the peer wants the invalidation_callback then all memory users + * linked to that peer must support invalidation. + */ + if (invalidate_callback) { + *invalidate_callback = ib_invalidate_peer_memory; + ib_peer_client->invalidation_required = true; + } + + mutex_lock(&peer_memory_mutex); + if (!peers_kobj) { + /* Created under /sys/kernel/mm */ + peers_kobj = kobject_create_and_add("memory_peers", mm_kobj); + if (!peers_kobj) + goto err_unlock; + } + + ret = kobject_add(&ib_peer_client->kobj, peers_kobj, peer_client->name); + if (ret) + goto err_parent; + + ret = sysfs_create_group(&ib_peer_client->kobj, + &peer_mem_attr_group); + if (ret) + goto err_parent; + list_add_tail(&ib_peer_client->core_peer_list, &peer_memory_list); + mutex_unlock(&peer_memory_mutex); + return ib_peer_client; + +err_parent: + if (list_empty(&peer_memory_list)) { + kobject_put(peers_kobj); + peers_kobj = NULL; + } +err_unlock: + mutex_unlock(&peer_memory_mutex); + kobject_put(&ib_peer_client->kobj); + return NULL; +} +EXPORT_SYMBOL(ib_register_peer_memory_client); + +void ib_unregister_peer_memory_client(void *reg_handle) +{ + struct ib_peer_memory_client *ib_peer_client = reg_handle; + + mutex_lock(&peer_memory_mutex); + list_del(&ib_peer_client->core_peer_list); + if (list_empty(&peer_memory_list)) { + kobject_put(peers_kobj); + peers_kobj = NULL; + } + mutex_unlock(&peer_memory_mutex); + + /* + * Wait for all umems to be destroyed before returning. Once + * ib_unregister_peer_memory_client() returns no umems will call any + * peer_mem ops. + */ + if (refcount_dec_and_test(&ib_peer_client->usecnt)) + complete(&ib_peer_client->usecnt_zero); + wait_for_completion(&ib_peer_client->usecnt_zero); + + kobject_put(&ib_peer_client->kobj); +} +EXPORT_SYMBOL(ib_unregister_peer_memory_client); + +static struct ib_peer_memory_client * +ib_get_peer_client(unsigned long addr, size_t size, + unsigned long peer_mem_flags, void **peer_client_context) +{ + struct ib_peer_memory_client *ib_peer_client; + int ret = 0; + + mutex_lock(&peer_memory_mutex); + list_for_each_entry(ib_peer_client, &peer_memory_list, + core_peer_list) { + if (ib_peer_client->invalidation_required && + (!(peer_mem_flags & IB_PEER_MEM_INVAL_SUPP))) + continue; + ret = ib_peer_client->peer_mem->acquire(addr, size, NULL, NULL, + peer_client_context); + if (ret > 0) { + refcount_inc(&ib_peer_client->usecnt); + mutex_unlock(&peer_memory_mutex); + return ib_peer_client; + } + } + mutex_unlock(&peer_memory_mutex); + return NULL; +} + +static void ib_put_peer_client(struct ib_peer_memory_client *ib_peer_client, + void *peer_client_context) +{ + if (ib_peer_client->peer_mem->release) + ib_peer_client->peer_mem->release(peer_client_context); + if (refcount_dec_and_test(&ib_peer_client->usecnt)) + complete(&ib_peer_client->usecnt_zero); +} + +static void ib_peer_umem_kref_release(struct kref *kref) +{ + struct ib_umem_peer *umem_p = + container_of(kref, struct ib_umem_peer, kref); + + mutex_destroy(&umem_p->mapping_lock); + kfree(umem_p); +} + +static void ib_unmap_peer_client(struct ib_umem_peer *umem_p, + enum ib_umem_mapped_state cur_state, + enum ib_umem_mapped_state to_state) +{ + struct ib_peer_memory_client *ib_peer_client = umem_p->ib_peer_client; + const struct peer_memory_client *peer_mem = ib_peer_client->peer_mem; + struct ib_umem *umem = &umem_p->umem; + + if (cur_state == UMEM_PEER_MAPPED && + (to_state == UMEM_PEER_UNMAPPED || + to_state == UMEM_PEER_INVALIDATED)) { + /* + * In the invalidated state we will never touch the sg again, + * but the client might, so fix it anyhow. + */ + if (umem_p->last_sg) { + umem_p->last_sg->length = umem_p->last_length; + sg_dma_len(umem_p->last_sg) = umem_p->last_dma_length; + } + + if (umem_p->first_sg) { + umem_p->first_sg->dma_address = + umem_p->first_dma_address; + umem_p->first_sg->length = umem_p->first_length; + sg_dma_len(umem_p->first_sg) = umem_p->first_dma_length; + } + + if (to_state == UMEM_PEER_UNMAPPED) { + peer_mem->dma_unmap(&umem_p->umem.sg_head, + umem_p->peer_client_context, + umem_p->umem.ibdev->dma_device); + peer_mem->put_pages(&umem_p->umem.sg_head, + umem_p->peer_client_context); + } + + memset(&umem->sg_head, 0, sizeof(umem->sg_head)); + atomic64_inc(&ib_peer_client->stats.num_dealloc_mrs); + } + + if ((cur_state == UMEM_PEER_MAPPED && to_state == UMEM_PEER_UNMAPPED) || + (cur_state == UMEM_PEER_INVALIDATED && + to_state == UMEM_PEER_UNMAPPED)) { + atomic64_add(umem->sg_head.nents, + &ib_peer_client->stats.num_dereg_pages); + atomic64_add(umem->length, + &ib_peer_client->stats.num_dereg_bytes); + } + umem_p->mapped_state = to_state; +} + +/* + * True if the client should do unmap itself after the invalidate callback + * returns. Clients operating in this mode need to use this locking pattern: + * + * client_invalidate: + * mutex_lock(&client_lock) + * invalidate_callback(): + * mutex_lock(mapping_lock) + * mutex_unlock(mapping_lock) + * client_dma_unmap() + * client_put_pages() + * mutex_unlock(&client_lock) + * + * ib_umem_stop_invalidation_notifier(): + * mutex_lock(mapping_lock) + * mutex_unlock(mapping_lock) + * peer_mem->dma_unmap(): + * mutex_lock(&client_lock) + * client_dma_unmap() + * mutex_unlock(&client_lock) + * peer_mem->put_pages(): + * mutex_lock(&client_lock) + * client_put_pages() + * mutex_unlock(&client_lock) + * + * ib_peer_umem_release(): + * peer_mem->release(): + * mutex_lock(&client_lock) + * mutex_unlock(&client_lock) + * + * Noting that dma_unmap/put_pages can be called even though invalidate has + * already done the unmap, and release() can be called concurrently with + * invalidate. The client must protect itself against these races. + */ +static bool ib_peer_unmap_on_invalidate(struct ib_umem_peer *umem_p) +{ + const struct peer_memory_client *peer_mem = + umem_p->ib_peer_client->peer_mem; + const struct peer_memory_client_ex *peer_mem_ex; + + if (peer_mem->version[IB_PEER_MEMORY_VER_MAX - 1] == 0) + return false; + peer_mem_ex = container_of(peer_mem, const struct peer_memory_client_ex, + client); + if (peer_mem_ex->ex_size < + offsetofend(struct peer_memory_client_ex, flags)) + return false; + return peer_mem_ex->flags & PEER_MEM_INVALIDATE_UNMAPS; +} + +static int ib_invalidate_peer_memory(void *reg_handle, u64 core_context) +{ + struct ib_peer_memory_client *ib_peer_client = reg_handle; + struct ib_umem_peer *umem_p; + + /* + * The client is not required to fence against invalidation during + * put_pages() as that would deadlock when we call put_pages() here. + * Thus the core_context cannot be a umem pointer as we have no control + * over the lifetime. Since we won't change the kABI for this to add a + * proper kref, an xarray is used. + */ + xa_lock(&ib_peer_client->umem_xa); + ib_peer_client->stats.num_free_callbacks += 1; + umem_p = xa_load(&ib_peer_client->umem_xa, core_context); + if (!umem_p) + goto out_unlock; + kref_get(&umem_p->kref); + xa_unlock(&ib_peer_client->umem_xa); + + mutex_lock(&umem_p->mapping_lock); + /* + * For flows that require invalidation the invalidation_func should not + * be NULL while the device can be doing DMA. The mapping_lock ensures + * that the device is ready to receive an invalidation before one is + * triggered here. + */ + if (umem_p->mapped_state == UMEM_PEER_MAPPED && + umem_p->invalidation_func) + umem_p->invalidation_func(&umem_p->umem, + umem_p->invalidation_private); + if (ib_peer_unmap_on_invalidate(umem_p)) + ib_unmap_peer_client(umem_p, umem_p->mapped_state, + UMEM_PEER_INVALIDATED); + else + ib_unmap_peer_client(umem_p, umem_p->mapped_state, + UMEM_PEER_UNMAPPED); + mutex_unlock(&umem_p->mapping_lock); + kref_put(&umem_p->kref, ib_peer_umem_kref_release); + return 0; + +out_unlock: + xa_unlock(&ib_peer_client->umem_xa); + return 0; +} + +void ib_umem_activate_invalidation_notifier(struct ib_umem *umem, + umem_invalidate_func_t func, + void *priv) +{ + struct ib_umem_peer *umem_p = + container_of(umem, struct ib_umem_peer, umem); + + if (WARN_ON(!umem->is_peer)) + return; + if (umem_p->xa_id == PEER_NO_INVALIDATION_ID) + return; + + umem_p->invalidation_func = func; + umem_p->invalidation_private = priv; + /* Pairs with the lock in ib_peer_umem_get() */ + mutex_unlock(&umem_p->mapping_lock); + + /* At this point func can be called asynchronously */ +} +EXPORT_SYMBOL(ib_umem_activate_invalidation_notifier); + +/* + * Caller has blocked DMA and will no longer be able to handle invalidate + * callbacks. Callers using invalidation must call this function before calling + * ib_peer_umem_release(). ib_umem_activate_invalidation_notifier() is optional + * before doing this. + */ +void ib_umem_stop_invalidation_notifier(struct ib_umem *umem) +{ + struct ib_umem_peer *umem_p = + container_of(umem, struct ib_umem_peer, umem); + bool unmap_on_invalidate = ib_peer_unmap_on_invalidate(umem_p); + enum ib_umem_mapped_state cur_state; + + if (umem_p->invalidation_func) { + mutex_lock(&umem_p->mapping_lock); + umem_p->invalidation_func = NULL; + } else if (umem_p->xa_id != PEER_NO_INVALIDATION_ID) { + mutex_lock(&umem_p->mapping_lock); + } else { + /* + * Haven't called ib_umem_activate_invalidation_notifier() yet, + * still have the lock + */ + } + + if (!unmap_on_invalidate) { + ib_unmap_peer_client(umem_p, umem_p->mapped_state, + UMEM_PEER_UNMAPPED); + } else { + /* Block ib_invalidate_peer_memory() */ + cur_state = umem_p->mapped_state; + umem_p->mapped_state = UMEM_PEER_UNMAPPED; + } + mutex_unlock(&umem_p->mapping_lock); + + if (unmap_on_invalidate) + ib_unmap_peer_client(umem_p, cur_state, UMEM_PEER_UNMAPPED); +} +EXPORT_SYMBOL(ib_umem_stop_invalidation_notifier); + +static void fix_peer_sgls(struct ib_umem_peer *umem_p, unsigned long peer_page_size) +{ + struct ib_umem *umem = &umem_p->umem; + struct scatterlist *sg; + int i; + + for_each_sg(umem_p->umem.sg_head.sgl, sg, umem_p->umem.nmap, i) { + if (i == 0) { + unsigned long offset; + + umem_p->first_sg = sg; + umem_p->first_dma_address = sg->dma_address; + umem_p->first_dma_length = sg_dma_len(sg); + umem_p->first_length = sg->length; + + offset = ALIGN_DOWN(umem->address, PAGE_SIZE) - + ALIGN_DOWN(umem->address, peer_page_size); + sg->dma_address += offset; + sg_dma_len(sg) -= offset; + sg->length -= offset; + } + + if (i == umem_p->umem.nmap - 1) { + unsigned long trim; + + umem_p->last_sg = sg; + umem_p->last_dma_length = sg_dma_len(sg); + umem_p->last_length = sg->length; + + trim = ALIGN(umem->address + umem->length, + peer_page_size) - + ALIGN(umem->address + umem->length, PAGE_SIZE); + sg_dma_len(sg) -= trim; + sg->length -= trim; + } + } +} + +struct ib_umem *ib_peer_umem_get(struct ib_umem *old_umem, int old_ret, + unsigned long peer_mem_flags) +{ + struct ib_peer_memory_client *ib_peer_client; + unsigned long peer_page_size; + void *peer_client_context; + struct ib_umem_peer *umem_p; + int ret; + + ib_peer_client = + ib_get_peer_client(old_umem->address, old_umem->length, + peer_mem_flags, &peer_client_context); + if (!ib_peer_client) + return ERR_PTR(old_ret); + + umem_p = kzalloc(sizeof(*umem_p), GFP_KERNEL); + if (!umem_p) { + ret = -ENOMEM; + goto err_client; + } + + kref_init(&umem_p->kref); + umem_p->umem = *old_umem; + memset(&umem_p->umem.sg_head, 0, sizeof(umem_p->umem.sg_head)); + umem_p->umem.is_peer = 1; + umem_p->ib_peer_client = ib_peer_client; + umem_p->peer_client_context = peer_client_context; + mutex_init(&umem_p->mapping_lock); + umem_p->xa_id = PEER_NO_INVALIDATION_ID; + + mutex_lock(&umem_p->mapping_lock); + if (ib_peer_client->invalidation_required) { + ret = xa_alloc_cyclic(&ib_peer_client->umem_xa, &umem_p->xa_id, + umem_p, + XA_LIMIT(0, PEER_NO_INVALIDATION_ID - 1), + &ib_peer_client->xa_cyclic_next, + GFP_KERNEL); + if (ret < 0) + goto err_umem; + } + + /* + * We always request write permissions to the pages, to force breaking + * of any CoW during the registration of the MR. For read-only MRs we + * use the "force" flag to indicate that CoW breaking is required but + * the registration should not fail if referencing read-only areas. + */ + ret = ib_peer_client->peer_mem->get_pages(umem_p->umem.address, + umem_p->umem.length, 1, + !umem_p->umem.writable, NULL, + peer_client_context, + umem_p->xa_id); + if (ret) + goto err_xa; + + ret = ib_peer_client->peer_mem->dma_map(&umem_p->umem.sg_head, + peer_client_context, + umem_p->umem.ibdev->dma_device, + 0, &umem_p->umem.nmap); + if (ret) + goto err_pages; + + peer_page_size = ib_peer_client->peer_mem->get_page_size(peer_client_context); + if (peer_page_size != PAGE_SIZE) + fix_peer_sgls(umem_p, peer_page_size); + + umem_p->mapped_state = UMEM_PEER_MAPPED; + atomic64_add(umem_p->umem.nmap, &ib_peer_client->stats.num_reg_pages); + atomic64_add(umem_p->umem.length, &ib_peer_client->stats.num_reg_bytes); + atomic64_inc(&ib_peer_client->stats.num_alloc_mrs); + + /* + * If invalidation is allowed then the caller must call + * ib_umem_activate_invalidation_notifier() or ib_peer_umem_release() to + * unlock this mutex. This call should be done after the last read to + * sg_head, once the caller is ready for the invalidation function to be + * called. + */ + if (umem_p->xa_id == PEER_NO_INVALIDATION_ID) + mutex_unlock(&umem_p->mapping_lock); + + /* + * On success the old umem is replaced with the new, larger, allocation + */ + kfree(old_umem); + return &umem_p->umem; + +err_pages: + ib_peer_client->peer_mem->put_pages(&umem_p->umem.sg_head, + umem_p->peer_client_context); +err_xa: + if (umem_p->xa_id != PEER_NO_INVALIDATION_ID) + xa_erase(&umem_p->ib_peer_client->umem_xa, umem_p->xa_id); +err_umem: + mutex_unlock(&umem_p->mapping_lock); + kref_put(&umem_p->kref, ib_peer_umem_kref_release); +err_client: + ib_put_peer_client(ib_peer_client, peer_client_context); + return ERR_PTR(ret); +} + +void ib_peer_umem_release(struct ib_umem *umem) +{ + struct ib_umem_peer *umem_p = + container_of(umem, struct ib_umem_peer, umem); + + /* + * If ib_umem_activate_invalidation_notifier() is called then + * ib_umem_stop_invalidation_notifier() must be called before release. + */ + WARN_ON(umem_p->invalidation_func); + + /* For no invalidation cases, make sure it is unmapped */ + ib_unmap_peer_client(umem_p, umem_p->mapped_state, UMEM_PEER_UNMAPPED); + + if (umem_p->xa_id != PEER_NO_INVALIDATION_ID) + xa_erase(&umem_p->ib_peer_client->umem_xa, umem_p->xa_id); + ib_put_peer_client(umem_p->ib_peer_client, umem_p->peer_client_context); + umem_p->ib_peer_client = NULL; + + /* Must match ib_umem_release() */ + atomic64_sub(ib_umem_num_pages(umem), &umem->owning_mm->pinned_vm); + mmdrop(umem->owning_mm); + + kref_put(&umem_p->kref, ib_peer_umem_kref_release); +} + +/* Use it like this: + * struct peer_memory_client_ex peer_memory_test = { + * .client = { + * .version = "1.0", + * .version[IB_PEER_MEMORY_VER_MAX-1] = 1, + * }, + * .ex_size = sizeof(struct peer_memory_client_ex), + * .flags = PEER_MEM_INVALIDATE_UNMAPS, + * }; + */ diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c index deda9eef5bd62928d8fd31189f2f23ffd44cab70..2ed282cd2940e5e48a7502d2bc3ae0596a454e66 100644 --- a/drivers/infiniband/core/umem.c +++ b/drivers/infiniband/core/umem.c @@ -44,6 +44,10 @@ #include "uverbs.h" +#ifdef CONFIG_INFINIBAND_PEER_MEMORY +#include "ib_peer_mem.h" +#endif + static void __ib_umem_release(struct ib_device *dev, struct ib_umem *umem, int dirty) { bool make_dirty = umem->writable && dirty; @@ -131,15 +135,17 @@ unsigned long ib_umem_find_best_pgsz(struct ib_umem *umem, EXPORT_SYMBOL(ib_umem_find_best_pgsz); /** - * ib_umem_get - Pin and DMA map userspace memory. + * __ib_umem_get - Pin and DMA map userspace memory. * * @device: IB device to connect UMEM * @addr: userspace virtual address to start at * @size: length of region to pin * @access: IB_ACCESS_xxx flags for memory being pinned + * @peer_mem_flags: IB_PEER_MEM_xxx flags for memory being used */ -struct ib_umem *ib_umem_get(struct ib_device *device, unsigned long addr, - size_t size, int access) +static struct ib_umem *__ib_umem_get(struct ib_device *device, + unsigned long addr, size_t size, int access, + unsigned long peer_mem_flags) { struct ib_umem *umem; struct page **page_list; @@ -249,6 +255,26 @@ struct ib_umem *ib_umem_get(struct ib_device *device, unsigned long addr, umem_release: __ib_umem_release(device, umem, 0); +#ifdef CONFIG_INFINIBAND_PEER_MEMORY + /* + * If the address belongs to peer memory client, then the first + * call to get_user_pages will fail. In this case, try to get + * these pages from the peers. + */ + if (ret < 0 && peer_mem_flags & IB_PEER_MEM_ALLOW) { + struct ib_umem *new_umem; + + new_umem = ib_peer_umem_get(umem, ret, peer_mem_flags); + if (IS_ERR(new_umem)) { + ret = PTR_ERR(new_umem); + goto vma; + } + umem = new_umem; + ret = 0; + goto out; + } +vma: +#endif atomic64_sub(ib_umem_num_pages(umem), &mm->pinned_vm); out: free_page((unsigned long) page_list); @@ -259,8 +285,25 @@ struct ib_umem *ib_umem_get(struct ib_device *device, unsigned long addr, } return ret ? ERR_PTR(ret) : umem; } + +struct ib_umem *ib_umem_get(struct ib_device *device, unsigned long addr, + size_t size, int access) +{ + return __ib_umem_get(device, addr, size, access, 0); +} EXPORT_SYMBOL(ib_umem_get); +#ifdef CONFIG_INFINIBAND_PEER_MEMORY +struct ib_umem *ib_umem_get_peer(struct ib_device *device, unsigned long addr, + size_t size, int access, + unsigned long peer_mem_flags) +{ + return __ib_umem_get(device, addr, size, access, + IB_PEER_MEM_ALLOW | peer_mem_flags); +} +EXPORT_SYMBOL(ib_umem_get_peer); +#endif + /** * ib_umem_release - release memory pinned with ib_umem_get * @umem: umem struct to release @@ -272,6 +315,10 @@ void ib_umem_release(struct ib_umem *umem) if (umem->is_odp) return ib_umem_odp_release(to_ib_umem_odp(umem)); +#ifdef CONFIG_INFINIBAND_PEER_MEMORY + if (umem->is_peer) + return ib_peer_umem_release(umem); +#endif __ib_umem_release(umem->ibdev, umem, 1); atomic64_sub(ib_umem_num_pages(umem), &umem->owning_mm->pinned_vm); diff --git a/drivers/infiniband/hw/xsc/ib_umem_ex.c b/drivers/infiniband/hw/xsc/ib_umem_ex.c index 71c823bdc0f6e6e8fa41cefcee3259d4648a0263..c60afbc0270ce8963f2b4018c9e4a9db70458ba5 100644 --- a/drivers/infiniband/hw/xsc/ib_umem_ex.c +++ b/drivers/infiniband/hw/xsc/ib_umem_ex.c @@ -5,11 +5,11 @@ */ #include - -#include "ib_peer_mem.h" - #include #include "ib_umem_ex.h" +#ifndef CONFIG_INFINIBAND_PEER_MEMORY +#include "ib_peer_mem.h" +#endif #if defined(IB_CORE_UMEM_EX_V1) #define get_mm(umem_ctx) ((umem_ctx)->mm) @@ -88,12 +88,16 @@ struct ib_umem_ex *ib_umem_ex(struct ib_umem *umem) if (!umem) return ERR_PTR(-EINVAL); +#ifdef CONFIG_INFINIBAND_PEER_MEMORY + ret_umem = (struct ib_umem_ex *)umem; +#else ret_umem = kzalloc(sizeof(*ret_umem), GFP_KERNEL); if (!ret_umem) return ERR_PTR(-ENOMEM); ret_umem->umem = *umem; kfree(umem); +#endif return ret_umem; } diff --git a/drivers/infiniband/hw/xsc/ib_umem_ex.h b/drivers/infiniband/hw/xsc/ib_umem_ex.h index b73eecf3cd90280c120767a253299e0eb803a89b..034d1c55e5aa647a8f50a4e06b032d074e451143 100644 --- a/drivers/infiniband/hw/xsc/ib_umem_ex.h +++ b/drivers/infiniband/hw/xsc/ib_umem_ex.h @@ -15,14 +15,17 @@ struct invalidation_ctx; // ib umem ex ib_umem add peer memory support struct ib_umem_ex { struct ib_umem umem; +#ifndef CONFIG_INFINIBAND_PEER_MEMORY struct ib_peer_memory_client *ib_peer_mem; struct invalidation_ctx *invalidation_ctx; void *peer_mem_client_context; +#endif }; // expand ib_umem to ib_umem_ex by reallocate struct ib_umem_ex *ib_umem_ex(struct ib_umem *umem); +#ifndef CONFIG_INFINIBAND_PEER_MEMORY typedef void (*umem_invalidate_func_t)(void *invalidation_cookie, struct ib_umem_ex *umem_ex, unsigned long addr, size_t size); @@ -36,6 +39,7 @@ struct invalidation_ctx { int peer_invalidated; struct completion comp; }; +#endif struct ib_umem_ex *ib_client_umem_get(struct ib_ucontext *context, unsigned long addr, size_t size, int access, diff --git a/drivers/infiniband/hw/xsc/mr.c b/drivers/infiniband/hw/xsc/mr.c index 8b716392a964c8fcc5b25784bbb84c2a21a34719..013f470e0f940131b1488e388ccfb97ac1eb675a 100644 --- a/drivers/infiniband/hw/xsc/mr.c +++ b/drivers/infiniband/hw/xsc/mr.c @@ -14,9 +14,11 @@ #include "ib_umem_ex.h" #include "xsc_ib.h" +#ifndef CONFIG_INFINIBAND_PEER_MEMORY static void xsc_invalidate_umem(void *invalidation_cookie, struct ib_umem_ex *umem, unsigned long addr, size_t size); +#endif enum { DEF_CACHE_SIZE = 10, @@ -164,8 +166,16 @@ struct ib_mr *xsc_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, xsc_ib_dbg(dev, "start 0x%llx, virt_addr 0x%llx, length 0x%llx\n", start, virt_addr, length); +#ifdef CONFIG_INFINIBAND_PEER_MEMORY + umem = ib_umem_get_peer(&dev->ib_dev, start, length, + access_flags, IB_PEER_MEM_INVAL_SUPP); +#else umem = ib_umem_get(&dev->ib_dev, start, length, access_flags); +#endif if (IS_ERR(umem)) { +#ifdef CONFIG_INFINIBAND_PEER_MEMORY + return (void *)umem; +#else // check client peer memory u8 peer_exists = 0; @@ -188,13 +198,17 @@ struct ib_mr *xsc_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, if (err) goto error; using_peer_mem = 1; - +#endif } else { umem_ex = ib_umem_ex(umem); if (IS_ERR(umem_ex)) { err = -ENOMEM; goto error; } +#ifdef CONFIG_INFINIBAND_PEER_MEMORY + if (umem->is_peer) + using_peer_mem = 1; +#endif } umem = &umem_ex->umem; @@ -294,6 +308,7 @@ xsc_ib_dereg_mr_def() return 0; } +#ifndef CONFIG_INFINIBAND_PEER_MEMORY static void xsc_invalidate_umem(void *invalidation_cookie, struct ib_umem_ex *umem, unsigned long addr, @@ -320,6 +335,7 @@ static void xsc_invalidate_umem(void *invalidation_cookie, xsc_core_dereg_mr(dev->xdev, &mr->mmr); complete(&mr->invalidation_comp); } +#endif xsc_ib_alloc_mr_def() { diff --git a/include/rdma/ib_umem.h b/include/rdma/ib_umem.h index 70597508c7656b6dbd30626000e4521f0c04e6e1..164fa0bfacbe3432b3fdc7ff6f3cca433b9013eb 100644 --- a/include/rdma/ib_umem.h +++ b/include/rdma/ib_umem.h @@ -22,12 +22,26 @@ struct ib_umem { unsigned long address; u32 writable : 1; u32 is_odp : 1; +#ifdef CONFIG_INFINIBAND_PEER_MEMORY +#ifndef __GENKSYMS__ + /* Placing at the end of the bitfield list is ABI preserving on LE */ + u32 is_peer : 1; +#endif +#endif struct work_struct work; struct sg_table sg_head; int nmap; unsigned int sg_nents; }; +#ifdef CONFIG_INFINIBAND_PEER_MEMORY +typedef void (*umem_invalidate_func_t)(struct ib_umem *umem, void *priv); +enum ib_peer_mem_flags { + IB_PEER_MEM_ALLOW = 1 << 0, + IB_PEER_MEM_INVAL_SUPP = 1 << 1, +}; +#endif + /* Returns the offset of the umem start relative to the first page. */ static inline int ib_umem_offset(struct ib_umem *umem) { @@ -79,6 +93,15 @@ int ib_umem_copy_from(void *dst, struct ib_umem *umem, size_t offset, unsigned long ib_umem_find_best_pgsz(struct ib_umem *umem, unsigned long pgsz_bitmap, unsigned long virt); +#ifdef CONFIG_INFINIBAND_PEER_MEMORY +struct ib_umem *ib_umem_get_peer(struct ib_device *device, unsigned long addr, + size_t size, int access, + unsigned long peer_mem_flags); +void ib_umem_activate_invalidation_notifier(struct ib_umem *umem, + umem_invalidate_func_t func, + void *cookie); +void ib_umem_stop_invalidation_notifier(struct ib_umem *umem); +#endif /* CONFIG_INFINIBAND_PEER_MEMORY */ #else /* CONFIG_INFINIBAND_USER_MEM */ @@ -102,6 +125,26 @@ static inline unsigned long ib_umem_find_best_pgsz(struct ib_umem *umem, return 0; } +#ifdef CONFIG_INFINIBAND_PEER_MEMORY +static inline struct ib_umem *ib_umem_get_peer(struct ib_device *device, + unsigned long addr, size_t size, + int access, + unsigned long peer_mem_flags) +{ + return ERR_PTR(-EINVAL); +} + +static inline void ib_umem_activate_invalidation_notifier(struct ib_umem *umem, + umem_invalidate_func_t func, + void *cookie) +{ +} + +static inline void ib_umem_stop_invalidation_notifier(struct ib_umem *umem) +{ +} +#endif /* CONFIG_INFINIBAND_PEER_MEMORY */ + #endif /* CONFIG_INFINIBAND_USER_MEM */ #endif /* IB_UMEM_H */ diff --git a/include/rdma/peer_mem.h b/include/rdma/peer_mem.h new file mode 100644 index 0000000000000000000000000000000000000000..e3dc2ca630d65727d2975ef5679ac8d6b8a247e0 --- /dev/null +++ b/include/rdma/peer_mem.h @@ -0,0 +1,175 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ +/* + * Copyright (c) 2014-2020, Mellanox Technologies. All rights reserved. + */ +#ifndef RDMA_PEER_MEM_H +#define RDMA_PEER_MEM_H + +#include + +#define IB_PEER_MEMORY_NAME_MAX 64 +#define IB_PEER_MEMORY_VER_MAX 16 + +/* + * Prior versions used a void * for core_context, at some point this was + * switched to use u64. Be careful if compiling this as 32 bit. To help the + * value of core_context is limited to u32 so it should work OK despite the + * type change. + */ +#define PEER_MEM_U64_CORE_CONTEXT + +struct device; + +/** + * struct peer_memory_client - registration information for user virtual + * memory handlers + * + * The peer_memory_client scheme allows a driver to register with the ib_umem + * system that it has the ability to understand user virtual address ranges + * that are not compatible with get_user_pages(). For instance VMAs created + * with io_remap_pfn_range(), or other driver special VMA. + * + * For ranges the interface understands it can provide a DMA mapped sg_table + * for use by the ib_umem, allowing user virtual ranges that cannot be + * supported by get_user_pages() to be used as umems. + */ +struct peer_memory_client { + char name[IB_PEER_MEMORY_NAME_MAX]; + char version[IB_PEER_MEMORY_VER_MAX]; + + /** + * acquire - Begin working with a user space virtual address range + * + * @addr - Virtual address to be checked whether belongs to peer. + * @size - Length of the virtual memory area starting at addr. + * @peer_mem_private_data - Obsolete, always NULL + * @peer_mem_name - Obsolete, always NULL + * @client_context - Returns an opaque value for this acquire use in + * other APIs + * + * Returns 1 if the peer_memory_client supports the entire virtual + * address range, 0 or -ERRNO otherwise. If 1 is returned then + * release() will be called to release the acquire(). + */ + int (*acquire)(unsigned long addr, size_t size, + void *peer_mem_private_data, char *peer_mem_name, + void **client_context); + /** + * get_pages - Fill in the first part of a sg_table for a virtual + * address range + * + * @addr - Virtual address to be checked whether belongs to peer. + * @size - Length of the virtual memory area starting at addr. + * @write - Always 1 + * @force - 1 if write is required + * @sg_head - Obsolete, always NULL + * @client_context - Value returned by acquire() + * @core_context - Value to be passed to invalidate_peer_memory for + * this get + * + * addr/size are passed as the raw virtual address range requested by + * the user, it is not aligned to any page size. get_pages() is always + * followed by dma_map(). + * + * Upon return the caller can call the invalidate_callback(). + * + * Returns 0 on success, -ERRNO on failure. After success put_pages() + * will be called to return the pages. + */ + int (*get_pages)(unsigned long addr, size_t size, int write, int force, + struct sg_table *sg_head, void *client_context, + u64 core_context); + /** + * dma_map - Create a DMA mapped sg_table + * + * @sg_head - The sg_table to allocate + * @client_context - Value returned by acquire() + * @dma_device - The device that will be doing DMA from these addresses + * @dmasync - Obsolete, always 0 + * @nmap - Returns the number of dma mapped entries in the sg_head + * + * Must be called after get_pages(). This must fill in the sg_head with + * DMA mapped SGLs for dma_device. Each SGL start and end must meet a + * minimum alignment of at least PAGE_SIZE, though individual sgls can + * be multiples of PAGE_SIZE, in any mixture. Since the user virtual + * address/size are not page aligned, the implementation must increase + * it to the logical alignment when building the SGLs. + * + * Returns 0 on success, -ERRNO on failure. After success dma_unmap() + * will be called to unmap the pages. On failure sg_head must be left + * untouched or point to a valid sg_table. + */ + int (*dma_map)(struct sg_table *sg_head, void *client_context, + struct device *dma_device, int dmasync, int *nmap); + /** + * dma_unmap - Unmap a DMA mapped sg_table + * + * @sg_head - The sg_table to unmap + * @client_context - Value returned by acquire() + * @dma_device - The device that will be doing DMA from these addresses + * + * sg_head will not be touched after this function returns. + * + * Must return 0. + */ + int (*dma_unmap)(struct sg_table *sg_head, void *client_context, + struct device *dma_device); + /** + * put_pages - Unpin a SGL + * + * @sg_head - The sg_table to unpin + * @client_context - Value returned by acquire() + * + * sg_head must be freed on return. + */ + void (*put_pages)(struct sg_table *sg_head, void *client_context); + /* Client should always return PAGE_SIZE */ + unsigned long (*get_page_size)(void *client_context); + /** + * release - Undo acquire + * + * @client_context - Value returned by acquire() + * + * If acquire() returns 1 then release() must be called. All + * get_pages() and dma_map()'s must be undone before calling this + * function. + */ + void (*release)(void *client_context); +}; + +enum { + PEER_MEM_INVALIDATE_UNMAPS = 1 << 0, +}; + +struct peer_memory_client_ex { + struct peer_memory_client client; + size_t ex_size; + u32 flags; +}; + +/* + * If invalidate_callback() is non-NULL then the client will only support + * umems which can be invalidated. The caller may call the + * invalidate_callback() after acquire() on return the range will no longer + * have DMA active, and release() will have been called. + * + * Note: The implementation locking must ensure that get_pages(), and + * dma_map() do not have locking dependencies with invalidate_callback(). The + * ib_core will wait until any concurrent get_pages() or dma_map() completes + * before returning. + * + * Similarly, this can call dma_unmap(), put_pages() and release() from within + * the callback, or will wait for another thread doing those operations to + * complete. + * + * For these reasons the user of invalidate_callback() must be careful with + * locking. + */ +typedef int (*invalidate_peer_memory)(void *reg_handle, u64 core_context); + +void * +ib_register_peer_memory_client(const struct peer_memory_client *peer_client, + invalidate_peer_memory *invalidate_callback); +void ib_unregister_peer_memory_client(void *reg_handle); + +#endif