Skip to contentSkip to conversation

vmxnet3: Remove buf_info from device accessible structures

PATCH In review Ronak Doshi 3 participants 5 messages Jan 20, 2021 – Jan 22, 2021

Subsystems (4)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 vmxnet3: Remove buf_info from device accessible structures

From: Petr Vandrovec <[email protected]>

vmxnet3: Remove buf_info from device accessible structures

buf_info structures in RX & TX queues are private driver data that do not need to be visible to the device. Although there is physical address and length in the queue descriptor that points to these structures, their layout is not standardized, and device never looks at them.

So lets allocate these structures in non-DMA-able memory, and fill physical address as all-ones and length as zero in the queue descriptor.

That should alleviate worries brought by Martin Radev in
https://lists.osuosl.org/pipermail/intel-wired-lan/Week-of-Mon-20210104/022829.html
that malicious vmxnet3 device could subvert SVM/TDX guarantees.

---
 drivers/net/vmxnet3/vmxnet3_drv.c | 45 ++++++++++++++++-----------------------
 drivers/net/vmxnet3/vmxnet3_int.h |  2 --
 2 files changed, 18 insertions(+), 29 deletions(-)
drivers/net/vmxnet3/vmxnet3_drv.c +18 -27
index 336504b7531d..8f5e1e363598 100644
@@ -452,9 +452,7 @@ vmxnet3_tq_destroy(struct vmxnet3_tx_queue *tq,
 		tq->comp_ring.base = NULL;
 	}
 	if (tq->buf_info) {
-		dma_free_coherent(&adapter->pdev->dev,
-				  tq->tx_ring.size * sizeof(tq->buf_info[0]),
-				  tq->buf_info, tq->buf_info_pa);
+		kfree(tq->buf_info);
 		tq->buf_info = NULL;
 	}
 }
@@ -505,8 +503,6 @@ static int
 vmxnet3_tq_create(struct vmxnet3_tx_queue *tq,
 		  struct vmxnet3_adapter *adapter)
 {
-	size_t sz;
-
 	BUG_ON(tq->tx_ring.base || tq->data_ring.base ||
 	       tq->comp_ring.base || tq->buf_info);
 
@@ -534,11 +530,13 @@ vmxnet3_tq_create(struct vmxnet3_tx_queue *tq,
 		goto err;
 	}
 
-	sz = tq->tx_ring.size * sizeof(tq->buf_info[0]);
-	tq->buf_info = dma_alloc_coherent(&adapter->pdev->dev, sz,
-					  &tq->buf_info_pa, GFP_KERNEL);
-	if (!tq->buf_info)
+	tq->buf_info = kmalloc_array_node(tq->tx_ring.size, sizeof(tq->buf_info[0]),
+					  GFP_KERNEL | __GFP_ZERO,
+					  dev_to_node(&adapter->pdev->dev));
+	if (!tq->buf_info) {
+		netdev_err(adapter->netdev, "failed to allocate tx buffer info\n");
 		goto err;
+	}
 
 	return 0;
 
@@ -1738,10 +1736,7 @@ static void vmxnet3_rq_destroy(struct vmxnet3_rx_queue *rq,
 	}
 
 	if (rq->buf_info[0]) {
-		size_t sz = sizeof(struct vmxnet3_rx_buf_info) *
-			(rq->rx_ring[0].size + rq->rx_ring[1].size);
-		dma_free_coherent(&adapter->pdev->dev, sz, rq->buf_info[0],
-				  rq->buf_info_pa);
+		kfree(rq->buf_info[0]);
 		rq->buf_info[0] = rq->buf_info[1] = NULL;
 	}
 }
@@ -1883,12 +1878,13 @@ vmxnet3_rq_create(struct vmxnet3_rx_queue *rq, struct vmxnet3_adapter *adapter)
 		goto err;
 	}
 
-	sz = sizeof(struct vmxnet3_rx_buf_info) * (rq->rx_ring[0].size +
-						   rq->rx_ring[1].size);
-	bi = dma_alloc_coherent(&adapter->pdev->dev, sz, &rq->buf_info_pa,
-				GFP_KERNEL);
-	if (!bi)
+	bi = kmalloc_array_node(rq->rx_ring[0].size + rq->rx_ring[1].size,
+				sizeof(rq->buf_info[0][0]), GFP_KERNEL | __GFP_ZERO,
+				dev_to_node(&adapter->pdev->dev));
+	if (!bi) {
+		netdev_err(adapter->netdev, "failed to allocate rx buffer info\n");
 		goto err;
+	}
 
 	rq->buf_info[0] = bi;
 	rq->buf_info[1] = bi + rq->rx_ring[0].size;
@@ -2522,14 +2518,12 @@ vmxnet3_setup_driver_shared(struct vmxnet3_adapter *adapter)
 		tqc->txRingBasePA   = cpu_to_le64(tq->tx_ring.basePA);
 		tqc->dataRingBasePA = cpu_to_le64(tq->data_ring.basePA);
 		tqc->compRingBasePA = cpu_to_le64(tq->comp_ring.basePA);
-		tqc->ddPA           = cpu_to_le64(tq->buf_info_pa);
+		tqc->ddPA           = cpu_to_le64(~0ULL);
 		tqc->txRingSize     = cpu_to_le32(tq->tx_ring.size);
 		tqc->dataRingSize   = cpu_to_le32(tq->data_ring.size);
 		tqc->txDataRingDescSize = cpu_to_le32(tq->txdata_desc_size);
 		tqc->compRingSize   = cpu_to_le32(tq->comp_ring.size);
-		tqc->ddLen          = cpu_to_le32(
-					sizeof(struct vmxnet3_tx_buf_info) *
-					tqc->txRingSize);
+		tqc->ddLen          = cpu_to_le32(0);
 		tqc->intrIdx        = tq->comp_ring.intr_idx;
 	}
 
@@ -2541,14 +2535,11 @@ vmxnet3_setup_driver_shared(struct vmxnet3_adapter *adapter)
 		rqc->rxRingBasePA[0] = cpu_to_le64(rq->rx_ring[0].basePA);
 		rqc->rxRingBasePA[1] = cpu_to_le64(rq->rx_ring[1].basePA);
 		rqc->compRingBasePA  = cpu_to_le64(rq->comp_ring.basePA);
-		rqc->ddPA            = cpu_to_le64(rq->buf_info_pa);
+		rqc->ddPA            = cpu_to_le64(~0ULL);
 		rqc->rxRingSize[0]   = cpu_to_le32(rq->rx_ring[0].size);
 		rqc->rxRingSize[1]   = cpu_to_le32(rq->rx_ring[1].size);
 		rqc->compRingSize    = cpu_to_le32(rq->comp_ring.size);
-		rqc->ddLen           = cpu_to_le32(
-					sizeof(struct vmxnet3_rx_buf_info) *
-					(rqc->rxRingSize[0] +
-					 rqc->rxRingSize[1]));
+		rqc->ddLen           = cpu_to_le32(0);
 		rqc->intrIdx         = rq->comp_ring.intr_idx;
 		if (VMXNET3_VERSION_GE_3(adapter)) {
 			rqc->rxDataRingBasePA =
drivers/net/vmxnet3/vmxnet3_int.h +0 -2
index d958b92c9429..e910596b79cf 100644
@@ -240,7 +240,6 @@ struct vmxnet3_tx_queue {
 	spinlock_t                      tx_lock;
 	struct vmxnet3_cmd_ring         tx_ring;
 	struct vmxnet3_tx_buf_info      *buf_info;
-	dma_addr_t                       buf_info_pa;
 	struct vmxnet3_tx_data_ring     data_ring;
 	struct vmxnet3_comp_ring        comp_ring;
 	struct Vmxnet3_TxQueueCtrl      *shared;
@@ -298,7 +297,6 @@ struct vmxnet3_rx_queue {
 	u32 qid2;           /* rqID in RCD for buffer from 2nd ring */
 	u32 dataRingQid;    /* rqID in RCD for buffer from data ring */
 	struct vmxnet3_rx_buf_info     *buf_info[2];
-	dma_addr_t                      buf_info_pa;
 	struct Vmxnet3_RxQueueCtrl            *shared;
 	struct vmxnet3_rq_driver_stats  stats;
 } __attribute__((__aligned__(SMP_CACHE_BYTES)));
Signature · 1 line
2.11.0
maintainer
#1 lore raw
Quoted 3 lines · On Tue, 19 Jan 2021 18:19:40 -0800 Ronak Doshi wrote:

From: Petr Vandrovec <[email protected]>

vmxnet3: Remove buf_info from device accessible structures

Something happened to the posting, looks like the subject is listed twice?

Quoted 18 lines

buf_info structures in RX & TX queues are private driver data that do not need to be visible to the device. Although there is physical address and length in the queue descriptor that points to these structures, their layout is not standardized, and device never looks at them.

So lets allocate these structures in non-DMA-able memory, and fill physical address as all-ones and length as zero in the queue descriptor.

That should alleviate worries brought by Martin Radev in
https://lists.osuosl.org/pipermail/intel-wired-lan/Week-of-Mon-20210104/022829.html
that malicious vmxnet3 device could subvert SVM/TDX guarantees.

drivers/net/vmxnet3/vmxnet3_drv.c vmxnet3_tq_create(struct vmxnet3_tx_queue *tq, · line 535

Quoted 10 lines
drivers/net/vmxnet3/vmxnet3_drv.c +3 -4
@@ -534,11 +530,13 @@ vmxnet3_tq_create(struct vmxnet3_tx_queue *tq,
 		goto err;
 	}
 
-	sz = tq->tx_ring.size * sizeof(tq->buf_info[0]);
-	tq->buf_info = dma_alloc_coherent(&adapter->pdev->dev, sz,
-					  &tq->buf_info_pa, GFP_KERNEL);
-	if (!tq->buf_info)
+	tq->buf_info = kmalloc_array_node(tq->tx_ring.size, sizeof(tq->buf_info[0]),
+					  GFP_KERNEL | __GFP_ZERO,
+					  dev_to_node(&adapter->pdev->dev));

kcalloc_node()

drivers/net/vmxnet3/vmxnet3_drv.c vmxnet3_tq_create(struct vmxnet3_tx_queue *tq, · line 537

Quoted 2 lines
drivers/net/vmxnet3/vmxnet3_drv.c +2 -0
+	if (!tq->buf_info) {
+		netdev_err(adapter->netdev, "failed to allocate tx buffer info\n");

Please drop the message, OOM splat will be visible enough. checkpatch usually points this out

Review vmxnet3_tq_create(struct vmxnet3_tx_queue *tq, · line 539

Quoted 2 lines
(quoted diff) +1 -0
 		goto err;
+	}

Same comments for vmxnet3_rq_create()

#2 lore raw
Quoted 7 lines · Jakub Kicinski
Quoted 3 lines · On Tue, 19 Jan 2021 18:19:40 -0800 Ronak Doshi wrote:

From: Petr Vandrovec <[email protected]>

vmxnet3: Remove buf_info from device accessible structures

Something happened to the posting, looks like the subject is listed
twice?

It got sent twice. Please ignore.

Quoted 6 lines

drivers/net/vmxnet3/vmxnet3_drv.c vmxnet3_tq_create(struct vmxnet3_tx_queue *tq, · line 535

Quoted 4 lines
drivers/net/vmxnet3/vmxnet3_drv.c +3 -1
-	if (!tq->buf_info)
+	tq->buf_info = kmalloc_array_node(tq->tx_ring.size, sizeof(tq->buf_info[0]),
+					  GFP_KERNEL | __GFP_ZERO,
+					  dev_to_node(&adapter->pdev->dev));

kcalloc_node()

Sure, will use this callback.

Quoted 5 lines

Review

Quoted 2 lines
(quoted diff) +2 -0
+	if (!tq->buf_info) {
+		netdev_err(adapter->netdev, "failed to allocate tx buffer info\n")

Please drop the message, OOM splat will be visible enough. checkpatch usually points this out

Okay, will drop it. Checkpatch did not complain about the error message though.

Thanks,
Ronak

maintainer
#3 lore raw
Quoted 11 lines · On Fri, 22 Jan 2021 08:24:59 +0000 Ronak Doshi wrote:
Quoted 9 lines · Jakub Kicinski

drivers/net/vmxnet3/vmxnet3_drv.c vmxnet3_tq_create(struct vmxnet3_tx_queue *tq,

Quoted 5 lines · On Tue, 19 Jan 2021 18:19:40 -0800 Ronak Doshi wrote:
drivers/net/vmxnet3/vmxnet3_drv.c +5 -0
+	tq->buf_info = kmalloc_array_node(tq->tx_ring.size, sizeof(tq->buf_info[0]),
+					  GFP_KERNEL | __GFP_ZERO,
+					  dev_to_node(&adapter->pdev->dev));  
+	if (!tq->buf_info) {
+		netdev_err(adapter->netdev, "failed to allocate tx buffer info\n")  

Please drop the message, OOM splat will be visible enough. checkpatch usually points this out

Quoted 1 line

Okay, will drop it. Checkpatch did not complain about the error message though.

Looks like it matches alloc_node or alloc_array, but not alloc_array_node ?

our $allocFunctions = qr{(?x:

	(?:(?:devm_)?
		(?:kv|k|v)[czm]alloc(?:_node|_array)? |

CCing Joe in case he thinks it's worth fixing

maintainer
#4 lore raw

PATCH 1/1 checkpatch: Add kmalloc_array_node to unnecessary OOM message check

commit 5799b255c491 ("include/linux/slab.h: add kmalloc_array_node() and kcalloc_node()") was added in 2017. Update the unnecessary OOM message test to include it.

---

Maybe not worth fixing, but no real effort to fix either.

 scripts/checkpatch.pl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
scripts/checkpatch.pl +1 -1
index 4f8494527139..8dbf1986a8de 100755
@@ -487,7 +487,7 @@ our $logFunctions = qr{(?x:
 
 our $allocFunctions = qr{(?x:
 	(?:(?:devm_)?
-		(?:kv|k|v)[czm]alloc(?:_node|_array)? |
+		(?:kv|k|v)[czm]alloc(?:_array)?(?:_node)? |
 		kstrdup(?:_const)? |
 		kmemdup(?:_nul)?) |
 	(?:\w+)?alloc_skb(?:_ip_align)? |