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

Revert "Optimize cacheline locking in ocf_engine_prepare_clines"

Browse Source 
This change introduced a race condition. In some code paths after
cacheline trylock failed, hash bucket lock needed bo be upgraded
in order to obtain asynchronous lock. During hash bucket lock
upgrade, hash read locks were released followed by obtaining
hash write locks. After read locks were released, concurrent
thread could obtain hash bucket locks and modify cacheline
state. The thread upgrading hash bucket lock would need to
repeat traversation in order to safely continue.

This reverts commit 30f22d4f47.

Signed-off-by: Adam Rutkowski <adam.j.rutkowski@intel.com>
This commit is contained in:
Adam Rutkowski
2019-09-30 23:15:35 -04:00
parent 944d70288e
commit 09b68297b2
8 changed files with 101 additions and 151 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
src/engine/engine_pt.c
View File

@@ -102,7 +102,7 @@ static const struct ocf_io_if _io_if_pt_resume = {
int ocf_read_pt(struct ocf_request *req)
{
bool use_cache = false;
int lock = OCF_LOCK_ACQUIRED;
int lock = OCF_LOCK_NOT_ACQUIRED;
OCF_DEBUG_TRACE(req->cache);
@@ -127,17 +127,14 @@ int ocf_read_pt(struct ocf_request *req)
/* There are mapped cache line,
* lock request for READ access
*/
lock = ocf_req_trylock_rd(req);
lock = ocf_req_async_lock_rd(req, ocf_engine_on_resume);
} else {
/* No mapped cache lines, no need to get lock */
lock = OCF_LOCK_ACQUIRED;
}
}
if (lock != OCF_LOCK_ACQUIRED) {
ocf_req_hash_lock_upgrade(req);
lock = ocf_req_async_lock_rd(req, ocf_engine_on_resume);
ocf_req_hash_unlock_wr(req);
} else {
ocf_req_hash_unlock_rd(req);
}
ocf_req_hash_unlock_rd(req);
if (use_cache) {
/*