xhci: fix off by one error in TRB DMA address boundary check
Mathias Nyman [Mon, 3 Aug 2015 13:07:48 +0000 (16:07 +0300)]
We need to check that a TRB is part of the current segment
before calculating its DMA address.

Previously a ring segment didn't use a full memory page, and every
new ring segment got a new memory page, so the off by one
error in checking the upper bound was never seen.

Now that we use a full memory page, 256 TRBs (4096 bytes), the off by one
didn't catch the case when a TRB was the first element of the next segment.

This is triggered if the virtual memory pages for a ring segment are
next to each in increasing order where the ring buffer wraps around and
causes errors like:

[  106.398223] xhci_hcd 0000:00:14.0: ERROR Transfer event TRB DMA ptr not part of current TD ep_index 0 comp_code 1
[  106.398230] xhci_hcd 0000:00:14.0: Looking for event-dma fffd3000 trb-start fffd4fd0 trb-end fffd5000 seg-start fffd4000 seg-end fffd4ff0

The trb-end address is one outside the end-seg address.

Bug 1730718

Change-Id: I27bff8497493ea94b357184bd321e7dc478a0a1a
Cc: <stable@vger.kernel.org>
Tested-by: Arkadiusz Miƛkiewicz <arekm@maven.pl>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Ajay Gupta <ajayg@nvidia.com>
Reviewed-on: http://git-master/r/1020811
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vinayak Pane <vpane@nvidia.com>

drivers/usb/host/xhci-ring.c

index 06388f5..2088258 100644 (file)
@@ -86,7 +86,7 @@ dma_addr_t xhci_trb_virt_to_dma(struct xhci_segment *seg,
                return 0;
        /* offset in TRBs */
        segment_offset = trb - seg->trbs;
-       if (segment_offset > TRBS_PER_SEGMENT)
+       if (segment_offset >= TRBS_PER_SEGMENT)
                return 0;
        return seg->dma + (segment_offset * sizeof(*trb));
 }