* [Qemu-devel] [PATCH v4 1/2] sheepdog: refactor do_sd_create()
2013-10-31 5:49 [Qemu-devel] [PATCH v4 0/2] sheepdog: add user-defined redundancy option Liu Yuan
@ 2013-10-31 5:49 ` Liu Yuan
2013-10-31 5:49 ` [Qemu-devel] [PATCH v4 2/2] sheepdog: support user-defined redundancy option Liu Yuan
2013-11-01 8:35 ` [Qemu-devel] [PATCH v4 0/2] sheepdog: add " Liu Yuan
2 siblings, 0 replies; 5+ messages in thread
From: Liu Yuan @ 2013-10-31 5:49 UTC (permalink / raw)
To: qemu-devel; +Cc: Kevin Wolf, sheepdog, Stefan Hajnoczi
We can actually use BDRVSheepdogState *s to pass most of the parameters.
Cc: Kevin Wolf <kwolf@redhat.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Benoit Canet <benoit@irqsave.net>
Signed-off-by: Liu Yuan <namei.unix@gmail.com>
---
block/sheepdog.c | 37 +++++++++++++++----------------------
1 file changed, 15 insertions(+), 22 deletions(-)
diff --git a/block/sheepdog.c b/block/sheepdog.c
index ef387de..66b3ea8 100644
--- a/block/sheepdog.c
+++ b/block/sheepdog.c
@@ -1464,9 +1464,7 @@ 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 +1481,11 @@ 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 +1493,8 @@ 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;
ret = do_req(fd, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
@@ -1507,7 +1505,7 @@ 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;
}
@@ -1568,23 +1566,21 @@ 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 +1588,7 @@ 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)) {
@@ -1610,7 +1606,7 @@ static int sd_create(const char *filename, QEMUOptionParameter *options,
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 +1641,11 @@ 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 +1828,7 @@ 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 +2091,7 @@ 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));
--
1.7.9.5
^ permalink raw reply related [flat|nested] 5+ messages in thread
* [Qemu-devel] [PATCH v4 2/2] sheepdog: support user-defined redundancy option
2013-10-31 5:49 [Qemu-devel] [PATCH v4 0/2] sheepdog: add user-defined redundancy option Liu Yuan
2013-10-31 5:49 ` [Qemu-devel] [PATCH v4 1/2] sheepdog: refactor do_sd_create() Liu Yuan
@ 2013-10-31 5:49 ` Liu Yuan
2013-11-01 14:20 ` [Qemu-devel] [sheepdog] " MORITA Kazutaka
2013-11-01 8:35 ` [Qemu-devel] [PATCH v4 0/2] sheepdog: add " Liu Yuan
2 siblings, 1 reply; 5+ messages in thread
From: Liu Yuan @ 2013-10-31 5:49 UTC (permalink / raw)
To: qemu-devel; +Cc: Kevin Wolf, sheepdog, Stefan Hajnoczi
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)
E.g, to convert a vdi into sheepdog vdi 'test' with 8:3 erasure coding scheme
$ qemu-img convert -o redundancy=8:3 linux-0.2.img sheepdog:test
Cc: Kevin Wolf <kwolf@redhat.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Benoit Canet <benoit@irqsave.net>
Signed-off-by: Liu Yuan <namei.unix@gmail.com>
---
block/sheepdog.c | 90 ++++++++++++++++++++++++++++++++++++++++++++-
include/block/block_int.h | 1 +
2 files changed, 90 insertions(+), 1 deletion(-)
diff --git a/block/sheepdog.c b/block/sheepdog.c
index 66b3ea8..6f5a523 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 trips 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
@@ -1495,6 +1503,7 @@ static int do_sd_create(BDRVSheepdogState *s, uint32_t *vdi_id, int snapshot)
hdr.data_length = wlen;
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);
@@ -1562,6 +1571,76 @@ out:
return ret;
}
+static bool is_numeric(const char *s)
+{
+ const char *p = s;
+
+ if (*p) {
+ char c;
+
+ while ((c = *p++))
+ if (!isdigit(c)) {
+ return false;
+ }
+ return true;
+ }
+ return false;
+}
+
+/*
+ * 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;
+ uint8_t copy, parity;
+ char p[10];
+
+ strncpy(p, opt, sizeof(p));
+ n1 = strtok(p, ":");
+ n2 = strtok(NULL, ":");
+
+ if ((n1 && !is_numeric(n1)) || (n2 && !is_numeric(n2))) {
+ return -EINVAL;
+ }
+
+ copy = strtol(n1, NULL, 10);
+ if (copy > SD_MAX_COPIES) {
+ 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 == 0) {
+ 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)
{
@@ -1602,6 +1681,11 @@ 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++;
}
@@ -1644,7 +1728,6 @@ static int sd_create(const char *filename, QEMUOptionParameter *options,
bdrv_unref(bs);
}
- /* TODO: allow users to specify copy number */
ret = do_sd_create(s, &vid, 0);
if (!prealloc || ret) {
goto out;
@@ -2416,6 +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 }
};
diff --git a/include/block/block_int.h b/include/block/block_int.h
index a48731d..b90862f 100644
--- a/include/block/block_int.h
+++ b/include/block/block_int.h
@@ -53,6 +53,7 @@
#define BLOCK_OPT_COMPAT_LEVEL "compat"
#define BLOCK_OPT_LAZY_REFCOUNTS "lazy_refcounts"
#define BLOCK_OPT_ADAPTER_TYPE "adapter_type"
+#define BLOCK_OPT_REDUNDANCY "redundancy"
typedef struct BdrvTrackedRequest {
BlockDriverState *bs;
--
1.7.9.5
^ permalink raw reply related [flat|nested] 5+ messages in thread