[PATCH] swiotlb: fix the check whether a device has used software IO TLB

[Petr Tesarik]

When CONFIG_SWIOTLB_DYNAMIC=y, devices which do not use the software IO TLB
can avoid swiotlb lookup. A flag is added by commit 1395706a1490 ("swiotlb:
search the software IO TLB only if the device makes use of it"), the flag
is correctly set, but it is then never checked. Add the actual check here.

Note that this code is an alternative to the default pool check, not an
additional check, because:

1. swiotlb_find_pool() also searches the default pool;
2. if dma_uses_io_tlb is false, the default swiotlb pool is not used.

Tested in a KVM guest against a QEMU RAM-backed SATA disk over virtio and
*not* using software IO TLB, this patch increases IOPS by something between
2% and 3%.

Fixes: 1395706a1490 ("swiotlb: search the software IO TLB only if the device makes use of it")
Reported-by: Jonathan Corbet <corbet@xxxxxxx>
Closes: https://lore.kernel.org/linux-iommu/87a5uz3ob8.fsf@xxxxxxxxxxxx/
Signed-off-by: Petr Tesarik <petr@xxxxxxxxxxx>
---
 include/linux/swiotlb.h | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h
index b4536626f8ff..c94f7f205174 100644
--- a/include/linux/swiotlb.h
+++ b/include/linux/swiotlb.h
@@ -172,14 +172,16 @@ static inline bool is_swiotlb_buffer(struct device *dev, phys_addr_t paddr)
 	if (!mem)
 		return false;
 
-	if (IS_ENABLED(CONFIG_SWIOTLB_DYNAMIC)) {
-		/* Pairs with smp_wmb() in swiotlb_find_slots() and
-		 * swiotlb_dyn_alloc(), which modify the RCU lists.
-		 */
-		smp_rmb();
-		return swiotlb_find_pool(dev, paddr);
-	}
+#ifdef CONFIG_SWIOTLB_DYNAMIC
+	/* Pairs with smp_wmb() in swiotlb_find_slots() and
+	 * swiotlb_dyn_alloc(), which modify the RCU lists.
+	 */
+	smp_rmb();
+	return READ_ONCE(dev->dma_uses_io_tlb) &&
+		swiotlb_find_pool(dev, paddr);
+#else
 	return paddr >= mem->defpool.start && paddr < mem->defpool.end;
+#endif
 }
 
 static inline bool is_swiotlb_force_bounce(struct device *dev)
-- 
2.42.0