Skip to contentSkip to conversation

riscv/atomic: Strengthen implementations with fences

PATCH In review Andrea Parri 6 participants 12 messages Mar 9, 2018 – Mar 13, 2018

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH v2 2/2 riscv/atomic: Strengthen implementations with fences

Atomics present the same issue with locking: release and acquire variants need to be strengthened to meet the constraints defined by the Linux-kernel memory consistency model [1].

Atomics present a further issue: implementations of atomics such as atomic_cmpxchg() and atomic_add_unless() rely on LR/SC pairs, which do not give full-ordering with .aqrl; for example, current implementations allow the "lr-sc-aqrl-pair-vs-full-barrier" test below to end up with the state indicated in the "exists" clause.

In order to "synchronize" LKMM and RISC-V's implementation, this commit strengthens the implementations of the atomics operations by replacing .rl and .aq with the use of ("lightweigth") fences, and by replacing .aqrl LR/SC pairs in sequences such as:

0: lr.w.aqrl %0, %addr

          bne        %0, %old, 1f
          ...
          sc.w.aqrl  %1, %new, %addr
          bnez       %1, 0b

1:

with sequences of the form:

0: lr.w %0, %addr

          bne        %0, %old, 1f
          ...
          sc.w.rl    %1, %new, %addr   /* SC-release   */
          bnez       %1, 0b
          fence      rw, rw            /* "full" fence */

1:

following Daniel's suggestion.

These modifications were validated with simulation of the RISC-V memory consistency model.

C lr-sc-aqrl-pair-vs-full-barrier

{}

P0(int *x, int *y, atomic_t *u)

{
	int r0;
	int r1;

	WRITE_ONCE(*x, 1);
	r0 = atomic_cmpxchg(u, 0, 1);
	r1 = READ_ONCE(*y);
}

P1(int *x, int *y, atomic_t *v)

{
	int r0;
	int r1;

	WRITE_ONCE(*y, 1);
	r0 = atomic_cmpxchg(v, 0, 1);
	r1 = READ_ONCE(*x);
}

exists (u=1 /\ v=1 /\ 0:r1=0 /\ 1:r1=0)

[1] https://marc.info/?l=linux-kernel&m=151930201102853&w=2

    https://groups.google.com/a/groups.riscv.org/forum/#!topic/isa-dev/hKywNHBkAXM
    https://marc.info/?l=linux-kernel&m=151633436614259&w=2
Cc Palmer Dabbelt <[email protected]>
Cc Albert Ou <[email protected]>
Cc Daniel Lustig <[email protected]>
Cc Alan Stern <[email protected]>
Cc Will Deacon <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Boqun Feng <[email protected]>
Cc Nicholas Piggin <[email protected]>
Cc David Howells <[email protected]>
Cc Jade Alglave <[email protected]>
Cc Luc Maranget <[email protected]>
Cc "Paul E. McKenney" <[email protected]>
Cc Akira Yokosawa <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc linux-riscv at lists.infradead.org
Cc linux-kernel at vger.kernel.org
---
 arch/riscv/include/asm/atomic.h  | 417 +++++++++++++++++++++++++--------------
 arch/riscv/include/asm/cmpxchg.h | 391 +++++++++++++++++++++++++++++-------
 2 files changed, 588 insertions(+), 220 deletions(-)
arch/riscv/include/asm/atomic.h +268 -149
index e65d1cd89e28b..855115ace98c8 100644
@@ -24,6 +24,20 @@
 #include <asm/barrier.h>
 
 #define ATOMIC_INIT(i)	{ (i) }
+
+#define __atomic_op_acquire(op, args...)				\
+({									\
+	typeof(op##_relaxed(args)) __ret  = op##_relaxed(args);		\
+	__asm__ __volatile__(RISCV_ACQUIRE_BARRIER "" ::: "memory");	\
+	__ret;								\
+})
+
+#define __atomic_op_release(op, args...)				\
+({									\
+	__asm__ __volatile__(RISCV_RELEASE_BARRIER "" ::: "memory");	\
+	op##_relaxed(args);						\
+})
+
 static __always_inline int atomic_read(const atomic_t *v)
 {
 	return READ_ONCE(v->counter);
@@ -50,22 +64,23 @@ static __always_inline void atomic64_set(atomic64_t *v, long i)
  * have the AQ or RL bits set.  These don't return anything, so there's only
  * one version to worry about.
  */
-#define ATOMIC_OP(op, asm_op, I, asm_type, c_type, prefix)				\
-static __always_inline void atomic##prefix##_##op(c_type i, atomic##prefix##_t *v)	\
-{											\
-	__asm__ __volatile__ (								\
-		"amo" #asm_op "." #asm_type " zero, %1, %0"				\
-		: "+A" (v->counter)							\
-		: "r" (I)								\
-		: "memory");								\
-}
+#define ATOMIC_OP(op, asm_op, I, asm_type, c_type, prefix)		\
+static __always_inline							\
+void atomic##prefix##_##op(c_type i, atomic##prefix##_t *v)		\
+{									\
+	__asm__ __volatile__ (						\
+		"	amo" #asm_op "." #asm_type " zero, %1, %0"	\
+		: "+A" (v->counter)					\
+		: "r" (I)						\
+		: "memory");						\
+}									\
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, I)			\
+#define ATOMIC_OPS(op, asm_op, I)					\
         ATOMIC_OP (op, asm_op, I, w,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, I)			\
-        ATOMIC_OP (op, asm_op, I, w,  int,   )	\
+#define ATOMIC_OPS(op, asm_op, I)					\
+        ATOMIC_OP (op, asm_op, I, w,  int,   )				\
         ATOMIC_OP (op, asm_op, I, d, long, 64)
 #endif
 
@@ -79,75 +94,115 @@ ATOMIC_OPS(xor, xor,  i)
 #undef ATOMIC_OPS
 
 /*
- * Atomic ops that have ordered, relaxed, acquire, and relese variants.
+ * Atomic ops that have ordered, relaxed, acquire, and release variants.
  * There's two flavors of these: the arithmatic ops have both fetch and return
  * versions, while the logical ops only have fetch versions.
  */
-#define ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, asm_type, c_type, prefix)				\
-static __always_inline c_type atomic##prefix##_fetch_##op##c_or(c_type i, atomic##prefix##_t *v)	\
-{													\
-	register c_type ret;										\
-	__asm__ __volatile__ (										\
-		"amo" #asm_op "." #asm_type #asm_or " %1, %2, %0"					\
-		: "+A" (v->counter), "=r" (ret)								\
-		: "r" (I)										\
-		: "memory");										\
-	return ret;											\
+#define ATOMIC_FETCH_OP(op, asm_op, I, asm_type, c_type, prefix)	\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op##_relaxed(c_type i,			\
+					     atomic##prefix##_t *v)	\
+{									\
+	register c_type ret;						\
+	__asm__ __volatile__ (						\
+		"	amo" #asm_op "." #asm_type " %1, %2, %0"	\
+		: "+A" (v->counter), "=r" (ret)				\
+		: "r" (I)						\
+		: "memory");						\
+	return ret;							\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op(c_type i, atomic##prefix##_t *v)	\
+{									\
+	register c_type ret;						\
+	__asm__ __volatile__ (						\
+		"	amo" #asm_op "." #asm_type ".aqrl  %1, %2, %0"	\
+		: "+A" (v->counter), "=r" (ret)				\
+		: "r" (I)						\
+		: "memory");						\
+	return ret;							\
 }
 
-#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, asm_type, c_type, prefix)			\
-static __always_inline c_type atomic##prefix##_##op##_return##c_or(c_type i, atomic##prefix##_t *v)	\
-{													\
-        return atomic##prefix##_fetch_##op##c_or(i, v) c_op I;						\
+#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_type, c_type, prefix)	\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return_relaxed(c_type i,			\
+					      atomic##prefix##_t *v)	\
+{									\
+        return atomic##prefix##_fetch_##op##_relaxed(i, v) c_op I;	\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return(c_type i, atomic##prefix##_t *v)	\
+{									\
+        return atomic##prefix##_fetch_##op(i, v) c_op I;		\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, c_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, w,  int,   )
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_FETCH_OP( op, asm_op,       I, w,  int,   )		\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I, w,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, c_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, d, long, 64)	\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, d, long, 64)
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_FETCH_OP( op, asm_op,       I, w,  int,   )		\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I, w,  int,   )		\
+        ATOMIC_FETCH_OP( op, asm_op,       I, d, long, 64)		\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I, d, long, 64)
 #endif
 
-ATOMIC_OPS(add, add, +,  i,      , _relaxed)
-ATOMIC_OPS(add, add, +,  i, .aq  , _acquire)
-ATOMIC_OPS(add, add, +,  i, .rl  , _release)
-ATOMIC_OPS(add, add, +,  i, .aqrl,         )
+ATOMIC_OPS(add, add, +,  i)
+ATOMIC_OPS(sub, add, +, -i)
+
+#define atomic_add_return_relaxed	atomic_add_return_relaxed
+#define atomic_sub_return_relaxed	atomic_sub_return_relaxed
+#define atomic_add_return		atomic_add_return
+#define atomic_sub_return		atomic_sub_return
 
-ATOMIC_OPS(sub, add, +, -i,      , _relaxed)
-ATOMIC_OPS(sub, add, +, -i, .aq  , _acquire)
-ATOMIC_OPS(sub, add, +, -i, .rl  , _release)
-ATOMIC_OPS(sub, add, +, -i, .aqrl,         )
+#define atomic_fetch_add_relaxed	atomic_fetch_add_relaxed
+#define atomic_fetch_sub_relaxed	atomic_fetch_sub_relaxed
+#define atomic_fetch_add		atomic_fetch_add
+#define atomic_fetch_sub		atomic_fetch_sub
+
+#ifndef CONFIG_GENERIC_ATOMIC64
+#define atomic64_add_return_relaxed	atomic64_add_return_relaxed
+#define atomic64_sub_return_relaxed	atomic64_sub_return_relaxed
+#define atomic64_add_return		atomic64_add_return
+#define atomic64_sub_return		atomic64_sub_return
+
+#define atomic64_fetch_add_relaxed	atomic64_fetch_add_relaxed
+#define atomic64_fetch_sub_relaxed	atomic64_fetch_sub_relaxed
+#define atomic64_fetch_add		atomic64_fetch_add
+#define atomic64_fetch_sub		atomic64_fetch_sub
+#endif
 
 #undef ATOMIC_OPS
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, w,  int,   )
+#define ATOMIC_OPS(op, asm_op, I)					\
+        ATOMIC_FETCH_OP(op, asm_op, I, w,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, d, long, 64)
+#define ATOMIC_OPS(op, asm_op, I)					\
+        ATOMIC_FETCH_OP(op, asm_op, I, w,  int,   )			\
+        ATOMIC_FETCH_OP(op, asm_op, I, d, long, 64)
 #endif
 
-ATOMIC_OPS(and, and, i,      , _relaxed)
-ATOMIC_OPS(and, and, i, .aq  , _acquire)
-ATOMIC_OPS(and, and, i, .rl  , _release)
-ATOMIC_OPS(and, and, i, .aqrl,         )
+ATOMIC_OPS(and, and, i)
+ATOMIC_OPS( or,  or, i)
+ATOMIC_OPS(xor, xor, i)
 
-ATOMIC_OPS( or,  or, i,      , _relaxed)
-ATOMIC_OPS( or,  or, i, .aq  , _acquire)
-ATOMIC_OPS( or,  or, i, .rl  , _release)
-ATOMIC_OPS( or,  or, i, .aqrl,         )
+#define atomic_fetch_and_relaxed	atomic_fetch_and_relaxed
+#define atomic_fetch_or_relaxed		atomic_fetch_or_relaxed
+#define atomic_fetch_xor_relaxed	atomic_fetch_xor_relaxed
+#define atomic_fetch_and		atomic_fetch_and
+#define atomic_fetch_or			atomic_fetch_or
+#define atomic_fetch_xor		atomic_fetch_xor
 
-ATOMIC_OPS(xor, xor, i,      , _relaxed)
-ATOMIC_OPS(xor, xor, i, .aq  , _acquire)
-ATOMIC_OPS(xor, xor, i, .rl  , _release)
-ATOMIC_OPS(xor, xor, i, .aqrl,         )
+#ifndef CONFIG_GENERIC_ATOMIC64
+#define atomic64_fetch_and_relaxed	atomic64_fetch_and_relaxed
+#define atomic64_fetch_or_relaxed	atomic64_fetch_or_relaxed
+#define atomic64_fetch_xor_relaxed	atomic64_fetch_xor_relaxed
+#define atomic64_fetch_and		atomic64_fetch_and
+#define atomic64_fetch_or		atomic64_fetch_or
+#define atomic64_fetch_xor		atomic64_fetch_xor
+#endif
 
 #undef ATOMIC_OPS
 
@@ -157,22 +212,24 @@ ATOMIC_OPS(xor, xor, i, .aqrl,         )
 /*
  * The extra atomic operations that are constructed from one of the core
  * AMO-based operations above (aside from sub, which is easier to fit above).
- * These are required to perform a barrier, but they're OK this way because
- * atomic_*_return is also required to perform a barrier.
+ * These are required to perform a full barrier, but they're OK this way
+ * because atomic_*_return is also required to perform a full barrier.
+ *
  */
-#define ATOMIC_OP(op, func_op, comp_op, I, c_type, prefix)			\
-static __always_inline bool atomic##prefix##_##op(c_type i, atomic##prefix##_t *v) \
-{										\
-	return atomic##prefix##_##func_op##_return(i, v) comp_op I;		\
+#define ATOMIC_OP(op, func_op, comp_op, I, c_type, prefix)		\
+static __always_inline							\
+bool atomic##prefix##_##op(c_type i, atomic##prefix##_t *v)		\
+{									\
+	return atomic##prefix##_##func_op##_return(i, v) comp_op I;	\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, func_op, comp_op, I)			\
-        ATOMIC_OP (op, func_op, comp_op, I,  int,   )
+#define ATOMIC_OPS(op, func_op, comp_op, I)				\
+        ATOMIC_OP(op, func_op, comp_op, I,  int,   )
 #else
-#define ATOMIC_OPS(op, func_op, comp_op, I)			\
-        ATOMIC_OP (op, func_op, comp_op, I,  int,   )		\
-        ATOMIC_OP (op, func_op, comp_op, I, long, 64)
+#define ATOMIC_OPS(op, func_op, comp_op, I)				\
+        ATOMIC_OP(op, func_op, comp_op, I,  int,   )			\
+        ATOMIC_OP(op, func_op, comp_op, I, long, 64)
 #endif
 
 ATOMIC_OPS(add_and_test, add, ==, 0)
@@ -182,51 +239,87 @@ ATOMIC_OPS(add_negative, add,  <, 0)
 #undef ATOMIC_OP
 #undef ATOMIC_OPS
 
-#define ATOMIC_OP(op, func_op, I, c_type, prefix)				\
-static __always_inline void atomic##prefix##_##op(atomic##prefix##_t *v)	\
-{										\
-	atomic##prefix##_##func_op(I, v);					\
+#define ATOMIC_OP(op, func_op, I, c_type, prefix)			\
+static __always_inline							\
+void atomic##prefix##_##op(atomic##prefix##_t *v)			\
+{									\
+	atomic##prefix##_##func_op(I, v);				\
 }
 
-#define ATOMIC_FETCH_OP(op, func_op, I, c_type, prefix)					\
-static __always_inline c_type atomic##prefix##_fetch_##op(atomic##prefix##_t *v)	\
-{											\
-	return atomic##prefix##_fetch_##func_op(I, v);					\
+#define ATOMIC_FETCH_OP(op, func_op, I, c_type, prefix)			\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op##_relaxed(atomic##prefix##_t *v)	\
+{									\
+	return atomic##prefix##_fetch_##func_op##_relaxed(I, v);	\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op(atomic##prefix##_t *v)		\
+{									\
+	return atomic##prefix##_fetch_##func_op(I, v);			\
 }
 
-#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, c_type, prefix)				\
-static __always_inline c_type atomic##prefix##_##op##_return(atomic##prefix##_t *v)	\
-{											\
-        return atomic##prefix##_fetch_##op(v) c_op I;					\
+#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, c_type, prefix)		\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return_relaxed(atomic##prefix##_t *v)	\
+{									\
+        return atomic##prefix##_fetch_##op##_relaxed(v) c_op I;		\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return(atomic##prefix##_t *v)		\
+{									\
+        return atomic##prefix##_fetch_##op(v) c_op I;			\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, c_op, I)						\
-        ATOMIC_OP       (op, asm_op,       I,  int,   )				\
-        ATOMIC_FETCH_OP (op, asm_op,       I,  int,   )				\
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_OP(       op, asm_op,       I,  int,   )			\
+        ATOMIC_FETCH_OP( op, asm_op,       I,  int,   )			\
         ATOMIC_OP_RETURN(op, asm_op, c_op, I,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, c_op, I)						\
-        ATOMIC_OP       (op, asm_op,       I,  int,   )				\
-        ATOMIC_FETCH_OP (op, asm_op,       I,  int,   )				\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I,  int,   )				\
-        ATOMIC_OP       (op, asm_op,       I, long, 64)				\
-        ATOMIC_FETCH_OP (op, asm_op,       I, long, 64)				\
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_OP(       op, asm_op,       I,  int,   )			\
+        ATOMIC_FETCH_OP( op, asm_op,       I,  int,   )			\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I,  int,   )			\
+        ATOMIC_OP(       op, asm_op,       I, long, 64)			\
+        ATOMIC_FETCH_OP( op, asm_op,       I, long, 64)			\
         ATOMIC_OP_RETURN(op, asm_op, c_op, I, long, 64)
 #endif
 
 ATOMIC_OPS(inc, add, +,  1)
 ATOMIC_OPS(dec, add, +, -1)
 
+#define atomic_inc_return_relaxed	atomic_inc_return_relaxed
+#define atomic_dec_return_relaxed	atomic_dec_return_relaxed
+#define atomic_inc_return		atomic_inc_return
+#define atomic_dec_return		atomic_dec_return
+
+#define atomic_fetch_inc_relaxed	atomic_fetch_inc_relaxed
+#define atomic_fetch_dec_relaxed	atomic_fetch_dec_relaxed
+#define atomic_fetch_inc		atomic_fetch_inc
+#define atomic_fetch_dec		atomic_fetch_dec
+
+#ifndef CONFIG_GENERIC_ATOMIC64
+#define atomic64_inc_return_relaxed	atomic64_inc_return_relaxed
+#define atomic64_dec_return_relaxed	atomic64_dec_return_relaxed
+#define atomic64_inc_return		atomic64_inc_return
+#define atomic64_dec_return		atomic64_dec_return
+
+#define atomic64_fetch_inc_relaxed	atomic64_fetch_inc_relaxed
+#define atomic64_fetch_dec_relaxed	atomic64_fetch_dec_relaxed
+#define atomic64_fetch_inc		atomic64_fetch_inc
+#define atomic64_fetch_dec		atomic64_fetch_dec
+#endif
+
 #undef ATOMIC_OPS
 #undef ATOMIC_OP
 #undef ATOMIC_FETCH_OP
 #undef ATOMIC_OP_RETURN
 
-#define ATOMIC_OP(op, func_op, comp_op, I, prefix)				\
-static __always_inline bool atomic##prefix##_##op(atomic##prefix##_t *v)	\
-{										\
-	return atomic##prefix##_##func_op##_return(v) comp_op I;		\
+#define ATOMIC_OP(op, func_op, comp_op, I, prefix)			\
+static __always_inline							\
+bool atomic##prefix##_##op(atomic##prefix##_t *v)			\
+{									\
+	return atomic##prefix##_##func_op##_return(v) comp_op I;	\
 }
 
 ATOMIC_OP(inc_and_test, inc, ==, 0,   )
@@ -238,19 +331,19 @@ ATOMIC_OP(dec_and_test, dec, ==, 0, 64)
 
 #undef ATOMIC_OP
 
-/* This is required to provide a barrier on success. */
+/* This is required to provide a full barrier on success. */
 static __always_inline int __atomic_add_unless(atomic_t *v, int a, int u)
 {
        int prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.w.aqrl  %[p],  %[c]\n\t"
-		"beq        %[p],  %[u], 1f\n\t"
-		"add       %[rc],  %[p], %[a]\n\t"
-		"sc.w.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc], 0b\n\t"
-		"1:"
+		"0:	lr.w     %[p],  %[c]\n"
+		"	beq      %[p],  %[u], 1f\n"
+		"	add      %[rc], %[p], %[a]\n"
+		"	sc.w.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [a]"r" (a), [u]"r" (u)
 		: "memory");
@@ -263,13 +356,13 @@ static __always_inline long __atomic64_add_unless(atomic64_t *v, long a, long u)
        long prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.d.aqrl  %[p],  %[c]\n\t"
-		"beq        %[p],  %[u], 1f\n\t"
-		"add       %[rc],  %[p], %[a]\n\t"
-		"sc.d.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc], 0b\n\t"
-		"1:"
+		"0:	lr.d     %[p],  %[c]\n"
+		"	beq      %[p],  %[u], 1f\n"
+		"	add      %[rc], %[p], %[a]\n"
+		"	sc.d.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [a]"r" (a), [u]"r" (u)
 		: "memory");
@@ -300,37 +393,63 @@ static __always_inline long atomic64_inc_not_zero(atomic64_t *v)
 
 /*
  * atomic_{cmp,}xchg is required to have exactly the same ordering semantics as
- * {cmp,}xchg and the operations that return, so they need a barrier.
- */
-/*
- * FIXME: atomic_cmpxchg_{acquire,release,relaxed} are all implemented by
- * assigning the same barrier to both the LR and SC operations, but that might
- * not make any sense.  We're waiting on a memory model specification to
- * determine exactly what the right thing to do is here.
+ * {cmp,}xchg and the operations that return, so they need a full barrier.
  */
-#define ATOMIC_OP(c_t, prefix, c_or, size, asm_or)						\
-static __always_inline c_t atomic##prefix##_cmpxchg##c_or(atomic##prefix##_t *v, c_t o, c_t n) 	\
-{												\
-	return __cmpxchg(&(v->counter), o, n, size, asm_or, asm_or);				\
-}												\
-static __always_inline c_t atomic##prefix##_xchg##c_or(atomic##prefix##_t *v, c_t n) 		\
-{												\
-	return __xchg(n, &(v->counter), size, asm_or);						\
+#define ATOMIC_OP(c_t, prefix, size)					\
+static __always_inline							\
+c_t atomic##prefix##_xchg_relaxed(atomic##prefix##_t *v, c_t n)		\
+{									\
+	return __xchg_relaxed(&(v->counter), n, size);			\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_xchg_acquire(atomic##prefix##_t *v, c_t n)		\
+{									\
+	return __xchg_acquire(&(v->counter), n, size);			\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_xchg_release(atomic##prefix##_t *v, c_t n)		\
+{									\
+	return __xchg_release(&(v->counter), n, size);			\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_xchg(atomic##prefix##_t *v, c_t n)			\
+{									\
+	return __xchg(&(v->counter), n, size);				\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg_relaxed(atomic##prefix##_t *v,		\
+				     c_t o, c_t n)			\
+{									\
+	return __cmpxchg_relaxed(&(v->counter), o, n, size);		\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg_acquire(atomic##prefix##_t *v,		\
+				     c_t o, c_t n)			\
+{									\
+	return __cmpxchg_acquire(&(v->counter), o, n, size);		\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg_release(atomic##prefix##_t *v,		\
+				     c_t o, c_t n)			\
+{									\
+	return __cmpxchg_release(&(v->counter), o, n, size);		\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg(atomic##prefix##_t *v, c_t o, c_t n)	\
+{									\
+	return __cmpxchg(&(v->counter), o, n, size);			\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(c_or, asm_or)			\
-	ATOMIC_OP( int,   , c_or, 4, asm_or)
+#define ATOMIC_OPS()							\
+	ATOMIC_OP( int,   , 4)
 #else
-#define ATOMIC_OPS(c_or, asm_or)			\
-	ATOMIC_OP( int,   , c_or, 4, asm_or)		\
-	ATOMIC_OP(long, 64, c_or, 8, asm_or)
+#define ATOMIC_OPS()							\
+	ATOMIC_OP( int,   , 4)						\
+	ATOMIC_OP(long, 64, 8)
 #endif
 
-ATOMIC_OPS(        , .aqrl)
-ATOMIC_OPS(_acquire,   .aq)
-ATOMIC_OPS(_release,   .rl)
-ATOMIC_OPS(_relaxed,      )
+ATOMIC_OPS()
 
 #undef ATOMIC_OPS
 #undef ATOMIC_OP
@@ -340,13 +459,13 @@ static __always_inline int atomic_sub_if_positive(atomic_t *v, int offset)
        int prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.w.aqrl  %[p],  %[c]\n\t"
-		"sub       %[rc],  %[p], %[o]\n\t"
-		"bltz      %[rc],    1f\n\t"
-		"sc.w.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc],    0b\n\t"
-		"1:"
+		"0:	lr.w     %[p],  %[c]\n"
+		"	sub      %[rc], %[p], %[o]\n"
+		"	bltz     %[rc], 1f\n"
+		"	sc.w.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [o]"r" (offset)
 		: "memory");
@@ -361,13 +480,13 @@ static __always_inline long atomic64_sub_if_positive(atomic64_t *v, int offset)
        long prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.d.aqrl  %[p],  %[c]\n\t"
-		"sub       %[rc],  %[p], %[o]\n\t"
-		"bltz      %[rc],    1f\n\t"
-		"sc.d.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc],    0b\n\t"
-		"1:"
+		"0:	lr.d     %[p],  %[c]\n"
+		"	sub      %[rc], %[p], %[o]\n"
+		"	bltz     %[rc], 1f\n"
+		"	sc.d.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [o]"r" (offset)
 		: "memory");
arch/riscv/include/asm/cmpxchg.h +320 -71

430 lines not on this page · raw message

Signature · 1 line
2.7.4
#1 lore raw
Quoted 61 lines · Andrea Parri

Atomics present the same issue with locking: release and acquire variants need to be strengthened to meet the constraints defined by the Linux-kernel memory consistency model [1].

Atomics present a further issue: implementations of atomics such as atomic_cmpxchg() and atomic_add_unless() rely on LR/SC pairs, which do not give full-ordering with .aqrl; for example, current implementations allow the "lr-sc-aqrl-pair-vs-full-barrier" test below to end up with the state indicated in the "exists" clause.

In order to "synchronize" LKMM and RISC-V's implementation, this commit strengthens the implementations of the atomics operations by replacing .rl and .aq with the use of ("lightweigth") fences, and by replacing .aqrl LR/SC pairs in sequences such as:

0: lr.w.aqrl %0, %addr

          bne        %0, %old, 1f
          ...
          sc.w.aqrl  %1, %new, %addr
          bnez       %1, 0b

1:

with sequences of the form:

0: lr.w %0, %addr

          bne        %0, %old, 1f
          ...
          sc.w.rl    %1, %new, %addr   /* SC-release   */
          bnez       %1, 0b
          fence      rw, rw            /* "full" fence */

1:

following Daniel's suggestion.

These modifications were validated with simulation of the RISC-V memory consistency model.

C lr-sc-aqrl-pair-vs-full-barrier

{}

P0(int *x, int *y, atomic_t *u)

{
	int r0;
	int r1;

	WRITE_ONCE(*x, 1);
	r0 = atomic_cmpxchg(u, 0, 1);
	r1 = READ_ONCE(*y);
}

P1(int *x, int *y, atomic_t *v)

{
	int r0;
	int r1;

	WRITE_ONCE(*y, 1);
	r0 = atomic_cmpxchg(v, 0, 1);
	r1 = READ_ONCE(*x);
}
Quoted 1 line

exists (u=1 /\ v=1 /\ 0:r1=0 /\ 1:r1=0)

There's another aspect to this imposed by the LKMM, and I'm not sure whether your patch addresses it. You add a fence after the cmpxchg operation but nothing before it. So what would happen with the following litmus test (which the LKMM forbids)?

C SB-atomic_cmpxchg-mb

{}

P0(int *x, int *y)

{
	int r0;

	WRITE_ONCE(*x, 1);
	r0 = atomic_cmpxchg(y, 0, 0);
}

P1(int *x, int *y)

{
	int r1;

	WRITE_ONCE(*y, 1);
	smp_mb();
	r1 = READ_ONCE(*x);
}

exists (0:r0=0 /\ 1:r1=0)

This is yet another illustration showing that full fences are stronger than cominations of release + acquire.

Alan Stern

#2 lore raw
Quoted 63 lines · Alan Stern
Quoted 61 lines · Andrea Parri

Atomics present the same issue with locking: release and acquire variants need to be strengthened to meet the constraints defined by the Linux-kernel memory consistency model [1].

Atomics present a further issue: implementations of atomics such as atomic_cmpxchg() and atomic_add_unless() rely on LR/SC pairs, which do not give full-ordering with .aqrl; for example, current implementations allow the "lr-sc-aqrl-pair-vs-full-barrier" test below to end up with the state indicated in the "exists" clause.

In order to "synchronize" LKMM and RISC-V's implementation, this commit strengthens the implementations of the atomics operations by replacing .rl and .aq with the use of ("lightweigth") fences, and by replacing .aqrl LR/SC pairs in sequences such as:

0: lr.w.aqrl %0, %addr

          bne        %0, %old, 1f
          ...
          sc.w.aqrl  %1, %new, %addr
          bnez       %1, 0b

1:

with sequences of the form:

0: lr.w %0, %addr

          bne        %0, %old, 1f
          ...
          sc.w.rl    %1, %new, %addr   /* SC-release   */
          bnez       %1, 0b
          fence      rw, rw            /* "full" fence */

1:

following Daniel's suggestion.

These modifications were validated with simulation of the RISC-V memory consistency model.

C lr-sc-aqrl-pair-vs-full-barrier

{}

P0(int *x, int *y, atomic_t *u)

{
	int r0;
	int r1;

	WRITE_ONCE(*x, 1);
	r0 = atomic_cmpxchg(u, 0, 1);
	r1 = READ_ONCE(*y);
}

P1(int *x, int *y, atomic_t *v)

{
	int r0;
	int r1;

	WRITE_ONCE(*y, 1);
	r0 = atomic_cmpxchg(v, 0, 1);
	r1 = READ_ONCE(*x);
}
Quoted 6 lines
Quoted 1 line

exists (u=1 /\ v=1 /\ 0:r1=0 /\ 1:r1=0)

There's another aspect to this imposed by the LKMM, and I'm not sure whether your patch addresses it. You add a fence after the cmpxchg operation but nothing before it. So what would happen with the following litmus test (which the LKMM forbids)?

Available RISC-V memory model formalizations forbid it; an intuitive explanation could probably be derived by paralleling the argument for arm64, as pointed out by Daniel at:

https://marc.info/?l=linux-kernel&m=151994289015267&w=2

Andrea

Quoted 29 lines

C SB-atomic_cmpxchg-mb

{}

P0(int *x, int *y)

{
	int r0;

	WRITE_ONCE(*x, 1);
	r0 = atomic_cmpxchg(y, 0, 0);
}

P1(int *x, int *y)

{
	int r1;

	WRITE_ONCE(*y, 1);
	smp_mb();
	r1 = READ_ONCE(*x);
}

exists (0:r0=0 /\ 1:r1=0)

This is yet another illustration showing that full fences are stronger than cominations of release + acquire.

Alan Stern

#3 lore raw
In reply to Andrea Parri
Quoted 129 lines · parri.andrea at gmail.com

Atomics present the same issue with locking: release and acquire variants need to be strengthened to meet the constraints defined by the Linux-kernel memory consistency model [1].

Atomics present a further issue: implementations of atomics such as atomic_cmpxchg() and atomic_add_unless() rely on LR/SC pairs, which do not give full-ordering with .aqrl; for example, current implementations allow the "lr-sc-aqrl-pair-vs-full-barrier" test below to end up with the state indicated in the "exists" clause.

In order to "synchronize" LKMM and RISC-V's implementation, this commit strengthens the implementations of the atomics operations by replacing .rl and .aq with the use of ("lightweigth") fences, and by replacing .aqrl LR/SC pairs in sequences such as:

0: lr.w.aqrl %0, %addr

          bne        %0, %old, 1f
          ...
          sc.w.aqrl  %1, %new, %addr
          bnez       %1, 0b

1:

with sequences of the form:

0: lr.w %0, %addr

          bne        %0, %old, 1f
          ...
          sc.w.rl    %1, %new, %addr   /* SC-release   */
          bnez       %1, 0b
          fence      rw, rw            /* "full" fence */

1:

following Daniel's suggestion.

These modifications were validated with simulation of the RISC-V memory consistency model.

C lr-sc-aqrl-pair-vs-full-barrier

{}

P0(int *x, int *y, atomic_t *u)

{
	int r0;
	int r1;

	WRITE_ONCE(*x, 1);
	r0 = atomic_cmpxchg(u, 0, 1);
	r1 = READ_ONCE(*y);
}

P1(int *x, int *y, atomic_t *v)

{
	int r0;
	int r1;

	WRITE_ONCE(*y, 1);
	r0 = atomic_cmpxchg(v, 0, 1);
	r1 = READ_ONCE(*x);
}

exists (u=1 /\ v=1 /\ 0:r1=0 /\ 1:r1=0)

[1] https://marc.info/?l=linux-kernel&m=151930201102853&w=2

    https://groups.google.com/a/groups.riscv.org/forum/#!topic/isa-dev/hKywNHBkAXM
    https://marc.info/?l=linux-kernel&m=151633436614259&w=2
Cc Palmer Dabbelt <[email protected]>
Cc Albert Ou <[email protected]>
Cc Daniel Lustig <[email protected]>
Cc Alan Stern <[email protected]>
Cc Will Deacon <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Boqun Feng <[email protected]>
Cc Nicholas Piggin <[email protected]>
Cc David Howells <[email protected]>
Cc Jade Alglave <[email protected]>
Cc Luc Maranget <[email protected]>
Cc "Paul E. McKenney" <[email protected]>
Cc Akira Yokosawa <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc linux-riscv at lists.infradead.org
Cc linux-kernel at vger.kernel.org
---
 arch/riscv/include/asm/atomic.h  | 417 +++++++++++++++++++++++++--------------
 arch/riscv/include/asm/cmpxchg.h | 391 +++++++++++++++++++++++++++++-------
 2 files changed, 588 insertions(+), 220 deletions(-)
arch/riscv/include/asm/atomic.h +14 -9
index e65d1cd89e28b..855115ace98c8 100644
@@ -24,6 +24,20 @@
 #include <asm/barrier.h>
 
 #define ATOMIC_INIT(i)	{ (i) }
+
+#define __atomic_op_acquire(op, args...)				\
+({									\
+	typeof(op##_relaxed(args)) __ret  = op##_relaxed(args);		\
+	__asm__ __volatile__(RISCV_ACQUIRE_BARRIER "" ::: "memory");	\
+	__ret;								\
+})
+
+#define __atomic_op_release(op, args...)				\
+({									\
+	__asm__ __volatile__(RISCV_RELEASE_BARRIER "" ::: "memory");	\
+	op##_relaxed(args);						\
+})
+
 static __always_inline int atomic_read(const atomic_t *v)
 {
 	return READ_ONCE(v->counter);
@@ -50,22 +64,23 @@ static __always_inline void atomic64_set(atomic64_t *v, long i)
  * have the AQ or RL bits set.  These don't return anything, so there's only
  * one version to worry about.
  */
-#define ATOMIC_OP(op, asm_op, I, asm_type, c_type, prefix)				\
-static __always_inline void atomic##prefix##_##op(c_type i, atomic##prefix##_t *v)	\
-{											\
-	__asm__ __volatile__ (								\
-		"amo" #asm_op "." #asm_type " zero, %1, %0"				\
-		: "+A" (v->counter)							\
-		: "r" (I)								\
-		: "memory");								\
-}

arch/riscv/include/asm/atomic.h static __always_inline void atomic64_set(atomic64_t *v, long i) · line 76

Quoted 10 lines
arch/riscv/include/asm/atomic.h +10 -0
+#define ATOMIC_OP(op, asm_op, I, asm_type, c_type, prefix)		\
+static __always_inline							\
+void atomic##prefix##_##op(c_type i, atomic##prefix##_t *v)		\
+{									\
+	__asm__ __volatile__ (						\
+		"	amo" #asm_op "." #asm_type " zero, %1, %0"	\
+		: "+A" (v->counter)					\
+		: "r" (I)						\
+		: "memory");						\
+}									\

Just to be sure: this is just a whitespace change, right?

Quoted 89 lines
arch/riscv/include/asm/atomic.h +44 -26
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, I)			\
+#define ATOMIC_OPS(op, asm_op, I)					\
         ATOMIC_OP (op, asm_op, I, w,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, I)			\
-        ATOMIC_OP (op, asm_op, I, w,  int,   )	\
+#define ATOMIC_OPS(op, asm_op, I)					\
+        ATOMIC_OP (op, asm_op, I, w,  int,   )				\
         ATOMIC_OP (op, asm_op, I, d, long, 64)
 #endif
 
@@ -79,75 +94,115 @@ ATOMIC_OPS(xor, xor,  i)
 #undef ATOMIC_OPS
 
 /*
- * Atomic ops that have ordered, relaxed, acquire, and relese variants.
+ * Atomic ops that have ordered, relaxed, acquire, and release variants.
  * There's two flavors of these: the arithmatic ops have both fetch and return
  * versions, while the logical ops only have fetch versions.
  */
-#define ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, asm_type, c_type, prefix)				\
-static __always_inline c_type atomic##prefix##_fetch_##op##c_or(c_type i, atomic##prefix##_t *v)	\
-{													\
-	register c_type ret;										\
-	__asm__ __volatile__ (										\
-		"amo" #asm_op "." #asm_type #asm_or " %1, %2, %0"					\
-		: "+A" (v->counter), "=r" (ret)								\
-		: "r" (I)										\
-		: "memory");										\
-	return ret;											\
+#define ATOMIC_FETCH_OP(op, asm_op, I, asm_type, c_type, prefix)	\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op##_relaxed(c_type i,			\
+					     atomic##prefix##_t *v)	\
+{									\
+	register c_type ret;						\
+	__asm__ __volatile__ (						\
+		"	amo" #asm_op "." #asm_type " %1, %2, %0"	\
+		: "+A" (v->counter), "=r" (ret)				\
+		: "r" (I)						\
+		: "memory");						\
+	return ret;							\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op(c_type i, atomic##prefix##_t *v)	\
+{									\
+	register c_type ret;						\
+	__asm__ __volatile__ (						\
+		"	amo" #asm_op "." #asm_type ".aqrl  %1, %2, %0"	\
+		: "+A" (v->counter), "=r" (ret)				\
+		: "r" (I)						\
+		: "memory");						\
+	return ret;							\
 }
 
-#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, asm_type, c_type, prefix)			\
-static __always_inline c_type atomic##prefix##_##op##_return##c_or(c_type i, atomic##prefix##_t *v)	\
-{													\
-        return atomic##prefix##_fetch_##op##c_or(i, v) c_op I;						\
+#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_type, c_type, prefix)	\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return_relaxed(c_type i,			\
+					      atomic##prefix##_t *v)	\
+{									\
+        return atomic##prefix##_fetch_##op##_relaxed(i, v) c_op I;	\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return(c_type i, atomic##prefix##_t *v)	\
+{									\
+        return atomic##prefix##_fetch_##op(i, v) c_op I;		\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, c_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, w,  int,   )
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_FETCH_OP( op, asm_op,       I, w,  int,   )		\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I, w,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, c_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, d, long, 64)	\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, d, long, 64)
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_FETCH_OP( op, asm_op,       I, w,  int,   )		\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I, w,  int,   )		\

arch/riscv/include/asm/atomic.h ATOMIC_OPS(xor, xor, i) · line 152

Quoted 10 lines
arch/riscv/include/asm/atomic.h +4 -4
+        ATOMIC_FETCH_OP( op, asm_op,       I, d, long, 64)		\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I, d, long, 64)
 #endif
 
-ATOMIC_OPS(add, add, +,  i,      , _relaxed)
-ATOMIC_OPS(add, add, +,  i, .aq  , _acquire)
-ATOMIC_OPS(add, add, +,  i, .rl  , _release)
-ATOMIC_OPS(add, add, +,  i, .aqrl,         )
+ATOMIC_OPS(add, add, +,  i)
+ATOMIC_OPS(sub, add, +, -i)

Sorry, I must be missing something, but how do the acquire and release atomics get generated with this new code?

Quoted 843 lines
arch/riscv/include/asm/atomic.h +196 -110
+
+#define atomic_add_return_relaxed	atomic_add_return_relaxed
+#define atomic_sub_return_relaxed	atomic_sub_return_relaxed
+#define atomic_add_return		atomic_add_return
+#define atomic_sub_return		atomic_sub_return
 
-ATOMIC_OPS(sub, add, +, -i,      , _relaxed)
-ATOMIC_OPS(sub, add, +, -i, .aq  , _acquire)
-ATOMIC_OPS(sub, add, +, -i, .rl  , _release)
-ATOMIC_OPS(sub, add, +, -i, .aqrl,         )
+#define atomic_fetch_add_relaxed	atomic_fetch_add_relaxed
+#define atomic_fetch_sub_relaxed	atomic_fetch_sub_relaxed
+#define atomic_fetch_add		atomic_fetch_add
+#define atomic_fetch_sub		atomic_fetch_sub
+
+#ifndef CONFIG_GENERIC_ATOMIC64
+#define atomic64_add_return_relaxed	atomic64_add_return_relaxed
+#define atomic64_sub_return_relaxed	atomic64_sub_return_relaxed
+#define atomic64_add_return		atomic64_add_return
+#define atomic64_sub_return		atomic64_sub_return
+
+#define atomic64_fetch_add_relaxed	atomic64_fetch_add_relaxed
+#define atomic64_fetch_sub_relaxed	atomic64_fetch_sub_relaxed
+#define atomic64_fetch_add		atomic64_fetch_add
+#define atomic64_fetch_sub		atomic64_fetch_sub
+#endif
 
 #undef ATOMIC_OPS
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, w,  int,   )
+#define ATOMIC_OPS(op, asm_op, I)					\
+        ATOMIC_FETCH_OP(op, asm_op, I, w,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, d, long, 64)
+#define ATOMIC_OPS(op, asm_op, I)					\
+        ATOMIC_FETCH_OP(op, asm_op, I, w,  int,   )			\
+        ATOMIC_FETCH_OP(op, asm_op, I, d, long, 64)
 #endif
 
-ATOMIC_OPS(and, and, i,      , _relaxed)
-ATOMIC_OPS(and, and, i, .aq  , _acquire)
-ATOMIC_OPS(and, and, i, .rl  , _release)
-ATOMIC_OPS(and, and, i, .aqrl,         )
+ATOMIC_OPS(and, and, i)
+ATOMIC_OPS( or,  or, i)
+ATOMIC_OPS(xor, xor, i)
 
-ATOMIC_OPS( or,  or, i,      , _relaxed)
-ATOMIC_OPS( or,  or, i, .aq  , _acquire)
-ATOMIC_OPS( or,  or, i, .rl  , _release)
-ATOMIC_OPS( or,  or, i, .aqrl,         )
+#define atomic_fetch_and_relaxed	atomic_fetch_and_relaxed
+#define atomic_fetch_or_relaxed		atomic_fetch_or_relaxed
+#define atomic_fetch_xor_relaxed	atomic_fetch_xor_relaxed
+#define atomic_fetch_and		atomic_fetch_and
+#define atomic_fetch_or			atomic_fetch_or
+#define atomic_fetch_xor		atomic_fetch_xor
 
-ATOMIC_OPS(xor, xor, i,      , _relaxed)
-ATOMIC_OPS(xor, xor, i, .aq  , _acquire)
-ATOMIC_OPS(xor, xor, i, .rl  , _release)
-ATOMIC_OPS(xor, xor, i, .aqrl,         )
+#ifndef CONFIG_GENERIC_ATOMIC64
+#define atomic64_fetch_and_relaxed	atomic64_fetch_and_relaxed
+#define atomic64_fetch_or_relaxed	atomic64_fetch_or_relaxed
+#define atomic64_fetch_xor_relaxed	atomic64_fetch_xor_relaxed
+#define atomic64_fetch_and		atomic64_fetch_and
+#define atomic64_fetch_or		atomic64_fetch_or
+#define atomic64_fetch_xor		atomic64_fetch_xor
+#endif
 
 #undef ATOMIC_OPS
 
@@ -157,22 +212,24 @@ ATOMIC_OPS(xor, xor, i, .aqrl,         )
 /*
  * The extra atomic operations that are constructed from one of the core
  * AMO-based operations above (aside from sub, which is easier to fit above).
- * These are required to perform a barrier, but they're OK this way because
- * atomic_*_return is also required to perform a barrier.
+ * These are required to perform a full barrier, but they're OK this way
+ * because atomic_*_return is also required to perform a full barrier.
+ *
  */
-#define ATOMIC_OP(op, func_op, comp_op, I, c_type, prefix)			\
-static __always_inline bool atomic##prefix##_##op(c_type i, atomic##prefix##_t *v) \
-{										\
-	return atomic##prefix##_##func_op##_return(i, v) comp_op I;		\
+#define ATOMIC_OP(op, func_op, comp_op, I, c_type, prefix)		\
+static __always_inline							\
+bool atomic##prefix##_##op(c_type i, atomic##prefix##_t *v)		\
+{									\
+	return atomic##prefix##_##func_op##_return(i, v) comp_op I;	\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, func_op, comp_op, I)			\
-        ATOMIC_OP (op, func_op, comp_op, I,  int,   )
+#define ATOMIC_OPS(op, func_op, comp_op, I)				\
+        ATOMIC_OP(op, func_op, comp_op, I,  int,   )
 #else
-#define ATOMIC_OPS(op, func_op, comp_op, I)			\
-        ATOMIC_OP (op, func_op, comp_op, I,  int,   )		\
-        ATOMIC_OP (op, func_op, comp_op, I, long, 64)
+#define ATOMIC_OPS(op, func_op, comp_op, I)				\
+        ATOMIC_OP(op, func_op, comp_op, I,  int,   )			\
+        ATOMIC_OP(op, func_op, comp_op, I, long, 64)
 #endif
 
 ATOMIC_OPS(add_and_test, add, ==, 0)
@@ -182,51 +239,87 @@ ATOMIC_OPS(add_negative, add,  <, 0)
 #undef ATOMIC_OP
 #undef ATOMIC_OPS
 
-#define ATOMIC_OP(op, func_op, I, c_type, prefix)				\
-static __always_inline void atomic##prefix##_##op(atomic##prefix##_t *v)	\
-{										\
-	atomic##prefix##_##func_op(I, v);					\
+#define ATOMIC_OP(op, func_op, I, c_type, prefix)			\
+static __always_inline							\
+void atomic##prefix##_##op(atomic##prefix##_t *v)			\
+{									\
+	atomic##prefix##_##func_op(I, v);				\
 }
 
-#define ATOMIC_FETCH_OP(op, func_op, I, c_type, prefix)					\
-static __always_inline c_type atomic##prefix##_fetch_##op(atomic##prefix##_t *v)	\
-{											\
-	return atomic##prefix##_fetch_##func_op(I, v);					\
+#define ATOMIC_FETCH_OP(op, func_op, I, c_type, prefix)			\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op##_relaxed(atomic##prefix##_t *v)	\
+{									\
+	return atomic##prefix##_fetch_##func_op##_relaxed(I, v);	\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op(atomic##prefix##_t *v)		\
+{									\
+	return atomic##prefix##_fetch_##func_op(I, v);			\
 }
 
-#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, c_type, prefix)				\
-static __always_inline c_type atomic##prefix##_##op##_return(atomic##prefix##_t *v)	\
-{											\
-        return atomic##prefix##_fetch_##op(v) c_op I;					\
+#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, c_type, prefix)		\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return_relaxed(atomic##prefix##_t *v)	\
+{									\
+        return atomic##prefix##_fetch_##op##_relaxed(v) c_op I;		\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return(atomic##prefix##_t *v)		\
+{									\
+        return atomic##prefix##_fetch_##op(v) c_op I;			\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, c_op, I)						\
-        ATOMIC_OP       (op, asm_op,       I,  int,   )				\
-        ATOMIC_FETCH_OP (op, asm_op,       I,  int,   )				\
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_OP(       op, asm_op,       I,  int,   )			\
+        ATOMIC_FETCH_OP( op, asm_op,       I,  int,   )			\
         ATOMIC_OP_RETURN(op, asm_op, c_op, I,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, c_op, I)						\
-        ATOMIC_OP       (op, asm_op,       I,  int,   )				\
-        ATOMIC_FETCH_OP (op, asm_op,       I,  int,   )				\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I,  int,   )				\
-        ATOMIC_OP       (op, asm_op,       I, long, 64)				\
-        ATOMIC_FETCH_OP (op, asm_op,       I, long, 64)				\
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_OP(       op, asm_op,       I,  int,   )			\
+        ATOMIC_FETCH_OP( op, asm_op,       I,  int,   )			\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I,  int,   )			\
+        ATOMIC_OP(       op, asm_op,       I, long, 64)			\
+        ATOMIC_FETCH_OP( op, asm_op,       I, long, 64)			\
         ATOMIC_OP_RETURN(op, asm_op, c_op, I, long, 64)
 #endif
 
 ATOMIC_OPS(inc, add, +,  1)
 ATOMIC_OPS(dec, add, +, -1)
 
+#define atomic_inc_return_relaxed	atomic_inc_return_relaxed
+#define atomic_dec_return_relaxed	atomic_dec_return_relaxed
+#define atomic_inc_return		atomic_inc_return
+#define atomic_dec_return		atomic_dec_return
+
+#define atomic_fetch_inc_relaxed	atomic_fetch_inc_relaxed
+#define atomic_fetch_dec_relaxed	atomic_fetch_dec_relaxed
+#define atomic_fetch_inc		atomic_fetch_inc
+#define atomic_fetch_dec		atomic_fetch_dec
+
+#ifndef CONFIG_GENERIC_ATOMIC64
+#define atomic64_inc_return_relaxed	atomic64_inc_return_relaxed
+#define atomic64_dec_return_relaxed	atomic64_dec_return_relaxed
+#define atomic64_inc_return		atomic64_inc_return
+#define atomic64_dec_return		atomic64_dec_return
+
+#define atomic64_fetch_inc_relaxed	atomic64_fetch_inc_relaxed
+#define atomic64_fetch_dec_relaxed	atomic64_fetch_dec_relaxed
+#define atomic64_fetch_inc		atomic64_fetch_inc
+#define atomic64_fetch_dec		atomic64_fetch_dec
+#endif
+
 #undef ATOMIC_OPS
 #undef ATOMIC_OP
 #undef ATOMIC_FETCH_OP
 #undef ATOMIC_OP_RETURN
 
-#define ATOMIC_OP(op, func_op, comp_op, I, prefix)				\
-static __always_inline bool atomic##prefix##_##op(atomic##prefix##_t *v)	\
-{										\
-	return atomic##prefix##_##func_op##_return(v) comp_op I;		\
+#define ATOMIC_OP(op, func_op, comp_op, I, prefix)			\
+static __always_inline							\
+bool atomic##prefix##_##op(atomic##prefix##_t *v)			\
+{									\
+	return atomic##prefix##_##func_op##_return(v) comp_op I;	\
 }
 
 ATOMIC_OP(inc_and_test, inc, ==, 0,   )
@@ -238,19 +331,19 @@ ATOMIC_OP(dec_and_test, dec, ==, 0, 64)
 
 #undef ATOMIC_OP
 
-/* This is required to provide a barrier on success. */
+/* This is required to provide a full barrier on success. */
 static __always_inline int __atomic_add_unless(atomic_t *v, int a, int u)
 {
        int prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.w.aqrl  %[p],  %[c]\n\t"
-		"beq        %[p],  %[u], 1f\n\t"
-		"add       %[rc],  %[p], %[a]\n\t"
-		"sc.w.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc], 0b\n\t"
-		"1:"
+		"0:	lr.w     %[p],  %[c]\n"
+		"	beq      %[p],  %[u], 1f\n"
+		"	add      %[rc], %[p], %[a]\n"
+		"	sc.w.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [a]"r" (a), [u]"r" (u)
 		: "memory");
@@ -263,13 +356,13 @@ static __always_inline long __atomic64_add_unless(atomic64_t *v, long a, long u)
        long prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.d.aqrl  %[p],  %[c]\n\t"
-		"beq        %[p],  %[u], 1f\n\t"
-		"add       %[rc],  %[p], %[a]\n\t"
-		"sc.d.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc], 0b\n\t"
-		"1:"
+		"0:	lr.d     %[p],  %[c]\n"
+		"	beq      %[p],  %[u], 1f\n"
+		"	add      %[rc], %[p], %[a]\n"
+		"	sc.d.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [a]"r" (a), [u]"r" (u)
 		: "memory");
@@ -300,37 +393,63 @@ static __always_inline long atomic64_inc_not_zero(atomic64_t *v)
 
 /*
  * atomic_{cmp,}xchg is required to have exactly the same ordering semantics as
- * {cmp,}xchg and the operations that return, so they need a barrier.
- */
-/*
- * FIXME: atomic_cmpxchg_{acquire,release,relaxed} are all implemented by
- * assigning the same barrier to both the LR and SC operations, but that might
- * not make any sense.  We're waiting on a memory model specification to
- * determine exactly what the right thing to do is here.
+ * {cmp,}xchg and the operations that return, so they need a full barrier.
  */
-#define ATOMIC_OP(c_t, prefix, c_or, size, asm_or)						\
-static __always_inline c_t atomic##prefix##_cmpxchg##c_or(atomic##prefix##_t *v, c_t o, c_t n) 	\
-{												\
-	return __cmpxchg(&(v->counter), o, n, size, asm_or, asm_or);				\
-}												\
-static __always_inline c_t atomic##prefix##_xchg##c_or(atomic##prefix##_t *v, c_t n) 		\
-{												\
-	return __xchg(n, &(v->counter), size, asm_or);						\
+#define ATOMIC_OP(c_t, prefix, size)					\
+static __always_inline							\
+c_t atomic##prefix##_xchg_relaxed(atomic##prefix##_t *v, c_t n)		\
+{									\
+	return __xchg_relaxed(&(v->counter), n, size);			\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_xchg_acquire(atomic##prefix##_t *v, c_t n)		\
+{									\
+	return __xchg_acquire(&(v->counter), n, size);			\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_xchg_release(atomic##prefix##_t *v, c_t n)		\
+{									\
+	return __xchg_release(&(v->counter), n, size);			\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_xchg(atomic##prefix##_t *v, c_t n)			\
+{									\
+	return __xchg(&(v->counter), n, size);				\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg_relaxed(atomic##prefix##_t *v,		\
+				     c_t o, c_t n)			\
+{									\
+	return __cmpxchg_relaxed(&(v->counter), o, n, size);		\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg_acquire(atomic##prefix##_t *v,		\
+				     c_t o, c_t n)			\
+{									\
+	return __cmpxchg_acquire(&(v->counter), o, n, size);		\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg_release(atomic##prefix##_t *v,		\
+				     c_t o, c_t n)			\
+{									\
+	return __cmpxchg_release(&(v->counter), o, n, size);		\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg(atomic##prefix##_t *v, c_t o, c_t n)	\
+{									\
+	return __cmpxchg(&(v->counter), o, n, size);			\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(c_or, asm_or)			\
-	ATOMIC_OP( int,   , c_or, 4, asm_or)
+#define ATOMIC_OPS()							\
+	ATOMIC_OP( int,   , 4)
 #else
-#define ATOMIC_OPS(c_or, asm_or)			\
-	ATOMIC_OP( int,   , c_or, 4, asm_or)		\
-	ATOMIC_OP(long, 64, c_or, 8, asm_or)
+#define ATOMIC_OPS()							\
+	ATOMIC_OP( int,   , 4)						\
+	ATOMIC_OP(long, 64, 8)
 #endif
 
-ATOMIC_OPS(        , .aqrl)
-ATOMIC_OPS(_acquire,   .aq)
-ATOMIC_OPS(_release,   .rl)
-ATOMIC_OPS(_relaxed,      )
+ATOMIC_OPS()
 
 #undef ATOMIC_OPS
 #undef ATOMIC_OP
@@ -340,13 +459,13 @@ static __always_inline int atomic_sub_if_positive(atomic_t *v, int offset)
        int prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.w.aqrl  %[p],  %[c]\n\t"
-		"sub       %[rc],  %[p], %[o]\n\t"
-		"bltz      %[rc],    1f\n\t"
-		"sc.w.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc],    0b\n\t"
-		"1:"
+		"0:	lr.w     %[p],  %[c]\n"
+		"	sub      %[rc], %[p], %[o]\n"
+		"	bltz     %[rc], 1f\n"
+		"	sc.w.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [o]"r" (offset)
 		: "memory");
@@ -361,13 +480,13 @@ static __always_inline long atomic64_sub_if_positive(atomic64_t *v, int offset)
        long prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.d.aqrl  %[p],  %[c]\n\t"
-		"sub       %[rc],  %[p], %[o]\n\t"
-		"bltz      %[rc],    1f\n\t"
-		"sc.d.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc],    0b\n\t"
-		"1:"
+		"0:	lr.d     %[p],  %[c]\n"
+		"	sub      %[rc], %[p], %[o]\n"
+		"	bltz     %[rc], 1f\n"
+		"	sc.d.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [o]"r" (offset)
 		: "memory");
arch/riscv/include/asm/cmpxchg.h +320 -71
index db249dbc7b976..c12833f7b6bd1 100644
@@ -17,45 +17,153 @@
 #include <linux/bug.h>
 
 #include <asm/barrier.h>
+#include <asm/fence.h>
 
-#define __xchg(new, ptr, size, asm_or)				\
-({								\
-	__typeof__(ptr) __ptr = (ptr);				\
-	__typeof__(new) __new = (new);				\
-	__typeof__(*(ptr)) __ret;				\
-	switch (size) {						\
-	case 4:							\
-		__asm__ __volatile__ (				\
-			"amoswap.w" #asm_or " %0, %2, %1"	\
-			: "=r" (__ret), "+A" (*__ptr)		\
-			: "r" (__new)				\
-			: "memory");				\
-		break;						\
-	case 8:							\
-		__asm__ __volatile__ (				\
-			"amoswap.d" #asm_or " %0, %2, %1"	\
-			: "=r" (__ret), "+A" (*__ptr)		\
-			: "r" (__new)				\
-			: "memory");				\
-		break;						\
-	default:						\
-		BUILD_BUG();					\
-	}							\
-	__ret;							\
-})
-
-#define xchg(ptr, x)    (__xchg((x), (ptr), sizeof(*(ptr)), .aqrl))
-
-#define xchg32(ptr, x)				\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 4);	\
-	xchg((ptr), (x));			\
-})
-
-#define xchg64(ptr, x)				\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 8);	\
-	xchg((ptr), (x));			\
+#define __xchg_relaxed(ptr, new, size)					\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(new) __new = (new);					\
+	__typeof__(*(ptr)) __ret;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.w %0, %2, %1\n"			\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.d %0, %2, %1\n"			\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define xchg_relaxed(ptr, x)						\
+({									\
+	__typeof__(*(ptr)) _x_ = (x);					\
+	(__typeof__(*(ptr))) __xchg_relaxed((ptr),			\
+					    _x_, sizeof(*(ptr)));	\
+})
+
+#define __xchg_acquire(ptr, new, size)					\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(new) __new = (new);					\
+	__typeof__(*(ptr)) __ret;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.w %0, %2, %1\n"			\
+			RISCV_ACQUIRE_BARRIER				\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.d %0, %2, %1\n"			\
+			RISCV_ACQUIRE_BARRIER				\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define xchg_acquire(ptr, x)						\
+({									\
+	__typeof__(*(ptr)) _x_ = (x);					\
+	(__typeof__(*(ptr))) __xchg_acquire((ptr),			\
+					    _x_, sizeof(*(ptr)));	\
+})
+
+#define __xchg_release(ptr, new, size)					\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(new) __new = (new);					\
+	__typeof__(*(ptr)) __ret;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			RISCV_RELEASE_BARRIER				\
+			"	amoswap.w %0, %2, %1\n"			\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			RISCV_RELEASE_BARRIER				\
+			"	amoswap.d %0, %2, %1\n"			\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define xchg_release(ptr, x)						\
+({									\
+	__typeof__(*(ptr)) _x_ = (x);					\
+	(__typeof__(*(ptr))) __xchg_release((ptr),			\
+					    _x_, sizeof(*(ptr)));	\
+})
+
+#define __xchg(ptr, new, size)						\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(new) __new = (new);					\
+	__typeof__(*(ptr)) __ret;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.w.aqrl %0, %2, %1\n"		\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.d.aqrl %0, %2, %1\n"		\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define xchg(ptr, x)							\
+({									\
+	__typeof__(*(ptr)) _x_ = (x);					\
+	(__typeof__(*(ptr))) __xchg((ptr), _x_, sizeof(*(ptr)));	\
+})
+
+#define xchg32(ptr, x)							\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 4);				\
+	xchg((ptr), (x));						\
+})
+
+#define xchg64(ptr, x)							\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 8);				\
+	xchg((ptr), (x));						\
 })
 
 /*
@@ -63,7 +171,51 @@
  * store NEW in MEM.  Return the initial value in MEM.  Success is
  * indicated by comparing RETURN with OLD.
  */
-#define __cmpxchg(ptr, old, new, size, lrb, scb)			\
+#define __cmpxchg_relaxed(ptr, old, new, size)				\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(*(ptr)) __old = (old);				\
+	__typeof__(*(ptr)) __new = (new);				\
+	__typeof__(*(ptr)) __ret;					\
+	register unsigned int __rc;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"0:	lr.w %0, %2\n"				\
+			"	bne  %0, %z3, 1f\n"			\
+			"	sc.w %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"0:	lr.d %0, %2\n"				\
+			"	bne %0, %z3, 1f\n"			\
+			"	sc.d %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define cmpxchg_relaxed(ptr, o, n)					\
+({									\
+	__typeof__(*(ptr)) _o_ = (o);					\
+	__typeof__(*(ptr)) _n_ = (n);					\
+	(__typeof__(*(ptr))) __cmpxchg_relaxed((ptr),			\
+					_o_, _n_, sizeof(*(ptr)));	\
+})
+
+#define __cmpxchg_acquire(ptr, old, new, size)				\
 ({									\
 	__typeof__(ptr) __ptr = (ptr);					\
 	__typeof__(*(ptr)) __old = (old);				\
@@ -73,24 +225,24 @@
 	switch (size) {							\
 	case 4:								\
 		__asm__ __volatile__ (					\
-		"0:"							\
-			"lr.w" #scb " %0, %2\n"				\
-			"bne         %0, %z3, 1f\n"			\
-			"sc.w" #lrb " %1, %z4, %2\n"			\
-			"bnez        %1, 0b\n"				\
-		"1:"							\
+			"0:	lr.w %0, %2\n"				\
+			"	bne  %0, %z3, 1f\n"			\
+			"	sc.w %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			RISCV_ACQUIRE_BARRIER				\
+			"1:\n"						\
 			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
 			: "rJ" (__old), "rJ" (__new)			\
 			: "memory");					\
 		break;							\
 	case 8:								\
 		__asm__ __volatile__ (					\
-		"0:"							\
-			"lr.d" #scb " %0, %2\n"				\
-			"bne         %0, %z3, 1f\n"			\
-			"sc.d" #lrb " %1, %z4, %2\n"			\
-			"bnez        %1, 0b\n"				\
-		"1:"							\
+			"0:	lr.d %0, %2\n"				\
+			"	bne %0, %z3, 1f\n"			\
+			"	sc.d %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			RISCV_ACQUIRE_BARRIER				\
+			"1:\n"						\
 			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
 			: "rJ" (__old), "rJ" (__new)			\
 			: "memory");					\
@@ -101,34 +253,131 @@
 	__ret;								\
 })
 
-#define cmpxchg(ptr, o, n) \
-	(__cmpxchg((ptr), (o), (n), sizeof(*(ptr)), .aqrl, .aqrl))
+#define cmpxchg_acquire(ptr, o, n)					\
+({									\
+	__typeof__(*(ptr)) _o_ = (o);					\
+	__typeof__(*(ptr)) _n_ = (n);					\
+	(__typeof__(*(ptr))) __cmpxchg_acquire((ptr),			\
+					_o_, _n_, sizeof(*(ptr)));	\
+})
 
-#define cmpxchg_local(ptr, o, n) \
-	(__cmpxchg((ptr), (o), (n), sizeof(*(ptr)), , ))
+#define __cmpxchg_release(ptr, old, new, size)				\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(*(ptr)) __old = (old);				\
+	__typeof__(*(ptr)) __new = (new);				\
+	__typeof__(*(ptr)) __ret;					\
+	register unsigned int __rc;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			RISCV_RELEASE_BARRIER				\
+			"0:	lr.w %0, %2\n"				\
+			"	bne  %0, %z3, 1f\n"			\
+			"	sc.w %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			RISCV_RELEASE_BARRIER				\
+			"0:	lr.d %0, %2\n"				\
+			"	bne %0, %z3, 1f\n"			\
+			"	sc.d %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define cmpxchg_release(ptr, o, n)					\
+({									\
+	__typeof__(*(ptr)) _o_ = (o);					\
+	__typeof__(*(ptr)) _n_ = (n);					\
+	(__typeof__(*(ptr))) __cmpxchg_release((ptr),			\
+					_o_, _n_, sizeof(*(ptr)));	\
+})
+
+#define __cmpxchg(ptr, old, new, size)					\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(*(ptr)) __old = (old);				\
+	__typeof__(*(ptr)) __new = (new);				\
+	__typeof__(*(ptr)) __ret;					\
+	register unsigned int __rc;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"0:	lr.w %0, %2\n"				\
+			"	bne  %0, %z3, 1f\n"			\
+			"	sc.w.rl %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"	fence rw, rw\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"0:	lr.d %0, %2\n"				\
+			"	bne %0, %z3, 1f\n"			\
+			"	sc.d.rl %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"	fence rw, rw\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
 
-#define cmpxchg32(ptr, o, n)			\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 4);	\
-	cmpxchg((ptr), (o), (n));		\
+#define cmpxchg(ptr, o, n)						\
+({									\
+	__typeof__(*(ptr)) _o_ = (o);					\
+	__typeof__(*(ptr)) _n_ = (n);					\
+	(__typeof__(*(ptr))) __cmpxchg((ptr),				\
+				       _o_, _n_, sizeof(*(ptr)));	\
 })
 
-#define cmpxchg32_local(ptr, o, n)		\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 4);	\
-	cmpxchg_local((ptr), (o), (n));		\
+#define cmpxchg_local(ptr, o, n)					\
+	(__cmpxchg_relaxed((ptr), (o), (n), sizeof(*(ptr))))
+
+#define cmpxchg32(ptr, o, n)						\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 4);				\
+	cmpxchg((ptr), (o), (n));					\
 })
 
-#define cmpxchg64(ptr, o, n)			\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 8);	\
-	cmpxchg((ptr), (o), (n));		\
+#define cmpxchg32_local(ptr, o, n)					\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 4);				\
+	cmpxchg_relaxed((ptr), (o), (n))				\
 })
 
-#define cmpxchg64_local(ptr, o, n)		\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 8);	\
-	cmpxchg_local((ptr), (o), (n));		\
+#define cmpxchg64(ptr, o, n)						\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 8);				\
+	cmpxchg((ptr), (o), (n));					\
+})
+
+#define cmpxchg64_local(ptr, o, n)					\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 8);				\
+	cmpxchg_relaxed((ptr), (o), (n));				\
 })
 
 #endif /* _ASM_RISCV_CMPXCHG_H */
#4 lore raw
Quoted 134 lines · Palmer Dabbelt
Quoted 133 lines · parri.andrea at gmail.com

Atomics present the same issue with locking: release and acquire variants need to be strengthened to meet the constraints defined by the Linux-kernel memory consistency model [1].

Atomics present a further issue: implementations of atomics such as atomic_cmpxchg() and atomic_add_unless() rely on LR/SC pairs, which do not give full-ordering with .aqrl; for example, current implementations allow the "lr-sc-aqrl-pair-vs-full-barrier" test below to end up with the state indicated in the "exists" clause.

In order to "synchronize" LKMM and RISC-V's implementation, this commit strengthens the implementations of the atomics operations by replacing .rl and .aq with the use of ("lightweigth") fences, and by replacing .aqrl LR/SC pairs in sequences such as:

0: lr.w.aqrl %0, %addr

         bne        %0, %old, 1f
         ...
         sc.w.aqrl  %1, %new, %addr
         bnez       %1, 0b

1:

with sequences of the form:

0: lr.w %0, %addr

         bne        %0, %old, 1f
         ...
         sc.w.rl    %1, %new, %addr   /* SC-release   */
         bnez       %1, 0b
         fence      rw, rw            /* "full" fence */

1:

following Daniel's suggestion.

These modifications were validated with simulation of the RISC-V memory consistency model.

C lr-sc-aqrl-pair-vs-full-barrier

{}

P0(int *x, int *y, atomic_t *u)

{
	int r0;
	int r1;

	WRITE_ONCE(*x, 1);
	r0 = atomic_cmpxchg(u, 0, 1);
	r1 = READ_ONCE(*y);
}

P1(int *x, int *y, atomic_t *v)

{
	int r0;
	int r1;

	WRITE_ONCE(*y, 1);
	r0 = atomic_cmpxchg(v, 0, 1);
	r1 = READ_ONCE(*x);
}

exists (u=1 /\ v=1 /\ 0:r1=0 /\ 1:r1=0)

[1] https://marc.info/?l=linux-kernel&m=151930201102853&w=2
https://groups.google.com/a/groups.riscv.org/forum/#!topic/isa-dev/hKywNHBkAXM
https://marc.info/?l=linux-kernel&m=151633436614259&w=2

Cc Palmer Dabbelt <[email protected]>
Cc Albert Ou <[email protected]>
Cc Daniel Lustig <[email protected]>
Cc Alan Stern <[email protected]>
Cc Will Deacon <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Boqun Feng <[email protected]>
Cc Nicholas Piggin <[email protected]>
Cc David Howells <[email protected]>
Cc Jade Alglave <[email protected]>
Cc Luc Maranget <[email protected]>
Cc "Paul E. McKenney" <[email protected]>
Cc Akira Yokosawa <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc linux-riscv at lists.infradead.org
Cc linux-kernel at vger.kernel.org

---
arch/riscv/include/asm/atomic.h | 417 +++++++++++++++++++++++++-------------- arch/riscv/include/asm/cmpxchg.h | 391 +++++++++++++++++++++++++++++-------

2 files changed, 588 insertions(+), 220 deletions(-)
arch/riscv/include/asm/atomic.h +18 -9
index e65d1cd89e28b..855115ace98c8 100644
@@ -24,6 +24,20 @@
 #include <asm/barrier.h>
 
 #define ATOMIC_INIT(i)	{ (i) }
+
+#define __atomic_op_acquire(op, args...)				\
+({									\
+	typeof(op##_relaxed(args)) __ret  = op##_relaxed(args);		\
+	__asm__ __volatile__(RISCV_ACQUIRE_BARRIER "" ::: "memory");	\
+	__ret;								\
+})
+
+#define __atomic_op_release(op, args...)				\
+({									\
+	__asm__ __volatile__(RISCV_RELEASE_BARRIER "" ::: "memory");	\
+	op##_relaxed(args);						\
+})
+
 static __always_inline int atomic_read(const atomic_t *v)
 {
 	return READ_ONCE(v->counter);
@@ -50,22 +64,23 @@ static __always_inline void atomic64_set(atomic64_t *v, long i)
 * have the AQ or RL bits set.  These don't return anything, so there's only
 * one version to worry about.
 */
-#define ATOMIC_OP(op, asm_op, I, asm_type, c_type, prefix)				\
-static __always_inline void atomic##prefix##_##op(c_type i, atomic##prefix##_t *v)	\
-{											\
-	__asm__ __volatile__ (								\
-		"amo" #asm_op "." #asm_type " zero, %1, %0"				\
-		: "+A" (v->counter)							\
-		: "r" (I)								\
-		: "memory");								\
-}
+#define ATOMIC_OP(op, asm_op, I, asm_type, c_type, prefix)		\
+static __always_inline							\
+void atomic##prefix##_##op(c_type i, atomic##prefix##_t *v)		\
+{									\
Quoted 8 lines

arch/riscv/include/asm/atomic.h static __always_inline void atomic64_set(atomic64_t *v, long i) · line 76

Quoted 6 lines
arch/riscv/include/asm/atomic.h +6 -0
+	__asm__ __volatile__ (						\
+		"	amo" #asm_op "." #asm_type " zero, %1, %0"	\
+		: "+A" (v->counter)					\
+		: "r" (I)						\
+		: "memory");						\
+}									\

Just to be sure: this is just a whitespace change, right?

This belongs to the "few style fixes" (in the specific, 80-chars lines) mentioned in the cover letter; I could not resist ;-), but I'll remove them in v3 if you like so.

Quoted 94 lines
Quoted 94 lines
arch/riscv/include/asm/atomic.h +46 -26
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, I)			\
+#define ATOMIC_OPS(op, asm_op, I)					\
        ATOMIC_OP (op, asm_op, I, w,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, I)			\
-        ATOMIC_OP (op, asm_op, I, w,  int,   )	\
+#define ATOMIC_OPS(op, asm_op, I)					\
+        ATOMIC_OP (op, asm_op, I, w,  int,   )				\
        ATOMIC_OP (op, asm_op, I, d, long, 64)
 #endif
 
@@ -79,75 +94,115 @@ ATOMIC_OPS(xor, xor,  i)
 #undef ATOMIC_OPS
 
 /*
- * Atomic ops that have ordered, relaxed, acquire, and relese variants.
+ * Atomic ops that have ordered, relaxed, acquire, and release variants.
 * There's two flavors of these: the arithmatic ops have both fetch and return
 * versions, while the logical ops only have fetch versions.
 */
-#define ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, asm_type, c_type, prefix)				\
-static __always_inline c_type atomic##prefix##_fetch_##op##c_or(c_type i, atomic##prefix##_t *v)	\
-{													\
-	register c_type ret;										\
-	__asm__ __volatile__ (										\
-		"amo" #asm_op "." #asm_type #asm_or " %1, %2, %0"					\
-		: "+A" (v->counter), "=r" (ret)								\
-		: "r" (I)										\
-		: "memory");										\
-	return ret;											\
+#define ATOMIC_FETCH_OP(op, asm_op, I, asm_type, c_type, prefix)	\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op##_relaxed(c_type i,			\
+					     atomic##prefix##_t *v)	\
+{									\
+	register c_type ret;						\
+	__asm__ __volatile__ (						\
+		"	amo" #asm_op "." #asm_type " %1, %2, %0"	\
+		: "+A" (v->counter), "=r" (ret)				\
+		: "r" (I)						\
+		: "memory");						\
+	return ret;							\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op(c_type i, atomic##prefix##_t *v)	\
+{									\
+	register c_type ret;						\
+	__asm__ __volatile__ (						\
+		"	amo" #asm_op "." #asm_type ".aqrl  %1, %2, %0"	\
+		: "+A" (v->counter), "=r" (ret)				\
+		: "r" (I)						\
+		: "memory");						\
+	return ret;							\
 }
 
-#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, asm_type, c_type, prefix)			\
-static __always_inline c_type atomic##prefix##_##op##_return##c_or(c_type i, atomic##prefix##_t *v)	\
-{													\
-        return atomic##prefix##_fetch_##op##c_or(i, v) c_op I;						\
+#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_type, c_type, prefix)	\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return_relaxed(c_type i,			\
+					      atomic##prefix##_t *v)	\
+{									\
+        return atomic##prefix##_fetch_##op##_relaxed(i, v) c_op I;	\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return(c_type i, atomic##prefix##_t *v)	\
+{									\
+        return atomic##prefix##_fetch_##op(i, v) c_op I;		\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, c_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, w,  int,   )
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_FETCH_OP( op, asm_op,       I, w,  int,   )		\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I, w,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, c_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, d, long, 64)	\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, d, long, 64)
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_FETCH_OP( op, asm_op,       I, w,  int,   )		\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I, w,  int,   )		\
+        ATOMIC_FETCH_OP( op, asm_op,       I, d, long, 64)		\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I, d, long, 64)
 #endif
 
Quoted 9 lines

arch/riscv/include/asm/atomic.h ATOMIC_OPS(xor, xor, i) · line 152

Quoted 6 lines
arch/riscv/include/asm/atomic.h +2 -4
-ATOMIC_OPS(add, add, +,  i,      , _relaxed)
-ATOMIC_OPS(add, add, +,  i, .aq  , _acquire)
-ATOMIC_OPS(add, add, +,  i, .rl  , _release)
-ATOMIC_OPS(add, add, +,  i, .aqrl,         )
+ATOMIC_OPS(add, add, +,  i)
+ATOMIC_OPS(sub, add, +, -i)

Sorry, I must be missing something, but how do the acquire and release atomics get generated with this new code?

This proposal relies on the generic definition,

include/linux/atomic.h ,

and on the

__atomic_op_acquire()
__atomic_op_release()

above to build the acquire/release atomics (except for the xchg,cmpxchg, where the ACQUIRE_BARRIER is inserted conditionally/on success).

Andrea

Quoted 844 lines
Quoted 844 lines
arch/riscv/include/asm/atomic.h +196 -110
+
+#define atomic_add_return_relaxed	atomic_add_return_relaxed
+#define atomic_sub_return_relaxed	atomic_sub_return_relaxed
+#define atomic_add_return		atomic_add_return
+#define atomic_sub_return		atomic_sub_return
 
-ATOMIC_OPS(sub, add, +, -i,      , _relaxed)
-ATOMIC_OPS(sub, add, +, -i, .aq  , _acquire)
-ATOMIC_OPS(sub, add, +, -i, .rl  , _release)
-ATOMIC_OPS(sub, add, +, -i, .aqrl,         )
+#define atomic_fetch_add_relaxed	atomic_fetch_add_relaxed
+#define atomic_fetch_sub_relaxed	atomic_fetch_sub_relaxed
+#define atomic_fetch_add		atomic_fetch_add
+#define atomic_fetch_sub		atomic_fetch_sub
+
+#ifndef CONFIG_GENERIC_ATOMIC64
+#define atomic64_add_return_relaxed	atomic64_add_return_relaxed
+#define atomic64_sub_return_relaxed	atomic64_sub_return_relaxed
+#define atomic64_add_return		atomic64_add_return
+#define atomic64_sub_return		atomic64_sub_return
+
+#define atomic64_fetch_add_relaxed	atomic64_fetch_add_relaxed
+#define atomic64_fetch_sub_relaxed	atomic64_fetch_sub_relaxed
+#define atomic64_fetch_add		atomic64_fetch_add
+#define atomic64_fetch_sub		atomic64_fetch_sub
+#endif
 
 #undef ATOMIC_OPS
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, w,  int,   )
+#define ATOMIC_OPS(op, asm_op, I)					\
+        ATOMIC_FETCH_OP(op, asm_op, I, w,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, d, long, 64)
+#define ATOMIC_OPS(op, asm_op, I)					\
+        ATOMIC_FETCH_OP(op, asm_op, I, w,  int,   )			\
+        ATOMIC_FETCH_OP(op, asm_op, I, d, long, 64)
 #endif
 
-ATOMIC_OPS(and, and, i,      , _relaxed)
-ATOMIC_OPS(and, and, i, .aq  , _acquire)
-ATOMIC_OPS(and, and, i, .rl  , _release)
-ATOMIC_OPS(and, and, i, .aqrl,         )
+ATOMIC_OPS(and, and, i)
+ATOMIC_OPS( or,  or, i)
+ATOMIC_OPS(xor, xor, i)
 
-ATOMIC_OPS( or,  or, i,      , _relaxed)
-ATOMIC_OPS( or,  or, i, .aq  , _acquire)
-ATOMIC_OPS( or,  or, i, .rl  , _release)
-ATOMIC_OPS( or,  or, i, .aqrl,         )
+#define atomic_fetch_and_relaxed	atomic_fetch_and_relaxed
+#define atomic_fetch_or_relaxed		atomic_fetch_or_relaxed
+#define atomic_fetch_xor_relaxed	atomic_fetch_xor_relaxed
+#define atomic_fetch_and		atomic_fetch_and
+#define atomic_fetch_or			atomic_fetch_or
+#define atomic_fetch_xor		atomic_fetch_xor
 
-ATOMIC_OPS(xor, xor, i,      , _relaxed)
-ATOMIC_OPS(xor, xor, i, .aq  , _acquire)
-ATOMIC_OPS(xor, xor, i, .rl  , _release)
-ATOMIC_OPS(xor, xor, i, .aqrl,         )
+#ifndef CONFIG_GENERIC_ATOMIC64
+#define atomic64_fetch_and_relaxed	atomic64_fetch_and_relaxed
+#define atomic64_fetch_or_relaxed	atomic64_fetch_or_relaxed
+#define atomic64_fetch_xor_relaxed	atomic64_fetch_xor_relaxed
+#define atomic64_fetch_and		atomic64_fetch_and
+#define atomic64_fetch_or		atomic64_fetch_or
+#define atomic64_fetch_xor		atomic64_fetch_xor
+#endif
 
 #undef ATOMIC_OPS
 
@@ -157,22 +212,24 @@ ATOMIC_OPS(xor, xor, i, .aqrl,         )
 /*
 * The extra atomic operations that are constructed from one of the core
 * AMO-based operations above (aside from sub, which is easier to fit above).
- * These are required to perform a barrier, but they're OK this way because
- * atomic_*_return is also required to perform a barrier.
+ * These are required to perform a full barrier, but they're OK this way
+ * because atomic_*_return is also required to perform a full barrier.
+ *
 */
-#define ATOMIC_OP(op, func_op, comp_op, I, c_type, prefix)			\
-static __always_inline bool atomic##prefix##_##op(c_type i, atomic##prefix##_t *v) \
-{										\
-	return atomic##prefix##_##func_op##_return(i, v) comp_op I;		\
+#define ATOMIC_OP(op, func_op, comp_op, I, c_type, prefix)		\
+static __always_inline							\
+bool atomic##prefix##_##op(c_type i, atomic##prefix##_t *v)		\
+{									\
+	return atomic##prefix##_##func_op##_return(i, v) comp_op I;	\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, func_op, comp_op, I)			\
-        ATOMIC_OP (op, func_op, comp_op, I,  int,   )
+#define ATOMIC_OPS(op, func_op, comp_op, I)				\
+        ATOMIC_OP(op, func_op, comp_op, I,  int,   )
 #else
-#define ATOMIC_OPS(op, func_op, comp_op, I)			\
-        ATOMIC_OP (op, func_op, comp_op, I,  int,   )		\
-        ATOMIC_OP (op, func_op, comp_op, I, long, 64)
+#define ATOMIC_OPS(op, func_op, comp_op, I)				\
+        ATOMIC_OP(op, func_op, comp_op, I,  int,   )			\
+        ATOMIC_OP(op, func_op, comp_op, I, long, 64)
 #endif
 
 ATOMIC_OPS(add_and_test, add, ==, 0)
@@ -182,51 +239,87 @@ ATOMIC_OPS(add_negative, add,  <, 0)
 #undef ATOMIC_OP
 #undef ATOMIC_OPS
 
-#define ATOMIC_OP(op, func_op, I, c_type, prefix)				\
-static __always_inline void atomic##prefix##_##op(atomic##prefix##_t *v)	\
-{										\
-	atomic##prefix##_##func_op(I, v);					\
+#define ATOMIC_OP(op, func_op, I, c_type, prefix)			\
+static __always_inline							\
+void atomic##prefix##_##op(atomic##prefix##_t *v)			\
+{									\
+	atomic##prefix##_##func_op(I, v);				\
 }
 
-#define ATOMIC_FETCH_OP(op, func_op, I, c_type, prefix)					\
-static __always_inline c_type atomic##prefix##_fetch_##op(atomic##prefix##_t *v)	\
-{											\
-	return atomic##prefix##_fetch_##func_op(I, v);					\
+#define ATOMIC_FETCH_OP(op, func_op, I, c_type, prefix)			\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op##_relaxed(atomic##prefix##_t *v)	\
+{									\
+	return atomic##prefix##_fetch_##func_op##_relaxed(I, v);	\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op(atomic##prefix##_t *v)		\
+{									\
+	return atomic##prefix##_fetch_##func_op(I, v);			\
 }
 
-#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, c_type, prefix)				\
-static __always_inline c_type atomic##prefix##_##op##_return(atomic##prefix##_t *v)	\
-{											\
-        return atomic##prefix##_fetch_##op(v) c_op I;					\
+#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, c_type, prefix)		\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return_relaxed(atomic##prefix##_t *v)	\
+{									\
+        return atomic##prefix##_fetch_##op##_relaxed(v) c_op I;		\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return(atomic##prefix##_t *v)		\
+{									\
+        return atomic##prefix##_fetch_##op(v) c_op I;			\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, c_op, I)						\
-        ATOMIC_OP       (op, asm_op,       I,  int,   )				\
-        ATOMIC_FETCH_OP (op, asm_op,       I,  int,   )				\
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_OP(       op, asm_op,       I,  int,   )			\
+        ATOMIC_FETCH_OP( op, asm_op,       I,  int,   )			\
        ATOMIC_OP_RETURN(op, asm_op, c_op, I,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, c_op, I)						\
-        ATOMIC_OP       (op, asm_op,       I,  int,   )				\
-        ATOMIC_FETCH_OP (op, asm_op,       I,  int,   )				\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I,  int,   )				\
-        ATOMIC_OP       (op, asm_op,       I, long, 64)				\
-        ATOMIC_FETCH_OP (op, asm_op,       I, long, 64)				\
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_OP(       op, asm_op,       I,  int,   )			\
+        ATOMIC_FETCH_OP( op, asm_op,       I,  int,   )			\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I,  int,   )			\
+        ATOMIC_OP(       op, asm_op,       I, long, 64)			\
+        ATOMIC_FETCH_OP( op, asm_op,       I, long, 64)			\
        ATOMIC_OP_RETURN(op, asm_op, c_op, I, long, 64)
 #endif
 
 ATOMIC_OPS(inc, add, +,  1)
 ATOMIC_OPS(dec, add, +, -1)
 
+#define atomic_inc_return_relaxed	atomic_inc_return_relaxed
+#define atomic_dec_return_relaxed	atomic_dec_return_relaxed
+#define atomic_inc_return		atomic_inc_return
+#define atomic_dec_return		atomic_dec_return
+
+#define atomic_fetch_inc_relaxed	atomic_fetch_inc_relaxed
+#define atomic_fetch_dec_relaxed	atomic_fetch_dec_relaxed
+#define atomic_fetch_inc		atomic_fetch_inc
+#define atomic_fetch_dec		atomic_fetch_dec
+
+#ifndef CONFIG_GENERIC_ATOMIC64
+#define atomic64_inc_return_relaxed	atomic64_inc_return_relaxed
+#define atomic64_dec_return_relaxed	atomic64_dec_return_relaxed
+#define atomic64_inc_return		atomic64_inc_return
+#define atomic64_dec_return		atomic64_dec_return
+
+#define atomic64_fetch_inc_relaxed	atomic64_fetch_inc_relaxed
+#define atomic64_fetch_dec_relaxed	atomic64_fetch_dec_relaxed
+#define atomic64_fetch_inc		atomic64_fetch_inc
+#define atomic64_fetch_dec		atomic64_fetch_dec
+#endif
+
 #undef ATOMIC_OPS
 #undef ATOMIC_OP
 #undef ATOMIC_FETCH_OP
 #undef ATOMIC_OP_RETURN
 
-#define ATOMIC_OP(op, func_op, comp_op, I, prefix)				\
-static __always_inline bool atomic##prefix##_##op(atomic##prefix##_t *v)	\
-{										\
-	return atomic##prefix##_##func_op##_return(v) comp_op I;		\
+#define ATOMIC_OP(op, func_op, comp_op, I, prefix)			\
+static __always_inline							\
+bool atomic##prefix##_##op(atomic##prefix##_t *v)			\
+{									\
+	return atomic##prefix##_##func_op##_return(v) comp_op I;	\
 }
 
 ATOMIC_OP(inc_and_test, inc, ==, 0,   )
@@ -238,19 +331,19 @@ ATOMIC_OP(dec_and_test, dec, ==, 0, 64)
 
 #undef ATOMIC_OP
 
-/* This is required to provide a barrier on success. */
+/* This is required to provide a full barrier on success. */
 static __always_inline int __atomic_add_unless(atomic_t *v, int a, int u)
 {
       int prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.w.aqrl  %[p],  %[c]\n\t"
-		"beq        %[p],  %[u], 1f\n\t"
-		"add       %[rc],  %[p], %[a]\n\t"
-		"sc.w.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc], 0b\n\t"
-		"1:"
+		"0:	lr.w     %[p],  %[c]\n"
+		"	beq      %[p],  %[u], 1f\n"
+		"	add      %[rc], %[p], %[a]\n"
+		"	sc.w.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [a]"r" (a), [u]"r" (u)
 		: "memory");
@@ -263,13 +356,13 @@ static __always_inline long __atomic64_add_unless(atomic64_t *v, long a, long u)
       long prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.d.aqrl  %[p],  %[c]\n\t"
-		"beq        %[p],  %[u], 1f\n\t"
-		"add       %[rc],  %[p], %[a]\n\t"
-		"sc.d.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc], 0b\n\t"
-		"1:"
+		"0:	lr.d     %[p],  %[c]\n"
+		"	beq      %[p],  %[u], 1f\n"
+		"	add      %[rc], %[p], %[a]\n"
+		"	sc.d.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [a]"r" (a), [u]"r" (u)
 		: "memory");
@@ -300,37 +393,63 @@ static __always_inline long atomic64_inc_not_zero(atomic64_t *v)
 
 /*
 * atomic_{cmp,}xchg is required to have exactly the same ordering semantics as
- * {cmp,}xchg and the operations that return, so they need a barrier.
- */
-/*
- * FIXME: atomic_cmpxchg_{acquire,release,relaxed} are all implemented by
- * assigning the same barrier to both the LR and SC operations, but that might
- * not make any sense.  We're waiting on a memory model specification to
- * determine exactly what the right thing to do is here.
+ * {cmp,}xchg and the operations that return, so they need a full barrier.
 */
-#define ATOMIC_OP(c_t, prefix, c_or, size, asm_or)						\
-static __always_inline c_t atomic##prefix##_cmpxchg##c_or(atomic##prefix##_t *v, c_t o, c_t n) 	\
-{												\
-	return __cmpxchg(&(v->counter), o, n, size, asm_or, asm_or);				\
-}												\
-static __always_inline c_t atomic##prefix##_xchg##c_or(atomic##prefix##_t *v, c_t n) 		\
-{												\
-	return __xchg(n, &(v->counter), size, asm_or);						\
+#define ATOMIC_OP(c_t, prefix, size)					\
+static __always_inline							\
+c_t atomic##prefix##_xchg_relaxed(atomic##prefix##_t *v, c_t n)		\
+{									\
+	return __xchg_relaxed(&(v->counter), n, size);			\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_xchg_acquire(atomic##prefix##_t *v, c_t n)		\
+{									\
+	return __xchg_acquire(&(v->counter), n, size);			\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_xchg_release(atomic##prefix##_t *v, c_t n)		\
+{									\
+	return __xchg_release(&(v->counter), n, size);			\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_xchg(atomic##prefix##_t *v, c_t n)			\
+{									\
+	return __xchg(&(v->counter), n, size);				\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg_relaxed(atomic##prefix##_t *v,		\
+				     c_t o, c_t n)			\
+{									\
+	return __cmpxchg_relaxed(&(v->counter), o, n, size);		\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg_acquire(atomic##prefix##_t *v,		\
+				     c_t o, c_t n)			\
+{									\
+	return __cmpxchg_acquire(&(v->counter), o, n, size);		\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg_release(atomic##prefix##_t *v,		\
+				     c_t o, c_t n)			\
+{									\
+	return __cmpxchg_release(&(v->counter), o, n, size);		\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg(atomic##prefix##_t *v, c_t o, c_t n)	\
+{									\
+	return __cmpxchg(&(v->counter), o, n, size);			\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(c_or, asm_or)			\
-	ATOMIC_OP( int,   , c_or, 4, asm_or)
+#define ATOMIC_OPS()							\
+	ATOMIC_OP( int,   , 4)
 #else
-#define ATOMIC_OPS(c_or, asm_or)			\
-	ATOMIC_OP( int,   , c_or, 4, asm_or)		\
-	ATOMIC_OP(long, 64, c_or, 8, asm_or)
+#define ATOMIC_OPS()							\
+	ATOMIC_OP( int,   , 4)						\
+	ATOMIC_OP(long, 64, 8)
 #endif
 
-ATOMIC_OPS(        , .aqrl)
-ATOMIC_OPS(_acquire,   .aq)
-ATOMIC_OPS(_release,   .rl)
-ATOMIC_OPS(_relaxed,      )
+ATOMIC_OPS()
 
 #undef ATOMIC_OPS
 #undef ATOMIC_OP
@@ -340,13 +459,13 @@ static __always_inline int atomic_sub_if_positive(atomic_t *v, int offset)
       int prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.w.aqrl  %[p],  %[c]\n\t"
-		"sub       %[rc],  %[p], %[o]\n\t"
-		"bltz      %[rc],    1f\n\t"
-		"sc.w.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc],    0b\n\t"
-		"1:"
+		"0:	lr.w     %[p],  %[c]\n"
+		"	sub      %[rc], %[p], %[o]\n"
+		"	bltz     %[rc], 1f\n"
+		"	sc.w.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [o]"r" (offset)
 		: "memory");
@@ -361,13 +480,13 @@ static __always_inline long atomic64_sub_if_positive(atomic64_t *v, int offset)
       long prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.d.aqrl  %[p],  %[c]\n\t"
-		"sub       %[rc],  %[p], %[o]\n\t"
-		"bltz      %[rc],    1f\n\t"
-		"sc.d.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc],    0b\n\t"
-		"1:"
+		"0:	lr.d     %[p],  %[c]\n"
+		"	sub      %[rc], %[p], %[o]\n"
+		"	bltz     %[rc], 1f\n"
+		"	sc.d.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [o]"r" (offset)
 		: "memory");
arch/riscv/include/asm/cmpxchg.h +320 -71
index db249dbc7b976..c12833f7b6bd1 100644
@@ -17,45 +17,153 @@
 #include <linux/bug.h>
 
 #include <asm/barrier.h>
+#include <asm/fence.h>
 
-#define __xchg(new, ptr, size, asm_or)				\
-({								\
-	__typeof__(ptr) __ptr = (ptr);				\
-	__typeof__(new) __new = (new);				\
-	__typeof__(*(ptr)) __ret;				\
-	switch (size) {						\
-	case 4:							\
-		__asm__ __volatile__ (				\
-			"amoswap.w" #asm_or " %0, %2, %1"	\
-			: "=r" (__ret), "+A" (*__ptr)		\
-			: "r" (__new)				\
-			: "memory");				\
-		break;						\
-	case 8:							\
-		__asm__ __volatile__ (				\
-			"amoswap.d" #asm_or " %0, %2, %1"	\
-			: "=r" (__ret), "+A" (*__ptr)		\
-			: "r" (__new)				\
-			: "memory");				\
-		break;						\
-	default:						\
-		BUILD_BUG();					\
-	}							\
-	__ret;							\
-})
-
-#define xchg(ptr, x)    (__xchg((x), (ptr), sizeof(*(ptr)), .aqrl))
-
-#define xchg32(ptr, x)				\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 4);	\
-	xchg((ptr), (x));			\
-})
-
-#define xchg64(ptr, x)				\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 8);	\
-	xchg((ptr), (x));			\
+#define __xchg_relaxed(ptr, new, size)					\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(new) __new = (new);					\
+	__typeof__(*(ptr)) __ret;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.w %0, %2, %1\n"			\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.d %0, %2, %1\n"			\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define xchg_relaxed(ptr, x)						\
+({									\
+	__typeof__(*(ptr)) _x_ = (x);					\
+	(__typeof__(*(ptr))) __xchg_relaxed((ptr),			\
+					    _x_, sizeof(*(ptr)));	\
+})
+
+#define __xchg_acquire(ptr, new, size)					\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(new) __new = (new);					\
+	__typeof__(*(ptr)) __ret;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.w %0, %2, %1\n"			\
+			RISCV_ACQUIRE_BARRIER				\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.d %0, %2, %1\n"			\
+			RISCV_ACQUIRE_BARRIER				\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define xchg_acquire(ptr, x)						\
+({									\
+	__typeof__(*(ptr)) _x_ = (x);					\
+	(__typeof__(*(ptr))) __xchg_acquire((ptr),			\
+					    _x_, sizeof(*(ptr)));	\
+})
+
+#define __xchg_release(ptr, new, size)					\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(new) __new = (new);					\
+	__typeof__(*(ptr)) __ret;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			RISCV_RELEASE_BARRIER				\
+			"	amoswap.w %0, %2, %1\n"			\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			RISCV_RELEASE_BARRIER				\
+			"	amoswap.d %0, %2, %1\n"			\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define xchg_release(ptr, x)						\
+({									\
+	__typeof__(*(ptr)) _x_ = (x);					\
+	(__typeof__(*(ptr))) __xchg_release((ptr),			\
+					    _x_, sizeof(*(ptr)));	\
+})
+
+#define __xchg(ptr, new, size)						\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(new) __new = (new);					\
+	__typeof__(*(ptr)) __ret;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.w.aqrl %0, %2, %1\n"		\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.d.aqrl %0, %2, %1\n"		\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define xchg(ptr, x)							\
+({									\
+	__typeof__(*(ptr)) _x_ = (x);					\
+	(__typeof__(*(ptr))) __xchg((ptr), _x_, sizeof(*(ptr)));	\
+})
+
+#define xchg32(ptr, x)							\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 4);				\
+	xchg((ptr), (x));						\
+})
+
+#define xchg64(ptr, x)							\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 8);				\
+	xchg((ptr), (x));						\
 })
 
 /*
@@ -63,7 +171,51 @@
 * store NEW in MEM.  Return the initial value in MEM.  Success is
 * indicated by comparing RETURN with OLD.
 */
-#define __cmpxchg(ptr, old, new, size, lrb, scb)			\
+#define __cmpxchg_relaxed(ptr, old, new, size)				\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(*(ptr)) __old = (old);				\
+	__typeof__(*(ptr)) __new = (new);				\
+	__typeof__(*(ptr)) __ret;					\
+	register unsigned int __rc;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"0:	lr.w %0, %2\n"				\
+			"	bne  %0, %z3, 1f\n"			\
+			"	sc.w %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"0:	lr.d %0, %2\n"				\
+			"	bne %0, %z3, 1f\n"			\
+			"	sc.d %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define cmpxchg_relaxed(ptr, o, n)					\
+({									\
+	__typeof__(*(ptr)) _o_ = (o);					\
+	__typeof__(*(ptr)) _n_ = (n);					\
+	(__typeof__(*(ptr))) __cmpxchg_relaxed((ptr),			\
+					_o_, _n_, sizeof(*(ptr)));	\
+})
+
+#define __cmpxchg_acquire(ptr, old, new, size)				\
 ({									\
 	__typeof__(ptr) __ptr = (ptr);					\
 	__typeof__(*(ptr)) __old = (old);				\
@@ -73,24 +225,24 @@
 	switch (size) {							\
 	case 4:								\
 		__asm__ __volatile__ (					\
-		"0:"							\
-			"lr.w" #scb " %0, %2\n"				\
-			"bne         %0, %z3, 1f\n"			\
-			"sc.w" #lrb " %1, %z4, %2\n"			\
-			"bnez        %1, 0b\n"				\
-		"1:"							\
+			"0:	lr.w %0, %2\n"				\
+			"	bne  %0, %z3, 1f\n"			\
+			"	sc.w %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			RISCV_ACQUIRE_BARRIER				\
+			"1:\n"						\
 			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
 			: "rJ" (__old), "rJ" (__new)			\
 			: "memory");					\
 		break;							\
 	case 8:								\
 		__asm__ __volatile__ (					\
-		"0:"							\
-			"lr.d" #scb " %0, %2\n"				\
-			"bne         %0, %z3, 1f\n"			\
-			"sc.d" #lrb " %1, %z4, %2\n"			\
-			"bnez        %1, 0b\n"				\
-		"1:"							\
+			"0:	lr.d %0, %2\n"				\
+			"	bne %0, %z3, 1f\n"			\
+			"	sc.d %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			RISCV_ACQUIRE_BARRIER				\
+			"1:\n"						\
 			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
 			: "rJ" (__old), "rJ" (__new)			\
 			: "memory");					\
@@ -101,34 +253,131 @@
 	__ret;								\
 })
 
-#define cmpxchg(ptr, o, n) \
-	(__cmpxchg((ptr), (o), (n), sizeof(*(ptr)), .aqrl, .aqrl))
+#define cmpxchg_acquire(ptr, o, n)					\
+({									\
+	__typeof__(*(ptr)) _o_ = (o);					\
+	__typeof__(*(ptr)) _n_ = (n);					\
+	(__typeof__(*(ptr))) __cmpxchg_acquire((ptr),			\
+					_o_, _n_, sizeof(*(ptr)));	\
+})
 
-#define cmpxchg_local(ptr, o, n) \
-	(__cmpxchg((ptr), (o), (n), sizeof(*(ptr)), , ))
+#define __cmpxchg_release(ptr, old, new, size)				\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(*(ptr)) __old = (old);				\
+	__typeof__(*(ptr)) __new = (new);				\
+	__typeof__(*(ptr)) __ret;					\
+	register unsigned int __rc;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			RISCV_RELEASE_BARRIER				\
+			"0:	lr.w %0, %2\n"				\
+			"	bne  %0, %z3, 1f\n"			\
+			"	sc.w %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			RISCV_RELEASE_BARRIER				\
+			"0:	lr.d %0, %2\n"				\
+			"	bne %0, %z3, 1f\n"			\
+			"	sc.d %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define cmpxchg_release(ptr, o, n)					\
+({									\
+	__typeof__(*(ptr)) _o_ = (o);					\
+	__typeof__(*(ptr)) _n_ = (n);					\
+	(__typeof__(*(ptr))) __cmpxchg_release((ptr),			\
+					_o_, _n_, sizeof(*(ptr)));	\
+})
+
+#define __cmpxchg(ptr, old, new, size)					\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(*(ptr)) __old = (old);				\
+	__typeof__(*(ptr)) __new = (new);				\
+	__typeof__(*(ptr)) __ret;					\
+	register unsigned int __rc;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"0:	lr.w %0, %2\n"				\
+			"	bne  %0, %z3, 1f\n"			\
+			"	sc.w.rl %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"	fence rw, rw\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"0:	lr.d %0, %2\n"				\
+			"	bne %0, %z3, 1f\n"			\
+			"	sc.d.rl %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"	fence rw, rw\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
 
-#define cmpxchg32(ptr, o, n)			\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 4);	\
-	cmpxchg((ptr), (o), (n));		\
+#define cmpxchg(ptr, o, n)						\
+({									\
+	__typeof__(*(ptr)) _o_ = (o);					\
+	__typeof__(*(ptr)) _n_ = (n);					\
+	(__typeof__(*(ptr))) __cmpxchg((ptr),				\
+				       _o_, _n_, sizeof(*(ptr)));	\
 })
 
-#define cmpxchg32_local(ptr, o, n)		\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 4);	\
-	cmpxchg_local((ptr), (o), (n));		\
+#define cmpxchg_local(ptr, o, n)					\
+	(__cmpxchg_relaxed((ptr), (o), (n), sizeof(*(ptr))))
+
+#define cmpxchg32(ptr, o, n)						\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 4);				\
+	cmpxchg((ptr), (o), (n));					\
 })
 
-#define cmpxchg64(ptr, o, n)			\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 8);	\
-	cmpxchg((ptr), (o), (n));		\
+#define cmpxchg32_local(ptr, o, n)					\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 4);				\
+	cmpxchg_relaxed((ptr), (o), (n))				\
 })
 
-#define cmpxchg64_local(ptr, o, n)		\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 8);	\
-	cmpxchg_local((ptr), (o), (n));		\
+#define cmpxchg64(ptr, o, n)						\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 8);				\
+	cmpxchg((ptr), (o), (n));					\
+})
+
+#define cmpxchg64_local(ptr, o, n)					\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 8);				\
+	cmpxchg_relaxed((ptr), (o), (n));				\
 })
 
 #endif /* _ASM_RISCV_CMPXCHG_H */
#5 lore raw
Quoted 144 lines · parri.andrea at gmail.com
Quoted 134 lines · Palmer Dabbelt
Quoted 133 lines · parri.andrea at gmail.com

Atomics present the same issue with locking: release and acquire variants need to be strengthened to meet the constraints defined by the Linux-kernel memory consistency model [1].

Atomics present a further issue: implementations of atomics such as atomic_cmpxchg() and atomic_add_unless() rely on LR/SC pairs, which do not give full-ordering with .aqrl; for example, current implementations allow the "lr-sc-aqrl-pair-vs-full-barrier" test below to end up with the state indicated in the "exists" clause.

In order to "synchronize" LKMM and RISC-V's implementation, this commit strengthens the implementations of the atomics operations by replacing .rl and .aq with the use of ("lightweigth") fences, and by replacing .aqrl LR/SC pairs in sequences such as:

0: lr.w.aqrl %0, %addr

         bne        %0, %old, 1f
         ...
         sc.w.aqrl  %1, %new, %addr
         bnez       %1, 0b

1:

with sequences of the form:

0: lr.w %0, %addr

         bne        %0, %old, 1f
         ...
         sc.w.rl    %1, %new, %addr   /* SC-release   */
         bnez       %1, 0b
         fence      rw, rw            /* "full" fence */

1:

following Daniel's suggestion.

These modifications were validated with simulation of the RISC-V memory consistency model.

C lr-sc-aqrl-pair-vs-full-barrier

{}

P0(int *x, int *y, atomic_t *u)

{
	int r0;
	int r1;

	WRITE_ONCE(*x, 1);
	r0 = atomic_cmpxchg(u, 0, 1);
	r1 = READ_ONCE(*y);
}

P1(int *x, int *y, atomic_t *v)

{
	int r0;
	int r1;

	WRITE_ONCE(*y, 1);
	r0 = atomic_cmpxchg(v, 0, 1);
	r1 = READ_ONCE(*x);
}

exists (u=1 /\ v=1 /\ 0:r1=0 /\ 1:r1=0)

[1] https://marc.info/?l=linux-kernel&m=151930201102853&w=2
https://groups.google.com/a/groups.riscv.org/forum/#!topic/isa-dev/hKywNHBkAXM
https://marc.info/?l=linux-kernel&m=151633436614259&w=2

Cc Palmer Dabbelt <[email protected]>
Cc Albert Ou <[email protected]>
Cc Daniel Lustig <[email protected]>
Cc Alan Stern <[email protected]>
Cc Will Deacon <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Boqun Feng <[email protected]>
Cc Nicholas Piggin <[email protected]>
Cc David Howells <[email protected]>
Cc Jade Alglave <[email protected]>
Cc Luc Maranget <[email protected]>
Cc "Paul E. McKenney" <[email protected]>
Cc Akira Yokosawa <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc linux-riscv at lists.infradead.org
Cc linux-kernel at vger.kernel.org

---
arch/riscv/include/asm/atomic.h | 417 +++++++++++++++++++++++++-------------- arch/riscv/include/asm/cmpxchg.h | 391 +++++++++++++++++++++++++++++-------

2 files changed, 588 insertions(+), 220 deletions(-)
arch/riscv/include/asm/atomic.h +18 -9
index e65d1cd89e28b..855115ace98c8 100644
@@ -24,6 +24,20 @@
 #include <asm/barrier.h>
 
 #define ATOMIC_INIT(i)	{ (i) }
+
+#define __atomic_op_acquire(op, args...)				\
+({									\
+	typeof(op##_relaxed(args)) __ret  = op##_relaxed(args);		\
+	__asm__ __volatile__(RISCV_ACQUIRE_BARRIER "" ::: "memory");	\
+	__ret;								\
+})
+
+#define __atomic_op_release(op, args...)				\
+({									\
+	__asm__ __volatile__(RISCV_RELEASE_BARRIER "" ::: "memory");	\
+	op##_relaxed(args);						\
+})
+
 static __always_inline int atomic_read(const atomic_t *v)
 {
 	return READ_ONCE(v->counter);
@@ -50,22 +64,23 @@ static __always_inline void atomic64_set(atomic64_t *v, long i)
 * have the AQ or RL bits set.  These don't return anything, so there's only
 * one version to worry about.
 */
-#define ATOMIC_OP(op, asm_op, I, asm_type, c_type, prefix)				\
-static __always_inline void atomic##prefix##_##op(c_type i, atomic##prefix##_t *v)	\
-{											\
-	__asm__ __volatile__ (								\
-		"amo" #asm_op "." #asm_type " zero, %1, %0"				\
-		: "+A" (v->counter)							\
-		: "r" (I)								\
-		: "memory");								\
-}
+#define ATOMIC_OP(op, asm_op, I, asm_type, c_type, prefix)		\
+static __always_inline							\
+void atomic##prefix##_##op(c_type i, atomic##prefix##_t *v)		\
+{									\
Quoted 8 lines

arch/riscv/include/asm/atomic.h static __always_inline void atomic64_set(atomic64_t *v, long i) · line 76

Quoted 6 lines
arch/riscv/include/asm/atomic.h +6 -0
+	__asm__ __volatile__ (						\
+		"	amo" #asm_op "." #asm_type " zero, %1, %0"	\
+		: "+A" (v->counter)					\
+		: "r" (I)						\
+		: "memory");						\
+}									\

Just to be sure: this is just a whitespace change, right?

Quoted 3 lines

This belongs to the "few style fixes" (in the specific, 80-chars lines) mentioned in the cover letter; I could not resist ;-), but I'll remove them in v3 if you like so.

No problem, just next time it's a bit easier to not mix the really complicated stuff (memory model changes) with the really simple stuff (whitespace changes).

Quoted 108 lines
Quoted 94 lines
Quoted 94 lines
arch/riscv/include/asm/atomic.h +46 -26
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, I)			\
+#define ATOMIC_OPS(op, asm_op, I)					\
        ATOMIC_OP (op, asm_op, I, w,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, I)			\
-        ATOMIC_OP (op, asm_op, I, w,  int,   )	\
+#define ATOMIC_OPS(op, asm_op, I)					\
+        ATOMIC_OP (op, asm_op, I, w,  int,   )				\
        ATOMIC_OP (op, asm_op, I, d, long, 64)
 #endif
 
@@ -79,75 +94,115 @@ ATOMIC_OPS(xor, xor,  i)
 #undef ATOMIC_OPS
 
 /*
- * Atomic ops that have ordered, relaxed, acquire, and relese variants.
+ * Atomic ops that have ordered, relaxed, acquire, and release variants.
 * There's two flavors of these: the arithmatic ops have both fetch and return
 * versions, while the logical ops only have fetch versions.
 */
-#define ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, asm_type, c_type, prefix)				\
-static __always_inline c_type atomic##prefix##_fetch_##op##c_or(c_type i, atomic##prefix##_t *v)	\
-{													\
-	register c_type ret;										\
-	__asm__ __volatile__ (										\
-		"amo" #asm_op "." #asm_type #asm_or " %1, %2, %0"					\
-		: "+A" (v->counter), "=r" (ret)								\
-		: "r" (I)										\
-		: "memory");										\
-	return ret;											\
+#define ATOMIC_FETCH_OP(op, asm_op, I, asm_type, c_type, prefix)	\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op##_relaxed(c_type i,			\
+					     atomic##prefix##_t *v)	\
+{									\
+	register c_type ret;						\
+	__asm__ __volatile__ (						\
+		"	amo" #asm_op "." #asm_type " %1, %2, %0"	\
+		: "+A" (v->counter), "=r" (ret)				\
+		: "r" (I)						\
+		: "memory");						\
+	return ret;							\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op(c_type i, atomic##prefix##_t *v)	\
+{									\
+	register c_type ret;						\
+	__asm__ __volatile__ (						\
+		"	amo" #asm_op "." #asm_type ".aqrl  %1, %2, %0"	\
+		: "+A" (v->counter), "=r" (ret)				\
+		: "r" (I)						\
+		: "memory");						\
+	return ret;							\
 }
 
-#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, asm_type, c_type, prefix)			\
-static __always_inline c_type atomic##prefix##_##op##_return##c_or(c_type i, atomic##prefix##_t *v)	\
-{													\
-        return atomic##prefix##_fetch_##op##c_or(i, v) c_op I;						\
+#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_type, c_type, prefix)	\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return_relaxed(c_type i,			\
+					      atomic##prefix##_t *v)	\
+{									\
+        return atomic##prefix##_fetch_##op##_relaxed(i, v) c_op I;	\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return(c_type i, atomic##prefix##_t *v)	\
+{									\
+        return atomic##prefix##_fetch_##op(i, v) c_op I;		\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, c_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, w,  int,   )
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_FETCH_OP( op, asm_op,       I, w,  int,   )		\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I, w,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, c_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_FETCH_OP (op, asm_op,       I, asm_or, c_or, d, long, 64)	\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I, asm_or, c_or, d, long, 64)
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_FETCH_OP( op, asm_op,       I, w,  int,   )		\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I, w,  int,   )		\
+        ATOMIC_FETCH_OP( op, asm_op,       I, d, long, 64)		\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I, d, long, 64)
 #endif
 
Quoted 9 lines

arch/riscv/include/asm/atomic.h ATOMIC_OPS(xor, xor, i) · line 152

Quoted 6 lines
arch/riscv/include/asm/atomic.h +2 -4
-ATOMIC_OPS(add, add, +,  i,      , _relaxed)
-ATOMIC_OPS(add, add, +,  i, .aq  , _acquire)
-ATOMIC_OPS(add, add, +,  i, .rl  , _release)
-ATOMIC_OPS(add, add, +,  i, .aqrl,         )
+ATOMIC_OPS(add, add, +,  i)
+ATOMIC_OPS(sub, add, +, -i)

Sorry, I must be missing something, but how do the acquire and release atomics get generated with this new code?

This proposal relies on the generic definition,

include/linux/atomic.h ,

Quoted 7 lines

and on the

   __atomic_op_acquire()
   __atomic_op_release()

above to build the acquire/release atomics (except for the xchg,cmpxchg, where the ACQUIRE_BARRIER is inserted conditionally/on success).

I thought we wanted to use the AQ and RL bits for AMOs, just not for LR/SC sequences. IIRC the AMOs are safe with the current memory model, but I might just have some version mismatches in my head.

Quoted 844 lines
Quoted 844 lines
Quoted 844 lines
arch/riscv/include/asm/atomic.h +196 -110
+
+#define atomic_add_return_relaxed	atomic_add_return_relaxed
+#define atomic_sub_return_relaxed	atomic_sub_return_relaxed
+#define atomic_add_return		atomic_add_return
+#define atomic_sub_return		atomic_sub_return
 
-ATOMIC_OPS(sub, add, +, -i,      , _relaxed)
-ATOMIC_OPS(sub, add, +, -i, .aq  , _acquire)
-ATOMIC_OPS(sub, add, +, -i, .rl  , _release)
-ATOMIC_OPS(sub, add, +, -i, .aqrl,         )
+#define atomic_fetch_add_relaxed	atomic_fetch_add_relaxed
+#define atomic_fetch_sub_relaxed	atomic_fetch_sub_relaxed
+#define atomic_fetch_add		atomic_fetch_add
+#define atomic_fetch_sub		atomic_fetch_sub
+
+#ifndef CONFIG_GENERIC_ATOMIC64
+#define atomic64_add_return_relaxed	atomic64_add_return_relaxed
+#define atomic64_sub_return_relaxed	atomic64_sub_return_relaxed
+#define atomic64_add_return		atomic64_add_return
+#define atomic64_sub_return		atomic64_sub_return
+
+#define atomic64_fetch_add_relaxed	atomic64_fetch_add_relaxed
+#define atomic64_fetch_sub_relaxed	atomic64_fetch_sub_relaxed
+#define atomic64_fetch_add		atomic64_fetch_add
+#define atomic64_fetch_sub		atomic64_fetch_sub
+#endif
 
 #undef ATOMIC_OPS
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, w,  int,   )
+#define ATOMIC_OPS(op, asm_op, I)					\
+        ATOMIC_FETCH_OP(op, asm_op, I, w,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, I, asm_or, c_or)				\
-        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, w,  int,   )	\
-        ATOMIC_FETCH_OP(op, asm_op, I, asm_or, c_or, d, long, 64)
+#define ATOMIC_OPS(op, asm_op, I)					\
+        ATOMIC_FETCH_OP(op, asm_op, I, w,  int,   )			\
+        ATOMIC_FETCH_OP(op, asm_op, I, d, long, 64)
 #endif
 
-ATOMIC_OPS(and, and, i,      , _relaxed)
-ATOMIC_OPS(and, and, i, .aq  , _acquire)
-ATOMIC_OPS(and, and, i, .rl  , _release)
-ATOMIC_OPS(and, and, i, .aqrl,         )
+ATOMIC_OPS(and, and, i)
+ATOMIC_OPS( or,  or, i)
+ATOMIC_OPS(xor, xor, i)
 
-ATOMIC_OPS( or,  or, i,      , _relaxed)
-ATOMIC_OPS( or,  or, i, .aq  , _acquire)
-ATOMIC_OPS( or,  or, i, .rl  , _release)
-ATOMIC_OPS( or,  or, i, .aqrl,         )
+#define atomic_fetch_and_relaxed	atomic_fetch_and_relaxed
+#define atomic_fetch_or_relaxed		atomic_fetch_or_relaxed
+#define atomic_fetch_xor_relaxed	atomic_fetch_xor_relaxed
+#define atomic_fetch_and		atomic_fetch_and
+#define atomic_fetch_or			atomic_fetch_or
+#define atomic_fetch_xor		atomic_fetch_xor
 
-ATOMIC_OPS(xor, xor, i,      , _relaxed)
-ATOMIC_OPS(xor, xor, i, .aq  , _acquire)
-ATOMIC_OPS(xor, xor, i, .rl  , _release)
-ATOMIC_OPS(xor, xor, i, .aqrl,         )
+#ifndef CONFIG_GENERIC_ATOMIC64
+#define atomic64_fetch_and_relaxed	atomic64_fetch_and_relaxed
+#define atomic64_fetch_or_relaxed	atomic64_fetch_or_relaxed
+#define atomic64_fetch_xor_relaxed	atomic64_fetch_xor_relaxed
+#define atomic64_fetch_and		atomic64_fetch_and
+#define atomic64_fetch_or		atomic64_fetch_or
+#define atomic64_fetch_xor		atomic64_fetch_xor
+#endif
 
 #undef ATOMIC_OPS
 
@@ -157,22 +212,24 @@ ATOMIC_OPS(xor, xor, i, .aqrl,         )
 /*
 * The extra atomic operations that are constructed from one of the core
 * AMO-based operations above (aside from sub, which is easier to fit above).
- * These are required to perform a barrier, but they're OK this way because
- * atomic_*_return is also required to perform a barrier.
+ * These are required to perform a full barrier, but they're OK this way
+ * because atomic_*_return is also required to perform a full barrier.
+ *
 */
-#define ATOMIC_OP(op, func_op, comp_op, I, c_type, prefix)			\
-static __always_inline bool atomic##prefix##_##op(c_type i, atomic##prefix##_t *v) \
-{										\
-	return atomic##prefix##_##func_op##_return(i, v) comp_op I;		\
+#define ATOMIC_OP(op, func_op, comp_op, I, c_type, prefix)		\
+static __always_inline							\
+bool atomic##prefix##_##op(c_type i, atomic##prefix##_t *v)		\
+{									\
+	return atomic##prefix##_##func_op##_return(i, v) comp_op I;	\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, func_op, comp_op, I)			\
-        ATOMIC_OP (op, func_op, comp_op, I,  int,   )
+#define ATOMIC_OPS(op, func_op, comp_op, I)				\
+        ATOMIC_OP(op, func_op, comp_op, I,  int,   )
 #else
-#define ATOMIC_OPS(op, func_op, comp_op, I)			\
-        ATOMIC_OP (op, func_op, comp_op, I,  int,   )		\
-        ATOMIC_OP (op, func_op, comp_op, I, long, 64)
+#define ATOMIC_OPS(op, func_op, comp_op, I)				\
+        ATOMIC_OP(op, func_op, comp_op, I,  int,   )			\
+        ATOMIC_OP(op, func_op, comp_op, I, long, 64)
 #endif
 
 ATOMIC_OPS(add_and_test, add, ==, 0)
@@ -182,51 +239,87 @@ ATOMIC_OPS(add_negative, add,  <, 0)
 #undef ATOMIC_OP
 #undef ATOMIC_OPS
 
-#define ATOMIC_OP(op, func_op, I, c_type, prefix)				\
-static __always_inline void atomic##prefix##_##op(atomic##prefix##_t *v)	\
-{										\
-	atomic##prefix##_##func_op(I, v);					\
+#define ATOMIC_OP(op, func_op, I, c_type, prefix)			\
+static __always_inline							\
+void atomic##prefix##_##op(atomic##prefix##_t *v)			\
+{									\
+	atomic##prefix##_##func_op(I, v);				\
 }
 
-#define ATOMIC_FETCH_OP(op, func_op, I, c_type, prefix)					\
-static __always_inline c_type atomic##prefix##_fetch_##op(atomic##prefix##_t *v)	\
-{											\
-	return atomic##prefix##_fetch_##func_op(I, v);					\
+#define ATOMIC_FETCH_OP(op, func_op, I, c_type, prefix)			\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op##_relaxed(atomic##prefix##_t *v)	\
+{									\
+	return atomic##prefix##_fetch_##func_op##_relaxed(I, v);	\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_fetch_##op(atomic##prefix##_t *v)		\
+{									\
+	return atomic##prefix##_fetch_##func_op(I, v);			\
 }
 
-#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, c_type, prefix)				\
-static __always_inline c_type atomic##prefix##_##op##_return(atomic##prefix##_t *v)	\
-{											\
-        return atomic##prefix##_fetch_##op(v) c_op I;					\
+#define ATOMIC_OP_RETURN(op, asm_op, c_op, I, c_type, prefix)		\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return_relaxed(atomic##prefix##_t *v)	\
+{									\
+        return atomic##prefix##_fetch_##op##_relaxed(v) c_op I;		\
+}									\
+static __always_inline							\
+c_type atomic##prefix##_##op##_return(atomic##prefix##_t *v)		\
+{									\
+        return atomic##prefix##_fetch_##op(v) c_op I;			\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(op, asm_op, c_op, I)						\
-        ATOMIC_OP       (op, asm_op,       I,  int,   )				\
-        ATOMIC_FETCH_OP (op, asm_op,       I,  int,   )				\
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_OP(       op, asm_op,       I,  int,   )			\
+        ATOMIC_FETCH_OP( op, asm_op,       I,  int,   )			\
        ATOMIC_OP_RETURN(op, asm_op, c_op, I,  int,   )
 #else
-#define ATOMIC_OPS(op, asm_op, c_op, I)						\
-        ATOMIC_OP       (op, asm_op,       I,  int,   )				\
-        ATOMIC_FETCH_OP (op, asm_op,       I,  int,   )				\
-        ATOMIC_OP_RETURN(op, asm_op, c_op, I,  int,   )				\
-        ATOMIC_OP       (op, asm_op,       I, long, 64)				\
-        ATOMIC_FETCH_OP (op, asm_op,       I, long, 64)				\
+#define ATOMIC_OPS(op, asm_op, c_op, I)					\
+        ATOMIC_OP(       op, asm_op,       I,  int,   )			\
+        ATOMIC_FETCH_OP( op, asm_op,       I,  int,   )			\
+        ATOMIC_OP_RETURN(op, asm_op, c_op, I,  int,   )			\
+        ATOMIC_OP(       op, asm_op,       I, long, 64)			\
+        ATOMIC_FETCH_OP( op, asm_op,       I, long, 64)			\
        ATOMIC_OP_RETURN(op, asm_op, c_op, I, long, 64)
 #endif
 
 ATOMIC_OPS(inc, add, +,  1)
 ATOMIC_OPS(dec, add, +, -1)
 
+#define atomic_inc_return_relaxed	atomic_inc_return_relaxed
+#define atomic_dec_return_relaxed	atomic_dec_return_relaxed
+#define atomic_inc_return		atomic_inc_return
+#define atomic_dec_return		atomic_dec_return
+
+#define atomic_fetch_inc_relaxed	atomic_fetch_inc_relaxed
+#define atomic_fetch_dec_relaxed	atomic_fetch_dec_relaxed
+#define atomic_fetch_inc		atomic_fetch_inc
+#define atomic_fetch_dec		atomic_fetch_dec
+
+#ifndef CONFIG_GENERIC_ATOMIC64
+#define atomic64_inc_return_relaxed	atomic64_inc_return_relaxed
+#define atomic64_dec_return_relaxed	atomic64_dec_return_relaxed
+#define atomic64_inc_return		atomic64_inc_return
+#define atomic64_dec_return		atomic64_dec_return
+
+#define atomic64_fetch_inc_relaxed	atomic64_fetch_inc_relaxed
+#define atomic64_fetch_dec_relaxed	atomic64_fetch_dec_relaxed
+#define atomic64_fetch_inc		atomic64_fetch_inc
+#define atomic64_fetch_dec		atomic64_fetch_dec
+#endif
+
 #undef ATOMIC_OPS
 #undef ATOMIC_OP
 #undef ATOMIC_FETCH_OP
 #undef ATOMIC_OP_RETURN
 
-#define ATOMIC_OP(op, func_op, comp_op, I, prefix)				\
-static __always_inline bool atomic##prefix##_##op(atomic##prefix##_t *v)	\
-{										\
-	return atomic##prefix##_##func_op##_return(v) comp_op I;		\
+#define ATOMIC_OP(op, func_op, comp_op, I, prefix)			\
+static __always_inline							\
+bool atomic##prefix##_##op(atomic##prefix##_t *v)			\
+{									\
+	return atomic##prefix##_##func_op##_return(v) comp_op I;	\
 }
 
 ATOMIC_OP(inc_and_test, inc, ==, 0,   )
@@ -238,19 +331,19 @@ ATOMIC_OP(dec_and_test, dec, ==, 0, 64)
 
 #undef ATOMIC_OP
 
-/* This is required to provide a barrier on success. */
+/* This is required to provide a full barrier on success. */
 static __always_inline int __atomic_add_unless(atomic_t *v, int a, int u)
 {
       int prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.w.aqrl  %[p],  %[c]\n\t"
-		"beq        %[p],  %[u], 1f\n\t"
-		"add       %[rc],  %[p], %[a]\n\t"
-		"sc.w.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc], 0b\n\t"
-		"1:"
+		"0:	lr.w     %[p],  %[c]\n"
+		"	beq      %[p],  %[u], 1f\n"
+		"	add      %[rc], %[p], %[a]\n"
+		"	sc.w.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [a]"r" (a), [u]"r" (u)
 		: "memory");
@@ -263,13 +356,13 @@ static __always_inline long __atomic64_add_unless(atomic64_t *v, long a, long u)
       long prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.d.aqrl  %[p],  %[c]\n\t"
-		"beq        %[p],  %[u], 1f\n\t"
-		"add       %[rc],  %[p], %[a]\n\t"
-		"sc.d.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc], 0b\n\t"
-		"1:"
+		"0:	lr.d     %[p],  %[c]\n"
+		"	beq      %[p],  %[u], 1f\n"
+		"	add      %[rc], %[p], %[a]\n"
+		"	sc.d.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [a]"r" (a), [u]"r" (u)
 		: "memory");
@@ -300,37 +393,63 @@ static __always_inline long atomic64_inc_not_zero(atomic64_t *v)
 
 /*
 * atomic_{cmp,}xchg is required to have exactly the same ordering semantics as
- * {cmp,}xchg and the operations that return, so they need a barrier.
- */
-/*
- * FIXME: atomic_cmpxchg_{acquire,release,relaxed} are all implemented by
- * assigning the same barrier to both the LR and SC operations, but that might
- * not make any sense.  We're waiting on a memory model specification to
- * determine exactly what the right thing to do is here.
+ * {cmp,}xchg and the operations that return, so they need a full barrier.
 */
-#define ATOMIC_OP(c_t, prefix, c_or, size, asm_or)						\
-static __always_inline c_t atomic##prefix##_cmpxchg##c_or(atomic##prefix##_t *v, c_t o, c_t n) 	\
-{												\
-	return __cmpxchg(&(v->counter), o, n, size, asm_or, asm_or);				\
-}												\
-static __always_inline c_t atomic##prefix##_xchg##c_or(atomic##prefix##_t *v, c_t n) 		\
-{												\
-	return __xchg(n, &(v->counter), size, asm_or);						\
+#define ATOMIC_OP(c_t, prefix, size)					\
+static __always_inline							\
+c_t atomic##prefix##_xchg_relaxed(atomic##prefix##_t *v, c_t n)		\
+{									\
+	return __xchg_relaxed(&(v->counter), n, size);			\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_xchg_acquire(atomic##prefix##_t *v, c_t n)		\
+{									\
+	return __xchg_acquire(&(v->counter), n, size);			\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_xchg_release(atomic##prefix##_t *v, c_t n)		\
+{									\
+	return __xchg_release(&(v->counter), n, size);			\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_xchg(atomic##prefix##_t *v, c_t n)			\
+{									\
+	return __xchg(&(v->counter), n, size);				\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg_relaxed(atomic##prefix##_t *v,		\
+				     c_t o, c_t n)			\
+{									\
+	return __cmpxchg_relaxed(&(v->counter), o, n, size);		\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg_acquire(atomic##prefix##_t *v,		\
+				     c_t o, c_t n)			\
+{									\
+	return __cmpxchg_acquire(&(v->counter), o, n, size);		\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg_release(atomic##prefix##_t *v,		\
+				     c_t o, c_t n)			\
+{									\
+	return __cmpxchg_release(&(v->counter), o, n, size);		\
+}									\
+static __always_inline							\
+c_t atomic##prefix##_cmpxchg(atomic##prefix##_t *v, c_t o, c_t n)	\
+{									\
+	return __cmpxchg(&(v->counter), o, n, size);			\
 }
 
 #ifdef CONFIG_GENERIC_ATOMIC64
-#define ATOMIC_OPS(c_or, asm_or)			\
-	ATOMIC_OP( int,   , c_or, 4, asm_or)
+#define ATOMIC_OPS()							\
+	ATOMIC_OP( int,   , 4)
 #else
-#define ATOMIC_OPS(c_or, asm_or)			\
-	ATOMIC_OP( int,   , c_or, 4, asm_or)		\
-	ATOMIC_OP(long, 64, c_or, 8, asm_or)
+#define ATOMIC_OPS()							\
+	ATOMIC_OP( int,   , 4)						\
+	ATOMIC_OP(long, 64, 8)
 #endif
 
-ATOMIC_OPS(        , .aqrl)
-ATOMIC_OPS(_acquire,   .aq)
-ATOMIC_OPS(_release,   .rl)
-ATOMIC_OPS(_relaxed,      )
+ATOMIC_OPS()
 
 #undef ATOMIC_OPS
 #undef ATOMIC_OP
@@ -340,13 +459,13 @@ static __always_inline int atomic_sub_if_positive(atomic_t *v, int offset)
       int prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.w.aqrl  %[p],  %[c]\n\t"
-		"sub       %[rc],  %[p], %[o]\n\t"
-		"bltz      %[rc],    1f\n\t"
-		"sc.w.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc],    0b\n\t"
-		"1:"
+		"0:	lr.w     %[p],  %[c]\n"
+		"	sub      %[rc], %[p], %[o]\n"
+		"	bltz     %[rc], 1f\n"
+		"	sc.w.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [o]"r" (offset)
 		: "memory");
@@ -361,13 +480,13 @@ static __always_inline long atomic64_sub_if_positive(atomic64_t *v, int offset)
       long prev, rc;
 
 	__asm__ __volatile__ (
-		"0:\n\t"
-		"lr.d.aqrl  %[p],  %[c]\n\t"
-		"sub       %[rc],  %[p], %[o]\n\t"
-		"bltz      %[rc],    1f\n\t"
-		"sc.d.aqrl %[rc], %[rc], %[c]\n\t"
-		"bnez      %[rc],    0b\n\t"
-		"1:"
+		"0:	lr.d     %[p],  %[c]\n"
+		"	sub      %[rc], %[p], %[o]\n"
+		"	bltz     %[rc], 1f\n"
+		"	sc.d.rl  %[rc], %[rc], %[c]\n"
+		"	bnez     %[rc], 0b\n"
+		"	fence    rw, rw\n"
+		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [o]"r" (offset)
 		: "memory");
arch/riscv/include/asm/cmpxchg.h +320 -71
index db249dbc7b976..c12833f7b6bd1 100644
@@ -17,45 +17,153 @@
 #include <linux/bug.h>
 
 #include <asm/barrier.h>
+#include <asm/fence.h>
 
-#define __xchg(new, ptr, size, asm_or)				\
-({								\
-	__typeof__(ptr) __ptr = (ptr);				\
-	__typeof__(new) __new = (new);				\
-	__typeof__(*(ptr)) __ret;				\
-	switch (size) {						\
-	case 4:							\
-		__asm__ __volatile__ (				\
-			"amoswap.w" #asm_or " %0, %2, %1"	\
-			: "=r" (__ret), "+A" (*__ptr)		\
-			: "r" (__new)				\
-			: "memory");				\
-		break;						\
-	case 8:							\
-		__asm__ __volatile__ (				\
-			"amoswap.d" #asm_or " %0, %2, %1"	\
-			: "=r" (__ret), "+A" (*__ptr)		\
-			: "r" (__new)				\
-			: "memory");				\
-		break;						\
-	default:						\
-		BUILD_BUG();					\
-	}							\
-	__ret;							\
-})
-
-#define xchg(ptr, x)    (__xchg((x), (ptr), sizeof(*(ptr)), .aqrl))
-
-#define xchg32(ptr, x)				\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 4);	\
-	xchg((ptr), (x));			\
-})
-
-#define xchg64(ptr, x)				\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 8);	\
-	xchg((ptr), (x));			\
+#define __xchg_relaxed(ptr, new, size)					\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(new) __new = (new);					\
+	__typeof__(*(ptr)) __ret;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.w %0, %2, %1\n"			\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.d %0, %2, %1\n"			\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define xchg_relaxed(ptr, x)						\
+({									\
+	__typeof__(*(ptr)) _x_ = (x);					\
+	(__typeof__(*(ptr))) __xchg_relaxed((ptr),			\
+					    _x_, sizeof(*(ptr)));	\
+})
+
+#define __xchg_acquire(ptr, new, size)					\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(new) __new = (new);					\
+	__typeof__(*(ptr)) __ret;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.w %0, %2, %1\n"			\
+			RISCV_ACQUIRE_BARRIER				\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.d %0, %2, %1\n"			\
+			RISCV_ACQUIRE_BARRIER				\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define xchg_acquire(ptr, x)						\
+({									\
+	__typeof__(*(ptr)) _x_ = (x);					\
+	(__typeof__(*(ptr))) __xchg_acquire((ptr),			\
+					    _x_, sizeof(*(ptr)));	\
+})
+
+#define __xchg_release(ptr, new, size)					\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(new) __new = (new);					\
+	__typeof__(*(ptr)) __ret;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			RISCV_RELEASE_BARRIER				\
+			"	amoswap.w %0, %2, %1\n"			\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			RISCV_RELEASE_BARRIER				\
+			"	amoswap.d %0, %2, %1\n"			\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define xchg_release(ptr, x)						\
+({									\
+	__typeof__(*(ptr)) _x_ = (x);					\
+	(__typeof__(*(ptr))) __xchg_release((ptr),			\
+					    _x_, sizeof(*(ptr)));	\
+})
+
+#define __xchg(ptr, new, size)						\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(new) __new = (new);					\
+	__typeof__(*(ptr)) __ret;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.w.aqrl %0, %2, %1\n"		\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"	amoswap.d.aqrl %0, %2, %1\n"		\
+			: "=r" (__ret), "+A" (*__ptr)			\
+			: "r" (__new)					\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define xchg(ptr, x)							\
+({									\
+	__typeof__(*(ptr)) _x_ = (x);					\
+	(__typeof__(*(ptr))) __xchg((ptr), _x_, sizeof(*(ptr)));	\
+})
+
+#define xchg32(ptr, x)							\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 4);				\
+	xchg((ptr), (x));						\
+})
+
+#define xchg64(ptr, x)							\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 8);				\
+	xchg((ptr), (x));						\
 })
 
 /*
@@ -63,7 +171,51 @@
 * store NEW in MEM.  Return the initial value in MEM.  Success is
 * indicated by comparing RETURN with OLD.
 */
-#define __cmpxchg(ptr, old, new, size, lrb, scb)			\
+#define __cmpxchg_relaxed(ptr, old, new, size)				\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(*(ptr)) __old = (old);				\
+	__typeof__(*(ptr)) __new = (new);				\
+	__typeof__(*(ptr)) __ret;					\
+	register unsigned int __rc;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"0:	lr.w %0, %2\n"				\
+			"	bne  %0, %z3, 1f\n"			\
+			"	sc.w %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"0:	lr.d %0, %2\n"				\
+			"	bne %0, %z3, 1f\n"			\
+			"	sc.d %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define cmpxchg_relaxed(ptr, o, n)					\
+({									\
+	__typeof__(*(ptr)) _o_ = (o);					\
+	__typeof__(*(ptr)) _n_ = (n);					\
+	(__typeof__(*(ptr))) __cmpxchg_relaxed((ptr),			\
+					_o_, _n_, sizeof(*(ptr)));	\
+})
+
+#define __cmpxchg_acquire(ptr, old, new, size)				\
 ({									\
 	__typeof__(ptr) __ptr = (ptr);					\
 	__typeof__(*(ptr)) __old = (old);				\
@@ -73,24 +225,24 @@
 	switch (size) {							\
 	case 4:								\
 		__asm__ __volatile__ (					\
-		"0:"							\
-			"lr.w" #scb " %0, %2\n"				\
-			"bne         %0, %z3, 1f\n"			\
-			"sc.w" #lrb " %1, %z4, %2\n"			\
-			"bnez        %1, 0b\n"				\
-		"1:"							\
+			"0:	lr.w %0, %2\n"				\
+			"	bne  %0, %z3, 1f\n"			\
+			"	sc.w %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			RISCV_ACQUIRE_BARRIER				\
+			"1:\n"						\
 			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
 			: "rJ" (__old), "rJ" (__new)			\
 			: "memory");					\
 		break;							\
 	case 8:								\
 		__asm__ __volatile__ (					\
-		"0:"							\
-			"lr.d" #scb " %0, %2\n"				\
-			"bne         %0, %z3, 1f\n"			\
-			"sc.d" #lrb " %1, %z4, %2\n"			\
-			"bnez        %1, 0b\n"				\
-		"1:"							\
+			"0:	lr.d %0, %2\n"				\
+			"	bne %0, %z3, 1f\n"			\
+			"	sc.d %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			RISCV_ACQUIRE_BARRIER				\
+			"1:\n"						\
 			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
 			: "rJ" (__old), "rJ" (__new)			\
 			: "memory");					\
@@ -101,34 +253,131 @@
 	__ret;								\
 })
 
-#define cmpxchg(ptr, o, n) \
-	(__cmpxchg((ptr), (o), (n), sizeof(*(ptr)), .aqrl, .aqrl))
+#define cmpxchg_acquire(ptr, o, n)					\
+({									\
+	__typeof__(*(ptr)) _o_ = (o);					\
+	__typeof__(*(ptr)) _n_ = (n);					\
+	(__typeof__(*(ptr))) __cmpxchg_acquire((ptr),			\
+					_o_, _n_, sizeof(*(ptr)));	\
+})
 
-#define cmpxchg_local(ptr, o, n) \
-	(__cmpxchg((ptr), (o), (n), sizeof(*(ptr)), , ))
+#define __cmpxchg_release(ptr, old, new, size)				\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(*(ptr)) __old = (old);				\
+	__typeof__(*(ptr)) __new = (new);				\
+	__typeof__(*(ptr)) __ret;					\
+	register unsigned int __rc;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			RISCV_RELEASE_BARRIER				\
+			"0:	lr.w %0, %2\n"				\
+			"	bne  %0, %z3, 1f\n"			\
+			"	sc.w %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			RISCV_RELEASE_BARRIER				\
+			"0:	lr.d %0, %2\n"				\
+			"	bne %0, %z3, 1f\n"			\
+			"	sc.d %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
+
+#define cmpxchg_release(ptr, o, n)					\
+({									\
+	__typeof__(*(ptr)) _o_ = (o);					\
+	__typeof__(*(ptr)) _n_ = (n);					\
+	(__typeof__(*(ptr))) __cmpxchg_release((ptr),			\
+					_o_, _n_, sizeof(*(ptr)));	\
+})
+
+#define __cmpxchg(ptr, old, new, size)					\
+({									\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(*(ptr)) __old = (old);				\
+	__typeof__(*(ptr)) __new = (new);				\
+	__typeof__(*(ptr)) __ret;					\
+	register unsigned int __rc;					\
+	switch (size) {							\
+	case 4:								\
+		__asm__ __volatile__ (					\
+			"0:	lr.w %0, %2\n"				\
+			"	bne  %0, %z3, 1f\n"			\
+			"	sc.w.rl %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"	fence rw, rw\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	case 8:								\
+		__asm__ __volatile__ (					\
+			"0:	lr.d %0, %2\n"				\
+			"	bne %0, %z3, 1f\n"			\
+			"	sc.d.rl %1, %z4, %2\n"			\
+			"	bnez %1, 0b\n"				\
+			"	fence rw, rw\n"				\
+			"1:\n"						\
+			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
+			: "rJ" (__old), "rJ" (__new)			\
+			: "memory");					\
+		break;							\
+	default:							\
+		BUILD_BUG();						\
+	}								\
+	__ret;								\
+})
 
-#define cmpxchg32(ptr, o, n)			\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 4);	\
-	cmpxchg((ptr), (o), (n));		\
+#define cmpxchg(ptr, o, n)						\
+({									\
+	__typeof__(*(ptr)) _o_ = (o);					\
+	__typeof__(*(ptr)) _n_ = (n);					\
+	(__typeof__(*(ptr))) __cmpxchg((ptr),				\
+				       _o_, _n_, sizeof(*(ptr)));	\
 })
 
-#define cmpxchg32_local(ptr, o, n)		\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 4);	\
-	cmpxchg_local((ptr), (o), (n));		\
+#define cmpxchg_local(ptr, o, n)					\
+	(__cmpxchg_relaxed((ptr), (o), (n), sizeof(*(ptr))))
+
+#define cmpxchg32(ptr, o, n)						\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 4);				\
+	cmpxchg((ptr), (o), (n));					\
 })
 
-#define cmpxchg64(ptr, o, n)			\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 8);	\
-	cmpxchg((ptr), (o), (n));		\
+#define cmpxchg32_local(ptr, o, n)					\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 4);				\
+	cmpxchg_relaxed((ptr), (o), (n))				\
 })
 
-#define cmpxchg64_local(ptr, o, n)		\
-({						\
-	BUILD_BUG_ON(sizeof(*(ptr)) != 8);	\
-	cmpxchg_local((ptr), (o), (n));		\
+#define cmpxchg64(ptr, o, n)						\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 8);				\
+	cmpxchg((ptr), (o), (n));					\
+})
+
+#define cmpxchg64_local(ptr, o, n)					\
+({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 8);				\
+	cmpxchg_relaxed((ptr), (o), (n));				\
 })
 
 #endif /* _ASM_RISCV_CMPXCHG_H */
#6 lore raw
Quoted 1 line · Palmer Dabbelt

On Fri, 09 Mar 2018 10:36:44 PST (-0800), parri.andrea at gmail.com wrote:

[...]

Quoted 6 lines
Quoted 3 lines

This belongs to the "few style fixes" (in the specific, 80-chars lines) mentioned in the cover letter; I could not resist ;-), but I'll remove them in v3 if you like so.

No problem, just next time it's a bit easier to not mix the really complicated stuff (memory model changes) with the really simple stuff (whitespace changes).

Got it.

Quoted 9 lines
Quoted 9 lines

This proposal relies on the generic definition,

include/linux/atomic.h ,

and on the

__atomic_op_acquire()
__atomic_op_release()

Quoted 6 lines
Quoted 2 lines

above to build the acquire/release atomics (except for the xchg,cmpxchg, where the ACQUIRE_BARRIER is inserted conditionally/on success).

I thought we wanted to use the AQ and RL bits for AMOs, just not for LR/SC sequences. IIRC the AMOs are safe with the current memory model, but I might just have some version mismatches in my head.

AMO.aqrl are "safe" w.r.t. the LKMM (as they provide "full-ordering"); OTOH, AMO.aq and AMO.rl present weaknesses that LKMM (and some kernel developers) do not "expect". I was probing this issue in:

https://marc.info/?l=linux-kernel&m=151930201102853&w=2

(c.f., e.g., test "RISCV-unlock-lock-read-ordering" from that post).

Quoting from the commit message of my patch 1/2:

"Referring to the "unlock-lock-read-ordering" test reported below,
Daniel wrote:

I think an RCpc interpretation of .aq and .rl would in fact allow the two normal loads in P1 to be reordered [...]

[...]

Likewise even if the unlock()/lock() is between two stores. A control dependency might originate from the load part of the amoswap.w.aq, but there still would have to be something to ensure that this load part in fact performs after the store part of the amoswap.w.rl performs globally, and that's not automatic under RCpc.

Simulation of the RISC-V memory consistency model confirmed this expectation."

I have just (re)checked these observations against the latest specification, and my results _confirmed_ these verdicts.

Andrea

#7 lore raw
Quoted 57 lines · parri.andrea at gmail.com
Quoted 1 line · Palmer Dabbelt

On Fri, 09 Mar 2018 10:36:44 PST (-0800), parri.andrea at gmail.com wrote:

[...]

Quoted 6 lines
Quoted 3 lines

This belongs to the "few style fixes" (in the specific, 80-chars lines) mentioned in the cover letter; I could not resist ;-), but I'll remove them in v3 if you like so.

No problem, just next time it's a bit easier to not mix the really complicated stuff (memory model changes) with the really simple stuff (whitespace changes).

Got it.

Quoted 9 lines
Quoted 9 lines

This proposal relies on the generic definition,

include/linux/atomic.h ,

and on the

__atomic_op_acquire()
__atomic_op_release()

Quoted 6 lines
Quoted 2 lines

above to build the acquire/release atomics (except for the xchg,cmpxchg, where the ACQUIRE_BARRIER is inserted conditionally/on success).

I thought we wanted to use the AQ and RL bits for AMOs, just not for LR/SC sequences. IIRC the AMOs are safe with the current memory model, but I might just have some version mismatches in my head.

AMO.aqrl are "safe" w.r.t. the LKMM (as they provide "full-ordering"); OTOH, AMO.aq and AMO.rl present weaknesses that LKMM (and some kernel developers) do not "expect". I was probing this issue in:

https://marc.info/?l=linux-kernel&m=151930201102853&w=2

(c.f., e.g., test "RISCV-unlock-lock-read-ordering" from that post).

Quoting from the commit message of my patch 1/2:

"Referring to the "unlock-lock-read-ordering" test reported below,
Daniel wrote:

I think an RCpc interpretation of .aq and .rl would in fact allow the two normal loads in P1 to be reordered [...]

[...]

Likewise even if the unlock()/lock() is between two stores. A control dependency might originate from the load part of the amoswap.w.aq, but there still would have to be something to ensure that this load part in fact performs after the store part of the amoswap.w.rl performs globally, and that's not automatic under RCpc.

Quoted 5 lines

Simulation of the RISC-V memory consistency model confirmed this expectation."

I have just (re)checked these observations against the latest specification, and my results _confirmed_ these verdicts.

Thanks, I must have just gotten confused about a draft spec or something. I'm pulling these on top or your other memory model related patch. I've renamed the branch "next-mm" to be a bit more descriptiove.

Thanks!

#8 lore raw
Quoted 50 lines · Palmer Dabbelt
Quoted 49 lines · parri.andrea at gmail.com
Quoted 1 line · Palmer Dabbelt

On Fri, 09 Mar 2018 10:36:44 PST (-0800), parri.andrea at gmail.com wrote:

[...]

Quoted 9 lines
Quoted 9 lines

This proposal relies on the generic definition,

?? include/linux/atomic.h ,

and on the

?? __atomic_op_acquire()
?? __atomic_op_release()

Quoted 6 lines
Quoted 2 lines

above to build the acquire/release atomics (except for the xchg,cmpxchg, where the ACQUIRE_BARRIER is inserted conditionally/on success).

I thought we wanted to use the AQ and RL bits for AMOs, just not for LR/SC sequences.? IIRC the AMOs are safe with the current memory model, but I might just have some version mismatches in my head.

AMO.aqrl are "safe" w.r.t. the LKMM (as they provide "full-ordering"); OTOH, AMO.aq and AMO.rl present weaknesses that LKMM (and some kernel developers) do not "expect".? I was probing this issue in:

? https://marc.info/?l=linux-kernel&m=151930201102853&w=2

(c.f., e.g., test "RISCV-unlock-lock-read-ordering" from that post).

Quoting from the commit message of my patch 1/2:

? "Referring to the "unlock-lock-read-ordering" test reported below, ?? Daniel wrote:

???? I think an RCpc interpretation of .aq and .rl would in fact ???? allow the two normal loads in P1 to be reordered [...]

???? [...]

???? Likewise even if the unlock()/lock() is between two stores. ???? A control dependency might originate from the load part of ???? the amoswap.w.aq, but there still would have to be something ???? to ensure that this load part in fact performs after the store ???? part of the amoswap.w.rl performs globally, and that's not ???? automatic under RCpc.

?? Simulation of the RISC-V memory consistency model confirmed this ?? expectation."

Quoted 6 lines
Quoted 2 lines

I have just (re)checked these observations against the latest specification, and my results _confirmed_ these verdicts.

Thanks, I must have just gotten confused about a draft spec or something.? I'm pulling these on top or your other memory model related patch.? I've renamed the branch "next-mm" to be a bit more descriptiove.

(Sorry for being out of the loop this week, I was out to deal with a family matter.)

I assume you're using the herd model? Luc's doing a great job with that, but even so, nothing is officially confirmed until we ratify the model. In other words, the herd model may end up changing too. If something is broken on our end, there's still time to fix it.

Regarding AMOs, let me copy from something I wrote in a previous offline conversation:

Quoted 6 lines

it seems to us that pairing a store-release of "amoswap.rl" with a "ld; fence r,rw" doesn't actually give us the RC semantics we've been discussing for LKMM. For example:

(a) sd t0,0(s0)
(b) amoswap.d.rl x0,t1,0(s1)

Quoted 7 lines
    ...

(c) ld a0,0(s1)
(d) fence r,rw
(e) sd t2,0(s2)

There, we won't get (a) ordered before (e) regardless of whether
(b) is RCpc or RCsc. Do you agree?

At the moment, only the load part of (b) is in the predecessor set of (d), but the store part of (b) is not. Likewise, the .rl annotation applies only to the store part of (b), not the load part.

This gets back to a question Linus asked last week about whether the AMO is a single unit or whether it can be considered to split into a load and a store part (which still perform atomically). For RISC-V, for right now at least, the answer is the latter. Is it still the latter for Linux too?

https://lkml.org/lkml/2018/2/26/606

Quoted 3 lines

So I think we'll need to make sure we pair .rl with .aq, or that we pair fence-based mappings with fence-based mappings, in order to make the acquire/release operations work.

This assumes we'll say that .aq and .rl are RCsc, not RCpc. But in this case, I think .aq and .rl could still be safe to use, as long as you don't ever try to mix in a fence-based mapping on the same data structure like in the example above. That might be important if we want to find the most compact legal implementation, and hence do want to use .aq and .rl after all.

Quoted 4 lines

And since we don't have native "ld.aq" today in RISC-V, that would mean smp_store_release would have to remain implemented as "fence rw,w; s{w|d}", rather than "amoswap.{w|d}.rl", for example.

Thoughts?

Thanks,
Dan

Quoted 2 lines

Thanks!

#9 lore raw
Quoted 58 lines · Daniel Lustig
Quoted 50 lines · Palmer Dabbelt
Quoted 49 lines · parri.andrea at gmail.com
Quoted 1 line · Palmer Dabbelt

On Fri, 09 Mar 2018 10:36:44 PST (-0800), parri.andrea at gmail.com wrote:

[...]

Quoted 9 lines
Quoted 9 lines

This proposal relies on the generic definition,

?? include/linux/atomic.h ,

and on the

?? __atomic_op_acquire()
?? __atomic_op_release()

Quoted 6 lines
Quoted 2 lines

above to build the acquire/release atomics (except for the xchg,cmpxchg, where the ACQUIRE_BARRIER is inserted conditionally/on success).

I thought we wanted to use the AQ and RL bits for AMOs, just not for LR/SC sequences.? IIRC the AMOs are safe with the current memory model, but I might just have some version mismatches in my head.

AMO.aqrl are "safe" w.r.t. the LKMM (as they provide "full-ordering"); OTOH, AMO.aq and AMO.rl present weaknesses that LKMM (and some kernel developers) do not "expect".? I was probing this issue in:

? https://marc.info/?l=linux-kernel&m=151930201102853&w=2

(c.f., e.g., test "RISCV-unlock-lock-read-ordering" from that post).

Quoting from the commit message of my patch 1/2:

? "Referring to the "unlock-lock-read-ordering" test reported below, ?? Daniel wrote:

???? I think an RCpc interpretation of .aq and .rl would in fact ???? allow the two normal loads in P1 to be reordered [...]

???? [...]

???? Likewise even if the unlock()/lock() is between two stores. ???? A control dependency might originate from the load part of ???? the amoswap.w.aq, but there still would have to be something ???? to ensure that this load part in fact performs after the store ???? part of the amoswap.w.rl performs globally, and that's not ???? automatic under RCpc.

?? Simulation of the RISC-V memory consistency model confirmed this ?? expectation."

Quoted 6 lines
Quoted 2 lines

I have just (re)checked these observations against the latest specification, and my results _confirmed_ these verdicts.

Thanks, I must have just gotten confused about a draft spec or something.? I'm pulling these on top or your other memory model related patch.? I've renamed the branch "next-mm" to be a bit more descriptiove.

Quoted 7 lines

(Sorry for being out of the loop this week, I was out to deal with a family matter.)

I assume you're using the herd model? Luc's doing a great job with that, but even so, nothing is officially confirmed until we ratify the model. In other words, the herd model may end up changing too. If something is broken on our end, there's still time to fix it.

No need to say :) if you look back at the LKMM as from 2 years ago or as presented last year in LWN, you won't recognize it as such ;-) Spec. do change/evolve, and so do implementations: if ratifications of the RISC-V memory model (or of the LKMM) will enable optimizations/modifications to these implementations (while preserving correctness), I'll be glad to do or to help with them.

To answer your question: I used both the herd-based model from INRIA and the operational model from the group in Cambridge (these are referred to in the currently available RISC-V spec.).

Quoted 15 lines

Regarding AMOs, let me copy from something I wrote in a previous offline conversation:

Quoted 11 lines

it seems to us that pairing a store-release of "amoswap.rl" with a "ld; fence r,rw" doesn't actually give us the RC semantics we've been discussing for LKMM. For example:

(a) sd t0,0(s0)
(b) amoswap.d.rl x0,t1,0(s1)

    ...

(c) ld a0,0(s1)
(d) fence r,rw
(e) sd t2,0(s2)

Quoted 7 lines
Quoted 2 lines

There, we won't get (a) ordered before (e) regardless of whether
(b) is RCpc or RCsc. Do you agree?

At the moment, only the load part of (b) is in the predecessor set of (d), but the store part of (b) is not. Likewise, the .rl annotation applies only to the store part of (b), not the load part.

Indeed. (If you want, this is one reason why, with these patches, ".rl" and "fence r,rw" are never "mixed" as in your snipped above unless there is also a "fence rw,rw" in between.)

Quoted 6 lines

This gets back to a question Linus asked last week about whether the AMO is a single unit or whether it can be considered to split into a load and a store part (which still perform atomically). For RISC-V, for right now at least, the answer is the latter. Is it still the latter for Linux too?

Yes: (successful) atomic RMWs all generate (at least) one load event and one store event in LKMM. (This same approach is taken by other hardware models as well...)

Quoted 7 lines

https://lkml.org/lkml/2018/2/26/606

Quoted 3 lines

So I think we'll need to make sure we pair .rl with .aq, or that we pair fence-based mappings with fence-based mappings, in order to make the acquire/release operations work.

Quoted 11 lines

This assumes we'll say that .aq and .rl are RCsc, not RCpc. But in this case, I think .aq and .rl could still be safe to use, as long as you don't ever try to mix in a fence-based mapping on the same data structure like in the example above. That might be important if we want to find the most compact legal implementation, and hence do want to use .aq and .rl after all.

Quoted 4 lines

And since we don't have native "ld.aq" today in RISC-V, that would mean smp_store_release would have to remain implemented as "fence rw,w; s{w|d}", rather than "amoswap.{w|d}.rl", for example.

And these seem to be further valid/strong arguments for these patches ;) (independently of how you'll end up ratifying .aq and .rl).

Andrea

Quoted 8 lines

Thoughts?

Thanks,
Dan

Quoted 2 lines

Thanks!

#10 lore raw
In reply to Daniel Lustig
Quoted 83 lines · Daniel Lustig
Quoted 50 lines · Palmer Dabbelt
Quoted 49 lines · On Fri, 09 Mar 2018 13:30:08 PST (-0800), [email protected] wrote:
Quoted 1 line · Palmer Dabbelt

On Fri, 09 Mar 2018 10:36:44 PST (-0800), [email protected] wrote:

[...]

Quoted 9 lines
Quoted 9 lines

This proposal relies on the generic definition,

   include/linux/atomic.h ,

and on the

   __atomic_op_acquire()
   __atomic_op_release()

Quoted 6 lines
Quoted 2 lines

above to build the acquire/release atomics (except for the xchg,cmpxchg, where the ACQUIRE_BARRIER is inserted conditionally/on success).

I thought we wanted to use the AQ and RL bits for AMOs, just not for LR/SC sequences.  IIRC the AMOs are safe with the current memory model, but I might just have some version mismatches in my head.

AMO.aqrl are "safe" w.r.t. the LKMM (as they provide "full-ordering"); OTOH, AMO.aq and AMO.rl present weaknesses that LKMM (and some kernel developers) do not "expect".  I was probing this issue in:

  https://marc.info/?l=linux-kernel&m=151930201102853&w=2

(c.f., e.g., test "RISCV-unlock-lock-read-ordering" from that post).

Quoting from the commit message of my patch 1/2:

  "Referring to the "unlock-lock-read-ordering" test reported below, Daniel wrote:

     I think an RCpc interpretation of .aq and .rl would in fact allow the two normal loads in P1 to be reordered [...]

     [...]

     Likewise even if the unlock()/lock() is between two stores. A control dependency might originate from the load part of the amoswap.w.aq, but there still would have to be something to ensure that this load part in fact performs after the store part of the amoswap.w.rl performs globally, and that's not automatic under RCpc.

   Simulation of the RISC-V memory consistency model confirmed this expectation."

Quoted 6 lines
Quoted 2 lines

I have just (re)checked these observations against the latest specification, and my results _confirmed_ these verdicts.

Thanks, I must have just gotten confused about a draft spec or something.  I'm pulling these on top or your other memory model related patch.  I've renamed the branch "next-mm" to be a bit more descriptiove.

(Sorry for being out of the loop this week, I was out to deal with a family matter.)

I assume you're using the herd model? Luc's doing a great job with that, but even so, nothing is officially confirmed until we ratify the model. In other words, the herd model may end up changing too. If something is broken on our end, there's still time to fix it.

Regarding AMOs, let me copy from something I wrote in a previous offline conversation:

Quoted 6 lines

it seems to us that pairing a store-release of "amoswap.rl" with a "ld; fence r,rw" doesn't actually give us the RC semantics we've been discussing for LKMM. For example:

(a) sd t0,0(s0)
(b) amoswap.d.rl x0,t1,0(s1)

Quoted 7 lines
    ...

(c) ld a0,0(s1)
(d) fence r,rw
(e) sd t2,0(s2)

There, we won't get (a) ordered before (e) regardless of whether
(b) is RCpc or RCsc. Do you agree?

Quoted 7 lines

At the moment, only the load part of (b) is in the predecessor set of (d), but the store part of (b) is not. Likewise, the .rl annotation applies only to the store part of (b), not the load part.

This gets back to a question Linus asked last week about whether the AMO is a single unit or whether it can be

You mean AMO or RmW atomic operations?

Quoted 4 lines

considered to split into a load and a store part (which still perform atomically). For RISC-V, for right now at least, the answer is the latter. Is it still the latter for Linux too?

I think for RmW atomics it's still the latter, the acquire or release is for the load part or the store part of an RmW. For example, ppc uses lwsync as acquire/release barriers, and lwsync could not order write->read.

Regards,
Boqun

Quoted 25 lines

https://lkml.org/lkml/2018/2/26/606

Quoted 3 lines

So I think we'll need to make sure we pair .rl with .aq, or that we pair fence-based mappings with fence-based mappings, in order to make the acquire/release operations work.

This assumes we'll say that .aq and .rl are RCsc, not RCpc. But in this case, I think .aq and .rl could still be safe to use, as long as you don't ever try to mix in a fence-based mapping on the same data structure like in the example above. That might be important if we want to find the most compact legal implementation, and hence do want to use .aq and .rl after all.

Quoted 4 lines

And since we don't have native "ld.aq" today in RISC-V, that would mean smp_store_release would have to remain implemented as "fence rw,w; s{w|d}", rather than "amoswap.{w|d}.rl", for example.

Thoughts?

Thanks,
Dan

Quoted 2 lines

Thanks!

#11 lore raw
Quoted 99 lines · Daniel Lustig
Quoted 83 lines · Palmer Dabbelt
Quoted 50 lines · parri.andrea at gmail.com
Quoted 49 lines · Palmer Dabbelt
Quoted 1 line · parri.andrea at gmail.com

[...]

Quoted 9 lines
Quoted 9 lines

This proposal relies on the generic definition,

?? include/linux/atomic.h ,

and on the

?? __atomic_op_acquire()
?? __atomic_op_release()

Quoted 6 lines
Quoted 2 lines

above to build the acquire/release atomics (except for the xchg,cmpxchg, where the ACQUIRE_BARRIER is inserted conditionally/on success).

I thought we wanted to use the AQ and RL bits for AMOs, just not for LR/SC sequences.? IIRC the AMOs are safe with the current memory model, but I might just have some version mismatches in my head.

AMO.aqrl are "safe" w.r.t. the LKMM (as they provide "full-ordering"); OTOH, AMO.aq and AMO.rl present weaknesses that LKMM (and some kernel developers) do not "expect".? I was probing this issue in:

? https://marc.info/?l=linux-kernel&m=151930201102853&w=2

(c.f., e.g., test "RISCV-unlock-lock-read-ordering" from that post).

Quoting from the commit message of my patch 1/2:

? "Referring to the "unlock-lock-read-ordering" test reported below, ?? Daniel wrote:

???? I think an RCpc interpretation of .aq and .rl would in fact ???? allow the two normal loads in P1 to be reordered [...]

???? [...]

???? Likewise even if the unlock()/lock() is between two stores. ???? A control dependency might originate from the load part of ???? the amoswap.w.aq, but there still would have to be something ???? to ensure that this load part in fact performs after the store ???? part of the amoswap.w.rl performs globally, and that's not ???? automatic under RCpc.

?? Simulation of the RISC-V memory consistency model confirmed this ?? expectation."

Quoted 6 lines
Quoted 2 lines

I have just (re)checked these observations against the latest specification, and my results _confirmed_ these verdicts.

Thanks, I must have just gotten confused about a draft spec or something.? I'm pulling these on top or your other memory model related patch.? I've renamed the branch "next-mm" to be a bit more descriptiove.

(Sorry for being out of the loop this week, I was out to deal with a family matter.)

I assume you're using the herd model? Luc's doing a great job with that, but even so, nothing is officially confirmed until we ratify the model. In other words, the herd model may end up changing too. If something is broken on our end, there's still time to fix it.

Regarding AMOs, let me copy from something I wrote in a previous offline conversation:

Quoted 6 lines

it seems to us that pairing a store-release of "amoswap.rl" with a "ld; fence r,rw" doesn't actually give us the RC semantics we've been discussing for LKMM. For example:

(a) sd t0,0(s0)
(b) amoswap.d.rl x0,t1,0(s1)

Quoted 7 lines
    ...

(c) ld a0,0(s1)
(d) fence r,rw
(e) sd t2,0(s2)

There, we won't get (a) ordered before (e) regardless of whether
(b) is RCpc or RCsc. Do you agree?

Quoted 7 lines

At the moment, only the load part of (b) is in the predecessor set of (d), but the store part of (b) is not. Likewise, the .rl annotation applies only to the store part of (b), not the load part.

This gets back to a question Linus asked last week about whether the AMO is a single unit or whether it can be

You mean AMO or RmW atomic operations?

Quoted 4 lines

considered to split into a load and a store part (which still perform atomically). For RISC-V, for right now at least, the answer is the latter. Is it still the latter for Linux too?

Quoted 4 lines

I think for RmW atomics it's still the latter, the acquire or release is for the load part or the store part of an RmW. For example, ppc uses lwsync as acquire/release barriers, and lwsync could not order write->read.

You are correct LKMM represent read-modify-write constructs with 2 events, one read and one write. Those are connected by the special "rmw" relation.

Quoted 3 lines

Regards,
Boqun

Regards,

--Luc

Quoted 26 lines
Quoted 26 lines

https://lkml.org/lkml/2018/2/26/606

Quoted 3 lines

So I think we'll need to make sure we pair .rl with .aq, or that we pair fence-based mappings with fence-based mappings, in order to make the acquire/release operations work.

This assumes we'll say that .aq and .rl are RCsc, not RCpc. But in this case, I think .aq and .rl could still be safe to use, as long as you don't ever try to mix in a fence-based mapping on the same data structure like in the example above. That might be important if we want to find the most compact legal implementation, and hence do want to use .aq and .rl after all.

Quoted 4 lines

And since we don't have native "ld.aq" today in RISC-V, that would mean smp_store_release would have to remain implemented as "fence rw,w; s{w|d}", rather than "amoswap.{w|d}.rl", for example.

Thoughts?

Thanks,
Dan

Quoted 2 lines

Thanks!