#include "migration/migration.h"
#include "migration/qemu-file.h"
#include "exec/cpu-common.h"
+#include "qemu/error-report.h"
#include "qemu/main-loop.h"
#include "qemu/sockets.h"
#include "qemu/bitmap.h"
* the information. It's small anyway, so a list is overkill.
*/
typedef struct RDMALocalBlock {
- uint8_t *local_host_addr; /* local virtual address */
- uint64_t remote_host_addr; /* remote virtual address */
- uint64_t offset;
- uint64_t length;
- struct ibv_mr **pmr; /* MRs for chunk-level registration */
- struct ibv_mr *mr; /* MR for non-chunk-level registration */
- uint32_t *remote_keys; /* rkeys for chunk-level registration */
- uint32_t remote_rkey; /* rkeys for non-chunk-level registration */
- int index; /* which block are we */
- bool is_ram_block;
- int nb_chunks;
+ char *block_name;
+ uint8_t *local_host_addr; /* local virtual address */
+ uint64_t remote_host_addr; /* remote virtual address */
+ uint64_t offset;
+ uint64_t length;
+ struct ibv_mr **pmr; /* MRs for chunk-level registration */
+ struct ibv_mr *mr; /* MR for non-chunk-level registration */
+ uint32_t *remote_keys; /* rkeys for chunk-level registration */
+ uint32_t remote_rkey; /* rkeys for non-chunk-level registration */
+ int index; /* which block are we */
+ unsigned int src_index; /* (Only used on dest) */
+ bool is_ram_block;
+ int nb_chunks;
unsigned long *transit_bitmap;
unsigned long *unregister_bitmap;
} RDMALocalBlock;
* corresponding RDMALocalBlock with
* the information needed to perform the actual RDMA.
*/
-typedef struct QEMU_PACKED RDMARemoteBlock {
+typedef struct QEMU_PACKED RDMADestBlock {
uint64_t remote_host_addr;
uint64_t offset;
uint64_t length;
uint32_t remote_rkey;
uint32_t padding;
-} RDMARemoteBlock;
+} RDMADestBlock;
static uint64_t htonll(uint64_t v)
{
return ((uint64_t)ntohl(u.lv[0]) << 32) | (uint64_t) ntohl(u.lv[1]);
}
-static void remote_block_to_network(RDMARemoteBlock *rb)
+static void dest_block_to_network(RDMADestBlock *db)
{
- rb->remote_host_addr = htonll(rb->remote_host_addr);
- rb->offset = htonll(rb->offset);
- rb->length = htonll(rb->length);
- rb->remote_rkey = htonl(rb->remote_rkey);
+ db->remote_host_addr = htonll(db->remote_host_addr);
+ db->offset = htonll(db->offset);
+ db->length = htonll(db->length);
+ db->remote_rkey = htonl(db->remote_rkey);
}
-static void network_to_remote_block(RDMARemoteBlock *rb)
+static void network_to_dest_block(RDMADestBlock *db)
{
- rb->remote_host_addr = ntohll(rb->remote_host_addr);
- rb->offset = ntohll(rb->offset);
- rb->length = ntohll(rb->length);
- rb->remote_rkey = ntohl(rb->remote_rkey);
+ db->remote_host_addr = ntohll(db->remote_host_addr);
+ db->offset = ntohll(db->offset);
+ db->length = ntohll(db->length);
+ db->remote_rkey = ntohl(db->remote_rkey);
}
/*
* Description of ram blocks used throughout the code.
*/
RDMALocalBlocks local_ram_blocks;
- RDMARemoteBlock *block;
+ RDMADestBlock *dest_blocks;
+
+ /* Index of the next RAMBlock received during block registration */
+ unsigned int next_src_index;
/*
* Migration on *destination* started.
*/
typedef struct QEMU_PACKED {
union QEMU_PACKED {
- uint64_t current_addr; /* offset into the ramblock of the chunk */
+ uint64_t current_addr; /* offset into the ram_addr_t space */
uint64_t chunk; /* chunk to lookup if unregistering */
} key;
uint32_t current_index; /* which ramblock the chunk belongs to */
uint64_t chunks; /* how many sequential chunks to register */
} RDMARegister;
-static void register_to_network(RDMARegister *reg)
+static void register_to_network(RDMAContext *rdma, RDMARegister *reg)
{
+ RDMALocalBlock *local_block;
+ local_block = &rdma->local_ram_blocks.block[reg->current_index];
+
+ if (local_block->is_ram_block) {
+ /*
+ * current_addr as passed in is an address in the local ram_addr_t
+ * space, we need to translate this for the destination
+ */
+ reg->key.current_addr -= local_block->offset;
+ reg->key.current_addr += rdma->dest_blocks[reg->current_index].offset;
+ }
reg->key.current_addr = htonll(reg->key.current_addr);
reg->current_index = htonl(reg->current_index);
reg->chunks = htonll(reg->chunks);
typedef struct QEMU_PACKED {
uint32_t value; /* if zero, we will madvise() */
uint32_t block_idx; /* which ram block index */
- uint64_t offset; /* where in the remote ramblock this chunk */
+ uint64_t offset; /* Address in remote ram_addr_t space */
uint64_t length; /* length of the chunk */
} RDMACompress;
-static void compress_to_network(RDMACompress *comp)
+static void compress_to_network(RDMAContext *rdma, RDMACompress *comp)
{
comp->value = htonl(comp->value);
+ /*
+ * comp->offset as passed in is an address in the local ram_addr_t
+ * space, we need to translate this for the destination
+ */
+ comp->offset -= rdma->local_ram_blocks.block[comp->block_idx].offset;
+ comp->offset += rdma->dest_blocks[comp->block_idx].offset;
comp->block_idx = htonl(comp->block_idx);
comp->offset = htonll(comp->offset);
comp->length = htonll(comp->length);
static inline uint8_t *ram_chunk_start(const RDMALocalBlock *rdma_ram_block,
uint64_t i)
{
- return (uint8_t *) (((uintptr_t) rdma_ram_block->local_host_addr)
- + (i << RDMA_REG_CHUNK_SHIFT));
+ return (uint8_t *)(uintptr_t)(rdma_ram_block->local_host_addr +
+ (i << RDMA_REG_CHUNK_SHIFT));
}
static inline uint8_t *ram_chunk_end(const RDMALocalBlock *rdma_ram_block,
return result;
}
-static int __qemu_rdma_add_block(RDMAContext *rdma, void *host_addr,
+static int rdma_add_block(RDMAContext *rdma, const char *block_name,
+ void *host_addr,
ram_addr_t block_offset, uint64_t length)
{
RDMALocalBlocks *local = &rdma->local_ram_blocks;
- RDMALocalBlock *block = g_hash_table_lookup(rdma->blockmap,
- (void *) block_offset);
+ RDMALocalBlock *block;
RDMALocalBlock *old = local->block;
- assert(block == NULL);
-
local->block = g_malloc0(sizeof(RDMALocalBlock) * (local->nb_blocks + 1));
if (local->nb_blocks) {
int x;
- for (x = 0; x < local->nb_blocks; x++) {
- g_hash_table_remove(rdma->blockmap, (void *)old[x].offset);
- g_hash_table_insert(rdma->blockmap, (void *)old[x].offset,
- &local->block[x]);
+ if (rdma->blockmap) {
+ for (x = 0; x < local->nb_blocks; x++) {
+ g_hash_table_remove(rdma->blockmap,
+ (void *)(uintptr_t)old[x].offset);
+ g_hash_table_insert(rdma->blockmap,
+ (void *)(uintptr_t)old[x].offset,
+ &local->block[x]);
+ }
}
memcpy(local->block, old, sizeof(RDMALocalBlock) * local->nb_blocks);
g_free(old);
block = &local->block[local->nb_blocks];
+ block->block_name = g_strdup(block_name);
block->local_host_addr = host_addr;
block->offset = block_offset;
block->length = length;
block->index = local->nb_blocks;
+ block->src_index = ~0U; /* Filled in by the receipt of the block list */
block->nb_chunks = ram_chunk_index(host_addr, host_addr + length) + 1UL;
block->transit_bitmap = bitmap_new(block->nb_chunks);
bitmap_clear(block->transit_bitmap, 0, block->nb_chunks);
block->is_ram_block = local->init ? false : true;
- g_hash_table_insert(rdma->blockmap, (void *) block_offset, block);
+ if (rdma->blockmap) {
+ g_hash_table_insert(rdma->blockmap, (void *) block_offset, block);
+ }
- trace___qemu_rdma_add_block(local->nb_blocks,
- (uint64_t) block->local_host_addr, block->offset,
- block->length,
- (uint64_t) (block->local_host_addr + block->length),
- BITS_TO_LONGS(block->nb_chunks) *
- sizeof(unsigned long) * 8,
- block->nb_chunks);
+ trace_rdma_add_block(block_name, local->nb_blocks,
+ (uintptr_t) block->local_host_addr,
+ block->offset, block->length,
+ (uintptr_t) (block->local_host_addr + block->length),
+ BITS_TO_LONGS(block->nb_chunks) *
+ sizeof(unsigned long) * 8,
+ block->nb_chunks);
local->nb_blocks++;
* in advanced before the migration starts. This tells us where the RAM blocks
* are so that we can register them individually.
*/
-static void qemu_rdma_init_one_block(void *host_addr,
+static int qemu_rdma_init_one_block(const char *block_name, void *host_addr,
ram_addr_t block_offset, ram_addr_t length, void *opaque)
{
- __qemu_rdma_add_block(opaque, host_addr, block_offset, length);
+ return rdma_add_block(opaque, block_name, host_addr, block_offset, length);
}
/*
RDMALocalBlocks *local = &rdma->local_ram_blocks;
assert(rdma->blockmap == NULL);
- rdma->blockmap = g_hash_table_new(g_direct_hash, g_direct_equal);
memset(local, 0, sizeof *local);
qemu_ram_foreach_block(qemu_rdma_init_one_block, rdma);
trace_qemu_rdma_init_ram_blocks(local->nb_blocks);
- rdma->block = (RDMARemoteBlock *) g_malloc0(sizeof(RDMARemoteBlock) *
+ rdma->dest_blocks = (RDMADestBlock *) g_malloc0(sizeof(RDMADestBlock) *
rdma->local_ram_blocks.nb_blocks);
local->init = true;
return 0;
}
-static int __qemu_rdma_delete_block(RDMAContext *rdma, ram_addr_t block_offset)
+/*
+ * Note: If used outside of cleanup, the caller must ensure that the destination
+ * block structures are also updated
+ */
+static int rdma_delete_block(RDMAContext *rdma, RDMALocalBlock *block)
{
RDMALocalBlocks *local = &rdma->local_ram_blocks;
- RDMALocalBlock *block = g_hash_table_lookup(rdma->blockmap,
- (void *) block_offset);
RDMALocalBlock *old = local->block;
int x;
- assert(block);
-
+ if (rdma->blockmap) {
+ g_hash_table_remove(rdma->blockmap, (void *)(uintptr_t)block->offset);
+ }
if (block->pmr) {
int j;
g_free(block->remote_keys);
block->remote_keys = NULL;
- for (x = 0; x < local->nb_blocks; x++) {
- g_hash_table_remove(rdma->blockmap, (void *)old[x].offset);
+ g_free(block->block_name);
+ block->block_name = NULL;
+
+ if (rdma->blockmap) {
+ for (x = 0; x < local->nb_blocks; x++) {
+ g_hash_table_remove(rdma->blockmap,
+ (void *)(uintptr_t)old[x].offset);
+ }
}
if (local->nb_blocks > 1) {
local->block = NULL;
}
- trace___qemu_rdma_delete_block(local->nb_blocks,
- (uint64_t)block->local_host_addr,
+ trace_rdma_delete_block(block, (uintptr_t)block->local_host_addr,
block->offset, block->length,
- (uint64_t)(block->local_host_addr + block->length),
+ (uintptr_t)(block->local_host_addr + block->length),
BITS_TO_LONGS(block->nb_chunks) *
sizeof(unsigned long) * 8, block->nb_chunks);
local->nb_blocks--;
- if (local->nb_blocks) {
+ if (local->nb_blocks && rdma->blockmap) {
for (x = 0; x < local->nb_blocks; x++) {
- g_hash_table_insert(rdma->blockmap, (void *)local->block[x].offset,
- &local->block[x]);
+ g_hash_table_insert(rdma->blockmap,
+ (void *)(uintptr_t)local->block[x].offset,
+ &local->block[x]);
}
}
verbs->device->ibdev_path,
port.link_layer,
(port.link_layer == IBV_LINK_LAYER_INFINIBAND) ? "Infiniband" :
- ((port.link_layer == IBV_LINK_LAYER_ETHERNET)
+ ((port.link_layer == IBV_LINK_LAYER_ETHERNET)
? "Ethernet" : "Unknown"));
}
* and validate what time of hardware it is.
*
* Unfortunately, this puts the user in a fix:
- *
+ *
* If the source VM connects with an IPv4 address without knowing that the
* destination has bound to '[::]' the migration will unconditionally fail
- * unless the management software is explicitly listening on the the IPv4
+ * unless the management software is explicitly listening on the IPv4
* address while using a RoCE-based device.
*
* If the source VM connects with an IPv6 address, then we're OK because we can
* throw an error on the source (and similarly on the destination).
- *
+ *
* But in mixed environments, this will be broken for a while until it is fixed
* inside linux.
*
* We do provide a *tiny* bit of help in this function: We can list all of the
* devices in the system and check to see if all the devices are RoCE or
- * Infiniband.
+ * Infiniband.
*
* If we detect that we have a *pure* RoCE environment, then we can safely
* thrown an error even if the management software has specified '[::]' as the
/* This bug only exists in linux, to our knowledge. */
#ifdef CONFIG_LINUX
- /*
+ /*
* Verbs are only NULL if management has bound to '[::]'.
- *
+ *
* Let's iterate through all the devices and see if there any pure IB
* devices (non-ethernet).
- *
+ *
* If not, then we can safely proceed with the migration.
* Otherwise, there are no guarantees until the bug is fixed in linux.
*/
if (!verbs) {
- int num_devices, x;
+ int num_devices, x;
struct ibv_device ** dev_list = ibv_get_device_list(&num_devices);
bool roce_found = false;
bool ib_found = false;
for (x = 0; x < num_devices; x++) {
verbs = ibv_open_device(dev_list[x]);
+ if (!verbs) {
+ if (errno == EPERM) {
+ continue;
+ } else {
+ return -EINVAL;
+ }
+ }
if (ibv_query_port(verbs, 1, &port_attr)) {
ibv_close_device(verbs);
/*
* If we have a verbs context, that means that some other than '[::]' was
- * used by the management software for binding. In which case we can actually
- * warn the user about a potential broken kernel;
+ * used by the management software for binding. In which case we can
+ * actually warn the user about a potentially broken kernel.
*/
/* IB ports start with 1, not 0 */
* This search cannot fail or the migration will fail.
*/
static int qemu_rdma_search_ram_block(RDMAContext *rdma,
- uint64_t block_offset,
+ uintptr_t block_offset,
uint64_t offset,
uint64_t length,
uint64_t *block_index,
* to perform the actual RDMA operation.
*/
static int qemu_rdma_register_and_get_keys(RDMAContext *rdma,
- RDMALocalBlock *block, uint8_t *host_addr,
+ RDMALocalBlock *block, uintptr_t host_addr,
uint32_t *lkey, uint32_t *rkey, int chunk,
uint8_t *chunk_start, uint8_t *chunk_end)
{
if (!block->pmr[chunk]) {
perror("Failed to register chunk!");
fprintf(stderr, "Chunk details: block: %d chunk index %d"
- " start %" PRIu64 " end %" PRIu64 " host %" PRIu64
- " local %" PRIu64 " registrations: %d\n",
- block->index, chunk, (uint64_t) chunk_start,
- (uint64_t) chunk_end, (uint64_t) host_addr,
- (uint64_t) block->local_host_addr,
+ " start %" PRIuPTR " end %" PRIuPTR
+ " host %" PRIuPTR
+ " local %" PRIuPTR " registrations: %d\n",
+ block->index, chunk, (uintptr_t)chunk_start,
+ (uintptr_t)chunk_end, host_addr,
+ (uintptr_t)block->local_host_addr,
rdma->total_registrations);
return -1;
}
/*
* Perform a non-optimized memory unregistration after every transfer
- * for demonsration purposes, only if pin-all is not requested.
+ * for demonstration purposes, only if pin-all is not requested.
*
* Potential optimizations:
* 1. Start a new thread to run this function continuously
rdma->total_registrations--;
reg.key.chunk = chunk;
- register_to_network(®);
+ register_to_network(rdma, ®);
ret = qemu_rdma_exchange_send(rdma, &head, (uint8_t *) ®,
&resp, NULL, NULL);
if (ret < 0) {
RDMALocalBlock *block = &(rdma->local_ram_blocks.block[index]);
trace_qemu_rdma_poll_write(print_wrid(wr_id), wr_id, rdma->nb_sent,
- index, chunk,
- block->local_host_addr, (void *)block->remote_host_addr);
+ index, chunk, block->local_host_addr,
+ (void *)(uintptr_t)block->remote_host_addr);
clear_bit(chunk, block->transit_bitmap);
RDMAWorkRequestData *wr = &rdma->wr_data[RDMA_WRID_CONTROL];
struct ibv_send_wr *bad_wr;
struct ibv_sge sge = {
- .addr = (uint64_t)(wr->control),
+ .addr = (uintptr_t)(wr->control),
.length = head->len + sizeof(RDMAControlHeader),
.lkey = wr->control_mr->lkey,
};
{
struct ibv_recv_wr *bad_wr;
struct ibv_sge sge = {
- .addr = (uint64_t)(rdma->wr_data[idx].control),
+ .addr = (uintptr_t)(rdma->wr_data[idx].control),
.length = RDMA_CONTROL_MAX_BUFFER,
.lkey = rdma->wr_data[idx].control_mr->lkey,
};
return -EIO;
}
if (head->len > RDMA_CONTROL_MAX_BUFFER - sizeof(*head)) {
- error_report("too long length: %d\n", head->len);
+ error_report("too long length: %d", head->len);
return -EINVAL;
}
if (sizeof(*head) + head->len != byte_len) {
};
retry:
- sge.addr = (uint64_t)(block->local_host_addr +
+ sge.addr = (uintptr_t)(block->local_host_addr +
(current_addr - block->offset));
sge.length = length;
- chunk = ram_chunk_index(block->local_host_addr, (uint8_t *) sge.addr);
+ chunk = ram_chunk_index(block->local_host_addr,
+ (uint8_t *)(uintptr_t)sge.addr);
chunk_start = ram_chunk_start(block, chunk);
if (block->is_ram_block) {
* memset() + madvise() the entire chunk without RDMA.
*/
- if (can_use_buffer_find_nonzero_offset((void *)sge.addr, length)
- && buffer_find_nonzero_offset((void *)sge.addr,
+ if (can_use_buffer_find_nonzero_offset((void *)(uintptr_t)sge.addr,
+ length)
+ && buffer_find_nonzero_offset((void *)(uintptr_t)sge.addr,
length) == length) {
RDMACompress comp = {
.offset = current_addr,
trace_qemu_rdma_write_one_zero(chunk, sge.length,
current_index, current_addr);
- compress_to_network(&comp);
+ compress_to_network(rdma, &comp);
ret = qemu_rdma_exchange_send(rdma, &head,
(uint8_t *) &comp, NULL, NULL, NULL);
trace_qemu_rdma_write_one_sendreg(chunk, sge.length, current_index,
current_addr);
- register_to_network(®);
+ register_to_network(rdma, ®);
ret = qemu_rdma_exchange_send(rdma, &head, (uint8_t *) ®,
&resp, ®_result_idx, NULL);
if (ret < 0) {
}
/* try to overlap this single registration with the one we sent. */
- if (qemu_rdma_register_and_get_keys(rdma, block,
- (uint8_t *) sge.addr,
+ if (qemu_rdma_register_and_get_keys(rdma, block, sge.addr,
&sge.lkey, NULL, chunk,
chunk_start, chunk_end)) {
error_report("cannot get lkey");
block->remote_host_addr = reg_result->host_addr;
} else {
/* already registered before */
- if (qemu_rdma_register_and_get_keys(rdma, block,
- (uint8_t *)sge.addr,
+ if (qemu_rdma_register_and_get_keys(rdma, block, sge.addr,
&sge.lkey, NULL, chunk,
chunk_start, chunk_end)) {
error_report("cannot get lkey!");
} else {
send_wr.wr.rdma.rkey = block->remote_rkey;
- if (qemu_rdma_register_and_get_keys(rdma, block, (uint8_t *)sge.addr,
+ if (qemu_rdma_register_and_get_keys(rdma, block, sge.addr,
&sge.lkey, NULL, chunk,
chunk_start, chunk_end)) {
error_report("cannot get lkey!");
rdma->connected = false;
}
- g_free(rdma->block);
- rdma->block = NULL;
+ g_free(rdma->dest_blocks);
+ rdma->dest_blocks = NULL;
for (idx = 0; idx < RDMA_WRID_MAX; idx++) {
if (rdma->wr_data[idx].control_mr) {
if (rdma->local_ram_blocks.block) {
while (rdma->local_ram_blocks.nb_blocks) {
- __qemu_rdma_delete_block(rdma,
- rdma->local_ram_blocks.block->offset);
+ rdma_delete_block(rdma, &rdma->local_ram_blocks.block[0]);
}
}
+ if (rdma->qp) {
+ rdma_destroy_qp(rdma->cm_id);
+ rdma->qp = NULL;
+ }
if (rdma->cq) {
ibv_destroy_cq(rdma->cq);
rdma->cq = NULL;
ibv_dealloc_pd(rdma->pd);
rdma->pd = NULL;
}
- if (rdma->listen_id) {
- rdma_destroy_id(rdma->listen_id);
- rdma->listen_id = NULL;
- }
if (rdma->cm_id) {
- if (rdma->qp) {
- rdma_destroy_qp(rdma->cm_id);
- rdma->qp = NULL;
- }
rdma_destroy_id(rdma->cm_id);
rdma->cm_id = NULL;
}
+ if (rdma->listen_id) {
+ rdma_destroy_id(rdma->listen_id);
+ rdma->listen_id = NULL;
+ }
if (rdma->channel) {
rdma_destroy_event_channel(rdma->channel);
rdma->channel = NULL;
goto err_rdma_source_init;
}
+ /* Build the hash that maps from offset to RAMBlock */
+ rdma->blockmap = g_hash_table_new(g_direct_hash, g_direct_equal);
+ for (idx = 0; idx < rdma->local_ram_blocks.nb_blocks; idx++) {
+ g_hash_table_insert(rdma->blockmap,
+ (void *)(uintptr_t)rdma->local_ram_blocks.block[idx].offset,
+ &rdma->local_ram_blocks.block[idx]);
+ }
+
for (idx = 0; idx < RDMA_WRID_MAX; idx++) {
ret = qemu_rdma_reg_control(rdma, idx);
if (ret) {
if (ret) {
perror("rdma_connect");
ERROR(errp, "connecting to destination!");
- rdma_destroy_id(rdma->cm_id);
- rdma->cm_id = NULL;
goto err_rdma_source_connect;
}
perror("rdma_get_cm_event after rdma_connect");
ERROR(errp, "connecting to destination!");
rdma_ack_cm_event(cm_event);
- rdma_destroy_id(rdma->cm_id);
- rdma->cm_id = NULL;
goto err_rdma_source_connect;
}
perror("rdma_get_cm_event != EVENT_ESTABLISHED after rdma_connect");
ERROR(errp, "connecting to destination!");
rdma_ack_cm_event(cm_event);
- rdma_destroy_id(rdma->cm_id);
- rdma->cm_id = NULL;
goto err_rdma_source_connect;
}
rdma->connected = true;
static int qemu_rdma_dest_init(RDMAContext *rdma, Error **errp)
{
- int ret = -EINVAL, idx;
+ int ret, idx;
struct rdma_cm_id *listen_id;
char ip[40] = "unknown";
- struct rdma_addrinfo *res;
+ struct rdma_addrinfo *res, *e;
char port_str[16];
for (idx = 0; idx < RDMA_WRID_MAX; idx++) {
rdma->wr_data[idx].control_curr = NULL;
}
- if (rdma->host == NULL) {
+ if (!rdma->host || !rdma->host[0]) {
ERROR(errp, "RDMA host is not set!");
rdma->error_state = -EINVAL;
return -1;
snprintf(port_str, 16, "%d", rdma->port);
port_str[15] = '\0';
- if (rdma->host && strcmp("", rdma->host)) {
- struct rdma_addrinfo *e;
+ ret = rdma_getaddrinfo(rdma->host, port_str, NULL, &res);
+ if (ret < 0) {
+ ERROR(errp, "could not rdma_getaddrinfo address %s", rdma->host);
+ goto err_dest_init_bind_addr;
+ }
- ret = rdma_getaddrinfo(rdma->host, port_str, NULL, &res);
- if (ret < 0) {
- ERROR(errp, "could not rdma_getaddrinfo address %s", rdma->host);
- goto err_dest_init_bind_addr;
+ for (e = res; e != NULL; e = e->ai_next) {
+ inet_ntop(e->ai_family,
+ &((struct sockaddr_in *) e->ai_dst_addr)->sin_addr, ip, sizeof ip);
+ trace_qemu_rdma_dest_init_trying(rdma->host, ip);
+ ret = rdma_bind_addr(listen_id, e->ai_dst_addr);
+ if (ret) {
+ continue;
}
-
- for (e = res; e != NULL; e = e->ai_next) {
- inet_ntop(e->ai_family,
- &((struct sockaddr_in *) e->ai_dst_addr)->sin_addr, ip, sizeof ip);
- trace_qemu_rdma_dest_init_trying(rdma->host, ip);
- ret = rdma_bind_addr(listen_id, e->ai_dst_addr);
- if (!ret) {
- if (e->ai_family == AF_INET6) {
- ret = qemu_rdma_broken_ipv6_kernel(errp, listen_id->verbs);
- if (ret) {
- continue;
- }
- }
-
- goto listen;
+ if (e->ai_family == AF_INET6) {
+ ret = qemu_rdma_broken_ipv6_kernel(errp, listen_id->verbs);
+ if (ret) {
+ continue;
}
}
+ break;
+ }
+ if (!e) {
ERROR(errp, "Error: could not rdma_bind_addr!");
goto err_dest_init_bind_addr;
- } else {
- ERROR(errp, "migration host and port not specified!");
- ret = -EINVAL;
- goto err_dest_init_bind_addr;
}
-listen:
rdma->listen_id = listen_id;
qemu_rdma_dump_gid("dest_init", listen_id);
if (host_port) {
rdma = g_malloc0(sizeof(RDMAContext));
- memset(rdma, 0, sizeof(RDMAContext));
rdma->current_index = -1;
rdma->current_chunk = -1;
* SEND messages for control only.
* VM's ram is handled with regular RDMA messages.
*/
-static int qemu_rdma_put_buffer(void *opaque, const uint8_t *buf,
- int64_t pos, int size)
+static ssize_t qemu_rdma_put_buffer(void *opaque, const uint8_t *buf,
+ int64_t pos, size_t size)
{
QEMUFileRDMA *r = opaque;
QEMUFile *f = r->file;
r->len = MIN(remaining, RDMA_SEND_INCREMENT);
remaining -= r->len;
- head.len = r->len;
+ /* Guaranteed to fit due to RDMA_SEND_INCREMENT MIN above */
+ head.len = (uint32_t)r->len;
head.type = RDMA_CONTROL_QEMU_FILE;
ret = qemu_rdma_exchange_send(rdma, &head, data, NULL, NULL, NULL);
}
static size_t qemu_rdma_fill(RDMAContext *rdma, uint8_t *buf,
- int size, int idx)
+ size_t size, int idx)
{
size_t len = 0;
* RDMA links don't use bytestreams, so we have to
* return bytes to QEMUFile opportunistically.
*/
-static int qemu_rdma_get_buffer(void *opaque, uint8_t *buf,
- int64_t pos, int size)
+static ssize_t qemu_rdma_get_buffer(void *opaque, uint8_t *buf,
+ int64_t pos, size_t size)
{
QEMUFileRDMA *r = opaque;
RDMAContext *rdma = r->rdma;
*/
static size_t qemu_rdma_save_page(QEMUFile *f, void *opaque,
ram_addr_t block_offset, ram_addr_t offset,
- size_t size, int *bytes_sent)
+ size_t size, uint64_t *bytes_sent)
{
QEMUFileRDMA *rfile = opaque;
RDMAContext *rdma = rfile->rdma;
}
}
- qemu_set_fd_handler2(rdma->channel->fd, NULL, NULL, NULL, NULL);
+ qemu_set_fd_handler(rdma->channel->fd, NULL, NULL, NULL);
ret = rdma_accept(rdma->cm_id, &conn_param);
if (ret) {
return ret;
}
+static int dest_ram_sort_func(const void *a, const void *b)
+{
+ unsigned int a_index = ((const RDMALocalBlock *)a)->src_index;
+ unsigned int b_index = ((const RDMALocalBlock *)b)->src_index;
+
+ return (a_index < b_index) ? -1 : (a_index != b_index);
+}
+
/*
* During each iteration of the migration, we listen for instructions
* by the source VM to perform dynamic page registrations before they
*
* Keep doing this until the source tells us to stop.
*/
-static int qemu_rdma_registration_handle(QEMUFile *f, void *opaque,
- uint64_t flags)
+static int qemu_rdma_registration_handle(QEMUFile *f, void *opaque)
{
RDMAControlHeader reg_resp = { .len = sizeof(RDMARegisterResult),
.type = RDMA_CONTROL_REGISTER_RESULT,
CHECK_ERROR_STATE();
do {
- trace_qemu_rdma_registration_handle_wait(flags);
+ trace_qemu_rdma_registration_handle_wait();
ret = qemu_rdma_exchange_recv(rdma, &head, RDMA_CONTROL_NONE);
trace_qemu_rdma_registration_handle_compress(comp->length,
comp->block_idx,
comp->offset);
+ if (comp->block_idx >= rdma->local_ram_blocks.nb_blocks) {
+ error_report("rdma: 'compress' bad block index %u (vs %d)",
+ (unsigned int)comp->block_idx,
+ rdma->local_ram_blocks.nb_blocks);
+ ret = -EIO;
+ goto out;
+ }
block = &(rdma->local_ram_blocks.block[comp->block_idx]);
host_addr = block->local_host_addr +
case RDMA_CONTROL_RAM_BLOCKS_REQUEST:
trace_qemu_rdma_registration_handle_ram_blocks();
+ /* Sort our local RAM Block list so it's the same as the source,
+ * we can do this since we've filled in a src_index in the list
+ * as we received the RAMBlock list earlier.
+ */
+ qsort(rdma->local_ram_blocks.block,
+ rdma->local_ram_blocks.nb_blocks,
+ sizeof(RDMALocalBlock), dest_ram_sort_func);
if (rdma->pin_all) {
ret = qemu_rdma_reg_whole_ram_blocks(rdma);
if (ret) {
* their "local" descriptions with what was sent.
*/
for (i = 0; i < local->nb_blocks; i++) {
- rdma->block[i].remote_host_addr =
- (uint64_t)(local->block[i].local_host_addr);
+ rdma->dest_blocks[i].remote_host_addr =
+ (uintptr_t)(local->block[i].local_host_addr);
if (rdma->pin_all) {
- rdma->block[i].remote_rkey = local->block[i].mr->rkey;
+ rdma->dest_blocks[i].remote_rkey = local->block[i].mr->rkey;
}
- rdma->block[i].offset = local->block[i].offset;
- rdma->block[i].length = local->block[i].length;
+ rdma->dest_blocks[i].offset = local->block[i].offset;
+ rdma->dest_blocks[i].length = local->block[i].length;
- remote_block_to_network(&rdma->block[i]);
+ dest_block_to_network(&rdma->dest_blocks[i]);
+ trace_qemu_rdma_registration_handle_ram_blocks_loop(
+ local->block[i].block_name,
+ local->block[i].offset,
+ local->block[i].length,
+ local->block[i].local_host_addr,
+ local->block[i].src_index);
}
blocks.len = rdma->local_ram_blocks.nb_blocks
- * sizeof(RDMARemoteBlock);
+ * sizeof(RDMADestBlock);
ret = qemu_rdma_post_send_control(rdma,
- (uint8_t *) rdma->block, &blocks);
+ (uint8_t *) rdma->dest_blocks, &blocks);
if (ret < 0) {
error_report("rdma migration: error sending remote info");
trace_qemu_rdma_registration_handle_register_loop(count,
reg->current_index, reg->key.current_addr, reg->chunks);
+ if (reg->current_index >= rdma->local_ram_blocks.nb_blocks) {
+ error_report("rdma: 'register' bad block index %u (vs %d)",
+ (unsigned int)reg->current_index,
+ rdma->local_ram_blocks.nb_blocks);
+ ret = -ENOENT;
+ goto out;
+ }
block = &(rdma->local_ram_blocks.block[reg->current_index]);
if (block->is_ram_block) {
+ if (block->offset > reg->key.current_addr) {
+ error_report("rdma: bad register address for block %s"
+ " offset: %" PRIx64 " current_addr: %" PRIx64,
+ block->block_name, block->offset,
+ reg->key.current_addr);
+ ret = -ERANGE;
+ goto out;
+ }
host_addr = (block->local_host_addr +
(reg->key.current_addr - block->offset));
chunk = ram_chunk_index(block->local_host_addr,
chunk = reg->key.chunk;
host_addr = block->local_host_addr +
(reg->key.chunk * (1UL << RDMA_REG_CHUNK_SHIFT));
+ /* Check for particularly bad chunk value */
+ if (host_addr < (void *)block->local_host_addr) {
+ error_report("rdma: bad chunk for block %s"
+ " chunk: %" PRIx64,
+ block->block_name, reg->key.chunk);
+ ret = -ERANGE;
+ goto out;
+ }
}
chunk_start = ram_chunk_start(block, chunk);
chunk_end = ram_chunk_end(block, chunk + reg->chunks);
if (qemu_rdma_register_and_get_keys(rdma, block,
- (uint8_t *)host_addr, NULL, ®_result->rkey,
+ (uintptr_t)host_addr, NULL, ®_result->rkey,
chunk, chunk_start, chunk_end)) {
error_report("cannot get rkey");
ret = -EINVAL;
goto out;
}
- reg_result->host_addr = (uint64_t) block->local_host_addr;
+ reg_result->host_addr = (uintptr_t)block->local_host_addr;
trace_qemu_rdma_registration_handle_register_rkey(
reg_result->rkey);
return ret;
}
+/* Destination:
+ * Called via a ram_control_load_hook during the initial RAM load section which
+ * lists the RAMBlocks by name. This lets us know the order of the RAMBlocks
+ * on the source.
+ * We've already built our local RAMBlock list, but not yet sent the list to
+ * the source.
+ */
+static int rdma_block_notification_handle(QEMUFileRDMA *rfile, const char *name)
+{
+ RDMAContext *rdma = rfile->rdma;
+ int curr;
+ int found = -1;
+
+ /* Find the matching RAMBlock in our local list */
+ for (curr = 0; curr < rdma->local_ram_blocks.nb_blocks; curr++) {
+ if (!strcmp(rdma->local_ram_blocks.block[curr].block_name, name)) {
+ found = curr;
+ break;
+ }
+ }
+
+ if (found == -1) {
+ error_report("RAMBlock '%s' not found on destination", name);
+ return -ENOENT;
+ }
+
+ rdma->local_ram_blocks.block[curr].src_index = rdma->next_src_index;
+ trace_rdma_block_notification_handle(name, rdma->next_src_index);
+ rdma->next_src_index++;
+
+ return 0;
+}
+
+static int rdma_load_hook(QEMUFile *f, void *opaque, uint64_t flags, void *data)
+{
+ switch (flags) {
+ case RAM_CONTROL_BLOCK_REG:
+ return rdma_block_notification_handle(opaque, data);
+
+ case RAM_CONTROL_HOOK:
+ return qemu_rdma_registration_handle(f, opaque);
+
+ default:
+ /* Shouldn't be called with any other values */
+ abort();
+ }
+}
+
static int qemu_rdma_registration_start(QEMUFile *f, void *opaque,
- uint64_t flags)
+ uint64_t flags, void *data)
{
QEMUFileRDMA *rfile = opaque;
RDMAContext *rdma = rfile->rdma;
* First, flush writes, if any.
*/
static int qemu_rdma_registration_stop(QEMUFile *f, void *opaque,
- uint64_t flags)
+ uint64_t flags, void *data)
{
Error *local_err = NULL, **errp = &local_err;
QEMUFileRDMA *rfile = opaque;
if (flags == RAM_CONTROL_SETUP) {
RDMAControlHeader resp = {.type = RDMA_CONTROL_RAM_BLOCKS_RESULT };
RDMALocalBlocks *local = &rdma->local_ram_blocks;
- int reg_result_idx, i, j, nb_remote_blocks;
+ int reg_result_idx, i, nb_dest_blocks;
head.type = RDMA_CONTROL_RAM_BLOCKS_REQUEST;
trace_qemu_rdma_registration_stop_ram();
return ret;
}
- nb_remote_blocks = resp.len / sizeof(RDMARemoteBlock);
+ nb_dest_blocks = resp.len / sizeof(RDMADestBlock);
/*
* The protocol uses two different sets of rkeys (mutually exclusive):
* and then propagates the remote ram block descriptions to his local copy.
*/
- if (local->nb_blocks != nb_remote_blocks) {
- ERROR(errp, "ram blocks mismatch #1! "
+ if (local->nb_blocks != nb_dest_blocks) {
+ ERROR(errp, "ram blocks mismatch (Number of blocks %d vs %d) "
"Your QEMU command line parameters are probably "
- "not identical on both the source and destination.");
+ "not identical on both the source and destination.",
+ local->nb_blocks, nb_dest_blocks);
+ rdma->error_state = -EINVAL;
return -EINVAL;
}
qemu_rdma_move_header(rdma, reg_result_idx, &resp);
- memcpy(rdma->block,
+ memcpy(rdma->dest_blocks,
rdma->wr_data[reg_result_idx].control_curr, resp.len);
- for (i = 0; i < nb_remote_blocks; i++) {
- network_to_remote_block(&rdma->block[i]);
-
- /* search local ram blocks */
- for (j = 0; j < local->nb_blocks; j++) {
- if (rdma->block[i].offset != local->block[j].offset) {
- continue;
- }
-
- if (rdma->block[i].length != local->block[j].length) {
- ERROR(errp, "ram blocks mismatch #2! "
- "Your QEMU command line parameters are probably "
- "not identical on both the source and destination.");
- return -EINVAL;
- }
- local->block[j].remote_host_addr =
- rdma->block[i].remote_host_addr;
- local->block[j].remote_rkey = rdma->block[i].remote_rkey;
- break;
- }
-
- if (j >= local->nb_blocks) {
- ERROR(errp, "ram blocks mismatch #3! "
- "Your QEMU command line parameters are probably "
- "not identical on both the source and destination.");
+ for (i = 0; i < nb_dest_blocks; i++) {
+ network_to_dest_block(&rdma->dest_blocks[i]);
+
+ /* We require that the blocks are in the same order */
+ if (rdma->dest_blocks[i].length != local->block[i].length) {
+ ERROR(errp, "Block %s/%d has a different length %" PRIu64
+ "vs %" PRIu64, local->block[i].block_name, i,
+ local->block[i].length,
+ rdma->dest_blocks[i].length);
+ rdma->error_state = -EINVAL;
return -EINVAL;
}
+ local->block[i].remote_host_addr =
+ rdma->dest_blocks[i].remote_host_addr;
+ local->block[i].remote_rkey = rdma->dest_blocks[i].remote_rkey;
}
}
.get_buffer = qemu_rdma_get_buffer,
.get_fd = qemu_rdma_get_fd,
.close = qemu_rdma_close,
- .hook_ram_load = qemu_rdma_registration_handle,
+ .hook_ram_load = rdma_load_hook,
};
static const QEMUFileOps rdma_write_ops = {
static void *qemu_fopen_rdma(RDMAContext *rdma, const char *mode)
{
- QEMUFileRDMA *r = g_malloc0(sizeof(QEMUFileRDMA));
+ QEMUFileRDMA *r;
if (qemu_file_mode_is_not_valid(mode)) {
return NULL;
}
+ r = g_malloc0(sizeof(QEMUFileRDMA));
r->rdma = rdma;
if (mode[0] == 'w') {
QEMUFile *f;
Error *local_err = NULL, **errp = &local_err;
- trace_qemu_dma_accept_incoming_migration();
+ trace_qemu_rdma_accept_incoming_migration();
ret = qemu_rdma_accept(rdma);
if (ret) {
return;
}
- trace_qemu_dma_accept_incoming_migration_accepted();
+ trace_qemu_rdma_accept_incoming_migration_accepted();
f = qemu_fopen_rdma(rdma, "rb");
if (f == NULL) {
trace_rdma_start_incoming_migration_after_rdma_listen();
- qemu_set_fd_handler2(rdma->channel->fd, NULL,
- rdma_accept_incoming_migration, NULL,
- (void *)(intptr_t) rdma);
+ qemu_set_fd_handler(rdma->channel->fd, rdma_accept_incoming_migration,
+ NULL, (void *)(intptr_t)rdma);
return;
err:
error_propagate(errp, local_err);
projects / qemu.git / blobdiff