From 7b7a8665edd8db733980389b098530f9e4f630b2 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 4 Sep 2013 15:04:39 +0200 Subject: direct-io: Implement generic deferred AIO completions Add support to the core direct-io code to defer AIO completions to user context using a workqueue. This replaces opencoded and less efficient code in XFS and ext4 (we save a memory allocation for each direct IO) and will be needed to properly support O_(D)SYNC for AIO. The communication between the filesystem and the direct I/O code requires a new buffer head flag, which is a bit ugly but not avoidable until the direct I/O code stops abusing the buffer_head structure for communicating with the filesystems. Currently this creates a per-superblock unbound workqueue for these completions, which is taken from an earlier patch by Jan Kara. I'm not really convinced about this use and would prefer a "normal" global workqueue with a high concurrency limit, but this needs further discussion. JK: Fixed ext4 part, dynamic allocation of the workqueue. Signed-off-by: Christoph Hellwig Signed-off-by: Jan Kara Signed-off-by: Al Viro --- fs/super.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'fs/super.c') diff --git a/fs/super.c b/fs/super.c index 68307c029228..5536a95186e2 100644 --- a/fs/super.c +++ b/fs/super.c @@ -152,15 +152,9 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags) static const struct super_operations default_op; if (s) { - if (security_sb_alloc(s)) { - /* - * We cannot call security_sb_free() without - * security_sb_alloc() succeeding. So bail out manually - */ - kfree(s); - s = NULL; - goto out; - } + if (security_sb_alloc(s)) + goto out_free_sb; + #ifdef CONFIG_SMP s->s_files = alloc_percpu(struct list_head); if (!s->s_files) @@ -228,6 +222,7 @@ err_out: free_percpu(s->s_files); #endif destroy_sb_writers(s); +out_free_sb: kfree(s); s = NULL; goto out; @@ -414,6 +409,11 @@ void generic_shutdown_super(struct super_block *sb) evict_inodes(sb); + if (sb->s_dio_done_wq) { + destroy_workqueue(sb->s_dio_done_wq); + sb->s_dio_done_wq = NULL; + } + if (sop->put_super) sop->put_super(sb); -- cgit v1.2.3 From d040790391f292bbe5bc6b990c66af9787c855a1 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 19 Jul 2013 21:12:31 +0400 Subject: prune_super(): sb->s_op is never NULL Signed-off-by: Al Viro --- fs/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/super.c') diff --git a/fs/super.c b/fs/super.c index 5536a95186e2..f6961ea84c56 100644 --- a/fs/super.c +++ b/fs/super.c @@ -71,7 +71,7 @@ static int prune_super(struct shrinker *shrink, struct shrink_control *sc) if (!grab_super_passive(sb)) return -1; - if (sb->s_op && sb->s_op->nr_cached_objects) + if (sb->s_op->nr_cached_objects) fs_objects = sb->s_op->nr_cached_objects(sb); total_objects = sb->s_nr_dentry_unused + -- cgit v1.2.3 From 55f841ce9395a72c6285fbcc4c403c0c786e1c74 Mon Sep 17 00:00:00 2001 From: Glauber Costa Date: Wed, 28 Aug 2013 10:17:53 +1000 Subject: super: fix calculation of shrinkable objects for small numbers MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The sysctl knob sysctl_vfs_cache_pressure is used to determine which percentage of the shrinkable objects in our cache we should actively try to shrink. It works great in situations in which we have many objects (at least more than 100), because the aproximation errors will be negligible. But if this is not the case, specially when total_objects < 100, we may end up concluding that we have no objects at all (total / 100 = 0, if total < 100). This is certainly not the biggest killer in the world, but may matter in very low kernel memory situations. Signed-off-by: Glauber Costa Reviewed-by: Carlos Maiolino Acked-by: KAMEZAWA Hiroyuki Acked-by: Mel Gorman Cc: Dave Chinner Cc: Al Viro Cc: "Theodore Ts'o" Cc: Adrian Hunter Cc: Al Viro Cc: Artem Bityutskiy Cc: Arve Hjønnevåg Cc: Carlos Maiolino Cc: Christoph Hellwig Cc: Chuck Lever Cc: Daniel Vetter Cc: David Rientjes Cc: Gleb Natapov Cc: Greg Thelen Cc: J. Bruce Fields Cc: Jan Kara Cc: Jerome Glisse Cc: John Stultz Cc: KAMEZAWA Hiroyuki Cc: Kent Overstreet Cc: Kirill A. Shutemov Cc: Marcelo Tosatti Cc: Mel Gorman Cc: Steven Whitehouse Cc: Thomas Hellstrom Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Al Viro --- fs/super.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'fs/super.c') diff --git a/fs/super.c b/fs/super.c index f6961ea84c56..63b6863bac7b 100644 --- a/fs/super.c +++ b/fs/super.c @@ -82,13 +82,13 @@ static int prune_super(struct shrinker *shrink, struct shrink_control *sc) int inodes; /* proportion the scan between the caches */ - dentries = (sc->nr_to_scan * sb->s_nr_dentry_unused) / - total_objects; - inodes = (sc->nr_to_scan * sb->s_nr_inodes_unused) / - total_objects; + dentries = mult_frac(sc->nr_to_scan, sb->s_nr_dentry_unused, + total_objects); + inodes = mult_frac(sc->nr_to_scan, sb->s_nr_inodes_unused, + total_objects); if (fs_objects) - fs_objects = (sc->nr_to_scan * fs_objects) / - total_objects; + fs_objects = mult_frac(sc->nr_to_scan, fs_objects, + total_objects); /* * prune the dcache first as the icache is pinned by it, then * prune the icache, followed by the filesystem specific caches @@ -104,7 +104,7 @@ static int prune_super(struct shrinker *shrink, struct shrink_control *sc) sb->s_nr_inodes_unused + fs_objects; } - total_objects = (total_objects / 100) * sysctl_vfs_cache_pressure; + total_objects = vfs_pressure_ratio(total_objects); drop_super(sb); return total_objects; } -- cgit v1.2.3 From 19156840e33a23eeb1a749c0f991dab6588b077d Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 28 Aug 2013 10:17:55 +1000 Subject: dentry: move to per-sb LRU locks MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit With the dentry LRUs being per-sb structures, there is no real need for a global dentry_lru_lock. The locking can be made more fine-grained by moving to a per-sb LRU lock, isolating the LRU operations of different filesytsems completely from each other. The need for this is independent of any performance consideration that may arise: in the interest of abstracting the lru operations away, it is mandatory that each lru works around its own lock instead of a global lock for all of them. [glommer@openvz.org: updated changelog ] Signed-off-by: Dave Chinner Signed-off-by: Glauber Costa Reviewed-by: Christoph Hellwig Acked-by: Mel Gorman Cc: "Theodore Ts'o" Cc: Adrian Hunter Cc: Al Viro Cc: Artem Bityutskiy Cc: Arve Hjønnevåg Cc: Carlos Maiolino Cc: Christoph Hellwig Cc: Chuck Lever Cc: Daniel Vetter Cc: David Rientjes Cc: Gleb Natapov Cc: Greg Thelen Cc: J. Bruce Fields Cc: Jan Kara Cc: Jerome Glisse Cc: John Stultz Cc: KAMEZAWA Hiroyuki Cc: Kent Overstreet Cc: Kirill A. Shutemov Cc: Marcelo Tosatti Cc: Mel Gorman Cc: Steven Whitehouse Cc: Thomas Hellstrom Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Al Viro --- fs/super.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs/super.c') diff --git a/fs/super.c b/fs/super.c index 63b6863bac7b..3c5318694ccd 100644 --- a/fs/super.c +++ b/fs/super.c @@ -176,6 +176,7 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags) INIT_HLIST_BL_HEAD(&s->s_anon); INIT_LIST_HEAD(&s->s_inodes); INIT_LIST_HEAD(&s->s_dentry_lru); + spin_lock_init(&s->s_dentry_lru_lock); INIT_LIST_HEAD(&s->s_inode_lru); spin_lock_init(&s->s_inode_lru_lock); INIT_LIST_HEAD(&s->s_mounts); -- cgit v1.2.3 From 0a234c6dcb79a270803f5c9773ed650b78730962 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 28 Aug 2013 10:17:57 +1000 Subject: shrinker: convert superblock shrinkers to new API MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Convert superblock shrinker to use the new count/scan API, and propagate the API changes through to the filesystem callouts. The filesystem callouts already use a count/scan API, so it's just changing counters to longs to match the VM API. This requires the dentry and inode shrinker callouts to be converted to the count/scan API. This is mainly a mechanical change. [glommer@openvz.org: use mult_frac for fractional proportions, build fixes] Signed-off-by: Dave Chinner Signed-off-by: Glauber Costa Acked-by: Mel Gorman Cc: "Theodore Ts'o" Cc: Adrian Hunter Cc: Al Viro Cc: Artem Bityutskiy Cc: Arve Hjønnevåg Cc: Carlos Maiolino Cc: Christoph Hellwig Cc: Chuck Lever Cc: Daniel Vetter Cc: David Rientjes Cc: Gleb Natapov Cc: Greg Thelen Cc: J. Bruce Fields Cc: Jan Kara Cc: Jerome Glisse Cc: John Stultz Cc: KAMEZAWA Hiroyuki Cc: Kent Overstreet Cc: Kirill A. Shutemov Cc: Marcelo Tosatti Cc: Mel Gorman Cc: Steven Whitehouse Cc: Thomas Hellstrom Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Al Viro --- fs/super.c | 80 ++++++++++++++++++++++++++++++++++++++------------------------ 1 file changed, 49 insertions(+), 31 deletions(-) (limited to 'fs/super.c') diff --git a/fs/super.c b/fs/super.c index 3c5318694ccd..8aa2660642b9 100644 --- a/fs/super.c +++ b/fs/super.c @@ -53,11 +53,15 @@ static char *sb_writers_name[SB_FREEZE_LEVELS] = { * shrinker path and that leads to deadlock on the shrinker_rwsem. Hence we * take a passive reference to the superblock to avoid this from occurring. */ -static int prune_super(struct shrinker *shrink, struct shrink_control *sc) +static unsigned long super_cache_scan(struct shrinker *shrink, + struct shrink_control *sc) { struct super_block *sb; - int fs_objects = 0; - int total_objects; + long fs_objects = 0; + long total_objects; + long freed = 0; + long dentries; + long inodes; sb = container_of(shrink, struct super_block, s_shrink); @@ -65,11 +69,11 @@ static int prune_super(struct shrinker *shrink, struct shrink_control *sc) * Deadlock avoidance. We may hold various FS locks, and we don't want * to recurse into the FS that called us in clear_inode() and friends.. */ - if (sc->nr_to_scan && !(sc->gfp_mask & __GFP_FS)) - return -1; + if (!(sc->gfp_mask & __GFP_FS)) + return SHRINK_STOP; if (!grab_super_passive(sb)) - return -1; + return SHRINK_STOP; if (sb->s_op->nr_cached_objects) fs_objects = sb->s_op->nr_cached_objects(sb); @@ -77,33 +81,46 @@ static int prune_super(struct shrinker *shrink, struct shrink_control *sc) total_objects = sb->s_nr_dentry_unused + sb->s_nr_inodes_unused + fs_objects + 1; - if (sc->nr_to_scan) { - int dentries; - int inodes; - - /* proportion the scan between the caches */ - dentries = mult_frac(sc->nr_to_scan, sb->s_nr_dentry_unused, - total_objects); - inodes = mult_frac(sc->nr_to_scan, sb->s_nr_inodes_unused, - total_objects); - if (fs_objects) - fs_objects = mult_frac(sc->nr_to_scan, fs_objects, - total_objects); - /* - * prune the dcache first as the icache is pinned by it, then - * prune the icache, followed by the filesystem specific caches - */ - prune_dcache_sb(sb, dentries); - prune_icache_sb(sb, inodes); + /* proportion the scan between the caches */ + dentries = mult_frac(sc->nr_to_scan, sb->s_nr_dentry_unused, + total_objects); + inodes = mult_frac(sc->nr_to_scan, sb->s_nr_inodes_unused, + total_objects); - if (fs_objects && sb->s_op->free_cached_objects) { - sb->s_op->free_cached_objects(sb, fs_objects); - fs_objects = sb->s_op->nr_cached_objects(sb); - } - total_objects = sb->s_nr_dentry_unused + - sb->s_nr_inodes_unused + fs_objects; + /* + * prune the dcache first as the icache is pinned by it, then + * prune the icache, followed by the filesystem specific caches + */ + freed = prune_dcache_sb(sb, dentries); + freed += prune_icache_sb(sb, inodes); + + if (fs_objects) { + fs_objects = mult_frac(sc->nr_to_scan, fs_objects, + total_objects); + freed += sb->s_op->free_cached_objects(sb, fs_objects); } + drop_super(sb); + return freed; +} + +static unsigned long super_cache_count(struct shrinker *shrink, + struct shrink_control *sc) +{ + struct super_block *sb; + long total_objects = 0; + + sb = container_of(shrink, struct super_block, s_shrink); + + if (!grab_super_passive(sb)) + return 0; + + if (sb->s_op && sb->s_op->nr_cached_objects) + total_objects = sb->s_op->nr_cached_objects(sb); + + total_objects += sb->s_nr_dentry_unused; + total_objects += sb->s_nr_inodes_unused; + total_objects = vfs_pressure_ratio(total_objects); drop_super(sb); return total_objects; @@ -211,7 +228,8 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags) s->cleancache_poolid = -1; s->s_shrink.seeks = DEFAULT_SEEKS; - s->s_shrink.shrink = prune_super; + s->s_shrink.scan_objects = super_cache_scan; + s->s_shrink.count_objects = super_cache_count; s->s_shrink.batch = 1024; } out: -- cgit v1.2.3 From bc3b14cb2d505dda969dbe3a31038dbb24aca945 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 28 Aug 2013 10:17:58 +1000 Subject: inode: convert inode lru list to generic lru list code. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [glommer@openvz.org: adapted for new LRU return codes] Signed-off-by: Dave Chinner Signed-off-by: Glauber Costa Cc: "Theodore Ts'o" Cc: Adrian Hunter Cc: Al Viro Cc: Artem Bityutskiy Cc: Arve Hjønnevåg Cc: Carlos Maiolino Cc: Christoph Hellwig Cc: Chuck Lever Cc: Daniel Vetter Cc: David Rientjes Cc: Gleb Natapov Cc: Greg Thelen Cc: J. Bruce Fields Cc: Jan Kara Cc: Jerome Glisse Cc: John Stultz Cc: KAMEZAWA Hiroyuki Cc: Kent Overstreet Cc: Kirill A. Shutemov Cc: Marcelo Tosatti Cc: Mel Gorman Cc: Steven Whitehouse Cc: Thomas Hellstrom Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Al Viro --- fs/super.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) (limited to 'fs/super.c') diff --git a/fs/super.c b/fs/super.c index 8aa2660642b9..aa7995d73bcc 100644 --- a/fs/super.c +++ b/fs/super.c @@ -78,14 +78,13 @@ static unsigned long super_cache_scan(struct shrinker *shrink, if (sb->s_op->nr_cached_objects) fs_objects = sb->s_op->nr_cached_objects(sb); - total_objects = sb->s_nr_dentry_unused + - sb->s_nr_inodes_unused + fs_objects + 1; + inodes = list_lru_count(&sb->s_inode_lru); + total_objects = sb->s_nr_dentry_unused + inodes + fs_objects + 1; /* proportion the scan between the caches */ dentries = mult_frac(sc->nr_to_scan, sb->s_nr_dentry_unused, total_objects); - inodes = mult_frac(sc->nr_to_scan, sb->s_nr_inodes_unused, - total_objects); + inodes = mult_frac(sc->nr_to_scan, inodes, total_objects); /* * prune the dcache first as the icache is pinned by it, then @@ -119,7 +118,7 @@ static unsigned long super_cache_count(struct shrinker *shrink, total_objects = sb->s_op->nr_cached_objects(sb); total_objects += sb->s_nr_dentry_unused; - total_objects += sb->s_nr_inodes_unused; + total_objects += list_lru_count(&sb->s_inode_lru); total_objects = vfs_pressure_ratio(total_objects); drop_super(sb); @@ -194,8 +193,7 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags) INIT_LIST_HEAD(&s->s_inodes); INIT_LIST_HEAD(&s->s_dentry_lru); spin_lock_init(&s->s_dentry_lru_lock); - INIT_LIST_HEAD(&s->s_inode_lru); - spin_lock_init(&s->s_inode_lru_lock); + list_lru_init(&s->s_inode_lru); INIT_LIST_HEAD(&s->s_mounts); init_rwsem(&s->s_umount); lockdep_set_class(&s->s_umount, &type->s_umount_key); -- cgit v1.2.3 From f604156751db77e08afe47ce29fe8f3d51ad9b04 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 28 Aug 2013 10:18:00 +1000 Subject: dcache: convert to use new lru list infrastructure MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [glommer@openvz.org: don't reintroduce double decrement of nr_unused_dentries, adapted for new LRU return codes] Signed-off-by: Dave Chinner Signed-off-by: Glauber Costa Cc: "Theodore Ts'o" Cc: Adrian Hunter Cc: Al Viro Cc: Artem Bityutskiy Cc: Arve Hjønnevåg Cc: Carlos Maiolino Cc: Christoph Hellwig Cc: Chuck Lever Cc: Daniel Vetter Cc: David Rientjes Cc: Gleb Natapov Cc: Greg Thelen Cc: J. Bruce Fields Cc: Jan Kara Cc: Jerome Glisse Cc: John Stultz Cc: KAMEZAWA Hiroyuki Cc: Kent Overstreet Cc: Kirill A. Shutemov Cc: Marcelo Tosatti Cc: Mel Gorman Cc: Steven Whitehouse Cc: Thomas Hellstrom Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Al Viro --- fs/super.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) (limited to 'fs/super.c') diff --git a/fs/super.c b/fs/super.c index aa7995d73bcc..cd3c2cd9144d 100644 --- a/fs/super.c +++ b/fs/super.c @@ -79,11 +79,11 @@ static unsigned long super_cache_scan(struct shrinker *shrink, fs_objects = sb->s_op->nr_cached_objects(sb); inodes = list_lru_count(&sb->s_inode_lru); - total_objects = sb->s_nr_dentry_unused + inodes + fs_objects + 1; + dentries = list_lru_count(&sb->s_dentry_lru); + total_objects = dentries + inodes + fs_objects + 1; /* proportion the scan between the caches */ - dentries = mult_frac(sc->nr_to_scan, sb->s_nr_dentry_unused, - total_objects); + dentries = mult_frac(sc->nr_to_scan, dentries, total_objects); inodes = mult_frac(sc->nr_to_scan, inodes, total_objects); /* @@ -117,7 +117,7 @@ static unsigned long super_cache_count(struct shrinker *shrink, if (sb->s_op && sb->s_op->nr_cached_objects) total_objects = sb->s_op->nr_cached_objects(sb); - total_objects += sb->s_nr_dentry_unused; + total_objects += list_lru_count(&sb->s_dentry_lru); total_objects += list_lru_count(&sb->s_inode_lru); total_objects = vfs_pressure_ratio(total_objects); @@ -191,8 +191,7 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags) INIT_HLIST_NODE(&s->s_instances); INIT_HLIST_BL_HEAD(&s->s_anon); INIT_LIST_HEAD(&s->s_inodes); - INIT_LIST_HEAD(&s->s_dentry_lru); - spin_lock_init(&s->s_dentry_lru_lock); + list_lru_init(&s->s_dentry_lru); list_lru_init(&s->s_inode_lru); INIT_LIST_HEAD(&s->s_mounts); init_rwsem(&s->s_umount); -- cgit v1.2.3 From 9b17c62382dd2e7507984b9890bf44e070cdd8bb Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 28 Aug 2013 10:18:05 +1000 Subject: fs: convert inode and dentry shrinking to be node aware MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Now that the shrinker is passing a node in the scan control structure, we can pass this to the the generic LRU list code to isolate reclaim to the lists on matching nodes. Signed-off-by: Dave Chinner Signed-off-by: Glauber Costa Acked-by: Mel Gorman Cc: "Theodore Ts'o" Cc: Adrian Hunter Cc: Al Viro Cc: Artem Bityutskiy Cc: Arve Hjønnevåg Cc: Carlos Maiolino Cc: Christoph Hellwig Cc: Chuck Lever Cc: Daniel Vetter Cc: David Rientjes Cc: Gleb Natapov Cc: Greg Thelen Cc: J. Bruce Fields Cc: Jan Kara Cc: Jerome Glisse Cc: John Stultz Cc: KAMEZAWA Hiroyuki Cc: Kent Overstreet Cc: Kirill A. Shutemov Cc: Marcelo Tosatti Cc: Mel Gorman Cc: Steven Whitehouse Cc: Thomas Hellstrom Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Al Viro --- fs/super.c | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-) (limited to 'fs/super.c') diff --git a/fs/super.c b/fs/super.c index cd3c2cd9144d..181d42e2abff 100644 --- a/fs/super.c +++ b/fs/super.c @@ -76,10 +76,10 @@ static unsigned long super_cache_scan(struct shrinker *shrink, return SHRINK_STOP; if (sb->s_op->nr_cached_objects) - fs_objects = sb->s_op->nr_cached_objects(sb); + fs_objects = sb->s_op->nr_cached_objects(sb, sc->nid); - inodes = list_lru_count(&sb->s_inode_lru); - dentries = list_lru_count(&sb->s_dentry_lru); + inodes = list_lru_count_node(&sb->s_inode_lru, sc->nid); + dentries = list_lru_count_node(&sb->s_dentry_lru, sc->nid); total_objects = dentries + inodes + fs_objects + 1; /* proportion the scan between the caches */ @@ -90,13 +90,14 @@ static unsigned long super_cache_scan(struct shrinker *shrink, * prune the dcache first as the icache is pinned by it, then * prune the icache, followed by the filesystem specific caches */ - freed = prune_dcache_sb(sb, dentries); - freed += prune_icache_sb(sb, inodes); + freed = prune_dcache_sb(sb, dentries, sc->nid); + freed += prune_icache_sb(sb, inodes, sc->nid); if (fs_objects) { fs_objects = mult_frac(sc->nr_to_scan, fs_objects, total_objects); - freed += sb->s_op->free_cached_objects(sb, fs_objects); + freed += sb->s_op->free_cached_objects(sb, fs_objects, + sc->nid); } drop_super(sb); @@ -115,10 +116,13 @@ static unsigned long super_cache_count(struct shrinker *shrink, return 0; if (sb->s_op && sb->s_op->nr_cached_objects) - total_objects = sb->s_op->nr_cached_objects(sb); + total_objects = sb->s_op->nr_cached_objects(sb, + sc->nid); - total_objects += list_lru_count(&sb->s_dentry_lru); - total_objects += list_lru_count(&sb->s_inode_lru); + total_objects += list_lru_count_node(&sb->s_dentry_lru, + sc->nid); + total_objects += list_lru_count_node(&sb->s_inode_lru, + sc->nid); total_objects = vfs_pressure_ratio(total_objects); drop_super(sb); @@ -228,6 +232,7 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags) s->s_shrink.scan_objects = super_cache_scan; s->s_shrink.count_objects = super_cache_count; s->s_shrink.batch = 1024; + s->s_shrink.flags = SHRINKER_NUMA_AWARE; } out: return s; -- cgit v1.2.3 From 5ca302c8e502ca53b7d75f12127ec0289904003a Mon Sep 17 00:00:00 2001 From: Glauber Costa Date: Wed, 28 Aug 2013 10:18:18 +1000 Subject: list_lru: dynamically adjust node arrays MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We currently use a compile-time constant to size the node array for the list_lru structure. Due to this, we don't need to allocate any memory at initialization time. But as a consequence, the structures that contain embedded list_lru lists can become way too big (the superblock for instance contains two of them). This patch aims at ameliorating this situation by dynamically allocating the node arrays with the firmware provided nr_node_ids. Signed-off-by: Glauber Costa Cc: Dave Chinner Cc: Mel Gorman Cc: "Theodore Ts'o" Cc: Adrian Hunter Cc: Al Viro Cc: Artem Bityutskiy Cc: Arve Hjønnevåg Cc: Carlos Maiolino Cc: Christoph Hellwig Cc: Chuck Lever Cc: Daniel Vetter Cc: David Rientjes Cc: Gleb Natapov Cc: Greg Thelen Cc: J. Bruce Fields Cc: Jan Kara Cc: Jerome Glisse Cc: John Stultz Cc: KAMEZAWA Hiroyuki Cc: Kent Overstreet Cc: Kirill A. Shutemov Cc: Marcelo Tosatti Cc: Mel Gorman Cc: Steven Whitehouse Cc: Thomas Hellstrom Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Al Viro --- fs/super.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'fs/super.c') diff --git a/fs/super.c b/fs/super.c index 181d42e2abff..269d96857caa 100644 --- a/fs/super.c +++ b/fs/super.c @@ -195,8 +195,12 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags) INIT_HLIST_NODE(&s->s_instances); INIT_HLIST_BL_HEAD(&s->s_anon); INIT_LIST_HEAD(&s->s_inodes); - list_lru_init(&s->s_dentry_lru); - list_lru_init(&s->s_inode_lru); + + if (list_lru_init(&s->s_dentry_lru)) + goto err_out; + if (list_lru_init(&s->s_inode_lru)) + goto err_out_dentry_lru; + INIT_LIST_HEAD(&s->s_mounts); init_rwsem(&s->s_umount); lockdep_set_class(&s->s_umount, &type->s_umount_key); @@ -236,6 +240,9 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags) } out: return s; + +err_out_dentry_lru: + list_lru_destroy(&s->s_dentry_lru); err_out: security_sb_free(s); #ifdef CONFIG_SMP -- cgit v1.2.3 From f5e1dd34561e0fb06400b378d595198918833021 Mon Sep 17 00:00:00 2001 From: Glauber Costa Date: Wed, 28 Aug 2013 10:18:18 +1000 Subject: super: fix for destroy lrus This patch adds the missing call to list_lru_destroy (spotted by Li Zhong) and moves the deletion to after the shrinker is unregistered, as correctly spotted by Dave Signed-off-by: Glauber Costa Cc: Michal Hocko Cc: Dave Chinner Signed-off-by: Andrew Morton Signed-off-by: Al Viro --- fs/super.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs/super.c') diff --git a/fs/super.c b/fs/super.c index 269d96857caa..3a96c9783a8b 100644 --- a/fs/super.c +++ b/fs/super.c @@ -323,6 +323,9 @@ void deactivate_locked_super(struct super_block *s) /* caches are now gone, we can safely kill the shrinker now */ unregister_shrinker(&s->s_shrink); + list_lru_destroy(&s->s_dentry_lru); + list_lru_destroy(&s->s_inode_lru); + put_filesystem(fs); put_super(s); } else { -- cgit v1.2.3 From c2d22ecd3cf3c49aeecdc10ffd7af10ec1f7eab6 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 1 Oct 2013 13:11:21 -0400 Subject: fs/super.c: fix lru_list leak for real Freeing ->s_{inode,dentry}_lru in deactivate_locked_super() is wrong; the right place is destroy_super(). As it is, we leak them if sget() decides that new superblock it has allocated (and never shown to anybody) isn't needed and should be freed. Signed-off-by: Al Viro --- fs/super.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/super.c') diff --git a/fs/super.c b/fs/super.c index 3a96c9783a8b..0225c20f8770 100644 --- a/fs/super.c +++ b/fs/super.c @@ -264,6 +264,8 @@ out_free_sb: */ static inline void destroy_super(struct super_block *s) { + list_lru_destroy(&s->s_dentry_lru); + list_lru_destroy(&s->s_inode_lru); #ifdef CONFIG_SMP free_percpu(s->s_files); #endif @@ -323,8 +325,6 @@ void deactivate_locked_super(struct super_block *s) /* caches are now gone, we can safely kill the shrinker now */ unregister_shrinker(&s->s_shrink); - list_lru_destroy(&s->s_dentry_lru); - list_lru_destroy(&s->s_inode_lru); put_filesystem(fs); put_super(s); -- cgit v1.2.3