IB/srp: Fix infinite loop when FMR sg[0].offset != 0

Avoid that mapping an sg-list in which the first element has a
non-zero offset triggers an infinite loop when using FMR. This
patch makes the FMR mapping code similar to that of ib_sg_to_pages().

Note: older Mellanox HCAs do not support non-zero offsets for FMR.
See also commit 8c4037b501ac ("IB/srp: always avoid non-zero offsets
into an FMR").

Reported-by: Alex Estrin <alex.estrin@intel.com>
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>

1 files changed, 5 insertions, 3 deletions

diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c
index 60031b9159d9..d980fb458ad4 100644
--- a/drivers/infiniband/ulp/srp/ib_srp.c
+++ b/drivers/infiniband/ulp/srp/ib_srp.c
@@ -1402,7 +1402,9 @@ static int srp_map_sg_entry(struct srp_map_state *state,
 
         while (dma_len) {
                 unsigned offset = dma_addr & ~dev->mr_page_mask;
-                if (state->npages == dev->max_pages_per_mr || offset != 0) {
+
+                if (state->npages == dev->max_pages_per_mr ||
+                    (state->npages > 0 && offset != 0)) {
                         ret = srp_map_finish_fmr(state, ch);
                         if (ret)
                                 return ret;
@@ -1419,12 +1421,12 @@ static int srp_map_sg_entry(struct srp_map_state *state,
         }
 
         /*
-         * If the last entry of the MR wasn't a full page, then we need to
+         * If the end of the MR is not on a page boundary then we need to
          * close it out and start a new one -- we can only merge at page
          * boundaries.
          */
         ret = 0;
-        if (len != dev->mr_page_size)
+        if ((dma_addr & ~dev->mr_page_mask) != 0)
                 ret = srp_map_finish_fmr(state, ch);
         return ret;
 }