linux-kernel - [PATCH] vfio/type1: optimize vfio_pin_pages

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <20250513035730.96387-1-lizhe.67@bytedance.com>
Date: Tue, 13 May 2025 11:57:30 +0800
From: lizhe.67@...edance.com
To: alex.williamson@...hat.com
Cc: lizhe.67@...edance.com,
	kvm@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	muchun.song@...ux.dev
Subject: [PATCH] vfio/type1: optimize vfio_pin_pages_remote() for hugetlbfs folio

From: Li Zhe <lizhe.67@...edance.com>

When vfio_pin_pages_remote() is called with a range of addresses that
includes hugetlbfs folios, the function currently performs individual
statistics counting operations for each page. This can lead to significant
performance overheads, especially when dealing with large ranges of pages.

This patch optimize this process by batching the statistics counting
operations.

The performance test results for completing the 8G VFIO IOMMU DMA mapping,
obtained through trace-cmd, are as follows. In this case, the 8G virtual
address space has been mapped to physical memory using hugetlbfs with
pagesize=2M.

Before this patch:
funcgraph_entry:      # 33813.703 us |  vfio_pin_map_dma();

After this patch:
funcgraph_entry:      # 15635.055 us |  vfio_pin_map_dma();

Signed-off-by: Li Zhe <lizhe.67@...edance.com>
---
 drivers/vfio/vfio_iommu_type1.c | 49 +++++++++++++++++++++++++++++++++
 1 file changed, 49 insertions(+)

diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type1.c
index 0ac56072af9f..bafa7f8c4cc6 100644
--- a/drivers/vfio/vfio_iommu_type1.c
+++ b/drivers/vfio/vfio_iommu_type1.c
@@ -337,6 +337,30 @@ static struct vfio_pfn *vfio_find_vpfn(struct vfio_dma *dma, dma_addr_t iova)
 	return NULL;
 }
 
+/*
+ * Find a random vfio_pfn that belongs to the range
+ * [iova, iova + PAGE_SIZE * npage)
+ */
+static struct vfio_pfn *vfio_find_vpfn_range(struct vfio_dma *dma,
+		dma_addr_t iova, unsigned long npage)
+{
+	struct vfio_pfn *vpfn;
+	struct rb_node *node = dma->pfn_list.rb_node;
+	dma_addr_t end_iova = iova + PAGE_SIZE * npage;
+
+	while (node) {
+		vpfn = rb_entry(node, struct vfio_pfn, node);
+
+		if (end_iova <= vpfn->iova)
+			node = node->rb_left;
+		else if (iova > vpfn->iova)
+			node = node->rb_right;
+		else
+			return vpfn;
+	}
+	return NULL;
+}
+
 static void vfio_link_pfn(struct vfio_dma *dma,
 			  struct vfio_pfn *new)
 {
@@ -670,6 +694,31 @@ static long vfio_pin_pages_remote(struct vfio_dma *dma, unsigned long vaddr,
 				iova += (PAGE_SIZE * ret);
 				continue;
 			}
+
+		}
+		/* Handle hugetlbfs page */
+		if (likely(!disable_hugepages) &&
+				folio_test_hugetlb(page_folio(batch->pages[batch->offset]))) {
+			if (pfn != *pfn_base + pinned)
+				goto out;
+
+			if (!rsvd && !vfio_find_vpfn_range(dma, iova, batch->size)) {
+				if (!dma->lock_cap &&
+				    mm->locked_vm + lock_acct + batch->size > limit) {
+					pr_warn("%s: RLIMIT_MEMLOCK (%ld) exceeded\n",
+						__func__, limit << PAGE_SHIFT);
+					ret = -ENOMEM;
+					goto unpin_out;
+				}
+				pinned += batch->size;
+				npage -= batch->size;
+				vaddr += PAGE_SIZE * batch->size;
+				iova += PAGE_SIZE * batch->size;
+				lock_acct += batch->size;
+				batch->offset += batch->size;
+				batch->size = 0;
+				continue;
+			}
 		}
 
 		/*
-- 
2.20.1