Skip to contentSkip to conversation

iommu/dma: Fix calculation overflow in __finalise_sg()

PATCH In review Nicolin Chen 3 participants 7 messages Jun 22, 2019 – Jul 25, 2019

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 iommu/dma: Fix calculation overflow in __finalise_sg()

The max_len is a u32 type variable so the calculation on the left hand of the last if-condition will potentially overflow when a cur_len gets closer to UINT_MAX -- note that there're drivers setting max_seg_size to UINT_MAX:
drivers/dma/dw-edma/dw-edma-core.c:745:

    dma_set_max_seg_size(dma->dev, U32_MAX);
  drivers/dma/dma-axi-dmac.c:871:
    dma_set_max_seg_size(&pdev->dev, UINT_MAX);
  drivers/mmc/host/renesas_sdhi_internal_dmac.c:338:
    dma_set_max_seg_size(dev, 0xffffffff);
  drivers/nvme/host/pci.c:2520:
    dma_set_max_seg_size(dev->dev, 0xffffffff);

So this patch just casts the cur_len in the calculation to a size_t type to fix the overflow issue, as it's not necessary to change the type of cur_len after all.

Fixes 809eac54cdd6 ("iommu/dma: Implement scatterlist segment merging")
---
 drivers/iommu/dma-iommu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iommu/dma-iommu.c +1 -1
index a9f13313a22f..676b7ecd451e 100644
@@ -764,7 +764,7 @@ static int __finalise_sg(struct device *dev, struct scatterlist *sg, int nents,
 		 * - and wouldn't make the resulting output segment too long
 		 */
 		if (cur_len && !s_iova_off && (dma_addr & seg_mask) &&
-		    (cur_len + s_length <= max_len)) {
+		    ((size_t)cur_len + s_length <= max_len)) {
 			/* ...then concatenate it with the previous one */
 			cur_len += s_length;
 		} else {
Signature · 1 line
2.17.1
maintainer
#1 lore raw
Quoted 13 lines · Nicolin Chen

The max_len is a u32 type variable so the calculation on the left hand of the last if-condition will potentially overflow when a cur_len gets closer to UINT_MAX -- note that there're drivers setting max_seg_size to UINT_MAX:
drivers/dma/dw-edma/dw-edma-core.c:745:

    dma_set_max_seg_size(dma->dev, U32_MAX);
  drivers/dma/dma-axi-dmac.c:871:
    dma_set_max_seg_size(&pdev->dev, UINT_MAX);
  drivers/mmc/host/renesas_sdhi_internal_dmac.c:338:
    dma_set_max_seg_size(dev, 0xffffffff);
  drivers/nvme/host/pci.c:2520:
    dma_set_max_seg_size(dev->dev, 0xffffffff);
Quoted 7 lines

So this patch just casts the cur_len in the calculation to a size_t type to fix the overflow issue, as it's not necessary to change the type of cur_len after all.

Fixes 809eac54cdd6 ("iommu/dma: Implement scatterlist segment merging")

Looks good to me, but I let Robin take a look too before I apply it, Robin?

_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu

maintainer
#2 lore raw
Quoted 18 lines · Joerg Roedel
Quoted 17 lines · Nicolin Chen

The max_len is a u32 type variable so the calculation on the left hand of the last if-condition will potentially overflow when a cur_len gets closer to UINT_MAX -- note that there're drivers setting max_seg_size to UINT_MAX:

   drivers/dma/dw-edma/dw-edma-core.c:745:
     dma_set_max_seg_size(dma->dev, U32_MAX);
   drivers/dma/dma-axi-dmac.c:871:
     dma_set_max_seg_size(&pdev->dev, UINT_MAX);
   drivers/mmc/host/renesas_sdhi_internal_dmac.c:338:
     dma_set_max_seg_size(dev, 0xffffffff);
   drivers/nvme/host/pci.c:2520:
     dma_set_max_seg_size(dev->dev, 0xffffffff);

So this patch just casts the cur_len in the calculation to a size_t type to fix the overflow issue, as it's not necessary to change the type of cur_len after all.

Quoted 6 lines
Quoted 3 lines
Fixes 809eac54cdd6 ("iommu/dma: Implement scatterlist segment merging")

Looks good to me, but I let Robin take a look too before I apply it, Robin?

I'll need to take a closer look at how exactly an overflow would happen here (just got back off some holiday), but my immediate thought is that if this is a real problem, then what about 32-bit builds where size_t would still overflow?

Robin.
_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu

#3 lore raw

Hi Robin,

Quoted 17 lines · Robin Murphy
Quoted 17 lines
Quoted 17 lines

The max_len is a u32 type variable so the calculation on the left hand of the last if-condition will potentially overflow when a cur_len gets closer to UINT_MAX -- note that there're drivers setting max_seg_size to UINT_MAX:

   drivers/dma/dw-edma/dw-edma-core.c:745:
     dma_set_max_seg_size(dma->dev, U32_MAX);
   drivers/dma/dma-axi-dmac.c:871:
     dma_set_max_seg_size(&pdev->dev, UINT_MAX);
   drivers/mmc/host/renesas_sdhi_internal_dmac.c:338:
     dma_set_max_seg_size(dev, 0xffffffff);
   drivers/nvme/host/pci.c:2520:
     dma_set_max_seg_size(dev->dev, 0xffffffff);

So this patch just casts the cur_len in the calculation to a size_t type to fix the overflow issue, as it's not necessary to change the type of cur_len after all.

Quoted 7 lines
Quoted 6 lines
Quoted 3 lines
Fixes 809eac54cdd6 ("iommu/dma: Implement scatterlist segment merging")

Looks good to me, but I let Robin take a look too before I apply it, Robin?

I'll need to take a closer look at how exactly an overflow would happen here

It was triggered by a test case that was trying to map a 4GB dma_buf (1000+ nents in the scatterlist). This function then seemed to reduce the nents by merging most of them, probably because they were contiguous.

Quoted 3 lines

(just got back off some holiday), but my immediate thought is that if this is a real problem, then what about 32-bit builds where size_t would still overflow?

I think most of callers are also using size_t type for their size parameters like dma_buf, so the cur_len + s_length will unlikely go higher than UINT_MAX. But just in case that some driver allocates a large sg with a size parameter defined in 64-bit and uses this map() function, so it might be safer to change to "size_t" here to "u64"?

Thank you
Nicolin
_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu

maintainer
#4 lore raw
Quoted 24 lines · Nicolin Chen

Hi Robin,

Quoted 21 lines · Robin Murphy
Quoted 21 lines
Quoted 20 lines

The max_len is a u32 type variable so the calculation on the left hand of the last if-condition will potentially overflow when a cur_len gets closer to UINT_MAX -- note that there're drivers setting max_seg_size to UINT_MAX:

    drivers/dma/dw-edma/dw-edma-core.c:745:
      dma_set_max_seg_size(dma->dev, U32_MAX);
    drivers/dma/dma-axi-dmac.c:871:
      dma_set_max_seg_size(&pdev->dev, UINT_MAX);
    drivers/mmc/host/renesas_sdhi_internal_dmac.c:338:
      dma_set_max_seg_size(dev, 0xffffffff);
    drivers/nvme/host/pci.c:2520:
      dma_set_max_seg_size(dev->dev, 0xffffffff);

So this patch just casts the cur_len in the calculation to a size_t type to fix the overflow issue, as it's not necessary to change the type of cur_len after all.

Fixes 809eac54cdd6 ("iommu/dma: Implement scatterlist segment merging")
Quoted 8 lines
Quoted 3 lines
Quoted 2 lines

Looks good to me, but I let Robin take a look too before I apply it, Robin?

I'll need to take a closer look at how exactly an overflow would happen here

It was triggered by a test case that was trying to map a 4GB dma_buf (1000+ nents in the scatterlist). This function then seemed to reduce the nents by merging most of them, probably because they were contiguous.

So, looking at the code now, the actual problem seems to be that this overflow allows segments to be merged to or beyond 4GB total length, thus sg->dma_length for the resulting segment also ends up wrapped and truncated, and the caller will end up going downhill from there; is that correct?

Quoted 10 lines
Quoted 3 lines

(just got back off some holiday), but my immediate thought is that if this is a real problem, then what about 32-bit builds where size_t would still overflow?

I think most of callers are also using size_t type for their size parameters like dma_buf, so the cur_len + s_length will unlikely go higher than UINT_MAX. But just in case that some driver allocates a large sg with a size parameter defined in 64-bit and uses this map() function, so it might be safer to change to "size_t" here to "u64"?

On reflection, I don't really think that size_t fits here anyway, since all the members of the incoming struct scatterlist are unsigned int too. Does the patch below work?

Robin.

----->8-----
From: Robin Murphy <[email protected]>
Subject: [PATCH] iommu/dma: Handle SG length overflow better

Since scatterlist dimensions are all unsigned ints, in the relatively rare cases where a device's max_segment_size is set to UINT_MAX, then the "cur_len + s_length <= max_len" check in __finalise_sg() will always return true. As a result, the corner case of such a device mapping an excessively large scatterlist which is mergeable to or beyond a total length of 4GB can lead to overflow and a bogus truncated dma_length in the resulting segment.

As we already assume that any single segment must be no longer than max_len to begin with, this can easily be addressed by reshuffling the comparison.

Fixes 809eac54cdd6 ("iommu/dma: Implement scatterlist segment merging")
---
 drivers/iommu/dma-iommu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iommu/dma-iommu.c +1 -1
index 129c4badf9ae..8de6cf623362 100644
@@ -721,7 +721,7 @@ static int __finalise_sg(struct device *dev, struct scatterlist *sg, int nents,
 		 * - and wouldn't make the resulting output segment too long
 		 */
 		if (cur_len && !s_iova_off && (dma_addr & seg_mask) &&
-		    (cur_len + s_length <= max_len)) {
+		    (max_len - cur_len >= s_length)) {
 			/* ...then concatenate it with the previous one */
 			cur_len += s_length;
 		} else {
author Tested-by
#5 lore raw
Quoted 3 lines · Robin Murphy

On reflection, I don't really think that size_t fits here anyway, since all the members of the incoming struct scatterlist are unsigned int too. Does the patch below work?

Yes.

Quoted 12 lines

----->8-----
From: Robin Murphy <[email protected]>
Subject: [PATCH] iommu/dma: Handle SG length overflow better

Since scatterlist dimensions are all unsigned ints, in the relatively rare cases where a device's max_segment_size is set to UINT_MAX, then the "cur_len + s_length <= max_len" check in __finalise_sg() will always return true. As a result, the corner case of such a device mapping an excessively large scatterlist which is mergeable to or beyond a total length of 4GB can lead to overflow and a bogus truncated dma_length in the resulting segment.

Quoted 7 lines

As we already assume that any single segment must be no longer than max_len to begin with, this can easily be addressed by reshuffling the comparison.

Fixes 809eac54cdd6 ("iommu/dma: Implement scatterlist segment merging")

Thank you!

Quoted 9 lines
---
 drivers/iommu/dma-iommu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iommu/dma-iommu.c +0 -0
index 129c4badf9ae..8de6cf623362 100644

drivers/iommu/dma-iommu.c static int __finalise_sg(struct device *dev, struct scatterlist *sg, int nents, · line 727

Quoted 8 lines
drivers/iommu/dma-iommu.c +1 -1
@@ -721,7 +721,7 @@ static int __finalise_sg(struct device *dev, struct scatterlist *sg, int nents,
 		 * - and wouldn't make the resulting output segment too long
 		 */
 		if (cur_len && !s_iova_off && (dma_addr & seg_mask) &&
-		    (cur_len + s_length <= max_len)) {
+		    (max_len - cur_len >= s_length)) {
 			/* ...then concatenate it with the previous one */
 			cur_len += s_length;
 		} else {

_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu

#6 lore raw

Sorry to ping this but it's been a while.

Robin, did you get a chance to resend your version?

Thanks
Nicolin

Quoted 35 lines · Nicolin Chen
Quoted 3 lines · Robin Murphy

On reflection, I don't really think that size_t fits here anyway, since all the members of the incoming struct scatterlist are unsigned int too. Does the patch below work?

Yes.

Quoted 12 lines

----->8-----
From: Robin Murphy <[email protected]>
Subject: [PATCH] iommu/dma: Handle SG length overflow better

Since scatterlist dimensions are all unsigned ints, in the relatively rare cases where a device's max_segment_size is set to UINT_MAX, then the "cur_len + s_length <= max_len" check in __finalise_sg() will always return true. As a result, the corner case of such a device mapping an excessively large scatterlist which is mergeable to or beyond a total length of 4GB can lead to overflow and a bogus truncated dma_length in the resulting segment.

Quoted 7 lines

As we already assume that any single segment must be no longer than max_len to begin with, this can easily be addressed by reshuffling the comparison.

Fixes 809eac54cdd6 ("iommu/dma: Implement scatterlist segment merging")

Thank you!

Quoted 4 lines
---
 drivers/iommu/dma-iommu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Quoted 13 lines

drivers/iommu/dma-iommu.c static int __finalise_sg(struct device *dev, struct scatterlist *sg, int nents, · line 727

Quoted 13 lines
drivers/iommu/dma-iommu.c +1 -1
index 129c4badf9ae..8de6cf623362 100644
@@ -721,7 +721,7 @@ static int __finalise_sg(struct device *dev, struct scatterlist *sg, int nents,
 		 * - and wouldn't make the resulting output segment too long
 		 */
 		if (cur_len && !s_iova_off && (dma_addr & seg_mask) &&
-		    (cur_len + s_length <= max_len)) {
+		    (max_len - cur_len >= s_length)) {
 			/* ...then concatenate it with the previous one */
 			cur_len += s_length;
 		} else {

_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu