iommu/dma: Plumb in the per-CPU IOVA caches

With IOVA allocation suitably tidied up, we are finally free to opt in
to the per-CPU caching mechanism. The caching alone can provide a modest
improvement over walking the rbtree for weedier systems (iperf3 shows
~10% more ethernet throughput on an ARM Juno r1 constrained to a single
650MHz Cortex-A53), but the real gain will be in sidestepping the rbtree
lock contention which larger ARM-based systems with lots of parallel I/O
are starting to feel the pain of.
Reviewed-by: Nate Watterson <nwatters@codeaurora.org>
Tested-by: Nate Watterson <nwatters@codeaurora.org>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>

drivers/iommu/dma-iommu.c

@@ -361,8 +361,7 @@ static dma_addr_t iommu_dma_alloc_iova(struct iommu_domain *domain,
 {
 	struct iommu_dma_cookie *cookie = domain->iova_cookie;
 	struct iova_domain *iovad = &cookie->iovad;
-	unsigned long shift, iova_len;
-	struct iova *iova = NULL;
+	unsigned long shift, iova_len, iova = 0;
 
 	if (cookie->type == IOMMU_DMA_MSI_COOKIE) {
 		cookie->msi_iova += size;
@@ -371,41 +370,39 @@ static dma_addr_t iommu_dma_alloc_iova(struct iommu_domain *domain,
 
 	shift = iova_shift(iovad);
 	iova_len = size >> shift;
+	/*
+	 * Freeing non-power-of-two-sized allocations back into the IOVA caches
+	 * will come back to bite us badly, so we have to waste a bit of space
+	 * rounding up anything cacheable to make sure that can't happen. The
+	 * order of the unadjusted size will still match upon freeing.
+	 */
+	if (iova_len < (1 << (IOVA_RANGE_CACHE_MAX_SIZE - 1)))
+		iova_len = roundup_pow_of_two(iova_len);
 
 	if (domain->geometry.force_aperture)
 		dma_limit = min(dma_limit, domain->geometry.aperture_end);
 
 	/* Try to get PCI devices a SAC address */
 	if (dma_limit > DMA_BIT_MASK(32) && dev_is_pci(dev))
-		iova = alloc_iova(iovad, iova_len, DMA_BIT_MASK(32) >> shift,
-				  true);
-	/*
-	 * Enforce size-alignment to be safe - there could perhaps be an
-	 * attribute to control this per-device, or at least per-domain...
-	 */
+		iova = alloc_iova_fast(iovad, iova_len, DMA_BIT_MASK(32) >> shift);
+
 	if (!iova)
-		iova = alloc_iova(iovad, iova_len, dma_limit >> shift, true);
+		iova = alloc_iova_fast(iovad, iova_len, dma_limit >> shift);
 
-	return (dma_addr_t)iova->pfn_lo << shift;
+	return (dma_addr_t)iova << shift;
 }
 
 static void iommu_dma_free_iova(struct iommu_dma_cookie *cookie,
 		dma_addr_t iova, size_t size)
 {
 	struct iova_domain *iovad = &cookie->iovad;
-	struct iova *iova_rbnode;
+	unsigned long shift = iova_shift(iovad);
 
 	/* The MSI case is only ever cleaning up its most recent allocation */
-	if (cookie->type == IOMMU_DMA_MSI_COOKIE) {
+	if (cookie->type == IOMMU_DMA_MSI_COOKIE)
 		cookie->msi_iova -= size;
-		return;
-	}
-
-	iova_rbnode = find_iova(iovad, iova_pfn(iovad, iova));
-	if (WARN_ON(!iova_rbnode))
-		return;
-
-	__free_iova(iovad, iova_rbnode);
+	else
+		free_iova_fast(iovad, iova >> shift, size >> shift);
 }
 
 static void __iommu_dma_unmap(struct iommu_domain *domain, dma_addr_t dma_addr,