}
#ifdef DEBUG_ALLOC2
- printf("grow l1_table from %d to %d\n", s->l1_size, new_l1_size);
+ fprintf(stderr, "grow l1_table from %d to %d\n", s->l1_size, new_l1_size);
#endif
new_l1_size2 = sizeof(uint64_t) * new_l1_size;
- new_l1_table = qemu_mallocz(align_offset(new_l1_size2, 512));
+ new_l1_table = g_malloc0(align_offset(new_l1_size2, 512));
memcpy(new_l1_table, s->l1_table, s->l1_size * sizeof(uint64_t));
/* write new table (align to cluster) */
BLKDBG_EVENT(bs->file, BLKDBG_L1_GROW_ALLOC_TABLE);
new_l1_table_offset = qcow2_alloc_clusters(bs, new_l1_size2);
if (new_l1_table_offset < 0) {
- qemu_free(new_l1_table);
+ g_free(new_l1_table);
return new_l1_table_offset;
}
- bdrv_flush(bs->file);
+
+ ret = qcow2_cache_flush(bs, s->refcount_block_cache);
+ if (ret < 0) {
+ goto fail;
+ }
BLKDBG_EVENT(bs->file, BLKDBG_L1_GROW_WRITE_TABLE);
for(i = 0; i < s->l1_size; i++)
/* set new table */
BLKDBG_EVENT(bs->file, BLKDBG_L1_GROW_ACTIVATE_TABLE);
cpu_to_be32w((uint32_t*)data, new_l1_size);
- cpu_to_be64w((uint64_t*)(data + 4), new_l1_table_offset);
+ cpu_to_be64wu((uint64_t*)(data + 4), new_l1_table_offset);
ret = bdrv_pwrite_sync(bs->file, offsetof(QCowHeader, l1_size), data,sizeof(data));
if (ret < 0) {
goto fail;
}
- qemu_free(s->l1_table);
+ g_free(s->l1_table);
qcow2_free_clusters(bs, s->l1_table_offset, s->l1_size * sizeof(uint64_t));
s->l1_table_offset = new_l1_table_offset;
s->l1_table = new_l1_table;
s->l1_size = new_l1_size;
return 0;
fail:
- qemu_free(new_l1_table);
+ g_free(new_l1_table);
qcow2_free_clusters(bs, new_l1_table_offset, new_l1_size2);
return ret;
}
-void qcow2_l2_cache_reset(BlockDriverState *bs)
-{
- BDRVQcowState *s = bs->opaque;
-
- memset(s->l2_cache, 0, s->l2_size * L2_CACHE_SIZE * sizeof(uint64_t));
- memset(s->l2_cache_offsets, 0, L2_CACHE_SIZE * sizeof(uint64_t));
- memset(s->l2_cache_counts, 0, L2_CACHE_SIZE * sizeof(uint32_t));
-}
-
-static inline int l2_cache_new_entry(BlockDriverState *bs)
-{
- BDRVQcowState *s = bs->opaque;
- uint32_t min_count;
- int min_index, i;
-
- /* find a new entry in the least used one */
- min_index = 0;
- min_count = 0xffffffff;
- for(i = 0; i < L2_CACHE_SIZE; i++) {
- if (s->l2_cache_counts[i] < min_count) {
- min_count = s->l2_cache_counts[i];
- min_index = i;
- }
- }
- return min_index;
-}
-
-/*
- * seek_l2_table
- *
- * seek l2_offset in the l2_cache table
- * if not found, return NULL,
- * if found,
- * increments the l2 cache hit count of the entry,
- * if counter overflow, divide by two all counters
- * return the pointer to the l2 cache entry
- *
- */
-
-static uint64_t *seek_l2_table(BDRVQcowState *s, uint64_t l2_offset)
-{
- int i, j;
-
- for(i = 0; i < L2_CACHE_SIZE; i++) {
- if (l2_offset == s->l2_cache_offsets[i]) {
- /* increment the hit count */
- if (++s->l2_cache_counts[i] == 0xffffffff) {
- for(j = 0; j < L2_CACHE_SIZE; j++) {
- s->l2_cache_counts[j] >>= 1;
- }
- }
- return s->l2_cache + (i << s->l2_bits);
- }
- }
- return NULL;
-}
-
/*
* l2_load
*
uint64_t **l2_table)
{
BDRVQcowState *s = bs->opaque;
- int min_index;
int ret;
- /* seek if the table for the given offset is in the cache */
-
- *l2_table = seek_l2_table(s, l2_offset);
- if (*l2_table != NULL) {
- return 0;
- }
-
- /* not found: load a new entry in the least used one */
-
- min_index = l2_cache_new_entry(bs);
- *l2_table = s->l2_cache + (min_index << s->l2_bits);
-
- BLKDBG_EVENT(bs->file, BLKDBG_L2_LOAD);
- ret = bdrv_pread(bs->file, l2_offset, *l2_table,
- s->l2_size * sizeof(uint64_t));
- if (ret < 0) {
- return ret;
- }
-
- s->l2_cache_offsets[min_index] = l2_offset;
- s->l2_cache_counts[min_index] = 1;
+ ret = qcow2_cache_get(bs, s->l2_table_cache, l2_offset, (void**) l2_table);
- return 0;
+ return ret;
}
/*
static int l2_allocate(BlockDriverState *bs, int l1_index, uint64_t **table)
{
BDRVQcowState *s = bs->opaque;
- int min_index;
uint64_t old_l2_offset;
uint64_t *l2_table;
int64_t l2_offset;
if (l2_offset < 0) {
return l2_offset;
}
- bdrv_flush(bs->file);
+
+ ret = qcow2_cache_flush(bs, s->refcount_block_cache);
+ if (ret < 0) {
+ goto fail;
+ }
/* allocate a new entry in the l2 cache */
- min_index = l2_cache_new_entry(bs);
- l2_table = s->l2_cache + (min_index << s->l2_bits);
+ ret = qcow2_cache_get_empty(bs, s->l2_table_cache, l2_offset, (void**) table);
+ if (ret < 0) {
+ return ret;
+ }
+
+ l2_table = *table;
if (old_l2_offset == 0) {
/* if there was no old l2 table, clear the new table */
memset(l2_table, 0, s->l2_size * sizeof(uint64_t));
} else {
+ uint64_t* old_table;
+
/* if there was an old l2 table, read it from the disk */
BLKDBG_EVENT(bs->file, BLKDBG_L2_ALLOC_COW_READ);
- ret = bdrv_pread(bs->file, old_l2_offset, l2_table,
- s->l2_size * sizeof(uint64_t));
+ ret = qcow2_cache_get(bs, s->l2_table_cache, old_l2_offset,
+ (void**) &old_table);
+ if (ret < 0) {
+ goto fail;
+ }
+
+ memcpy(l2_table, old_table, s->cluster_size);
+
+ ret = qcow2_cache_put(bs, s->l2_table_cache, (void**) &old_table);
if (ret < 0) {
goto fail;
}
}
+
/* write the l2 table to the file */
BLKDBG_EVENT(bs->file, BLKDBG_L2_ALLOC_WRITE);
- ret = bdrv_pwrite_sync(bs->file, l2_offset, l2_table,
- s->l2_size * sizeof(uint64_t));
+
+ qcow2_cache_entry_mark_dirty(s->l2_table_cache, l2_table);
+ ret = qcow2_cache_flush(bs, s->l2_table_cache);
if (ret < 0) {
goto fail;
}
goto fail;
}
- /* update the l2 cache entry */
-
- s->l2_cache_offsets[min_index] = l2_offset;
- s->l2_cache_counts[min_index] = 1;
-
*table = l2_table;
return 0;
fail:
+ qcow2_cache_put(bs, s->l2_table_cache, (void**) table);
s->l1_table[l1_index] = old_l2_offset;
- qcow2_l2_cache_reset(bs);
return ret;
}
}
-static int qcow_read(BlockDriverState *bs, int64_t sector_num,
- uint8_t *buf, int nb_sectors)
+static int qcow2_read(BlockDriverState *bs, int64_t sector_num,
+ uint8_t *buf, int nb_sectors)
{
BDRVQcowState *s = bs->opaque;
int ret, index_in_cluster, n, n1;
if (n <= 0)
return 0;
BLKDBG_EVENT(bs->file, BLKDBG_COW_READ);
- ret = qcow_read(bs, start_sect + n_start, s->cluster_data, n);
+ ret = qcow2_read(bs, start_sect + n_start, s->cluster_data, n);
if (ret < 0)
return ret;
if (s->crypt_method) {
* For a given offset of the disk image, find the cluster offset in
* qcow2 file. The offset is stored in *cluster_offset.
*
- * on entry, *num is the number of contiguous clusters we'd like to
+ * on entry, *num is the number of contiguous sectors we'd like to
* access following offset.
*
- * on exit, *num is the number of contiguous clusters we can read.
+ * on exit, *num is the number of contiguous sectors we can read.
*
* Return 0, if the offset is found
* Return -errno, otherwise.
&l2_table[l2_index], 0, QCOW_OFLAG_COPIED);
}
+ qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
+
nb_available = (c * s->cluster_sectors);
out:
if (nb_available > nb_needed)
return ret;
}
} else {
- if (l2_offset)
- qcow2_free_clusters(bs, l2_offset, s->l2_size * sizeof(uint64_t));
+ /* First allocate a new L2 table (and do COW if needed) */
ret = l2_allocate(bs, l1_index, &l2_table);
if (ret < 0) {
return ret;
}
+
+ /* Then decrease the refcount of the old table */
+ if (l2_offset) {
+ qcow2_free_clusters(bs, l2_offset, s->l2_size * sizeof(uint64_t));
+ }
l2_offset = s->l1_table[l1_index] & ~QCOW_OFLAG_COPIED;
}
cluster_offset = qcow2_alloc_bytes(bs, compressed_size);
if (cluster_offset < 0) {
+ qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
return 0;
}
/* compressed clusters never have the copied flag */
BLKDBG_EVENT(bs->file, BLKDBG_L2_UPDATE_COMPRESSED);
+ qcow2_cache_entry_mark_dirty(s->l2_table_cache, l2_table);
l2_table[l2_index] = cpu_to_be64(cluster_offset);
- if (bdrv_pwrite_sync(bs->file,
- l2_offset + l2_index * sizeof(uint64_t),
- l2_table + l2_index,
- sizeof(uint64_t)) < 0)
- return 0;
-
- return cluster_offset;
-}
-
-/*
- * Write L2 table updates to disk, writing whole sectors to avoid a
- * read-modify-write in bdrv_pwrite
- */
-#define L2_ENTRIES_PER_SECTOR (512 / 8)
-static int write_l2_entries(BlockDriverState *bs, uint64_t *l2_table,
- uint64_t l2_offset, int l2_index, int num)
-{
- int l2_start_index = l2_index & ~(L1_ENTRIES_PER_SECTOR - 1);
- int start_offset = (8 * l2_index) & ~511;
- int end_offset = (8 * (l2_index + num) + 511) & ~511;
- size_t len = end_offset - start_offset;
- int ret;
-
- BLKDBG_EVENT(bs->file, BLKDBG_L2_UPDATE);
- ret = bdrv_pwrite(bs->file, l2_offset + start_offset,
- &l2_table[l2_start_index], len);
+ ret = qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
if (ret < 0) {
- return ret;
+ return 0;
}
- return 0;
+ return cluster_offset;
}
int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, QCowL2Meta *m)
int i, j = 0, l2_index, ret;
uint64_t *old_cluster, start_sect, l2_offset, *l2_table;
uint64_t cluster_offset = m->cluster_offset;
+ bool cow = false;
if (m->nb_clusters == 0)
return 0;
- old_cluster = qemu_malloc(m->nb_clusters * sizeof(uint64_t));
+ old_cluster = g_malloc(m->nb_clusters * sizeof(uint64_t));
/* copy content of unmodified sectors */
start_sect = (m->offset & ~(s->cluster_size - 1)) >> 9;
if (m->n_start) {
+ cow = true;
ret = copy_sectors(bs, start_sect, cluster_offset, 0, m->n_start);
if (ret < 0)
goto err;
if (m->nb_available & (s->cluster_sectors - 1)) {
uint64_t end = m->nb_available & ~(uint64_t)(s->cluster_sectors - 1);
+ cow = true;
ret = copy_sectors(bs, start_sect + end, cluster_offset + (end << 9),
m->nb_available - end, s->cluster_sectors);
if (ret < 0)
goto err;
}
- /* update L2 table */
+ /*
+ * Update L2 table.
+ *
+ * Before we update the L2 table to actually point to the new cluster, we
+ * need to be sure that the refcounts have been increased and COW was
+ * handled.
+ */
+ if (cow) {
+ qcow2_cache_depends_on_flush(s->l2_table_cache);
+ }
+
+ qcow2_cache_set_dependency(bs, s->l2_table_cache, s->refcount_block_cache);
ret = get_cluster_table(bs, m->offset, &l2_table, &l2_offset, &l2_index);
if (ret < 0) {
goto err;
}
+ qcow2_cache_entry_mark_dirty(s->l2_table_cache, l2_table);
for (i = 0; i < m->nb_clusters; i++) {
/* if two concurrent writes happen to the same unallocated cluster
(i << s->cluster_bits)) | QCOW_OFLAG_COPIED);
}
- /*
- * Before we update the L2 table to actually point to the new cluster, we
- * need to be sure that the refcounts have been increased and COW was
- * handled.
- */
- bdrv_flush(bs->file);
- ret = write_l2_entries(bs, l2_table, l2_offset, l2_index, m->nb_clusters);
+ ret = qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
if (ret < 0) {
- qcow2_l2_cache_reset(bs);
goto err;
}
* Also flush bs->file to get the right order for L2 and refcount update.
*/
if (j != 0) {
- bdrv_flush(bs->file);
for (i = 0; i < j; i++) {
qcow2_free_any_clusters(bs,
be64_to_cpu(old_cluster[i]) & ~QCOW_OFLAG_COPIED, 1);
ret = 0;
err:
- qemu_free(old_cluster);
+ g_free(old_cluster);
return ret;
}
* m->depends_on is set to NULL and the other fields in m are meaningless.
*
* If the cluster is newly allocated, m->nb_clusters is set to the number of
- * contiguous clusters that have been allocated. This may be 0 if the request
- * conflict with another write request in flight; in this case, m->depends_on
- * is set and the remaining fields of m are meaningless.
+ * contiguous clusters that have been allocated. In this case, the other
+ * fields of m are valid and contain information about the first allocated
+ * cluster.
*
- * If m->nb_clusters is non-zero, the other fields of m are valid and contain
- * information about the first allocated cluster.
+ * If the request conflicts with another write request in flight, the coroutine
+ * is queued and will be reentered when the dependency has completed.
*
* Return 0 on success and -errno in error cases
*/
return ret;
}
+again:
nb_clusters = size_to_clusters(s, n_end << 9);
nb_clusters = MIN(nb_clusters, s->l2_size - l2_index);
}
if (nb_clusters == 0) {
- /* Set dependency and wait for a callback */
- m->depends_on = old_alloc;
- m->nb_clusters = 0;
- *num = 0;
- return 0;
+ /* Wait for the dependency to complete. We need to recheck
+ * the free/allocated clusters when we continue. */
+ qemu_co_mutex_unlock(&s->lock);
+ qemu_co_queue_wait(&old_alloc->dependent_requests);
+ qemu_co_mutex_lock(&s->lock);
+ goto again;
}
}
}
cluster_offset = qcow2_alloc_clusters(bs, nb_clusters * s->cluster_size);
if (cluster_offset < 0) {
- QLIST_REMOVE(m, next_in_flight);
- return cluster_offset;
+ ret = cluster_offset;
+ goto fail;
}
/* save info needed for meta data update */
m->nb_clusters = nb_clusters;
out:
+ ret = qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
+ if (ret < 0) {
+ goto fail_put;
+ }
+
m->nb_available = MIN(nb_clusters << (s->cluster_bits - 9), n_end);
m->cluster_offset = cluster_offset;
*num = m->nb_available - n_start;
return 0;
+
+fail:
+ qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
+fail_put:
+ QLIST_REMOVE(m, next_in_flight);
+ return ret;
}
static int decompress_buffer(uint8_t *out_buf, int out_buf_size,
BLKDBG_EVENT(bs->file, BLKDBG_READ_COMPRESSED);
ret = bdrv_read(bs->file, coffset >> 9, s->cluster_data, nb_csectors);
if (ret < 0) {
- return -1;
+ return ret;
}
if (decompress_buffer(s->cluster_cache, s->cluster_size,
s->cluster_data + sector_offset, csize) < 0) {
- return -1;
+ return -EIO;
}
s->cluster_cache_offset = coffset;
}
return 0;
}
+
+/*
+ * This discards as many clusters of nb_clusters as possible at once (i.e.
+ * all clusters in the same L2 table) and returns the number of discarded
+ * clusters.
+ */
+static int discard_single_l2(BlockDriverState *bs, uint64_t offset,
+ unsigned int nb_clusters)
+{
+ BDRVQcowState *s = bs->opaque;
+ uint64_t l2_offset, *l2_table;
+ int l2_index;
+ int ret;
+ int i;
+
+ ret = get_cluster_table(bs, offset, &l2_table, &l2_offset, &l2_index);
+ if (ret < 0) {
+ return ret;
+ }
+
+ /* Limit nb_clusters to one L2 table */
+ nb_clusters = MIN(nb_clusters, s->l2_size - l2_index);
+
+ for (i = 0; i < nb_clusters; i++) {
+ uint64_t old_offset;
+
+ old_offset = be64_to_cpu(l2_table[l2_index + i]);
+ old_offset &= ~QCOW_OFLAG_COPIED;
+
+ if (old_offset == 0) {
+ continue;
+ }
+
+ /* First remove L2 entries */
+ qcow2_cache_entry_mark_dirty(s->l2_table_cache, l2_table);
+ l2_table[l2_index + i] = cpu_to_be64(0);
+
+ /* Then decrease the refcount */
+ qcow2_free_any_clusters(bs, old_offset, 1);
+ }
+
+ ret = qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
+ if (ret < 0) {
+ return ret;
+ }
+
+ return nb_clusters;
+}
+
+int qcow2_discard_clusters(BlockDriverState *bs, uint64_t offset,
+ int nb_sectors)
+{
+ BDRVQcowState *s = bs->opaque;
+ uint64_t end_offset;
+ unsigned int nb_clusters;
+ int ret;
+
+ end_offset = offset + (nb_sectors << BDRV_SECTOR_BITS);
+
+ /* Round start up and end down */
+ offset = align_offset(offset, s->cluster_size);
+ end_offset &= ~(s->cluster_size - 1);
+
+ if (offset > end_offset) {
+ return 0;
+ }
+
+ nb_clusters = size_to_clusters(s, end_offset - offset);
+
+ /* Each L2 table is handled by its own loop iteration */
+ while (nb_clusters > 0) {
+ ret = discard_single_l2(bs, offset, nb_clusters);
+ if (ret < 0) {
+ return ret;
+ }
+
+ nb_clusters -= ret;
+ offset += (ret * s->cluster_size);
+ }
+
+ return 0;
+}
projects / qemu.git / blobdiff