set_msi_sid_cb() is used to determine whether device aliases share the
same bus, but it can provide false indications that aliases use the same
bus when in fact they do not. The reason is that set_msi_sid_cb()
assumes that pdev is fixed, while actually pci_for_each_dma_alias() can
call fn() when pdev is set to a subordinate device.

As a result, running an VM on ESX with VT-d emulation enabled can
results in the log warning such as:

  DMAR: [INTR-REMAP] Request device [00:11.0] fault index 3b [fault reason 38] 
Blocked an interrupt request due to source-id verification failure

This seems to cause additional ata errors such as:
  ata3.00: qc timeout (cmd 0xa1)
  ata3.00: failed to IDENTIFY (I/O error, err_mask=0x4)

These timeouts also cause boot to be much longer and other errors.

Fix it by checking comparing the alias with the previous one instead.

Fixes: 3f0c625c6ae71 ("iommu/vt-d: Allow interrupts from the entire bus for 
aliased devices")
Cc: sta...@vger.kernel.org
Cc: Logan Gunthorpe <log...@deltatee.com>
Cc: David Woodhouse <dw...@infradead.org>
Cc: Joerg Roedel <j...@8bytes.org>
Cc: Jacob Pan <jacob.jun....@linux.intel.com>
Signed-off-by: Nadav Amit <na...@vmware.com>
---
 drivers/iommu/intel_irq_remapping.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/iommu/intel_irq_remapping.c 
b/drivers/iommu/intel_irq_remapping.c
index 4786ca061e31..81e43c1df7ec 100644
--- a/drivers/iommu/intel_irq_remapping.c
+++ b/drivers/iommu/intel_irq_remapping.c
@@ -376,13 +376,13 @@ static int set_msi_sid_cb(struct pci_dev *pdev, u16 
alias, void *opaque)
 {
        struct set_msi_sid_data *data = opaque;
 
+       if (data->count == 0 || PCI_BUS_NUM(alias) == PCI_BUS_NUM(data->alias))
+               data->busmatch_count++;
+
        data->pdev = pdev;
        data->alias = alias;
        data->count++;
 
-       if (PCI_BUS_NUM(alias) == pdev->bus->number)
-               data->busmatch_count++;
-
        return 0;
 }
 
-- 
2.17.1

Reply via email to