From: Pan Xinhui <[email protected]>
This patch aims to enable qspinlock on PPC. And on pseries platform, it also support
paravirt qspinlock.
Signed-off-by: Pan Xinhui <[email protected]>
---
arch/powerpc/include/asm/qspinlock.h | 37 +++++++++++++++
arch/powerpc/include/asm/qspinlock_paravirt.h | 36 +++++++++++++++
.../powerpc/include/asm/qspinlock_paravirt_types.h | 13 ++++++
arch/powerpc/include/asm/spinlock.h | 31 ++++++++-----
arch/powerpc/include/asm/spinlock_types.h | 4 ++
arch/powerpc/kernel/paravirt.c | 52 ++++++++++++++++++++++
arch/powerpc/lib/locks.c | 32 +++++++++++++
arch/powerpc/platforms/pseries/setup.c | 5 +++
8 files changed, 198 insertions(+), 12 deletions(-)
create mode 100644 arch/powerpc/include/asm/qspinlock.h
create mode 100644 arch/powerpc/include/asm/qspinlock_paravirt.h
create mode 100644 arch/powerpc/include/asm/qspinlock_paravirt_types.h
create mode 100644 arch/powerpc/kernel/paravirt.c
diff --git a/arch/powerpc/include/asm/qspinlock.h b/arch/powerpc/include/asm/qspinlock.h
new file mode 100644
index 0000000..d4e4dc3
--- /dev/null
+++ b/arch/powerpc/include/asm/qspinlock.h
@@ -0,0 +1,37 @@
+#ifndef _ASM_POWERPC_QSPINLOCK_H
+#define _ASM_POWERPC_QSPINLOCK_H
+
+#include <asm-generic/qspinlock_types.h>
+
+#define SPIN_THRESHOLD (1 << 15)
+#define queued_spin_unlock queued_spin_unlock
+
+static inline void native_queued_spin_unlock(struct qspinlock *lock)
+{
+ /* no load/store can be across the unlock()*/
+ smp_store_release((u8 *)lock, 0);
+}
+
+#ifdef CONFIG_PARAVIRT_SPINLOCKS
+
+#include <asm/qspinlock_paravirt.h>
+
+static inline void queued_spin_lock_slowpath(struct qspinlock *lock, u32 val)
+{
+ pv_queued_spin_lock(lock, val);
+}
+
+static inline void queued_spin_unlock(struct qspinlock *lock)
+{
+ pv_queued_spin_unlock(lock);
+}
+#else
+static inline void queued_spin_unlock(struct qspinlock *lock)
+{
+ native_queued_spin_unlock(lock);
+}
+#endif
+
+#include <asm-generic/qspinlock.h>
+
+#endif /* _ASM_POWERPC_QSPINLOCK_H */
diff --git a/arch/powerpc/include/asm/qspinlock_paravirt.h b/arch/powerpc/include/asm/qspinlock_paravirt.h
new file mode 100644
index 0000000..86e81c3
--- /dev/null
+++ b/arch/powerpc/include/asm/qspinlock_paravirt.h
@@ -0,0 +1,36 @@
+#ifndef CONFIG_PARAVIRT_SPINLOCKS
+#error "do not include this file"
+#endif
+
+#ifndef _ASM_QSPINLOCK_PARAVIRT_H
+#define _ASM_QSPINLOCK_PARAVIRT_H
+
+#include <asm/qspinlock_paravirt_types.h>
+
+extern void pv_lock_init(void);
+extern void native_queued_spin_lock_slowpath(struct qspinlock *lock, u32 val);
+extern void __pv_init_lock_hash(void);
+extern void __pv_queued_spin_lock_slowpath(struct qspinlock *lock, u32 val);
+extern void __pv_queued_spin_unlock(struct qspinlock *lock);
+
+static inline void pv_queued_spin_lock(struct qspinlock *lock, u32 val)
+{
+ pv_lock_op.lock(lock, val);
+}
+
+static inline void pv_queued_spin_unlock(struct qspinlock *lock)
+{
+ pv_lock_op.unlock(lock);
+}
+
+static inline void pv_wait(u8 *ptr, u8 val)
+{
+ pv_lock_op.wait(ptr, val, -1);
+}
+
+static inline void pv_kick(int cpu)
+{
+ pv_lock_op.kick(cpu);
+}
+
+#endif
diff --git a/arch/powerpc/include/asm/qspinlock_paravirt_types.h b/arch/powerpc/include/asm/qspinlock_paravirt_types.h
new file mode 100644
index 0000000..e1fdeb0
--- /dev/null
+++ b/arch/powerpc/include/asm/qspinlock_paravirt_types.h
@@ -0,0 +1,13 @@
+#ifndef _ASM_QSPINLOCK_PARAVIRT_TYPES_H
+#define _ASM_QSPINLOCK_PARAVIRT_TYPES_H
+
+struct pv_lock_ops {
+ void (*lock)(struct qspinlock *lock, u32 val);
+ void (*unlock)(struct qspinlock *lock);
+ void (*wait)(u8 *ptr, u8 val, int cpu);
+ void (*kick)(int cpu);
+};
+
+extern struct pv_lock_ops pv_lock_op;
+
+#endif
diff --git a/arch/powerpc/include/asm/spinlock.h b/arch/powerpc/include/asm/spinlock.h
index 523673d..3b65372 100644
--- a/arch/powerpc/include/asm/spinlock.h
+++ b/arch/powerpc/include/asm/spinlock.h
@@ -52,6 +52,24 @@
#define SYNC_IO
#endif
+#if defined(CONFIG_PPC_SPLPAR)
+/* We only yield to the hypervisor if we are in shared processor mode */
+#define SHARED_PROCESSOR (lppaca_shared_proc(local_paca->lppaca_ptr))
+extern void __spin_yield(arch_spinlock_t *lock);
+extern void __spin_yield_cpu(int cpu);
+extern void __spin_wake_cpu(int cpu);
+extern void __rw_yield(arch_rwlock_t *lock);
+#else /* SPLPAR */
+#define __spin_yield(x) barrier()
+#define __spin_yield_cpu(x) barrier()
+#define __spin_wake_cpu(x) barrier()
+#define __rw_yield(x) barrier()
+#define SHARED_PROCESSOR 0
+#endif
+
+#ifdef CONFIG_QUEUED_SPINLOCKS
+#include <asm/qspinlock.h>
+#else
static __always_inline int arch_spin_value_unlocked(arch_spinlock_t lock)
{
return lock.slock == 0;
@@ -106,18 +124,6 @@ static inline int arch_spin_trylock(arch_spinlock_t *lock)
* held. Conveniently, we have a word in the paca that holds this
* value.
*/
-
-#if defined(CONFIG_PPC_SPLPAR)
-/* We only yield to the hypervisor if we are in shared processor mode */
-#define SHARED_PROCESSOR (lppaca_shared_proc(local_paca->lppaca_ptr))
-extern void __spin_yield(arch_spinlock_t *lock);
-extern void __rw_yield(arch_rwlock_t *lock);
-#else /* SPLPAR */
-#define __spin_yield(x) barrier()
-#define __rw_yield(x) barrier()
-#define SHARED_PROCESSOR 0
-#endif
-
static inline void arch_spin_lock(arch_spinlock_t *lock)
{
CLEAR_IO_SYNC;
@@ -169,6 +175,7 @@ extern void arch_spin_unlock_wait(arch_spinlock_t *lock);
do { while (arch_spin_is_locked(lock)) cpu_relax(); } while (0)
#endif
+#endif /* !CONFIG_QUEUED_SPINLOCKS */
/*
* Read-write spinlocks, allowing multiple readers
* but only one writer.
diff --git a/arch/powerpc/include/asm/spinlock_types.h b/arch/powerpc/include/asm/spinlock_types.h
index 2351adc..bd7144e 100644
--- a/arch/powerpc/include/asm/spinlock_types.h
+++ b/arch/powerpc/include/asm/spinlock_types.h
@@ -5,11 +5,15 @@
# error "please don't include this file directly"
#endif
+#ifdef CONFIG_QUEUED_SPINLOCKS
+#include <asm-generic/qspinlock_types.h>
+#else
typedef struct {
volatile unsigned int slock;
} arch_spinlock_t;
#define __ARCH_SPIN_LOCK_UNLOCKED { 0 }
+#endif
typedef struct {
volatile signed int lock;
diff --git a/arch/powerpc/kernel/paravirt.c b/arch/powerpc/kernel/paravirt.c
new file mode 100644
index 0000000..355c9fb
--- /dev/null
+++ b/arch/powerpc/kernel/paravirt.c
@@ -0,0 +1,52 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/spinlock.h>
+
+static void __native_queued_spin_unlock(struct qspinlock *lock)
+{
+ native_queued_spin_unlock(lock);
+}
+
+static void __native_wait(u8 *ptr, u8 val, int cpu)
+{
+}
+
+static void __native_kick(int cpu)
+{
+}
+
+static void __pv_wait(u8 *ptr, u8 val, int cpu)
+{
+ HMT_low();
+ __spin_yield_cpu(cpu);
+ HMT_medium();
+}
+
+static void __pv_kick(int cpu)
+{
+ __spin_wake_cpu(cpu);
+}
+
+struct pv_lock_ops pv_lock_op = {
+ .lock = native_queued_spin_lock_slowpath,
+ .unlock = __native_queued_spin_unlock,
+ .wait = __native_wait,
+ .kick = __native_kick,
+};
+EXPORT_SYMBOL(pv_lock_op);
+
+void __init pv_lock_init(void)
+{
+ if (SHARED_PROCESSOR) {
+ __pv_init_lock_hash();
+ pv_lock_op.lock = __pv_queued_spin_lock_slowpath;
+ pv_lock_op.unlock = __pv_queued_spin_unlock;
+ pv_lock_op.wait = __pv_wait;
+ pv_lock_op.kick = __pv_kick;
+ }
+}
diff --git a/arch/powerpc/lib/locks.c b/arch/powerpc/lib/locks.c
index f7deebd..6e9d3bb 100644
--- a/arch/powerpc/lib/locks.c
+++ b/arch/powerpc/lib/locks.c
@@ -23,6 +23,35 @@
#include <asm/hvcall.h>
#include <asm/smp.h>
+void __spin_yield_cpu(int cpu)
+{
+ unsigned int holder_cpu = cpu, yield_count;
+
+ if (cpu == -1) {
+ plpar_hcall_norets(H_CEDE);
+ return;
+ }
+ BUG_ON(holder_cpu >= nr_cpu_ids);
+ yield_count = be32_to_cpu(lppaca_of(holder_cpu).yield_count);
+ if ((yield_count & 1) == 0)
+ return; /* virtual cpu is currently running */
+ rmb();
+ plpar_hcall_norets(H_CONFER,
+ get_hard_smp_processor_id(holder_cpu), yield_count);
+}
+EXPORT_SYMBOL_GPL(__spin_yield_cpu);
+
+void __spin_wake_cpu(int cpu)
+{
+ unsigned int holder_cpu = cpu;
+
+ BUG_ON(holder_cpu >= nr_cpu_ids);
+ plpar_hcall_norets(H_PROD,
+ get_hard_smp_processor_id(holder_cpu));
+}
+EXPORT_SYMBOL_GPL(__spin_wake_cpu);
+
+#ifndef CONFIG_QUEUED_SPINLOCKS
void __spin_yield(arch_spinlock_t *lock)
{
unsigned int lock_value, holder_cpu, yield_count;
@@ -42,6 +71,7 @@ void __spin_yield(arch_spinlock_t *lock)
get_hard_smp_processor_id(holder_cpu), yield_count);
}
EXPORT_SYMBOL_GPL(__spin_yield);
+#endif
/*
* Waiting for a read lock or a write lock on a rwlock...
@@ -69,6 +99,7 @@ void __rw_yield(arch_rwlock_t *rw)
}
#endif
+#ifndef CONFIG_QUEUED_SPINLOCKS
void arch_spin_unlock_wait(arch_spinlock_t *lock)
{
smp_mb();
@@ -84,3 +115,4 @@ void arch_spin_unlock_wait(arch_spinlock_t *lock)
}
EXPORT_SYMBOL(arch_spin_unlock_wait);
+#endif
diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
index 6e944fc..c9f056e 100644
--- a/arch/powerpc/platforms/pseries/setup.c
+++ b/arch/powerpc/platforms/pseries/setup.c
@@ -547,6 +547,11 @@ static void __init pSeries_setup_arch(void)
"%ld\n", rc);
}
}
+
+#ifdef CONFIG_PARAVIRT_SPINLOCKS
+ pv_lock_init();
+#endif
+
}
static int __init pSeries_init_panel(void)
-- 2.4.3
On 04/28/2016 06:55 AM, Pan Xinhui wrote:
> From: Pan Xinhui<[email protected]>
>
> This patch aims to enable qspinlock on PPC. And on pseries platform, it also support
> paravirt qspinlock.
>
> Signed-off-by: Pan Xinhui<[email protected]>
> ---
> arch/powerpc/include/asm/qspinlock.h | 37 +++++++++++++++
> arch/powerpc/include/asm/qspinlock_paravirt.h | 36 +++++++++++++++
> .../powerpc/include/asm/qspinlock_paravirt_types.h | 13 ++++++
> arch/powerpc/include/asm/spinlock.h | 31 ++++++++-----
> arch/powerpc/include/asm/spinlock_types.h | 4 ++
> arch/powerpc/kernel/paravirt.c | 52 ++++++++++++++++++++++
> arch/powerpc/lib/locks.c | 32 +++++++++++++
> arch/powerpc/platforms/pseries/setup.c | 5 +++
> 8 files changed, 198 insertions(+), 12 deletions(-)
> create mode 100644 arch/powerpc/include/asm/qspinlock.h
> create mode 100644 arch/powerpc/include/asm/qspinlock_paravirt.h
> create mode 100644 arch/powerpc/include/asm/qspinlock_paravirt_types.h
> create mode 100644 arch/powerpc/kernel/paravirt.c
>
>
This is just an enablement patch. You will also need a patch to activate
qspinlock for, at lease, some PPC configs. Right?
It has dependency on the pv_wait() patch that I sent out extend the
parameter list. Some performance data on how PPC system will perform
with and without qspinlock will also be helpful data points.
Cheers,
Longman
On 2016年04月29日 05:07, Waiman Long wrote:
> On 04/28/2016 06:55 AM, Pan Xinhui wrote:
>> From: Pan Xinhui<[email protected]>
>>
>> This patch aims to enable qspinlock on PPC. And on pseries platform, it also support
>> paravirt qspinlock.
>>
>> Signed-off-by: Pan Xinhui<[email protected]>
>> ---
>> arch/powerpc/include/asm/qspinlock.h | 37 +++++++++++++++
>> arch/powerpc/include/asm/qspinlock_paravirt.h | 36 +++++++++++++++
>> .../powerpc/include/asm/qspinlock_paravirt_types.h | 13 ++++++
>> arch/powerpc/include/asm/spinlock.h | 31 ++++++++-----
>> arch/powerpc/include/asm/spinlock_types.h | 4 ++
>> arch/powerpc/kernel/paravirt.c | 52 ++++++++++++++++++++++
>> arch/powerpc/lib/locks.c | 32 +++++++++++++
>> arch/powerpc/platforms/pseries/setup.c | 5 +++
>> 8 files changed, 198 insertions(+), 12 deletions(-)
>> create mode 100644 arch/powerpc/include/asm/qspinlock.h
>> create mode 100644 arch/powerpc/include/asm/qspinlock_paravirt.h
>> create mode 100644 arch/powerpc/include/asm/qspinlock_paravirt_types.h
>> create mode 100644 arch/powerpc/kernel/paravirt.c
>>
>>
>
> This is just an enablement patch. You will also need a patch to activate qspinlock for, at lease, some PPC configs. Right?
>
yep, I want to enable these config and makefile at last.
it just looks like
diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile
index 2da380f..ae7c2f1 100644
--- a/arch/powerpc/kernel/Makefile
+++ b/arch/powerpc/kernel/Makefile
@@ -50,6 +50,7 @@ obj-$(CONFIG_PPC_970_NAP) += idle_power4.o
obj-$(CONFIG_PPC_P7_NAP) += idle_power7.o
procfs-y := proc_powerpc.o
obj-$(CONFIG_PROC_FS) += $(procfs-y)
+obj-$(CONFIG_PARAVIRT_SPINLOCKS) += paravirt.o
rtaspci-$(CONFIG_PPC64)-$(CONFIG_PCI) := rtas_pci.o
obj-$(CONFIG_PPC_RTAS) += rtas.o rtas-rtc.o $(rtaspci-y-y)
obj-$(CONFIG_PPC_RTAS_DAEMON) += rtasd.o
diff --git a/arch/powerpc/platforms/pseries/Kconfig b/arch/powerpc/platforms/pseries/Kconfig
index bec90fb..46632e4 100644
--- a/arch/powerpc/platforms/pseries/Kconfig
+++ b/arch/powerpc/platforms/pseries/Kconfig
@@ -21,6 +21,7 @@ config PPC_PSERIES
select HOTPLUG_CPU if SMP
select ARCH_RANDOM
select PPC_DOORBELL
+ select ARCH_USE_QUEUED_SPINLOCKS
default y
config PPC_SPLPAR
@@ -127,3 +128,11 @@ config HV_PERF_CTRS
systems. 24x7 is available on Power 8 systems.
If unsure, select Y.
+
+config PARAVIRT_SPINLOCKS
+ bool "Paravirtialization support for qspinlock"
+ depends on PPC_SPLPAR && QUEUED_SPINLOCKS
+ default y
+ help
+ If platform supports virtualization, for example PowerVM, this option
+ can let guest have a better performace.
--
2.4.3
> It has dependency on the pv_wait() patch that I sent out extend the parameter list. Some performance data on how PPC system will perform with and without qspinlock will also be helpful data points.
>
For now, pv_wait defined in ppc is static inline void pv_wait(u8 *ptr, u8 val)
My plan is that waiting your patch goes in kernel tree first, then I send out another patch to extend the parameter list.
yes, I need copy some performance data in my patch's comments.
thanks
xinhui
> Cheers,
> Longman
>