Commit 52c6912f authored by Brian King's avatar Brian King Committed by Jeff Kirsher

i40e: Use smp_rmb rather than read_barrier_depends

The original issue being fixed in this patch was seen with the ixgbe
driver, but the same issue exists with i40e 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: default avatarBrian King <brking@linux.vnet.ibm.com>
Acked-by: default avatarJesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: default avatarAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: default avatarJeff Kirsher <jeffrey.t.kirsher@intel.com>
parent 0a9a17e3
...@@ -3966,7 +3966,7 @@ static bool i40e_clean_fdir_tx_irq(struct i40e_ring *tx_ring, int budget) ...@@ -3966,7 +3966,7 @@ static bool i40e_clean_fdir_tx_irq(struct i40e_ring *tx_ring, int budget)
break; break;
/* prevent any other reads prior to eop_desc */ /* prevent any other reads prior to eop_desc */
read_barrier_depends(); smp_rmb();
/* if the descriptor isn't done, no work yet to do */ /* if the descriptor isn't done, no work yet to do */
if (!(eop_desc->cmd_type_offset_bsz & if (!(eop_desc->cmd_type_offset_bsz &
......
...@@ -759,7 +759,7 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi, ...@@ -759,7 +759,7 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi,
break; break;
/* prevent any other reads prior to eop_desc */ /* prevent any other reads prior to eop_desc */
read_barrier_depends(); smp_rmb();
i40e_trace(clean_tx_irq, tx_ring, tx_desc, tx_buf); i40e_trace(clean_tx_irq, tx_ring, tx_desc, tx_buf);
/* we have caught up to head, no work left to do */ /* we have caught up to head, no work left to do */
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment