[PATCH] iommu/amd: Use try_cmpxchg64 in alloc_pte and free_clear_pte

[Uros Bizjak]

Use try_cmpxchg64 instead of cmpxchg64 (*ptr, old, new) != old in
alloc_pte and free_clear_pte.  cmpxchg returns success in ZF flag, so this
change saves a compare after cmpxchg (and related move instruction
in front of cmpxchg). Also, remove racy explicit assignment to pteval
when cmpxchg fails, this is what try_cmpxchg does implicitly from
*pte in an atomic way.

Signed-off-by: Uros Bizjak <ubizjak@xxxxxxxxx>
Cc: Joerg Roedel <joro@xxxxxxxxxx>
Cc: Suravee Suthikulpanit <suravee.suthikulpanit@xxxxxxx>
Cc: Will Deacon <will@xxxxxxxxxx>
---
 drivers/iommu/amd/io_pgtable.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/iommu/amd/io_pgtable.c b/drivers/iommu/amd/io_pgtable.c
index 6608d1717574..7d4b61e5db47 100644
--- a/drivers/iommu/amd/io_pgtable.c
+++ b/drivers/iommu/amd/io_pgtable.c
@@ -258,7 +258,7 @@ static u64 *alloc_pte(struct protection_domain *domain,
 			__npte = PM_LEVEL_PDE(level, iommu_virt_to_phys(page));
 
 			/* pte could have been changed somewhere. */
-			if (cmpxchg64(pte, __pte, __npte) != __pte)
+			if (!try_cmpxchg64(pte, &__pte, __npte))
 				free_page((unsigned long)page);
 			else if (IOMMU_PTE_PRESENT(__pte))
 				*updated = true;
@@ -341,10 +341,8 @@ static void free_clear_pte(u64 *pte, u64 pteval, struct list_head *freelist)
 	u64 *pt;
 	int mode;
 
-	while (cmpxchg64(pte, pteval, 0) != pteval) {
+	while (!try_cmpxchg64(pte, &pteval, 0))
 		pr_warn("AMD-Vi: IOMMU pte changed since we read it\n");
-		pteval = *pte;
-	}
 
 	if (!IOMMU_PTE_PRESENT(pteval))
 		return;
-- 
2.35.3