Re: DMA error when sg->offset value is greater than PAGE_SIZE in Intel IOMMU

[Casey Leedom]

| From: Robin Murphy <robin.murphy@xxxxxxx>
| Sent: Tuesday, September 26, 2017 7:22 AM
|
| On 26/09/17 13:21, Harsh Jain wrote:
| > Find attached new set of log. After repeated tries it panics.
|
| Thanks, that makes things a bit clearer - looks like fixing the physical
| address/pteval calculation to not be off by a page in one direction wasn't
| helping much because the returned DMA address is actually also off by a
| page in the other direction, and thus overflowing past the allocated IOVA
| into whoever else's mapping happened to be there; complete carnage ensues.
|
| After another look through the intel_map_sg() path, here's my second (still
| completely untested) guess at a possible fix.
|
| Robin.
|
| ----->8-----
| diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
| index 6784a05dd6b2..d7f7def81613 100644
| --- a/drivers/iommu/intel-iommu.c
| +++ b/drivers/iommu/intel-iommu.c
| @@ -2254,10 +2254,12 @@ static int __domain_mapping(struct dmar_domain *doma!
|                  uint64_t tmp;
|
|                  if (!sg_res) {
| +                       size_t off = sg->offset & ~PAGE_MASK;
| +
|                          sg_res = aligned_nrpages(sg->offset, sg->length);
| -                       sg->dma_address = ((dma_addr_t)iov_pfn << VTD_PAGE_S!
| +                       sg->dma_address = ((dma_addr_t)iov_pfn << VTD_PAGE_S!
|                          sg->dma_length = sg->length;
| -                       pteval = page_to_phys(sg_page(sg)) | prot;
| +                       pteval = (page_to_phys(sg_page(sg)) + sg->offset - o!
|                          phys_pfn = pteval >> VTD_PAGE_SHIFT;
|                  }

Robin,

  Harsh was able to do an initial test of you proposed patch above and his
test setup survived for 2 minutes with iperf traffic before his PEER went
belly up.  Since his PEER isn't on a Remote Power IPMI, we'll have to wait
till tomorrow for further tests -- Harsh does have to sleep some time ...
:-) But the 2 minutes his test machine did survive were Very Promising!
 
  And I see that Raj has also sent a couple of proposed fixes now.
It's just as well that Harsh is off to bed now so we can hash things out
while he sleeps ...

Casey