summaryrefslogtreecommitdiffstats
path: root/block
diff options
context:
space:
mode:
authorStefan Hajnoczi <stefanha@linux.vnet.ibm.com>2012-07-27 09:05:22 +0100
committerKevin Wolf <kwolf@redhat.com>2012-08-06 22:39:14 +0200
commitbfe8043e9214d2fc6572cc72b5f2218308747acd (patch)
tree40d56873d87524c6b4b062f43851e44ededcae9c /block
parent91cf8a35e777cdbdc3bf0d39c319ac9f76b7a296 (diff)
downloadhqemu-bfe8043e9214d2fc6572cc72b5f2218308747acd.zip
hqemu-bfe8043e9214d2fc6572cc72b5f2218308747acd.tar.gz
qcow2: implement lazy refcounts
Lazy refcounts is a performance optimization for qcow2 that postpones refcount metadata updates and instead marks the image dirty. In the case of crash or power failure the image will be left in a dirty state and repaired next time it is opened. Reducing metadata I/O is important for cache=writethrough and cache=directsync because these modes guarantee that data is on disk after each write (hence we cannot take advantage of caching updates in RAM). Refcount metadata is not needed for guest->file block address translation and therefore does not need to be on-disk at the time of write completion - this is the motivation behind the lazy refcount optimization. The lazy refcount optimization must be enabled at image creation time: qemu-img create -f qcow2 -o compat=1.1,lazy_refcounts=on a.qcow2 10G qemu-system-x86_64 -drive if=virtio,file=a.qcow2,cache=writethrough Update qemu-iotests 031 and 036 since the extension header size changes when we add feature bit table entries. Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Diffstat (limited to 'block')
-rw-r--r--block/qcow2-cluster.c5
-rw-r--r--block/qcow2.c73
-rw-r--r--block/qcow2.h13
3 files changed, 86 insertions, 5 deletions
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index d7e0e19..e179211 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -662,7 +662,10 @@ int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, QCowL2Meta *m)
qcow2_cache_depends_on_flush(s->l2_table_cache);
}
- qcow2_cache_set_dependency(bs, s->l2_table_cache, s->refcount_block_cache);
+ if (qcow2_need_accurate_refcounts(s)) {
+ qcow2_cache_set_dependency(bs, s->l2_table_cache,
+ s->refcount_block_cache);
+ }
ret = get_cluster_table(bs, m->offset, &l2_table, &l2_index);
if (ret < 0) {
goto err;
diff --git a/block/qcow2.c b/block/qcow2.c
index 7fe1567..fd5e214 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -215,6 +215,41 @@ static void report_unsupported_feature(BlockDriverState *bs,
}
/*
+ * Sets the dirty bit and flushes afterwards if necessary.
+ *
+ * The incompatible_features bit is only set if the image file header was
+ * updated successfully. Therefore it is not required to check the return
+ * value of this function.
+ */
+static int qcow2_mark_dirty(BlockDriverState *bs)
+{
+ BDRVQcowState *s = bs->opaque;
+ uint64_t val;
+ int ret;
+
+ assert(s->qcow_version >= 3);
+
+ if (s->incompatible_features & QCOW2_INCOMPAT_DIRTY) {
+ return 0; /* already dirty */
+ }
+
+ val = cpu_to_be64(s->incompatible_features | QCOW2_INCOMPAT_DIRTY);
+ ret = bdrv_pwrite(bs->file, offsetof(QCowHeader, incompatible_features),
+ &val, sizeof(val));
+ if (ret < 0) {
+ return ret;
+ }
+ ret = bdrv_flush(bs->file);
+ if (ret < 0) {
+ return ret;
+ }
+
+ /* Only treat image as dirty if the header was updated successfully */
+ s->incompatible_features |= QCOW2_INCOMPAT_DIRTY;
+ return 0;
+}
+
+/*
* Clears the dirty bit and flushes before if necessary. Only call this
* function when there are no pending requests, it does not guard against
* concurrent requests dirtying the image.
@@ -752,6 +787,11 @@ static coroutine_fn int qcow2_co_writev(BlockDriverState *bs,
goto fail;
}
+ if (l2meta.nb_clusters > 0 &&
+ (s->compatible_features & QCOW2_COMPAT_LAZY_REFCOUNTS)) {
+ qcow2_mark_dirty(bs);
+ }
+
cluster_offset = l2meta.cluster_offset;
assert((cluster_offset & 511) == 0);
@@ -994,6 +1034,11 @@ int qcow2_update_header(BlockDriverState *bs)
.bit = QCOW2_INCOMPAT_DIRTY_BITNR,
.name = "dirty bit",
},
+ {
+ .type = QCOW2_FEAT_TYPE_COMPATIBLE,
+ .bit = QCOW2_COMPAT_LAZY_REFCOUNTS_BITNR,
+ .name = "lazy refcounts",
+ },
};
ret = header_ext_add(buf, QCOW2_EXT_MAGIC_FEATURE_TABLE,
@@ -1176,6 +1221,11 @@ static int qcow2_create2(const char *filename, int64_t total_size,
header.crypt_method = cpu_to_be32(QCOW_CRYPT_NONE);
}
+ if (flags & BLOCK_FLAG_LAZY_REFCOUNTS) {
+ header.compatible_features |=
+ cpu_to_be64(QCOW2_COMPAT_LAZY_REFCOUNTS);
+ }
+
ret = bdrv_pwrite(bs, 0, &header, sizeof(header));
if (ret < 0) {
goto out;
@@ -1289,6 +1339,8 @@ static int qcow2_create(const char *filename, QEMUOptionParameter *options)
options->value.s);
return -EINVAL;
}
+ } else if (!strcmp(options->name, BLOCK_OPT_LAZY_REFCOUNTS)) {
+ flags |= options->value.n ? BLOCK_FLAG_LAZY_REFCOUNTS : 0;
}
options++;
}
@@ -1299,6 +1351,12 @@ static int qcow2_create(const char *filename, QEMUOptionParameter *options)
return -EINVAL;
}
+ if (version < 3 && (flags & BLOCK_FLAG_LAZY_REFCOUNTS)) {
+ fprintf(stderr, "Lazy refcounts only supported with compatibility "
+ "level 1.1 and above (use compat=1.1 or greater)\n");
+ return -EINVAL;
+ }
+
return qcow2_create2(filename, sectors, backing_file, backing_fmt, flags,
cluster_size, prealloc, options, version);
}
@@ -1485,10 +1543,12 @@ static coroutine_fn int qcow2_co_flush_to_os(BlockDriverState *bs)
return ret;
}
- ret = qcow2_cache_flush(bs, s->refcount_block_cache);
- if (ret < 0) {
- qemu_co_mutex_unlock(&s->lock);
- return ret;
+ if (qcow2_need_accurate_refcounts(s)) {
+ ret = qcow2_cache_flush(bs, s->refcount_block_cache);
+ if (ret < 0) {
+ qemu_co_mutex_unlock(&s->lock);
+ return ret;
+ }
}
qemu_co_mutex_unlock(&s->lock);
@@ -1603,6 +1663,11 @@ static QEMUOptionParameter qcow2_create_options[] = {
.type = OPT_STRING,
.help = "Preallocation mode (allowed values: off, metadata)"
},
+ {
+ .name = BLOCK_OPT_LAZY_REFCOUNTS,
+ .type = OPT_FLAG,
+ .help = "Postpone refcount updates",
+ },
{ NULL }
};
diff --git a/block/qcow2.h b/block/qcow2.h
index b5fefc0..b4eb654 100644
--- a/block/qcow2.h
+++ b/block/qcow2.h
@@ -118,6 +118,14 @@ enum {
QCOW2_INCOMPAT_MASK = QCOW2_INCOMPAT_DIRTY,
};
+/* Compatible feature bits */
+enum {
+ QCOW2_COMPAT_LAZY_REFCOUNTS_BITNR = 0,
+ QCOW2_COMPAT_LAZY_REFCOUNTS = 1 << QCOW2_COMPAT_LAZY_REFCOUNTS_BITNR,
+
+ QCOW2_COMPAT_FEAT_MASK = QCOW2_COMPAT_LAZY_REFCOUNTS,
+};
+
typedef struct Qcow2Feature {
uint8_t type;
uint8_t bit;
@@ -245,6 +253,11 @@ static inline int qcow2_get_cluster_type(uint64_t l2_entry)
}
}
+/* Check whether refcounts are eager or lazy */
+static inline bool qcow2_need_accurate_refcounts(BDRVQcowState *s)
+{
+ return !(s->incompatible_features & QCOW2_INCOMPAT_DIRTY);
+}
// FIXME Need qcow2_ prefix to global functions
OpenPOWER on IntegriCloud