Date: Sat, 27 Sep 2014 19:59:47 -0500 From: Bryan Venteicher <bryanv@daemoninthecloset.org> To: "freebsd-hackers@freebsd.org" <freebsd-hackers@freebsd.org> Subject: Change uma_mtx to rwlock Message-ID: <CAMo0n6Q=P5H3%2BCqr8KjFRVLvWHZfJYnROVe4xF3DmKw95D%2B5zQ@mail.gmail.com>
next in thread | raw e-mail | index | archive | help
[-- Attachment #1 --]
Hi,
I'd appreciate some comments attached patch that changes the uma_mtx to a
rwlock.
At $JOB, we have machines with ~400GB RAM, with much of that being
allocated through UMA zones. We've observed that timeouts were sometimes
unexpectedly delayed by a half second or more. We tracked one of the
reasons for this down to when the page daemon was running, calling
uma_reclaim() -> zone_foreach(). zone_foreach() holds the uma_mtx while
zone_drain()'ing each zone. If uma_timeout() fires, it will block on the
uma_mtx when it tries to zone_timeout() each zone.
[-- Attachment #2 --]
From 9560c308229fe88c90114afb207fd2a96ed6324f Mon Sep 17 00:00:00 2001
From: Bryan Venteicher <bryanv@daemoninthecloset.org>
Date: Tue, 1 Jul 2014 16:04:23 -0500
Subject: [PATCH] Make the UMA lock a rwlock instead of a mutex
The zone_foreach() call that is done in uma_timeout() may block on the
UMA mutex while the each zone is drained in uma_reclaim(). This may
stall timeouts for an unacceptable amount of time if the draining takes
a long time.
---
sys/vm/uma_core.c | 49 +++++++++++++++++++++++++++----------------------
1 file changed, 27 insertions(+), 22 deletions(-)
diff --git a/sys/vm/uma_core.c b/sys/vm/uma_core.c
index 81b714a..30ba90a 100644
--- a/sys/vm/uma_core.c
+++ b/sys/vm/uma_core.c
@@ -135,8 +135,8 @@ static LIST_HEAD(,uma_keg) uma_kegs = LIST_HEAD_INITIALIZER(uma_kegs);
static LIST_HEAD(,uma_zone) uma_cachezones =
LIST_HEAD_INITIALIZER(uma_cachezones);
-/* This mutex protects the keg list */
-static struct mtx_padalign uma_mtx;
+/* This RW lock protects the keg list */
+static struct rwlock_padalign uma_rwlock;
/* Linked list of boot time pages */
static LIST_HEAD(,uma_slab) uma_boot_pages =
@@ -886,6 +886,7 @@ finished:
static void
zone_drain_wait(uma_zone_t zone, int waitok)
{
+ int wlock;
/*
* Set draining to interlock with zone_dtor() so we can release our
@@ -897,16 +898,20 @@ zone_drain_wait(uma_zone_t zone, int waitok)
while (zone->uz_flags & UMA_ZFLAG_DRAINING) {
if (waitok == M_NOWAIT)
goto out;
- mtx_unlock(&uma_mtx);
+ wlock = rw_wowned(&uma_rwlock);
+ rw_unlock(&uma_rwlock);
msleep(zone, zone->uz_lockptr, PVM, "zonedrain", 1);
- mtx_lock(&uma_mtx);
+ if (wlock != 0)
+ rw_wlock(&uma_rwlock);
+ else
+ rw_rlock(&uma_rwlock);
}
zone->uz_flags |= UMA_ZFLAG_DRAINING;
bucket_cache_drain(zone);
ZONE_UNLOCK(zone);
/*
* The DRAINING flag protects us from being freed while
- * we're running. Normally the uma_mtx would protect us but we
+ * we're running. Normally the uma_rwlock would protect us but we
* must be able to release and acquire the right lock for each keg.
*/
zone_foreach_keg(zone, &keg_drain);
@@ -1542,9 +1547,9 @@ keg_ctor(void *mem, int size, void *udata, int flags)
LIST_INSERT_HEAD(&keg->uk_zones, zone, uz_link);
- mtx_lock(&uma_mtx);
+ rw_wlock(&uma_rwlock);
LIST_INSERT_HEAD(&uma_kegs, keg, uk_link);
- mtx_unlock(&uma_mtx);
+ rw_wunlock(&uma_rwlock);
return (0);
}
@@ -1594,9 +1599,9 @@ zone_ctor(void *mem, int size, void *udata, int flags)
zone->uz_release = arg->release;
zone->uz_arg = arg->arg;
zone->uz_lockptr = &zone->uz_lock;
- mtx_lock(&uma_mtx);
+ rw_wlock(&uma_rwlock);
LIST_INSERT_HEAD(&uma_cachezones, zone, uz_link);
- mtx_unlock(&uma_mtx);
+ rw_wunlock(&uma_rwlock);
goto out;
}
@@ -1613,7 +1618,7 @@ zone_ctor(void *mem, int size, void *udata, int flags)
zone->uz_fini = arg->fini;
zone->uz_lockptr = &keg->uk_lock;
zone->uz_flags |= UMA_ZONE_SECONDARY;
- mtx_lock(&uma_mtx);
+ rw_wlock(&uma_rwlock);
ZONE_LOCK(zone);
LIST_FOREACH(z, &keg->uk_zones, uz_link) {
if (LIST_NEXT(z, uz_link) == NULL) {
@@ -1622,7 +1627,7 @@ zone_ctor(void *mem, int size, void *udata, int flags)
}
}
ZONE_UNLOCK(zone);
- mtx_unlock(&uma_mtx);
+ rw_wunlock(&uma_rwlock);
} else if (keg == NULL) {
if ((keg = uma_kcreate(zone, arg->size, arg->uminit, arg->fini,
arg->align, arg->flags)) == NULL)
@@ -1720,9 +1725,9 @@ zone_dtor(void *arg, int size, void *udata)
if (!(zone->uz_flags & UMA_ZFLAG_INTERNAL))
cache_drain(zone);
- mtx_lock(&uma_mtx);
+ rw_wlock(&uma_rwlock);
LIST_REMOVE(zone, uz_link);
- mtx_unlock(&uma_mtx);
+ rw_wunlock(&uma_rwlock);
/*
* XXX there are some races here where
* the zone can be drained but zone lock
@@ -1744,9 +1749,9 @@ zone_dtor(void *arg, int size, void *udata)
* We only destroy kegs from non secondary zones.
*/
if (keg != NULL && (zone->uz_flags & UMA_ZONE_SECONDARY) == 0) {
- mtx_lock(&uma_mtx);
+ rw_wlock(&uma_rwlock);
LIST_REMOVE(keg, uk_link);
- mtx_unlock(&uma_mtx);
+ rw_wunlock(&uma_rwlock);
zone_free_item(kegs, keg, NULL, SKIP_NONE);
}
ZONE_LOCK_FINI(zone);
@@ -1768,12 +1773,12 @@ zone_foreach(void (*zfunc)(uma_zone_t))
uma_keg_t keg;
uma_zone_t zone;
- mtx_lock(&uma_mtx);
+ rw_rlock(&uma_rwlock);
LIST_FOREACH(keg, &uma_kegs, uk_link) {
LIST_FOREACH(zone, &keg->uk_zones, uz_link)
zfunc(zone);
}
- mtx_unlock(&uma_mtx);
+ rw_runlock(&uma_rwlock);
}
/* Public functions */
@@ -1789,7 +1794,7 @@ uma_startup(void *bootmem, int boot_pages)
#ifdef UMA_DEBUG
printf("Creating uma keg headers zone and keg.\n");
#endif
- mtx_init(&uma_mtx, "UMA lock", NULL, MTX_DEF);
+ rw_init(&uma_rwlock, "UMA lock");
/* "manually" create the initial zone */
memset(&args, 0, sizeof(args));
@@ -3364,12 +3369,12 @@ sysctl_vm_zone_count(SYSCTL_HANDLER_ARGS)
int count;
count = 0;
- mtx_lock(&uma_mtx);
+ rw_rlock(&uma_rwlock);
LIST_FOREACH(kz, &uma_kegs, uk_link) {
LIST_FOREACH(z, &kz->uk_zones, uz_link)
count++;
}
- mtx_unlock(&uma_mtx);
+ rw_runlock(&uma_rwlock);
return (sysctl_handle_int(oidp, &count, 0, req));
}
@@ -3394,7 +3399,7 @@ sysctl_vm_zone_stats(SYSCTL_HANDLER_ARGS)
sbuf_new_for_sysctl(&sbuf, NULL, 128, req);
count = 0;
- mtx_lock(&uma_mtx);
+ rw_rlock(&uma_rwlock);
LIST_FOREACH(kz, &uma_kegs, uk_link) {
LIST_FOREACH(z, &kz->uk_zones, uz_link)
count++;
@@ -3470,7 +3475,7 @@ skip:
ZONE_UNLOCK(z);
}
}
- mtx_unlock(&uma_mtx);
+ rw_runlock(&uma_rwlock);
error = sbuf_finish(&sbuf);
sbuf_delete(&sbuf);
return (error);
--
1.8.5.4
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAMo0n6Q=P5H3%2BCqr8KjFRVLvWHZfJYnROVe4xF3DmKw95D%2B5zQ>