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

Use management queue for parallelized management operations

Browse Source 
When IO queues are used for parallelized management operations,
e.g. changing cleaning policy, a deadlock may occur due to global
metadata lock interfering with taking request from IO queue,
as they might be run on the same thread. As a workaround using
a management queue specifically for such operations eliminates
this problem.

Signed-off-by: Rafal Stefanowski <rafal.stefanowski@huawei.com>
This commit is contained in:
Rafal Stefanowski
2024-09-23 12:13:33 +02:00
parent d8994e886e
commit 97ee3af8f7
7 changed files with 15 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
src/metadata/metadata.c
View File

@@ -1,5 +1,6 @@
/*
* Copyright(c) 2012-2022 Intel Corporation
* Copyright(c) 2024 Huawei Technologies
* SPDX-License-Identifier: BSD-3-Clause
*/
@@ -882,7 +883,7 @@ void ocf_metadata_init_collision(ocf_pipeline_t pipeline, void *priv,
result = ocf_parallelize_create(&parallelize, cache,
ocf_cache_get_queue_count(cache), sizeof(*context),
ocf_metadata_init_collision_handle,
ocf_metadata_init_finish);
ocf_metadata_init_finish, false);
if (result)
OCF_PL_FINISH_RET(pipeline, result);
@@ -934,7 +935,7 @@ void ocf_metadata_init_hash_table(ocf_pipeline_t pipeline, void *priv,
result = ocf_parallelize_create(&parallelize, cache,
ocf_cache_get_queue_count(cache), sizeof(*context),
ocf_metadata_init_hash_table_handle,
ocf_metadata_init_finish);
ocf_metadata_init_finish, false);
if (result)
OCF_PL_FINISH_RET(pipeline, result);