fm10k: Use smp_rmb rather than read_barrier_depends
authorBrian King <brking@linux.vnet.ibm.com>
Fri, 17 Nov 2017 17:05:48 +0000 (11:05 -0600)
committerJeff Kirsher <jeffrey.t.kirsher@intel.com>
Wed, 22 Nov 2017 07:48:39 +0000 (23:48 -0800)
The original issue being fixed in this patch was seen with the ixgbe
driver, but the same issue exists with fm10k as well, as the code is
very similar. read_barrier_depends is not sufficient to ensure
loads following it are not speculatively loaded out of order
by the CPU, which can result in stale data being loaded, causing
potential system crashes.

Cc: stable <stable@vger.kernel.org>
Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
drivers/net/ethernet/intel/fm10k/fm10k_main.c

index dbd69310f263ea5e4815a020bd1ec8a6113a85e5..538b42d5c1874870f865b7b217a328fb04e2b252 100644 (file)
@@ -1231,7 +1231,7 @@ static bool fm10k_clean_tx_irq(struct fm10k_q_vector *q_vector,
                        break;
 
                /* prevent any other reads prior to eop_desc */
                        break;
 
                /* prevent any other reads prior to eop_desc */
-               read_barrier_depends();
+               smp_rmb();
 
                /* if DD is not set pending work has not been completed */
                if (!(eop_desc->flags & FM10K_TXD_FLAG_DONE))
 
                /* if DD is not set pending work has not been completed */
                if (!(eop_desc->flags & FM10K_TXD_FLAG_DONE))