linux_dsm_epyc7002/fs
Jason Baron 67347fe4e6 epoll: do not take global 'epmutex' for simple topologies
When calling EPOLL_CTL_ADD for an epoll file descriptor that is attached
directly to a wakeup source, we do not need to take the global 'epmutex',
unless the epoll file descriptor is nested.  The purpose of taking the
'epmutex' on add is to prevent complex topologies such as loops and deep
wakeup paths from forming in parallel through multiple EPOLL_CTL_ADD
operations.  However, for the simple case of an epoll file descriptor
attached directly to a wakeup source (with no nesting), we do not need to
hold the 'epmutex'.

This patch along with 'epoll: optimize EPOLL_CTL_DEL using rcu' improves
scalability on larger systems.  Quoting Nathan Zimmer's mail on SPECjbb
performance:

"On the 16 socket run the performance went from 35k jOPS to 125k jOPS.  In
addition the benchmark when from scaling well on 10 sockets to scaling
well on just over 40 sockets.

...

Currently the benchmark stops scaling at around 40-44 sockets but it seems like
I found a second unrelated bottleneck."

[akpm@linux-foundation.org: use `bool' for boolean variables, remove unneeded/undesirable cast of void*, add missed ep_scan_ready_list() kerneldoc]
Signed-off-by: Jason Baron <jbaron@akamai.com>
Tested-by: Nathan Zimmer <nzimmer@sgi.com>
Cc: Eric Wong <normalperson@yhbt.net>
Cc: Nelson Elhage <nelhage@nelhage.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Davide Libenzi <davidel@xmailserver.org>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-11-13 12:09:25 +09:00
..
9p
adfs
affs
afs Merge branch 'fscache' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs into linux-next 2013-10-28 19:36:46 -04:00
autofs4
befs
bfs
btrfs
cachefiles
ceph
cifs Merge branch 'for-linus' of git://git.samba.org/sfrench/cifs-2.6 2013-11-08 06:01:47 +09:00
coda
configfs
cramfs cramfs: mark as obsolete 2013-11-13 12:09:12 +09:00
debugfs debugfs: use list_next_entry() in debugfs_remove_recursive() 2013-11-13 12:09:24 +09:00
devpts
dlm
ecryptfs
efivarfs
efs
exofs
exportfs
ext2
ext3
ext4
f2fs
fat
freevxfs
fscache
fuse
gfs2 The main feature of interest this time is quota updates. There are 2013-11-11 07:11:00 +09:00
hfs
hfsplus
hostfs
hpfs
hppfs
hugetlbfs
isofs
jbd
jbd2
jffs2
jfs
lockd
logfs
minix
ncpfs
nfs NFSv4.2: Remove redundant checks in nfs_setsecurity+nfs4_label_init_security 2013-11-04 16:42:52 -05:00
nfs_common
nfsd
nilfs2
nls
notify
ntfs
ocfs2 ocfs2: simplify ocfs2_invalidatepage() and ocfs2_releasepage() 2013-11-13 12:09:02 +09:00
omfs
openpromfs
proc mm: factor commit limit calculation 2013-11-13 12:09:11 +09:00
pstore
qnx4
qnx6
quota
ramfs
reiserfs
romfs
squashfs
sysfs Revert "sysfs: drop kobj_ns_type handling" 2013-11-07 20:47:28 +09:00
sysv
ubifs
udf
ufs
xfs writeback: do not sync data dirtied after sync start 2013-11-13 12:09:07 +09:00
aio.c
anon_inodes.c
attr.c
bad_inode.c
binfmt_aout.c
binfmt_elf_fdpic.c
binfmt_elf.c
binfmt_em86.c
binfmt_flat.c
binfmt_misc.c
binfmt_script.c
binfmt_som.c
bio-integrity.c
bio.c
block_dev.c
buffer.c
char_dev.c
compat_binfmt_elf.c
compat_ioctl.c
compat.c
coredump.c
coredump.h
dcache.c vfs: decrapify dput(), fix cache behavior under normal load 2013-10-31 15:43:02 -07:00
dcookies.c
direct-io.c
drop_caches.c
eventfd.c
eventpoll.c epoll: do not take global 'epmutex' for simple topologies 2013-11-13 12:09:25 +09:00
exec.c
fcntl.c
fhandle.c
file_table.c
file.c
filesystems.c
fs_struct.c
fs-writeback.c writeback: do not sync data dirtied after sync start 2013-11-13 12:09:07 +09:00
generic_acl.c
inode.c
internal.h
ioctl.c
ioprio.c
Kconfig
Kconfig.binfmt
libfs.c
locks.c
Makefile
mbcache.c
mount.h
mpage.c
namei.c
namespace.c
no-block.c
open.c
pipe.c
pnode.c
pnode.h
posix_acl.c
proc_namespace.c
read_write.c
readdir.c
select.c Revert "select: use freezable blocking call" 2013-10-30 15:28:35 +01:00
seq_file.c
signalfd.c
splice.c
stack.c
stat.c
statfs.c
super.c
sync.c writeback: do not sync data dirtied after sync start 2013-11-13 12:09:07 +09:00
timerfd.c
utimes.c
xattr_acl.c
xattr.c