ARC: atomics: Implement arch_atomic64_cmpxchg using _relaxed
Subsystems (2)
PATCH 1/1 ARC: atomics: Implement arch_atomic64_cmpxchg using _relaxed
The core atomic code has a number of macros where it elaborates architecture primitives into more functions. ARC uses arch_atomic64_cmpxchg() as it's architecture primitive which disable alot of the additional functions.
Instead provide arch_cmpxchg64_relaxed() as the primitive and rely on the core macros to create arch_cmpxchg64().
The macros will also provide other functions, for instance, try_cmpxchg64_release(), giving a more complete implementation.
--- arch/arc/include/asm/atomic64-arcv2.h | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-)
arch/arc/include/asm/atomic64-arcv2.h +5 -10
@@ -137,12 +137,9 @@ ATOMIC64_OPS(xor, xor, xor)#undef ATOMIC64_OP_RETURN#undef ATOMIC64_OP -static inline s64 -arch_atomic64_cmpxchg (atomic64_t *ptr,s64 expected,s64 new) +static inline u64 __arch_cmpxchg64_relaxed (volatile void *ptr,u64 old,u64 new) { -s64 prev; - -smp_mb (); +u64 prev; __asm__ __volatile__( "1: llockd %0, [%1] \n"
@@ -152,14 +149,12 @@ arch_atomic64_cmpxchg(atomic64_t *ptr, s64 expected, s64 new) " bnz 1b \n" "2: \n" : "=&r"(prev) - : "r"(ptr), "ir"(expected), "r"(new) - : "cc"); /* memory clobber comes from smp_mb() */ - -smp_mb (); + : "r"(ptr), "ir"(old), "r"(new) + : "memory", "cc");return prev; } -#define arch_atomic64_cmpxchg arch_atomic64_cmpxchg +#define arch_cmpxchg64_relaxed __arch_cmpxchg64_relaxedstatic inline s64 arch_atomic64_xchg (atomic64_t *ptr,s64 new) {
base-commit: ea8f6ee2111cd78b32d0363ea630ba9b08ada22d
Signature · 1 line
2.43.0
Quoted 8 lines · Jason Gunthorpe
The core atomic code has a number of macros where it elaborates architecture primitives into more functions. ARC uses
arch_atomic64_cmpxchg()as it's architecture primitive which disable alot of the additional functions.Instead provide
arch_cmpxchg64_relaxed()as the primitive and rely on the core macros to createarch_cmpxchg64().
Quoted 6 lines
The macros will also provide other functions, for instance,
try_cmpxchg64_release(), giving a more complete implementation.Suggested-by Mark Rutland <[email protected]>Signed-off-by Jason Gunthorpe <[email protected]>
Quoted 42 lines
--- arch/arc/include/asm/atomic64-arcv2.h | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-)arch/arc/include/asm/atomic64-arcv2.h +5
-10@@ -137,12 +137,9 @@ ATOMIC64_OPS(xor, xor, xor)#undef ATOMIC64_OP_RETURN#undef ATOMIC64_OP -static inline s64 -arch_atomic64_cmpxchg (atomic64_t *ptr,s64 expected,s64 new) +static inline u64 __arch_cmpxchg64_relaxed (volatile void *ptr,u64 old,u64 new) { -s64 prev; - -smp_mb (); +u64 prev; __asm__ __volatile__( "1: llockd %0, [%1] \n"@@ -152,14 +149,12 @@ arch_atomic64_cmpxchg(atomic64_t *ptr, s64 expected, s64 new) " bnz 1b \n" "2: \n" : "=&r"(prev) - : "r"(ptr), "ir"(expected), "r"(new) - : "cc"); /* memory clobber comes from smp_mb() */ - -smp_mb (); + : "r"(ptr), "ir"(old), "r"(new) + : "memory", "cc");return prev; } -#define arch_atomic64_cmpxchg arch_atomic64_cmpxchg +#define arch_cmpxchg64_relaxed __arch_cmpxchg64_relaxedstatic inline s64 arch_atomic64_xchg (atomic64_t *ptr,s64 new) {
Quoted 1 line
base-commit:
ea8f6ee2111cd78b32d0363ea630ba9b08ada22d
Thx for this.
FWIW I missed the cmpxchg API when the rest of atomics were relaxed in commit 301014cf6d728 ("ARC: atomic_cmpxchg/atomic_xchg: implement relaxed variants")
Added to ARC for-curr (for next)
Thx,
-Vineet
Quoted 6 lines · Vineet Gupta
Thx for this.
FWIW I missed the cmpxchg API when the rest of atomics were relaxed in commit
301014cf6d728("ARC: atomic_cmpxchg/atomic_xchg: implement relaxed variants")Added to ARC for-curr (for next)
I'm looking at v6.16-rc1 and I don't see this patch???
It is in linux-next though:
commit ef7647b215c919a82ee3235b80998c9bb3a915a3
Author: Jason Gunthorpe <[email protected]> Date: Tue Apr 8 14:22:56 2025 -0300
ARC: atomics: Implement arch_atomic64_cmpxchg using _relaxed
Jason