[dpdk-dev] ixgbe_rx_scan_hw_ring: Fix initializing id and hash fields in flow director mode.
diff mbox

Message ID 1408520788-20380-1-git-send-email-pawelx.wodkowski@intel.com
State Accepted, archived
Headers show

Commit Message

Wodkowski, PawelX Aug. 20, 2014, 7:46 a.m. UTC
When Flow Director was used together with bulk alloc, id and hash was swapped
when packet matches flow director filter due to improper fdir field initialization.


Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
---
 lib/librte_pmd_ixgbe/ixgbe_rxtx.c |   24 +++++++++++++++++-------
 1 file changed, 17 insertions(+), 7 deletions(-)

Comments

Thomas Monjalon Aug. 29, 2014, 3:41 p.m. UTC | #1
Hi,

2014-08-20 08:46, Pawel Wodkowski:
> When Flow Director was used together with bulk alloc, id and hash was swapped
> when packet matches flow director filter due to improper fdir field initialization.
> 
> Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>

I'd really like someone else review this touchy patch for integration
in 1.7.1.

Thanks
Wodkowski, PawelX Sept. 8, 2014, 7:18 a.m. UTC | #2
> -----Original Message-----
> From: Thomas Monjalon [mailto:thomas.monjalon@6wind.com]
> Sent: Friday, August 29, 2014 17:41
> To: dev@dpdk.org
> Cc: Wodkowski, PawelX
> Subject: Re: [dpdk-dev] [PATCH] ixgbe_rx_scan_hw_ring: Fix initializing id and
> hash fields in flow director mode.
> 
> Hi,
> 
> 2014-08-20 08:46, Pawel Wodkowski:
> > When Flow Director was used together with bulk alloc, id and hash was
> swapped
> > when packet matches flow director filter due to improper fdir field
> initialization.
> >
> > Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
> 
> I'd really like someone else review this touchy patch for integration
> in 1.7.1.
> 
> Thanks
> --
> Thomas

Reviewed-by: Helin Zhang <helin.zhang@intel.com>
Reviewed-by: Jingjing Wu <jingjing.wu@intel.com>
Bruce Richardson Sept. 25, 2014, 4:07 p.m. UTC | #3
On Mon, Sep 08, 2014 at 07:18:10AM +0000, Wodkowski, PawelX wrote:
> > -----Original Message-----
> > From: Thomas Monjalon [mailto:thomas.monjalon@6wind.com]
> > Sent: Friday, August 29, 2014 17:41
> > To: dev@dpdk.org
> > Cc: Wodkowski, PawelX
> > Subject: Re: [dpdk-dev] [PATCH] ixgbe_rx_scan_hw_ring: Fix initializing id and
> > hash fields in flow director mode.
> > 
> > Hi,
> > 
> > 2014-08-20 08:46, Pawel Wodkowski:
> > > When Flow Director was used together with bulk alloc, id and hash was
> > swapped
> > > when packet matches flow director filter due to improper fdir field
> > initialization.
> > >
> > > Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
> > 
> > I'd really like someone else review this touchy patch for integration
> > in 1.7.1.
> > 
> > Thanks
> > --
> > Thomas
> 
> Reviewed-by: Helin Zhang <helin.zhang@intel.com>
> Reviewed-by: Jingjing Wu <jingjing.wu@intel.com>

Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Thomas Monjalon Sept. 25, 2014, 8:57 p.m. UTC | #4
> > > > When Flow Director was used together with bulk alloc, id and hash was
> > > > swapped when packet matches flow director filter due to improper fdir
> > > > field initialization.
> > > >
> > > > Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
> > 
> > Reviewed-by: Helin Zhang <helin.zhang@intel.com>
> > Reviewed-by: Jingjing Wu <jingjing.wu@intel.com>
> 
> Acked-by: Bruce Richardson <bruce.richardson@intel.com>

Applied with some changes to fit with recent mbug refactoring.

Thanks

Patch
diff mbox

diff --git a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
index aaf46d4..796e5a4 100644
--- a/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
+++ b/lib/librte_pmd_ixgbe/ixgbe_rxtx.c
@@ -902,6 +902,7 @@  ixgbe_rx_scan_hw_ring(struct igb_rx_queue *rxq)
 	struct igb_rx_entry *rxep;
 	struct rte_mbuf *mb;
 	uint16_t pkt_len;
+	uint16_t pkt_flags;
 	int s[LOOK_AHEAD], nb_dd;
 	int i, j, nb_rx = 0;
 
@@ -935,21 +936,30 @@  ixgbe_rx_scan_hw_ring(struct igb_rx_queue *rxq)
 		/* Translate descriptor info to mbuf format */
 		for (j = 0; j < nb_dd; ++j) {
 			mb = rxep[j].mbuf;
-			pkt_len = (uint16_t)(rxdp[j].wb.upper.length -
-							rxq->crc_len);
+			pkt_len = (uint16_t)(rxdp[j].wb.upper.length - rxq->crc_len);
 			mb->pkt.data_len = pkt_len;
 			mb->pkt.pkt_len = pkt_len;
-			mb->pkt.vlan_macip.f.vlan_tci = rxdp[j].wb.upper.vlan;
-			mb->pkt.hash.rss = rxdp[j].wb.lower.hi_dword.rss;
+			mb->pkt.vlan_macip.f.vlan_tci =
+				rte_le_to_cpu_16(rxdp[j].wb.upper.vlan);
 
 			/* convert descriptor fields to rte mbuf flags */
-			mb->ol_flags  = rx_desc_hlen_type_rss_to_pkt_flags(
+			pkt_flags  = rx_desc_hlen_type_rss_to_pkt_flags(
 					rxdp[j].wb.lower.lo_dword.data);
 			/* reuse status field from scan list */
-			mb->ol_flags = (uint16_t)(mb->ol_flags |
+			pkt_flags = (uint16_t)(pkt_flags |
 					rx_desc_status_to_pkt_flags(s[j]));
-			mb->ol_flags = (uint16_t)(mb->ol_flags |
+			pkt_flags = (uint16_t)(pkt_flags |
 					rx_desc_error_to_pkt_flags(s[j]));
+			mb->ol_flags = pkt_flags;
+
+			if (likely(pkt_flags & PKT_RX_RSS_HASH))
+				mb->pkt.hash.rss = rxdp[j].wb.lower.hi_dword.rss;
+			else if (pkt_flags & PKT_RX_FDIR) {
+				mb->pkt.hash.fdir.hash =
+					(uint16_t)((rxdp[j].wb.lower.hi_dword.csum_ip.csum)
+						& IXGBE_ATR_HASH_MASK);
+				mb->pkt.hash.fdir.id = rxdp[j].wb.lower.hi_dword.csum_ip.ip_id;
+			}
 		}
 
 		/* Move mbuf pointers from the S/W ring to the stage */