[PATCH v4] arm: remove !CPU_V6 and !GENERIC_ATOMIC64 build dependencies for XEN

[Stefano Stabellini]

Remove !GENERIC_ATOMIC64 build dependency:
- introduce xen_atomic64_xchg
- use it to implement xchg_xen_ulong

Remove !CPU_V6 build dependency:
- introduce __cmpxchg8 and __cmpxchg16, compiled even ifdef
  CONFIG_CPU_V6
- implement sync_cmpxchg using __cmpxchg8 and __cmpxchg16

Signed-off-by: Stefano Stabellini <stefano.stabellini at eu.citrix.com>
CC: arnd at arndb.de
CC: linux at arm.linux.org.uk
CC: will.deacon at arm.com
CC: catalin.marinas at arm.com
CC: linux-arm-kernel at lists.infradead.org
CC: linux-kernel at vger.kernel.org
CC: xen-devel at lists.xenproject.org

---

Changes in v4:
- avoid moving and renaming atomic64_xchg
- introduce xen_atomic64_xchg
- fix asm comment in __cmpxchg8 and __cmpxchg16.

---
 arch/arm/Kconfig                   |    3 +-
 arch/arm/include/asm/cmpxchg.h     |   60 ++++++++++++++++++++++++------------
 arch/arm/include/asm/sync_bitops.h |   24 ++++++++++++++-
 arch/arm/include/asm/xen/events.h  |   32 ++++++++++++++++++-
 4 files changed, 95 insertions(+), 24 deletions(-)

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index c1f1a7e..ae54ae0 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -1881,8 +1881,7 @@ config XEN_DOM0
 config XEN
 	bool "Xen guest support on ARM (EXPERIMENTAL)"
 	depends on ARM && AEABI && OF
-	depends on CPU_V7 && !CPU_V6
-	depends on !GENERIC_ATOMIC64
+	depends on CPU_V7
 	select ARM_PSCI
 	select SWIOTLB_XEN
 	help
diff --git a/arch/arm/include/asm/cmpxchg.h b/arch/arm/include/asm/cmpxchg.h
index df2fbba..a17cff1 100644
--- a/arch/arm/include/asm/cmpxchg.h
+++ b/arch/arm/include/asm/cmpxchg.h
@@ -133,6 +133,44 @@ extern void __bad_cmpxchg(volatile void *ptr, int size);
  * cmpxchg only support 32-bits operands on ARMv6.
  */
 
+static inline unsigned long __cmpxchg8(volatile void *ptr, unsigned long old,
+				      unsigned long new)
+{
+	unsigned long oldval, res;
+
+	do {
+		asm volatile("@ __cmpxchg8\n"
+		"	ldrexb	%1, [%2]\n"
+		"	mov	%0, #0\n"
+		"	teq	%1, %3\n"
+		"	strexbeq %0, %4, [%2]\n"
+			: "=&r" (res), "=&r" (oldval)
+			: "r" (ptr), "Ir" (old), "r" (new)
+			: "memory", "cc");
+	} while (res);
+
+	return oldval;
+}
+
+static inline unsigned long __cmpxchg16(volatile void *ptr, unsigned long old,
+				      unsigned long new)
+{
+	unsigned long oldval, res;
+
+	do {
+		asm volatile("@ __cmpxchg16\n"
+		"	ldrexh	%1, [%2]\n"
+		"	mov	%0, #0\n"
+		"	teq	%1, %3\n"
+		"	strexheq %0, %4, [%2]\n"
+			: "=&r" (res), "=&r" (oldval)
+			: "r" (ptr), "Ir" (old), "r" (new)
+			: "memory", "cc");
+	} while (res);
+
+	return oldval;
+}
+
 static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
 				      unsigned long new, int size)
 {
@@ -141,28 +179,10 @@ static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
 	switch (size) {
 #ifndef CONFIG_CPU_V6	/* min ARCH >= ARMv6K */
 	case 1:
-		do {
-			asm volatile("@ __cmpxchg1\n"
-			"	ldrexb	%1, [%2]\n"
-			"	mov	%0, #0\n"
-			"	teq	%1, %3\n"
-			"	strexbeq %0, %4, [%2]\n"
-				: "=&r" (res), "=&r" (oldval)
-				: "r" (ptr), "Ir" (old), "r" (new)
-				: "memory", "cc");
-		} while (res);
+		oldval = __cmpxchg8(ptr, old, new);
 		break;
 	case 2:
-		do {
-			asm volatile("@ __cmpxchg1\n"
-			"	ldrexh	%1, [%2]\n"
-			"	mov	%0, #0\n"
-			"	teq	%1, %3\n"
-			"	strexheq %0, %4, [%2]\n"
-				: "=&r" (res), "=&r" (oldval)
-				: "r" (ptr), "Ir" (old), "r" (new)
-				: "memory", "cc");
-		} while (res);
+		oldval = __cmpxchg16(ptr, old, new);
 		break;
 #endif
 	case 4:
diff --git a/arch/arm/include/asm/sync_bitops.h b/arch/arm/include/asm/sync_bitops.h
index 63479ee..942659a 100644
--- a/arch/arm/include/asm/sync_bitops.h
+++ b/arch/arm/include/asm/sync_bitops.h
@@ -21,7 +21,29 @@
 #define sync_test_and_clear_bit(nr, p)	_test_and_clear_bit(nr, p)
 #define sync_test_and_change_bit(nr, p)	_test_and_change_bit(nr, p)
 #define sync_test_bit(nr, addr)		test_bit(nr, addr)
-#define sync_cmpxchg			cmpxchg
 
+static inline unsigned long sync_cmpxchg(volatile void *ptr,
+										 unsigned long old,
+										 unsigned long new)
+{
+	unsigned long oldval;
+	int size = sizeof(*(ptr));
+
+	smp_mb();
+	switch (size) {
+	case 1:
+		oldval = __cmpxchg8(ptr, old, new);
+		break;
+	case 2:
+		oldval = __cmpxchg16(ptr, old, new);
+		break;
+	default:
+		oldval = __cmpxchg(ptr, old, new, size);
+		break;
+	}
+	smp_mb();
+
+	return oldval;
+}
 
 #endif
diff --git a/arch/arm/include/asm/xen/events.h b/arch/arm/include/asm/xen/events.h
index 8b1f37b..2032ee6 100644
--- a/arch/arm/include/asm/xen/events.h
+++ b/arch/arm/include/asm/xen/events.h
@@ -16,7 +16,37 @@ static inline int xen_irqs_disabled(struct pt_regs *regs)
 	return raw_irqs_disabled_flags(regs->ARM_cpsr);
 }
 
-#define xchg_xen_ulong(ptr, val) atomic64_xchg(container_of((ptr),	\
+#ifdef CONFIG_GENERIC_ATOMIC64
+/* if CONFIG_GENERIC_ATOMIC64 is defined we cannot use the generic
+ * atomic64_xchg function because it is implemented using spin locks.
+ * Here we need proper atomic instructions to read and write memory
+ * shared with the hypervisor.
+ */
+static inline u64 xen_atomic64_xchg(atomic64_t *ptr, u64 new)
+{
+	u64 result;
+	unsigned long tmp;
+
+	smp_mb();
+
+	__asm__ __volatile__("@ xen_atomic64_xchg\n"
+"1:	ldrexd	%0, %H0, [%3]\n"
+"	strexd	%1, %4, %H4, [%3]\n"
+"	teq	%1, #0\n"
+"	bne	1b"
+	: "=&r" (result), "=&r" (tmp), "+Qo" (ptr->counter)
+	: "r" (&ptr->counter), "r" (new)
+	: "cc");
+
+	smp_mb();
+
+	return result;
+}
+#else
+#define xen_atomic64_xchg atomic64_xchg
+#endif
+
+#define xchg_xen_ulong(ptr, val) xen_atomic64_xchg(container_of((ptr),	\
 							    atomic64_t,	\
 							    counter), (val))
 
-- 
1.7.10.4