arm64: atomics: fix grossly inconsistent asm constraints for exclusives
Our uses of inline asm constraints for atomic operations are fairly
wild and varied. We basically need to guarantee the following:
1. Any instructions with barrier implications
(load-acquire/store-release) have a "memory" clobber
2. When performing exclusive accesses, the addresing mode is generated
using the "Q" constraint
3. Atomic blocks which use the condition flags, have a "cc" clobber
This patch addresses these concerns which, as well as fixing the
semantics of the code, stops GCC complaining about impossible asm
constraints.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
diff --git a/arch/arm64/include/asm/cmpxchg.h b/arch/arm64/include/asm/cmpxchg.h
index e0e65b0..968b5cb 100644
--- a/arch/arm64/include/asm/cmpxchg.h
+++ b/arch/arm64/include/asm/cmpxchg.h
@@ -29,39 +29,39 @@
switch (size) {
case 1:
asm volatile("// __xchg1\n"
- "1: ldaxrb %w0, [%3]\n"
- " stlxrb %w1, %w2, [%3]\n"
+ "1: ldaxrb %w0, %2\n"
+ " stlxrb %w1, %w3, %2\n"
" cbnz %w1, 1b\n"
- : "=&r" (ret), "=&r" (tmp)
- : "r" (x), "r" (ptr)
- : "memory", "cc");
+ : "=&r" (ret), "=&r" (tmp), "+Q" (*(u8 *)ptr)
+ : "r" (x)
+ : "cc", "memory");
break;
case 2:
asm volatile("// __xchg2\n"
- "1: ldaxrh %w0, [%3]\n"
- " stlxrh %w1, %w2, [%3]\n"
+ "1: ldaxrh %w0, %2\n"
+ " stlxrh %w1, %w3, %2\n"
" cbnz %w1, 1b\n"
- : "=&r" (ret), "=&r" (tmp)
- : "r" (x), "r" (ptr)
- : "memory", "cc");
+ : "=&r" (ret), "=&r" (tmp), "+Q" (*(u16 *)ptr)
+ : "r" (x)
+ : "cc", "memory");
break;
case 4:
asm volatile("// __xchg4\n"
- "1: ldaxr %w0, [%3]\n"
- " stlxr %w1, %w2, [%3]\n"
+ "1: ldaxr %w0, %2\n"
+ " stlxr %w1, %w3, %2\n"
" cbnz %w1, 1b\n"
- : "=&r" (ret), "=&r" (tmp)
- : "r" (x), "r" (ptr)
- : "memory", "cc");
+ : "=&r" (ret), "=&r" (tmp), "+Q" (*(u32 *)ptr)
+ : "r" (x)
+ : "cc", "memory");
break;
case 8:
asm volatile("// __xchg8\n"
- "1: ldaxr %0, [%3]\n"
- " stlxr %w1, %2, [%3]\n"
+ "1: ldaxr %0, %2\n"
+ " stlxr %w1, %3, %2\n"
" cbnz %w1, 1b\n"
- : "=&r" (ret), "=&r" (tmp)
- : "r" (x), "r" (ptr)
- : "memory", "cc");
+ : "=&r" (ret), "=&r" (tmp), "+Q" (*(u64 *)ptr)
+ : "r" (x)
+ : "cc", "memory");
break;
default:
BUILD_BUG();
@@ -82,14 +82,14 @@
case 1:
do {
asm volatile("// __cmpxchg1\n"
- " ldxrb %w1, [%2]\n"
+ " ldxrb %w1, %2\n"
" mov %w0, #0\n"
" cmp %w1, %w3\n"
" b.ne 1f\n"
- " stxrb %w0, %w4, [%2]\n"
+ " stxrb %w0, %w4, %2\n"
"1:\n"
- : "=&r" (res), "=&r" (oldval)
- : "r" (ptr), "Ir" (old), "r" (new)
+ : "=&r" (res), "=&r" (oldval), "+Q" (*(u8 *)ptr)
+ : "Ir" (old), "r" (new)
: "cc");
} while (res);
break;
@@ -97,29 +97,29 @@
case 2:
do {
asm volatile("// __cmpxchg2\n"
- " ldxrh %w1, [%2]\n"
+ " ldxrh %w1, %2\n"
" mov %w0, #0\n"
" cmp %w1, %w3\n"
" b.ne 1f\n"
- " stxrh %w0, %w4, [%2]\n"
+ " stxrh %w0, %w4, %2\n"
"1:\n"
- : "=&r" (res), "=&r" (oldval)
- : "r" (ptr), "Ir" (old), "r" (new)
- : "memory", "cc");
+ : "=&r" (res), "=&r" (oldval), "+Q" (*(u16 *)ptr)
+ : "Ir" (old), "r" (new)
+ : "cc");
} while (res);
break;
case 4:
do {
asm volatile("// __cmpxchg4\n"
- " ldxr %w1, [%2]\n"
+ " ldxr %w1, %2\n"
" mov %w0, #0\n"
" cmp %w1, %w3\n"
" b.ne 1f\n"
- " stxr %w0, %w4, [%2]\n"
+ " stxr %w0, %w4, %2\n"
"1:\n"
- : "=&r" (res), "=&r" (oldval)
- : "r" (ptr), "Ir" (old), "r" (new)
+ : "=&r" (res), "=&r" (oldval), "+Q" (*(u32 *)ptr)
+ : "Ir" (old), "r" (new)
: "cc");
} while (res);
break;
@@ -127,14 +127,14 @@
case 8:
do {
asm volatile("// __cmpxchg8\n"
- " ldxr %1, [%2]\n"
+ " ldxr %1, %2\n"
" mov %w0, #0\n"
" cmp %1, %3\n"
" b.ne 1f\n"
- " stxr %w0, %4, [%2]\n"
+ " stxr %w0, %4, %2\n"
"1:\n"
- : "=&r" (res), "=&r" (oldval)
- : "r" (ptr), "Ir" (old), "r" (new)
+ : "=&r" (res), "=&r" (oldval), "+Q" (*(u64 *)ptr)
+ : "Ir" (old), "r" (new)
: "cc");
} while (res);
break;