X-Git-Url: http://git.droids-corp.org/?a=blobdiff_plain;f=lib%2Flibrte_eal%2Fcommon%2Finclude%2Frte_memory.h;h=3d8d0bd6976d64ee26fd8462c55ba37ac646ec93;hb=bdc993fa3dc38d6ae2d4d1480604556bc20a5522;hp=01e7548dc4eb9c5908a599d87eba94edd87f4416;hpb=2e378ff29740013d3b0e2bbae1f31c9509d0dfd2;p=dpdk.git diff --git a/lib/librte_eal/common/include/rte_memory.h b/lib/librte_eal/common/include/rte_memory.h index 01e7548dc4..3d8d0bd697 100644 --- a/lib/librte_eal/common/include/rte_memory.h +++ b/lib/librte_eal/common/include/rte_memory.h @@ -22,9 +22,7 @@ extern "C" { #include #include #include - -/* forward declaration for pointers */ -struct rte_memseg_list; +#include __extension__ enum rte_page_sizes { @@ -41,48 +39,12 @@ enum rte_page_sizes { }; #define SOCKET_ID_ANY -1 /**< Any NUMA socket. */ -#define RTE_CACHE_LINE_MASK (RTE_CACHE_LINE_SIZE-1) /**< Cache line mask. */ - -#define RTE_CACHE_LINE_ROUNDUP(size) \ - (RTE_CACHE_LINE_SIZE * ((size + RTE_CACHE_LINE_SIZE - 1) / RTE_CACHE_LINE_SIZE)) -/**< Return the first cache-aligned value greater or equal to size. */ - -/**< Cache line size in terms of log2 */ -#if RTE_CACHE_LINE_SIZE == 64 -#define RTE_CACHE_LINE_SIZE_LOG2 6 -#elif RTE_CACHE_LINE_SIZE == 128 -#define RTE_CACHE_LINE_SIZE_LOG2 7 -#else -#error "Unsupported cache line size" -#endif - -#define RTE_CACHE_LINE_MIN_SIZE 64 /**< Minimum Cache line size. */ - -/** - * Force alignment to cache line. - */ -#define __rte_cache_aligned __rte_aligned(RTE_CACHE_LINE_SIZE) - -/** - * Force minimum cache line alignment. - */ -#define __rte_cache_min_aligned __rte_aligned(RTE_CACHE_LINE_MIN_SIZE) - -typedef uint64_t phys_addr_t; /**< Physical address. */ -#define RTE_BAD_PHYS_ADDR ((phys_addr_t)-1) -/** - * IO virtual address type. - * When the physical addressing mode (IOVA as PA) is in use, - * the translation from an IO virtual address (IOVA) to a physical address - * is a direct mapping, i.e. the same value. - * Otherwise, in virtual mode (IOVA as VA), an IOMMU may do the translation. - */ -typedef uint64_t rte_iova_t; -#define RTE_BAD_IOVA ((rte_iova_t)-1) /** * Physical memory segment descriptor. */ +#define RTE_MEMSEG_FLAG_DO_NOT_FREE (1 << 0) +/**< Prevent this segment from being freed back to the OS. */ struct rte_memseg { RTE_STD_C11 union { @@ -99,8 +61,30 @@ struct rte_memseg { int32_t socket_id; /**< NUMA socket ID. */ uint32_t nchannel; /**< Number of channels. */ uint32_t nrank; /**< Number of ranks. */ + uint32_t flags; /**< Memseg-specific flags */ } __rte_packed; +/** + * memseg list is a special case as we need to store a bunch of other data + * together with the array itself. + */ +struct rte_memseg_list { + RTE_STD_C11 + union { + void *base_va; + /**< Base virtual address for this memseg list. */ + uint64_t addr_64; + /**< Makes sure addr is always 64-bits */ + }; + uint64_t page_sz; /**< Page size for all memsegs in this list. */ + int socket_id; /**< Socket ID for all memsegs in this list. */ + volatile uint32_t version; /**< version number for multiprocess sync. */ + size_t len; /**< Length of memory area covered by this memseg list. */ + unsigned int external; /**< 1 if this list points to external memory */ + unsigned int heap; /**< 1 if this list points to a heap */ + struct rte_fbarray memseg_arr; +}; + /** * Lock page in physical memory and prevent from swapping. * @@ -126,6 +110,11 @@ phys_addr_t rte_mem_virt2phy(const void *virt); /** * Get IO virtual address of any mapped virtual address in the current process. * + * @note This function will not check internal page table. Instead, in IOVA as + * PA mode, it will fall back to getting real physical address (which may + * not match the expected IOVA, such as what was specified for external + * memory). + * * @param virt * The virtual address. * @return @@ -145,7 +134,8 @@ rte_iova_t rte_mem_virt2iova(const void *virt); * Virtual address corresponding to iova address (or NULL if address does not * exist within DPDK memory map). */ -__rte_experimental void * +__rte_experimental +void * rte_mem_iova2virt(rte_iova_t iova); /** @@ -159,7 +149,8 @@ rte_mem_iova2virt(rte_iova_t iova); * @return * Memseg pointer on success, or NULL on error. */ -__rte_experimental struct rte_memseg * +__rte_experimental +struct rte_memseg * rte_mem_virt2memseg(const void *virt, const struct rte_memseg_list *msl); /** @@ -170,7 +161,8 @@ rte_mem_virt2memseg(const void *virt, const struct rte_memseg_list *msl); * @return * Memseg list to which this virtual address belongs to. */ -__rte_experimental struct rte_memseg_list * +__rte_experimental +struct rte_memseg_list * rte_mem_virt2memseg_list(const void *virt); /** @@ -185,7 +177,7 @@ typedef int (*rte_memseg_walk_t)(const struct rte_memseg_list *msl, /** * Memseg contig walk function prototype. This will trigger a callback on every - * VA-contiguous are starting at memseg ``ms``, so total valid VA space at each + * VA-contiguous area starting at memseg ``ms``, so total valid VA space at each * callback call will be [``ms->addr``, ``ms->addr + len``). * * Returning 0 will continue walk @@ -212,6 +204,9 @@ typedef int (*rte_memseg_list_walk_t)(const struct rte_memseg_list *msl, * @note This function read-locks the memory hotplug subsystem, and thus cannot * be used within memory-related callback functions. * + * @note This function will also walk through externally allocated segments. It + * is up to the user to decide whether to skip through these segments. + * * @param func * Iterator function * @param arg @@ -221,7 +216,8 @@ typedef int (*rte_memseg_list_walk_t)(const struct rte_memseg_list *msl, * 1 if stopped by the user * -1 if user function reported error */ -int __rte_experimental +__rte_experimental +int rte_memseg_walk(rte_memseg_walk_t func, void *arg); /** @@ -230,6 +226,9 @@ rte_memseg_walk(rte_memseg_walk_t func, void *arg); * @note This function read-locks the memory hotplug subsystem, and thus cannot * be used within memory-related callback functions. * + * @note This function will also walk through externally allocated segments. It + * is up to the user to decide whether to skip through these segments. + * * @param func * Iterator function * @param arg @@ -239,7 +238,8 @@ rte_memseg_walk(rte_memseg_walk_t func, void *arg); * 1 if stopped by the user * -1 if user function reported error */ -int __rte_experimental +__rte_experimental +int rte_memseg_contig_walk(rte_memseg_contig_walk_t func, void *arg); /** @@ -248,6 +248,9 @@ rte_memseg_contig_walk(rte_memseg_contig_walk_t func, void *arg); * @note This function read-locks the memory hotplug subsystem, and thus cannot * be used within memory-related callback functions. * + * @note This function will also walk through externally allocated segments. It + * is up to the user to decide whether to skip through these segments. + * * @param func * Iterator function * @param arg @@ -257,9 +260,294 @@ rte_memseg_contig_walk(rte_memseg_contig_walk_t func, void *arg); * 1 if stopped by the user * -1 if user function reported error */ -int __rte_experimental +__rte_experimental +int rte_memseg_list_walk(rte_memseg_list_walk_t func, void *arg); +/** + * Walk list of all memsegs without performing any locking. + * + * @note This function does not perform any locking, and is only safe to call + * from within memory-related callback functions. + * + * @param func + * Iterator function + * @param arg + * Argument passed to iterator + * @return + * 0 if walked over the entire list + * 1 if stopped by the user + * -1 if user function reported error + */ +__rte_experimental +int +rte_memseg_walk_thread_unsafe(rte_memseg_walk_t func, void *arg); + +/** + * Walk each VA-contiguous area without performing any locking. + * + * @note This function does not perform any locking, and is only safe to call + * from within memory-related callback functions. + * + * @param func + * Iterator function + * @param arg + * Argument passed to iterator + * @return + * 0 if walked over the entire list + * 1 if stopped by the user + * -1 if user function reported error + */ +__rte_experimental +int +rte_memseg_contig_walk_thread_unsafe(rte_memseg_contig_walk_t func, void *arg); + +/** + * Walk each allocated memseg list without performing any locking. + * + * @note This function does not perform any locking, and is only safe to call + * from within memory-related callback functions. + * + * @param func + * Iterator function + * @param arg + * Argument passed to iterator + * @return + * 0 if walked over the entire list + * 1 if stopped by the user + * -1 if user function reported error + */ +__rte_experimental +int +rte_memseg_list_walk_thread_unsafe(rte_memseg_list_walk_t func, void *arg); + +/** + * Return file descriptor associated with a particular memseg (if available). + * + * @note This function read-locks the memory hotplug subsystem, and thus cannot + * be used within memory-related callback functions. + * + * @note This returns an internal file descriptor. Performing any operations on + * this file descriptor is inherently dangerous, so it should be treated + * as read-only for all intents and purposes. + * + * @param ms + * A pointer to memseg for which to get file descriptor. + * + * @return + * Valid file descriptor in case of success. + * -1 in case of error, with ``rte_errno`` set to the following values: + * - EINVAL - ``ms`` pointer was NULL or did not point to a valid memseg + * - ENODEV - ``ms`` fd is not available + * - ENOENT - ``ms`` is an unused segment + * - ENOTSUP - segment fd's are not supported + */ +__rte_experimental +int +rte_memseg_get_fd(const struct rte_memseg *ms); + +/** + * Return file descriptor associated with a particular memseg (if available). + * + * @note This function does not perform any locking, and is only safe to call + * from within memory-related callback functions. + * + * @note This returns an internal file descriptor. Performing any operations on + * this file descriptor is inherently dangerous, so it should be treated + * as read-only for all intents and purposes. + * + * @param ms + * A pointer to memseg for which to get file descriptor. + * + * @return + * Valid file descriptor in case of success. + * -1 in case of error, with ``rte_errno`` set to the following values: + * - EINVAL - ``ms`` pointer was NULL or did not point to a valid memseg + * - ENODEV - ``ms`` fd is not available + * - ENOENT - ``ms`` is an unused segment + * - ENOTSUP - segment fd's are not supported + */ +__rte_experimental +int +rte_memseg_get_fd_thread_unsafe(const struct rte_memseg *ms); + +/** + * Get offset into segment file descriptor associated with a particular memseg + * (if available). + * + * @note This function read-locks the memory hotplug subsystem, and thus cannot + * be used within memory-related callback functions. + * + * @param ms + * A pointer to memseg for which to get file descriptor. + * @param offset + * A pointer to offset value where the result will be stored. + * + * @return + * Valid file descriptor in case of success. + * -1 in case of error, with ``rte_errno`` set to the following values: + * - EINVAL - ``ms`` pointer was NULL or did not point to a valid memseg + * - EINVAL - ``offset`` pointer was NULL + * - ENODEV - ``ms`` fd is not available + * - ENOENT - ``ms`` is an unused segment + * - ENOTSUP - segment fd's are not supported + */ +__rte_experimental +int +rte_memseg_get_fd_offset(const struct rte_memseg *ms, size_t *offset); + +/** + * Get offset into segment file descriptor associated with a particular memseg + * (if available). + * + * @note This function does not perform any locking, and is only safe to call + * from within memory-related callback functions. + * + * @param ms + * A pointer to memseg for which to get file descriptor. + * @param offset + * A pointer to offset value where the result will be stored. + * + * @return + * Valid file descriptor in case of success. + * -1 in case of error, with ``rte_errno`` set to the following values: + * - EINVAL - ``ms`` pointer was NULL or did not point to a valid memseg + * - EINVAL - ``offset`` pointer was NULL + * - ENODEV - ``ms`` fd is not available + * - ENOENT - ``ms`` is an unused segment + * - ENOTSUP - segment fd's are not supported + */ +__rte_experimental +int +rte_memseg_get_fd_offset_thread_unsafe(const struct rte_memseg *ms, + size_t *offset); + +/** + * @warning + * @b EXPERIMENTAL: this API may change without prior notice + * + * Register external memory chunk with DPDK. + * + * @note Using this API is mutually exclusive with ``rte_malloc`` family of + * API's. + * + * @note This API will not perform any DMA mapping. It is expected that user + * will do that themselves. + * + * @note Before accessing this memory in other processes, it needs to be + * attached in each of those processes by calling ``rte_extmem_attach`` in + * each other process. + * + * @param va_addr + * Start of virtual area to register. Must be aligned by ``page_sz``. + * @param len + * Length of virtual area to register. Must be aligned by ``page_sz``. + * @param iova_addrs + * Array of page IOVA addresses corresponding to each page in this memory + * area. Can be NULL, in which case page IOVA addresses will be set to + * RTE_BAD_IOVA. + * @param n_pages + * Number of elements in the iova_addrs array. Ignored if ``iova_addrs`` + * is NULL. + * @param page_sz + * Page size of the underlying memory + * + * @return + * - 0 on success + * - -1 in case of error, with rte_errno set to one of the following: + * EINVAL - one of the parameters was invalid + * EEXIST - memory chunk is already registered + * ENOSPC - no more space in internal config to store a new memory chunk + */ +__rte_experimental +int +rte_extmem_register(void *va_addr, size_t len, rte_iova_t iova_addrs[], + unsigned int n_pages, size_t page_sz); + +/** + * @warning + * @b EXPERIMENTAL: this API may change without prior notice + * + * Unregister external memory chunk with DPDK. + * + * @note Using this API is mutually exclusive with ``rte_malloc`` family of + * API's. + * + * @note This API will not perform any DMA unmapping. It is expected that user + * will do that themselves. + * + * @note Before calling this function, all other processes must call + * ``rte_extmem_detach`` to detach from the memory area. + * + * @param va_addr + * Start of virtual area to unregister + * @param len + * Length of virtual area to unregister + * + * @return + * - 0 on success + * - -1 in case of error, with rte_errno set to one of the following: + * EINVAL - one of the parameters was invalid + * ENOENT - memory chunk was not found + */ +__rte_experimental +int +rte_extmem_unregister(void *va_addr, size_t len); + +/** + * @warning + * @b EXPERIMENTAL: this API may change without prior notice + * + * Attach to external memory chunk registered in another process. + * + * @note Using this API is mutually exclusive with ``rte_malloc`` family of + * API's. + * + * @note This API will not perform any DMA mapping. It is expected that user + * will do that themselves. + * + * @param va_addr + * Start of virtual area to register + * @param len + * Length of virtual area to register + * + * @return + * - 0 on success + * - -1 in case of error, with rte_errno set to one of the following: + * EINVAL - one of the parameters was invalid + * ENOENT - memory chunk was not found + */ +__rte_experimental +int +rte_extmem_attach(void *va_addr, size_t len); + +/** + * @warning + * @b EXPERIMENTAL: this API may change without prior notice + * + * Detach from external memory chunk registered in another process. + * + * @note Using this API is mutually exclusive with ``rte_malloc`` family of + * API's. + * + * @note This API will not perform any DMA unmapping. It is expected that user + * will do that themselves. + * + * @param va_addr + * Start of virtual area to unregister + * @param len + * Length of virtual area to unregister + * + * @return + * - 0 on success + * - -1 in case of error, with rte_errno set to one of the following: + * EINVAL - one of the parameters was invalid + * ENOENT - memory chunk was not found + */ +__rte_experimental +int +rte_extmem_detach(void *va_addr, size_t len); + /** * Dump the physical memory layout to a file. * @@ -300,6 +588,49 @@ unsigned rte_memory_get_nchannel(void); */ unsigned rte_memory_get_nrank(void); +/** + * @warning + * @b EXPERIMENTAL: this API may change without prior notice + * + * Check if all currently allocated memory segments are compliant with + * supplied DMA address width. + * + * @param maskbits + * Address width to check against. + */ +__rte_experimental +int rte_mem_check_dma_mask(uint8_t maskbits); + +/** + * @warning + * @b EXPERIMENTAL: this API may change without prior notice + * + * Check if all currently allocated memory segments are compliant with + * supplied DMA address width. This function will use + * rte_memseg_walk_thread_unsafe instead of rte_memseg_walk implying + * memory_hotplug_lock will not be acquired avoiding deadlock during + * memory initialization. + * + * This function is just for EAL core memory internal use. Drivers should + * use the previous rte_mem_check_dma_mask. + * + * @param maskbits + * Address width to check against. + */ +__rte_experimental +int rte_mem_check_dma_mask_thread_unsafe(uint8_t maskbits); + +/** + * @warning + * @b EXPERIMENTAL: this API may change without prior notice + * + * Set dma mask to use once memory initialization is done. Previous functions + * rte_mem_check_dma_mask and rte_mem_check_dma_mask_thread_unsafe can not be + * used safely until memory has been initialized. + */ +__rte_experimental +void rte_mem_set_dma_mask(uint8_t maskbits); + /** * Drivers based on uio will not load unless physical * addresses are obtainable. It is only possible to get @@ -327,7 +658,7 @@ enum rte_mem_event { * Function typedef used to register callbacks for memory events. */ typedef void (*rte_mem_event_callback_t)(enum rte_mem_event event_type, - const void *addr, size_t len); + const void *addr, size_t len, void *arg); /** * Function used to register callbacks for memory events. @@ -336,19 +667,28 @@ typedef void (*rte_mem_event_callback_t)(enum rte_mem_event event_type, * therefore some functions (e.g. `rte_memseg_walk()`) will cause a * deadlock when called from within such callbacks. * + * @note mem event callbacks not being supported is an expected error condition, + * so user code needs to handle this situation. In these cases, return + * value will be -1, and rte_errno will be set to ENOTSUP. + * * @param name * Name associated with specified callback to be added to the list. * * @param clb * Callback function pointer. * + * @param arg + * Argument to pass to the callback. + * * @return * 0 on successful callback register * -1 on unsuccessful callback register, with rte_errno value indicating * reason for failure. */ -int __rte_experimental -rte_mem_event_callback_register(const char *name, rte_mem_event_callback_t clb); +__rte_experimental +int +rte_mem_event_callback_register(const char *name, rte_mem_event_callback_t clb, + void *arg); /** * Function used to unregister callbacks for memory events. @@ -356,13 +696,17 @@ rte_mem_event_callback_register(const char *name, rte_mem_event_callback_t clb); * @param name * Name associated with specified callback to be removed from the list. * + * @param arg + * Argument to look for among callbacks with specified callback name. + * * @return * 0 on successful callback unregister * -1 on unsuccessful callback unregister, with rte_errno value indicating * reason for failure. */ -int __rte_experimental -rte_mem_event_callback_unregister(const char *name); +__rte_experimental +int +rte_mem_event_callback_unregister(const char *name, void *arg); #define RTE_MEM_ALLOC_VALIDATOR_NAME_LEN 64 @@ -389,6 +733,10 @@ typedef int (*rte_mem_alloc_validator_t)(int socket_id, * therefore some functions (e.g. `rte_memseg_walk()`) will cause a * deadlock when called from within such callbacks. * + * @note validator callbacks not being supported is an expected error condition, + * so user code needs to handle this situation. In these cases, return + * value will be -1, and rte_errno will be set to ENOTSUP. + * * @param name * Name associated with specified callback to be added to the list. * @@ -406,7 +754,8 @@ typedef int (*rte_mem_alloc_validator_t)(int socket_id, * -1 on unsuccessful callback register, with rte_errno value indicating * reason for failure. */ -int __rte_experimental +__rte_experimental +int rte_mem_alloc_validator_register(const char *name, rte_mem_alloc_validator_t clb, int socket_id, size_t limit); @@ -424,7 +773,8 @@ rte_mem_alloc_validator_register(const char *name, * -1 on unsuccessful callback unregister, with rte_errno value indicating * reason for failure. */ -int __rte_experimental +__rte_experimental +int rte_mem_alloc_validator_unregister(const char *name, int socket_id); #ifdef __cplusplus