public inbox for linux-nvme@lists.infradead.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: Keith Busch <kbusch@kernel.org>, Sagi Grimberg <sagi@grimberg.me>
Cc: linux-nvme@lists.infradead.org
Subject: [PATCH 1/2] nvme-pci: fix freeing of the HMB descriptor table
Date: Fri,  1 Nov 2024 05:40:04 +0100	[thread overview]
Message-ID: <20241101044016.405265-2-hch@lst.de> (raw)
In-Reply-To: <20241101044016.405265-1-hch@lst.de>

The HMB descriptor table is sized to the maximum number of descriptors
that could be used for a given device, but __nvme_alloc_host_mem could
break out of the loop earlier on memory allocation failure and end up
using less descriptors than planned for, which leads to an incorrect
size passed to dma_free_coherent.

In practice this was not showing up because the number of descriptors
tends to be low and the dma coherent allocator always allocates and
frees at least a page.

Fixes: 87ad72a59a38 ("nvme-pci: implement host memory buffer support")
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 drivers/nvme/host/pci.c | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index 4b9fda0b1d9a..34daf6d8db07 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -153,6 +153,7 @@ struct nvme_dev {
 	/* host memory buffer support: */
 	u64 host_mem_size;
 	u32 nr_host_mem_descs;
+	u32 host_mem_descs_size;
 	dma_addr_t host_mem_descs_dma;
 	struct nvme_host_mem_buf_desc *host_mem_descs;
 	void **host_mem_desc_bufs;
@@ -1966,10 +1967,10 @@ static void nvme_free_host_mem(struct nvme_dev *dev)
 
 	kfree(dev->host_mem_desc_bufs);
 	dev->host_mem_desc_bufs = NULL;
-	dma_free_coherent(dev->dev,
-			dev->nr_host_mem_descs * sizeof(*dev->host_mem_descs),
+	dma_free_coherent(dev->dev, dev->host_mem_descs_size,
 			dev->host_mem_descs, dev->host_mem_descs_dma);
 	dev->host_mem_descs = NULL;
+	dev->host_mem_descs_size = 0;
 	dev->nr_host_mem_descs = 0;
 }
 
@@ -1977,7 +1978,7 @@ static int __nvme_alloc_host_mem(struct nvme_dev *dev, u64 preferred,
 		u32 chunk_size)
 {
 	struct nvme_host_mem_buf_desc *descs;
-	u32 max_entries, len;
+	u32 max_entries, len, descs_size;
 	dma_addr_t descs_dma;
 	int i = 0;
 	void **bufs;
@@ -1990,8 +1991,9 @@ static int __nvme_alloc_host_mem(struct nvme_dev *dev, u64 preferred,
 	if (dev->ctrl.hmmaxd && dev->ctrl.hmmaxd < max_entries)
 		max_entries = dev->ctrl.hmmaxd;
 
-	descs = dma_alloc_coherent(dev->dev, max_entries * sizeof(*descs),
-				   &descs_dma, GFP_KERNEL);
+	descs_size = max_entries * sizeof(*descs);
+	descs = dma_alloc_coherent(dev->dev, descs_size, &descs_dma,
+			GFP_KERNEL);
 	if (!descs)
 		goto out;
 
@@ -2020,6 +2022,7 @@ static int __nvme_alloc_host_mem(struct nvme_dev *dev, u64 preferred,
 	dev->host_mem_size = size;
 	dev->host_mem_descs = descs;
 	dev->host_mem_descs_dma = descs_dma;
+	dev->host_mem_descs_size = descs_size;
 	dev->host_mem_desc_bufs = bufs;
 	return 0;
 
@@ -2034,8 +2037,7 @@ static int __nvme_alloc_host_mem(struct nvme_dev *dev, u64 preferred,
 
 	kfree(bufs);
 out_free_descs:
-	dma_free_coherent(dev->dev, max_entries * sizeof(*descs), descs,
-			descs_dma);
+	dma_free_coherent(dev->dev, descs_size, descs, descs_dma);
 out:
 	dev->host_mem_descs = NULL;
 	return -ENOMEM;
-- 
2.45.2



  reply	other threads:[~2024-11-01  4:40 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-11-01  4:40 create single-segment HMBs when using IOMMU Christoph Hellwig
2024-11-01  4:40 ` Christoph Hellwig [this message]
2024-11-01  4:40 ` [PATCH 2/2] nvme-pci: use dma_alloc_noncontigous if possible Christoph Hellwig
2024-12-02 19:05   ` Leon Romanovsky
2024-12-02 19:15     ` Keith Busch
2024-12-02 19:20       ` Leon Romanovsky
2024-12-03  0:33     ` Christoph Hellwig
2024-11-05 15:59 ` create single-segment HMBs when using IOMMU Keith Busch

Reply instructions:

You may reply publicly 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=20241101044016.405265-2-hch@lst.de \
    --to=hch@lst.de \
    --cc=kbusch@kernel.org \
    --cc=linux-nvme@lists.infradead.org \
    --cc=sagi@grimberg.me \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox