[PATCH v5 0/7] riscv: Add qspinlock/qrwlock

[PATCH v5 0/7] riscv: Add qspinlock/qrwlock

[guoren]

From: Guo Ren <guoren@linux.alibaba.com>

Current riscv is still using baby spinlock implementation. It'll cause
fairness and cache line bouncing problems. Many people are involved
and pay the efforts to improve it:

 - The first version of patch was made in 2019.1:
   https://lore.kernel.org/linux-riscv/20190211043829.30096-1-michaeljclark@mac.com/#r

 - The second version was made in 2020.11:
   https://lore.kernel.org/linux-riscv/1606225437-22948-2-git-send-email-guoren@kernel.org/

 - A good discussion at Platform HSC.2021-03-08:
   https://drive.google.com/drive/folders/1ooqdnIsYx7XKor5O1XTtM6D1CHp4hc0p

Hope your comments and Tested-by or Co-developed-by or Reviewed-by ...

Let's kick the qspinlock into riscv right now (Also for the
architecture which hasn't xchg16 atomic instruction.)

Change V5:
 - Fixup #endif comment typo by Waiman
 - Remove cmpxchg coding convention patches which will get into a
   separate patchset later by Arnd's advice
 - Try to involve more architectures in the discussion

Change V4:
 - Remove custom sub-word xchg implementation
 - Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32 in locking/qspinlock

Change V3:
 - Coding convention by Peter Zijlstra's advices

Change V2:
 - Coding convention in cmpxchg.h
 - Re-implement short xchg
 - Remove char & cmpxchg implementations

Guo Ren (6):
  locking/qspinlock: Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32
  csky: Convert custom spinlock/rwlock to generic qspinlock/qrwlock
  powerpc/qspinlock: Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32
  openrisc: qspinlock: Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32
  sparc: qspinlock: Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32
  xtensa: qspinlock: Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32

Michael Clark (1):
  riscv: Convert custom spinlock/rwlock to generic qspinlock/qrwlock

 arch/csky/Kconfig                       |   2 +
 arch/csky/include/asm/Kbuild            |   2 +
 arch/csky/include/asm/spinlock.h        |  82 +--------------
 arch/csky/include/asm/spinlock_types.h  |  16 +--
 arch/openrisc/Kconfig                   |   1 +
 arch/powerpc/Kconfig                    |   1 +
 arch/riscv/Kconfig                      |   3 +
 arch/riscv/include/asm/Kbuild           |   3 +
 arch/riscv/include/asm/spinlock.h       | 126 +-----------------------
 arch/riscv/include/asm/spinlock_types.h |  15 +--
 arch/sparc/Kconfig                      |   1 +
 arch/xtensa/Kconfig                     |   1 +
 kernel/Kconfig.locks                    |   3 +
 kernel/locking/qspinlock.c              |  46 +++++----
 14 files changed, 49 insertions(+), 253 deletions(-)

-- 
2.17.1

[PATCH v5 1/7] locking/qspinlock: Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32

[guoren]

From: Guo Ren <guoren@linux.alibaba.com>

Some architectures don't have sub-word swap atomic instruction,
they only have the full word's one.

The sub-word swap only improve the performance when:
NR_CPUS < 16K
 *  0- 7: locked byte
 *     8: pending
 *  9-15: not used
 * 16-17: tail index
 * 18-31: tail cpu (+1)

The 9-15 bits are wasted to use xchg16 in xchg_tail.

Please let architecture select xchg16/xchg32 to implement
xchg_tail.

Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Waiman Long <longman@redhat.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Anup Patel <anup@brainfault.org>
---
 kernel/Kconfig.locks       |  3 +++
 kernel/locking/qspinlock.c | 46 +++++++++++++++++++++-----------------
 2 files changed, 28 insertions(+), 21 deletions(-)

diff --git a/kernel/Kconfig.locks b/kernel/Kconfig.locks
index 3de8fd11873b..d02f1261f73f 100644
--- a/kernel/Kconfig.locks
+++ b/kernel/Kconfig.locks
@@ -239,6 +239,9 @@ config LOCK_SPIN_ON_OWNER
 config ARCH_USE_QUEUED_SPINLOCKS
 	bool
 
+config ARCH_USE_QUEUED_SPINLOCKS_XCHG32
+	bool
+
 config QUEUED_SPINLOCKS
 	def_bool y if ARCH_USE_QUEUED_SPINLOCKS
 	depends on SMP
diff --git a/kernel/locking/qspinlock.c b/kernel/locking/qspinlock.c
index cbff6ba53d56..4bfaa969bd15 100644
--- a/kernel/locking/qspinlock.c
+++ b/kernel/locking/qspinlock.c
@@ -163,26 +163,6 @@ static __always_inline void clear_pending_set_locked(struct qspinlock *lock)
 	WRITE_ONCE(lock->locked_pending, _Q_LOCKED_VAL);
 }
 
-/*
- * xchg_tail - Put in the new queue tail code word & retrieve previous one
- * @lock : Pointer to queued spinlock structure
- * @tail : The new queue tail code word
- * Return: The previous queue tail code word
- *
- * xchg(lock, tail), which heads an address dependency
- *
- * p,*,* -> n,*,* ; prev = xchg(lock, node)
- */
-static __always_inline u32 xchg_tail(struct qspinlock *lock, u32 tail)
-{
-	/*
-	 * We can use relaxed semantics since the caller ensures that the
-	 * MCS node is properly initialized before updating the tail.
-	 */
-	return (u32)xchg_relaxed(&lock->tail,
-				 tail >> _Q_TAIL_OFFSET) << _Q_TAIL_OFFSET;
-}
-
 #else /* _Q_PENDING_BITS == 8 */
 
 /**
@@ -206,6 +186,30 @@ static __always_inline void clear_pending_set_locked(struct qspinlock *lock)
 {
 	atomic_add(-_Q_PENDING_VAL + _Q_LOCKED_VAL, &lock->val);
 }
+#endif /* _Q_PENDING_BITS == 8 */
+
+#if _Q_PENDING_BITS == 8 && !defined(CONFIG_ARCH_USE_QUEUED_SPINLOCKS_XCHG32)
+/*
+ * xchg_tail - Put in the new queue tail code word & retrieve previous one
+ * @lock : Pointer to queued spinlock structure
+ * @tail : The new queue tail code word
+ * Return: The previous queue tail code word
+ *
+ * xchg(lock, tail), which heads an address dependency
+ *
+ * p,*,* -> n,*,* ; prev = xchg(lock, node)
+ */
+static __always_inline u32 xchg_tail(struct qspinlock *lock, u32 tail)
+{
+	/*
+	 * We can use relaxed semantics since the caller ensures that the
+	 * MCS node is properly initialized before updating the tail.
+	 */
+	return (u32)xchg_relaxed(&lock->tail,
+				 tail >> _Q_TAIL_OFFSET) << _Q_TAIL_OFFSET;
+}
+
+#else
 
 /**
  * xchg_tail - Put in the new queue tail code word & retrieve previous one
@@ -236,7 +240,7 @@ static __always_inline u32 xchg_tail(struct qspinlock *lock, u32 tail)
 	}
 	return old;
 }
-#endif /* _Q_PENDING_BITS == 8 */
+#endif
 
 /**
  * queued_fetch_set_pending_acquire - fetch the whole lock value and set pending
-- 
2.17.1

[PATCH v5 2/7] riscv: Convert custom spinlock/rwlock to generic qspinlock/qrwlock

[guoren]

From: Michael Clark <michaeljclark@mac.com>

Update the RISC-V port to use the generic qspinlock and qrwlock.

This patch requires support for xchg_xtail for full-word which
are added by a previous patch:

Guo added select ARCH_USE_QUEUED_SPINLOCKS_XCHG32 in Kconfig

Guo fixed up compile error which made by below include sequence:
+#include <asm/qrwlock.h>
+#include <asm/qspinlock.h>

Signed-off-by: Michael Clark <michaeljclark@mac.com>
Co-developed-by: Guo Ren <guoren@linux.alibaba.com>
Tested-by: Guo Ren <guoren@linux.alibaba.com>
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Link: https://lore.kernel.org/linux-riscv/20190211043829.30096-3-michaeljclark@mac.com/
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Anup Patel <anup@brainfault.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Palmer Dabbelt <palmerdabbelt@google.com>
---
 arch/riscv/Kconfig                      |   3 +
 arch/riscv/include/asm/Kbuild           |   3 +
 arch/riscv/include/asm/spinlock.h       | 126 +-----------------------
 arch/riscv/include/asm/spinlock_types.h |  15 +--
 4 files changed, 11 insertions(+), 136 deletions(-)

diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
index 87d7b52f278f..67cc65ba1ea1 100644
--- a/arch/riscv/Kconfig
+++ b/arch/riscv/Kconfig
@@ -33,6 +33,9 @@ config RISCV
 	select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT if MMU
 	select ARCH_WANT_FRAME_POINTERS
 	select ARCH_WANT_HUGE_PMD_SHARE if 64BIT
+	select ARCH_USE_QUEUED_RWLOCKS
+	select ARCH_USE_QUEUED_SPINLOCKS
+	select ARCH_USE_QUEUED_SPINLOCKS_XCHG32
 	select CLONE_BACKWARDS
 	select CLINT_TIMER if !MMU
 	select COMMON_CLK
diff --git a/arch/riscv/include/asm/Kbuild b/arch/riscv/include/asm/Kbuild
index 445ccc97305a..750c1056b90f 100644
--- a/arch/riscv/include/asm/Kbuild
+++ b/arch/riscv/include/asm/Kbuild
@@ -3,5 +3,8 @@ generic-y += early_ioremap.h
 generic-y += extable.h
 generic-y += flat.h
 generic-y += kvm_para.h
+generic-y += mcs_spinlock.h
+generic-y += qrwlock.h
+generic-y += qspinlock.h
 generic-y += user.h
 generic-y += vmlinux.lds.h
diff --git a/arch/riscv/include/asm/spinlock.h b/arch/riscv/include/asm/spinlock.h
index f4f7fa1b7ca8..a557de67a425 100644
--- a/arch/riscv/include/asm/spinlock.h
+++ b/arch/riscv/include/asm/spinlock.h
@@ -7,129 +7,7 @@
 #ifndef _ASM_RISCV_SPINLOCK_H
 #define _ASM_RISCV_SPINLOCK_H
 
-#include <linux/kernel.h>
-#include <asm/current.h>
-#include <asm/fence.h>
-
-/*
- * Simple spin lock operations.  These provide no fairness guarantees.
- */
-
-/* FIXME: Replace this with a ticket lock, like MIPS. */
-
-#define arch_spin_is_locked(x)	(READ_ONCE((x)->lock) != 0)
-
-static inline void arch_spin_unlock(arch_spinlock_t *lock)
-{
-	smp_store_release(&lock->lock, 0);
-}
-
-static inline int arch_spin_trylock(arch_spinlock_t *lock)
-{
-	int tmp = 1, busy;
-
-	__asm__ __volatile__ (
-		"	amoswap.w %0, %2, %1\n"
-		RISCV_ACQUIRE_BARRIER
-		: "=r" (busy), "+A" (lock->lock)
-		: "r" (tmp)
-		: "memory");
-
-	return !busy;
-}
-
-static inline void arch_spin_lock(arch_spinlock_t *lock)
-{
-	while (1) {
-		if (arch_spin_is_locked(lock))
-			continue;
-
-		if (arch_spin_trylock(lock))
-			break;
-	}
-}
-
-/***********************************************************/
-
-static inline void arch_read_lock(arch_rwlock_t *lock)
-{
-	int tmp;
-
-	__asm__ __volatile__(
-		"1:	lr.w	%1, %0\n"
-		"	bltz	%1, 1b\n"
-		"	addi	%1, %1, 1\n"
-		"	sc.w	%1, %1, %0\n"
-		"	bnez	%1, 1b\n"
-		RISCV_ACQUIRE_BARRIER
-		: "+A" (lock->lock), "=&r" (tmp)
-		:: "memory");
-}
-
-static inline void arch_write_lock(arch_rwlock_t *lock)
-{
-	int tmp;
-
-	__asm__ __volatile__(
-		"1:	lr.w	%1, %0\n"
-		"	bnez	%1, 1b\n"
-		"	li	%1, -1\n"
-		"	sc.w	%1, %1, %0\n"
-		"	bnez	%1, 1b\n"
-		RISCV_ACQUIRE_BARRIER
-		: "+A" (lock->lock), "=&r" (tmp)
-		:: "memory");
-}
-
-static inline int arch_read_trylock(arch_rwlock_t *lock)
-{
-	int busy;
-
-	__asm__ __volatile__(
-		"1:	lr.w	%1, %0\n"
-		"	bltz	%1, 1f\n"
-		"	addi	%1, %1, 1\n"
-		"	sc.w	%1, %1, %0\n"
-		"	bnez	%1, 1b\n"
-		RISCV_ACQUIRE_BARRIER
-		"1:\n"
-		: "+A" (lock->lock), "=&r" (busy)
-		:: "memory");
-
-	return !busy;
-}
-
-static inline int arch_write_trylock(arch_rwlock_t *lock)
-{
-	int busy;
-
-	__asm__ __volatile__(
-		"1:	lr.w	%1, %0\n"
-		"	bnez	%1, 1f\n"
-		"	li	%1, -1\n"
-		"	sc.w	%1, %1, %0\n"
-		"	bnez	%1, 1b\n"
-		RISCV_ACQUIRE_BARRIER
-		"1:\n"
-		: "+A" (lock->lock), "=&r" (busy)
-		:: "memory");
-
-	return !busy;
-}
-
-static inline void arch_read_unlock(arch_rwlock_t *lock)
-{
-	__asm__ __volatile__(
-		RISCV_RELEASE_BARRIER
-		"	amoadd.w x0, %1, %0\n"
-		: "+A" (lock->lock)
-		: "r" (-1)
-		: "memory");
-}
-
-static inline void arch_write_unlock(arch_rwlock_t *lock)
-{
-	smp_store_release(&lock->lock, 0);
-}
+#include <asm/qspinlock.h>
+#include <asm/qrwlock.h>
 
 #endif /* _ASM_RISCV_SPINLOCK_H */
diff --git a/arch/riscv/include/asm/spinlock_types.h b/arch/riscv/include/asm/spinlock_types.h
index f398e7638dd6..d033a973f287 100644
--- a/arch/riscv/include/asm/spinlock_types.h
+++ b/arch/riscv/include/asm/spinlock_types.h
@@ -6,20 +6,11 @@
 #ifndef _ASM_RISCV_SPINLOCK_TYPES_H
 #define _ASM_RISCV_SPINLOCK_TYPES_H
 
-#ifndef __LINUX_SPINLOCK_TYPES_H
+#if !defined(__LINUX_SPINLOCK_TYPES_H) && !defined(_ASM_RISCV_SPINLOCK_H)
 # error "please don't include this file directly"
 #endif
 
-typedef struct {
-	volatile unsigned int lock;
-} arch_spinlock_t;
-
-#define __ARCH_SPIN_LOCK_UNLOCKED	{ 0 }
-
-typedef struct {
-	volatile unsigned int lock;
-} arch_rwlock_t;
-
-#define __ARCH_RW_LOCK_UNLOCKED		{ 0 }
+#include <asm-generic/qspinlock_types.h>
+#include <asm-generic/qrwlock_types.h>
 
 #endif /* _ASM_RISCV_SPINLOCK_TYPES_H */
-- 
2.17.1

[PATCH v5 3/7] csky: Convert custom spinlock/rwlock to generic qspinlock/qrwlock

[guoren]

From: Guo Ren <guoren@linux.alibaba.com>

Update the C-SKY port to use the generic qspinlock and qrwlock.

C-SKY only support ldex.w/stex.w with word(double word) size &
align access. So it must select XCHG32 to let qspinlock only use
word atomic xchg_tail.

Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Waiman Long <longman@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Cc: Arnd Bergmann <arnd@arndb.de>
---
 arch/csky/Kconfig                      |  2 +
 arch/csky/include/asm/Kbuild           |  2 +
 arch/csky/include/asm/spinlock.h       | 82 +-------------------------
 arch/csky/include/asm/spinlock_types.h | 16 +----
 4 files changed, 6 insertions(+), 96 deletions(-)

diff --git a/arch/csky/Kconfig b/arch/csky/Kconfig
index 34e91224adc3..5910eb6ddde2 100644
--- a/arch/csky/Kconfig
+++ b/arch/csky/Kconfig
@@ -8,6 +8,8 @@ config CSKY
 	select ARCH_HAS_SYNC_DMA_FOR_DEVICE
 	select ARCH_USE_BUILTIN_BSWAP
 	select ARCH_USE_QUEUED_RWLOCKS
+	select ARCH_USE_QUEUED_SPINLOCKS
+	select ARCH_USE_QUEUED_SPINLOCKS_XCHG32
 	select ARCH_WANT_FRAME_POINTERS if !CPU_CK610
 	select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT
 	select COMMON_CLK
diff --git a/arch/csky/include/asm/Kbuild b/arch/csky/include/asm/Kbuild
index cc24bb8e539f..2a2d09963bb9 100644
--- a/arch/csky/include/asm/Kbuild
+++ b/arch/csky/include/asm/Kbuild
@@ -2,6 +2,8 @@
 generic-y += asm-offsets.h
 generic-y += gpio.h
 generic-y += kvm_para.h
+generic-y += mcs_spinlock.h
 generic-y += qrwlock.h
+generic-y += qspinlock.h
 generic-y += user.h
 generic-y += vmlinux.lds.h
diff --git a/arch/csky/include/asm/spinlock.h b/arch/csky/include/asm/spinlock.h
index 69f5aa249c5f..fcff36753c25 100644
--- a/arch/csky/include/asm/spinlock.h
+++ b/arch/csky/include/asm/spinlock.h
@@ -3,87 +3,7 @@
 #ifndef __ASM_CSKY_SPINLOCK_H
 #define __ASM_CSKY_SPINLOCK_H
 
-#include <linux/spinlock_types.h>
-#include <asm/barrier.h>
-
-/*
- * Ticket-based spin-locking.
- */
-static inline void arch_spin_lock(arch_spinlock_t *lock)
-{
-	arch_spinlock_t lockval;
-	u32 ticket_next = 1 << TICKET_NEXT;
-	u32 *p = &lock->lock;
-	u32 tmp;
-
-	asm volatile (
-		"1:	ldex.w		%0, (%2) \n"
-		"	mov		%1, %0	 \n"
-		"	add		%0, %3	 \n"
-		"	stex.w		%0, (%2) \n"
-		"	bez		%0, 1b   \n"
-		: "=&r" (tmp), "=&r" (lockval)
-		: "r"(p), "r"(ticket_next)
-		: "cc");
-
-	while (lockval.tickets.next != lockval.tickets.owner)
-		lockval.tickets.owner = READ_ONCE(lock->tickets.owner);
-
-	smp_mb();
-}
-
-static inline int arch_spin_trylock(arch_spinlock_t *lock)
-{
-	u32 tmp, contended, res;
-	u32 ticket_next = 1 << TICKET_NEXT;
-	u32 *p = &lock->lock;
-
-	do {
-		asm volatile (
-		"	ldex.w		%0, (%3)   \n"
-		"	movi		%2, 1	   \n"
-		"	rotli		%1, %0, 16 \n"
-		"	cmpne		%1, %0     \n"
-		"	bt		1f         \n"
-		"	movi		%2, 0	   \n"
-		"	add		%0, %0, %4 \n"
-		"	stex.w		%0, (%3)   \n"
-		"1:				   \n"
-		: "=&r" (res), "=&r" (tmp), "=&r" (contended)
-		: "r"(p), "r"(ticket_next)
-		: "cc");
-	} while (!res);
-
-	if (!contended)
-		smp_mb();
-
-	return !contended;
-}
-
-static inline void arch_spin_unlock(arch_spinlock_t *lock)
-{
-	smp_mb();
-	WRITE_ONCE(lock->tickets.owner, lock->tickets.owner + 1);
-}
-
-static inline int arch_spin_value_unlocked(arch_spinlock_t lock)
-{
-	return lock.tickets.owner == lock.tickets.next;
-}
-
-static inline int arch_spin_is_locked(arch_spinlock_t *lock)
-{
-	return !arch_spin_value_unlocked(READ_ONCE(*lock));
-}
-
-static inline int arch_spin_is_contended(arch_spinlock_t *lock)
-{
-	struct __raw_tickets tickets = READ_ONCE(lock->tickets);
-
-	return (tickets.next - tickets.owner) > 1;
-}
-#define arch_spin_is_contended	arch_spin_is_contended
-
+#include <asm/qspinlock.h>
 #include <asm/qrwlock.h>
 
 #endif /* __ASM_CSKY_SPINLOCK_H */
diff --git a/arch/csky/include/asm/spinlock_types.h b/arch/csky/include/asm/spinlock_types.h
index 8ff0f6ff3a00..757594760e65 100644
--- a/arch/csky/include/asm/spinlock_types.h
+++ b/arch/csky/include/asm/spinlock_types.h
@@ -7,21 +7,7 @@
 # error "please don't include this file directly"
 #endif
 
-#define TICKET_NEXT	16
-
-typedef struct {
-	union {
-		u32 lock;
-		struct __raw_tickets {
-			/* little endian */
-			u16 owner;
-			u16 next;
-		} tickets;
-	};
-} arch_spinlock_t;
-
-#define __ARCH_SPIN_LOCK_UNLOCKED	{ { 0 } }
-
+#include <asm-generic/qspinlock_types.h>
 #include <asm-generic/qrwlock_types.h>
 
 #endif /* __ASM_CSKY_SPINLOCK_TYPES_H */
-- 
2.17.1

[PATCH v5 4/7] powerpc/qspinlock: Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32

[guoren]

From: Guo Ren <guoren@linux.alibaba.com>

We don't have native hw xchg16 instruction, so let qspinlock
generic code to deal with it.

Using the full-word atomic xchg instructions implement xchg16 has
the semantic risk for atomic operations.

This patch cancels the dependency of on qspinlock generic code on
architecture's xchg16.

Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 386ae12d8523..69ec4ade6521 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -151,6 +151,7 @@ config PPC
 	select ARCH_USE_CMPXCHG_LOCKREF		if PPC64
 	select ARCH_USE_QUEUED_RWLOCKS		if PPC_QUEUED_SPINLOCKS
 	select ARCH_USE_QUEUED_SPINLOCKS	if PPC_QUEUED_SPINLOCKS
+	select ARCH_USE_QUEUED_SPINLOCKS_XCHG32	if PPC_QUEUED_SPINLOCKS
 	select ARCH_WANT_IPC_PARSE_VERSION
 	select ARCH_WANT_IRQS_OFF_ACTIVATE_MM
 	select ARCH_WANT_LD_ORPHAN_WARN
-- 
2.17.1

Re: [PATCH v5 4/7] powerpc/qspinlock: Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32

[Christophe Leroy]

Le 28/03/2021 à 08:30, guoren@kernel.org a écrit :
> From: Guo Ren <guoren@linux.alibaba.com>
> 
> We don't have native hw xchg16 instruction, so let qspinlock
> generic code to deal with it.

We have lharx/sthcx pair on some versions of powerpc.

See https://patchwork.ozlabs.org/project/linuxppc-dev/patch/20201107032328.2454582-1-npiggin@gmail.com/

Christophe

> 
> Using the full-word atomic xchg instructions implement xchg16 has
> the semantic risk for atomic operations.
> 
> This patch cancels the dependency of on qspinlock generic code on
> architecture's xchg16.
> 
> Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
> Cc: Michael Ellerman <mpe@ellerman.id.au>
> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
> Cc: Paul Mackerras <paulus@samba.org>
> ---
>   arch/powerpc/Kconfig | 1 +
>   1 file changed, 1 insertion(+)
> 
> diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
> index 386ae12d8523..69ec4ade6521 100644
> --- a/arch/powerpc/Kconfig
> +++ b/arch/powerpc/Kconfig
> @@ -151,6 +151,7 @@ config PPC
>   	select ARCH_USE_CMPXCHG_LOCKREF		if PPC64
>   	select ARCH_USE_QUEUED_RWLOCKS		if PPC_QUEUED_SPINLOCKS
>   	select ARCH_USE_QUEUED_SPINLOCKS	if PPC_QUEUED_SPINLOCKS
> +	select ARCH_USE_QUEUED_SPINLOCKS_XCHG32	if PPC_QUEUED_SPINLOCKS
>   	select ARCH_WANT_IPC_PARSE_VERSION
>   	select ARCH_WANT_IRQS_OFF_ACTIVATE_MM
>   	select ARCH_WANT_LD_ORPHAN_WARN
> 

Re: [PATCH v5 4/7] powerpc/qspinlock: Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32

[Guo Ren]

On Sun, Mar 28, 2021 at 7:14 PM Christophe Leroy
<christophe.leroy@csgroup.eu> wrote:
>
>
>
> Le 28/03/2021 à 08:30, guoren@kernel.org a écrit :
> > From: Guo Ren <guoren@linux.alibaba.com>
> >
> > We don't have native hw xchg16 instruction, so let qspinlock
> > generic code to deal with it.
>
> We have lharx/sthcx pair on some versions of powerpc.
>
> See https://patchwork.ozlabs.org/project/linuxppc-dev/patch/20201107032328.2454582-1-npiggin@gmail.com/
Got it, thx for the information.

>
> Christophe
>
> >
> > Using the full-word atomic xchg instructions implement xchg16 has
> > the semantic risk for atomic operations.
> >
> > This patch cancels the dependency of on qspinlock generic code on
> > architecture's xchg16.
> >
> > Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
> > Cc: Michael Ellerman <mpe@ellerman.id.au>
> > Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
> > Cc: Paul Mackerras <paulus@samba.org>
> > ---
> >   arch/powerpc/Kconfig | 1 +
> >   1 file changed, 1 insertion(+)
> >
> > diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
> > index 386ae12d8523..69ec4ade6521 100644
> > --- a/arch/powerpc/Kconfig
> > +++ b/arch/powerpc/Kconfig
> > @@ -151,6 +151,7 @@ config PPC
> >       select ARCH_USE_CMPXCHG_LOCKREF         if PPC64
> >       select ARCH_USE_QUEUED_RWLOCKS          if PPC_QUEUED_SPINLOCKS
> >       select ARCH_USE_QUEUED_SPINLOCKS        if PPC_QUEUED_SPINLOCKS
> > +     select ARCH_USE_QUEUED_SPINLOCKS_XCHG32 if PPC_QUEUED_SPINLOCKS
> >       select ARCH_WANT_IPC_PARSE_VERSION
> >       select ARCH_WANT_IRQS_OFF_ACTIVATE_MM
> >       select ARCH_WANT_LD_ORPHAN_WARN
> >



-- 
Best Regards
 Guo Ren

ML: https://lore.kernel.org/linux-csky/

[PATCH v5 5/7] openrisc: qspinlock: Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32

[guoren]

From: Guo Ren <guoren@linux.alibaba.com>

We don't have native hw xchg16 instruction, so let qspinlock
generic code to deal with it.

Using the full-word atomic xchg instructions implement xchg16 has
the semantic risk for atomic operations.

This patch cancels the dependency of on qspinlock generic code on
architecture's xchg16.

Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Jonas Bonn <jonas@southpole.se>
Cc: Stefan Kristiansson <stefan.kristiansson@saunalahti.fi>
Cc: Stafford Horne <shorne@gmail.com>
Cc: openrisc@lists.librecores.org
---
 arch/openrisc/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/openrisc/Kconfig b/arch/openrisc/Kconfig
index 591acc5990dc..b299e409429f 100644
--- a/arch/openrisc/Kconfig
+++ b/arch/openrisc/Kconfig
@@ -33,6 +33,7 @@ config OPENRISC
 	select OR1K_PIC
 	select CPU_NO_EFFICIENT_FFS if !OPENRISC_HAVE_INST_FF1
 	select ARCH_USE_QUEUED_SPINLOCKS
+	select ARCH_USE_QUEUED_SPINLOCKS_XCHG32
 	select ARCH_USE_QUEUED_RWLOCKS
 	select OMPIC if SMP
 	select ARCH_WANT_FRAME_POINTERS
-- 
2.17.1

[PATCH v5 6/7] sparc: qspinlock: Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32

[guoren]

From: Guo Ren <guoren@linux.alibaba.com>

We don't have native hw xchg16 instruction, so let qspinlock
generic code to deal with it.

Using the full-word atomic xchg instructions implement xchg16 has
the semantic risk for atomic operations.

This patch cancels the dependency of on qspinlock generic code on
architecture's xchg16.

Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: David S. Miller <davem@davemloft.net>
Cc: Rob Gardner <rob.gardner@oracle.com>
---
 arch/sparc/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 164a5254c91c..1079fe3f058c 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -91,6 +91,7 @@ config SPARC64
 	select HAVE_REGS_AND_STACK_ACCESS_API
 	select ARCH_USE_QUEUED_RWLOCKS
 	select ARCH_USE_QUEUED_SPINLOCKS
+	select ARCH_USE_QUEUED_SPINLOCKS_XCHG32
 	select GENERIC_TIME_VSYSCALL
 	select ARCH_CLOCKSOURCE_DATA
 	select ARCH_HAS_PTE_SPECIAL
-- 
2.17.1

[PATCH v5 7/7] xtensa: qspinlock: Add ARCH_USE_QUEUED_SPINLOCKS_XCHG32

[guoren]

From: Guo Ren <guoren@linux.alibaba.com>

We don't have native hw xchg16 instruction, so let qspinlock
generic code to deal with it.

Using the full-word atomic xchg instructions implement xchg16 has
the semantic risk for atomic operations.

This patch cancels the dependency of on qspinlock generic code on
architecture's xchg16.

Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Chris Zankel <chris@zankel.net>
Cc: Max Filippov <jcmvbkbc@gmail.com>
---
 arch/xtensa/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig
index 9ad6b7b82707..f19d780638f7 100644
--- a/arch/xtensa/Kconfig
+++ b/arch/xtensa/Kconfig
@@ -9,6 +9,7 @@ config XTENSA
 	select ARCH_HAS_DMA_SET_UNCACHED if MMU
 	select ARCH_USE_QUEUED_RWLOCKS
 	select ARCH_USE_QUEUED_SPINLOCKS
+	select ARCH_USE_QUEUED_SPINLOCKS_XCHG32
 	select ARCH_WANT_FRAME_POINTERS
 	select ARCH_WANT_IPC_PARSE_VERSION
 	select BUILDTIME_TABLE_SORT
-- 
2.17.1