[dpdk-dev] [PATCH] bnx2x: set Ethernet address type during transmit for VF's

[dpdk-dev] [PATCH] bnx2x: set Ethernet address type during transmit for VF's

[Chas Williams]

The original was always setting unicast.  While here, clean up some
other references that also point into the Ethernet header.

Signed-off-by: Chas Williams <3chas3@gmail.com>
---
 drivers/net/bnx2x/bnx2x.c     | 23 +++++++++++++++--------
 drivers/net/bnx2x/ecore_hsi.h |  5 +++--
 2 files changed, 18 insertions(+), 10 deletions(-)

diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c
index 76444eb..294711f 100644
--- a/drivers/net/bnx2x/bnx2x.c
+++ b/drivers/net/bnx2x/bnx2x.c
@@ -2177,25 +2177,32 @@ int bnx2x_tx_encap(struct bnx2x_tx_queue *txq, struct rte_mbuf **m_head, int m_p
 		bd_prod = NEXT_TX_BD(bd_prod);
 		if (IS_VF(sc)) {
 			struct eth_tx_parse_bd_e2 *tx_parse_bd;
-			uint8_t *data = rte_pktmbuf_mtod(m0, uint8_t *);
+			const struct ether_hdr *eh = rte_pktmbuf_mtod(m0, struct ether_hdr *);
+			uint8_t mac_type = UNICAST_ADDRESS;
 
 			tx_parse_bd =
 			    &txq->tx_ring[TX_BD(bd_prod, txq)].parse_bd_e2;
+			if (is_multicast_ether_addr(&eh->d_addr)) {
+				if (is_broadcast_ether_addr(&eh->d_addr))
+					mac_type = BROADCAST_ADDRESS;
+				else
+					mac_type = MULTICAST_ADDRESS;
+			}
 			tx_parse_bd->parsing_data =
-			    (1 << ETH_TX_PARSE_BD_E2_ETH_ADDR_TYPE_SHIFT);
+			    (mac_type << ETH_TX_PARSE_BD_E2_ETH_ADDR_TYPE_SHIFT);
 
 			rte_memcpy(&tx_parse_bd->data.mac_addr.dst_hi,
-				   &data[0], 2);
+				   &eh->d_addr.addr_bytes[0], 2);
 			rte_memcpy(&tx_parse_bd->data.mac_addr.dst_mid,
-				   &data[2], 2);
+				   &eh->d_addr.addr_bytes[2], 2);
 			rte_memcpy(&tx_parse_bd->data.mac_addr.dst_lo,
-				   &data[4], 2);
+				   &eh->d_addr.addr_bytes[4], 2);
 			rte_memcpy(&tx_parse_bd->data.mac_addr.src_hi,
-				   &data[6], 2);
+				   &eh->s_addr.addr_bytes[0], 2);
 			rte_memcpy(&tx_parse_bd->data.mac_addr.src_mid,
-				   &data[8], 2);
+				   &eh->s_addr.addr_bytes[2], 2);
 			rte_memcpy(&tx_parse_bd->data.mac_addr.src_lo,
-				   &data[10], 2);
+				   &eh->s_addr.addr_bytes[4], 2);
 
 			tx_parse_bd->data.mac_addr.dst_hi =
 			    rte_cpu_to_be_16(tx_parse_bd->data.mac_addr.dst_hi);
diff --git a/drivers/net/bnx2x/ecore_hsi.h b/drivers/net/bnx2x/ecore_hsi.h
index a4ed9b5..6c11c5a 100644
--- a/drivers/net/bnx2x/ecore_hsi.h
+++ b/drivers/net/bnx2x/ecore_hsi.h
@@ -4029,7 +4029,7 @@ struct double_regpair
 
 
 /*
- * Ethernet address typesm used in ethernet tx BDs
+ * Ethernet address types used in ethernet tx BDs
  */
 enum eth_addr_type
 {
@@ -4037,7 +4037,8 @@ enum eth_addr_type
 	UNICAST_ADDRESS,
 	MULTICAST_ADDRESS,
 	BROADCAST_ADDRESS,
-	MAX_ETH_ADDR_TYPE};
+	MAX_ETH_ADDR_TYPE
+};
 
 
 /*
-- 
2.1.0

Re: [dpdk-dev] [PATCH] bnx2x: set Ethernet address type during transmit for VF's

[Thomas Monjalon]

2015-11-18 15:58, Chas Williams:
> The original was always setting unicast.  While here, clean up some
> other references that also point into the Ethernet header.
> 
> Signed-off-by: Chas Williams <3chas3@gmail.com>

Applied, thanks

Re: [dpdk-dev] [PATCH] bnx2x: set Ethernet address type during transmit for VF's

[Thomas Monjalon]

2015-12-06 23:20, Thomas Monjalon:
> 2015-11-18 15:58, Chas Williams:
> > The original was always setting unicast.  While here, clean up some
> > other references that also point into the Ethernet header.
> > 
> > Signed-off-by: Chas Williams <3chas3@gmail.com>
> 
> Applied, thanks

No sorry, not applied, it needs review.

Re: [dpdk-dev] [PATCH] bnx2x: set Ethernet address type during transmit for VF's

[Harish Patil]

>
>The original was always setting unicast.  While here, clean up some
>other references that also point into the Ethernet header.
>
>Signed-off-by: Chas Williams <3chas3@gmail.com>
>---
> drivers/net/bnx2x/bnx2x.c     | 23 +++++++++++++++--------
> drivers/net/bnx2x/ecore_hsi.h |  5 +++--
> 2 files changed, 18 insertions(+), 10 deletions(-)
>
>diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c
>index 76444eb..294711f 100644
>--- a/drivers/net/bnx2x/bnx2x.c
>+++ b/drivers/net/bnx2x/bnx2x.c
>@@ -2177,25 +2177,32 @@ int bnx2x_tx_encap(struct bnx2x_tx_queue *txq,
>struct rte_mbuf **m_head, int m_p
>               bd_prod = NEXT_TX_BD(bd_prod);
>               if (IS_VF(sc)) {
>                       struct eth_tx_parse_bd_e2 *tx_parse_bd;
>-                      uint8_t *data = rte_pktmbuf_mtod(m0, uint8_t *);
>+                      const struct ether_hdr *eh = rte_pktmbuf_mtod(m0, struct ether_hdr *);
>+                      uint8_t mac_type = UNICAST_ADDRESS;
>
>                       tx_parse_bd =
>                           &txq->tx_ring[TX_BD(bd_prod, txq)].parse_bd_e2;
>+                      if (is_multicast_ether_addr(&eh->d_addr)) {

Minor comment. unlikely() may be used here to keep it consistent with base
driver.

>+                              if (is_broadcast_ether_addr(&eh->d_addr))
>+                                      mac_type = BROADCAST_ADDRESS;
>+                              else
>+                                      mac_type = MULTICAST_ADDRESS;
>+                      }
>                       tx_parse_bd->parsing_data =
>-                          (1 << ETH_TX_PARSE_BD_E2_ETH_ADDR_TYPE_SHIFT);
>+                          (mac_type << ETH_TX_PARSE_BD_E2_ETH_ADDR_TYPE_SHIFT);
>
>                       rte_memcpy(&tx_parse_bd->data.mac_addr.dst_hi,
>-                                 &data[0], 2);
>+                                 &eh->d_addr.addr_bytes[0], 2);
>                       rte_memcpy(&tx_parse_bd->data.mac_addr.dst_mid,
>-                                 &data[2], 2);
>+                                 &eh->d_addr.addr_bytes[2], 2);
>                       rte_memcpy(&tx_parse_bd->data.mac_addr.dst_lo,
>-                                 &data[4], 2);
>+                                 &eh->d_addr.addr_bytes[4], 2);
>                       rte_memcpy(&tx_parse_bd->data.mac_addr.src_hi,
>-                                 &data[6], 2);
>+                                 &eh->s_addr.addr_bytes[0], 2);
>                       rte_memcpy(&tx_parse_bd->data.mac_addr.src_mid,
>-                                 &data[8], 2);
>+                                 &eh->s_addr.addr_bytes[2], 2);
>                       rte_memcpy(&tx_parse_bd->data.mac_addr.src_lo,
>-                                 &data[10], 2);
>+                                 &eh->s_addr.addr_bytes[4], 2);
>
>                       tx_parse_bd->data.mac_addr.dst_hi =
>                           rte_cpu_to_be_16(tx_parse_bd->data.mac_addr.dst_hi);
>diff --git a/drivers/net/bnx2x/ecore_hsi.h b/drivers/net/bnx2x/ecore_hsi.h
>index a4ed9b5..6c11c5a 100644
>--- a/drivers/net/bnx2x/ecore_hsi.h
>+++ b/drivers/net/bnx2x/ecore_hsi.h
>@@ -4029,7 +4029,7 @@ struct double_regpair
>
>
> /*
>- * Ethernet address typesm used in ethernet tx BDs
>+ * Ethernet address types used in ethernet tx BDs
>  */
> enum eth_addr_type
> {
>@@ -4037,7 +4037,8 @@ enum eth_addr_type
>       UNICAST_ADDRESS,
>       MULTICAST_ADDRESS,
>       BROADCAST_ADDRESS,
>-      MAX_ETH_ADDR_TYPE};
>+      MAX_ETH_ADDR_TYPE
>+};
>
>
> /*
>--
>2.1.0
>
>

Acked-by: Harish Patil <harish.patil@qlogic.com>


Thanks,
Harish


________________________________

This message and any attached documents contain information from the sending company or its parent company(s), subsidiaries, divisions or branch offices that may be confidential. If you are not the intended recipient, you may not read, copy, distribute, or use this information. If you have received this transmission in error, please notify the sender immediately by reply e-mail and then delete this message.

Re: [dpdk-dev] [PATCH] bnx2x: set Ethernet address type during transmit for VF's

[Charles (Chas) Williams]

On Sun, 2015-12-06 at 23:34 +0000, Harish Patil wrote:
> >
> >The original was always setting unicast.  While here, clean up some
> >other references that also point into the Ethernet header.
> >
> >Signed-off-by: Chas Williams <3chas3@gmail.com>
> >---
> > drivers/net/bnx2x/bnx2x.c     | 23 +++++++++++++++--------
> > drivers/net/bnx2x/ecore_hsi.h |  5 +++--
> > 2 files changed, 18 insertions(+), 10 deletions(-)
> >
> >diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c
> >index 76444eb..294711f 100644
> >--- a/drivers/net/bnx2x/bnx2x.c
> >+++ b/drivers/net/bnx2x/bnx2x.c
> >@@ -2177,25 +2177,32 @@ int bnx2x_tx_encap(struct bnx2x_tx_queue *txq,
> >struct rte_mbuf **m_head, int m_p
> >               bd_prod = NEXT_TX_BD(bd_prod);
> >               if (IS_VF(sc)) {
> >                       struct eth_tx_parse_bd_e2 *tx_parse_bd;
> >-                      uint8_t *data = rte_pktmbuf_mtod(m0, uint8_t *);
> >+                      const struct ether_hdr *eh = rte_pktmbuf_mtod(m0, struct ether_hdr *);
> >+                      uint8_t mac_type = UNICAST_ADDRESS;
> >
> >                       tx_parse_bd =
> >                           &txq->tx_ring[TX_BD(bd_prod, txq)].parse_bd_e2;
> >+                      if (is_multicast_ether_addr(&eh->d_addr)) {
> 
> Minor comment. unlikely() may be used here to keep it consistent with base
> driver.

It wasn't clear to me that this code path is all that unlikely().
 
> >+                              if (is_broadcast_ether_addr(&eh->d_addr))
> >+                                      mac_type = BROADCAST_ADDRESS;
> >+                              else
> >+                                      mac_type = MULTICAST_ADDRESS;
> >+                      }

Re: [dpdk-dev] [PATCH] bnx2x: set Ethernet address type during transmit for VF's

[Harish Patil]

>
>On Sun, 2015-12-06 at 23:34 +0000, Harish Patil wrote:
>> >
>> >The original was always setting unicast.  While here, clean up some
>> >other references that also point into the Ethernet header.
>> >
>> >Signed-off-by: Chas Williams <3chas3@gmail.com>
>> >---
>> > drivers/net/bnx2x/bnx2x.c     | 23 +++++++++++++++--------
>> > drivers/net/bnx2x/ecore_hsi.h |  5 +++--
>> > 2 files changed, 18 insertions(+), 10 deletions(-)
>> >
>> >diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c
>> >index 76444eb..294711f 100644
>> >--- a/drivers/net/bnx2x/bnx2x.c
>> >+++ b/drivers/net/bnx2x/bnx2x.c
>> >@@ -2177,25 +2177,32 @@ int bnx2x_tx_encap(struct bnx2x_tx_queue *txq,
>> >struct rte_mbuf **m_head, int m_p
>> >               bd_prod = NEXT_TX_BD(bd_prod);
>> >               if (IS_VF(sc)) {
>> >                       struct eth_tx_parse_bd_e2 *tx_parse_bd;
>> >-                      uint8_t *data = rte_pktmbuf_mtod(m0, uint8_t *);
>> >+                      const struct ether_hdr *eh =
>>rte_pktmbuf_mtod(m0, struct ether_hdr *);
>> >+                      uint8_t mac_type = UNICAST_ADDRESS;
>> >
>> >                       tx_parse_bd =
>> >                           &txq->tx_ring[TX_BD(bd_prod,
>>txq)].parse_bd_e2;
>> >+                      if (is_multicast_ether_addr(&eh->d_addr)) {
>>
>> Minor comment. unlikely() may be used here to keep it consistent with
>>base
>> driver.
>
>It wasn't clear to me that this code path is all that unlikely().

Its an optional comment, unlikely() is because fast path traffic is mostly
unicast.
BTW, have you tested the patches? Another question, not related to your
change though.
I guess this block of code does not have to been under the IS_VF() check.

>
>> >+                              if
>>(is_broadcast_ether_addr(&eh->d_addr))
>> >+                                      mac_type = BROADCAST_ADDRESS;
>> >+                              else
>> >+                                      mac_type = MULTICAST_ADDRESS;
>> >+                      }
>
>
>



________________________________

This message and any attached documents contain information from the sending company or its parent company(s), subsidiaries, divisions or branch offices that may be confidential. If you are not the intended recipient, you may not read, copy, distribute, or use this information. If you have received this transmission in error, please notify the sender immediately by reply e-mail and then delete this message.

Re: [dpdk-dev] [PATCH] bnx2x: set Ethernet address type during transmit for VF's

[Charles (Chas) Williams]

On Mon, 2015-12-07 at 17:29 +0000, Harish Patil wrote:
> >
> >On Sun, 2015-12-06 at 23:34 +0000, Harish Patil wrote:
> >> >
> >> >The original was always setting unicast.  While here, clean up some
> >> >other references that also point into the Ethernet header.
> >> >
> >> >Signed-off-by: Chas Williams <3chas3@gmail.com>
> >> >---
> >> > drivers/net/bnx2x/bnx2x.c     | 23 +++++++++++++++--------
> >> > drivers/net/bnx2x/ecore_hsi.h |  5 +++--
> >> > 2 files changed, 18 insertions(+), 10 deletions(-)
> >> >
> >> >diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c
> >> >index 76444eb..294711f 100644
> >> >--- a/drivers/net/bnx2x/bnx2x.c
> >> >+++ b/drivers/net/bnx2x/bnx2x.c
> >> >@@ -2177,25 +2177,32 @@ int bnx2x_tx_encap(struct bnx2x_tx_queue *txq,
> >> >struct rte_mbuf **m_head, int m_p
> >> >               bd_prod = NEXT_TX_BD(bd_prod);
> >> >               if (IS_VF(sc)) {
> >> >                       struct eth_tx_parse_bd_e2 *tx_parse_bd;
> >> >-                      uint8_t *data = rte_pktmbuf_mtod(m0, uint8_t *);
> >> >+                      const struct ether_hdr *eh =
> >>rte_pktmbuf_mtod(m0, struct ether_hdr *);
> >> >+                      uint8_t mac_type = UNICAST_ADDRESS;
> >> >
> >> >                       tx_parse_bd =
> >> >                           &txq->tx_ring[TX_BD(bd_prod,
> >>txq)].parse_bd_e2;
> >> >+                      if (is_multicast_ether_addr(&eh->d_addr)) {
> >>
> >> Minor comment. unlikely() may be used here to keep it consistent with
> >>base
> >> driver.
> >
> >It wasn't clear to me that this code path is all that unlikely().
> 
> Its an optional comment, unlikely() is because fast path traffic is mostly
> unicast.

Multicast traffic isn't all that uncommon.  I also don't see that we gain
much from branch prediction here regardless.  My understanding is that
unlikely() should be used for really unlikely situations since the branch
will not be optimized.

> BTW, have you tested the patches? Another question, not related to your
> change though.

Yes.  This patch is necessary (for VF anyway).  Without it, you need to
manually assign arp addresses.

> I guess this block of code does not have to been under the IS_VF() check.

I don't know.  It is under IS_VF() in the linux driver and it looks like
an attempt to "program" some internal switch on the card.  I don't have
a programmer's guide for this card so it's a complete guess.

> >> >+                              if
> >>(is_broadcast_ether_addr(&eh->d_addr))
> >> >+                                      mac_type = BROADCAST_ADDRESS;
> >> >+                              else
> >> >+                                      mac_type = MULTICAST_ADDRESS;
> >> >+                      }

Re: [dpdk-dev] [PATCH] bnx2x: set Ethernet address type during transmit for VF's

[Bruce Richardson]

On Tue, Dec 08, 2015 at 09:27:05AM -0500, Charles (Chas) Williams wrote:
> On Mon, 2015-12-07 at 17:29 +0000, Harish Patil wrote:
> > >
> > >On Sun, 2015-12-06 at 23:34 +0000, Harish Patil wrote:
> > >> >
> > >> >The original was always setting unicast.  While here, clean up some
> > >> >other references that also point into the Ethernet header.
> > >> >
> > >> >Signed-off-by: Chas Williams <3chas3@gmail.com>
> > >> >---
> > >> > drivers/net/bnx2x/bnx2x.c     | 23 +++++++++++++++--------
> > >> > drivers/net/bnx2x/ecore_hsi.h |  5 +++--
> > >> > 2 files changed, 18 insertions(+), 10 deletions(-)
> > >> >
> > >> >diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c
> > >> >index 76444eb..294711f 100644
> > >> >--- a/drivers/net/bnx2x/bnx2x.c
> > >> >+++ b/drivers/net/bnx2x/bnx2x.c
> > >> >@@ -2177,25 +2177,32 @@ int bnx2x_tx_encap(struct bnx2x_tx_queue *txq,
> > >> >struct rte_mbuf **m_head, int m_p
> > >> >               bd_prod = NEXT_TX_BD(bd_prod);
> > >> >               if (IS_VF(sc)) {
> > >> >                       struct eth_tx_parse_bd_e2 *tx_parse_bd;
> > >> >-                      uint8_t *data = rte_pktmbuf_mtod(m0, uint8_t *);
> > >> >+                      const struct ether_hdr *eh =
> > >>rte_pktmbuf_mtod(m0, struct ether_hdr *);
> > >> >+                      uint8_t mac_type = UNICAST_ADDRESS;
> > >> >
> > >> >                       tx_parse_bd =
> > >> >                           &txq->tx_ring[TX_BD(bd_prod,
> > >>txq)].parse_bd_e2;
> > >> >+                      if (is_multicast_ether_addr(&eh->d_addr)) {
> > >>
> > >> Minor comment. unlikely() may be used here to keep it consistent with
> > >>base
> > >> driver.
> > >
> > >It wasn't clear to me that this code path is all that unlikely().
> > 
> > Its an optional comment, unlikely() is because fast path traffic is mostly
> > unicast.
> 
> Multicast traffic isn't all that uncommon.  I also don't see that we gain
> much from branch prediction here regardless.  My understanding is that
> unlikely() should be used for really unlikely situations since the branch
> will not be optimized.

+1 to this. 

Let the cpu branch predictor do the work at run-time picking the
correct leg of the branch. likely()/unlikely() should only ever be used for
things like fatal error conditions where it doesn't matter how optimized the
unlikely branch part is. For normal code where there is no error, don't try and
optimize it using likely()/unlikely() unless you can prove there is a definite
performance improvement by doing so.

/Bruce

Re: [dpdk-dev] [PATCH] bnx2x: set Ethernet address type during transmit for VF's

[Harish Patil]

>On Tue, Dec 08, 2015 at 09:27:05AM -0500, Charles (Chas) Williams wrote:
>> On Mon, 2015-12-07 at 17:29 +0000, Harish Patil wrote:
>> > >
>> > >On Sun, 2015-12-06 at 23:34 +0000, Harish Patil wrote:
>> > >> >
>> > >> >The original was always setting unicast.  While here, clean up
>>some
>> > >> >other references that also point into the Ethernet header.
>> > >> >
>> > >> >Signed-off-by: Chas Williams <3chas3@gmail.com>
>> > >> >---
>> > >> > drivers/net/bnx2x/bnx2x.c     | 23 +++++++++++++++--------
>> > >> > drivers/net/bnx2x/ecore_hsi.h |  5 +++--
>> > >> > 2 files changed, 18 insertions(+), 10 deletions(-)
>> > >> >
>> > >> >diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c
>> > >> >index 76444eb..294711f 100644
>> > >> >--- a/drivers/net/bnx2x/bnx2x.c
>> > >> >+++ b/drivers/net/bnx2x/bnx2x.c
>> > >> >@@ -2177,25 +2177,32 @@ int bnx2x_tx_encap(struct bnx2x_tx_queue
>>*txq,
>> > >> >struct rte_mbuf **m_head, int m_p
>> > >> >               bd_prod = NEXT_TX_BD(bd_prod);
>> > >> >               if (IS_VF(sc)) {
>> > >> >                       struct eth_tx_parse_bd_e2 *tx_parse_bd;
>> > >> >-                      uint8_t *data = rte_pktmbuf_mtod(m0,
>>uint8_t *);
>> > >> >+                      const struct ether_hdr *eh =
>> > >>rte_pktmbuf_mtod(m0, struct ether_hdr *);
>> > >> >+                      uint8_t mac_type = UNICAST_ADDRESS;
>> > >> >
>> > >> >                       tx_parse_bd =
>> > >> >                           &txq->tx_ring[TX_BD(bd_prod,
>> > >>txq)].parse_bd_e2;
>> > >> >+                      if (is_multicast_ether_addr(&eh->d_addr)) {
>> > >>
>> > >> Minor comment. unlikely() may be used here to keep it consistent
>>with
>> > >>base
>> > >> driver.
>> > >
>> > >It wasn't clear to me that this code path is all that unlikely().
>> >
>> > Its an optional comment, unlikely() is because fast path traffic is
>>mostly
>> > unicast.
>>
>> Multicast traffic isn't all that uncommon.  I also don't see that we
>>gain
>> much from branch prediction here regardless.  My understanding is that
>> unlikely() should be used for really unlikely situations since the
>>branch
>> will not be optimized.
>
>+1 to this.
>
>Let the cpu branch predictor do the work at run-time picking the
>correct leg of the branch. likely()/unlikely() should only ever be used
>for
>things like fatal error conditions where it doesn't matter how optimized
>the
>unlikely branch part is. For normal code where there is no error, don't
>try and
>optimize it using likely()/unlikely() unless you can prove there is a
>definite
>performance improvement by doing so.
>
>/Bruce
>

I agree too, lets leave it that way. As I said its optional, to be same as
linux base driver which has:
3809         /* set flag according to packet type (UNICAST_ADDRESS is
default)*/
   3810         if (unlikely(is_multicast_ether_addr(eth->h_dest))) {



Thanks,
Harish


________________________________

This message and any attached documents contain information from the sending company or its parent company(s), subsidiaries, divisions or branch offices that may be confidential. If you are not the intended recipient, you may not read, copy, distribute, or use this information. If you have received this transmission in error, please notify the sender immediately by reply e-mail and then delete this message.