aboutsummaryrefslogtreecommitdiff
path: root/block/sheepdog.c
diff options
context:
space:
mode:
Diffstat (limited to 'block/sheepdog.c')
-rw-r--r--block/sheepdog.c130
1 files changed, 107 insertions, 23 deletions
diff --git a/block/sheepdog.c b/block/sheepdog.c
index ef387de71f..b4ae50f44d 100644
--- a/block/sheepdog.c
+++ b/block/sheepdog.c
@@ -91,6 +91,14 @@
#define SD_NR_VDIS (1U << 24)
#define SD_DATA_OBJ_SIZE (UINT64_C(1) << 22)
#define SD_MAX_VDI_SIZE (SD_DATA_OBJ_SIZE * MAX_DATA_OBJS)
+/*
+ * For erasure coding, we use at most SD_EC_MAX_STRIP for data strips and
+ * (SD_EC_MAX_STRIP - 1) for parity strips
+ *
+ * SD_MAX_COPIES is sum of number of data strips and parity strips.
+ */
+#define SD_EC_MAX_STRIP 16
+#define SD_MAX_COPIES (SD_EC_MAX_STRIP * 2 - 1)
#define SD_INODE_SIZE (sizeof(SheepdogInode))
#define CURRENT_VDI_ID 0
@@ -1464,9 +1472,7 @@ out:
return ret;
}
-static int do_sd_create(BDRVSheepdogState *s, char *filename, int64_t vdi_size,
- uint32_t base_vid, uint32_t *vdi_id, int snapshot,
- uint8_t copy_policy)
+static int do_sd_create(BDRVSheepdogState *s, uint32_t *vdi_id, int snapshot)
{
SheepdogVdiReq hdr;
SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr;
@@ -1483,11 +1489,11 @@ static int do_sd_create(BDRVSheepdogState *s, char *filename, int64_t vdi_size,
* does not fit in buf? For now, just truncate and avoid buffer overrun.
*/
memset(buf, 0, sizeof(buf));
- pstrcpy(buf, sizeof(buf), filename);
+ pstrcpy(buf, sizeof(buf), s->name);
memset(&hdr, 0, sizeof(hdr));
hdr.opcode = SD_OP_NEW_VDI;
- hdr.vdi_id = base_vid;
+ hdr.vdi_id = s->inode.vdi_id;
wlen = SD_MAX_VDI_LEN;
@@ -1495,8 +1501,9 @@ static int do_sd_create(BDRVSheepdogState *s, char *filename, int64_t vdi_size,
hdr.snapid = snapshot;
hdr.data_length = wlen;
- hdr.vdi_size = vdi_size;
- hdr.copy_policy = copy_policy;
+ hdr.vdi_size = s->inode.vdi_size;
+ hdr.copy_policy = s->inode.copy_policy;
+ hdr.copies = s->inode.nr_copies;
ret = do_req(fd, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
@@ -1507,7 +1514,7 @@ static int do_sd_create(BDRVSheepdogState *s, char *filename, int64_t vdi_size,
}
if (rsp->result != SD_RES_SUCCESS) {
- error_report("%s, %s", sd_strerror(rsp->result), filename);
+ error_report("%s, %s", sd_strerror(rsp->result), s->inode.name);
return -EIO;
}
@@ -1564,27 +1571,79 @@ out:
return ret;
}
+/*
+ * Sheepdog support two kinds of redundancy, full replication and erasure
+ * coding.
+ *
+ * # create a fully replicated vdi with x copies
+ * -o redundancy=x (1 <= x <= SD_MAX_COPIES)
+ *
+ * # create a erasure coded vdi with x data strips and y parity strips
+ * -o redundancy=x:y (x must be one of {2,4,8,16} and 1 <= y < SD_EC_MAX_STRIP)
+ */
+static int parse_redundancy(BDRVSheepdogState *s, const char *opt)
+{
+ struct SheepdogInode *inode = &s->inode;
+ const char *n1, *n2;
+ long copy, parity;
+ char p[10];
+
+ pstrcpy(p, sizeof(p), opt);
+ n1 = strtok(p, ":");
+ n2 = strtok(NULL, ":");
+
+ if (!n1) {
+ return -EINVAL;
+ }
+
+ copy = strtol(n1, NULL, 10);
+ if (copy > SD_MAX_COPIES || copy < 1) {
+ return -EINVAL;
+ }
+ if (!n2) {
+ inode->copy_policy = 0;
+ inode->nr_copies = copy;
+ return 0;
+ }
+
+ if (copy != 2 && copy != 4 && copy != 8 && copy != 16) {
+ return -EINVAL;
+ }
+
+ parity = strtol(n2, NULL, 10);
+ if (parity >= SD_EC_MAX_STRIP || parity < 1) {
+ return -EINVAL;
+ }
+
+ /*
+ * 4 bits for parity and 4 bits for data.
+ * We have to compress upper data bits because it can't represent 16
+ */
+ inode->copy_policy = ((copy / 2) << 4) + parity;
+ inode->nr_copies = copy + parity;
+
+ return 0;
+}
+
static int sd_create(const char *filename, QEMUOptionParameter *options,
Error **errp)
{
int ret = 0;
- uint32_t vid = 0, base_vid = 0;
- int64_t vdi_size = 0;
+ uint32_t vid = 0;
char *backing_file = NULL;
BDRVSheepdogState *s;
- char vdi[SD_MAX_VDI_LEN], tag[SD_MAX_VDI_TAG_LEN];
+ char tag[SD_MAX_VDI_TAG_LEN];
uint32_t snapid;
bool prealloc = false;
Error *local_err = NULL;
s = g_malloc0(sizeof(BDRVSheepdogState));
- memset(vdi, 0, sizeof(vdi));
memset(tag, 0, sizeof(tag));
if (strstr(filename, "://")) {
- ret = sd_parse_uri(s, filename, vdi, &snapid, tag);
+ ret = sd_parse_uri(s, filename, s->name, &snapid, tag);
} else {
- ret = parse_vdiname(s, filename, vdi, &snapid, tag);
+ ret = parse_vdiname(s, filename, s->name, &snapid, tag);
}
if (ret < 0) {
goto out;
@@ -1592,7 +1651,7 @@ static int sd_create(const char *filename, QEMUOptionParameter *options,
while (options && options->name) {
if (!strcmp(options->name, BLOCK_OPT_SIZE)) {
- vdi_size = options->value.n;
+ s->inode.vdi_size = options->value.n;
} else if (!strcmp(options->name, BLOCK_OPT_BACKING_FILE)) {
backing_file = options->value.s;
} else if (!strcmp(options->name, BLOCK_OPT_PREALLOC)) {
@@ -1606,11 +1665,16 @@ static int sd_create(const char *filename, QEMUOptionParameter *options,
ret = -EINVAL;
goto out;
}
+ } else if (!strcmp(options->name, BLOCK_OPT_REDUNDANCY)) {
+ ret = parse_redundancy(s, options->value.s);
+ if (ret < 0) {
+ goto out;
+ }
}
options++;
}
- if (vdi_size > SD_MAX_VDI_SIZE) {
+ if (s->inode.vdi_size > SD_MAX_VDI_SIZE) {
error_report("too big image size");
ret = -EINVAL;
goto out;
@@ -1645,12 +1709,10 @@ static int sd_create(const char *filename, QEMUOptionParameter *options,
goto out;
}
- base_vid = s->inode.vdi_id;
bdrv_unref(bs);
}
- /* TODO: allow users to specify copy number */
- ret = do_sd_create(s, vdi, vdi_size, base_vid, &vid, 0, 0);
+ ret = do_sd_create(s, &vid, 0);
if (!prealloc || ret) {
goto out;
}
@@ -1833,8 +1895,7 @@ static int sd_create_branch(BDRVSheepdogState *s)
* false bail out.
*/
deleted = sd_delete(s);
- ret = do_sd_create(s, s->name, s->inode.vdi_size, s->inode.vdi_id, &vid,
- !deleted, s->inode.copy_policy);
+ ret = do_sd_create(s, &vid, !deleted);
if (ret) {
goto out;
}
@@ -2097,8 +2158,7 @@ static int sd_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
goto cleanup;
}
- ret = do_sd_create(s, s->name, s->inode.vdi_size, s->inode.vdi_id, &new_vid,
- 1, s->inode.copy_policy);
+ ret = do_sd_create(s, &new_vid, 1);
if (ret < 0) {
error_report("failed to create inode for snapshot. %s",
strerror(errno));
@@ -2407,6 +2467,22 @@ sd_co_get_block_status(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
return ret;
}
+static int64_t sd_get_allocated_file_size(BlockDriverState *bs)
+{
+ BDRVSheepdogState *s = bs->opaque;
+ SheepdogInode *inode = &s->inode;
+ unsigned long i, last = DIV_ROUND_UP(inode->vdi_size, SD_DATA_OBJ_SIZE);
+ uint64_t size = 0;
+
+ for (i = 0; i < last; i++) {
+ if (inode->data_vdi_id[i] == 0) {
+ continue;
+ }
+ size += SD_DATA_OBJ_SIZE;
+ }
+ return size;
+}
+
static QEMUOptionParameter sd_create_options[] = {
{
.name = BLOCK_OPT_SIZE,
@@ -2423,6 +2499,11 @@ static QEMUOptionParameter sd_create_options[] = {
.type = OPT_STRING,
.help = "Preallocation mode (allowed values: off, full)"
},
+ {
+ .name = BLOCK_OPT_REDUNDANCY,
+ .type = OPT_STRING,
+ .help = "Redundancy of the image"
+ },
{ NULL }
};
@@ -2436,6 +2517,7 @@ static BlockDriver bdrv_sheepdog = {
.bdrv_create = sd_create,
.bdrv_has_zero_init = bdrv_has_zero_init_1,
.bdrv_getlength = sd_getlength,
+ .bdrv_get_allocated_file_size = sd_get_allocated_file_size,
.bdrv_truncate = sd_truncate,
.bdrv_co_readv = sd_co_readv,
@@ -2465,6 +2547,7 @@ static BlockDriver bdrv_sheepdog_tcp = {
.bdrv_create = sd_create,
.bdrv_has_zero_init = bdrv_has_zero_init_1,
.bdrv_getlength = sd_getlength,
+ .bdrv_get_allocated_file_size = sd_get_allocated_file_size,
.bdrv_truncate = sd_truncate,
.bdrv_co_readv = sd_co_readv,
@@ -2494,6 +2577,7 @@ static BlockDriver bdrv_sheepdog_unix = {
.bdrv_create = sd_create,
.bdrv_has_zero_init = bdrv_has_zero_init_1,
.bdrv_getlength = sd_getlength,
+ .bdrv_get_allocated_file_size = sd_get_allocated_file_size,
.bdrv_truncate = sd_truncate,
.bdrv_co_readv = sd_co_readv,