github/ocf
4
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Introduce hash bucket locks

Browse Source 
There is one RW lock per hash bucket. Write lock is required
to map cacheline, read lock is sufficient for traversing.
Hash bucket locks are always acquired under global metadata
read lock. This assures mutual exclusion with eviction and
management paths, where global metadata write lock is held.

Signed-off-by: Adam Rutkowski <adam.j.rutkowski@intel.com>
This commit is contained in:
Adam Rutkowski
2019-08-01 16:57:14 -04:00
parent 42f65c3fbb
commit d91012f4b4
7 changed files with 298 additions and 87 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
src/mngt/ocf_mngt_cache.c
View File

@@ -986,7 +986,8 @@ static void _ocf_mngt_attach_prepare_metadata(ocf_pipeline_t pipeline,
context->flags.attached_metadata_inited = true;
if (ocf_metadata_concurrency_attached_init(cache)) {
if (ocf_metadata_concurrency_attached_init(&cache->metadata.lock,
cache->device->hash_table_entries)) {
ocf_cache_log(cache, log_err, "Failed to initialize attached "
"metadata concurrency\n");
OCF_PL_FINISH_RET(context->pipeline, -OCF_ERR_START_CACHE_FAIL);
@@ -1734,6 +1735,7 @@ static void _ocf_mngt_cache_unplug_complete(void *priv, int error)
ocf_volume_close(&cache->device->volume);
ocf_metadata_concurrency_attached_deinit(&cache->metadata.lock);
ocf_metadata_deinit_variable_size(cache);
ocf_concurrency_deinit(cache);
ocf_freelist_deinit(cache->freelist);