Module Name: src
Committed By: ad
Date: Sun Dec 1 13:39:53 UTC 2019
Modified Files:
src/sys/kern: vfs_cache.c
src/sys/rump/include/rump: rump_namei.h
src/sys/sys: namei.h namei.src
Log Message:
namecache changes:
- Delete the per-entry lock, and borrow the associated vnode's v_interlock
instead. We need to acquire it during lookup anyway. We can revisit this
in the future but for now it's a stepping stone, and works within the
quite limited context of what we have (BSD namecache/lookup design).
- Implement an idea that Mateusz Guzik (m...@freebsd.org) gave me. In
cache_reclaim(), we don't need to lock out all of the CPUs to garbage
collect entries. All we need to do is observe their locks unheld at least
once: then we know they are not in the critical section, and no longer
have visibility of the entries about to be garbage collected.
- The above makes it safe for sysctl to take only namecache_lock to get stats,
and we can remove all the crap dealing with per-CPU locks.
- For lockstat, make namecache_lock a static now we have __cacheline_aligned.
- Avoid false sharing - don't write back to nc_hittime unless it has changed.
Put a a comment in place explaining this. Pretty sure this was there in
2008/2009 but someone removed it (understandably, the code looks weird).
- Use a mutex to protect the garbage collection queue instead of atomics, and
adjust the low water mark up so that cache_reclaim() isn't doing so much
work at once.
To generate a diff of this commit:
cvs rdiff -u -r1.123 -r1.124 src/sys/kern/vfs_cache.c
cvs rdiff -u -r1.33 -r1.34 src/sys/rump/include/rump/rump_namei.h
cvs rdiff -u -r1.99 -r1.100 src/sys/sys/namei.h
cvs rdiff -u -r1.43 -r1.44 src/sys/sys/namei.src
Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.
Modified files:
Index: src/sys/kern/vfs_cache.c
diff -u src/sys/kern/vfs_cache.c:1.123 src/sys/kern/vfs_cache.c:1.124
--- src/sys/kern/vfs_cache.c:1.123 Sun Sep 15 17:37:25 2019
+++ src/sys/kern/vfs_cache.c Sun Dec 1 13:39:53 2019
@@ -1,7 +1,7 @@
-/* $NetBSD: vfs_cache.c,v 1.123 2019/09/15 17:37:25 maya Exp $ */
+/* $NetBSD: vfs_cache.c,v 1.124 2019/12/01 13:39:53 ad Exp $ */
/*-
- * Copyright (c) 2008 The NetBSD Foundation, Inc.
+ * Copyright (c) 2008, 2019 The NetBSD Foundation, Inc.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
@@ -58,7 +58,7 @@
*/
#include <sys/cdefs.h>
-__KERNEL_RCSID(0, "$NetBSD: vfs_cache.c,v 1.123 2019/09/15 17:37:25 maya Exp $");
+__KERNEL_RCSID(0, "$NetBSD: vfs_cache.c,v 1.124 2019/12/01 13:39:53 ad Exp $");
#define __NAMECACHE_PRIVATE
#ifdef _KERNEL_OPT
@@ -130,7 +130,7 @@ __KERNEL_RCSID(0, "$NetBSD: vfs_cache.c,
* - Invalidate: active--->queued
*
* Done by cache_invalidate. If not already invalidated, nullify
- * ncp->nc_dvp and ncp->nc_vp, and add to cache_gcqueue. Called,
+ * ncp->nc_dvp and ncp->nc_vp, and add to namecache_gc_queue. Called,
* among various other places, in cache_lookup(dvp, name, namelen,
* nameiop, cnflags, &iswht, &vp) when MAKEENTRY is missing from
* cnflags.
@@ -145,16 +145,17 @@ __KERNEL_RCSID(0, "$NetBSD: vfs_cache.c,
* Locking.
*
* L namecache_lock Global lock for namecache table and queues.
+ * G namecache_gc_lock Global lock for garbage collection.
* C struct nchcpu::cpu_lock Per-CPU lock to reduce read contention.
- * N struct namecache::nc_lock Per-entry lock.
- * V struct vnode::v_interlock Vnode interlock.
+ * N struct namecache::nc_lock Per-entry lock, matching nc_vp->v_interlock.
+ * If nc_vp==NULL, lock is private / not shared.
*
- * Lock order: L -> C -> N -> V
+ * Lock order: L -> C -> N
*
* Examples:
* . L->C: cache_reclaim
- * . C->N->V: cache_lookup
- * . L->N->V: cache_purge1, cache_revlookup
+ * . C->N: cache_lookup
+ * . L->N: cache_purge1, cache_revlookup
*
* All use serialized by namecache_lock:
*
@@ -167,8 +168,9 @@ __KERNEL_RCSID(0, "$NetBSD: vfs_cache.c,
* - Insertion serialized by namecache_lock,
* - read protected by per-CPU lock,
* - insert/read ordering guaranteed by memory barriers, and
- * - deletion allowed only under namecache_lock and *all* per-CPU locks
- * in CPU_INFO_FOREACH order:
+ * - deletion allowed only under namecache_lock, with namecache_gc_lock
+ * taken to chop out the garbage collection list, and *all* per-CPU locks
+ * observed as "unowned" at least once:
*
* nchashtbl / struct namecache::nc_hash
*
@@ -180,11 +182,13 @@ __KERNEL_RCSID(0, "$NetBSD: vfs_cache.c,
*
* struct namecache::nc_dvp
* struct namecache::nc_vp
- * struct namecache::nc_gcqueue (*)
- * struct namecache::nc_hittime (**)
+ * struct namecache::nc_hittime (*)
*
- * (*) Once on the queue, only cache_thread uses this nc_gcqueue, unlocked.
- * (**) cache_prune reads nc_hittime unlocked, since approximate is OK.
+ * All use serialized by struct namecache_gc_lock:
+ *
+ * struct namecache::nc_gclist
+ *
+ * (*) cache_prune reads nc_hittime unlocked, since approximate is OK.
*
* Unlocked because stable after initialization:
*
@@ -257,7 +261,7 @@ typedef u_long nchash_t;
* Structures associated with name cacheing.
*/
-static kmutex_t *namecache_lock __read_mostly;
+static kmutex_t namecache_lock __cacheline_aligned;
static pool_cache_t namecache_cache __read_mostly;
static TAILQ_HEAD(, namecache) nclruhead __cacheline_aligned;
@@ -276,8 +280,9 @@ static u_long ncvhash __read_mostly;
static long numcache __cacheline_aligned;
/* Garbage collection queue and number of entries pending in it. */
-static void *cache_gcqueue;
-static u_int cache_gcpend;
+static kmutex_t namecache_gc_lock __cacheline_aligned;
+static SLIST_HEAD(namecache_gc_queue, namecache) namecache_gc_queue;
+static u_int namecache_gc_pend;
/* Cache effectiveness statistics. This holds total from per-cpu stats */
struct nchstats nchstats __cacheline_aligned;
@@ -287,8 +292,6 @@ struct nchstats nchstats __cacheline_ali
* values and add current per-cpu increments to the subsystem total
* last collected by cache_reclaim().
*/
-#define CACHE_STATS_CURRENT /* nothing */
-
#define COUNT(cpup, f) ((cpup)->cpu_stats.f++)
#define UPDATE(cpup, f) do { \
@@ -298,15 +301,10 @@ struct nchstats nchstats __cacheline_ali
Xcpup->cpu_stats_last.f = Xcnt; \
} while (/* CONSTCOND */ 0)
-#define ADD(stats, cpup, f) do { \
- struct nchcpu *Xcpup = (cpup); \
- stats.f += Xcpup->cpu_stats.f - Xcpup->cpu_stats_last.f; \
-} while (/* CONSTCOND */ 0)
-
/* Do unlocked stats the same way. Use a different name to allow mind changes */
#define COUNT_UNL(cpup, f) COUNT((cpup), f)
-static const int cache_lowat = 95;
+static const int cache_lowat = 97;
static const int cache_hiwat = 98;
static const int cache_hottime = 5; /* number of seconds */
static int doingcache = 1; /* 1 => enable the cache */
@@ -369,15 +367,12 @@ cache_hash(const char *name, size_t name
/*
* Invalidate a cache entry and enqueue it for garbage collection.
- * The caller needs to hold namecache_lock or a per-cpu lock to hold
- * off cache_reclaim().
*/
static void
cache_invalidate(struct namecache *ncp)
{
- void *head;
- KASSERT(mutex_owned(&ncp->nc_lock));
+ KASSERT(mutex_owned(ncp->nc_lock));
if (ncp->nc_dvp != NULL) {
SDT_PROBE(vfs, namecache, invalidate, done, ncp->nc_dvp,
@@ -385,11 +380,10 @@ cache_invalidate(struct namecache *ncp)
ncp->nc_vp = NULL;
ncp->nc_dvp = NULL;
- do {
- head = cache_gcqueue;
- ncp->nc_gcqueue = head;
- } while (atomic_cas_ptr(&cache_gcqueue, head, ncp) != head);
- atomic_inc_uint(&cache_gcpend);
+ mutex_enter(&namecache_gc_lock);
+ SLIST_INSERT_HEAD(&namecache_gc_queue, ncp, nc_gclist);
+ namecache_gc_pend++;
+ mutex_exit(&namecache_gc_lock);
}
}
@@ -401,7 +395,7 @@ static void
cache_disassociate(struct namecache *ncp)
{
- KASSERT(mutex_owned(namecache_lock));
+ KASSERT(mutex_owned(&namecache_lock));
KASSERT(ncp->nc_dvp == NULL);
if (ncp->nc_lru.tqe_prev != NULL) {
@@ -424,7 +418,8 @@ cache_disassociate(struct namecache *ncp
/*
* Lock all CPUs to prevent any cache lookup activity. Conceptually,
- * this locks out all "readers".
+ * this locks out all "readers". This is a very heavyweight operation
+ * that we only use for nchreinit().
*/
static void
cache_lock_cpus(void)
@@ -433,6 +428,9 @@ cache_lock_cpus(void)
struct cpu_info *ci;
struct nchcpu *cpup;
+ /* Not necessary but don't want more than one LWP trying this. */
+ KASSERT(mutex_owned(&namecache_lock));
+
/*
* Lock out all CPUs first, then harvest per-cpu stats. This
* is probably not quite as cache-efficient as doing the lock
@@ -485,6 +483,7 @@ cache_lookup_entry(const struct vnode *d
struct nchashhead *ncpp;
struct namecache *ncp;
nchash_t hash;
+ int ticks;
KASSERT(dvp != NULL);
hash = cache_hash(name, namelen);
@@ -496,15 +495,22 @@ cache_lookup_entry(const struct vnode *d
ncp->nc_nlen != namelen ||
memcmp(ncp->nc_name, name, (u_int)ncp->nc_nlen))
continue;
- mutex_enter(&ncp->nc_lock);
+ mutex_enter(ncp->nc_lock);
if (__predict_true(ncp->nc_dvp == dvp)) {
- ncp->nc_hittime = hardclock_ticks;
+ ticks = hardclock_ticks;
+ if (ncp->nc_hittime != ticks) {
+ /*
+ * Avoid false sharing on MP: do not store
+ * to *ncp unless the value changed.
+ */
+ ncp->nc_hittime = ticks;
+ }
SDT_PROBE(vfs, namecache, lookup, hit, dvp,
name, namelen, 0, 0);
return ncp;
}
/* Raced: entry has been nullified. */
- mutex_exit(&ncp->nc_lock);
+ mutex_exit(ncp->nc_lock);
}
SDT_PROBE(vfs, namecache, lookup, miss, dvp,
@@ -573,7 +579,6 @@ cache_lookup(struct vnode *dvp, const ch
int error;
bool hit;
-
/* Establish default result values */
if (iswht_ret != NULL) {
*iswht_ret = 0;
@@ -610,12 +615,13 @@ cache_lookup(struct vnode *dvp, const ch
* want cache entry to exist.
*/
cache_invalidate(ncp);
- mutex_exit(&ncp->nc_lock);
+ mutex_exit(ncp->nc_lock);
mutex_exit(&cpup->cpu_lock);
/* found nothing */
return false;
}
- if (ncp->nc_vp == NULL) {
+ vp = ncp->nc_vp;
+ if (__predict_false(vp == NULL)) {
if (iswht_ret != NULL) {
/*
* Restore the ISWHITEOUT flag saved earlier.
@@ -642,14 +648,11 @@ cache_lookup(struct vnode *dvp, const ch
/* found nothing */
hit = false;
}
- mutex_exit(&ncp->nc_lock);
+ mutex_exit(ncp->nc_lock);
mutex_exit(&cpup->cpu_lock);
return hit;
}
-
- vp = ncp->nc_vp;
- mutex_enter(vp->v_interlock);
- mutex_exit(&ncp->nc_lock);
+ KASSERT(vp->v_interlock == ncp->nc_lock);
mutex_exit(&cpup->cpu_lock);
/*
@@ -724,13 +727,12 @@ cache_lookup_raw(struct vnode *dvp, cons
*iswht_ret = (ncp->nc_flags & ISWHITEOUT) != 0;
}
COUNT(cpup, ncs_neghits);
- mutex_exit(&ncp->nc_lock);
+ mutex_exit(ncp->nc_lock);
mutex_exit(&cpup->cpu_lock);
/* found negative entry; vn is already null from above */
return true;
}
- mutex_enter(vp->v_interlock);
- mutex_exit(&ncp->nc_lock);
+ KASSERT(vp->v_interlock == ncp->nc_lock);
mutex_exit(&cpup->cpu_lock);
/*
@@ -775,6 +777,7 @@ cache_revlookup(struct vnode *vp, struct
struct nchcpu *cpup;
char *bp;
int error, nlen;
+ bool locked, again;
if (!doingcache)
goto out;
@@ -787,10 +790,12 @@ cache_revlookup(struct vnode *vp, struct
* is the only place these counters are incremented so no one
* will be racing with us to increment them.
*/
+ again = false;
+ retry:
cpup = curcpu()->ci_data.cpu_nch;
- mutex_enter(namecache_lock);
+ mutex_enter(&namecache_lock);
LIST_FOREACH(ncp, nvcpp, nc_vhash) {
- mutex_enter(&ncp->nc_lock);
+ mutex_enter(ncp->nc_lock);
if (ncp->nc_vp == vp &&
(dvp = ncp->nc_dvp) != NULL &&
dvp != vp) { /* avoid pesky . entries.. */
@@ -813,8 +818,8 @@ cache_revlookup(struct vnode *vp, struct
bp -= nlen;
if (bp <= bufp) {
*dvpp = NULL;
- mutex_exit(&ncp->nc_lock);
- mutex_exit(namecache_lock);
+ mutex_exit(ncp->nc_lock);
+ mutex_exit(&namecache_lock);
SDT_PROBE(vfs, namecache, revlookup,
fail, vp, ERANGE, 0, 0, 0);
return (ERANGE);
@@ -823,9 +828,18 @@ cache_revlookup(struct vnode *vp, struct
*bpp = bp;
}
- mutex_enter(dvp->v_interlock);
- mutex_exit(&ncp->nc_lock);
- mutex_exit(namecache_lock);
+
+ KASSERT(ncp->nc_lock != dvp->v_interlock);
+ locked = mutex_tryenter(dvp->v_interlock);
+ mutex_exit(ncp->nc_lock);
+ mutex_exit(&namecache_lock);
+ if (!locked) {
+ if (again) {
+ kpause("nchrace", false, 1, NULL);
+ }
+ again = true;
+ goto retry;
+ }
error = vcache_tryvget(dvp);
if (error) {
KASSERT(error == EBUSY);
@@ -841,10 +855,10 @@ cache_revlookup(struct vnode *vp, struct
0, 0, 0);
return (0);
}
- mutex_exit(&ncp->nc_lock);
+ mutex_exit(ncp->nc_lock);
}
COUNT(cpup, ncs_revmiss);
- mutex_exit(namecache_lock);
+ mutex_exit(&namecache_lock);
out:
*dvpp = NULL;
return (-1);
@@ -873,10 +887,10 @@ cache_enter(struct vnode *dvp, struct vn
SDT_PROBE(vfs, namecache, enter, done, vp, name, namelen, 0, 0);
if (numcache > desiredvnodes) {
- mutex_enter(namecache_lock);
+ mutex_enter(&namecache_lock);
cache_ev_forced.ev_count++;
cache_reclaim();
- mutex_exit(namecache_lock);
+ mutex_exit(&namecache_lock);
}
if (namelen > NCHNAMLEN) {
@@ -885,7 +899,7 @@ cache_enter(struct vnode *dvp, struct vn
} else
ncp = pool_cache_get(namecache_cache, PR_WAITOK);
- mutex_enter(namecache_lock);
+ mutex_enter(&namecache_lock);
numcache++;
/*
@@ -895,24 +909,27 @@ cache_enter(struct vnode *dvp, struct vn
oncp = cache_lookup_entry(dvp, name, namelen);
if (oncp) {
cache_invalidate(oncp);
- mutex_exit(&oncp->nc_lock);
+ mutex_exit(oncp->nc_lock);
}
/* Grab the vnode we just found. */
- mutex_enter(&ncp->nc_lock);
ncp->nc_vp = vp;
ncp->nc_flags = 0;
ncp->nc_hittime = 0;
- ncp->nc_gcqueue = NULL;
if (vp == NULL) {
/*
* For negative hits, save the ISWHITEOUT flag so we can
* restore it later when the cache entry is used again.
*/
ncp->nc_flags = cnflags & ISWHITEOUT;
+ ncp->nc_lock = mutex_obj_alloc(MUTEX_DEFAULT, IPL_NONE);
+ } else {
+ ncp->nc_lock = vp->v_interlock;
+ mutex_obj_hold(ncp->nc_lock);
}
/* Fill in cache info. */
+ mutex_enter(ncp->nc_lock);
ncp->nc_dvp = dvp;
LIST_INSERT_HEAD(&VNODE_TO_VIMPL(dvp)->vi_dnclist, ncp, nc_dvlist);
if (vp)
@@ -959,8 +976,8 @@ cache_enter(struct vnode *dvp, struct vn
nvcpp = &ncvhashtbl[NCVHASH(vp)];
LIST_INSERT_HEAD(nvcpp, ncp, nc_vhash);
}
- mutex_exit(&ncp->nc_lock);
- mutex_exit(namecache_lock);
+ mutex_exit(ncp->nc_lock);
+ mutex_exit(&namecache_lock);
}
/*
@@ -977,8 +994,8 @@ nchinit(void)
cache_dtor, NULL);
KASSERT(namecache_cache != NULL);
- namecache_lock = mutex_obj_alloc(MUTEX_DEFAULT, IPL_NONE);
-
+ mutex_init(&namecache_lock, MUTEX_DEFAULT, IPL_NONE);
+ mutex_init(&namecache_gc_lock, MUTEX_DEFAULT, IPL_NONE);
nchashtbl = hashinit(desiredvnodes, HASH_LIST, true, &nchash);
ncvhashtbl =
#ifdef NAMECACHE_ENTER_REVERSE
@@ -1009,10 +1026,6 @@ nchinit(void)
static int
cache_ctor(void *arg, void *obj, int flag)
{
- struct namecache *ncp;
-
- ncp = obj;
- mutex_init(&ncp->nc_lock, MUTEX_DEFAULT, IPL_NONE);
return 0;
}
@@ -1020,10 +1033,7 @@ cache_ctor(void *arg, void *obj, int fla
static void
cache_dtor(void *arg, void *obj)
{
- struct namecache *ncp;
- ncp = obj;
- mutex_destroy(&ncp->nc_lock);
}
/*
@@ -1060,7 +1070,7 @@ nchreinit(void)
#else
hashinit(desiredvnodes/8, HASH_LIST, true, &mask2);
#endif
- mutex_enter(namecache_lock);
+ mutex_enter(&namecache_lock);
cache_lock_cpus();
oldhash1 = nchashtbl;
oldmask1 = nchash;
@@ -1083,7 +1093,7 @@ nchreinit(void)
}
}
cache_unlock_cpus();
- mutex_exit(namecache_lock);
+ mutex_exit(&namecache_lock);
hashdone(oldhash1, HASH_LIST, oldmask1);
hashdone(oldhash2, HASH_LIST, oldmask2);
}
@@ -1097,16 +1107,16 @@ cache_purge1(struct vnode *vp, const cha
{
struct namecache *ncp, *ncnext;
- mutex_enter(namecache_lock);
+ mutex_enter(&namecache_lock);
if (flags & PURGE_PARENTS) {
SDT_PROBE(vfs, namecache, purge, parents, vp, 0, 0, 0, 0);
for (ncp = LIST_FIRST(&VNODE_TO_VIMPL(vp)->vi_nclist);
ncp != NULL; ncp = ncnext) {
ncnext = LIST_NEXT(ncp, nc_vlist);
- mutex_enter(&ncp->nc_lock);
+ mutex_enter(ncp->nc_lock);
cache_invalidate(ncp);
- mutex_exit(&ncp->nc_lock);
+ mutex_exit(ncp->nc_lock);
cache_disassociate(ncp);
}
}
@@ -1115,9 +1125,9 @@ cache_purge1(struct vnode *vp, const cha
for (ncp = LIST_FIRST(&VNODE_TO_VIMPL(vp)->vi_dnclist);
ncp != NULL; ncp = ncnext) {
ncnext = LIST_NEXT(ncp, nc_dvlist);
- mutex_enter(&ncp->nc_lock);
+ mutex_enter(ncp->nc_lock);
cache_invalidate(ncp);
- mutex_exit(&ncp->nc_lock);
+ mutex_exit(ncp->nc_lock);
cache_disassociate(ncp);
}
}
@@ -1126,11 +1136,11 @@ cache_purge1(struct vnode *vp, const cha
ncp = cache_lookup_entry(vp, name, namelen);
if (ncp) {
cache_invalidate(ncp);
- mutex_exit(&ncp->nc_lock);
+ mutex_exit(ncp->nc_lock);
cache_disassociate(ncp);
}
}
- mutex_exit(namecache_lock);
+ mutex_exit(&namecache_lock);
}
/*
@@ -1143,19 +1153,19 @@ cache_purgevfs(struct mount *mp)
struct namecache *ncp, *nxtcp;
SDT_PROBE(vfs, namecache, purge, vfs, mp, 0, 0, 0, 0);
- mutex_enter(namecache_lock);
+ mutex_enter(&namecache_lock);
for (ncp = TAILQ_FIRST(&nclruhead); ncp != NULL; ncp = nxtcp) {
nxtcp = TAILQ_NEXT(ncp, nc_lru);
- mutex_enter(&ncp->nc_lock);
+ mutex_enter(ncp->nc_lock);
if (ncp->nc_dvp != NULL && ncp->nc_dvp->v_mount == mp) {
/* Free the resources we had. */
cache_invalidate(ncp);
cache_disassociate(ncp);
}
- mutex_exit(&ncp->nc_lock);
+ mutex_exit(ncp->nc_lock);
}
cache_reclaim();
- mutex_exit(namecache_lock);
+ mutex_exit(&namecache_lock);
}
/*
@@ -1170,7 +1180,7 @@ cache_prune(int incache, int target)
struct namecache *ncp, *nxtcp, *sentinel;
int items, recent, tryharder;
- KASSERT(mutex_owned(namecache_lock));
+ KASSERT(mutex_owned(&namecache_lock));
SDT_PROBE(vfs, namecache, prune, done, incache, target, 0, 0, 0);
items = 0;
@@ -1198,13 +1208,13 @@ cache_prune(int incache, int target)
TAILQ_INSERT_TAIL(&nclruhead, ncp, nc_lru);
continue;
}
- mutex_enter(&ncp->nc_lock);
+ mutex_enter(ncp->nc_lock);
if (ncp->nc_dvp != NULL) {
cache_invalidate(ncp);
cache_disassociate(ncp);
incache--;
}
- mutex_exit(&ncp->nc_lock);
+ mutex_exit(ncp->nc_lock);
}
cache_ev_scan.ev_count += items;
}
@@ -1215,17 +1225,21 @@ cache_prune(int incache, int target)
static void
cache_reclaim(void)
{
+ CPU_INFO_ITERATOR cii;
+ struct cpu_info *ci;
+ struct nchcpu *cpup;
+ struct namecache_gc_queue queue;
struct namecache *ncp, *next;
int items;
- KASSERT(mutex_owned(namecache_lock));
+ KASSERT(mutex_owned(&namecache_lock));
/*
* If the number of extant entries not awaiting garbage collection
* exceeds the high water mark, then reclaim stale entries until we
* reach our low water mark.
*/
- items = numcache - cache_gcpend;
+ items = numcache - namecache_gc_pend;
if (items > (uint64_t)desiredvnodes * cache_hiwat / 100) {
cache_prune(items, (int)((uint64_t)desiredvnodes *
cache_lowat / 100));
@@ -1233,31 +1247,73 @@ cache_reclaim(void)
} else
cache_ev_under.ev_count++;
+ /* Chop the existing garbage collection list out. */
+ mutex_enter(&namecache_gc_lock);
+ queue = namecache_gc_queue;
+ items = namecache_gc_pend;
+ SLIST_INIT(&namecache_gc_queue);
+ namecache_gc_pend = 0;
+ mutex_exit(&namecache_gc_lock);
+
/*
- * Stop forward lookup activity on all CPUs and garbage collect dead
- * entries.
+ * Now disassociate all entries. We haven't locked out the reader
+ * side (cache_lookup_entry()) but the "next" pointers in the hash
+ * list will remain sufficiently consitent regardless of which
+ * version of the list the reader sees (see defs of LIST_FOREACH,
+ * LIST_NEXT);
*/
- cache_lock_cpus();
- ncp = cache_gcqueue;
- cache_gcqueue = NULL;
- items = cache_gcpend;
- cache_gcpend = 0;
- while (ncp != NULL) {
- next = ncp->nc_gcqueue;
+ SLIST_FOREACH(ncp, &queue, nc_gclist) {
cache_disassociate(ncp);
KASSERT(ncp->nc_dvp == NULL);
if (ncp->nc_hash.le_prev != NULL) {
LIST_REMOVE(ncp, nc_hash);
ncp->nc_hash.le_prev = NULL;
}
+ }
+
+ /*
+ * With that done, make sure our updates are visible on the bus, and
+ * make a pass to observe the status of all of the CPU locks. If we
+ * see a lock is unheld, we know the garbage collected entries can
+ * no longer visible to that CPU. If we see a lock IS held, we need
+ * to acquire and release it once to make sure that CPU is out of
+ * cache_lookup_entry(). Take the opportunity to refresh stats.
+ */
+ membar_sync(); /* stores above vs. reads below */
+ for (CPU_INFO_FOREACH(cii, ci)) {
+ cpup = ci->ci_data.cpu_nch;
+ if (__predict_false(mutex_owner(&cpup->cpu_lock) != NULL)) {
+ mutex_enter(&cpup->cpu_lock);
+ /* nothing */
+ mutex_exit(&cpup->cpu_lock);
+ }
+ UPDATE(cpup, ncs_goodhits);
+ UPDATE(cpup, ncs_neghits);
+ UPDATE(cpup, ncs_badhits);
+ UPDATE(cpup, ncs_falsehits);
+ UPDATE(cpup, ncs_miss);
+ UPDATE(cpup, ncs_long);
+ UPDATE(cpup, ncs_pass2);
+ UPDATE(cpup, ncs_2passes);
+ UPDATE(cpup, ncs_revhits);
+ UPDATE(cpup, ncs_revmiss);
+ }
+ membar_sync(); /* reads above vs. stores below */
+
+ /*
+ * Nobody else can see the cache entries any more. Make a final
+ * pass over the list and toss the contents.
+ */
+ SLIST_FOREACH_SAFE(ncp, &queue, nc_gclist, next) {
+ mutex_obj_free(ncp->nc_lock);
+ ncp->nc_lock = NULL;
if (ncp->nc_nlen > NCHNAMLEN) {
cache_dtor(NULL, ncp);
kmem_free(ncp, sizeof(*ncp) + ncp->nc_nlen);
} else
pool_cache_put(namecache_cache, ncp);
- ncp = next;
}
- cache_unlock_cpus();
+
numcache -= items;
cache_ev_gc.ev_count += items;
}
@@ -1273,10 +1329,10 @@ static void
cache_thread(void *arg)
{
- mutex_enter(namecache_lock);
+ mutex_enter(&namecache_lock);
for (;;) {
cache_reclaim();
- kpause("cachegc", false, hz, namecache_lock);
+ kpause("cachegc", false, hz, &namecache_lock);
}
}
@@ -1330,53 +1386,40 @@ namecache_count_2passes(void)
static int
cache_stat_sysctl(SYSCTLFN_ARGS)
{
- struct nchstats stats;
- struct nchcpu *my_cpup;
-#ifdef CACHE_STATS_CURRENT
CPU_INFO_ITERATOR cii;
struct cpu_info *ci;
-#endif /* CACHE_STATS_CURRENT */
if (oldp == NULL) {
- *oldlenp = sizeof(stats);
+ *oldlenp = sizeof(nchstats);
return 0;
}
- if (*oldlenp < sizeof(stats)) {
+ if (*oldlenp < sizeof(nchstats)) {
*oldlenp = 0;
return 0;
}
- /*
- * Take this CPU's per-cpu lock to hold off cache_reclaim()
- * from doing a stats update while doing minimal damage to
- * concurrent operations.
- */
sysctl_unlock();
- my_cpup = curcpu()->ci_data.cpu_nch;
- mutex_enter(&my_cpup->cpu_lock);
- stats = nchstats;
-#ifdef CACHE_STATS_CURRENT
+ mutex_enter(&namecache_lock);
for (CPU_INFO_FOREACH(cii, ci)) {
struct nchcpu *cpup = ci->ci_data.cpu_nch;
- ADD(stats, cpup, ncs_goodhits);
- ADD(stats, cpup, ncs_neghits);
- ADD(stats, cpup, ncs_badhits);
- ADD(stats, cpup, ncs_falsehits);
- ADD(stats, cpup, ncs_miss);
- ADD(stats, cpup, ncs_long);
- ADD(stats, cpup, ncs_pass2);
- ADD(stats, cpup, ncs_2passes);
- ADD(stats, cpup, ncs_revhits);
- ADD(stats, cpup, ncs_revmiss);
+ UPDATE(cpup, ncs_goodhits);
+ UPDATE(cpup, ncs_neghits);
+ UPDATE(cpup, ncs_badhits);
+ UPDATE(cpup, ncs_falsehits);
+ UPDATE(cpup, ncs_miss);
+
+ UPDATE(cpup, ncs_pass2);
+ UPDATE(cpup, ncs_2passes);
+ UPDATE(cpup, ncs_revhits);
+ UPDATE(cpup, ncs_revmiss);
}
-#endif /* CACHE_STATS_CURRENT */
- mutex_exit(&my_cpup->cpu_lock);
+ mutex_exit(&namecache_lock);
sysctl_relock();
- *oldlenp = sizeof(stats);
- return sysctl_copyout(l, &stats, oldp, sizeof(stats));
+ *oldlenp = sizeof(nchstats);
+ return sysctl_copyout(l, &nchstats, oldp, sizeof(nchstats));
}
static void
Index: src/sys/rump/include/rump/rump_namei.h
diff -u src/sys/rump/include/rump/rump_namei.h:1.33 src/sys/rump/include/rump/rump_namei.h:1.34
--- src/sys/rump/include/rump/rump_namei.h:1.33 Fri Sep 13 13:59:31 2019
+++ src/sys/rump/include/rump/rump_namei.h Sun Dec 1 13:39:53 2019
@@ -1,4 +1,4 @@
-/* $NetBSD: rump_namei.h,v 1.33 2019/09/13 13:59:31 christos Exp $ */
+/* $NetBSD: rump_namei.h,v 1.34 2019/12/01 13:39:53 ad Exp $ */
/*
Index: src/sys/sys/namei.h
diff -u src/sys/sys/namei.h:1.99 src/sys/sys/namei.h:1.100
--- src/sys/sys/namei.h:1.99 Fri Sep 13 13:59:31 2019
+++ src/sys/sys/namei.h Sun Dec 1 13:39:53 2019
@@ -1,4 +1,4 @@
-/* $NetBSD: namei.h,v 1.99 2019/09/13 13:59:31 christos Exp $ */
+/* $NetBSD: namei.h,v 1.100 2019/12/01 13:39:53 ad Exp $ */
/*
@@ -211,6 +211,7 @@ struct nameidata {
*
* - stable after initialization
* L namecache_lock
+ * G namecache_gc_lock
* C struct nchcpu::cpu_lock
* L/C insert needs L, read needs L or any C,
* must hold L and all C after (or during) delete before free
@@ -222,10 +223,10 @@ struct namecache {
TAILQ_ENTRY(namecache) nc_lru; /* L pseudo-lru chain */
LIST_ENTRY(namecache) nc_dvlist;/* L dvp's list of cache entries */
LIST_ENTRY(namecache) nc_vlist; /* L vp's list of cache entries */
+ SLIST_ENTRY(namecache) nc_gclist;/*G queue for garbage collection */
struct vnode *nc_dvp; /* N vnode of parent of name */
struct vnode *nc_vp; /* N vnode the name refers to */
- void *nc_gcqueue; /* N queue for garbage collection */
- kmutex_t nc_lock; /* lock on this entry */
+ kmutex_t *nc_lock; /* - lock on this entry */
int nc_hittime; /* N last time scored a hit */
int nc_flags; /* - copy of componentname ISWHITEOUT */
u_short nc_nlen; /* - length of name */
Index: src/sys/sys/namei.src
diff -u src/sys/sys/namei.src:1.43 src/sys/sys/namei.src:1.44
--- src/sys/sys/namei.src:1.43 Fri Sep 13 13:58:53 2019
+++ src/sys/sys/namei.src Sun Dec 1 13:39:53 2019
@@ -1,4 +1,4 @@
-/* $NetBSD: namei.src,v 1.43 2019/09/13 13:58:53 christos Exp $ */
+/* $NetBSD: namei.src,v 1.44 2019/12/01 13:39:53 ad Exp $ */
/*
* Copyright (c) 1985, 1989, 1991, 1993
@@ -203,6 +203,7 @@ NAMEIFL PARAMASK 0x02ee300 /* mask of pa
*
* - stable after initialization
* L namecache_lock
+ * G namecache_gc_lock
* C struct nchcpu::cpu_lock
* L/C insert needs L, read needs L or any C,
* must hold L and all C after (or during) delete before free
@@ -214,10 +215,10 @@ struct namecache {
TAILQ_ENTRY(namecache) nc_lru; /* L pseudo-lru chain */
LIST_ENTRY(namecache) nc_dvlist;/* L dvp's list of cache entries */
LIST_ENTRY(namecache) nc_vlist; /* L vp's list of cache entries */
+ SLIST_ENTRY(namecache) nc_gclist;/*G queue for garbage collection */
struct vnode *nc_dvp; /* N vnode of parent of name */
struct vnode *nc_vp; /* N vnode the name refers to */
- void *nc_gcqueue; /* N queue for garbage collection */
- kmutex_t nc_lock; /* lock on this entry */
+ kmutex_t *nc_lock; /* - lock on this entry */
int nc_hittime; /* N last time scored a hit */
int nc_flags; /* - copy of componentname ISWHITEOUT */
u_short nc_nlen; /* - length of name */
