The usage of a cmpxchg loop in get_next_handle() is unnecessary, as the same
can be achieved with an atomic increment, which is both simpler to read, and
avoid any need for a loop.
The cmpxchg usage is likely a remnant of 32bit support, which didn't have an
instruction to do an atomic 64bit add, and instead a cmpxchg had to be used.
Suggested-by: Andrew Cooper <andrew.coop...@citrix.com>
Signed-of-by: Roger Pau Monné <roger....@citrix.com>
---
xen/arch/x86/mm/mem_sharing.c | 8 +-------
1 file changed, 1 insertion(+), 7 deletions(-)
diff --git a/xen/arch/x86/mm/mem_sharing.c b/xen/arch/x86/mm/mem_sharing.c
index 4f810706a315..fe299a2bf9aa 100644
--- a/xen/arch/x86/mm/mem_sharing.c
+++ b/xen/arch/x86/mm/mem_sharing.c
@@ -179,13 +179,7 @@ static void mem_sharing_page_unlock(struct page_info *pg)
static shr_handle_t get_next_handle(void)
{
- /* Get the next handle get_page style */
- uint64_t x, y = next_handle;
- do {
- x = y;
- }
- while ( (y = cmpxchg(&next_handle, x, x + 1)) != x );
- return x + 1;
+ return arch_fetch_and_add(&next_handle, 1) + 1;
}
static atomic_t nr_saved_mfns = ATOMIC_INIT(0);
--
2.43.0
