[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-devel] [PATCH 1/2] rwlock: add per-cpu reader-writer locks



Per-cpu read-write locks allow for the fast path read case to have low overhead
by only setting/clearing a per-cpu variable for using the read lock.
The per-cpu read fast path also avoids locked compare swap operations which can
be particularly slow on coherent multi-socket systems, particularly if there is
heavy usage of the read lock itself.

The per-cpu reader-writer lock uses a global variable to control the read lock
fast path. This allows a writer to disable the fast path and ensure the readers
use the underlying read-write lock implementation.

Once the writer has taken the write lock and disabled the fast path, it must
poll the per-cpu variable for all CPU's which have entered the critical section
for the specific read-write lock the writer is attempting to take. This design
allows for a single per-cpu variable to be used for read/write locks belonging
to seperate data structures as long as multiple per-cpu read locks are not
simultaneously held by one particular cpu. This also means per-cpu
reader-writer locks are not recursion safe.

Slow path readers which are unblocked set the per-cpu variable and drop the
read lock. This simplifies the implementation and allows for fairness in the
underlying read-write lock to be taken advantage of.

There may be slightly more overhead on the per-cpu write lock path due to
checking each CPUs fast path read variable but this overhead is likely be hidden
by the required delay of waiting for readers to exit the critical section.
The loop is optimised to only iterate over the per-cpu data of active readers
of the rwlock.

Signed-off-by: Malcolm Crossley <malcolm.crossley@xxxxxxxxxx>
---
 xen/common/spinlock.c        | 32 ++++++++++++++++++++++++++++++++
 xen/include/asm-arm/percpu.h |  5 +++++
 xen/include/asm-x86/percpu.h |  6 ++++++
 xen/include/xen/percpu.h     |  4 ++++
 xen/include/xen/spinlock.h   | 37 +++++++++++++++++++++++++++++++++++++
 5 files changed, 84 insertions(+)

diff --git a/xen/common/spinlock.c b/xen/common/spinlock.c
index 7f89694..a526216 100644
--- a/xen/common/spinlock.c
+++ b/xen/common/spinlock.c
@@ -492,6 +492,38 @@ int _rw_is_write_locked(rwlock_t *lock)
     return (lock->lock == RW_WRITE_FLAG); /* writer in critical section? */
 }
 
+void percpu_write_lock(rwlock_t **per_cpudata, bool_t *writer_activating,
+               rwlock_t *rwlock)
+{
+    int cpu;
+    cpumask_t active_readers;
+
+    cpumask_copy(&active_readers, &cpu_online_map);
+    /* First take the write lock to protect against other writers */
+    write_lock(rwlock);
+
+    /* Now set the global variable so that readers start using read_lock */
+    *writer_activating = true;
+    smp_mb();
+
+    /* Check if there are any percpu readers in progress on our rwlock*/
+    do
+    {
+        for_each_cpu(cpu, &active_readers)
+        {
+            /* Remove any percpu readers not contending 
+             * from our check mask */
+            if ( per_cpu_ptr(per_cpudata, cpu) != rwlock )
+                cpumask_clear_cpu(cpu, &active_readers);
+        }
+        /* Check if we've cleared all percpu readers */
+        if ( cpumask_empty(&active_readers) )
+            break;
+        /* Give the coherency fabric a break */
+        cpu_relax();
+    } while ( 1 );
+}
+
 #ifdef LOCK_PROFILE
 
 struct lock_profile_anc {
diff --git a/xen/include/asm-arm/percpu.h b/xen/include/asm-arm/percpu.h
index 71e7649..c308a56 100644
--- a/xen/include/asm-arm/percpu.h
+++ b/xen/include/asm-arm/percpu.h
@@ -27,6 +27,11 @@ void percpu_init_areas(void);
 #define __get_cpu_var(var) \
     (*RELOC_HIDE(&per_cpu__##var, READ_SYSREG(TPIDR_EL2)))
 
+#define per_cpu_ptr(var, cpu)  \
+    (*RELOC_HIDE(&var, __per_cpu_offset[cpu]))
+#define __get_cpu_ptr(var) \
+    (*RELOC_HIDE(&var, READ_SYSREG(TPIDR_EL2)))
+
 #define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu__##name
 
 DECLARE_PER_CPU(unsigned int, cpu_id);
diff --git a/xen/include/asm-x86/percpu.h b/xen/include/asm-x86/percpu.h
index 604ff0d..51562b9 100644
--- a/xen/include/asm-x86/percpu.h
+++ b/xen/include/asm-x86/percpu.h
@@ -20,4 +20,10 @@ void percpu_init_areas(void);
 
 #define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu__##name
 
+#define __get_cpu_ptr(var) \
+    (*RELOC_HIDE(var, get_cpu_info()->per_cpu_offset))
+
+#define per_cpu_ptr(var, cpu)  \
+    (*RELOC_HIDE(var, __per_cpu_offset[cpu]))
+
 #endif /* __X86_PERCPU_H__ */
diff --git a/xen/include/xen/percpu.h b/xen/include/xen/percpu.h
index abe0b11..c896863 100644
--- a/xen/include/xen/percpu.h
+++ b/xen/include/xen/percpu.h
@@ -16,6 +16,10 @@
 /* Preferred on Xen. Also see arch-defined per_cpu(). */
 #define this_cpu(var)    __get_cpu_var(var)
 
+#define this_cpu_ptr(ptr)    __get_cpu_ptr(ptr)
+
+#define get_per_cpu_var(var)  (per_cpu__##var)
+
 /* Linux compatibility. */
 #define get_cpu_var(var) this_cpu(var)
 #define put_cpu_var(var)
diff --git a/xen/include/xen/spinlock.h b/xen/include/xen/spinlock.h
index fb0438e..f929f1b 100644
--- a/xen/include/xen/spinlock.h
+++ b/xen/include/xen/spinlock.h
@@ -3,6 +3,7 @@
 
 #include <asm/system.h>
 #include <asm/spinlock.h>
+#include <xen/stdbool.h>
 
 #ifndef NDEBUG
 struct lock_debug {
@@ -261,4 +262,40 @@ int _rw_is_write_locked(rwlock_t *lock);
 #define rw_is_locked(l)               _rw_is_locked(l)
 #define rw_is_write_locked(l)         _rw_is_write_locked(l)
 
+static inline void percpu_read_lock(rwlock_t **per_cpudata, bool_t 
*writer_activating,
+               rwlock_t *rwlock)
+{
+    /* Indicate this cpu is reading */
+    this_cpu_ptr(per_cpudata) = rwlock;
+    smp_mb();
+    /* Check if a writer is waiting */
+    if ( unlikely(*writer_activating) )
+    {
+        /* Let the waiting writer know we aren't holding the lock */
+        this_cpu_ptr(per_cpudata) = NULL;
+        /* Wait using the read lock to keep the lock fair */
+        read_lock(rwlock);
+        /* Set the per CPU data again and continue*/
+        this_cpu_ptr(per_cpudata) = rwlock;
+        /* Drop the read lock because we don't need it anymore */
+        read_unlock(rwlock);
+    }
+}
+
+static inline void percpu_read_unlock(rwlock_t **per_cpudata)
+{
+    this_cpu_ptr(per_cpudata) = NULL;
+    smp_wmb();
+}
+
+/* Don't inline percpu write lock as it's a complex function */
+void percpu_write_lock(rwlock_t **per_cpudata, bool_t *writer_activating,
+               rwlock_t *rwlock);
+
+static inline void percpu_write_unlock(bool_t *writer_activating, rwlock_t 
*rwlock)
+{
+    *writer_activating = false;
+    write_unlock(rwlock);
+}
+
 #endif /* __SPINLOCK_H__ */
-- 
1.7.12.4


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.