* Copyright(c) 2010-2014 Intel Corporation
*/
+#include <fcntl.h>
#include <errno.h>
#include <stdio.h>
#include <stdint.h>
#include <string.h>
#include <unistd.h>
#include <inttypes.h>
-#include <sys/mman.h>
#include <sys/queue.h>
#include <rte_fbarray.h>
#include <rte_memory.h>
#include <rte_eal.h>
#include <rte_eal_memconfig.h>
+#include <rte_eal_paging.h>
#include <rte_errno.h>
#include <rte_log.h>
#include "eal_memalloc.h"
#include "eal_private.h"
#include "eal_internal_cfg.h"
+#include "eal_memcfg.h"
+#include "eal_options.h"
+#include "malloc_heap.h"
/*
* Try to mmap *size bytes in /dev/zero. If it is successful, return the
#define MEMSEG_LIST_FMT "memseg-%" PRIu64 "k-%i-%i"
-static uint64_t baseaddr_offset;
+static void *next_baseaddr;
static uint64_t system_page_sz;
+#define MAX_MMAP_WITH_DEFINED_ADDR_TRIES 5
void *
eal_get_virtual_area(void *requested_addr, size_t *size,
- size_t page_sz, int flags, int mmap_flags)
+ size_t page_sz, int flags, int reserve_flags)
{
bool addr_is_hint, allow_shrink, unmap, no_align;
uint64_t map_sz;
void *mapped_addr, *aligned_addr;
+ uint8_t try = 0;
+ struct internal_config *internal_conf =
+ eal_get_internal_configuration();
if (system_page_sz == 0)
- system_page_sz = sysconf(_SC_PAGESIZE);
-
- mmap_flags |= MAP_PRIVATE | MAP_ANONYMOUS;
+ system_page_sz = rte_mem_page_size();
RTE_LOG(DEBUG, EAL, "Ask a virtual area of 0x%zx bytes\n", *size);
allow_shrink = (flags & EAL_VIRTUAL_AREA_ALLOW_SHRINK) > 0;
unmap = (flags & EAL_VIRTUAL_AREA_UNMAP) > 0;
- if (requested_addr == NULL && internal_config.base_virtaddr != 0) {
- requested_addr = (void *) (internal_config.base_virtaddr +
- (size_t)baseaddr_offset);
+ if (next_baseaddr == NULL && internal_conf->base_virtaddr != 0 &&
+ rte_eal_process_type() == RTE_PROC_PRIMARY)
+ next_baseaddr = (void *) internal_conf->base_virtaddr;
+
+#ifdef RTE_ARCH_64
+ if (next_baseaddr == NULL && internal_conf->base_virtaddr == 0 &&
+ rte_eal_process_type() == RTE_PROC_PRIMARY)
+ next_baseaddr = (void *) eal_get_baseaddr();
+#endif
+ if (requested_addr == NULL && next_baseaddr != NULL) {
+ requested_addr = next_baseaddr;
requested_addr = RTE_PTR_ALIGN(requested_addr, page_sz);
addr_is_hint = true;
}
- /* if requested address is not aligned by page size, or if requested
- * address is NULL, add page size to requested length as we may get an
- * address that's aligned by system page size, which can be smaller than
- * our requested page size. additionally, we shouldn't try to align if
- * system page size is the same as requested page size.
+ /* we don't need alignment of resulting pointer in the following cases:
+ *
+ * 1. page size is equal to system size
+ * 2. we have a requested address, and it is page-aligned, and we will
+ * be discarding the address if we get a different one.
+ *
+ * for all other cases, alignment is potentially necessary.
*/
no_align = (requested_addr != NULL &&
- ((uintptr_t)requested_addr & (page_sz - 1))) ||
+ requested_addr == RTE_PTR_ALIGN(requested_addr, page_sz) &&
+ !addr_is_hint) ||
page_sz == system_page_sz;
do {
return NULL;
}
- mapped_addr = mmap(requested_addr, (size_t)map_sz, PROT_READ,
- mmap_flags, -1, 0);
- if (mapped_addr == MAP_FAILED && allow_shrink)
+ mapped_addr = eal_mem_reserve(
+ requested_addr, (size_t)map_sz, reserve_flags);
+ if ((mapped_addr == NULL) && allow_shrink)
*size -= page_sz;
- } while (allow_shrink && mapped_addr == MAP_FAILED && *size > 0);
+
+ if ((mapped_addr != NULL) && addr_is_hint &&
+ (mapped_addr != requested_addr)) {
+ try++;
+ next_baseaddr = RTE_PTR_ADD(next_baseaddr, page_sz);
+ if (try <= MAX_MMAP_WITH_DEFINED_ADDR_TRIES) {
+ /* hint was not used. Try with another offset */
+ eal_mem_free(mapped_addr, map_sz);
+ mapped_addr = NULL;
+ requested_addr = next_baseaddr;
+ }
+ }
+ } while ((allow_shrink || addr_is_hint) &&
+ (mapped_addr == NULL) && (*size > 0));
/* align resulting address - if map failed, we will ignore the value
* anyway, so no need to add additional checks.
if (*size == 0) {
RTE_LOG(ERR, EAL, "Cannot get a virtual area of any size: %s\n",
- strerror(errno));
- rte_errno = errno;
+ rte_strerror(rte_errno));
return NULL;
- } else if (mapped_addr == MAP_FAILED) {
+ } else if (mapped_addr == NULL) {
RTE_LOG(ERR, EAL, "Cannot get a virtual area: %s\n",
- strerror(errno));
- /* pass errno up the call chain */
- rte_errno = errno;
+ rte_strerror(rte_errno));
return NULL;
} else if (requested_addr != NULL && !addr_is_hint &&
aligned_addr != requested_addr) {
RTE_LOG(ERR, EAL, "Cannot get a virtual area at requested address: %p (got %p)\n",
requested_addr, aligned_addr);
- munmap(mapped_addr, map_sz);
+ eal_mem_free(mapped_addr, map_sz);
rte_errno = EADDRNOTAVAIL;
return NULL;
} else if (requested_addr != NULL && addr_is_hint &&
RTE_LOG(WARNING, EAL, "WARNING! Base virtual address hint (%p != %p) not respected!\n",
requested_addr, aligned_addr);
RTE_LOG(WARNING, EAL, " This may cause issues with mapping memory into secondary processes\n");
+ } else if (next_baseaddr != NULL) {
+ next_baseaddr = RTE_PTR_ADD(aligned_addr, *size);
}
RTE_LOG(DEBUG, EAL, "Virtual area found at %p (size = 0x%zx)\n",
aligned_addr, *size);
if (unmap) {
- munmap(mapped_addr, map_sz);
+ eal_mem_free(mapped_addr, map_sz);
} else if (!no_align) {
void *map_end, *aligned_end;
size_t before_len, after_len;
/* unmap space before aligned mmap address */
before_len = RTE_PTR_DIFF(aligned_addr, mapped_addr);
if (before_len > 0)
- munmap(mapped_addr, before_len);
+ eal_mem_free(mapped_addr, before_len);
/* unmap space after aligned end mmap address */
after_len = RTE_PTR_DIFF(map_end, aligned_end);
if (after_len > 0)
- munmap(aligned_end, after_len);
+ eal_mem_free(aligned_end, after_len);
}
- baseaddr_offset += *size;
+ if (!unmap) {
+ /* Exclude these pages from a core dump. */
+ eal_mem_set_dump(aligned_addr, *size, false);
+ }
return aligned_addr;
}
-static uint64_t
-get_mem_amount(uint64_t page_sz, uint64_t max_mem)
-{
- uint64_t area_sz, max_pages;
-
- /* limit to RTE_MAX_MEMSEG_PER_LIST pages or RTE_MAX_MEM_MB_PER_LIST */
- max_pages = RTE_MAX_MEMSEG_PER_LIST;
- max_mem = RTE_MIN((uint64_t)RTE_MAX_MEM_MB_PER_LIST << 20, max_mem);
-
- area_sz = RTE_MIN(page_sz * max_pages, max_mem);
-
- /* make sure the list isn't smaller than the page size */
- area_sz = RTE_MAX(area_sz, page_sz);
-
- return RTE_ALIGN(area_sz, page_sz);
-}
-
-static int
-free_memseg_list(struct rte_memseg_list *msl)
-{
- if (rte_fbarray_destroy(&msl->memseg_arr)) {
- RTE_LOG(ERR, EAL, "Cannot destroy memseg list\n");
- return -1;
- }
- memset(msl, 0, sizeof(*msl));
- return 0;
-}
-
-static int
-alloc_memseg_list(struct rte_memseg_list *msl, uint64_t page_sz,
- uint64_t max_mem, int socket_id, int type_msl_idx)
+int
+eal_memseg_list_init_named(struct rte_memseg_list *msl, const char *name,
+ uint64_t page_sz, int n_segs, int socket_id, bool heap)
{
- char name[RTE_FBARRAY_NAME_LEN];
- uint64_t mem_amount;
- int max_segs;
-
- mem_amount = get_mem_amount(page_sz, max_mem);
- max_segs = mem_amount / page_sz;
-
- snprintf(name, sizeof(name), MEMSEG_LIST_FMT, page_sz >> 10, socket_id,
- type_msl_idx);
- if (rte_fbarray_init(&msl->memseg_arr, name, max_segs,
+ if (rte_fbarray_init(&msl->memseg_arr, name, n_segs,
sizeof(struct rte_memseg))) {
RTE_LOG(ERR, EAL, "Cannot allocate memseg list: %s\n",
rte_strerror(rte_errno));
msl->page_sz = page_sz;
msl->socket_id = socket_id;
msl->base_va = NULL;
+ msl->heap = heap;
- RTE_LOG(DEBUG, EAL, "Memseg list allocated: 0x%zxkB at socket %i\n",
- (size_t)page_sz >> 10, socket_id);
+ RTE_LOG(DEBUG, EAL,
+ "Memseg list allocated at socket %i, page size 0x%"PRIx64"kB\n",
+ socket_id, page_sz >> 10);
return 0;
}
-static int
-alloc_va_space(struct rte_memseg_list *msl)
+int
+eal_memseg_list_init(struct rte_memseg_list *msl, uint64_t page_sz,
+ int n_segs, int socket_id, int type_msl_idx, bool heap)
{
- uint64_t page_sz;
- size_t mem_sz;
- void *addr;
- int flags = 0;
+ char name[RTE_FBARRAY_NAME_LEN];
-#ifdef RTE_ARCH_PPC_64
- flags |= MAP_HUGETLB;
-#endif
+ snprintf(name, sizeof(name), MEMSEG_LIST_FMT, page_sz >> 10, socket_id,
+ type_msl_idx);
+
+ return eal_memseg_list_init_named(
+ msl, name, page_sz, n_segs, socket_id, heap);
+}
+
+int
+eal_memseg_list_alloc(struct rte_memseg_list *msl, int reserve_flags)
+{
+ size_t page_sz, mem_sz;
+ void *addr;
page_sz = msl->page_sz;
mem_sz = page_sz * msl->memseg_arr.len;
- addr = eal_get_virtual_area(msl->base_va, &mem_sz, page_sz, 0, flags);
+ addr = eal_get_virtual_area(
+ msl->base_va, &mem_sz, page_sz, 0, reserve_flags);
if (addr == NULL) {
+#ifndef RTE_EXEC_ENV_WINDOWS
+ /* The hint would be misleading on Windows, because address
+ * is by default system-selected (base VA = 0).
+ * However, this function is called from many places,
+ * including common code, so don't duplicate the message.
+ */
if (rte_errno == EADDRNOTAVAIL)
- RTE_LOG(ERR, EAL, "Could not mmap %llu bytes at [%p] - please use '--base-virtaddr' option\n",
+ RTE_LOG(ERR, EAL, "Cannot reserve %llu bytes at [%p] - "
+ "please use '--" OPT_BASE_VIRTADDR "' option\n",
(unsigned long long)mem_sz, msl->base_va);
- else
- RTE_LOG(ERR, EAL, "Cannot reserve memory\n");
+#endif
return -1;
}
msl->base_va = addr;
+ msl->len = mem_sz;
- return 0;
-}
-
-static int __rte_unused
-memseg_primary_init_32(void)
-{
- struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
- int active_sockets, hpi_idx, msl_idx = 0;
- unsigned int socket_id, i;
- struct rte_memseg_list *msl;
- uint64_t extra_mem_per_socket, total_extra_mem, total_requested_mem;
- uint64_t max_mem;
-
- /* no-huge does not need this at all */
- if (internal_config.no_hugetlbfs)
- return 0;
-
- /* this is a giant hack, but desperate times call for desperate
- * measures. in legacy 32-bit mode, we cannot preallocate VA space,
- * because having upwards of 2 gigabytes of VA space already mapped will
- * interfere with our ability to map and sort hugepages.
- *
- * therefore, in legacy 32-bit mode, we will be initializing memseg
- * lists much later - in eal_memory.c, right after we unmap all the
- * unneeded pages. this will not affect secondary processes, as those
- * should be able to mmap the space without (too many) problems.
- */
- if (internal_config.legacy_mem)
- return 0;
-
- /* 32-bit mode is a very special case. we cannot know in advance where
- * the user will want to allocate their memory, so we have to do some
- * heuristics.
- */
- active_sockets = 0;
- total_requested_mem = 0;
- if (internal_config.force_sockets)
- for (i = 0; i < rte_socket_count(); i++) {
- uint64_t mem;
-
- socket_id = rte_socket_id_by_idx(i);
- mem = internal_config.socket_mem[socket_id];
-
- if (mem == 0)
- continue;
-
- active_sockets++;
- total_requested_mem += mem;
- }
- else
- total_requested_mem = internal_config.memory;
-
- max_mem = (uint64_t)RTE_MAX_MEM_MB << 20;
- if (total_requested_mem > max_mem) {
- RTE_LOG(ERR, EAL, "Invalid parameters: 32-bit process can at most use %uM of memory\n",
- (unsigned int)(max_mem >> 20));
- return -1;
- }
- total_extra_mem = max_mem - total_requested_mem;
- extra_mem_per_socket = active_sockets == 0 ? total_extra_mem :
- total_extra_mem / active_sockets;
-
- /* the allocation logic is a little bit convoluted, but here's how it
- * works, in a nutshell:
- * - if user hasn't specified on which sockets to allocate memory via
- * --socket-mem, we allocate all of our memory on master core socket.
- * - if user has specified sockets to allocate memory on, there may be
- * some "unused" memory left (e.g. if user has specified --socket-mem
- * such that not all memory adds up to 2 gigabytes), so add it to all
- * sockets that are in use equally.
- *
- * page sizes are sorted by size in descending order, so we can safely
- * assume that we dispense with bigger page sizes first.
- */
-
- /* create memseg lists */
- for (i = 0; i < rte_socket_count(); i++) {
- int hp_sizes = (int) internal_config.num_hugepage_sizes;
- uint64_t max_socket_mem, cur_socket_mem;
- unsigned int master_lcore_socket;
- struct rte_config *cfg = rte_eal_get_configuration();
- bool skip;
-
- socket_id = rte_socket_id_by_idx(i);
-
-#ifndef RTE_EAL_NUMA_AWARE_HUGEPAGES
- if (socket_id > 0)
- break;
-#endif
-
- /* if we didn't specifically request memory on this socket */
- skip = active_sockets != 0 &&
- internal_config.socket_mem[socket_id] == 0;
- /* ...or if we didn't specifically request memory on *any*
- * socket, and this is not master lcore
- */
- master_lcore_socket = rte_lcore_to_socket_id(cfg->master_lcore);
- skip |= active_sockets == 0 && socket_id != master_lcore_socket;
-
- if (skip) {
- RTE_LOG(DEBUG, EAL, "Will not preallocate memory on socket %u\n",
- socket_id);
- continue;
- }
-
- /* max amount of memory on this socket */
- max_socket_mem = (active_sockets != 0 ?
- internal_config.socket_mem[socket_id] :
- internal_config.memory) +
- extra_mem_per_socket;
- cur_socket_mem = 0;
-
- for (hpi_idx = 0; hpi_idx < hp_sizes; hpi_idx++) {
- uint64_t max_pagesz_mem, cur_pagesz_mem = 0;
- uint64_t hugepage_sz;
- struct hugepage_info *hpi;
- int type_msl_idx, max_segs, total_segs = 0;
-
- hpi = &internal_config.hugepage_info[hpi_idx];
- hugepage_sz = hpi->hugepage_sz;
-
- /* check if pages are actually available */
- if (hpi->num_pages[socket_id] == 0)
- continue;
-
- max_segs = RTE_MAX_MEMSEG_PER_TYPE;
- max_pagesz_mem = max_socket_mem - cur_socket_mem;
-
- /* make it multiple of page size */
- max_pagesz_mem = RTE_ALIGN_FLOOR(max_pagesz_mem,
- hugepage_sz);
-
- RTE_LOG(DEBUG, EAL, "Attempting to preallocate "
- "%" PRIu64 "M on socket %i\n",
- max_pagesz_mem >> 20, socket_id);
-
- type_msl_idx = 0;
- while (cur_pagesz_mem < max_pagesz_mem &&
- total_segs < max_segs) {
- if (msl_idx >= RTE_MAX_MEMSEG_LISTS) {
- RTE_LOG(ERR, EAL,
- "No more space in memseg lists, please increase %s\n",
- RTE_STR(CONFIG_RTE_MAX_MEMSEG_LISTS));
- return -1;
- }
-
- msl = &mcfg->memsegs[msl_idx];
-
- if (alloc_memseg_list(msl, hugepage_sz,
- max_pagesz_mem, socket_id,
- type_msl_idx)) {
- /* failing to allocate a memseg list is
- * a serious error.
- */
- RTE_LOG(ERR, EAL, "Cannot allocate memseg list\n");
- return -1;
- }
-
- if (alloc_va_space(msl)) {
- /* if we couldn't allocate VA space, we
- * can try with smaller page sizes.
- */
- RTE_LOG(ERR, EAL, "Cannot allocate VA space for memseg list, retrying with different page size\n");
- /* deallocate memseg list */
- if (free_memseg_list(msl))
- return -1;
- break;
- }
-
- total_segs += msl->memseg_arr.len;
- cur_pagesz_mem = total_segs * hugepage_sz;
- type_msl_idx++;
- msl_idx++;
- }
- cur_socket_mem += cur_pagesz_mem;
- }
- if (cur_socket_mem == 0) {
- RTE_LOG(ERR, EAL, "Cannot allocate VA space on socket %u\n",
- socket_id);
- return -1;
- }
- }
-
- return 0;
-}
-
-static int __rte_unused
-memseg_primary_init(void)
-{
- struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
- int i, socket_id, hpi_idx, msl_idx = 0;
- struct rte_memseg_list *msl;
- uint64_t max_mem, total_mem;
+ RTE_LOG(DEBUG, EAL, "VA reserved for memseg list at %p, size %zx\n",
+ addr, mem_sz);
- /* no-huge does not need this at all */
- if (internal_config.no_hugetlbfs)
- return 0;
-
- max_mem = (uint64_t)RTE_MAX_MEM_MB << 20;
- total_mem = 0;
-
- /* create memseg lists */
- for (hpi_idx = 0; hpi_idx < (int) internal_config.num_hugepage_sizes;
- hpi_idx++) {
- struct hugepage_info *hpi;
- uint64_t hugepage_sz;
-
- hpi = &internal_config.hugepage_info[hpi_idx];
- hugepage_sz = hpi->hugepage_sz;
-
- for (i = 0; i < (int) rte_socket_count(); i++) {
- uint64_t max_type_mem, total_type_mem = 0;
- int type_msl_idx, max_segs, total_segs = 0;
-
- socket_id = rte_socket_id_by_idx(i);
-
-#ifndef RTE_EAL_NUMA_AWARE_HUGEPAGES
- if (socket_id > 0)
- break;
-#endif
-
- if (total_mem >= max_mem)
- break;
-
- max_type_mem = RTE_MIN(max_mem - total_mem,
- (uint64_t)RTE_MAX_MEM_MB_PER_TYPE << 20);
- max_segs = RTE_MAX_MEMSEG_PER_TYPE;
-
- type_msl_idx = 0;
- while (total_type_mem < max_type_mem &&
- total_segs < max_segs) {
- uint64_t cur_max_mem;
- if (msl_idx >= RTE_MAX_MEMSEG_LISTS) {
- RTE_LOG(ERR, EAL,
- "No more space in memseg lists, please increase %s\n",
- RTE_STR(CONFIG_RTE_MAX_MEMSEG_LISTS));
- return -1;
- }
-
- msl = &mcfg->memsegs[msl_idx++];
-
- cur_max_mem = max_type_mem - total_type_mem;
- if (alloc_memseg_list(msl, hugepage_sz,
- cur_max_mem, socket_id,
- type_msl_idx))
- return -1;
-
- total_segs += msl->memseg_arr.len;
- total_type_mem = total_segs * hugepage_sz;
- type_msl_idx++;
-
- if (alloc_va_space(msl)) {
- RTE_LOG(ERR, EAL, "Cannot allocate VA space for memseg list\n");
- return -1;
- }
- }
- total_mem += total_type_mem;
- }
- }
return 0;
}
-static int
-memseg_secondary_init(void)
+void
+eal_memseg_list_populate(struct rte_memseg_list *msl, void *addr, int n_segs)
{
- struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
- int msl_idx = 0;
- struct rte_memseg_list *msl;
+ size_t page_sz = msl->page_sz;
+ int i;
- for (msl_idx = 0; msl_idx < RTE_MAX_MEMSEG_LISTS; msl_idx++) {
+ for (i = 0; i < n_segs; i++) {
+ struct rte_fbarray *arr = &msl->memseg_arr;
+ struct rte_memseg *ms = rte_fbarray_get(arr, i);
- msl = &mcfg->memsegs[msl_idx];
+ if (rte_eal_iova_mode() == RTE_IOVA_VA)
+ ms->iova = (uintptr_t)addr;
+ else
+ ms->iova = RTE_BAD_IOVA;
+ ms->addr = addr;
+ ms->hugepage_sz = page_sz;
+ ms->socket_id = 0;
+ ms->len = page_sz;
- /* skip empty memseg lists */
- if (msl->memseg_arr.len == 0)
- continue;
+ rte_fbarray_set_used(arr, i);
- if (rte_fbarray_attach(&msl->memseg_arr)) {
- RTE_LOG(ERR, EAL, "Cannot attach to primary process memseg lists\n");
- return -1;
- }
-
- /* preallocate VA space */
- if (alloc_va_space(msl)) {
- RTE_LOG(ERR, EAL, "Cannot preallocate VA space for hugepage memory\n");
- return -1;
- }
+ addr = RTE_PTR_ADD(addr, page_sz);
}
-
- return 0;
}
static struct rte_memseg *
/* a memseg list was specified, check if it's the right one */
start = msl->base_va;
- end = RTE_PTR_ADD(start, (size_t)msl->page_sz * msl->memseg_arr.len);
+ end = RTE_PTR_ADD(start, msl->len);
if (addr < start || addr >= end)
return NULL;
msl = &mcfg->memsegs[msl_idx];
start = msl->base_va;
- end = RTE_PTR_ADD(start,
- (size_t)msl->page_sz * msl->memseg_arr.len);
+ end = RTE_PTR_ADD(start, msl->len);
if (addr >= start && addr < end)
break;
}
return msl;
}
-__rte_experimental struct rte_memseg_list *
+struct rte_memseg_list *
rte_mem_virt2memseg_list(const void *addr)
{
return virt2memseg_list(addr);
return 0;
}
-__rte_experimental void *
+void *
rte_mem_iova2virt(rte_iova_t iova)
{
struct virtiova vi;
+ const struct internal_config *internal_conf =
+ eal_get_internal_configuration();
memset(&vi, 0, sizeof(vi));
/* for legacy mem, we can get away with scanning VA-contiguous segments,
* as we know they are PA-contiguous as well
*/
- if (internal_config.legacy_mem)
+ if (internal_conf->legacy_mem)
rte_memseg_contig_walk(find_virt_legacy, &vi);
else
rte_memseg_walk(find_virt, &vi);
return vi.virt;
}
-__rte_experimental struct rte_memseg *
+struct rte_memseg *
rte_mem_virt2memseg(const void *addr, const struct rte_memseg_list *msl)
{
return virt2memseg(addr, msl != NULL ? msl :
{
uint64_t *total_len = arg;
+ if (msl->external)
+ return 0;
+
*total_len += msl->memseg_arr.count * msl->page_sz;
return 0;
void *arg)
{
struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
- int msl_idx, ms_idx;
+ int msl_idx, ms_idx, fd;
FILE *f = arg;
msl_idx = msl - mcfg->memsegs;
if (ms_idx < 0)
return -1;
+ fd = eal_memalloc_get_seg_fd(msl_idx, ms_idx);
fprintf(f, "Segment %i-%i: IOVA:0x%"PRIx64", len:%zu, "
"virt:%p, socket_id:%"PRId32", "
"hugepage_sz:%"PRIu64", nchannel:%"PRIx32", "
- "nrank:%"PRIx32"\n",
+ "nrank:%"PRIx32" fd:%i\n",
msl_idx, ms_idx,
ms->iova,
ms->len,
ms->socket_id,
ms->hugepage_sz,
ms->nchannel,
- ms->nrank);
+ ms->nrank,
+ fd);
return 0;
}
* Defining here because declared in rte_memory.h, but the actual implementation
* is in eal_common_memalloc.c, like all other memalloc internals.
*/
-int __rte_experimental
+int
rte_mem_event_callback_register(const char *name, rte_mem_event_callback_t clb,
void *arg)
{
+ const struct internal_config *internal_conf =
+ eal_get_internal_configuration();
+
/* FreeBSD boots with legacy mem enabled by default */
- if (internal_config.legacy_mem) {
+ if (internal_conf->legacy_mem) {
RTE_LOG(DEBUG, EAL, "Registering mem event callbacks not supported\n");
rte_errno = ENOTSUP;
return -1;
return eal_memalloc_mem_event_callback_register(name, clb, arg);
}
-int __rte_experimental
+int
rte_mem_event_callback_unregister(const char *name, void *arg)
{
+ const struct internal_config *internal_conf =
+ eal_get_internal_configuration();
+
/* FreeBSD boots with legacy mem enabled by default */
- if (internal_config.legacy_mem) {
+ if (internal_conf->legacy_mem) {
RTE_LOG(DEBUG, EAL, "Registering mem event callbacks not supported\n");
rte_errno = ENOTSUP;
return -1;
return eal_memalloc_mem_event_callback_unregister(name, arg);
}
-int __rte_experimental
+int
rte_mem_alloc_validator_register(const char *name,
rte_mem_alloc_validator_t clb, int socket_id, size_t limit)
{
+ const struct internal_config *internal_conf =
+ eal_get_internal_configuration();
+
/* FreeBSD boots with legacy mem enabled by default */
- if (internal_config.legacy_mem) {
+ if (internal_conf->legacy_mem) {
RTE_LOG(DEBUG, EAL, "Registering mem alloc validators not supported\n");
rte_errno = ENOTSUP;
return -1;
limit);
}
-int __rte_experimental
+int
rte_mem_alloc_validator_unregister(const char *name, int socket_id)
{
+ const struct internal_config *internal_conf =
+ eal_get_internal_configuration();
+
/* FreeBSD boots with legacy mem enabled by default */
- if (internal_config.legacy_mem) {
+ if (internal_conf->legacy_mem) {
RTE_LOG(DEBUG, EAL, "Registering mem alloc validators not supported\n");
rte_errno = ENOTSUP;
return -1;
rte_memseg_walk(dump_memseg, f);
}
+static int
+check_iova(const struct rte_memseg_list *msl __rte_unused,
+ const struct rte_memseg *ms, void *arg)
+{
+ uint64_t *mask = arg;
+ rte_iova_t iova;
+
+ /* higher address within segment */
+ iova = (ms->iova + ms->len) - 1;
+ if (!(iova & *mask))
+ return 0;
+
+ RTE_LOG(DEBUG, EAL, "memseg iova %"PRIx64", len %zx, out of range\n",
+ ms->iova, ms->len);
+
+ RTE_LOG(DEBUG, EAL, "\tusing dma mask %"PRIx64"\n", *mask);
+ return 1;
+}
+
+#define MAX_DMA_MASK_BITS 63
+
+/* check memseg iovas are within the required range based on dma mask */
+static int
+check_dma_mask(uint8_t maskbits, bool thread_unsafe)
+{
+ struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
+ uint64_t mask;
+ int ret;
+
+ /* Sanity check. We only check width can be managed with 64 bits
+ * variables. Indeed any higher value is likely wrong. */
+ if (maskbits > MAX_DMA_MASK_BITS) {
+ RTE_LOG(ERR, EAL, "wrong dma mask size %u (Max: %u)\n",
+ maskbits, MAX_DMA_MASK_BITS);
+ return -1;
+ }
+
+ /* create dma mask */
+ mask = ~((1ULL << maskbits) - 1);
+
+ if (thread_unsafe)
+ ret = rte_memseg_walk_thread_unsafe(check_iova, &mask);
+ else
+ ret = rte_memseg_walk(check_iova, &mask);
+
+ if (ret)
+ /*
+ * Dma mask precludes hugepage usage.
+ * This device can not be used and we do not need to keep
+ * the dma mask.
+ */
+ return 1;
+
+ /*
+ * we need to keep the more restricted maskbit for checking
+ * potential dynamic memory allocation in the future.
+ */
+ mcfg->dma_maskbits = mcfg->dma_maskbits == 0 ? maskbits :
+ RTE_MIN(mcfg->dma_maskbits, maskbits);
+
+ return 0;
+}
+
+int
+rte_mem_check_dma_mask(uint8_t maskbits)
+{
+ return check_dma_mask(maskbits, false);
+}
+
+int
+rte_mem_check_dma_mask_thread_unsafe(uint8_t maskbits)
+{
+ return check_dma_mask(maskbits, true);
+}
+
+/*
+ * Set dma mask to use when memory initialization is done.
+ *
+ * This function should ONLY be used by code executed before the memory
+ * initialization. PMDs should use rte_mem_check_dma_mask if addressing
+ * limitations by the device.
+ */
+void
+rte_mem_set_dma_mask(uint8_t maskbits)
+{
+ struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
+
+ mcfg->dma_maskbits = mcfg->dma_maskbits == 0 ? maskbits :
+ RTE_MIN(mcfg->dma_maskbits, maskbits);
+}
+
/* return the number of memory channels */
unsigned rte_memory_get_nchannel(void)
{
rte_eal_memdevice_init(void)
{
struct rte_config *config;
+ const struct internal_config *internal_conf;
if (rte_eal_process_type() == RTE_PROC_SECONDARY)
return 0;
+ internal_conf = eal_get_internal_configuration();
config = rte_eal_get_configuration();
- config->mem_config->nchannel = internal_config.force_nchannel;
- config->mem_config->nrank = internal_config.force_nrank;
+ config->mem_config->nchannel = internal_conf->force_nchannel;
+ config->mem_config->nrank = internal_conf->force_nrank;
return 0;
}
int
rte_mem_lock_page(const void *virt)
{
- unsigned long virtual = (unsigned long)virt;
- int page_size = getpagesize();
- unsigned long aligned = (virtual & ~(page_size - 1));
- return mlock((void *)aligned, page_size);
+ uintptr_t virtual = (uintptr_t)virt;
+ size_t page_size = rte_mem_page_size();
+ uintptr_t aligned = RTE_PTR_ALIGN_FLOOR(virtual, page_size);
+ return rte_mem_lock((void *)aligned, page_size);
}
-int __rte_experimental
-rte_memseg_contig_walk(rte_memseg_contig_walk_t func, void *arg)
+int
+rte_memseg_contig_walk_thread_unsafe(rte_memseg_contig_walk_t func, void *arg)
{
struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
int i, ms_idx, ret = 0;
- /* do not allow allocations/frees/init while we iterate */
- rte_rwlock_read_lock(&mcfg->memory_hotplug_lock);
-
for (i = 0; i < RTE_MAX_MEMSEG_LISTS; i++) {
struct rte_memseg_list *msl = &mcfg->memsegs[i];
const struct rte_memseg *ms;
len = n_segs * msl->page_sz;
ret = func(msl, ms, len, arg);
- if (ret < 0) {
- ret = -1;
- goto out;
- } else if (ret > 0) {
- ret = 1;
- goto out;
- }
+ if (ret)
+ return ret;
ms_idx = rte_fbarray_find_next_used(arr,
ms_idx + n_segs);
}
}
-out:
- rte_rwlock_read_unlock(&mcfg->memory_hotplug_lock);
+ return 0;
+}
+
+int
+rte_memseg_contig_walk(rte_memseg_contig_walk_t func, void *arg)
+{
+ int ret = 0;
+
+ /* do not allow allocations/frees/init while we iterate */
+ rte_mcfg_mem_read_lock();
+ ret = rte_memseg_contig_walk_thread_unsafe(func, arg);
+ rte_mcfg_mem_read_unlock();
+
return ret;
}
-int __rte_experimental
-rte_memseg_walk(rte_memseg_walk_t func, void *arg)
+int
+rte_memseg_walk_thread_unsafe(rte_memseg_walk_t func, void *arg)
{
struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
int i, ms_idx, ret = 0;
- /* do not allow allocations/frees/init while we iterate */
- rte_rwlock_read_lock(&mcfg->memory_hotplug_lock);
-
for (i = 0; i < RTE_MAX_MEMSEG_LISTS; i++) {
struct rte_memseg_list *msl = &mcfg->memsegs[i];
const struct rte_memseg *ms;
while (ms_idx >= 0) {
ms = rte_fbarray_get(arr, ms_idx);
ret = func(msl, ms, arg);
- if (ret < 0) {
- ret = -1;
- goto out;
- } else if (ret > 0) {
- ret = 1;
- goto out;
- }
+ if (ret)
+ return ret;
ms_idx = rte_fbarray_find_next_used(arr, ms_idx + 1);
}
}
-out:
- rte_rwlock_read_unlock(&mcfg->memory_hotplug_lock);
+ return 0;
+}
+
+int
+rte_memseg_walk(rte_memseg_walk_t func, void *arg)
+{
+ int ret = 0;
+
+ /* do not allow allocations/frees/init while we iterate */
+ rte_mcfg_mem_read_lock();
+ ret = rte_memseg_walk_thread_unsafe(func, arg);
+ rte_mcfg_mem_read_unlock();
+
return ret;
}
-int __rte_experimental
-rte_memseg_list_walk(rte_memseg_list_walk_t func, void *arg)
+int
+rte_memseg_list_walk_thread_unsafe(rte_memseg_list_walk_t func, void *arg)
{
struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
int i, ret = 0;
- /* do not allow allocations/frees/init while we iterate */
- rte_rwlock_read_lock(&mcfg->memory_hotplug_lock);
-
for (i = 0; i < RTE_MAX_MEMSEG_LISTS; i++) {
struct rte_memseg_list *msl = &mcfg->memsegs[i];
continue;
ret = func(msl, arg);
- if (ret < 0) {
- ret = -1;
- goto out;
- }
- if (ret > 0) {
- ret = 1;
- goto out;
- }
+ if (ret)
+ return ret;
+ }
+ return 0;
+}
+
+int
+rte_memseg_list_walk(rte_memseg_list_walk_t func, void *arg)
+{
+ int ret = 0;
+
+ /* do not allow allocations/frees/init while we iterate */
+ rte_mcfg_mem_read_lock();
+ ret = rte_memseg_list_walk_thread_unsafe(func, arg);
+ rte_mcfg_mem_read_unlock();
+
+ return ret;
+}
+
+int
+rte_memseg_get_fd_thread_unsafe(const struct rte_memseg *ms)
+{
+ struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
+ struct rte_memseg_list *msl;
+ struct rte_fbarray *arr;
+ int msl_idx, seg_idx, ret;
+
+ if (ms == NULL) {
+ rte_errno = EINVAL;
+ return -1;
+ }
+
+ msl = rte_mem_virt2memseg_list(ms->addr);
+ if (msl == NULL) {
+ rte_errno = EINVAL;
+ return -1;
+ }
+ arr = &msl->memseg_arr;
+
+ msl_idx = msl - mcfg->memsegs;
+ seg_idx = rte_fbarray_find_idx(arr, ms);
+
+ if (!rte_fbarray_is_used(arr, seg_idx)) {
+ rte_errno = ENOENT;
+ return -1;
+ }
+
+ /* segment fd API is not supported for external segments */
+ if (msl->external) {
+ rte_errno = ENOTSUP;
+ return -1;
+ }
+
+ ret = eal_memalloc_get_seg_fd(msl_idx, seg_idx);
+ if (ret < 0) {
+ rte_errno = -ret;
+ ret = -1;
+ }
+ return ret;
+}
+
+int
+rte_memseg_get_fd(const struct rte_memseg *ms)
+{
+ int ret;
+
+ rte_mcfg_mem_read_lock();
+ ret = rte_memseg_get_fd_thread_unsafe(ms);
+ rte_mcfg_mem_read_unlock();
+
+ return ret;
+}
+
+int
+rte_memseg_get_fd_offset_thread_unsafe(const struct rte_memseg *ms,
+ size_t *offset)
+{
+ struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
+ struct rte_memseg_list *msl;
+ struct rte_fbarray *arr;
+ int msl_idx, seg_idx, ret;
+
+ if (ms == NULL || offset == NULL) {
+ rte_errno = EINVAL;
+ return -1;
+ }
+
+ msl = rte_mem_virt2memseg_list(ms->addr);
+ if (msl == NULL) {
+ rte_errno = EINVAL;
+ return -1;
+ }
+ arr = &msl->memseg_arr;
+
+ msl_idx = msl - mcfg->memsegs;
+ seg_idx = rte_fbarray_find_idx(arr, ms);
+
+ if (!rte_fbarray_is_used(arr, seg_idx)) {
+ rte_errno = ENOENT;
+ return -1;
+ }
+
+ /* segment fd API is not supported for external segments */
+ if (msl->external) {
+ rte_errno = ENOTSUP;
+ return -1;
+ }
+
+ ret = eal_memalloc_get_seg_fd_offset(msl_idx, seg_idx, offset);
+ if (ret < 0) {
+ rte_errno = -ret;
+ ret = -1;
}
-out:
- rte_rwlock_read_unlock(&mcfg->memory_hotplug_lock);
return ret;
}
+int
+rte_memseg_get_fd_offset(const struct rte_memseg *ms, size_t *offset)
+{
+ int ret;
+
+ rte_mcfg_mem_read_lock();
+ ret = rte_memseg_get_fd_offset_thread_unsafe(ms, offset);
+ rte_mcfg_mem_read_unlock();
+
+ return ret;
+}
+
+int
+rte_extmem_register(void *va_addr, size_t len, rte_iova_t iova_addrs[],
+ unsigned int n_pages, size_t page_sz)
+{
+ struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
+ unsigned int socket_id, n;
+ int ret = 0;
+
+ if (va_addr == NULL || page_sz == 0 || len == 0 ||
+ !rte_is_power_of_2(page_sz) ||
+ RTE_ALIGN(len, page_sz) != len ||
+ ((len / page_sz) != n_pages && iova_addrs != NULL) ||
+ !rte_is_aligned(va_addr, page_sz)) {
+ rte_errno = EINVAL;
+ return -1;
+ }
+ rte_mcfg_mem_write_lock();
+
+ /* make sure the segment doesn't already exist */
+ if (malloc_heap_find_external_seg(va_addr, len) != NULL) {
+ rte_errno = EEXIST;
+ ret = -1;
+ goto unlock;
+ }
+
+ /* get next available socket ID */
+ socket_id = mcfg->next_socket_id;
+ if (socket_id > INT32_MAX) {
+ RTE_LOG(ERR, EAL, "Cannot assign new socket ID's\n");
+ rte_errno = ENOSPC;
+ ret = -1;
+ goto unlock;
+ }
+
+ /* we can create a new memseg */
+ n = len / page_sz;
+ if (malloc_heap_create_external_seg(va_addr, iova_addrs, n,
+ page_sz, "extmem", socket_id) == NULL) {
+ ret = -1;
+ goto unlock;
+ }
+
+ /* memseg list successfully created - increment next socket ID */
+ mcfg->next_socket_id++;
+unlock:
+ rte_mcfg_mem_write_unlock();
+ return ret;
+}
+
+int
+rte_extmem_unregister(void *va_addr, size_t len)
+{
+ struct rte_memseg_list *msl;
+ int ret = 0;
+
+ if (va_addr == NULL || len == 0) {
+ rte_errno = EINVAL;
+ return -1;
+ }
+ rte_mcfg_mem_write_lock();
+
+ /* find our segment */
+ msl = malloc_heap_find_external_seg(va_addr, len);
+ if (msl == NULL) {
+ rte_errno = ENOENT;
+ ret = -1;
+ goto unlock;
+ }
+
+ ret = malloc_heap_destroy_external_seg(msl);
+unlock:
+ rte_mcfg_mem_write_unlock();
+ return ret;
+}
+
+static int
+sync_memory(void *va_addr, size_t len, bool attach)
+{
+ struct rte_memseg_list *msl;
+ int ret = 0;
+
+ if (va_addr == NULL || len == 0) {
+ rte_errno = EINVAL;
+ return -1;
+ }
+ rte_mcfg_mem_write_lock();
+
+ /* find our segment */
+ msl = malloc_heap_find_external_seg(va_addr, len);
+ if (msl == NULL) {
+ rte_errno = ENOENT;
+ ret = -1;
+ goto unlock;
+ }
+ if (attach)
+ ret = rte_fbarray_attach(&msl->memseg_arr);
+ else
+ ret = rte_fbarray_detach(&msl->memseg_arr);
+
+unlock:
+ rte_mcfg_mem_write_unlock();
+ return ret;
+}
+
+int
+rte_extmem_attach(void *va_addr, size_t len)
+{
+ return sync_memory(va_addr, len, true);
+}
+
+int
+rte_extmem_detach(void *va_addr, size_t len)
+{
+ return sync_memory(va_addr, len, false);
+}
+
/* init memory subsystem */
int
rte_eal_memory_init(void)
{
struct rte_mem_config *mcfg = rte_eal_get_configuration()->mem_config;
+ const struct internal_config *internal_conf =
+ eal_get_internal_configuration();
+
int retval;
RTE_LOG(DEBUG, EAL, "Setting up physically contiguous memory...\n");
return -1;
/* lock mem hotplug here, to prevent races while we init */
- rte_rwlock_read_lock(&mcfg->memory_hotplug_lock);
+ rte_mcfg_mem_read_lock();
- retval = rte_eal_process_type() == RTE_PROC_PRIMARY ?
-#ifndef RTE_ARCH_64
- memseg_primary_init_32() :
-#else
- memseg_primary_init() :
-#endif
- memseg_secondary_init();
-
- if (retval < 0)
+ if (rte_eal_memseg_init() < 0)
goto fail;
if (eal_memalloc_init() < 0)
if (retval < 0)
goto fail;
- if (internal_config.no_shconf == 0 && rte_eal_memdevice_init() < 0)
+ if (internal_conf->no_shconf == 0 && rte_eal_memdevice_init() < 0)
goto fail;
return 0;
fail:
- rte_rwlock_read_unlock(&mcfg->memory_hotplug_lock);
+ rte_mcfg_mem_read_unlock();
return -1;
}
git.droids-corp.org / dpdk.git / blobdiff