IOMMU Archive on lore.kernel.org
 help / color / Atom feed
From: Filippo Sironi via iommu <iommu@lists.linux-foundation.org>
To: sironi@amazon.de, joro@8bytes.org,
	iommu@lists.linux-foundation.org, linux-kernel@vger.kernel.org
Subject: [PATCH 3/5] iommu/amd: Hold the domain lock when calling __unmap_single
Date: Tue, 10 Sep 2019 19:49:23 +0200
Message-ID: <1568137765-20278-4-git-send-email-sironi@amazon.de> (raw)
In-Reply-To: <1568137765-20278-1-git-send-email-sironi@amazon.de>

__unmap_single makes several calls to __domain_flush_pages, which
traverses the device list that is protected by the domain lock.
__attach_device and __detach_device).

Also, this is in line with the comment on top of __unmap_single, which
says that the domain lock should be held when calling.

Signed-off-by: Filippo Sironi <sironi@amazon.de>
---
 drivers/iommu/amd_iommu.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c
index 8e3664821b3c..d4f25767622e 100644
--- a/drivers/iommu/amd_iommu.c
+++ b/drivers/iommu/amd_iommu.c
@@ -2508,6 +2508,7 @@ static void unmap_page(struct device *dev, dma_addr_t dma_addr, size_t size,
 {
 	struct protection_domain *domain;
 	struct dma_ops_domain *dma_dom;
+	unsigned long flags;
 
 	domain = get_domain(dev);
 	if (IS_ERR(domain))
@@ -2515,7 +2516,9 @@ static void unmap_page(struct device *dev, dma_addr_t dma_addr, size_t size,
 
 	dma_dom = to_dma_ops_domain(domain);
 
+	spin_lock_irqsave(&domain->lock, flags);
 	__unmap_single(dma_dom, dma_addr, size, dir);
+	spin_unlock_irqrestore(&domain->lock, flags);
 }
 
 static int sg_num_pages(struct device *dev,
@@ -2645,6 +2648,7 @@ static void unmap_sg(struct device *dev, struct scatterlist *sglist,
 	struct dma_ops_domain *dma_dom;
 	unsigned long startaddr;
 	int npages;
+	unsigned long flags;
 
 	domain = get_domain(dev);
 	if (IS_ERR(domain))
@@ -2654,7 +2658,9 @@ static void unmap_sg(struct device *dev, struct scatterlist *sglist,
 	dma_dom   = to_dma_ops_domain(domain);
 	npages    = sg_num_pages(dev, sglist, nelems);
 
+	spin_lock_irqsave(&domain->lock, flags);
 	__unmap_single(dma_dom, startaddr, npages << PAGE_SHIFT, dir);
+	spin_unlock_irqrestore(&domain->lock, flags);
 }
 
 /*
@@ -2726,6 +2732,7 @@ static void free_coherent(struct device *dev, size_t size,
 	struct protection_domain *domain;
 	struct dma_ops_domain *dma_dom;
 	struct page *page;
+	unsigned long flags;
 
 	page = virt_to_page(virt_addr);
 	size = PAGE_ALIGN(size);
@@ -2736,7 +2743,9 @@ static void free_coherent(struct device *dev, size_t size,
 
 	dma_dom = to_dma_ops_domain(domain);
 
+	spin_lock_irqsave(&domain->lock, flags);
 	__unmap_single(dma_dom, dma_addr, size, DMA_BIDIRECTIONAL);
+	spin_unlock_irqrestore(&domain->lock, flags);
 
 free_mem:
 	if (!dma_release_from_contiguous(dev, page, size >> PAGE_SHIFT))
-- 
2.7.4

_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

  parent reply index

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-09-10 17:49 iommu/amd: Flushing and locking fixes Filippo Sironi via iommu
2019-09-10 17:49 ` [PATCH 1/5] iommu/amd: Wait for completion of IOTLB flush in attach_device Filippo Sironi via iommu
2019-09-24  9:41   ` Joerg Roedel
2019-09-10 17:49 ` [PATCH 2/5] iommu/amd: Hold the domain lock when calling __map_single Filippo Sironi via iommu
2019-09-10 17:49 ` Filippo Sironi via iommu [this message]
2019-09-10 17:49 ` [PATCH 4/5] iommu/amd: Hold the domain lock when calling iommu_map_page Filippo Sironi via iommu
2019-09-20 18:05   ` Sironi, Filippo via iommu
2019-09-10 17:49 ` [PATCH 5/5] iommu/amd: Hold the domain lock when calling domain_flush_tlb[_pde] Filippo Sironi via iommu
2019-09-11 11:34 ` iommu/amd: Flushing and locking fixes Joerg Roedel
2019-09-13 17:37   ` Sironi, Filippo via iommu

Reply instructions:

You may reply publically to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1568137765-20278-4-git-send-email-sironi@amazon.de \
    --to=iommu@lists.linux-foundation.org \
    --cc=joro@8bytes.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sironi@amazon.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

IOMMU Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-iommu/0 linux-iommu/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-iommu linux-iommu/ https://lore.kernel.org/linux-iommu \
		iommu@lists.linux-foundation.org
	public-inbox-index linux-iommu

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.linux-foundation.lists.iommu


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git