Fix comment atomics.h.

I appear to accidentally have switched the comments for
pg_atomic_write_u32 and pg_atomic_read_u32 around. Also fix some minor
typos I found while fixing.

Noticed-By: Amit Kapila
Backpatch: 9.5

1 files changed, 10 insertions, 10 deletions

diff --git a/src/include/port/atomics.h b/src/include/port/atomics.h
index 97a0064111..d94fea60f1 100644
--- a/src/include/port/atomics.h
+++ b/src/include/port/atomics.h
@@ -18,7 +18,7 @@
  * new platform. If no such generic implementation is available spinlocks (or
  * even OS provided semaphores) will be used to implement the API.
  *
- * Implement the _u64 variantes if and only if your platform can use them
+ * Implement the _u64 variants if and only if your platform can use them
  * efficiently (and obviously correctly).
  *
  * Use higher level functionality (lwlocks, spinlocks, heavyweight locks)
@@ -46,7 +46,7 @@
  *
  * These files can provide the full set of atomics or can do pretty much
  * nothing if all the compilers commonly used on these platforms provide
- * useable generics.
+ * usable generics.
  *
  * Don't add an inline assembly of the actual atomic operations if all the
  * common implementations of your platform provide intrinsics. Intrinsics are
@@ -280,10 +280,12 @@ pg_atomic_init_u32(volatile pg_atomic_uint32 *ptr, uint32 val)
 }
 
 /*
- * pg_atomic_write_u32 - unlocked write to atomic variable.
+ * pg_atomic_read_u32 - unlocked read from atomic variable.
  *
- * The write is guaranteed to succeed as a whole, i.e. it's not possible to
- * observe a partial write for any reader.
+ * The read is guaranteed to return a value as it has been written by this or
+ * another process at some point in the past. There's however no cache
+ * coherency interaction guaranteeing the value hasn't since been written to
+ * again.
  *
  * No barrier semantics.
  */
@@ -295,12 +297,10 @@ pg_atomic_read_u32(volatile pg_atomic_uint32 *ptr)
 }
 
 /*
- * pg_atomic_read_u32 - unlocked read from atomic variable.
+ * pg_atomic_write_u32 - unlocked write to atomic variable.
  *
- * The read is guaranteed to return a value as it has been written by this or
- * another process at some point in the past. There's however no cache
- * coherency interaction guaranteeing the value hasn't since been written to
- * again.
+ * The write is guaranteed to succeed as a whole, i.e. it's not possible to
+ * observe a partial write for any reader.
  *
  * No barrier semantics.
  */