[Sound-open-firmware] [PATCH 09/10] arch: spinlock: fix value sharing between cores

[Liam Girdwood]

From: Tomasz Lauda <tomasz.lauda at linux.intel.com>

Fixes xtensa spinlock implementation.
Previous one caused deadlock between cores.

Signed-off-by: Tomasz Lauda <tomasz.lauda at linux.intel.com>
---
 src/arch/xtensa/include/arch/spinlock.h | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/src/arch/xtensa/include/arch/spinlock.h b/src/arch/xtensa/include/arch/spinlock.h
index 13bf4c1..3bb6b15 100644
--- a/src/arch/xtensa/include/arch/spinlock.h
+++ b/src/arch/xtensa/include/arch/spinlock.h
@@ -49,29 +49,29 @@ static inline void arch_spinlock_init(spinlock_t *lock)
 
 static inline void arch_spin_lock(spinlock_t *lock)
 {
-	uint32_t result;
+	uint32_t result, current;
 
 	__asm__ __volatile__(
-		"       movi    %0, 0\n"
-		"       wsr     %0, scompare1\n"
-		"1:     movi    %0, 1\n"
-		"       s32c1i  %0, %1, 0\n"
-		"       bnez    %0, 1b\n"
-		: "=&a" (result)
+		"1:     l32i    %1, %2, 0\n"
+		"       wsr     %1, scompare1\n"
+		"       movi    %0, 1\n"
+		"       s32c1i  %0, %2, 0\n"
+		"       bne     %0, %1, 1b\n"
+		: "=&a" (result), "=&a" (current)
 		: "a" (&lock->lock)
 		: "memory");
 }
 
 static inline int arch_try_lock(spinlock_t *lock)
 {
-	uint32_t result;
+	uint32_t result, current;
 
 	__asm__ __volatile__(
-		"       movi    %0, 0\n"
-		"       wsr     %0, scompare1\n"
+		"       l32i    %1, %2, 0\n"
+		"       wsr     %1, scompare1\n"
 		"       movi    %0, 1\n"
-		"       s32c1i  %0, %1, 0\n"
-		: "=&a" (result)
+		"       s32c1i  %0, %2, 0\n"
+		: "=&a" (result), "=&a" (current)
 		: "a" (&lock->lock)
 		: "memory");
 
-- 
2.17.0