mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2025-03-11 23:27:42 +07:00
8c0a853770
There's no reason to call rcu_barrier() on every deactivate_locked_super(). We only need to make sure that all delayed rcu free inodes are flushed before we destroy related cache. Removing rcu_barrier() from deactivate_locked_super() affects some fast paths. E.g. on my machine exit_group() of a last process in IPC namespace takes 0.07538s. rcu_barrier() takes 0.05188s of that time. Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> |
||
|---|---|---|
| .. | ||
| budget.c | ||
| commit.c | ||
| compress.c | ||
| debug.c | ||
| debug.h | ||
| dir.c | ||
| file.c | ||
| find.c | ||
| gc.c | ||
| io.c | ||
| ioctl.c | ||
| journal.c | ||
| Kconfig | ||
| key.h | ||
| log.c | ||
| lprops.c | ||
| lpt_commit.c | ||
| lpt.c | ||
| Makefile | ||
| master.c | ||
| misc.h | ||
| orphan.c | ||
| recovery.c | ||
| replay.c | ||
| sb.c | ||
| scan.c | ||
| shrinker.c | ||
| super.c | ||
| tnc_commit.c | ||
| tnc_misc.c | ||
| tnc.c | ||
| ubifs-media.h | ||
| ubifs.h | ||
| xattr.c | ||