Thread (27 messages) 27 messages, 4 authors, 2016-06-15

Re: [PATCH v2] mbuf: new flag when Vlan is stripped

From: Ananyev, Konstantin <hidden>
Date: 2016-06-13 14:42:38

Hi Olivier,
The behavior of PKT_RX_VLAN_PKT was not very well defined, resulting in
PMDs not advertising the same flags in similar conditions.

Following discussion in [1], introduce 2 new flags PKT_RX_VLAN_STRIPPED
and PKT_RX_QINQ_STRIPPED that are better defined:

  PKT_RX_VLAN_STRIPPED: a vlan has been stripped by the hardware and its
  tci is saved in mbuf->vlan_tci. This can only happen if vlan stripping
  is enabled in the RX configuration of the PMD.

For now, the old flag PKT_RX_VLAN_PKT is kept but marked as deprecated.
It should be removed from applications and PMDs in a future revision.
I am not sure it has to be deprecated & removed.
ixgbe (and igb as I can read the specs) devices can provide information is that
a vlan packet or not even when vlan stripping is disabled. 
Right now ixgbe PMD do carry thins information to the user,
and I suppose igb could be improved to carry it too.
So obviously we need a way to pass that information to the upper layer.
I remember it was a discussion about introducing new packet_type
instead of ol_flag value PKT_RX_VLAN_PKT.
But right now it is not there, and again I don't know how easy it would be to replace
one with another without performance considering that packet_type is not supported
now by ixgbe vRX.
If we would be able to replace it, then yes we can deprecate and drop the   PKT_RX_VLAN_PKT.
But till then, I think we'd better keep it.
quoted hunk ↗ jump to hunk
diff --git a/drivers/net/ixgbe/ixgbe_rxtx_vec.c b/drivers/net/ixgbe/ixgbe_rxtx_vec.c
index e97ea82..d895bf1 100644
--- a/drivers/net/ixgbe/ixgbe_rxtx_vec.c
+++ b/drivers/net/ixgbe/ixgbe_rxtx_vec.c
@@ -140,10 +140,9 @@ ixgbe_rxq_rearm(struct ixgbe_rx_queue *rxq)
  */
 #ifdef RTE_IXGBE_RX_OLFLAGS_ENABLE

-#define VTAG_SHIFT     (3)
-
 static inline void
-desc_to_olflags_v(__m128i descs[4], struct rte_mbuf **rx_pkts)
+desc_to_olflags_v(__m128i descs[4], uint8_t vlan_flags,
+	struct rte_mbuf **rx_pkts)
 {
 	__m128i ptype0, ptype1, vtag0, vtag1;
 	union {
@@ -151,12 +150,6 @@ desc_to_olflags_v(__m128i descs[4], struct rte_mbuf **rx_pkts)
 		uint64_t dword;
 	} vol;

-	/* pkt type + vlan olflags mask */
-	const __m128i pkttype_msk = _mm_set_epi16(
-			0x0000, 0x0000, 0x0000, 0x0000,
-			PKT_RX_VLAN_PKT, PKT_RX_VLAN_PKT,
-			PKT_RX_VLAN_PKT, PKT_RX_VLAN_PKT);
-
 	/* mask everything except rss type */
 	const __m128i rsstype_msk = _mm_set_epi16(
 			0x0000, 0x0000, 0x0000, 0x0000,
@@ -168,6 +161,19 @@ desc_to_olflags_v(__m128i descs[4], struct rte_mbuf **rx_pkts)
 			PKT_RX_RSS_HASH, 0, PKT_RX_RSS_HASH, 0,
 			PKT_RX_RSS_HASH, PKT_RX_RSS_HASH, PKT_RX_RSS_HASH, 0);

+	/* mask everything except vlan present bit */
+	const __m128i vlan_msk = _mm_set_epi16(
+			0x0000, 0x0000,
+			0x0000, 0x0000,
+			IXGBE_RXD_STAT_VP, IXGBE_RXD_STAT_VP,
+			IXGBE_RXD_STAT_VP, IXGBE_RXD_STAT_VP);
+	/* map vlan present (0x8) to ol_flags */
+	const __m128i vlan_map = _mm_set_epi8(
+		0, 0, 0, 0,
+		0, 0, 0, vlan_flags,
+		0, 0, 0, 0,
+		0, 0, 0, 0);
+
 	ptype0 = _mm_unpacklo_epi16(descs[0], descs[1]);
 	ptype1 = _mm_unpacklo_epi16(descs[2], descs[3]);
 	vtag0 = _mm_unpackhi_epi16(descs[0], descs[1]);
@@ -178,8 +184,8 @@ desc_to_olflags_v(__m128i descs[4], struct rte_mbuf **rx_pkts)
 	ptype0 = _mm_shuffle_epi8(rss_flags, ptype0);

 	vtag1 = _mm_unpacklo_epi32(vtag0, vtag1);
-	vtag1 = _mm_srli_epi16(vtag1, VTAG_SHIFT);
-	vtag1 = _mm_and_si128(vtag1, pkttype_msk);
+	vtag1 = _mm_and_si128(vtag1, vlan_msk);
+	vtag1 = _mm_shuffle_epi8(vlan_map, vtag1);

 	vtag1 = _mm_or_si128(ptype0, vtag1);
 	vol.dword = _mm_cvtsi128_si64(vtag1);
@@ -221,6 +227,7 @@ _recv_raw_pkts_vec(struct ixgbe_rx_queue *rxq, struct rte_mbuf **rx_pkts,
 				0, 0            /* ignore pkt_type field */
 			);
 	__m128i dd_check, eop_check;
+	uint8_t vlan_flags;

 	/* nb_pkts shall be less equal than RTE_IXGBE_MAX_RX_BURST */
 	nb_pkts = RTE_MIN(nb_pkts, RTE_IXGBE_MAX_RX_BURST);
@@ -270,6 +277,11 @@ _recv_raw_pkts_vec(struct ixgbe_rx_queue *rxq, struct rte_mbuf **rx_pkts,
 	 */
 	sw_ring = &rxq->sw_ring[rxq->rx_tail];

+	/* ensure these 2 flags are in the lower 8 bits */
+	RTE_BUILD_BUG_ON(((PKT_RX_VLAN_PKT | PKT_RX_VLAN_STRIPPED) &
+			0xffffffffffffff00ULL) != 0ULL);

I suppose your need here:
RTE_BUILD_BUG_ON(((PKT_RX_VLAN_PKT | PKT_RX_VLAN_STRIPPED) & UINT8_MAX) == 
PKT_RX_VLAN_PKT | PKT_RX_VLAN_STRIPPED));

To make sure both flags are inside first 8 bits?

			
+	vlan_flags = rxq->vlan_flags & 0xff;
+
Probably better to do that check/ AND at setup  phase, not run-time?
As a nit: s/0xff/UINT8_MAX/.
Konstantin
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help