d/numam-dpdk
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

net/i40e: fix out of order Rx read

Browse Source 
In vPMD, when we load Rx desc with _mm_loadu_si128, the volatile
modifier will be cast away, allowing the compiler to reorder the
load instructions.

The Rx recv function's correctness is relying on these load
instructions following a strict sequence, reading the descriptors
in reverse order, so we add compiler barrier to prevent compiler
reorder.

Fixes: 9ed94e5bb0 ("i40e: add vector Rx")

Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
This commit is contained in:
Qi Zhang 2016-10-18 02:29:40 +08:00 committed by Bruce Richardson
parent 7832fe9819
commit ff6edd3886
1 changed files with 3 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
drivers/net/i40e/i40e_rxtx_vec_sse.c
View File

@ -306,6 +306,7 @@ _recv_raw_pkts_vec(struct i40e_rx_queue *rxq, struct rte_mbuf **rx_pkts,
/* Read desc statuses backwards to avoid race condition */
/* A.1 load 4 pkts desc */
descs[3] = _mm_loadu_si128((__m128i *)(rxdp + 3));
rte_compiler_barrier();
/* B.2 copy 2 mbuf point into rx_pkts */
_mm_storeu_si128((__m128i *)&rx_pkts[pos], mbp1);
@ -314,8 +315,10 @@ _recv_raw_pkts_vec(struct i40e_rx_queue *rxq, struct rte_mbuf **rx_pkts,
mbp2 = _mm_loadu_si128((__m128i *)&sw_ring[pos+2]);
descs[2] = _mm_loadu_si128((__m128i *)(rxdp + 2));
rte_compiler_barrier();
/* B.1 load 2 mbuf point */
descs[1] = _mm_loadu_si128((__m128i *)(rxdp + 1));
rte_compiler_barrier();
descs[0] = _mm_loadu_si128((__m128i *)(rxdp));
/* B.2 copy 2 mbuf point into rx_pkts */